Studies on Iran and The Caucasus

Presented to Prof. Garnik S. Asatrian on the Occasion of his 60th Birthday

Studies on Iran and The Caucasus
Presented to Prof. Garnik S. Asatrian on the
Occasion of his 60th Birthday

Edited by

Uwe Bläsing, Victoria Arakelova and Matthias Weinreich

With the assistance of

Khachik Gevorgian

LEIDEN | BOSTON
Library of Congress Control Number: 2015941783

This publication has been typeset in the multilingual ‘Brill’ typeface. With over 5,100 characters covering
Latin, ipa, Greek, and Cyrillic, this typeface is especially suitable for use in the humanities.
For more information, please see www.brill.com/brill-typeface.

isbn 978-90-04-30201-3 (hardback)

isbn 978-90-04-30206-8 (e-book)

Copyright 2015 by Koninklijke Brill nv, Leiden, The Netherlands.

Koninklijke Brill NV incorporates the imprints Brill, Brill Hes & De Graaf, Brill Nijhoff, Brill Rodopi and
Hotei Publishing.
All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system,
or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise,
without prior written permission from the publisher.
Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill nv provided
that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive,
Suite 910, Danvers, ma 01923, usa. Fees are subject to change.

This book is printed on acid-free paper.

 TABLE OF CONTENTS

INTRODUCTION
by Prof. Dr. Boghos Zekian
GARNIK ASATRIAN – A Scholar, a Master, a Friend
–1–

HISTORY AND TEXTS

I. Early Mediaeval Period

Marco Bais – “Like a Flame Through the Reeds”: An Iranian Image in the
Buzandaran Patmut‘iwnk‘ . . . . . . . . . . . . . . . 9
Jost Gippert – The “Bun-Turks” in Ancient Georgia . . . . . . . . 25
Dan Shapira – On the Relative Value of Armenian Sources for the Khazar
Studies: The Case of the Siege of Tbilisi . . . . . . . . . . . 45
Giusto Traina – Some Remarks on the Inscription of Maris, Casit filius
(Classical-Oriental Notes, 9) . . . . . . . . . . . . . . 63

II. Late Mediaeval Period

Kaveh Farrokh – The Military Campaigns of Shah Abbas I in Azerbaijan
and the Caucasus (1603-1618) . . . . . . . . . . . . . . 75
Aldo Ferrari – Persia and Persians in Raffi’s Xamsayi Melikʻutiwnnerə . . 97
Hirotake Maeda – New Information on the History of the Caucasus in the
Third Volume of Afzal al-tavarikh . . . . . . . . . . . . . 107
Irène Natchkebia – Unrealized Project: Rousseaus’ Plan of Franco-Persian
Trade in the Context of the Indian Expedition (1807) . . . . . . . 115
Roman Smbatian – Nadir’s Religious Policy Towards Armenians . . . . 131

RELIGION AND ETHNOGRAPHY

Victoria Arakelova – The Song Unveiling the Hidden . . . . . . . 139

 Koninklijke Brill NV, Leiden, 2015
Viacheslav A. Chirikba – Between Christianity and Islam: Heathen Heritage
in the Caucasus . . . . . . . . . . . . . . . . . . 145
Matteo Compareti – Armenian Pre-Christian Divinities: Some Evidence
from the History of Art and Archaeological Investigation . . . . 193
Peter Nicolaus – The Taming of the Fairies . . . . . . . . . . 205
Antonio Panaino – The Classification of Astral Bodies in the Framework
of a Historical Survey of Iranian Traditions . . . . . . . . . . 229
Vahe S. Boyajian – From Muscat to Sarhadd: Remarks on gwātī Healing
Ritual within the Social Context . . . . . . . . . . . . . 245

LINGUISTICS

Uwe Bläsing – Georgische Gewächse auf türkischer Erde: Ein Beitrag zur
Phytonomie in Nordostanatolien . . . . . . . . . . . . . 253
Johnny Cheung – The Persian Verbal Suffixes -ān and -andeh (-andag) . . 271
Claudia A. Ciancaglini – Allomorphic Variability in the Middle Persian
Continuants of the Old Iranian suffix *-ka- . . . . . . . . . . 291
Desmond Durkin-Meisterernst – Vowel Length in Middle Persian Verbal
Endings . . . . . . . . . . . . . . . . . . . . . 309
Vladimir Livshits – Some Khwarezmian Names . . . . . . . . . 317
Ela Filippone – Kurdish bažn, Persian bašn and Other Iranian Cognates . 325
Adriano V. Rossi – Once Again on Iranian *kund . . . . . . . . 351
James R. Russell – A Note on Armenian hrmštk-el . . . . . . . . 365
Wolfgang Schulze – Aspects of Udi-Iranian Language Contact . . . . 373
Martin Schwarz – Armenian varkaparazi and Its Iranian Background . . 403
Donald Stilo – The Poligenetic Origins of the Northern Talishi Language . 411
Matthias Weinreich – Not only in the Caucasus: Ethno-linguistic Diversity
on the Roof of the World . . . . . . . . . . . . . . . 455

LITERATURE AND FOLKLORE

Anna Krasnawolska – Hedayat’s Nationalism and His Concepts of Folklore 473

Mikhail Pelevin – Early Specimens of Pashto Folklore . . . . . . . 479
Nagihan Haliloğlu – Activist, Professional, Family Man: Masculinities in
Marjane Satrapi’s Work . . . . . . . . . . . . . . . . 495
Khachik Gevorgian – On the Interpretation of the Term “Futuwwa” in
Persian Fotovvatnamehs . . . . . . . . . . . . . . . 507

HISTORICO-POLITICAL ISSUES

Çakır Ceyhan Suvari, Elif Kanca – The Alevi Discourse in Turkey . . . . 511
Pascal Kluge – Turkey’s Border with Armenia: Obstacle and Chance for
Turkish Politics . . . . . . . . . . . . . . . . . . 521
Irina Morozova – On the Causes of Socialism’s Deconstruction:
Conventional Debates and Popular Rhetoric in Contemporary
Kazakhstan and Mongolia . . . . . . . . . . . . . . . 551
Caspar ten Dam – The Limitations of Military Psychology: Combat-stress
and Violence-values among the Chechens and Albanians . . . . . 577
Garry W. Trompf – The Ararat Factor: Moral Basics in Western Political
Theory from Isaac Newton to John Stuart Mill . . . . . . . . . 629
Eberhard Werner – Communication and the Oral-Aural Traditions of an
East-Anatolian Ethnicity: What us Stories tell! . . . . . . . . . 667

SELECT BIBLIOGRAPHY
of Prof. Garnik Asatrian
– 693 –
 GARNIK ASATRIAN
A Scholar, a Master, a Friend

Իրանի արեւն է իմ հոգում

The sun of Iran is in my soul
Yeghishe Charents

While I consider him a great scholar and a true master, Garnik Asatrian is and
remains for me, above all, a friend in every sense of the word. This is why I was
initially hesitant to accept the request to write the preface for the Festschrift, yet
felt very honoured by the invitation.
I still remember the first time I met Garnik Asatrian in 1985. It was during the
noon banquet at the 3rd International Symposium on Armenian Linguistics. This
first occasion was rather superficial, when people exchange only a few niceties.
Some years had passed since when we met again, this time at the entrance of the
National Academy of Sciences, which was then the Academy of Sciences of
ArmSSR (Armenian Soviet Socialist Republic). He shared his desire to establish
academic contacts with Italian Iranological centres. In the space of half an hour,
there standing in the Academy’s court, he presented an outline of his projects for
Armenia. Ambitious, indeed, he shared his vision to establish contacts through-
out the world. I had not yet known him very well. While his speech was inter-
rupted by my short and somehow hesitant questions, my thoughts alternated be-
tween admiration, amazement, and suspicions of utopia. I promised him, how-
ever, to do my best to help him in establishing contacts.
Our subsequent meetings took place at the height of the Karabagh Move-
ment. Garnik was, among my younger friends, one of those very few, indeed,
who did not give up their scientific career, even if he too followed, in his own
way, the Movement. Today I can better appreciate Garnik’s stance, when I look
at that period with the distance of time. Political passions were high during
those years, but they caused a deep void in the scientific and scholarly potential
of Armenia, one of the traditionally greatest sources of richness and power in
this country.

 Koninklijke Brill NV, Leiden, 2015

2 Boghos Levon Zekiyan

I never asked Garnik about the reasons for his choice. I know, however, over
the years his deepest concerns for human rights and justice, but also his view of
the longue durée. Garnik engaged boldly his own way. Garnik developed his own
vision of Armenia’s historical vocation, of Armenia’s role and place in the con-
cert of the Middle and Near Eastern peoples, nations, and cultures and of those
of the Caucasian region as well. He dedicated particular attention to the rela-
tionship between Armenia’s past and present. Consequently to this vision and
coherently with it, Garnik conceived his academic programme whose realisation
he pursued with the passion of his convictions and methodical consistency.
Wishing to sum up in a few words the leitmotiv of Garnik’s vision, I would
employ the established and well-known French expression, sortie du ghetto, that
is “way out from the ghetto”. With great consternation, he foresaw the trends in
late Soviet and independent Armenia that could entangle Armenian scholarship
and intellectual elite in ingenuously nationalistic or aggressively isolationist po-
sitions. With sagacity beyond his age, he warned that such intellectual streams
would, instead of making stronger, rather weaken the historical and cultural
identity of a nation, the self-consciousness that it has of its own being, of its roots
and values, and finally its international prestige. This has been and is Garnik’s
steady conviction. Conveying it with pupils, collaborators, colleagues, friends,
and anyone he meets along the path. He often makes use of two key-concepts:
“ghetto” and “metropolis”, in striking contrast one with the other, to express the
above-mentioned paradox and to explain how to break its deadlock. While
writing these lines, I hear those words resounding in my ears, which I have heard
repeated by Garnik for many years and so often in our conversations.
Besides these two concepts, the former being of a basically negative, and the
latter of a substantially positive valence in the context of his discourse, Garnik
uses frequently a metaphor to put in a still clearer light what he means by “me-
tropolis”: “wide horizon”, “enlarging horizons”. The continuous search for wide
and wider horizons has been, we can say without any undue emphasis, a leitmo-
tiv in Garnik’s personal yearning and academic ambitions, as well as in his schol-
arly programme and research. Such dynamics require the refusal of any collec-
tive narcissistic attitude, excessive national navel-gazing, and the intellectual at-
rophy that comes from self-contemplation, self-satisfaction, self-concentration.
To achieve this kind of opening towards the “Others”, the ideal has been transna-
tional and transcultural communication, to build new bridges or, rather, to re-
store the old, millenary bridges between cultures and peoples, to re-open con-
 Garnik Asatrian – A Scholar, a Master, a Friend 3

versations interrupted, to re-ignite a new, fresh dialogue between communities

and cultures. This had been Garnik’s prescription. Garnik looked to re-establish
intellectual exchanges in the great Iranian oikumene, which Gianroberto Scarcia,
one of the foremost Italian Iranologists, considers as the counterpart of the Hel-
lenistic oikumene, the more familiar one to the average Westerner.
Garnik’s encyclopaedic knowledge, intellectual curiosity, and scholarly com-
petence in various fields of Iranian civilization, grant him the possibility of being
a first class actor and, even, a protagonist in managing Armenia’s opening to
worldwide scholarship through the mighty filter of the Iranian oikumene.
One of the most conspicuous elements of Garnik’s world outlook is tradition-
alism or, even, a sort of conservatism – conservatism in the best sense of the
word. It is seen in some of his main attitudes: in his political approach to the pre-
sent and future of his country, in his special accent on the role of traditional val-
ues in the perpetuation of the Armenian culture and identity, in his admiration
of Classical German Orientalists, in his mode of life, manners and even instru-
ments of writing (he writes exclusively on a high quality paper with an old-fash-
ion fountain pen and never approaches a computer or typewriter), in his genu-
ine interest to nomadic and tribal societies and admiration of their archaic cul-
tures. After all, his proverbial passion to rural life finds realisation in his country-
house and garden in the historical village of Karbi in Ashtarak region at the
piedmont of Mount Aragats, not far from Oshakan, where St. Mesrob Mashtots is
buried. However, in his perception of Armenia and Armenian-ness Garnik has
the courage and conviction of a realist, and has little time for jingoistic slogan-
eering, puerile romanticism, or pathetic prolixity. Regarding the Armenian na-
tional strivings, Garnik often says: “The way to Mount Ararat lies only through
the plateau of Ararat valley, and no one can ever reach Ararat, if he is not able to
see the lantern of St. Gregory the Illuminator on the top of Aragats”. For Garnik,
the main ideal remains the Republic of Armenia with Artsakh, a material home-
land, requiring constant care and concern, and the point of “primary landmark”
for all the Armenians. Thus he is an avid follower of the Armenian Case, and has,
in fact, done a lot for the formation of a realistic national ideology in present-day
Armenia. His “sovereign” way of thinking, “sovereign” being understood here, of
course, in the sense of his independent, yet broad-minded nature, is clearly ex-
pressed in his approaches to the ideas of statehood and nation-building. Various
aspects of Garnik’s ideology and Weltanschauung are expressed in hundreds of
4 Boghos Levon Zekiyan

his publications, interviews and essays on the socio-political and cultural devel-
opments of Armenia and the region as a whole.
The erudition of Garnik Asatrian has made him a world-renowned scholar in
a wide variety of fields. His perfect command of a whole group of Iranian lan-
guages and dialects (Persian, Kurdish, Zaza, Talishi, Gabri, Central Iranian idi-
oms) is an extraordinary phenomenon in our time of narrow specialisation in
the Oriental Studies, and generally in the Humanities. Garnik’s scholarly contri-
butions range from Middle Iranian linguistics and philology (Middle Persian,
Parthian, Sogdian), Iranian etymology, toponymy, Iranian religions, dialectology,
ethnography, ethno-demography, ethnic history of the Iranian tribal societies
(Kurds, Zazas, Talishis, etc.), and social anthropology to contemporary socio-po-
litical issues of the Near East and the Caucasus. Many of his works, such as “Ver-
bal Nouns in the Manichaean Middle Persian and Parthian”, “Comparative Vo-
cabulary of Central Iranian Dialects”, “Ethnic Composition of Iran”, series of
books and article on the origin of the Kurds, history of the Kurdish consonant
system, Kurdish historical lexicology, the Yezidi religion, the Zazas and their
language, Bakhtiaris, Talishis, the Zoroastrians of Iran, etc. are of utmost schol-
arly value. His “Prolegomena to the Study of the Kurds”, appearing in 2009, is
now the bedrock for future studies on the origin, history, culture, and language
of the Kurds. Garnik’s approach to the ethnic situation in Iran and Iranian de-
mography is highly original and innovative.
The most emblematic achievement of the work Garnik has led and inspired
is the prestigious international journal Iran and the Caucasus, published since
1996 (since 2000 by BRILL Academic Publishers, Leiden-Boston), of which he is
both the founder and chief editor. This high standard peer-reviewed journal,
currently one of the leading Orientalistic periodicals, especially in Caucasian and
Iranian Studies, has been the result of an evolution, which quickly reached its
full maturity. It started, indeed, with Iran-Namē: Armenian Journal of Oriental
Studies, a review, founded in 1993 by Garnik and edited by him and published
mainly in Armenian. This was followed one year later by Acta Kurdica. Interna-
tional Journal of Kurdish and Iranian Studies at the Curzon Press, London, also
founded and edited by Garnik himself.1 Finally, Iran and the Caucasus appeared,

1
It would be apt, I think, to quote here the evaluation of this phenomenon by Prof. Rüdiger
Schmitt given in his review of Acta Kurdica (Kratylos, Bd 40: 197): “Mit Überraschung und Erstau-
nen nimmt der Beobachter zur Kenntnis, daβ in der jetzt unabhängigen Republik Armenien trotz
des seit Jahren währenden Krieges um ‘Berg-Karabach’, die historische ermenische Provinz Arc‘ax,
 Garnik Asatrian – A Scholar, a Master, a Friend 5

in all three years after Iran-Namē, with an internationally selected board of advi-
sors of a wide renown.
Another important achievement of Garnik is his establishing a prominent
role for Armenia towards neighbouring cultures and international scholarship.
This is illustrated in the numerous conferences he has hosted in Yerevan. Let us
mention first the “Second International Convention on Iranian Studies” of the
“Association for the Study of Persianate Societies” (April 2-5, 2004); this was pre-
ceded by two important international conferences: “Armenia-Iran: Yesterday,
Today, Tomorrow” (Tehran, December, 1998, in collaboration with the Foreign
Office of the Islamic Republic of Iran), and “Russia, Armenia and Iran: Dialogue
of Civilisations” (Yerevan, May 1999), and was followed by “The International
Conference on Talishi Studies” (Tsakhkadzor, May, 2005); “Iran and the Cauca-
sus: Unity and Diversity” (Yerevan, 6-8 June 2008); “The Turkic World, the Cauca-
sus, and Iran: Civilisational Crossroads of Interactions” (Tsakhkadzor, July 10-12,
2009), “Second International Conference on Talishi Studies” (Yerevan, October,
2011); “Iran and the Caucasus: 15 Years of Achievements” (July 2011), “Interna-
tional Conference on Zaza Studies” (Yerevan, November, 2011), and just recently
“The International Conference on the Autochthonous Peoples of the Caucasian-
Caspian Region” (Yerevan, October, 2012).
I cannot help mentioning in this context Garnik’s enthusiastic participation
in a European international research project under the auspices of INTAS (In-
ternational Association for the promotion of co-operation with scientists from
the New Independent States of the former Soviet Union), having as subject “Tbi-
lisi in the 19th Century: History and Culture”. Garnik headed the team of Arme-
nia’s Iranologists, while the writer of these lines was the scientific “Coordinator”,
responsible for the execution and the overall management of the biennial Pro-
ject (2001-2003), which included a great number of scientists from both Georgia
and Armenia (eleven persons from each country) and a few others from France

und ungeachtet der Auflösung der ehemaligen Sowjetunion und des daraus resultierenden
Zusammenbruchs der Wirtschaft und überhaupt vieler Bereiche des öffentlichen Lebens weiterhin
wissenschaftliche Forschung aktiv betrieben wird, ja mitunter völlig neue Initiativen ergriffen
werden und Realität annehmen. So sind in den letzten beiden Jahren, hauptsächlich dank des
Einsatzes von Garnik Asatrian allein zwei iranistische Periodika ins Leben gerufen worden: seit
Juni 1993 erscheint monatlich in armenischer Sprache die (wie es im Untertitel heiβt) wissen-
schaftliche und gesellschaftspolitische Zeitschrift Iran-Namē; und im Herbst 1994 folgte der erste
Band der vorliegenden Acta Kurdica”.
6 Boghos Levon Zekiyan

and Italy. The project was a resounding success and culminated in two scientific
colloquia, held in Tbilisi (June 9-12, 2002) and in Venice, Italy (June 28-30, 2003),
and a publication: La Géorgie entre Perse et Europe, sous la direction de Florence
Hellot-Bellier et Irène Natchkebia, L’Harmattan, Paris, 2009.
A remarkable side of Garnik’s academic personality, which distinguishes this
figure from even other outstanding scholars is his special capacity to inspire
around him young researchers of great promise, and to help them foster their
skills in a rigorous and impressive way. This important feature of Garnik’s activ-
ity had its concrete expression in the creation of the “Caucasian Centre for Ira-
nian Studies” (with its own publishing house and imprimerie) that was founded
by Garnik in Yerevan in 1996. Moreover, the Department of Iranian Studies at
Yerevan State University, during the past 15 years, under his guidance, has turned
from a small division with a limited staff into an internationally renown higher
education and research centre, incorporating more than 30 scholars and well-
trained specialists in various fields of Oriental Studies, and training around 200
students of BA, MA, and PhD levels. The teaching curriculum at the Department
is a unicum of its kind. It covers a wide spectrum of disciplines: history and
culture of the region (including, apart from proper Iran, Central Asia, Afghani-
stan, the North Caucasus, India, etc.), Iranian religions, Old and Middle Iranian
languages, Sanskrit, Persian, Kurdish, Pashto, Baluchi, Ossetic, etc. I will not ex-
ceed in words if I will say that Garnik is already the leader of an Armenian school
of Iranologists in the lager sense of its definition, bearing his brand. Garnik him-
self is the bearer of the best traditions of the Leningrad school of Oriental Stud-
ies, where he studied at the Leningrad Branch of the Institute of Oriental Studies,
Academy of Sciences of the USSR, in its peak period of flourishing scholarship,
from 1976 to 1984, under the guidance of such prominent masters as Prof. Anahit
Georgievna Perikhanian and Prof. Vladimir Aronovich Livshits. Before that (1971-
1976), he had studied Kurdology at the Yerevan State University, with such con-
noisseurs of Kurdish language and culture as Prof. Hajie Jindi, Prof. Gurgen
Akopov, Prof. Kanat Kurdoev, and Tital Chatoev.
Garnik was born on the 7th of March 1953, in Tehran. He is a descendant of
an old Armenian clan having migrated from Upper Armenia (Barjr Hayk‘), the
northern areas of lake Van, to Fereydan region of the province of Isfahan in Iran
exactly 400 years ago, in 1610. His ancestor, Asatur, founded the village, Boloran,
which so far remains the single stronghold of the Armenian-ness and ancient
Armenian traditions in the region, once almost totally inhabited by Armenians.
 Garnik Asatrian – A Scholar, a Master, a Friend 7

While Garnik was still a child, his family repatriated to Armenia. Even if he
passed the greatest part of his conscious life and personal activity in Armenia, in
the shade of the great Masis, inspired by Armenia’s legendary heroes, its eternal
myths and perennial tales, he too could affirm with the great Charents: Իրանի
արեւն է իմ հոգում / The sun of Iran is in my soul. That sun of Iran, which gave
rise to one of the brightest and most refined expressions of human civilisation.
The unique complex of Apadana, which is, in the amazing plasticity of its lifelike
forms, a symbol of that marvelous light and refinement, is at once a very elo-
quently archē-ic witness of the most ancient Armenian identity, as well as of
those millenary, deep-rooted ties, which link Armenia to the Iranian oikumene.
Garnik’s scholarly work is a splendid monument of our days, which honours this
rich heritage, old and living together.
These few words hardly will give a full profile of Prof. Dr. Garnik Serobi Asa-
trian’s figure as a man, a scholar, a researcher, a public leader, and a teacher. He
is already known to both national and international forums alike. I hope, how-
ever, to have given some basic and appropriate inputs for a more adequate and
deeper appreciation of this singular personality of Armenian academy. When the
road behind us has been so fecund, our expectations of him in the future are no
less profound.

Boghos Levon Zekiyan

Member of the Armenian National Academy of Sciences,
Università Ca' Foscari Venezia
Venice, 31st of October, 2012
 “Like a Flame Through the Reeds”:
An Iranian Image in the Buzandaran Patmut‘iwnk‘

Marco Bais
Pontifical Oriental Institute (Rome)

Abstract
The paper discusses the simile of the fire running through the reeds, which is well attested in clas-
sical Armenian literature and is used to describe a formidable assault of an army. The connection
of the image with the Iranian concept of kingship and supernatural valour is confirmed by the oc-
currence of the same simile in the Iranian epic. The original value of the image still survives in
some later attestations, though in a less clear and conscious form, and sometimes it is reinter-
preted with a Christian context. In most of the later occurrences, anyway, this meaning is com-
pletely lost, and the image is employed just as a simile for the violence and destructiveness of an
attack. That is the reason why it is applied not only to Armenian troops but also to the enemies,
and some later authors show a contact between the original image and a similar simile occurring
in the Bible. However, a closer analysis of the texts and a comparison with the Septuagint shows
that the interference of the Bible with the traditional Armenian image is just a formal one.

Keywords
Armeno-Iranian Literary Ties, Iranian Elements in Armenian Culture, Classical Armenian Litera-
ture

Some years ago, I met Prof. Garnik Asatrian at a conference on the Christians in
the East (1st-7th century) held in Rome. On that occasion he invited me to partici-
pate in the International Conference The Turkic World, the Caucasus, and Iran:
Civilisational Crossroads of Interactions scheduled to take place later that year, an
invitation which I gladly accepted, because I had a good subject for a paper. The
topic was about a simile found in the Buzandaran Patmut‘iwnk ‘‘The Epic Histo-
ries’ attributed to P‘awstos and in other Armenian works, and its relation with a
similar comparison attested in the Pahlavi epic Ayādgār ī Zarērān‘ Memorial of
Zarēr’. I had worked on this theme with a view to participating in the previous
year’s conference Iran and the Caucasus: Unity and Diversity, which in the end I
could not attend. Unfortunately, not even this time I was able to go to Yerevan

 Koninklijke Brill NV, Leiden, 2015

10 Marco Bais

and the dossier on this subject ended up in a drawer. It was 2009. When I re-
ceived the invitation to contribute to the Festschrift dedicated to the 60th anni-
versary of Prof. Garnik Asatrian, I felt that the time had come to get my papers
out of the drawer and bring the work to a conclusion.
According to the Buzandaran Patmut‘iwnk‘, the Persian king Šapuh was
greatly astonished at the bravery displayed by the Armenians in the battle of
Ganjak. When Šapuh reached Persia with few survivors, he spoke his thoughts in
a sort of monologue. At first Šapuh’s speech focuses on the battle tactic employed
by the Armenians at Ganjak. They advanced and attacked the Persians, but as
soon as they were driven off, they took refuge in the Roman legion, which opened
its shield wall receiving them as into a fortified city. After resting a little, they sal-
lied forth again and fought until the Persians were routed. Then Šapuh’s thoughts
turn to the Armenians’ loyalty and devotion to their king Aršak. The simile that
attracted my attention occurs in this part of the monologue. The Persian king
praises the Armenian army led by Mušeł Mamikonean and expresses his amaze-
ment that the Armenians dedicated to Aršak all the champions whom they slew.
Aršak had been captured by the Persians and imprisoned in the Castle of Obliv-
ion (Andməš). In spite of this, the Armenians fought valiantly in battle and slew
countless enemies, behaving as if Aršak were there on the battlefield, at the head
of his army. Here are Šapuh’s words:
And at this too do I marvel, at the steadfast loyalty of the Armenian army in its devo-
tion to its lord. For so many years have passed since Aršak their lord has been lost to
them, and yet, they gave each other encouragement in his name. And whenever they
struck down champions, they ever called out: “To Aršak”, and yet he was not among
them. And in the reverent devotion to their own true lord, they dedicated all the
champions whom they killed to him. And the frenzied army of Mušeł! It seemed to
me that a blaze of fire arose from that army and from its standards, as in the blaze
of a conflagration, it coursed through the army like a flame through the reeds (hur
boc‘ i gndēn elanic‘ēr1 ew i nšanakac‘, ibrew i hur hrdehi aynpēs anc‘anēr ənd gund
orpēs boc‘ ənd ełēgn). And so much time has passed since Aršak their lord has been
lost to them, for he lies in the fortress of Andməš in the land of Xužastan, yet they, in
their piety, believed that he stood at their head as their king; that he really stood with

1
Both the edition of the Buzandaran (Buzandac‘i 1883: 170; idem 1933: 207) have elanic‘ēr.
Though typical of the latinizing Armenian (lat‘inaban hayerēn), this form is recorded by Bagratuni
(1852: 195 §454) as attested in some works which, at least in part, date back to the 5th century (I am
grateful to Prof. Alessandro Orengo for this reference). Of course it can be a later form integrated
into the manuscript tradition. This issue needs further investigation.
 An Iranian image in the Buzandaran Patmut‘iwnk‘ 11

them in the midst of the army, at the head of the battle, and that they performed
their service to him in his presence. Lo,—he said—blessed be he who may be lord of
the Armenian army, of such faithful, and loyal troops devoted to their lord (Buzan-
dac‘i 1933: 207).2
Commenting on this chapter of the Buzandaran Patmut‘iwnk‘, Nina Garsoïan
pointed out that it “leaves its factual, historical setting to rely on an oral source
with characteristic repetitions, refrains, and anachronisms; its imaginary solilo-
quy attributed to Šāhpuhr II; and a familiarity with the ethos and traditions of Zo-
roastrian Iran and Pre-Christian Armenia” (Garsoïan 1989: 310-311, n. 4). In other
words, she highlighted the epic character of this passage, assuming that it “is
composed of a paraphrase and possibly even of quotations from an epic or paean
to Aršak II that was still current in the oral literature of the period” (ibid.: 35, 311,
n. 8). The insistence on Aršak’s k‘aǰut‘iwn ‘valour’ is an essential part of the Ira-
nian ethos. In P‘awstos, in fact, k‘aǰut‘iwn has to be understood in its transcenden-
tal value, rather than in its most trivial sense of ‘courage, bravery.’ It is a super-
natural valour “distinguishing the legitimate ruler of Iran in the Zoroastrian tradi-
tion. This quality was bestowed on them by the god Vərəθraγna (Arm. Vahagn),
who was himself ‘created victorious’ (pērōzgar), as it is explicitly stated by Aa,
cxxvii (the reference is to Agat‘angełos 1983: 80, cf. Thomson 1976: 138), ‘k‘aǰut‘iwn
hasc‘ē jez i k‘aǰēn Vahagnē’ (may valor come to you from valorous Vahagn)” (Gar-
soïan 1989: 535 (s.v. k‘aǰ/k‘aǰut‘iwn)).
According to Garsoïan, P‘awstos’ image of the fire running through the reeds
is associated precisely with the Zoroastrian concept of supernatural valour,
therefore it has a strong symbolic meaning: it “is an evocation of the fiery epiph-
any of the god Vahagn from the scarlet reed in the sea, described in the fragment
of a Pre-Christian poem3[...] it is yet another recall of Armenia’s Zoroastrian past
and of the ‘valor’ (k‘aǰut‘iwn) that Vahagn bestowed on its kings” (Garsoïan 1989:
311-312, n. 16). This claim is substantiated by comparing the wording of the Song
on the Birth of Vahagn—“... ənd ełēgan p‘oł boc‘ elanēr (from the tube of the reed
a flame came forth)”—with P‘awstos’ description of the Armenian army, “hur boc‘
i gndēn elanic‘ēr...orpēs boc‘ ənd ełēgn.” The close similarity seems to hint that the
verse of the older poem is echoed in Pawstos’ passage. Garsoïan observes that the
same image is preserved in the Venice edition of the Life of St. Nersēs (Ališan

2
Cf. Engl. trans. by Garsoïan 1989: 195-196.
3
The fragment is preserved in Movsēs Xorenac‘i I.31 (1913: 86). Cf. Engl. trans. by Thomson
2006: 119.
12 Marco Bais

1853: 73), a work that draws heavily on the Buzandaran. However, the Life “no
longer understands the symbolic meaning of Vahagn passing through the fiery
reed at his birth and gives the more prosaic explanation of a blaze devouring
reeds” (Garsoïan 1989: 312, n. 16), whereas, according to Garsoïan, a reminiscent
image, but without symbolic meaning, occurs in Ełišē and Sebēos.
Nina Garsoïan seems not to be aware of the occurrence of the same image in
the Iranian epic Ayādgār ī Zarērān (Monchi-Zadeh 1981).4 Yet, it is perhaps one of
the strongest evidences in favour of the epic character and Iranian ethos of
P‘awstos’ chapter describing the battle of Ganjak. The Ayādgār ī Zarērān cele-
brates the bitter battle in which Wištāsp defended the newly adopted mazdeism
against Arjāsp, lord of the Hyōns, who wanted him to give up the new religion.
The battle between Wištāsp and Arjāsp has just begun. The two kings watch the
fray from the top of a mountain, and the superiority of the Kayanians led by
Zarēr, Wištāsp’s brother, is immediately evident: “And the strong commander,
the brave Zarēr fights so bravely as when the deity of fire plunges into a reed bed
and the wind assists him (ud ān tahm spāh-bed ī nēw zarēr kārezār ōwōn nēw
kunēd čiyōn ka ādur yazd andar ō nayestān ōftēd u-š wād-iz ayār bawēd)”.5 He
slaughters many enemies and eventually is treacherously killed by the sorcerer
Wīdrafš.
The text has come down to us in a Pahlavi recension of the Sasanian period,
but various linguistic features point to an older Parthian version (Cf. Cereti 2001:
200). Thus, the story of Zarēr must have gained fame among Parthian minstrels,
who eventually taught it to Sasanian singers. Since Parthians exerted a strong cul-
tural influence on Armenia, which was ruled by a branch of the same Arsacid
dynasty ruling the Parthian Empire, it is plausible that Parthian epic heritage was
known to Arsacid Armenia.
Russell has called attention to some echoes of the Ayādgār ī Zarērān in
P‘awstos’ work. According to him, the two texts share the following “broad simi-
larities”: the fall of the Armenian Arsacids is only foreshadowed, but not de-
scribed, by P‘awstos, like that of the Kayanians in the Ayādgār ī Zarērān; P‘awstos
sets “weak, indecisive, lustful, angry kings” side by side with true heroes, i.e. the
Mamikoneans and the catholicoi, especially the visionary Nersēs the Great, just as
the king Wištāspis juxtaposed with the two heroes Zarēr and the visionary

4
The text is quoted from the on-line version in TITUS Text Database: http://titus.uni-frankfurt.
de/texte/etcs/iran/miran/mpers/jamasp/jamas.htm. Accessed September 10, 2012.
5
Sentence 70 (on-line version in TITUS Text Database).
 An Iranian image in the Buzandaran Patmut‘iwnk‘ 13

Jāmāsp in the Iranian epic; both works describe “the just but hopeless struggle of
a dynasty about to crash, whose newly-acquired religion will, by virtue of its self-
sacrifice, remain. “Among the smaller episodes which Russell regards as identical
in the two works, the only one to bear striking similarities is that of the young boy
forbidden to fight because of his youth and sneaking into battle to distinguish
himself, while his father is killed in the same fight. The Iranian child-hero is
Bastwar, son of Zarēr, and his Armenian counterpart is Artawazd, son of Vač‘ē
Mamikonean. The remaining “identical” episodes in Russell’s view are: Arjāsp,
enemy of the mazdeism, is replaced in the Armenian work by the Sasanian king
Šapuh, persecutor of Christians; Mušeł Mamikonean refuses to kill Šapuh, as
Spandiyād does not kill Arjāsp; both Mušeł and Zarēr are murdered by treachery
(Russell 1996: 30-31).
In my opinion, all the features listed by Russell are too broad and vague to be
taken as echoes of the Ayādgār ī Zarērān in P‘awtos’ work, with the sole excep-
tion of the child-hero tale. Yet, it is undeniable that they evoke guiding beliefs
and values that are a common heritage of the Arsacid lands and survived well
into Christian Armenia, adapting themselves to the new Christian framework of
values. In this respect, the strong religious connection of the Iranian story made it
even more appealing in Armenian eyes. In fact, just as Wištāsp and his heroes
fought against Arjāsp in defence of the “pure religion of the Mazda-worshippers
(dēn ī abēzag ī māzdēsnān),” so did Christian Armenia fight against the Zoroas-
trian Sasanians to preserve its faith. P‘awstos composed his work some decades
after the most celebrated battle of Awarayr (451), which is traditionally regarded
as the climax of the Armeno-Persian confrontation. He wrote in a period when
the relations between Armenia and Persia were still hostile and permeated with
religious hatred. Though he dealt with the wars which took place between the
two countries in the preceding century, he inevitably looked back on these events
through the lens of the situation of his time. The 4th century wars against Persia
appeared to him as a struggle to ensure the survival of Christianity in Armenia,
and king Šapuh II (309-379), persecutor of Christians in Iran, must have been
considered as a sort of prefiguration of Yazdgird II (439-457), whose zeal in perse-
cuting Caucasian Christianity reminds Arjāsp’s invite to Wištāsp to give up maz-
deism and become his coreligionist. Therefore, the ethos of the Ayādgār ī Zarērān
is in tune with ideology of at least some sections of the Buzandaran and Russell is
right when he says that “it is likely that some form of the epic tale of Zarēr was
known to the Armenians of the Arsacid period, probably in oral form, since it
14 Marco Bais

contributed [...] to the formation of the Armenian Arsacid epic narratives of the
fourth century A.D., as P‘awstos set them down in the fifth” (ibid.: 22).
The image of the fire through the reeds occurring in a battle context in the
Ayādgār ī Zarērān and in the Buzandaran may thus be considered as a vestige of
the common imagery shared by Parthian epic and P‘awstos’ work. In both cases,
the simile compares the violence of fire raging through the reeds with the
strength of an army attacking the enemy. Nevertheless, as we have seen, Garsoïan
emphasized the symbolic value of the simile in the Buzandaran, where a special
stress is laid on Aršak’s k‘aǰut‘iwn. In that passage it evokes the god Vahagn and
consequently the k‘aǰut‘iwn that he bestows on the Arsacid kings of Armenia, and
therefore on Aršak as well. In spite of his forced physical absence from the bat-
tlefield, Aršak’s k‘aǰut‘iwn run through the army in the form of fire and led it to
victory.
The simile found in the Iranian epic is more complex. It does not simply rep-
resent a fire raging through the reeds, it alludes, rather, to a lightning falling over
a reed bed and spreading through it driven by the wind. The simile obviously
compares the destructive fury of fire and its appalling speed with Zarēr’s blood-
thirsty vehemence. Moreover, the Iranian text explicitly refers to the divine na-
ture of the fire by calling it ādur yazd, deity of fire. We have here possibly a remi-
niscence of the Kayanian concept of xᵛarənah-, “a magic force or power of lumi-
nous and fiery nature” (Gnoli 1999: 314), mentioned in a number of passages of
the Avesta. Gnoli refers to Yašt 10.127, where “the ‘strong’ (uγra-) xᵛarənah- of the
kauui- is identified with a ‘blazing fire’ (ātarš yōupa.suxtō) that precedes Mithra
in his chariot,” whereas “in Zādspram... Zoroaster’s xwarrah is said to have de-
scended from heaven and become manifest ‘in the form of fire’ (pad ātaxš
ēwēnag) at the moment of his birth (5.1, 8.8)” (ibid.). This magic force is the he-
reditary dynastic charisma lying at the roots of the Iranian idea of kingship,
which mingled with the concept of royal fortune in the Hellenistic period (ibid.).
Thus, it is no accident that the image of the fire plunging into the reeds is associ-
ated with Zarēr, the “strong commander” of the Kayanian army. He is the brother
of king Wištāsp, and therefore a member of the Kayanian royal family, and is
usually described as nēw ‘brave,’6 an attribute applied also to Spandyād,7 another
Kayanian hero who is destined to become the main character of the story in its

6
Cf. sentences 14, 17, 55, 64, 70, 71, 79, 83, 89 (on-line version in TITUS Text Database, cf. supra).
Accessed September 10, 2012.
7
Cf. sentences 61, 110.
 An Iranian image in the Buzandaran Patmut‘iwnk‘ 15

later reworkings, as for instance in the Šāhnāme. In the Ayādgār ī Zarērān, how-
ever, Zarēr still plays a leading role, in spite of a clear tendency to draw increasing
attention to Spandyād. After receiving Arjāsp’s request to choose between abjura-
tion or war, Wištāsp slums in dejection, and it is his brother Zarēr who responds
to the threat choosing to join battle. Likewise, it is Zarēr who makes preparations
for fight and he is the first Kayanian champion to die in defence of his religion,
thereby fulfilling Jāmāsp’s prophecy: victory is certain, but many heroes of
Wištāsp’s family will die. The brave Zarēr seems to be endowed with the spiritual
force which enables him to achieve his mission in battle. Therefore, the image of
the deity of fire plunging into the reeds can be interpreted on different levels: it is
a simile comparing the destructive force of the fire with the fury displayed by
Zarēr on the battlefield, and at the same time it is a tangible representation of the
hero’s transcendental power.
In her commentary on P‘awstos, Garsoïan pays almost no attention to the oc-
currence of the simile in other works of Armenian literature. As we have seen, she
quotes the Life of St. Nersēs and a “reminiscent image” in Sebēos and Ełišē.
The latter, however, has no trace of a simile comparable to that found in
P‘awstos. Ełišē’s passage referred to by Garsoïan deals with the pivotal battle of
Awarayr and describes the din of the battle and the glitter of arms: “Their melee
caused a roar like the thundering in turbulent clouds, and the echoing of their
shouts made the caverns of the mountains shake. From the multitude of helmets
and shining armour of the soldiers light flashed like that of the rays of the sun
(ibrew nšoylk‘ jaŕagayt‘ic‘ aregakan hatanēin). Likewise from the glittering of the
many swords and the waving of the massed lances there came a blaze like a
fearful lightning from the heaven (ibrew yerknust ahagin hrajgut‘iwnk‘ eŕayin)”
(Ełišē 1957: 116-117, cf. Engl. trans. by Thomson 1982: 169).
A simile loosely resembling that of P‘awstos occurs, on the other hand, in
Sebēos’ description of battle scene: “From on high there will be a fearsome
crashing and flashing (ahagin čayt‘ē ew p‘aylatakē); warriors will assail you on
white horses with heavy lances and will penetrate your host like thunderbolts of
flashing fire, that will drop down from heaven to earth and burn up the brush-
wood of the plains and the forest, green and dry alike (anc‘aneni mêǰ bazmu-
t‘eann, ibrew zp‘aylatakunk‘ hroy šant‘eloy, zi c‘awłanayc‘en i yerknic‘ i yerkir, ew
ayresc‘en zxŕiws daštac‘, zantaŕs zdalars ew zgawsac‘eals)” (Sebēos 1979: 78-79;
Engl. trans. by Thomson 1999, v. 1: 22). The attack of the Armenian troops is here
compared with a lightning that strikes the earth and causes a fire. The basic idea
16 Marco Bais

of the comparison is the same as in P‘awstos’ simile, but it is expressed in differ-

ent words. The quotation is drawn from the reply of Mušeł Mamikonean to the
letters of the Persian usurper Vahram. Mušeł refuses to side with him, predicting
his defeat in battle. What is interesting here is the first sentence of Mušeł’s reply:
“Kingship is from God, and He gave it to whom He whished.” Therefore, Mušeł
challenges the legitimacy of Vahram’s power and claims the supernatural origin
of kingship, though in a Christian perspective. Even if this passage has to do with
the concept of legitimate kingship, the image of a lightning burning the brush-
wood is too generic to be compared to the simile employed in the Buzandaran.
Contrariwise, in the Life of St. Nersēs we find almost the same image as in
P‘awstos: the commander Mušeł Mamikonean led his troops against the enemy
and “like a fire raging in the reeds, he broke through the troops of Persians and
Honk‘ (orpēs zhur p‘ot‘orkeal yełēgn, čełk‘ēr zzōrsn Parsic‘ ew zHonac‘)” (Ališan
1853: 73; Tēr-Vardanean 2010: 709). This passage confuses or conflates two distinct
battles fought by the Armenians against the Persians, that of Mount Npat and
that of Ganjak, which are described separately in P‘awstos (P‘awstos V.4 (Npat)
and V.5 (Ganjak), (Buzandac‘i 1933: 197-207; Cf. Garsoïan1989: 308, n. 1)). Thus, the
simile occurs in the same context as in the Buzandaran. The reason for this coin-
cidence is that the Life draws heavily on the work of P‘awstos or on the same
source used by P‘awstos. Nevertheless, the Life does not mention king Aršak in
connection with the battle, nor does it allude to the legitimacy of royal power,
thus depriving the simile of its symbolic meaning.
What Garsoïan seems to have overlooked, however, is that another occur-
rence of the same simile can be found in the first part of the Life. After his acces-
sion to the throne, king Aršak reorganized his kingdom and entrusted again the
command-in-chief of the Armenian army to the Mamikoneans, who had quitted
that office due to disagreements with the former king Tiran. The Mamikoneans
are celebrated as k‘aǰasirtk ‘‘brave-hearted’ and tirasēr ‘devoted to their lords’, and
as soon as they were appointed commanders, “they rushed to battle like a fire
through the reeds (i marts paterazmac‘ ibrew hur ənd ełēgn ənt‘anayin)” (Ališan
1853: 17-18; Tēr-Vardanean 2010: 672). The praise of the Mamikoneans has a par-
allel in the Buzandaran (Buzandac‘i 1933: 76-77), although P‘awstos does not em-
ploy here the simile of the fire in the reeds. Commenting on this passage of
P‘awstos, Garsoïan observes that “the exaltation of the Mamikonean house with
the transfer to it of the royal epithet, k‘aǰ, ‘valiant’ is especially manifest in this
chapter and from here on” (Garsoïan 1989: 268, n. 2). In the Life the adjective k‘aǰ
 An Iranian image in the Buzandaran Patmut‘iwnk‘ 17

is almost invariably applied to the Mamikoneans, and particularly to Mušeł, but it

has probably lost its transcendental value. It is not clear whether the purported
royal descent of the Mamikoneans (Buzandac‘i 1933: 204, 244; Cf. Garsoïan 1989:
366 (s.v. Čenk‘)) played a role in their being qualified with this attribute (Garsoïan
1989: 34, n. 153). Be it as it may, we can assume that the use of the simile in rela-
tion to the Mamikoneans’ exaltation is perhaps an echo from the epic back-
ground of the Buzandaran and the Life. It could possibly go back to an epic cycle
dealing with the Mamikoneans’ deeds, where the image of a fire running through
the reeds could have had a supernatural overtone.
The use of the simile, however, is not confined to the aforementioned works.
It is well attested throughout classical Armenian literature.
We find it twice in the second book of the Patmut‘iwn Ałuanic‘ ‘History of Al-
bania’, attributed to a certain Movsēs Dasxuranc‘i or Kałankatuac‘i. Though the
final redaction of this work is not earlier than the 10th century (Akopjan 1987: 166-
177; 212-216; Aleksidze-Mahé 2008: x-xi), the chapters which interest us (II.11 and
16) seem to go back to the 7th century (Akopjan 1987: 188-196; Zuckerman 2007:
404-405). Ch. 11 provides an account of the invasion of Albania by the Khazars,
Heraclius’ allies, in the last year (38th) of Xusraw II (590-628). They came and
mercilessly destroyed the town of Č‘ołay “like a fire lit in the reeds, they entered
at one gate and emerged through the other (ibrew zhur vaŕeal yełēgn, mtanēin ənd
duŕn ew elanēin ənd miwsn) (Kałankatuac‘i 1983: 136; Cf. Engl. trans. by Dowsett
1961: 84). In ch. 16 we find the description of an engagement between the Persians
and the Khazars, who surrounded them “like a fire in the reeds on the bank of the
Lake of Gełam (orpēs hur yełēgn zezerb covun Gełamay)” (Kałankatuac‘i 1983: 169;
Cf. Engl. trans. by Dowsett 1961: 105), and slaughtered them. In the Patmut‘iwn
Ałuanic‘ the simile compares the rapid and violent attack of the enemy, unlike
what happened in P‘awstos and the Life, where it referred always to the Arme-
nian army.
The simile is particularly dear to T‘ovma Arcruni, an Armenian historian who
lived at the turn of 9th and 10th centuries. It occurs three times in his Patmut‘iwn
Arcruneac‘ tan ‘History of the House of the Arcunis.’ A fourth occurrence is in the
last Book of the Patmut‘iwn, which is attributed to anonymous continuators.
T‘ovma’s work refers twice to the 5th century struggle of the Armenians for
religious freedom. The author exalted the glory of the Arcrunis claiming that at
the battle of Awarayr “Vahan Arcruni, with splendid and outstanding valour
(k‘aǰamartut‘eamb), fought side by side with saint Vardan, pressing into the midst
18 Marco Bais

of the valiant Persian champions like a fire through the reeds (ibrew hurənd ełēgn,
čemelov i mēǰ k‘aǰ axoyeanac‘n Parsic‘)” (T‘ovma Arcruni 1887: 79-80; cf. Engl.
trans. by Thomson 1985: 146). A few pages later T‘ovma gives an account of the
battle between Vahan Mamikonean and the Persians led by Hazarawuxt, some
thirty years after the events of Awarayr. Faced with the enemy’s numerical supe-
riority, many Armenians fled away, and only thirty men remained with the val-
iant Vahan (ənd k‘aǰin Vahanay). The battle took place in the village of Eriz, and
the Armenians put to the sword and routed most of the Persians, “like a tempest
whirling around dust, like a fire running through the reeds (ibrew mrrik zp‘oši ptu-
teal, ibrew hur ənd ełēgn ənt‘ac‘eal)” (T‘ovma Arcruni 1887: 84; cf. Engl. trans. by
Thomson 1985: 151).
The third occurrence of the simile concerns the devastation caused by the
campaigns of Bugha, who was entrusted by the caliph with the suppression of the
Armenian revolt in the early 850s. The leading members of the Arcruni house
were imprisoned and exiled, and the Tačik‘ (Arabs) spread over Armenia taking
possession of the land and everything was ruined: “As the prophet Joel lamented
over the misery which had befallen, saying: ‘The land was like a garden of delight
before, and behind them a plain of desolation, and like a fire running through the
reeds (ibrew hur ənd ełēgn ənt‘ac‘eal), so they set upon us, like the locust, like the
caterpillar, and like the grub with the grasshopper’, set upon the fruit-bearing
trees and the sensitive pasture, in such fashion they destroyed and consumed the
resources of the human race and their property, just as this is described in the
book of the prophet Joel’s vision” (T‘ovma Arcruni 1887: 152; cf. Engl. trans. by
Thomson 1985: 218).
T‘ovma compares the miserable condition of Armenia with the devastating
locust plague described by Joel but he does not quote verbatim the prophet’s
words. What we have here is a conflation of part of verses and reminiscences
from Joel 2: “the land was like a garden of delight before them, and behind them a
plain of desolation (ibrew zdraxt p‘ap‘kut‘ean erkird aŕaǰi nora, ew zkni nora, dašt
apakanut‘ean)” (part of 2.3);8 “And I will restore to you the years, which the locust
and the caterpillar and the grub has eaten, my great army which I sent upon you
(ew hatuc‘ic‘ jez p‘oxanak, amac‘n, yors eker maraxn, ew ǰoreakn ew t‘rt‘urn, zōr im
mec zor aŕak‘ec‘i i veray jer)” (2.25). The image of the fire in the reeds, which
T‘ovma inserted between these two quotations, has no real equivalent in Joel.

8
The Armenian Bible is quoted after the edition of Zōhrapean 1805.
 An Iranian image in the Buzandaran Patmut‘iwnk‘ 19

Probably its use here is an echo of the verse 2.5: “And like the rumbling of chariots
they stormed on the tops of the mountains, like the sound of a flame of fire,
which devours the reeds, like a numerous and strong host ready for battle (ew
ibrew zšač‘iwn kaŕac‘ i veray glxoc‘ leranc‘ aspatakesc‘en, ibrew zjayn boc‘oy hroy or
utic‘ē zełēgn, ibrew zōr bazum ew hzōr patrasteal i paterazm).” Joel describes here
the massive invasion of locusts and the noise produced by their advance, similar
to the crackling of a flame burning reeds. Although the Tačik‘’s devastating attack
on Armenia could easily be compared with the calamity of locust, T‘ovma’s
wording is identical to the other occurrences of the image in his own work and is
closer to Movsēs Dasxuranc‘i and ultimately to P‘awstos. Thus, we can assume in
this case an overlapping of the biblical passage and the simile found in P‘awstos
and well attested in T‘ovma’s work.
The last occurrence of the simile in the Patmut‘iwn Arcruneac‘ tan is particu-
larly significant in this respect. Xač‘ik-Gagik, king of Vaspurakan, joins battle with
the Arabs at the gates of Duin: “[...] crashings and thunderings and flashings ap-
peared as darting downwards from the clouds, and the day started to become
dreadful. Then, the king, together with his troops, was strengthened by help from
on High, especially because the holy patriarch of Armenia, the great Ełishē, going
to the summit of the hill of Gēn with groups of priests, held aloft his hands like
saint Nersēs until the second Amałek was defeated. Hence, the Armenian army
surpassed a contingent after the other, the king came and attacked the middle of
the foreigners’ forces (= Muslims), where there were about four thousand cham-
pion armed foot soldiers. The king, striking them frontally, routed the forces of
the foreigners’ army like a lightning striking through the reeds (ibrew kaycakn ənd
ełēgn hareal)” (T‘ovma Arcruni 1887: 301, cf. Engl. trans. by Thomson 1985: 363-
364). It appears clearly that the fight is described according to the standard battle
imagery we find, for instance, in Ełišē. Besides, there is a direct reference to the
prayer of Nersēs the Great on behalf of the Armenians fighting the Persians at
Npat. As we have seen, P‘awstos describes the battle of Npat and the prayer of
Nersēs immediately before the battle of Ganjak, while the Life of Nersēs conflates
the two battles. Thus, the presence of the simile in the Patmut‘iwn Arcruneac‘ tan
may be a reminiscence of P‘awstos and/or of his cultural milieu closely linked to
Arsacid Iran (Cf. Russell 2002-2003: 4 for the closeness of ancient Armenian histo-
riography to the epics of the Iranian world.), especially since the ethos of the two
passages is similar. The simile, in fact, is used of Gagik leading the Armenian
army to victory against the Arabs. Gagik, king of Vaspurakan, is styled “great king
20 Marco Bais

of Armenia (meci t‘agaworin Hayoc‘)” (T‘ovma Arcruni 1887: 300; cf. Engl. trans. by
Thomson 1985: 362) by Tovma’s continuator and is described as endowed with
the traditional attribute of k‘aǰut‘iwn (T‘ovma Arcruni 1887: 301; cf. Engl. trans. by
Thomson 1985: 363), though reinterpreted from a Christian point of view. Never-
theless, the wording of this passage recalls Wisd. of Sol. 3.7, where we are told that
the souls of the just ones “shall shine, and run like a spark through the reeds (pay-
caŕasc‘in, ew ibrew kaycakn ənd ełēgn ənt‘asc‘in).” The same figure occurs in Isa.
5.24, where it introduces the punishment for the sins listed in the preceding
verses: “therefore, as a reed (stubble) is set alight by a fire of coal, so they will be
burnt by a blazing flame (Vasn aysorik orpēs vaŕi ełēgn i kaycakanc‘ hroy, aynpēs
ayresc‘in i borbok‘eal boc‘oy).” The Armenian word kaycakn renders both σπινθῆ-
ρες ‘sparks’ (Wisd. of Sol. 3.7) (Ziegler 1980: 103) and ἄνθρακος ‘of coal’ (Isa. 5.24)
(idem 1983: 140) of the Septuagint, but it means also ‘lightning’ (NBHL, vol. 1: 1045
(s.v. kaycakn)), and this seems to be the sense it has in the Patmut‘iwn, as the use
of the verb harkanel ‘to strike’ suggests. Once again the Patmut‘iwn Arcruneac‘ tan
preserves, to a certain extent, the original significance of the simile and recalls
the image of the lightning alluded to in the Iranian epic. It shows the influence of
the scriptural model as well, but this influence seems to be merely a formal one.
For the meaning of the simile in Bible changes according to the context: it may
describe the joy of the just ones in the Day of Judgement or the punishment of
the sinners.
Smbat Sparapet employs the same image in his Taregirk‘ (Annals), composed
in the 13th century. Describing the siege of Antioch in 1099 (Armenian Era 548),
Smbat says: “In that year all the Christians called upon God to help them, and at-
tacked with faith the foreigners, and with the power of God they overcame them
and they burned the army of the foreigners like a fire that consumes the reeds
(orpēs hur or hrdehē zełegn)[...]” (Smbat Sparapet 1957: 109). The Christian troops
were not led by a sovereign endowed with supernatural power, but they were as-
sisted by the power of God himself, represented by the Holy Lance that pierced
Christ’s side. They brought it with them in battle and erected opposite the stand-
ards of the foreigners (Muslims). Thus the simile has to be interpreted in a fully
Christian context, and even if it has lost much of its original value, it seems to
hint to the strength of the Christian army divinely inspired by God.
No trace of transcendental implications remains in the later occurrences of
the simile, though in some instances the way the image is expressed sounds close
to P‘awstos’ model. Grigor Kamaxec‘i, known also as Daranałc‘i, (16th-17th cen-
 An Iranian image in the Buzandaran Patmut‘iwnk‘ 21

tury), Aŕak‘el Davrižec‘i (17th century) and Simēon Erewanc‘i (18th century) in-
variably used the image to depict the violent fights which took place in the tur-
bulent years of the Ottoman and Persian domination. In Grigor’s Žamanaka-
grut‘iwn ‘Chronicle’ we read of two armies clashing with each other “like a fire
which runs in the reeds (orpēs zhur or yełēgn ənt‘asc‘i)” (Grigor Kamaxec‘i 1915:
41). Exactly the same image is used by Grigor to describe the devastation of a vil-
lage by a troop of horsemen (ibid.: 563), whereas Simēon speaks of an analogous
situation using almost the same words “they run like a fire through the reeds
(ibrew zhur ənd ełēgn ənt‘anayin)” (Simēon Erewanc‘i 40). Even more interesting
are the two occurrences of the simile in Aŕak‘el’s Girk‘ Patmut‘eanc‘ (Book of His-
tories). In ch. 4 we are told that the troops of Shah ‘Abbās were ordered to deport
the entire population of the land of Ararat and its surrounding regions, and “like a
wind-fueled fire running through the reeds (ibrew hołmnaxaŕn hur ənd ełēgn
ənt‘ac‘eal),” they drove the inhabitants out of their homes (Aŕak‘el Davrižec‘i
1990: 78; cf. Engl. trans. by Bournoutian 2005: 38). The second passage deals with
the forced conversion to Islam of the Jews living under the dominion of the
Persian king. As soon as the decree of the king was dispatched to the governors,
they, “like a fire blazing through the reeds (orpēs hur borbok‘eal ənd ełēgn),”
assembled all the Jews and forced them to apostatize (Aŕak‘el Davrižec‘i 1990:
370; cf. Engl. trans. by Bournoutian 2006: 359). Even if the adjective hołmnaxaŕn
lit. “mixed with the wind” reminds the Ayādgār ī Zarērān (“the deity of fire
plunges into a reed bed and the wind assists him”), the point of view of the
author is completely different. The positive energy which was represented by the
fire burning the reeds in the Iranian Epic, in P‘awstos and in most of the ancient
Armenian works, has turned into the enemy’s destructive strength.
Two later authors show the same Biblical overtones present in Tovma’s Pat-
mut‘iwn. In his Ołb mayrak‘ałak‘in Stəmpolu ‘Elegy on the Capital City of Stəm-
pol/Istanbul’ Aŕak‘el Bałišec‘i (14th-15th century) dreams of a future crusade by
the Franks against the Muslims: “like a lightning in the reeds (orpēs kaycakn iyełe-
gan), they (= the Franks) shall descend upon the foreigners, and they shall drive
out the Tačiks, for they are against the Holy Cross” (Anasyan 1957: 72, vv. 213-216).
The Patmut‘iwn paterazmin Xotinu ‘History of the battle of Xot‘in’ by Yovhannēs
Kamenac‘i (16th-17th century), on the other hand, focuses on the battle of Khotin
(present-day Ukraine) between the Ottomans and the Polish-Lithuanian army,
which took place in 1621. In ch. 8 Yovhannēs describes the Christian forces strik-
ing the Muslim enemies “like a lightning through the reeds (orpēs zkaycakn ənd
22 Marco Bais

ełēgsn)” (Yovhannēs Kamenac‘i 1964: 52). In both instances, the word z-kaycakn
evokes T‘ovma Arcruni’s IV.9, with its Biblical implications, but, exactly as it was
the case in that passage, it means ‘lightning.’ The simile, in fact, compares the
swift and powerful military action of the Christians against the Muslims with a
lightning falling upon the reeds. We have to do with a positive strength, but there
seems to be no supernatural power leading the troops to victory. There is no
reminiscence of the metaphysical meaning of the epic image.
In conclusion, the simile of the fire running through the reeds is well attested
in ancient Armenian literature and it is used to describe a formidable assault of
an army. The connection of the image with the Iranian concept of kingship and
supernatural valour pointed out by N. Garsoïan on the base of evidences internal
to Armenian tradition is confirmed by the occurrence of the same simile in the
Iranian epic. The original value of the image still survives in some later attesta-
tions (Life of Nersēs ch. 2, T‘ovma Arcruni and possibly Smbat Sparapet), though
in a less clear and conscious form, and sometimes it is reinterpreted with a
Christian context. In most of the later occurrences, anyway, this meaning is
completely lost and the image is employed just as simile for the violence and de-
structiveness of an attack. That is the reason why it is applied not only to Arme-
nian troops but also to the enemies, like for instance in Aŕak‘el Davrižec‘i, but
this is the case already in Movsēs Kałankatuac‘i. T‘ovma Arcruni and some later
authors show a contact between the original image and a similar simile occur-
ring in the Bible. However, a closer analysis of the texts and a comparison with
the Septuagint shows that the interference of the Bible with the traditional Ar-
menian image is just a formal one.

BIBLIOGRAPHY
Agat‘angełos (1983), Patmut‘iwn Hayoc‘/Hayoc‘ Patmut‘yun, K‘nnakan bnagirə G. Ter-Mkrtč‘yani ev
S. Kanayanc‘i, ašxarhabar t‘argmanut‘yunə ev canot‘agrut‘yunnerə Aram Ter-Łevondyani,
Erevan.
Akopjan, A. A. (1987), Albanija-Aluank v greko-latinskih i drevnearmjanskih istočnikah, Erevan.
Aleksidze, Z. / J.-P. Mahé (2008), “Introduction”, J. Gippert; W. Schulze; Z. Aleksidze; J.-P. Mahé
(eds.), The Caucasian Albanian Palimpsests of Mt. Sinai, vol. 1, Turnhout, vii-xxiv.
Ališan, Ł. (1853), “Patmut‘iwn srboyn Nersisi Part‘ewi Hayoc‘ Hayrapeti. Yałags zaprmic‘ srboyn Gri-
gori Hayoc‘ Lusaworč‘i ew patmut‘iwn srboyn Nersisi Hayoc‘ Hayrapeti”, Sop‘erk‘ Hayka-
kank‘, vol. 6, Venetik: 9-147.
 An Iranian image in the Buzandaran Patmut‘iwnk‘ 23

Anasyan, H. S. (1957), Haykakan ałbyurnerə Byuzandiayi ankman masin, Erevan.

Aŕak‘el Davrižec‘i (1990), Girk‘ Patmut‘eanc‘. Ašxatasirut‘eamb L. A. Xanlariani, Erevan.
Bagratuni, A. (1852), Hayerēn k‘erakanut‘iwn i pēts zargac‘eloc‘, Venetik.
Bournoutian, G. A. (2005), The History of Vardapet Arakel of Tabriz, vol. 1, Costa Mesa, CA.
―― (2006), The History of Vardapet Arakel of Tabriz, vol. 2, Costa Mesa, CA.
Cereti, C. G. (2001), La letteratura pahlavi. Introduzione ai testi con riferimenti alla storia degli studi e
alla tradizione manoscritta, Milano.
Dowsett, C. J. F. (1961), The History of the Caucasian Albanians by Movses Dasxuranci, translated by
C. J. F. Dowsett, London.
Ełišē (1957), Vasn Vardanay ew Hayoc‘ paterazmin, i loys aceal bałdatut‘eamb jeŕagrac‘ ašxatu-
t‘eamb E. Tēr-Minasean, Erevan.
Garsoïan, N. G. (1989), The Epic Histories Attributed to P‘awstos Buzand (Buzandaran Patmut‘iwnk‘),
Cambridge MA.
Gnoli, G. (1999), “Farra(ah)”, Encyclopædia Iranica, vol. 9, New York: 312-319.
Grigor Kamaxec‘i (1915), Žamanakagrut‘iwn Grigor vardapeti Kamaxec‘woy kam Daranałc‘woy, hra-
tarakec‘ Mesrop Nšanean, Erusałēm.
Monchi-Zadeh, D. (1981), Die Geschichte Zarēr’s, ausführlich kommentiert von Davoud Monchi-Za-
deh, Uppsala.
Movsēs Kałankatuac‘i (1983), Patmut‘iwn Ałuanic‘ ašxarhi, K‘nnakan bnagirə ew neracut‘yunə Va-
rag Aŕak‘elyani, Erevan.
Movsēs Xorenac‘i (1913), Patmut‘iwn Hayoc‘, ašxatut‘eambm. Abełean ew S. Yarut‘iwnean, Tp‘łis.
Nor Baŕgirk‘ Haykazean Lezui (1836-1837) NBHL, erkasirut‘iwn eric‘ vardapetac‘ yašakertut‘enē me-
cin Mxit‘ay abbahōr h. Gabriēli Awetik‘ean, h. Xač‘atroy Siwrmēlean, h. Mkrtč‘i Awgerean,
2 vols., Venetik.
P‘awstos Buzandac‘i (1883), Patmut‘iwn Hayoc‘, S. Peterburg.
―― (1933), Patmut‘iwn Hayoc‘ i č‘ors dprut‘iwns, Venetik.
Russell, J. R. (1996), “A Parthian Bhagavad Gītā and its Echoes”, J.-P. Mahé; R. Thomson (eds.), From
Byzantium to Iran: In Honour of Nina Garsoian, Atlanta: 17-35.
―― (2002-2003), “Epic in the Armeno-Iranian Marchlands: The Metamorphoses of a Genre”, Jour-
nal of Armenian Studies 7, 1: 3-20.
Sebēos (1979), Patmut‘iwn Sebēosi, ašxatasirut‘yamb G. V. Abgaryani, Erevan.
Simēon Erewanc‘i, Yišatakaran, vol. 1, (quoted from the on-line text: http://digilib.am/digilib/?me
nu=&wrk=277&wrpg=0&aupg=0).Smbat Sparapet (1956), Taregirk‘, Venetik-S. Łazar.
Tēr-Vardanean, G. (2010), “Mesrop Vayoc‘jorec‘u Nersēs A. Part‘ew kat‘ołikosi patmut‘iwnə”, Mate-
nagirk‘ Hayoc‘, vol. 11, Ant‘ilias-Libanan.
Thomson, R. W. (1976), Agathangelos. History of the Armenians, Translation and Commentary by R.
W. Thomson, Albany.
―― (1982), Elishē, The History of Vardan and the Armenian War, Cambridge, MA.
―― (1985), Thomas Artsruni, History of the House of the Artsrunik‘, Translation and Commentary
by R. W. Thomson, Detroit.
24 Marco Bais

―― (1999), The Armenian History attributed to Sebeos, translated, with notes, by R. W. Thomson.
Historical Commentary by J. Howard-Johnston. Assistance from T. Greenwood, 2 vols, Liv-
erpool.
―― (2006), Moses Khorenatsi, History of the Armenians. Translation and Commentary on the Lit-
erary Sources by R. W. Thomson, Revised Edition, Ann Arbor.
T‘ovma Arcruni (1887), T‘ovmayi vardapeti Arcrunwoy patmut‘iwn tann Arcruneac‘, i loys ēac K‘.
P‘[atkanean], S. Peterburg.
Yovhannēs Kamenac‘i (1964), Patmut‘iwn paterazmin Xotinu. Ašxatasirut‘eamb H. A. Anasyani,
Erevan.
Ziegler, J. (1980), Septuaginta. Vetus Testamentum Graecum Auctoritate Academiae Scientiarum
Gottingensis editum, vol. XII,1, Sapientia Salomonis, edidit J. Ziegler, Göttingen.
―― (1983), Septuaginta. Vetus Testamentum Graecum Auctoritate Academiae Scientiarum Gottin-
gensis editum, vol. XIV, Isaias, edidit J. Ziegler, Göttingen.
Zōhrapean, Y. (1805), Astuacašunč‘ Matean Hin ew Nor Ktakaranac‘, 4 vols, Venetik.
Zuckerman, C. (2007), “The Khazars and Byzantium – The First Encounter”, P. B. Golden; H. Ben-
Shammai; A. Róna-Tas (eds.), The World of the Khazars. New Perspectives. Selected Papers
from the Jerusalem 1999 International Khazar Colloquium Hosted by the Ben Zvi Institute,
[Handbook of Oriental Studies 17], Leiden-Boston: 399-432.
 The “Bun-Turks” in Ancient Georgia*

Jost Gippert
Goethe Universität, Frankfurt

Abstract
The paper deals with the identification of the so-called “Bun-Turks” that are mentioned in several
historical texts as a tribe which settled in Georgia in prehistoric times. On the basis of a thorough
comparison of the relevant Georgian and other sources, the term is shown to have emerged from a
corruption of the name of the Huns, which occurs in similar contexts, together with other designa-
tions of Turkic tribes. The available text materials further suggest that the historical basis for the
mentioning of the “Bun-Turks” as settlers in Georgia was the Khazar attacks of the VIth-VIIth cen-
turies, which were secondarily re-projected into prehistoric times.

Keywords
Bun-Turks, Huns, Qypčaqs, Khazars; Kartlis Cxovreba; Mokceva Kartlisa; Alexander Romance; Šāh-
nāme

Turkic languages are an integral part of the linguistic landscape of present-day

Caucasia, both north and south of the mountain ridge. However, different from
the so-called “autochthonous” Caucasian languages, i.e. the languages pertaining
to the Kartvelian (South Caucasian), (North-) West Caucasian and (North-)East
Caucasian families, both the southern (Oghuz) and the northern (Qypchaq) idi-
oms of Turkic stock are generally believed to have entered the area in relatively
recent times. Nevertheless there are explicit indications of ancient contacts be-
tween Caucasian and Turkic peoples in historical sources from the area itself.
The present paper deals with one of these traditions, viz. that of the “Bun-Turks”
mentioned in Old Georgian historiography.
Even though there is good reason to believe that Old Georgian literacy
emerged about the same time as that of Old Armenian, by the beginning of the

*
The main points of the present article were first presented on the conference “Anatolia –
Melting Pot of Languages” in Istanbul on May 28, 2005.
 Koninklijke Brill NV, Leiden, 2015
26 Jost Gippert

Vth c. A.D., Georgian historiography differs from that of its neighbours in that
the sources that have come down to us are of a much later origin. As a matter of
fact, no Georgian fellow-historians are known of authors such as Koriwn, Aga-
thangelos, Łazar Parpec‘i (all Vth c.), Ełišē, Sebēos (VIIth c.), Łewond (VIIIth c.),
or Movsēs Xorenac‘i (IXth c.). In Georgian tradition, we must wait until the Xth c.
for the first noteworthy account of the history of the country to be written down;
this is the anonymous text on the “Conversion of Kartli (East-Georgia)”, Mokce-
vay Kartlisay, which contains, beside the legend of the conversion of King Mirian
by a captive woman called Nino, a brief chronicle extending from prehistorical
times to the IVth c. A.D. The “Conversion”, existing in four different versions1 and
representing a compilation of various older sources,2 was later used by the
bishop Leonṭi Mroveli (Leontius of Ruisi) who in the XIth c. authored the initial
parts of Kartlis Cxovreba, the Georgian “Chronicle”, which was steadily contin-
ued until the XVIIth c.3 Apart from these works, it is only a few hagiographic
texts that may be regarded as authentic historical sources of first millenium
Georgia.4
Within the “Conversion of Kartli”, the people called bun-turkni, i.e. “Bun-
Turks”, play a prominent rôle indeed. In the most comprehensive version of the
text, that of the Šaṭberd codex of the late Xth c., they are mentioned as inhabit-
ants of East Georgia right at the beginning, in connection with an enigmatical
account of a king named Alexander:

1
The versions of Mokcevay Kartlisay (MK) are contained in one codex each of Šaṭberd (Xth c.)
and Č̣ eliši (XIIIth c., cf. Lerner 2004a), and two manuscripts of St. Catherine’s monastery on Mt. Si-
nai (Xth c.). The text of the Šaṭberd codex has been edited in Gigineišvili/Giunašvili (1979: 320-355)
and, in parallel with the Č̣ eliši codex, in Abuladze (1963: 82-163); of the two Sinai manuscripts (N48
and N50), only the latter is available via the facsimile edition in Aleksidzé (2001: 73-215). The (frag-
mentary) second Sinai manuscript (N48) was investigated in situ by the present author in 2010; it
does not overlap with N50.
2
Cf. Gippert (2006) for a discussion of several relevant cases.
3
For Kartlis Cxovreba (KC) cf. the edition Q̣ auxčišvili (1955-9). – The question of the dating of
the model used by Leonṭi cannot be discussed here (the VIIIth c. has recently been proposed in
Rapp 1999: 80 and 2006: 175).
4
These are the legends of St. Šušaniḳ (VIth c.), St. Evsṭati of Mcxeta (VIth c.), St. Habo of Ṭpilisi
(VIIIth c.), and a few other ones, all edited in Abuladze 1963.
 The “Bun-Turks” in Ancient Georgia 27

MK.S 320,2–6:5
odes aleksandre mepeman natesavni igi lotis šviltani c̣ arikcina da šeqadna igini
ḳedarsa mas kueq̇ anasa, ixilna natesavni sasṭiḳni bun-turkni, msxdomareni mdi-
naresa zeda mṭḳuarsa mixuevit, otx kalakad, da dabnebi mati: sarḳine-kalaki,
ḳasṗi, urbnisi da oʒraqe.
“After Alexander the King (had) conquered the descendants of the children of Lot
and dispelled them into the land Ḳedar (?), he saw the fierce tribes (of) the Bun-
Turks who resided along the river Kur, in four cities, and their villages (were)
Sarḳine-City, Ḳasṗi, Urbnisi and Oʒraqe”.6
From Leonṭi Mroveli’s adaptation of the passage it is clear that the king in
question is Alexander the Great, but neither the “children of Lot” nor the “land
Ḳedar” appear here:
KC. L.Mr. 17,6–87:
aman aleksandre daiṗq̇ rna q̇ ovelni ḳideni kueq̇ anisani. ese gamovida dasavlit, da
ševida samqrit, šemovida črdilot, gardamovlna ḳavḳasni da movida kartlad...
“That Alexander conquered all the edges of the land. He started from the west,
and went south, entered northwards, transgressed the Caucasus (mountains) and
came to Kartli...”
It is but a vague idea that the “land Ḳedar” of the “Conversion”, ḳedarsa mas
kueq̇ anasa, might have been replaced by the “edges of the land”, ḳideni kueq̇ ani-
sani, in this text,8 and that the “children of Lot” have their counterpart in the
“northward” direction, črdilot, of Alexander’s progression. As both the “descend-
ants of Lot” and a land (or, rather, tribe) named “Kedar” are Biblical topoi,9 it may

5
MK.S = the text as appearing in the Šaṭberd version of the legend (here quoted by pages and
lines of the edition Gigineišvili/Giunašvili 1979); of the other versions, none has the initial para-
graphs forming the “Primary History of Georgia” (thus the term introduced by Rapp 1999: 82).
6
For the place names concerned cf. the map (by Robert H. Hewsen) in Rapp (1999: 128).
7
Here quoted by pages and lines of the edition Q̣ auxčišvili (1955).
8
Gertrud Pätsch in her German translation of the “Conversion” (1975: 290 n. 2, referring to
Čikobava 1955: 1120) considers to see Modern Georgian ḳedaro- “side, edge” (“Seite, Rand” / “მხარე,
კიდე”) here, suggesting that ḳedar- in the “Conversion” might be interpreted as “outlying” (“könnte
auch in diesem Sinne als «abseits gelegen» gedeutet werden”). The stem ḳedaro- seems not be at-
tested anywhere in Old Georgian, however, so that we should rather assume a corruption of *ḳide-
instead. – Old Georgian ḳedar-, the name of the “cedar tree”, can be excluded in the given context
as we have an appositive construction “in the land ḳ.” in “ḳedarsa mas kueq̇ anasa”, not a genitival
syntagm “in the land of the cedar(s)” (*kueq̇ anasa mas ḳedarisasa/ḳedartasa).
9
Cf. Deut. 2,9.19 and Ps. 83,8 (82,9) for the “children of Lot”. Note that in the Old Georgian Bible
tradition, it is not švil- “child” but ʒe- “son” that is used in these passages. Cf. below for another sug-
gestion as to the “children of Lot”. – For the “land Kedar” cf. the “tents of Kedar” mentioned in Ps.
28 Jost Gippert

well be the text of the “Conversion” that has undergone changes here, rather
than Leonṭi’s which must have relied upon a model quite distant from the
Šaṭberd version of the legend.
Different from the “Conversion”, Leonṭi Mroveli continues not with “Bun-
Turks” but with “Kartvelians”, i.e., Georgians, in the present context, and in a
very unfavourable manner indeed:
KC. L.Mr. 17,8–11:
... da ṗovna q̇ ovelni kartvelni uboroṭes q̇ ovelta natesavta sǯulita. rametu col-
kmrobisa da siʒvisatws ara učnda natesaoba, q̇ ovelsa suliersa č̣ amdes, mḳudarsa
šesč̣ amdes, vitarca mqecni da ṗiruṭq̇ uni, romelta kcevisa c̣ armotkma uqm ars ...
“... and he found all (the) Georgians worse than all tribes by (their) faith. For they
did not care of (sanguinal) relations in marriage and matrimony, used to eat eve-
rything living and (even) dead, just like beasts and wild animals, whose customs
are impossible to describe...”
However, in Leonṭi Mroveli’s treatise, it is not the Kartvelians alone that are
ascribed these raw manners. Immediately afterwards, the author agrees with the
“Conversion” again in introducing the “Bun-Turks”, too. But different from the
latter text, the term is here combined with another designation of a Turkic tribe,
viz. q̇ ivčaq̇ -, i.e., Qypchaqs:
KC. L.Mr. 17,11–13:
da ixilna ra ese natesavni sasṭiḳni c̣ armartni, romelta-igi čuen bunturkad da
q̇ ivčaq̇ ad uc̣ odt, msxdomareni mdinaresa mas mṭḳurisasa mixvevit, dauḳwrda ese
aleksandres, rametu ara romelni natesavni ikmodes mas.
“And when he saw these fierce pagan tribes, whom we call Bun-Turks and Qyp-
chaqs, who resided along the river Kur, Alexander was astonished, for no (other)
tribes would do the (same).”

119 (120), 5; it is this verse that is quoted s.v. ḳedari in the XVIIth c. Georgian lexicon by Sulkhan-
Saba Orbeliani (1965: 367). Saba’s translation “ბნელი საჭმუნავი”, i.e. “woeful dark”, adapted by
Pätsch (1975: 290 with n. 2: “das dunkle Land”), is obviously based on an etymological connection
of the Biblical name of the tribe of the sons of Ishmael, ‫ = ֵקדָ ר‬qedār (Gen. 25,13 etc.) with the root
‫ = קדר‬qdr “to be dark, darken”. Given that both the Septuagint and the Armenian Bible leave the
name of the tribe untranslated in Ps. 119,5 (Κηδαρ / kedar-), there is no reason to believe that ḳedar-
existed as a common noun meaning “dark” in Old Georgian, even though Saba’s entry seems to be
supported by the “Conversion” itself which has the sentence c̣ arvedit bnelta črdiloysata mtata mat
ḳedarisata “go away into the darknesses of the North, into the mountains of Kedar” later (MK.S
341,42-345,1). This, however, only proves that the etymological connection of the name with Hebr.
qdr was widespread long before Saba. – The translation “land of midnight” proposed by Rapp with
reference to “Khurâsân, the great eastern province of Persia” denoting the “east” (1999: 94) has no
basis whatsoever.
 The “Bun-Turks” in Ancient Georgia 29

Who, then, are the “Bun-Turks” who are reported here to have lived along
with Kartvelians and Qypchaqs in East Georgia by the time of Alexander? As a
matter of fact, several explanations have been proposed for their name, which
seems not to be attested as such outside of Old Georgian sources. The first pro-
posal was made by Marie-Félicité Brosset (1849: 33) who regarded bun-turk- as a
compound denoting “Turks primitifs”. This assumption is in accordance with the
use of the word bun- in Old Georgian, esp. of its derivative buneba- which is the
general term for “basis” or “nature”. It is further supported by two later revisions
of St. Nino’s legend10 which allude to buneba- explicitly in the given context, in a
sort of lucus a non lucendo argumentation:
N.A. 46,15–18:
ixilna natesavni igi sasṭiḳni c̣ armarttani, romelta čuen ac̣ bun-turkad da q̇ ivčaq̇ ad
uc̣ odt, msxdomareni mdinaresa zeda mṭḳurisasa mixuevit, da kalakni matni ʒlierni
da cixeni priad magarni, da cxondebodes igini q̇ ovlad ucxod ḳacta bunebisagan, vi-
tarca mqecni da ṗiruṭq̇ uni, romelta kcevisa c̣ armotkumay uqmar ars.
“And he saw the fierce tribes of the pagans, whom we now call Bun-Turks and
Qypchaqs, residing along the river Kur, and their strong cities and very firm
strongholds, and they lived (in a way) totally deviant from the nature of men, like
beasts and animals, whose customs are impossible to describe.”
N.B. 79,26–80,3:
da ixilna natesavni sasṭiḳni c̣ armartni, romelta čuen ačat-bun-turkad uc̣ est, rametu
ixilvebodes igini q̇ ovlad ucxod ḳacta bunebisagan, vitarca iq̇ vnes mqecni rayme
saʒulvelni.
“And he saw the fierce pagan tribes, whom we call Ačat-Bun-Turks,11 for they
looked totally deviant from the nature of men, because they were somewhat ugly
beasts.”
Brosset’s proposal was but slightly altered by Nikolai Marr who suggested a
translation “коренной турокъ”, i.e., “original” or “old-established Turk”, assum-
ing “корень, основаніе” (“root, basis”) to be the underlying meaning of bun-
(Marr 1901: LXII). At the same time, Marr rejected the interpretation published
by Ekvtime Taq̇ aišvili in the first edition of the “Conversion”, according to whom
the word might denote Turks as “spear-bearers” (“будет означать турка-копье-
носца”) (Takajšvili 1900: 1–2 n. 2). As Marr correctly observed, bun- nowhere

10
N.A. (metaphrastic version by Arsen Beri): ca. XIIth c.; N.B. (anonymous metaphrastic ver-
sion): ca. XIIIth c.; both quoted by pages and lines after the edition Abuladze (1971).
11
Note that ačat- is unexplained. Should this be a corruption of q̇ ivčaq̇ - rather than of ac̣ “now”
as in N.A.?
30 Jost Gippert

means “spear-bearer” nor even “spear” alone; in the combination bun- horolisa-
appearing, e.g., in the Šaṭberd codex within the Treatise on David and Goliath by
Hippolytus (243,26; 244,33) as a quotation from II Kings (II Sam.) 21,19, it is horol-
which denotes the weapon, while bun- designates the “shaft” (“ратовище”) as its
“basis” or “handle” (“основаніе, рукоятка”).12 Marr was also right in underlining
the coincidence with Armenian which has bown gełardan in I Kings (I Sam.) 17,7
as a perfect equivalent of bun- horolisa-. And there is hardly any room for doubt
that both Armenian bown and Georgian bun- lastly reflect Middle Persian bun
with its meanings “base, foundation, bottom” as proposed by Heinrich Hübsch-
mann (1897: 123–4), Ilia Abuladze (1944: 085), and Mzia Androniḳašvili (1966:
297). Thus the assumption that the term “Bun-Turks” means something like
“primeval” or “original” Turkic inhabitants of Kartli seems to be well founded.13
However, a different view suggests itself when we consider the information
provided in Mokcevay Kartlisay and Kartlis Cxovreba in a broader context. As a
matter of fact, Leonṭi’s text strongly reminds of a certain type of medieval leg-
ends on Alexander the Great that have come down to us in other languages, viz.
Greek, Armenian, and Syriac. As a close parallel we may quote the prose version
of the “Christian Legend”, which is preserved in the latter language as an appen-
dix to the Alexander Romance proper.14 Here, both Alexander’s travels into the
Caucasus and the wild appearance of the people living there are described in a
very similar way:
CL. 260,15–264,2 / 148,35–151,7:
“And Alexander looked towards the west ... then they went down to the source of
the Euphrates ... and they came to the confines of the north, and entered Arme-

12
“Šubis ṭari”, i.e. “spear shaft”, is noted as the meaning of buni in Sulkhan-Saba’s lexicon (Orbe-
liani 1965: 124); in a second entry, the same word is translated by “saqelsakmre”, i.e. “(tool) for
handicraft” (?; ib.; correspondingly in Čubinašvili 1887: 123).
13
Marr’s interpretation “original Turk” has recently been sustained by Rapp (1999: 95). Accord-
ing to Culaja (1979: 60 nn. 85 and 89), the term was used in referring to the pre-Hellenic period. – A
different solution has recently been published by K. Lerner (2004b: 224) who proposed to see the
influence of a “supposed Semitic substratum” here, deducing the term from “Hebrew, bney-Turks =
‘seed, sons of the Turks’”. It seems, however, that the “Old Hebrew Romance” on Alexander Lerner
refers to does not contain this notion, and the phonetic reshaping to be assumed in this case is not
paralleled anywhere else.
14
Cf. the edition and translation in Budge (1889: 255-275 and 144-158); in the present paper,
only the translation will be quoted (as CL). For an account of the Syriac manuscripts containing
the Alexander Romance and a summary of the “Legend” cf. Hunnius (1904: 9) and, more recently,
Ciancaglini (2001). The XIIth c. “Book of the Bee” referred to by Rapp (1999: 98) stands farther off.
 The “Bun-Turks” in Ancient Georgia 31

nia and Âdarbaijân and Inner Armenia ... and he went and encamped by the gate
of the great mountain. ... Alexander said, “This mountain is higher and more ter-
rible than all the mountains which I have seen. ... Who are the nations within this
mountain upon which we are looking? ... What is their appearance, and their
clothings, and their languages?” ... “They wear dressed skins; and they eat the raw
flesh of everything which dies of theirs; and they drink the blood of men and
animals. ...”
And of course, Alexander’s question as to what nations he is looking at is an-
swered as well:
CL. 263,2–5 / 150,20–24:
“Alexander said, “Who are the nations within this mountain upon which we are
looking? ...” The natives of the land said, “They are the Huns.” He said to them,
“Who are their kings?” The old men said: “Gôg and Mâgôg and Nâwâl the kings of
the sons of Japhet ...”
This parallel suggests off-hand that the name of the “Bun-Turks” might have
emerged from a corruption of the name of the “Huns”, which would presuppose
a confusion of h- and b- if Syriac )YNwh = hunāyē as occurring in the given pas-
sage (263,4) was the model. The same would hold true if bun- should reflect
Greek οὗνν- still spoken hun- with initial aspiration; this assumption is valid even
though none of the existing Greek versions of the legend seems to use this name.
The closest parallel we find among these versions is surely that of recension λ of
the Alexander Romance ascribed to (Pseudo-)Kallisthenes (cf. the edition in
Thiel 1959). This text does agree with the Syriac legend in denoting the “tribes of
the North” as descendants of Japhet,15 thus suggesting that the enigmatical “chil-
dren of Lot” we found in the Georgian “Conversion” might have emerged from a
corruption of “children of Japhet” (*iapetis švilni).
Ps.-Kall. Rec. λ, III,29 (51,10–53,7 ed.Thiel)
᾽Εξελϑὼν δὲ ᾽Αλέξανδρος ... ἔδοξεν αὐτῷ πορευϑῆναι ἐπὶ τὰ βόρεια µέρη. εὗρε δὲ ἐκεῖ
ἔϑνη πονηρὰ ἐσϑίοντας σάρκας ἀνϑρώπων καὶ πίνοντας αἷµα ζώων [καὶ ϑηρίων] ὥσπερ
ὕδωρ. ἰδὼν δὲ ὁ ᾽Αλέξανδρος ἐϕοβήϑη αὐτούς· ἦσαν γὰρ οἱ τοῦ ᾽Ιάϕεϑ ἀπόγονοι ... τοὺς
νεκροὺς οὐκ ἔϑαπτον, ἀλλ’ ἤσϑιον αὐτούς. ...
“But Alexander went off ... and it seemed good to him to travel into the northern
lands. There he found worthless people eating human flesh and drinking the

15
The same notion is also found in the parallel passage of two redactions of the Apocalypsis by
(Pseudo-) Methodius (edited in Thiel 1959: 72-75); here we read: ἔνθα καὶ ἑώρακεν ἔθνη ἀκάθαρτα καὶ
δυσειδῆ ἅ εἰσι τῶν υἱῶν Ἰάφεθ ἀπόγονοι / ἔνθα καὶ ἑώρακεν ἔθνη ἀκάθαρτα καὶ εἶδεν ἐκεῖ ἐκ τῶν υἱῶν
Ἰάφεθ ἀπογόνους (72, 5-7/ 73, 5-6).
32 Jost Gippert

blood of living beings [and animals] like water. When Alexander saw them, he
was affrighted, for they were descendants of Japhet ... they did not bury the dead
but ate them. ...”
In ending up the account of Alexander’s journey to the North, the same text
introduces the ethnonym of the “Turks”, too, thus indicating a possible source for
the second part of the quasi-compound bun-turk- of the Georgian tradition:
Ps.-Kall. Rec. λ, III,29 (57,4–6 ed.Thiel):
ἐκκαϑάρας οὖν τὰ µέρη τοῦ βορρᾶ ἐκ τῶν µιαρῶν ἐκείνων ἐϑνῶν ἔπτισα δὲ τοῖχον πρὸς
ἀνατολὰς πηχέων π’ τὸ ὕψος καὶ πηχῶν κ’ τὸ πλάτος. καὶ διεχώρησα ἀναµέσον
Τούρκων καὶ ᾽Αρµενίων.
“Having cleansed the lands of the North from the defilements of those people, he
built a wall against the north, 70 cubits high and 20 cubits wide, and passed
through the Turks and the Armenians.”
The identification of bun- with the name of the Huns still hits on two prob-
lems. First, the replacement of h- by b- can by no means be motivated phoneti-
cally,16 and we must assume some sort of paleographic confusion instead. This
assumption is equally hard to prove but not improbable. If the replacement took
place within Georgian, we must presuppose that the script involved was the an-
cient majuscule script, Asomtavruli, as only in this script the letters <b> and <h>
are similar enough to be confusable;17 cp. the two letters in VIIIth c. Asomtavruli
(Ⴁ vs. Ⴠ), XIth c. minuscule script (Nuskhuri: ⴁ vs. ⴠ), and Modern Mkhedruli
script (ბ vs. ჰ). It must be admitted in any way that a common prototype of the
“Conversion of Kartli” and Leonṭi Mroveli’s account was written in Asomtavruli
majuscules, given that similar confusions must be assumed for other passages of
the “Conversion”, too.18

16
A. Vovin (personal communication of 2005) drew my attention to the Greek ethnonym ϕρυν-
which occurs in Strabo’s Geography (11,11,1,15) and denotes a people in the neighbourhood of the
Chinese (σηρ-) and Bactria; this might represent an older variant of the name of Huns (< *hwrung)
and underlie Georgian bun-. The sound substitution involved (*fr > b) would be unparalleled, how-
ever, even though Old Georgian does possess examples of Middle Iranian fr- being substituted by
br- (e.g., *frazēn “wise” > brʒen-i, cf. Gippert 1993: 223-4 and 267-8) as well as hr- being substituted
by pr- (e.g., prom- “Rome” vs. hrom- “id.”, via *fr-; cf. Blake 1923: 84-7; Peeters 1926: 76-7).
17
“A conflation of the designations Hun (Honi) and Turk (T‘urk‘i)” was also considered hesitat-
ingly by Rapp (2003: 149 n. 185) but rejected on paleographic grounds.
18
Cf. Gippert 2006: 114-6 for an example. – Note that the Sinai manuscript N48 (cf. note 1
above) is peculiar for the fact that it contains various lines written in Asomtavruli letters in an oth-
erwise Nuskha-Khutsuri based context (cf. Gippert 2010, n. 23), thus proving that older versions of
the “Conversion” written in Asomtavruli letters may well have existed.
 The “Bun-Turks” in Ancient Georgia 33

The second problem consists in the fact that the name of the Huns does oc-
cur in Old Georgian sources in a different form, viz. hon-. As a matter of fact, this
form is met with in the “Conversion” itself, side by side with bun-turk-, in the
continuation of the passage treated above:
MK.S. 320,7–16:
dauḳwrda aleksandres da cna, rametu ieboselta natesavni iq̇ vnes: q̇ ovelsa qorcielsa
č̣ amdes da samare mati ara iq̇ o, mḳudarsa šešč̣ amdes. da ver eʒlo brʒolay mati
mepesa da c̣ arvida. mašin movides natesavni mbrʒolni, kaldeveltagan gamo-
sxmulni, honni, da itxoves bun-turkta uplisagan kueq̇ anay xarḳita. da dasxdes igini
zanavs. ... da šemdgomad raodenisa-me žamisa movida aleksandre, mepē q̇ ovlisa
kueq̇ anisay, da dalec̣ na samni ese kalakni da cixeni, da honta dasca maxwli.
“And Alexander was astonished and realised that they were descendants of the
Jebusites: they used to eat all (kinds of) meat and had no cemeteries, (because)
they used to eat the dead. And the king could not fight against them and went
away. Then came martial tribes, an offspring of the Chaldees, Hons, and they
asked the ruler of the Bun-Turks for tributed land. And they settled in Zanavi. ...
And after some time, Alexander, the king of all the land, came (again) and de-
stroyed these three cities and fortresses and defeated the Hons with the sword.”
This notion seems to imply that the hon-ni and the bun-turkni cannot be the
same people. Leonṭi Mroveli’s account is not helpful in this context at first
glance as it mentions only the “tribes of the Chaldees”:
KC. L.Mr. 17,14–16:
aramed mas žamsa ver uʒlo, rametu ṗovna cixeni magarni da kalakni ʒlierni.
ḳualad gamovides sxuani natesavni kaldevelni, da daešennes igini-ca kartls.
šemdgomad amissa ganʒlierda aleksandre da daiṗq̇ ra q̇ oveli kueq̇ ana, da
aġmovida kueq̇ anasa kartlisasa. da ṗovna cixe-kalakni ese ʒlierni šua-kartl: ...
urbnisi, ḳasṗi ... sarḳine, da zanavi, ubani uriata ...
“But at that time, he was not able (to fight against them), because he detected
(their) firm fortresses and strong cities. (And) again, other Chaldean tribes came,
and they, too, settled in Kartli. After this, Alexander gained strength and con-
quered all the land, and he came to the land of Kartli. And he found these strong
fortified cities in Inner-Kartli: ... Urbnisi, Ḳasṗi ... Saṛkine, and Zanavi, the quarter
of the Jews ...”
We must note, however, that Leonṭi’s text contains another type of infor-
mation that might be decisive here. Based on his equation of Zanavi with a
“quarter of the Jews”, Ekvtime Taq̇ aišvili proposed to read ho~nni as an abbrevi-
ated form of “ჰურიანნი, т.е. евреи” in the Conversion (Takajšvili 1900: 5 n. 1). As
the Georgian Jews are generally believed to be of Babylonian provenance, this
34 Jost Gippert

explanation seems to have a good deal in its favour,19 even though it does not ac-
count for the double n, the plural nominative of huria- “Jew” being huriani
throughout.
On the other hand, this explanation will not work for the dative plural form
honta appearing at the end of the passage, as the corresponding form of huria-
would be huriata, with no n at all. What is more, there can be no doubt that Old
Georgian did possess a stem hon- denoting the “Huns”. This is attested, e.g., in
the legend of St. Šušaniḳ,20 allegedly an authentic report of the late Vth or early
VIth century written by a contemporary of the Saint, and generally assumed to
be the oldest extant non-translated literary text in Old Georgian. Here, the peo-
ple named hon- are the adversaries of the vicegerent (ṗiṭiaxši, vitaxa) of Kartli:
Šuš. VII: 19,2-5:
da man mrkua me: “uc̣ q̇ ia, xuces, me brʒolad c̣ arval honta zeda. da čemi samḳauli
mas ara dauṭeo, odes igi ara čemi coli ars – iṗoos vinme, romelman ganḳapos igi.” ...
da vitar moic̣ ia aġvsebisa oršabati da movida ṗiṭiaxši brʒolisa misgan hontaysa,
ešmaḳi txrida gulsa missa.
“And he (the vitaxa) told me: ‘Do you know, priest, I am going to fight against the
Huns, And don’t leave her my jewellery as long as she isn’t my wife – someone
will be found who will wear it out.’ ... And as Easter Monday came and the vitaxa
returned from the fight against the Huns, the devil was stirring (lit. digging) his
heart up.”
There can be no doubt that the “Huns” here referred to are the same as those
mentioned, under the same name, hon-, in Armenian historiographic texts such
as Agathangelos’ History of the Armenians:21
Agath. 19: 16,6–10:
... sksanēr Xosrov tʿagaworn Hayocʿ gownd kazmel ... gowmarel zzōrs Ałowanicʿ ew
Vracʿ, ew banal zdrowns Alanacʿ ew zČoray pahakin, hanel zzōrs Honacʿ, aspatak
dnel i kołmans Parsicʿ, aršawel i kołmans Asorestani, minčʿew i drowns Tisboni:
“... Xosrov, the king of the Armenians, began to assemble an army, ... to take to-
gether the troops of the Albanians and the Georgians, and to open the gates of
the Alans and the guard of Čor (Derbent),22 to extract the troops of the Huns,

19
Cf. Bielmeier (1990: 32) who connects the name Zanavi with the Hebrew place name zānūḥ
(Zanoah) appearing in the Old Testament (Jos. 15,34 etc.).
20
Here quoted by pages and lines of the edition in Abuladze 1963.
21
Here quoted by paragraph numbers, pages and lines after the edition Tēr-Mkrtčean/Kana-
yeancʿ (1909).
22
Note that the historical setting of St. Šušaniḳ’s legend agrees with that of Agathangelos’ His-
tory in mentioning Derbent under the name Čor-: xolo ṗiṭiaxši čord c̣ aremarta da ǯoǯiḳ, ʒmay misi,
 The “Bun-Turks” in Ancient Georgia 35

make an attack into the regions of the Persians, to invade the regions of Assyria
up to the gates of Ktesiphon.”
Unfortunately, the XIIIth c. Armenian translation of the Georgian Chronicle,
Patmowtʿiwn Vracʿ, does not contain any information on the “Bun-Turks” or
“Huns” in the passage corresponding to Leonṭi Mroveli’s treatise on Alexander’s
journey, its text being abridged in the present context as elsewhere:
PV 24,1–25,3:23
Yaynm žamanaki ambarjaw mecn Ałekʿsandr... Sa ekn yarewmticʿ ar̄ hiwsisiw, ew
šrǰeal ənd arewels, emowt i cmakayin erkirn. ew ēancʿ ənd Kovkasow lear̄ n yašxarhn
Vracʿ. ew hiacʿaw ənd zazir keans nocʿa: Ew zi etes amrocʿs bazowms, ew ašxateacʿ
zzawrs iwr amiss vecʿ yar̄ nowln znosa‘ zCownda, zXertʿwis, ZOwnjrxē, kar̄ owcʿealn
ənd kʿarin Ladasoy, zTʿowłars i veray getoyn Speroy, or asi Čorox, zOwrbnis, zKasb,
Owpʿliscʿixē, or asi Tear̄ n-berd, zMcʿxetʿa – ztʿałkʿn‘ or Sarkina kočʿecʿaw, zCʿixēdid,
or ē berd mec, ew Zawanoy tʿałn Hrēicʿ ...
“At that time, the Great Alexander arose... He came from the west to the north,
and having travelled through the east, he entered the land of the shadow.24 And
he went from Mt. Caucasus into the land of the Georgians. And he was aston-
ished about their disgustful life. And as he saw many strongholds (there) and he
was busy for six months with his troops (trying) to conquer them, (viz.) C̣ unda,
Xertvisi, Oʒrqe which was built at the rock of Ladasi, Tuġarisi above the river
Sṗeri, which is (also) called Čoroxi, Urbnisi, Ḳasṗi, Upliscixe which means For-
tress of the Lord, Mcxeta – (its) quarters which are called Sarḳine, Cixedidi,
which means Big Fortress, and Zavani (= Zanavi!), the quarter of the Jews ...”

In another passage, however, the Patmowtʿiwn Vracʿ does use the term hon-.
This passage is concerned with King David the Builder (Davit Aġmašenebeli) and
his wife Guaranduxṭ, and the period in question is the XIth–XIIth century A.D.:
PV 244,3–9:
Ew kin nora Gorandowxt dowstr ēr Kiwčʿałacʿ glxaworin, aysinkʿn Honacʿ` Atʿrakay.
ew nocʿa awgnowtʿeambn hnazandeacʿ ztʿagaworsn Awsetʿoy. ew ar̄ patands i
nocʿanē, ew arar xałałowtʿiwn i mēǰ Awsacʿ ew Honacʿ.

ara daxuda, odes sakmē ese ikmna c̣ midisa šušaniḳis zeda “But the vitaxa had moved off to Čor, and
Ǯoǯiḳ, his brother, was not present when this affair happened to St. Šušaniḳ” (ch. X: 22,11-12). For
other peculiarities of the Old Georgian legend agreeing with features of Old Armenian cf. Gippert
1991: 82-84.
23
Cf. the edition in Abuladze (1953), here quoted by page and lines (as PV).
24
Cmakayin erkir is a literal translation of *kueq̇ ana- črdiloysa-, lit. “land of the shadow”, the
term underlying Georgian črdiloet-i “North”.
36 Jost Gippert

“And his wife Gorandowxt was the daughter of the head of the Kipchaks, i.e., the
Huns, Atʿrakay. And with their help, he subdued the kings of Ossetia. And he took
hostages from them, and he made peace between the Ossetes and the Huns.”
Equating the hons with the kiwčʿałs, i.e., the Qypchaqs, the Armenian text
differs considerably from its Georgian model, the chapter on Davit Aġmašene-
beli of Kartlis Cxovreba,25 which uses only the term q̇ ivčaq̇ - here:
KC. D.A. 336,4–18:
moeq̇ vana sanaṭreli da q̇ ovlad gantkmuli siḳetita guaranduxṭ dedopali, švili
q̇ ivčaq̇ ta umtavresisa atraka šaraġanis ʒisa ... amistws-ca c̣ aravlinna ḳacni
sarc̣ munoni da mouc̣ oda q̇ ivčaq̇ ta da simamrsa twssa. ševides ovsets da moegebnes
mepeni ovsetisani da q̇ ovelni mtavarni matni, da vitarca monani dadges c̣ inaše
missa. da aġixunes mʒevalni ortagan-ve, ovsta da q̇ ivčaq̇ ta, da esret advilad še-
aertna orni-ve natesavni. da q̇ o šoris matsa siq̇ uaruli da mšwdoba vitarca ʒmata.
“He had married the blessed and very beautiful queen Guaranduxṭ, a child of the
leader of the Qypchaqs, Atraka the son of Šaraġan ... Therefore he sent out faithful
men and invited the Qypchaqs and his father-in-law. They entered Ossetia, and
the kings of Ossetia and all their leaders approached them and stood like servants
in front of them. And they took hostages from both the Ossetes and the Qyp-
chaqs, and in this way he easily reunited them. And he made happiness and
peace between them like brothers.”
On the other hand, it is just this equation which is reminiscent of the “Bun-
Turks” and “Qypchaqs” being named side by side in Leonṭi Mroveli’s account of
Alexander, and it is highly probable that the two passages are linked to each
other, given that King David is explicitly compared with the Greek emperor right
before:
KC D.A. 335,16–336,1:
da msgavsad aleksandressa kmna ... amistwsca aman meoreman aleksandre ganiz-
raxa sivrcita gonebisata, rametu sxuaebr ara iq̇ o ġone, dauc̣ q̇ oda ḳetilad q̇ ivčaq̇ ta
natesavisa simravle ...
“And he acted similarly to Alexander ... and therefore this second Alexander con-
sidered with the width of (his) wit that there was no other means, (for) he knew
the size of the tribe of the Qypchaqs well ...”
In this way, even the later text tradition supports the assumption that the
term bun-turk- of the “Conversion of Kartli” and its adaptations emerged from a
contamination of the ethnonyms of “Huns” and “Turks” appearing in a legend on

25
The chapter is entitled Cxovrebay mepet-mepisay davitisi; for a special edition cf. Šaniʒe 1992:
182-3.
 The “Bun-Turks” in Ancient Georgia 37

Alexander that must have been its source. This assumption implies the mis-
reading or misspelling of the former term in an Asomtavruli manuscript source
of the “Conversion” and its spread into all later text variants (as secondary attes-
tations); all this may well have been facilitated by popular etymology associating
bun- with notions of “nature”, “ground”, or “origin(ality)”. At the same time, the
term hon- in the Šaṭberd-version of Mokcevay Kartlisay may be regarded as being
re-introduced into the legend on the basis of a parallel source, possibly as an (in-
terlinear) gloss. As to the coexistence of hon- and *hun-, we should keep in mind
that the latter stem was partly homonymous with that of hune- “horse”, which
might have led to confusion; cf. the text on the destruction of Jerusalem in 614
A.D. ascribed to a certain Antiokhos Strategos, where the form honebi appears
instead of huneebi “horses” in an allusion to the submersion of the Pharao’s
troops in the exodus of the Israelites’ (Ex. 14,18–28):
Ant.Strat. Exp.Ier. V,1826:
da merme, odes ǯer učnda ġmertsa damqobay mati, eṭlebi igi da honebi mati dai-
qsna da sṗarazenebay aġč̣ urvilta matta daintka.
“And then, when it seemed appropriate to God to destroy them, their chariots
and horses were dissolved and the equipment of their armed (forces) was swal-
lowed.”
It depends on the reliability of the alleged sources then, i.e, the Alexander
Romance and its derivates, whether the existence of “Hunnic Turks” in Southern
Caucasia can be assumed for the time of the Macedonian emperor. As a matter
of fact, it is anything but certain that we have reliable historical information
here. Instead it is highly probable that the items concerning the “Huns” were in-
tegrated into the Alexander tradition not earlier than the year 515 A.D., possibly
even about a hundred years later, in 628 A.D., when there were actual “Hunnic”
or, rather, Khazar attacks in the Caucasus. At least for the Syriac “Christian Leg-
end” there are clear indications of its having been compiled by that period.27 The
connection of “Huns” with Alexander’s conquest thus remains a mere anachro-
nism, and it is by far not the only anachronism we find in the “Conversion of
Kartli”28 or Leonṭi Mroveli’s chronicle. And indeed, there is at least one more
coincidence that must be dealt with in this context.

26
Thus according to both editions: Garitte 1960: 13, 24-26; Marr 1909: 11, 3-5.
27
Cf. Hunnius (1904: 31) in dispute with Nöldeke (1890); Ciancaglini (2001: 138) accepts the lat-
ter date.
28
Cf. Gippert (2006: 108-114) for several anachronisms in the legend.
38 Jost Gippert

Different from the “Conversion”, Leonṭi begins his account of the history of
Georgia not with Alexander but in much earlier times. After speculations about
the descendance of the Caucasian peoples from Targamos, i.e. the Biblical patri-
arch Togarmah, a grandson of Japhet (Gen. 10,2–3; I Chr. 1,5–6), he deals in exten-
so about prehellenic times, and it is within this context that he first introduces
the “Turks”:
KC. L.Mr. 14,13-14:
da šemdgomad amissa raodentame c̣ elic̣ adta ucalo ikmna keḳaṗos, mepe sṗarsta,
rametu ic̣ q̇ o brʒola turkta.
“And several years after that Keḳaṗos, the Persian king, became busy, for he be-
gan to struggle against the ‘Turks’.”
In the passage in question Leonṭi is declaredly referring to a source he used,
viz. a text styled “The Life of Persia” which must be some prototype of Firdausī’s
Šāhnāme, and the “Turks” mentioned must be the “Turanians” of the Iranian tra-
dition (Culaja 1979: 58 n. 79):
KC. L.Mr. 14,21-23:
šemdgomad amissa mciredta c̣ elta ḳualad gamogzavna amanve keḳaṗos ʒis-c̣ uli
misi, ʒe šioš bednierisa, romeli moiḳla turkets, vitarca c̣ eril ars c̣ ignsa sṗarsta
cxovrebisasa.
“A few years after that, the same Keḳaṗos sent away his grandson, the son of Šioš
the Lucky, who was killed in the Turks’ country, as it is written in the book of the
Life of the Persians.”
It is obvious that the persons named here are the Iranian heroes Kai Kawūs,
his son Siyāwuš/Siyāwaxš (Abuladze 1916: 3 n. 2), and his grandson Firōd,29 all fig-
uring in Firdausi’s Šāhnāme30 as Iranian kings who were involved in struggles
with the Turanians under Afrāsyāb.31 The reason why this episode is quoted in
Leonti Mroveli’s chronicle is that it contains the report about another son of

29
For the death of Firōd cf. Šn. 13, 843 [830/426] ff. (references to verses of the Šāhnāme are
here given in accordance with the system used in Wolff 1935; corresponding verse numbers of the
editions Bertel’s 1960-1971 and Khaleghi-Motlagh 1988-2009 are added in square brackets). – The
epithet bednieri “lucky” should refer to Firōd rather than to his father Siāvuš as he is named farrux
Firōd e.g. in Šn. 13, 913 [892/486].
30
In its chapter on Alexander, the Šāhnāme does contain the episode on Yāǧūǧ and Māǧūǧ =
Gog and Magog (20, 1450 [1421] ff.), but in a much divergent form and without mentioning the
name of the “Huns”.
31
Cp. str. 791-2 of the Middle Georgian metrical adaptation of the Šāhnāme (ed. Abuladze 1916:
210) where kekaoz = keḳaṗos = Kai Kawūs and aprasiob = Afrāsyāb are mentioned side by side in
connection with the birth of rosṭom = Rostam.
 The “Bun-Turks” in Ancient Georgia 39

Keḳaṗos/Kai Kawūs, viz. Paraboroṭ = Farīburz,32 who was sent out by his father
into a struggle against the Caucasian peoples, Armenians, Georgians, and “all the
descendants of Targamos”:
KC. L.Mr. 14,14-20:
šemdgomad amissa raodentame c̣ elic̣ adta gamogzavna keḳaṗos, sṗarsta mepe-
man, ʒe misi, romelsa erkua paraboroṭ, sṗita didita somexta da kartvelta da q̇ ovelta
targamosianta zeda. xolo šeḳrbes ese q̇ ovelni targamosianni, miegebnes da daec̣ ev-
nes adarbadagans, da ioṭes paraboroṭ, da mosres sṗa misi.
“Some years after that Keḳaṗos, the king of the Persians, sent his son, who was
called Paraboroṭ, with a big army against the Armenians and Georgians and all
the descendants of Targamos. But all these descendants of Targamos gathered,
moved off to Azerbaijan and ravaged it, and they drove Paraboroṭ away and de-
feated his army.”
This episode may well refer to the defeat of the Iranians under Farīburz by
the Turanians reported in the Šāhnāme (13, 1343 [1314/905] ff.). In a similar way,
the Georgian text alludes to another grandson of Kai Kawūs struggling against
the “Turks” = Turanians, viz. Kaixosro = Kai Xosrow; here, the “Turks” are even
reported to have entered Mcxeta, the capital of Georgia:
KC. L.Mr. 15,5-17:
da šemdgomad amissa raodentame c̣ elic̣ adta ucalo ikmna kaixosro mepe, da ic̣ q̇ o
brʒolad turkta, eʒiebda sisxlsa mamisa matisasa. da ṗoves žami somexta da
kartvelta, ganudges sṗarsta da mosrnes eristavni sṗarstani, da gantavisupldes. xolo
masve žamsa movides turkni, oṭebulni misve kaixosrosgan, gamovles zġua gurgan-
isi, aġmoq̇ ves mṭḳuarsa da movides mcxetas saxli ocdarva. ezraxnes mamasaxlissa
mcxetisasa, aġutkues šec̣ evna sṗarsta zeda. xolo mamasaxlisman mcxetelman
auc̣ q̇ a q̇ ovelta kartvelta. inebes damegobreba mat turkta, rametu akunda šiši
sṗarsta, da šemc̣ eobisatws daimegobrnes turkni igi gamosxmulni, da ganiq̇ vanes
q̇ ovelta kalakta šina. xolo umravlesni matganni movides da ṗoves adgili erti mcxe-
tas, dasavlit ḳerʒo ḳldeta šoris gamoḳuetili, ġrma, da moitxoves adgili igi mcxetelta
mamasaxlisisagan. misca da aġašenes igi, mozġudes mṭḳiced, da ec̣ oda mas
adgilsa sarḳine.
“And some years after that, Kaixosro the king became busy, and he began to
struggle against the Turks, seeking (revenge for) the blood of their (!) father. And
the Armenians and Georgians grasped the opportunity (lit. found the time), re-
belled against the Persians and defeated the generals of the Persians and freed
themselves. But at the same time came the Turks (who had been) defeated by the

32
The Georgian form of the name is likely to have been influenced by another name frequent
in the Šāhnāme, Farāmarz, as in the Persian manuscript tradition itself.
40 Jost Gippert

same Kaixosro, crossed the Sea of Gurgan (i.e., the Caspian Sea), went up the Kur
and came to Mcxeta, 28 families (lit. houses). They negotiated with the mayor (lit.
housefather) of Mcxeta (and) promised to help him against the Persians. And the
mayor of Mcxeta informed all Georgians (telling them that) they should want to
become friends with those Turks, because he was afraid of the Persians, and be-
cause of the help (they had offered) they made friends with those Turkic refugees,
and they distributed them over all the cities. But most of them came and found
one place in Mcxeta, in its western side, deeply enclosed in the rocks, and they
asked the mayor of Mcxeta for that place. He gave it to them and they built it up,
encircled it with a strong wall, and that place was called Sarḳine.”

This report – though not identifiable as such within the Šāhnāme33 – strongly
reminds us not only of the settlement of “Chaldean Huns” thematised in Mokce-
vay Kartlisay, but also, in mentioning Mcxeta and Sarḳine, of the “Bun-Turks”
and their dwelling places along the Kur river.34 The identification of “Huns” and
“Turks” as presupposed by the latter designation may thus reflect two projec-
tions of the same historical event, the Khazar attacks of the VIth–VIIth centuries,
into two different periods of prehistory. This view is corroborated by the fact
that in Kartlis Cxovreba, the chapter in question is entitled gamoslva xazarta, i.e.,
the “Coming of the Khazars”. And indeed, Leonṭi Mroveli’s chronicle deals with
Khazars and their attacks in both its “pre-Alexandrian” and “Alexandrian” parts:
KC. L.Mr. 11,1 ff.: “The coming of the Khazars”
mas žamsa šina ganʒlierdes xazarni da dauc̣ q̇ es brʒolad natesavta leḳisata da
ḳavḳasiosta ... amissa šemdgomad xazarta ičines mepe ... da gamovles zġws-ḳari,
romelsa ac̣ hkwan darubandi. ... da šemusrnes q̇ ovelni kalakni araraṭisani da ma-
sisisani da črdilosani da daurčes cixe-kalakni tuxarisi, samšwlde da mṭueris-cixe,
romel ars xunani, šida-kartli da egrisi. da isc̣ aves xazarta orni-ve ese gzani, romel
ars zġws-ḳari darubandi da aragws-ḳari, romel ars dariala. ...
“At that time, the Khazars gained strength and began to fight against the de-
scendants of Leḳ and Ḳavḳasios ... after that the Khazars chose a king ... and
passed through the sea-gate which is now called Daruband ... and they destroyed
all cities of the Ararat and Masis and of the North, and (only) the fortified cities of
Tuxarisi, Samšwlde and the fortress of Mṭueri, which is Xunani, Inner-Kartli and

33
Most probably, the episode in question is adapted from the story of the defeat of Afrāsyāb by
Kai Xosrow (Šn. 13g); the crossing of the “sea of Gurgan” may reflect the crossing of the river Ǧey-
ḥūn by Afrāsyāb’s troops (13g, 345 [330/336] ff.).
34
Note that it was M. F. Brosset (1849: 33 n. 3) who first equated the “Bounthourki” with the
“Touraniens”.
 The “Bun-Turks” in Ancient Georgia 41

Egrisi persisted. And the Khazars usurped both these ways, viz. the sea-gate (of)
Daruband and the Aragvi-gate, which is Dariala(n). ...”
KC. L.Mr. 19,2–10:
da c̣ arvida aleksandre. xolo aman azon moarġwvna zġudeni kalaksa mcxetas
sapuʒvliturt ... da daiṗq̇ ra kartlsa zeda egrisi-ca, da moxarḳe q̇ vna osni, leḳni da xa-
zarni.
“And Alexander went off. But that Azon destroyed the walls in the city of Mcxeta
with its foundation ... and after Kartli, he took over Egrisi, and he laid Ossetes,
Lezgians and Khazars under tribute.”
It is another autochthonous hagiographical text, the VIIIth c. legend of St.
Habo of Tbilisi, which provides final evidence of the Khazar attacks of the first
millennium A.D. being the background of the Georgian “Bun-Turks”, as it uses
the same epithets for the former as those assigned to the latter in the Alexander
Romance:
Habo 2: 58,1-11:
da iq̇ o dġeta mat šina ḳualad ganrisxebay qelmc̣ ipeta mat sarḳinoztay nersē eri-
stvisa zeda da ivlṭoda igi ... da ganvlo man ḳari igi ovsetisay, romelsa darialan erku-
mis. da mat tana-ve iq̇ o sanaṭreli-ca ese monay krisṭēsi habo. xolo nerse ... ševida
kueq̇ anasa mas črdiloysasa, sada igi ars sadguri da sabanaḳē ʒeta magogistay, ro-
mel arian xazarni, ḳac velur, sašinel ṗirita, mqecis buneba, sisxlis mč̣ amel, romelta
šǯuli ara akus, garna ġmerti xolo šemokmedi ician.
“And in those days, the rulers of the Saracens became angry again about Nersē,
the leader (of Kartli), and he fled ... and transgressed the gate of Ossetia, which is
called Darialan. And the blessed servant of Christ, Habo, was with him. But Nerse
... entered the land of the north, which (lit. where) is the abode and dwelling
place of the sons of Magog, who are the Khazars: wild men, terribly looking,
(with) the nature of beasts, blood eaters, who have no faith except for knowing a
God-Creator ...”
We may conclude that the information on pre-Christian times provided in
the “Conversion of Georgia” and in Leonṭi Mroveli’s chronicle has no historical
value as such. Nonetheless, with the equation of Huns, Turks-Turanians, Khazars
and, lastly, Qypchaqs, the medieval Georgian sources do give us remarkable in-
sights into the late first millennium perception of ethnic strata of Turkic stock in
and around Caucasia.
42 Jost Gippert

BIBLIOGRAPHY
Abuladze, I. (ed.) (1916), Šah-names anu mepeta c̣ ignis kartuli versiebi, Tbilisi.
―― (1944), Kartuli da somxuri liṭeraṭuruli urtiertoba IX–X ss-ši, Tbilisi.
―― (ed.) (1953), Kartlis cxovrebis ʒveli somxuri targmani, Tbilisi.
―― (ed.) (1963), Ʒveli kartuli agiograpiuli liṭeraṭuris ʒeglebi, c̣ . I, dasabeč̣ dad moamzades..., Tbilisi.
―― (ed.) (1971), Ʒveli kartuli agiograpiuli liṭeraṭuris ʒeglebi, c̣ . III, dasabeč̣ dad moamzades..., Tbilisi.
Aleksidzé, Z. (ed.) (2001), Le nouveau manuscrit géorgien sinaïtique N Sin 50, Édition en facsimilé,
Introduction par Z. A., traduite du géorgien par J.-P. Mahé, Lovanii.
Androniḳašvili, M. (1966), Narḳvevebi iranul-kartul enobrivi urtiertobidan I, Tbilisi.
Bertel’s, E. (ed.) (1960–1971), Firdousī, Šāx-nāme. Kritičeskij tekst. T. 1-9, Moskva.
Bielmeier, R. (1990), “Sprachkontakt in der Bekehrung Kartlis”, eds. R. Schulz; M. Görg, Lingua
Restituta Orientalis. Festgabe für Julius Assfalg, Wiesbaden: 30–44.
Blake, R.-P. (1923), Note supplémentaire sur Fou-Lin, Journal Asiatique 202: 83–88.
Brosset, M.-F. (1849), Histoire de la Géorgie depuis l'antiquité jusqu'au XIXe siècle, St. Pétersbourg.
Budge, E. W. (ed.) (1889), The History of Alexander The Great, Cambridge.
Ciancaglini, C. A. (2001), “The Syriac Version of the Alexander Romance”, Le Muséon 114: 121–140.
Čubinašvili, D. (1887), Kartul-rusuli leksiḳoni, S.-Peterburg.
Culaja, G. V. (ed.) (1979), Mroveli Leonti: Žisn’ kartlijskix carej, Moskva.
Čikobava, A, (ed.) (1955), Kartuli enis ganmarṭebiti leksiḳoni, ṭ. IV, Tbilisi.
Garitte, G. (ed.) (1960), La prise de Jérusalem par les Perses en 614, [Edition:] Corpus Scriptorum
Christianorum Orientalium, 202/Scriptores Iberici, 11, Louvain.
Gigineišvili, B. / E. Giunašvili (1979), Šaṭberdis ḳrebuli X sauḳunisa, gamosacemad (moamzades...,
Tbilisi.
Gippert, J. (1991), “Daemonica Irano-Caucasica”, ed. P. Vavroušek, Iranian and Indo-European
Studies. Memorial Volume of Otakar Klíma. Praha 1994: 53–88.
―― (1993), Iranica Armeno-Iberica. Vienna.
―― (2006), “C̣ m. Ninos legenda: gansxvavebul c̣ q̇ arota ḳvali”, Enatmecnierebis saḳitxebi 1-2: 104-
122.
―― (2010), Hexaplaric material in the Albano-Armenian palimpsests from Mt. Sinai. To appear in a
festschrift, Tbilisi.
Hübschmann, H. (1897), Armenische Grammatik. I. Theil: Armenische Etymologie, Leipzig.
Hunnius, C. (1904), Das syrische Alexanderlied, Göttingen.
Kaleghi-Motlagh, Dj. (1988–2009), Abu’l Qasem Ferdowsi, The Shahnameh, Vol. 1–8. New York.
Lerner, K. (2004a), “On the Origin of the “Chelishi” Manuscript of the Conversion of Kartli”, Le
Muséon 117: 131–136.
―― (2004b), The Wellspring of Georgian Historiography, The Early Medieval Historical Chronicle
The Conversion of Kartli and The Life of St. Nino, translated with introduction, commen-
tary and indices by Constantine B. Lerner, London.
Marr, N. (ed.) (1901), Ippolit, Tolkovanie pěsni pěsnej, Gruzinskij tekst..., S.-Peterburg.
―― (ed.) (1909), Antiox Stratig, Plěnenie Ierusalima Persami v 614 g., S.-Peterburg.
 The “Bun-Turks” in Ancient Georgia 43

Nöldeke, Th. (1890), Beiträge zur Geschichte des Alexanderromans, Denkschrift, Vienna.
Orbeliani, S.-S. (1965), Leksiḳoni kartuli. = Txzulebani, ṭ. IV/1, Tbilisi.
Pätsch, G. (1975), “Die Bekehrung Georgiens Mokcevay Kartlisay (Verfasser unbekannt)”, Bedi
Kartlisa 33: 288–337.
Peeters, P. (1926), “La passion géorgienne des Ss. Théodore, Julien, Eubulus, Malcamon, Mocimus
et Salamanes”, Analecta Bollandiana 44: 70–101.
Q̣ auxčišvili, S. (ed.) (1955–1959), Kartlis cxovreba. Ṭeksṭi dadgenili..., Tbilisi.
Rapp, St. H. (1999), “Pre-Christian History in the Georgian Shatberdi Codex: A Translation of the
Initial Texts of Mokʿcʿevay Kʿartʿlisay (“The Conversion of Kʿartʿli“)”, Le Muséon 112: 79–128.
―― (2003), Studies in Medieval Georgian Historiography: Early Texts and Eurasian Contexts, (Cor-
pus Scriptorum Christianorum Orientalium, 601/Subsidia, 113), Lovanii.
―― (2006), “The Conversion of Kʿartʿli: The Shatberdi Variant”, Kek. Inst. S–1141, Le Muséon 119:
169–225.
Šanidze, M. (ed.) (1992), Cxovrebay mepet-mepisa davitisi, Tbilisi.
Takajšvili, E. S. (1900), “Istočniki gruzinskix lětopisej. Tri xroniki. 1. Obraščenie Gruzii (v xristian-
stvo)”, Sbornik materialov dlja opisanija městnostej i plemen Kavkaza 28: 1–216.
Tēr-Mkrtčean, G. / St. Kanayeancʿ (eds.) (1909), Agatʿangełay Patmowtʿiwn Hayocʿ, Tpʿłis.
Thiel, H. van (ed.) (1959), Die Rezension λ des Pseudo-Kallisthenes, Bonn.
Wolff, F. (1935), Glossar zu Firdosis Schahname, Berlin.
 On the Relative Value of Armenian Sources for the
Khazar Studies: The Case of the Siege of Tbilisi

Dan Shapira
Bar-Ilan University, Ramat Gan

Abstract
In this paper, the author examines two contemporary Byzantine-Greek sources on one episode of
the last Byzantine-Sasanian war (602-628), the siege of Tbilisi in 627 (?) CE by the combined forces
of Heraclius, the Byzantine Emperor, and those of a West-Turkic (called “Khazar”) leader. Two of
the sources are Byzantine-Greek, one by Theophanes the Confessor who wrote between 811-3 and
based himself, for the siege of Tbilisi, on a source not later than ca. 720 CE, and one by Theopha-
nes’ contemporary Nicephoros the Patriarch, whose source was different; another source is an Ar-
menian chronicle from Caucasian Albania attributed to Movsēs Dasxuranc‘i/Kałankatuac‘i, who
wrote between the first years of the eighth century (very close to the time of Theophanes’ source)
and 958 CE and whose version of the events is the fullest, but badly arranged by a later editor; then
come two Georgian versions, that of the “Life of Kartli” and that of the older “Conversion of Kartli”.
The conclusion of the author is that the Albano-Armenian author worked, while describing the
siege of Tbilisi and the role of the Western Turks, from more than one source.

Keywords
Heraclius, Khazars, Western Turkic Qaganate, Tbilisi, Byzantine-Sasanian War of 602-628, Cauca-
sian Albania, Theophanes the Confessor, Nicephoros the Patriarch, Movsēs Dasxuranc‘i, Georgian
Chronicles “Life of Kartli” and “Conversion of Kartli”

This is a great honour to take part in this book celebrating our esteemed friend
Garnik Asatrian. The following contains a heavily reworked and updated version
of some parts of a previous study of mine.1 The present essay is historiographical
and focuses on several sources describing one historical event. The case of de-
scriptions of the common Byzantine-Turkic/“Khazar” siege on the Iranian-held
Tbilisi at the final stages of the last Byzantine-Sasanian war (602-628) is a good

1
D. Shapira, “Armenian and Georgian Sources on the Khazars: A Re-Evaluation”, The World of
the Khazars: New Perspectives. Selected Papers from the Jerusalem 1999 International Khazar Collo-
quium, ed. H. Ben-Shammai, P. B. Golden, A. Roná-Tas, Leiden 2007, pp. 307-352.
 Koninklijke Brill NV, Leiden, 2015
46 Dan Shapira

example of using Armenian and other Transcaucasian written texts as historical

sources for reconstruction historical events.
Our best source for this siege is the Byzantine- Greek of Theophanes the Con-
fessor (752/758-818), a monastic and iconophilic chronicler, who continued the
work of George Syncellus (d. 810/811). Syncellus’ Chronicle begins from Adam
and goes up to 284 CE, whilst that of Theophanes describes the events between
284-813 CE.
Theophanes was not an eyewitness of the siege, of course (he was born more
than a century after Heraclius’ triumph), and his source on the Khazars and the
siege of Tbilisi was composed ca. 720 at the latest. This source might have been
some form of an official Byzantine report. However, the work of Theophanes is
wanting in chronological accuracy; on page 316, he compressed all the following
events in one year, 6117 AM / Heraclius’ 16th year / Chosroes’ 38th year [/ 625/6
CE].2
He reports the events that led to the siege on Tbilisi as follows: the Persians
sought to orchestrate an attack on Constantinople by the Western Huns, called
Avars, and by the Bulgars, Slavs and Gepids. In respond, Heraclius invited, while
in Lazica, “the Turks from the East called Khazars”3 to become his allies.

2
Chronografía. Theophanis Chronographia, ed. C. de Boor, 2 vols., Lipsiae 1883 (reprint: Hildes-
heim 1963), pp. 308ff., 315 l. 1-15; p. 315 l. 20 -p. 316 l. 16 (the events of 624/5, including the siege on
Tbilisi); pp. 328.13-329.1 (the events of 627/8, including, Heraclius in Palestine and persecution of
the Jews); translations: The Chronicle of Theophanes. An English Translation of anni mundi 6095-
6305 (A.D. 602-813), with introduction and notes, by Harry Turtledove, Philadelphia, 1982; The Chroni-
cle of Theophanes Confessor. Byzantine and Near Eastern History AD 284-813, Translated with Intro-
duction and Commentary by Cyril Mango and Roger Scott with the assistance of Geoffrey Greatrex,
Oxford 1997. On Heraclius see now Walter Emil Kaegi, Heraclius: Emperor of Byzantium, Cambridge
2003. On chronological problems in Heraclius’ Caucasian campaigns, see E. Geland, “Die persische
Feldzüge des Kaisers Heracleios”, Byzantinische Zeitschrift, Vol. 3 (1894), pp. 330-373; V. V. Bolotov,
“K istorii imperatora Iraklija”, Vizantijskij Vremennik, Vol. XIV (1909), pp. 68-124; Ja. A. Manandjan,
“Maršruty persidskix poxodov imp. Iraklija”, Vizantijskij Vremennik (NS), Vol. III (1950), pp. 133-153;
A. N. Stratos, Byzantium in the Seveneth Century, I: 602-634, Amsterdam 1968, p. 150ff; J. Howard-
Johnston, “The Official History of Heraclius’ Persian Campaigns”, The Roman and Byzantine Army
in the East, ed. E. Dabrowa, Cracow 1994, pp. 57-87; idem., “Heraclius’ Persian Campaigns and the
Revival of the East Roman Empire”, War in History 6 (1999), pp. 1-44; C. Zuckerman, “Heraclius in
625”, Revue des Études Byzantines 60 (2002), pp. 189-197.
3
It was observed that Kartlis Cxovreba [=KC]/the Georgian Chronicle, p. 223, identifies them in
the same context as “Turks from the West”, see P. B. Golden, Khazar Studies. An Historico-Philologi-
cal Inquiry into the Origins of the Khazars, Vol. I, Budapest 1980, Vol. I, pp. 58, 188. For KC, see
Kartlis Cxovreba, I-II, ed. S. Qauxčišvili, Tbilisi 1955; quoted (vol. I) as KC, p. ... . See also F. Brosset,
 On the Relative Value of Armenian Sources for the Khazar Studies 47

Theophanes calls them indiscriminately either “Khazars”, or “Turks”;4 they broke

through the Caspian Gates and, under their stratēgos Ziebēl,5 who was second to
their Xagan, they invaded Adharbagan (Adragan). Heraclius set out from Lazica
and joined them. Both leaders met under the walls of Tiphilios (Tbilisi); Ziebēl,
seeing the Emperor, rushed to him, kissed his neck (or, stretched his neck, ka-
taspázetai autou tòn tráchēlon), and prostrated himself, while the Persians
looked on from the town of Tiphilios.
It is clear from both Theophanes’ own words and from the context that Zie-
bēl was thus not the Khazar Qagan, but second in rank, surely not on the same
royal footing with Heraclius. The entire army of the Turks fell flat with their faces
downwards and stretched out on their faces, reverenced the emperor with honor
unknown among alien nations (but, in fact, similar descriptions of Turkic rever-
ence offered to foreign rulers are found elsewhere in Chinese sources), etc.; Zie-
bēl, seeking perhaps the royal charisma of Heraclius, presented to the Emperor
his son (called in an Armenian text—see further—šat‘ easily identified with the
šad known from other sources6); Ziebēl gave the Emperor 40,000 brave men and
returned to his country. Heraclius, with the Khazars, proceeded to Persia, etc.
Theophanes (Bonn edition, p. 316; translation by Mango & Scott, pp. 446-447,
with minor changes), actually, said:
… As for Sarbaros [*Sharhwarāz], he dispatched him with his remaining army
against Constantinople with a view to establishing an alliance between the west-
ern Huns (who are called Avars) and the Bulgars, Slavs, and Gepids, and so ad-
vancing on the City and laying siege to it. When the emperor learnt of this, he di-
vided his army into three contingents: the first he sent to protect the City, the se-
cond he entrusted to his own brother Theodore, whom he ordered to fight Sain
[*Shāhīn]; the third he took himself and advanced to Lazica. During his stay there

Histoire de la Géorgie depuis l’Antiquité jusqu’au XIXe siècle, trans., St Petersburg 1856; cf. also W. E.
D. Allen, History of the Georgian People, London 1932 (reprint: New York 1971); G. Pätsch, Das Leben
Kartli’s. Eine Chronik aus Georgien, 300-1200, Leipzig 1985; R. W. Thomson, Rewriting Caucasian His-
tory. The Medieval Armenian Adaptation of the Georgian Chronicles. The Original Georgian Texts
and the Armenian Adaptation. Translated with Introduction and Commentary, Oxford 1996.
4
Such an earlier author as [Pseudo-]Sebēos (ed. E. Patkanov, St. Peterburg 1879, p. 22) names
them T‘eatalac‘ik‘, “Hephtalites”. [Pseudo-]Sebēos, Part I (Sebēos: Patmut‘iwn Sebēosi Episkoposi i
Herakln, ed. T. Mihrdatian, Constantinopole 1851; Sebēos, Patmout‘iwn, ed. G. Abgaryan, Erevan
1979; translations: K. Patkanjan, Istorija imperatora Irakla. Sočinenije jepiskopa Sebeosa pisatelja VII
vēka, St. Peterburg 1862; F. Macler, Histoire d’Héraclius, Paris 1904.
5
On this title and possible identifications, see Golden, Khazar Studies, Vol. I, pp. 187ff.; 218-9.
6
See P. B. Golden, Khazar Studies, p. 206ff.
48 Dan Shapira

he invited the eastern Turks, who are called Chazars, to become his allies. … Now
the Chazars broke through the Caspian Gates and invaded Persia, that is the land
of *Abadraigan, under their commander Ziebēl who was second in rank after the
Chagan. And in all the lands they traversed they made the Persians captive and
burnt the towns and villages. The emperor, too, set out from Lazica and joined
them. When Ziebēl saw him, he rushed to meet him, kissed his neck (or bowed
with his face downwards), and did obeisance to him, while the Persians were
looking on from the town of Tiphlios. And the entire army of the Turks fell flat on
the ground and, stretched out on their faces, reverenced the emperor with an
honour that is unknown among other nations. Likewise, their commanders
climbed on rocks and fell flat in the same manner. Ziebēl also brought before the
emperor his adolescent son, and he took as much pleasure in the emperor’s con-
versation as he was astonished by his appearance and wisdom. After picking
40,000 brace men, Ziebēl gave them to the emperor as allies, while he himself re-
turned to his own land. Taking these men along, the emperor advanced on
Chosroes.
There is a description of the siege of Tbilisi here. However, another Byzantine
source does not mention the siege at all and portraits the Turkic leader as a ruler
of the same standing as Heraclius: this is chronicle of Nicephoros (758-828/9),
the iconodule Patriarch of Constantinople (806-815) appointed by his namesake,
Emperor Nicephoros (802-811), but dismissed by the iconoclastic Leon V the Ar-
menian (813-820) in 815. Nicephoros belonged to the same generation as The-
ophanes. Nicephoros used the same source as Theophanes for the years 668-769,
but a different source for 602-641,7 and for the period 609-799 CE, his chronicle is
very poor.
In 622 CE, being pressed by the Avars and Persians and with the Empire be-
ing tortured by the heavy famine and plague, Heraclius decided to attack the
Persians via the Black Sea and Lazica; while in Lazica, he contacted the Lord of
the Turks (tôn tourkôn kurios) and invited him into an alliance against the Per-
sians; the Lord of the Turks accepted and promised to act as an ally; having
heard thus, Heraclius proceeded to the Lord of the Turks in person; the Lord of
the Turks moved unto Heraclius with a host of his people, descended his horse
and fell to the earth before the Emperor, and so did all the host of the Turks. Re-
joicing, Heraclius called upon the Turkic leader to draw closer and called him
his, Heraclius’, son. He embraced the Lord of the Turks, took down his crown

7
Nikephoros Patriarch of Constantinople. Short History, Text, Translation and Commentary by
Cyril Mango, Dumbarton Oaks, Washington D.C. 1990, pp. 12-13.
 On the Relative Value of Armenian Sources for the Khazar Studies 49

and put it on the head of the Turk, inviting him to a banquette. While on the
banquette, the Emperor gave the Turk all the utensils, the royal wardrobes and
earrings of pearls, and likewise he adorned with earrings other Turkic archons.
To make the alliance stronger, the Emperor showed the Turk a portrait of his
daughter, Eudocia, and said: “God made us unite when He called thee my son.
This is my daughter, the Augusta of the Romans (Augusta Rômaiôn). If thee as-
sisstest me and givest me auxiliary troops against my enemies, I shall give thee
her as a wife”. The barbarian, of course, fell in love with the portrait of the prin-
cess on the spot and gave the Emperor an archon with a host of Turks.8 Heraclius
invaded Persia with them and began to destroy her cities and fire-temples (ta
pureia).9 The text says:
Once again Chosroes, king of Persia, made war on the Romans, having placed his
army under the command of Sarbaros, who devastated all the eastern lands. This

8
Nothing came out of the proposed marriage, as the Turkic ruler was slain in 629. On this epi-
sode, cf. C. Zuckerman, “La petite augusta et le Turc. Epiphania-Eudocie sur les monnaies
d’Héraclius”, Revue Numismatique 150 (1995), pp. 113-126; compare the critical reviews published in
Revue Numismatique 152 (1997), pp. 453-472, and Zuckerman’s answer on pp. 473-478. Later, Ni-
cephoros mentioned that Heraclius suggested marrying the same Eudocia to the Muslim-Arab
general Ambros (*‘Amr[u]), if only the Muslim would be baptized. A Georgian source of a much
later date (the extant edition is from the 12th century), the Vita of David and Constantine
(C’amebai da ɣuac’li c’midata da didebulta moc’ameta Davit da K’ost’ant’inesi (The Martyrdom and
Heroism of the Holy and Glorious Martyrs, David and Constantine), in Čveni Saunǰe I, ed. K. Keke-
lidze (K’. K’ek’elije), Tbilisi 1960, pp. 435-46), tells the tragic story of two Georgian princes killed in
741 by Marwān ibn Muḥammad, while referring to the same earlier events: “but the servant of God,
the Greek king Heraclius ... was commanded by God to go to the land of K’omans who are
Q’ivčaq’s, and he gave his daughter to the king of the Q’ivčaq’s as a wife, and then took him with
his entire army to strengthen him” (ševida kue q’anasa K’omantasa romel arian Q’ivčaq’ni, da misca
asuli twisi colad mepesa Q’ivčaq’tasa). This text indicates that Khazars were no longer known to—
or relevant for—the 12th century redactor.
This is the only source (edited in the extant form more that half a millennium after the events
described!) to make mention of the proposed marriage of the Emperor’s daughter to the Tur-
kic/Khazar ruler, cf. M. van Esbroeck, “Une chronique de Maurice a Héraclius dans une récit des
sièges de Constantinople”, Bedi Kartlisa 34 (1976), pp. 74-96, pp. 93 at the bottom; see also M. Bíró,
“Georgian Sources on the Caucasian Campaign of Heracleios”, Acta Orientalia Academiae Scien-
tiarum Hungariae, XXXV (1) 1981, pp. 121-132, p. 129.
9
See Nicephorus Patriarcha, Historía síntomos. Breviarium rerum post Mauricium gestarum, ed.
I. Bekker, Bonae, 1837, 15.20.16.20, p. 78; Nikephoros Patriarch of Constantinople. Short History, Text,
Translation and Commentary by Cyril Mango, Dumbarton Oaks, Washington D.C. 1990, pp. 56-57,
66-67; cmp. K. Czeglédy, “Herakleios török szövetségesei”, Magyar Nyelv XLIX (1953), 319-323, p.
323.
50 Dan Shapira

man seized in the Holy Places the life-giving relic of the Cross of redemption,
Modestos being at that time bishop of Jersualem. Finding himself troubled by
both the Persians and the Avars, and the Roman State hard pressed by famine
and decimated by plague, Herakleios called in Sergius, the bishop of the City,
along with noblemen and the rest of the people, and placed his children in their
care. He entrusted the administration of affairs to the patrician Bonos and, set-
ting out by way of the Black Sea, he attempted to invade Persia through Lazica. At
this juncture a son was born to him by his wife Martina (for he had taken her
along), and he named him Herakleios. From there <Lazica> he sent gifts to the
chieftain of the Turks, whom he urged to enter on an alliance against the Per-
sians. The latter accepted <the gifts> and promised to be an ally. Herakleios was
pleased at this and set forth in his direction; and he, on being informed of the
emperor’s presence, met him with a great multitude of Turks and, dismounting
from his horse, prostrated himself on the ground before the emperor, while his
entire host did the same. On perceiving this exceedingly great honor, the em-
peror declared to him that if their friendship was steadfast, he could draw nigh
even on horseback; and he called him his own son. Taking off the crown from his
head, he placed it on the Turk’s and, after serving a banquet, presented to him all
the utensils of the table as well as the imperial garment and earrings adorned
with pearls. He likewise decorated with his own hand the noblemen of <the
Turk’s> suite with similar earrings. Fearing, however, lest he suffer the same fate
as with the Avar <chief>, and with a view to making the agreement more binding,
he showed him the portrait of his daughter Eudokia and said to him: God has
joined us and made you my son. Behold, this is my daughter, the Roman Augusta.
If you espouse my cause and help me against my enemies, I shall give her to you
in marriage”. <The Turk> was so struck by the beauty of the picture and its
adornment that he fell in love with the person represented and held fast to the
alliance all the more. Straightaway he delivered to the emperor a multitude of
Turks under a commander. Taking these along, <Heraclius> invaded Persia and
set about destroying cities and overturning fire temples. In one of the temples it
was discovered that Chosroes, making himself a god, had put up his own picture
on the ceiling, as if he were seated in heaven, and had fabricated stars, the sun,
the moon, and the angels standing around him, and a mechanism for producing
thunder and rain whenever he so wished.10

10
Nikephoros Patriarch of Constantinople. Short History, Text, Translation and Commentary by
Cyril Mango, Dumbarton Oaks, Washington D.C. 1990, pp. 54-57 (§12). The throne tāgdēs is de-
scribed here. Husraw II, “who made himself a god”, was called Aparwēz, “victorious” or “holy”. Her-
aclius, on the other hand, identified himself with Alexander the Great (cf. G. J. Reinink, “Die Ent-
stehung der syrischen Alexanderlegende als politisch-religiöse Propagandaschrift für Herakleios
 On the Relative Value of Armenian Sources for the Khazar Studies 51

Another source is the Armenian History of the Caucasian Albanians by

Movsēs Dasxuranc‘i (or Kałankatuac‘i, or Kałankaytuac‘i) of Uti.11 Caucasian
Albania (or Ałuank‘) corresponds roughly to the northern parts of the present
Republic of Azerbayjan, where a small Udi (or Udin, from *Uti) minority still
survives, belonging to the Armenian Church and speaking a “Daghestani” lan-
guage. Their Udi language is believed to be derived from one of the dialects of
the now extinct Caucasian Albanian/Ałuanian.12

Kirchenpolitik”, After Chaldecon. Studies in Theology and Church History. Offered to Professor Albert
van Roey For His Seventieth Birthday (Orientalia Lovaniensia Analecta, 18), Leuven 1985, pp. 263-
281, p. 267), adding thus another motivation for the destruction he caused in Iran.
11
Text: Movsēsi Kałankatuac‘woy Patmut‘iwn Ałuanic‘ ašxarhi, ed. M. Emin, Moscow 1860 [re-
printed Tiflis 1912]; Movsēsi Kałankatuac‘woy Patmut‘iwn Ałuanic‘ ašxarhi. K‘nnakan bnagirĕ ev
nēražurjunĕ Varaz Arak‘ēliani, Erevan 1983/Movses Kalankatuaci, Istorija strany Aluank. Kritičeskij
tekst i predislovije V. D. Arakeljana, Erevan 1983. Translations: Movses Kalankatuaci, Istorja strany
Aluank (trnsl. by Š. V. Smbatjan), Matenadaran, AN ArmSSR, Jerevan 1984; C. J. F. Dowsett, The His-
tory of the Caucasian Albanians by Movsēs Dasxuranc’i, London Oriental Series. Volume 8, London
1961.
12
In fact, all the extant Christian literature from Caucasian Albania known until recently is in
Armenian. There are remnants of the Albanian alphabet and some inscriptions in it, see I. V. Abu-
ladze, “K otkrytiju alfavita kavkazskix albancev”, Izvestija Instituta Jazyka, Istorii i Material'noj
Kul'tury Gruzinskogo Filiala AN SSSR IV 4 (Tbilisi 1938), pp. 69-71; A. G. Šanidze, “Novootkrytyj
alfavit kavkazskix albancev i jego značenie dlja nauki”, Izvestija Instituta Jazyka, Istorii i Materi-
al'noj Kul'tury im. N. Ja. Marra Gruzinskogo Filiala AN SSSR IV.1, Tbilisi 1938; H. Kurdian, “The Newly
Discovered Alphabet of the Caucasian Albanians”, Journal of the Royal Asiatic Society 1956, pp. 81-
83; A. G. Šanidze, “Jazyk i pis'mo kavkazskix albancev”, Sakartvelos SSR Mecnierebata Ak’ademiis
Sazogadoebriv Mecnierebata Ganq’opilebis Moambe I, Tbilisi 1960; R. H. Hewsen, “On the Alphabet
of the Caucasian Albanians”, Revue des études arméniennes NS, Vol. I (1964), pp. 427-32; this latter
article was published also in Russian, with a “Commentary” by G. A. Klimov, in Tajny drevnix
pis’mion. Problemy dešifrovki, “Progress”, Moscow 1976, pp. 444-452; two new Albanian palimpsest
MSs were found during the last decade among the Georgian MSs in the Mt. Sinai’s St Catherine
monastery library, dated to the 10th century, see Z. Aleksidzé & J. P. Mahé, “Decouverte d’une texte
albanien: une langue ancienne du Caucase retrouvée”, Comptes rendus de l’Académie des Inscrip-
tions et Belles-Lettre (1997), pp. 517-532; Z. Aleksidze, “Novyje pamjatniki pis'mennosti Kavkazskoj
Albanii”, Xristianskij Vostok NS 1 (7), St. Peterburg - Moscow 1999, pp. 3-13, and a short note in the
same volume by S. A. Starostin, “Fonetičeskij kommentarij k stat'je Z. N. Aleksidze”, pp. 13-14; Z.
Aleksidzé & J.-P. Mahé, Le nouveau manuscript géorgien sinaïtique N°50. N Sin 50, Series: Corpus
Scriptorum Christianorum Orientalium, 586; Subsidia, 108, Leuven : Peeters, 2001; Jost Gippert and
Wolfgang Schulze, “Some Remarks on the Caucasian Albanian Palimpsests”, Iran and the Caucasus
11 (2007), pp. 201-212; Jost Gippert, Wolfgang Schulze, Zaza Aleksidze and Jean-Pierre Mahé, eds.,
The Caucasian Albanian Palimpsests of Mount Sinai, Monumenta Palaeographica Medii Aevi. Series
Ibero-Caucasica 2, vols. I-II, Turnhout 2009; Jost Gippert, “Hexaplaric material in the Albano-Ar-
52 Dan Shapira

Movsēs’ work was composed between the first years of the eighth century
and 958 CE (this is to say, between the period in which Theophanes and Niceph-
oros wrote and between the very last years of the Khazar statehood) and finally
edited at the turn of the 11th-12th centuries.13 The chronology of this valuable Al-
bano-Armenian source is, however, blurred, and the text is arranged badly in the
present manuscripts, where the description of the siege of Tbilisi is the most
striking example of chronological confusions.14 Indeed, we have in our Albano-
Armenian source more than one different account combined mechanically.15
The order of events given in the Albano-Armenian composition is as follows:
Heraclius attacked the Persians in Transcaucasia and Atropatena, but they
struck back. Heraclius invited the Khazars to attack Caucasian Albania, they rav-
aged it and, having seen the abundance of booty captured, “the prince their lord”
(išxann tērn noc‘a) decided to return the next year. Indeed, it occurred in the
38th year of Husraw II (628 CE), “the year of his murder”, when J̌ ebow Xak‘an16 ar-
rived with his son and an immense army of Mongoloid-looking hordes of the
Northern barbarians (“of that ugly, insolent, broad-faced, eyelashless mob in the
shape of women with dishevelled hair”, žantatesil žprheres laynadēm anartewa-
noun bazmout‘eann i jew igakan gisarjaks dimeals).17 Having destroyed the walls
of Č‘ołay (in the vicinity of Derbend?) like a flood, he attacked Partaw and pil-

menian palimpsests from Mt. Sinai””, Caucasus between East and West. Historical and Philological
Studies in Honour of Zaza Aleksidze, ed. by Dali Čitunašvili, Tbilisi 2012, pp. 205-211.
13
Dowsett, The History of the Caucasian Albanians, pp. xx.
14
M. I. Artamonov, Očerki drevnejšej istorii xazar, Leningrad 1936, pp. 52ff.; Dowsett, The History
of the Caucasian Albanians, pp. xiv-xv; J. Howard-Johnston, “Heraclius’ Persian Campaigns and the
Revival of the East Roman Empire”, War in History 6 (1999), pp. 1-44, see pp. 12-13. A different atti-
tude is expressed in A. Akopjan [Hakobyan], Albania-Aluank v greko-latinskix i drevnearmjanskix
istočnikax, Jerevan 1987, pp. 188-196; cf. K. Cukerman [C. Zuckerman], “Xazary i Vizantija: pervyje
kontakty”, Materialy po Arxeologii, Istorii i Étnografii Tavrii, VIII (Simferopol’ 2001), pp. 312-333;
Constantin Zuckerman, “The Khazars and Byzantium – The First Encounters”, The World of the
Khazars: New Perspectives. Selected Papers from the Jerusalem 1999 International Khazar Collo-
quium, ed. H. Ben-Shammai, P. B. Golden, A. Roná-Tas, Leiden 2007, pp. 399-432, pp. 404-417, esp.
p. 405.
15
C. Zuckerman (cf. the previous note) discerns between Source A (chapters 9-11; the Eulogy of
J̌ uanšēr, written in 670), whose author does not pretend to be an eyewitness, and Source B (chap-
ters 12-16).
16
J. Marquart, Osteuropäische und ostasiatische Streifzüge, Lepzig 1903, p. 498, identified him as
T‘ong Che-hou.
17
Armenian text, p. 135 l. 22 - p. 136 l.2; the translation is slightly altered from that given in
Dowsett, The History of the Caucasian Albanians, p. 83.
 On the Relative Value of Armenian Sources for the Khazar Studies 53

laged Albania. Then they turned to Georgia and besieged “the luxurious, com-
mercial, famous, and great city of Tbilisi” (zp‘ap‘kasoun vačaṙašah hṙč‘akawor
mec‘ k‘ałak‘n Tp‘łis).18 It was at this stage that “the Great Emperor” (Heraclius)
joined his ally19—our Albano-Armenian source describes their meeting in one
short phrase (“exchanging royal gifts, they greatly rejoiced to see each other”).
Husraw sent an army under Šahrapłakan (Σαραβλαγγᾶς of the Greek sources) for
the defense of the besieged city, and when the townfolk saw the approaching
Persians, they began to mock the two kings. It is not clear from this account
whether the Persians were successful in entering the besieged city, which the
Byzantine engineers tried in the meanwhile to destroy by using ballistae and
other siege machines and attempting the River Kura to overflow into the city.
Exhausted, Heraclius tells the Khazars to lift the siege for the time being and
return the next year. Seeing that, the besieged citizens began to parody the de-
feat; they brought a huge pumpkin upon which they drew a caricaturesque im-
age of the “king of the Huns”, which stressed his Mongoloid features, and placed
this offensive image on the city wall; they also thrust a spear into the pumpkin,
calling out to the Northern armies: “Behold Caesar, your king, turn and worship
him, for he is J̌ ebow Xak‘an” (ahawasik Kaysr t‘agawors jer, owr kays, darjarouk‘
erkir pagēk‘ sma, J̌ ebow Xak‘an ē ays); they also called “the other king”, appar-
ently, Heraclius, “impure/filthy and pederast”;20 this wording is similar to that of
the corresponding slurs found in the Armenian translation of Kartlis Cxovreba
and in Mokcevay Kartlisay (see further). However, the siege was lifted.
The two kings withdrew; after that, according to Movsēs Dasxuranc‘i, “in the
36th year of Husraw” (626 CE),21 Heraclius sent Andrē, one of his nobles, to the
viceroy of the king of the North who was second to him in kingship and was
called Jebou Xak‘an,22 urging him to invade Persia via the gates of Č‘ołay (in the
vicinity of Derbend?) and promising them rich loot; the Northmen sent back to
the royal palace an embassy with an élite force of a thousand warriors, broke
through the passage of Č‘ołay ignored by the Persian garrison, and arrived at the

18
Armenian text, p. 137 l. 21-2; the translation is slightly altered from that given in Dowsett, The
History of the Caucasian Albanians, p. 85.
19
According to Byzantine sources, Heraclius’s allies were Lazes, Abasges, Iberes; al-Mas‘ūdī
mentions Alans, Khazars, Abkhazes, Sarir, Georgians, Armenians, etc.
20
piłc ew arouazełc, p. 140.
21
This is where Source B begins, see note 15 above.
22
yaǰord ark‘ayin Hiwsisoy wor ēr yerkord t‘agaworout‘ean nora anoun iwr J̌ ebou Xak‘an, p. 141.
54 Dan Shapira

place where Heraclius was based to conclude an agreement with him.23 It is clear
that the same story is repeated here, under different dates (unequivocally dated
by the author himself, with the events of 628 CE preceding those of 626 CE) and
in two different versions.24
At the beginning of the 37th year of Husraw (627), the king of the North sent
the army he had promised “appointing his nephew (brother’s son), whom they
call, in honour of his princely rank among them, by the name Šat‘“.25 This army
invaded Albania and Atropatena, Husraw of Persia tried, in vain, to prevent the
Khazars from siding with Heraclius, while Heraclius was advancing towards
Ctesiphon. Then, the murder of Husraw by his son Kawād and Kawād’s en-
thronement are described. There then follows the accounts about the Catholico-
sate of Viroy of Albania (released by Kawād from his prison at Ctesiphon) and
his return to Albania, and Viroy’s assistance to the victims of the Khazar Šat‘‘s as-
sault in Albania.26 It is only then that the Khazars’ siege of Tbilisi, their capture of

23
This should be a reference to the Persian-Avar attack on Constantinople, when Heraclius,
who was then in Asia Minor, succeeded in lifting the siege from his capital with the aid provided
by the North-Caucasian barbarians, see M. I. Artamonov, Istorija Xazar, Leningrad 1962,, p. 145 and.
n. 11. Cf. also M. van Esbroeck, “Une chronique de Maurice a Héraclius dans une récit des sièges de
Constantinople”, Bedi Kartlisa 34 (1976), pp. 74-96. Note that [Pseudo-]Sebēos, the Armenian au-
thor of the mid-seventh century, mentioned Čepetux, or Čēnastan-Čepetux, as a general of the
xak‘an of the North, who helped an Armenian army, which revolted against the Persian overlords,
to pass from Khorasan via Derbend to assist Heraclius (see Artamonov, Istorija Xazar, p. 147). This
perplexing title/name (cf. R. Bedrosian, “China and the Chinese according to 5-13th Century Classi-
cal Armenian Sources”, Armenian Review, Vol. 34 No.1-133 (1981), pp.17-24) is probably connected to
the Arabic form sīnǰibū found in Ṭabarī’s History, ed. de Goeje, I, p. 895, cf. J. Markwart, Wehrōt und
Arang, Untersuchungen zur mythischen und geschichtlischen Landeskunde von Ostiran, hrsg. v. H. H.
Schaeder, Leiden 1938, p. 142; cf. D. M. Dunlop, The History of the Jewish Khazars, Princeton Univer-
sity Press 1954, p. 35. Yabgū-Xāqān is mentioned in an epic context in a Pahlavi text of late prove-
nance, see J. Markwart, A Catalogue of the Provincial Capitals of Ērānšahr (Pahlavi Text, Version and
Commentary), ed. by G. Messina S. I., Pontificio Istitituto Biblico, Roma 1931, § 35 (pp. 17, 85).
24
Cf. M. I. Artamonov, Očerki drevnejšej istorii xazar, Leningrad 1936, pp. 51-54; M. I. Artamo-
nov, Istorija Xazar, Leningrad 1962, p. 145 n. 11; p. 162 n. 15; cf. Constantin Zuckerman, “The Khazars
and Byzantium – The First Encounters”, The World of the Khazars: New Perspectives. Selected Papers
from the Jerusalem 1999 International Khazar Colloquium, ed. H. Ben-Shammai, P. B. Golden, A.
Roná-Tas, Leiden 2007, pp. 399-432, esp. p. 406-7.
25
zełborordin iwr woroum i patiw išxanout‘eann iwreanc‘ Šat‘ anoun kardayin, p. 142.
26
Cf. also Kirakos Ganjakec‘i, Patmout‘iwn Hayoc‘, ed. K. Melik‘-Ohanǰanjan, Erevan 1961, p. 195.
There is a rather wide-spread modern myth according to which Movsēs used an Eastern-Slavic
word salo in his description of the foodstuffs of the 7th century Khazar warriors who put siege on
Tbilisi; as the authority, Patkanjan’s 1861 Russian translation, pp. 125-126, is quoted; however, the
 On the Relative Value of Armenian Sources for the Khazar Studies 55

the city and the atrocities committed by the Khazar “king” are mentioned: the
Khazars “brought the two princes, one the chief-governor of the Persian king-
dom, and the other a local native, from the princely family of the Georgian coun-
try, as captives before the king, who commanded that their eyes be put out as
retribution for having insultingly represented his image as blind. And with dire
tortures he strangled them to death, then stripped their skin from their bodies,
stretched it, stood it up, filled it with straw and suspended it from the top of the
wall”,27 and returned home, leaving his forces in the hands of his son Šat‘; then
follows a very long description of the woes wrecked by the Khazar in Albania in
the next years. Later, the Albanian author tells us about Orthodox coercion in
Albania launched by Heraclius. The approaching Byzantine armies posed a mor-
tal threat to the non-Orthodox population of Albania, and, during the tumult of
war, many Albanians, Christian and other, fled from their country to the Persian
territory; however,
ew k‘ahanay womn anoun Zak‘aria ayr sourb wor ēr hamakan Partaway
ekełec‘wojn hez ew handart wor ed ganjn iwr i weray noc‘a ew erdmambk‘ ew azgi
azgi hnariwk‘ zerjoyc‘ zbazoum anjins K‘ristonēic‘ ałot‘iwk‘ iwrovk‘ erašxaworeal
znosa, na ew vasn Hrēic‘ ew Het‘anosac‘owsti ew yetpy govec‘aw gorj nora ew
vkayeal yamenec‘ownc‘ kargec‘aw yepiskoposapetowt‘iwn at‘oṙayn Ałownic‘,
a certain priest called Zakaria, a holy man, who was a monk (obedient) at the
church of Partaw, a meek and humble fellow, took command; he saved many
Christians by oaths and various other means, by his prayers and guarantees,
which he also made on behalf of the Jews and pagans; his deeds were afterwards
praised by everyone, and he was appointed to a bishopric in the See of Albania.28

text has no such statements (and in 1984 Smbatjan’s Russian translation the whole end of II.14 is
absent). It was, in fact, the interpretation of Nikolai Marr of the passage is question that ascribed
our Albano-Armenian author the use of the word salo, see N. Ja. Marr, “Po povodu russkogo slova
«salo» v drevnearmjanskom opisanii xazarskoj trapezy VII v.”, in: N. Ja. Marr, Izbrannye raboty, T.
V. M.; L., 1935, 97-113. This Eastern-Slavic form, salo, is impossible for the early 7th century, compare
Polish sadło (the root being sad-), let alone the fact the the Armenian text has nothing similar.
27
Armenian text, p. 153: acin ew zerkosin išxansn zmi išxann pet kołmnakal t‘agaworout‘eann
Parsic‘, ew boun bnakč‘ac‘ iwroc‘, i tohmē išxanout‘eann ašxarhin Vrac‘, zerkosean ounelov; work‘
ibrew acan jerbakalk‘ aṙaǰi t‘agaworin, hramajeac‘ p‘orel zač‘s noc‘a p‘oxanak zi koyr nkarec‘in zpat-
ker nora i naxatel noc‘ zna; ew daṙn č‘arč‘aranok‘ hełjamah arareal znosa wołǰoyn varec‘in zmort‘s
noc‘a jandamoc‘ noc‘a, ew prkeal hagneal lc‘eal znosa xotov kaxec‘in i veroust zparspēn; I have
adopted, with slight alterations, the translation given in Dowsett, The History of the Caucasian Al-
banians, p. 95.
28
Ed. Arak‘ēlian [Arakeljan], Erevan 1983, p. 132 l. 16-19; Dowsett, The History of the Caucasian
Albanians, p. 80. This seems to be the only reference to Jews in Albania in the first millennium of
56 Dan Shapira

It is not difficult to see that this last account is a repetition, taken uncritically
from one of the sources which were before our compiler, resulting thus in cre-
ating the false impression that there were two sieges. This Albano-Armenian
version maintains that the defenders of the citadel mocked two rulers; there
were two leaders of the defenders, and both were executed in a similar fashion.29
However, in 625/6, when Heraclius and his “Khazar” allies besieged Tbilisi,
the city was defended by its local Georgian Christian ruler subject to Iran, and by
a Persian garrison under a Persian officer. It seems that the genuine Georgian
account of these events was lost, if it ever existed in a written form in Georgian,
because for the later Georgian readers the prospect that their ruler had probably
abandoned what was seen since then as the core of Georgian identity, was
deemed to be offensive. The period 606-629 CE was a religiously embarrassing
period for them, thus the sources would have been rewritten anew. It is to be ob-
served that such an excellent Armenian source as Sebēos, writing only a couple
of decades after the Byzantine-“Khazar” entente in Transcaucasia, does not men-
tion the sack of Tbilisi at all.30 It seems that the two existing Georgian accounts
about the joint Byzantine-“Khazar” siege of Tbilisi drew upon Armenian sources

the Common Era. From the newly found Albanian texts of the New Testament, we know that the
Albanian word for “Jew” was derived from the Middle Iranian *wāzāragān, testifying thus to the
trade of the Jews in Caucasian Albania. This is at odds with the late 19th/twentieth centuries leg-
ends of the Mountain Jews about their warlike origins from the Sasanian military mercenaries. On
Viroy’s ransoming the Armenian, Georgian and Albanian prisoners taken by xazir šat‘, cf. also
Kirakos Ganjakec‘i, Patmout‘iwn Hayoc‘, ed. K. Melik‘-Ohanǰanjan, Erevan 1961, p. 195.
29
While the parallel Georgian accounts (see further) avoided mentioning the humiliating
death of their ruler (“the local native, from the princely family of the Georgian country”, which can
be identified as Stepanoz), the Albano-Armenian source had no reason to pass over the tragic fate
of Stepanoz in silence, rather omitting any reference to the humiliation forced on the Albanian
king by Heraclius: Movsēs claims that the stuffed skin of the neighbor and relative of the Albanian
king was suspended from the top of the wall in Tbilisi, not sent to the Albanian king as a warning.
30
It is impossible to relate here to the complicated problems surrounding this work; in fact, in
the end of the short Part II, which actually is the preface to the work of Sebēos proper (Part III),
[Pseudo-]Sebēos promises to relate the addressing of Heraclius to the Northern Countries, to the
T‘et‘al (*Hephthalite) king, the sending of innumerable hordes of armies, the Greek campaign in
Atrpatakan, their pillage and their return through Paytakaran, arrival of the Persian troops from
the East to combat them, the war in Albania, the Emperor’s return to Naxčavan, the Ačeš battle,
Emperor’s return to his realm, the second campaign against Husrau, the Nineveh battle, the attack
upon Ctesiphon, etc. In reality, Ch. 26 of Sebēos proper does not contain much of the accounts
promised by the later redactor of the Preface (neither the siege on Tbilisi, nor the entente with the
Northern barbarians), who apparently drew upon the tradition current in the 9th century.
 On the Relative Value of Armenian Sources for the Khazar Studies 57

(like those of the Albano-Armenian Chronicle?) and were composed long after
the events they describe.31
According to the version of Kartlis Cxovreba (= KC), the eristavta mtavari
(“the leader of the heads of the people”) of Kartli, Stepanoz son of Guaram,32
trapped between the Byzantines and the Persians, did not dare to adopt a royal
title. It seems that his religious politics were non-Orthodox and he presumably
returned to Monophysite Christianity after the Sasanian government issued a
decree ordering all Christians to embrace it, for it is said that he was impious and
did not increase the religion33 and did not serve God. In Georgia, which only a
few years earlier had gone through a painful split with the mother-Church of
Armenia and jointed the previously hated Chalcedonians, the situation was ex-
tremely sensitive.34 In the war between Byzantium and Persia, the presumably
Monophysite Stepanoz supported the Persians against the Byzantines supported
by his Chalcedonian compatriots, who were led by the erismtavar (“ethnarch”)
Adarnase/Adrnerse, Stepanoz’ relative from the “older” line of the Chosroid
House. According to the 11th century Chronicle of J̌ uanšer (KC pp. 223ff.), in 627
CE Heraclius brought Turks from the West, gathered innumerable troops and at-
tacked Persia, coming first to Kartli. Stepanoz, loyal to his Persian masters, forti-
fied the citadel of Tbilisi (known later as Qal‘a)35 and, though the Byzantines laid

31
J. Marquart, Osteuropäische und ostasiatische Streifzüge, Lepzig 1903, p. 394 n. 2, noted that
the account in the Georgian Chronicle is secondary. See also R. W. Thomson, “The Anonymous
Story-Teller (Also Known as “Pseudo-Šapuh”)”, Revue des études arméniennes NS 21 (1988-89), 171-
232.
32
Gorgenēs the Curopalates, ruled ca. 588-602, the first holder of the Iberian Principate,
founder of the Guaramid line of the Chosroid House, see C. Toumanoff, “Iberia on the Eve of
Bagratid Rule. An Inquiry into the Political History of Eastern Georgia between the VIth and the IXth
century”, Le Muséon, Vol. LXV (1952), pp. 17-149, pp. 199-259, p. 199ff. and the table on p. 259. Cf. also
C. Toumanoff, “The Princely Nobility of Georgia”, From Byzantium to Iran. Armenian Studies in
Honour of Nina G. Garsoïan, ed. J.-P. Mahé & R. W. Thomson, Atlanta, GA 1997, pp. 37-46.
33
A semantic calque from Persian, found also in Syriac; for an example of the Syriac usage, cf. J.
P. Asmussen, “Christians in Iran”, The Cambridge History of Iran 3(2), The Seleucid, Parthian and
Sasanian Periods, ed. by Ehsan Yarshater, Cambridge, London etc. 1983, pp. 924-948, p. 944.
34
Cf. M. Bíró, “Georgian Sources on the Caucasian Campaign of Heracleios”, Acta Orientalia
Academiae Scientiarum Hungariae, XXXV (1) 1981, pp. 121-132, p. 127.
35
This Arabic name for Tbilisi’s citadel is anachronistic here, as it was given to the citadel only
after Tbilisi became a Muslim city in the mid-7th century and remained so for the following centu-
ries. For another example of terms used anachronistically, cf. Ja. A. Borovskij, “Vizantijskije, staro-
slavjanskije i starogruzinskije istočniki o poxode rusov v VII v. na Car'grad”, Drevnosti Slavjan i Rusi,
ed. B. A. Timoščuk, “Nauka”, Moscow 1988, pp. 114-119.
58 Dan Shapira

siege to Tbilisi, made daily sorties out of the city gates and fought the Greeks, un-
til he was killed, and the Emperor seized Tbilisi. But in the citadel there re-
mained a few who did not surrender. Obviously, most, if not all of them, were
members of the Persian garrison, although the text does not state this explicitly:
“The commander of the citadel insulted the king, that is, the Emperor, calling
down from the citadel: “You have the beard of a billy-goat, and you have the neck
of a he-goat” [in the 12th century Armenian translation of the Georgian, which is
the oldest attestation of the text:36 “Up and depart, you smelly goat”), accusing
him thus of pederasty (cf. further)]. The king commanded:37 “Although this man
scornfully calls me a he-goat, yet his remark is not false”. He took then the book
of Daniel, and found it written thus: “The goat of the West will come forth, and
he will destroy the horns of the ram of the East” [a paraphrase of Daniel 8.3-10].38
Then the king rejoiced, and was convinced that everything would succeed for
him against the Persians. The king [Emperor] then summoned the son of Bakur,
king of the Georgians [i.e., Bakur III, d. 580, a descendant of Dači (r. c. 522-534)
the son of Vaxtang (r. 447-502/522?)], who was eristavi of Kaxeti, named Adar-
nase,39 and gave him Tbilisi and the principality of Kartli. He left with him an
eristavi who was called J̌ ibɣo,40 and ordered him to attack Qal‘a. The king himself

36
KC in Old Armenian was published in Tbilisi in 1953 by I. Abuladze, Kartli Cxovrebis Jveli
Somxuri Targmani. See also an important study by S. S. Kakabadze, “Ustanovlenije kritičeskogo
teksta načal'noj časti ‘Kartlis Cxovreba’”, Palestinskij Sbornik 15 (78), 1966, pp. 172-180; also R. W.
Thomson, “The Armenian Version of the Georgian Chronicles”, Journal of the Society for Armenian
Studies 5 (1990-1991), pp. 81-90. For an English translation of the Armenian and Georgian texts, see
R. W. Thomson, Rewriting Caucasian History. The Medieval Armenian Adaptation of the Georgian
Chronicles. The Original Georgian Texts and the Armenian Adaptation. Translated with Introduction
and Commentary, Oxford 1996, p. 233 (Georgian), p. 234 (Armenian).
37
The verb used is a semantic calque of Persian farmūdan, used for both “to command” and “to
speak”.
38
K. Czeglédy, “Herakleios török szövetségesei”, Magyar Nyelv XLIX (1953), pp. 319-323, p. 322,
remarked that the Byzantine Emperors tended to consider themselves as the “He-goats of the
West” (see also M. Bíró, “Georgian Sources on the Caucasian Campaign of Heracleios”, Acta Orien-
talia Academiae Scientiarum Hungariae, XXXV (1) 1981, pp. 121-132, p. 129).
39
Ruled 627-642, see Thomson, Rewriting Caucasian History, p. 380, or, according to the table
on p. 259 in C. Toumanoff, “Iberia on the Eve of Bagratid Rule. An Inquiry into the Political History
of Eastern Georgia between the VIth and the IXthh century”, Le Muséon, Vol. LXV (1952), pp. 17-149,
pp. 199-259, 627-630/4 CE.
40
See P. B. Golden, Khazar Studies. An Historico-Philological Inquiry into the Origins of the
Khazars, Vol. I, Budapest 1980, p. 187ff. M. I. Artamonov, Istorija Xazar, Leningrad 1962, p. 146, iden-
tified him with Mo-ho-šad, the younger brother of T‘ong Yabɣu (who became Qaɣan not later than
 On the Relative Value of Armenian Sources for the Khazar Studies 59

set off to wage war on the Persians. Within a few days they captured Qal‘a and
seized its commander. The eristavi [i.e., as it seems, J̌ ibɣo] first filled his [Ste-
panoz’s, as it appears from the following] mouth with drahkanis, since the Em-
peror had been pleased with his words. But afterwards he had him flayed, and
had it [the skin] sent back to the Emperor at Gardaban [in Albania]. For his pre-
sumption against the king he was slain. By such a death Stepanoz and his adher-
ents perished. God destroyed the mtavari Stepanoz, because he did not live with
trust in God. He was an enemy to the faithful, and loved the impious” [i.e., the
Persians or anti-Chalcedonites].41
Here the sequence is as follows: Heraclius allied himself with the Turks/Kha-
zars, assembled an army and went to Georgia. There is not a single word about
Turkic, or Khazar, troops arriving in Georgia, however; the role of the Persians is
downplayed; it was Stepanoz who organized the defenses, fought vigorously, but
was killed in battle. After his death, Tbilisi was captured, but some people, obvi-
ously, Persians, kept defending the citadel (at this point we can assume that this
continuation of the siege could have been seen as two different sieges, as it be-
came, seemingly, in the case of the Albano-Armenian Chronicle). Their com-
mander mocked the appearance of Heraclius, his beard and his neck. The Em-
peror read aloud from Daniel, appointed a new ruler of Kartli from the local dyn-
asty and a relative to Stepanoz who had been killed, and departed to fight the
main forces of the Persians. He left with the new Georgian ruler a J̌ ibɣo, who was
described as merely an eristavi, “prince”, and who had been successful in cap-
turing the citadel. There is nothing in the text that could lead one to realize that
this J̌ ibɣo was the mighty ally of the Emperor, and not merely one of his generals
(in the Armenian translation of KC he is called zoragloyx, “the head of the
army”). The whole setting of the version in KC is such that it attributes the sack
of the Georgian capital to the Orthodox Emperor, not to a Northern barbarian.
This J̌ ibɣo takes revenge for the Emperor’s insult: he first fills the mouth of the
captured officer with drahkanis, then has him flayed. Then it is concluded that

618 CE, see Golden, Khazar Studies, Vol. I, p. 188); however, Pritsak identified the Turkic ally of
Heraclius with “T‘ong Še-hu (yabɣu), ruler from 618 to 630, [who] acquired the high title of šad in
627. His son established the new realm in the 630s and 640s” (O. Pritsak, “The Khazar Kingdom’s
Conversion to Judaism”, Harvard Ukranian Studies 2.3 (1978), pp. 261-281, p. 261). Cf. also A. Bom-
baci, “Qui était Jebu Xak‘an?”, Turcica 2 (1970), pp.7-24.
41
KC, pp. 224-6; the translation adopted from Thomson, Rewriting Caucasian History, p. 235.
60 Dan Shapira

“by such a death perished Stepanoz and his adherents”, though we have believed
Stepanoz had already died a warrior’s death.
The version of another Georgian text, which was older than KC, namely, The
Conversion of Kartli, or Mokcevai Kartlisai [henceforth: MK],42 25-26 is an
abridgement, albeit a coherent one, of a source common with KC:
da misa šemdgomad eristvobda Stepanoz, jē misi, jmai Demetresi, da ikmoda
ek’lesiasa ǰuarisasa.
Mašin čamovlo Herak’le mepeman Berjentaman. Da uqmo cixis tavman k’alayt
Tpilisisayt mepesa Herakles vac-bot’obit, xolo man perqi daap’qra da Daniel moiɣo
da moijia saxe ese: “movides vaci igi mzis dasavalisay da šemusrnes rkani verjisa
mis mzis aɣmosavalisani”,43 da mepeman hrkua: “esret iq’os sit’q’uay, me migago
misagebeli šeni”. Da daut’eva J̌ ibɣo eristavi brjolad da tvit c’arvida Baɣdads brjolad
Xuasro mepisa. Xolo aman J̌ ibɣo mcireta dɣeta šemdgomad k’alay gamoiɣo da
cixistavi igi šeip’q’ra dap’iri drahk’anita aɣuvso. Da merme mrtels t’q’avi gahqada
da mepesa uk’uana mis c’ia Gardbans Varaz-Grigolissa šina.
“Then after him (Guaram) eristavi was his son Stepanoz, brother of Demetre, and
he was building the Church of the Cross.44 Then Heraclius, King of the Greeks,
swept (Kartli). Then the commander of the citadel Qal‘a of Tbilisi called King
Heraclius “a goat”. Then he (the King) put forth his leg firmly, took the book of
Daniel, and found it written thus: “The goat of the West will come forth, and he
will destroy the horns of the ram of the East”, and the king said: “Let it be so, I will
reward you”. He left J̌ ibɣo the eristavi to wage war, and went to Baɣdad to fight
King Husraw. And after a few days J̌ ibɣo seized Qal‘a, caught the commander of
the citadel and filled his mouth with drahkanis. Then he flayed him while still
alive, and it (his skin) was sent back to the king at Gardaban, to Varaz-Grigol’s”.
In this account one could not know who this J̌ ibɣo is, for he acts as one of the
generals of Heraclius. The commander of the citadel calls the Byzantine king just
“a goat”, which is more similar to the old Armenian testimony of the Georgian
text of KC. Ctesiphon is called Baɣdad, not Babylon; there is no bad word to say

42
See E. S. Takaišvili, “Obraščenije Gruzii”, Sbornik Materialov dlja opisanija mestnostej i plemën
Kavkaza XXVIII, Tiflis 1900; D. M. Lang, Lives and Legends of the Georgian Saints, London 1956, pp.
13-39; G. Pätsch, “Die Bekehrung Georgiens Mokcevay Kartlisay (Verfasser unbekannt)”, Bedi Kartli-
sa. Revue de kartvélogie, 33 (1975), pp. 288-337.
43
In the text of KC the quotation appears in a slightly different form: gamovides vaci dasavlisa
da šemusrnes rkani verjisa aɣmosavlisani.
44
The building of this important church is attributed now to his namesake, Stepanoz II (r.
637/642-650?), cf. Stephen H. Rapp, Studies in Medieval Georgian Historiography: Early Texts And
Eurasian Contexts, CSCO, Vol. 601, Subsidia Tomus 113, Peeters: Lovanii 2003, p. 344; cf. n. 47.
 On the Relative Value of Armenian Sources for the Khazar Studies 61

about Stepanoz, and his death is not mentioned at any stage of the story: it is the
commander of the citadel who was executed. But let see how the account contin-
ues.
After that, the text tells about Heraclius’ victories in Persia. The text adds that
the Emperor returned to Tbilisi, where he assembled all the Christians in the
churches and forcibly converted Magi and fire-worshippers (moɣuni da cecxlisa
msaxurni) and has slain those refusing to be baptized, purified/sanctified the re-
ligion of Christ (xolo Herak’le gancmida šǰuli Krist’esi), and streams of blood
washed the churches (da ek’lesiata šina mdinareni sisxlisani diodes). Surprisingly,
the text adds: “and the eristavi was the same Stepanoz the Great, and the Ca-
tholicos was Bartholomew for the second time” (da eristavobda igive didi Ste-
panoz da k’atalik’ozi iq’o Bartlome meored).
It seems that Heraclius carried out a massacre not only of Zoroastrians45 (and,
perhaps, other non-Christians—note that Jews are nowhere mentioned!—per-
haps because there were no Jews in town?),46 but also of Christians: all the Chris-
tians were rounded up into the churches which were washed with rivers of
blood. I would suggest that this is an indication that many Georgians in Tbilisi at
that point were Monophysite or Nestorian. And after all that, Stepanoz became
the ruler, and the Catholicos Bartholomew returned to his office for the second
time!47
The reference to the skin of the commander of the citadel sent to the king at
Gardaban, to the place of Varaz-Grigol, is interesting. Varaz-Grigol (r. 628-636)
was the Christian ruler of the Caucasian Albania who collaborated with the Per-
sians, just as Stepanoz had, but unlike him, Varaz-Grigol crossed the lines. When

45
On Heraclius’ agenda of converting the Persians, see C. Mango, “Deux études sur Byzance et
la Perse sassanide”, Travaux et Mémoires (Collège de France, Centre des recherches d’histoire et de
civilisation byzantines), 9, Paris 1985, pp. 91-118 (pp. 105-118: “Heraclius, Šahravaraz et la Vraix
Croix”). On some attitudes hostile to Heraclius, see Antiox Stratig. Plenenije Ierusalima persami v
614 g., gruzinskij tekst issledoval, perevel, izdal i arabskoje izvlečenie priložil N. Marr, (St. Peterburg
1909), p. 60.
46
On Jews in Sasanian Kartli, cf. D. Shapira, “Gleanings on Jews of Greater Iran under the Sasa-
nians (According to the Oldest Armenian and Georgian Texts)”, Iran & the Caucasus 12.2 (2008),
pp. 191-216.
47
One might suggest that the pro-Iranian and non-Orthodox Stepanoz I (590-627) was con-
fused with Adarnase’s son, Stepanoz II (642-650), the Kartli ruler who capitulated in 645 to the Ar-
abs, securing thus his country. Bartholomew held the office between 591-595; in 627, the Catholicos
was Babila (619-629), who was followed by Tabor (629-634).
62 Dan Shapira

Heraclius stayed at Gardaban, he forced Varaz-Grigol to accept the King’s (Mo-

noenergetic) “Orthodoxy”.48 What was the point of sending the skin of the Per-
sian officer to (the place of) Varaz-Grigol? Was it not rather the skin of Stepanoz
which was sent (as the version of KC would suggests), in order to intimidate the
shaky Christian ally?
The Armenian versions (both in the translation of KC and in the Albanian
chronicle) have better readings of the insult addressed to Heraclius, while in the
Georgian sources, especially in the extant text of KC, the defenders mock the
Emperor’s appearance, not his sexual behavior; this change of context is stressed
by putting a Biblical quotation into the mouth of the Emperor.49 According to
the Georgian sources, Qal‘a, the fortress of Tbilisi, was captured in a few days af-
ter Heraclius and his army departed for Babylon/Baɣdad, i.e., Ctesiphon, while
according to the afore-mentioned last account of the Albano-Armenian Chroni-
cle, the capture of the city by the Šat‘ was achieved in two months and this hap-
pened a year later (629 CE), only after Husraw has been killed in a conspiracy.
To sum up, it is clear that the Albano-Armenian author had more than one
source at his disposal, and that his text, whilst garbled, is more complete.

48
After the Byzantine victories, Heraclius proceeded baptizing anew the local Monophysite
Christians into his Monoenergism; according to Sambat, the son of David, Varaz-Grigol was bap-
tized anew by the Emperor, and it was only later that the Albanian Catholicos Viroy re-baptized
him back into the Armenian brand (M. I. Artamonov, Očerki drevnejšej istorii xazar, Leningrad
1936, pp. 59-60; C. J. F. Dowsett, The History of the Caucasian Albanians by Movsēs Dasxuranc’i,
London Oriental Series. Volume 8, London 1961, p. 109; M. Bíró, “Georgian Sources on the Cauca-
sian Campaign of Heracleios”, Acta Orientalia Academiae Scientiarum Hungariae, XXXV (1) 1981, pp.
121-132, pp. 131-2). My understanding of the episode is in agreement of C. Toumanoff, in his review
of Dowsett in BSOAS 25 (1962), pp. 364-366. C. Zuckerman (see K. Cukerman, “Xazary i Vizantija:
pervyje kontakty”, Materialy po Arxeologii, Istorii i Étnografii Tavrii, VIII (Simferopol' 2001), pp. 312-
333 (p. 319) treats the episode with Varaz-Grigol’s baptism very differently.
49
Heraclius was the first emperor (caesar, imperator, augustus) to adopt the ancient Hellenic
title of Basileus (first documented in 629, see I. Zepos and P. Zepos, Jus Graecoromanum, Vol. I,
Athenai 1939, p. 39). On the motives to adopt this title, see I. Shahid, “The Iranian Factor in Byzan-
tium during the Reign of Heraclius”, Dumbarton Oaks Papers 26, Wash. 1972, pp. 295-312.
 Some Remarks on the Inscription of Maris, Casiti filius
(Classical-Oriental notes, 9)*

Giusto Traina
University of Paris-Sorbonne

Abstract
The funerary inscription of Maris Casiti f(ilius) from Mainz (AÉ 1959.188; 1967.339) gives evidence of
several Oriental names. The auxiliary mounted archer Maris, as well as his father Casitus and his
brother (and also brother in arms) Masicates bear Semitic names, but Tigranus, another soldier
who contributed to this monument, is possibly Armenian, as well as Variagnes, the commander of
their auxiliary unit, whose name is the first dated occurrence of the name later attested in the form
Vahagn.

Keywords
Latin epigraphy, Roman military history, Parthian history, Semitic onomastics, Armenian onomas-
tics

The funerary monument of Maris Casiti f(ilius), now in Landesmuseum Mainz, is

one of the best pieces of evidence of Oriental auxiliary troops in the Roman West
during the Principate (Fig. 1).1 The relief shows him mounting a horse, in the act
of pulling an arrow (or several arrows?), assisted by a servant afoot ready to sup-
ply him with fresh arrows. His inscription reads:
Maris Casiti f(ilius) anno(rum) L / stip(endiorum) XXX ala Part(h)o(rum) et /
Araborum turma / Variagnis Masicates / frater et Tigranus / posierunt. 2

*
I would thank Claudia Ciancaglini, Giulia Francesca Grassi, Mark Geller, Patrick Le Roux and
James Russell for their valuable comments. But above all, I would thank Garnik Asatrian who sent
me some important considerations on the name Variagnes. He may well be the person celebrated
by this Festschrift, nonetheless he was the best authority on this question: let it be a sort of tribute
in the tribute. For a former German version of this article see ZPE 185 (2013): 279-285.
1
Landesmuseum Mainz, Inv-Nr. S 634, height 210 cm, broadth 92 cm, thickness 28 cm.
2
AÉ 1959.188; Nesselhauf/Lieb 1959, n° 169 = AÉ 1967.339. Boppert reads Vartagnis (see also
Herz 1982: 174), but this reading is highly unlikely.
 Koninklijke Brill NV, Leiden, 2015
64 Giusto Traina

“Maris, son of Casitus, aged fifty. He served for thirty years. (A cavalryman) in the
ala of the Parthians and of the Arabs, in the turma commanded by Variagnes/is.
Masicates, his brother, and Tigranus, made (this monument)”.
The finding of this text eventually caught the attention of the Oriental schol-
ars Hans Petersen and Helmut Humbach (Petersen 1966; Humbach 1968; see also
Ziethen 1997); yet some new considerations may be added. Maris, son of Casitus,
died at the age of fifty, after thirty years of service.3 Therefore, after a training as a
mounted archer, he became an auxiliary soldier at the age of twenty. For both
stylistic and epigraphical reasons, the monument seems to date before the reign
of Claudius;4 so, Maris (and most probably, his brother Masicates) may have
joined the Roman army by the end of the principate of Augustus, eventually be-
ginning his service under the command of Tiberius in 10-12 AD, when Mongon-
tiacum increased its strategic importance as a consequence of the disaster of the
Teutoburger Wald. There is no evidence, however, that he served in Germany all
the time).5 At any rate, when Maris died he was serving in the ala “of the Parthi-
ans and of the Arabs”.
This ala has been identified with the same unit attested in Germania Inferior
and Dalmatia, possibly established after Teutoburg, with the goal of sending
there experienced soldiers with special know-how.6

3
Possibly, these figures are round numbers. But a certain number of inscriptions shows that a
great deal of auxilia were recruited at the age of twenty or twenty-one (see Gallet 2012: 55).
4
Selzer 1988: 158, n° 91; Boppert 1992: 130f., pl. 27, who dates the stele before the reign of Clau-
dius.
5
Petersen 1966: 68f. dates the establishment of this unit after 38 AD, but see the criticism of
Kennedy 1977: 528. On military life in Mogontiacum see Schuhmacher 2003 and, most recently, the
catalogue of the exhibition Im Dienst des Kaisers, Rheinisches Landesmuseum Trier, 8/24/2012-4/7
2013 (non vidi).
6
See the general conclusions of Herz 1982: 182. On the history of this unit, see Kennedy 1977:
526; Herz 1982. Spaul (1994: 176-178) claims the existence of only one unit called Ala I Augusta Par-
thorum et Araborum Sagittaria, but see the considerations of Le Roux 2005: 480. For the Ala Par-
thorum veterana, see Alföldy 1968: 28f. Some brick stamps found in Cantabria attest the stationing
of an ala Parthorum between the reigns of Claudius and Vespasian, which does not seem to coin-
cide with the unit stationing in Germany. On the rise of the ‘ethnic’ units, see Speidel 1975. On the
diffusion of Semitic names in the Roman world, see Solin 1980. A list of units with double ethnic
names is drawn by Holder 1980: 22-23.
 Some Remarks on the Inscription of Maris, Casiti filius 65

Figure 1. The funerary stele of Maris (Courtesy of Landesmuseum Mainz)

Maris and his father Casitus bear Semitic names, most likely Aramaic.7 If the
name Maris (and the derived composite names) is very common, the form Ca-
situs is apparently less clear.8 In fact, this is the Latin form of Palmyrene qšṭ’, qšṭy

7
Grassi 2012b: “il figlio di Casitus si chiama Maris, che è un nome aramaico e questo potrebbe
deporre a favore di un’origine aramaica anche del nome del padre”. Maris is attested in several in-
scriptions of Roman Syria (see also Intinsky 1958: 76; Petersen 1966: 64; Solin 1980: 323; Herz 1982:
178; Grassi 2012a: 222). The form Mārē is also known from a Christian Pahlavi inscription dating
from the Sasanian period (see Shaked 1995: 249 (who does not exclude a possible reference to Syr.
Māryā, The Lord); Jullien/Jullien 2003: 4; Grassi 2012a: 222).
8
Petersen (1967: 64) suggested that the name “is probably not Persian, but rather Semitic”.
66 Giusto Traina

and qšty, a name usually transcribed in Greek as Χασετος (Negev 1991: 36; Sartre
1985: 243f.; Grassi 2012a: 212, and idem 2012b).9 It would be very tempting to con-
nect it with Aramaic qšṭ/t(’), “arch”, “archer”.10 The name Masicates is Semitic as
well.11 A tribe of Māsikat is attested by a Safaitic graffito found near Ǧawā, in the
Eastern desert of Jordan (Hackl et alia 2003: 160); a form msykt is also attested in
Qatabānic (Hayajneh 1998: 234), and a name Μασιχος, with several variants, is at-
tested in Syria (Grassi 2012a: 226). Although the term frater can also signify “fel-
low soldier” in a military context,12 Masicates seems to be another son of Casitus:
otherwise we need to explain why Tigranus, the other contributor to the erec-
tion of the monument, is not also designed as a frater.
The names Variagnes and Tigranus are connected with quite different ethnic
contexts. In his seminal contribution of 1966, Petersen concluded that “in all
probability, therefore, Casitus and his two sons Maris and Masicates were Arabs;
Variagnes and Tigranus were Parthians. Consequently, the turma of Variagnes
consisted of both Arabs and Parthians. In this connexion it should not be over-
looked that the Semitic names of the inscription point to Syria and its border-
lands rather than to the Arabian peninsula” (Petersen 1967: 64).
We may add some further considerations. Although the ala Parthorum et
Araborum was ‘ethnically’ made up of Parthians and Arabs,13 the name Tigranus
is undoubtedly Armenian (Justi 1895: 324f.; Traina 2005). An Armenian connec-
tion could be also possible for the name of Variagnes, the officer in charge of the
command of the turma.14 Although we cannot exclude a Western Iranian form
(Ciancaglini 2012; see also Nyberg 1964: 276; Schmitt 1989: 99; Ciancaglini 1998:

9
See also the Latin form Chasetus, attested by the rosters of the Cohors XX Palmyrenorum:
PDura 100, col. XXXIX, 14, AD 210 (Welles et alia 1959: 335); PDura 101, col. XXXIX, 17, AD 222 (ibid.: 360).
10
Grassi 2012a: 212, and idem 2012b, who also considers a Safaitic form ksṭ (of uncertain etymol-
ogy) attested in some fifty inscriptions (see Ryckmans 1934: 115; Harding 1971: 500; Negev 1991: 36
(also on Nabataean kšyṭw); Grassi 2012b).
11
Winnett 1957: 49, n° 287; Instinsky 1958: 76, who mentions the Μασικας of IGLS II 264; Peter-
sen 1967: 63 (also proposing, as an alternative, an Iranian name meaning “Massagete”); Harding
1971: 545; Solin 1980: 323 (who holds Masicates as an Iranian name); Hackl et alia 2003: 160; Grassi
2012a: 226.
12
For example, the Vindolanda tablets give important evidence of “brothers in arms” (see
Birley 2002).
13
According to Kennedy 1977: 526, this unit is “probably the result of an amalgamation of a unit
of Arabs with an ala Parthorum”.
14
Humbach 1968: 320 (“Variagnes, der Führer der Turma des Bestattenen”); Traina 2005: 260 n.
6. On the units named after their commanders, see Holder 1980: 21-22.
 Some Remarks on the Inscription of Maris, Casiti filius 67

63, n. 15; Russell 2013:1), on the other hand it is worth noting that this theophoric
name recalls Vahagn, an Armenian warlike god whose typology is connected
with the Vǝrǝθraγna- attested in the Avesta, and also with the Greek Herakles
(Humbach 1968 (referring to the inscription of Maris); Russell 1989; Mahé 1994:
782-783; Petrosyan 2007; Ivanov 2011; Russell 2013).15 The Old Iranian form of this
name can be reconstructed as *Wṛθragna-; the corresponding forms in Middle
Iranian are, in chronological order, Varhrān [wlhl’n'], Varhrām [wlhl’m], Vahrām
[w’hl’m] and, after the 3rd century AD, Bahrām. All these Middle Iranian forms
belong to the Northwestern branch of the Iranian languages, where Old Iranian
nexus -θr- becomes -hr- with a later metathesis (Russell 2012; idem 2013).16 The
Latin inscription of Maris gives the first attestation of Vahagn in any form out-
side Armenian (with the addition of the usual suffix -es, from Greek -ης). In pre-
Christian Greater Armenia, this form is unprecedented and of particular value.17
But what does “Parthian” actually mean in this military context? In his pre-
liminary study on the Parthian units in the Roman army under the Principate,
Hugh Kennedy took into consideration the inscription of Antiochus, son of Anti-
ochus, another officer of the same unit of Maris, designed as Parthus Anazarba-
eus.18 Kennedy pondered whether to consider Parthus “as a cognomen or an eth-

15
Asatrian (2012) remarks: “This is the first part of a compound name (shortened later, with the
loss of the second component, type of, say, Warhrag/γn-bōžan, etc.), going back to Old Iranian
*Wṛθragna- (Avestan Vəәrəәθraγna-)”. On the religious aspects of Roman auxilia, see Haynes 1993.
16
Asatrian (2012) remarks : “it must be noticed that Arm. Vahagn in its present form can come
only from *Vahragn (with a later metathesis in Armenian of Iran. -rh- to -hr-); otherwise we would
have had *Varagn in Armenian. This metathesis could have taken place much later, in the Sasa-
nian time. Armenians took Parthian Warhragn in the early Parthian period (when the final -gn-
was clearly pronounced) and adapted it to *Varhagn (*Varhragn in Armenian is impossible: Ar-
menian phonetics in most cases simplifies -rhr- cluster to -rh- or -hr-), as in the first century AD the
same name in Parthian probably was pronounced already as Warhraγn or Warhraŋ, although in
historical writing it was depicted as wrtrgn (cf. also Parth. inscriptional Wryhr’n)”.
17
As remarked by Russell 2012. Asatrian 2012 remarks: “The Latin Variagnes can equally be the
reflex of either Armenian *Varhagn or Parth. Warhragn―with an important reservation: if the
name Variagnes dates from I BC to I AD, then it comes definitely from Armenian (*Varhagn), if it is
of an earlier coinage, it must be of Parthian origin (from Warhragn)”.
18
Antiochus’ funerary inscription reads:
Antiochus, | Antiochi f(ilius), | Parthus, Anaz|arbaeus, eques | ala(e) Parthorum | et Araborum,
euo|catus triplicarius, | stip(endiorum) X, donis don|atus, Belessippus | frater posuit.
“Antiochus, the son of Antiochus, the Parthian, from Anazarbus, a cavalryman of the ala of the
Parthians and the Arabs. He was promoted to triplicarius. He served for ten years and received
decorations. Belessippus, his brother, made (this monument)”.
68 Giusto Traina

nic”, finally stating: “the fact remains that he is of Parthian stock and regarded
his home as Anazarbus” (Kennedy 1977: 528). Yet Antiochus bore the same Greek
name (a very evocative one) of his father, and he resided in a Cilician city, and
his brother (either actual or in arms) Belessippus (Μελήσιππος?) had a Greek
name as well.19 Was he a descendant of Greek troops serving under the Parthi-
ans? (Dąbrowa 2006).
The epigraphic evidence, though relatively scarce (and somehow puzzling),
may nonetheless tell us something more about the composition of ‘Parthian’
units. Certainly, in the Julio-Claudian period, the auxiliaries designed or recog-
nizable as Parthians seem to belong to a warrior elite. This could explain the
honour of evocatus triplicarius given to Antiochus.20
Another interesting case is given by the inscription of Klis (Clissa) in Dalma-
tia: […] | I[…]L[…] | C(aio) Iul(io), Thridatis f(ilio), | dec(urioni) ala(e) Phar-
tho(rum), | an(norum) XXVI, dom(o) | Roma. H(ic) s(itus) e(st).21 We are dealing
here with an outstanding character. His father Tiridates might be identified ei-
ther with Tiridates III of Parthia, one of the sons of Phraates IV who were sent to
Rome as hostages (Dąbrowa 1987; Nedergaard 1988), or, more likely, with an
usurper (not necessarily an Arsacid) backed by Octavian after 32/1 BC, who died
in Rome.22 We do not know if Tiridates obtained the Roman citizenship from Au-
gustus, but his son was a civis Romanus domo Roma. So, although he was not a
Parthus anymore, his Parthian descent gave him the necessary charisma to
command a turma despite his young age.
As a matter of fact, the definition of a Parthian identity is less than clear. In
her dissertation, Charlotte Lerouge collected the evidence for the portrait of the
rival empire given by the classical sources : one of the most striking features
seems to be its ambiguity. In fact, the Romans recognized as Parthians a melting-
pot of several peoples and tribes more than an actual ethnic reality (Lerouge
2007; Muccioli 2007; Traina 2012).

AÉ 1976.495. See Schillinger-Häfele 1997, n° 99; Ziethen 1997: 127f.; Sayar 2000: 14.
19
The oscillation m/b is a very common one.
20
This is apparently the only “uncontroversial example of a decorated, non-citizen, auxiliary
soldier” (Roxan 1995: 143).
21
CIL 3.8746 = ILS 2532.
22
D.C. 51.18; 53.33; 55.10a.5; Iust. 42.5. On this interesting character, see Gaslain/Maleuvre 2007.
 Some Remarks on the Inscription of Maris, Casiti filius 69

Figure 2. The funerary stele of Antiochus (Courtesy of Landesmuseum Mainz).

Of course, the logic of the Roman army justified the creation of an ethnic la-
bel for the warriors serving as mercenaries for Rome, at least as early as 48 BC.23
However, we cannot be confident that this label was so precise: at least the aux-

23
We have evidence of mercenary troops fighting for Pompey at Pharsalus and Cassius at Phi-
lippi: Caes. BC 3.82; Luc. Phars. 8.294-304; App. BC 4.8.59; 4.11.88. See Kennedy 1977: 530: “It must be
borne in mind too, that Parthia was organized on a fundamentally different basis from the Roman
empire. Without a standing army, it was left to would-be mercenaries to join a noble or to seek a
living where pay was offered”.
70 Giusto Traina

iliary Tigranus attested by the inscription of Maris was Armenian, not Iranian,
nor it is to be forgotten that Variagnes, the commander of the turma where Ma-
ris served, was an Armenian as well.24
To sum up. In the first half of the 1st century AD, Roman auxiliary units gen-
erally preserved their original ethnic affiliation (Gallet 2012, passim). However,
the case of ‘Parthian’ contingents shows the limits of this ethnicity. In the case of
the ala Parthorum et Araborum, we suggest that the soldiers―Syrians, Arabs,
Armenians, Iranians―shared more or less a common language (Aramaic?), and
above all had in common the typical fighting techniques, which characterised
the Parthians, widely renowned as the best mounted archers in the world (Trai-
na 2011: 74-78).25

BIBLIOGRAPHY
Alföldy, G. (1968), Die Hilfstruppen der römischen Provinz Germania Inferior, Düsseldorf.
Asatrian, G. (2012), private e-mail of 9/21/2012.
Birley, A. R. (2002), Garrison Life at Vindolanda: a Band of Brothers, Oxford.
Boppert, W. (1992), Militärische Denkmäler aus Mainz und Umgebung, Bonn.
Cantineau, J. (1932), Le nabatéen. II. Choix de textes, Paris.
Ciancaglini, C. (1998), “Gli antecedenti del Romanzo siriaco di Alessandro”, La diffusione dell’ere-
dità classica nell’età tardoantica e medievale, eds. R. B. Finazzi, A. Valvo, Alessandria: 55-93.
―― (2012), private e-mail of 9/11/2012.
Cuff, D. B. (2010), The Auxilia in Roman Germany and the Two Germanies from Augustus to Caracal-
la: Family, Religion and ‘Romanization’, Ph.D. Dissertation, University of Toronto.
Dąbrowa, E. (1987), “Les premiers ‘otages’ parthes à Rome”, FolOr 24: 63-71.
―― (2006), “Les Grecs sous les drapeaux des Arsacides” = Idem, Studia Graeco-Parthica. Political
and Cultural Relations between Greeks and Parthians, Wiesbaden: 75-81.
Facella, M. (2006), La dinastia degli Orontidi nella Commagene ellenistico-romana, Pisa.
Gallet, S. (2012), Le recrutement des auxiliaires de l’armée romaine sous le Haut-Empire dans l’Occi-
dent romain, Ph.D. Dissertation, University of Paris-Sorbonne.
Gaslain, J. / J.-Y. Maleuvre (2007), “Auguste et les Arsacides, ou le prix des enseignes”, Parthica 8,
2006 [2007]: 169-194.
Grassi, G. F. (2012a), Semitic Onomastics from Dura Europos. The Names in Greek Script and from

24
According to Spaul (1994: 176-178, followed by Cuff 2010: 137-139), the Ala Parthorum et Arabo-
rum was originally raised in Armenia minor, but we need more evidence to support this hypothe-
sis.
25
For an evolution of Parthian (but also Armenian and Oshroenic) mercenary units, see Petit-
jean 2012: 32. The question of language in the auxilia still awaits a thorough investigation (see Hay-
nes 1999: 169ff.).
 Some Remarks on the Inscription of Maris, Casiti filius 71

Latin Epigraphs, Padova.

Grassi, G. F. (2012b), private e-mail of 9/12/2012.
Hackl, U. / H. Jenni; Ch. Schneider (eds.) (2003), Quellen zur Geschichte der Nabatäer. Textsamm-
lung mit Übersetzung und Kommentar, Freiburg (Schweiz).
Harding, G. L. (1971), An Index and Concordance of Pre-Islamic Arabian Names and Inscriptions, To-
ronto.
Hayajneh, H. (1998), Die Personennamen in den qatabānischen Inschriften, Hildesheim.
Haynes, I. P. (1993), “The Romanization of Religion in the ‘Auxilia’ of the Roman Imperial Army
from Augustus to Septimius Severus”, Britannia 24: 141-157.
―― (1999), “Military Service and Cultural Identity in the Auxilia”, The Roman Army as a Commu-
nity [JRA Suppl. 34], ed. A. Goldsworthy, Id.: 165-174.
Herz, P. (1982), “Die Ala Parthorum et Araborum. Bemerkungen zur römischen Heeresgeschichte”,
Germania 60: 173-182.
Holder, P. A. (1980), Studies in the Auxilia of the Roman Army from Augustus to Trajan, Oxford.
Humbach, H. (1968), “Variagnes und Barzimeres”, Germania 46: 320f.
Instinsky, H. U. (1958), “Grabstein eines berittenen Bogenschützen der Ala Parthorum et Arabo-
rum”, Germania 36: 72-77.
Ivanov, V. I. (2011), “A Probable Structure of a Protoform of the Ancient Armenian Song of Vahagn”,
Aramazd 6.1: 7-23.
Jullien, Ch. / F. Jullien (2003), Aux origines de l’Église de Perse : les Actes de Mār Māri CSCO, Sub-
sidia, t. 114, Leuven.
Justi, F. (1895), Iranisches Namenbuch, Marburg.
Kennedy, H. (1977), “Parthian Regiments in the Roman Army”, Limes. Acta XI Limeskongresses, ed.
J. Fitz, Budapest: 521-531.
Lerouge, Ch. (2007), L’image des Parthes dans le monde gréco-romain. Du début du Ier siècle av. J.-C.
jusqu’à la fin du Haut-Empire romain, Stuttgart.
Le Roux, P. (2005), “Armées et contrôle des territoires en Aquitaine et en péninsule Ibérique sous
les Julio-Claudiens” = Le Roux 2011: 471-502 (originally published in English, “Roman Mili-
tary Epigraphy”, The Roman Army in Hispania. An Archaeological Guide, eds. Á. Morillo; J.
León Aurrecoechea: University of León 2006: 451-471).
―― (2006), “Les inscriptions militaires” = Le Roux 2011, 489-502.
―― (2011), La toge et les armes. Rome entre Méditerranée et Océan. Scripta Varia I, Rennes.
Mahé, J.-P. (1994), “Un dieu guerrier à la campagne : l’exemple du Vahagn arménien”, CRAI: 779-
804.
Muccioli, F. (2007), “La rappresentazione dei Parti nelle fonti tra II e I secolo a.C. e la polemica di
Livio contro i levissimi ex Graecis”, Incontri tra culture nell'Oriente ellenistico e romano, eds.
T. Gnoli; Id., Milano: 87-115.
Nedergaard, E. (1988), “The four Sons of Phraates IV in Rome”, Acta Hyperborea 1: 102-115.
Nyberg, H. S. (1964), A Manual of Pahlavi. Part II. Glossary, Wiesbaden.
Negev, A. (1991), Personal Names in the Nabatean Realm, Jerusalem.
72 Giusto Traina

Nesselhauf, H. / H. Lieb (1959), “Dritter Nachtrag zu CIL XIII: Inschriften aus den germanischen
Provinzen und dem Treverergebiet”, BRGK 40: 120-228.
Petersen, H. (1966), “New Evidence for the Relations Between Romans and Parthians”, Berytus 16:
61-69.
Petitjean, M. (2012), Virtus equitum. Les transformations de la cavalerie romaine au IIIe siècle et la
crise du modèle classique de la guerre, M.A. Dissertation, University of Paris-Sorbonne.
Petrosyan, A. (2007), “State Pantheon of Greater Armenia”, Aramazd 2: 174-201.
Roxan, M. M. (1995), “The Hierarchy of the Auxilia. Promotion Prospects in the Auxilia and Work
Done in the Last Twenty Years”, La hiérarchie (Rangordnung de l’armée romaine sous le
Haut-Empire) ed. Y. Le Bohec, Paris: 139-146.
Russell, J. R. (1989), “Carmina Vahagni” = Idem., Armenian and Iranian Studies, Cambridge (Ma.),
2004.
―― (2012), private e-mail of 9/19/2012.
―― (2013), “Magic Mountains, Milky Seas, Dragon Slayers, and Other Zoroastrian Archetypes”,
Bulletin of the Asia Institute 22: 1-21.
Ryckmans, G. (1934) Les noms propres sud-sémitiques, I, Leuven.
Sartre, M., (1985), Bostra. Des origines à l’Islam, Paris.
Sayar, M. H. (2000), Die Inschriften von Anazarbos und Umgebung. Teil I. Inschriften aus dem Stadt-
gebiet und der nächsten Umgebung der Stadt [IK 56], Bonn.
Schillinger-Häfele, U. (1977), “Vierter Nachtrag zu CIL XIII und zweiter Nachtrag zu Fr. Vollmer,
Inscriptiones Bavariae Romanae. Inschriften aus dem deutschen Anteil der germanischen
Provinzen und des Treverergebiets sowie Raetiens und Noricums”, BRGK 58: 452-603.
Schmitt, R. (1989), “Die mitteliranischen Sprachen im Überblick”, in Idem (ed.), Compendium
linguarum Iranicarum, Wiesbaden: 95-105.
Schuhmacher, L. (2003), “Mogontiacum. Garnison und Zivilsiedlung im Rahmen der Reichsge-
schichte”, Die Römer und ihr Erbe. Fortschritt durch Innovation und Integration, ed. M. J.
Klein, Mainz: 1-28.
Selzer, W. et al. (1988), Römische Steindenkmäler. Mainz in Römischer Zeit, Mainz.
Shaked, Sh. (1995), “Jewish Sasanian Sigillography”, Au carrefour des religions. Mélanges offerts à
Philippe Gignoux [Res Orientales VII], ed. R. Gyselen, Bures-sur-Yvette: Groupe pour
l’Étude de la Civilisation du Moyen-Orient: 239-256.
Solin, H. (1980), “Juden und Syrer im römischen Reich”, Die Sprachen im römischen Reich der Kai-
serzeit [BJ, Beiheft 40], Köln: 301-330.
Spaul, J. E. H. (1994), Ala.2 The Auxiliary Cavalry Units of the Pre-Diocletianic Imperial Roman Army,
Andover.
Speidel, M. (1975), “The Rise of Ethnic Units in the Roman Imperial Army” = Idem, Roman Army
Studies. Volume One, Amsterdam, 1984: 117-148.
Traina, G. (2005), “Des affranchis arméniens à Arretium?”, Esclavage antique et discriminations ocio-
culturelles, eds. V. I. Anastasiadis; P. N. Doukellis, Bern: 259-267 [see also Traina (2009),
“Tigranus e Bargathes: due armeni ad Arretium?”, Arezzo nell’Antichità, eds. G. Campo-
reale, G. Firpo, G. Arezzo: 217f.].
 Some Remarks on the Inscription of Maris, Casiti filius 73

―― (2011), Carrhes. 9 juin 53 av. J.-C. Anatomie d’une bataille, Paris.

―― (2012), “Letteratura classica e spazio geografico partico: alcune osservazioni”, Geographia
Antiqua, 19 (2010) [2012]: 109-112.
Welles, C. B. / R. O. Fink; J. F. Gilliam (eds.) (1959), The Excavations at Dura Europos, Final Report V,
I: The Parchments and Papyri, New Haven.
Winnett, F. V. (1957), Safaitic Inscriptions from Jordan, Toronto.
Ziethen, G. (1997), “Ex Oriente ad Rhenum – Orientalen im Römischen Mainz”, Mainzer Archäolo-
gische Zeitschrift 4: 111-186.
 The Military Campaigns of Shah Abbas I in Azerbaijan and
the Caucasus (1603-1618)

Kaveh Farrokh
University of British Columbia Continuing Studies

Abstract
This paper provides an overview of the Safavid military forces and reforms at the time of Shah Ab-
bas I (r.1587-1629), especially with the promotion of the new Ghulam units to counterbalance the
traditional Qizlibash forces which had bought the Safavids to power at the time of Shah Ismail (r.
1502-1524). Other significant military reforms were the introduction of firearm units such as the
Tofanchi (musketeers), Jazayerchis (bearers of larger and heavier muskets) and Toopchis (artillery-
men). The introduction of these reforms proved instrumental in Shah Abbas I’s successes in expel-
ling the Ottomans from Tabriz (1603) and Yerevan (1604), defeating the Ottoman counteroffensive
in Azerbaijan (1605) and the capture of Shamakhi (1606) and Ganja (1606). Large numbers of Ar-
menians, Kurds and Azeris had been displaced from their homelands by Shah Abbas I as a result of
his scorched earth tactics against invading Ottoman forces in 1606-1607. Shah Abbas’ military suc-
cesses led to the Ottoman-Safavid peace treaty of 1612 which affired all of the Iranian conquests
since the recapture of Tabriz. The Safavid army had to fight a series of battles in Georgia (1613-1623)
which led to a new Ottoman war (1616). The Safavid army defeated the Ottoman offensives in Ye-
revan and Ardabil (1616-1618) obliging the Ottomans to negotiate a new peace treaty which af-
firmed all of Shah Abbas’ conquests since 1603.

Keywords
Safavids, Ottomans, Battle of Chaldiran, Shah Abbas I, Qizilbash, Qoorchi, Gholam, Shahsevan, Ja-
zayerchi, Toopchi, Yesanchi, Jabadar-bashi, Azerbaijan, Tabriz, Caucasus, Yerevan, Ganja, Shama-
khi, Ardebil, Kakhetia, Meskhetia

MILITARY LEADERSHIP OF SHAH ABBAS

The military successes of Shah Abbas I (r.1587-1629) against the Ottoman Empire
was largely due to a reformed Safavid military machine. The Shah was in fact to
remain closely engaged in the supervision of the military affairs of the Safavid
state throughout his reign. In Shah Abbas’ view “…the king must live a soldier’s life
and always move alongside his troops to be able to overcome any difficulties and to
be successful at any task” (Ravandī 1973, vol. II: 398). Shah Abbas often joined his
 Koninklijke Brill NV, Leiden, 2015
76 Kaveh Farrokh

troops in battle and displayed concern for the morale and welfare of his troops
(ibid.: 409).1 In peacetime, the Shah would inspect Isfahan’s weapons factories on
a daily basis (Taheri 1985: 159).
SAFAVID ARMY AND REFORMS: AN OVERVIEW
The Safavid army wielded by Shah Abbas had changed considerably from its be-
ginnings in the late 15th and early 16th centuries. Many of the units and designa-
tions cited below had their origins in reforms implemented during Shah Abbas’
reign, with a number of other reforms (i.e. the Gholam units) having their origins
earlier during the reign of Shah Tahmasp (r. 1524-1576).
Military consequences of defeat at Chaldiran
The disastrous defeat of Shah Ismail (r. 1502-1524) at the battle of Chaldiran in
1514, forced the Safavid leadership to institute significant military reforms. Is-
mail’s successor, Tahmasp I ensured that these reforms began in earnest during
his reign. Tahmasp and the Safavid military realized that the primary factors for
Ottoman success at Chaldiran pertained to their superb organization, elite units
(notably the Yenicheri or Janissaries) and especially their firearms (muskets and
cannon).
Vincenzo D’Alessandri (Amiri 1970: 448-449) a European visitor to Iran in 1571
has provided valuable observations into the state of Iranian armies in the later
years of Tahmasp’s rule. First, D’Alessandri reported Iran having been adminis-
tratively divided into five regions (ibid.). The Safavid military had thus rational-
ized Iran’s defense against attacks by the Uzbeks (northeast), Ottomans (from
the west and northwest), and their Crimean Tatar allies (from the north) as well
as other potential enemies. Fifty Sultans were designated for the defense of these
zones, with each Sultan in command of 500-3000 professional troops. In case of
war, the Sultans would recruit additional men from their districts.2 With the
maximum number of troops assembled, the Sultans would march to a pre-desig-
nated assembly area to link up with the Shah and the Royal Guards.

1
This is shown in the case of the military leader Allah-Beg Qajar and a number of Qizilbash
warriors who had been severely wounded during an attack against one of Ganja’s Ottoman held
towers in 1606 (discussed later in this paper). Shah Abbas ordered the medical tents tending the
wounded to be erected next to his own (royal) tent. The Shah would then visit the wounded twice
a day. The Shah was accompanied by the medical staff of the Safavid army during these visits.
2
Mobilization of troops began at least two months before the onset of battles. Mobilization or-
ders would be sent by messengers arriving from the Royal Palace.
 The Military Campaigns of Shah Abbas I in Azerbaijan and the Caucasus 77

According to D’Alessandri, Iran’s standing professional cavalry corps stood at

60,000 troops (ibid.), which could be expanded in wartime. Iranian cavalry in
battle were described as resolute and highly disciplined. D’Alessandri recorded
that “Persians are tall and strong… commonly use swords, lances and guns on the
battlefield… Persian Musketeers use their muskets so adeptly…they will draw the
sword at times of necessity…muskets are slung to the back as to not interfere with
the usage of bows and swords…their horses are very well trained and they [the Ira-
nians] have no need to import horses…” (ibid.). D’Alessandri’s report suggests that
the Iranians had attained sufficient proficiency in the usage of firearms, before
the arrival of the Shirley mission to Iran at the time of Shah Abbas I. The Portu-
guese had formed the major pool of firearms instructors during the reign of Tah-
masp; they were also occupying Iran’s Persian Gulf islands in the Straits of Hor-
muz (Khorasani 2010: 20).
Rising Iranian capabilities in firearms alongside their proficiency in close-
quarter combat using traditional weapons enabled Iranian troops to successfully
partake in combat against the highly resilient and well-trained Ottoman troops.
The professional nature of Safavid troops by Shah Abbas’ time helps explain why
despite wielding superior numbers of high-quality troops and devastating fire-
power, the Ottoman Empire failed to overthrow the Safavids and conquer Iran.

Qoorchi
Tahmasp I had continued his late father’s project of forming an elite armed royal
guard known as the Qoorchi (Turkic plural Qoorchilar, singular Qoorchi; original
Mongolian term for “archer”). Being under the command of the Qoorchi-Bashi
(Bābāie 2005: 39), these troops were recruited from among Iran’s Turkmen war-
riors. The latter were often from the Qizilbash clans who had assisted Shah Is-
mail in founding the Safavid dynasty. The Safavid royal house directly paid and
armed the Qoorchi troops, who were subject to the direct orders of the king. As a
result, the office of the Qoorchi-Bashi become one of the most important posts
in the Safavid state (Sīaqī 1989: 7; Abbasi 1956, vol. VIII: 239). As professional
royal guards for the Safavid dynasty, the Qoorchis were distinct from the regular
Qizilbash units of the Safavid army.3 By the 1570s, Qoorchi forces stood at ap-
proximately 4500 to 5000 men,4 this then tripling to 12,000 men by the end of

3
Nevertheless, the Qoorchi would often don the distinctive 12-gore Qizilbash hat in battle.
4
Šokrī (1971: 609) cites 4,500 with Matofī (1999: 656) citing 5000.
78 Kaveh Farrokh

Shah Abbas I’s reign in 1629 (Šafā 1969: 343-348; Šokrī 1971: 609).5 Qoorchi units
(both cavalry and infantry) were armed with a combination of muskets and tra-
ditional weapons (i.e. curved Shamshir swords, axes, etc.).
Gholams
By the second half of Tahmasp`s reign, the foundations of a new elite military
unit were being established. The Safavids were bringing Caucasian Christians
(Georgians, Armenians) and Circassians from the Caucasus to form the Ghollar
(Turkic plural for Gholam or “slaves/servants”). Like the Ottoman practice of
bringing Christian boys from Europe to be raised as Janissaries, the Safavids im-
ported the Caucasians into Iran from an early age to be raised and trained as
Gholam troops.
Like the Qoorchis, the Gholams were highly trained in the use of muskets
(Matofī 1999: 659) and traditional weapons (Shamshir swords, spears, axes,
maces, and archery) (Šafā 1969: 343-348). Gholam units (like the Qoorchis) could
be infantry or cavalry, totaling 15,000 men at the time of Shah Abbas I (Bābāie
2005: 39) and 18,000 men at the time of Shah Abbas II (1642-1666) (Šīrānī 1957).6
The office of Ghollar-Aghassi (Turkic: General of the Gholams) became second
only to that of the Qoorchi-Bashi (Sīaqī 1989: 8).
The Caucasians who arrived into Iran’s Gholam corps were very soon ab-
sorbed into Iranian society. This was due to the long-standing cultural and his-
torical links that have existed between Iran and the Caucasus as far back as the
Sassanian era (224-651 CE) (Whittow 1996: 203-204) and even earlier into the
Bronze Age. Armenia’s Bagrationi clan for example had connections with Iran
(Farrokh 2007: 54, 60, 277-278, 281) with researchers reporting a “cultural synthe-
sis” (Farmanfarmain 2009: 24) between Iran and the Caucasus (Rayfield 2000: 58;
Frye 1977: 19, 145; Gvakharia 1995). Persian literary traditions exerted a mighty
influence on the Caucasus right up to the Russian invasions in the early 19th cen-
tury CE (Chardin 1983: 268, 290; Gvakharia 1995: 241; Farmanfarmaian 2009: 20-
23; see also Andronikashvili 1966, Lang 1966). This “cultural synthesis” occurred
irrespective of the Georgians’ and Armenians’ Christian faith, which may partly
explain why the Caucasians became such effective troops against the Ottomans.

5
By the time of Shah Abbas II in 1642-1666, Jean Baptitse Tavernier (1605-1689), records a total
of 22,000 Qoorchis in military service. A comprehensive description of the armies of Shah Abbas II
by Tavenier is provided by Šīrānī (1957: 256, 278, 548-581, 594, 599-600).
6
Note that the total number of Gholam troops remained consistent at 15-18,000 men during
the reign of Shah Safi II/Suleiman (1666-1694) (Matofī 1999: 672).
 The Military Campaigns of Shah Abbas I in Azerbaijan and the Caucasus 79

Georgian influence became significant, especially from the 16th century. Many
prominent Safavids hailed from a Georgian or part-Georgian background with
many being settled in Isfahan (Roemer 1991: 272) and northern Iran (consult Na-
sidze, et. al. 2006).
Decline of Qizilbash and rise of Gholams
The Qizilbash had been instrumental in bringing Shah Ismail to power and
played a major military role in Safavid armies. They were formidable warriors
but highly conservative in their military doctrine. They tended to favor tradi-
tional weaponry and eschewed modern firearms. Their organization was based
on the Turcomen manner of tribal units, with their provincial military contin-
gents being led by their respective khans. The Yuz-Bashi (Turkic: commander/
leader of one Hundred) commanded one hundred men, much like the Tofangchi
musketeers. The Qizilbash Khans exerted a powerful influence in the military
leadership, government and even the royal succession. Shah Ismail did elevate a
number of local Iranian nobles to positions of upper leadership after the Battle
of Chaldiran, but this did little to diminish the continuing Qizilbash influence in
military and civilian affairs.
Tahmasp I (Ismail’s son and successor) worked to reduce Qizilbash influence
by recruiting a large new class of Gholams into the Safavid military and civilian
administration. Gholam contingents had greatly expanded in scope and im-
portance by the time of Shah Abbas I. These never came to threaten or dispro-
portionately influence the political processes of the Safavid state. Nevertheless,
the Qizilbash were still able to mobilize up to a maximum of 60,000 troops (al-
most all cavalry) at the beginning of Shah Abbas’ reign in 1587 (Matofī 1999: 658).
This changed with the implementation of Shah Abbas’ reforms. A numerical
comparison of the numbers of Qizilbash chiefs in 1578 and 1629 provides a statis-
tical measure of their decline. A listing of Safavid military commanders in 1576
(last year of Tahmasp I’s rule), reports a total of 114 Amirs (military command-
ers), almost all Qizilbash. This changed significantly by 1629 (last year of Shah
Abbas I’s rule) where 25 out of 90 Amirs listed were Qizilbash (Pigulevskaya 1975:
525). By the end of Shah Abbas I’s reign, Qizlibash warriors had been halved to
30,000 troops.
Changes did take place in Qizilbash armaments that had traditionally been
swords, maces, lances and archery equipment. A number of Qizilbash adopted
light metallic helmets in battle (Falsafī 1965, vol. I: 175-178) and (by the time of
Shah Abbas’ reforms) muskets. Top level Qizilbash commanders still refused to
80 Kaveh Farrokh

adopt firearms in favor of traditional fighting methods (Šafā 1969: 343-348).

Chardin has noted of the Qizilbash derision of firearms, viewing these as the an-
tithesis of manliness and bravery on the battlefield (Falsafī 1965, vol. I: 175-178).
The reduction of Qizilbash influence removed a primary obstacle for the full-
scale development of firearm units in the Safavid military.

Regular, Provincial and Tribal Troops and Cavalry

By the time of Shah Abbas there were 20,000 regular cavalry (intended to replace
the Qizilbash) and 12,000 regular infantry armed and paid by the state and under
the direct command of the Shah (Matofī 1999: 658). The Safavid military was able
to raise very large numbers of cavalry from the country’s towns, tribal regions
and provincial districts to augment its professional cavalry. One of the European
translators for the Shirley brothers in Iran later wrote in Rome that Shah Abbas I
could raise a maximum force of 100,000 cavalry (Sīaqī 1989: 51).
The cost of maintaining tribal and provincial infantry and cavalry units
(which were distinct from regular Safavid standing forces) were often incurred
by the commanding Khans. By later Safavid times (i.e. Safi II/Suleiman II) Iran
was (nominally) able to call upon an impressive array of provincial contingents
and local tribal forces from Kandahar (12,000), Khorasan (20,000), Gilan and Ma-
zandaran (15,000), Darband and Shirvan in the Caucasus (12,000), Azerbaijan
(20,000), Yerevan in Armenia (12,000), Luristan (12,000), Susa in the southwest
(15,000), south of Iran and Kerman (12,000) (Šīrānī 1957: 43-48, 106-107, 137-144,
185-186). European accounts describe regular units deploying a variety of weap-
ons (i.e. Shamshir curved swords, spears, archery equipment, firearms, etc.) with
effectiveness (Amiri 1970: 217-218). Cavalry troops were described as capable of
quick dispersal and coalescing abilities at points of their own choosing during
battle (ibid.). Qajar tribes in particular had militarily distinguished themselves
against the Ottomans throughout the Safavid dynasty.7
The Shahsevan or “Friends of the Shah” was an important all-volunteer na-
tional force composed primarily of Azeris, Kurds, Turkmens and other tribal-
provincial elements. Iranian historians trace their origins to the latter days of
Shah Abbas I, numbering up to 120,000 men at that time.8 Of these 45,000 were
permanent troops serving within elite contingents (Qoorchi and Gholam) and

7
Tarikh-e Rozat ol Safa, (Original in 1851, Reprinted 1960-1961), Volume VIII, pp.212.
8
Consult discussions by Rajab-Nīyā (1955: 50ff.). See also synopsis by Ravandī (1973, vol. II:
398).
 The Military Campaigns of Shah Abbas I in Azerbaijan and the Caucasus 81

firearm units (Tofangchis and Toopchis). They were also instrumental in reduc-
ing the influence of the Qizilbash (Bābāie 2005: 28). The Shahsevan have often
served as a bulwark of Iran’s defense against foreign invasion through Azerbai-
jan, as seen for example in the 19th century during the Russian invasions of the
Caucasus and in the twentieth century when they cooperated with the Iranian
army in ejecting pro-Soviet separatist militias from Iran’s northwest (Farrokh
2011: 290-291).

Tofangchi
The Tofangchilar (Turkic plural: gunmen/musketeers, singular Tofangchi) were
the formal firearm units of the Safavid military. Tofangchi units of one hundred
troops were led by the Yuz-Bashi (Turkic: commander/leader of one Hundred).
The Tofangchis were an infantry force using horses for long distance marches
and campaigns (Falsafī 1965: 177-178). Paid on a regular salary basis throughout
the year, these units were expected to report to duty at times of war.
Italian traveler Pietro Della Valle (1586-1652) has provided a vivid description
of the Tofanghchis at the time of Shah Abbas I.9 Della Valle noted that the
Tofangchis were mainly recruited from the physically resilient Iranian-speaking
peasants (Mīrxand/Hedāyat 1960-1961, vol. IX: 359, 438), in accordance with the
advice of the Shirley brothers. Valle was especially impressed by the Tofangchis’
ability to set up or leave camp at great speed without detection by the enemy.
The Tofangchis were noted for maintaining their weapons and uniforms in a
high state of cleanliness and preparedness (see Šīrānī 1957). These units, ac-
cording to Valle, were among Shah Abbas’ most favorite units with the best
Tofangchis hailing from Iran’s northern Mazandaran province (Mīrxand/Hedā-
yat 1960-1961, vol. IX: 359, 438). Interestingly, two centuries later, Agha Moham-
mad Khan (1742-1797), the first Qajar monarch, considered his Mazandarani
Tofangchis as being among his most trusted and capable warriors.
Reporting during the reign of Shah Abbas II Jean Baptitse Tavernier (1605-
1689) described the Tofangchis as having a high level accuracy and agility.10
Tavernier reports one of their exercises involving a line of ten men aiming their
muskets towards a line of ten spears thrust into the ground, each with an apple
placed at its top. Every apple would be accurately hit by a single shot by the ten

9
These have been compiled by Šafā (1969: 343-348).
10
The entire description of the armies of Shah Abbas II by Tavenier is provided by Šīrānī (1957:
256, 278, 548-581, 594, 599-600).
82 Kaveh Farrokh

corresponding Tofangchis firing at a distance of one hundred paces (see Šīrānī

1957). Numerical estimates of Tofangchis at the time of Shah Abbas I vary from
20,000, 50,000 and 60,000.11 By the time of Shah Abbas II a total of 40-50,000
Tofangchis are reported in regular service (see ibid.). Like the Safavids before
him, Nader Shah Afshar would also recruit his Tofangchis from Iran’s rural peas-
ant base.
Jazayerchi
Shah Abbas II was the first to introduce the Jazayerchi units. A new elite troop
numbering at just 600 troops (Abbasi 1956, vol. VIII: 214-216) the Jazayerchis were
selected from among the most capable and resilient warriors of the Iranian mili-
tary. These were later increased to 2000 men (Roemer 1991: 291). The primary
weapon of the Jazayerchi was the Jazayer that was a larger and heavier hand-
held form of musket (Matofī 1999: 667). Due to its size and weight, the Jazayer
had to be fired by placing it upon a tripod. The Jazayerchis were also armed with
Shamshir swords and daggers for hand to hand combat (Sīaqī 1989: 57).
Toopchi
Military personnel manning cannon units were the Toopchis with the Toopchi-
Bashi (lit. cannon master/leader) in command (idem 1985: 14). Amiri (Amiri 1970:
448-449) and Monshi (Monšī 2003, vol. II: 653) provide a summary of the appro-
priation of cannon into the Safavid army. The first of these came from battlefield
captures of Ottoman and later of Portuguese cannon. The flight of the Ottoman
Prince Bayezid into Iran (who bought numbers of cannon with him) may also be
viewed in this context. The next source of cannon came from the factories es-
tablished with the assistance of the Shirley brothers during the reign of Shah Ab-
bas I. There were also a number of mobile workshops capable of producing can-
non at military camps during campaigns. Exact statistics of the size of the Sa-
favid artillery corps cannot be provided with certainty, which most likely stood
at its strongest level at the time of Shah Abbas I. It is estimated that the Safavids
probably fielded a maximum of 500 cannon (Bābāie 2005: 40) and up to 12,000
artillerymen (Toopchis) (Khorasani 2010: 21).
Safavid cannon were often used for destroying enemy towers and walls dur-
ing sieges (Matofī 1999: 734). For battlefield protection against enemy firearms

11
Valle reports these at 20,000 with the Tazakor o Molook (compiled by Sīaqī) which cites “the
Europeans” as reporting 60,000 and a European translator of the Shirley Brothers reporting 50,000
(1989: 51).
 The Military Campaigns of Shah Abbas I in Azerbaijan and the Caucasus 83

during sieges, Toopchis often laid sandbags around their artillery nests (ibid.:
702). The capture of cities and fortresses usually entailed the practice of cutting
off the besieged locale from all sources of supplies. Western observers note of the
Iranians’ lack of knowledge in their use of cannon in siege warfare (consult Mat-
thee 1999: 394-395). Despite this liability, the Iranians proved adept in engineer-
ing operations and the digging of tunnels under enemy walls (Abbasi 1956, vol.
VIII: 219-222). Excepting reforms during the reign of Shah Abbas I, the Iranians
do not appear to have integrated cannon into their regular battlefield order
(Matthee 1999: 394-396).
The office of the Toopchi-Bashi, which was of considerable importance during
Shah Abbas I’s era, witnessed a steady decline by the middle of the 17th century.
The reports of Tavernier cite the artillery’s low level of efficiency and prepared-
ness, with claims even being made of numbers of cannon buried in the grounds
of Isfahan (consult Šīrānī 1957). The Toopchi-Bashi office was abolished (albeit
temporarily) in 1655 (during Shah Abbas II’s reign) due to a combination of few
operational cannon, financial difficulties, and a relatively peaceful state of rela-
tions with Iran’s immediate neighbors (Matthee 1999: 395). Negative reports of
Iranian artillery performance continued into the reign of Shah Safi II/Suleiman
(Matofī 1999: 672). By this time the Toopchi-Bashi office had been reinstated.
This was in nominal command of 4000 artillerymen (Toopchis)12 but in practice
that force existed mostly on paper. This was due to the consistent neglect af-
forded to the artillery arm after the death of Shah Abbas I. Suleiman had con-
templated reforms to the artillery arm at first, but this was abandoned (Khoras-
ani 2010: 21-22).

Yesanchi
The Yesanchi were auxiliary units responsible for guarding important buildings
and installations. These also cleared the streets for the passage of the Shah and
his entourage. They were also involved in elaborate ceremonies for newly arrived
dignitaries from important states. One such example occurred when new am-
bassadors from the Ottoman Empire, Russia and Moghul India were greeted
with a highly disciplined honor guard of up to 60,000 armed troops lined up on
two sides of a road from the Shah’s palace in Isfahan to the village of Dolatabad

12
For description of armies of Shah Safi II/Suleiman see Šīrānī 1957: 43-48, 106-107, 137-144, 185-
186.
84 Kaveh Farrokh

(Ravandī 1973, vol. II: 419). The Yesanchi played a major role in ceremonies and
celebrations following battlefield victories (Šokrī 1971: 156).
Flags and banners
The Alamdar (lit. Banner/flag bearer) carried flags in battle with the Alamdar-
Bashi (lit. Chief/head banner/flag bearer) having the privilege of bearing the
largest and most prestigious banners. Iranian flags and banners generally in-
cluded a number of motifs including Qoranic inscriptions, the double-headed
Zulfaqar sword of Imam Ali, the names of past fallen warriors, and the Lion and
Sun emblem (Taqzelī 1346 vol. VIII: 222). This is evidenced in a print in the Ga-
zette de France depicting the arrival of an Iranian delegation at Versailles ac-
companied by the Alamdar-Bashi in August 1715. The Alamdar-Bashi carried the
Iranian flag bearing the Lion and Sun emblem whose motifs may be traced to
ancient (pre-Islamic) Iran’s past (Nayernuri 1965: 78; Khorasani 2006: 320).
Jabadar-Bashi
The Jabadar-Bashi ensured that (a) arsenals throughout the country were well-
stocked with weapons (traditional weapons, firearms and gunpowder) and (b)
was responsible for the supervision of armament production in factories (Falsafī
1965, vol. II: 412). In addition to factories, the army also had mobile field produc-
tion and repair workshops to ensure that troops were consistently equipped with
armaments during campaigns. If defeat was imminent or if a city was about to be
captured, the army would destroy any weapons production facilities that could
be captured by the enemy.
Military Music
The role of music in the ancient pre and post-Islamic Iranian armies has been
discussed by Mašoun (2001: 29-75, 77-737). Safavid armies used a variety of
drums, percussion instruments, and trumpets. Music was intended to build up
the martial spirit of troops and to demoralize the enemy (Matofī 1999: 708).
Mašoun's historical analysis notes that music and poetry were vital for the over-
all morale of the peoples of the Safavid realm.13
SHAH ABBAS’ BATTLES AGAINST THE OTTOMANS IN AZERBAIJAN AND THE CAUCASUS
(1603-1626)
The restoration of peace to Iran’s eastern marches allowed Shah Abbas and the
Safavids to cast their military attention toward those territories in the Caucasus

13
Consult Mašoun 2001: 280-338 regarding role of music during the Safavid era.
 The Military Campaigns of Shah Abbas I in Azerbaijan and the Caucasus 85

and Iran’s northwest, west, and southwest that had been captured by the Otto-
mans (Savory 1980: 85; Matofī 1999: 646). Shah Abbas first proposed an alliance
with the Russians against the Ottomans which was not accepted (Pigulevskaya
1975: 51). Russian action became confined to small-scale raids along Daghestan’s
borders in the northern Caucasus (Matofī 1999: 646). Shah Abbas thus had to
face the mighty Ottoman military machine by himself but by this time a number
of Safavid military reforms had taken effect. Shah Abbas’ advisors had cautioned
him on Ottoman numerical superiority over Iran’s military (Savory 1980: 85).
A major factor inciting the Safavids’ to action was the arrival of a number of
Khans from Iran’s occupied northwest regions (especially Azerbaijan) asking
Shah Abbas to attack the Ottomans. Notable was the role of Ghazi Beg, the
leader of the Kurds of Maku who complained of the excesses of Tabriz’s Ottoman
governor and his harsh maltreatment of local citizens (Bayānī 1974: 149). A major
factor favoring the Safavids was the rule of the Ottoman governor of Baghdad,
Uzun Ahmad who refused to send supplies to the Ottoman garrison at Nahavand
in modern-day Iraq. Nahavand’s local Ottoman garrison then resorted to pillag-
ing local civilians to secure supplies (Matofī 1999: 647). These actions led to anti-
Ottoman discontent prompting many locals to request Safavid military assis-
tance. Shah Abbas’ response was to storm and capture Nahavand (Savory 1980:
85).

Liberation of Tabriz (October 21, 1603)

The newly reformed Iranian army deployed out of Isfahan on September 14, 1603
towards Ottoman-occupied Azerbaijan in northwest Iran. Shah Abbas deceived
the Ottomans by “officially” declaring that he and his troops were leaving Isfahan
to Iran’s northern Mazandaran’s forests to engage in a hunting expedition (Na-
havandi/Bomati 1998: 148-149).
In practice, Shah Abbas had ordered Zolfaqar Khan and Amir Goneh Beg
Qajar (governors of Ardabil and Qazvin respectively), to quickly link up their
forces with the Gholam corps and Qoorchian royal guards. The main army led by
Shah Abbas reached Tabriz in 11 days. The Ottoman Turkish garrison was una-
ware of their arrival. Shah Abbas’ tactics have been described by Chardan (Ab-
basi 1956 vol. II: 419).14 The first step was the selection of the best troops to en-

14
These accounts (and those of Matofī 1999: 647) are at variance to those described by Savory
(1980: 85-86) who makes no mention of an advance guard silencing the Ottoman sentries or the ar-
rival of 500 troops masquerading as merchants.
86 Kaveh Farrokh

gage in what may be characterized as a “commando” operation. This succeeded

in quietly infiltrating Tabriz to slay a number of Ottoman sentries. This was
achieved with considerable stealth as the main Ottoman garrison remained un-
aware of the infiltration of Tabriz by Safavid troops. The next stage was the arri-
val to Tabriz of a group of 500 Iranian troops attired in civilian clothing posing as
merchants15 Their mission was to distract Ottoman troops from the main Safavid
army now assembling just outside of Tabriz. With the Iranian “Trojan horse” (i.e.
the “commandos” and the “merchants”) in place within Tabriz, Shah Abbas
launched his primary assault. Thanks to the earlier elimination of Ottoman sen-
tries at Tabriz’s critical ingress points, 6000 Iranian troops stormed into the city.
This was coordinated with the 500 “merchant” troops already in Tabriz who had
also engaged in fighting against the local Ottoman garrison. After twenty days of
close quarter combat in Tabriz’s streets the Ottomans surrendered (Matofī 1999:
647) on October 21, 1603 (Clodfelter 2008: 61; Dupuy/Dupuy 1977: 586).
The liberation of Tabriz was of major importance. Tabriz is Azerbaijan’s chief
city and the venue where Shah Ismail had been crowned. This victory was a
major morale booster for the Iranian army, which had suffered many defeats at
Ottoman hands. The Ottomans paid dearly for their defeat at Tabriz. In a letter
written by Shah Abbas to the Moghul emperor, Jalal e Din Mohammad Akbar
(1542-1605), he noted that the Ottomans in Tabriz had “…200 cannon, 5000 mus-
keteers…supplies lasting for ten years and much equipment for the holding of for-
tresses…” (Falsafī 1965, vol. IV: 22-23). Many ex-Ottoman troops joined the Sa-
favid army. This may have been partly due to Safavid troops being paid double
the salary and benefits of their Ottoman counterparts at the time of Shah Abbas
(Savory 1980: 86).
Local Azerbaijanis who had assisted the Iranian army celebrated by donning
their 12-gore Safavid hats, signaling their loyalty to the Iranian throne and Shi-
ism. Tabriz’s citizens also took reprisals against a number of captured Ottoman
troops who had committed offenses against civilians during the occupation
(ibid.). Many buildings and houses in the city had been severely damaged during
the Ottoman occupation (Nahavandi/Bomati 1998: 149-150), with Ottoman
troops having taken at will many of Tabriz’s womenfolk (Savory 1980: 86).

15
These had claimed to the authorities that they were awaiting the arrival of their merchandise
into Tabriz in a number of days.
 The Military Campaigns of Shah Abbas I in Azerbaijan and the Caucasus 87

Ottoman defeat at Yerevan (June 1604)

The liberation of Tabriz allowed the Safavid army to recapture Nakhchevan from
Ottoman troops (Roemer 1991: 267). Shah Abbas had also sent messages to the
remaining Ottoman garrisons in Azerbaijan to surrender (Matofī 1999: 647-648).
Many of these did evacuate but instead of marching west into the Ottoman Em-
pire, these coalesced (12,000 in all) (Savory 1980: 86) in the region of Yerevan (in
Armenia) north of the Araxes River (Matofī 1999: 648). The Ottomans then began
to organize their defense around three of that area’s fortresses: Koozchi, Yerevan
city and Eshiq. Koozchi was the first to fall with Safavid forces entering through
its main gates. Safavid forces had taken careful precautions to not expose them-
selves to the deadly musket and cannon fire of the Ottomans during this battle
and the following battle of Yerevan City.16 The occupying Ottoman garrison at
Yerevan City proved most obstinate by holding out against besieging Safavid
forces for over six months into the summer of 1604. The ground was too frozen in
winter to allow the Safavids to dig trenches (Savory 1980: 86). As the winter
passed, Safavid sappers began to dig under the city’s walls. A number of trenches
were also constructed allowing the Safavids to launch a number of attacks which
the Ottomans repelled.
With more supplies having arrived, Safavid forces were able to launch a new
and more powerful assault using cannon and muskets. They also burrowed un-
der a number of walls to place explosives under these. The explosions destroyed
a number of Yerevan’s walls, allowing Iranian assault parties to pour into the for-
tress (Matofī 1999: 702). This forced the Ottomans into close-quarter fighting
with the Iranians. After losing 2000 of their troops in hand to hand combat in a
few hours, the Ottoman garrison was forced to surrender (ibid.: 648). The fall of
Yerevan in June 1604 (Burton 1997: 119) was followed by Eshiq which surrendered
without battle. The fall of Eshiq was a major blow against the Ottomans as this
had been a major center of their operations against the Iranians in the Caucasus.
Among the most effective troops during these campaigns had been the musket-
eers of Tabriz (Matofī 1999: 648).

16
Monšī 2003 (Iskander Beg Torkaman, 1628), vol. II: 637. It is not clear how this was imple-
mented, but there is mention of using local trees and foliage for cover.
88 Kaveh Farrokh

Pursuit of Ottoman Forces (1604-1605)

Concerned with the prospect of an Ottoman counterattack, Shah Abbas and his
top commanders pursued the fleeing Ottoman forces with 20,000 troops.17 The
commanders of these operations were Allahverdi Khan (Beglerbegi and Gover-
nor of Fars province), Zolfaqhar Khan Gharamanlu (Beglerbegi of Azerbaijan),
Ganjali Khan (Governor of Kerman), and Hassan Khan (Governor of Alishokr).
Safavid forces finally caught up with a dangerous Ottoman force led by Cheqhal
Oghlu at the Battle of Ati (Matofī 1999: 648). Once again the Ottomans were de-
cisively defeated with a large haul of war materiel, supplies, and important Ot-
toman documents being captured from the abandoned Ottoman camps (Matofī
1999: 648-649), indicative that the Ottomans had retreated in haste.
Defeat of Ottoman counteroffensive in Azerbaijan (November 6, 1605)
Ottoman Sultan Ahmed organized a major counteroffensive to retake Tabriz and
crush the Safavid army. A very large Ottoman army of 100,000 troops (Clodfelter
2008: 61) reached Sufiyan (near Tabriz) to confront Shah Abbas leading 62,000
men (Dupuy/Dupuy 1977: 586) on November 6, 1605 (Clodfelter 2008: 61) or No-
vember 9, 1605 (Tucker 2010: 560). The Turks resorted to their traditional battle
tactics against the Iranians. This was to first engage in a cavalry duel to draw the
Iranian cavalry towards the Ottoman lines. The aim would be to bring them
within range of the powerful Ottoman cannon and musketeers (Dupuy/Dupuy
1977: 586). Ottoman tactics failed to draw the Iranian cavalry into the trap. In-
stead, Shah Abbas ordered a detachment of his troops to attack the Ottomans
from the rear. Shah Abbas’s tactic was intended to appear as the primary Iranian
attack. The Ottomans then swung their entire force to their rear. Shah Abbas
then struck with his primary force of artillery, infantry and cavalry, shattering
the Ottoman army, which suffered 20,000 men killed (Clodfelter 2008: 61). This
defeat finalized the Ottoman expulsion out of Azerbaijan and allowed the Ira-
nian army to focus on the ejection of Ottoman forces still ensconced in the Cau-
casus.
Ottomans ejected from Ganja (1606)
With Azerbaijan secured, Shah Abbas focused his efforts on Ganja, an important
and strategic Caucasian fortress which had fallen to the Ottomans. Shah Abbas
arrived at Ganja to make heavy use of trenches, siege works and cannon fire to

17
The operations are described by Monšī 2003 (Iskander Beg Torkaman, 1628), vol. II: 639-645,
654, 681, 683, 687, 697.
 The Military Campaigns of Shah Abbas I in Azerbaijan and the Caucasus 89

subdue its Ottoman garrison. Iranian engineers succeeded in crossing Ganja’s

surrounding ditch to then dig out a large amount of soil from under one of its
defensive walls. A large amount of chopped wood was stuffed into the excavated
area and set on fire. This decisively damaged a section of Ganja’s walls, prompt-
ing the Ottoman garrison to finally surrender.
When Safavid forces entered the city they obliged its citizens to tabulate their
names on a royal registry. The aim of this was to identify collaborators who had
cooperated with the Ottomans during their occupation. Up to 2500 Iranian
“fifth-column” collaborators were soon identified and put to death (Monšī 2003
(Iskander Beg Torkaman, 1628), vol. II: 709-715).
Capture of Shamakhi (June 27, 1606)
The fall of Ganja posed a mortal threat to the Ottoman garrison at Shamakhi,
capital of the Shirvan khanate. The Ottomans responded by destroying a key
bridge over the Kura River in the Caucasus, but this failed to stem the arrival of
Iranian forces. Safavid engineers built a pontoon bridge over the Kura River by
attaching boats together with planks, a tactic used by Iranian armies since
Achaemenid times (Farrokh 2007: 68-69). Other points of the Kura were made
traversable by sinking many pieces of lumber into the river. The main body of
Iranian troops finally crossed the Kura in late 1605 or early 1606, but lost many
horses and supplies. This forced some Safavid troops to halt for the arrival of new
supplies before proceeding to rejoin the main Safavid army marching ahead of
them to Shamakhi.18 Supply shortages also delayed the march of the main army
but this was ameliorated with the arrival of fresh supplies. The Safavids finally
arrived at the end of the winter in early 1606 to besiege Shamakhi. This was a
tough and obstinate fortress with very well constructed walls and towers with a
deep surrounding ditch.19 The only way that ditch could be crossed was by
means of a mechanical bridge that could be retracted into Shamakhi castle
during sieges.
The assault forces at Shamakhi were divided in two: one to its north led by
Allahverdi Khan and the other to its west commanded by Shah Abbas.20 Both
flanks were supported by a number of important units such as the Azerbaijani
and Kermani detachments. The eastern flank of Shamakhi could not be utilized

18
These troops had been obliged to halt in a region named Karka to replenish their stocks be-
fore proceeding to Shamakhi.
19
Matofī (1999: 649) does not report the specific dimensions of the Shamakhi ditch.
20
A complete account of the fighting at Shamakhi has been provided by Matofī (1999: 649-651).
90 Kaveh Farrokh

for the siege. This was due to heavy rains which had turned the ground into im-
passible mud. This meant that heavy equipment such as siege engines, cannon,
horses and pack animals would quickly bog down and sink. The southern flank
of Shamakhi was also left unmanned. The decision to leave the eastern and
southern flanks proved especially hazardous as the Ottomans made sorties from
those same sectors to launch dangerous raids into Safavid positions. Despite this
advantage, the Ottomans failed to disrupt Safavid operations (i.e. engineering
works and the build-up of supplies), which continued despite heavy rains.
The main Safavid strategy was to build a bridge across the ditch surrounding
Shamakhi. To protect his engineers against the muskets and archery of Shama-
khi’s defenders, Shah Abbas built a series of “mini-forts” manned by a combina-
tion of musketeers, archers and infantry. Each of these was surrounded by a
ditch situated at those sites where engineering crews were at work. The mini-
forts were also a deterrent against Ottoman hit and run attacks.
Crucial was Hossein Qoli Beg Qajar’s success in bringing forward heavy siege
cannon from Ganja.21 The cannon were distributed to Allahverdi Khan’s north-
ern flank and to Shah Abbas’ western flank. Both flanks had already prepared
their artillery pits for the arriving cannon. Soon more troops led by Pir Bodaqh
Khan arrived from Tabriz.
Shah Abbas was finally ready to commence his main effort to capture
Shamakhi. His heavy stone-firing cannon pounded the fortress on June 27, 1606,
destroying a number of its towers and structures. Some Iranian engineers man-
aged to land near the fortress and began digging underneath its walls. At this
juncture, a number of troops led by Qaracheqai Beg from Shah Abbas’ western
flank, stormed and captured one of Shamakhi’s towers. Soon Zolfaqhar Khan’s
Azerbaijani troops forced a number of Ottoman troops defending another tower
to surrender. The turning point came when 150 Safavid soldiers broke into the
fortress and signaled Qaracheqai Beg for assistance. As Qaracheqai Beg sent his
troops, Shah Abbas ordered the entire army to converge for the final attack into
Shamakhi. The Ottomans lost 2000-3000 men in the first round of close quarter
fighting inside the fortress. Safavid troops then succeeded in opening Shamakhi’s
gates, allowing the main Safavid force to stream into Shamakhi. The conquest of
Shamakhi secured the battle for the Shirvan khanate for the Safavids.

21
These fired cannon balls made of heavy stone which had been transported by specially de-
signed wagons.
 The Military Campaigns of Shah Abbas I in Azerbaijan and the Caucasus 91

Scorched-Earth tactics (1606-1607)

Despite the recent Iranian victories, the Ottomans soon recuperated from their
losses to launch more counterattacks into Iran’s northwest and the Caucasus.
Shah Abbas embarked on an expanded scorched-earth program in anticipation
of the Ottoman arrival. If the Ottomans chose to invade with very large forces
they then faced the prospect of being starved out of supplies inside Iran.22 They
would also be subjected to harassment raids by the Safavid cavalry (Abbasi 1956,
vol. VIII: 219-222). If the Ottomans chose to invade with smaller armies instead,
they would then risk being outnumbered by hostile Safavid forces.
Shah Abbas I ordered the entire region between Tabriz in Iran’s Azerbaijan
province and Erzerum in eastern Anatolia to be systematically depopulated of
up to 300,000 people (Clodfelter 2002: 60) these being overwhelmingly Armeni-
ans, Kurds and Azeris.23 The uprooted peoples were mostly re-settled in Iran’s
interior (Izady 1992: 102). Many Kurds were sent to Khorasan with three thou-
sand Armenians from Julfa in Azerbaijan being relocated to Isfahan (Roemer
1991: 271). Chardin has reported on the destruction of farms and buildings and
the poisoning of wells along possible Ottoman invasion routes (Abbasi 1956 vol.
II: 293-294). Shah Abbas also engaged in a number of military operations in the
north of Iran, notably Gilan (1592), Lahijan (1595), and Mazandaran (1596-1598),
followed by Lar (1601-1602) (Roemer 1991: 269). This was part of his overall policy
of bringing semi-autonomous provinces under the centralized power of the Sa-
favid state.
PEACE TREATY WITH THE OTTOMAN EMPIRE (1612)
The Ottoman army’s repeated defeats obliged the Ottoman Empire to arrive at a
negotiated settlement with Shah Abbas, culminating in the peace treaty of 1612.
This forced the Ottomans to renounce their previous Iranian conquests and to
recognize the validity of the old Irano-Ottoman frontiers (ibid.: 267). In diplo-
matic gesture on the Ottoman’s behalf, Shah Abbas was to send an annual trib-
ute of 200 camels laden with silk to the Ottoman Sultan (Dupuy/Dupuy 1977:
586).

22
Whole cities, towns, villages and farms had their populations relocated away from the Otto-
man path of advance. All buildings and food supplies were destroyed with water wells poisoned.
This ensured that invading Ottoman armies would be prevented from securing supplies from re-
cently conquered territories. This often wreaked havoc with Ottoman logistics, especially when
they had to supply their armies inside hostile Iranian territory.
23
Clodfelter (2002: 60) notes that half of the Armenians died on the trek towards Isfahan.
92 Kaveh Farrokh

Battles in Georgia (1613-1623)

Shah Abbas was obliged to engage in a number of battles in Georgia in 1614-1615
to bring the area under Iranian suzerainty (Dupuy/Dupuy 1977: 586). Iranian
sources report the deaths of 70,000 Georgians during the course of the fighting
and the rounding up of 130,000 of them as prisoners (Pārīzī 1999: 224). The Ot-
tomans, who considered Georgia within their zone of influence, mobilized for
another war against Iran.
Resistance in Georgia certainly continued into the 1620s, as Shah Abbas an-
nexed Kakhetia (1620-1621) and Meskhetia (1622-1624) (Roemer 1991: 269). A rare
glimpse into the operations in Meskheti has been provided by the relatively re-
cent discovery of a gilded bronze winged lion,24 dated to 1622-1623 by the Geor-
gian Institute of Manuscripts of the Georgian Academy of Sciences.25 This lion
was part of a secret directive sent by Shah Abbas26 to the local Safavid com-
mander Khorji Khan who was encamped with his army on the valley of Mark-
topi. The orders were for him to defeat the region’s Kakhs (led by Giorgi Saaka-
dze) and then deport them (Abuladze 2000: 2).27 Khorji Khan neither received
the lion nor the message from Shah Abbas. These were intercepted by Saakadze
who then launched a successful surprise attack to defeat and kill Khorji Khan at
Marktopi. Despite this success, Saakadze proved unable to maintain his position
against the full might of the Safavid army and retreated towards the Ottoman
frontier.
DEFEAT OF RENEWED OTTOMAN OFFENSIVES IN YEREVAN AND ARDABIL (1616-1618)
The Ottomans launched a powerful offensive into the Caucasus in 1616 (Tucker
2010: 566). Pietro Della Valle, an Italian traveler in Iran during the time of Shah
Abbas, reports the Turkish invasion force standing at 200-300,000 troops (Šafā
1969: 343-348). The aim of this force was to capture Yerevan defended by 60,000

24
Discovered on August 14, 2000 in the Meskheti region of the Georgian republic.
25
A report was submitted by Professor Tsisania Abuladze to the Institute of Manuscripts of the
Georgian Academy of Sciences, Tbilisi, Republic of Georgia in 2000. According to the Georgian
Academy of Sciences “The statue is a mythical one and represents a lion, a symbol of Persia at the
time [Safavid era]” (see also full contents of that report and analysis in English by Farrokh, 2009).
26
The Georgian Academy of Sciences examined these documents in collaboration with Profes-
sors S. Janashi and N. Berdzenishvili.
27
The documents do not make clear where the Kartlians were to be deported, but judging by
past precedent, Iran would have been the most likely destination.
 The Military Campaigns of Shah Abbas I in Azerbaijan and the Caucasus 93

Iranian troops (Goodrich 1896: 127).28 The Ottoman siege of Yerevan was a fail-
ure, forcing the Ottomans into a costly withdrawal in early 1617 (Clodfelter 2008:
61). Valle reports the Iranian Tofangchi musketeers as having played a key role in
repelling the Ottoman assault (Šafā 1969: 343-348).
After their defeat at Yerevan, retreating Ottoman troops suffered more losses
to the bitter winter cold and constant attacks of Safavid forces (Dupuy/Dupuy
1977: 586). Despite their recent failure, the Ottomans prepared for another expe-
dition in 1618. This time an Ottoman army of 50,000 troops, led by Khalil Pasha,
advanced towards Tabriz in Azerbaijan province (Maškūr 1999: 277). Shah Ab-
bas, now in Ardabil, allowed the Ottoman army to march into Azerbaijan and
seize Tabriz (which had already been evacuated). This allowed Shah Abbas to
conserve his forces for the counterattack at a time and place of his choosing.
The Ottomans, now perceiving themselves as having the upper hand, sent
messages to Shah Abbas to yield all territories he had recaptured from the Otto-
mans since 1603. Shah Abbas refused and “officially” declared that he was to set
fire to Ardabil before retreating. The Ottomans marched towards Ardabil, not
realizing that Shah Abbas was waiting for them with 40,000 troops. The Safavid
army set a deadly ambush (Dupuy/Dupuy 1977: 586) and defeated the Ottomans
(Clodfelter 2008: 61) who then sued for peace. The Ottomans reaffirmed the Ira-
nian conquests and agreed to reduce the annual Iranian tribute to the Sultan to
just one hundred silk laden camels (ibid.). Political disarray within the Ottoman
state had also contributed to their recent military defeats.

BIBLIOGRAPHY
Abbasi, M. (1956), Siyāhnāmeye Chardin, (10 volumes), Tehran.
Abuladze, Ts. (2000), A Legend: About a Golden Lion, Report issued for the Institute of Manuscripts
of the Georgian Academy of Sciences, Tbilisi.
Amiri, M. (1970), Safarnāmeye Venezian dar Irān, Tehran.
Andronikashvili, M. (1966), Narkvevebi Iranul-Kartul enobrivi Urtiertobidan, Tbilisi.
Bābāie, A. (2005), Tārix-e arteše Irān, Tehran.
Bayānī, J. (1974), Tārix-e neẓāmi-e Irān (Janghā-ye dorre-ye Safaviye), Tehran.
Burton, A. (1997), The Bukharans: A Dynastic, Diplomatic, and Commercial History 1550-1702, Rich-
mond.
Chardin, J. (1983), Voyage de Paris à Ispahan I: De Paris à Tiflis, Paris.

28
Goodrich also reports the Ottoman force as having stood at 100,000 troops.
94 Kaveh Farrokh

Clodfelter, M. (2002), Warfare and Armed Conflicts: A Statistical Reference to Casualty and Other
Figures, 1500-2000, Jefferson NC.
Dupuy, R. E. / T. N. Dupuy (1986), The Encyclopedia of Military History: From 3500 B.C. to the Present,
New York.
Falsafī, N. (1965), Zendegānī-ye Šāh Abbas Avval, (6 Volumes), Tehran.
Farmanfarmaian, F. S. (2009); “Georgia and Iran: Three Millennia of Cultural Relations An Over-
view”, Journal of Persianate Studies 2.1: 1-43.
Farrokh, K. (2005), Elite Sassanian Cavalry, Oxford.
―― (2007), Shadows in the Desert: Ancient Persia at War, Oxford.
―― (2009), “The Winged Lion of Meskheti: a pre- or post-Islamic Iranian Legacy in Georgia?”, Sci-
entific Paradigms, Studies in Honour of Professor Natela Vachnadze. St. Andrew the First-
Called Georgian University of the Patriarchy of Georgia, Tbilisi: 455-492 (volume in Geor-
gian, Farrokh article in English).
―― (2011), Iran at War: 1500-1988, Oxford.
Frye, R. (1977), The Golden Age of Persia: Arabs in the East, London.
Goodrich, S. G. (1896), A History of all Nations, from the earliest periods to the Present, New York and
Auburn, Miller, Orton, Mulligan.
Gvakharia, A. (1995), “On the History of Persian-Georgian Contacts”, Proceedings of the Second Eu-
ropean Conference of Iranian Studies, ed. B. G. Fragner et al., Rome.
Izady, M. (1992), The Kurds: A Concise History and Fact Book, Abingdon.
Khorasani, M. M. (2006), Arms and Armor from Iran: The Bronze Age to the End of the Qajar Period,
Tübingen.
―― (2010), “Persian Firepower: Artillery”, Classic Arms and Militaria XVI.1: 19-25.
Lang, D. M. (1966), The Georgians, New York, Washington.
Maškūr, M. J. (1999), Tārix-e Irānzamīn az rūzegār bāstān tā Qājārīeh, Tehran.
Mašoun, H. (2001), Tārix-e Mūsīqī-ye Irān, Tehran.
Matofī, A. (1999), Tārix-e čahār hezār sāle-ye arteš-e Irān: Jeld-e dovom, Tehran.
Matthee, R. P. (1999), The Politics of Trade in Safavid Iran, Cambridge.
Mīrxand / R. Gh. Hedāyat (Original in 1851, Reprinted 1960-1961); Tārix-e Rozat ol Safa, Tehran.
Monšī, I. B. (2003), Tārix-e Alam Araye Abbasi, Tehran.
Nahavandi, H. / Y. Bomati (1998), Shah Abbas, Empereur de Perse (1587-1629), Paris.
Nasidze, I. / D. Quinque; M. Rahmani; S. Y. Alemohamad; M. Stoneking (2006), “Concomitant Re-
placement of Language and mtDNA in South Caspian Populations”, Current Biology 16:
668-673.
Nayernuri, H. (1965), Tārixče-ye beyraq-e Irān va Šīr o Xoršīd, Tehran.
Pārīzī, M. (1999); Siāsat va eqtesad dar ‘asr-e Safavi, Tehran.
Pigulevskaya, N. (1975), Tārix-e Irān az dorān bāstān tā pāyān-e sade-ye hejdah mīlādī, translated to
persian by K. Keshavarz, Tehran.
Rajab-Nīyā, R. (1955), Sazmān-e edārī-ye hokūmat-e Safavī, Tehran.
Ravandī, M. (1973), Tārix-e ejtemāyī-e Irān (8 Volumes), Tehran.
 The Military Campaigns of Shah Abbas I in Azerbaijan and the Caucasus 95

Rayfield, M. (2000), The Literature of Georgia: A History, Richmond.

Roemer, H. R. (1991), “The Safavid Period”, eds. P. Jackson; L. Lockhart, The Cambridge History of
Iran Volume 6: The Timurid and Safavid Period, Cambridge: 189-347.
Savory, R. M. (1980), Iran under the Safavids, Cambridge.
Šafā, Sh. (1969), Safarnāme-ye Pietro Della Valle-Qesmat-e marbūd be Irān, Tehran.
Šīrānī, H. (1957), Safārname-ye Jean baptiste Tavernier, Tehran.
Šokrī, Y. (1971), Alam Araye Safavi, Tehran.
Sīaqī, M. D. (1989), Tazakor ol Molook, Tehran.
Taqzelī, N. (1967), Safarnāme-ye Sanson, Tehran.
Taheri, A. (1985), The Spirit of Allah, Hutchinson.
Tucker, S., et. al. (2010), A Global Chronology of Conflict: From the Ancient World to the Modern Mid-
dle East, Santa-Barbara.
Whittow, M. (1996), The Making of Byzantium: 600-1025, Berkley.
 Persia and Persians in Raffi’s Xamsayi Melikʻutiwnnerə

Aldo Ferrari
Ca’ Foscari University, Venice

Abstract
This paper takes into consideration the image of Persia and Persians in Raffi’s Xamsayi melikʻu–
tʻiwnnerə. It is not a literary work, but a kind of history of Łarabał/Arcʻax, especially of the Arme-
nian nobility of that region, the so-called meliks. During almost the whole period described by
Raffi—1600-1827—Łarabał was part of the Persian empire, although in a position of strong auton-
omy. Therefore it is not surprising that we find in Xamsayi melikʻutʻiwnnerə many remarkable con-
siderations about Persia and Persians, obviously according to the peculiar ideological perspective
of Raffi. As a matter of fact, the western-minded Armenian writer often addresses to the Persians
the “orientalist” biases of “religious fanaticism”, “cruelty” and “barbarism”. Nevertheless his text
provides an interesting description of the secular relations between Persia and the Armenians,
mainly but not only those of Łarabał, in the fields of politics, religion and customs. Xamsayi meli-
kʻutʻiwnnerə can also be considered an important evaluation of the transition from the Persian in-
fluence to the Russian one undergone by Eastern Armenia in XVIII-XIX centuries. In spite of his
warm, but not uncritical, support to the Tsarist conquest of Transcaucasia, Raffi was indeed able to
give a multisided picture of this process.

Keywords
Armeno-Persian Relations, Armenian Nobility, History of Łarabał/Arcʻax, Russian Empire, Russian
Orientalism

RAFFI AND THE MELIKʻS OF ŁARABAŁ

Raffi (Yakob Melikʻ-Yakobean, 1835-1888) is one of the most famous Armenian
writers.1 My paper will take into consideration some aspects of his attitude to-
wards Persia and Persians, which is an interesting page of a millenary relation
(on this topic see Zekiyan 2005).
First of all we have to remember that he was born near Salmast, in north-
western Persia, and spent there his youth. Later, he visited Persia in 1857-58 and,

1
For a first approach to this author see (1937), Raffi (Yakob Melikʻ-Yakobean). Keankʻ, grakanu-
tʻiwnə, yišołutʻiwnnerə, Paris.; Hay nor grakanutʻyan patmutʻyun, III, Erevan, 1964: 327-420; Čanašean
1973: 86-94; Bardakjian 2000: 144-148.
 Koninklijke Brill NV, Leiden, 2015
98 Aldo Ferrari

on the occasion, wrote a Trip to Persia (Čanaparhordutʻiwn Parskastani) and a

collection of articles about Persia and the Armenians—A letter from Tabriz (Na-
mak Davrežicʻ)—whose ethnographical value has recently been studied (Petros-
yan 2007). My presentation does not deal with these texts or the literary works
devoted by Raffi to the Armenians in Persia, but I’m taking into account the im-
age of the Persians we can find in another text of this author, i.e. Xamsayi meli-
kʻutʻiwnnerə. 1600- 1827. Niwtʻer hayocʻ nor patmut'ean hamar.
Xamsayi melikʻutʻiwnnerə is something like a history of Łarabał/Arcʻax, espe-
cially of the Armenian nobility of that region, the so-called meliks (melikʻner).2 As
part of a work in progress of mine on Armenian nobility in modern times,3 I pre-
pared an Italian translation of this text (Raffi 2008).4
Although from a historical point of view, this work is largely obsolete con-
tains a number of mistakes, it can be extremely useful in illustrating not only the
history of this region, but also an important page of modern Armenian self-con-
sciousness.
One must consider Xamsayi melikʻutʻwnnerə within the ideological frame of
Raffi, who was persuaded that the political rebirth of Armenia could not be
achieved through the intervention of the great powers, but only by the autono-
mous action of the people. Such an action, however, depended in his opinion on
a large work of self-education, mostly of historical nature. In this sense Xamsayi
melikʻutʻiwnnerə is strictly connected with Raffi’s historical novels: indeed, the
building of a modern Armenian nation needed a deep consciousness of the past.
From this point of view, the meliks of Łarabał, who still in the eighteenth century
conserved a remarkable political and military role, were very important in Raffi’s

2
A part from Raffi’s Xamsayi melikəutʻiwnnerə. 1600- 1827. Niwtʻer hayocʻ nor patmut'ean hamar,
in Tsarist period this topic has been studied by Beknazareancʻ A. (1886); Kostaneancʻ A. (1913); Tēr-
Mkrtčʻean K. (1914). Among the few works dedicated to the melikʻs in Soviet time see Barxudaryan
S. (1967) and Sargsyan M. (1987. In the West Hewsen R. H. published a number of articles to this
subject, see Hewsen 1972; idem 1973-74; idem 1975-76; idem 1980; idem 1981-82; idem 1984. In con-
temporary Armenia, besides the monograph of Łulyan A. (2001), Małalyan A. has devoted to this
topic some articles, see Małalyan 2003; idem 2003a; idem 2004; idem 2004a; idem 2005; idem 2006;
idem 2007.
3
See also my translation of the XVIII century chronicle about Dawitʻ Bēk written in Venice by
the Mekhitarist Łukas Sebastacʻi (Sebastacʻi 1997). Later on I devoted to this topic some articles
and a monograph , see Ferrari 2004; idem 2004a; idem 2006; idem 2007; idem 2009; idem 2011a.
4
A Russian translation of this text has been published in 1991 (Melikstva Chamsy, Erevan),
while in 2010 an English one appeared, see Raffi 2010.
 Persia and Persians in Raffi’s Xamsayi Melikʻutiwnnerə 99

“ideological use of history” (Sarkisyanz 1985: 99). Besides, as his true name (Me-
likʻ-Yakobean) seems to show, Raffi descended from a family of the eastern Ar-
menian nobility (Hewsen 1972: 308).
During the summer of 1881, Raffi left Tʻiflis—where he lived as most part of
the eastern Armenian intelligencija—and travelled for two months in Łarabał.
He visited almost all the region and collected a multitude of written documents
and oral reports concerning the meliks. In 1882 he published Xamsayi melikʻu-
tʻiwnnerə in the newspaper “Mšak”.

PERSIA UNDER RAFFI’S EYES

This text is interesting in many ways. Raffi takes into account the persistence of
an indigenous nobility, gives a critical evaluation of the Armenian Church, ex-
amines the Armenian relations with Russia and the Muslim peoples and, finally,
produces a facinating representation of Persia and the Persians. We have to re-
member that during almost the whole period described by Raffi—1600-1827—
Łarabał was part of the Persian Empire, although in a position of large auton-
omy. Therefore it is not surprising that many a remarkable consideration about
Persia and the Persians can be found in Xamsayi melikʻutʻiwnnerə.
First of all, Raffi recognizes that the legitimacy of melikʻs power came from
Persia:
In the last centuries instead of the ancient naxarar houses appeared the melikʻs,
whose power was more legally sanctioned by Šah Abbas (1603). Unlike his predeces-
sors, this creative Persian king understood the relations with the foreign subjects in a
wholly different way and allowed them to be ruled by their representatives. So he
succeeded in strengthening the inner cohesion of the State. Šah Abbas was the first
to confirm the title of melikʻ that the Armenian princes had used since ancient times.
Thus, he also rewarded the Armenian melikʻs for the important services rendered to
him at the time of his victory over the Ottomans (Raffi 1987: 417).

A century later, their position was recognized also by Nadir-Šah. Raffi re-
marks that
… The Armenians, who chased off the Ottomans with the sword from their father-
land, greatly helped Nadir’s victory (ibid.: 442).

The new Persian king did not forget Armenian melikʻs services and, unlike
Christian—i.e. Russian—emperors he rewarded them. Once he ascended the
throne, Nadir-Šah
100 Aldo Ferrari

… confirmed them in their domains and gave each of them the right to rule autono-
mously in their countries, paying an annual fixed tribute. Since then the melikʻs were
subject only to the šah… (ibid.: 442, 444).

In Xamsayi melikʻutʻiwnnwə Raffi points out that the Armenians were often in
good terms with the Persians. For example, the melikʻ of Łarabał usually had
friendly relations to the khans of Ganjak. About one of them, Šahverdi-xan, Raffi
writes that he
… was a good hearted man, as often were the khans of Ganjak. He was of Persian,
not Turkish origin. And the Persians proved themselves comparitavly more benevo-
lent toward the Christians than the wild Turks-Mongols (ibid.: 466).

As a matter of fact, among the Muslim peoples Raffi considers the Persians
much more civilized than the Turks. The cooperation between Armenians and
Persians has been indeed lasting and profitable for both. Still at the eve of the
Russian conquest of Transcaucasia, the heir to Persian throne, Abas-Mirza, con-
sidered the Armenians a useful element of Persia:
Therefore he wanted to improve their situation and granted them several privileges.
His father, the good and philanthropic Fatali-šah, shared this benevolence and
during his reign the Armenians of Persia could live in an enviable situation. As Šah
Abas the Great founded in Isfahan the companies of Armenian merchants to pro-
mote Persian commerce, so Abas-Mirza kept Armenian merchants, who used his
own capital. He also strengthened the Armenian melikʻs by granting them rights
and honours in order to make the Armenian population of the region safer against
the abuses of the Muslim governors. Besides, Abas-Mirza knew well the strong con-
nection of the Armenians with their Church and clergy; thus he not only began to
lessen the religious persecution of the Christians, but even tried to sustain and en-
sure Christendom. Abas-Mirza frequented Christian churches and attended the Ar-
menian religious feasts, thus showing with his personal example that Armenian
worship was highly respectable. At the time of Abas-Mirza the bells of the Armenian
churches could ring again, where as previously it had been forbidden.
… Abas-Mirza and his predecessors appreciated the Armenians not only as farmers,
merchants and craftsmen, but also for their military and administrative qualities….
Many Armenians attained high military ranks, ruled whole regions as viceroys,
served in the diplomacy, controlled the treasury and even looked after the Šah's
harem (ibid.: 593).
In this point of Xamsayi melikʻutʻiwnnerə Raffi inserts also an extremely long
note in which he enumerates many Armenians who distinguished themselves in
the Persian empire (Raffi 1987: 593-595). Raffi finally remarks that
 Persia and Persians in Raffi’s Xamsayi Melikʻutiwnnerə 101

… the line of conduct of Abas-Mirza had his political aims. He desperately tried to
bind Armenian hearts to Persia, cooling their sympathy toward Russia. Neverthe-
less, apart from his political aims Abas-Mirza’s benevolence toward the Armenians
is unquestionable (ibid.: 593).

However, this benevolent policy of the late Persian Empire could not dis-
suade many Armenians from backing the Russians. One should not forget that
the Russian conquest of Transcaucasia depended also on Armenian demand,
since Israyēl Ōri’s first famous mission to Moscow in 1701(Johannissjan 1913;
Kʻiwrtean 1960; Essefian 1972). Also in the time of the Persian expedition of Peter
the Great (1721-22) several melikʻ and the katʻołikos of Ganjasar backed the Rus-
sians and began the so-called Armenian Liberation Movemenent.5
According to Raffi,
… In the face of the dissolution of Persia, the melikʻs of Łarabał decided to exploit the
situation. Until then they had considered themselves Persian subjects, but then they
tried to reverse this yoke to build an independent Armenian state (Raffi 1987: 430).

So, in spite of the large self-government of the melikʻs and the already men-
tioned good terms with the Persians, Raffi calls their domination “a yoke”. Why?
Although we can’t consider him a devout Christian such a definition partially
depends on religious and moral considerations. For example Raffi severely
blames Melikʻ-Šahnazar of Varanda not only for helping Pʻanah-xan in building a
Muslim Khanate in Łarabał, but also for being influenced by Persian customs. In
Chapter XI he writes that
… Melikʻ- Šahnazar was a completely immoral man who followed Persian customs
and had many concubines … So doing, he introduced in his house the polygamy of
the Muslims. Such a behavior deeply offended the religious feelings of the people and
made him odious to the other meliks of Łarabał (ibid.: 452).

The Muslim religion of the Persians is for Raffi a tremendous barrier between
them and the Armenians. He often highlights the religious fanaticism of the Per-
sians, who from this point of view are considered to be even more intolerant
than the Turks. For example, according to Raffi, Pʻanah-xan was morally better
than his son, Ibrahim-xan, because he had preserved the simplicity of his (Turk-
ish) stock. On the contrary, Ibrahim-xan
… had been educated in Persia where he had learned all the fanaticism of the Mus-
lim religion (ibid.: 496).

5
On this remarkable episode of modern Armenian history see Arutjunjan 1954; Aivazian 1997.
102 Aldo Ferrari

The religious fanaticism of the Persians is stigmatized also in the chapter de-
voted to the cruel execution of the young Safareli-Bēk, who had murdered Ała-
Mamad, the founder of the Qajar dynasty:
The Armenians [of Tabriz], who knew that he was the son of Christians and believed
in Christ, asked his body for burial. The Persian dignitaries told them that they could
bury his body according to their contemptible rite. Indeed, they considered the burial
according to the Armenian rite more shameful than being fed to the animals (ibid.:
526).
Therefore, in Raffi’s perspective, Armenians could certainly be in good terms
with Persians on the basis of personal links but only in spite of their respective
religion. As a matter of fact, Raffi considered Islam a fanatical and oppressive re-
ligion, but at the same time he blamed the Armenian Church for passivity, lack
of culture and insufficient national spirit (Bardakjian 2000: 145; Ferrari 2010). For
Raffi, who was indeed a progressive intellectual deeply influenced by Russian
radicalism, the basic questions of his time didn’t have religious, but cultural
character. So, when in chapter XXXVIII he writes that after the Russian conquest
of Transcaucasia
… for the Armenians began a new life: the Persian tyranny … gave way to a Christian
state (Raffi 1987: 539).
we ought to understand his thought correctly. Russia was a Christian country,
but first of all a European and modern state. According to Raffi, while Persia was
an Asian, Eastern and backward country, Russia represented for the Armenians a
model of Western progress and development.6
From this point of view Raffi’s description of the last Russo-Persian war (1826-
1827) is very interesting. As a matter of fact the result of that war was the com-
plete Tsarist conquest of Eastern Armenia, which for centuries had been a Per-
sian domination. In the Xamsayi melikʻutʻiwnnerə the main hero of the Russo-
Persian war is General Madatʻov, an Armenian from Łarabał. Raffi describes his
victory near Šamkʻor as a kind of colonial battle:
His [Madatʻov’s] threatening name, that had already became legendary among the
Muslims, sufficed to terrify them. Besides, in his military operations he resorted to
tricks, which strongly impressed the imagination of the Orientals (arm. are-
welkʻcʻiner). Like Homer’s heroes, who built a huge wooden horse to conquest Troy,
Prince Madatʻov ordered to make a big cart, a kind of infernal machine, which was
slowly pushed by his soldiers and equipped with cannons. The enemy was frightened

6
On this topic see Ferrari 2011.
 Persia and Persians in Raffi’s Xamsayi Melikʻutiwnnerə 103

by the fire and the cannon balls emerging from this monster. In the result, the Per-
sians were completely defeated (ibid.: 587-588).

In this description we can find not only the affirmation of the technological
superiority of modern European Russia upon backward Asian Persia, but also a
new version of the archetypical contrast between East and West, between Asia
and Europe. Madatʻov is depicted as a new Ulysses and his Western reason inevi-
tably prevails upon Eastern emotionalism. I think that this passage could be
quoted as a perfect example of Western Orientalism, in Edward Said’s sense.7
However, we must also take Raffi’s peculiar perspective into consideration. He
was not a Europe-born “Orientalist”, but an (Eastern) Armenian who, like Xačʻa-
tur Abovean before him, was indebted to Tsarist Russia for a new “western” per-
spective. Thus, the Western-minded Armenian writer can look at the Persians,
the traditional neighbours of his people, not only as adherants to a different reli-
gion, but also as representatives of a backward, Eastern and “Oriental” world.
From this point of view, expressions linked to the Persians like “religious fanati-
cism”, “cruelty” and “barbarism” which we can find rather frequently in the Xam-
sayi melikʻutʻiwnnerə are largely connected to the new political and cultural situ-
ation embraced by the Eastern Armenians after the Russian conquest.
At the same time, one must remember that Raffi was not dogmatic in his pro-
Russian stance. For example, he understood Melikʻ-Meǰlum’s decision to help
Ała-Mamad-xan, the founder of the Qajar dynasty, who in 1795 invaded Trans-
caucasia. According to Raffi,
Ała-Mamad-xan was a barbar, but also an intelligent man and a good politician. He
understood very well that with the Armenian help it would have been easier for the
Russians to penetrate into his dominions. Therefore, granting concessions to the
Armenians, he blocked the Russians' way to Persia. But the Armenian meliks pre-
ferred to remain loyal to the Russians and to resist Ała-Mamad-xan and even joined
the enemy of old, Ibrahim-xan. They thought that doing this it would have been easy
to destroy him afterwards, while following a submission to Ała-Mamad-xan the lib-
eration of the motherland would have become impossible. Only Melikʻ-Meǰlum’s Is-
rayēlean didn’t agree … Melikʻ-Meǰlum was an intelligent young. He was aware that
his ancestors could hold and defend an autonomous princedom in the mountains of
Łarabał only thanks to the Persian Šahs, not to a Christian kingdom (ibid.: 508-509).

7
For a picture of the issue of Russian Orientalism see Knight 2000; Khalid/Knight/Todorova
2000; Jobst 2000; Ferrari 2003/2012; Meaux De 2010); Schimmelpenninck 2010; Tolz 2011. On the
Russian “Orientalist” approach towards Persia in Tsarist period see Andreeva 2007.
104 Aldo Ferrari

Thus, Raffi did not refuse a priori Melikʻ-Meǰlum's pro-Persian option, which
might have a firmer historical bases than the pro-Russian one. At the same time
Raffi often criticizes Russia for its ingratitude shown towards the Armenians. As
a matter of fact, the melikʻs of Łarabał received a poor reward for their services;
unlike the Georgian nobility, they were not recognized as princes by the Russian
government (Hewsen 1972: 295). From this point of view Madatʻov was an excep-
tion and Raffi openly criticizes him for trying to introduce serfdom in his lands
in Łarabał where such an institution had never existed under Persian domina-
tion (Raffi 1987: 580).
CONCLUSION
Apart from its importance for the study of the Armenian nobility, Raffi’s Xamsayi
melikʻutʻiwnnerə can be considered as an important description of the transition
from Persian influence to Russian dominion undergone by Łarabał and Eastern
Armenia in the XVIII-XIX centuries. In spite of his warm, but not uncritical, sup-
port to the Tsarist conquest of Transcaucasia, Raffi was able to give a multisided
assessment of this historic process. Therefore, the Xamsayi melikʻutʻiwnnrə pro-
vides an interesting description of the secular relations between Persia and the
Armenians, mainly, but not only, those of Łarabał, in the fields of politics, reli-
gion and custom. To a certain extent this text can also be interpreted as a reflec-
tion of the Orientalist approach borrowed after the Tsarist conquest of the Cau-
casus by the Eastern Armenians from Russian culture.

BIBLIOGRAPHY
Aivazian, A. (1997), The Armenian Rebellion of the 1720s and the Threat of Genocidal Reprisal, Yere-
van.
Andreeva, E. (2007), Russia and Iran in the Great Game. Travelogue and Orientalism, London-New
York.
Arutjunjan, P. T. (1954), Osvoboditel'noe dviženie armjanskogo naroda v pervoj četverti XVIII veka,
Moskva.
Bardakjian, K. B. (2000), A Reference Guide to Modern Armenian Literature. 1500-1920, Detroit.
Barxudaryan, S. (1967), “Gełarkʻunikʻi melikʻnern u tanuterə əst Tatʻevi vankʻi mi pʻastatʻułti”, Ban-
ber Matenadarani 8: 191-227.
Beknazareancʻ, A. (1886), Gałtnikʻ Łarabałi, Sankt Peterburg.
Calzolari, V. / A. Sirinian / B. L. Zekiyan (eds.) (2004), Dall’Italia e dall’Armenia. Studi in onore di
Gabriella Uluhogian, Bologna: 181-205.
Čanašean, M. (1973), Hay grakanutʻean nor šrǰani hamaṙōt patmutʻiwn, Venetik.
 Persia and Persians in Raffi’s Xamsayi Melikʻutiwnnerə 105

Essefian, A. (1972),”The Mission of Israel Ori for the Liberation of Armenia”, Recent Studies in Mod-
ern Armenian History, Cambridge: 1-10.
Ferrari, A. (2004), “Nobility and Monarchy in Eighteenth Century Armenia. Introduction to a New
Study”, Iran and the Caucasus 8.1: 53-63.
―― (2004a), “Menkʻ mec Hayastaneacʻs išxankners ew melikʻners”, Introduzione allo studio della
nobiltà armena in Transcaucasia nel XVIII secolo”, Dall’Italia e dall’Armenia. Studi in onore
di Gabriella Uluhogian, eds. V. Calzolari , A. Sirinian, B. L. Zekiyan, Bologna: 181-205.
―― (2006), “Ean Potockin ew nra Hayastani azatagrman cragirə”, Bazmavep 2006: 552-566.
―― (2007), “Raffin ew Łarabałi hay aznvakanutʻiwnə”, Republic of Nagorno-Karabakh. Past, Pre-
sent, Future. International Scientific Conference, Yerevan: 229-240.
―― (2009), “La nobiltà georgiana e armena nell’Impero russo”, Semantiche dell’Impero, Atti del
Convegno della Facoltà di Lingue e Letterature Straniere, eds. A. Ferrari, F. Fiorani, F. Passi,
B. Ruperti, Scriptaweb, Napoli: 377-396.
―― (2010), “Vojna i mir v armjanskoj kul’ture Novogo vremeni”, Vojna i sakral’nost’. Materialy Čet-
vertych meždunarodnych naučnych čtenij, eds. O. Ermačenko, S. M. Capilupi, Sankt Peter-
burg: 298-317.
―― (2011) Alla frontiera dell’impero. Gli armeni in Russia (1801-1917), Milano.
―― (2011a), In cerca di un regno. Profezia, nobiltà e monarchia in Armenia tra Settecento e Otto-
cento, Milano.
―― (2012), La foresta e la steppa. Il mito dell’Eurasia nella cultura russa, Milano.
Sarinyan, S. N. (ed.) (1964) Hay nor grakanutʻyan patmutʻyun, III, Erevan.
Hewsen, R. H. (1972), “The Meliks of Eastern Armenia: A Preliminar Study”, Revue des Etudes Armé-
niens 9: 285-329.
―― (1973-1974), “The Meliks of Eastern Armenia II”, Revue des Etudes Arméniens 10: 282-300.
―― (1975-1976), “The Meliks of Eastern Armenia III”: Revue des Etudes Arméniens 12: 219-243.
―― (1980), “The Meliks of Eastern Armenia IV”, Revue des Etudes Arméniens: (1980): 14: 459-470.
―― (1981-1982), “Three Armenian Noble Families of the Russian Empire [The Meliks of Eastern
Armenia V]”, Hask: 389-400.
―― (1984), “The Meliks of Eastern Armenia VI: the House of Aghamaleanc”, Bazmavēp: 319-333.
Jobst, K. S. (2000), “Orientalism, E. W. Said und die Osteuropäische Geschichte”, Saeculum 51.2: 250-
266.
Johannissjan, A. (1913), Israel Ory und die armenische Befreiungsidee, München.
Kʻiwrtean, Y. (1960), Israyēl Ori, Venetik.
Khalid, A. / N. Knight / M. Todorova (2000), “Ex tempore: Orientalism and Russia”, Kritika: Explora-
tions in Russian and Eurasian History 1.4: 691-727.
Knight, N. (2000), “Grigor’ev in Orenburg, 1851-1862: Russian Orientalism in the Service of Empire?”,
Slavic Review 1: 108-141.
Kostaneancʻ, A. (1913), Niwtʻer hay melikʻutʻean masin. I prak Dizaki melikutʻiwnə, Vałaršapat.
Łulyan, A. (2001), Arcʻaxi ev Syunikʻi melikʻakan aparankʻnerə, Erevan.
Małalyan, A. (2003), “Gyulistan gavaṙi tirakal Melikʻ-Beglaryani tohmacaṙə”, Krtʻutʻyunə ev gitutʻyu-
nə Arcʻaxum: 1-2, 11-14.
―― (2003), “Ĵraberdi gavaṙi tirakal Melikʻ-Israyel tohmacaṙə”, Hayocʻ patmutʻyan harcʻer 4: 36-40.
106 Aldo Ferrari

―― (2004), “Arcʻaxi Giwlistan gavaṙi tirakal Melikʻ-Beglareanneri Hndkastani čiwłə”, Handes Am-
sorya CXVIII: 475-491.
―― (2004a), “Arcʻaxi Gyulistan, Ĵraberdi ev Dizaki gavaṙner melikʻakan tneri tohmacaṙə”, Bazma-
vēp CLXII: 93-121.
―― (2005), “Arcʻaxi melikʻakan tneri tohmacaṙerə (niwtʻer hayocʻ melikʻutʻiwnneri masin) I, Baz-
mavēp CLXIII: 153-193.
―― (2006), “Arcʻaxi melikʻakan tneri tohmacaṙerə (niwtʻer hayocʻ melikʻutʻiwnneri masin) II, Baz-
mavēp CLXIV: 262-292.
―― (2007), Arcʻaxi melikʻutʻiunnerə ev melikʻakan tnerə XVII-XIX dd., Erevan.
Meaux, L. De (2010), La Russie et la tentation de l’Orient, Paris.
Minasjan, S. (2003), “Rol’ armjan v persidskoj armii v pervoj polovine XIX veka”, Patma-banasira-
kan handes 1.262: 231-242.
Parsamean, M. et alii (eds.), (1937), Raffi (Yakob Melikʻ-Yakobean). Keankʻ, grakanutʻiwnə, yišołu-
tʻiwnnerə, Paris.
Petrosyan, A. (2007), “Raffu parskakan ułevorutʻyunnerə orpes Parskahayeri azgagrakan usumnasi-
rutʻyan skizbałbyur”, Patma-banasirakan handes 3 (176): 100-120.
Raffi (1987), Erkeri žoŀovacu, IX, Erevan.
―― (2008), I melikʻ del Łarabał. 1600-1827. Materiali per la storia moderna degli armeni,
(traduzione, introduzione e note di A. Ferrari), Milano.
―― (2010), The Five Melikdoms of Karabagh, (translated from Armenian by A. S. Melkonian),
London.
Sarkisyanz, M. (1985), A Modern History of Transcaucasian Armenia, Nagpur.
Sargsyan, M. (1987), “Melikʻakan bnakeli hamkaruycʻ Toł avanum”, Patma-banasirakan handes 3:
132-140.
Schimmelpenninck van der Oye, D. (2010), Russian Orientalism. Asia in the Russian Mind from Peter
the Great to the Emigration, New Haven-London.
Sebastacʻi, Ł. (1997), Le guerre di Dawit’ Bēk, un eroe armeno del XVIII secolo, (a cura di A. Ferrari),
Milano.
Tēr-Mkrtčʻean, K. (1914), Hay melikutʻean masin. II prak. Dōpʻeankʻ ew melikʻ Šahnazareancʻ, Ēǰmia-
cin.
Tolz, V. (2011), Russia’s Own Orient. The Politics of Identity and Oriental Studies in the late Imperial
and Early Soviet Period, Oxford.
Zekiyan, B. L. (2005), “The Iranian Oikumene and Armenia”, Iran and the Caucasus 9.2: 231-256.
 New Information on the History of the Caucasus in the
Third Volume of Afzal al-tavarikh

Hirotake Maeda
Tokyo Metropolitan University

Abstract
The purpose of the article is to examine the account of a newly discovered Persian chronicle, the
third volume of Afzal al-tavarikh by Fazli Khuzani al-Isfahani. In the reign of Shah ‘Abbas, for a
decade, Fazli worked in the Caucasus as the top Safavid administrator, vizier of Barda‘ and Kakhe-
ti. Thus, his account has particular importance for the history of the Caucasus. He conveys new
and detailed information about the local administrative units and local representatives. He even
refers to a Georgian word and narrates the uprising in Georgia in 1625 led by Giorgi Saakadze (Mu-
rav Beg) as a witness to the event. His descriptions of Qizilbash representatives, Armenians, Jews,
and North Caucasians are also worth mentioning.

Keywords
Safavid Iran, Caucasus, Persian Chronicles, Fazli Khuzani, Georgia, Azerbaijan, Armenia, North
Caucasus

INTRODUCTION
In late 1990’s a Persian chronicle fully covering the 42-year reign of Shah ‘Ab-
bas was discovered in the library of Christ College of Cambridge University. It is
the third volume of Afzal al-tavarikh. The author, Fazli b. Zayn al-‘Abidin Beg
Khuzani (born around 1593 and died after 1639), was from a bureaucrat family of
Isfahan which produced several high ranking officials in the 16th century. The
third volume deals precisely with the period of the author (Library of Christ’s
College, Cambridge, Ms.Dd.5.6). Thus it constitutes the most important part of
the chronicle. Ch. Melville, who discovered the manuscript, revealed the history
and the features of this manuscript (Melville 1998, 2003). The aim of this paper is
to supplement his observations and further clarify the character of the source for
successive studies. Particular attention is paid to the information about the
Caucasus where he spent most of his time as a local bureaucrat and an admin-
istrator.
 Koninklijke Brill NV, Leiden, 2015
108 Hirotake Maeda

DESCRIPTIONS OF LOCAL ADMINISTRATIVE UNITS

Fazli spent most of his life as a provincial officer and left many notes on the local
administration and bureaucrats. His relatives occupied many important offices
in the central administration. Naturally most of his description is dedicated to
northwestern Iran and southeastern Caucasus.
In comparison with the well-known chronicle of Iskandar Beg, Fazli’s de-
tailed description of political events in the provinces are extremely important
because it not only gives us many new knowledge but also shows local dynamics
and their relations with the imperial center. In other words, the information of
the chronicle possesses immense possibilities for exploring the local political dy-
namics and the imperial policy towards it. For recognizing the specific character
of the description, we cite one example on Shirvan just after the (re)conquest in
1607-08.
Shamakhi and the surrounding area was the former fief (tiyul) of Beylarbey (of the
Ottoman army). The shah gave the military leadership to Zul Faqar Khan Qara-
manlu. Khwaja Sultan Hanan Qazvini, uncle of Saru Khwaja, was promoted as its
vizier. Khwaja Muhammand became the royal attendant and was entrusted with all
the sheep tax (chupanbeygi) in Azarbaijan. The accountant (mastaufi) of Ardabil
Shrine was entrusted to Abdal Beg Talish through the request of Zul Faqar Khan.
Abdal recognized himself as the descendant of Shaykh Zahid Gilani and received
royal favor in the expedition toward Hirat. Around this time he was ranked with the
qurchis together with his brother Shaykh Sharif Beg and nephew Chiragh Beg. Arda-
bil became the royal property and Abdal Beg became governor too. Revenue from
the royal property (khalisa) was allotted to the fief of the royal army.

This description contains information not only on individuals and their posi-
tions but also the career and connections each representative had. For example
Abdal Beg Talish was appointed as the accountant of Ardabil through the re-
quest of Zul Faqar Khan. His nephew Chiragh later played an important political
role as qurchibashi in Safi’s early reign. Allocation of royal property as the fief of
the royal army is a clear testimony of Shah ‘Abbas’s effort for constructing a
centralized empire. Through his description, we observe the imperial dynamics
of the household empire made by ‘Abbas.
Then the description turns to localities in Azarbaijan.
Sarab became the fief of Shah Nazar Sultan Bapardalu. Unkut was given to Mansur
Sultan Chamshqazaq. Mughan, the former fief of Zul Faqar Khan, was converted
into royal property and its revenue was the responsibility of Chupan Beg, the kalan-
 New Information on the History of the Caucasus 109

tar. The position of darugha, which supervised eighty thousand families, was en-
trusted to Mansur Khan Beg Shamlu.

These descriptions enrich our knowledge of local bureaucratic processes

which were embedded within the mechanism of imperial administration.
Considering the size of the chronicle, the wealth of information is remarkable.
We can assess the deliberate policies of ‘Abbas by following the information
on the new appointments and the distribution of administrative posts in the
provinces. In 1598 Farhad Khan Qaramanlu, a powerful Qizilbash representative,
was executed by the order of the shah. Nonetheless he did not purge the entire
clan. The shah calmed the unrest of Farhad Khan’s relative and former retainers
by bestowing positions.
Qizilaghaj was a main base of Farhad Khan. Zul Faqar Khan was given that region
with Langar Konan as his fief. Yadgar ‘Ali was once a deputy (vakil) of Farhad Khan
and at that time became centurion of the qurchis of Talish. He was given Safid Dasht
and Ujrud. Astara was given to Akhi Sultan Chakarlu, the close retainer of Farhad
Khan, together with the rank of sultan. Sayyidi Sultan Khabushlu received the
governorship of Anhar and Mazar-i Mughanat. The former fief of Farhad Khan ex-
isted in Azarbaijan were all divided as such.

Then Fazli turns to the area connected with northern Azarbaijan and Daghe-
stan.
The Governorship of Darband, the city called “gate among gates”, was given to Chi-
ragh Sultan Gilanpa Ustajlu. Shah Nazar Beg Tavakkuli, centurion of the musket-
eers (tufangchi), was ordered to guard the Darband castle with Rumlu and Bayat
soldiers. Their salary was allotted from the sheep tax of Azarbaijan and was to be
sent to the castle every year. Khwaja Muhammad Riza bears the responsibility. Um-
mant Sultan was appointed as the governor of Baku with three oil fields. His father
Rustam Sultan Suklun was killed in the Ganja castle by the hands of Ahmad Pasha.
The Khargi oil field which earns 800 tumans every year and became royal property
(Fazli, Afzal III, 218b-219b).

The description of local administrators and their functions, as well as

other specific information, including the oil of Baku, will be of interest to any
historian working on the region.

FAZLI’S INFORMATION ON THE CAUCASUS: A GENERAL VIEW

Fazli not only mentioned the administrative appointments but described how
the territory was divided and distributed. We also find significant information on
110 Hirotake Maeda

the administrative units and concrete identities of the officials in charge. It is im-
portant to note that Fazli was always keen to describe specific numbers. It is, of
course, dangerous to rely exclusively on his information. But by closely scrutiniz-
ing and analyzing the details, his account is valuable for its exposure of the real
picture of local management.
In fact Iskandar Beg usually referred only to the top persons, powerful gener-
als and governors. We expect to observe the process of centralization through
the khassa polity to integrate the household empire more tightly. The rich Azar-
baijan province certainly attracted ‘Abbas’s attention. In this regard Fazli’s de-
scription is highly valuable.
Ardabil 55b; 64b; 87b; 95a; 180a; 185b; 220a; 283b; 299b; 381b; 422a;
550a
Qarachedagh 28a; 55b; 63a; 272b
Azarbayjan (Tabriz) 28a; 49a; 52b; 151a-b; 180a; 181b; 188b; 244a; 263b; 267b; 271b;
279a; 280a; 295b; 316b; 360a; 381b; 386a; 425a; 454b; 478b; 484a;
488a-b; 515a; 518a
Chukhur-i Sa’d (Īrevan) 155a; 156a; 161b; 169b; 176a; 203a; 211a; 244a; 269b; 275b; 361b;
362a; 365b; 369b; 371a; 380b; 386a; 484a; 513a
Qarabagh (Ganja, 28a; 34b; 151ab; 156a; 165a; 194b; 197a; 203a; 244a; 252a; 269b;
Arasbar) 335a-b; 336a; 346a; 348b; 349b; 354b; 357b; 371a; 372a; 379a;
496b; 507b
Shirvan (Shamakhi, 51b; 57a; 171a; 178a; 185b; 203a; 206a; 215b; 220a; 245b; 258a;
Shakki) 339b; 348b; 349b; 378a; 515a; 518a
Gurjestan (Tiflis, Kartil, 156a; 171a; 178a; 185b; 200b; 203a; 241b; 244a; 267b; 284b; 307a;
Kakht, Suram, Akhisqa, 311b; 316b; 320b; 322a; 323b; 327a; 328a; 329a; 330b; 331b; 332a;
Bashiachuq) 334a; 335a-b; 346a; 349b; 354b; 373a; 380; 427a; 491b; 494a;
496b; 498a-b; 500b; 504a; 505a; 507b; 509a; 515a
North Caucasus (dasht-i 84b; 206a; 245b; 258a; 299b; 544a; 546b
Qipchaq, Daghistan,
Kuh-i Alburz, Shapuran,
Tabarsaran)

Other regional centres such as Qizilagaj (Fazli, Afzal III, 95a), Shuragul (Fazli,
Afzal III, 168b), Van (Fazli, Afzal III, 169b,181b), Nakhjivan (Fazli, Afzal III,
185b,352b), Baku (Fazli, Afzal III, 206a), Qars (Fazli, Afzal III, 211a) also appear.

FAZLI’S INFORMATION ON THE POLITICAL SITUATION IN GEORGIA

Fazli described in detail the activities of Caucasian elements at the Safavid court.
In particular, we find much original information on the life of Giorgi Saakdze
 New Information on the History of the Caucasus 111

Georgian origin. He changed his loyalty three times, i.e. to the Georgian king, the
Safavid shah and the Ottoman sultan. His fate is closely connected with Shah
‘Abbas. He climbed to one of the most influential position in Kartli kingdom of
eastern Georgia serving as the governor of capital Tbilisi (Mouravi means gover-
nor/mayor in Georgian and is equivalent to darugha) and was the father-in-law
of King Luarsab II. However having lost the struggle with the powerful tavadi
class (great nobles), he found himself in the court of Shah ‘Abbas. Serving the
shah for a decade in the complicated relations of Safavid authority with Georgia,
Saakadze suddenly betrayed and killed Qarchaqay and other prominent Safavid
generals in Georgia in 1625. Once again he lost his position in Georgia and
Saakadze became an Ottoman general. However he was captured and executed
by the Ottoman authority when he abandoned the Ottoman camp on their way
to the Safavid front and left for Georgia without notice (Jamburia 1964; Maeda
2011a).
Iskandar Beg gives little mention of him before the revolt in Georgia in 1625.
In fact he acknowledged Girogi Saakadze as an important political figure who
had become a trusted retainer of the shah (Iskandar Beg Turkman /Muhammad
Yusuf, Zayl-i ‘alam-ara, 29-30). We find such a reference in later description of
the early reign of Shah Safi. A Georgian source written several decades later
stressed how Saakadze was active as a Safavid courtier. Fazli incorporated this
contemporary episode and referred to Giorgi Saakadze’s exile to the shah’s court
in his narrative of the events of 1612-13 (Fazli, Afzal III, 311b). The internal situa-
tion of the Kartli kingdom and Saakadze’s downfall is described in detail. After
Saakdze started to work at the Safavid court, Fazli regarded him as the closest
servant of the shah after the death of Isfandiyar Beg Arabgirlu. Whether this was
really the case, the description is specific and thus worth attention (Fazli, Afzal
III, 482b).
The reference to Saakadze’s sister and brother are unique in contemporary
sources. He praised the beauty of Saakadze’s sister when he visited a Georgian
church as an envoy of his patron Paykar Khan, governor of Barda and Kakheti.
He also witnessed the murder of Kaykhosraw, brother of Saakadze and a servant
of the shah in the military camp near Baghdad (Fazli, Afzal III, 311b, 461a). Thus,
new information is now available about Giorgi Saakdze from the imperial court
in Isfahan, from the military camp in the frontier, from inside Georgia and the
Caucasus. Fazli’s unusual career and provenance allowed him to craft a multi-
dimensional and nuanced description.
112 Hirotake Maeda

In addition, Fazli has left various descriptions of Georgians and their culture.
He was interested in their customs and religious ceremonies as well as their in-
stitutions. The account is more generic but of course it possesses a certain im-
portance. In the spring of 1625 the shah decided to give his granddaughter to Si-
mon II, vali of Kartli. Fazli recorded the transfer of the bride from Mazandaran to
Georgia via the Caspian coast. The journey took several weeks. After the splen-
did feast of their marriage near Tbilisi, Saakadze suddenly revolted. Georgian re-
bel troops even plundered as far away as Ganja. Fazli barely escaped with his life.
Then he lost his position and never managed to return to the Caucasus (he
worked as vizier of Kirman for a short period and then, after serving Zaynal Khan
Shamlu, moved to India where he supposedly died). Thus the Saakdze’s revolt is
described by an eyewitness of the events, an author having certain dignity and
high rank, which made the report highly valuable (Fazli, Afzal III, 491b-509a).

FAZLI’S INFORMATION ON ARMENIANS AND JEWS

Fazli described the activities of not only Georgians but Armenians and Jews. One
of the most important descriptions of Caucasian peoples concerns the fate of
forcefully displaced peoples. Elsewhere I have already discussed the issue to
some extent citing Fazli’s description of an Armenian religious feast, etc (Maeda
2006). In a royal edict issued in 1619 the shah declared a state monopoly on the
silk trade (Melville 2003, 79). Fazli cited many royal edicts found only in this
chronicle. There are several descriptions on the relations of Armenian notables
with Safavid authority.
We find ample information about the activities of the influential Armenian
ghulam Qarchaqay. Iskandar Beg briefly refers to his career and gives us substan-
tial information only at the time of his death in 1625 (Iskandar Beg, ‘Alam-ara,
1039-40). Fazli wrote about his career in detail already in his first reference (Fazli,
Afzal III, 47b). Qarchaqay assassinated Zul Faqar Khan in Shirvan by the order of
the shah. After this incident he temporarily left Safavid domains and stayed in
the Northern Caucasus for a while. According to Fazli the reason for this move
was his dissatisfaction for the appointment of Yusuf Khan, mirshekarbashi, his
close friend from childhood and of the same Armenian provenance as the gover-
nor general of Shirvan. Iskandar Beg never refers to this incident but Mirza Beg
notes Qarchaqay’s defection to the Ottoman court (Mirza Beg, Rauzat, 813-814).
Whatever the truth may have been, the news of Qarchaqay’s escape and return
would have been known to contemporaries.
 New Information on the History of the Caucasus 113

Fazli possessed certain knowledge about local groups in the North Caucasus
and reveals their connections with Safavid authority. In comparison to Iskandar
Beg Fazli left a relatively detailed description of Armenians and Georgians who
became victims of the forcible transfer to the Caspian coast (Fazli, Afzal III, 367b-
370a). As he was directly involved in the Safavid policy towards the area, he was
well informed of the customs and successive histories of the local peoples.

CONCLUSION
In the third volume of Afzal al-tavarikh, we have a lot of new local information
on the Caucasus and northwestern Iran including bureaucratic and administra-
tive units. The author Fazli Khuzani narrates many local incidents as an eyewit-
ness of the events. He had an elite bureaucratic background. Thus, he was well
informed of both the central and local administrations.
For historians working on the Caucasus, Fazli’s account is of immense im-
portance. The heyday of his working career was from about 1616 to 1625 when he
was the vizier of Barda‘ and Kakheti. This was a time when the regional order
was rapidly changing and it gave great impulse to the remaking of Safavid Em-
pire. The third volume of his chronicle is dedicated to the entire reign of Shah
‘Abbas, thus the account is not limited to local information on the Caucasus. But,
because the peoples of the Caucasus were deeply involved in/suffered from ‘Ab-
bas’s state reforms, Fazli’s account provides many new insights on the local as
well as the central political society of the Safavid dynasty as it was in transition.

BIBLIOGRAPHY
Falsafi, N. (1371), Zandagani-yi Šah ‘Abbas-i avval, 3 vols. Tehran.
Fazli Khuzani al-Isfahani, Afzal al-tavarikh, Library of Christ’s College, Cambridge, Ms.Dd.5.6.
Iskandar Beg Turkman / Muhammad Yusuf Muvarrix (1317), Zayl-i tarix-i ‘alam-ara-yi ‘Abbasi, ed.
Suhayla Khwansari, Tehran.
Iskandar Beg Turkman Munshi (1350), Tarix-i ‘alam-ara-yi ‘Abbasi, ed. I. Afshar, 2vols., Tehran.
Islam, R. (1970), Indo-Persian Relations: A Study of the Political and Diplomatic Relations Between the
Mughul Empire and Iran, Tehran.
Jalal al-Din Munajjim Yazdi (1366), Tarix-i ‘Abbasi, ed. Sayf-allah Vahidniya, Tehran.
Jamburia, G. (1964), Giorgi Saakadze, Tbilisi.
Haneda, M. (1989), “La Famille Khuzani d’Isfahan”, Studia Iranica 18, 77-92.
Maeda, H. (2001), “Hamza Mirza and the ‘Caucasian Elements’ at the Safavid Court: A Path toward
the Reforms of Shah ‘Abbas I” Orientalist 1, Tbilisi: 155-171.
114 Hirotake Maeda

―― (2003), “On the Ethno-Social Background of Four Gholam Families from Georgia in Safavid
Iran”, Studia Iranica 32: 243-278.
―― (2006), “The Forced Migrations and Reorganisation of the Regional Order in the Caucasus by
Safavid Iran: Preconditions and Developments Described by Fazli Khuzani”, eds. O. Ieda;
T. Uyma, Reconstruction and Interaction of Slavic Eurasia and Its Neighboring Worlds, Sap-
poro: 237-273
―― (2011a), “Breaking through the Boundaries: From the Activities of a Georgian Mercenary who
Served the Safavid Dynasty,” Historical Studies 881: 22-33 (in Japanese).
―― (2011b), “Slave Elites Who Returned Home: Georgian Vali-king Rostom and the Safavid
Household Empire”, Memoirs of the Research Department of the Toyo Bunko 69: 97-127.
―― (2012), “Exploitation of the Frontier: The Caucasus Policy of Shah 'Abbas I”, eds. W. Floor; E.
Herzig, Iran and the World in the Safavid Age, London: 471-489.
Matthee, Rudolph P. (1999), The Politics of Trade in Safavid Iran: Silk for Silver, 1600-1730, Cambridge.
―― (2003), “Anti-Ottoman Concerns and Caucasian Interests: Dipolomatic Relations between
Iran and Russia, 1587-1639”, ed. Michel Mazzaoui, Safavid Iran and Her Neighbors, Salt Lake
City: 101-128.
Melville, C. (1998), “A Lost Source for the Reign of Shah ‘Abbas: the Afzal al-tawarikh of Fazli Khu-
zani Isfahani”, Iranian Studies 31.ii: 263-265.
―― (2003), “New Light on the Reign of Shah ‘Abbas: Volume III of the Afzal al-Tawarikh”, ed. A. J.
Newman, Society and Culture in the Early Modern Middle East. Studies on Iran in the Safavid
Period, Leiden-Boston: 63-96, 67.
Mirza Beg Junabadi (1378), Rauzat al-Safaviya, ed. Ghulamriza Tabatabai-Majd, Tehran.
Storey, C. A. (1972), Persian Literature: A Bio-Bibliographical Survey; Translated into Russian and
Revised by Yu. E. Bregel, Moskva.
Szuppe, M. (1995), “La participation des femmes de la famille royale à l’exercice du pouvoir en Iran
safavide au XVIe siècle (II)”, Studia Iranica 24: 100-102.
 Unrealized Project: Rousseaus’ Plan of Franco-Persian Trade
in the Context of the Indian Expedition (1807)

Irène Natchkebia
G. Tsereteli Institute of Oriental Studies of the Ilia State University, Tbilisi

Abstract
The paper focuses on Tableau général de la Perse moderne ou mémoire géographique, historique et
politique sur la situation actuelle de cet Empire―an unpublished manuscript by Joseph Rousseau,
a French diplomat and orientalist. Dated 1807 and kept in the archives of the Ministry of For-
eign Affairs of France, this document was compiled by Napoléon’s order, and was supposed to
give complete information about the situation in Persia, in regard with Napoléon’s planned In-
dian campaign. Although the Indian expedition was not realized, the manuscript became an im-
portant document of its epoch. It consists of fourteen chapters and covers such aspects as his-
tory, geography, administrative decision, religion, traditions, art, Fath-Ali Shah and his circle,
military potential, foreign relations, trade etc..

Keywords
Tableau général de la Perse moderne, Joseph Rousseau, Napoléon, the Indian expedition, Franco-
Persian trade relations, the trade goods

RELATIONS BETWEEN FRANCE AND PERSIA AT THE BEGINNING OF THE 19TH CENTURY
In the beginning of the 19th century Napoleon sought to ally with Persia in his ex-
pedition against India. In October 1803, he ordered the French ambassador in
Constantinople, Marshal Brune, to investigate the attitude of the court of Tehran
towards France (Dehérain 1923: 257).
For this project, the information provided by the commissar of commercial
relations of France in Baghdad, Jean-François Xavier Rousseau, born and raised
in Julfa, deserves special attention (Sacy 1810: 2-3). In a letter dated October 22th,
1804, Rousseau informs Talleyrand that he had contacted influential people in
Persia, including Shaiykh ol-Eslam of Isfahan Mirza Morteza, his longtime
friend.1 Mirza Morteza confided the eagerness of Fath-‘Ali Shah (1797-1834) to es-

1
AMFAE, MD/Perse, vol. 6, doc. 19, fol. 105r.
 Koninklijke Brill NV, Leiden, 2015
116 Irène Natchkebia

tablish relations with Napoleon. The Shah was tired of vain promises from the
English with regards to Russia and dissatisfied with the imperative request by
England to join future expeditions against the Afghans, whom the Shah did not
consider enemies.
J.-F. Rousseau argued that it could be profitable to take advantage of the
Shah’s negative attitude towards the English and drew up a plan. He called for
Napoleon to send a letter to the Shah, followed by an embassy to Tehran and
Kandahar for secret negotiations to contract a tripartite alliance between France,
Persia and Kandahar, with Fath-‘Ali Shah as the mediator between Kandahar
and France. The king of Kandahar would declare war on the English, due to the
Afghans’ hostility towards the English. The king of Kandahar would gather Af-
ghan and 70 thousand Sindhi cavalrymen, who should be joined by 2-3 thousand
French soldiers. In addition, the Marathas would join. These troops would pass
through Delhi, Agra and Lahore and seize Benares and Patna, towns on the River
Ganges with little resistance. Finally, they would continue along the Ganges to
conquer Bengal once and for all, and occupy all the places ‘suffering under the
yoke’ of England.
If Napoleon approved this plan, J.-F. Rousseau, as a person with great
knowledge of the habits and traditions of Persia, was ready ‘despite his old age’, to
leave for Tehran as Ambassador. His 24-year-old son (Joseph Rousseau–I. N.),
who knew many oriental languages (Persian, Turkish and Arab–I. N.), was pre-
pared for diplomatic service and would accompany him.2 Furthermore, J.-F. Rous-
seau would raise funds for travel and gifts for the Persian court representatives.3
This idea agreed with Napoleon’s preparation against the Third Coalition.
Despite the great contribution of Jean-Francois Rousseau in preparing the
ground for the Franco-Persian relationship, in March 1805 Napoleon sent 25-
year-old Amédée Jaubert,4 his private secretary-interpreter, and an experienced
officer in 41-year-old Adjutant Commandant Alexandre-Antoine Romieu,5 to in-

2
In many documents kept in the Ministry of Foreign Affaires of France Jean-François Xavier
Rousseau and J.-B. (Joseph) Rousseau are mentioned as Rousseau, the father, and Rousseau, the
son–I. N.
3
AMFAE, CP/Perse, vol. 8, doc. 64, fol. 182r-189v.
4
Pierre-Amédée-Émilien-Probe Jaubert (1779-1847)—famous French orientalist, diplomat and
statesman (Jaubert 1860).
5
Aantoine-Alexandre Romieu de Sorgues (1764-1805)—Officer of French National Guard
(1789), Member of the Directory of the District of Nione (1791-1793). In the years 1793-1801 he was in
 Unrealized Project: Rousseaus’ Plan of Franco-Persian Trade 117

vestigate the possibilities of an expedition against India and for negotiations

with Fath-‘Ali Shah. Romieu, a military man, was to investigate the military pre-
paredness of Persia in detail.6
From this point of view, both reports: one written by Alexandre Romieu, be-
fore his death (Tehran, 12.11.1805),7 and the other, of March 6, 1807, received by
Napoléon from Amédée Jaubert present interesting details.8 Unlike the report by
J.-F. Rousseau, Romieu evaluated the outlook of the expedition against India
negatively. Furthermore, differing from Amédée Jaubert, he doubted the benefits
of an alliance with Persia (Natchkebia 2005: 23).

J.-F. ROUSSEAU’S PROJECT ON PERSIAN-FRENCH RELATIONS

Georges Outrey, Romieu’s interpreter, brought the mentioned report of Romieu
to Baghdad. Having read this letter, J.-F. Rousseau made an attempt to neutralize
the negative impressions. In a letter to Talleyrand (May 1, 1806), he criticized
Romieu’s opinion and attributed his attitude to his ignorance of the Persian lan-
guage. J.-F. Rousseau again emphasized the need to send a French ambassador to
Persia and if the treaty with Persia was concluded, he advised that the following
be taken into consideration: 1. One of the islands of the Persian Gulf should be
handed over to France in order to eventually set up a French colony. In this case,
the Persian Gulf would be closed to the English and the French would be able to
export goods from Persia, India and Arabia to France. 2. To help the Shah form a
navy to suppress Russian trade on the Caspian Sea. 3. To start the expedition
against India from the Persian Gulf, where the Marathas would join the Afghans.
At the end of the letter J.-F. Rousseau again referred to the negative attitude of
Fath-‘Ali Shah towards the English, and noted that his son was ready to prove his
devotion to the Emperor.9
The contents of J.-F. Rousseau’s letters can be considered a program for Per-
sian-French relations. His family’s economic interests were focused in the East. It
is worth noting that Rousseau’s family had settled in Persia at the beginning of
the 18th century. Jacques Rousseau (1679-1753), J.-F. Rousseau’s father (1738-1808),

service of the Italian army, in 1802-1804—in Corfu he was the General Commisssary of the French
Trade Relations (Gotteri 1993: 412-413).
6
AMFAE, MD/Perse, vol. 8, doc. 2, fol. 45r; doc. 25, fol. 238r.
7
AMFAE, vol. 9, doc. 27, fol. 54r-60r.
8
AN, AF IV, 1686, doc. 16, fol. 1-25; Natchkebia 2002: 317-338.
9
AMFAE, CP/Perse, vol. 9, doc. 26, fol. 52r-53v.
118 Irène Natchkebia

Jean-Jacques Rousseau’s (1712-1778) uncle, and Joseph Rousseau’s grandfather

had left for Persia together with the Embassy, sent by King Louis XIV to the Sa-
favid court of Shah Soltan-Hosayn in 1705. A watch-maker by profession, Jacques
Rousseau stayed in Persia and worked as a jeweler for the last Safavids and Nader
Shah Afshar. Jacques Rousseau was entrusted by Nader Shah to choose and pro-
cess the diamonds that he had brought from India for his crown. It is also worth
mentioning, that Adil-Shah (1749), Ibrahim (1749) and Shahrukh Shah (1750) had
exempted the population of Julfa from taxation several times, due to their admi-
ration for Jacques Rousseau. From 1753, after the death of his father, J.-F. Rous-
seau engaged in commerce at Bandar-e ‘Abbas and was considered to be an ex-
perienced merchant. He doubled his father’s fortune and later arrived to Al-
Basra to join the service of France.
We can assume that the plan for the Indian Expedition was conditioned by J.-
F. Rousseau’s subjective attitude. Jean Raymond, an ex-sergeant of the East India
Company and an eye-witness, mentions that the primary concern of Harford
Johns, the Resident of Great Britain in Baghdad since 1798, was the urgent depor-
tation of J.-F. Rousseau from this town.10 After the French invasion of Egypt, J.-F.
Rousseau was put in shackles and thrown into the prison of Mardin. He re-
mained in prison for eleven months before his friend, Suleiman, the Pasha of
Baghdad (1777-1802) could liberate him (Rousseau 1899: iv-x). For this reason J.-F.
Rousseau, who served France in the Orient for four decades, was displeased with
Napoleon for entrusting others to conduct negotiations with Fath-‘Ali Shah. His
family had established a century-old contact with the Persian court, in addition
to his friends among the Ottoman high officials.11 He tried again to actively in-
volve his son in Napoleon’s Oriental policy and to secure a diplomatic career for
him. He wished to promote his son to coordinate commercial affairs of the
French in the Orient. On February 17, 1806 Pierre Ruffin,12 Chargé d’affaire of
France in Constantinople, with the help of Talleyrand, presented the Emperor
with an ode dedicated to him, penned in Persian by Joseph Rousseau, and its
French translation. J.-F. Rousseau wrote the following to Ruffin about the success
of his son in Oriental languages: “During a year he learned Persian perfectly. He
creates and writes very subtly in Turkish too. The people in Baghdad, who know

10
AMFAE, CP/Perse, vol. 8, doc. 165, fol. 399r.
11
J.-F. Rousseau became acquainted with young Agha Mohammad Khan at the court of Karim
Khan Zend (AMFAE, MD/Perse, vol. 1, fol. 139r-139v).
12
Pierre-Jean-Marie Ruffin (1742-1824)—famous French diplomate and prominent orientalist.
 Unrealized Project: Rousseaus’ Plan of Franco-Persian Trade 119

these two languages, are really surprised that a young Frenchman has learned
their different sciences so well” (Dehérain 1929/30: 258). It is known that Jean-
François Rousseau knew Persian so well that Kerim-Khan (1704-1779) himself re-
quested his recitation of the poetry by Ferdousi, Saadi, Hafez and other Persian
poets (Sacy 1810: 10).
To convince Napoleon that his son was an expert on the Orient, Jean-
François Rousseau presented his son, Joseph Rousseau as the sole author of the
work on Persia at Talleyrand’s request, though, in a letter (summer, 1805), he
wrote that he was busy writing (Gottery 1993a: 517-518). Thus, Rousseau the fa-
ther tried by all means to attract the attention of the Emperor to his son.

THE CONTENT OF THE TABLEAU GENERAL DE LA PERSE MODERN

A 222 page manuscript entitled—Tableau général de la Perse moderne ou mé-
moire géographique, historique et politique sur la situation actuelle de cet Em-
pire13—is kept in the archives of the Ministry of Foreign Affaires of France. It is
dated 1807, and signed by J.-B. (dit Joseph) Rousseau fils (J.-B. Rousseau son). The
manuscript’s Prologue consists of Rousseau-son’s French versions of Saadi’s
words on how to govern (Max. Polit. de Saadi). The prologue is followed by an
“Address” to Talleyrand, “Preface”, the translation of a Persian ode in praise of
Napoleon, and fourteen chapters. In the “Address” young Rousseau thanks Tal-
leyrand for his kindness. He is honored that Talleyrand, the greatest minister of
Europe, presented the Emperor of France “the results of his modest attempt to
describe the contemporary situation in Persia”.
In the “Preface” Rousseau notes that the “Tableau” was written in accordance
with Napoleon’s desire to obtain complete information about the situation in
Persia. He “durst to give complete guarantee” that the information collected on
geography, administration, customs, habits and commerce was reliable. He
maintained that in order to compile all the necessary information, he studied
manuscripts, spoke with competent people and relied on the knowledge of his
father Jean-François Rousseau, who had forty years of experience in the Orient.
He noted that the map of Persia, made by him, and the portraits of Persians and
Afghans in their national clothes were copied from the paintings of Persian art-
ists. We have not found this very interesting annex yet.

13
General picture of the modern Persia or geographical, historical and political memory on the
current situation of this Empire (AMFAE, MD/Perse, vol. 6, doc. 19, fol. 70r-169r)
120 Irène Natchkebia

In his odes, dedicated to the enthronement of Napoleon, written in Persian

verse, Joseph Rousseau praises the Emperor of France and calls him the liberator
of France from tyranny, whose name has reached the borders of Kandahar and
Sistan. In these odes he did not forget the most important theme for the Rous-
seaus—that of the Indian expedition. The author writes that though the English
are trembling in fear from Napoleon, they are yet to be conquered. The last para-
graph ends with ambitious notes—‘Europe waits for new victories from Napoleon
again’.14
The fourteen chapters of the “Tableau” cover various aspects of the history,
geography, historical administrative units, religion, traditions, education, art,
Fath-‘Ali Shah and his circle, the state of military affairs, foreign relations, trade,
etc. Having studied the “Tableau” we believe it was completed in the middle of
March, 1806. In August 1806 (the day is not indicated–I. N.) in the report of
Talleyrand, submitted to Napoleon, there were notes about Persia made from
the second letter (1.05.1806) of J.-F. Rousseau. At the end of this report the
Minister noted that Rousseau the father’s letter was attached to the Rousseau
son’s voluminous notes about the contemporary situation in Persia and
Romieu’s writings.15 Because of this we suppose that the date, indicated in the
“Tableau”—1807, is the date of its being received in the Archives of the Ministry
of Foreign Relations of France.
While rich in subjects, we will focus on those related to the theme of the
Indian expedition—the trade between France and Persia via Baghdad to India.
In this work young Rousseau often repeats the profit to be gained from the
Indian expedition, and his ideas about the need of sending a magnificent
embassy to Fath-‘Ali Shah for secret negotiations. He notes that, according to the
letter of Prince ‘Abbas Mirza to his father, and the letter of Fath-‘Ali Shah’s
trusted vezier Mirza-Riza Khan to Talleyrand, the Shah desired to establish
friendly relations with the Emperor and receiving a plenipotentiery
representative of France for establishing close contacts between these two
Empires. According to Joseph Rousseau’s explanation, at a court of Asian
monarchs, nothing evoked such respect as the magnificence and brilliance of
diplomatic representations. An Eastern ruler to whom a European monarch
stretched out his friendly hand and sent precious gifts, would be pleased by this

14
Ibid., fol. 65r-67r.
15
AMFAE, CP/Perse, vol. 9, doc. 41, fol. 83r-83v.
 Unrealized Project: Rousseaus’ Plan of Franco-Persian Trade 121

courtesy and obliged to agree with all the propositions. For this reason, an
ambassador with knowledge of Persian etiquettes and language should be taken
into consideration. This would be of great advantage, because he would not need
a dragoman, whose translations often were vague and inaccurate.16 Thus, young
Rousseau unequivocally offered to the Government of France either his or his
father’s service.

NOTES ABOUT TRADE OF PERSIA

Information and considerations regarding future trade relations between France
and Persia are presented in the XII chapter of the “Tableau”.17 Joseph Rousseau
noted that trade in Persia was such a respectful and profitable business that
great seigniors and even kings did not shy away from trading with merchants.
The author explained that trade could even offer one protection, as ancient
superstitions considered a merchant’s person as sacred. The merchant took care
of satisfying his compatriots’ demands. And those travellers, who arrived to
Persia with important trade plans also ‘deserved the title of the Shah’s personal
guest’.18 The young Rousseau underlines the fact that Fath-‘Ali Shah habitually
chooses ambassadors to be sent to foreign courts from among merchants.19 By
mentioning this, he alluded to the advantage of him being sent to Tehran.
The rule of Shah-‘Abbas I is the symbol of mighty Persia for Joseph Rousseau.
In different chapters of this work he refers to this sovereign as ‘great’ and
‘immortal’ and of his rule as ‘peaceful’ and ‘happy’.20 In his opinion, Fath-‘Ali Shah
was trying to return the glory of Shah ‘Abbas I to Persia, and this gave the hope
that trade and handicraft would regain their old importance. The author saw the
foundation for this progress in the extensive territory of Persia and its richness in
natural products and merchandises. However, having said this, he noted that
Persia’s wealth did not compare to India. It seems that for both Rousseaus, India
was the real prize.
Rousseau the son hoped that with improving administration, Persians would
become the richest and mightiest of all Asian peoples. He repeated his father’s

16
AMFAE, MD/Perse, vol. 6, doc. 19, fol. 139r-140r.
17
Ibid., fol. 152v-158v.
18
Alexandre Romieu noted in his letter that Fath Ali Shah patronized foreign merchants, and
they glorified him. He was known as an honest man who was able to keep his word (AMFAE,
CP/Perse, vol. 9, doc. 27, fol. 57r).
19
AMFAE, MD/Perse, vol. 6, doc. 19, fol. 152r-153r.
20
Ibid., fol. 67v, 74r, 77v, 81v, 82r, 87r, 101v, 152v, 165r.
122 Irène Natchkebia

opinion that the Persians had no navy and used foreign ships for trade. Aiding
Fath-‘Ali Shah create a commercial fleet would contribute to the success of his
subjects in trade. This was realizable, because the Shah’s treasury was full, Persia
had forests for wood, and the Shah only needed competent European naval
architects.21
The trade center of Persia was Tehran, its capital. Every year caravans moved
to and fro, arriving from Kandahar, Lahore, India and Turkey. Rousseau specified
the goods, imported from Europe to be sold in Persia. Imported goods included:
woolens, twill, silk fabrics and fabrics decorated with golden and silken threads,
glass, arms, metals, paper, fancy goods, jewelry, hardware, braids and Cashmere
shawls. Moreover, American goods—cochineal, indigo, etc.—were also im-
ported. The goods exported from Persia were: silk, tobacco, canvas, china, dried
fruits, wine, vodka (eau distillées–J. R.), leather, Persian lamb skin, furs, all kinds
of fabrics, carpets, fancy cloths, felt, mat, nut-gall, roots of plants, drogues, pre-
cious resin, lapis lazuli, and turquoise. Jewels from India were an important
trade, as were pearls from Bahrain. Such trade was very lucrative.
In spite of the strained relations with Turkey, Afghanistan and the war with
Russia, Persian merchants still imported many Indian goods, including the most
beautiful and precious Kashmir shawls; from Baghdad, they brought European
goods such as sextants, achromatic eyepieces and telescopes. The monopoly of
trade with Persia via the Caspian Sea was in the hands of Russian merchants.
They exported different kinds Persian goods and Indian raw materials from Ma-
zandaran to Astrakhan. In addition to this, the author indicates that, unlike Eu-
rope, in Persia there were neither exchange markets, nor banks, and in order to
transfer their money, merchants used bills of exchange and, in internal and for-
eign trade, services of Hebrew, Indian and Armenian brokers.22
For Joseph Rousseau, the main purpose of the treaty between the Emperor of
France and Fath-‘Ali Shah was the facilitation of Franco-Persian trade. He hoped
that France would take profit away from England. He argued that trade through
Persia and the Persian Gulf to India would be more profitable than that around
the Cape of Good Hope. It would be necessary to establish storehouses in Mar-

21
Ibid., fol. 153r-154r. According to the information provided by Lieutenant Trézel, a member of
the mission of General Gardane, the Minister Plenipotentiary of France in Persia in the years 1807-
1809, the Persians were tolerant towards the Arabs, who lived in the Persian Gulf only because they
had no fleet and could not use the ports (AMFAE, MD/Perse, vol. 3, doc. 3, fol. 36r).
22
AMFAE, MD/Perse, vol. 6, doc. 19, fol. 76v, 89v, 93v, 95r, 113r, 154r-155v 166r.
 Unrealized Project: Rousseaus’ Plan of Franco-Persian Trade 123

seille, Aleppo, Baghdad and if needed, in Al-Basra. In this last town it was possi-
ble to acquire Indian goods at very convenient prices via Persian, Turkish and
Arabian resellers.
Transportation of French goods from Marseille to Alexandrette (Iskenderun)
was possible in 20-25 days. From there to Aleppo by caravans would take only 7-
8 days. From Aleppo to Baghdad, a route of 180 leagues,23 caravans should be ac-
companied by an escort of 250-300 armed people to protect against Wahabi at-
tacks. Carrying goods from Baghdad to Persia would be easy. However, in Rous-
seau the son’s opinion, Persian merchants would flock to Baghdad to purchase
French goods.24
Joseph Rousseau also calculated the possible profit of this trade for French
merchants at 70% to 100% profit. New networks would connect French trade
with India. He proposed that this trade should be led by a company, specially
created for this purpose and to whose shareholders the Government of France
would give exclusive, comparatively long-term privileges. The initial capital nec-
essary for starting this project should not be less than 6 millions of livres.25 For
raising this money it would be necessary to issue 2000 actions, with share de-
nomination of 3000 livres each.26
Joseph Rousseau repeated his father’s idea, “I think soon I’ll submit to His Ex-
cellent the Minister of Foreign Affairs, my second detailed essay, the subject of
which will be the trade via the Persian Gulf. I’ll explain the advantage of this
trade for the nation and the importance of establishing a permanent settlement
there. I’ll prove that the best thing we can obtain is Karg Island, because for a
long time it was occupied by the Portuguese and the king of Persia will cede it to
France without complications. Karim Khan agreed with this demand of the
government of France of the old regime.27 The original of the statement of this
cession shall be kept in the chancellory of the Ministry”.28 We have not yet found
this manuscript in the archives of Paris.

23
League was old french unit of distance, which is about 4 km. 180 leagues were about 720 km.
24
AMFAE, MD/Perse, vol. 6, doc. 19, fol. 156v-157v.
25
Livres a French monetary unit, which was replaced by Frank in 1799. Arguably, Joseph Rous-
seau, who lived in the east, instead of Franks, took into consideration livres.
26
AMFAE, MD/Perse, vol. 6, doc. 19, fol. 157v-158r.
27
The rule time of Louis XVI (1754 –1793).
28
Ibid., fol. 158v.
124 Irène Natchkebia

In a document kept at the Ministry of Foreign Affairs of France, we read this

assessment on the Rousseaus—“The Republic of France in the East had the
agents, whose thoughts were completely different from appreciations of Olivier
and Brugnières… Our Consul in Baghdad, citizen Rousseau and his son said oth-
erwise. They, who were grown up in the East, had well adopted eastern tradi-
tions, elevated style and inborn respect for the king’s power. (…) They consid-
ered the king of Persia quite strong, so that he would be able to eliminate do-
mains of the English in India”.29
Napoleon considered it urgent to organise an expedition against India. On
April 12, 1807 he assigned Brigadier-General Claude Mathieu Gardane as Minister
Plenipotentiary to the court of Persia (1766-1818), his brother, Paul-Ange-Louis
Gardane, as first secretary, and Rousseau the son as second secretary.30 The Rous-
seaus were disappointed again. The Gardane brothers were grandsons of Ange
Gardane, Consul of France, who lived in Isfahan from 1718 until 1730 (Touzard
2002: 297-316). It seems Napoleon gave the assignment of Gardane, his aide-de-
camp and experienced general as ambassador to add glory to the mission in the
eyes of Persians, who had great respect for military men and, on the other hand,
the appointment of the Gardanes as highest functionnaries, was aimed at
emphasising former diplomatic relations between France and Persia.
In December 1807 Joseph Rousseau met the French mission in Tehran. In De-
cember 1807 Fath-‘Ali Shah honored Joseph Rousseau with Decorations of the
Sun of the Second Degree, and conferred upon him the hereditary title of Khan.31
In May 1808 Rousseau left Persia for Aleppo due to his father’s death (Rousseau,
1899: xvi). He realised his father’s dream—the expedition to India—would not
bear fruition.

CONCLUSION
The goals of the Rousseaus, penned in their Tableau général de la Perse moderne
ou mémoire géographique, historique et politique sur la situation actuelle de cet
Empire, includes the expedition against India, domination of the Persian Gulf,
and increased French trade. A second book by Rousseau, Notice historique sur la
Perse ancienne et moderne et sur ses peuples en général edited in 1818 in Marseille,

29
In the opinion of French envoys, conclusion of an alliance with Persia for accomplishing
great plans ‘would be equal to relying on the slush’ (AMFAE, MD/Perse, vol. 8, doc. 2. fol. 40r-40v).
30
AMFAE, CP/Peres, 9, doc. 85, fol. 184r-184v.
31
AMFAE, CP/Perse, vol. 11, doc. 105, fol. 173r-173v.
 Unrealized Project: Rousseaus’ Plan of Franco-Persian Trade 125

differs from the “Tableau” which has never been published and stayed beyond
the attention of researchers for two centuries. The “Tableau”, the first work about
Persia written in French after the enthronment of Fath-‘Ali Shah, enables us a
social-economic picture of Persia in the first decade of the 19th century.
Although Jean-Baptist-Louis-Jacques (Joseph) Rousseau (1780-1831) did not
receive the post of Ambassador of France in Persia, his subsequent diplomatic
career and scientific activities were quite successful. He was Consul in Aleppo
(1808-1814) and Consul-General in Baghdad (1814-1824). He collected many man-
uscripts. Their historical and scientific significance was highly appreciated by
the Count Sergey Semionovich Uvarov—President of the Russian Academy
(1818-1855). He bought 500 manuscripts from Rousseau for the Imperial Academy
of St. Petersburg. Joseph Rousseau’s diplomatic career continued in North Africa.
On December 15, 1824 he was appointed Consul-General and Charges d'Affaires
of France in Tripoli. For his scholarly activities in Oriental studies, he was elected
the correspondent of the l’Académie des Inscriptions et Belles-Lettres of France. In
addition to this, for the years of faithful diplomatic service for France in the Ori-
ent, the government of France conferred upon him the title of Baron. Joseph
Rousseau died in 1831 in Tripoli, at the age of 51.

BIBLIOGRAPHY
Archives du ministère français des Affaires étrangères (AMFAE): Correspondance politique (CP),
Perse, vol. 8: 64, 165; vol. 9: doc. 26, 27, 41, 85; vol. 11, doc 105.
Archives du ministère français des Affaires étrangères (AMFAE) : Correspondance politique (MD),
Perse, vol: 3, doc. 3; vol. 6, doc. 19; vol. 8: doc. 2, 25.
Archives Nationales (AN), Relations extérieures, Perse, AF IV, 1686, doc. 16.
Debidour, A. (1904), Le général Fabvier, sa vie militaire et politique, Paris.
Dehérain, H. (1923), “Letters inédites de membres de la mission Gardane”, Revue de l’histoire des
colonies Françaises 15.1: 247-282.
―― (1929-1930), La vie de Pierre Ruffin, orientaliste et diplomate, vol. I-II, Paris.
Gaffarel, P. (1908), La politique coloniale en France de 1789 à 1830, Paris.
Gotteri, N. (1993), “Antoine-Alexandre Romieu (1764-1805), général et diplomat”, Revue dromoise,
archéologie, histoire, géographie 88.468: 411-456.
―― (1993a), “Antoine-Alexandre Romieu (1764-1805), général et diplomat”, Revue dromoise, ar-
chéologie, histoire, géographie 88.469: 476-564.
Hellot-Bellier, F. (2007), France–Iran, quatre cents ans de dialogue. Avec le concurs de L’UMR 7528
“Mondes iranien et indien”, Paris.
Jaubert, P. Am. (1860), Voyage en Arménie et en Perse. Précédé d’une notice sur l’auteur par M. Sédil-
lot, Paris.
126 Irène Natchkebia

Natchkebia, I. (2002), “La Perse vue par deux émissaires de Napoléon”, Studia iranica 26.2: 317-338.
―― (2005), “Unfinished Project: Napoleon's Policy in Persia in the Context of the Indian Expedi-
tion and Georgia”, Qajar Studies 5: 17-39, Rotterdam/Santa Barbara/Tehran.
Rousseau, J.-B. Louis Jacques (1899), Voyage de Bagdad à Alep (1808). Publié d’après le manuscrits
inédit de l’auteur, par Louis Poinssot, Membre de la Société de Géographie de Paris, Paris.
Sacy, Sylvestre de, (1810), Eloge historique de feu Jean-François Rousseau ancien consul général de
France à Bagdad et à Bassora, mort à alep le 12 mai 1808. Précédé de quelques détails curieux
et intéressans sur le voyage de son père à la cour de Perse, au commencement du dix-huitième
siècle.
Touzard, A.-M. (2002), “Un drogman Grec consul de France en Perse”, Studia iranica 26.2: 297-316.
 Unrealized Project: Rousseaus’ Plan of Franco-Persian Trade 127

Illustrations from Tableau général de la Perse... by Joseph Rousseau, 1807

Archives du Ministère français des Affaires étrangères,

Mémoires et Documents, Perse, vol. 6, doc. 19, fol. 152r.
128 Irène Natchkebia

Archives du Ministère français des Affaires étrangères,

Mémoires et Documents, Perse, vol. 6, doc. 19, fol. 156v.
Unrealized Project: Rousseaus’ Plan of Franco-Persian Trade 129

Archives du Ministère français des Affaires étrangères,

Mémoires et Documents, Perse, vol. 6, doc. 19, fol. 158v.
 Nadir’s Religious Policy Towards Armenians

Roman Smbatian
Yerevan State University

Abstract
This paper aims at shedding light on one of the most sensitive aspects of Nadir Shah’s policy to-
wards Armenians. Nadir’s religious policy towards Armenians derives from his overall religious
worldview and political expediency of taking control over Armenia and the Caucasus as a whole.
Before the rise of Nadir Shah, eastern Armenia and most of the Caucasus was under the yoke of the
Ottomans and the Caspian western littoral was controlled by Russians. Both the Ottomans and the
Russians introduced themselves as the protectors of their coreligionists in the region: the Otto-
mans of the Sunnis and the Russians of Armenians and Georgians. Thus Nadir realized the im-
portance of religion in politics and trying to legitimize his invasion of the region, the Iranian Shah
adopted tolerance towards all religions and ethnicities, especially Armenians by honoring the Ar-
menian Archbishop Abraham Kretatsi and favoring special status to the church of Ejmiatsin and
Armenian Meliks. Nadir realized their significant role in local partisanship and defeating the Ot-
tomans. Nadir also opposed European missionaries in their efforts to convert Armenians and
aimed at closely recognizing Christianity and other religions of his empire by ordering to translate
Bible, Quran and Torah into Persian. In conclusion the current paper states that there was a com-
plete convergence of interests between the Shah and the Armenian religious and political leader-
ship which resulted in strong Armenian-Iranian cooperation in defeating the Ottomans as well as
re-establishment of Iran’s control over eastern Armenia and the Caucasus.

Keywords
Nadir Shah, Armenians, Religious Policy, Armenian Archbishop, Meliks, Convergence of Interests

NADIR’S RELIGIOUS WORLDVIEW

Many observations have been made about Nadir’s attitude towards religion and
his religious policy. Some characterize him as a Sunni taking into consideration
his Turkmen ancestors (Astarābādī 1962-1963: 269; Ša’banī 1970: 1132); the famous
contemporary historian Mohammad Kazem Marvi due to Nadir’s demonstrative
religious actions, such as visiting Najaf and Karbala, renewing the tomb of Emam
Reza in Mashhad and also the naming of his sons, considers him to be a devoted
Shiite (Marvī 1979: 826-827, 921-925). Hanway and Kishmishev thought he was

 Koninklijke Brill NV, Leiden, 2015

132 Roman Smbatian

inclined to found a new world religion (Hanway 1986: 262-265; Kišmišev 1889:
213). Bazin believes that the only thing Nadir believed was his own personality
and wasn’t a follower of any religion, merely using it as means to reach his per-
sonal goals (Bazin 1961: 42).
According to the studies of American orientalist Ernest Tucker, for Nadir re-
ligion was an important factor to legitimize his political power both in Iran and
in relation to foreign states (Tucker 2006: 2-12). The newest and the most realistic
opinion on this issue is suggested by Michael Axworthy, who emphasizes the
physiological factor of the formation of Nadir’s religious worldview. He mentions
the fact that Nadir was fatherless from the early age, the widowhood of his moth-
er and indifference of Muslim clergy towards their state could have influenced
negatively on his attitude towards religion (Axworthy 2006: 20). It should be
added that another important event that could have affected his religious views
was his captivity by Sunni Uzbeks in his childhood. His life experience taught
him that religious rivalry and even enmity between Iran and its Sunni neighbors
had direct repercussions both for the domestic and foreign affairs of the state
and only by solving of these disputes could Iran become a prosperous country.
Taking into consideration all the above-mentioned opinions, a conclusion
can be made that Nadir believed at least in a Deity and his actions in this sphere
cannot be considered to only be demonstrative.

NADIR’S POLICY TOWARDS THE ARMENIAN CHURCH AND MELIKS

When Nadir, for the first time, in 1734 launched a military campaign to retake the
Caucasus from the Ottoman and Russian Empires, he pursued policy of religious
tolerance towards all non-Muslim ethnic groups, especially the Armenians. Ac-
cording to the Persian documents kept in Matenadaran that relate to this period,
there was complete convergence of interests between the Armenian Church of
Ejmiatsin and the new ruler of Iran. Contemporary Armenian historians such as
Abraham Kretatsi and Abraham Yerevantsi also prove this point (Abraham
Kretatsi 1870; Abraham Yerevantsi 1977). Prior, Armenian saghnaghs1 of Zangezur
and Gharabagh had established relations with Shah Tahmasb II. These relations
turned into real cooperation only in 1727, when the leader of the Armenian na-
tional resistance movement, David Beg, began negotiations with Shah Tahmasb’s
officials (Leo 1946: 692). Both Tahmasb and David Beg realized that the Otto-

1
Arduous mountainous fortifications.
 Nadir’s Religious Policy Towards Armenians 133

mans were their common enemy and that they should unite their forces of war
against them. Thus the roots of the cooperation between the Armenian Meliks
and Iranian court extend prior to the period of Nadir’s joining Tahmasb.
I In 1734, Armenian Meliks hurried to help Nadir and put their forces under
his service. Abraham Yerevantsi mentions that during the campaign against the
Ottoman army of Tupal Osman Pasha there were 6 Armenian regiments in Na-
dir’s forces, under the leadership of Armenian commanders (Abraham Yere-
vantsi 1977: 69-70). In return, Nadir treated Armenian Meliks respectfully. For in-
stance his attitude towards Melik Shahnazar of Gegharkunik, who joined the
Iranian army immediately after Nadir entered Armenia, was full of respect and
the ruler of Iran not only reestablished and recognized all his political and eco-
nomic rights, but even advised all Armenian Meliks to keep solidarity (Hay Zho-
ghovrdi Patmutyun, 1972: 182).
Iranian contemporary historian Marvi mentions that one of the most influen-
tial Armenian Meliks named Yegan, advised Nadir about the importance of
having good relationship with the Armenian Archbishop (Marvī 1979: 410). Nadir
realized the importance of local partisanship to his success, when he launched
his military campaigns. The Armenian Archbishop, Abraham Kretatsi, states that
due to the recommendations of Armenian Meliks, he went to welcome Nadir as
he reached Ejmiatsin and gained his respect. Nadir attended a mass in Ejmiatsin
church and followed it with special interest (Abraham Kretatsi 1870: 198-200).
Before leaving, he ordered famous masters to renovate the church and dedicated
one thousand tumans for the work. He also urged his officials to bring some fa-
mous carpets from Kerman for the church (Marvi 1979: 410) and told the Arch-
bishop: “Never worry and fear of anything. This is our home, you are a good old
man and may your church thrive and flourish” (Abraham Kretatsi 1870: 202).
At the request of Abraham Kretatsi, Nadir issued new firmans to reaffirm the
rights of the church on its properties (Matenadaran, Archbishop’s Archive, cata-
logue №29/1g, document № 353). He also issued another firman according to
which taxes paid by church remained on the level of previous years (ibid, docu-
ment № 358).
Thereafter when Kretatsi accompanied Nadir to Tiflis, the Iranian ruler gath-
ered together all the khans, Meliks, kalantars and kadkhodas of the region and
ordered them to support the church of Ejmiatsin and never interfere with the
rights of the Archbishop. He especially warned all the officials of Armenian re-
gions to fully obey the Archbishop and consult his opinion (Leo 1946: 70).
134 Roman Smbatian

Kretatsi mentions in his book that Nadir was satisfied with the Armenians and
commanded the officials to never interfere with them, warning those who don’t
obey his order would be punished accordingly. He also ordered that except for
the jezye, Armenians should pay only half of the amount of all taxes levied from
Armenian regions and the other half to be paid by the Muslim inhabitants of
those regions. Trying to regulate all his orders Nadir issued 15 new firmans on the
above-mentioned matters that aimed at improving the situation in the regions
inhabited by the Armenians (Abraham Kretatsi 1870: 210-212). Most of these fir-
mans are related to the spheres under the Ejmiatsin Church’s authority, includ-
ing marriage, divorce, inheritance (Matenadaran, opt. cit., documents № 374,
378, 379).
The study of the documents that relate to the problems of inheritance and
marriage shows that Nadir was concordant with Kretatsi and prohibited all
Muslim officials and regional rulers from compulsory marriage and conversion
of Armenian girls (ibid, documents № 370, 377). According to another firman is-
sued by Nadir, those Armenians converting into Islam, had no right to inherit
property from their Christian families. One document states that no Muslim of-
ficial without permission from Nadir could enter Ejmiatsin (Abraham Kretatsi
1870: 200).
Despite his kind attitude towards the religious and political institutions of
the Armenians, Nadir ordered 600 Armenian families from Tiflis and the Ararat
region to expatriate to Khorasan. As a result of Kretatsi’s request not to under-
take that action, Nadir exiled only 300 families from the Ararat region (Leo 1946:
70).
Nadir also expatriated a large number of Armenians from the Van region to
Khorasan in 1735, when he laid the siege of Kars (Abraham Yerevantsi 1977: 83).
With this step Nadir aimed at building a town in Mashhad inhabited by Armeni-
ans and named New Nakhijevan, like that of Shah Abbas’s New Julfa in Isfahan.
However the Armenian representative of the Ottoman Empire Tambur Kuchuk
Arutin, who visited Mashhad, says that not a single wealthy Armenian lived
there and the church that was built by the order of Nadir had no roof (Tucker
2006: 68-71). The reason of Nadir’s failure to build prosperous Armenian town in
Mashhad was that most Armenians brought to this city were villagers and had
no skills in trade and craftsmanship.
Nadir didn’t oppose the policy of expatriation, but regional leaders and
khans, who saw opportunities with such policies in gain through the selling of
 Nadir’s Religious Policy Towards Armenians 135

these exiled people as slaves, pursued it most actively. In the Mughan plain dur-
ing the quriltai when Nadir was declared Shah of Iran, some 700 Georgians and
Armenians slaves were presented as gifts to the Iranian nobles. Thus it seems
Nadir had no special regard towards all Christians of his Empire, but rather only
with religious and political grandees. This is why it is not surprising that some
sources state that Armenian Meliks helped Nadir not only by providing military
forces, but also by levying taxes and supplying provisions (Leo 1946: 181-182).
At any rate Nadir remembered Abraham Kretsatsi’s support during the wars
against the Ottoman Empire and vowed to always protect the church and the
Archbishop (ibid: 182). With Nadir’s support, Armenian Meliks of Gharabagh
strengthened their position and in late 1730s, they founded a new administrative
organization called Moluke Khamsa2 that united five Armenian Melik authori-
ties of Gharabagh (ibid: 185-186). Nadir assigned the most influential Armenian
Melik named Yegan as a head of Moluke Khamsa. Yegan gained all the rights of
an administrative ruler, as a biglerbeg, the local head of a region in the Iranian
state, who was in charge of collecting taxes (Mirza Sami’a 1998: 186).
Moreover in recognition of Armenian Meliks’ services, Nadir reaffirmed all
the political and economic rights of Meliks of Gharabagh and Yerevan and even
granted them greater rights than local Muslim leaders (Abraham Kretatsi 1870:
244).
More than the Armenian Meliks, it was Archbishop Abraham Kretatsi who
was held in great esteem by Nadir. Simeon Yerevantsi, a later Archbishop, re-
flected that Nadir treated Abraham Kretatsi very kindly and as long as the Arch-
bishop was alive the church of Ejmiatsin was protected and kept all its privileges
(Simeon Yerevantsi 2003: 209). When Nadir introduced the new biglerbeg of Ye-
revan to the Armenian Archbishop, he said to him: “Pirmohammad Khan is not
aware of the affairs of this region and you are in charge of informing and helping
him during his reign and I assign you as the head of all aghayan of Iravan” (Abra-
ham Kretatsi 1870: 243).
Nadir’s policy towards Muslims and especially Shiite clergymen differed
from that of the Armenian Archbishop. After his crowning, he ordered the con-
fiscation of much property of the Shiite clergymen and levied taxes upon the
seyyeds. Nadir even mentioned that if people needed the clergy’s support, they
should pay for their expenses not the state (Axworthy 2006: 171). By this, Nadir

2
Five Melik authorities.
136 Roman Smbatian

sought to gain additional financial sources for his army and military campaigns.
In contrast, the Armenian Church was much tolerated. According to one of his
firmans, the Church of Ejmiatsin could have properties (moqufāt) even outside
the Yerevan region and not only could repair old churches, but build new ones,
which was generally prohibited under Islamic laws (Matenadaran, opt. cit., doc-
uments № 886).
It is interesting that these rights were granted only during the rule of Abra-
ham Kretatsi and didn’t relate to the archbishop as an institution in general. Na-
dir’s relationship with the next Armenian Archbishop, Ghazar Jahketsi, was ra-
ther tense (Banber Yerevani Hamalsarani 1994: 94) and he even forced his resig-
nation in 1744 (Simeon Yerevantsi 2003: 44-45).
It should be mentioned that during the reign of Nadir, the situation of Arme-
nians in Eastern Armenia and the Caucasus was much better than that of the
Armenians of Isfahan.

NADIR’S AIM TO GET ACQUAINTED WITH DIFFERENT RELIGIONS AND HIS DISAPPROVAL
OF THE ACTIVITIES OF FOREIGN MISSIONARIES WITH THE ARMENIANS
Nadir realized the important role of religion in ruling and, although he was not a
religious person, he was eager to know the main religions of his empire. To fulfill
his goal Nadir ordered translations of the Bible, Quran and Torah into Persian
and assigned his official historian Mirza Mehdi Khan Astarabadi in charge of the
work (ibid.: 91). Mirza Mehdi formed a group of religious experts, consisting of
eight Armenian priests, four orthodox and four Catholic, four Muslim and four
Jewish clergymen. They translated the three holy books into Persian and brought
it before Nadir Shah. After becoming acquainted with their contents, Nadir is
said to have been surprised and asked why all the religious rivalry and disputes,
when all three books shared a monotheistic God. Hanway mentions that Nadir
was contemplating the founding of a new world religion that would be complete
and solve all the disputes between various religious identities (Hanway 1986:
262). But this statement of Hanway is denied by a newly found source in Mate-
nadaran called “Qnnikon”. The author of this manuscript is one of those Arme-
nian priests, who participated in the work of translation. He states that after they
had finished the work at Nadir’s court in Qazvin, the Shah praised the clergy-
men, and said that as God is one, the holy book also should be one, so all three
holy books are the same. Then he gave a thousand tumans and six hundred
golden coins to each and dismissed them. The author mentions that Nadir’s
 Nadir’s Religious Policy Towards Armenians 137

main goal was to get acquainted with these religious books (Banber Yerevani
Hamalsarani 1994: 92). Taking this into consideration, it can be concluded that
Nadir was curious about these religions, but he had no aim to create a new reli-
gion.
Another issue that can be considered to be important in the context of Na-
dir’s religious policy towards the Armenians is his negative attitude towards Eu-
ropean Catholic missionaries proselytizing among the Armenians. There are two
documents in Matenadaran that relate to this issue and both are dated to 1741.
According to these raqams, the Armenian Archbishop Ghazar Jahketsi com-
plained before Nadir that European missionaries, by cheating and dividing
money, are making many Armenians in Tiflis, Ganja, Nakhijevan, Tabriz, Esfa-
han, and Hamedan convert to their religion and immigrate to European coun-
tries. Nadir’s reaction was clear. He condemned those actions and issued a fir-
man stating no missionary could interfere in the religious affairs of the Armeni-
ans (Matenadaran, Archbishop’s Archive, opt. cit., documents № 375, 380). An
important related factor was economic interests, as many European missionaries
aimed at spreading their influence among the Armenians, whom they consid-
ered to be their eastern allies, and believed by having them immigrate to Euro-
pean countries, the Armenians would transfer their financial wealth from Iran
and the Caucasus to Europe.

CONCLUSION
During the reign of Nadir, religion was regarded as an important source for
power, legitimacy and to regulate relations among various ethno-religious com-
munities. Nadir pursued a pragmatic and realist religious policy toward the
Christians and especially the Armenians, whom he considered to be his natural
allies against the Ottoman Empire. Religious tolerance, a respectful attitude to-
wards the Armenian Archbishop and Holy Ejmiatsin in general were diplomatic
acts, as Nadir Shah in rallying the support of Armenian Meliks and Archbishop.
Nadir believed the Armenians were reliable and would not betray him in war
against the Ottomans. Caucasian Armenia bordered Anatolia and was of great
importance to Nadir’s plans to conquer the Caucasus. As for the Armenian Me-
liks, they needed the support of a strong Iranian ruler to oust the Ottomans from
the region. Thus it can be stated that during the reign of Nadir, there existed a
convergence of interests between the Iranian Shah and the Armenian political
and religious nobility that resulted in strong cooperation and the re-establish-
138 Roman Smbatian

ment of Iran’s power in the region, as well as the affirmation of the previous po-
litical, economic and religious rights of the Armenian Church and Meliks.

BIBLIOGRAPHY
Abraham Yerevantsi, A. (1977), Patmutyun Paterazmtsn 1721-1736 tui, patrastets Sahak Mkhitaretsi,
Venetik-Surb Ghazar.
Abraham Kretatsi (1870), Patmagrutyun Antsitsn Yurots ev Nadr-Shahin Parsits, Vagharshapat.
Astarābādī, M. M. (1962-1963), Jahāngošāy-e nāderī, ed. S. ʿA. Anwār, Tehran.
Axworthy, M. (2006), The Sword of Persia: Nader Shah from Tribal Warrior to Conquering Tyrant,
London- New York.
“Banber Yerevani Hamalsarani”, hasarakakan gitutyunner, Arandznatip # 2(83), 1994.
Bazin, P. (1961), Nāmeha-ye Tabib-e Nāder Šāh, Tehran.
Hanway, J. (1986), Zendegi-e Nāder Šāh, Tehran.
Kišmišev S. O. (1889), Poxody Nadir Šaxa v Herat, Kandagar, Indiyu i sobytiya v Persii posle ego smer-
ti, Tiflis.
Leo, (1946), Hayots Patmutyun, v. III, Yerevan.
Marvī, M. K. (1979), Alamaraye Nāderī, Tehran.
Mirza Sami’a, M. S. (1998), Tazkirat al-Moluk, Tehran.
Simeon Yerevantsi (2003), Jambr, Yerevan.
Ša’banī, R. (1970), “Siāsat-e Mazhabi-e Nāder Šāh Afšār”, Vahid 9.
Tucker, E. (2006), Nadir Shah’s Quest for Legitimacy in Post-Safavid Iran, Gainesville.
Hay Zhoghovrdi Patmutyun,(1972) v. 4, Yerevan.
Matenadaran, Archbishop’s Archive, catalogue №29/1g, documents № 353, 358, 370, 374, 375, 377,
378, 379, 380, 886.
 The Song Unveiling the Hidden

Victoria Arakelova
Yerevan State University

Abstract
The paper discusses a Yezidi qawl recited by a sheikh both during the procedure of revealing
sacred objects and as part of the funeral ceremony. It is an attempt to analyse the parallel be-
tween a sacred object and a corpse as two phenomena, which do not belong to the profane real-
ity.

Keywords
Yezidi qawl, Sacred Objects

Qawlē Ēzdīdī Ĉuk,1 “The Small Song of [Sultan] Yezid”,2 is a text of considerable
interest both in terms of obscure passages requiring analysis and, particularly, in
terms of its specific use. This qawl recited by a Yezidi sheikh, accompanies the
procedure of revealing, on a special occasion, a sacred object usually kept con-
cealed from extraneous view.3
The general meaning of the qawl, apart from praising Sultan Yezid in a typical
maddāḥī (praise of god or rulers) manner, is an appeal to holy men to take out
the chained hula4 “from the treasury of strength”. What is peculiar is that the per-
formance of this qawl is part of the Yezidi funeral ceremony. A sheikh recites it
when the dead body is being removed from the house.

1
The Text tof the qawl without translation and commentaries was published in Nure Jawari
(1983): 63.
2
On this character, his place in the Yezidi triad, as well as his genesis in the Yezidi tradition see
Arakelova 2008.
3
Such an object, e.g., can be the sanjak – a stand with a fixed peacock on it, symbolysing Ma-
lak-Tawus, the Peacock Angel, the central figure of the Yezidi religion (See on him Asatrian/Arake-
lova 2004a).
4
Hula – kind of a woolen robe, similar to xirqa traditionally worn by the Yezidi faqirs and ap-
proached as a sacred object in the Yezidi tradition.

 Koninklijke Brill NV, Leiden, 2015

140 Victoria Arakelova

The relation between the content (the appeal to the holy) and the context
(the removal of the corpse) seems rather curios. However, there are at least two
levels connecting a corpse and the mentioned objects. First, both do not belong
to the profane world: sacred objects per se represent the sacral sphere, while the
corpse is no longer intramundane and thus cannot be identified with the
“thingish” reality. The second level is the idea of concealment itself. Both the
mentioned objects and the dead body, as parts of the non-profane sphere, are
supposed to be hidden.
For further evidence we refer to another Yezidi text, Du’āya t’asmīlī a’rdē,5 the
prayer of committal to the earth addressed to Dawrēš-a’rd (“Saint (Master or
Host) of the earth”), the Yezidi lord of the earth, whose important function is the
preservation of people’s possessions either hidden or pawned. Yā Dawrēšē-a’rd,
tasmīlī ta av āmanata, “O lord of the earth, (I commit) to you this pledge”, is the
formula uttered when hiding some object in a secret place (see Asatrian/Arake-
lova 2004b: 260-264). The tradition also prescribes to call for Dawrēšē-a’rd during
grain planting. Finally, another marked mandatory use of Du’āya t’asmīlī a’rdē is
its performance during the funeral rite, which also points to the phenomenon, in
which the idea of concealment connects hidden possessions, seeds and dead
bodies on the chthonic level. All of them enter the nether dimension and all are
entrusted to the earth temporarily, possessions waiting for their owners, while
seeds and bodies, for the mystery of revival.
Thus, similar to Du’āya t’asmīlī a’rdē covering the sphere of the three men-
tioned phenomena, the Qawlē Ēzdīdī Ĉuk is outspread to both the revealing of
concealed sacred objects and the concealment of a no longer profane dead body.
1. Āqāyē min Ēzdīda,
Bayrāqa sori-sipī p’īdā,
Čārda tavak a’rd, a’zmān k’afā dastē Sul’tān Ēzdīdā
My lord is Ezdid,
A white-red flag is in his hand,
The fourteen spheres of the earth and heaven are in the palm of Sultan
Ezdid’s hand.
2. Galī mērā, dinyā bū danga,
Dahar bū Sul’tān Ēzdīdē minī p’ar řanga,
Bar dičē šā, xwandk’ār, fařanga.

5
For the publication of this text see Rudenko 1982: 132; Voskanian 1999-2000: 160-162.
 The Song Unveiling the Hidden 141

Oh saints, news (lit. voice) was spread in the world,

My [lord] Ezdid with colourful wings has appeared,
Shahs, sovereigns, Europeans (here, rather Christians) make way [for him].
3. Galī mērā, dinē bū ba’sa,
Dahar bū Sul’tān Ēzdīdē minī xāsa,
Digařa hazār yēk livāsē.
Oh saints, news (lit. talks) was spread,
My noble Sultan Ezdid has appeared,
He walks [dressed] in a thousand and one clothes.
4. Galī mērā, Šām šaharē bar āvē,
Suřa Sul’tān Ēzdīd hēwirī nāvē,
Wa’da hātīya−kalā bāžēř bistīna, řūnī nāva.
Oh saints, Sham (Syria) is a town by the sea,
The mystery (or essence) of Sultan Ezdid settled in it.
Time has come [for him] to take the castle, to sit inside.
5. Galī mērā, Šām šaharē girāna,
Suřā Sul’tān Ēzdīd dahar būya nāva,
Wē’da hātīya, kalā bāžēř wē bistīna, řūnī nāva.
Oh saints, Sham is a big town,
The mystery of Sultan Ezdid has appeared there,
Time has come [for him] to take the castle,6 to sit inside.
6. Galī mērā, k’āsa sōra,
Nāzilbūyē a’smān, a’ršē harī žōra,
Aw k’ās lāyiqī wī mērīya, řūništīya dīvānā Qubldōra.
Oh saint, this cup is red,
It descended from heaven, from the upper throne,
This cup is worthy of the saint, who has sat in the council of Qubldōr.7
7. Galī mērā, k’āsa zara,
Nāzilbūyē a’smān, a’ršē harī žōra,
Aw k’ās lāyiqī wī mērīya, řūništīya dīwāna Šamsē t’at’awara.8

6
Lit. “the castle of the city”
7
Kreyenbroek (1995: 88, nt.78) interprets this term as “the qibla of the full moon” (from Arab.
qiblat al-bi-dawr , lit. “the qibla which rotates”). In any case, here Qubldōr can be interpreted as the
“centre of the religion”, or even as the Yezidi axis mundi, i.e. Lalesh or its central temple−Sheikh
‘Adi’s shrine.
142 Victoria Arakelova

Oh saints, this cup is yellow (golden),

It descended from the highest sky, from the throne,
This cup is worthy of the saint, who has sat in the divan of Shams, the king.
8. Galī mērā, k’āsa sipīya,
Nāzilbūyē a’smān, a’ršē harī hēlīya,
Aw k’ās lāyiqī wī mērīyē, řūništīya dīwāna Šīxādīya.
Oh saints, this cup is white,
It descended from the highest sky, from the throne.
This cup is worthy of the saint, who has sat in the divan of Sheikh ‘Adi.9
9. Galī mērā, k’āsa řaša,
Nāzilbūyē a’smān-a’rša,
Mirā misk’an girtīya Qubldōr, Lāliša.
Oh saints, this cup is black,
It descended from heaven, from the throne,
The saints took up residence in Qubldōr, in Lalesh.
10. Galī mērā, bik’in k’ārī,
Darē jamā xazina qudrata vakin, hulakī řašī nūrānī, zanǰabārī lē darīnin,
Alīfa bar mīrē wē bimaykin.
Oh saints, try (lit. work),
Open the door of the treasury of strength, take out the black shining hula,
[Like the letter] Alif stay in front of your master.10
11. Galī mērā, havřā bikin k’ārē,
Darē ǰamā h’izanīyā qudratē vakin, h’ulēki zōrī, zanǰabārī,
Lē dērīnin suřā Sul’tān Ēzdīd balābū filān šaharī.

8
Another form of t’at’awar, Sheikh Shams’ epithet occurring in the qawls is t’art’ar . The form
used in this qawl is another proof that t’art’ar is a corrupted form of the Persian takfūr “king, lord”
(attested also in Kurdish dialects), which in its turn is borrowed from the Armenian dialectal t‘äk‘-
fur (classical t‘agawor “king”) (Arakelova 2002: 63-65).
9
Sheikh ‘Adi is a legendary founder of the first Yezidi community, considered the incarnation
of god in Yezidism. (See on this character Asatrian/Arakelova 2014: Part I, Chapter 3).
10
The last line of this verse is obscure and seems to be corrupted. Sheikh Hasan Tamoyan (Ye-
revan) interprets it as “Stay in front of your lord”, and also gives another version of this verse,
learned from his father:
Galī mērā, bikin k’ārī, Oh saints, try,
Darē xēzinā qodratī vēkin, Open the door of the treasury of strength,
Jē dērxin h’ulēkī řašī (nirānīī) zanjēbārī. Take out of there the chained black hula.
(Black is considered among the most sacred colours in Yezidism, being an attribute of some
Yezidi deities and saints, etc., see, e.g., Arakelova/Voskanian 2007: 155-156).
 The Song Unveiling the Hidden 143

Oh saints, try (lit. work) all together,

Open the door of the strong, chained hula,
Take the soul of Sultan Ezdid from there; [the soul] is spread in this town.
12. Šām šaharē mazina,
H’arĉkā řīā mālā ādīyā bika bīrē,
Řōžā āxratē nāvīna dīndāra mīra.
Sham is a big town,
The one who will not remember the house of ‘Adi,
He will not see the face of Lord on the Doomsday.
13. Qāsidē ĉīyākī hūra,
H’arĉkā dastē xwa sar mālā ādīādā bih’ažīna,
Řōžā āxratē řūnānī bar dīndāra mīra.
[It is] the messenger of a small mount,11
The one who will trespass on (lit. shake his arms on) the house of ‘Adi,
He will not sit in front of the Lord’s face on the Doomsday.
Refrain:
Hay ǰān, hay ǰān, hay ǰāna,
Šēxē H’asan12 Sultāna,
Sultān, Sultān mērāna.
Hey jan, hey jan, hey jan,
Sheikh Hasan is Sultan,
[He is] The lord, the lord of the [holy] men.

11
The “small mount” is, most probably, Sinjar, of the centre of the Yezidi spiritual life, while its
messenger is likely a symbolic designation of the unveiled sacred object. In case with, let us say,
the sanjak (whose revealing has been considered a sacred procedure (see, e.g., Nicolaus 2008: 247-
250), and which was used by the qawals while conducting their service and simultaneously col-
lecting alms on behalf of the sanjak as the guarantor of legality of their actions (Guest 1987: 34),
this particular object can definitely represent “the messenger” of Sinjar. Similarly, any sacred, from
the point of view of Yezidism, object can be approached the same way, as kind of a messenger of
the Yezidis’spiritual centre.
12
Sheikh Hasan (Sheikh ‘Adi’s relative and follower), according to the Yezidi tradition, is the in-
carnation of angel Dardail (Mash’afē řaš, part I.3, see Asatrian/Arakelova 2004a: 14).
144 Victoria Arakelova

BIBLIOGRAPHY
Arakelova, V. (2008), “Sultan Yezid in the Yezidi Religion: Genesis of the Character”, Journal of Per-
sianate Societies, vol. 3 (2005-1384), 2008: 198-202.
―― / V. Voskanian (2007), [Review of:] “Ph. Kreyenbroek, Kh. Rashow, God and Sheikh Adi are
Perfect: Sacred Poems and Religious Narratives from the Yezidi Tradition, Harrassowitz Ver-
lag”, Wiesbaden, Iran and the Caucasus, vol. 11.1: 141-148.
Asatrian, G. / V. Arakelova (2004a), “Malak-Tawus – The Peacock Angel of the Yezidis”, Iran and the
Caucasus, vol. 8/1: 19-28.
―― (2004b), “The Yezidi Pantheon”, Iran and the Caucasus, vol. 8/2: 231-279.
―― (2014), The Religion of the Peacock Angel: The Yezidis and their Spirit World, Durham.
Guest, J. (1987), The Yezidis: A Study in Survival, New York.
Jawari, N. (1983), Kurdskaya narodnaya pesnya, Yerevan.
Kreyenbroek, Ph. (1995), Yezidism−Its Background, Observances and Textual Tradition, New York.
Nicolaus, P. (2008), “The Lost Sanjak”, Iran and the Caucasus, vol.12.2: 57-91.
Rudenko, M. (1982), Kurdskaya obryadovaya poeziya, Moskva.
Voskanian, V. (1999-2000), “Dewrēš E’rd: The Yezidi Lord of the Earth,” Iran and the Caucasus 3-4:
159-166.
 Between Christianity and Islam:
Heathen Heritage in the Caucasus*

Viacheslav A. Chirikba
Leiden University

Abstract
The paper presents a brief survey of the traditional religious practices as still, or until recent times,
observed in the Caucasus. I postulate the possibility of a pan-Caucasian “mythological union”
formed over centuries between all the Caucasian communities, and discuss in some detail a local
“mythological union” on the example of the lightning ritual Čoppa.
Although the pre-monotheistic heritage, partially intertwining with the official religions, still
constitutes an intimate part of the identity of some Caucasians communities, it is slowly fading in
the shadow of the mainstream religions—Christianity and Islam, which have become a strong
unifying factor in the post-Soviet period.

Keywords
Religions in the Caucasus, Caucasian mythology

THE CAUCASUS AS A SOCIO-CULTURAL SETTING

The Caucasus represents a specific geographic area with a variety of contrasting
landscapes and climatic zones. The traditional economies were based on agricul-
ture, animal husbandry, hunting and crafts. Similar patriarchal feudal systems
were typical for most parts of the Caucasus. Despite some differences in eco-
nomic, cultural and geographic environment, the Caucasian peoples, irrespec-
tive of the languages they speak, are characterized by many common traits in so-
cial and cultural life. In the words of Vasily Abaev (1949: 89): “One gets an im-
pression that irrespective of a wholly impenetrable multilingualism in the Cau-

*
The present article is an updated version of the paper read at the conference “Ethnic and Re-
ligious Communities in the Caucasus” (27-28 March, 2006) organized by The Royal Anthropological
Institute of Great Britain and Ireland (Warburg Institute), London. I thank V. G. Ardzinba (†), A. A.
Ankvab, B. G. Hewitt, M. Costello, H. Martirosyan and A. Petrosian for valuable suggestions on the
text of the paper.
 Koninklijke Brill NV, Leiden, 2015
146 Viacheslav A. Chirikba

casus, there was formed in significant features a common cultural world”. As to

religion, in general, it rarely played an antagonizing role in relations between the
various Caucasian communities. In the cosmopolitan urban centres like Der-
bent, Tbilisi or Baku peoples of various ethnic backgrounds and confessions eas-
ily mixed and lived peacefully side by side.
Linguistically, the peoples of the Caucasus speak languages belonging to
various families. The Abkhazian, Abaza, Circassian (i.e. Adyghe and Kabardian)
and Ubykh languages belong to the western branch of the North Caucasian lin-
guistic family, its eastern branch being Nakh-Dagestanian (including Chechen,
Ingush, Bats and some thirty languages of Dagestan, such as Avar, Dargwa, Lezgi,
Tabasaran, Lak, Andi, Tsez, Khvarshi, Rutul, Udi, etc.). The Georgians, Megreli-
ans, Laz and Svans speak the languages of the Kartvelian family, unrelated to
North Caucasian. The Caucasian peoples speaking Turkic languages are Azeris,
Kumyks, Nogays, Karachays, Balkars and Meskhetian Turks. The Armenians, Os-
setians, Kurds, Tats and Talysh speak languages belonging to the vast Indo-Eu-
ropean linguistic family. A small Assyrian community, mainly in the Southern
Caucasus, speaks a Semitic language.
From the point of view of official religious denominations, the Caucasus is
divided into two main realms: that of Christianity (Georgia, Armenia, major part
of Ossetia and Abkhazia) and that of Islam (the Northern Caucasus, Azerbaijan,
partially Ossetia and Abkhazia). But in the shadow of these two religious main-
streams many cultures of the Caucasus preserved, to a greater or lesser extent,
vestiges of older, pre-monotheistic beliefs, which are polytheistic and animistic
in nature, and which can be subsumed under the term “traditional religion”. The
extent of the preservation of this pre-Christian and pre-Islamic heritage is differ-
ent in various parts of the Caucasus: it is much stronger in Abkhazia, Ossetia and
parts of Georgia, and is less visible in such monotheistic strongholds as Chech-
nya, Dagestan, Azerbaijan and Armenia. The vestiges of pre-monotheistic reli-
gious paradigm form an integral part of the spiritual life and cultural identity of
many Caucasian communities, and in some regions the traditional religion un-
dergoes a certain revival and even gains partial official recognition. In general,
this often-overlooked spiritual heritage undoubtedly warrants special attention
and investigation.1

1
The main sources for the present survey are as follows. On Abkhazian mythology: the works
by Adzhindzhal (1969), Akaba (1984), Chursin (1957), Inal-ipa (1965), and results of my own field-
 Heathen Heritage in the Caucasus 147

MAIN OFFICIAL RELIGIONS PRACTISED IN THE CAUCASUS

Christianity
Christianity appeared in the Caucasus at the very beginning of our era, entering
first Abkhazia, coastal Circassia, Armenia, Georgia and Caucasian Albania (i.e.
modern Azerbaijan and parts of Georgia), whence it later partially spread to the
Northern Caucasus. In the Middle Ages, the Georgians, with limited success,
tried to spread Christianity to some of the Northern Caucasus areas adjoining
Georgia, including present-day Chechnya, Ingushetia and Dagestan, but these
activities were stopped with the destruction of the mediaeval Georgian state by
the Mongols in the 13th century. The missionary activities of Georgian priests in
the Northern Caucasus were not very effective due, to a large extent, to linguistic
problems: the liturgical language, in which the Christian creed was preached,
was Georgian, unknown to the overwhelming majority of the Nakhs and the Da-
gestanians. Less is known about the missionary activity in the Northern Cauca-
sus (including Dagestan) of Armenian and (Caucasian) Albanian priests.2
Megrelians, Svans, most Georgians, major part of Ossetians, (Vartashen
Orthodox and Oktomber) Udis, the Bats, Pontic and Tsalka Greeks, a part of
Abkhazians, Mozdok Kabardians, a part of Assyrians
Gregorian the majority of Armenians, Nidzh Udis, Armeno-Tats
Catholic a small part of Armenians, a small part of Georgians, a part of Assyrians
Table 1. Christianity in the Caucasus

Islam
Islam was first brought to the Caucasus by the Arabs in the middle of the 7th
century, mainly to southern Dagestan. By the end of the 15th century Islam es-
tablished itself in Dagestan and in the period between 16th and 19th centuries it
spread from Dagestan to Chechnya and Ingushetia, from the Crimea to Circassia
(16th c.), from Ottoman Turkey to Circassia, Abkhazia (since the 15th c.) and
parts of Georgia, and from Iran to Azerbaijan. Adherence to Islam has always
been weaker in the Western Caucasus and stronger in Chechnya and especially

research; on Circassian mythology: Jaimoukha (2001), Mizhaev (1973); on Karachay mythology:

Karaketov (1995), Shamanov (1982); on Kumyk mythology: Gadzhieva (1961); on Ossetian mythol-
ogy: Abaev (1949; 1958), Chibirov (2008); on Nakh mythology: Jaimoukha (2005); on Dagestanian
mythology: Gadzhiev (1991), Jusupova (1988), Seferbekov (2001; 2009); on Georgian mythology:
Bardavelidze (1957), Virsaladze (1976); on Svan mythology: Bardavelidze (1957); on Armenian my-
thology: Karapetian (1979), Petrosian (2007).
2
See the recent account of the history of Christianity in Dagestan in Gasanov (2001).
148 Viacheslav A. Chirikba

in Dagestan. Used as a unifying pan-North Caucasian ideology, Islam attained its

peak in the period of the Russo-Caucasian wars in the 19th century, which led to
the deep Islamization of the majority of North Caucasian and Dagestanian
communities and at the same time to the elimination of much of the traditional
beliefs and/or Christianity.
In Chechnya, Ingushetia and Dagestan the mystical branch of Islam, Sufism,
evolved into a national form of religion. As explained by the Chechen author
Vakhit Akaev, “Sufi ideology easily lends itself to popular beliefs, customs and
traditions. This peculiarity, enabling the incorporation into Islam of elements of
popular culture related to the cult of ancestors, native land and etiquette, led to
its massive dissemination among the Nakh and Dagestanis” (cited in Jaimoukha
2005: 118).

Dagestanians, Ingush, Chechens, a (small) part of Ossetians, Abazas,

Adyghes, Kabardians, Karachays, Balkarians, Nogays, Kumyks, a part
Sunni of Talysh, Ajar Georgians, Laz, Meskhetians (originally southern
Georgia), Muslim Armenians (Khemshils), a part of Abkhazians,
Armenian and Georgian Kurds
Azeris, Azeri Kurds, Azeri Tats, most Talysh, Fereydan Georgians (in
Shia
Iran)
Sufi Ingush, Chechens, Dagestanis
Table 2. Islam in the Caucasus

Judaism
Judaism is practised in the Caucasus by three small groups: Tats, or Mountain
Jews (mainly in Dagestan, also in Azerbaijan and Kabarda-Balkaria), Georgian
Jews (Georgia, South Ossetia) and Russian/Ashkenazi Jews (mainly in urban cen-
tres across the Caucasus). The Tats speak a south-western Iranian language, the
Georgian Jews—Georgian, and Ashkenazi Jews—Russian and, probably, to some
extent Yiddish. Both Mountain and Georgian Jews are attested to the Caucasus
from early times (the Tats—since the 5th-6th cc.), though the exact time of their
settlement there is not known.
Yezidism—a syncretic religion
Yezidis (in total some 100,000) mostly live in Northern Iraq, Turkey, Syria, Yemen
and Iran. In the Caucasus, they are a very small group, and live mainly in Arme-
nia and Georgia. Speaking the Northern, Kurmanji dialect of Kurdish, most of the
Yezidis claim to be a separate nation and are in fact an ethno-religious group
 Heathen Heritage in the Caucasus 149

with a unique complex identity based on the religious belonging. Their religion
rooted in Sufism, was shaped also under the influenced of Gnostic ideas, Christi-
anity, local pagan cults, etc.3
“Traditional religion”
The terms “paganism” or “heathenry” can sometimes bear negative or pejorative
connotations, reflecting the monotheistic (Christian or Islamic) attitudes. Prob-
ably, better terms are polytheism or animism. In this paper I shall mostly employ
the term “traditional religion”, as often used by the respective communities
themselves.
Elements of traditional religions in the Caucasus are hidden behind the
mainstream confessions, and no sizable group claims to profess one or another
“traditional religion” as opposed to Islam or Christianity. Nevertheless, some
Caucasian communities have preserved vivid vestiges of pre-monotheistic reli-
gions and have retained the institutes of pagan priests or custodians taking care
of the shrines. The traditional religions are intimately interwoven into the tradi-
tional lifestyle, customs and etiquette and are syncretic in nature, as they also
include elements of Christian and (more rarely) Islamic beliefs and rites. In some
cases, it is possible to speak about traditional religion forming the substrate for
the official confessions.

The Religious System of Abkhazia

In this section I shall describe to some extent the system of traditional beliefs as
preserved up to the present time in rural Abkhazia. Here, unlike in neighbouring
Georgia or in Armenia, Christianity failed to become a dominant ideology. The
reasons are both historical and political. In Georgia and Armenia the Christian
creed was used as an ideological tool in the struggle against the political domi-
nance of Zoroastrian Iran and later of Islamic Caliphate, and as such became an
integral part of the ethnic identity of these nations. In contrast, the numerous
invaders who tried to control Abkhazia’s territory, used to stay mainly in the
coastal area, not, as a rule, penetrating deep into mainland Abkhazia, whereas
the main bulk of the Abkhazians lived in the foothills and in the mountains,
finding there a relatively safe shelter from any potential intruder. The other im-
portant reason was that Christian priests used to address the Abkhazians in
Greek, which was unknown to them, and then (after the 10th c.) in the equally

3
Among recent publications on the Yezidi religion, see: Asatrian, Arakelova (2014).
150 Viacheslav A. Chirikba

unknown Georgian language. Neither Greeks, nor Georgians showed any interest
in translating the Christian texts into the vernacular to make them comprehen-
sible for the Abkhaz population. In the apparent absence of any literacy in the
Abkhaz language, and no translations of Christian texts into Abkhaz, this ren-
dered the ideology inaccessible for the broad masses, beyond some sections of
the upper classes. Greek or Georgian sermons could leave little impression on
Abkhazians, who mostly borrowed the outer, ritual side of Christianity, not
bothering much about its theological contents.
Islam, which entered Abkhazia in 15th-16th centuries thanks to the hegem-
ony of Ottoman Turkey, was installed only in sections of the population, mainly
the aristocracy, who maintained close commercial and cultural ties with the Ot-
tomans. On the other hand, it proved to become one of the important factors
causing the effective decay of the previously imported Christianity. But Islam,
with its strict observance of rituals and restrictions, did not appeal to the tradi-
tional mentality and lifestyle of the Abkhazians and failed, as earlier Christianity,
to convert the masses to its fold. Paradoxically, by virtually eliminating Christian
institutions in Abkhazia (apart from several remaining centers like Lykhny or
Iylyr), Islam inadvertently caused the reinvigoration of the never really vanished
paganism, which enriched itself by absorbing certain Christian saints and rites.
Characterizing the religious situation of the Abkhazian society of his time, i.e.
the first part of the 19th century, the early Abkhazian ethnologist Solomon Zvan-
ba (Zvanbaj 1855; cited from the 1955 edition) wrote: “Apart from the Princely
House,4 whose members profess Orthodoxy, a very small part of the population
of Abkhazia are regarded as Christians and a very limited number of its inhabit-
ants regard themselves as Muslim; the rest are all pagans… all inhabitants of Ab-
khazia, irrespective of their [official] creed, worship some deities and conduct
heathen rituals” (p. 65). The Russian author of the beginning of the 19th century
E. Vejdenbaum remarked: “The majority of the Abkhazians regard themselves as
Muslims, though there is no mosque in the region and there are no priests…
Equally indifferent to the questions of religion in their majority are those Ab-
khazians who regard themselves as Christians” (cited in Inal-ipa 1965: 587).
Probably due to this habit of integrating elements of Islam or Christianity
into their overall system of beliefs, religion never divided the Abkhazians, and in

4
The princely house of the Chachbas, who ruled Abkhazia from Middle Ages until 1864, when
the Principality was abolished by the Tsarist administration.
 Heathen Heritage in the Caucasus 151

one family there peacefully co-existed both Christians and Muslims, who jointly
observed also the traditional rituals. Regardless of whether they referred to
themselves as Christians or Muslims, the Abkhazians in general never expressed
any religious fervor, let alone fanaticism, proving to be quite indifferent to theo-
logical issues and tolerant of any religion.
However, it would be equally wrong to claim that the Abkhazians are reli-
giousless, atheist or agnostic. The population at large is still, after decades of liv-
ing in an atheistic state, deeply religious, but in a different, more covert or eso-
teric way. For the Abkhazians, religion is a matter of very personal, intimate
communication with God, for which one does not need an intermediary, as a
priest, or a special place or institution, like a church or a mosque. The mystical
element is very strong in the Abkhazian religious psyche. The succession of
events is seen in their cause and effect relationship, everything being in the
hands of God, the ultimate supervisor and moral judge. Another very important
aspect interwoven into the canvas of the religious system, apart from its mysti-
cism, is the moral and ethical aspect, the unwritten code of traditional values
and behavioural prescriptions, etiquette, called Apswara (“Abkhazianness”)—a
proper way of being a decent and morality-ruled person on the levels of the fam-
ily, of the community and of the whole nation.
One of the reasons why the traditional religion managed to survive amazingly
well amongst the rural Abkhazians even in the conditions of the Communist
state was probably the fact that the Soviets were interested in controlling the sit-
uation mainly around Christianity or Islam, fearing their competing influence on
the minds and hearts of the population. As to the traditional practices, they, with
very little exception, did not attract much attention of the ever-vigilant authori-
ties, who looked at them indulgently, as the remnants of the traditional patriar-
chal lifestyle.

The Abkhazian pantheon

The traditional Abkhazian pantheon is characterized by a considerable sophisti-
cation: nearly all more or less important aspects of life had their own gods, god-
desses or patron saints. But, above all, there is also a clear idea of a supreme god,
called Antsva (Anc⁰á ),5 the almighty god, to whom daily prayers are addressed

5
To render the linguistic data, including the names of the deities, I use the phonetic transcrip-
tion used in Caucasology; some explanations for non-linguists: the sign x = kh (like ch in Scottish
loch ‘lake’), c⁰ = tsv, č⁰ = chv, j = y (as in yoke), y = Russian ы, c = ts, ǯ = dzh (as in joke), č = ch, ɣ̂ , γ =
152 Viacheslav A. Chirikba

and with the mention of whose name all toasts are started (cf. the traditional ini-
tial formula of such prayers or toasts: Wa Anc⁰a, wǝlǝpxa hat! ‘Oh Lord, give us
Your mercy!’). According to traditional views, God is infinitely plural in its vari-
ous aspects or spheres of responsibility. “Each natural phenomenon, as well as
each clan, family or individual has its own portion of god”, wrote N. Dzhanashia
(1915: 75); this phenomenon has its own term jəә-nc⁰a-x⁰əә ‘his-god’s-part’.
Beside the Supreme God Antsva, the pantheon includes quite a number of
high- and lower-ranking deities. The high-ranking deities were responsible for
crucial aspects of traditional life: household, cattle-breeding, fertility, agriculture,
hunting and warfare. Some of them are anthropomorphic and regarded either as
male or female. Others have the guise of an animal, and others are rather
amorphous, appearing, e.g. as thick fog. Typically, the majority of the personifi-
cations of the hunting, warfare, thunder and lightning gods, as well as the chief
god of cattle were regarded as male, whereas many agricultural and household
cults, and the “portions” of chief gods, were imagined as female. The following
schemes give an impression of the elaborate nature of the Abkhazian pantheon.

Supreme God (Anc⁰a)

High-ranking deities: god of the reproduction of cattle (Ajtar), goddess of bees
and women (Anana-G⁰əәnda), god of hunting and warfare (Ajergj), god of thunder
and lightning (Afǝ), god of metallurgy and smithy (Šjaš⁰ǝ), god of hunting (Až⁰ej-
pšjaa), half-forgotten deity Wašx⁰a, half-forgotten great god Awəәblaa.
Lower rank deities: goddess of harvest and fertility (Ǯaǯa), goddess of milling
and millstones (Sawnaw), goddess of weaving (Jarəәš//Jarəәšq’an), deity of cotton
and flax (K⁰əәk⁰əәn), the divine smith (Ajnar-(jəә)žʲəәj), deity of hearth (Č’ʲap’).
The creator-gods: Mother-creator, goddess of fertility (Anana-Šac⁰a), God-
creator (Anc⁰a Šana), the creators, who determine fate (Ašac⁰a//Ačʲapac⁰a//Aša-
c⁰a-Čʲapac⁰a), patron of procreation and family (Ažʲahara), deity of marital hap-
piness (Ləәməәjrah//Nəәməәjrah//Šəәməәjrah), deity patron of married women (Ačʲa-
pac⁰a).
Deities of animals, birds: deity of buffalos (Mkamgarəәja), deity of goats (Aca-
baxʲ), deity of cats (Acxʲ), patron of birds (Až⁰ejpšʲaa anč’ʲa-k’⁰əәnč’ʲa), patron of
quails (Ačʲarah//Ačʲaj⁰rəә).

gh-like sounds, ś = sound between s and sh, ź = sound between z and zh, ž = zh (as s in usual), ʒ =
dz, x̂ = kh-like sound, λ = voiceless lateral fricative similar to Welsh ll, L = voiced lateral fricative, ʲ =
sign for palatalization, ⁰ = sign for labialization, ’ = sign for glottalization.
 Heathen Heritage in the Caucasus 153

Deities of diseases and death: deity of death, which takes souls (Apsc⁰aha),
patron of internal diseases and the sick (Anapra), patron of smallpox (Axʲəә Zo-
shan).
Lower rank deities, demons,and mythological creatures: angel (amaaləәkʲ),
goddess of dreams and nightmares (C⁰əәblaq’⁰), incubus (Nap’k’əәlc’⁰a), dead
souls’ spirits/vampires (Ad⁰nəәq’⁰a ‘ones(s) walking in the field/outside’), devils
(Aj⁰staa; Agəәzmal; Aǯʲnəәš)6, woodgoblins (akač’ʲaa), demons (akaǯʲk⁰a), mer-
maid (Ʒəәzlan), monster/dragon eating the sun or the moon causing the eclipse
(at’⁰əәj⁰) , the giants (adawəә), the forest man (abnawaj⁰əә).
Special cults: the calling for rains (Ʒəәjwaw), the rite of new-built house (Aʒar-
x⁰ma), the cult of the hearth-chain (Arxəәšʲna), the procreation cult (Ažʲrac⁰ara),
the New Year ritual (Xʲačʲx⁰ama) (in the Bzyp region), the cult of spirits (Ad⁰n-
jəәq’aw ‘the ones who are outside’, referring to dead people’s spirits which abide
outside the homes), deity Targʲalaz7, shrines of thieves, robbers and bandits
(Ajrəәx-Aac-nəәx; Čʲəәg⁰əәrx-nəәx).
Witches: witch (arəәwp’ap’), wolf-riders/werewolves (abga jak⁰t’⁰aw ‘the
one(s) sitting on a wolf’).
Taking into account that Abkhazia is traditionally an agrarian and cattle-
breeding country, it is not surprising to find here the relevant cults, the main one
of which was the cult of the god of procreation, cattle and renovation, Ajtar, re-
garded as being seven-fold, i.e. consisting of seven parts or fractions, represented
by deities responsible for separate branches of cattle-breeding, as well as by the
harvesting goddess and two astral deities. Another seven-fold god was the
smithy god Shashwy (cf. a typical address to this god: Šjaš⁰ǝ axʲahdǝw, abžʲnǝxa!
‘Shashwy great golden prince, seven shrines!’). The digit seven bears, of course,
special significance in mythology and magic practices of many peoples of the
world.8

6
A-ǯʲnəә-š – lit. ‘white (š) jinn’, i.e. originally a beneficent spirit.
7
From Georgian mtavar angelozi ‘chief angel/Archangel’, whence also Ossetic Taranǵeloz, the
name of the deity and its shrine, and the Svan deity Taringzel//Tarinӡel//Targlezer (Abaev 1979:
232).
8
Cf. especially the Indo-Iranian tradition: seven gods (Amesha Spenta) in Zoroastrianism,
seven celestial deities Āditya in Rigveda, Ardavda ‘Seven gods’ as the old Alanian (Ossetic) name of
the Crimean city of Feodosiya, and in modern Ossetic: Avd Wacillaji ‘seven Watsilla (deity of thun-
der)’, the mythological creature Ævdiv < Old Iranian hafta-daiva ‘seven gods’ (cf. Abaev 1990), Avd
ʒuary ‘seven shrines’, deity combining the functions of seven gods, etc.
154 Viacheslav A. Chirikba

Ž⁰abran female deity of cows (also used to designate the month of February)
Ǯjabran female deity of goats
Ačǝšjašjana female deity of horses
Alǝšjk’jant’ǝr deity of dogs
Anapa-Naga female deity of millet and harvesting
Amra Sun
Amza Moon
Table 4. Seven fractions of Ajtar

Another crucially important occupation was hunting. The process of hunting

was regarded as a magic act, during which the hunters were not allowed to use a
usual language and had to communicate in a so-called “forest tongue” (abna
bǝzš⁰a), which was lexically different from the common language, lest the ani-
mals understand the intentions of the hunters (cf. Khiba 1980).
Až⁰ejpšʲaa’s mother Mč’ʲəәxan//Məәč’ʲxan
Až⁰ejpšʲaa’s son Jǝwana//Dad Jǝwana
‘John//Father John’
Až⁰ejpšʲaa’s son Pəәšʲq’an
Až⁰ejpšʲaa’s daughters Až⁰ejpšʲaa rtəәphac⁰a
Až⁰ejpšʲaa’s uncle Ž⁰ejpšʲəәrq’an
Až⁰ejpšʲaa’s servant Š⁰əәk’az
Table 5. The family of the hunting god Až⁰ejpšʲaa

Abkhazians also worshipped sacred groves, as well as spirits or patrons of

mountains (e.g. Ašʲxa jəәnc⁰ax⁰əә, lit. «god’s mountain fraction»), and of other
natural objects, on whom the good luck of hunters, shepherds and travellers de-
pended. The low-ranking gods included even the special patrons of thieves, rob-
bers and bandits (Ajrəәx-Aac-nəәx; Čʲəәg⁰əәrx-nəәx).
Ašʲxa jəәnc⁰ax⁰əә patron of mountains9
Axra jəәnc⁰ax⁰əә patron of rocks
Apsta jəәnc⁰ax⁰əә patron of gorges
Abna jəәnc⁰ax⁰əә patron of woods
Et-nəәx=Aga-nəәx patron of seashore
Adgʲəәl-čʲčʲa/əә goddess of earth

9
The 3rd person singular masculine possessive prefix jəә- ‘his’ in jəә-nc⁰a-x⁰əә ‘his-god-part’ may
probably indicate a missing word a-nc⁰a ‘god’, from a fuller phrase *a-nc⁰a a-šʲxa jəә-nc⁰a-x⁰əә ‘god’s
mountain part’.
 Heathen Heritage in the Caucasus 155

Aʒəә jəәnc⁰ax⁰əә goddess of water

Ʒəәzlan//Ʒəәzlan-ʒahk’⁰až⁰ the Mistress of the Waters
//Aʒ jəәtaw//Axʲəә-psha-g⁰ašʲa10
Table 6. Deities of natural objects

The important female deity of the agrarian cult was the goddess of harvest
and fertility Ǯaǯa, imagined as a dumpy old woman who was roaming the fields
with crops: depending on her moods, she could give a blessing to the future har-
vest, or, on the contrary, put damnation on it. The other female agricultural dei-
ties were the spirit of millet Anapa-Naga, of cotton and flax K⁰əәk⁰əәn, of milling
Sawnaw. The goddess of female handicraft and knitting was Jarəәš. The beautiful
and insidious gold-haired mermaid Ʒəәzlan, with the heels of her feet turned for-
ward, was the goddess of rivers and the seducer of lonely travellers.
During the drought, the peasants used to organize the ritual of rain sum-
moning called Ʒəәjwaw, where figured a specially decorated doll, which they
brought to the river. The doll was laid on an improvised raft, put on fire and
given to the streaming water, whereas the participants were entering the water,
catching and killing the frogs, turning them on their backs.
The other female deities were the female deity of dreams (C⁰əәblaq’⁰) and
even the female demon of nightmare (Nap’k’əәlc’⁰a, lit. «(with) a hole in the palm
of the hand»11), an analogue of the incubus.
The smithy (a-žʲəәjra) and its instruments played an important role in tradi-
tional (especially family) cults. As N. Dzhanashia wrote: “In the eyes of the Ab-
khazians the smithy occupies a higher place than the church” (cited in Adzhin-
dzhal 1969: 235). The Abkhazians regarded the smithy god Shashwy as one of the
most powerful gods, and some even placed him second after the supreme god
Antsva. The blacksmith (a-žʲəәj) was respected, but also somehow feared, being
perceived as a priest of the great god Shashwy.
The anykhas (a-nəәxa ‘icon’, ‘shrine’) have always been objects of great respect
among the Abkhazians. Many publicly important events used to take place near
the shrines. Thus, an accused person had an obligation to make an oath at an
anykha in order to prove his innocence. One could use an anykha to put damna-
tion or a curse on a person, or group of persons, which was always a public act.

10
Aʒ jəәtaw ‘the one who is in the water’, ʒahk’⁰až⁰ ‘water princess’, Axʲəә-psha-g⁰ašʲa ‘Golden
Queen Lady’.
11
The same mythological creature with the hole in the palm is known among the Armenians
(see below).
156 Viacheslav A. Chirikba

There is a certain hierarchy of anykhas, divided into chief (for a certain region)
and junior. The chief anykhas in Abkhazia are seven and they are called accord-
ingly Abžʲnəxa ‘seven anykhas’. These are Dəәdrəәpšʲ-nəәxa, Psh⁰əә-nəәxa, Jəәləәr-nəәxa,
Lašʲk’ʲəәndar, Lʒaa-nəәxa, Ləәx-nəәxa and Bəәtxa (in neighbouring Ubykhia). The
anykha in the village of Achandara, situated on the Dydrypsh mountain (whence
its name: Dəәdrəәpšʲ-nəәxa, containing a-dəәd ‘thunder’) is still revered as the chief
and most important shrine of Abkhazia. Some of the anykhas are regarded as
being in family relations to each other, and are thought to be visiting each other
periodically, flying as fireballs. Thus, Pshu-nykha and Jylyr-nykha were regarded
as brothers, Jylyr-nykha—the elder brother and Christian, Pshu-nykha—the
younger brother and Muslim. The shrine of the Dydrypsh mountain was regard-
ed as being brother and sister, the sacred grave of Inal in Pshu and the shrine
Anyps-nyx (anǝ ‘mother’, ps ‘dead’, i.e. the anykha of the Assumption of God’s
Mother)—also as brother and sister (Chursin 1957: 27).
There were regional, communal, clan and family shrines. E.g. the shrine of
the Dydrypsh mountain—Anǝps-nǝxa, was worshipped in the whole of Bzyp or
Western Abkhazia, Ajlǝr- or Jǝlǝr-nǝxa was the chief shrine of the Abzhywa or the
eastern region of Abkhazia, whilst Jǝnal-q’⁰ǝba was revered mainly in the
mountainous Abkhazia. Lejaa-rnǝxa was respected in the community of the vil-
lage of Mgudzyrkhwa, which members regard their anykha as a protecting deity
which is invisibly present amongst them, especially when they are in the sacred
grove. The family clan Ampar had as their patron saint the shrine Etnǝx-Ag(a)-
nǝx, which is regarded as being connected to the sea and the prayer to it is car-
ried out at the seashore (Chursin 1957: 25-6). The latter two shrines are usually
situated in the ritual smithy (a-žjǝjra), where the representatives of the clan meet
once a year, usually on New Year’s Eve and engage in special rituals. Beside
purely religious functions, these gatherings serve to strengthen the bond be-
tween the members of the clan and create a feeling of participation and com-
mon responsibility for the affairs within the clan.
The Abkhazian traditional cult does not know the institute of priests, their
functions being fulfilled by the custodians of the shrines, or by old, experienced
and ritually pure (i.e. sexually inactive) men or women. Each anykha used to
have its own custodian (anəәxapaaj⁰əә), usually belonging to a certain family, of
which a chosen male representative functioned as a priest. Besides, there were
several categories of augurs, who could carry out rituals resembling shamanistic
(cf. Johansons 1972). For instance, ac’aaj⁰əә ‘the questioner’ is a woman who asks
 Heathen Heritage in the Caucasus 157

God about the reason of His wrath towards a person fallen ill; she can as well use
for predicting future events beans or cards.

ac’aaj⁰əә ‘the questioner’12

aq’⁰əәdəәrpaj⁰əә ‘one who makes the beans jump’
amacarpaj⁰ǝ ‘one who makes the cards jump’
apšj⁰əә ‘one who looks’, a man who sees the
future by looking at the shoulder-blade
of an animal like sheep or a rooster
adǝrj⁰ǝ ‘one who knows’ (a soothsayer who
tells future by looking, e.g., at a
person’s palm or lobe of the ear)
axalaj⁰əә ‘one who climbs’
Table 7. Abkhazian augurs and clairvoyants

Interestingly, during the act of prophecy, the questioner woman (ac’aaj⁰əә)

was addressed to as a man rather than as a woman, which implies a ritual
change of gender or transvestism, typical for shamanistic practices in many
traditional cultures (V. Ardzinba, p.c.; Johansons 1972: 253).
Typically, when asked, the religious Abkhazians usually declare their adher-
ence either to Christianity or Islam, not to traditional religion, which until re-
cently had no special term in the language.13 Nevertheless, significant parts of
“Christians” and “Muslims” continue to venerate old shrines and practise tradi-
tional cults. The adherents of the traditional religion prefer now to call it “Ab-
khazian religion”, and some refuse to qualify it as “pagan”, citing the presence of
the single almighty God—Antsva.
The question of to which extent the Christianity influenced Abkhazian tradi-
tional religion is not well studied, but one can surmise that the impact was ra-
ther substantial. This is attested, in particular, by the fact that the Abkhazian
pantheon includes a number of deities having their origin in Christian saints. In-
terestingly, the major pagan shrines are situated at the sites or near the ruins of
Christian churches. Thus, on the top of the Dydrypsh mountain, which is the seat
of the much revered and even feared shrine Dydrypsh-nykha, there are ruins of a

12
Cf. typologically Armenian harc‘-uk ‘sorcerer, magician’, from harc‘anem ‘to ask, question, in-
quire’; cp. also Old English freht ‘clairvoyance’ from the same Proto-Indo-European etymon (H.
Martirosyan, p.c.).
13
However, there is a verb apaara, which means ‘to perform traditional priesthood’.
158 Viacheslav A. Chirikba

Christian church (whence the second name of the shrine: Anǝps-nǝxa ‘The
shrine of the Assumption of God’s Mother’). It is difficult to say, however, which
preceded which: the pagan shrine being associated with the Christian church, or
the latter being built on the place of the old shrine.

Pagan deity Origin/Prototype

Ajtar (god of cattle) St Theodore Tyro (Greek Άγιος Θεόδωρος ο
Τήρων)14
Anan//Anana G⁰ǝnda (female deity of St Mary, God’s Mother, associated in
bees; cf. an//nan ‘mother’) popular mythology with bees
Ajəәrgʲ (god of warfare and hunting) St George (Gr Άγιος Γεώργιος)15
Mkamgarəәja (god of buffalos)16 Archangels St Michael and St Gabriel
Anapra (patron of internal diseases and St. Onuphrius (Greek Ονούφριος)17
the sick)
Dad Jəәwana ‘father (dad) John’ St John the Baptist18
Table 8. Abkhaz pagan deities originating from Christian saints

At present, traditional Abkhazian religion is undergoing a certain revival.

Some time ago it gained official acknowledgment by the state, which equated it
to the two official denominations, Christianity and Islam. In August 2012 the
“Union of the Traditional Priests of Abkhazia” was formed by custodians of the
seven shrines of Abkhazia (Abžʲnəxa) with the aim to be officially registered and
function as a religious institution. Zaur Chichba, the custodian of the main
shrine in Western Abkhazia, Dydrypsh-nykha, was elected the chief priest of Ab-
khazia (traditionally it is one of the Chichba family from the village of Achan-
dara who is elected as a custodian of this important shrine). However, traditional

14
The form Ajtar//Ajtǝr is due to haplology from *Aj-Tǝta/ǝr, cf. the Ossetic deity Tutǝr from
the same Greek source, but without the Greek prefixed Άγιος ‘saint’ (cf. Chirikba 2006: 55).
15
S. Zvanba (Zvanbaj 1853, reprinted in 1955, p. 58) was probably the first to propose the deriva-
tion of the name of this deity from Greek Georgios, though his analysis of the Abkhaz form is not
quite correct: “A, Ergi”, where A- was probably perceived by him as a (vocative) interjection,
whereas we have here obviously Greek Άγιος ‘saint’ (> Aj); thus, Ajergʲ < *Aj-Gʲergʲ < *Aj-Gʲeorgʲi (cf.
also Aj- in Aj-tar above). Cf. a perception of the initial a- in Ajergʲ as the definite article in Tuite
(2004: 152).
16
From Megrelian mikam-gario < Georgian mikel-gabrieli ‘(archangels) St Michael and St Ga-
briel’ (Bgazhba 1964: 225).
17
Cf. Chirikba (2006: 56).
18
Cf. Ossetic Fyduani, name of a festival and of a saint, from Fyd-Iuane ‘Father John’, i.e. John
the Baptist, compared with its Abkhaz counterpart by V. Abaev (1949: 316).
 Heathen Heritage in the Caucasus 159

religion remains mostly an oral tradition, not codified in the form of a book or
collection of texts. Besides, the majority of the believers, even those who practice
traditional rituals, often identify themselves either with Christianity or Islam.
The Circassian pantheon
In Circassia pantheistic beliefs were dominant until the early 19th century, de-
spite the nominal adherence to (Sunni) Islam and the presence of small Ortho-
dox Christian groups. The remnants of these beliefs are to some extent preserved
in rural areas in the form of traditional rituals, and are discernible in mythologi-
cal and folklore texts. The main elements of the Circassian pantheon include the
supreme god The, the god of the soul Pse-the, the hunters’ god Mezǝ-the, the god-
dess of apiculture Merjem (i.e. St Mary), and the smithy god λepś. Revered natu-
ral objects included sacred trees and groves. Small iron objects were used to pro-
tect against evil forces.
At present there is some revival of interest among Circassian intellectuals in
the pre-Islamic tradition, which is seen from a number of relevant publications,
but this interest so far has not taken on the same organizational character, as in
Ossetia or especially in Abkhazia.
The ‘God’ // Pse-the ‘the Soul-God’// The supreme god
The-šx̂ ⁰e ‘the Great God’
ŠəәbLe (Adyghe)//JeLe19 god of thunder and lightning
Sewəәzerešʲ (Adyghe), Sozeraś god of fertility, agriculture and the family
(Kabardian) hearth
λepś god of metallurgy and the smithy
Žʲəә-the (Adyghe žʲəә ‘wind’) lord of the wind
Ax̂ əәn god of cattle, god of the sea
Mezəә-the (‘Forest-god’) god of hunting (sometimes female)
A(j)məәšʲ god of cattle20
The-ɣeLeǯʲ god of crops and the harvest
Zəәjwəәzhan (lit. ‘the one who carries god of smallpox
the disease’)
Awəәś-gʲergʲ (< St George) god of warriors, weapons and fortitude, god of
(Kabardian) hunters
Wašx⁰e god of heavens
Table 9. Male gods of the Circassian Pantheon

19
From Greek Elias.
20
A(j)məәšʲ and The-ɣeLeǯʲ are regarded as brothers (cf. Mizhaev 1973: 25).
160 Viacheslav A. Chirikba

Hance-g⁰ašʲ (Adyghe; ‘Shovel-Lady’) goddess of rain

Psəә-the ‘Water-God’//Psəәthe g⁰ašʲe (Adyghe; the water goddess
‘Water-God Lady’) //Psəәx⁰e-g⁰aśe (Kabardian;
‘River-Lady’)
x̂ əә-g⁰ašʲ (Adyghe ‘Sea-Lady’) goddess of the sea
žəәɣ̂ -g⁰ašʲe (Kabardian ‘Tree-Lady’) goddess of trees
Mez-g⁰ašʲ (Adyghe ‘Forest-Lady’) goddess of forests and trees
Merise//Merjem (< St Mary) goddess of bees and herbs
Pšəәšan21 goddess of small cattle, cows
Table 10. Female gods of the Circassian Pantheon

Wəәdəә (mainly) female witch

mezəәλ’ forest man
neɣ⁰əәč’əәc (Adyghe), neźɣ⁰əәś’əәʒe evil creature regarded as an ugly decrepit
(Kabardian) humpbacked old woman with iron teeth
and with long breasts brushed behind her
back
almestəә evil creature regarded as a naked woman
with long hair
Jəәnəәź mythological giant
bLaɣ⁰e dragon (lit. ‘yellow snake’)
Table 11. Lower rank deities or mythological creatures

The elements of the Abaza pantheon

The small North Caucasian Abaza people are linguistically and culturally the
closest kin of the Abkhazians. Islamization influenced them more profoundly
than the Abkhazians, and only few vestiges of older cults can be reconstructed.
Thus, among the Ashkharywa Abazas, a group especially close to the Abkhazi-
ans, there is still some memory among the old people about pluvial magic (the
ritual of calling for rains); the relevant deity is called Ʒǝjwara (cf. Abkhaz Ʒǝj-
waw); it contains the word ӡǝ ‘water’ as the first element. During the ritual, the
procession would bring a decorated doll to the river, the blacksmith was pulled
into the water, and then the rest of the procession also entered the river. The
main characters of the traditional cult are the Supreme god (Nč⁰a), the patron of
woods (Abna-ʕ⁰əә ‘Forest Man’), the goddess of waters (Aʒ-tas or Aʒ-nanəә ‘Water-
mother’). The cult of the smithy and smith-craft had as its patrons faγəәmbar
(‘prophet’) and the white (benevolent) jinn, which inhabited each smithy. The

21
Cf. the Abkhazian female deity of horses Ačǝ-šʲašʲ-ana (a-čǝ ‘horse’, ana ‘mother’)
 Heathen Heritage in the Caucasus 161

anvil was especially respected by the Abazas, sacrifices were made to it, and an
oath made at the anvil was regarded as inviolable. Moreover, Abaza folklore has
preserved stories about demons, about witches appearing in the guise of cats,
etc.
The elements of the Ubykh pantheon
Very little is known about the traditional religion of the Ubykhs, who were de-
ported by the Tsarist Russian administration to Turkey in their entirety in the
middle of the 19th century, before elements of their traditional beliefs could have
been recorded. One of the few known deities is the supreme god Wa//Waba.
Some gods were borrowed from the neighbouring Circassians (as the smithy cult
of λepś), or Abkhazians (nc⁰a ‘god’). Also known is the Ubykh pagan shrine Bəәtxa
(revered, by some accounts, also by the Abkhazians).
The Karachay traditional beliefs
Before the spread of Islam among the Western Caucasian Turkic-speaking Kara-
chays (mainly from the fellow-Turkic Kumyks) at the beginning of the 18th cen-
tury, their religious system consisted of a mixture of traditional and Christian
beliefs. The traditional pantheon, beside the Supreme God Tejri, is presented in
the following table.
Kek Tejresi Sky god
Ǯer Tejri Earth god
Suu Tejrisi Water god
Suu anası Mother of the Water
Gol god of Spring
Dolaj god of cattle
Bajrım (St Mary) goddess of (children’s) birth
Erirej God of cereals
Apsatı God of hunting and prey
Beautiful Fatima daughter of hunting god Apsaty
Table 12. Elements of the Karachay traditional pantheon

Cattle-breeding and agriculture were, as with fellow-North Caucasians, of

prime importance to the Karachays, and most important rituals were connected
with these sectors of the economy. During the spring festival Elliri Čoppa, which
celebrated the coming of spring and the beginning of agricultural works, spe-
cially dressed people used to organize dramatized rituals. The Karachays also
celebrated the first furrow day, the village festival saban ojun ‘the field game’ and
162 Viacheslav A. Chirikba

the ritual guguk xorlau ‘victory over the cuckoo’. The harvest (October) festival
was called Čoppa, and the December ritual game, dedicated to insemination of
sheep in barns, was called Boran keldi ojun. The worshipped natural objects in-
cluded the sacred stone in the village of Uchkulan (called Bajrım taš “the Stone
of Bajrym”, i.e. of St Mary) and the sacred tree to which infertile women ad-
dressed their prayers (cf. Shamanov 1982).
Elements of Kumyk cults
Kumyks, a Turkic-speaking Islamic people mainly in Dagestan (but also in some
other places in the North Caucasus) have preserved some vestiges of pre-Islamic
cults, such as fertility and agricultural cults. Among the latter is, for example, the
spring ritual of the “burning” of the winter, during which young people played
ritual games, like jumping over bonfires. The rain goddess Zemire22 was regarded
as a stout crummy woman, who was addressed during the procession for sum-
moning rains. The forest woman Sut-qatın (qatın ‘woman’), who was believed to
be roaming in the vicinity of the villages during the night-time, is also mentioned
in the special song performed during this ritual. The ritual doll which figured in
the ritual was made either of a shovel painted as a woman or of dough and was
called qaqıj or určuqan. The fertility and harvest ritual Gudurbaj (= South Kumyk
Güssemej) was carried out by young men in the evening time, during which they
sang ritual songs and prayed for good harvests, prosperity and happy life. The
Kumyk female water deity was Suvanası, lit. ‘water mother’, who was regarded as
a giant and exceptionally strong woman living in rivers; she could do harm or
even kill people coming down to the water. Especially feared was the giant
demonic woman Albaslı qatın with thick loose hair and big hanging breasts,
brushed back on her shoulders, a character typical for Caucasian demonology.
Her male counterpart was Temir-töš (‘iron breast’), who was sometimes regarded
as Albasly’s husband. The other demons were the malignant spirits jinn and
shajtans, the demon ilbis-shajtan. The evil incubus bastrıq was believed to
smother people during their sleep (Gadzhieva 1961: 322-327).

Er anası Mother of the Earth

Suv-anası Mother of the Water
Zemire rain/water goddess

22
One can note the formal resemblance of this name with the name of the Abkhaz-Abaza rit-
ual of rain summoning called Ʒəәjwaw/Ʒǝjwara. Cf. also the Dargwa rain deity Zaburaj and Lak Züvil
below.
 Heathen Heritage in the Caucasus 163

Sutqatın forest woman

Gudurbaj (South Kumyk Güssemej) fertility and harvest deity
Avrum anası female deity of illness
peri benevolent jinn
albaslı qatın demonic woman
temir-töš demonic man
ilbis-shajtan Devil
Bastrıq Incubus
Table 13. Elements of Kumyk traditional religion

The Ossetian traditional cult

The majority of Ossetians are nominally Orthodox Christians, whilst a smaller
part of them are Sunni Muslims. Ossetians have preserved vivid vestiges of their
traditional cults. Many heathen deities are still well known to traditional rural
Ossetians, and some of the shrines are still worshipped. According to Ossetian
beliefs, each place has its own spirit-protector, “the lord of the place” (bınatı-xi-
cau), and each village has a sacred stone called Madae Majram, i.e. ‘Mother
Mary’. There is a certain revival of the traditional heritage and its partial official
recognition by the authorities, who attend popular festivals devoted to some of
the shrines. The most popular of such traditional cults has become the one de-
voted to the sacred Grove of Khetag, and it has been made an official national
festival.

Avriaɣd god of natural forces, like clouds, rain showers

and hail
Don-bettır (lit. ‘Water-Peter’, i.e. St patron of waters
Peter)
Barduag god of thunder, fogs and clouds
Wac-illa (Divine Elias) god of thunder and lightning
Kar, Kæræf demons of winter and severe cold
Gætæg god of rivers
Galægon god of winds
Buduri izæd patron of the plain
Baragǯın goddess of forest, forest woman
Donı čızǯı-tæ23 mermaids, daughters of Don-bettır
Æfcæǯi ʒuar deity of mountain gorges and passes

23
-tæ is the Ossetian plural suffix.
164 Viacheslav A. Chirikba

Wad-æxsīn(æ) goddess of winds

zæd-tæ, dauǯı-tæ heavenly spirits, patrons of people, animals,
woods, waters and winds
Table 14. Ossetian deities of natural objects and phenomena

Anigol goddess of bees

Æfsati god of prey and hunting
Tutır (< St Theodore Tyro, Greek patron of wolves
Hagios Theódōros)
Burxorali//Borxwarali god of crops and cereals, regarded as son of
Xwari ældar (see below)
Mıkalgabır-tæ (< archangels Michael god of agriculture
and Gabriel)
Nikkola (< St Nicholas) god of fertility
Zegiman god of bread
Dauǯıtı ʒuar deity of fertility and harvest
G⁰ıdırtı kom demon of crops
Karčikloj patron of foxes and hares
Kæftısar master of fishes
Nogbon patron of the new day of the year
K⁰ırdalægon god of smithy
Was-tırǯi//Was-Gergi (< St George) patron saint of men, war god
Xwari ældar//Xwareldar deity of the harvest
Table 15. Ossetian cults of agriculture, hunting and crafts

Æfsin goddess of the hearth

Bındur patron of the hearth
(Madæ) Majrem (Mother patroness of children and marriage
Mayrem < St Mary)
Naf patron of the clans
Nælı Azuar saint of the sun, patron of male infants
Ærtxuron deity of the hearth-chain
Atınæg24 deity of fertility
Nıxı ʒuar patron of the newly-wed

24
From Greek St Athenogene, whence also Georgian Atenagani ‘St Thomas Sunday’, Ingush
Eting ‘name of a summer month’ (Abaev 1958: 82), cf. also the Khevsur name of the popular festival
Atengena.
 Heathen Heritage in the Caucasus 165

Ič’ına patron of new-born children and newly-wed girls

Bınatı xicau anthropomorphic deity, protector of the home;

often turns into a snake
Table 16. Ossetian deities of household and family

The Ossetian pandemonium was inhabited by quite a number of scary crea-

tures, to name a few of them:

Axærı dur stone-eater of the earth

Aminon god(ess)-doorkeeper of the Land of Dead
Barastır god of the Land of the Dead
Odesæg Death Angel (‘bereaving of the soul’)
Kæfqujndar Dragon
Kanʒargas seven-headed winged dragon
Mæjxor-tæ monster eating moon during the eclipse
Zaliag kalm a snake-like monster
Ʒinqur; Kaʒi Devil
K’ulıbadæg Witch
Wæju/ug demon, devil, evil giant often regarded as having
one eye and seven or sometimes hundred heads
Zın-tæ Jinn
Table 17. The Ossetian Pandemonium

Like in Abkhazia, the Ossetians have a concept of a deity combining the func-
tions of seven gods (Avd ʒuary ‘seven shrines’).
Nakh (Ingush and Chechen) traditional pantheon
Despite the fact that Islam is exceptionally deep-rooted in Chechnya and In-
gushetia, having become, according to specialists, an integral part of the national
identity, this state of affairs is historically speaking relatively recent, and the
strengthening of Islam in this part of the Caucasus is explained mainly by its
mobilizing role in the fight against Russia, both in the second part of the 19th
century, and during the conflicts in Chechnya in the post-Soviet period.
The elements of the pre-monotheistic religious system in Chechnya can be
summarized as follows. The Chechens and Ingush connected the appearance of
life with a gigantic white bird, from whose excrement there evolved water and
various plants. Others attribute the creation of the earth to the activity of the su-
preme god Deela. According to popular beliefs, the Sun and the Moon are two
166 Viacheslav A. Chirikba

brothers born of different mothers, who are being chased by their demonic sister
Mož; when she reaches them, there occurs an eclipse.
The gods Erd and Tusholi were protectors of agriculture, fertility and harvest.
The protector-spirits were taram, while there were also quite a number of evil
spirits or demons—almaz, ubur, gamsilg, etc. (Tokarev 1991: 604, 605) A more
detailed classification of the Nakh pantheon and pandemonium can be pre-
sented in the following tables.

Deela Supreme deity

Seela God of thunder and lightning
Maetcil God of agriculture and harvest
Gal-Jerdi God of cattle and breeders
Ištar-Deela Lord of life and death
Moliz-Jerdi God of war and victory
one-eyed Elta, son of Deela God of hunting
Taamaš-Jerdi (< St Thomas ?) Lord of fate
Table 18. Male gods of the Nakh Pantheon

Tušoli Goddess of spring and fertility

Darca-Naana Blizzard-Mother
Mokh-Naana Wind-Mother
Seelasat (lit. ‘Oriole’) Patroness of virgins
Table 19. Female gods of the Nakh Pantheon

The Nakh peoples also knew male (c’uu) and female (maelkh-aeznii, lit. ‘sun-
maidens’) pagan priests.

Elements of traditional cults in Dagestan

Islam entered Dagestan as early as in the middle of the 7th century. It embraced
all Dagestani communities and became the official religion in the 10th – 15th cc.
(Gadzhiev 1991: 6), earlier than in any other part of the North Caucasus. Apart
from Islam, in this mountainous region the other monotheistic religions—Chris-
tianity, Judaism and Zoroastrianism—were also known and practised. Due to
this factor, many elements of traditional (pre-monotheistic) religion have been
lost. However, some of the traditional beliefs and rites, which often operated un-
der a superficial Muslim guise, accompanied by Muslim prayers and terminol-
ogy, did survive (e.g. the cult of the saints, of the (formerly) pagan shrines, etc.).
 Heathen Heritage in the Caucasus 167

That is why ethnologists speak of “popular Islam” (народный ислам) in relation

to Dagestani syncretic traditional culture.
The Dagestanis worshipped sacred mountains (e.g. Axulgo, Shalbuzdag—in
Southern Dagestan and Northern Azerbaijan), sacred caves (as the cave of Di-
urk), as well as sacred groves, trees, stones, rivers, etc. Also revered were shrines
(called p’ir in Southern Dagestan), often dedicated to local saints.
The Islamic-inspired demonology distinguished between good (jinn) and evil
spirits (šajt’an). The spirits are believed to be able to turn into people, animals,
monsters, or become invisible. Some of them are envisaged as being small in
size, with the heels of their feet turned forward. The giant Eprit, borrowed from
the Koran (< Arabic Ifrit), inhabits calm waters; it can change its guise and in the
night-time it catches and eats people. A giant dragon-like creature called Azhda-
ha (< Persian) is believed to have one, three or seven-heads, projecting fire from
its maw; it sits near water springs, protecting them from intruders and demand-
ing in exchange for access to the water human sacrifices, mostly young women,
on whom it feeds.
Witchcraft and exorcism were a common practice. The Tsez quack doctor
used to make two anthropomorphic dolls (male and female) of old cloth, hang
them together with bread and cheese on a twigged stick and put them in the
abode of spirits called šatanisaλ (‘village of the spirits’). In order to do harm to
hateful people, it was a common practice to make a doll of cloth (as with the
Avars) or of the fatty tail of a sheep (as with the Dargwas or Laks), pierce it with
sharp objects (nails, needles) and hide it at the threshold of the persons(s) con-
cerned (Gadzhiev 1991: 24, 25, 40).
Belief in the evil eye, which can make people sick or even die, is a supersti-
tion shared by many cultures in the Caucasus and beyond. Especially feared
from this point of view were blond people with blue or green eyes (as with the
Avars), or red-haired people with blue eyes (as with the Laks), as these features
were probably associated with devils.25
In Mountainous Dagestan they believed in incubus—a creature abiding in
the houses which approaches sleepers during the night and smothers them by
closing their nostrils with its own nose without nostrils. By doing this, the crea-

25
This belief is shared by other Caucasian peoples as well; e.g. the Georgians were afraid of
people with grey eyes (cudi tvali ‘bad eye’) (Gadzhiev 1991: 39). The Abkhazian Prometheus-like
mythical hero Abrsk’ʲǝl fought against the vicious red-haired people with grey eyes, probably evil
demons. Grey, blue or green eyes were the feature attributed to witches (wǝd) by the Circassians.
168 Viacheslav A. Chirikba

ture can paralyze sleepers and hamper their blood circulation, releasing them
only at dawn. It can be invisible, but it can also appear as an ugly fat old woman
with big limp breasts and wearing rags (as with Avars and Laks), while others re-
garded it as a cloth ball (the Laks), a tri-cornered hat (the Avars), a buffalo (the
Dargwas), a cat (the Dargwas, the Laks, the Akhvakhs), a donkey or a billy-goat
(the Laks). It has various names in different Dagestanian languages and dialects:
Kibisxan, Ilbalxan, Kibiran, Čikabilsal, Kibils,26 K’vibilsi, Sijha, Simahada (Darg-
wa); Q’ehel, Hic’, Nisus, Tamiho, Rǝsisa (Avar); Suxasulu, Suxalutu, Accalav (Lak),
Rehe (incubus with one nostril), Ciccixaro (incubus-like demon appearing as a
skinny man) (Akhvakh), etc.
There is a certain ambivalence concerning the nature of the house spirits.
Thus, the Laks believed that the incubus-like creature Suxasulu can protect food
storages and the hearth (Gadzhiev 1991: 26-27, 29). The same can be said of the
Tabasaran Rux, a creature with one nostril and one eye in the forehead, regarded
by some as female. She combined features of an incubus and a boggard: it can
smother people during their sleep, or can make them rich or abundant in food-
stuffs (Seferbekov 2001: 141).
All Dagestani communities share the belief in a demonic woman, by some
accounts with big hanging breasts, one of which is brushed over her shoulder.
The Laks believed in a vampire-like creature that came out of graves and harmed
people: it could smother people during their sleep, could steal an unborn infant
from the womb, etc. In different Lak dialects it was called xxurttama, q’urgala-
ma, q’urtama, q’urt’ma, xartama. The popular belief had it that people who
during their life-time led an indecent way of life, were greedy, or exploited their
relatives, after their death turned into such creatures.
The Dagestanis believed also in good holy spirits, who live in difficult to ac-
cess places. They could take the guise of animals, reptiles or beautiful women,
and could help people or protect cattle. The houses were believed to be inhab-
ited by benevolent zoomorphic spirits, who protected the family and the hearth.
They were called kaž (Avar), q’une, tar, malč’un (‘snake’) (Dargwa), kini (Lak),
kine (Tsez), šibrit’arin bit’ar ‘basement snakes’ or šavɣar-bit’ ‘snake Shavghar’ (Ta-
basaran), mǝkmagar ‘ceiling snake’ (Rutul), ɣassagesse beka (Akhvakh), ƛ’uč’alul
berka ‘basement snake’ (Karata) and often imagined as legged snakes with

26
Cp. Armenian xipil-ik (with the diminutive suffix -ik) ‘nightmare’ (the parallel was pointed
out to me by H. Martirosyan).
 Heathen Heritage in the Caucasus 169

golden horns, or sometimes as frogs or lizards. Cf. also Agul ɣvadikken ‘boggard’,
which can be translated as ‘living under the ceiling’. But the Dargwa q’une was
imagined as a tall woman with a big bosom and long red hair, while the Taba-
saran word for ‘boggard’, Rux bab, contains bab/bav ‘(grand)mother’ (Seferbekov
2001: 140, 141). In some villages these creatures were regarded as kind and
attractive females, whilst in others they were of an uncertain gender that could
harm people. Snakes establishing a nest in the house were regarded as spirits of
dead relatives.
Hunting was one of the favourite occupations of men, and there are beliefs
associated with the cult of hunting. For example, Lak avdal//ovrdal huxču
(‘hunter’) or Avar budual are sometimes imagined as a deer, which is regarded as
a sacred animal; in traditional songs deer horns are likened to the sun. Such crea-
tures as Andi gogoči, Lezgi alpab, Lak almas xatun (< Turkic) are believed to be
able to live with the hunters (Gadzhiev 1991: 30). The Avars believed in a Forest
Beauty with long loose golden hair, who wears a white dress and does harm to
people, while the Dargwas believed in anthropomorphic Forest Men. Rutul
Kaškaftar was the god-protector of woods and nature, who lived in sacred
woods. It was imagined as a two-legged creature with one eye as big as a dish,
with red tongue, the shining torches as ears, and with all its body iridescent. The
Rutuls believed in T’urun q’ari—the spirit of the hills, regarded as a giant woman
wearing rags, with dishevelled hair, who was able to harm people (Jusupova
1988: 157, 158, 159, 166).
The Avar demonic female character Untul ebel ‘Mother of Illnesses’ was im-
agined as a naked child, who could make people fall ill or, on the contrary, could
heal them (Gadzhiev 1991: 22). Her Rutul counterpart Jadlanin was imagined as a
beautiful giant woman with long loose hair, who normally wore white clothes
and could either heal sick persons, or (especially if dressed in black) make
healthy people fall ill (Jusupova 1988: 159-160).
The essential part of the Dagestani traditional economy has always been ag-
riculture, hence the existence of various agricultural rituals and cults. Thus,
during the festival welcoming the spring, the Avars made ritual breads in the
form of a human figure, a ram, a horse, a bull, or a rooster. In Southern Dagestan,
during the processions dedicated to summoning the sun, they used a ritual
wooden doll (günü), dressed and decorated as a woman. In the ritual of sum-
170 Viacheslav A. Chirikba

moning rains there figured a specially made doll called Dodola;27 its Nogay name
is Ändir-Šopaj, in Dargwa—Vassamaj, or k’at’a bah, in Lak—ɣaral sihu. The
Dargwas used to make female-shaped dolls called Sutqatun (‘Slow woman’), or
Aminala k’at’a (‘Shovel of Amina’), which were used in processions during the
ritual of summoning rains.
The pan-Dagestanian festival of the first plough or the first furrow is now
being restored in some parts of Dagestan as a popular spring festival. Tradition-
ally, in the first furrow festival there figured people posing as wolves or bears and
using special clothes and masks. The procedure used to include such elements as
ritual nakedness or even sexual intercourse in the first-made furrow, symboliz-
ing fertility and the regeneration of the nature.28
People Deity Description
Avar Raƛ’ul ebel Mother of the Earth
Avar, bečed supreme god, god-creator
Avar, Andi, c’ob
Tsez, Akhvakh
Dargwa
Tabasaran c’alla
Lak umčar
Rutul Tsakhur zal
Tat jiniš
gǝniš
ofirogor
Lezgi Alpan god of fire
Lak Vilax
Avar, Karata λadal ebel Mother of the water
Lak Ass god of thunder and lightning
Tsakhur Arš
Lezgi Ɣuc’ar
Lak Idor deity of harvest
Lak Züvil deity of rain

27
The Dagestanian doll Dodola and the ritual strikingly resemble the Balkan rituals for sum-
moning rain, whereby girls called Dodola would undressed and put on leaves, flowers and herbs to
perform the rainmaking ceremony. The Balkan Dodola is regarded as being connected with the
Slavic cult of the thunder-god Perun (cf. Tokarev 1991: 391). I thank H. Martirosyan for drawing my
attention to this Balkan parallel.
28
Cp. a similar sexual ritual of the first furrow in Armenia (H. Martirosyan, p.c.).
 Heathen Heritage in the Caucasus 171

Tabasaran Gudil
Rutul Gudi
Tsakhur Godej
Lezgi Pešapaj
Dargwa Zaburaj female deity of rain, or the doll
Vassamaj used in the ritual of summoning
k’at’a bah rains
Sutqatun
Aminala k’at’a
Avar Dodola
Nogay Ändir-Šopaj
Lak ɣaral sihu
Karata Mučulal ila ‘Mother of the the deity of the wind
wind’
Mučulal dada ‘Father of the
wind’
Avar Temirhojas goddess of weavers
Lak Avdal//Ovrdal huxču (‘hunter’) hunting deity, protector of wild
Tsakhur Abdal beasts
Avar Budual
Khvarshi Budalla
Akhvakh Budalaal
Karata Budal(d)i
Rutul Xidir-Nebi
Akhvakh T’at’aha the lord of the forest
Avar Untul ebel ‘mother of illnesses’ goddess of diseases
Roqdulaj
Khvarshi Leλλnas (išu) ‘mother of
Akhvakh illnesses’
Roƛ’eroλλi ila ‘mother of
Karata illnesses’
Ruƛ’aril ila ‘mother of illnesses’
Rutul Jadlanin
Zalzanagij
Lak Acalov
Avar kaž spirits protecting the family and
Dargwa q’une, tar, malč’un the hearth
Lak kini
Tsez kine
172 Viacheslav A. Chirikba

Tabasaran šavgär
benegjut
Rutul tušedrǝšbe
Tat numnegir
Lak Suxasulu protector of food storages and
the hearth
Rutul Kaškaftar the god-protector of woods and
nature
Rutul Sifi-Havalǝj goddess-protector
Table 20. Elements of the Dagestani Pantheon

Khvarshi qubal demon, devil

Khvarshi Risisan goblin, house-spirit, boggard which
strikes an awakening man with his
paw
Avar Rox-dulaj demonic woman
Untul-ebel (ebel ‘mother’)
Al-bab (bab ‘mother’)
Khvarshi Qartaj
Karata Albastǝ/i ,
Rutul T’urun q’ari
Accalav
Lak
Akhvakh λeλλise ‘one who lives in the water demon
water’
Akhvakh Xxatu beautiful mermaid, who seduces
men
Dargwa Kibisxan, Ilbalxan, Kibiran, Incubus
Čikabilsal, Kibils, K’vibilsi,
Sijha, Simahada
Q’ehel, Hic’, Nisus, Tamiho,
Avar Rǝsisa, Suxasulu, Suxalutu
Karata Risidobo
Lak Accalav
Tabasaran Rux
Lak xxurttama, q’urgalama, vampire-like creature, graveyard
q’urtama, q’urt’ma, xartama demon, who scares people who lead
Lezgi malkamut a sinful life
Avar q’av
 Heathen Heritage in the Caucasus 173

Bezhta q’av
Tabasaran Kuruzai//Kuruzan
Tabasaran kaftar-janavar werewolf29
Avar Avliħune demon stealing a foetus from the
Lak Almas-xatun mother’s womb, or doing harm to
Lezgi Al-pab the baby
Agul Albasti
Andi Gogoči
Tat Dedej-ol (dedej ‘mother’)
Tabasaran Jarsel demon killing a young mother after
Al childbirth, when she is left alone in
the birthing room
Tabasaran Meze demon impersonating children’s
skin diseases
Lak Kaftar-kari anthropomorphic evil demon
Tsakhur Alna jed
Tabasaran Kuš-kaftar
Lezgi, Rutul Kaš-kaftar
Table 21. Elements of the Dagestani Pandemonium

Georgian traditional cults

Despite early Christianity (from the 4th c.), the Georgians never really forgot
their traditional cults, merging some of the Christian elements and characters
with the earlier beliefs to produce, typically for many Caucasian traditions, an
amalgam of elements of both pantheistic and monotheistic systems. Especially
well preserved were traditional cults in conservative mountainous regions of
Georgia, such as Tusheti, Pshavi, or Khevsureti. All deities of the East Georgian
pantheon were perceived as anthropomorphic creatures with a strong propen-
sity to metamorphosis (Bardavelidze 1957: 10).
The Supreme God Ghmerti (ɣmerti) was imagined as abiding in the skies, sit-
ting on a golden throne by the gates of the sky. Ghmerti was the supreme law-
maker and maintained order both in the skies and on the earth, hence his epi-
thet Morige (“On Duty”). By the ‘gate(s) of God’ (ɣvtis k’ari), always after sun-set,
there sometimes assembled the other deities and even the evil spirits kadzhis,
and it is there where the Supreme God would hold a court examining the argu-
ments between the subordinate deities. Moreover, Ghmerti decided the fate of

29
Cp. kaftar in various Dagestanian languages with Arm. k‘awt‘ar ‘hyena’, ‘witch’, Azeri kaftar,
all from Persian kaftar ‘hyena’ (H. Martirosyan, p.c.).
174 Viacheslav A. Chirikba

the yearly harvest; if angry, he could send hail on the crops or illnesses on peo-
ple.
The sun-goddess Mze (also known as Mze-kali ‘sun-woman’, Dɣe dɣesindeli
‘present day’ in Khevsurian, and p’irimze ‘sun-faced’ in Mtiulian) was accompa-
nied by angels (angelozni//c’mida angelozni ‘angels//holy (‘clean’) angels’). Her
epithet, reflecting her main function, was ‘the shrine of the field’ (mindvris ǯvari);
she could rule the clouds, and the grain harvest was largely dependent on her
benevolence. If she was angry with people, she could destroy the harvest, but in
a good mood she would send good weather and protect the harvest (Bardaveli-
dze 1957: 2, 10, 15, 17-8, 31).
The other powerful god was K’viria. His epithets are xmelt mouravi ‘the ruler
of the dry lands’ or k’araviani ‘the one having a marquee’. Kviria was door-keeper
to Ghmerti and used fire to punish sinful people. He was an intermediary be-
tween Ghmerti and his sons (ɣvtis-švili) and served as the chief of the angels.
We have thus a triad of high ranking gods, headed by the supreme god
Ghmerti. It is supposed that Ghmerti was the god of some nightly luminary and
regarded as male, whereas Mze (‘sun’)—was the deity of the daily luminary and
regarded as female.
The lower ranking deities included communal/local patrons and territorial-
ethnic or tribal deities. Local deities could be subdivided into ǯvari (lit. ‘shrine,
cross’), xat’i (lit. ‘image, icon’), and ɣvisšvili (< ɣmrtis švili ‘god’s son’), or ɣvtis-
nasaxi ‘one created in the image of god’, ɣvtisnabadebi (< ɣmrtis nabadebi ‘born
of god’). This shows that local deities were perceived as being in kin relationship
to God (Bardavelidze 1957: 2, 3).
An example of a communal deity was the Pshav female deity Tamar akim-
dedupali (Tamar, the patroness (dedupali) of doctors (akimi) and healing). More
local territorial Pshav communes had eleven patron saints (xat’i). Still lower in
rank were the deities of separate village communes, as well as family and clan
communal units, with strict observance of the hierarchy between them.
The tribal deities included, for example, the pan-Pshav deity Lašaris ǯvari, the
pan-Khevsurian deity saɣmrto gudanis ǯvari, the pan-Mokhevian deity sameba//
samebis xat’i (lit. ‘Trinity’, ‘Trinity’s icon’, under Christian influence), and the
pan-Mtiulian and Gudamaqarian deity Lomisas xat’i (“Lion’s shrine’). The major
deities of the Khevsurian pantheon were K’op’ala, his blood brothers Šubnuri
and Xaxmatis ǯvari. The Khevsurian lower deities included dobilni in the guise of
very small children, who accompanied sun-beams during the sun-rise and sun-
 Heathen Heritage in the Caucasus 175

set; they attacked mainly women and children and were the cause of their
illnesses.
Pshav hunters dedicated themselves to the patron spirit of the place, called
“mother of this place” (Čursin 1957: 38). The Pshav and Khevsurian deity-black-
smith P’irkuši(s ǯvari) was associated also with thunder clouds and the lightning
(Bardavelidze 1957: 5, 12, 17, 23, 111).
The will of the god was made known through meene or kadagi, the custodians
of tribal or communal shrines. A person called meene or kadagi used to sit down
in the sanctuary (ǯvari) during the New Year’s festival and make prophecy on
behalf of god’s sons (ɣvtis-švili) about what happened by the gates of Ghmerti,
whether the community had to face hail, bad harvest or illnesses, enemy attack,
bloodshed or death. The custodians of the shrines had to know sacred texts
(xucoba, sadidebeltaj ‘glorification’), which were transmitted orally from one
generation of custodians to the other; in these texts the gods were mentioned in
strict hierarchy, starting from Ghmerti, then Mze and Kviria. The soothsayer
(xuces-meene) was feared, as his words could be fulfilled, as believed, by god’s
sons (ɣvtis-švilni). After his death, the soothsayer’s grave could become a shrine
(Bardavelidze 1957: 4, 5, 16, 32).
The gods were assisted by spirits having the appearance of wolves (esaul),
snakes (gvelisperni), hunting-dogs (mc’evarni), or very small children (dobilni);
the other assistants included laškarni ‘fighters’, various angels, etc. The Great
Mother and goddess of fertility Nana was imagined as having fractions repre-
sented by bat’onebi ‘masters’, or by angels.
The pagan sanctuaries (ǯvar-i) were situated in sacred trees/groves; they rep-
resented constructions built on more or less flat surfaces and consisted of a ritual
tower (k’ošk’i) containing niches or bays, the central room/temple (darbazi), a
barn (beɣeli) or a brewery (salude), some other apartments for the personnel
serving the temple and finally ‘the gate of the shrine’ (ǯvris//xat’is k’ari). The East
Georgians regarded falling meteorites as the flying xat’i/ǯvari, which move from
one place to another.
The important element of the agricultural cult was the ritual of summoning
rain, where there figured a cross, clothed as a woman or a bride, called in the
Tush dialect sac’vimara gugai ‘rain doll’.
The hunting cult included such characters as the Rachan ‘Angel of the rocks’
and the ‘shepherd of the beasts’—the hunting god, sometimes identical with St
George (Virsaladze 1976: 30), cf. also the Khevsurian god of hunting Očopintre.
176 Viacheslav A. Chirikba

East Georgian Xaxmatis ǯvari god-patron of fertility and reproduction

of humans and animals
Pshav, Khevsurian P’irkuši(sǯvari) deity-blacksmith
Gurian Aguna, Racha, Lechkhumi deity-patron of wine-growing
Angura30
Pshav Tamar akimdedupali Tamar, the patroness of doctors and
healing
Bombɣa//Basila (probably < St Basil) phallic deity, patron of fertility,
responsible for the proliferation of house
animals and men, and for the harvest of
cereals
Elia god of thunder and lightning
Nana Great Mother, goddess of fertility, who
had fractions (bat’onebi or angels)
Kal-Babar//Barbare (St Barbara) goddess of sun
Khevsurian Očopintre god of hunting
angelozni//c’mida angelozni ‘angels//holy angels’
Giorgi (< St George) pagan deity
Gudani pagan deity
Iaxsar pagan deity
Samʒimari (‘necklace-wearer’)31 pagan goddess with golden hair
Table 22. The Georgian Pantheon

kaǯi evil spirit

devi evil mythical giant
ali demonic woman
K’op’ala the lord of demons
Pridon Demon
Muza mythological giant
Table 23. The Georgian Demons

The Svan traditional cults

Of all Kartvelian peoples the Svans, thanks no doubt to their isolated mountain-
ous lifestyle, maximally close to, and dependent on, Nature, have preserved their

30
Bardavelidze (1957: 74) associates these names with Circassian aguna, magical cask with the
drink of the Narts and with Persian angu:r ‘grapes’.
31
Etymology by G. Charachidzé (cited in Tuite 2007: 169).
 Heathen Heritage in the Caucasus 177

traditional religious system much better as compared to any other region of con-
temporary Georgia. The male triad of the Svan pantheon is the supreme god
Xoša ɣermet//Xoša ɣērbät, his vizier Tarinӡel//Targlezer and the moon god
Ǯgǝræg, patron and protector of people (especially males), augmented by the
female goddess of motherhood, fertility of the land and cereals Lamaria (St
Mary). The supreme god is regarded as the lord of the skies (upal deceš) and the
universe (Pusnabuāsdiš), the great creator of the world (Bardavelidze 1957: 14, 44,
171, 172).
Each Svan village or even a quarter of a community had its own shrine. The
sanctuary was looked after by its own custodian called ‘key-keeper’
(mok’il//mek’il), who kept the key of the shrine and its treasury. The key-keeper
had a rather high status, he could replace the priest (bap’), and sometimes he
himself was a legitimate priest.

ɣerbet// xoša ɣermet//xoša ɣērbät Supreme God

Barbol//Barbal//Kal-Babar goddess of sun (St Barbara), also of contagious
diseases and the fertility of cattle
Taringzel//Tarinӡel//Targlezer deity, vizier of the great god
Pust/d//Pusnabuasd/Pusnabuāsdiš supreme deity of big cattle//great god
Ǯgǝræg//Molli Ǯgǝræg the moon god, patron and protector of people,
(< St George)32 especially males, patron of hunters
Tulepia-melia//Melia-Telepia deity of fertility
Tar-Bednier33 deity of sheep
Lamaria (St Mary) goddess of motherhood, fertility of land and
cereals, of the hearth
Daal goddess of wild animals in the mountains (cf.
Nakh Deela)
Q’ur-ša (‘black-ear’) dog of the goddess of hunting
Apsat patron of birds and fish
Ber-šišwliš supreme god of hunting and wild animals
Ali female evil forest spirit
Mamberi patron of wolves
Table 24. The traditional Svan Pantheon

32
From Megrelian ǯgiri givargi ‘Good St George’ (A. Shanidze’s etymology, cited in Tuite 2005:
165).
33
Probably of the same origin as Abx Ajtar, cf. above.
178 Viacheslav A. Chirikba

The Svans had various calendar festivals, like Lipanali—a New Year’s animis-
tic festival of spirits of ancestors, the phallic ritual Melia-Telepia, connected with
the fertility and nature cycles: the rite of sakmisaj, where figured a wooden
phallus (Chartolani 1988: 188, 189, 190), etc.
Hunting played a crucial role in the Svan economy and was a regular occupa-
tion of males. The Svans know the Goddess of wild animals Daal, imagined as a
beautiful naked woman with long red hair, who can adopt the guise of an ani-
mal. She was accompanied and assisted by her dog Qursha, born from the egg of
an eagle. By cutting Daal’s hair, the hunter could escape her (Virsaladze 1976: 30).
The other characters of the hunting cult are the supreme god of hunting and
wild animals Ber-šišwliš, the god of forest animals—the forest angel cxekiš an-
gelwez, the patron of birds and fish Apsat. In prayers and ritual incantations the
Svans mentioned the names of some of the gods together with the indication of
the location of their abode: Daal of the rocks, Apsat of the rocks (Virsaladze 1976:
30). It is believed that the female evil forest spirit ali can, if content with the
people, tie up the wolf’s jaws, thus protecting the cattle, or, when angry with
them, she can set the wolf on their cattle (Bardavelidze 1957: 47).
Typically for a hunting society, the Svans worshipped the big predator, the
wolf, regarding it as a totem or a sacred animal, Mamberi being the patron of
wolves. It was forbidden to kill the wolf, and the one who by chance happened to
have killed one, had to apologize for its death, bewail it as a member of his own
family and bury it (Bardavelidze 1957: 45). On the other hand, the Svans prac-
tised the ritual killing of the wolf with the subsequent procession in the village;
the ritual was called Ašangelo//Šašangelo (cf. Georgian samgelo ‘for the wolf’,
‘pertaining to wolf’) (Bardavelidze 1957: 45).

Megrelian traditional cults

The Megrelians, like their neighbours the Abkhazians, Svans and Western Geor-
gians, are characterized by a natural syncretism of traditional pantheistic cults
with superimposed Christianity. Some elements of their pantheon are depicted
in the following table.

ɣoronti Supreme God

tkaši-mapa goddess of hunting (lit. ‘the queen of the tree(s)’)
mesepi masters of animals
galenišiši orta deity (galenišiši ‘of the exterior’)
 Heathen Heritage in the Caucasus 179

Ǯaǯu stone woman34

očo-k’oči the forest (lit. ‘goat’) man
č’ink’a goblin
Table 25. Elements of Megrelian Pantheon

Armenian traditional cults

Armenians embraced Christianity quite early (4th century), and the Christian
creed became the integral part of their ethnic identity, the Armenian Apostolic
Church providing spiritual guidance for the whole nation. Such profound devo-
tion to the monotheistic doctrine, which acquired the status of a pan-national
ideology over the periods of struggle against Zoroastrian Iran and later numer-
ous Muslim incursions, left little place for the remnants of the heathen past,
pushed firmly into the realm of folklore, superstitions and folk beliefs, or at best
acquiring the superficial attire of Christian rites.
Vahagn god of thunder and lightning, warrior god, dragon-slayer35
Aramazd Supreme god, god of thunder and lightning
Anahit goddess, daughter/wife of Aramazd
Nanē daughter of Anahit
Mihr fire-god, Aramazd’s son (< Iranian Mithra)
Astłik (‘little star’) goddess of love, beauty and water, goddess of Venus, the
mistress of Vahagn
Tir//Tiwr pagan god of scribes, with some features of a sun-god
(Petrosian 2007: 184)
Table 26. Elements of Armenian Pantheon

Some of the formerly pagan rites became associated with Christian festivities
and are still widely celebrated—for example, the pagan festival of water, plants
and fertility Vardavaṙ, later associated with Christ’s Transfiguration. Cf. also the
popular cult of Barekendan, associated with Christian Shrove, the popular Festi-
val of Flowers or the harvest cult, called Hambarjum and associated by the
church with the Ascension of Christ.
dew mythical monster
ar(a)lez half-men, half dogs, spirits, who descend from the sky in
order to lick the wounds of those killed in the battle, thus
reviving them

34
Cf. the name of the Abkhazian goddess of harvest Ǯaǯa.
35
Cf. A. Petrosian (2007: 181).
180 Viacheslav A. Chirikba

kʽaǯ(kʽ)36 demons
oguz Tork‘ (or Turk‘) mythical giants
Angeł(eay)
xipilik incubus with hollowed palm of the hand, who causes
nightmares
xoeǯeloz incubus, nightmare
al(kʽ) evil spirit
ǯanavar vampire
Lekion monster-fish
višap ichthonic monster, resembling a winged dragon or a
winged snake and associated with the cult of water; višap
eats the sun, which causes eclipses
Table 27. Elements of the Armenian Pandemonium
The other characters of the traditional Pandemonium are devil-blacksmiths,
mermaids, half-people, half-fishes, witches. The vampire-like creature gornapštik
was imagined as a dog or a cat, which came out of grave and frightened people,
sometimes turning them mad or causing their death. Those who were buried
without the observance of Christian ceremony or the souls of non-Christians
(Turks or Kurds) could become such creatures. The Armenians also worshipped
sacred trees. As in other parts of the Caucasus, in the ritual of summoning rain
they used a specially made doll called Nurin.
A common Caucasian mythological system?
It would certainly be a challenging task to investigate whether it is possible to
reconstruct the contours of a pan-Caucasian pre-Christian and pre-Islamic reli-
gious system, or whether such a system ever existed. On this theme the early
20th century Russian-German Caucasologist Adolf Dirr (1915: 13) wrote: “In gen-
eral, while studying the most ancient mythological concepts and beliefs of the
Caucasians, one cannot but think that there was once in the Caucasus one reli-
gion, which was then changed and partially replaced by the historical religions;
but it is still preserved by many Caucasian peoples in the guise of some rem-
nants, superstitions and in folklore”. The Russian Soviet ethnologist Grigorij
Chursin (1957: 63) also noted that “we have some grounds to claim that the sig-
nificant part of the Caucasian peoples, on both sides of the Main Chain, includ-
ing here also the Kartvelians, had in the past similar religious beliefs, connected
with the worship of thunder and lightning, with similar forms of the cult”.

36
-kʽ here and below in al(kʽ) is a (petrified) plural suffix.
 Heathen Heritage in the Caucasus 181

Indeed, the comparison of mythologies in various parts of the Caucasus does

reveal striking parallels. As expected, we certainly observe interesting and close
affinities between the traditional systems of the neighbouring communities. But
even comparing the traditions of the Caucasian communities situated much far-
ther from each other—taking, for instance, Abkhazia and Armenia—we some-
times find amazing parallels, some of which, admittedly, can be of a broader ar-
eal nature.
Comparative analysis of the mythological systems of all Caucasian communi-
ties does yield interesting results and can bring us to the necessity of postulating
a largely common pre-monotheistic heritage of the Caucasian peoples, irrespec-
tive of their ethnic and linguistic affiliations. It may well be that, parallel to the
postulation of the Caucasian language union or Sprachbund (Chirikba 2008), we
have no less reason to consider the reality of a common Caucasian mythological
paradigm. A pan-Caucasian (pre-Christian and pre-Islamic) mythological and
cultural union, or, in other words, a Caucasian mythological area unites both
close and distant parts of the Caucasus within a specific set of mythological fea-
tures shared by the majority of Caucasian cultures and traditions. Though many
of these features can also be found beyond the geographical limits of the Cauca-
sus, it is their set or cluster which is specific to the Caucasus area and which al-
lows us to speak of the Caucasian mythological area.
In order to substantiate the idea of a pan-Caucasian mythological system,
one has to set up an inventory of the traits common to traditional systems of all
the Caucasian peoples, including those speaking non-Caucasian languages (Os-
setians, Armenians, Kumyks, Karachay-Balkar, Azeris, etc.), and reconstruct the
main elements of such a paradigm. Many elements undoubtedly have a broader
areal context, going beyond the geographical confines of the Caucasus area.
The tentative contours of such a pan-Caucasian system would include the
following traits:
1. The separation of the world into three horizontally situated realms: up-
per/celestial (populated by gods), middle (populated by humans, animals and
plants), and the lower world (or underworld, populated by demons, devils,
dragons and other chthonic creatures).
2. The world tree, which connects the three worlds.
3. The presence of a Supreme god, aided by his lieutenants.
4. Sacred groves, trees, woods and mountains.
5. The Moon as a male deity and the Sun as a female deity.
182 Viacheslav A. Chirikba

6. A dragon or a demon eats the sun or the moon, thus causing eclipses. People
scare the dragon by creating much noise.
7. Deities such as patrons of the sea, rivers, ravines, caves, woods, mountains.
8. The thunder-god.
9. The hunting god.
10. The cattle and procreation god.
11. The rain god, and the ritual of summoning rain using a doll brought to the
river as a probable replacement of human sacrifice.
12. The smithy god and the important role of the smith.
13. The harvest and agricultural god or goddess.
14. The mermaid who seduces lonely travellers.
15. The forest man and woman (almasty, al).
16. The sacred animal (e.g. cow or bull) sent to people as a divine sacrifice.
17. Incubus with a hole in the palm of the hand or with one or no nostrils.
18. The house snakes as spirits protecting the home.
19. The vampire-like creatures.
20. The birth of a hero from a rock or a stone.
21. A Prometheus-like hero punished by god by chaining him in the cave or on a
rock (Abkhaz Abrsk’ʲǝl, Circassian Nasren-žač’ʲe, Georgian Amirani, Lak Amir,
Ossetian Amran, Armenian Mher).

The main mechanism for the formation of the Caucasian mythological area
was the dispersal of religious beliefs, cults and rituals across the areas populated
by the Caucasian communities living in a similar geographical, cultural and eco-
nomic environment, characterized by similar sedentary patriarchal systems and
the important role of agriculture, animal husbandry, hunting and warfare. There
is no doubt that convergence between the various Caucasian peoples on linguis-
tic and mythological levels involved also such systems as folklore, customary law
and behavioural patterns. This allows us to see the Caucasus area holistically, as
a well-defined union of linguistic, mythological, cultural and folklore systems,
which can be assumed under the common name “Caucasian civilization”.
Another additional factor, which strengthened the similarities between the
various Caucasian communities, was the commonality of external influences. All
North Caucasian and some South Caucasian cultures share a common Islamic
superstrate religious and cultural system, and the same can be said of a common
Christian superstrate layer for some North Caucasian and South Caucasian cul-
tures. It is remarkable that after the spread of the monotheistic religions, many
traditional systems of the Caucasus, functioning parallel to the mainstream de-
 Heathen Heritage in the Caucasus 183

nominations, incorporated rather harmoniously some of the elements of the

new, superstrate systems.

Local mythological systems: the case of the lightning deity Čop(p)a

Like the Caucasian linguistic union, whereby apart from the pan-Caucasian lin-
guistic area there exist also parallel local Sprachbünde (cf. Chirikba 2008: 68), we
can clearly see in the Caucasus also the local mythological areas having traits
common to neighbouring cultures, which form a lower-level mythological union.
These local areas are:
(a) The Western Caucasus area (Circassia, Abkhazia, North Ossetia, Karachay-
Balkaria);
(b) The Central Caucasus area (Chechnya, Ingushetia, partially North Ossetia);
(c) The Eastern Caucasus area (Dagestan);
(d) The Kartvelian area (Georgia, Svanetia, Megrelia, partially South Ossetia and
Lazistan);
(e) The South Caucasus area (Azerbaijan, Armenia).
As an example of a local mythological union I shall take the cult of lightning,
widespread in the past in the Western Caucasus, including Circassia, Abkhazia,
Karachay-Balkaria and (North) Ossetia. This is the area where the influence of ei-
ther Christianity or Islam has always been weak and where the traditional beliefs
have been preserved much better than in the majority of the other regions of the
Caucasus. Though modern-day Ossetia in sensu stricto is not a part of the West-
ern Caucasus area, historically speaking, until the Mongol invasion in the 13th
century, the Ossetians used to live much more to the west, in the Western Cau-
casus, directly bordering on Circassians and Abkhazians.
Let us turn to the Abkhazian ritual Atlar-čʲopa, carried out when a person
was killed by lightning. The victim’s body was put on a specially made high
wooden platform, and a round dance was performed by people all dressed in
white clothes. The dance was accompanied by the song “Atlar-čʲopa”, sang in two
parties.37 One party sang “Ajtar atlar”, while the other replied by “atlar čʲowpar”

37
The name of the song has variants: A-f-r-aš⁰a ‘the song of Afy (the thunder-god)’, or A-nc⁰a-r-
aš⁰a ‘the song of God’; note the 3rd p. plural possessive prefix r-: ‘thunder-god/god-their-song’,
which indicates that gods were regarded as plural personalities. Besides the ritual consecrated to
the lightning strike, this song was also used during the ritual connected with popular healing of the
people sick with neurological disorders known as St-Vitus-Dance or chorea (an abnormal involun-
tary movement disorder) and called in Abkhaz aršǝšra ‘fever’, lit. ‘boiling’. Cf. its description in Ak-
aba (1984: 71).
184 Viacheslav A. Chirikba

(the words atlar and čʲopa, čʲowpar have no meaning in Abkhaz). During the
procedure, the people were not supposed to show any sign of grief, in order not
to cause the anger of the thunder god Afy. The same ritual was performed if the
man was wounded by lightning, or if an animal was killed by it (cf. Chursin 1957:
55-63). As a sacrifice, a white he-goat was slain (Akaba 1984: 74).
This ritual has obvious areal dimensions, as it reveals close parallels in the
neighbouring (West) Caucasian communities.38 G. Chursin noted a similar ritual
reflecting the cult of the lightning and thunder as performed by the Circassians,
in which the thunder god Shible and Jalija (St Elias) were mentioned. According
to him, the Kabardians used to perform the round dance čoppa when somebody
was struck by lightning (Chursin 1957: 58, 61). As described by M. Kantaria (cited
from Tuite 2004: 147), ‘Should a person be slain by a thunderbolt, no signs of
mourning were permitted; the survivors consoled themselves with the
knowledge that [thunder-god. – V. Ch.] Shible had brought good fortune to their
family by his touch’. <…> The womenfolk performed a round dance in honor of
Shible over 7 days, while singing a song to this deity including the refrain: cop’ai,
elari, ilia’.
A similar cult existed among the Turkic-speaking neighbours of the Kabardi-
ans, the Karachays and Balkars. When somebody was struck by the lightning, the
people would organize a ritual procedure accompanied with a song, which in-
cluded the words “čoppa”,39 or “Elliri (St Elias) čoppa” (Chursin 1957: 61). There
exists a sacred grove in Karachay called Čoppa-čila ‘[belonging to] the perform-
ers of the dance choppa’, which is revered by the inhabitants of the Chegem
gorge. Also known are sacred stones called čoppanı tašı ‘stone of choppa.40 It
seems that of all the Caucasian cultures it is Karachay and Balkaria where the
cult of choppa played the most prominent role, as witnessed by its description in
Karaketov (1995).
The Ossetians used to perform a very similar ritual, which included the dance
around the victim struck by the lightning and the singing of the song coppaj. The
song was sung in chorus and comprised the following words: “O, Elia, Elia, ældari
coppaj”; Elia is St Elias, and ældar means ‘lord, prince’. It was not allowed to
show any sign of grief, as it would anger Elia. The victim’s coffin was put on a
platform and kept there for 8 days (Abaev 1958: 314).

38
Cf. detailed descriptions of this West Caucasian cult in Karaketov (1995) and Tuite (2004).
39
Some Karachays pronounce the word as šoppa (Karaketov 1995: 119, 150).
40
Cf. http://karachays.com/index/0-265.
 Heathen Heritage in the Caucasus 185

It is probable that, due to a much better attestation of the cult of the thunder-
and lightning god čoppa(j) among the Karachays and Balkars, the čop(p)a//
čop(p)aj ritual reflects an old Bulgarian or Qypchak/Bulgarian thunder or rain
deity, which through the intermediary of Hazars/Huns and/or Karachay-Balkars
spread in the Western Caucasus among the Ossetians, Abkhazians and partially
the Kabardians. Yet the word čoppa(j) does not seem to have any obvious Turkic
etymology.
One of the possible connections is seen by some in the Iranian languages (cf.
Kurdish čōpī, čūpī ‘mass dance, in which participants are dancing hand in hand,
forming a circle or a semi-circle’,41 čōpīčamar bastin ‘to form a circle around the
dead body in order to perform a funeral dance’, and with a metathesis: čamar-
čōpī ‘a funeral round dance performed together by men and women around the
deceased’s body’ (Tsabolov 2001: 244)). R. Tsabolov further connects the Kurdish
word with Baluchi čāp ‘dance’, Persian čūpī ‘round dance with kerchiefs, per-
formed by men and women in the western part of Iran’, Luri čupī ‘local Luri
dance’, Bakhtiari čūpī ‘mourning ritual round dance’, Laki čȗpī ‘mourning mel-
ody’. He tentatively mentions here also Sanskrit cópati ‘(it) moves’.
Apart from the Western Caucasus, the term čop(p)a is attested in Nogay
Ändir-Šopaj, designating the doll used in the ritual of summoning rains. Note in
this connection that besides being used in the ritual devoted to the lightning’s
victim, the Ossetian word coppaj was also used in the song performed during the
ritual of summoning rains in the times of drought (cf. Chursin 1957: 61). The Ka-
rachays too addressed the prayer for rain to the deity Čoppa. Nogay Šopaj is re-
garded as being connected with the West Caucasian name of the god Čoppa,
whereas the part Ändir is unclear; Jarlykapov (1998: 44) suggests that it may
somehow be related to the name of Indra, the Indian God of War, Storms, and
Rainfall.42
M. Karaketov points out that outside the Caucasus a ritual for summoning
rain called čupke-kiju (translated as ‘the pouring water on Chupke’) is known
among a branch of the Volga Tatars known as Kryashens (Russian кряшены, i.e.
‘Baptized’). Also, a similar ritual čup-botkası was known among the other group
of Orthodox Christian Tatars called Nagajbak, or Nağajbäk (Karaketov 1995: 102),

41
The resemblance of the Caucasian čop(p)a with the name of the Kurdish round dance čopi
was noticed by me in 1976; it was also noted in Tuite (2004: 152) and in Tsabolov (2001: 244). Cf.
tentative etymologies of čop(p)a in Tuite (ibid.).
42
But cf. probably Karachay ındır ‘threshing floor’ (?).
186 Viacheslav A. Chirikba

who live now in the Urals and are regarded as descendants of either a Nogay-
Qypchak tribe, or of Kazan Tatars.
The Old Armenian text of “The History of the Country Aluank” (i.e. Caucasian
Albania), attributed (in its main part) to the author of the middle of the 8th c.
Movses Kalankatuatsi (or Dasxurantsi), contains a word which can be connected
with the term under discussion. While referring to the baptizing of the Turkic-
speaking Huns (or Khazars ?), living in the North Caucasus, by Albanian bishop
Israyel, the author mentions the heathen “thunderous roar of the graveyard
čopa” (cf. Kalankatuatsi 1984: 131), which had to be demolished. The English
translation of this passage made by C. Dowsett reads somewhat differently: “the
so-called royal graves of the thunder č‘op‘ayk‘ ” (cit. from Kalankatuatsi 1984: 213).
Note also the more recent R. Bedrosian’s translation: “the graves of the thunder-
ch‘op‘ayk‘//the royal graves of the thunder-ch‘op‘ayk‘.”43
In the Armenian original, the word is used in the form of the plural genitive
(č‘op‘ayicd), which, according to Sh. Smbatian (cf. his comment on p. 213 in
Kalankatuatsi 1984), presupposes the singular č‘op‘ay. The Armenian commenta-
tors of this passage, including Smbatian (ibid.), explain the word č‘op‘ay as refer-
ring either to a horse scull, or to the head, skin and bones of a horse given as sac-
rifice to sacred trees.
In the beliefs of the Karachays, probably the close linguistic relatives of the
Huns, we can see an obvious parallel to the connection of Hun čopay with the
sacred grove/graveyard in Kalankatuatsi’s account. One may compare the de-
scription of the Karachay ritual čoppa performed around the sacred pine-tree
ǯangız-terek. During the dance around the tree, the participants asked it to send
down to them a rich harvest, the rainy spring, the sunny summer and warm au-
tumn. The ritual was accompanied by the leading of a goat kid around the sacred
tree. At the end of the ritual the priest slew the goat and it was prepared and
eaten by all the participants. Its head and skin were hanged at the tree and left
there.44
Among the Karachays the ritual song choppa was performed also around a
mad person (Karaketov 1995: 45); cf. the association of the Abkhazian dance/
song čʲopa with mental illness chorea. The Ossetians, in their turn, associated
this term with compulsive movements (Abaev 1958: 315). In Abkhazia, in case of

43
http://rbedrosian.com/md14.htm#42.
44
http://karachays.com/index/0-265.
 Heathen Heritage in the Caucasus 187

a mental illness, they invited the augur (ac’aaj⁰ə), who would confirm that the
family was visited by the “great lord Afy” and give recommendation on the cure.
Then they performed a ritual devoted to Afy, which consisted in prayers, the
slaying of a white he-goat, its eating by the members of the procedure, the
singing of “the Gods’ song’ (Anc⁰a raš⁰a), the raising of a pole with the skin of the
slain goat attached to it, and the sacred round dance, with people jumping and
circling to the left and to the right (Inal-ipa 1965: 532-533). The attaching of the
goat skin to the raised pole is similar to its hanging on the tree by the Karachays
and to the horse skins given to sacred trees by the Huns.
Another enigmatic mythological term contained in the book by Kalankatua-
tsi, which can have relevance to the topic under discussion, is the heathen god of
thunder and lightning Kuar, worshipped by the Huns. This is the relevant pas-
sage from the “History”: “if flashes of thundering fiery lightning and ethereal fire
struck a man or some material object, they considered him or it to be some sort
of sacrifice to a god K‘u(w)ar” (Dasxuranc‘i 1961: 155-156). There is no consensus
among the commentators on this passage as to the etymology of the name
Ku(w)ar. P. Golden (2007: 131, fn 38) suggests that it is either a corruption of Ira-
nian xwar ‘sun’, or Turkic *köğer < kök ‘sky’, er ‘man’.
A striking isogloss to this Hun thunder-god is the enigmatic phrase Temǝr-
q’⁰ara used in the Abkhaz ritual song “Atlar čʲopa”45 mentioned above: “Atlar
čʲopa Temǝr-q’⁰ara”. The first part of the compound is obviously Turkic temir
‘iron’, to yield ‘the iron Quara”, the name of the metal being probably used as epi-
thet. It is quite tempting to connect Hun Ku(w)ar with Abkhaz Quara, as it is
being used in the ritual song where the term čʲopa is also used, which parallels
Hun čopay. The 19th century author in the newspaper “Kavkaz” (1873, no. 150)
gives valuable information about the meaning of this now forgotten word: “The
word Temurgvara [Temǝr-q’⁰ara - V.Ch.] is the name of the deity imagined as a
venerable white-haired old man, riding a winged horse, which by the thumping
of its powerful hooves produces thunder, while the bare curved sword flashes
with lightning. The Abkhaz thunderer is constantly after the evil spirit, unceas-
ingly flying from one object to another in the guise of a fly, and if it happens to

45
The forms of this term differ in various isolects as to the absence or presence of glottaliza-
tion, as well as some other phonetic features: čʲop’a, čʲopa, čʲowpa, at’lar-čʲowpa, at’lar-čʲopa, atlar-
čʲopa, ot’lar-čʲop’a, etlar čʲowpar; the -r in čʲowpar is probably due to the alliteration with at’lar. The
part at’lar//atlar is plausibly explained by V. Abaev (1958: 316) as Ossetic aldar ‘lord, prince’ (cf. the
phrase aldary coppaj ‘the Lord Coppaj’).
188 Viacheslav A. Chirikba

sit on a man, then the curved sword of Temurgvara will not spare him too,
though involuntarily. The one struck by lightning, as a victim of the deity, if he
dies, is regarded as being blessed; otherwise he is brought to consciousness in
the following way”. And then follows the ritual described above (Aguazhba,
Achugba 2005: 74-75).
Though the lexical counterpart of Hun Kuar, Abkhaz Temǝr Q’⁰ara does not
seem to be attested in Ossetic, it can tentatively be seen in the Karachay deity of
thunder Qauar-xan//Kürüju//Krüu//Kürüu-xan (xan ‘lord’), whose name is used
in the ritual song addressed to the deity of wind Gorij.46 The deity is asked by the
power of Choppa not to make thunder, to stop lightning from flashing, and to
drive away the Old Woman-Drought (Karaketov 1995: 96). In Dagestan the Lak
supreme god Kuara (Seferbekov 2009: 34) can also have the same origin. N.
Dzhidalaev (1970) suggested the connection of Lak Kuara with the Georgian de-
ity of harvest K’viria and the Urartian deity Kvera. Yet, the comparison of all
these theonyms is so far only hypothetical.
Interestingly, my mother (born in 1939 in the village of Zwandrypsh of the
Gudauta region of Abkhazia) remembers the song performed during the ritual
around a lightning-struck person with somewhat different words: ót’lar čʲóp’a,
ssir t’ap’ánčʲa ‘Otlar47 Chopa, the wonderful gun”48. She indicated that the gun
could be a metaphor for the lightning. I so far was unable to find these same
words in any folklore texts at my disposal, whether published or unpublished,
but they may well represent the remains of the original song in which (the god ?)
Q’uara also figured. For a semantic parallel to the metaphoric depiction of light-
ning as a weapon compare the Abkhaz expression Šʲaš⁰ə r-x⁰əmp’al ‘the arrow of
[smithy-god] Shashwy’, which, according to Ardzinba (1988: 277), implies a
lightning. Cp. also Nogay jasın ok (ok ‘arrow’) ‘lightning as a fiery arrow of the
celestial deity’, ajındırık < *ajındır ok, interpreted by Jarlykapov (1998: 45) as ‘ar-
row of Indra’.

46
Another hypothesis is that the part Temǝr in Temǝr Q’⁰ara could be an Abkhaz mispercep-
tion of Karachay/Turkic tejri/tengri ‘god’, i.e. ‘God Quara’ (?).
47
The form with the initial o (ót’lar) is probably due to the contraction of the vocative interjec-
tion wa/o with at’lar, i.e. wa/o, at’lar.
48
The preposing of the adjective in the phrase ssir t’ap’ánčʲa is not typical for normal Abkhaz
speech (as noted to me also by A. Ankvab; postposed order is the norm: a-t’ap’ánčʲa ssir), but this
may reflect either an (alternative) older word-order, or a deviation from normal speech permitted
in the style of magic folk-poetry.
 Heathen Heritage in the Caucasus 189

CONCLUSION
As we have seen from the (of necessity) limited material presented above, all
Caucasian peoples have preserved elements of pre-monotheistic religious prac-
tices. The extent of the preservation of the pre-monotheistic systems varies from
strong (Abkhazians, Ossetians, less so Circassians, Svans, Megrelians and Moun-
tainous Georgians groups) to moderate (Chechens, Dagestanis) or minimal (Aze-
ris, Armenians). The attitude of modern “official” religious communities to this
pagan heritage is also very different. Whereas many Abkhazians, Circassians and
Ossetians are proud of their ancient traditions and take measures to maintain
and sometimes (as in Abkhazia and Ossetia) even revive some of them, in other
Caucasian communities there prevails a stronger adherence to monotheistic re-
ligions and in general a negative (or at best indifferent) attitude to the heathen
past.
Despite differences in the extent of the preservation and of the attitudes, one
can say that elements of pre-monotheistic heritage do indeed form an intimate
part of the identity of many Caucasian communities, partially intertwining with
the official religions and with the native ethical systems, like Apswara (i.e. the
Abkhazianness) in Abkhazia, or Adǝɣ̂ e x̂ abze//Adǝɣ̂ aɣe (‘the Circasian Law’//
‘Circassianness’) in Circassia. As the Jordanian Circassian author Amjad Jaimou-
kha (2001: 137) writes in his excellent book “The Circassians”: “From the cradle to
the grave, the Circassian native creed—intertwined with the code of conduct
Adige Xabze—dictated the way an individual behaved, formed their system of
values and certainly influenced the way they conceived the world. Religion and
customs and traditions were the dual formers of the Circassian outlook on life
and they meshed perfectly together. Rejecting one of these intimately associated
components would have entailed forsaking the other and ultimately compro-
mising the essence of Circassianism”.
Although certain elements of the traditional systems are still maintained by
some Caucasian communities, the post-Soviet period is witnessing a powerful
revival of interest in monotheistic religions, which seek to fill the vacuum cre-
ated by the collapse of the Communist ideology. In the condition of the multi-
ethnic and multilingual Caucasus, Christianity—as in Armenia and in Georgia,
and especially Islam—as in the Northern Caucasus and Dagestan, are perceived
as unifying and mobilizing factors, and in the majority of cases they leave no
place for traditional beliefs, which will undoubtedly continue to fade and slowly
vanish in the shadow of the mainstream religions. The awareness of this factor,
190 Viacheslav A. Chirikba

of course, necessitates a better study and better understanding of this ancient

cultural lore.

BIBLOGRAPHY
Abaev, V. I. (1949), Osetinskij yazyk i fol'klor. Moskva-Leningrad.
―― (1958), Istoriko-etimologičeskij slovar' osetinskogo yazyka, Tom I. A – K’, Moskva-Leningrad.
―― (1979), Istoriko-etimologičeskij slovar' osetinskogo yazyka ,Tom III. S – T’, Moskva-Leningrad.
―― (1990), “Kul’t semi bogov u skifov”, Abaev, V. I. Izbrannye trudy. Religiya, fol’klor, literatura,
Vladikavkaz: 89-96.
Adzhindzhal, I. A. (1969), Iz etnografii Abxazii. Materialy i issledovaniya, Sukhum.
Aguazhba, R. X. / T. A. Achugba (eds.) (2005), Abxazija i abxazy v rossijskoj periodike (XIX – nač. XX
v.). Kniga I, Sukhum.
Akaba, K. X. (1984), Istoricheskie korni arxaicheskix ritualov abxazov, Sukhum.
Ardzinba, V. G. (1988), “K istorii kul’ta železa i kuznečnogo remesla (počitanie kuznicy u abxazov)”,
Drevnij Vostok: etnokul’turnye svyazi, Moskva: 263-306.
Asatrian, G. S. / Arakelova V. (2014), The Religion of the Peacock Angel: The Yezidis and their Spirit
World, Durham.
Bardavelidze, V. V. (1957), Drevnejshie religioznye verovaniya i obryadovoe grafičeskoe iskusstvo gru-
zinskix plemen, Tbilisi.
Bgazhba, Kh. S. (1964), Bzybskij dialekt abxazskogo yazyka (Issledovanie i teksty), Tbilisi.
Chartolani, Sh. (1989), K istorii nagorja Zapadnoj Gruzii doklassovoj epoxi, Tbilisi.
Chibirov, L. A. (2008), Tradicionnaya duxovnaya kul’tura osetin, Moskva.
Chirikba, V. A. (2006), “Abkhaz Loans in Megrelian”, Iran and the Caucasus 10.1: 25-76.
―― (2008), “The Problem of the Caucasian Sprachbund”, ed. Pieter Muysken, From Linguistic Ar-
eas to Areal Linguistics. – Studies in Language Companion Series (SLCS). Volume 90: 25-93.
Chursin, G. F. (1957), Materialy po etnografii Abxazii, Sukhum.
Dasxuranc‘i, M. (1961), The history of the Caucasian Albanians, transl. by C. J. F. Dowsett, London.
Dirr, A. (1915), “Božestva oxoty i oxotnichij yazyk u kavkazcev”, Sbornik materialov dlja opisaniya
mestnostej i plemen Kavkaza 44, section 4: 1-16.
Dzhanashia, N. S. (1915), “Iz religioznyx verovanij abxazov”. Xristianskij Vostok IV,1: 72-112.
Dzhidalaev, N. S. (1970), “K voprosu ob istoricheskoj obščnosti kavkazskogo i peredneaziatskogo
mira”, Doklady AN Azerbajdžanskoj SSR XXVI: 3: 83-84.
Gadzhiev, G. А. (1991), Doislamskie verovaniya i obryady narodov Nagornogo Dagestana, Moskva.
Gadzhieva, S. Sh. (1961), Kumyki. Istoriko-etnografičeskoe issledovanie, Moskva.
Gasanov, M. (2001), “On Christianity in Dagestan”, Iran and the Caucasus 5: 79-84.
Golden, P. B. (2007), “The Conversion of the Khazars to the Judaism”, P. B. Golden, H. Ben-Shammai,
A. Róna-Tas. The World of the Khazars. New Perspectives. Selected Papers from the Jerusalem
1999 International Khazar Colloquium, Leiden-Boston: 123-162.
Inal-ipa, Sh. D. (1965), Abkhazy. (Istoriko-etnografičeskie očerki). Vtoroe pererabotannoe, dopolnen-
noe izdanie, Sukhum.
 Heathen Heritage in the Caucasus 191

―― (1976), Voprosy ėtnokul’turnoj istorii abxazov, Sukhum.

Jaimoukha, A. (2001), The Circassians, A Handbook (Peoples of the Caucasus), Richmond.
―― (2005), The Chechens, A Handbook (Peoples of the Caucasus). London, New York.
Jarlykapov, A. A. (1998), “Tradicionnaya xozyajstvennaya kul’tura nogajcev: obryad vyzyvaniya
dozhdya”, Naučnaya mysl’ Kavkaza, № 1: 43-48.
Johansons, A. (1972), “The Shamaness of the Abkhazians”, History of Religions 11.3: 251-256.
Jusupova, Ch. S. (ed.) (1988), Problemy mifologii i verovanij narodov Dagestana. Sbornik statej, Ma-
khachkala.
Kalankatuatsi, M. (1984), Istoriya strany Aluank. Perevod s drevnearmyanskogo, predislovie i kom-
mentarij Sh. V. Smbatjan, Yerevan.
Karaketov, M. D. (1995), Iz tradicionnoj obryadovo-kul’tovoj žizni karačaevcev, Moskva.
Karapetian, G. O. (1979), Armianskij fol’klor. Sostavlenie i perevod G. O. Karapetiana, Moskva.
Khiba, Z. K. (1980), “A contribution to Abkhaz lexicography: the secret language of the hunters”,
Bedi Kartlisa XXXVIII: 269-277.
Mizhaev, M. I. (1973), Mifologicheskaya i obryadovaya poeziya adygov, Cherkessk.
Petrosian, A. (2007), “State pantheon of Greater Armenia: earliest sources” Aramazd: Armenian
Journal of Near Eastern Studies 2: 174-201.
Seferbekov, R. (2001), “On the Demonology of the Tabasaranians. Typology and Description”, Iran
and the Caucasus 5: 139-148.
―― (2009) Panteon yazycheskix božestv narodov Dagestana (tipologiya, xarakteristika, personi-
fikacii), Avtoreferat, Makhachkala.
Shamanov, I. M. (1982), “Drevnetyurkskoe verxovnoe božestvo Tengri (Teyri) v Karačae i Balkarii”,
ed. E. P. Alekseeva, Problemy arxeologii i etnografii Karačaevo-Čerkesii (material’naya i du-
xovnaya kul’tura), Cherkessk: 155-170.
Tokarev, S. A. (ed.) (1991), Mify narodov mira, Enciklopediya, vol. 1 (A – K). Moskva.
Tsabolov, R. L. (2001), Etimologicheskij slovar’ kurdskogo yazyka. Tom I, A – M., Moskva.
Tuite, K. (2004), “Lightning, Sacrifice, and Possession in the Traditional Religions of the Caucasus”,
Anthropos 99.1: 143-159.
―― (2005), “The meaning of Dæl. Symbolic and Spatial Associations of the South Caucasian God-
dess of Game Animals”, Language, Culture and the Individual: A Tribute to Paul Friedrich.
Edited by C. O’Neil, M. Scoggin and K. Tuite. – Lincom Studies in Anthropology 03: 165-188.
Virsaladze, E. B. (1976), Gruzinskij oxotničij mif i poeziya, Moskva.
Zvanbaj, S. T. (1853), “Obryad žertvoprinošeniya sv. Pobedonoscu Georgiyu soveršaemyj ežegodno
abxazami”, Newspaper “Kavkaz”, no. 90, Tiflis.
―― (1855), “Abxazskaya mifologiya i religioznye poverya i obryadnosti meždu žitelyami Abxazii”,
Newspaper “Kavkaz”, no. 81, Tiflis.
―― (1955), Etnografičeskie etyudy, Sukhum.
 Armenian Pre-Christian Divinities:
Some Evidence from the History of Art and
Archaeological Investigation

Matteo Compareti
University of New York, ISAW-NYU

Abstract
The present paper is an attempt to investigate pre-Christian Armenian religious iconography. It is
highly probable that pre-Christian Armenian deities presented some connections with pre-Islamic
Persian divinities. As it is suggested in early Christian Armenian and in Middle Persian literature,
all these divinities concerned Zoroastrian religion. The main idea is that not only some images
could have been re-used by the Christians to represent their holy images (several saints and the
Virgin Mary) but that a common visual language probably had existed on a very wide area from
Anatolia to Central Asia through the Caucasus and Persia. Some other scholars had already sug-
gested a connection between pre-Christian Armenian terracotta statuettes and the production of
those same objects in Central Asia. Some other common characteristics could be found even in
much farer away Iranian lands such as in Khotan.

Keywords
Armenian Art and Archaeology, Zoroastrianism, Terracotta Statuettes, Sasanian Persia, Sogdiana,
Khotan

Armenian literary evidence on the religious situation in the country before

Christianization by Gregory the Illuminator at the beginning of the 4th century
CE is, fortunately, considerable. Scholars have written quite extensively on the
subject. A great number of these literary sources and oral traditions were col-
lected by J. Russell.1 In several cases, information on ancient Armenian religion
can be found in the works of Christian polemists (mainly Armenian and Byzan-
tine) who were openly hostile to everything which was not Christian or that had
been imported from the land of the fiercest enemies of the Romano-Byzantines,

1
Russell, 1987; Russell, 1990. See also: Garsoyan, 1982. For a diverging opinion on the pre-Chris-
tian Armenian pantheon, see Petrosyan, 2007.
 Koninklijke Brill NV, Leiden, 2015
194 Matteo Compareti

that is to say, the Persians.2 During the immediate pre-Islamic period, the state of
conflict between the Romans and the Persians was near constant and competi-
tion for control over Armenia, together with the whole Caucasian region, was
one of the reasons of that conflict.
The Achaemenids (539-323 BCE), the Parthian Arsacids (c. 250 BCE-226 CE)
and the Sasanians (224-651) conquered Armenia or parts of it at different times.3
The local dynasties which ruled Armenia during the period of the Arsacids and
the Sasanians were a branch of the former ruling Parthian family of Persia. Pre-
Christian Armenian religion reflected this situation. For example, the local cal-
endar in use until 1084 CE employed names for the months that were rooted in
the pre-Christian local cults and Iranian religion.4 Both the Persians and the
Parthians presented very strong cultural similarities and were both committed
to the so-called Zoroastrian (or Mazdean) religion. In fact, the names of four di-
vinities mentioned in the calendar months were definitely rooted in Persian and
Parthian Zoroastrianism.
Armenian written sources also recorded further information about the fate of
pre-Christian local temples and their decoration. They were systematically de-
stroyed or converted into churches. Gregory the Illuminator himself was a great
persecutor of paganism.5 This is probably one of the reasons why very few pre-
Christian monuments survived in Armenia. Among the most well-known is the
temple at Garni (1st century CE) which represents an emblematic specimen of
local paganism despite its clear Hellenistic forms and extensive use of the Greek
language in inscriptions found at the site.6 However, in this paper, (which is
dedicated to Garnik Asatryan whose research into the Yazidi religion has re-
vealed more than one point of coincidence with Zoroastrianism), only small ob-
jects of art that have been proposed to be representations of pre-Christian Ar-
menian divinities (yazatas) will be mentioned.7

2
Der Nersessian, 1944-1945.
3
On this early period of Armenian history, see: Hewsen, 2001: 29-91.
4
Panaino, 1990: 664.
5
Prudhomme, 1863.
6
Kanetsian, 1998: 26-29, 37-40, 43-44, 58-59, 74. See also: Russell, 1987: 268-270, 523 with literary
evidence about other temples that did not survive. Another interesting building that has been con-
sidered to be a Zoroastrian temple can be observed among the ruins of the ancient Armenian capi-
tal of Ani, now just beyond the border in Turkey: Thierry, 1997.
7
Non-scientifically excavated stone sculptures inscribed in Aramaic possibly alluding to Maz-
dean divinities have been considered by Russell, 1987: pls. II-III; id. 1990: pls. I-IV; Lemaire, 2003.
 Armenian Pre-Christian Divinities 195

Not many archeological sites in the territory of the modern Republic of Ar-
menia are able to offer important insights into the religious situation of the pre-
Christian (and post-Urartean) eras. The most important site is that of Artaxata,
the capital of Hellenistic Armenia, which has been the object of excavations for
many years. The other important site is Armavir, in the Ararat Valley.8
Fragmentary terracotta statuettes are particularly interesting for the present
study since some of them show clear Hellenistic features that can be associated
with Greek divinities. This is the case with the Aphrodite, Eros and Silenus statu-
ettes from Artaxata. Heracles too should be added to the group even though his
image is common only on coins. Some classical Armenian writers referred to
their own ancient gods using both local and Greek names.9 It is well established
that classical writers (both Greek and Roman) associated foreign divinities with
their own pantheon according to coincidences of divine attributes (Interpretatio
Graeca). Such an association appeared explicitly when the statues of gods at
Nimrud Daği, in ancient Commagene, were discovered. In fact, the local divini-
ties were accompained by Greek inscriptions referring to gods called Zeus-Oro-
masdes, Artagnes-Heracles-Ares, Apollo-Mithras-Helios-Hermes.10 Iranians and
Iranic peoples were often very receptive when it came to Hellenistic art and cul-
ture. Some very interesting Sogdian mural paintings and sculptures in Penjikent
(Tajikistan) also allow us to propose that Indian religious iconography could be
adopted and adapted to local divinities (belonging to a variant of Zoroastrian-
ism) in certain circumstances.11
However, it is extremely difficult to persuasively propose that there could be
any identification rooted in Armenian culture and religion for the three statu-
ettes from Artaxata, because their attributes and poses seem to indicate that
they represent the original Greek divinities themselves. It is not even clear if they
were the objects of veneration for a local cult. In any case, the iconography of

8
Kanetsian, 1998.
9
Russell, 1990: 2682-2688; De Jong, 1997: 251-316; Khachatrian, 1998: 153. The statues of Aphro-
dite are considered to be representations of the local goddess Astłik (of Semitic origins), an identi-
fication supported by Armenian written sources. However, Anahita too was associated with Aph-
rodite in some written sources, specifically Herodotus: Chaumont, 1985: 1006. According to Arme-
nian written sources, Astłik was worshipped in the form of a golden statue that was called “the
golden mother” in the city of Axtixat: Chaumont, 1985: 1007. On the association of Armenian divini-
ties with Greek ones in Movses K’orenaci book II, chap. 8 and 12: trans. Mahé, 1993: 351, n. 10.
10
Colledge, 1986: 28; Boyce, Grenet, 1991: 315-346.
11
Compareti, 2009; Grenet, 2010.
196 Matteo Compareti

Greek divinities was very well-known in ancient Armenia, as the numismatic

finds at several Armenian archaeological sites testify. Scholars propose an asso-
ciation between Heracles and Verethragna, the Zoroastrian god of victory, who
the Armenians called Vahagn mainly on the basis of the inscribed statues of
Commagene.12 An inscribed bronze statue of Heracles was found during the Ital-
ian excavations at Seleucia-Ctesiphon. His name was inscribed in both Greek
and Parthian on his legs. In Greek he was definitely called Heracles while in the
Parthian language it appears he was called Verethragna.13
Traditionally, the terracotta statuettes of a horse rider and a standing figure
in Iranian dress have been identified with the Zoroastrian god Mithra (Armenian
Mihr). Representations of a woman sitting on a throne under an arch, possibly
suckling a naked child, have been proposed to be a local version of Anahita (fig.
1).14 Almost every female statuette that has been found in Armenia has been
identified with Anahita. This is also the case with the famous 2nd-1st century
BCE bronze head from Şahdağ (central Turkey) which is now kept in the British
Museum;15 one gilded bronze medallion from a royal tomb in Sisian; and, a bone
statuette of a woman holding a cup with both her hands from Artaxata (both
now preserved in the Museum of History of Armenia, Yerevan).16 There are few
elements to indicate that a religious identification for these statuettes is justified,
and they often appear to have been given arbitrary attributions by scholars.

12
Frey-Brönnimann, 1997; Khachatrian, 1998: 149, 154.
13
Invernizzi, 1989.
14
Russell, 1990: 2682; Au pied du Mont Ararat, 2007: fig. 154; Compareti, 2011: fig. 5. The arch
used as a frame for the woman in the terracotta statuette from Armavir would be an indication
that it is a holy image: Russell, 2001: 192, note 16.
15
Thierry, Donabédian, 1989: fig. 20. It is not clear why scholars proposed Anahita and not, for
example, Astłik since the bronze statue could be a local or Semitic version of Aphrodite too.
16
Khachatrian, 1998: 146, fig. 88a-b; Au pied du Mont Ararat, 2007: figs. 135, 159. A 1st century
BCE-1st century CE seal impression from Artaxata, representing a standing person holding a disc in
front of a sitting man (a king?) has been identified with Anahita: Khachatrian, 1998: fig. 66.4. The
seal impression does not seem to have any inscription and it is not even clear if the figure holding
the disc is a man or a woman. The disc could be considered an attribute that, at least in late Sasa-
nian art (such as at Taq-i Bustan), is conferred by a god on the king as a token of approbation:
Compareti, 2006. Other seal impressions from the same site depict more common Hellenistic di-
vinities such as Heracles and the winged victory (Nike): Khachatrian, 1998: fig. 66.1-2. Seal impres-
sions have been found in great number at Dvin, the Armenian capital during the Sasanian occupa-
tion period. Those seal impressions, however, depict characteristic Sasanian elements or, in case of
clear religious symbols, only crosses: Kalantarjan, 1982: pls. II.16, III.21-22, V.39, VI.47, 50, XI.107.
 Armenian Pre-Christian Divinities 197

However, if the interpretation of this woman as a goddess is correct, it is not im-

possible that she may have been a local divinity whose features had been assim-
ilated to the Zoroastrian Anahita and, possibly, other Mesopotamian and Greek
divinities. Representations of Isis lactans may even have contributed to the crea-
tion of this statuette which also calls to mind certain Hellenized Egyptian ele-
ments.

Figure 1. Terracotta statuette, 1st-2nd century CE from Armavir.

Museum of History of Armenia, Erevan. After: Au pied du Mont Ararat, 2007: fig. 154.
Is there any evidence to support the proposition that a type of iconography
depicting Anahita accompanied by a child existed? In fact, something more can
be added about Anahita, who was one of the most venerated yazatas in ancient
Armenia. In Mesopotamian culture (and also in Hellenized Egyptian one)
similar sorts of representations existed. The image of Aphrodite and Eros
(winged or not) was also very common in Greek art.17 Evidence from the Iranian
milieu may be more relevant for the present paper. One 7th century Christian
Armenian text is known as the Apology of the Images. Its author, while mocking
the Zoroastrian divinities which stood in many temples in Armenia, explicitly
mentions Anahita, as opposed to the Virgin Mary, “having Jesus Christ sitting on
her knees”.18 Such a parallel could suggest that Anahita too was originally repre-
sented together with a child. However, the few representations of a goddess who
could be identified with Anahita that have been found on Persian soil are all

17
Russell, 1990: 2682; Mahé, 2008. On a cautious approach to such an identification in Coptic
Egypt: Higgins, 2012.
18
Der Nersessian, 1944-1945: 64.
198 Matteo Compareti

concentrated at the late Sasanian site of Taq-i Bustan and no child appears to-
gether with her.19 Terracotta statuettes from Susa representing a mutilated
woman suckling a child are now preserved in the Archaeological Museum in
Tehran (fig. 2).20 Unfortunately, the statuette presented here is at display without
great information about its chronology and the excavation where it has been
found. It is just one piece of evidence to ascertain that in proper Persia too such
a subject was known.

Figure 2. Terracotta statuette from Susa, date unknown.

Archaeological Museum, Tehran. Photo: Ali Majdfar (courtesy Greg Watson).
Further evidence can be found in the production of early terracotta figurines
from the Iranian territories of Central Asia. Archaeological investigations during
the Soviet period in the territory corresponding to historical Khorasmia (specifi-
cally at Koy Krilgan-kala, fig. 3)21 and Sogdiana (at Afrasyab, fig. 4)22 revealed pro-
lific production of terracotta statuettes possibly destined for votive use at least
since the 1st-2nd century CE. For many of those statuettes, Hellenistic borrowing
appears very clear and it is possible that it was the concrete presence of Greeks
in Central Asia which gave impulse to the local tradition of votive offerings in

19
Teixidor, 1981; Bier, 1986; Compareti, 2006: 170-171; Compareti, 2011; Overlaet, 2013. At Taq-i
Bustan, the attributes of Anahita are the jar of water and the stars which embellish her garments.
At Naqsh-i Rustam, in the Sasanian stone relief reproducing Narseh facing a woman (identified
with Anahita or a queen: Overlaet, 2013: 314-315; Shenkar, 2013) presents also a child in front of her.
It is not clear who this child is: Soudavar, 2003: 72-77.
20
I wish to thank Greg Watson who called my attention on the terracotta statuettes under
exam and revised my English for the present paper (and many others).
21
Tolstov, 1958: fig. 164.
22
Meškeris, 1962: fig. 4.3; Meškeris, 1989: fig. 28.1.
 Armenian Pre-Christian Divinities 199

temples and at shrines in the form of terracotta (and, sometimes, metal) statu-
ettes that often featured Hellenistic traits.23 The image of a woman together with
a child has been found at both sites. It does not seem to have been a very widely
distributed motif since not many of these pieces have been found at the two sites
mentioned, and it is not even clear if it should be considered the representation
of a goddess or a simple tableau of everyday life. In any case, it is worth observ-
ing that the motif was known in Central Asia and it would furthermore be inter-
esting to investigate the situation in more oriental Iranian lands.

Figure 3. Terracotta statuette from Koy Krilgan-kala, 1st-2nd century CE,

location unknown. After: Tolstov, 1958: fig. 165.
Khotan was an Iranic kingdom in modern Xinjiang Uighur Autonomous
Province (Western China), which adopted Buddhism and disappeared at the be-
ginning of the 11th century with the Turcization and Islamization of Central Asia.
Several 8th century wooden votive tablets representing typically Indian (or, less
correctly, Hindu) divinities converted to Buddhism were found by Sir Aurel Stein
at the beginning of the last century and are now preserved at the National Mu-
seum of New Delhi and in the British Museum. Recent studies of some Buddhist
votive tablets strongly suggest that these images of divinities converted to Bud-
dhism actually represent local gods, most likely Iranian ones, with iconography
rooted in Indian art.24 The problem for the identification of such divinities is that
it was only their iconography that was predominantly Indian. As already ob-
served, for some reason, the Sogdians had started to adopt (and adapt) the ico-

23
Offerings to temples in the shape of terracotta anthopomorphic statuettes were very com-
mon in ancient Greco-Roman religion exactly as they can be observed in contemporary catholic
churches (the so-called ex voto). See, for example, the very ancient sanctuary of Hera not far from
Paestum known as the “Heraion of the Sele River”: Cipriani, 1997.
24
According to M. Mode, those wooden tablet should be attributed to Sogdian immigrants
while for B, Marshak those are representations of local divinities: Mode, 1991/92.
200 Matteo Compareti

nography of Indian divinities to represent their own local (Zoroastrian) gods. For
this reason it is not always clear if the representation of foreign gods in a Bud-
dhist sphere should be considered as belonging to an Iranian or Indian cultural
context. The question is even more problematic in Khotan, which was actually
within Iranian cultural frontiers where many Sogdian immigrants lived. Unfor-
tunately, there are no explicit traces of pre-Buddhist cults in Khotan since the
Khotanese people accepted that Indian religion early and many Sanskrit terms
into their own language. Their alphabet too was borrowed from the sub-conti-
nent.25

Figure 4. Terracotta statuette from Afrasyab, 1st-2nd century CE,

Samarkand Museum. After: Meškeris, 1989: fig. 28.1.
Approximately ten years ago, Swiss, Japanese and Chinese archaeologists
found some Buddhist mural paintings in the Khotan area which share more than
one point in common with the divinities on the wooden tablets recovered by
Stein. Among the representations of female divinities, is one with four hands
who is holding the sun and the moon and a second, who is always represented
with one or more children (fig. 5).26 While the goddess holding the sun and the

25
On ancient Khotanese culture, see the proceedings of the symposium “The Kingdom of Kho-
tan to AD 1000: A Meeting of Cultures” held on the 10th and 11th May 2004 at the British Library
and published in a special issue of: Journal of Inner Asian Art and Archaeology, ed. U. Sims-Wil-
liams, 3, 2008: 61-186.
26
C. Lo Muzio thinks that the divinities in the Khotanese paintings should be interpreted ac-
cording to Indian art and culture: Lo Muzio, 2006. However, some newly found Buddhist paintings
that show parades of horse riders and are accompanied by an inscription undoubtedly refer to
 Armenian Pre-Christian Divinities 201

moon could be associated with the Mesopotamian Nana (who was very well-
known among Parthians, Bactrians and Sogdians), the one with children may be
identified with the female ogre Hariti whose iconography could have, in Khotan,
been superimposed on that of an important pre-existing goddess who could, in
her turn, offer a parallel with the terracotta statuette from Armavir.

Figure 5. Detail of the mural painting on the western wall of Temple D13 from
Dandan Oylik, 8th century CE. After: Lo Muzio, 2006: fig. 3.
Of course, this is a theoretical identification which still leaves many gaps. For
example, not every scholar considers the Iranian reading of the Khotanese
wooden tablets and mural paintings to be persuasive. This paper simply aims to
call attention to a motif—the figure of a woman with child—that was quite
widespread across the ancient Near East and into Central Asian lands and which,
possibly, should be considered to be the representation of an important and
much venerated goddess. The main suggestion offered by this paper is that the
woman with child from Armavir could be a divine image which might offer an
Armenian parallel for the Khotanese female divinity with children. As it is well-
known, Buddhism had no problem in accepting divinities belonging to other re-
ligions and it is possible that something similar may have occurred in a (usually
less tolerant) Christian milieu. The acceptance of this kind of image could have

“eight gods”: Xinjiang Institute of Cultural Relics and Archaeology, The Academic Research Or-
ganization for the Niya Ruins of Bukkyo University, 2009: pls. 39-40. These are probably genuine
Khotanese deities (that is to say, Iranian ones) later accepted into the Buddhist system.
202 Matteo Compareti

been facilitated by some iconographic correspondences of this goddess with the

image of the Virgin Mary and child which made the later easier to accept for the
Christianizing people of Armenia. As observed, this might be what is being sug-
gested in the Armenian text Apology of the Images.

BIBLIOGRAPHY
Au pied du Mont Ararat. Splendeurs de l’Arménie antique (2007), Catalogue of the Exhibition, Arles.
Bier, C. A. (1986), “Anāhīd.iv. Anāhītā in the Arts”, Encyclopædia Iranica, vol. I, ed. E. Yarshater,
London, Boston, Henley: 1009-1011.
Boyce, M. / F. Grenet (1991), A History of Zoroastrianism. Volume Three. Zoroastrianism under Mac-
edonian and Roman Rule, Leiden, New York, Københaven, Köln.
Chaumont, M. L. (1985), “Anāhīd.iii. The Cult and Its Diffusion”, Encyclopædia Iranica, vol. I, ed. E.
Yarshater, London, Boston, Henley: 1006-1009.
Cipriani, M. (1997), “Il ruolo di Hera nel santuario meridionale di Poseidonia”, Héra. Images, espa-
ces, cultes, ed. J. de La Genière, Naples: 211-225.
Colledge, M. A. R. (1986), Iconography of Religions. Section XIV: Iran. The Parthian Period, Leiden.
Compareti, M. (2006), “Iconographical Notes on Some Recent Studies on Sasanian Religious Art
(With an Additional Note on an Ilkhanid Monument by Rudy Favaro)”, Annali di Ca’ Fos-
cari, XLV, 3: 163-200.
―― (2009), “The Indian Iconography of the Sogdian Divinities and the Role of Buddhism and
Hinduism in Its Transmission”, Annali dell’Istituto Orientale di Napoli, 69/1-4: 175-210.
―― (2011) “The Representation of Anāhitā in Sasanian Art: The Case of Taq-i Bustan Rock Reliefs
and Figurative Capitals”, Anahita Persian Goddess and Zoroastrian Yazata. A Special Issue
of Mithras Reader: An Academic and Religious Journal of Greek, Roman and Persian Studies,
ed. Payam Nabarz: 28-43.
De Jong, A. (1997), Traditions of the Magi. Zoroastrianism in Greek and Latin Literature, Leiden, New
York, Köln.
Frey-Brönnimann, J. (1997), “Verethragna”, Lexicon Iconographicum Mythologiae Classicae, VIII. 2,
Zürich, Düsseldorf: 233-235.
Garsoyan, N. G. (1982), “The Iranian Substratum of the «Agat’angełos» Cycle”, East of Byzantium:
Syria and Armenia in the Formative Period, eds. N. G. Garsoîan, Th. F. Mathews, R. W.
Thomson, Washington: 151-189.
Grenet, F. (2010), “Iranian Gods in Hindu Garb: The Zoroastrian Pantheon of the Bactrians and
Sogdians, Second-Eighth Centuries”, Bulletin of the Asia institute, 20: 87-99.
Hewsen, R. H. (2001), Armenia. A Historical Atlas, Chicago, London.
Higgins, S. (2012), “Divine Mothers: The Influence of Isis on the Virgin Mary in Egyptian Lactans-
Iconography, Journal of the Canadian Society for Coptic Studies, 3-4: 71-90.
Invernizzi, A. (1989), “Héracles a Séleucie du Tigre”, Rev. Arch., 1: 65-113.
Kalantarjan, A. A. (1982), Rannesrednevekovye bully Dvina, Arheologičeskie pamjatniki Armenii, 13,
Srednevekovye pamjatniki, V, Erevan.
 Armenian Pre-Christian Divinities 203

Kanetsian, A. G. (1998), “Città e insediamenti nell’Armenia di età classica”, Ai piedi dell’Ararart. Ar-
taxata e l’Armenia ellenistico-romana, ed. A. Invernizzi, Firenze: 3-94.
Khachatrian, Ž. D. (1998), “Artaxata capitale dell’Armenia antica (II sec. a.C.-IV d.C.)”, Ai piedi dell’
Ararart. Artaxata e l’Armenia ellenistico-romana, ed. A. Invernizzi, Firenze: 97-158.
Lemaire, A. (2003), “Les pierres et inscriptions araméenes d’Arebsun, nouvel examen”, Irano-Ju-
daica V. Studies Relating to Jewish Contacts with Persian Culture throughout the Ages, ed. Sh.
Shaked, A, Netzer, Jerusalem : 138-164.
Lo Muzio, C. (2006), “Culti brahmanici a Khotan: note sulle pitture del Tempio D13 a Dandan Oi-
liq”, Rivista degli Studi Orientali, 79, 1-4: 185-201.
Mahé, J.-P. (1993), Histoire de l’Armenie par Moïse de Khorène, Paris.
―― (2008), “Le mythe d’Ištar dans l’oralité caucasienne”, Comptes rendus de l’Académie des In-
scriptions et Belles-Lettres, I: 215-230.
Meškeris, V. A. (1989), Sogdijskaja terrakota, Dušanbe.
―― (1962), Terrakoty Samarkandskogo muzeja, Leningrad.
Mode, M. (1991/92), “Sogdian Gods in Exile. Some Iconographic Evidence from Khotan in the Light
of Recent Excavated Material from Sogdiana”, Silk Road Art and Archaeology, 2: 179-214.
der Nersessian, S. (1944-1945), “Une apologie des images du septième siècle”, Byzantion, XVII: 58-87.
Overlaet, B. (2013), “And Man Created God? Kings, Priests and Gods on Sasanian Investiture Re-
liefs”, Iranica Antiqua, XLVIII: 313-354.
Panaino, A. (1990), “Calendars. i. Pre-Islamic Calendars”, Encyclopædia Iranica, vol. IV, ed. E. Yar-
shater, London, New York: 658-668.
Petrosyan, A. (2007), “State Pantheon of Greater Armenia: Earliest Sources”, Aramazd, II: 174-201.
Prudhomme, E. (1863), “Histoire de Darôn par Zénob de Klag”, Journal Asiatique, II: 401-475.
Russell, J. (1987), Zoroastrianism in Armenia, Harvard.
―― (1990), “Pre-Christian Armenian Religion”, ANWR, II, 18, ed. W. Haase, Berlin, New York: 2679-
2692.
―― (2001), “The Scepter of Tiridates”, Le Muséon, 114, 1-2, 2001: 187-215.
Shenkar, M. A. (2013), “Boginja ili carica? K interpretacii ženskogo personaža na rel'efe Narse iz
Nakš-e Rustama”, Scripta Antiqua. Edward Rtveladze Felicitation Volume, ed. M. D. Bukha-
rin, Moskva: 614-634.
Soudavar, A. (2003), The Aura of the Kings. Legitimacy and Divine Sanction in Iranian Kingship,
Costa Mesa, CA.
Teixidor, J. (1981), “Anaeitis”, Lexicon Iconographicum Mythologiae Classicae, I.2, Zürich, Düsseldorf:
754-756.
Thierry, J.-M. (1997), “Le monument tétrapode d’Ani”, in: From Byzantium to Iran. Armenian Studies
in Honor of Nina G. Garsoyan, ed. J.-P. Mahé, R. W. Thompson, Atlanta: 425-451.
―― / P. Donabédian (1989), Armenian Art, New York.
Tolstov, S. P. (1958), “Les résultats des travaux de l’expédition archéologique et ethnographique au
Kharezm de l’Académie des Sciences de l’U.R.S.S. en 1951-1955”, Orientalia Romana. Essays
and Lectures, I, Roma: 119-168.
204 Matteo Compareti

Xinjiang Institute of Cultural Relics and Archaeology, The Academic Research Organization for the
Niya Ruins of Bukkyo University (2009), Dandan Oilik Site. Report of the Sino-Japanese Joint
Expedition, Beijing.
 The Taming of the Fairies

Peter Nicolaus
Office of the United Nations High Commissioner for Refugees, Kabul*

Abstract
The first part of this contribution provides an account of a rare initiation ceremony of a Hunza
shaman (daiyal). The initiation commences with the daiyal inhaling the smoke of burning juniper
branches putting him into a trance. This is followed by an enduring dance of the shaman, inter-
rupted only by his conversations with fairies (periye) and violent fits of rage against witches (ruiye).
The dance culminates with the daiyal sucking the blood from the severed head of a sacrificed goat.
Just before he slips into unconsciousness, his guiding shaman rewards him with an iron bangle
(kau). This symbolizes and confirms the bond between him and his periye, established through this
initiation. As long as the shaman wears the kau, the artifact will ward off the fairies, and allow the
shaman to draw on their magical powers.
The second part, based on interviews with 28 shamans (14 female), illustrates the danger that
periye present for potential shamans and the society, and examines how fairies are tamed. It fur-
ther analyzes the balance of power established between the ruler of Hunza (tham) and the sham-
ans representing his subjects. Even after the Islamisation of the Hunza-society, the tham had to
perform ancient religious rituals. Hence, he needed to have close relations with the spiritual world.
However, since he was unable to communicate with the fairies, he had to rely on the shamans,
who acquired a considerable degree of freedom to endorse or criticize decisions of the tham. This
shamanic clout certainly presented a challenge to the ruler and evidenced the need to establish a
social mechanism to hold the shamans at bay. Consequently, the tham assigned the Dom—an un-
derprivileged and ethnically different segment of the population completely dependent on him—
to perform as musicians in all shamanistic performances. In other words, by monopolizing the sa-
cred music, he could successfully counterbalance the shamans.

*
At the time of writing this article the author, served as the Representative of the United Na-
tions High Commissioner for Refugees (UNHCR) in Kabul (Afghanistan) and visited Northern Pa-
kistan several times in 2011 and 2012. In the context of these private journeys the author inter-
viewed 28 shamans, mainly in Lower Hunza, but also in Gilgit district, Nager and Central Hunza.
The author is obliged to his friend and guide, Mr. Muneer Alam, for arranging these expeditions, as
well as for interpreting and assisting with the interviews. The author is further grateful to his
daughter, Lisa Nicolaus, and Serkan Yuce, as well as to his colleagues Ms. Robin Ellis and Ms. Su-
zanne Murray-Jones for critically commenting on the text. The opinions expressed in this article
are those of the author and are not necessarily shared by the United Nations or UNHCR.
 Koninklijke Brill NV, Leiden, 2015
206 Peter Nicolaus

Keywords
Bericho, Burushaski, Dance of the Shaman, Demons, Dom, Fairies, Hunza, Initiation of the Sha-
man, Ismaili, Pakistan, Shamanism, Shia, Shina, Syncretism, Witches

The thermometer shows 24o C and small, light clouds move swiftly over the fir-
mament―it is a warm and pleasant spring day in Khana Abad,1 a small hamlet
in the mountains of Lower Hunza between Hussain Abad and Nasir Abad2 on the
Karakoram Highway in Pakistan. It is the 20th of May 2012 and a major event, a
chatti3, will soon be celebrated. The chatti will be held on a field at the eastern
end of the village, north of the main road. The rectangular compound stretches
from south to north and is enclosed by simple stonewalls. Its green grass gives
way only to a few huge erratic boulders, mainly in the northern part of the area,
whilst close to the south-eastern corner grows a large mulberry tree bestowing
its welcome shade.
Around noon the Dom or Bericho musicians arrive. They sit down under the
spreading branches of this tree and start playing their instruments: Abbas (two
small drums),4 Gul Baig (a bigger double-headed drum),5 as well as Ali Gohar and
Rahmanullah (their flutes).6 Already before the event, they had gathered in a
friend’s house where they had been welcomed with a very particular sheifial7 and

1
About 230 households with some 1,300 inhabitants. However, almost half of the population is
not living in the village, as they temporarily migrated to the urban centres in Punjab and Sind.
2
The location is still also known under its old name, Hindi.
3
Chatti denotes, in the Shina dialect spoken in Lower Hunza (in Burushaski: chato, and in
Shina of the Bagrot Valley: chatoki), a special kind of goat (see below) as well as the shaman cere-
mony, which culminates in the sacrifice of such a goat. In the context of this footnote, it should be
noted that all local terms are expressed in the Shina of the Lower Hunza valley, if not indicated
differently.
While Burushaski is an isolated language (no generally accepted connection has been demon-
strated between Burushaski and any other language or language family), Shina belongs to the
Indo-European family: Indo-European → Indo-Iranian → Indo-Aryan → Northwestern Zone → Dar-
dic → Shina → Shina (Lewis 2009).
4
Damal (Shina and Burushaski), a pair of kettledrums, shaped like melon halves, each about 18
to 20 cm in diameter, played with two drumsticks (Sidky 1994: 80).
5
Dadang (Shina and Burushaski), a two-sided, circular bass drum, nearly ½ m in diameter and
80 cm in length, which is beaten by the left hand on the left side and with a drumstick on the right
side.
6
Totek (Shina and Burushaski), a 30 cm long reed-pipe, is the wind-instrument preferred for
shamanistic rituals, although Sidky (1994: 80) mentions the gabi (Burushaski) or taroi (Shina) as
being used as the wind instrument during the dance of the shaman.
7
Sheifial is the traditional bread-and-butter-welcome in this region. For this very special event,
 The Taming of the Fairies 207

treated with great respect―although Dom musicians are generally considered to

be somewhat ‘unclean’ because of their low-caste origin. This quartet consists of
masters, who have performed in more than 20 chatti ceremonies over the last 35
years.
Meanwhile the first spectators start to appear; mainly married men and
women, as well as children. Then, after a short while, also the young men and
teenage girls begin to enter the grounds. All are wearing their best clothes; this is
certainly the event of the year―the occasion to see and to be seen. A few old
men start dancing while the audience is waiting for the arrival of the shamans or
daiyals.8 By half past one, they finally arrive at the scene where, meanwhile,
some 300 villagers have congregated. Among them are also most of the religious
dignitaries of the hamlet:9 two mukhi10 and one khalifa.11 Apparently, they do not
consider it a major problem to reconcile shamanistic practises with mainstream
Ismaili beliefs.12
The two shamans, dressed in white―even their shoes are white―are Bi Bi
Cha and Bul Bul Jan. The former, a 50-year-old lady from Khana Abad, is a
revered spiritual master; she has taught many young daiyal and is also the
teacher or guiding shaman of Bul Bul. He, a 36-year-old man from Nasir Abad,
is―what the people in the districts of Gilgit and Hunza-Nager call―an ‘imper-

seven year old (quite rancid) butter was served, which had been wrapped in leaves and birch-bark
and buried in the ground until used (see also Müller-Stellrecht 1980: 33).
8
The shaman is called daiyal (pl.: daiyali) or dinyal in Shina, while Burushaski speakers refer to
their shamans as bitan (pl.: bitayo).
9
Most Hunza adhere to the Ismaili religion, while most Nager people are Shia followers.
10
In the Ismaili tariqah, the guardian of each Jamatkhana (public or prayer house, i.e. the Isma-
ili mosque) is called mukhi, a word derived from mukhiy and that means (in Urdu and Hindi) “head
or the uppermost, foremost, or most important part of an inanimate object”. Since the Imam is not
physically present in the Jamatkhana all the time, the mukhi presents him, drawing on the Imam's
authority.
11
In the Hunza Valley the khalifa is seen as the Ismaili equivalent to the Shia akhound in Nager,
Gilgit and Chitral. Akhound is more or less equivalent to the term molla, although the latter is used
in a somewhat pejorative sense, as the akhound deals with petty day-to-day magic and many of
them are renowned for the tumar (amulets) they produce to combat the evil eye and other forms
of black magic (see also Müller-Stellrecht 1979: 221-223).
12
Snoy (1975: 88) observed the same lack of concern and remarked that notwithstanding their
opposition to shamanistic practices, Islamic dignitaries would nevertheless actively participate as
spectators in shamanistic séances. However, the environment for religious minorities and sects in
Pakistan is getting more and more tense (Rashid 2012: 14 and 162), and it is doubtful whether this
religious tolerance will prevail in the years to come.
208 Peter Nicolaus

fect shaman’ and needs to undergo this ritual to be accepted as a true daiyal.
A little red carpet13 is rolled out in front of the musicians, and an assistant
lights a small fire of dried juniper twigs in an aluminium bowl placed next to the
rug. Bul Bul kneels down on the carpet and starts inhaling the smoke. His head is
shaking, the expression on his face changes and his eyes roll back, revealing that
he has entered into a state of trance.
The musicians start playing the yudeni music14 or fairy-tunes. The first is Ghul
Ghool;15 and, indeed, seven periye or fairies16 arrive on a flying zan pan17attracted
by the music. They are Bul Bul’s rachiye18 and can only be seen by him, and by his
guiding shaman. The seven periye of Bi Bi Cha19 have also come in their own zan
pan. They too enjoy the music and the tryst with their sisters. All the fairies settle
comfortable in the branches of the tree, while Bul Bul stands and listens atten-
tively to the music: first, to the flutes and then to the drums―it looks like he is
communicating with his rachiye through these instruments.20 Then he starts
dancing counter clockwise within a circle, some 40 meters in diameter, tightly
enclosed by young men; at first it is a slow gyration that seems to increase expo-
nentially as the beat of the music becomes wild and demanding.

13
About 1 x 2 meters.
14
Yudeni is the music traditionally played during a chatti-ceremony. Some informants indi-
cated that the term would mean: the music played by the fairies through the instrument of the
Dom musicians.
15
Ghul Ghool is considerably slower than the other pieces of music typically and traditionally
performed during a chatti and it is usually played as the fourth tune, but Bi Bi Cha had insisted on
starting with it.
16
Peri (Shina pl.: periye; Burushaski pl.: perting). “The resemblance of the word ‘Peri’ to ‘fairy’ is of
course purely accidental, but there are actually points of similarity between Peris and western fairies”
(Lorimer 1929: 518). See also below.
17
The Hunza equivalent of a Central Asian takhta (Persian takht), a huge carpet covered back-
less divan bed (often located in the gardens), where people sit to talk, eat and drink green tea.
18
The shaman’s rachiye (sing.: rachi) are those periye who form a bond with him/her and who
protect and support the shaman (see below and Lorimer 1929: 522). Snoy (1975: 111) sees the rachiye
as the benevolent opponents of parian (fairies that harm humans).
19
One mother-fairy and six daughters only visible to Bi Bi Cha, who even knows their names,
and that they are from the Phoner (flower) family: Guli Amin (the mother), Zahivani, Chamblesh,
Anglesh, Chashmani, Dam Dodeli, and Zamrud (the daughters).
20
Leitner (1893: 8), as well as Sidky/Subedi (2000: 59) were told that the communication would,
indeed, flow through these instruments. Bul Bul just mentioned after the séances that the fairies
could hear better through his ears, and that he conveyed instructions from his rachiye to the musi-
cians.
 The Taming of the Fairies 209

He spots three local rui and a mitu, sworn enemies of the fairies;21 incited by
his rachiye, he wants to assault them and tries violently to leave the circle. In the
course of the séance he attempts three more times to break out and to attack
these opponents of the periye. However, the strong young men on the edges of
the circle are on alert and each time they push him back into the arena, where
he continues to perform his ecstatic dance.22
Faster and faster he whirls around until an invisible force brings him to an
abrupt halt in front of the musicians. Bul Bul bends over to listen again: first, to
the tune of the flutes and then to the beat of the big drum, the dadang. Mean-
while, Bi Bi Cha is throwing small juniper twigs on his back and mumbling unin-
telligible spells and incantations. She talks to her own fairies, who tell her how to
conjure up Bul Bul’s rachiye and make them agree to form a bond with this
young shaman.23
Through Bul Bul the rachiye dictate the performance by changing the rhythm
and speed or by even introducing another tune. After instructing the musicians,
a leap sends him bounding through the air with legs outstretched, a position
akin to performing the splits. He instructs the musicians; and then, with huge
jumps and legs outstretched, almost as if he is about to perform the splits in the
air, he spins away―counter clockwise, faster and faster. Five times he interrupts
his dance and changes the music. One after the other, Ghul Ghool, Pasul, Base-
mani or Dani, Lathasi, Saos, and Shireen Zuban are played.24
After completing his ninth round of dancing, Bul Bul halts again in front of
the musicians. However, this time he has not stopped to listen, but to recite. He
sits down on the bass drum, the music halts and the audience falls completely

21
Rui (pl.: ruiye) are dangerous and allegedly cannibalistic women, while the mitu is the male
assistant of these malicious witches. Although they are usually well-known to their neighbours,
this does not trigger hatred or even witch-hunts (see in this context also Lorimer 1929: 527-534;
Jettmar 1975: 272-276; Snoy 1975: 212).
22
Jettmar (1975: 278) describes a similar scene where during a shamanic dance in Danyor (Gil-
git district) in the year 1958 “der Daiyal den Kreis der Zuschauer zu durchbrechen [sucht, um] sich
auf ein Weib zu stürzen, das er als unrein erkennt”.
23
“I pray on the juniper to bring ferries closer to the new shaman and from time to time I throw
them on Bul Bul. My rachiye can help to bring the periye of the new shaman closer to him. I cannot tell
you the words—these prayers are between the daiyal and his rachiye; and no other person should
hear them” (interview with Bi Bi Cha, 21st May 2012).
24
They traditionally form, together with the first piece (Ghul Ghool), the musical framework for
chatti-séances.
210 Peter Nicolaus

silent. Then, in a monotonous and emotionless voice, whispering more than

singing, the daiyal begins:
Maga Harige, Maga Harige Bilchhar Dobani Chees Daro
(They took me, they took me, to the mountain of Bilchhar Dobani25)
Maga Harige, Maga Harige Deengi Jut Daro
(They took me, they took me, over the Deengi Jut26)
Maga Harige, Maga Harige Majini Chiliye Zcholi Daro
(They took me, they took me, through the branches of the Majini juniper27)
Maga Harige, Maga Harige Kili Harai Daro
(They took me, they took me, to the pasture of the female ibex)
Kilyi Sono Cheengaro Dodh Atege
(They put milk into the golden horn of the female ibex)
Maga Harige Maga Harige Bomi Harai Daro
(They took me, they took me, to the pasture of the male ibex)
Bomayo Sono Cheengaro Lail Atege
(They brought blood in the golden horn of the male ibex)
Lail Ga Dodh Mai Dodi Ga Ajai Se Gati Thigey Ne Badaro Lail Ga Dodh Niratege
(My fairies first mixed the milk and blood, then they separated the milk from the
blood)
Mai Dodi Ga Ajai Se Choshiye Doloija Phat Thege
(My fairies dropped me down on polyester strings28 and they went back)
Mai Sujo Dim Banalido
(My levitating body became unconscious).29
Slowly, very slowly, the shaman gets up and takes a few sluggish steps back
into the circle; then he jumps and is again caught up in his ecstatic dance. He
halts in front of the musicians, listens to their instruments, gives instructions to
change the tune, then whirls away again with an unbelievable speed. It is ex-
hausting, and he and the rachiye need to drink. A young girl, about eight years
old, and dressed in white clothes, enters the circle. She offers water in a poono, a
small calabash gourd decorated with a string of plastic pearls. Bul Bul takes a

25
A peak (6,134 m) in the Karakoram Range where, like on Rakaposhi and Nanga Parbat, the
periye live. Snoy’s informant describes in detail their dwelling on the Dobani with bridges of pure
gold and silver, as well as trees bearing pearls (Snoy 1975: 174).
26
Mythical green meadows.
27
The Russian Juniper (Juniperus semiglobosa) grows in abundance on the Karakoram moun-
tain range, but this one is a mythical tree with strong mythical powers.
28
Nylon or polyester has replaced cotton very recently.
29
I am obliged to Muneer Alam, who provided the translation.
 The Taming of the Fairies 211

deep gulp; then he spills water on the floor so that the fairies also get some of the
refreshing liquid.30
The dance continues, while the music is nearing its climax, and the chatti31 is
sacrificed at the eastside of the field, behind the musicians, about seven metres
away from the mulberry tree. The guiding shaman has left the circle to supervise
three young men; two of which are holding the goat while the third cuts its
throat with a knife. As the blood of the sacrificed goat spills like a fountain, the
one who held the knife collects the blood in a metal bowl. He then presents the
blood to Bi Bi Cha who sets the bowl on the ground and removes a kau,32 a sim-
ple bangle made of twisted iron, from inside her clothes. Chanting inaudible
spells she dips the kau into the blood, before moving back into the circle with
the bangle in her right hand; blood dripping from it.
There is Bul Bul struggling to free himself from a group of young men who are
preventing him from leaving the circle. This time he is not after any evil oppo-
nents of the rachiye―he himself desires blood. He desperately wants to drink
the blood gushing from the chatti’s torso.33 If Bul Bul drinks this blood, he will
turn into a male rui.34 Therefore, even before the ceremony, the guiding shaman
has instructed the young men to prevent him by all means possible from getting
close to the sacrificed animal. The young men use all their strength and try to
contain the daiyal―in vain! He is fighting and struggling with unnatural
force―with the vigour of a man in trance. Finally, when there is no more blood
flowing from the aorta of the goat-torso, Bi Bi Cha allows the men to let him go.
Running to the sacrifice, the daiyal snatches the chatti’s severed head and sucks
the blood dripping from it. For Bul Bul it is not blood, but ibex35 milk—the blood

30
There is no doubt that the offering of the water and the subsequent consumption of it are
integral parts of the séance, as also Snoy (1975: 195) reports on it.
31
A young female goat about one year old (10 to 14 months), which is of a very rare
grey―almost bluish―colour.
32
For the significance of the kau (pl.: kauey) (Burushaski: zumozch, pl.: zumianzch) in the fairy-
cult see below, as well as Snoy 1975: 193 and 200; Jettmar 1975: 279.
33
Jettmar sees the importance of the desire to drink the chatti’s blood, but interprets it too
superficilally: “Ein guter Daiyal stürzt sich auf den blutenden Rumpf und versucht, das hervorspru-
delnde Blut zu trinken―er hält es für Milch” (Jettmar 1975: 277).
34
Snoy (1975: 213) confirms this with a statement of one of his informants, emphasizing that a
man will turn into a male rui in case he drinks blood.
35
The ibex is the favourite domestic animal of the periye (for the ibex cult see Jettmar 1975:
245-250).
212 Peter Nicolaus

vessels of the goat-head are the udder from where the milk flows. Returning once
more to the circle, he dances around holding the goat’s cranium high above his
own head.
Still, the blood seems to increase the daiyal’s aggressivity and once again he
tries to break out of the circle. This time, the men forming the circle appear un-
comfortable having a severed goat-head pushed violently in their faces, and they
move backwards. But at this point Bi Bi Cha takes over. With one swift move-
ment she pulls up Bul Bul’s sleeve and quickly fixes the kau on his left forearm.
The moment the bangle is firmly placed around the shaman’s arm, he collapses
to the ground as if suddenly hit by lightning. The music stops abruptly, and the
men take the unconscious shaman to a mattress, which has been covered with a
white bed sheet. Gently they lay him down on his back with his head pointing
north, whilst others wrap him with white linen. The audience is absolutely silent;
mesmerised.
The guiding shaman is omnipresent, closely supervising everything. Sitting
Bul Bul’s left side, she stares at him and mumbles spells. On her order, a cup
filled with water, is brought and she places Bul Bul’s left hand into the cold liq-
uid. He reacts immediately. With his eyes still closed, he swiftly rolls onto his left
side to place also his right hand into the cooling water; only then he opens his
eyes. The insane expression has vanished; he is once more the joyful Bul Bul and
no longer a daiyal in trance obsessed and possessed by his rachiye.
He washes the blood from his face and returns to the circle. The music starts
again and he, and his guiding shaman, resume the dance. This time he proceeds
gently and with no signs of ecstasy. Soon he is joined by his relatives and then by
almost all the male spectators―even five-year-old boys. With the ceremony
complete the villagers remain to enjoy the festivities for another two hours.36

COMMENTARY AND FURTHER THOUGHTS

1) FAIRIES―THE UNWANTED POWERS
The term peri means ‘fairy’. It is most probably of Persian origin and has replaced
the previously used Shina word, rakshasi. The latter term, however, lives on as
name for those periye, who are particularly linked to humans (Jettmar 1975:

36
The actual ceremony lasted from 13:41 (when the shaman started to dance) to 14:56 (when
the shaman awoke from his trance and completed the séance). Bul Bul completed 32 circuits
within the circle when he was dancing, tried four times to break out of the circle, and stopped 10
times in front of the musicians to listen and/or change the music or to recite his song.
 The Taming of the Fairies 213

321).37 These periye are called rachiye; when interviewed, the shamans preferred
to use the term rachi, when talking about their fairies, rather than the more ge-
neric word peri.
These rachiye are―without exception―female,38 and they appear in groups
up to 44 strong. Hence, the female daiyal in Nasir Abad, Lower Hunza, who is
visited by only one rachi, seems to be a rare exception.39 In almost all cases these
groups consist of a makhakhar40 or mother-rachi and her daughters. However,
there are exceptions: during the interviews five shamans spoke about dealing
with two or three mother-rachiye;41 while the fairies of two other shamans in
Lower Hunza did not seem to be part of any family unit. The average age of a
daiyal’s rachiye was 40 years for the mother-fairies42 and 10 years for their daugh-
ters.43
Daiyali interviewed in Chalt and Chaprot (Nager), as well as one daiyal in
Lower Hunza, explained how their rachiye would appear together with a chalo or
de.u.44 They stated that the chalo protects the rachiye and accompanies them
everywhere.

37
However, compare also Lorimer (1929: 519) who suggests that the term peri could be a genu-
ine Shina word.
38
See, however, Jettmar/Kattner (2003: 96); and Lorimer (1929: 519) who suggests that “the Peris
of the Shina-speaking peoples are of both sexes. The female is called a Peri and the male a Perian”.
39
On average a daiyal has 15 rachiye (based on interviews with 28 shamans). However, the
strength of the groups varies from region to region: in Lower Hunza and the Gilgit district the
numbers of fairies per shaman is quite modest (7 or 9), while in Bagrot and Central Hunza most
shamans have 21 fairies. Only in Nager do shamans have more than 24 fairies (2 shamans in Chap-
rot were visited by 35 and 44 fairies, respectively). The age of the shamans seems irrelevant with
regard to the number of fairies he/she possesses: one of the youngest shamans (28) had 21 fairies,
while the oldest (85) was visited by 24 fairies.
40
The term is Shina and means “head”; in this context “head of the fairies”.
41
Three shamans (Bagrot, Central Hunza and Lower Hunza) had three mother-rachiye, while
two shamans (Lower Hunza and Nager) had two mother-rachiye.
42
The youngest mother-fairy (in Chaprot) was 15 years old, while the oldest (in Nasir Abad)
was 80 years of age. However, most of the shamans (14 out of 22) indicated that their mother-fairies
would be 30 to 50 years old. Two shamans did not have mother-fairies, and four did not provide
information on the age of their mother-fairies.
43
The age-range for child-fairies stretches from 1 to 30 years. However, there were only two
shamans that had child-fairies older than 14 (both indicated 30 years); if these were not considered
(as they are extreme exceptions), the average age would drop to 8 years only. The gender, age and
place of residence of the shamans interviewed seem to be irrelevant to the age of their fairies.
44
The Shina term chalo is only used in Chaprot and Chalt. They belong to the same category of
214 Peter Nicolaus

Surprisingly, most daiyali interviewed do not know the names of their ra-
chiye, or said that they generally have no names―at least, no names used in this
world.45 They seem, however, much better informed about the creed of their ra-
chiye. Logically rachiye share the religion of their daiyali; which means that the
rachiye in Lower and Upper Hunza are Ismailis, while those in Bagrot and Nager
are followers of the Shia. Yet, the tolerant character of the people living along the
Karakoram Highway seems to be reflected in the statements of a number of dai-
yali, underlining that they have rachiye of different Islamic denominations and
even religions.46 On the other hand, one female daiyal in Oshikhandas in Gilgit
district talked about religious quarrels and even wars in Koh-e-Qaf;47 mainly be-
tween Shia and Sunni rachiye―maybe a reflection of the increasing sectarian
violence in the region.48 The daiyali are also well informed about the language
spoken by their rachiye; they speak Shina and communicate exclusively in this
language, even if the daiyal―like the one interviewed at Ganish in Central
Hunza―does not speak Shina.49
Most descriptions of earlier travelers and scholars depict rachiye as young
females with a very fair complexion, golden hair and blue eyes―to a great ex-
tent reflecting the European beauty ideals adopted by the local population.50
They further reported that the rachiye, described by the daiyali they met, had
vertically shaped eyes and the feet pointing backwards.51 Obviously this is all
changing. Firstly, as mentioned above, present-day rachiye are either mature
mother-fairies or young children. This seems to reflect growing Islamic funda-

magical beings, namely the de.us and “play a prominent part in the folktales of the country as de-
mons, ogres, etc., usually of maleficent proclivities” (Lorimer 1929: 516).
45
Only three shamans provided the names of their fairies, among them Bi Bi Cha (see above).
46
13 shamans (mainly from Lower Hunza) said that they have a minority group of fairies with a
different belief (majority Ismailis and minority Shiites) and one shaman in Chaprot had, in addi-
tion to his Shia fairies, Hindu and Christian ones. For the tolerance prevailing in the area, see also
Baloch (2004: 83).
47
Koh-e-Qaf, the fairyland located in the Caucasus Mountains.
48
One of Snoy’s (1975, 176) informants reported about a war between Muslim and Hindu fairies
in 1947.
49
The neighbours confirmed that he only spoke Burushaski, but would communicate with his
fairies in Shina.
50
Jettmar (1975: 220) repeats the local opinion prevailing at the time, namely that the fairies
“sähen aus wie englische Ladies”. Snoy (1975: 176) confirms this and stresses that, according to his
informants, the fairies would wear European female dresses.
51
Ibid.
 The Taming of the Fairies 215

mentalism, the rules of which do not allow young females to be together with
men without the presence of witnesses―apparently this holds also true with re-
gard to fairies.52 Secondly, only very few daiyali reported that their fairies had
vertical eyes or feet pointing backwards; it seems that these features will soon be
attributed only to ruiye and other evil creatures from the demon-world. Thirdly,
the pendulum is swinging back from the European to the sub-continental beauty
ideal and many of the interviewed daiyali affirmed that their rachiye were not
blond and blue eyed, but rather look like beautiful Pakistani children. This also
affects the style of their clothes, with rachiye again seeming to prefer the tradi-
tional firaq to European dresses. The colours of the dresses, however, remained
the same; rachiye love white and green and abhor red and black. Apparently,
they do even not allow their daiyali to wear colours that they detest.53
As much as the appearance of the rachiye has changed during the last 50
years, their abodes remained unaffected. Rachiye continue to live, like other
periye, in palaces next to glaciers on the highest peaks (Nanga Parbat, Rakaposhi,
Diran etc.) or in the Caucasus (Koh-e-Qaf). The palaces are made out of precious
or shiny materials―like gold, mirrors, gemstones, mountain crystals or simply
ice54―and are surrounded by lush alpine meadows of flowers where ibexes
peacefully graze beside ponds filled with ibex milk. The fairies spend their days
lying on zan pans, chatting and bathing in these milk-ponds. Most of the female
daiyali interviewed report that rachiye do not eat, but only consume the fra-
grance of the alpine flowers. The male daiyali do not deny this, but add that ra-
chiye would also drink ibex milk. The intensity with which rachiye are fond of
the ibex, is only matched by their contempt and disgust for the cow and bovine
products.55 They associate the cow with pollution, dirt and uncleanness―in par-
ticular, a filthy daiyal―infuriates them the most.56

52
In other words, young (but adult) fairies should not exist as they could meet men (not only
their male daiyal) wherever they go. Another solution would be the afore-mentioned chalo or de.u,
as they accompany the fairies and protect them. Maybe these are two different concepts intending
to resolve the problem.
53
It is interesting that only female shamans made statements about the dress style, while men
referred only to the colour.
54
Leitner (1893: 6): “They call it ‘Shell-batte-kot[1]’ or Castle of Glass-stone”.
55
In a folktale reproduced by Lorimer (1935: 13) the smoke of burning cow dung prevents a
fairy from flying away.
56
For taboos related to cows and their changing patterns, see below.
216 Peter Nicolaus

Although some of the daiyali interviewed confirm that their rachiye have
wings,57 most rachiye do not use them when they visit their shamans. Instead
they just appear, arrive on flying zan pans, or come down on nylon or polyester
twines. Flying carpets and rachiye riding on the shoulders of a 12-headed de.u are
the more exotic exceptions.58
The majority of the shamans that were interviewed stated that their rachiye
do age,59 although very slowly, so that not even their daiyal would notice.60 Inter-
estingly, about half of the female daiyali61 emphasised that their rachiye do not
age. One young female daiyal from Bagrot stressed: “Rachis don’t age and don’t
die! That’s the reason they don’t want to live in this world―otherwise they would
age and die”. The majority of the shamans interviewed62 agreed with the latter
statement, saying that their rachiye―although aging―are immortal. This seems
to contradict Csáji’s informants in Central Hunza who lead him to believe that
the fairies were “vanishing” and “dying out of this world” (2011: 171); or it at least
indicates that there are regional differences, in particular between Central and
Lower Hunza.63 Those daiyali, who believe that their rachiye are mortal, men-
tioned graveyards for fairies in Paristan or the Caucasus mountains64 and some

57
While 12 shamans (7 females) insisted that their fairies had wings, seven shamans (3 females)
denied this and nine did not comment on this question.
58
A 28 year old female shaman in Bagrot (flying carpet) and a 44 year old male shaman in
Lower Hunza (12-headed de.u).
59
While 10 shamans indicated that their fairies do not age, 15 were of an opposite opinion (3
did not comment on this question).
60
Only 2 shamans indicated that they could observe their fairies aging and gradually losing
their power. See in this context Csáji (2011: 171) whose informants in Central Hunza advised him
that the fairies “are continuously losing their power as they grow older”.
61
Seven out of 14 shamans interviewed. It is further interesting that the majority (8) of those
who confirmed that their fairies do not age are younger than 50 (12 shamans interviewed were
younger than 50).
62
While 17 shamans stated that their fairies would not die (9 had aging fairies), 8 shamans re-
plied that their fairies were mortal (3 shamans had no opinion). Similar to the findings concerning
the aging of the fairies, the majority of female shamans (10 out of 14) stated that their fairies would
not die; the age of the shamans would seem irrelevantto the question of whether or not fairies
were mortal (8 below 50 years of age and 9 above).
63
This seems, indeed, quite likely as basically all shamans from Lower Hunza (12 out of 13; 1
had no opinion) held that their fairies were immortal, while those from Central Hunza (2 out of 3)
indicated that their fairies would die.
64
Paristan (Persian for “fairyland") is a name of the fairyland in the folklore of Middle East,
South Asia and Central Asia. Often the term Koh-e-Qaf (the Caucasus Mountains) is also used for
 The Taming of the Fairies 217

talked about having visited these sites with their rachiye. Others account that
fairies receive sea burials or vanish at the end of their lifespan, into the ice of a
glacier or the flames of a fire.
Rachiye are generally benevolent, but their powers and their contacts with
the human world are not always appreciated. Like gods, they are able to reward,
but may also destroy and kill. In addition, they are emotional! They support and
protect their favourites and even fall in love with handsome men.65 But they also
follow with their wrath those humans whom they dislike (Leitner 1893: 51,
Jettmar/Kattner 2003: 42)—sometimes for no reason. Shamans consider rachiye
to be a burden and it is not really seen as a privilege to be chosen by rachiye as
their tools. On the contrary, all daiyali interviewed tried to resist this ‘honour’.
Furthermore, it is not merely the selected individual who is hesitant, but in most
cases the whole village finds itself uncomfortable at the thought of harbouring a
new daiyal and his/her fairies.
2) SHAMANS―TAMING THE FAIRIES
The story of a 50-year-old male daiyal from the village of Ganish in Central
Hunza seems symptomatic of the personal evolution of many shamans:
“Already when I was one year old, I often fainted. I did not eat properly and was ill
most of the time. My parents became worried and took me to Rahimo, a daiyal from
Altit. Rahimo told my parents that I had rachiye visiting me and that it would be my
destiny to become a shaman. My parents did not like the idea and did not take any
further action. I used to jump from high walls and trees when I just started walking.
When I got older, I frequently run away from home during nighttime. My fairies hurt
me a lot! When I reached the age of 8, I became seriously ill and finally my parents
allowed Rahimo to treat me and to become my guiding shaman. However, my par-
ents were poor and could not afford the cost of a chatti. So they approached the
tham [Mir of Hunza] who agreed to support us financially. But the people from my
village were against a chatti―they did not want to have rachiye and a daiyal
around. Finally, I became even sicker and the villagers allowed my family to conduct
a chatti for me”.
In addition females, in addition, often have problems with their husbands,
and their husbands’ families, as a 28-year-old woman from Lower Hunza ac-
counts:

fairyland. An informant of Snoy (1975: 176) talked about fairy-graveyards in Chilas and in a valley
next to Karga near Gilgit.
65
See, for instance, the folktales reproduced by Leitner (1893: 7) and Lorimer (1929: 521), as well
as the account of Snoy (1975: 179).
218 Peter Nicolaus

“When I was 10 years old I became seriously ill. I frequently fell unconscious and was
attracted by traditional music and juniper smoke. I got married when I was 14 years
old, and after the marriage my fairies were hurting me a lot. Many times voices
called me during the night and I ran away from my husband’s house to the moun-
tains. One day I went to the garden with one of my friends. Nearby, some boys burnt
juniper branches. I dashed towards the smoke and put my hands into the fire, but
my hands remained unhurt. It was obvious: I was chosen to become a daiyal. When
my husband and my father-in-law learnt about it, they became very angry, as they
did not like me to become a shaman. Mercifully, my father took me back to his home
and told my husband’s family that he would only return me if they would organize a
proper chatti for me. They brought musicians from central Hunza and I had my
chatti; I was 15 years old”.
It seems that the initial reaction of the families and the villagers towards a
shaman-to-be is almost always negative. Although the benefits provided by a
shaman to society are generally recognised and welcomed. Periye usually try to
take possession of a person when he/she is quite young; most of the daiyali in-
terviewed had their first encounter with the fairy-world when they were around
11 years old.66 In most cases, this first contact was accompanied by severe illness
and aggressive hostility, not only towards others,67 but also against themselves.68
Furthermore, they develop an aversion to beef and associated products, includ-
ing manure. The family’s reaction is usually to take the person to a hospital, or if
violence is displayed, to isolate the individual. More often than not, this is in vain
and the family will then seek the advice of an Islamic cleric; an akhound (Shiites)
or a khalifa (Ismailis).69 These religious dignitaries will provide the person with a
tumar (amulet) against the negative influence of the periye,70 or, in rare cases, ad-

66
Amongst the 28 shamans interviewed, the age range for the first encounter was between 1
and 32 years. However, 21 shamans had their first contact at 15 or below and 6 shamans between 16
and 22.
67
A 61 year old female shaman from Lower Hunza accounts: “In my childhood I was very aggres-
sive. One day, I quarrelled with a girl and started beating her, though she was four years older than me
and much stronger and healthier. Other girls tried to stop me, but they couldn’t and finally I broke her
leg”.
68
Often by jumping from high walls or the roof of the family house. The joyful story of a
shaman from Central Hunza about his first encounter is certainly the exception. He flew on his bi-
cycle from Gilgit to his home village.
69
With regard to the role of the khalifas in Hunza, see Müller-Stellrecht 1979: 217-223.
70
A 29 year old female shaman from Lower Hunza accounts: “My grandfather was a khalifa. He
made a tumar for me. He copied, with the blood of a pigeon, spells from an ancient Persian book on a
sheet of paper. This was sewn into a piece of fabric, which I had to carry around my neck”.
 The Taming of the Fairies 219

minister as counter-charm a glass of cow urine.71 In many cases these actions

yield results and the periye leave that person alone.72 However, if the fairies are
very strong and determined then the protection provided by these actions is
temporary at best. With the daiyali who were interviewed, charms applied by
akhounds or khalifas either did not work at all, or faded away after some years.73
At this point the family usually agreed to consult a daiyal willing and capable to
act as a guiding shaman.
A number daiyali-to-be see and communicate with the periye visiting them
well before they begin to receive guidance from another shaman, while others
only become acquainted with their rachiye during this guiding process.
However, the pattern of appearance is almost always the same. First, the
shaman-to-be sees small flying animals―like mosquitoes, beetles, butterflies or
sparrows, then larger birds (such as pigeons and swallows), and finally the periye
appear. In those instances where the fairies show themselves at an early stage
they often try to enter into a milk-kinship with the individual concerned.74 If the
fairy is successful, the pressure on the shaman-to-be ceases to a large extent, but
on the other hand his/her chances to be spared the fate of a daiyal are waning.75
Fairies also often oppose the marriage of female daiyali-to-be and, almost all

71
This follows a common belief, that drinking cow urine blinds fairies and they can only be
cured when the person subsequently drinks water flavoured with special mountain flowers. Lori-
mer (apud Müller-Stellrecht (1979: 264) provides a similar account: “They employ a khalifa to ‘bind’
a bitan. They put cow’s milk, cow’s meat and blood, in his mouth and so bind him”.
72
Several daiyali emphasized that their siblings were saved through such interventions by Is-
lamic clergymen.
73
Sometimes the changes were brought about unintentionally, like in the case of a 29 year old
female shaman in Lower Hunza: “I was 13 years old when I got married. One year later my in-laws
threw my tumar into water―it lost its power”.
74
A young female shaman from Lower Hunza remembers the time she was four years old: “The
second day after I saw my rachi, she came again holding an iron rod and told me to drink from her
breast. She threatened me with the stick, but I did not take her milk. She became angry, beat me and
disappeared. After another two days, the rachi came back holding a baby in her arms. She said that
this was be her son and complained that he did not feed properly from her breast. She said: ‘Please
help me and push him towards my breast’. Thus, I pushed the mouth of the baby towards the breast of
the rachi, but suddenly the baby disappeared and her milk splashed into my mouth”.
75
In the Islamic context, as well as in the folktales of Bagrot, Gilgit, Hunza and Nager, the ties
between a woman who wet-nurses a foster child are stronger than the ones between a mother and
her natural child. The person concerned will become a part of the peri’s family and hence, will not
be able to escape her. See also Snoy’s chapter ‘Milchverwandtschaft’ (1975: 140-142) and Jettmar
(1975: 221).
220 Peter Nicolaus

women who had contact with fairies before their families requested the assis-
tance of a guiding shaman,76 reported that the periye threatened them saying
that they “would inflict even more pain if they got married”.77
The role of the guiding shaman is to influence and appease the periye―the
future rachiye of the daiyal-to-be. This is established through two processes: oyao
and ganao. Oyao is a Shina word, which means ‘to open’—in this context, to
allow good spirits to access the daiyal-to-be and to establish good relations with
them. Ganao denotes the opposite, namely ‘to close’ the paths leading to the in-
dividual concerned for evil spirits and to fix them in one location so that they
can no longer do harm.78
During this process the guiding shaman permanently blocks the access of
malicious periye to the daiyal-to-be (ganao).79 The guiding shaman will also
introduce the future rachiye to him/her and convince them to take―during the
forthcoming ceremony―an oath to support the new daiyal whenever he/she
will need help, to appear whenever the new daiyal calls them, and most im-
portantly, to not disturbed or harm the new daiyal (oyao). In return, the guiding
shaman agrees to vouch for the new daiyal during the above-mentioned cere-
mony, and to promise that he/she will heed the rachiye’s advices as well as live
according to their rules.
It can take months, if not years, for a guiding shaman to complete this task―
although, with most of the shamans interviewed it took just a few weeks, and in

76
10 out of the 14 female shamans interviewed.
77
Considering that all of the 14 interviewed female shamans got married at a very young age
(most of them married when they were 12 or 13), these fairy-threats could be seen as reflecting the
fear of the child-brides of being taken away from their parents.
78
A 45-year-old male shaman from Central Hunza provided the prayers or magic formulae for
oyao and ganao (however, in an unknown language, which the fairies did not allow him to trans-
late). The formula for oyao: “Ame moti, dame moti, Dame motura moti, Areli moti kareli moti, Kar ka-
reli moti, Daje Darda Moti, Dam te palaton moti, Jin Pari mote, Deo pari mote, Lailai Rathas moti, Sat
Pariye mote, Yachalange mote, Roye mote, Jun Jun Dara Muto, Sehro jado moti, Takhte Sulaiman Kul
Maqsood dam te riza hum moti”. The formula for ganao: “Ame Ganem, Dame ganem, Dame mutura
ganem, Areli ganem, Kareli Ganem, Kar kareli Ganem, Daje Darda Ganem, Dum te palatoon ganem,
Jin Pari deo pari Laiali rathas ganem, Sat Asmani Ganem, Sat Zamini ganem,Sat cheeye ganem, sat
chiliye ganem, Sat poniye ganem, Ganem ganem ganem bani adam ganem, Takhte Sulaiman Kol
Maqsood chode ban kapan them”.
79
The same shaman from Central Hunza stated in this context: “There are so many periye who
tried to be with one and the same shaman, some are evil and some are good. Thus, the guiding sha-
man stops the evil periye and supports the good ones”.
 The Taming of the Fairies 221

some instance, only days.80 As soon as the periye agree to become the new
shaman’s rachiye, and the afore-mentioned agreement is reached, then a chatti is
organised. Hence, the chatti does not seem to be a typical initiation ceremony,
even though it is the event where, for the first time, the new shaman dances and
publicly falls in trance and the rachiye teach him/her prayers and magic formu-
lae,81 while he/she is unconscious. Rather it is the occasion where the fairies are
tamed and where they have to take the afore-mentioned oath. This oath is sworn
on the kau, the iron82 bangle which is dipped into the blood of the sacrificed goat
during the chatti ceremony. Although the fairies may have consented to all the
conditions of the agreement during the oyao, they may change their mind and
ask for additional concessions. In this case the kau either disappears or breaks
when it is held in the blood,83 or it fails to render the dancing daiyal unconscious
when it is placed on his/her arm.84 Then, the oyao-process has to be continued,
and if a consensus with the periye is reached, the chatti-ceremony will either
continue or be repeated.
On the other hand, periye can be lenient and may not demand a chatti. In-
stead they may indicate that they would be satisfied with a sheifial. This cere-
mony is conducted in a much smaller circle involving only relatives and friends,
as well as some small girls. These girls, about eight years of age, are dressed in
white or green cloths. In many cases the number of girls reflects the number of

80
With regard to two of the shamans interviewed, one from Nasir Abad (Lower Hunza) and the
other one from Chaprot (Nager), such an agreement was never reached and until now they are still
at the mercy of some periye molesting and physically hurting them.
81
A 64-year-old female shaman from Lower Hunza told: “I had my chatti when I was 15 years old.
When the kau was put on me I became unconscious and my rachiye started teaching me the prayers.
However, I was woken up before they finished. My hands were put into cold water too early. Hence, my
rachiye became angry and had to be convinced again. We repeated the chatti after three days had
passed”.
82
One shaman from Central Hunza wears a silver kau.
83
Another 64-year-old female shaman from Lower Hunza reported that her kau broke twice
and disappeared during her chatti (when she was 20). Only the third iron bangle finally held. The
guiding shaman had already told her family that she had very strong fairies and that they should
bring a number of kauye to the ceremony.
84
A 60-year-old male shaman from Chaprot (Nager) explained: “The guiding shaman prays over
the kau while dipping it into the blood of the chatti. Then the teacher shaman puts the kau on the arm
of the dancing daiyal. If his rachis are fully convinced and ready to swear the oath he becomes uncon-
scious; if the guiding shaman was not able to convince the rachis then the daiyal continues dancing.
The guiding shaman has to put the kau into the blood again and continue to convince the rachis”.
222 Peter Nicolaus

rachiye adopting the new daiyal. Bread and goat-butter are put on a table, and
the kau is placed on top of the butter. The guiding shaman softly mutters prayers
and magic formulae over the bangle. Then the kau is fixed on the arm of the new
daiyal and―like in the chatti-ceremony―he/she becomes unconscious. Mean-
while, the girls start to eat the bread and butter. They are soon joined by the
other participants, as well as by the new daiyal when he/she awakes from the
trance-sleep. It seems that the sheifial-ceremony is mainly conducted for female
daiyali85 and one could speculate that the advancing Islamic fundamentalism
constitutes a major reason for this, as it is certainly opposed to young women
exposing themselves by dancing in public and drinking the blood of a goat.86
This speculation is supported by looking at the frequency of a chatti ceremony
versus a sheifal ceremony when compared with the age of the female shamans
interviewed: among the four oldest female shamans (61 to 68 years) only one had
a sheifial-ceremony, while among the four youngest (28 and 29 years) only one
did not have a sheifal ceremony.
All shamans interviewed agreed that a chatti or the substituting sheifial-cer-
emony can be held only once, namely, when the fairies take the oath on the kau.
However, there is disagreement as to whether or not a kau can be replaced, if it
gets lost or breaks. While a strong minority87 emphasizes that a kau can be re-
placed, the majority maintains that this would be impossible, or that the re-
placement would not be as protective as the original.88 Actually many of the mi-
nority group did lose their kauye and have―sometimes with assistance of an-
other shaman―fashioned a replacement either in the form of a new kau or an
embroidered ribbon to be worn on the wrist or around the neck.
The kau enables the daiyal to simply switch the rachiye and the entire magic
world off. In other words, as long as the shaman wears the kau, the fairies cannot
approach him/her and the shaman may do things, which are otherwise taboo:89

85
Out of 12 male and 14 female shamans, 7 females and only 1 male underwent the sheifial cere-
mony. 7 females and 11 males had a chatti. Two of the interviewed were denied any ceremony by
their fairies (see above).
86
The latter also holds true for male shamans, but there may be much more tolerance with re-
gard to males.
87
12 out of 28.
88
In summer 2011, the author bought a kau belonging to a female shaman in Lower Hunza for
300 USD. Then in spring 2012, she offered to buy it back for 350 USD, as the replacement was not
protecting her sufficiently against her rachiye.
89
As a 40-year-old female shaman in the Gilgit district phrases it: “I could not eat beef products
 The Taming of the Fairies 223

such as eating beef, drinking milk, meeting with menstruating women, wearing
red or black coloured clothes, visiting houses where people recently died or ba-
bies have been born, etc. The kau, however, does not protect the daiyal from the
rachiye’s wrath if he/she does not keep him/herself clean or drinks alcohol. This
is in strong contradiction with earlier accounts where the cow-taboo was re-
ported as the most important and absolute, while the consumption of alcohol
was seen as something almost indivisible from daiyali and the peri-cult—at least
in Hunza.90 Again, this seems to reflect progressing Islamic fundamentalism, as
well as a changing food economy, where-in beef has replaced mutton to a great
extent.
Nowadays, daiyali care more for individual clients than the society at large.
This means that they act as mediums between their fairies and an individual re-
questing them for help. Through the daiyal, the rachiye give counsel on all as-
pects and questions of life. Following their advice, the shaman prepares magic
potions and amulets. They cure diseases, provide love-charms and protective
amulets, predict the future, counter black magic, etc. Contact with the supernat-
ural world is usually established either through a trance, into which the shaman
falls by inhaling the smoke of burning juniper boughs, or through a dream the
shaman has after sleeping on an item that belongs to the person seeking advice
(e.g., a piece of fabric from his/her clothes, a button, or a coin).
Although these individual consultations flourish, the tradition of ashosch―
the traditional public dance of the shaman establishing a link between the
people and the fairy-world in order to receive prophecies―has meanwhile
become extinct. This shamanic dance should not be confused with the chatti-
ceremony. The shamans interviewed—without any exception—stressed that
the sacrifice of a goat,91 and the drinking of the blood dripping from the goat’s
head by a daiyal is essential for the chatti-ceremony. However, this was never
part of ashosch! This is confirmed by earlier reports;92 and if correct, would mean
that Sidky and Subedi believed the swindle against which they were cautioning.93

and I was not allowed to wear red and black clothes. This all changed after the sheifial and now I
can―as long as I carry my kau―eat everything and wear what I like”.
90
See Jettmar’s chapter ‘Religiöse Aspekte der Viehzucht’ (1975: 250-6).
91
This does not mean that there could not be a slaughtering of a goat on the fringes of an ash-
osch. On the contrary, rich people or the tham usually presented the participants with a goat in or-
der to provide the food for the festival.
92
Müller-Stellrecht (1979: 262); Jettmar (1975: 277) (“Die erste Séance wird oft mit besonderer
224 Peter Nicolaus

Sidky and Subedi were, however, perfectly right when they highlighted the
social power a daiyal could exercise during an ashosch. He/she would often vali-
date the activities and actions of the tham. Furthermore, during “their trances,
bitan would sometimes offer supernatural support for public grievances and allevi-
ate the anxieties of the people. For example, in the name of the pari they might voice
objections to state taxes, or challenge the improprieties of members of the upper
class” (Sidky/Subedi 2000: 79). This certainly challenged the power holders and
since the ruling class could not control the fairies, they and in particular the
tham, needed to establish a social mechanism that would at least ensure some
sort of control over the daiyali.
3) THE DOM MUSICIANS―CONTROLLING THE SHAMANS
No daiyal can perform the shaman-dance without a certain type of music. These
so-called ‘fairy-tunes’ may only be played by a special caste of musicians who are
linguistically and anthropologically distinct from the Shina and Burushaski
speaking population of the area. They constitute the lowest caste. In the past,
they have been discriminated against,94 because they traditionally worked as
blacksmiths, leatherworkers and musicians in the Hunza-Nager Valley.95 Even to-
day they are considered somewhat ‘unclean’. They call themselves Dom, while
Burushaski speakers refer to them as Bericho.96 The Shin also call them Dom, but

Feierlichkeit veranstaltet. In ihrem Verlauf wird eine Ziege herangeführt und ihr der Kopf abgeschla-
gen. Ein gutter Daiyal stürzt sich auf den blutenden Rumpf und versucht, das hervorsprudelnde Blut
zu trinken–er hältes für Milch.”); Durand (1899: 291)(“At the end of the performance a goat is brought
in and decapitated, and the novice has to seize the neck and drink the spouting blood.”). However,
Snoy (1975: 208-209) remarks in this context: “Leitner erwähnt dieses Trinken von Ziegenblut als
Abschnitt einer offensichtlich normalen Séance. […] Heute wird während normaler Seancen keine
Ziege getötet. Ob dies früher zu jeder Séance gehörte, wie es dem Bericht Leitners zu entnehmen ist
[…], läßt sich nicht zwingend folgern”.
93
These authors visited Central Hunza and witnessed shamanic dances resembling a chatti-
ceremony (i.e. the shaman drinking the blood of a goat). They considered these performances as
genuine, but stated that since “the completion of the Karakoram Highway in 1982 … a number of self-
professed bitan have appeared who earn hefty fees by putting on bogus shows for tourists” (Sidky;
Subedi 2000: 100).
94
For examples of this discrimination, see Müller-Stellrecht 1979: 145.
95
For more on their anthropological background, particularly with regard to the Dom migra-
tion and their demographic distribution outside the Hunza/Nager Valley, see Biddulph 1880: 39;
Müller-Stellrecht 1979: 144-145; and Sidky 1994: 80.
96
But this is changing too, as they disapprove more and more of being associated with a cer-
tain (low) cast and refute to be called Dom or Bericho. The musicians present in the above-de-
scribed ritual demanded: “Don’t call us Dom or Bericho―call us simply musicians”.
 The Taming of the Fairies 225

in Lower Hunza, which is also Shina speaking, they are known as Ustatey. Their
language is Domaaki97 “although the higher cast of Burushaski speakers refers to
this language in a derogatory way as Beriski”.98
The Dom “live segregated from the rest of the population in a village in Central
Hunza, called Berishal” (Sidky 1994: 80). There are actually two places called Beri-
shal: one is in Central Hunza near Karimabad, nowadays known as Mominabad;
the other one is located within Nagar town in Nager valley.99 Due to the public
demand for musicians and blacksmiths, the ruler of Hunza moved some families
from Berishal to Lower Hunza (Hussain Abad) and Gilgit (Domyal–a quarter of
Gilgit town).
About 30 years ago, the Dom families outside Central Hunza gradually
stopped performing as musicians, and now—although a kind of renaissance
with regard to traditional music amongst these families has begun–they no long-
er have the necessary experience to perform during a daiyal séance. Hence, for
the above described chatti-ritual, Bericho from Berishal were asked to perform.
In most societies, which are built upon, or include elements of shamanism,
the shamans play their music themselves. In fact, the shaman beating his/her
drum is considered an icon of shamanism.100 Therefore, entrusting the Dom (a
third party) to perform ritual music is atypical and appears to be a relatively new
development occurring following the acceptance of the tham by the local Shin
aristocracy in the 18th century. This seems to be confirmed by a legend accord-
ing to which the Dom learnt to play the fairy tunes from a famous daiyal (Sidky/
Subedi 2000: 42).
The Dom were closely linked with the Shin101 and the ruling houses in Hunza
and Nager. Some sources in Hunza, as well as some of the Dom themselves, men-
tioned that the Dom or Bericho once belonged to the tham.102 The tham gave

97
Indo-European → Indo-Iranian → Indo-Aryan → Central Zone → Domaaki (Lewis 2009).
98
Winter (2004: 293), briefly describes the situation of the Dom as a population/language that
is on the brink of extinction, but seems to survive.
99
Known as Oyum Nagar (Burushaski), Nagar Khas (Urdu) and Nagar Proper (English).
100
Csáji (2011: 177) very rightly remarks: The shamans “cannot play any musical instrument at
all―kind of taboo that is opposite of the customs of most other shamanic communities in Eurasia”.
101
It is widely assumed that the Dom already lived as a subordinate class under the Shin in
northern India and migrated together with them to the Hunza Valley where they prospered “under
the protection of the Shin rule” (Biddulph 1880: 39).
102
Snoy (1975: 51) confirms this for Bagrot when he states that Dom were part of the property,
which could be inherited.
226 Peter Nicolaus

them land, kept them as his servants, and selected musicians from amongst
them.103 These musicians were the only ones allowed to play the fairy tunes.
This could mean that there existed a balance of power between the tham and
the shamans. The tham was perceived as heaven-sent and the ruling dynasty was
known as Ayashkutz, (descended from heaven) (Sidky/Subedi 2000: 7) until the
last tham abdicated. Even after the Islamisation of the Hunza-society, the tham
had to perform ancient religious or quasi-religious rituals and hence, needed to
have close relations with the spiritual world, particularly with the periye. “It was
because of their special relationship with the pari that the rulers of Hunza claimed
to possess the power to make rain, melt glaciers, and quell storms” (Sidky/Subedi
2000: 69). In other words, due to the entwinement of his spiritual and worldly
duties, the tham needed the periye and the link to the supernatural world. How-
ever, since he was not able to communicate with these supernatural beings, he
had to rely on the daiyali, just like the rest of the population. As indicted above,
this situation gave the shamans a considerable degree of freedom to endorse or
to criticize, in the course of a public ashosch, decisions and actions of the tham
and the ruling class. In order to keep the daiyali at bay, it appears that the tham
introduced an underprivileged segment of the population, which was completely
dependent on him, to perform as musicians for all public daiyal-performances.
In other words, by monopolizing the sacred music in this way, he could success-
fully counterbalance the shamans and even indirectly control them.104
The final and complete integration of Hunza into the Islamic Republic of Pa-
kistan, in 1974, and the subsequent loss of power by the tham, made this need to
balance power redundant and should have triggered adjustments. However,
there was no need for readjustments, as at the same time the public shamanic
dance, the ashosch, was becoming extinct and the chatti was gradually being re-
placed by a sheifial-ceremony. This means that in the few chatti-ceremonies,
which are still performed, the Dom musicians continue to play an active role
when a new shaman’s fairies are tamed.

103
This is supported by Lorimer (apud Müller-Stellrecht 1979: 146).
104
See Sidky/Subedi 2000: 17: “In the past, the Bericho could perform only at the behest of the Mirs.
For this reason the Mirs exercised a considerable degree of control over the bitan”.
 The Taming of the Fairies 227

BIBLIOGRAPHY
Baig, R. K. (1997), Hindu Kush―Study Series, Vol. 2, Peshawar.
Baloch, S. K. (2004), Gilgit and Baltistan: In the Wonderland of Asia, Lahore.
Biddulph, J. (1880), Tribes of the Hindoo Koosh, Calcutta.
Csáji, L. K. (2011), “Flying with the Vanishing Fairies: Typology of the Shamanistic Traditions of the
Hunza”, Anthropology of Consciousness 22.2: 159–187.
Durand, A. (1899), The Making of a Frontier, London.
Jettmar, K. (1975), Die Religionen des Hindukusch, Stuttgart.
―― / E. Kattner (eds.) (2003), Die vorislamischen Religionen Mittelasiens, Stuttgart.
Leitner, G. W. (1893), Dardistan in 1866, 1886, and 1893. Being an Account of the History, Religions,
Customs, Legends, Fables and Songs of Gilgit, Chilas, Kandia (Gabrial), Dasin, Chitral,
Hanza, Aagyr, and other Parts of the Hindukush, as also a Supplement to the Second Edition
of The Hunza and Nagyr Handbook and an Epitome of Part III of the Authors ‘The Languages
and Races of Dardistan’, Woking.
Lewis, M. P. (ed.) (2009), Ethnologue: Languages of the World, 16th edition, Dallas (Online version:
http://www.ethnologue.com/16).
Lines, M. (1988), Beyond the North-West Frontiers: Travels in the Hindu Kush and the Karakorams,
Oxford.
Lorimer, D. L. R. (1935), Folk Tales From Hunza, Oslo.
―― (1929), “The Supernatural in the Popular Belief of the Gilgit Region”, Journal of the Royal
Asiatic Society (New Series) 61: 507-536.
Müller-Stellrecht, I. (1979), Materialien zur Ethnographie von Dardistan (Pakistan). Aus den nachge-
lassenen Aufzeichnungen von D. L. R. Lorimer, Teil I: Hunza, Graz.
―― (1980), Materialien zur Ethnographie von Dardistan (Pakistan). Aus den nachgelassenen Auf-
zeichnungen von D. L. R. Lorimer, Teil II: Gilgit/Teil III Chitral und Yasin, Graz.
Rashid, A. (2012), Pakistan on the Brink. The Future of America, Pakistan, and Afghanistan, New
York.
Sidky, M. H. (1994), “Shamans and Mountain Spirits in Hunza”, Asian Folklore Studies 53: 67-96.
―― / J. Subedi (2000), Bitan: Oracles and Healers in the Karakorams, Jaipur.
Snoy, P. (1975), Bagrot. Eine Dardische Talschaft im Karakorum, Graz.
Winter, D. (2004), Footprint: Northern Pakistan, Bath.
 The Classification of Astral Bodies in the framework
of an Historical Survey of Iranian traditions

Antonio Panaino
University of Bologna at Ravenna

Abstract
The current paper presents the status of our knowledge about the classification patterns adopted
in the earliest phases of the Avestan tradition with regard to the astral bodies and their supposed
identity, origin, function and role in the framework of the Mazdean religious tradition. Then, this
study underlines the fundamental importance of the single stars in the Zoroastrian cosmology and
the forms of distinction, individualization and categorization they were attributed, with particular
regard for the determination of some constellations. Remarkably interesting appears the distinc-
tion between the stars and the two luminaries, the Sun and the Moon, all of them celestial symbols
of the preservation of the correct cosmic order in the heavens as on the earth. Furthermore, the ar-
ticle emphasizes the problem of the subdivision of the heavens in three and more levels, and its
origins as well as the following introduction of new categories of stars. Very important is, at least
for the earliest periods of the Iranian civilization, the evident absence of the planets (and of the
planetary demons), whose clear identification was only a later phenomenon, probably derived by
contacts with the Mesopotamian traditions. On the contrary, the Old Iranian people focused on
their closer attention on astral bodies whose movement was unpredictable and without regularity,
as in the cases of falling stars, bolides, etc. This category of celestial bodies was later associated
with the planets and their negative role, as symbols of cosmic disorder, but framed in the new pat-
terns of the astrological tradition, which was completely unknown in the Old Iranian period.

Keywords
Stars, Constellations, Luminaries, Planets, Falling Stars, Cosmic Order and Disorder, Mazdeism,
Zoroastrianism, Cultural contacts

Our knowledge of the Iranian representation of the astral bodies is based, unfor-
tunately, on a limited number of sources,1 mostly of mythical and ritual nature,
so that we do not actually possess a real corpus written by specialists dealing

1
Essential bibliography and basic discussions in Khareghat 1914; Duchesne-Guillemin 1986; Ei-
lers 1967; 1976; 1987; Pingree 1963; 1983; Panaino 1990a; 1990b; 1995a; 1995b; 1995/96; 1998; 1999a;
1999b; 2004a; Raffaelli 2001; Taqizadeh 1937; 2010; Zaehner 1955 (reprint 1972).
 Koninklijke Brill NV, Leiden, 2015
230 Antonio Panaino

with heavenly matters in a more professional way, as it happened, for instance,

in the Mesopotamian framework.2 With “professional approach” I mean an ap-
proach that, albeit it had nothing to do with modern patterns, might be at least
of the importance assumed in some ancient cultures where astral divination im-
posed observations, recordings, comparisons among various phenomena and
real events, interpretations, and, then, consequently reports established on a
kind of para-scientific “system”. The theory behind such an enormous work was
not properly “scientific”, as we would define it today, because it was based on a
series of inferences emerging from speculative associations between phenomena
and realia, so that it produced the essential patterns for a never-ending process
of divination, which, in its turn, was basic for the birth of Hellenistic astrology,
an art that found an impressive evolution also in the Iranian framework.3 But,
although the art of astral divination cannot be nowadays considered as a rational
scientific model, the systematic observations of the recurrent astral phenomena
produced the basic pillars for the birth of a scientific method. In fact, when the
periodicity of the celestial phenomena was clearly detected, calculated, and by
means of mathematical parameters (and not by divination or magic) predicted,
the method it generated was surely fundamental for the foundation and the de-
velopment of mathematical astronomy.
As I noted before, apparently we have nothing of comparable in Iran, and in
the earliest Vedic sources as well, although a certain impact from the Mesopo-
tamian tradition should have played a reasonable influence also in these areas,
at least since the Achaemenian domination of Western India.
In writing this contribution dedicated to my friend and colleague, Prof. Gar-
nik Asatrian, I have considered two theoretical problems as a premise to the
whole work. First, any classification of the Iranian astral bodies should avoid to
be just framed, at least in principle, according to the patterns of the Western
classification,4 which partly were the same of the Greeks, or to be just referred to
the comparatively most advanced ones of the Mesopotamian cultures. In other
words, I will try to represent the Iranian patterns, starting from inside, or from
the innermost schemes of this tradition, although I know that this target will be

2
See Hunger/Pingree 1999. Cf. also Neugebauer 1957; 1975; 1983.
3
See Brunner 1987; MacKenzie 1964; Panaino 1993; 1996a; 1996b; 1996c; 1997; 1998; 2004a;
2004b; 2005a; 2005b; 2009a; 2013 Pingree 1963; 1983; 1989; Raffaelli 1998; 2001; Taqizadeh 1937; 2010;
Zaehner 1955 (reprint 1972).
4
See the caveat expressed by Pingree 1992.
 The Classification of Astral Bodies 231

difficult, if not partly impossible. I want also to restrict the borders of my investi-
gation to the sources of Mazdean derivation; this in order to avoid that too many
cosmological systems with their inner divisions and models overlap each other,
adding more confusion than a first basic clarification.5 Necessary comparisons,
or contrastive parallels will be evoked, but only when and if necessary. This will
happen in particular when we will consider the structure of the heaven, and the
discovering of the planets, but also in the distinction among inherited, i.e. Indo-
Iranian or, in few cases, Indo-Europaean common patterns and designations of
realia, proper Iranian developments, and external influences, in particular from
the Mesopotamian world, but in later times, also from the Greek and Indian tra-
ditions. For these reasons a chronological sequence will be necessary.
We may start by noting that, for instance, no ancient Iranian source apparen-
tly knows the existence of the planets. This, probably, does not mean that the an-
cient Iranians did not see them (they should, of course), but that their observa-
tion was not considered relevant to determine a special different or special clas-
sification, while a deeper focus was directed towards the single stars and the
main constellations. But, also in using the term “constellation”, I am implicitly
introducing an abusive classification, because a special technical word in order
to represent a particular group of stars did not exist, or, in any case, is not
properly attested in Avestan and Old Persian. Avestan sources just call the
“stars”, for which a stem of Indo-European derivation, stār-, is normally applied,
by name, when they appear of particular brightness or of a certain significance
(we will see later why), while for groups of them Avestan texts just used (1) a plu-
ral form, as we can do today, when we call an open cluster of stars in the con-
stellation Taurus as “Pleiades”, i.e. Av. paoiriiaēinī-, or again, (2) second possibil-
ity, a bahuvrīhi-compound,6 in which a numeral is attested as first member, e.g.
haptōiriṇga-, “the Big Dipper”, literally “which has seven signs” (Kellens thinks
that this is “la constellation de sept phallus”),7 or hapta.srū-, “Ursa Minor”, lit.
“who has seven horns”.
Thus, we may state that the knowledge of the stars was surely an inherited
fact, as well as the habit to group some particularly important stars in a larger
unit, simply designated with a plural name or with a special kind of nominal
composition.

5
In the case of the Manichaean framework see Panaino 1997; 1999d.
6
Duchesne-Guillemin 1936: 176-178.
7
Kellens 2010: 756, n. 21.
232 Antonio Panaino

The single stars, mentioned in Avestan sources, are:

1) Tištriia-, < *ti-str-ii̯a-s < *tri-str-ii̯a-s < *tri-str-ii̯o-s, built on *tri-str-(o-m)
“group of three stars”, Sirius α Canis Maioris, probably linguistically related with
Ved. Tiṣya-;8 it is probable that, at least in the Indo-Iranian framework, *tri-str-(o-
m) was referred to the “Orion’s Belt”, i.e. δ ε ζ Orionis. As shown by Forssman9
and later by the present author,10 the Indo-Iranian cycle of Tištriia and Tiṣya11 was
strictly connected with a widespread association between the star Sirius and an
arrow12 (Iran, Mesopotamia), an archer (India), or with the target of the arrow it-
self (China, Egypt).13 Already in Mesopotamia, Sirius was called the “arrow”
(KAK-SI-SÁ, glossed in Akkadian as šiltaḫu, šukūdu, “arrow”). In Vedic India the
Orion’s Belt was named iṣus trikāṇḍā “the arrow of the three knots”, which was
shot by Tiṣya/Rudra against Prajāpati.
2) Vanaṇt-, “the winner”,14 Vega, α Lyrae, whose denomination is semantically
comparable with Skt. abhijit-;15
3) Upa.paoiriia-, “who follows Paoiriia[ēinī]- (< *upopelu̯ im), i.e. the Pleiads”,16
thus corresponding to Aldebaran, α Tauri;
4) Satauuaēsa-, “who has hundred servants”, of debated identification, and
usually associated with Fomalhaut, α Piscis Austrini. On the other hand, this as-
sumption is now untenable, as I have shown in a work dedicated to this subject,17
while it probably designated α Cygni (i.e. Deneb). It is to be noted that it was
never called stār- in Avestan, but it was surely one of them at least according the
Middle Persian sources.18

8
Panaino 1986; 1985a.
9
1968.
10
Panaino 1995a. For the edition of the Avestan hymn to Sirius (Yašt 8) see Panaino 1990a.
11
With regard to Gr. Σείριος, Fischer (1969) suggested that it could be also connected with the
same Indo-European stem *tri-str-ii̯o-s (via a dissimilated form like *ti-šr-ii̯o-s > * tihrii̯o-s > Σῑ́ριος),
but I doubt of this solution (see Panaino 1995a: 33, n. 21).
12
Panaino 1987; 1995a.
13
For the later traditions about Sirius see Panaino 1995a; 1996b.
14
Panaino 1989.
15
Cf. Filliozat 1962.
16
See the discussion about the name of the Pleiads.
17
Panaino 2008b.
18
Eilers (1967) assumed that Av. satauuaēsa- would have been a denomination of the “rain-
bow”.
 The Classification of Astral Bodies 233

Groups of stars, or constellations:

1) Tištriiaēinī-, “the stars of/belonging to Tištriia-”, i.e. Canis Minor; probably
this name was a deformation, after paoiriiaēinī-, of an earlier (patronimic) stem
like *tištriianī-, “the wives/daughters of Tištriia”, “the stars who belong to Tiš-
triia”.19 In fact, as Bartholomae20 already noted, it is difficult to explain the pres-
ence of the thematization in -aēinī-, which usually produces Stoffadjectiva21 in
connection with a proper name like that of the star Tištriia-.22 Thus, we cannot
exclude that *tištriianī- became *tištriiainī- and, then, produced *tištriiaēinī- on
the model of paoiriiaēinī-.23

19
Bartholomae 1904: 652.
20
1912/13.
21
See the list of Avestan examples given by Taraf 1981: 117.
22
This argument is correct if we accept that tištriiaēinī- was just the name of a constellation, as
many scholars have assumed and as it seems to be. Darmesteter (1883: 174-175), on the contrary, in-
sisted on the fact that Avestan star names are not feminine, but this statement is not compellingly
true (see e.g. paoiriiaēinī-), while he correctly remarked that in the tradition on Ny. 1, 8, it was in-
terpreted as referred to the “waters of Tištriia”. See, in fact, Ny. 1, 8c, tištriiaēniiō yazamaide “we
worship Tištriiaē(i)nī”; Pahl. translation: wārišnīg tištar stārag rāy yazēm “we worship the raining
star Tištar”; Skt. translation: tistaratārakasya vr̥ ṣṭīh ārādhaye “I worship the rains of the star
Tistara” (cf. Taraf 1981: 28-31). Although also Taraf (1981: 117) seems to question Bartholomae’s solu-
tion, in favour of an implicit return to Darmesteter’s interpretation (“les eaux de Tištrya”), I se-
riouly doubt that we can translate Av. tištriiaēinī- as the “waters by which Tištrya is made”, as we
literally should make. More probably, the old patronimic in -anī-, f. (< -ana-, m.), transformed in -
aē(i)nī- according to the model of paoiriiaēinī-, the Pahlavi and Sanskrit translations are based on
the evident influence of the “new” form assumed by the stem by means of the derivation normally
attributed to the Stoffadjectiva. It is for this reason that the matter of the star Sirius was associated
with rains, and that his astronomical “house” was strictly connected with the waters, as it appears
in Ny. 1, 8b1, but only in the Pahl. and Skt. translations. Here, in fact, after tištar rāy yazēm /
tistaratārakaṃ ārādhaye, we find: in Pahlavi ē manāzil ī wārān; in Sanskrit: tistaraṃ iti vr̥ ṣṭinak-
ṣatram or (Pahl.) “(Tištar I worship), the (astronomical) house [manāzil] of rains” / (Skt.) “(The star
Tistara I worship), Tistara, i.e. the astral mansion (nakṣatra-) of the rains”. In the Avestan text these
references to the lunar stations or mansions (here called in Pahl. manāzil, a loanword from Arabic;
Skt. nakṣatra-) are not attested at all. We must also observe that in Pahlavi these constellations
were called xwurdag; see Panaino 2009c.
23
In the framework of a larger conspectus of the problem, we must observe that Benveniste
(1966: 19, n. 4) refused the attribution of stems like paoriiaēinī- and tištriiaēinī- to the category of
the “adjectifs de matière en -aēna-”, a category which, according to him, is not existing. Actually he
assumed that these feminine forms were directly derived from masculine stems like paoiriia- and
tištriia- by means of the suffix -(a)inī-, which Benveniste compared with the “suffixe de féminine
indien en -inī-, du type jārá- : jāriṇī ; yakṣa- : yakṣiṇī [...]”, and that he would like to identify also in
the Sogdian title γwt’yn “queen” < *xwatā̆ wyainī-. It is to be noted that Benveniste does not
234 Antonio Panaino

2) Paoiriiaēinī-, Pleiades. It has been suggested by Bartholomae24 that this

name could be connected with an Indo-European earlier stem such as *pelu̯ i-s,
corresponding to Lat. pulvis (but now it would be better to derive pulvis from
I.-E. *pe/olH-u- [> Pr.Latin *pe/olaw-, and not from *polHu- > PIE. *polu-, since, as
de Vaan fittingly remarks, “*polw- would yield Lat. *poll-”).25 Then, paoiriiaēinī-
would be a substantivized feminine plural deriving from a real Proto-Iranian
Stoffadjectiv such as *puru̯ (i)i̯aina-, and meaning “aus Staub bestehend”. New
Persian parv “Pleiads” seems to continue IE. *pe/olHu̯ (i)-, while Paštu pērūne
should derive from *parvyān.26 The Gr. stem Πλειάδες could be, in its turn, ac-
cording to Bartholomae, derived from a related stem like *pleu̯ ii̯- (*pleHu̯ ii̯-), alt-
hough this name was probably influenced through popular etymology27 by an
adjectival stem like πλέος or by the verb πλέω; its Ausgang in -αδ- should have
been made on the model of the name of another constellation like Υ ̒ άδες. On the
other hand, Frisk28 and Chantraine29 raised doubts about the existence of a com-
mon Indo-Europaean Grundform, and insisted on the possibility that the Iranian
and the Greek forms were simply derived by popular etymology. We cannot ex-
clude that this Avestan stem was actually reinterpreted as meaning the “the first
(group of stars)”, i.e. as whether it were connected with Av. paoiriia-. In fact, this
“constellation” assumed a particular importance in the observation of heaven,
and, then, it actually became the “first” in the enumeration of the lunar stations,
although this classification probably reflects a later phenomenon (reasonably of
a certain antiquity), comparable with the Indian example of Kr̥ ttikāh, the “Pleia-
des”, the first asterism among the 27 nakṣatras.

mention at all Bartholomae’s article about the Indo-European name of the Pleiades. In his turn,
Klingenschmitt (1967: 70, n. 11) just put in evidence that the Avestan suffix -aēnī- had fallen in a
“verblaßter Bedeutung wie -ānī- in ai. rājakulānī-) [...]. Hoffmann again in the same year (1967: 36 =
1976: 492-493) insisted on the fact that the suffix -aēna- “dient zwar im allgemeinen zur Bildung
von Stoffadjektiven, hat aber in einigen Bildungen eine allgemeine Bedeutung, etwa ‘von etwas
stammend, damit in Beziehung stehend [...]’, with reference, e.g., also to tištriiaēnī- (“zum Sirius
gehörend”), although in the case of paoiriiaēnī- Hoffmann (1967: 38, n. 11 = 1976: 493, n. 11) clearly
referred to Bartholomae’s article about the name of the Pleiades (1912/13). On this discussion see
also Kellens 1971: 9-10.
24
1912/13.
25
Cf. de Vaan 2008: 498.
26
Cf. Scherer 1953: 142-143.
27
See Scherer 1953: 143.
28
1970, II: 555.
29
2009: 913.
 The Classification of Astral Bodies 235

3) The circumpolar stars, as we have seen, were well known: they were: Hap-
tōiriṇga- and Hapta.srū-, the two Bears,30 which probably reflect in their denomi-
nations an inherited tradition, in which their number was taken as essential for
their correct identification, as it happens still today, when we observe the
heaven. See, in Latin, Septentriōnēs (“Ursa Maior”), Septentriōnēs minores or Sep-
tentriō minor (“Ursa Minor”), bis septem triōnes (Mart. Cap. VIII, 808, 5), septen-
triōnēs maiores et minores (Varro, apud Gell. VIII, N.A., III, 10, 2). Cf. also Skt.
Sapta R̥ ṣih.
On the other hand, we have no reason to consider Av. mərəzu-, m. “vertebra”,
as a star, although Henning31 suggested its identification with α Ursae Minoris,
i.e. with the (supposed) “Pole Star”. The serious (and unconsidered) problem, in
fact, is that no star was actually “Polar” since 2800 BC and before the 8th c. AD,
and accordingly also such a concept was relatively modern, while in antiquity
the image and role of a “Pole star” is not per se really attested. More reasonably,
mərəzu- was just a “peg” in the centre of the heaven, around which the two con-
stellations of seven stars each turned.32 Only in Sasanian times33 it was associated
with Mex ī Gāh, which corresponded to the “Pole Star”.
This overview of the data shows that the Old Iranian astral system was quite
archaic with regard to the knowledge and denomination34 of the main stars or
stars groups. More complex is the representation of the heaven, which is strictly
connected with the levels in which the stars were thought to stay. Although we
can find a very simple and primitive distinction between heaven and hearth,
very common in Old Persian, Avestan sources insist also on the existence of
three different heavens,35 the lowest one of the stars, the middle one of the
Moon, Māh-, the third one, of the Sun, Huuar-/Xvar-. On the top, there is the Par-
adise of Ahura Mazdā; this level was that of the anaγra raocā̊ , i.e. of the “lights
without beginning”.36 A threefold subdivision of the heaven is attested also in Ve-
dic sources, but in this case the order of the astral bodies is different. Contrari-
wise, the location of the stars in the lowest heaven is attested in some “mystical”

30
Panaino 1995/96.
31
1942 (= 1977).
32
Panaino 1995/96.
33
Cf. Brunner 1987.
34
See Duchesne-Guillemin 1986.
35
Panaino 1995b.
36
This is the literal translation; normally these stars are named “endless lights”.
236 Antonio Panaino

texts of Babylonian origin, belonging to the beginning of the 1st millennium, in

which three are the heavens, each one made of a different precious stone. The
peculiar order, star, Moon and Sun was shared also by some Pre-Socratic Greek
philosophers. In Iran, the essential idea was that light was increasing during the
ascent towards Paradise. This systematization, then, has an inner functional
meaning full of implications, which are visible in the after life ascent of the indi-
vidual soul. It is possible that a Mesopotamian model played a certain influence,
in particular in the location of the stars in the lowest heaven, and that from the
Iranian area this system was adopted by Greek scholars, although with new in-
terpretations. The Vedic witness, on the other hand, may be interpreted as a
simple resonance, but independent, while in this case the Iranian reorganization
will be an intermediate phenomenon, partly influenced by the Western (Meso-
potamian) traditions, partly perhaps inherited or independently developed. In
any case, the anaγra raocā̊ correspond to another kind of stars, although this de-
nomination is never used strictly for them. In the Pahlavi sources, these lights
will be considered as corresponding to those of the extragalactic sphere, or with
the so-called “unmixable” stars, which closed the hole produced in the heaven by
Ahreman himself at the moment of his first attack against the Good Creation.
This later systematization probably reflects an earlier idea, which considered the
(apparently) smaller heavenly lights, as for instance those belonging to the Milky
Way, as a sort of celestial belt or river, connected with the goddess Anāhīd (Av.
Anāhitā),37 but also located in a different level, closer to the Paradise. In the Bun-
dahišn, this distinction will be systematized in that between the spihr ī gumēziš-
nīg and the spihr ī agumēzišnīg or between the “sphere of mixture”, including the
12 constellations (Pahl. 12-axtarān) involved in the “mixture” with the demoniac
and evil forces (see Ir. Bd., 2, 8-9),38 and the “unmixable sphere” to which the
Milky Way was connected.
The lack of any denomination for the planets shows, as I noted before, the an-
tiquity of the Iranian systematization of the heavenly bodies. Apart from the
stars, the focus concerned the astral bodies, which were moving in a very irreg-
ular way, such as the falling stars and the bolides. These realia were classified as
pairikās, witches, i.e. as demons, and denominated stārō.kərəmā-, or “starred
worms”,39 and opposed to the (fixed) stars, whose movement was assumed to

37
Cf. Witzel 1984.
38
Henning, 1942: 232-33, 240; Belardi 1977: 125-26.
39
Panaino 2005c. Cf. also Panaino 2005a.
 The Classification of Astral Bodies 237

represent and confirm the cosmic order and its regular development. In this
framework, the stars were taken as divine beings, and some of them, like Tištriia,
transformed in astral protagonists of important myths, as that of the liberation of
the waters.40 In particular, the stars were considered as bringer of the waters and
rains, and for this reason denominated afš.ciθra-. It is difficult to understand the
reasons behind the later introduction of other two types of stars, called zəmas.
ciθra- and uruuarō.ciθra-. Apart from the difficulties concerning the interpreta-
tion of the real meaning of ciθra- in this contest, if “image”, “brightness” or “(fe-
male) semen/germen”,41 and, then, “origin”, the most serious problem lies in the
actual determination of which kind of classification was really meant. One pos-
sibility is that the earlier denomination was simply descriptive: thus, afš.ciθra-
would be the stars having the aspect of the water drops of rain, because the
movement of the stars was put in connection with the cosmic order and the arri-
val of rains (in contrast with that of the stārō.kərəmā-, the “starred-worms”, re-
sponsible of famine on the earth), then zəmas.ciθra- and uruuarō.ciθra- would be
those having the aspect of the earth and of the plants. While the first solution
seems to be semantically clear and the following interpretation quite possible,
the other two are less evident. Another possibility is that the two later categories
of stars were introduced when these stars were connected with some compo-
nents of the creation made by Ahura Mazdā, so that after the earlier pattern rep-
resented by afš.ciθra-, now reinterpreted as “having the origin/germen of the wa-
ter drops”, some stars were put in direct relation with the “earth” and with the
“plants”.
In the classification and cataloguing of the stars the idea of their size has as-
sumed a certain importance, but this distinction is surely later, and possibly it
was influenced by the Greek concept of stellar magnitudo (Pahl. wuzurgīh
[wcwlgyh]). In fact, it appears in Pahlavi texts, in particular in the Bundahišn,
although its adaptation to the Iranian culture reflects a primitive cataloguing.
This is the statement we find in the Great Bundahišn, II, 16:42 “Among these stars,
the large ones are like a piece of rock the size of a room, the medium-sized ones
are like a rolling wheel, the smallest ones like the head of the domesticated ox.
The moon is the size of a racecourse of two hāsars (= Av. hāθras), each geo-
graphical hāsar being about as much as a parasang of average length. The sun is

40
See Panaino 1990a; 1995a.
41
Panaino 2009a. See also Panaino 2005b.
42
Cf. Pakzad 2005: 40-41; here the chapter is edited as 2, 18 with some little differences.
238 Antonio Panaino

the size of Ērān-wēz” (az āwēšān stāragān, ān ī meh čand ī sang-ē(w) kadag-
masāy. ān ī mayānag čand *čahāragwān wafišn ud ān ī keh čand ī sar ī gāw ī ka-
dagīg. ud māh čand asprēs-ē(w) <ī> dō hāsar; čiyōn har(w) hāsar-ē(w) pad zamīg
frasang-ē(w) paymānīg homānāg, xwaršēd and-čand ī ērān-wēz).43
While the concept of “size” of the stars (expressed by the second element of
the compound -masāy)44 has to be interpreted as referred to their apparent
brightness, the antiquity of the pattern is clearly shown by an Avestan fragment
survived in the Avestan-Pahlavi Dictionary known as Frahang ī Ōīm (in a frame-
work referring to the contents of the Nīkādūm Nask):45 nitəmacit̰46 auuaēšąm stā-
rąm yaθa narš *maδəmiiehe47 vaγδanəm, translated in Pahlavi as ān-iz ī nidom az
awēšān starān čand mard ēk ī mayānag waγdān, “The smallest ones of those stars
(are) like the head48 of a middle-sized man”.
I have dealt in detail with this “categorization” in another study,49 but the
whole problem of the dimension of the astral bodies has to be just mentioned in
this context again, because it shows how new doctrines of scientific origin were
adapted to earlier traditions.
Now we must briefly mention the problem of the planetary denominations:
The names of the five visible planets are fully documented only in Middle-Ira-
nian sources, but their classification as astral bodies different from the “fixed”
stars should be surely earlier: in Pahlavi their names are: Anāhīd (Venus), Tīr
(Mercury), Wahrām (Mars), Ohrmazd (Jupiter), and Kēwān (Saturn). In the
framework of the Mazdean cosmology of the Sasanian times, the planets were

43
Translation by Henning 1942: 233-234; cf. Bailey 1971: 136-137; Anklesaria, Zand-Ākāsīh, p. 34-
35.
44
It has to be noted that Pahlavi kadag-masāy corresponds to an Av. katō.masah- (see Henning
1942: 233, n. 7; cf. Bartholomae 1904: 434, where katō°, the first member of the compound, has to be
derived from the stem kata-, m. “Kammer, Keller” [Bartholomae 1904: 432]; Duchesne-Guillemin
1936: 153, par. 193). Pahlavi kadag-masāy occurs also in Dēnkard, VII, 4, 41 (see Molé 1967: 48-49)
where it is referred to a stone that Zoroaster received from Ohrmazd. It is possible that the celestial
origin of such as stone is evoked thoughout the mention of this rare compound.
45
See also West 1892: 471-475.
46
Darmesteter (1893: III, p. 16, n. 3) reads nitəmcit̰ (thus, as acc. sg.), but he suggests a correc-
tion in nitəməcit̰.
47
MSS ma∂miiehe; correction suggested by Reichelt (1901: 176) and Klingenschmitt (1968: 81).
48
It is peculiar that here the Daēvic stem vaγ∂a-, n., is used, and not sāra-, n. (Ahuric), which
should be the expected one with reference to the stars, who are positive divine beings in the Zoro-
astrian tradition.
49
Panaino 2004c.
 The Classification of Astral Bodies 239

demonized.50 This phenomenon took place in the context of a new dualistic clas-
sification of the astral bodies, where the planets assumed the same role played in
Avestan astral mythology by the falling stars, or Pairikās, and, in fact, the planets
were exactly called, as demons, also Parigān. Furthermore, they were denomi-
nated abāxtar “retrograde” or nē axtar “not-star”, but in some cases the planets
were called also gēg “robbers, bandits” in direct opposition to the stars, consid-
ered as the “givers” (bayān) par excellence (ŠGW 4, 8-10).51 But it is the later Pah-
lavi denomination of the planets as abāxtar(ān), deriving from an Old-Iranian
stem such as *apāxtara- (“backward-turning, retrograde”), in its turn, a compara-
tive built on apāk-/apāŋk- “backward” (from the preposition apa “behind”),52
which deserves to be focused on in the present discussion. It is probable that
that such a denomination was put in connection with the Northern direction,
which, in Iran, was traditionally considered the side of the demons. In fact, an-
other meaning of *apāxtara- was actually “North”. Henning already noted that
the Greek denomination of the planets (πλανητής, “errans”, from πλανάοµαι “to
wander”, in direct opposition to ἀπλανής, i.e. inerrans) contains a patent refer-
ence to their irregular movement, by means of which these astral bodies were
classified. This observation was fittingly adopted in Pahlavi by means of a stem
such as *wiyābanīg (from the verb wiyābān-, “to deviate”),53 clearly attested in the
Middle-Persian “reversed” denomination of the stars as a-wiyābanīg, “inerrantes,
not deviating”,54 but it is hidden behind the apparently peculiar denomination of
a group of stars called beibenie in some astrological Medieval Latin sources.55
It is evident that the demonization of the planets turned out to be in evident
contradiction with the fact that some of them had been earlier given the same
names of the most important Mazdean gods.56 In fact, when the existence of the
planets was really taken into account also in the Iranian classificatory system of
the existing astral bodies, the basic Mesopotamian denominations were just fol-
lowed by Persian observers, exactly as the Greeks did it. In other words, the Mes-

50
See Panaino 2008a. Cf. also Panaino 1999c.
51
See de Menasce 1945: 50-51. Cf. now Panaino 2013.
52
Eilers 1976; 1987.
53
Cf. MacKenzie 1971: 14.
54
Henning 1942.
55
Kunitzsch 1964; 1968; 1971; 1972; 1974; 1981; 1993; Pingree 1989.
56
But we must remark that astrological doctrines produced a mass of additional contradic-
tions; see, e.g., Panaino 1995a, passim; 2004b.
240 Antonio Panaino

opotamian diviners and observers originally distinguished and denominated the

single planets, associating them with some of the highest divinities of their own
pantheon, and probably already the ancient Persians followed their example by
means of a sort of syncretistic association. Thus, the new names taken by the
planets became so popular that a systematic change of their denominations re-
sulted very difficult in later times, even when the planets were demonized.
Planet Akkadian Greek Old Persian Mid.-Persian Sogdian N. Persian
Mars Nergal Áres *Vr̥ θraγna- Wahrām Unxān Bahrām
Mercury Nabû Hermês *Tīriya- Tīr Tīr Tīr
Jupiter Marduk Zeús Ahuramazdā- Ohrmazd Urmazt Hormozd
Venus Ištar Aphrodítēs Anāhitā- Anāhīd Nāxid Anāhīd
Saturn Kajamānu Krónos *Kayvānu- Kēwān Kēwān Keyvān
(cf. Lat. Saturnus) (*Zruvan) (Zurwān)

The astral nomenclature57 in the Iranian sources shows that an earlier primi-
tive systematization and classification of the astral bodies underwent changes
and adaptations to foreign doctrines, although some dualistic patterns, very
common in the Mazdean folklore and cosmology, already working in the context
of the earlier Iranian astral mythology, imposed a re-organization of later catego-
ries. This process was, of course, full of contradictions, partly derived from the
need of finding a compromise between different astrological doctrines and an-
cestral Mazdean patterns, but the presence of all these problems just demon-
strate the complexity of the efforts developed in Pre-Islamic Iran during the
course of times in order to balance tradition and innovation.

BIBLIOGRAPHY
Bailey, H. W. (1971), Zoroastrian Problems in the Ninth-Century Books, New York.
Bartholomae, Chr. (1904), Altiranisches Wörterbuch, Strassburg.
―― (1912/13), “Der indogermanische Name der Plejaden”, Indogermanische Forschungen 31: 35-48.
Belardi, W. (1977), Studi Mithraici e Mazdei, Roma.
Benveniste, E. (1966), Titres et noms propres en Iranien ancient, (Travaux de l’Institut d’Études
Iraniennes de l’Université de Paris, 1), Paris.
Brunner, C. J. (1987), “Astrology and Astronomy in Iran. ii. Astronomy and astrology in the Sasanian
period”, Encyclopædia Iranica. Vol. II, 8, London/New York: 862-868.
Chantraine, P. (2009), Dictionnaire étymologique de la langue grecque, Histoire des mots. Nouvelle
edition, Paris.

57
See Panaino 1993.
 The Classification of Astral Bodies 241

Darmesteter, J. (1883), Études Iraniennes. Tome second. Études sur la langue, la littérature, les croy-
ances de la Perse ancienne, Paris.
―― (1893), Le Zend-Avesta, Vol. III, Paris.
Duchesne-Guillemin, J. (1936), Les Composés de l’Avesta, Liège/Paris.
―― (1986), “Origines iraniennes et babyloniennes de la nomenclature astral”, Académie des In-
scriptions et Belles Lettres, Comptes rendus des Séances (CRAI): 234-250.
Eilers, W. (1967), “Stern-Planet-Regenbogen, Zur Nomenklatur der orientalischen Himmelskunde”,
Der Orient in der Forschung. Festschrift für Otto Spies zum 5. April 1966. Herausgegeben von
W. Hoenerbach. Wiesbaden: 92-146.
―― (1976), Sinn und Herkunft der Planetennamen, München.
―― (1987), “Axtar”, Encyclopædia Iranica. Vol. 3/2, London/New York: 123-124.
Filliozat, J. (1962), “Notes d’astronomie ancienne de l’Iran et de l’Inde (I, II et III)”, JA 250: 325-350.
Fischer, H. (1969), “Griechisch Σείριος”, MSS, 26: 19-26.
Forssman, B. (1968), “Apaoša der Gegner des Tištrya”, KZ 82: 37-61.
Frisk, H. (1954-1970-1972), Griechisches Etymologisches Wörterbuch, 3 vols, Heidelberg.
Henning, W. B. (1942), “An Astronomical Chapter of the Bundahishn”, JRAS: 229-248 [= (1977) SP, II:
95-114].
―― (1977), Selected Papers I-II, (Acta Iranica, 14-15), Téhéran/Liège.
Hoffmann, K. (1967), “Drei indogermanische Tiernamen in einem Avesta-Fragment”, MSS 22: 29-28
( = Hoffmann 1976: 486-493).
―― (1976), Aufsätze zur Indoiranistik, Band 2, Herausgegeben von J. Narten. Wiesbaden.
Hunger, H. / D. Pingree (1999), Astral Sciences in Mesopotamia (Handbuch der Orientalistik. Erste
Abteilung. Der Nahe und Mittlere Osten), Leiden.
Kellens, J. (1971), “Forschungsbericht, L’avestique de 1962 à 1972”, Kratylos 16/1: 1-30.
―― (2010), “La notion d’âme préexistante, In Langues et religions indo-iraniennes”, Résumée des
Cours, Collège de France, Paris: 747-762.
Khareghat, M. P. (1914), “The identity of some heavenly bodies mentioned in the Old Iranian writ-
ings”, Sir Jamsetjee Jejeebhoy Madressa Jubilee Volume, Bombay: 116-58.
Klingenschmitt, G. (1967), “Die Erbtochter im zoroastrischen Recht nach dem Māδiyān ē hazār
dāδistān”, MSS, 21: 59-70.
―― (1968), Farhang-i ōīm, Edition und Kommentar, Erlangen (unpublished).
Kunitzsch, P. (1964), “Das Fixsternverzeichnis in der “Persischen Syntaxis” des Georgios Chryso-
kokkes”, BZ 57: 382-411.
―― (1968), “Zum liber hermetis de stellis beibeniis”, ZDMG 118: 62-74.
―― (1971), “Neues zum liber hermetis de stellis beibeniis”, ZDMG 120: 126-130.
―― (1972), “Zur Tradition der Unwettersterne”, ZDMG 122: 108-117.
―― (1974), Der Almagest. Die Syntaxis Mathematica des Claudius Ptolemäus in arabisch-
lateinischer Überlieferung, Wiesbaden.
―― (1981), “Stelle beibenie - al-kawākib al-biyābānīya”, ZDMG 131: 263-267.
―― (1993), “The Chapter of the Fixed Stars in Zarādusht’s Kitāb al-mawālīd”, Zeitschrift für Ge-
schichte der arabisch-islamischen Wissenschaften, Bd. 8: 241-249.
MacKenzie, D. N. (1964), “Zoroastrian Astrology in the Bundahišn”, BSOAS 27: 511-529.
242 Antonio Panaino

―― (1971), A Concise Pahlavi Dictionary, London.

de Menasce, J. (1945), Škand-Gumānīk vičar. La solution décisive des doutes. Texte pazand-pehlevi
traduit et commenté, Fribourg en Suisse.
Molé, M. (1967), La légénde de Zoroastre selon les textes pehlevis, Paris.
Neugebauer, O. (1957), The Exact Sciences in Antiquity, Providence (Rhode Island).
―― (1975), A History of Ancient Mathematical Astronomy, 3 voll, Berlin/Heidelberg/New York.
―― (1983), Astronomy and History, Selected Essays, Berlin.
Pakzad, F. (2005), Bundahišn: Zoroastrische Kosmogonie und Kosmologie, Band 1, Kritische Edition,
Tehran.
Panaino, A. (1986), “Vedico Tiṣya”, Incontri Linguistici 11: 71-75.
―― (1987), “Sirio stella-freccia nell'Oriente antico”, Atti della 4a giornata di Studi Camito-Semitici e
Indeuropei, A cura di G. Bernini; V. Brugnatelli, Milano: 139-155.
―― (1989), “L’inno avestico a Vanant”, Atti del Sodalizio Glottologico Milanese 28 (1987): 21-30.
―― (1990a), Tištrya, Part I, The Avestan Hymn to Sirius (Serie Orientale Roma, LXVIII, 1) Roma.
―― (1990b), “Contatti sino-iranici nella cultura astronomica dell’Iran preislamico”, Dall’Europa
alla Cina: contributi per una storia dell’Astronomia. A cura di I. Iannaccone e A. Tambu-
rello, Napoli: 41-54.
―― (1993), “Considerazioni sul lessico astronomico-astrologico medio-persiano”, Lingue e Culture
in Contatto nel Mondo Antico e Altomedievale. Atti dell’ VIII Convegno Internazionale dei Lin-
guisti, tenuto a Milano nei giorni 10-12 Settembre 1992, Testi raccolti a cura di R. B. Finazzi; P.
Tornaghi, Brescia: 417-433.
―― (1995a), Tištrya, Part II, The Iranian Myth of the star Sirius (Serie Orientale Roma, LXVIII, 2),
Roma: I-XVI; 150.
―― (1995b), “Uranographia Iranica I. The Three Heavens in the Zoroastrian Tradition and the
Mesopotamian Background”, Au carrefour des religions. Mélanges offerts à Philippe Gig-
noux, Bures-sur-Yvette: 205-225.
―― (1995/96), “Uranographia Iranica II. Avestan hapta.srū- and mərəzu- : Ursa Minor and the
North Pole?”, Archiv für Orientforschung 42-43: 190-207.
―― (1996a), “The two Astrological Reports of the Kārnāmag ī Ardašīr ī Pābagān (III, 4-7; IV, 6-7)”,
Die Sprache, 36, 2, (1994): 181-196.
―― (1996b), “Sopravvivenze del culto iranico della stella Sirio nel Kitāb al-mawalīd di Zarādušt ed
altre questioni di uranografia sasanide”, Alle soglie della Classicità. Il Mediterraneo tra
tradizione e innovazione. Studi in onore di Sabatino Moscati. Volume Primo. A cura di E.
Acquaro, Pisa/Roma: 343-354.
―― (1996c), “Saturn, the Lord of the Seventh Millennium”, In collaboration with D. Pingree, East
and West 46, Nos. 3-4: 235-250.
―― (1997), “Visione della volta celeste e Astrologia nel Manicheismo”, Atti del Terzo Congresso In-
ternazionale di Studi: «Manicheismo e Oriente Cristiano Antico». Arcavacata di Rende -
Amantea, 31 August - 5 September, A cura di L. Cirillo e A. van Tongerloo, Lovanii/Neapoli:
249-295.
―― (1998), Tessere il cielo, Considerazioni sulle Tavole astronomiche, gli Oroscopi e la Dottrina
dei Legamenti tra Induismo, Zoroastrismo e Mandeismo, Serie Orientale Roma LXXIX,
Roma.
 The Classification of Astral Bodies 243

―― (1999a), “The Cardinal Asterisms in the Sasanian Uranography”, La Science des cieux. Sages,
mages et astrologues, Contributions réunis par R. Gyselen, Res Orientales XII, Bures-sur-
Yvette: 183-190.
―― (1999b), “Le développement de l’uranographie iranienne, Conférence de M. A. Panaino, Di-
recteur d’études invité”, Annuaire EPHE, Section sciences religieuses, t. 106 (1997-1998): 211-
216.
―― (1999c), “An Avestan Planetary Order?”, Astronomical Amusements, Papers in Honor of Jean
Meeus, Milano: 183-190.
―― (1999d), “Manichaean Concepts in the Pahlavi Commentary of Māh Niyāyišn, par. 4”, Studia
Manichaica IV, Internationaler Kongreß zum Manichäismus, Berlin, 14.-18. Juli 1997, Heraus-
gegeben von R. E. Emmerick; W. Sundermann; P. Zieme, Berlin: 464-480.
―― (2004a), “Considerazioni sulla trasmissione delle scienze e delle cosiddette “pseudoscienze”
tra Grecia e Iran”, Varia Iranica, 7, (eds) C. G. Cereti; B. Melasecchi; F. Vajifdar, Roma: 213-
237.
―― (2004b), “Two Zoroastrian Nērangs and the Invocation to the Stars and the Planets”, The
Spirit of Wisdom [Mēnōg ī Xrad], Essays in Memory of Ahmad Tafazzoli, (eds) T. Daryaee;
M. Omidsalar, Costa Mesa (California): 196-218.
―― (2004c), “On the Dimension of the Astral Bodies in Zoroastrian Literature: Between Tradition
and Scientific Astronomy”, Studies in the History of the Exact Sciences in Honour of David
Pingree, (eds.) Ch. Burnett; J. P. Hogendijk; K. Plofker; M. Yano, Leiden: 267-286.
―― (2005a), “I Magi e la stella nei Sermoni di San Pier Crisologo. Qualche riflessione critica a
proposito di scienza, fede e metodo storico”, Ravenna da capitale imperiale a capitale esar-
cale. Atti del XVII Congresso internazionale di studio sull’alto medioevo, Ravenna, 6-12 giugno
2004, Spoleto: 559-592.
―― (2005b), “Pahlavi gwcyhl: gōzihr o gawčihr?”, Scritti in onore di Giovani M. D’Erme, A cura di
M. Bernardini e N. L. Tornesello, Napoli: 795-826.
―― (2005c), “Yt. 8, 8: stārō kərəmā̊ o stārō.kərəmā̊ ? Stelle infuocate o Stelle-verme?” Indogermani-
ca. Festschrift Gert Klingenschmitt. Indische, Iranische und Indogermanische Studien dem
verehrten Jubilar dargebracht zu seinem fünfundsechzigsten Geburtstag, Herausgegeben
von G. Schweiger. Taimering: 455-463 (published only in 2006).
―― (2008a), “The Zoroastrian Incestuous Unions in Christian Sources and Canonical Laws: their
(distorted) Aetiology and some other Problems”, Controverses des Chrétiens dans l’Iran sas-
sanide, (Studia Iranica, Cahier 36), Paris: 69-87.
―― (2008b), “Again about α Cygni in the Sasanian Astronomical Tradition”, Papers in Honour of
Professor B. Gharib, (eds) Z. Zarshenas; V. Naddaf, Tehran: 33-39.
―― (2009a), “Sasanian Astronomy and Astrology in the Contribution of David Pingree”, Kaid.
Studies in History of Mathematics, Astronomy and Astrology in Memory of David Pingree,
(eds) Gh. Gnoli; A. Panaino (SOR CII), Roma: 71-99.
―― (2009b), “Avestan daxšta- and ciθra-. I: The Semantic Field: Female Germen and Menstrua-
tion”, Zarathushtra entre l’Inde et l’Iran : études indo-iraniennes et indo-européennes offertes
à Jean Kellens à l'occasion de son 65e anniversaire. Èd. par É. Pirart; X. Tremblay (Beiträge
zur Iranistik. Bd. 30), Wiesbaden: 197-220.
―― (2009c), “Zodiac”, Encyclopædia Iranica (on line), Published: July 20, 2009. <http://www.
iranicaonline.org/articles/zodiac>.
244 Antonio Panaino

―― (2013), “The Gift of the Givers”, Gifts to a Magus: Indo-Iranian Studies Honoring Firoze Kotwal,
(eds.) J. K. Choksy; J. Dubeansky, New York: 137-145.
Pingree, D. (1963), “Astronomy and Astrology in India and Iran”, ISIS 54/2, 176: 229-246.
―― (1983), “Astrology and Astronomy in Iran. i. History of Astronomy in Iran, iii. Astrology in Is-
lamic Times”, Encyclopædia Iranica. II, 8, London/New York: 858-862; 868-871.
―― (1989), “Classical and Byzantine Astrology in Sassanian Persia”, DOP 43: 227-239.
―― (1992), “Hellenophilia versus History of Science”, ISIS, 83: 554-563.
Raffaelli, E. G. (1998), “The Diadrams of the zāyč ī gēhān”, EW, 49, 1-4: 285-291.
―― (2001), L’Oroscopo del mondo. Il tema di nascita del mondo e del primo uomo secondo l'astrolo-
gia zoroastriana, Milano.
Reichelt, H. (1900), “Der Frahang i oīm”, WZKM 14: 177-213.
―― (1901), Der Frahang i oīm, WZKM 151: 117-186.
Scherer, A. (1953), Gestirnnamen bei den indogermanischen Völkern, Heidelberg.
Taqizadeh, S. H. (1937), Gāh šomārī dar Īrān-e qadīm, Teheran.
―― (2010), Il computo del tempo nell’Iran antico, Edizione riveduta e integrata sulla base delle in-
dicazioni dell’autore, traduzione e cura di S. Cristoforetti. (Il Nuovo Ramusio, Strumenti,
2), Roma.
Taraf, Z. (1981), Der Awesta-Text Niyāyiš mit Pahlavi- und Sanskritübersetzung, (Münchener Studien
zur Sprachwissenschaft, Beiheft 10. Neue Folge), München.
de Vaan, M. (2008), Etymological Dictionary of Latin and the other Italic Languages, Leiden.
West, E. S. (1892), Pahlavi Texts, Part IV, SBE 24, Oxford.
Witzel, M. (1984), “Sur le chemin du ciel”, Bulletin d’Etudes Indiennes 2: 213-279.
Zaehner, R. C. (1955), Zurvan, a Zoroastrian Dilemma, Oxford (reprint New York 1972).
 From Muscat to Sarhadd:
Remarks on gwātī Healing Ritual within the Social Context*

Vahe S. Boyajian
Department of Social Anthropology, University of St. Andrews
Institute of Archaeology and Ethnography, Yerevan

Abstract
This paper focuses on the gwātī healing ritual practiced in the Sarhadd region of Iranian Baluchi-
stan. Though the gwātī ritual is widespread all over Baluchistan (both in Iranian and Pakistani
sides) and its central pattern is relatively common in different locations, however the homegrown
varieties of the ritual are quite interesting from the anthropological perspective. The data for this
article has been collected in the Sarhadd region of Iranian Baluchistan during the ethnographic
fieldwork in 2012.

Keywords
Gwātī, Baluchistan, Sarhadd, Baluch, Ritual, Tribal Identity

An attempt is made to cognize the changes the gwātī ritual has undergone, and
the social conditions that caused its adaptation to the Sarhaddi1 realities, thus
skipping the plain description of the ritual itself. The aim of the article, in fact, is
to portray the social niche, which is occupied by the phenomenon of the gwāt
and gwātī, and all the other “acting” characters within the ritual. My encounters
with the individuals associated with the ritual not only in Sarhadd but also in the
southern parts of Iranian Baluchistan were limited by the investigation of the so-
cial perceptions of the people about the ritual and its practical implications in
the daily life, hence the nuances of the ritual itself are beyond the scope of this
paper.

*
The preliminary version of this paper was presented at the international Workshop “Baluchi
Identity and Culture”, held at Brock University, St. Catharines, Canada, 8-9 September, 2012.
1
For the geographical and political peculiarities of the Sarhadd region, see Bosworth 2002: 79-
102.
 Koninklijke Brill NV, Leiden, 2015
246 Vahe S. Boyajian

According to the local narratives the gwātī ritual, originating in Oman (Mus-
cat), has entered the southern Makkuran (Makran) region of Baluchistan with
the migration waves of slaves brought by the Arab conquerors from Africa (Zan-
zibar, Tanzania,) through the Persian Gulf.2 It had long been considered within
the domain of pagan practices; still later, it was adapted to the everyday life of
the indigenous population of Baluchistan professing Islam. As the locals of Sar-
hadd believe, “it [the penetration of the ritual into the Sarhadd region] has been
inevitable, as the gwāt (Balochi “wind”) knows no limits and boundaries”.
The narrative of the origins tells that, at the very beginning of Islam there was
a Jewish woman in Muscat, allegedly in Arabia, and there was a wandering dar-
vish. The darvish used to sit under the dome and pray to god, what annoyed the
Jewish woman. So, she gathered children of the neighborhood and sent them to
destroy the beehives, bring all the bees and hide them inside the dome, so that
the bees could harm the darvish, and he would go away. It was wintertime. The
bees were frozen, so the children fulfilled the task easily. When the spring came,
the bees attacked the darvish and bit him. His body became swollen, as if there
was a wind inside. The darvish complained to God and asked him to send a kind
of disease to all the women that could be cured only with the help of music per-
formed by a string instrument and accompanied by songs. Since then the disease
is called the gwātī-e parī—‘the wind of the fairy’.
The gwātī in Sarhadd is a relatively new phenomenon; its appearance in the
region goes back to the 60s of the previous century. Those who have ever en-
countered with the ritual, assert that it has been always practiced “in harmony
with the Islamic traditions and faith”.3 During my several visits to the region and
long-time fieldwork in Sarhadd, I have met only one well-known sheykh,4 who is
said to have inherited the healing abilities and possesses a shajara (genealogical
tree) that goes back to one of the most prominent Sufis, the founder of the Qa-
diriyya order, Abd al-Qadir Gilani.

THE HEALERS
Two individuals in Sarhadd are well-known among the population as healers of
the obsession called gwāt, and they both have been my main informants. Both

2
See with full bibliography Sultana 1996: 28-50.
3
On the religious practices in Sarhadd, see Boyajian-Sureniants 2004: 199-213.
4
The names of the informants are either omitted or changed due to the ethical and security is-
sues.
 Remarks on gwātī Healing Ritual within the Social Context 247

can rather be described as local pundits, representatives of Iranian Baluch intel-

ligentsia, or, as they call themselves—Baluch cultural heritage preservers. Here-
after they will be referred as Waja5 A and Waja Z.
In Sarhadd, according to them, there is only one person who is able to seri-
ously deal (the outcome of the healing ritual is not discussed yet) with the gwātī,
a person possessed with the gwāt. They were asked this same question sepa-
rately during friendly meetings at their houses. Each of them modestly bowed
his head and hid his eyes from my curious glance, leaving no other choice for me
but to unconditionally accept that he is the one. If there is a need of a proof, then
no worries—dozens of the diseased gwātīs came every week for treatment from
almost all over Baluchistan province and beyond its borders. Meantime one of
the sheykhs, as they are called, admitted that currently there were two
Masters—Pakir Khodabakhsh in Southern Baluchistan (precisely in Bampur, a
small town near the city of Iranshahr in Iranian Baluchistan) and Pakir Nur-
bakhsh in Karachi, Pakistan.

THE CASE OF WAJA A

Waja A was a very busy man and, I am sure still is, considering the multifaceted
nature of his activities. I met him thrice at his home and every time our meeting
has been interrupted either with an unexpected phone call, which later resulted
in his immediate leaving, or he was in a hurry to complete an indispensable task.
All the promises that he would invite me to be present at a healing ritual were
not realized unfortunately. Instead, once he basically described the healing pro-
cess and performed the ritual music. He said:
“The human being has 366 veins and there are 366 different kinds of wind—gwāt.
The numbers’ correspondence is not accidental. There is something sacral in it.
The pari, in fact the Master of the winds, does not penetrate into a person’s body
itself; it’s only the wind that, by the order of the pari, goes inside the vein and the
person becomes obsessed – he starts to have hallucinations, ghosts appear in
front of his eyes in the shape of cat, sheep, hen, etc. When being asked what they
want (the obsessed person talks to those ghosts), the answer is: blood, blood, and
music, and songs”. (see also Darvīšī 1999; Riyāhī 1977).
Waja A was a sheykh, whose ancestry, according to his genealogical tree, goes
back to Abd al-Qader Gilani, the founder of the Qadiriyya Sufi brotherhood. In

5
Wāja in Balochi means ‘master’, a formal respectful way to refer to male individuals, and is
derived from the New Persian word khvāja.
248 Vahe S. Boyajian

this case, being a sheykh is defined both by the ancestry and by the supposed
possession of baraka (the divine gift to perform miraculous acts) for healing the
diseased people. His status as a sheykh cannot be disputed, as it is inherited and
is going to be transferred to one of his progenies. An interesting fact is that not
all the male descendants can inherit both the status and the baraka. Only one of
his sons (25 years old) demonstrates some “abilities” to become a healer. A cru-
cial condition for becoming a healer is that the healer has to be affected by the
gwāt himself. There are numerous stories how he first felt the ability to heal.
The statement about the terms and conditions for performing the ritual be-
ing the main motivation to heal the diseased person, evoke, of course, doubts
about the genuineness and sincerity of the healer.
Before visiting the sheykh, the obsessed or diseased person has to be dressed
fully in new garments; hands and feet in henna, fragranced. Unconditional belief
in the success of the treatment (which can last from 3 to 5, sometimes 7 nights in
a row) is a necessity. Later things become more specified and itemized—the dis-
eased has to bring one sheep, one goat, 7 kilos of rice (no maximum limits), and
from 200,000 up to 500,000 thousand toumans (app. 200-500 USD). The sheep
and the goat should be sacrificed on the first night of the ritual, and the diseased
person or his family should distribute the mutton among people, giving the pri-
ority to the neediest ones. From the money the sheykh buys 2 sheep, one for
himself, and the other for his children, but not for sacrifice—it is a contribution
to his flocks.6
If the diseased is a female, she should be accompanied by her husband, father
or brother. The ceremony of the first night starts only after 9 pm. The space is di-
vided into two parts by a curtain and the obsessed person stays behind the cur-
tain. The sheykh with his disciples—in the case of Waja A his younger son—
stays on the other side of the curtain and starts to play the soruz and the setar.7
In the case of a male or an old female diseased there is no need of a curtain. Un-
doubtedly this has to do with the notion of harām and halāl, and, as Waja A ex-
plained: “we are true Muslims and we obey the shari‘a”. The sheykh recites ex-
cerpts from the Holy Qur’an (what, he said, originally was not done among the
coastal Baluches in Makkuran) and systematically blows on the face of the ob-
sessed until the wind of the pari—gwāt— abandons the body of the person.

6
In the social context of Sarhadd where the unemployment has a quite high level, the perfor-
mance of the healing ritual serves as a stable source of income for the sheykh and his family.
7
Local Baluchi string musical instruments.
 Remarks on gwātī Healing Ritual within the Social Context 249

Waja A is a respected citizen, beloved cultural performer, in a way the pride

of the Baluches of Sarhadd. His activities as a sheykh and a gwātī (gwātī has be-
come his lable), though, are not that crucial for the maintenance of his social sta-
tus. Of course, it does not mean that a person who is known as sheykh, who
claims ancestry of a Sufi Master, should be restricted with boundaries of piety.
All that matters here, is his status within the society and his efforts to maintain
and exercise it, and the higher the better. The status of a sheykh undoubtedly
gives him credit in the eyes of the laymen coming for treatment from remote lo-
calities (however, there existed an opinion that people in Khash—the central
town of Sarhadd—hardly go for treatment to Waja A).
One day we had an appointment to meet, so I could be present at the gwātī
ceremony; but when I arrived at his place, he had already cancelled the ap-
pointment with the diseased and was in a rush to reach the ceremony organized
by the Islamic Revolutionary Guards Corps unit in the province, where Waja A
was going to be honored by an AK 47 as a gift for his devotional service in the
path of the unity of the Muslim people in the province.8

THE CASE OF WAJA Z

The case of Waja Z, who “provides treatment” for the gwātīs, is interesting from
another point of view. He considers himself a true follower of the Qadiriyya Sufi
brotherhood, which has not actively functioned in Sarhadd, as well as in Baluchi-
stan (in terms of organizing the khaneqahs and performing samā‘ ritual), since
the dawn of the Islamic Revolution in Iran.
Waja Z describing the ritual of healing, avoided using the word gwāt or gwātī,
although the symptoms of the obsessed people visiting Waja A and Waja Z are
entirely identical. Waja Z is believed to be specialized in curing the so-called
jennzade—literary ‘stricken by the jinn’. He specified that the terms gwāt, gwātī
and zār are used in Makkuran, Southern Baluchistan, and are alien to the Sar-
hadd region.
Quite a number of my informants in Khash considered Waja Z a lunatic,
many others referred to him as an extraordinary personality with higher educa-

8
The notion of Islamic unity (vahdat-e eslāmī) is a recent socio-political phenomenon in the
province of Sistan and Baluchistan in Iran. The idea has been launched by the central government
to bring the efforts of all the inhabitants of the province (the Sunni Baluches and the Persian
speaking Shiites) into the path of the religious cooperation and unity for the prosperous future of
the province.
250 Vahe S. Boyajian

tion. Waja Z himself was very proud to show me 2 pages of rather timeworn
copies from somewhat an unidentified periodical in English, which once had
published a note about him and his healing abilities. He assured that “many for-
eign, European specialists have shown interest in his activities, which proves
that the healing has nothing to do with supernatural phenomena… Yes, the way I
perform the ritual and the excerpts I use from the Qur’an are just auxiliary
‘tricks’ to make the ‘clients’ confident in their treatment”. Obviously, he was ra-
ther honest to tell that a small prayer written on a piece of paper with a calli-
graphic handwriting was not a cure itself, but it was necessary for the obsessed
ones to be sure their healing is in the hands of God. His main ‘identifier’ of the
type of the jinn inside the body of the diseased was a glass of ordinary water. By
drinking it and describing the taste, the ‘clients’ helped him to identify the jinns
and write the respective prayer and make a talisman, or perform the same prayer
orally.
As in the case with Waja A, Waja Z alike is entirely engaged in social, and
even political activities: once he has even run as an MP candidate for the Iranian
Parliament. These activities seem to secure a certain status of an honorable per-
son in the society, and the healing ritual occupies a central position here.

RITUAL AS AN IDENTITY MARKER

The socio-political context of Sarhadd, where everybody is in competition over
gaining a higher and a better social status provides certain hints to realize the
motivation of practicing this ritual. This context is mostly dependent on the
manifestations of the Baluchi, or rather Sarhaddi identity, which includes both
the traditional dimensions of their identity—tribal affiliation, segmentary line-
age system, kin relations, code of honor, language, and—a relatively recent
one—being the Sunni Muslims as opposed to the Persian speaking Shiites. Par-
ticularly in Sarhadd, where the Baluchi tribes (mainly the Shahnawazi,
Gamshadzehi, Kord, Mirbaluchzehi, Ismailzehi) position themselves as the true
heirs of the Baluchi culture and traditions, the boundaries of self-perceptions
and identity on several levels (group, ethnic, religious, etc.) are extremely flexi-
ble. They do separate themselves from the southern and northern parts of the
province, in certain cases even treat the southern Makkurani Baluches as non-
real.
In addition, two distinct models of social organization, such as the traditional
tribal system based on the segmentary lineage, the disappearing feudalistic
 Remarks on gwātī Healing Ritual within the Social Context 251

model (for instance, small town of Sangan with its vicinities under the rule of the
Kord tribe)9 with its slightly modified variants depending on geographical, socio-
economic and environmental conditions, and a relatively modern system with
urbanized population, contribute to the complexity of the current situation in
Sarhadd. On this very subject Salzman (2000:353) writes that the organizational
multiplicity was noteworthy as well in the political system of the Sarhaddi tribes:
“The Sarhaddi Baluch organized social control, coercion, leadership and peace-
making by ingeniously combining two diverse structures: the egalitarian, decen-
tralized, democratic, segmentary lineage system, on the one hand, and, on the
other hand, the hierarchical, centralizing system of political seniority and chief-
ship”.
Considering the abovementioned, nowadays with almost entire encapsula-
tion of the Baluchi tribes into the Iranian society, one can assume that alongside
the traditional markers of the Sarhaddi identity, new ones are being exercised
that are relatively “harmless” and not that sensitive in the eyes of the central
government of the Islamic Republic.
Those affiliated with the gwātī ritual are enormously proud of the fact that its
certain constituents (the music, instruments, and songs in Balochi) are being
presented as genuinely Baluchi. Adding several Islamic elements (mentioning
the names of Allah, the Prophet, etc.) to the ritual, they stress their adherence to
the Sunni Islam, yet keeping the ritual in the frames of the cultural heritage.
From the other hand, exercising the ritual, both Waja A and Waja Z create a
circle of socio-economic relations, based on mutual trust and loyalty—a vital
condition for strengthening their position in the fading tribal society in Sarhadd,
which is supposed to guide them towards the life of a ‘good Muslim’.

9
For the existence of a feudalistic model of political and socio-economic relations in Sarhadd,
see Bestor 1979.
252 Vahe S. Boyajian

BIBLIOGRAPHY
Bestor, J. F. (1979), The Kurds of Iranian Baluchistan: A Regional Elite, MA Thesis, McGill University,
Montreal.
Bosworth, E.C. (2002), “The Sarhadd Region of Persian Baluchistan: From Mediaeval Islamic Times
to the Mid-Twentieth Century”, Studia Iranica 31:79-102.
Boyajian-Sureniants, V. (2004), “Notes on the Religious Landscape of Iranian Baluchistan:
Observations from the Sarhadd Region”, Iran and the Caucasus, vol. 8.2:199-213.
Darvīšī, M. (1378/1999), Δekrhā-ye marāsem-e Gvātī-ye Balūčestān, Tehran.
Riyāhī, A. (1356/1977), Zār o bād o Balūč, Tehran.
Salzman, P. C. (2000), Black Tents of Baluchistan, Washington/London.
Sultana, F. (1996), “Gwāt and Gwāt-i-leb: Spirit healing and Social Change in Makran”, Marginality
and Modernity: Ethnicity and Change in Post-Colonial Baluchistan, P. Titus (ed.), Oxford
University Press, Karachi:28-50.
 Georgische Gewächse auf türkischer Erde:
Ein Beitrag zur Phytonomie in Nordostanatolien*

Uwe Bläsing
Universität Leiden

Abstract
The current paper is a contribution to the research of Georgian-Turkish language interaction. It is
especially focused on plant names used in north-east Anatolia, or more precisely speaking, in the
historic province Tao-Klardjeti of Georgia, now belonging to Turkey. The phytonyms in question
make part of the Georgian substate still present in this regions Turkish dialects. In particular two
terms will be discussed: A) Turkish abet otu, representing Georgian abed-i, the name of the tree-
fungus tinder and B) Turkish köndar ~ kondar from Georgian kondar-i ‘thyme’.

Keywords
Turkish, Georgian, Tao-Klarjeti, Plant Names, Phytonymy, Tinder, Thyme, Etymology, Language In-
teraction, Dialectology, Ethnobotany

Wie der Titel bereits erahnen läßt, geht es in den nachfolgenden Ausführungen
um Pflanzen, jedoch nicht unter echt botanischem Aspekt sondern viel mehr um
ihre Namen und deren Etymologie in einem breiteren Umfeld. Ausgangspunkt
hierfür sind Phytonyme im Türkeitürkischen mit einem unmittelbar georgischen
Hintergrund.
Bevor ich mich diesen direkt zuwende, seien einige Anmerkungen allgemei-
ner Natur vorausgeschickt. Georgische lexikale Elemente im Türkeitürkischen
beschränken sich weitestgehend auf die dialektale bzw. mundartliche Ebene.
Mit anderen Worten sie haben sich in solchen Gebieten etabliert, in denen geor-
gische und türkische Volksgruppen in unmittelbarem Kontakt miteinander leb-
ten und noch leben. Ausgehend hiervon muß unsere Reise unweigerlich in den
pontisch-anatolischen Raum führen. Denn hauptsächlich im äußersten Nordos-
ten, wo die heutige Türkei an Georgien grenzt, d. h. in einem mehr oder weniger

*
Diesen Beitrag habe ich in etwas kürzerer Form im Sommer 2005 bei der Deutschen Turkolo-
gen-Konferenz in Frankfurt präsentiert.
 Koninklijke Brill NV, Leiden, 2015
254 Uwe Bläsing

breiten Streifen beginnend etwa auf der Höhe Borçka/Murgul bis nach Artvin,
Yusufeli und Şavşat (s. EGRT 1989: 421f & EGRT 2002: 146), ist ein Teil der Bevöl-
kerung georgisch—und das seit alters her. Historisch nämlich gehörte diese Re-
gion lange Zeit zu Georgien (zemo kartli). Erst mit dem stetig zunehmenden
Machteinfluß der Osmanen auf diese Gebiete (Tao-Klardschetien, Schawsche-
tien) kam das Türkische hier überhaupt als sprachliche Komponente hinzu. Mit
dem Zuzug türkischer eth-
nischer Elemente und na-
türlich aufgrund der osma-
nisch-türkischen Verwal-
tung gewann das Türki-
sche gegenüber den loka-
len Sprachen, wozu neben
dem Georgischen vor al-
Abb. 1. Armenisch-Georgisch-Türkisches Grenzgebiet
lem das Armenische ge-
hörte, mehr und mehr an Boden. Dennoch ging die Türkisierung in diesem lange
Zeit zwischen Persien, dem Osmanischen Reich und auch Russland umkämpften
Grenzgebiet im Vergleich zu anderen Regionen eher schleppend vor sich. Das
Georgische ist daher in Resten bis dato präsent und auch in den lokalen türki-
schen Idiomen spiegelt sich noch eine gehörige Portion georgischer Einflüsse
wider, allem voran im Wortschatz. Betrachtet man diesen Wortschatz unter se-
mantischen Aspekten, fällt auf, daß er sich vor allem auf Themenbereiche er-
streckt, die ganz allgemein gesprochen dem täglichen Leben und der materiellen
Kultur zuzurechnen sind; hier einige Beispiele:1
ASTAM (Şavşat, Ardanuç) ‘eiserner Teigschaber; Glutschippe’ sowie AS-
DAM ~ ASTAN (Şavşat—Artvin) und OSTAM (Erzincan) ‘Glutschippe’ (DS 347b,
341b, 3291b; İlker 1989: 251; id. 1992: 243; Pehlivan 1993: 241b; Özkan 1994: 100a) <
grg. asṭam-i ‘Schab-, Kratzeisen, Spachtel des Bäckers’ etc. (KEGL 1: 620;
Tschenkéli 39b); für Einzelheiten s. die ausführliche Studie Bläsing 2010;
BAGA (Şavşat, Yusufeli, Ardanuç—Artvin; Kars), BAĞA (Artvin; Göle, Posof,
Ardahan—Kars), BAGE (Maraş), BEGE (Çıldır), BEYE (Sarıkamış—Kars) ‘Futter-
krippe’ (DS 471b; Pehlivan 1993: 243a; Özkan 1994: 101a; Polat 2000: 152) < grg. bag-

1
Die im Anschluß in sehr knapper Form aufgelisteten Wörter werde ich—soweit nicht schon
geschehen—zu einem späteren Zeitpunkt im Rahmen gesonderter Studie ausführlich erörtern;
dasgleiche gilt für die am Ende des Beitrages angeführten Phytonyme aus der Vogelschau.
 Georgische Gewächse auf türkischer Erde 255

a ‘Krippe, Futtertrog’ (KEGL 1: 942; Tschenkéli 57b); für detailierte Angaben s. Blä-
sing 2009: 18-23;
BUÇULA ~ BUCULA, VUÇULA (Şavşat, Ardanuç—Artvin) ‘kleine Wasser-
mühle’ (DS 777a; Tokdemir 1993: 668b; Pehlivan 1993: 246b; Özkan 1994: 103a) <
grg. buč̣ ula (~ burč̣ ula) ‘kleine Mühle’ (KEGL 1: 1187, 1198; Tschenkéli 125b, 122b);
GODOR (Ardanuç—Artvin) ‘große Korbart’ (DS 2092b) < grg. godor-i ‘großer
Korb’ (KEGL 2: 1468; Tschenkéli 200a);
ḪARO (Aşağı Maden, Yusufeli, Şavşat, Ardanuç—Artvin; Posof—Kars; Trab-
zon) ‘Mehlkiste, Abteilungen in der Kornkammer’ (İlker 1992: 168a, 212a, 239b; DS
2293b; Pehlivan 1993: 269a; Özkan 1994: 115a) < grg. xaro ‘tiefe Grube; Kornkasten;
abgeteilter Raum in einer Kornkammer’ (KEGL 8: 1347; Tschenkéli 2303a);
KİBE (Ardanuç—Artvin) ‘Leiter, Treppe’ sowie die türkische -lik-Ableitung
KİBELİK ‘Holz, das zum Bau einer Leiter, Treppe geeignet bzw. bestimmt ist’ (DS
2869a; Pehlivan 1993: 285a; Özkan 1994: 123b) < grg. ḳibe ‘Treppe, Steige, Leiter’
(KEGL 4: 1194; Tschenkéli 586a);
PORÇḪİ (Şavşat—Artvin) ‘Reisigbesen’ (DS 3469b) < grg. porčx-i ‘Rechen,
Harke’; (imeruli, guruli) ‘Stecken (für rankende Pflanzen)’; (guruli) ‘verzweigter
Ast, gestutzt und ohne Laub’ bzw. grg. po(r)cx-i ‘Rechen, Harke’ (KEGL 7: 154;
Tschenkéli 1490b);
SAHREÇ (Şavşat—Artvin) ‘eine Art Sieb zum Entkernen von Pflaumen, das
bei der Herstellung von Mus verwendet wird’ (DS 3516b) < grg. sa-xleč-i ‘ein Gerät
zum Spalten, Teilen’; formal ist dies eine sa-Ableitung (“Partizip Futur Passiv”;
Fähnrich 1986: 69f) von xleč- ‘spalten, abreisen’; vgl. dazu a. sa-xleč-i aṭam-i ‘Pfir-
sichart mit leichtablösendem Stein’ (KEGL 6: 969; id. 8: 1464; Tschenkéli 1203a,
2368);
SARTUMAL (Şavşat—Artvin), SARTUMEL (Ardanuç—Artvin) ‘ein Holz bzw.
ein mit Heu gestopftes, hartes Kissen, das man zur Erhöhung des Kissens am
Kopfende des Bettes unterlegt’ (DS 3549a; Pehlivan 1993: 305b; Özkan 1994: 131b)
< grg. sa-rtum-al-i (~ sa-stum-al-i, sa-stun-al-i) ‘Kopfkissen, -polster; Kopfende’
(KEGL 6: 784; 802; Tschenkéli 1160b, 1164b).
Ein nicht unerheblicher Teil dieses Wortschatzes bezieht sich auch auf die
Natur und ganz besonders auf Flora und Fauna. Gerade erstere ist in diesem Ge-
biet sowohl vom Klima als auch von vielen anderen Faktoren überaus begüns-
tigt, was sich in einem imensen und vielfältigen Artenreichtum widerspiegelt.
Eine ganze Reihe Phytonyme, vor allem solche für Nutzpflanzen oder—allge-
meiner ausgedrückt—für Gewächse, die in irgendeiner Form eine Rolle für die
Menschen dort spielen, sodaß diese sie mit einem Namen bedacht haben, sind
Übernahmen aus dem Georgischen, was jedoch nicht vorausetzt, daß die diesen
256 Uwe Bläsing

Namen zugrunde liegenden Etyma letztlich ur-georgischer bzw. kartwelischer

Provenienz sein müssen. Ähnliches gilt für die Pflanzen selbst, auch ihr allge-
meines Verbreitungsgebiet beschränkt sich keineswegs allein auf Georgien.
Dennoch kann man sagen, daß es sich meist um in diesem Florengebiet häufig
vorkommende bzw. für dieses typische Gewächse handelt.
Hinsichtlich der Form, können sich die Georgismen oft in einem vom georgi-
schen Standard mehr oder weniger abweichenden Gewand präsentieren. Ursa-
che hierfür ist einerseits, daß wir es nicht mit Übernahmen aus der georgischen
Literatursprache sondern aus lokalen Dialekten tun haben. Andererseits muß
man noch dem Umstand Rechnung tragen, daß georgische Lexeme ihrer lautli-
chen Struktur nach oft nicht mit der des Türkischen kompatibel sind.
Aus der großen Anzahl der von mir bis jetzt gesammelten und als Georgis-
men festgestellten Pfanzennamen möchte ich hier—quasi exemplarisch—eine
ganz kleine Auswahl vorstellen und erläutern.

ABET OTU (Yukarı Maden—Artvin).2 İlker (1989: 342) beschreibt die so bezeich-
nete Pflanze als ein im Wald einzeln aus dem Boden hervorkommendes Ge-
wächs (ot), dessen Blätter (yapraklar) sich durch eine graufarbene, weiche Au-
ßenhaut (kabuk) auszeichnen, welche man (mit den Fingern) abzieht, trocknen
läßt und schließlich als Zündstoff—genannt ABETA—beim Feuerentfachen mit
einem Feuerstein verwendet.
Die genannten Formen—abet und abeta—repräsentieren ohne jeden Zwei-
fel grg. abed-i bzw. abeda ‘Zunderpils, Feuerschwamm; Zunder’ (Tschenkéli 1b;
KEGL 1: 58), wozu sich als mehr oder weniger lautgesetzliche Entsprechungen
noch laz., ming. obed-i (Kipšidze 1914: 292b; Kadshaia & Fähnrich 2001: 234a)3 so-
wie svan. habed- ~ haböd-, hobed- id. (Nižaradze 1910: 467; Wardrop 1911: 630)
stellen lassen. Aus diesem Formenset hat Klimov (1964: 43) die Grundform krtw.
*abed- rekonstruiert.4 Im Altgeorgischen ist abed- scheinbar noch nicht belegt,

2
Aşağı und Yukarı Maden sind zwei winzige Ortschaften, die ganz im Südzipfel des Verwal-
tungsgebietes (İlçe) Artvin (im Bucak Zeytinlik) unmittelbar an der Grenze zum Verwaltungsge-
biet Yusufeli liegen.
3
Erten (2000: 243) überträgt laz. obed-i mit ttü. kütük ‘Baumstumpf’, was bekanntlich einer der
Orte ist, an denen dieser Pilz gewöhnlich wächst. Nach BLS (2007: 646b) bezeichnet obed-i darüber
hinaus ‘Verdickungen an Ästen etc.’ und ‘kleine, fleischige Geschwulste am Körper’ sowie ‘die nach
der Geburt auf dem Jungtier zurückgebliebenen Fruchtwasserreste’ (letzteres semantisch unklar).
4
Die Unregelmäßigkeiten finden sich im Vokalismus der zweiten Silbe bei den zanischen For-
men, die e anstatt a zeigen; zu krtw. *e > grg., svan. e vs. laz., ming. a s. Schmidt 1962: 31. Die Form
 Georgische Gewächse auf türkischer Erde 257

wohl aber ab dem 17./18. Jahrhundert bei Sulchan Saba Orbeliani (I: 37a; es ars
nivti msc̣ rapl cecxlis šemc̣ q̇ narebeli ‘dies ist etwas, das ganz schnell Feuer fängt)5
und später bei Čubinašvili (1840: 2b; “трутъ, amadou”) etc. Beide Autoren weisen
auch gleich auf das nahezu gleichlautende arm. apet‘ ‘Zunder’ (Malxaseanc‘ I: 1c;
ARS 1984: 1a) hin,6 welches im 14. Jh. bei Grigor Tat‘ewac‘i, vielleicht aber auch
schon ab dem 10./11. Jh. bei Grigor Narekac‘i7 vorkommt und das auch aus einer
Reihe von Dialekten bekannt ist: arm. (Hačin) ab‘b‘et‘, (Xarberd) ab‘ēt‘, (Hamšen)
ap‘ēt‘, (Zēyt‘un) ab‘it‘ id. und (Łarabał) abet‘ ‘eine Bergpflanze, woraus Zunder
gemacht wird’ (s. dazu Ačaṙyan I: 74a).
Wohl zur Unterscheidung der eigentlichen Pflanze vom Zündstoff dienen zu-
sammengestellte Formen wie grg. abed-is soḳo ‘Echter Zunderschwamm, Fomes
fomentarius (L.) Fr.’ (Maq̇ ašvili 1961: 9a)8 bzw. arm. abet‘-a-sunk und abet‘-eni

svan. hobed- ist zanisiert, wie das o der ersten Silbe verrät; s. Klimov ibid. Weiter zu nennen ist
noch ming. abed-i ‘Zunder; Schwiele’ (Kadshaia & Fähnrich 2001: 8a), an dessen Status als georgi-
sches Lehnwort kein Zweifel bestehen kann. Die neueren etymologischen Wörterbücher zum
Kartwelischen (Fähnrich & Sardshweladse 1995 und auch Klimov 1998) verzeichnen dieses Etymon
übrigens nicht mehr!
5
Als “türkische” Übersetzung für diesen Terminus gibt Orbeliani q̇ av (s. Abulaʒe 1968: 70a) =
ttü. (osm.) kav (qav) ~ (dial.) gav ‘Polyporus fomentarius Fr., Feuerschwamm; Zunder’ (Baytop
1997: 163; Steuerwald 1972: 502b; Redhouse 1890: 1426a; DS 1936f), gegenüber azb. ġov id. (ADİL I:
532a). Dieser Terminus erscheint erstmals bei Maḥmūd al-Kāšġarī (11. Jh.) als atü. qāw “The tinder
(ḥurrāqa) used as kindling for a firestick” (Dankoff & Kelly II: 228) und ist in vielen Türksprachen
(s. Clauson 1972: 579a; DTS 1969: 436b) sowie (als Lehnwort aus diesen) in einer Reihe von Nach-
barsprachen nachweisbar, u. a. im Iranischen, Arabischen und Samojedischen (s. dazu Doerfer Nr.
1543); weiterhin hinzuzufügen ist hier noch arm. ɣav ‘Zunder, Porling’ (Łazaryan 1981: 1a; Malxase-
anc‘ III: 185a). Als Basisform hat Räsänen (1969: 214b) gtü. (*qāb =) *qab rekonstruiert. Wie uns
Baytop (1999: 250) für die Türkei mitteilt, wird der im unteren Bereich an Baumstämmen vorkom-
mende kav mantarı (syn. ağaçmantarı) außer als Zündstoff auch officinell als blutstillendes Mittel
in Wundaufschlägen gebraucht.
6
Orbeliani (II: 475a) verzeichnet dieses Wort in der Form (“somxuri”) apiet (vgl. dazu die ar-
menische Dialektform aus Hamšen gleich im Anschluß).
7
Grigor Tat‘ewac‘i (*1340, †1411) wirkte als Lehrer im Kloster Tat‘ew, woher sein Beiname rührt.
Er ist bekannt als ein erbitterter Gegner der mit Rom unierten armenischen Geistlichen, den
fratres unitores. Grigor Narekac‘i (*945, †1010), “ein gelehrter und tieffrommer Mönch”, gilt als “der
größte Mystiker” Armeniens; er wirkte im Kloster Narek, wo er auch strab (Deeters 1963: 209, 186;
Thomson 1995: 134, 128). Der Beleg bei Narekac‘i ist insofern nicht gesichert, weil unklar ist, ob das
Werk Ban xratow vasn owłił hawatoy ew mak‘owr varowc‘ aṙak‘inowt‘ean, in dem er auftritt, auch
wirklich Narekac‘i zuzuschreiben ist; s. dazu Ačaṙyan I: 34.
8
Weitere hier zu nennende Dialektformen bei Maq̇ ašvili (ibd.) sind: grg. (pšauri) sa-abed-e
soḳo, (kartluri, lečxumuri) abeda-soḳo; ferner ming. obed-iši soḳo, svan. hobediak (ṭqubul), das we-
258 Uwe Bläsing

‘Porling, Löcherschwamm, Polyporus Fr.’ (Łazaryan 1981: 1a; Malxaseanc‘ I: 1c).9

Über das gegenseitige Verhältnis der kartwelischen und armenischen Formen
schreibt Ačaṙyan zunächst in einem Aufsatz (1898: 225b), daß es sich im Georgi-
schen um ein Lehnelement aus dem Armenischen handele. Später, in seinem
etymologischen Wörterbuch (I: 74a) hegt er
aber aufgrund der gemein-kartwelischen
Verbreitung des Stammes *abed- gewisse
Zweifel hieran und hält ein umgekehrtes
Lehnverhältnis für nicht ausgeschlossen.
J̌ ahukyan (1987: 595) vertritt ganz klar letz-
tere Auffassung, d. h. er beschreibt arm.
apet‘ als georgisches Lehnwort. Zusätzlich
deutet er an, ohne jedoch wirklich konkret
zu werden, daß es ein Substratwort aus dem
Urartäischen sein könne (“urarterenĕ?”).
Erwähnt werden sollte in diesem Zusam-
menhang noch das von Nikolayev & Staros-
tin (1994: 615) rekonstruierte PEC *HVṗitV
Abb. 2. Echter Zunderschwamm, “tree-fungus”, auf dessen Ähnlichkeit zu
Fomes fomentarius (L.) Fr.
krtw. *abed- die beiden Autoren hinweisen;
vgl. dazu besonders die lakische Realisation
aṗiṭa “mushroom” (ibid.), die den oben genannten Formen lautlich am nächsten
kommt.10
Fassen wir also zusammen: Der Umstand, daß das Etymon im Armenischen
völlig isoliert dasteht, während es als Wurzel mit Belegen in allen kartwelischen
Sprachen vertreten ist, schließt ein Lehnverhältnis grg. < arm. sicherlich aus. Die
Zurückführung von arm. apet‘ auf grg. abed-i ist hingegen sehr naheliegend. Zu
beachten ist dabei, daß, wie Ačaṙyan (1898: 225b) sonst richtig festgestellt hat, u.
a. die Lautkombination -ed im Auslaut im Armenischen vermieden wird und

gen seines o wiederum auf eine zanische (mingreliche) Form zurückgehen muß sowie svan. abed(-i
soḳ), welches aus dem Georgischen entlehnt ist (s. a. Onian 1917: 1a).
9
Zu grg. soḳo > ttü. soko ‘Pilz’ s. die Liste weiterer Pflanzennamen am Ende dieses Beitrages;
zum Formans arm. -eni, das in der Regel Baum- bzw. Strauchnamen bildet, s. Abeghian 1936: 44f.
10
Die PEC-Form beruht neben lak. aṗiṭa nur noch auf gunz. bodol (< proto-cez. *bot-ol “tree-
fungus, tinder”) und tab. bit (< proto-lzg. *ṗit(a))“tree-fungus”, was keineswegs eine reiche und
überzeugende Basis für ihr Zutreffen ist.
 Georgische Gewächse auf türkischer Erde 259

hierfür gewöhnlich -et‘ eintritt. Welche konkreten Beziehungen evt. zwischen

den beiden proto-sprachlichen Wurzeln (*HVṗitV : *abed-) und anderen ähnli-
chen Formen bestehen (s. dazu Nikolayev & Starostin ibid.), liegt vorerst noch
im Dunkeln. Ebensowenig läßt sich ein Nachweis für ein uraratäisches Substrat-
wort erbringen.
Was speziell die Dialektform abet aus Yukarı Maden angeht, kann diese
unmittelbar sowohl auf grg. abed-i als auch arm. apet‘ zurückgeführt werden. Im
ersten Fall wäre der stimmhafte auslautende Verschlußlaut (-d) stimmlos ge-
worden (-t), ein Phänomen das für das Türkeitürkische bekanntermaßen typisch
ist, vgl. hierzu ttü. (Borçka, Yusufeli) arnat ‘dreieckiges Gerät zum Anhäufen von
ausgedroschenem Stroh’ (DS 329b; Pehlivan 1993: 241b) < grg. arnad-i ‘eine Art
Holzrechen zum Einsammeln und Zusammenfegen des gedroschenen Korns
und der Spreu auf der Tenne’ (Tschenkéli 34b).
Was die semantische Seite betrifft, ist die eingangs referierte Beschreibung
von İlker (s. o.) kaum brauchbar, präzise Rückschlüsse auf die Art des Gewächses
zu ziehen. Um die Beschreibung des Feuerschwammes, der am unteren Teil von
Baumstämmen oder an Baumstümpen wächst, handelt es sich jedenfalls nicht,
zumal İlker abet otu auch in eine Reihe stellt mit Gewächsen (“otlar”) wie yara
otu ‘ein Hypericum-Gewächs’ oder ebegümeci ‘Malva’ (Baytop 1997: 152; 97) etc.!
Hinzu kommt ferner, daß der Feuerschwamm auch im Türkischen von Artvin
durch ttü. kav (s. Fußnote 4) bezeichnet wird (s. Tokdemir 1993: 649a).
In früherer Zeit kam dem Zunder auch in der Türkei eine wichtige Bedeutung
als Zündstoff zu. Heute ist er so gut wie außer Gebrauch. Nur alte Leuten bedie-
nen sich noch ganz vereinzelt eines sog. doğal çakmak (“Naturfeuerzeug”) das
ausgenommen vom Zunder selbst aus einem Feuerstein und einem kleinen
Schlageisen besteht. Um mit diesen Utensilien ein Feuer zu entfachen, häuft
man zunächst ein wenig Zunder an. Sodann schlägt man das Eisen fest und in
kurzen Intervallen gegen den Feuerstein, sodaß Funken entstehen, die auf den
Zunder überspringen und ihn langsam zum Brennen, oder besser gesagt zum
Glühen bringen. Unter Zufuhr von Luft—etwa durch Anblasen oder Fächeln—
beginnt dieser schließlich so heiß zu glühen, daß man damit beispielsweise tro-
ckene Blätter oder Holzsplitter in Brand setzen und somit langsam ein richtiges
Feuer aufbauen kann. Aufbewahrt werden Zunder und Utensilien meist einge-
wickelt in einem kleinen Lederläppchen oder einem Lederbeutelchen, das sich
überall leicht einstecken läßt. Feste Behälter, sog. Zunderbüchsen kommen
260 Uwe Bläsing

ebenfalls als kavlık (lokal kavluk) vor, sie sind jedoch wegen ihrer starren, festen
Form weniger beliebt.

KÖNDAR (Aşağı Maden, Şavşat), KONDAR (Ardanuç—Artvin). Die so bezeich-

nete Pflanze wird als ‘ein zwischen Felsen wachsendes, angenehm duftendes
Kraut’ beschrieben, das İlker (1992: 206b), Tokdemir (1993: 571), Pehlivan (1993:
276a) und Özkan (1994: 100a) übereinstimmend als kekik otu, also ‘Thymian’ in-
dentifizieren.
Es handelt sich hierbei um grg. kondar-i, (mtiuluri, moxevuri, rač̣ uli) kondara
~ beg-kondara (“Hügel-Kondar”), (tušuri) velt-kondara (“Felder-Kondar”), (ime-
ruli) ṭq̇ is kondari (“Wald-Kondar”)11 etc., welches in der Tat Namen für verschie-
dene Arten von ‘Thymian’ und ‘Bohnenkraut’ o. dgl. sind (Maq̇ ašvili 1961: 17a,
56a; Tschenkéli 1575a; Ɣlonṭi 1984: 574b). Dieser Pflanzenname tritt bereits im
Altgeorgischen einmal auf und zwar in der 9. Homilie zum Hexaemeron des Ba-
silius von Caesarea in der Verbindung kondar-i velur-i (“wildwachsendes Kon-
dar”), welches hier für gr. ὀρίγανον ‘Origanum vulgare’, also den ‘Oregano’ oder
‘Dost’ steht (s. Abulaʒe 1973: 455b).12 Sulxan Saba Orbeliani (II: 230a, 171a) verbin-
det kondari jedoch irgendwie mit usuṗi (usuṗi kondariao ‘Usup sagt man ist Kon-
dar’)13 und Čubinašvili (1840: 520b) verzeichnet kondari als “tymbre, sarriette

11
Hier die attributiven Elemente im einzelnen: grg. beg-i ‘Anhöhe, Hügel’, vel-i ‘Feld, Flur,
Ebene’ (-t, Pluralsuffix im Obliquus) und ṭq̇ e ‘Wald’ (-is, Genitiv Suffix); s. Tschenkéli 74a, 372,
1374a.
12
Die hier entsprechenden Textstellen sind: grg. da ḳuvi, ražams ganʒɣis qorcita ikednisajta,
č̣ amis kondari veluri da ganeris vnebisagan c̣ amlisa mis maḳudinebelisa (Abulaʒe 1964: 12626, 27); gr.
Χελώνη δὲ σαρκῶν ἐχίδνης ἐµφορθεῖσα διὰ τῆς τοῦ ὀριγάνου ἀντιπαθείας φεύγει τὴς βλάβην τοῦ ἰοβόλου
(Amand de Mendieta & Rudberg 1997: 15017, 18) “Die Schildkröte, die sich von Otternfleisch vollge-
fressen hat, entgeht der schädlichen Wirkung des Giftes durch die Gegenwirkung des Dosten”
(Stegmann 1925: 143u).
13
Grg. usuṗ-i ‘Hyssopus (officinalis), Ysop’ (Maq̇ ašvili 1961: 68a; Tschenkéli 1417a) < gr. ὕσσωπος,
lat. hyssōp-us, -um. Im fremdsprachigen Teil seines Werkes übersetzt Orbeliani (II: 572b) kondari
aber ganz passend mit (“latinuri”) timo (= ital. timo ‘Thymian’), (“turkuli’) marza (= azb. märzä
‘eine duftende Garten-, Nutzpflanze’, ADİL III: 299a = prs. marza “Origany”, Steingass 1214b, “Sa-
tureja hortensis”, PRS II: 494a; ferner vgl. krd. merze ‘Majoran’, Omar 1992: 395b, zur weiteren Ver-
breitung im Iranischen s. Cabolov I: 618) und (“somxuri”) ḳortin (~ ḳotrin) (= arm. kort‘in, kordiwn
etc. ‘Satureja (hortensis) L.’, Łazaryan 1981: Nr. 624, Ačaṙyan II: 647a). Ebenfalls unklar sind mir die
Hintergründe der Wiedergabe von gr. ὀρίγανον durch arm. dałjn, einem Wort für ‘Minze, Mentha’
(Łazaryan 1981: Nr. 44), an der entsprechenden Stelle in der armenischen Übersetzung des Hexae-
merons: darjeal ew ǰrayin kriay et‘ē keric‘ē i msoy iži ew tagnap inč‘ hasanic‘ē nma i kerakroyn, ert‘ay
owtē dałjn (Muradyan 1984: 29714, 15).
 Georgische Gewächse auf türkischer Erde 261

(plante)”. In den anderen südkaukasischen Sprachen läßt sich dieser Terminus

nur noch im Svanischen nachweisen und zwar als Lehnelement aus dem Georgi-
schen: svan. kondar ~ kondär, mǝ-kondr-ōl ‘Satureia spicigera C. Koch’; (Niža-
radze 1910: 503; Onian 1917: 30b; Maq̇ ašvili 1961: 56a). Als Lehnwort ist er ferner
noch ganz sporadisch im Armenischen, Ossetischen und Aserbaidschanischen
vertreten, hier die entsprechenden Belege: arm. dial. (ohne Hinweis auf die Her-
kunft) k‘ondar ‘Bohnenkraut, Satureja hortensis L.’ (Malxaseanc‘ IV: 584a), oss.
(Süd) kondari ‘eine Art Grünpflanze’ (Abaev
1949: 502) und azb. (Zaġatala) köndär ‘Name
einer Pflanze’ (ADL I: 288a). Die weitere
Herkunft von grg. kondara ist noch undeut-
lich. Erwähnenswert ist in diesem Zusam-
menhang aber die von Nikolayev & Starostin
(1994: 705) in Verband mit PEC *kwĕnVdV
(~-əә̆-) “bush, crown of a tree” rekonstruierte
proto-nachische Form *kondV-r (~-a-), die
sich begründet auf čeč. kondar “crown of a
tree, a kind of weed” und ing., bacb. kondar
“savory”, wozu die beiden Autoren mit Recht
einräumen, “forms with the meaning ‘weed,
savory’ may be loaned from Georgian (cf.
Georg. kondari); however, a reverse
Abb. 3. Thymian
direction of borrowing is also possible”.14
Ferner genannt werden sollten hier auch
die formal recht ähnlichen Formen ming. ḳoindar-i ~ ḳolandar-i, ḳolindar-i ‘Wie-
senlolch, Raygras; Lolium perenne L.’ (Kipšidze 1914: 260a), an die sich als Lehn-
elemente abx. (Bsyp) a-ḳɨǝndǝr, (Abshuj) a-mḳɨandǝr und (abžui) a-ḳɨǝlandǝr
‘Coriandrum sativum L.’ (AAIAM 1999: 28), svan. ḳoindar und grg. ḳoindar-i ~ ḳo-
lindar-i ‘Wiesenlolch, Raygras; Lolium perenne L.’ (Tschenkéli 606b; Maqašvili
1961, 40a; Onian 1917: 10b; KEGL 4: 1270) anschließen. Ein direkter etymologischer
Zusammenhang mit grg. kondara scheint allerdings nicht zu bestehen, und zwar
wegen des unterschiedlichen Anlauts (ḳ- : k-), insbesondere aber wegen des ‘zu-
sätzlichen’ -l-. Was speziell die Realisation ming. ḳoindar-i (> grg. ḳoindar-i, svan.

14
Wegen der geringen Verbreitung des Wortes in den ostkaukasischen Sprachen (lediglich in
der nachischen Gruppe) halte ich die Entlehnung aus dem Georgischen für naheliegender.
262 Uwe Bläsing

ḳoindar etc.) betrifft, so ist für sie am wahrscheinlichsten von einer Verdünnung
des -l- > -j- zwischen Vokalen und schließlichem Ausfall des letzteren vor -i aus-
zugehen, eine Erscheinung, die gerade in der martwilischen Mundart des sena-
kischen Dialekts des Mingrelischen beobachtet werden kann; vgl. dazu ming.
buleḳi ‘Radieschen’, teleba ‘unversehrt, gesund sein’, mela ‘Fuchs’ bzw. ḳalie
‘Simlax eclesa L.’, sali ‘Radspeiche’ ~ (martwilisch) bujeḳi, tejeba, meja bzw. ḳaie,
sai etc. (Kipšidze 1914: 208a, 241b, 279a; 251b, 315a).
Die Bevölkerung von Maden Köyü übrigens mißt der im freien Gelände zwi-
schen Steinen und Felsen wild wachsenden Thymianpflanze besondere Bedeu-
tung bei. Man sammelt sie, läßt sie trocknen und zerkleinert sie schließlich, um
damit Speisen zu würzen und zu verfeinern, dies häufig anstelle von Minze, die
nämlich in dieser Region nicht vorkommt. Gerade die für die Gegend typische
Suppe, lor aşı oder lor çorbası,15 wird mit köndar (oder nişoş) gereicht, wegen ih-
rer offizinellen Eigenschaften sogar an Kranke.16 Ferner gilt köndar als eines der
‘wertvollsten Geschenke’ (“en kıymeti armağanlar”; İlker ibid.), das die Dorfbe-
wohner an ihre Freunde und Verwandte in der Fremde schicken.

Hier weitere Phytonyme dieser Art aus der Vogelschau:

ABUTIRAK ~ ABUTIRAḪ (Ardanuç—Artvin) ‘eine auf Weiden und Wiesen
wachsende, dem Lauch ähnliche, giftige Pflanze’ (DS 35a; Pehlivan 1993: 237;
Özkan 1994: 97a) bzw. ABUDARAT (Artvin) ‘ein auf hohen Lagen vorkommendes
Gewächs, dessen Wurzeln giftig sind’ (Tokdemir 1993: 570) < grg. dial. aṗuṭraḳ-i ~
(mesxuri, ǯavaxuri) aṗuṭaraḳ-i, (imeruli) abudraḳ-i, (rač̣ uli, lečxumuri) aṗunṭraḳ-
i, (guruli) xaṗuṭraḳ-i ‘Germer-Art, Veratrum lobelianum Bernh.’ (Maq̇ ašvili 1961:
81b; Ɣlonṭi 1984: 38b, 739a; Tschenkéli 2299b; KEGL 1: 529, 1336). Bereits Orbeliani
(I: 38a) bezeugt diesen Pflanzennamen, allerdings in der etwas abweichenden
Form grg. abuṭaraṭ-i ‘Veratrum-Gewächs’, die sich interessanterweise bis heute
im Raum Artvin in ttü. dial. abudarat erhalten hat!

15
Die lor-Suppe, und so wird sie gemacht (İlker 1992, 257b): ‘Man gibt Mehl in siedendes Was-
ser und läßt es unter gleichmäßigem Umrühren aufkochen. Dann wird der zuvor in einer Schale
vorbereitete lor-Käse hinzugegeben, wonach man das Ganze noch etwas kochen läßt. Schließlich
wird die fertige Suppe in Schälchen gefüllt. Wenn sie etwas abgekühlt ist, zerkleinert man nişoş
und schmeckt sie damit ab’; bon appétit!
16
Der Volksmund drückt das dann so aus: Hele bir köndarlı, nişoşlu bir lor çorbası yapın yesin,
bir şeyciği kalmaz ‘Macht ihm doch eine lor-Suppe mit köndar oder nişoş, die soll er essen, und
nicht das Geringste bleibt von seiner Krankheit übrig’ bzw. (der Kranke selbst verlangt:) Bana kön-
darlı, nişoşlu bir lor çorbası yapın belki yerim ‘Macht mir eine lor-Suppe mit köndar oder nişoş, dann
eß ich vielleicht was!’ (İlker 1992: 207a).
 Georgische Gewächse auf türkischer Erde 263

ADOL (Ardanuç, Şavşat—Artvin; Ardahan, Sarıkamış—Kars; Erzurum; To-

kat) ‘eßbare, etwa kartoffelgroße Knollen eines im Frühjahr wachsenden Krautes,
die beim Pflügen des Feldes zutage kommen’ ~ ATOL (Ardanuç, Şavşat—Artvin;
Ardahan—Kars), ATTOL (Göle—Kars) ‘eine Art Erdapfel, der auf gepflügten Fel-
dern wild wächst und dessen Knollen roh oder gekocht verzehrt werden können,
während aus den Blättern turşu (sauer Eingelegtes, pickles) zubereitet wird’ (DS
68a, 372b; Tokdemir 1993: 571; Pehlivan 1993: 242a; Özkan 1994: 100a)17 < grg. (Tür-
kei: Borčxa, Murɣuli) aṭol-i ‘Name einer Pflanze, die gegessen wird und die be-
sonders in den Gebieten Meßchetien-Dschawachetien und Klardschetien-
Schawschetien verbreitet ist’, (mesxuri, ǯavaxuri) ‘Rübenkälberkropf, Kerbelrübe;
Chaerophyllum bulbosum L. (Ch. caucasicum (Fisch.) Schischk.)’ bzw. ‘Trüffel’
(Puṭḳaraʒe 1993: 384b; Maq̇ ašvili 1961: 75; Tschenkéli 40a; KEGL 1: 640). Außer ins
Türkeitürkische ist dieses Phytonym auch ins Aserbaidschanische entlehnt: azb.
dial. atdol (Kälbäǰär) ‘Name einer Pflanze’ (ADL I: 19b).
ANSLİ ~ ANSLI (Ardanuç, Şavşat) ‘eine ganz übel riechende, stinkende
Pflanze’ (DS 280a), ‘eine etwa ein Meter hohe Pflanze mit dicken Stengeln und
schirmähnlichen erhabenen Blütenständen’ (Tokdemir 1993: 571, 628b), ANTSLİ
(Şavşat) ‘einjährige Pflanze, ca. 1-2 Meter hoch’ (Polat 2000: 152) sowie ANSIL (Ar-
danuç—Artvin) ‘ein duftendes Kraut von kurzem Wuchs, auf dessen Blättern
zwirnartige Adern verlaufen; yedi ṭamar oti’ (Pehlivan 1993: 241a; Özkan 1994:
99b)18 und ANZİLİ (Artvin) ‘Sambucus ebulus L.’ (Baytop 1997: 295) < grg. anc̣ -l-i
‘Holunder, Attich; Sambucus ebulus L.’ bzw. (kvemo-imeruli) ‘Schwarzer Holun-
der; Sambucus nigra L.’ (Tschenkéli 29a, KEGL 1: 515; Maqašvili 1961: 12a, 24a).
Ebenfalls hiervon azb. dial. anc̣ li, anc̣ ili(ji) (Qaxbašı, Güllük, Sarıbaš, İlisu)
‘Wacholder’ (Džangidze 1978: 130).
BİRBAMBA (Şavşat—Artvin) ‘Pflanze mit rundlichen, weichen Blättern, die
an ihrer Unterseite behaart sind’ (DS 697b) < grg. ṗirbamba (kartluri, ač̣ aruli)
‘Huflattich, Tussilago farfara L.’, (mesxuri) ‘Phlomis pungens Willd.’ (Tschenkéli
1008a; Maq̇ ašvili 1961: 28, 100a).
BİRKİ (Ardanuç—Artvin) ‘ein auf den Feldern gedeihendes Gewächs mit
Kletten, aus dem man in noch jungem, frischem Zustand Pickles (turşu) macht’
(DS 702a; Pehlivan 1993: 245b; Özkan 1994: 102b). < grg. birḳ-i ~ birḳ-a, (Türkei:

17
Nicht ganz klar ist, mit welcher Pflanze wir es hier zu tun haben. Nach Hauenschild (1989: Nr
560) bezeichnet atol ‘Helianthus tuberosus L., Erdbirne, Topinambur’, nach Baytop (1997: 143 +
Abb. 56) aber ‘Bunium microcarpum (Boiss.) Freyn’. In beiden Fällen handelt es sich um Gewächse
mit eßbaren Knollen. Weiter verweisen Baytop und Pehlivan auf eine Dialektform mit protheti-
schem (vermutlich sekundärem) h-: ttü. HATOL.
18
Yedi damar otu ist im Volksmund in aller Regel ein Name von ‘Plantago L.; Wegerich’ (s. Bay-
top 1997: 40f.).
264 Uwe Bläsing

Murɣuli) birḳ-a-i ‘heilkräftige Pflanze’, (ingilouri) birḳ-a-j ‘stachelige Früchte ge-

wisser Pflanzen, welche an der Kleidung bzw. im Fell von Tieren haften bleiben;
Klette(n)’ (Ɣlonṭi 1984: 74b; Tschenkéli 87a; Rosṭiašvili 1978: 38b; Maq̇ ašvili 1961:
17b).
BİTNE ‘ein Kraut’ ~ BİDNE (Yusufeli—Artvin) ‘Blätterart, die man im Win-
ter dem Stallvieh unterlegt’ (DS 688b, 712a) < grg. ṗiṭna ‘Minze, Mentha’
(Tschenkéli 1018b; Maqašvili 1961: 58), welches in der Form ṗiṭnaḳ-i schon im Alt-
georgischen, u. a. im Lucas-Evangelium vorkommt (s. Abulaʒe 1973: 342a). Ferner
hieran anzuschließen ist noch azb. dial. pitnä (Zaġatala) “yarpız” (ADL II: 457a;
Džangidze 1978: 133).
CİNCAR (Şavşat, Ardanuç, Yusufeli—Artvin; Ardahan, Sarıkamış, Çıldır—
Kars) ~ ÇİNÇAR (Şavşat—Artvin), CINCAR (Şavşat—Artvin; Zara—Amasya)
und CIMCAR (Yusufeli—Artvin) ‘Brennessel’ (DS 975b; Tokdemir 1993: 571, 637a;
Pehlivan 1993: 254a) < grg. (Tupekči Konaki/Türkei, guruli) ǯinč̣ ar-i, das so schon
im Altgeorgischen erscheint (s. Abulaʒe 1973: 574b), bzw. (Schriftsprache) č̣ inč̣ ar-i
(durch Assimilation < ǯinč̣ ar-i) ‘Brennessel, Urtica’ (Puṭḳaraʒe 1993: 685b;
Tschenkéli 2254b, 2448b). Weiterhin finden wir grg. ǯinč̣ ar-i als Lehnwort in azb.
dial. ǰinǰar (Balakän, Ġax, Nuxa, Šäki, Zaġatala) (ADDL 1964: 459; ADL I: 80a).
ÇARḪALA (Şavşat) ~ CARHALA (Ardanuç, Şavşat, Yusufeli—Artvin; Göle,
Ardahan, Posof), CALHALA, ÇARGALA (Posof—Kars) ‘Rübe, Runkelrübe, Zu-
ckerrübe’ (DS 861b; 1081a; Tokdemir 1993: 571, 635b; Pehlivan 1993: 253a; Özkan
1994: 107a; Polat 2000: 153) < grg. č̣ arxal-i ‘Beta vulgaris L., Runkelrübe, rote Rübe’
(Maq̇ ašvili 1961: 95a; Tschenkéli 2234a).
ESER (Artvin) ‘Farnkraut’ (Tokdemir 1993: 571, 641b) < grg. (ač̣ aruli, guruli)
ec̣ eri, (Schriftsprache) ec̣ ris gvimra ‘Pteridium’ (Maq̇ ašvili 26b); zu weiteren
Einzelheiten siehe Bläsing 2001: 60f.
GOLO (Yusufeli) ‘eine Art Kraut’ (DS 2098b), ĠALO (Ardanuç, Şavşat—Art-
vin) ‘eßbares Kraut mit langem Stiel und großen Blättern’ (Pehlivan 1993: 264a;
Özkan 1994: 113a; Polat 2000: 156) < grg. ɣolo ‘Ampfer; Rumex sp.’, das in den Dia-
lekten sehr oft auch in der Form ɣvalo (mesxuri, imeruli, rač̣ uli, lečxumuri,
ač̣ aruli) id., (mesxuri, dasavluri) ‘Rumex conglomeratus Murray’ (Tschenkéli
1641a, 1622a; Maq̇ ašvili 1961: 75f; Ɣlonṭi 1984: 586b) vertreten ist.
İPNİ (Ardanuç, Şavşat—Artvin) ‘Esche’ (DS 2546a; Tokdemir 1993: 570;
Pehlivan 1993: 272a; Özkan 1994: 117a; Polat 2000: 157) < grg. (kartluri, ḳaxuri,
pšauri, moxevuri, ač̣ aruli) ipn-i ~ ipen-i, (kartluri, rač̣ uli, lečxumuri) ipn-a, das der
imerchewische Dialekt ebenso wie die moderne Literatursprache auch als ipani
‘Esche, Fraxinus excelsior L.’ bzw. ‘Fraxinus oxycarpa Willd.’ kennen (s.
Tschenkéli 535a; Maq̇ ašvili 1961: 36a; Puṭḳaraze 1993: 477b).
 Georgische Gewächse auf türkischer Erde 265

LEK ~ LEKİ (Ardanuç) ‘ein Baum mit kleinen Blättern’ (Pehlivan 1993: 289a;
Özkan 1994: 125b), LEK (Şavşat) ‘Esche’, (Artvin) ‘Ahorn’ sowie DAĞLEKI (Artvin)
‘eine auf höheren Lagen wachsende Ahornart’ (DS 3070b; Tokdemir 1993: 569f,
656a; Polat 2000: 160) < grg. (imerxevuli) leḳ-i 'Baum mit breiten Blättern'
(Puṭḳaraʒe 493b), (pšauri, xevsuruli, rač̣ uli, lečxumuri) leḳ-i, (ač̣ aruli) leḳixe,
(Schriftsprache) leḳa, leḳis-xe ‘Acer platanoides L., Spitz-Ahorn’ und (in einigen
Dialekten) ‘Acer laetum C. A. Mey. non Schwer., Kolchischer Ahorn’ (Maq̇ ašvili
1961: 43a: 74a; Tschenkéli 677). Die zweite Dialektform, dağleki sieht dagegen
nach einem gemischtsprachlichen Kompositum aus, das ttü. dağ ‘Berg’ und grg.
leḳ-i in sich vereint. Sachlich gestützt wird diese Herleitung jedenfalls durch den
entsprechenden Hinweis—‘eine auf höheren Lagen wachsende Ahornart’.
MAḲUVAL ~ MEḲVELA (Ardanuç), MAGVAL (Şavşat) ‘Brombeere’
(Pehlivan 1993: 290a; Özkan 1994: 126a; Polat 2000: 160), MAKVAL (Artvin) ‘Rubus
discolor Weihe et Nees’ (Baytop 1997: 297) < grg. maq̇ v-al-i ‘Brombeere’
(Tschenkéli 730b; Klimov 1998: 117).
MİCİVAŞLİ (Yukarı Maden-Artvin) ‘Topinambur, Erdapfel’ (İlker 1989: 211) <
grg. mic̣ avašla, (guruli) mic̣ is vašli ‘Topinambur; Helianthus tuberosus L.’
(Tschenkeli 782b; KEGL 5: 503; Maq̇ ašvili 1961: 50b), welches ein Kompositum ist,
bestehend aus grg. mic̣ a ‘Erde, Boden’ und vašl-i ‘Apfel’.
SASḪİ (Artvin) ‘Linde(nbaum)’ (Tokdemir 1993: 570). < grg. cacxv-i ‘Tilia L.,
Linde’ (Maq̇ ašvili 1961: 86a; Tschenkéli 1977b).
SİPELA (Şavşat—Artvin) “ihlamur ağacı” (DS 3644b) < grg. c̣ ip-el-a ~ c̣ ip-el-i
‘Buche, Fagus’ (Tschenkéli 2176a; Maq̇ ašvili 1961: 93a). Weiterhin liegt dieser
Baumname als Lehnwort in azb. (Zaġatala) c̣ ipli und cax. c̣ ipil ‘Buche’ (Aslanov
1974: 225) vor.
SİSMAT (OTU) (Aşağı Maden), TSİTSMAND (Şavşat—Artvin) ‘Kresse’ (İlker
1992: 206b; Tokdemir 1993: 571; Polat 2000: 163) < grg. c̣ ic̣ maṭ-i ‘Garten-, Salat-
kresse, Lepidium sativum L.’ (Tschenkéli 2178a; Maq̇ ašvili 1961: 93a).
SOKO ~ ZEKO (Ardanuç), SOKO (Şavşat) ~ ZOKO (Ardanuç, Yukarı Ma-
den—Artvin) ‘Pilz’ (Pehlivan 1993: 308a; DS 3658a; İlker 1989: 311b; Tokdemir 1993:
571, 664a) < grg. soḳo ~ dial. (kartluri, pšauri, kiziq̇ uri, moxevuri) zoḳo ‘Pilz’
(Tschenkéli 1270a; 413b; Ɣlonṭi 1984: 234a).
ŞİKAR ~ ŞİKĀR (Şavşat) “yabangülü” (DS 3776b), ŞIKER (Artvin) ‘eine Art
“orman gülü”’. Der Botaniker Baytop (1997: 298) belegt diesen Terminus ebenfalls
für Şavşat und bestimmt die damit bezeichnete Pflanze als ‘Daphne mezereum
L., Gewöhnlicher Seidelbast’. Gerade die letztgenannte Realisationen ist es, die
ganz eindeutig auf grg. šker-i ‘Rhododendron ponticum L., pontische Alpenrose’
(Tschenkéli 1857a) als Anküpfungspunkt für das türkeitürkische Set weist. Aus
den georgischen Dialekten sowie den anderen kartwelischen Sprachen lassen
266 Uwe Bläsing

sich für dieses Lexem reiche Belege zusammentragen, wie z. B.: grg. dial. (guruli)
šḳer-i, (lečxumuri) žgeri, (kvemo-rač̣ uli) lešḳeri, (zemo-rač̣ uli) lešḳi, laz. m-šker-i ~
m-ške-i, m-šḳer-i (Aṭina, Arkabi), p-šḳer-i (Vic̣ e-Arkabi); ming. (p-)šker-i ~ (p-)šḳer-
i; svan. šgeri ~ šgōri id. (Tschenkéli 1829a; Maq̇ ašvili 1961: 81b; Čikobava 1938: 131;
Marr 1910: 206b; Kipšidze 1914: 359a; Kadshaia & Fähnrich 2001: 383a, 463a; Gud-
jedjiani & Palmaitis 1985: 271) sowie grg. (imeruli, guruli) škeri, šḳeri ‘Rhododend-
ron caucasicum Pallas’ und (kvemo-rač̣ uli, lečxumuri) šḳeri, (zemo-rač̣ uli) lešḳeri
~ lešḳi, (kartluri) čkeri ‘Laurocerasus officinalis Roem., Kirschlorbeer’ (Maq̇ ašvili
1961: 23b, 93b).

ABKÜRZUNGEN
abx. Abchazisch grg. Georgisch laz. Lazisch
arm. Armenisch gtü. Gemein-Türkisch lzg. Lezgisch
atü. Alt-Türkisch gunz. Ghunzib ming. Mingrelisch
azb. Azerbaidschanisch ing. Inguschisch osm. Osmanisch
bacb. Batsbisch ital. Italienisch oss. Ossetisch
cax. Tsachurisch krd. Kurdisch prs. Persisch
čeč. Tschetschenisch krtw. Kartwelisch svan. Svanisch
cez. Tsezisch lak. Lakisch tab. Tabasaranisch
gr. Griechisch lat. Lateinisch ttü. Türkei-Türkisch
dial. dialektal

LITERATUR
AAIAM (1999) = Apsnǝ apsabarac’ara iazk’u až°art’° mat’erialk°a ‹Slovarnye materialy po priro-
dovedeniju Abxazii›, Aq’°a (Suxum).
Abaev, V. I. (1949), Osetinskij jazyk i fol'klor, I, Moskva & Leningrad.
Abeghian, A. (1936), Neuarmenische Grammatik; Ost- und Westarmenisch, mit Lesestücken und ei-
nem Wörterverzeichnis, Berlin & Leipzig.
Abulaʒe, C. A. (1968), Sulxan-Saba Orbelianis leksiḳonis siṭq̇ vanis turkuli targmanebi ‹Tjurkskie pere-
vody slovnika slovarja Sulaxan-Saba Orbeliani›, Tbilisi.
Abulaʒe, I. (1964), Uʒvelesi redakciebi Basili Ḳesarielis “ekusta dɣetajsa” da Grigol Noselis targmane-
bisa “ḳacisa agebulebisatwis” (gamoḳvleva da leksiḳoni daurto Ilia Abulaʒem), Tbilisi.
―― (1973), Ʒveli kartuli enos leksiḳoni ‹Slovar' drevnegruzinskogo jazyka›, Tbilisi.
Ačaṙyan, Hr. H. (1898), “Hay lezui e, ē, o, ō grerun artasanut‘iwnĕ”, Bazmavep, 1898: 220 - 227.
―― (1971-79), Hayeren armatakan baṙaran ‹Ėtimologičeskij korennoj slovar' armjanskogo jazyka›, I-
IV, Erevan. (Reprint of the edition in 7 volumes from 1926-35)
ADDL (1964) = Azärbayǰan dilinin dialektoloži lüɣäti (Redaktorları R. Ä. Rüstämov & M. Š. Širäliyev),
Bakı.
ADİL (1964-87) = Azärbayǰan dilinin izahlı lüɣäti ‹Tolkovyj slovar' azerbajdžanskogo jazyka›, I-IV,
Bakı.
 Georgische Gewächse auf türkischer Erde 267

ADL (1999-2003) = Azǝrbaycan dialektoloji lüğǝti, I-II, Ankara.

Amand de Mendieta, E. & St. Y. Rudberg (1997), Basilius von Caesarea, Homilien zum Hexaemeron,
Berlin.
ARS (1984) = Armjansko-russkij slovar' ‹Hay-ṙuseren baṙaran›. Erevan.
Aslanov, A. M. (1974), “Iberijsko-kavkazskie abruptivy v zakatal'sko-kaxskix govorax azerbajdžan-
skogo jazyka”, Ežegodnik Iberijsko-Kavkazskogo jazykoznanija, 1: 224-228.
Baytop, T. (1997), Türkçe Bitki Adları Sözlüğü, İkinci Baskı, Ankara.
―― (1999), Türkiye'de Bitkiler İle Tedavi, Geçmişte ve Bugün, İkinci Baskı, İstanbul.
Bläsing, U. (2001), “Arm. p‘ilunc‘ vs. laz. bilonc-, grg. blenc, Ein Beitrag zu den Bezeichnungen von
Farnen (Pteropsida) im Kaukasus und Anatolien”, Studia Etymologica Cracoviensi, 6: 15-78.
―― (2009), “Artvin Yöresel Sözlüğünden Örnekler: Türkiye Türkçesine Etimolojik Katkılar”,
Workshop on Turkish Dialects II, Orient Institute, 18-19 November 2005, İstanbul = Türk Dilleri
Araştırmaları, 19: 17-31.
―― (2010), “Bir Kazıyacağın Kanlı Geçmişinden”, Hacettepe Üniversitesi Türkiyat Araş̧tırmaları,
Sayı 13: 21-40.
BLS (2007) = İsmail Bucaḱlişi, Hasan Uzunhasanoğlu & İrfan Aleksiva; Büyük Lazca Sözlük ‹Didi La-
zuri Nenapuna›, İstanbul.
Cabolov, R. L. (2001), Ėtimologičeskij slovar' kurdskogo jazyka, I, Moskva.
Clauson, Sir Gerald (1972), An Etymological Dictionary of Pre-Thirteenth-Century Turkish, Oxford.
Čikobava, A. (1938), Č̣ anur-megrul-kartuli šedarebiti leksiḳoni ‹Čansko-megrel’sko-gruzinskij sravni-
tel’nyj slovar’›, Tpilisi.
Čubinašvili = David Čubinov, (1840), Gruzinsko-russko-francuzskij slovar' ‹Dictionnaire Géorgien-
Russe-Français›, Sanktpeterburg.
Dankoff, R. & J. Kelly (1982-85), Maḥmūd al-Kāšɣarī, Compendium of the Turkic Dialects (Dīwān
Luɣāt at-Turk), Harvard.
Deeters, G. (1963), “Georgische Literatur”, Handbuch der Orientalistik I, 7: Armenisch und Kaukasi-
sche Sprachen: 129-155, Leiden & Köln.
Doerfer, G. (1963-75), Türkische und Mongolische Elemente im Neupersischen, I-IV, Wiesbaden.
DS (1963-82) = Türkiye’de Halk Ağzından Derleme Sözlüğü, I- XII, Ankara.
DTS (1969) = Drevnetjurkskij slovar', Leningrad.
EGRT (1989) = Ethnic Groups in the Republic of Turkey, ed. Peter Alford Andrews, Wiesbaden.
―― (2002), Ethnic Groups in the Republic of Turkey, Supplement and Index, ed. Peter Alford
Andrews, Wiesbaden.
Erten, M. (2000), Lazca-Türkçe / Türkçe - Lazca Sözlük, İstanbul.
Fähnrich, H. (1986), Kurze Grammatik der georgischen Sprache, Leipzig.
Fähnrich, H. & S. Sardshweladse (1995), Etymologisches Wörterbuch der Kartwel-Sprachen, Leiden,
New York & Köln.
Ɣlonṭi, Al. (1984), Kartul ḳilo-temata siṭq̇ vis ḳona; meore gamocema, Tbilisi.
Gudjedjiani, Ch. & L. Palmaitis (1985), Svan-English Dictionary, Delmar/New York.
Hauenschild, I. (1989), Türksprachige Volksnamen für Kräuter und Stauden, Wiesbaden.
268 Uwe Bläsing

İlker, O. (1989), Yukarı Maden ve Yukarı Madenliler (Y. Hod ve Y. Hodlular), Köyün Doğal ve Top-
lumsal Yapısı, Cilt I, İstanbul.
―― (1992), Aşağı Maden ve Aşağı Madenliler (A. Hod ve A. Hodlular), Köyün Doğal ve Toplumsal
Yapısı, Cilt I, İstanbul.
J̌ ahukyan, G. B. (1987), Hayoc‘ lezvi patmut‘yun, naxagrayin žamanakašrǰan, Erevan.
Kadshaia, O. & H. Fähnrich (2001), Mingrelisch-Deutsches Wörterbuch, Wiesbaden.
KEGL (1950-64) = Kartuli enis ganmarṭebiti leksiḳoni ‹Tolkovyj slovar’ gruzinskogo jazyka›, 1-8, Tbi-
lisi.
Kipšidze, I. (1914), Grammatika mingrel’skago (iberskago) jazyka s xrestomatieju i slovarem, S.-Pe-
terburg.
Klimov, G. A. (1964), Ėtimologičeskij slovar' kartvel'skix jazykov, Moskva.
―― (1998), Etymological Dictionary of the Kartvelian Languages, Berlin & New York.
Łazaryan, Ṙ. S. (1981), Busanunneri hayeren-latineren-ṙuseren-angleren-franseren-germaneren baṙa-
ran ‹Armjano-latinsko-russko-anglo-francuzsko-nemeckij slovar' nazvanij rastenij›, Erevan.
Malxaseanc‘, St. (1944-45), Hayerēn bac‘atrakan baṙaran, I-IV, Erevan.
Maq̇ ašvili, A. (1961), Boṭaniḳuri leksiḳoni mcenareta saxelnodebani ‹Botaničeskij slovar' nazvanija
rastenij›, Tbilisi.
Marr, N. Ja. (1910), Grammatika čanskago (lazskago) jazyka s xrestomatieju i slovarem, S.-Peterburg.
Muradyan, K. (1984), Barseł Kesarac‘i, Yałags vec‘awreay ararč‘owt‘ean, Erevan.
Nikolayev, S. L. & S. A. Starostin (1994), A North Caucasian Etymological Dictionary, Moskow.
Nižaradze, I. I. (1910), Russko-svanskij slovar', Tiflis.
Omar, F. F. (1992), Kurdisch-Deutsches Wörterbuch, Berlin.
Onian, A. [Arsena Wonjān] (1917), “Sbornik svanskix nazvanij derev'ev i rastenij (na lāšxskom na-
rečij)”, Materialy po jafetičeskomu jazykoznaniju, VIII, Petrograd.
Orbeliani, Sulxan-Saba (1966-93), Leksiḳoni kartuli, I-II, Tbilisi.
Özkan, İ. E. (1994), Ardanuç ve yöresi ağızları, Kayseri. (unveröffentlichte Magisterarbeit, Erciyes
Üniversitesi)
Pehlivan, S. Ş. (1993), Artvin–Ardanuç Ağzından Derlemeler, Bursa.
Polat, F. (2000), Şavşat ve yöresi ağızları, Kayseri. (unveröffentlichte Magisterarbeit, Erciyes Üniver-
sitesi)
PRS (1985) = Persidsko-russkij slovar' ‹Farhang-i fārsī ba-rūsī›, Izdanie tret'e, stereotipnoe, I-II, Mos-
kva.
Puṭḳaraʒe, Š. (1993), Čveneburebis kartuli ‹The Georgian Language of “Chveneburebi” in Turkey›, Ba-
tumi.
Räsänen, M. (1969), Versuch eines etymologischen Wörterbuchs der Türksprachen, Helsinki.
Redhouse, Sir James W. (1890), A Turkish and English Lexicon, Constantinople. (Nachdruck Istan-
bul 1978).
Rosṭiašvili, N. (1978), Ingilouri leksiḳoni ‹Slovar’ ingilojskogo dialekta›, Tbilisi.
Schmidt, K. H. (1962), Studien zur Rekonstruktion des Lautstandes der Südkaukasischen Grundspra-
che, Wiesbaden.
 Georgische Gewächse auf türkischer Erde 269

Stegmann, A. (1925), Des heiligen Kirchenlehrers Basilius des Grossen, Bischofs von Cäsarea ausge-
wählte Homilien und Predigten, München.
Steingass, F. (1957), A Comprehensive Persian-English Dictionary, London.
Steuerwald, K. (1972), Türkisch-Deutsches Wörterbuch ‹Türkçe-Almanca sözlük›, Wiesbaden.
Thomson, R. W. (1995), A Bibliography of Classical Armenian Literature to 1500 AD, Turnhout.
Tokdemir, H. (1993), Artvin Yöresi Folkloru, Ankara.
Tschenkéli, K. (1965-74), Georgisch-Deutsches Wörterbuch, I-III, Zürich.
Wardrop, O. (1911), “English-Svanetian Vocabulary”, Journal of the Royal Asiatic Society, XVI: 589-
634.
Ǯangiʒe, V. (1978), “Kartuli leksiḳa azerbaiǯanuli enis črdili-dasavlur ḳiloḳavebši”, Sakartvelos SSR
mecnierebata aḳademiis macne, 4/1978: 124-140.

Nachweis der Abbildungen:

Abb. 1. Armenisch-Georgisch-Türkisches Grenzgebiet: Bläsing, privat.
Abb. 2. Echter Zunderschwamm, Fomes fomentarius: www.blumammu.de/ressourcen/zunder.
php (10. 10. 2005).
Abb. 3. Thymian: http://plantaardigheden.nl/plant/beschr/gonnve/tijm.htm (11. 10. 2005).
 The Persian Verbal Suffixes -ān and -andeh (-andag)

Johnny Cheung
Leiden University

Abstract
The present study discusses the differences in usage and meaning of the originally Middle/New
Persian present participles in -andag/-andeh and in -ān respectively. They are illustrated by ample
material cited from Manichaean & Pahlavi sources (in the case of Middle Persian) and the epic
work of the Šāhnāmeh (concerning New Persian). Formally, these participles reflected the Old Ira-
nian active (-andag/-andeh) and middle (-ān) voice/diathesis. The later Persian distribution of the
participles in -andag/-andeh and -ān respectively appear to be semantically marked and linked to
the Old Iranian employment of the active and middle voice.

Keywords
Historical Linguistics, Middle Persian, History of Persian, Syntax, Middle Voice (Diathesis), Parti-
cipial Constructions, Iranian Languages

INTRODUCTION
As a token of my utmost respect and admiration for Professor Asatrian’s schol-
arly output, the present contribution can be considered as an elaboration on two
of his earlier publications, viz. Asatrian 1983; idem 1989, that dealt with the parti-
ciple -ān in West Iranian. I expected that it could have been published a long
time ago, but alas, it has not seen the day on paper. For this reason I would like
to have it published in this festive volume. The present Festschrift is an excellent
venue to shine a light on one of the many important, sometimes pioneering con-
tributions in Iranology from the person we are celebrating his 60th birthday.
In this study the synchronic and diachronic aspects of two originally partici-
ple suffixes, -ān and -andeh (Middle Persian -andag) are being discussed. I shall
attempt to account for its usage and distribution and, ultimately, to provide a
possible explanation. The explanation may also have implications for other Ira-
nian languages, which often show a similar (morphological) development as
Persian.
 Koninklijke Brill NV, Leiden, 2015
272 Johnny Cheung

USAGE AND MEANING

The employment and occurrence of -ān and -andeh in modern Persian are fairly
straightforward. The suffix -andeh forms both adjectives, on a quite productive
scale, such as košandeh ‘lethal, deadly’ (koštan/koš- ‘to kill’), āyandeh ‘coming’
(āmadan/āy- ‘to come’) and, especially agent nouns, e.g. nevīsandeh ‘writer’
(nevīstan/nevīs-), rānandeh ‘driver’ (rāndan/rān- ‘to drive (away)’), konandeh
‘maker, doer’ (kardan/kon- ‘to do, make’). On the other hand, -ān used to form
present participles, the procedure of which is now defunct. In early Classical
Persian, however, -ān perhaps enjoyed a similar productivity as -andeh. It con-
veys the following functions, which Lazard 1963: 352 defines as follows:
1. “mot circonstantiel (gérondif)”, e.g. ğaltān ğaltān bāz zīr mī oftad ‘while roll-
ing down, it fell’
2. predicate, e.g. ādamī šab-o rūz davān-o pūyān az bahr-i nān ast ‘day and night
man is running and looking for his share of bread’
3. as qualifying adjective (“epithète”), e.g. šīr-i ğorrān ‘the roaring lion’.
In the contemporary language these ān-formations function almost exclu-
sively as an adverb of manner (cf. Lazard 1992: 168), expressing an action parallel
to the principal action as indicated by the main verb, i.e. cf. 1.
In the earlier, Middle Persian period the participle in -ān functions primarily
as a substantive, according to Brunner 1977: 31 sq. It is either a verbal adjective or
a noun comparable to the agentive noun in -āg. As a verbal noun the participle
may be construed in past impersonal sentences with verb nwystn ‘to begin’. The
ān-participle can refer to the subject of the sentence and to the logical object as
well (Henning 1933: 252). The statement is illustrated by the following examples:
’wš’n pyhw’n ’wh gw’nd ‘and imploringly they shall talk to them’ (M470 V.11); ’wš
dwdy xwr ’wd m’h phryz’n nyyšyd ‘and then she saw the sun and the moon going
protectingly’. Asatrian 1983: 53 (repeated 1989: 56 f.) describes the ān-participles
as having two semantic directions, co-determining the subject. This is explained
by the fact that the subject occurs as the agent of two actions simultaneously: the
main (predicate) and the secondary, associative one (gerund). On the basis of
the examples cited above, the ān-participles would fulfil the role of adjunct of as-
sociative action (“obstojatel'stva soputstvujuščego dejstvija”) in the sentence. In
other words, the employment of -ān would be more or less identical to that in
modern Persian (see above).
Henning further states that the ān-participle can also stand in lieu of a finite
verbal form, e.g. drxt wrzn s’n’n w p’dyz ‘šmyr’n ‘the trees shall produce fruit and
 The Persian Verbal Suffixes -ān and -andeh (-andag) 273

be reckoned as autumn’ (7981 II Vii,6-12/Andreas/Henning I: 190). This is mostly

found in constructions with nwystn ‘to begin’, which may carry an infinitive as
well.
Asatrian (1989: 56) considers the participle phryz’n in the sentence ’wš dwdy
xwr ’wd m’h phryz’n nyyšyd to refer to the “object” of nyyšyd ‘she saw’, similar to
what is found in several (older) languages, e.g. Av. yat̰ spāδǝm pairi.auuaēnat̰ dū-
rat̰ aiiaṇtǝm rasmaiiō ‘When he saw from afar the army going in order’ (Yt 5.68).
Compare modern English: I saw him approaching the house.
Asatrian (1983: 53) adds that these formations have “an active meaning” (pre-
sumably contrasting to the middle voice meaning, which it used to have, ety-
mologically speaking).
As for the -andag participle in Middle Persian, its occurrence is even rarer, cf.
Rastorgueva/Molčanova 1981: 73. The suff. -andag forms mainly adjectives, which
function independently from the verb: šāyendag ‘able, worthy’ (šāyistan ‘to be
able, worthy’), bowandag ‘complete, perfect’ (būdan ‘to be, become’), zī(wa)ndag
‘living, alive’ (zī(wi)stan ‘to live’). An occasional substantive is found: wāyendag
‘bird’ (wāy- ‘to fly’). In striking contrast to modern Persian the suffix is improduc-
tive in the Middle Persian period. The semantic aspects are hardly addressed, ex-
cept in Geiger/Kuhn 1895-1904: 145, which merely states that it indicates “a con-
tinuous action”.
Since both suffixes, -ān and -andag, are (originally) participles, we may ask
ourselves whether they are interchangeable and differ in usage and meaning,
especially in view of the transition from Middle to New Persian.

ORIGIN AND DISTRIBUTION

The origin of both suffixes is clear. The formations in -ān generally derive from
the present middle, athematic participle *-āna- (cf. Geiger/Kuhn 1895-1904: 109 f.,
II/2: 146; Henning, l.c., Jensen 1931: 154; Rastorgueva/Molčanova 1981: 73). In rare
instances they may also go back to the perfect middle formations in *-āna-,
which are not considered here (e.g. MP wiyābān ‘deluded, seduced, astray’).
Pers. -andeh (MP –andag), derives from the present active participle *-ant-, to
which the common suffix *-ka- has been added. Only a few examples without
the attachment of *-ka- can be found: MP nyāzand ‘needy, indigent, beggar’’
(verb not attested), tanand ‘spider, spinner’ (tadan ‘to spin (thread)’), NP parand
‘bird’ (more freq. parandeh; parīdan ‘to fly’), čarand ‘(grazing) animal’ (čarīdan
‘to graze’), navand ‘messenger, steed’ (navīdan ‘to shake, to move, especially
274 Johnny Cheung

when rising from a place’) (Vahman/Asatrian 1987: 109). The complex suffix with
*-ka- has become dominant in the Middle Iranian period. In the first place, Par-
thian, being not only a close sister-language of Persian, but also the language,
that has exercised a huge influence, shows a similar pattern. In Parthian, both
variants may be found too, e.g. (hapax) jw’ng and, much more frequently, jywndg
‘living, alive’. In the East Iranian group, Khotanese has a present participle in -
andaa-, which is usually found in active verbs, although in LKhot. the suffix -an-
daa- is also connected with middle verbs as well. This is evidently “in accordance
with the general tendency to confuse act. and mid. outside the pres. ind.” (Em-
merick 1968: 215; cf. Also Degener 1989: 29). Finally, also Sogdian has a old act.
pres. participle *-ant- with enlarged *-ka-, which has become (Buddh.) -ntk,
(Man.) -ndty, (Chr.) -nty (cf. Gershevitch 1961: 163, § 1068). The corresponding
(athematic) middle pres. participle *-āna- has been continued as -(y)ny (older: -
n’k, -n’k), with an additional *-ka- as well (cf. Gershevitch 1961: 132, § 889 ff). An
attempt to clarify their precise use and meaning has not been given to the best of
my knowledge.
In the transition from Old to Middle Iranian the originally athematically
middle forms in -ān entered the system of active participles (Asatrian 1983: 53).
Asatrian asserts that “such matter should not come as a surprise, since already in
Old Iranian the opposition of active and middle and also active and passive was
not precisely expressed, hence the given opinion may be considered as com-
pletely probable”.1 The notion that such an opposition in OIran. was expressed
imprecise (“nečetko vyraženo”) is not correct though: in most instances, with the
exception of very late attestations, the usage of the middle voice in Old Iranian
languages, such as Avestan is justified.2 But returning to Persian -ān, it is indenia-
bly true that the convergence of middle and active had taken place at a certain
stage of Persian. The question arises whether just like in Khotanese, the opposi-
tion may have lingered in the distribution of -ān and -andeh (-andag) being con-
fined to specific verbs.

1
“Takoe položenie ne dolžno vyzyvat' udivlenija, poskol'ku uže v drevneiranskom protivopo-
stavlenie aktiva i medija i daže aktiva i passiva bylo nečetko vyraženo, v silu čego dannoe mnenie
možno sčitat' vpolne verojatnym.”
2
Cf. Kellens 1984: 327: “Il [i.e. the pres. mid. participle] est pourvu, par le thème verbal dont il
dérive, d’une connotation temporelle et, par la nature de son suffix, d’une signification diathéthique
[italics are mine]”.
 The Persian Verbal Suffixes -ān and -andeh (-andag) 275

I. Attestation―the Middle Persian Evidence

In order to address these two questions we need to look into the attested exam-
ples from both the Middle and New Persian stage. Let us first look at the Middle
Persian data.3
1. The number of -ān-participles is relatively limited, in both Manichaean Per-
sian and Book Pahlavi. Three main categories of verbs can be distinguished:
i. The verb nwn- ‘to begin, start’ is usually followed by a participle in -ān, e.g.
- ‘spyz’n ‘blooming, blossoming’: .. k’ nwnyyd grm’g ’cyš prwd wyš’hyd 00 w: drxt
‘spyz’n ’wd wh’r ‘šmyr’n ’wd h’mšhr ’wšyb’m bw’n ‘... when, through it, the warmth
is unleashed downwards, and (when) the trees are blossoming, the spring is
being calculated and the whole realm is becoming dawnlike’ (Andreas/Henning
I: 191)
- ‘šmyr’n ‘being reckoned, calculated’: ... qš’’n nwnyd srd’g ‘zyš prwd wyš’h’n 00 w:
drxt wrzn s’n’n w: p’dyz ‘šmyr’n ’wd h’mqyšwr xwrwpr’n bw’’n ‘... when, through it,
the cold begins for them to be unleashed downwards and (when) the withering
trees are raising up and the autumn is being calculated and the whole clime is
becoming nocturnal’ (Andreas/Henning I: 190).
- bw’n (bw’’n) ‘being’: ... ’wd ’c s’r ’w s’r nwyst mhy bw’n ‘... and from year to year
she [s.c. soul ?] started to become bigger’ (Sundermann 1973: 34 f.); .. k’ nwnyyd
grm’g ’cyš prwd wyš’hyd 00 w: drxt ‘spyz’n ’wd wh’r ‘šmyr’n ’wd h’mšhr ’wšyb’m
bw’n ‘... when, through him, the warmth ... down(wards), und (when) the trees
are blossoming, the spring is being calculated and the whole realm is becoming
dawnlike’ (Andreas/Henning I: 191).
- hmwwc’’n ‘teaching, learning’: ’’wn ps’c dwdy nwyst ’’z ’wyn mzn’n ’wd ’sryšt’rn
’b’ryg’n nr’n ’wd m’yg’n .... ’wyš’nz hmgwng’wzm’h ’wd mrz‘yyšn hmwwc’n... ‘then,
again, Greed started to teach also these remaining Mazans and Asreshtars ...,
men and women, in the same way, the lust and copulation...’ (Andreas/Henning
I: 194).
- kwn’n ‘doing’: ’wd ’c dry’b b’ ‘škrwst ‘wš nwyst ’ndr šhr wyn’h kwn’n ‘and he stag-
gered from the sea and he started to commit sins on the land’ (ibid.: 181).
- MMP xwyr’y’n ‘preparing’: ’wd ’wyc zn nwyst h’mr’st xwyr’y’n ‘And each woman
started to prepare in the same way’ (Sundermann 1973: 90).

3
The Manichaean examples are mostly cited from texts that have been published. The pub-
lished Pahlavi texts from which the forms have been cited, are: Ardā Wirāz Nāmag (AW), Ayādgār ī
Wuzurg-Mihr (Ay.Wuz.Mihr), Bundahišn hindī (Bd.), Dādestān ī Dēnīg (Dād.Dēn), (H)andarz ī
Dānāgān ō Māzdēsnān (Hand.Dan.Mazd.), (H)andarz ī Dastwarān ō Weh-Dēnān (Hand.Dastw.),
Kārnāmag ī Arda(x)šīr ī Pābagān (KAP), Mēnōg ī Xrad (MX), Wizīdagīhā ī Zādspram (WZ), the Per-
sian Rivāyat accompanying the Dādestān ī Dēnīg (PR-Dād.Dēn). The Book Pahlavi forms, with a
few exceptions, are transcribed according to the system of D. N. MacKenzie.
276 Johnny Cheung

- xw’h’n ‘wanting, desiring’: š nwyst g’m xw’h’n ‘it [i.e. Greed] began to wish for a
step’ (Andreas/Henning I: 193).
- xyz’n ‘standing, rising’: ’wd nwyst hynd ’br zmyg xyz’n ’wš’n b’r ’wd myw ’c drxt’’n
xwrd ‘They started to rise on the earth. And they ate the harvest and fruits from
the trees’ (ibid.: 183).

ii. Verbs of Motion

- dawān/dww’n ‘running’: čiyōn-aš dīd kū pad zōr ud hunar ud nērōg ud šāyend-
agīh būd ī xwad hēzag az čāh ul āhixt dawān ō pēš šāhpuhr mad’ ‘No sooner did
she see this than she, with the strength, skill, and vigor that were purely estab-
lished (in her), drew up the bucket (full of water) from the well, and went run-
ning to Shahpuhr’ (KAP xvii.13 / Sanjana: 51); w mrdwhm’n ... dww’n pdyš ’wy
šw’nd ’wd nm’c br’nd ‘And the men ... will go running before him and pay hom-
mage (M473 c,11(36) ff./MacKenzie 1979: 504 f.).
- rawān [SGYTWN-’n] ‘current, running’: ēn ruwān ī ōy druwand mard kē pad
gētīg sar ud rōy ī xwēš ud dast ud šabīg ud abārīg rēmanīh ī handām ī xwēš pad
āb ī ēstādag ī wuzurg ud xānīg <ī> rawān was šust... ‘This is the soul of that
wicked man who in the world often washed his head and face and hands and
shirt and other filthy <parts> of his body in large pools or running water ...’
(Vahman 1986: 151, 209).
- tāzān/tz’n ‘running’: ud pas āb-kirb ud urwar-kirb pad ēwēnag ī bōyestān homā-
nāg be ēstēd <ī> čašmag čašmag awiš tāzān <ud> az-iš frāz tazēnd... ‘Et la forme
d’eau et la forme de plante sont semblables à une sorte d’(endroit) semblable à
un jardin dans lequel coulent différentes sources, et d’où s’écoulent...’ (Gignoux/
Tafazzoli 1993: 110 f.); w’d ’b w ’<dwr ...> tz’n ‘wind, water and <fire will cease ?>
running’ (damaged, M482 a a V.1(168) f./MacKenzie 1979: 508 f.).
- kešān/kš’n ‘drawing, pulling’: ēg-im dīd ruwān ī mard-ē kē kešān ō dušox nayēnd
ud hamē zanēnd ‘Then I saw the soul of a man whom they were dragging to
Hell and continuously beating.’ (Vahman 1986: 151, 209).

iii. Verbs of Emotion

- brāmān ‘weeping’: u-m dīd ruwān ī zan-ē kē griyān ud brāmān hamē āmad <ud>
šud ‘And I saw the soul of a woman who was coming and going, crying and
weeping’ (Vahman 1986: 155, 210 f.).
- griyān/gryy’n ‘weeping’: ’wd gryy’n ’c pyšyh <’wz>yd hynd (Sundermann 1981:
24.3); ēn ruwān ī ōy druwand zan kē kōdak ī xwēš niyāz <ud> gursagīh rāy griyān
hišt ‘This is the soul of that wicked woman who left her own child crying in
need and hunger.’ (Vahman 1986: 151, 209); u-m dīd ruwān ī zan-ē kē griyān ud
brāmān hamē āmad <ud> šud (ibid.: 155, 210 f.).
- garzān ‘se lamentant’: u-š Gōšurwan az pas garzān hamē šud u-š wāng burd kū-t
sālārīh ī pad dāmān pad kē be hišt ‘et Gōšurvan allait derrière lui en se lamen-
 The Persian Verbal Suffixes -ān and -andeh (-andag) 277

tant, et lui cria: «À qui as-tu abandonné la souveraineté sur les créatures ?»’
(Gignoux/Taffazoli 1993: 36, f.).
- nālān ‘crying, lamenting’: u-m dīd ruwānān kē-šān azēr ī pāy <ī> gāwān abgand
ēstād hēnd pad srū zad ud aškomb darrīd ud astuxwān be škast ud nālān būd
hēnd ‘And I saw the souls of those who were thrown under the hooves of cows.
<The cows> gored them with their horns, tore their bellies and broke their
bones, and they were lamenting’ (Vahman 1986: 165, 213).

iv. The other Verbs that do not belong to the aforementioned Categories are:
- ’’stw’’n ‘professing, pledging oneself to’: ’wd pd tw n’m ’’stw’n hwm (m’m ws’n oo
wc’rm’n ’c my’n bzd’n ’wm’n ’c my’n ’bd’g’n ‘bd’c) ‘And we are confessing our-
selves to Thy name. (Separate us from the midst of the bad and redeem us from
the midst of the attackers !)’ (Andreas/Henning II: 315).
- arzān/ ’rz’n ‘valuable, worthy’: mowbedān mowbed guft kū: anōšag bawēd, ō
kāmag rasēd, ōy kē pad gyān ī xwadāyān kōxšēd marg-arzān ud be ōzanišn ‘The
mobed [of mobeds] said: “May you be immortal! May you attain to your goal!
She who strives after the life of her lord is worthy of death (deserves to die) and
should be killed” (KAP xiv:16): čē agar man marg-arzān ham ēn frazand ī andar
škamb dāram pad marg-arzān abāyēd dāštan? ‘because if I am worthy of death,
this offspring that I have in my womb should not also be regarded as worthy of
death.’ (KAP xiv:20); passim.
- drāyān ‘chattering, (the sin of) speaking while eating’: ēn ruwān ī ōy druwand
mard kē-š pad gētīg Hordād ud Amurdād āb ud urwar drāyān jūd ud a-dādīhā
xward ud wāz nē dāšt... ‘This is the soul of that wicked man who while chatter-
ing devoured the water and the plant of Hōrdād and Amurdād, ate unlawfully
and did not keep the bāj ...’ (Vahman 1986: 125, 203).
- frāmōšān ‘forgetting’: tan hangār kū-m widerēd kār ī gētīg, ān ī grāmīg nāzuk
kālbod barēnd pad gyāg frāmōšān, ānōh an-ayād bē nihēnd ‘Oh body, consider:
when the wordly affairs pass me by, they carry the body that is dear and tender
to the place, being oblivious, there they shall put [the body] without remem-
brance’ (Hand.Dan.Mazd. 16).
- ? qxs’n ‘bending, turning ?’4: <qyrb>kr’n qxs’n d’hw’<’n> d’d ‘the Pious [pl.] gave
bending(ly) presents ?’ (Henning 1943: 62).
- ? nišīnān [YTYBWN’n] ‘sitting’: ka-š ranjagīhātar ka-z hamāg nišīnān gowēd ā-z
nē tar-menišnīhā ‘if (it is) more troublesome (still) for him even if one recites
everything sitting, even then (it is) not a disrespect(ful act)’ (Williams 1990: 207
(1), 98 (2)).

4
The meaning is not given by Henning, l.c. Perhaps it is an inchoative pres. formation of *kauč-
‘to bend, hang on, suspend’, cf. Oroshorvi kaxs-/kaxt ‘to curl up, turn’ (Morgenstierne 1973: 42a).
278 Johnny Cheung

- swc’n ‘burning’: swcyšn ‘y xyšmyn .... <’nd>r <’>brg w: xwr’s’n ‘yrg w: x<wr>nwr
<’w>d b’l’y w: zwpryh <phn’y> w: dr’z’y swc’n rw’d ‘And that raging fire ...., that
will go burning in north and east, south and west, and in the height and depth
<breadth> and length...’ (MacKenzie 1979: 514 f.).
- ? wys’n ‘coming to rest ?’5: ’wd xwrxšyd ’wd m’h ’wd yzd’n hsp’n ’wd wys’n bw’d
‘and the sun and the moon and gods shall come to rest’ (Andreas/Henning I: 19)
- widerān ‘passing away, dying’: ka ham zamān widerān bawēd ēg-iš bīm nēst... ‘if
he dies at the same time, even then there is no fear ...’ (Williams 1990: 223 (1),
107 (2)).

2. As mentioned above, the number of -andag-participle derivatives is even

more limited.
i. The most frequently attested forms are šāyendag ‘able, worthy’ (šāyistan/
šāy- ‘to be able, worthy’) and zīndag/jywndg ‘living, alive’:
- šāyendag ‘able, worthy’: pad kas-iz kas-afsōs mā kunēd čē afsōs-gar mardafsōs-
bar zad-xwarrah, nifrīdag bawēnd u-šān frazand i šāyendag ī arteštār kam bēd.
‘Do not make mock of anyone at all; for the man who mocks will himself be
mocked, will lose his dignity (khwarrah) and be execrated; and rarely indeed
will he have a decent or warlike son.’ (Cid.Hand.Por. 43; Zaehner: 26); šāyendag
ka ō a-šāyendagīh rasēd ud pādyāwand ka ō a-pādyāwandīh rasēd, ummēdwār
ka ō an-ummēdīh rasēd ud xwad-dōšag ka ō frazām ī kār mad bawēd. ‘[He is]
worthy when he reaches worthlessness and strong when he reaches weakness,
hopeful when he reaches hopelessness and self-indulgent when he has come to
the conclusion of the work’ (Ay.Wuz.Mihr 158); passim
- zīndag/jywndg ‘living, alive’: ‘st’wyšn ’w tw gryw jywndg q’dwš k’dwš bg m’ry m’ny
‘Praise to Thee, Living Self, holy, holy god, my Lord Mani’ (Waldschmitt/Lentz
1926: 126); ky šhyd ‘w tw ‘st’w’d ’wnglywn jywndg pscg ‘w tw ‘st’yšn jyryft yzd’n
‘Who can praise thee, Living Gospel? The wisdom of the gods is fitting for thy
praise’ (Sims-Williams 1989: 330 f.); passim.

ii. The remaining examples are:

- wārendag ‘raining, rainy’ (wārīdan/wār- ‘to rain’); ud hast ī gyāg ud rāyēnāg ud
nišēm ud awestām ī abr ī wārendag ‘there are some which are the place and
vent, the resting-place and support of the rainy cloud’ (MX 56.5/ West 1885: 98).
- wāyendag ‘air-(borne); bird’ (= JP w’ynd’ ; wāy- ‘to fly’): ud harw ēwēnag tōm ī
hamāg dām ud dahišn ī ohrmazd ī xwadāy az mardōm ud stōr ud gōspand ud
wāyendag, harw čē wehtar ud pad-wizēntar, ō ānōy burd ēstēd ‘And every species
and seed of all the creatures and creations of Ohrmazd, the lord, whatever is

5
The meaning is not entirely certain.
 The Persian Verbal Suffixes -ān and -andeh (-andag) 279

better and more select of man and beast of burden, of cattle and flying creatures
is brought thither’ (MX 62.16/ West 1885: 109); ... sidīgar ēwēnag ān ī panj-an-
gurāg panj<ag>, kē az awēšān sag meh, mušk farxaw nidom; čahārom ēwēnag
wāyendag, kē az awēšān sēn ī sē angurāg meh, ud tarw nidom ‘The third genus is
that of the five-dividing paw, of which the dog is the largest, and the civet-cat
the least. The fourth genus is the flying, of which the griffin of three natures is
the largest, and the chaffinch the least.’ (Bd. 13.39, l. 13 ff./ West 1880: 47); etc.
- ? xdyndg ‘injuring’: context unclear (Sundermann 1981: 175b).

3. Other -andag-formations also have a corresponding ān-participle:

- bowandag ‘complete, entire’ (verb. būdan/baw- ‘to be(come)’); paywast-axwīh
bowandag-ham-wēnišnīh spurīgīhā-dānišnīh ud purr-xwarrahīh ī ān haftān fra-
šagird-kardārān ēdōn abd kū az kišwar be ō kišwar har(w) ēk ō 6 āgenēn ēdōn
ham-pursēnd ... ‘The like-mindedness, perfect agreement, complete knowledge
and gloriousness of these seven renovation-makers are so wonderful that they
each consult with the six others, from continent to continent ...’ (Jaafari-Deha-
ghi 1998: 108 f.); pas pad dādīg abāz zanišnīh rāy bārīgīhā be ō dēwān wēmārīh
ud margīh be ō dām ī yazdān ašōgān bēšazišnīh zīndagīh ī bowandag ‘After-
wards, on account of lawfully smiting (i.e. druz) back, (he gave) subtly the sick-
ness and death to the demons and healing and perfect life to the righteous crea-
tures of yazdān.’ (Jaafari-Dehaghi 1998: 136 f.); ēk ēn kū ān ī ōstīgāntom gōhr ī
dēw [ī] xwad ast tārīkīh ī-š wattarīh owōn bowandag ī dēwān-iz az tom tōmagān
xwānānd ‘One is this, that the strongest substance of the demon, which itself is
darknessm (is) his evil; so all the demons are called those of the dark families’
(Jaafari-Dehaghi 1998: 136 f.); passim.
- griyendag ‘lamenting’: u-š guft kū ō garrānāg ud wāng ī xar wāng ī šagr wāng ī
uštar ī dēnōdag wāng ī griyendag wāng ī mard ī ahlaw... ‘And he said: ‘(It is sim-
ilar) to thunder and the cry of an ass, and the cry of a lion, and the cry of a fe-
male camel, the cry of the lamenting voice of the righteous man ...’ (Williams
1990: 145 (1), 62 (2)).
- swcndg ‘burning’: ky prystynd ’dwr swcndg ’c ’ydr xwd d’nynd... ‘who tend the
burning fire, from here they self (can) acknowledge ...’ (Müller 1904: 94).
- waxšendag ‘blazing (fire)’: ōy ī gētīg-ārāy ud mēnōg-wišōb mard andar pādifrāh ī
stōš ēdōn tabāhīhēd čiyōn ātaxš ī waxšendag ka+š āb abar rasēd. ‘he who is a
world-arranging and spirit-destroying man is as injured, in the punishment of
the three nights ..., as a raging fire when water comes upon it’ (MX 21.10/West
1885: 51).
- xxyzyndg ‘crawling around’: ... wd ky xxyzyndg ’wd dryst ‘n’nd ‘... and who/which
[s.c. Mazan-demons ?] were crawling around and healthy’ (previous passages
damaged, partially restored, Sundermann 1973: 40).
280 Johnny Cheung

4. Some observations.
We can make the following observations on the basis of the attested forms. In
the first place, the participle in -ān is used in certain syntactical constructions,
viz. as predicate complement of niwist ‘to begin’ and auxiliary verbs, such as
būdan ‘to be’, or rawastan ‘to go’, as already observed by Asatrian and Henning.
This should not surprise us, since the present (middle) participle is used with
these comparable auxiliary verbs to express continued duration, similar to, for
instance OP DNb15 uvaipašiyahyā daršam xšayamna ami ‘I am ruling very much
over my own possessions’, Vedic RV 2.40.5 víśvam anyó abhicákṣāṇa eti ‘the other
(Pūsan) goes on watching the universe’ (cf. Delbrück 1888: 390 f.). The corre-
sponding andag-participle appears to lack the function of predicate in most in-
stances. Secondly, it is striking that in the category of motion and emotion the
participle in -ān is most frequently found. However there are a few correspond-
ing andag-participles found in this category, viz. ’dwr swcndg ‘the burning fire’,
griyendag wāng ‘the lamenting voice’ and ātaxš ī waxšendag ‘the raging fire’. In
these attestations the participle refers to an inanimate noun. Another interesting
aspect is that the participle of widerdan ‘to pass away, die’ is in -ān, but the verb
zī(wi)stan ‘to live’ has a corresponding participle in -andag / NP -andeh.
How do we account for this distribution? The only way to explain this is that
it depends on the diathesis applied, in other words, the middle must have been a
functional category at an earlier stage. The tell-tale signs are there.
The middle voice that -ān used to express, emphasizes the subjectivity of the
internal state or the action of the qualified.6 Certain semantic categories of verbs
tend to be attested in this voice. Spontaneous processes, such as dying, tend to
be in the middle voice. In Avestan the verb for ‘to die’, √mar-, is attested exclu-
sively in the middle voice, whereas ‘to live’, √juua-, only in the active. This is mir-
rored in Vedic (pres. 3sg. mriyáte, but act. jīv́ ati) and other IE languages (Latin
1sg. morior vs. act. vivō). It is also understandable that with an inanimate subject,
voice is referred to by an (originally) active participle in -and(ag): subjectivity is
not likely to be emphasized. It is worth citing Gonda 1979: 102: “The actives ‘gov-

6
Cf. Lyons 1969: 373. This study does not intend to give an exhaustive survey of all the catego-
ries and functions that the middle voice may entail. It suffices here to mention two notable publi-
cations, Stempel 1996 (from an IE perspective) and Kemmer 1993 (from a typological point of
view), some of the definitions and examples have been adopted here. In addition, I also need to
mention the standard work of Gonda 1979 on the middle voice in (Rig-)Vedic, which has been
quoted in several instances.
 The Persian Verbal Suffixes -ān and -andeh (-andag) 281

ern’ objects denoting ideas with which one generally speaking has no normal,
personal or intimate relations”. Another case is arzān/šāyendag ‘worthy’: the ap-
plication of the middle voice to arzān is clear, since arzān refers to the intrinsic
value of an object, whereas šāyendag alludes to the external appearance (or be-
haviour).
It is perhaps no coincidence that ’’stw’’n ‘praising, praying to, professing/
pledging oneself to’ and framōšān ‘forgetting’ do not have a corresponding -and-
(ag). The Avestan cognate of ’’stw’’n, āstao- ‘to profess oneself to, pledge oneself
to’, has no active forms either (i.e. they are media tantum): the use of the middle
refers to the interested motive from the subject (cf. Gonda 1979: 164 f.). Compare,
for instance, Lat. fateor ‘I admit, acknowledge, profess’, which has passive end-
ings only (i.e. a deponens). As for framōšān, the Vedic correspondence √marṣ- ‘to
forget’ is almost exclusively attested in the middle voice: the process of ‘forget-
ting’ is usually something that cannot be verified factually or objectively (i.e.
bears no external relation or effect). Outside Indo-Iranian, we may cite the Lat.
deponens obliviscor ‘I forget’ as comparison.
The attestation of numerous ān-participles denoting (strong) emotion and
movement should not come as a surprise either. This is in line with a (universal)
tendency to express emotion and (translational) motion in the middle (or “pas-
sive”) voice, which, after all, expresses the feedback to the subject, being in such
a state. This is the reason that Av. garəz- ‘to lament’ and its Skt. cognate garh- ‘to
complain’ are media tantum. A prime example of a verb of movement is Ir.
*čyaw- ‘to go’,7 which is attested exclusively in the middle in Gatha-Avestan
(pres. ind. 3sg. ṧauuaitē Y 29.3, subj. 1sg. ṧauuāi Y 33.8) and its Indo-Aryan sister
language Vedic cyav-. The old (Indo-European) middle character of this root is
supported by the Greek cognate seúomai (medium tantum) ‘I am in commotion,
rush to’ and probably also Armenian č’ogan ‘they went’. Natural to the middle,
these verbs are generally intransitive, being closely associated with the subject.
An interesting shift in meaning shows bowandag, which derives from būdan
‘to be, become’. The corresponding ān-participle bw’n is, as expected, durative
with the meaning ‘becoming, being’, whereas bowandag means ‘complete’. Such

7
Remarkably enough, as to the Old Persian cognate šiyav-, only the active endings have been
attested. Although it is only natural that an active voice arises in due course, in this case, it may
also refer to a different aspect as well: ‘to set, go forth’, cf. Greek kínomai ‘to move [intr.]’, act. kíneō
‘to set in motion’. Not the actual, translational movement is envisaged, but the start or com-
mencement of the movement.
282 Johnny Cheung

a difference may be explained by the voice applied, for which we can compare
Greek mid. pres. téllomai ‘I come into being, am’, fut. télomai ‘I shall be’ vs. (act.
pres.) téllō ‘I fulfil, accomplish’. The focus of bowandag is similar to that of šāyen-
dag, not internal but rather external: the result of the “durative state” is en-
visaged.8
Although verbs of speaking do not necessarily entail the middle voice, drāyān
‘chattering, (the sin of) speaking while eating’ is a special case, as it draws the at-
tention to the (unlawful) state in which the subject is.
In the other cases a justification for the application of the middle voice is not
apparent. The passage in which swc’n ‘burning’ is attested is damaged, whereas
the meaning or the form of the other instances is disputable.

II. Attestation―the New Persian Evidence

Let us now look at the New Persian forms. I have limited myself to one single
corpus, viz. the Šāhnāmeh, since it is the largest and oldest (preserved) source of
early New Persian literature. Many more ān- and andeh-formations can be found
in the Šāhnāmeh epic than ever before. Moreover, we find a substantial number
of -ān/-andeh “doublets”.
1. The following -ān formations are attested, as can be gleaned from Wolff 1935:
i. from verbs of motion: ii. from verbs of emotion:
- bazān ‘blowing’ - biryān ‘roasted, tormented’
- čamān ‘treading’ - danān ‘boiling with emotion (joy,
- čarān ‘grazing’ esp. anger)’
- gorāzān ‘entering’ - geryān ‘weeping’
- jahān ‘galloping’ - jūšān ‘boiling [emotionally]’
- rīzān ‘flowing away, pouring out’ - larzān ‘shivering, shaking’
- šetābān ‘hastening’ - navān ‘lamenting’
- tāzān-tar ‘running faster’ - pīčān ‘twisting [especially in
emotional sense], tormenting’
- xandān ‘laughing’
- xorūšān ‘shouting’
iii. from verbs of activity:
- konān ‘doing’

8
In the case of Gr. téllō, the factitive-transitive result is due to the shift of the emphasis to the
goal, viz. the object.
 The Persian Verbal Suffixes -ān and -andeh (-andag) 283

2. The formations in -andeh are much more diverse:

i. from verbs of motion: ii: from verbs of giving:
- borandeh ‘carrying, carrier’ - baxšandeh ‘generous, bestowing’
- pāyandeh ‘firm, steady’ - dehandeh ‘giver’
- parastandeh ‘serving, servant’
iii. from verbs of miscellaneous activities:
- porsandeh ‘asker’
- bīnandeh ‘eye’ (lit. ‘looker’)
- parvarānandeh ‘feeder, nourisher’
3. In addition there are also some verbs that derive both formations.
i. seven “doublets” denote movement:
- davān ‘running’ / davandeh ‘runner, messenger’
- gardān / gardandeh ‘turning to, moving’
- gorīzān ‘fleeing’ / gorīzandeh ‘escaping, escaper’
- jahān / jahandeh ‘jumping, galopping’
- jombān ‘moving [intr.]’ / jombandeh ‘mover’
- pūyān ‘walking, running’ / pūyandeh ‘walking (animal); messenger’
- ravān ‘flowing’ / ravandeh ‘going, messenger’
ii. The remaining participles are:
- (tīğ-e) borrān / borrandeh ‘cutting (sword)’
- damān ‘screaming, loud’/ damandeh ‘screaming, snorting, rushing, hurrying’
- gūyān / gūyandeh ‘talking, speaking’
- ğorrān / ğorrandeh ‘roaring’
- roxšān ‘shining’ / roxšandeh ‘lighting up’
- tābān / tābandeh ‘shining’
- xorūšān/xorūšandeh ‘shouting’
4. In adjectival constructions both forms occur. In many instances, there is no
apparent distinction in usage between the ān-formation and its correspondence
in -andeh. The active meaning which the andeh-formations originally possessed
may have survived, being increasingly employed as agentive noun, whereas the
middle meaning is sometimes clear or understandable, expressing the intimate
connection between the subject (or qualified noun) and the qualifying ān-parti-
ciple, to the exclusion of a resultative, external effect.
We can illustrate this distinction with some doublet forms. It is impossible to
treat all the attested forms from the Šāhnāmeh, which would go far beyond the
scope of this article. A representative selection of forms has been given, which
may nevertheless give us some good indications of its meaning and employment
though.
284 Johnny Cheung

Since I have principally relied on Wolff 1935, Mohl’s edition of the Šāhnāmeh
is the reference in quoting passages. For an easier tracking of the passages, I have
also added the annotation of the two most recently published editions (if pre-
sent or traceable): Bertel’s and Motlağ. Some of the passages may have to be de-
leted or revised as the result of the new findings. But as far as I can assess, no
significant deviation in the use of the participles in -ān and -andeh in these,
probably later interpolations, can be noted.
a. (tīğ-e) borrān / borrandeh ‘cutting (sword)’:
i. keh man tīğ-e borrān nagīram bedast / garāmī tan-ū na-xvāham be-xast ‘je ne
prendrai pas une épée tranchante, je ne veux pas blesser son noble corps’ (Goš-
tāsp 3364 /Mohl 4: 322/516 f.;9 Motlağ 5: 373, l. 974;10 Bertel’s VI: 276, fn.11).
ii. zabāneš be-kardār borrandeh tīğ čo daryā bar-ū kaf čo bārandeh mīğ ‘..sa
langue qui était comme une épée qui déchire, comme la mer qui écume, comme
le nuage qui versa la pluie.’ (Nowzar 96/ Mohl 1: 196/308; Motlağ 1: 291, l. 84;12
Bertel’s II: 11, 84).
In these passages the middle/active distinction of borrān and borrandeh is
clearly illustrated. In the first passage the quality of the sword, tīğ, is stressed, i.e.
having the ability to cut. In the second passage the sword is compared to the
tongue of Pašang who has just offended Afrasiyāb with his words: the effect,
which Pašang’s tongue had on Afrasiāb is thus like the cutting effect of a sword.
Compare MP arzān vs. šāyendag.
b. davān ‘running’ / davandeh ‘runner, messenger’.
i. davān mādar āmad sū-ye marğzār / čonīn goft bā mard-e zenhārdār ‘... la mère
arriva en courant au jardin, et dit au protecteur ...’ (Zohhāk 142/Mohl 1: 41/58;
Motlağ 1: 63, l. 136; Bertel’s 1: 58, l. 133).
ii. bedeh dād man k’āmade-stam davān / hamī nālam az to beranj-e ravān ‘Rends-
moi justice; je suis venu en hâte, et c’est toi que j’accuse dans l’amertume de
mon âme.” (Zohhāk 212/Mohl 1: 44/62; Motlağ 1: 67, fn.; Bertel’s 1: 62, fn.).
- davān āmad az bahr-e āzār-e tān / hamān ārzūmand dīdār-e tān ‘... accourt au-
devant de vous à cause de votre affliction; et dans son désir de vous voir...’
(Fereydūn 467/Mohl 1: 77/117; Motlağ 1: 117, l. 444; Bertel’s I: 100, l. 351).

9
The page number after the slash refers to the French translation of Mohl 1876-1877. The trans-
lation merely serves as a reference of discussion.
10
To be revised to: keh man tīğ-e hendī nagīram bedast / sar-e tīr-o zeh tā bebandam bedast.
11
To be revised to: keh man tīğ-e hendī nagīram bedast / čo sāzam bar-ān kūh(e)peykar nešast.
12
Second mesra‘ : čo daryā del-ū kaf čo bārandeh mīğ.
 The Persian Verbal Suffixes -ān and -andeh (-andag) 285

- čo dīdand pormāyegān rū-ye šāh / peyādeh davān bar gereftand rāh ‘Lorsque les
princes illustres virent la face du roi, ils s’avancèrent vers lui à pied et en cou-
rant’ (Fereydūn 258/Mohl 1: 69/102; Motlağ . 1: 104, l. 241).
The participle davān often occurs with āmadan ‘to come’. This combination
is syntactically comparable to the construction with MMP nwystn ‘to begin,
start’. We may notice the actual running in which state the qualified subject is
taking place.
Compare these passages to the following instances.
- saxonhā ze har gūne-ī sāxtand / hayūnī tagāvar berūn tāxtand // davandeh hamī
tāxt tā nīmrūz / čo āmad bar-e zāl-e gītī ferūz ‘Ayant ainsi considéré la question
sous toutes ses faces, ils expédièrent un dromadaire de course. Le messager
s’élança et courut jusqu’à ce qu’il eût atteint le Nimrouz; et lorqu’il fut arrivé
devant Zal la lumière du monde’ (Key Kāvūs 68/Mohl 1: 236/388; Motlağ . 2: 6, l.
62; Bertel’s II: 79, l. 63).
- gar eydūn keh rāy-e šekār āyadet vo yūz-e davandeh bekār āyadet ‘si tu as envie
d’aller à la chasse, et si tes guépards aux pieds légers sont prêts ...’ (Key Kavūs
556/Mohl 2: 26/39; Motlağ . 2: 104, l. 1813).
- bey-ārīd goftā yekī pīl-e narr navandī davandeh čo morğī be-parr ‘ ... disant:
“Amenez un éléphant mâle qui coure comme un oiseau volant à tire d’aile” ’
(Goštāsp 986/Mohl 4: 222/353; Motlağ . 5: 168, l. 97514; Bertel’s VI: 132, l. 97115).
The formation davande in these passages acts as an agent, whose external be-
haviour is the main focus.
c. gūyān / gūyandeh
i. beraftand gūyān be īvān-e šāh čo xoršīd bar čarx-e gom kard rāh ‘En parlant ainsi
ils arrivèrent au palais du roi au moment où le soleil disparaissait de la voûte du
ciel.’ (Behrām Gūr 959/Mohl 5: 316/506; Bertel’s VII: 349, l. 77816).
- hamī rānad gūyān be-moškū-ye xvīš be-sū-ye botān-e samanbūy-e xvīš ‘Il alla en
causant jusqu’à ce qu’il arrivât dans l’appartement de ses femmes, auprès de ses
idoles au parfum de jasmin.’ (Behrām Gūr 1137/Mohl 5: 324/518; Bertel’s VII: 359,
l. 951).
- dabīrān-o zarvān-o dastūr-e šāh / beraftand yek rūz gūyān be-rāh // saxon raft
čandī ze afsūn-o band / ze jādū-vo āharman-e porgazand ‘Un jour les scribes,
Zerwan et le Dastour du roi cheminaient sur la route en conversant; on parla

13
To be revised to: gar eydūnak rāy-e šekār āyadet / keh yūz-e šekārī be kār-e āyadet.
14
Second mesra‘: davandeh navandī čo morğī be-parr.
15
Second mesra‘: davandeh parandeh čo morğī be-parr.
16
Second mesra‘: yekī goft xoršīd-e gom kard rāh.
286 Johnny Cheung

longuement d’incantation et de sorcelleries, de magiciens et d’Ahriman le per-

nicieux’ (Nūšervān 1700/Mohl 4: 151/238; Bertel’s VIII: 151, l. 164817).
ii. to nīz āfarīn kon keh gūyandeh-ī / bed-ū nām-e jāvīd-e jūyandeh-ī ‘Rends-lui
hommage, toi qui sais parler et qui cherches par lui un nom immortel’ (Āğāz
212/Mohl 1: 13/16; Motlağ . 1: 17, l. 185; Bertel’s I: 26, l. 203).
- čonīn goft gūyandeh bā pehlevān / keh az kāx-e mehrāb rowšan ravān // paras-
tandegān-rā sū-ye golestān / ferastad hamī māh-e kābūlestān ‘Celui à qui il avait
parlé lui répondit: “Ce sont des esclaves que la lune du Kaboulistan aura en-
voyées du palais de Mihrab à l’âme brillante, dans le jardin de roses.”’ (Manū-
čehrī 505/Mohl 1: 128/197 f.; Motlağ . 1: 191, l. 395;18 Bertel’s I: 164, l. 42319).
- bedīšān čonīn goft k’īman šavīd / ze gūyandeh goftār-e bad mašnavīd ‘Ayez con-
fiance et ne croyez pas ceux qui vous diraient de mauvaises paroles’ (P. Key Xos-
row 1475/Mohl 4: 65/102; Motlağ . 4: 263, l. 1446; Bertel’s V: 321, l. 1434) etc.
Similar to davān/davandeh, gūyān is often constructed with raftan ‘to go’ or
āmadan ‘to come’: gūyān refers to the state in which the subject occurs, viz. the
state of talking (to each other in a plural subject). Gūyandeh on the other hand
indicates the external action, which is the uttering of goftār ‘words, speech’.
d. roxšān/roxšande ‘shining’
Boths formations are attested abundantly. Again, it is not always comprehensi-
ble or explicable why ān-participle or its correspondence in -ande is used. The
epithetic employment of roxšān and roxšandeh is rather fixed or stereotypical. In
several passages we may provide a plausible explanation for a particular usage,
especially if the context is clear.
Compare the following passages:
i. In several passages the use and meaning of roxšān clearly transpire:
- yekī manzarī pīš-e īvān-e xvīš / barāvord čūn taxt-e roxšān-e xvīš // be-meydān
šodandī do dāmād-e ūy / bey-ārāstandī del-e šād-e ūy ‘[Le Kaisar] fit construire
devant son palais une tribune qui ressemblait à son trône brillant; ses deux
gendres se rendirent au cirque, et réjouirent son coeur enchanté’ (Lohrāsp 607/
Mohl 4: 166/265; Bertel’s VI: 45, l. 599).
The fact that the portico of the Qeysar is compared to his shining throne does
not carry any further consequences nor does the shining throne exhibit any ex-
ternal relations.

17
Second mesra‘: beraftand yek rūz pūyān be rāh.
18
Second mesra‘: ferastad hamī māh-e kāvūlestān.
19
Second mesra‘: ferastad hamī māh-e kābāstān.
 The Persian Verbal Suffixes -ān and -andeh (-andag) 287

- konūn mī gosārīm tā čāk-e rūz / čo roxšān šavad tāj-e gītī forūz // naxostīn be-
šamšīr šīr afganīm / ham ān aždahā-ye delīr afganīm ‘Nous allons boire jusqu’à
l’aube du jour, et quand la couronne du soleil qui éclaire le monde brillera, nous
abattrons d’abord les lions avec l’épée, nous abattrons ces vaillants dragons..’
(Behrām Gūr 1310/Mohl 5: 331/530; Bertel’s VII: 369, l. 1120).
The sun, tāj-e gītī-forūz ‘the crown, the world-enlighter’, refers to a state of
roxšān ‘shining’, which, again, does not further affect the next (planned) actions,
described in the subsequent line: čo būdī sar-e sāl-e now farvardīn keh roxšān šodī
dar del az hūr-e dīn ‘Lorsque la nouvelle année commençait au mois de Ferwer-
din, quand le soleil réveillait la foi dans les âmes’ (Xosrow Parvīz 3280/Mohl 7:
139/225; Bertel’s IX: 197, l. 3167). The effect of the “sunny” faith, hūr-e dīn, which is
put on the del ‘heart’ of the subject has no further implications for the sub-
sequent event or action, whereas the beginning of the New Year, sar-e sāl-e now,
does affect the subsequent action.20
ii. In contrast, the externality of the corresponding roxšandeh is noticeable:
- ferastādeh āmad bar-e zan čo gard / saxonhā-ye xosrow hame yād kard // zan-e
šīr az ān nāmeh-e šahr(e)yār / čo roxšandeh gol šod be-vaqt-e bahār ‘Le messager
alla rapidement, comme la poussière, auprès de Gordieh, et lui répéta toutes les
paroles de Khosrou. La lettre du roi rendit cette lionne rayonnante comme une
rose au printemps.’ (Xosrow Parvīz 3112/Mohl 7: 132/213; Bertel’s IX: 186, l. 3002).
The flower shedding light in the spring season, which is compared to the (ex-
ternal!) effect of the letter on the king Xosrow, is emphasized.
- čo goftār-e bahrām bešenīd šāh bexandīd-o roxšandeh šod pīšgāh ‘A ces paroles
de Bahrām, le roi sourit et le trône resplendit.’ (Hormozd 451/Mohl 6: 291/467;
Bertel’s VIII: 340, l. 434).
The effect of the laughing of king Hormozd is that the throne gives light, ra-
ther than being alight.
- jahān tāze gašt az sar-e tāj-e ūy / abā gorg-o mīš āb xordī be-jūy // az ān taxt-e
roxšandeh šādān šodand / hame kas bar-ū āfarīn xvānand ‘Le monde fut rajeuni
par sa couronne, la brebis et le loup s’abreuvèrent au même ruisseau; ce trône
brillant répandit le bonheur, et tous les hommes le comblèrent de bénédic-
tions...’ (Qobād P. 387/Mohl 6: 79/121; Bertel’s VIII: 51, fn.).

20
In the following line we are thus told: nehādī yekī ganj-e xosrow nehān keh neš<e>nāxtī keh-
tarī dar jahān ‘... il établissait chaque fois en secret [of Xosrow] un trésor, qu’aucun de ses sujets ne
connaissait’.
288 Johnny Cheung

The throne has regained its glory thanks to Kasrā’s ascension to the throne. In
return, the shining and the glory of the throne thus results in the rejuvenation of
the world, happiness and blessings to the people.

CONCLUSION
In short, the formal origin of the participle suffixes -andag (-andeh) and -ān is
clear, viz. from the present active *-ant- + *-ka- and middle *-āna- respectively.
Although both formations can be used adjectivally, the predicate employment is
mainly confined to ān-forms.
I hope that I have sufficiently demonstrated above that the functional appli-
cation of the active and middle voice accounts for both the meaning and func-
tions of these suffixes. This has continued right until the Middle Persian stage
and may also account for the subsequent functional developments in New Per-
sian. The middle voice emphasizes the subjectivity.
Certain categories of verbs are likely to be in the middle voice. Middle Per-
sian has many participles in -ān that denote (translational) motion and emotion.
The state in which the subject or qualified noun is occurring, is emphasized. This
appears to be more frequently envisaged in the earlier period. The predicate use
of -ān is also often found after certain auxiliary verbs, expressing the (inner) du-
rative state, thus betraying its former present tense nature. In contrast, the active
voice is more likely applied to the inanimate subject of a sentence or participle
construction, also when we are dealing with emotions. Whereas bramān, garzān,
etc. ‘weeping, lamenting, sim.’ are reserved for people (or personified beings),
the originally active voice is found in griyendag wāng ‘the lamenting voice’. The
application of the middle voice is also explicable in ān-participles such as arzān
worthy, of worth’, ’’stw’’n ‘praising, praying to, professing/pledging oneself to’ and
framōšān ‘forgetting’.
The original meaning of the active diathesis also transpires in the andeh-
/andag-participles denoting motion, when the goal or external effect to which
the motion is leading, is envisaged. In short, a choice is thus made whether to
emphasize the internal state of the qualified noun or sentence subject, or rather
its external quality or action, i.e. the choice between the middle vs. active voice.
It is perhaps not surprising either that -andeh in Modern Persian enjoys a greater
productivity, especially to form (deverbal) agent nouns, than in the past. In
Middle Persian, the most common way to derive agent nouns or adjectives with
an active meaning from verbs is by means of -āg (> NP -ā), e.g. dānāg ‘wise man’
 The Persian Verbal Suffixes -ān and -andeh (-andag) 289

(dānistan), sazāg ‘fitting, worthy’ (saz-), tarsāg ‘Christian’ (tarsīdan ‘to fear’), etc.
(cf. Rastorgueva/Molčanova 1981: 73). After the loss of final -g in New Persian, the
suffix may have lost its distinctive character, thus giving way to -andeh, which
has maintained its clarity. The formations in -ān has its quasi-adverbal or gerund
functions, which are intimately connected to the sentence subject. Their pro-
ductivity has perhaps faded due to the rise of alternative, syntactical means to
qualify the state or circumstance of this subject.

BIBLIOGRAPHY
Andreas, F. C. / W. B. Henning (1932-1934), “Mitteliranische Manichaica aus Chinesisch-Turkestan”,
Part I: SPAW 1932: 173-222: Part II: SPAW 1933: 292-363; Part III: SPAW 1934: 846-912.
Asatrian, G. S. (1983), “Pričastija na -ān v zapadnosredoiranskix jazykov (na materiale turfanskix
tekstov)”, Lraber hasarakakan gitowt’yownneri/Vestnik obščestvennyx nauk 5: 53-56, Erevan.
―― (1989), Otglagol'nie imena v srednepersidskom i parfjanskom, Erevan.
Bertel’s E. E. (ed.) (1960-1971), Šaxname, kritičeskij tekst, 9 vols, Moscow.
Brunner, Ch. J. (1977), A Syntax of Western Middle Iranian, Delmar-New York.
Degener, A. (1989), Khotanische Suffixe, Stuttgart-Wiesbaden.
Delbrück, B. (1888), Altindische Syntax, Halle a.d. Saar.
Emmerick, R. E. (1968), Saka Grammatical Studies, London-Oxford.
Gershevitch, I. (1961), A Grammar of Manichaean Sogdian, Oxford.
Gignoux, Ph. / A. Taffazoli (1993), Anthologie de Zādspram, Paris.
Geiger, W. / E. Kuhn (eds) (1895-1904), Grundriss der Iranischen Philologie, Strassburg.
Gonda, J. (1979), The Medium in the Rgveda, Leiden.
Henning, W. B. (1933), “Das Verbum des Mittelpersischen der Turfan-Fragmente”, Zeitschrift für In-
dologie und Iranistik 9: 158-253.
―― (1936), “Ein manichäisches Bet- und Beichtbuch”, Abhandlungen der Preussischen Akademie
der Wissenschaften. Philosophisch-historische Klasse (APAW) 10.
―― (1943), “The Book of Giants”, BSOAS: 52-74.
Jaafari-Dehaghi, M. (1998), Dādestān ī Dēnīg, Paris.
Jensen, H. (1931), Neupersische Grammatik: mit Berücksichtigung der historischen Entwicklung, Hei-
delberg.
Kellens, J. (1984), Le verbe avestique, Wiesbaden.
Kemmer, S. (1993), The Middle Voice, Amsterdam.
Lyons, J. (1969), Introduction to Theoretical Linguistics, Cambridge.
MacKenzie, D. N. (1979), “Mani’s Šābuhragān”, BSOAS 1979: 500-534.
Mohl, J. (1838-1878), Le livre des rois par Abou’lkasim Firdousi, 7 vols., Paris.
Morgenstierne, G. (1973), Etymological Vocabulary of the Šuγni Group, Wiesbaden.
Motlağ, Jalāl Xāleğī (ed.) (1988), Šāhnāme, Abū ol-Qasem Ferdowsī, vol. I-V, New York.
290 Johnny Cheung

Müller, F. W. K. (1904) “Handschriften-Reste in Estrangelo-Schrift aus Turfan, Chinesisch-Turkistan

Part II”, Anhang zu den Abhandlungen der Preussischen Akademie der Wissenschaften, vom
Jahre 1904, II, Berlin.
Rastorgueva, V. S. / A. A. Kerimova (1964), Sistema tadžikskogo glagola, Moskva.
―― / E. K. Molčanova (1981), “Srednepersidskij jazyk”, Osnovy iranskogo jazykoznanija, t. II: Sred-
neiranskie jazyki: 3-146, Moskva.
Sims-Williams, N. (1989), “A New Fragment from the Parthian Hymn-Cycle Huyadagmān”, Études
Irano-Aryennes offerts à Gilbert Lazard: 321-331.
Stempel, R. (1996), Die Diathese im Indogermanischen. Formen und Funktionen des Mediums und
ihre sprachhistorischen Grundlagen, Innsbruck.
Sundermann, W. (1973), Mittelpersische und parthische kosmogonische und Parabel-texte der Mani-
chäer, Berlin.
―― (1981), Mitteliranische manichäische Texte kirchengeschichtlichen Inhalts, Berlin.
Vahman, F. (1986), Ardā Wirāz Nāmag. The Iranian ‘Divina Commedia’, Copenhagen.
―― / G. S. Asatrian (1987), West Iranian Dialect Materials. From the Collection of D. L. Lorimer, vol.
I: Materials on the Ethnography of the Baxtiārīs, Copenhagen.
Waldschmitt E. / W. Lentz (1926), “Die Stellung Jesu im Manichäismus”, Abhandlungen der
Preussischen Akademie der Wissenschaften. Philosophisch-historische Klasse (APAW) 4.
West, E. W. (1880), Pahlavi Texts. Part I: The Bundahish, Bahman Yast, and Shâyast lâ-Shâyast. Sa-
cred Books of the East, vol. 5, Oxford.
―― (1885), Pahlavi Texts. Part III: Dînâ-î Maînôg-î Khirad, Sikand-Gûmânîk Vigâr, Sad Dar. Sacred
Books of the East, vol. 24, Oxford.
Williams A. V. (1990), The Pahlavi Rivāyat accompanying the Dādistān ī Dēnīg. 2 vols, Copenhagen.
Wolff, F. (1935), Glossar zu Firdosis Schahname, Berlin.
Zaehner R. C. (1956), The Teachings of the Magi. A Compendium of Zoroastrian Beliefs, London.
Allomorphic Variability in the Middle Persian Continuants of
the Old Iranian Suffix *-ka-

Claudia A. Ciancaglini
La Sapienza, Rome

Abstract
This paper investigates the Middle Persian continuants of a very widespread derivative suffix, OIr.
*-ka-, which is generally claimed to have had only one outcome in Middle Persian, namely -Vg,
written ‹-k'›. As is known, this outcome gave raise to a number of different Middle Persian suffixes,
for instance -ag, -āg, -īg etc., through the reanalysis of the preceding vowel as a part of the suffix. I
wish to demonstrate that already in Middle Persian, and not just in New Persian, OIr. *-ka- had
many other minority outcomes that have never been recognised in the previous studies. I also wish
to underline that these allotropes, not all of which are perceived, and function, as true suffixes in
Middle Persian synchronically, are sometimes explicitly marked in the Pahlavi script, notwith-
standing its well-known ambiguous and archaizing nature. Finally, I suggest that the presence al-
ready in Middle Persian of different outcomes of OIr. *-ka- can partly depend on early borrowing
processes among Middle Iranian dialects, and partly reflect different diachronic stages of the leni-
tion process undergone by the OIr. voiceless velar plosive in internal and final postvocalic position.

Keywords
Middle Persian, Morphology, Phonology, Suffixation

1. INTRODUCTION*
The Old Iranian (henceforth: OIr.) suffix *-ka- is continued by many Middle Per-
sian (henceforth: MP) suffixes, generally assuming the form -(V)g or -(V)k, for
example MP -ag, -āg, -ak, -īg, -īk, -ūg etc. The initial vowel of the suffix in the
majority of cases derives from the wrong segmentation of the stem vowel his-
torically preceding the suffix *-ka-, and only very rarely from genuine vocalic
“primary” suffixes such as *-aka- and the like. The quantity of the long vowels in
the MP suffixes results either from sandhi phenomena that lengthened an older

*
I am happy and honoured to contribute to this volume celebrating the 60th anniversary of
Prof. Dr. Garnik Asatrian, an excellent and renowned scholar in the field of Iranian studies. I offer
him this paper, together with my best wishes, as a little token of my esteem and gratitude.
 Koninklijke Brill NV, Leiden, 2015
292 Claudia A. Ciancaglini

short vowel, or from the reduction of various different phonic sequences.1 In

New Persian (henceforth: NP), these suffixes lost the final stop, if voiced, and
maintained it if voiceless.
For example, the derivational paths of MP -ag and -ak are the following:
OIr. *-a-ka- > Early MP -ak > MP -ag > Early NP -a > NP -e;
OIr. *-a-ka- > Early MP -ak > MP -ak > NP -ak.

From this traditional reconstruction it seems that the MP outcomes of OIr.

*-ka- are only two, namely -g and -k. Furthermore, it is generally assumed that
the difference between the two is unrecorded in the ambiguous and archaizing
Pahlavi script, where for instance both -ag and -ak are spelled <-k'>, and that the
voiced or voiceless nature of the final velar stop may be recovered only if one
takes into account: a) the Manichaean MP texts, because they are written in a
variation of the Aramaic script closely related to the Palmyrene script and to the
Estrangelo variety of the Syriac script—Manichean MP script distinguishes /g/
and /k/ in the final position, the former being written <-g> while the latter either
<-k> or <-q>;2 b) the NP continuants of these suffixes; c) the parallel traditions.
In this paper I shall argue that:
1) the MP outcomes of OIr. *-ka- are not limited to -g and -k, but include also
other, even if less frequently attested, outcomes;
2) sometimes the Pahlavi script provides more information than usually assumed
on the phonetic changes undergone by the OIr. velar stops.

2. THE PHONOLOGICAL DEVELOPMENT OF THE OIR. VOICELESS STOPS

MP preserves the OIr. voiceless stops */p t k/ only in initial position, rarely in
final position and also in internal position if geminated. They were voiced in the
other phonic contexts, changing to MP /b d g/ (see Weber 1997: 613 with bibl.;
Sundermann 1989: 144 f.; Skjærvø 2009: 201).
The voicing of OIr. */k/ took place later than that of the other two voiceless
stops (cf. Pisowicz 1985: 139; Ciancaglini 2008: 72 ff.). The exact period when the
process started and when it was accomplished is unknown, but the data at our

1
On the lengthening of -i- and -u- in MP see Korn 2009; regarding OIr. *-ka- cf. Ciancaglini 2012.
2
On the Manichaean system of writing, see Henning 1958: 72 f.; Sundermann 1989: 143; Naveh
1982: 149–153, who holds that the Palmyrene and the Syriac Estrangelo scripts go back to a common
prototype, Palmyrene being the more conservative one; Skjærvø 1996: 530 ff.; Idem 2009: 197, 199
table); Durkin-Meisterernst 2000; Idem 2005. Note that in some instances Manichaean MP <-q> is
also used to represent /g/: cf. Henning 1958: 75 fn. 2.
 Middle Persian Continuants of the Old Iranian Suffix *-ka- 293

disposal allow us to assume that the voicing of the OIr. voiceless velar stop in in-
ternal and postvocalic position took place at least one or two centuries after the
voicing of OIr. *-p- and *-t- in the same phonic context.
This delay in the voicing of OIr. *-k- is proven by many facts, the first of which
is that in the renderings of MP words containing suffixes tracing back to the OP
suffixes -a-ka-, -i-ka-, which occur in the Greek versions of the 3rd cent. Sasanian
multilingual inscriptions, these suffixes are consistently spelled as -ακ, -ικ.
The second piece of evidence is represented by the renderings of the MP ve-
lar stop continuing OIr. *-k- in MP loanwords in the parallel traditions, especially
in Armenian, Talmudic Aramaic, Syriac and Arabic: the rendering with <-g> is
relatively rare in Armenian and in Talmudic Aramaic, and becomes more fre-
quent (but not prevalent) in Syriac and Arabic (in the latter, MP or Early NP -g is
represented by <ğ>).3
The third indirect proof is the great variability of the outcomes of OIr. *-k- in
New Persian. Pisowicz (1985: 139 f.) lists three New Persian outcomes in intervo-
calic position, and seven in postvocalic final position, namely (Pisowicz 1985: 139
f.; cf. Ciancaglini 2008: 73 f.):
● intervocalic position: -g- (e.g. āgāh ‘conscious’ < OP *ākāθa-); -ɣ- (e.g. taɣār
‘through’, cf. Arm. loanword takaṙ); -k- (e.g. čakād ‘forehead’, cf. Arm. loanword
čakāt);
● postvocalic final position: -Ø (e.g. zarda ‘yolk’ < MP zardag); -g (e.g. sōg
‘mourning’); -k (e.g. zardak ‘safflower’, MP id.); -ɣ (e.g. rēɣ ‘enmity’ < OP araika-);4
-x (e.g. zanax ‘chin’ < OP *zanaka-); -y (e.g. ǰāy ‘place’ < MP gyāg); -h in some in-
stances of the suffix -āh (e.g. dībāh ‘brocade’ < MP dēbāg) or in final sequences in
which -āh is formed by the long vowel of the stem and -h < *k, for example siyāh
‘black, dark’ < OIr. *syāva-ka- < PIE *ḱyeh1-.

These different continuants in New Persian are due to interdialectal bor-

rowings from varieties where OP -k- was attested in different phases of the leni-
tion process. This dialectal variability, according to Pisowicz (1985: 140), “took

3
The great majority of the Iranian loanwords in Armenian is borrowed from Parthian, but
there are also words borrowed from MP, and in these latter loanwords we have <-k>, e.g. Arm. das-
tak ‘handle’ (Bolognesi 1960: 8) ← MP dastag ‘bunch, bundle, group’ (CPD 25), NP dasta ‘a handle,
helve, haft’ (Steingass 1892: 525); Syr. and Talm. Aram. dstqʾ ‘handle, hilt’ also point to a voiceless
final stop (Ciancaglini 2008: 155).
4
However, the OP word is to be read arīka-, as suggested by Mayrhofer (EWA 1: 128); note also
that he suggests that it does not contain the suffix *-ka- but perhaps derives from *h2li-h3kw-o-
‘turned towards another side’: in this case, it would be an -añc- formation.
294 Claudia A. Ciancaglini

place in the late period of MP following a relatively uniform status of koine of

the beginning of Sassanids’ reign (3rd c. A.D.)”. Importantly for our concern here,
this variability is also very evident in the NP allotropes of the suffix -āk, namely
-āɣ, -āx, -āh.
The mentioned fluctuations of the outcomes of OIr. *k in this suffix give rise
to New Persian allotropes of identical derivatives, for example NP dībā and dībāh
‘brocade’ < MP dēbāg, NP rama and ramak ‘herd, flock’ < MP ramag ‘id.’, or NP
istāk, sitāk and istāx, sitāx ‘branch’ < MP stāk ‘id.’.5

3. VOICING OF OIR. *-K- IN MIDDLE PERSIAN

As regards the MP phase, however, scholars do not agree: some of them hold that
the voicing of OIr. *-k- was already accomplished in the 3rd cent. AD. Such an
opinion is based almost exclusively on the data provided by MP texts written in
Manichaean script, where /g/ and /k/ are graphically distinguished. The basic as-
sumption is that Manichaean MP was exactly the same linguistic variety as the
MP attested in the Book Pahlavi texts of the 3rd cent. AD, only with the ad-
vantage of being written in a nearly phonetic, unambiguous writing system,
without heterograms and the historical spellings that characterize the Pahlavi
script (MacKenzie 1967: 18; Sundermann 1989: 139; Tedesco 1921: 249 f.; Skjærvø
2009: 200; Weber 1997: 603; Pisowicz 985: 133). This opinion is particularly rele-
vant because it is shared by MacKenzie, whose system of transcription of Pahlavi
script has been adopted by the majority of scholars.
In reality, the data provided by Manichaean Middle Persian texts are not
quite so unequivocal: the Manichaean texts were written during a time span of
many centuries, from the 3rd to the 8th cent. AD, and the manuscripts them-
selves were in their turn created much later (Skjærvø 2009: 197). During such
long period of time it is almost certain that the language underwent phonetic
changes. But on the graphic side Henning (1958: 73 f.) holds that the Manichaean
script did not undergo significant changes in course of time, and this is paral-
leled by Sundermann's observation (Sundermann 1989: 143) that the Manichaean
script shows historical spellings too: as is expected for a writing system used for a
long time without graphic change, it becomes archaizing vis-à-vis the spoken
language, which changes more quickly.

5
On MP dēbāg see Ciancaglini 2008: 150; for the allotropes of the NP word for ‘branch’ see
Steingass 1892: 49; 654; in general on the allotropes of the NP suffix -āk see Horn 1898: 175; on simi-
lar allotropes in MP see below, § 5.
 Middle Persian Continuants of the Old Iranian Suffix *-ka- 295

It is also noteworthy that the Manichaean writing system, though differenti-

ating <k> and <g>, often uses <q> for both MP k and g (from older k), and this
holds not only for MP texts, but also for those in Parthian and Sogdian written
with this script (Henning 1958: 75; Salemann 1908: 152 f.; regarding Sogdian, see
Sims-Williams 1989: 177). Exactly on this basis, and also on the basis of the Mani-
chaean use of <ṭ> both for Parthian /t/ and /d/ (< /t/), Henning (1958: 75) sug-
gested that in the Middle Iranian dialects adopted by the Manichaeans in the 3rd
cent. AD the change of the voicing of the voiceless stops was not still complete.
Finally, there are cases in which Manichaean MP has a final <-g> ( < OIr. *-k-),
but where New Persian retains the voiceless velar stop, and this seems to bewil-
der MacKenzie in his transcription choice: see, e.g., MP tārīg/k [tʾlyk'] ‘dark’,
ManMP tʾryg, NP tārīk (CPD 82); MP paristōg/k [plstwk'] ‘swallow’, ManParth.
prstwg, NP piristū, piristūk, piristuk (CPD 65); but MP nāzuk [nʾcwk'] ‘tender, gen-
tle; fickle’, ManMP nʾzwg, NP nāzuk (CPD 58); MP kanīzag [knyck'] ‘girl’, ManMP
knycg, NP kanīzak (CPD 49).
In any case, it is certain that the voicing of OIr. *-k- took place later than the
voicing of *-p- and *-t-, and that it was not homogeneous: the outcomes of OIr.
*-k- are different even within the western Middle Iranian dialects,6 and it is
highly probable that, even within Middle Persian, the voicing started in different
epochs according to the different syntagmatic contexts.
As regards this latter point, Back (1981: 179; fn. 8) claims that, diachronically,
the voicing of the Old Iranian voiceless plosives in intervocalic contexts is to be
distinguished from voicing after continuants, because this latter voicing assimi-
lation, at least in the case of a nasal preceding a voiceless stop, took place already
in Late Old Persian, whereas the former phenomenon is a Middle Persian inno-
vation. The Middle Persian voicing, which he describes as a Lautverschiebung,
according to him only concerns the older voiceless plosives in internal, intervo-
calic position and in final, postvocalic position. The sonorization after /n/ and
/r/, of which he explicitly attributes the first to the Late Old Persian period and
the second to the Middle Persian stage, would be a different and independent
phenomenon, because it affects only a specific segmental feature in an assimila-
tion process, whereas the sonorization in internal, intervocalic position and in
final, postvocalic position—according to Back—is included in a Lautverschie-
bung which involves many different reduction processes (for example the

6
See e.g. the example of NP payg and payk below, § 4.2.
296 Claudia A. Ciancaglini

change of the Old Iranian voiced plosives into glides, and the subsequent
changes affecting these glides in particular contexts), all of which are favoured, if
not caused, by phonic contexts with a maximum degree of voicing and charac-
terized by different segmental assimilations.7
In any case, notwithstanding the neutralization of the functional opposition
between /k/ and /g/ in internal and final postvocalic position, produced by the
voicing of the OIr. voiceless stops in these phonic contexts, Middle Persian still
shows a functional opposition of voiced and voiceless plosives also in final posi-
tion, which is testified not only by minimal pairs such as MP sāg ‘number’ ~ sāk
‘tribute’, MP zardag ‘yolk’ ~ zardak ‘safflower’, but also by the opposition of some
of the suffixes under consideration, for example -ag and -ak, -īg and -īk (e.g. MP
nazdīk ‘near’; MP, NP zandīk ‘heretic’), which are identically spelled in Middle
Persian, but graphically distinguished in Manichaean Middle Persian. There are
also further adjectival suffixes ending in -k (< OIr. *-ka-), for example -uk (e.g. MP
sabuk, nāzuk, tanuk), where the voiceless final plosive is also represented in
Manichaean script and preserved in New Persian, and often confirmed by the
parallel traditions. Finally, Weber (1997: 613) observes that the retention of MP
medial -k- in words like cakōk ‘lark’ is unclear, and hints at “semantic reasons”
that may have played a role here as well as in the diminutive suffix -ak.

4. MP -G AND -K < OIR. *-KA-

The most frequent outcomes of the OIr. suffix *-ka- are MP -g and -k, even if they
are not the only two developments witnessed by Middle Persian.
It is important to underline that the diachronic origin of a morpheme differs
from its synchronic function: in other words, a phonic sequence that diachroni-
cally continues the OIr. suffix *-ka- does not necessarily function as a suffix in
Middle Persian. That a given phonic sequence functions as a suffix in a dead lan-
guage and that it was perceived as such by its speakers may be argued on the ba-
sis of linguistic parameters, like frequency of occurrence, productivity, presence

7
Back (1981: 184): “Entdeutlichungsbewegungen werden durch eine extreme-sonore Umge-
bung unterstützt, wenn nicht sogar provoziert. Dabei sehe ich einen prinzipiellen Unterschied in
der Assimilation von spätap. N + T → N + D und vom mp. r + T → r + D […], die nicht in dieselbe
Entdeutlichungstendenz gehört, sondern einen von ihren speziellen segmentalen Merkmalen be-
stimmten unabhängigen Assimilationsprozeß zeigt (‘consonants in contact’). Die in dieser Studie
beschriebenen Reduktionsvorgänge dagegen sind durch eine gemeinsame Bedingung einerseits
und durch verschiedene segmentale Assimilationen andererseits gekennzeichnet”.
 Middle Persian Continuants of the Old Iranian Suffix *-ka- 297

or absence of the correspondent non-suffixed lexemes in language vocabulary,

degree of identifiability of the morpheme according to the language typology
and so on.8
In particular, in Middle Persian only some of the outcomes of OIr. *-ka- func-
tion as suffixes, notably those in which a preceding vowel (a thematic or a radi-
cal one) coalesced with the final velar stop, giving rise to MP -ag, -āg, -ak, -īg, -ūg
etc.
In this section we shall focus especially on the MP outcomes -g and -k that
formally continue OIr. *-ka- but do not behave as suffixes in Middle Persian. In
the following one (§ 5) we shall consider other MP outcomes of OIr. *-ka-, which
are very uncommon and whose presence in Middle Persian requires an explana-
tion from the point of view of historical phonology.

§ 4.1. MP -k < OIr. *-ka-

Among the cases in which OIr. *-ka- did not develop as an autonomous suffix in
Middle Persian there are those in which the suffix follows a stem ending with a
consonant. If this consonant is voiceless, especially a sibilant or a fricative, OIr.
*-ka- becomes MP -k; if it is a voiced continuant, especially a nasal or vibrant,
OIr. *-ka- becomes MP -g (see § 4.2).
Some examples of MP -k deriving from OIr. *-ka- and no longer recognizable
as a suffix are the following:
● MP hušk ‘dry, withered’ (MacKenzie 1971: 45, Nyberg 1974: 104), NP xušk ‘id.’
(Steingass 1892: 462) < OP uška- ‘id.’ (Kent 1953: 187); cf. YAv. huška- adj. ‘dried up’
< hauš- ‘to dry’ (Bartholomae 1904: 1840), Ved. śúṣka- ‘dried up’ < śoṣ- ‘to dry’
(EWA 2: 648; 658);
● MP bizešk [bcšk'] ‘doctor, physician’ (MacKenzie 1971: 18; Nyberg 1974: 48);
ManMP, ManParth. bzyšk; NP bizišk, pizišk (Horn 1893: 42 f. nr. 183); < OIr. *bišaz-
ka- whence in NW (with metathesis) *bizaška-; cf. Av. baēšaza- ‘healing’, OInd.
bhiṣáj- (EWA 2: 264);
● MP yask [ysk'] ‘illness, sickness’ (MacKenzie 1971: 97), ManMP ysk, YAv. yaska-
m. ‘disease’; the etymon is disputed: the Av. word may contain the suffix *-ka-
and be connected to OInd. yákṣma- m. ‘id.’, as Bartholomae (1904: 1269) and
Edgerton believe (1911: 309), only if we admit that it comes from OIr. *yakṣka- but
the similarity between the Avestan and the Old Indian forms is purely casual, ac-

8
On the theoretical problems of the concept of “productivity” applied to corpus-languages see,
e.g., Panagl 1987: 127–131.
298 Claudia A. Ciancaglini

cording to Mayrhofer (EWA 2: 392), so that it is uncertain whether this is indeed

the suffix *-ka-.9
Another small group of MP words continuing OIr. *-ka- derivatives in which
this suffix results in MP voiceless -k shows a preceding long vowel ā or, more
rarely, ē. This long vowel is either explainable as the contraction and/or loss of
older phonic sequences, or pertains to the original roots. These MP forms are
almost all monosyllabic. Synchronically, this final -k was not perceived as a suffix
in MP.
See the following examples:
● MP pāk [pʾk'] ‘clean, pure, holy’ (MacKenzie 1971: 64, Nyberg 1974: 150), ManMP
pʾk; ManParth. pwʾg; NP pāk ‘id.’ (Steingass 1892: 231) < OIr. *pavāka-, cf. OInd.
pāvaká- ‘clear, bright’, metrically to be read *pavāká- (cf. Debrunner 1954: 266 f. §
150; Whitney 1924: 446 § 1181), which is an instance of the primary suffix -āka- at-
tached to the OInd. verbal root pav- ‘to cleanse, purify’ (pres. pávate, punāti, pas-
sive pūyáte, part. pūtá-); all these forms trace back to PIE *pewH- ‘to purify,
cleanse’, from which also derives, e.g., YAv. pūitika- adj. ‘purifying, purificatory’,
Lat. pūrus ‘pure, clean’ etc. Weber (1997: 613) reconstructs OIr. *pāw(a)ka- as the
basis of MP, NP pāk, and Skjærvø (2009: 201) holds that the MP form comes from
a preceding *pāwk; nevertheless, it seems more probable that the OIr. antecedent
of MP pāk should be reconstructed as *pavāka-, with a preceding long *-ā-, not a
short one, if we also take into account, beside the abovementioned OInd. form,
the fact that the Parthian form (also mentioned by Skjærvø) is pʾwg /pawāg/, with
a long -ā- before the suffix: see also Bailey (1979: 228);
● MP nēk [nywk'] ‘good, beautiful’ (MacKenzie 1971: 58; Nyberg 1974: 137);
ManMP nyk; NP nēk ‘id.’ < OIr. *naib(a)ka; cf. Ciancaglini (2008: 74);
● MP šāk [šʾk'] ‘branch’ (Nyberg 1974: 184), NP šāx, cf. OInd. śākhā < IIr. *ḱākhā <
PIE *ḱok[w]-eh2- ~ ḱ[w]-h2-os; Nyberg negates that MP šāk be related to OInd. śākhā,
but see EWA 2: 628;
● MP stāk ‘branch’, NP istāk, istāx, sitāx, unrecorded in MacKenzie and Nyberg
but attested in the Bundahišn and mentioned by Hübschmann (1895: 245) and
Horn (1893: 20 nr. 82; cf. Id. 1898: 175), probably trace back to IIr. *sthā-, PIE *steh2-
‘to stay’ (see EWA 2: 764 ff.);
● MP xāk [hʾk'] ‘earth, dust’ (MacKenzie 1971: 93; Nyberg 1974: 218), NP xāk, but
the etymon and formation are unknown as far I know;
● MP sāk [sʾk'] ‘tax, tribute’ (MacKenzie 1971: 73, Nyberg 1974: 172); both MacKen-
zie and Nyberg refer to NP sāv, sā ‘tribute, revenue’ (Steingass 1892: 637; 644), but

9
Regarding YAv. yaska- see Ciancaglini 2012: 18.
 Middle Persian Continuants of the Old Iranian Suffix *-ka- 299

the former quotes only NP sāv and encloses it in parentheses, as a related but not
directly comparable form, being from a different dialect or in some way modified.
Moreover, MacKenzie lists sāk ‘tribute’ and sāg [sʾk] ‘number’ as two different
words and, consequently, as a minimal pair of -k ~ -g. On the other hand, Nyberg
compares sāk ‘tribute’ with ManParth. sʾg ‘number, part’, ManMP sʾgwmnd ‘cal-
culable’, ʾsʾg ‘innumerable’, evidently considering sāk ‘tribute’ and sāg ‘number’ to
be the same word (but see below). Nöldeke (1892: 20 fn. 4), followed by Horn
(1893: 154 fn. 1), holds that MP sāk ‘tribute’ as well as NP sāv, sā trace back to a
previous *sāwak,10 like pāk derives from *pāwak and nēk from MP nēwak (this lat-
ter is the old transcription of MP nēk [nywk'] ‘good, beautiful’, ManMP nyk).

There are many formal problems: first, we have seen that MP, NP pāk comes
from OIr. *pavāka-, not from pāvaka-, and MP, NP nēk derives from OIr.
*naib(a)ka-. Furthermore, the MP words ending in -āk generally maintain this fi-
nal sequence in NP, whereas MP -āg develops in NP -ā, and this excludes a direct
derivation of NP sā from MP sāk. Finally, the MP words ending in -āk generally
show a long -ā- that is part of the root (as in MP stāk ‘branch’ < *stāka-) or results
from the reduction of a previous sequence -avāka-, whereas -āvaka- usually pro-
duces MP -āg, NP -ā (see, for instance, MP niyāg ‘grandfather’, OIr. *ni-āva-ka-,
OP niyāka-, YAv. niiāka-, NP niyā).11
In other words, I believe that formally the antecedent of MP sāk should be
OIr. *savāka-, perhaps from PIE *ḱewh1-, cf. OInd. 1śav- ‘to become strong’ (EWA
2: 623). The two NP words sā and sāv may both derive from the same OIr. arche-
type, but represent its outcomes in different dialect varieties. NP sāv may also
derive from an antecedent not enlarged with the velar suffix; NP sā may formally
be the result of a previous MP *sāg, but this latter form, if it had existed, should
have been different from the attested MP sāg ‘number’, whose etymon is equally
uncertain.12

10
This reconstruction is dubitatively quoted also by Hübschmann (1895: 245 fn. 1).
11
As for the etymon, see Szemerényi 1950: 235 f., who suggests that OP niyāka- comes from *ni-
āvaka-, in which the root *ā̆ va- would be connected with Lat. avus ‘grandfather’. See also Ciancagli-
ni 2012.
12
Henning (1947: 308) suggested the derivation of MP sāg ‘number’ from *sāhak and compares
YAv. +sā̊ ŋha- ‘number’ (AirWb 1577). However, the Avestan word, though its Pahlavi translation is
marag ‘number’ (but its meaning was considered otherwise uncertain by Bartholomae), is actually
a proper name, comparable with OInd. śāsá- ‘commander, ruler’ (Mayrhofer 1977: 75; EWA 2, 632),
so that it does not mean ‘number’ and is not cognate with MP sāg, whose etymon remains doubt-
ful.
300 Claudia A. Ciancaglini

4.2. MP -g < OIr. *-ka-

If the preceding consonant is voiced, especially if it is a voiced continuant like
/n, r, l/, OIr. *-ka- generally developed into MP -g.
The voicing of the velar stop after /n/ is attributed by Back (1981: 179 fn. 5; 184)
already to the Late Old Persian period.13 His opinion is confirmed by the parallel
traditions (see, e.g., Hdt. 3, 93 Σαράγγαι, name of the inhabitants of the Drangi-
ana, OP Zranka-) and by the fact that this voicing is systematically recorded in
the Pahlavi script: see, for instance, MP tang [tng] ‘narrow, tight’ (MacKenzie
1971: 82, Nyberg 1974: 191), NP tang ‘id.’ (Steingass 1892: 330), cognate with OInd.
tañc- ‘to contract’, from PIE *tenk- ‘to contract, coagulate’ (EWA 1: 615).
The voicing of OIr. *-ka- after /r/ is dated by Back (1981: 184) to the MP period;
see the following examples:
● MP wuzurg [wc(w)lg] ‘big, great’ (MacKenzie 1971: 93; Nyberg 1974: 207 reads
vazurg) < OP vazra-ka-, ManMP and ManParth. wzrg; NP buzurg; Arm. LW vzurk,
vzǝrk, vzruk;
● MP xwarg [hwlg] ‘ember’ (CPD 95), Nyberg (220) translates ‘consuming, of the
fire’ < *xwarka-; it seems a case of primary *-ka- added to a verbal root, which
may be the same as in xwardan, xwar- ‘to eat, consume’ or, as suggested by Gnoli
(1996), *hwar- ‘to smoulder, burn’; Jud.-Pers. xwwrg.
Within the MP words ending in -rg there are instances in which -g does not
continue the OIr. suffix *-ka-, but is part of the root or stem, for example in MP
gurg [gwlg] ‘wolf’ (MacKenzie 1971: 38; Nyberg 1974: 86), NP id., YAv. vǝhrka-
(AirWb 1418 f.), OInd. vŕ̥ ka-, PIE *wlkw- (EWA 2, 570), or in MP marg ‘death’, cf.
YAv. mahrka-, OAv. maraka-, from the Av. verbal root marǝk- ‘to kill, destroy’ <
IIr. *mark-/marc- (EWA 2: 323 f.); sometimes it continues an older *-ga-, as in MP
sang [sng] ‘stone, rock’ (MacKenzie 1971: 74), ManParth. ʾsng /asang/ (Durkin-
Meisterernst 2004: 55); NP id. (Steingass 1892: 701), cf. YAv. asǝṇga-, OP aθanga-.
Nevertheless, it is worth noting that in the case of the final sequence -rg < -
rk-, of whatever origin, the Pahlavi script records the sonorization. If indeed this
voicing developed only in the MP period, as suggested by Back, we have to con-
clude that at least in this case the Pahlavi script is not as ambiguous and archa-
izing as generally stated.
A further question related to the voicing of *-k- after /r/ is represented by a
group of MP derivatives, historically containing the continuation of the OIr. suf-

13
See also Hübschmann (1895: 239); Weber (1997: 613); Sundermann (1989: 144 f.); Skjærvø
(2009: 201).
 Middle Persian Continuants of the Old Iranian Suffix *-ka- 301

fix *-ka-, which are differently transcribed by MacKenzie and Nyberg: MacKen-
zie reads <-rg> as /-rag/, Nyberg as /-rg/. See the following examples:
● MP andarag [ʾndlg] ‘in, between, among’ (MacKenzie 1971: 9); Nyberg (1974: 18)
reads andarg < *antarkā, cf. Av. (Bartholomae 1904: 133) antarǝca (for *antarcā)
‘between’;
● MP wastarag [wstlg], ‘clothing, garment’ (MacKenzie 1971: 88) < wastar ‘id.’ <
Av. (Bartholomae 1904: 1385) vastra- ‘clothing’; Nyberg (1974: 205) reads vastarg;
● MP widarag [wtlg] ‘path, passage’ (MacKenzie 1971: 90; Nyberg 1974: 216) < OIr.
*vitar-ka-, cf. MP widardan, wider- ‘to pass (by, away), to cross, to die’, ManMP
wdr; NP guδar; the MP word hardly derives from MP widār ‘path, passage’, be-
cause of the different quantity of /a/; MP widār is related to the causative verb
widārdan, widār- ‘to let pass’;
● MP wistarag [wstlg] ‘bedding, cover’ (MacKenzie 1971: 91; Nyberg 1974: 215) <
wistar ‘id.’; NP bistar; Nyberg reads vistarg ‘bed coverlet’ < OIr. *vi-star-ka-, cf.
vistartan ‘to spread’ < vi + star-.
The reason for these two different MacKenzie’s transcriptions, -ag and -g for
the same script <-g>, is unclear to me, especially in the cases in which this
spelling is the outcome of one and the same OIr. suffix *-ka-. If in MP wuzurg we
may believe that the NP outcome buzurg influenced MacKenzie's choice, the
reason for the readings widarag or wistarag (NP guδar and bistar) is nevertheless
unclear: it is known that the continuation in NP of MP -ag is -a, not -Ø. On the
other hand, the final -Ø is better explainable if we admit a MP final sequence
-arg: in other words, a change MP -arg > NP -ar is phonologically natural,
whereas a hypothetic change MP -arag > NP -ar (instead of -ara) would require
an explanation.
As opposed to MacKenzie, Nyberg transcribes almost all of the abovemen-
tioned cases with the final -g, namely wistarg, wastarg and so on, and it seems
clear that, in his opinion, this final -g corresponds to a MP suffix different from
the one that he usually transcribes as -ak (corresponding to MacKenzie's -ag).
More explicitly, Salemann (1895: 278), in his survey of the MP suffixes, distin-
guishes -k (which is the outcome of OIr. *-ka- in the MP suffixes -ag, -āg, -īg etc.)
from -g, and as instances of this latter suffix mentions MP wistarg, widarg, an-
darg and—wrongly, in my opinion—marg (where -g is part of the root and not a
suffix: see above).
Nevertheless, Salemann (1895: 278) suggests that this MP -g may trace back to
an older *-ka-, perhaps the same inherited suffix -ka- in “urir. *marka- ‘Tod’”.
302 Claudia A. Ciancaglini

Apart from the wrong example provided by Salemann, it is in my opinion

highly probable that MP -g results from the crystallization of a voiced allotrope
of IIr. *-ka-, as already hypothesised by Pisani (1935: 160) regarding OInd. -ga-:
according to Pisani, OInd. -ga- results from a voicing assimilation of -ka- in
voiced contexts; see, for example, the formal allotropes OInd. árbhaga- ‘youthful’
and arbhaká- ‘small, little’, both from árbha- ‘little’, and OInd. Nabhāka-, name of
the author of the R̥ gveda, and Nabhāga-, name of the son of Manu Vaivasvata
(this latter form, however, is attested only in the lexicographers).
This voiced allotrope of -ag <-k'>, spelled <-g>, merged with the outcome of
the older IIr. suffix *-ga-, occurring for instance in YAv. asǝṇga-, OP aθanga-, MP
sang ‘stone’. Furthermore, the general tendency toward the loss of the formal
transparency of the internal components of the sign also caused the merger of
this -g with final non-suffixal -g, such as in MP marg etc.
In any case, as far Middle Persian is concerned, the different spelling, even in
the absence of a detectable different function, does in my opinion indicate two
different suffixes. This formal distinction is also required by the different out-
comes in New Persian: MP -ag [-k'] is continued in New Persian by -a, whereas
MP -g [-g], in all the instances in which it is preceded by /r/, is continued by -g
(e.g. NP buzurg etc.) or -Ø (e.g. NP guδar, bidar etc.). This different New Persian
treatment seems to be the only reason why MacKenzie transcribes -ag (widarag,
wistarag etc.) in the first case and -g in the second case (wuzurg etc.).
Different is the case of MP duš-farrag [dwš-plg] ‘unfortunate’ (MacKenzie
1971: 29), read as dušxwarrah by Nyberg (1974: 69), who records only the writing
[dwš-GDH], namely the heterographic form of MP xwarrah ‘fortune, glory,
splendour’ (MacKenzie 1971: 96). If the writing [dwš-plg] is reliable, it surely rep-
resents duš-farrag, with the final sequence -ag, which may be an analogical lev-
elling on the model of the -ag adjectives, but historically this compound traces
back to Av. duš-xvarǝnah- ‘ill-famed’ (Bartholomae 1904: 755), so that in Middle
Persian it should be dušxwarrah or dušfarrah.14
The voiced velar stop not preceded by a vowel, beside the derivatives consid-
ered above where the suffix is preceded by /r/ (of the type MP buzurg, wistarg
etc.), also appears in words like payg [pdk'] ‘foot-soldier, courier’ (MacKenzie
1971: 67; Nyberg 1974: 147), where -g is the result of the sonorization of an early

14
On the etymon of MP xwarrah and the initial alternance xw- ~ f- see Gnoli 1999, with bibl.;
Gnoli 1996; Skjærvø 1983.
 Middle Persian Continuants of the Old Iranian Suffix *-ka- 303

MP payk, as is proven by the Pahlavi script, which—differently from the cases of

the type buzurg, wistarg—does not record the sonorization, and by the double
NP continuants payg ‘newsman, messenger’ and payk ‘running footman, carrier,
messenger’ (Steingass 1892: 268–269; Horn 1893: 80 nr. 359). This word, attested
in ManMP as pyg, in ManParth. as pdg and in the parallel traditions, for example
Arm. LW payik ‘foot-soldier’, payak ‘id.’ (both from MP, on account of the change
*-d- > -y-, see Bolognesi 1960: 43) and Syr. LW paygā ‘id.’ (Ciancaglini 2008: 231), is
of problematic formation. Nyberg, who reads padak [pdk'] and translates ‘step,
tread, (a horse's) run’, related this word to Av. (Bartholomae 1904: 842) 1paδa-, so
that he seems to suggest a derivation of the MP word from OIr. *pada-ka-, but
from this formation only MP padag is possible (which is in fact the reading pro-
vided by Nyberg), which would have developed to *payag (after the MP change
of internal and originary *-d- > -y-), not payg, which is attested by ManMP pyg
and Syr. LW paygā. Horn refers to Av. *padika-, but an OIr. form *padika- would
have developed to MP *padīg > *payīg. Mayrhofer (EWA 3: 303; cf. KEWA 2, 206),
regarding OInd. padika- ‘pedestrian’ and OInd. padga- ‘foot-soldier’, claims that
any closer relation with the Iranian words is uncertain, “weder als Zeugnis für
Urverwandtschaft (< iran. *padika-) noch als Lehnquelle” (whereas Pkt. pāyakka-
is surely a loanword from Iranian). In any case, OInd. has all the possible for-
mations, namely padika- ‘pedestrian’, padaka- ‘foot, step’ (Monier-Williams 1899:
584) and padga- ‘foot-soldier’.
In my opinion, it is probable that MP payg (together with NP payg, payk) de-
rives from an athematic formation like *pad-ka- ( > payk > payg), but it seems
that in MP there have also been other formations, as seems indicated by the
double forms of the Arm. loanwords payik and payak. Furthermore, this word is
also interesting because it testifies to the double treatment in the Western Mid-
dle Iranian varieties of the velar suffix, namely -g and -k: the form with the
voiceless velar stop probably does not only represent an early MP stage before
the sonorization, but an allotrope with a different areal or diaphasic distribution,
as is proven by its preservation in NP payk.
Finally, there are some cases in which MP -g < *-ka- no longer functions as a
suffix, its voicing is unrecorded in script and it is preceded by a long vowel: an
example is MP nibēg [npyk'] ‘writing, scripture, book’ (MacKenzie 1971: 59;
Nyberg 1974: 141), ManMP and ManParth. nybg, < *ni-paya-ka-, cf. Av. ni-pāy-
(pres. also pay-) ‘to keep, guard, preserve’ (Bartholomae 1904: 886).
304 Claudia A. Ciancaglini

5. MINORITY OUTCOMES OF OIR. *-KA-

As we have said above, MP also witnesses some less frequent outcomes of the
OIr. suffix *-ka-. These no longer function as autonomous suffixes in MP, in par-
ticular MP -x and -h (both written <-h>) and -Ø.
In a few cases Middle Persian continues OIr. *-ka- with a voiceless velar frica-
tive: an example, also interesting because the same word shows four different
allotropes, is the word for ‘hole, burrow’, namely MP sūrāx [swlʾh], flanked by
MP sūrag [swlk'] , sūrāg, sūlāg [swlʾk'].15
These allotropes, already occurring in Middle Persian, seem to point to an
early fricativization of the voiceless velar stop preceding its voicing.
A similar case is attested in the parallel traditions: MP warrag [wlk'] ‘lamb,
ram’, (ManMP wrg, NP barra < OIr. *warna-ka-) appears as loanword in Tal-
mudic Aramaic and in Syriac in the form brḥʾ (Ciancaglini 2008: 132). The final -ḥ
is considered inexplicable by Telegdi (1935: 235) and exceptional by Shaked
(1987: 260), but I believe that it represents the outcome -x of OIr. *-k-, an out-
come that appears in New Persian (e.g. NP zanax ‘chin’ < *zanaka-) and occa-
sionally in Middle Persian, as for instance in the abovementioned sūrāx [swlʾh]
‘hole, burrow’.
There are also a few cases in which the outcome of OIr. *-ka- is MP -h (writ-
ten <-h>) or -Ø.
The outcome -h is only to be assumed for the first and second plural personal
pronouns, which are written logographically in Pahlavi script, but the final -h is
attested in Manichaean Persian, namely MP amāh [LNE] ‘we, us’ (Nyberg 1974:
14; MacKenzie 1971: 7 reads amā) < OP ahmāxam, ManMP, ManParth. ʾmʾh, NP
mā and MP šmāh [LKWM] ‘you (pl.)’ (Nyberg 1974: 187; MacKenzie 1971: 13 reads
ašmā) < OP *xšmāxam, ManMP ʾšmʾ(h), ManParth. ʿšmʾh, NP šumā: in both
cases, MP -h is the result of OP -xa-, in its turn a lenition of OIr. *-ka- (cf. OInd.
asmā́ ka- ‘ours’, and OAv., YAv. yūšmāka-, xšmāka-, OInd. yuṣmā́ ka- ‘yours’).16
A further interesting case is the Iranian word for ‘black, dark’, namely NP
siyāh, whose final -h is an outcome of an older suffixal *-k and probably traces
back to OIr. *syāva-ka- (cf. Av. siiāva-, and OInd. śyāva-ka-, proper noun, EWA 2:

15
MacKenzie 1971: 78; Nyberg 1974: 181 reads sūlāk; Hübschmann (1895: 245) and Horn (1898:
175) record both sūrāk and sūlāk; cf. YAv. sūra- (Bartholomae 1904: 1585); this form is related to
OInd. śū́ na-, and traces back to PIE *ḱuH-r-/*ḱuH-r- (EWA 2, 650).
16
For these personal pronouns see EWA 1, 151 f.; 2, 415; note that the lenition phase -x- < IIr. *-k-
is witnessed, beside OP, by ManSogd. mʾx ‘we’ and ʾšmʾx ‘you (pl.)’.
 Middle Persian Continuants of the Old Iranian Suffix *-ka- 305

661). The velar suffix is witnessed by ManParth. syʾwg /syāwag/, attested in addi-
tion to the form without the velar suffix, namely ManParth. and ManMP syʾw
/syāw/ (Durkin-Meisterernst 2005: 311 f.), whereas the MP form siyā (or syā in
MacKenzie's reading, see MacKenzie 1971: 78) may be the continuation of an
otherwise unattested Iranian form without suffixes (< PIE *ḱyeh1-wo-) or again,
and more plausibly in my opinion, an early -Ø outcome of the velar suffix, trac-
ing back to a previous MP *siyāg (note that this latter form would be attested,
according to Hübschmann (1895: 245) and Horn (1898: 175), who mention Pahl.
siyā-k, but it is unrecorded in Nyberg 1974 and in MacKenzie 1971).
A case in which we have two allotropes, attesting the outcomes -Ø and -h, is
the word for ‘grass’, NP giyāh, beside giyā and giyāɣ (Steingass 1892: 1108; Horn
1898: 175), which occurs in MP as giyā [gbyʾ] and giyāh [gyʾh],17 and it is compara-
ble with ManMP gyʾw: the Middle Persian final -Ø and -h are different outcomes
of an older final *-k. The etymon of this word is unclear, but Horn (1898: 175)
holds that in NP giyāh an allotrope of the suffix -āk is involved.
Another instance of double outcome, in this case -g and -Ø, seems to be MP
padistāg [ptstʾk'] and padistā ‘promise, vow’ < padistādan, padist- ‘to promise,
vow’ or padist ‘promise, vow’ (MacKenzie 1971: 63); cf. ManMP pdystʾw; this word
is probably connected with MP stūdan, stāy- ‘to praise’, also stāyīdan, stāy-, NP
sutūdan or sitāyīdan, sitāy- (MacKenzie 1971: 77; Nyberg 1974: 180); OInd. stav-,
PIE *stew- (EWA 2: 757 f.).
A further possible case of -Ø is MP xormā [Tg, hwlmʾk'] ‘date’, NP ‘id.’ (Mac-
Kenzie 1971: 94), provided that we accept MacKenzie’s reading: the spelling
[hwlmʾk'] seems to suggest xormāg. Finally, Salemann (1895: 321) mentions the
MP writing [prtʾ], followed by (sic), for the MP word fradāg [prtʾk'] ‘tomorrow’,
which may be a mistake or a further instance of -Ø.
In any case, it is noticeable that the possible MP outcomes -Ø, -x and -h of
OIr. *‑ka- only appear after a long vowel, generally ā.
According to Horn (1898: 186), there is also the possibility of the outcome -h <
-k after a long -ō-: he refers to doublets like MP andōh [ʾndwh] ‘sorrow’ (Mac-
Kenzie 1971: 9) and andōg (but this latter is unrecorded in MacKenzie 1971 and
Nyberg 1974; see, however, andāg ‘id.’, MacKenzie 1971: 9) and anbōh ‘quantity’
and anbōg (both unrecorded in MacKenzie 1971 and Nyberg 1974). Horn (1898:

17
Nyberg 183 transliterates giyā [gbyʾ] and giyāh [gbʾh]; CPD 36 reads giyā(h) and transliterates
[gyʾh] and [gȳʾh].
306 Claudia A. Ciancaglini

186) suggests that MP andōh ~ andōg (NP andōh, ManMP ʾndwx-) may come
from OP *ha(n)dava-ka-, cognate to the OInd. root dav- ‘to burn’ (EWA 1: 707 f.),
and that MP anbōh ~ anbōg may derive from OP *ha(n)bava-ka- ‘the gathering,
collecting’, -h ~ -g being different outcomes of *-ka-.

6. CONCLUSIONS
As against the traditional view, we can conclude that the OIr. suffix *-ka- is
continued in MP by a great number of outcomes: apart from those enlarged with
the preceding vowel which gave rise to highly widespread and productive MP
suffixes (-ag, -āg, -ak, -īg etc.), there are minority outcomes that generally do not
function as autonomous suffixes in MP, in particular:
1) -k or -g preceded by a long vowel (e.g. pāk, nēk, nibēg), both written <-k'>;
2) -k <-k'> preceded by a voiceless consonant (e.g. hušk);
3) -g <-g> preceded by a voiced consonant (e.g. dāng, buzurg);
only after -ā- and -ō-:
4) -h <-h> (e.g. amāh, andōh);
only after -ā-:
5) -x <-h> (e.g. sūrāx);
6) -Ø (e.g. padistā).
The second point is that the Pahlavi system of writing is less ambiguous than
generally claimed and, as far as our topic is concerned, records some of these al-
lotropes, namely -g after voiced continuant, -Ø, -x and -h.
A final consideration is that this variability in the outcomes of the OIr. suffix
*-ka- does not appear only in the Early New Persian period (where the massive
presence of interdialectal borrowings has been recognized and studied by many
scholars) but already affects Middle Persian.
Nevertheless, this Middle Persian variability, in my opinion, cannot be ac-
counted for only as a trace of early dialectal variability. I believe that these dif-
ferent outcomes of OIr. *-ka- also represent the different diachronic stages of the
lenition process that affected OIr. *-k-. As we have seen, this lenition process
touched *-k- later than the other OIr. voiceless stops, and the traces of its differ-
ent stages—mainly fricativization, voicing, and loss—remain in Middle Persian
at the level of lexical stratification.
These two explanations, diachronic change and dialectal variability, are not
mutually exclusive. Moreover, it is not necessary for all the abovementioned
cases to be accounted for by a single explanation. For example, the MP outcomes
 Middle Persian Continuants of the Old Iranian Suffix *-ka- 307

-h and -x do not necessarily depend on interdialectal borrowings, given that

there is a trace of this fricativization already in Old Persian, at least as far the
personal pronouns OP ahmāxam ‘we’ and *xšmāxam ‘you (pl.)’ are concerned.

BIBLIOGRAPHY
Asatrian, G. (1989), Otglagol'nye imena v srednepersidskom i parfjanskom, Erevan.
Back, M. (1981), “Die mittelpersische Lautverschiebung: ein Stilwandel”, Sprache 27: 178–186.
Bailey, H. W. (1979), Dictionary of Khotan Saka, Cambridge.
Bartholomae, Ch. (1904), Altiranisches Wörterbuch, Strassburg.
Bolognesi, G. (1960), Le fonti dialettali degli imprestiti iranici in armeno, Milano.
Ciancaglini, C. A. (2008), Iranian Loanwords in Syriac, Wiesbaden.
―― (2012), “Il suffisso indo-ir. *-ka- nelle lingue iraniche antiche”, Archivio Glottologico Italiano
98/1: 3-33.
Debrunner, A. (1954), Altindische Grammatik. Band II.2: Die Nominalsuffixe, Göttingen.
Durkin-Meisterernst, D. (2000), “Erfand Mani die manichäische Schrift?”, Studia Manichaica. IV In-
ternationaler Kongreβ zum Manichäismus Berlin, 14.-18. Juli 1997, eds. R. W. Emmerick; W.
Sundermann; P. Zieme, Berlin: 161–178.
―― (2005), Manichean Script, Encyclopaedia Iranica, online (www.iranicaonline.org).
Edgerton, F. (1911), “The K-Suffixes of Indo-Iranian”, JAOS 31: 93–150; 296–342.
Gnoli, Gh. (1996), “Über das iranische hṷarnah-: lautliche, morphologische und etymologische Pro-
bleme. Zum Stand der Forschung”, Altorientalische Forschungen 23: 171–180.
―― (1999), “Farr(ah)”, Encyclopaedia Iranica 9: 312–319.
Henning, W. B. (1947), “A Sogdian Fragment of the Manichaean Cosmogony”, BSOAS 12: 306–318.
―― (1958), “Mitteliranisch”, Handbuch der Orientalistik, I Abt., IV Bd., Leiden–Köln.
Hinz, W. (1975), Altiranisches Sprachgut der Nebenüberlieferungen, Wiesbaden.
Horn, P. (1893), Grundriss der Neupersischen Etymologie, Strassburg.
―― (1898), “Neupersische Schriftsprache”, Grundriss der iranischen Philologie, I Bd., 2. Abt.,
Strassburg.
Hübschmann, H. (1895), Persische Studien, Strassburg.
Kent, R. G. (1953), Old Persian, New Haven.
Korn, A. (2009), “Lenghtening of i and u in Persian”, Exegisti monumenta. Festschrift in Honour of
Nicholas Sims-Williams, eds. W. Sundermann, A. Hintze, F. de Blois, Wiesbaden: 197–214.
MacKenzie D. N. (1971), A Concise Pahlavi Dictionary, London.
Mayrhofer, M. (1956–1980), Kurzgefasstes etymologisches Wörterbuch des Altindischen (KEWA), Hei-
delberg.
―― (1977), Die avestischen Namen, Wien 1977 (IPNB I/1).
―― (1992–2001), Etymologisches Wörterbuch des Altindoarischen (EWA), Heidelberg.
Monier-Williams, M. (1899), A Sanskrit–English Dictionary, London.
Naveh, J. (1982), Early History of the Alphabet. An Introduction to West Semitic Epigraphy and Palae-
ography, Jerusalem–Leiden.
308 Claudia A. Ciancaglini

Nöldeke, Th. (1892), Persische Studien II, SBWA, Phil.-hist. Kl. 126, Wien.
Nyberg, H. S. (1974), Manual of Pahlavi, II, Wiesbaden.
Panagl, O. (1987), “Productivity and Diachronic Change in Morphology”, Leitmotifs in Natural Mor-
phology, eds. W. U. Dressler et al., Amsterdam–Philadelphia: 127–151.
Pisani, V. (1935), “Ἑλληνοκελτικά”, Revue des études anciennes 37: 145–160.
Pisowicz, A. (1985), Origins of the New and Middle Persian Phonological Systems, Cracow.
Rastorgueva, V. S. / E. K. Molčanova (1981), “Srednepersidskij jazyk”, Osnovy iranskogo jazykoznani-
ja II. Sredneiranskie jazyki, Moskva: 147–232.
Salemann, C. (1895), “Mittelpersisch”, Grundriss der iranischen Philologie, eds. Geiger, W., Kuhn, E.
Strassburg 1895–1901, Bd. 1, Abt. 1: 249–332.
―― (1908), Manichaeische Studien I, St.-Pétersbourg.
Shaked, Sh. (1987), “Aramaic. III. Iranian Loanwords in Middle Aramaic”, Encyclopædia Iranica,
II/3: 259–261.
Sims-Williams, N. (1989), “Sogdian”, Compendium Linguarum Iranicarum, Wiesbaden: 173–203.
Skjærvø, P. O. (1996), “Aramaic Scripts for Iranian Languages”, The World's Writing Systems, eds. P.
T. Daniels, W. Bright, New York–Oxford: 513–535.
―― (2009), “Middle West Iranian”, The Iranian Languages, ed. G. Windfuhr, Oxford: 196–278.
Steingass, F. (1892), A Comprehensive Persian-English Dictionary, London.
Sundermann, W. (1989), “Mittelpersisch”, Compendium Linguarum Iranicarum, Wiesbaden: 138–
164.
Szemerényi, O. (1950), “Contributions to Iranian lexicography”, JAOS 70: 226–236.
Telegdi, S. (1935), “Essai sur la phonétique des emprunts iraniens en araméen talmudique”, JA 226:
177–256.
Weber, D. (1997), “Pahlavi Phonology”, Phonologies of Asia and Africa (Including the Caucasus), II,
ed. A. S. Kaye, Winona Lake, Ind.: 601–636.
Whitney, W. D. (1924), Sanskrit Grammar, Leipzig.
 Vowel Length in Middle Persian Verbal Endings*

Desmond Durkin-Meisterernst
Berlin-Brandenburgische Akademie der Wissenschaften, Berlin

Abstract
This paper deals with the question of vowel-length in Middle Persian verbal endings, e.g. the pres.
3rd sing. written in Manichaean script with -yd, which may indicate a long or a short e. A survey is
given of the Middle Iranian languages and the way(s) they deal with the main present formations
of Old Iranian. The author concludes that the vowel was likely to have been long and explains the
short-vowelled endings in Modern Persian as the result of shortening in the accent group formed
by the past participle together with the auxiliary verb.

Keywords
Middle Persian, Middle Iranian, Verbal Endings

The question of vowel length in Middle Persian verbal endings is an important

one, touching on the interrelation between Old and Middle Persian on the one
hand and Middle and Modern Persian on the other. In Old Persian, as in Old Ira-
nian in general, we can observe an extensive system of formations of the present
stem, some of which can be expected to yield long-vowelled endings in Middle
Persian. In Modern Persian the category of ‘vowel length’ is no longer present.
The question can be posed in the following manner: is the Middle Persian
ending of the 3rd singular present -yd in Manichaean script (e.g. kwnyd) to be in-
terpreted as /-ēd/ or possibly /-ed/ < Old Iranian -ati and, in the latter case, as-
suming continuity from Old Iranian to Modern Persian -ad (e.g. mi-konad)? Ad-
ditionally, is -yd, the spelling used for three endings, that of the 3rd singular pre-
sent, the 2nd plural present and the imperative plural, really uniform? In addi-

*
I presented a preliminary version of this paper at the board meeting of the Societas Irano-
logica Europea in Vienna on the 30th of March 2012 and would like to express my gratitude to the
colleagues who made comments then: Velizar Sadovski, Florian Schwarz, Heiner Eichner. I would
also like to thank Matthias Weinreich for a discussion on this topic. It is my great pleasure to dedi-
cate this version to Garnik Asatrian.
 Koninklijke Brill NV, Leiden, 2015
310 Desmond Durkin-Meisterernst

tion -yd also occurs as the spelling of the past stem in /-īd/ but that is not an is-
sue here.
Modern Persian has short-vowelled endings, but Modern Persian has lost the
category of vowel length in any case. The 3rd singular is: mi-konad. The 2nd plu-
ral present is: -īd. Dari and Tajiki have the 3rd singular in -ad, and the 2nd plural
in -ed. We can also compare the Kurdish 3rd sing. -ad, and 2nd pl. -êd. These ex-
amples show a consistent short-vowelled 3rd singular in -ad but indicate that the
2nd plural is different. Despite the fact that there is no basis in Old Iranian for
this and the fact that in Manichaean Middle Persian the same spelling is used for
both endings there is the possibility that it being used to write endings that the
speakers were able to distinguish. In Pāzand the 3rd singular is kunet and the 2nd
plural imperative kunēt; however, the situation is confused by examples, such as
agar gōêt, agar gōênd, with apparently long vowels. In the Middle Persian in-
scriptions, in Middle Persian Zoroastrian texts and in other forms of Pahlavi, ver-
bal endings are very often written without vowels, i.e. the script system notes
only the consonants. This is despite the fact that the vowels are important for
distinguishing the subjunctive (with -ā-) from the present (whether this was -a-,
-e- or -ē-). The reader has to decide which vowel to supply. This creates the im-
pression of short-vowelled endings, which, I think only accidentally, seems to
agree with the short vowels of Modern Persian and to confirm the Pahlavi read-
ing. Many older editions, therefore, transcribe the verbal forms with short vow-
els.
On the other hand Old Iranian has a complex system of present stem for-
mations, some of which can be expected not to have disappeared entirely with-
out a trace. Examples (3rd sg. only) are:
*as-ti athematic (no vowel between the root and the ending)
*bar-a-ti thematic (a vowel -a- between the root and the ending)
*dār-ay-a-ti iterative/causative and thematic (a suffix -ay- and a vowel -a-)
*grb-nā-ti several other formations also occur (e.g., suffix -nā-)

Variation in the vowel of the root (shortening, lengthening) also occur. The
Old Iranian suffix -aya- occurs both with and without lengthening of the vowel.
The lengthening seems to develop in order to indicate the causative or factitive
function.
The following table gives some examples for what happens to the first three
formations in this system in the Parthian and Middle Persian:
 Vowel Length in Middle Persian Verbal Endings 311

Old Iranian *asti *barati *dārayati

Parthian ʾst /ast/ bryd /barēd/ dʾryd /dārēd/
Middle Persian ʾst /ast/ kwnyd /kunēd/, kwnd /kund/, dt /dat/ dʾryd /dārēd/

Middle Persian and Parthian are quoted in Manichaean script, not Pahlavi.
Remarkably Parthian, usually regarded as being very close to Middle Persian,
seems to show that the *barati and *dārayati classes yield the same result, which
is spelled -yd. The spelling is ambiguous and could be /-ed/ for the barati class (<
*-a-ti) and /-ēd/ for the dārayati class (< *-aya-ti) or it could be uniformly /-ēd/
for both, meaning that the *barati class has taken over the endings of the *dāra-
yati class. The same situation may apply to Middle Persian: kwnyd could be /ku-
ned/ or /kunēd/. Additionally, kwnd could be /kund/ or even /kunad/, as an iso-
lated relict of the *barati class. The spelling kwnd is attested 11 times, the spelling
qwnd 12 times and, therefore, the form is not negligible. Only the verb kun- ‘to do’
has this behaviour. These spellings occur more than once in some texts and not
usually at the end of the line (where the scribe sometimes leaves letters out).
E.g., qwnd occurs twice in the Šābuhragān, which is an old text, the only 3rd sg.
pres. of this verb there. The same applies to M 7981. There is the possibility that
kun-, frequently used as an auxiliary verb together with a preceding noun, might
have been subjected to an accent rule indicating the unit formed by the noun
and the verb and weakening the verbal ending. On the other hand, some of the
attestations are in younger texts, such as MIK III 4983 V 4; M235 I R 4, which are
texts of the 9th-10th c. Perhaps here these do represent early Modern Persian
/kunad/. In all, this situation continues a pattern otherwise observable between
Middle Persian and Parthian. Despite the general agreement, the broader attes-
tation of Middle Persian allows us a glimpse of greater variety in Middle Persian
than that attested in Parthian. The Middle Persian imperfect ʾkylydy /*akirī/ “it
was done” in Kerdir’s inscriptions is a case in point. On the other unusual
spelling in Manichaean Middle Persian, dt (in M177 R 14; M32b B 11 and M627 R ii
7) Henning (apud Tsui 1943: 217) wrote “from OIran. *dadāti by haplology”. It is
rather from *daθt (cf. Pahl. dah-) < *daθati. The verb should have a long vowel s.
aw. dadāiti, jaw. daδāiti. However, the short-vowelled form is attested: jaw. da-
θaiti (Hoffmann/Forssman 1996: 98).
If Parthian and Middle Persian do show the extension of the Old Iranian
*-aya- class we can compare this structural development with Middle Indian
where, for example, Pālī shows the spread of a suffix <e> (i.e. /ē/) < *-aya at the
expense of other present stems. This is not just a mechanical simplification and
312 Desmond Durkin-Meisterernst

harmonising of the complex system of present stems, it can also be understood

as the extension of an iterative marker to reinforce the primary function of the
present stem similarly to what KuryƗowicz demonstrated for the Slavic lan-
guages. The extension is restricted to the suffix *-aya- > -ē- and does not include
the lengthened vowel of the stem, which in Middle Persian is a sign of a factitive
or causative verb.
Interestingly, the other Middle Iranian languages also present a complex sit-
uation:
Old Iranian *asti *barati *dārayati
Khotanese aśtä bīḍi paderīndä (pl.), panemäte
Sogdian ʾsty /asti/ βrty /βarti/ δʾrt /δārt/
Khwarezmian yit:i /yitti/ γwyc /γaweč/? δʾrycʾ /δārēč-ā/
Bactrian αστο /ast/ κιριδο /kirid/? ληρδο /lērd/

Khotanese bīḍi shows in -ḍ- a reflex of the r of the root bar; the vowel ī (prob-
ably indicating vowel quality rather than length) may reflect a modification of
the root vowel by a trace of the thematic vowel that in Old Iranian stood be-
tween the final consonant of the root and the ending. The form panemäte shows
a spelling with -e- in the stem. Since this is in Brāhmī script <e> stands for a long
ē, which here reflects a long ā and a modification of that by a front vowel in the
following syllable, such as -ē- from Old Iranian *-aya-, though various other de-
velopments are also possible, such as -aya- to -ya- by syncope, which then affects
the previous syllable. The spelling <ä> = /e/ (short) may indicate this. The ‘long’ ī
in the plural form paderīndä is not a reflection of Old Iranian *-aya- or a result-
ing -ē- because bīḍi from Old Iranian *barati also has a ‘long’ ī. In Sogdian βrty
from Old Iranian *barati could indicate /βarti/ or /βarati/ but Christian Sogdian,
the only variety which occasionally uses additional vowel signs, never indicates a
vowel between the root and the ending. More surprisingly, the form δʾrt also has
no vowel in that position despite its clear derivation from Old Iranian
*dārayati―the derivation being confirmed by the length of the stem vowel. The
accent rule in Sogdian obviously deletes all vowels after a long vowel and pre-
sumably affected the Old Iranian suffix *-aya- before a possible contraction to a
long -ē-. Khwarezmian may reflect a distribution -e- and -ē- for the two classes
but vowel length cannot be reliably noted in Arabo-Khwarezmian. The
Khwarezmian imperfects γwd/γwyd (the variance suggesting short e?) and impf.
δʾryd (long ē?) may also confirm the distribution; they also show non-palatalised
 Vowel Length in Middle Persian Verbal Endings 313

endings. Bactrian κιριδο might be /kered/ representing *karati. Ληρδο clearly

shows, with long ē, its origin from *dārayati; here too the -aya-suffix has resulted
in the change from ā to ē in the preceding syllable but there is also definitely no
vowel between the final root consonant and the ending. The most important
point to be gained from this survey of the developments in the other Middle Ira-
nian languages is that these probably do not show a uniform extension of the
Old Iranian *-aya- class to verbs of the *-a- class. If there has been such an exten-
sion in Parthian and Middle Persian it would, therefore, have to be a sign of the
less archaic nature of these languages contrasting with the more complex and
more archaic morphology of most of the Eastern Iranian languages―Bactrian
being somewhat of an exception and in many ways quite similar to Parthian.
But if Middle Persian has a 3rd sing. in -ēd how do the short-vowelled endings
in Modern Persian come about? Are they inherited? Do they reflect the type
*barati and, therefore, continue Middle Persian kwnd as /kund/ with loss of the
non-accentuated vowels or rather as /kunad/? I would like to suggest an alterna-
tive explanation, namely, that the Modern Persian verbal system is the result of a
re-allignment (see G. Haig’s term in a different context) from the split-ergative of
Middle Persian to the consistently non-ergative construction of Modern Persian.
For reference, see the following table of the Middle Persian split-ergative.
Forms in round brackets are expected but not (always) present. The first three
sentence types show the non-ergative structure; only the past transitive has the
ergative. In the non-ergative structure the first agent (the logical subject) agrees
with the verb; in the ergative structure the second agent (the logical object)
agrees with the verb. In both the intransitive and transitive past sentences the
verbal unit is composed of two parts, a semantic first part and a verbal second
part. Remarkably, the verbal second part is always in the present.
Middle Persian 1st agent 2nd agent verb translation
Present intransitive an hēm ‘I am’
Present transitive (an) (ō) tō xwānēm ‘I call you’
Past intransitive (an) āmad hēm ‘I came’
am/man (tō) xwand hē ‘I called you’
Past transitive u-t dēwān bast hēnd ‘You bound the dēw
(demons)’

In Modern Persian, despite the loss of the split-ergative, the basically present
system has been retained. Here too, the verb ‘to be’ is at the centre of the system.
Of the four sentence types, pres. intrans., pres. trans., past intrans., past trans.
314 Desmond Durkin-Meisterernst

only the pres. trans. has a finite verb with inflexional endings, all the other sen-
tences use the same finite verb ‘to be’:
Modern Persian 1st agent 2nd agent verb translation
Present intransitive man hast-am ‘I am’
Present transitive man kār mikonam ‘I do/am doing a task’
Past intransitive man āmad-am ‘I came’
Past transitive man kār kard-am ‘I did a task’

I suggest that in the intransitive present and in combinations with a past

stem (the intransitive and transitive past) the auxiliary verb is subjected to par-
ticular accent and intonation rules that must have reduced any long-vowelled
endings to short-vowelled ones, e.g., āmad hēm (two accents) > āmád-hēm >
āmád-ham > āmád-am. This results in a form that contains a short -a- vowel,
which is then used in the whole paradigm. The Pahlavi Psalter fragment shows
both formations side by side: bwhty HWEm PsB 12 V 5 = Psalm 136, 24 and bwhtm
PsB 5 V 16 = Psalm 123, 7, both meaning ‘(he) saved us’. It is surely relevant that
the first occurrence is not at the end of the sentence, whereas the second one is
in final position. Furthermore, the apparent identity of this form with that of the
enclitic personal pronoun often used in the Middle Persian ergative construction
contributed to the reinterpretation of the ergative structure of transitive sen-
tences in the past to the uniform non-ergative construction in Modern Persian.
Sentences, such as u-m .... kird ‘I made ...’ could, with an initial verb, have the
form kird-am .... In the Šāhnāma guft-aš ‘he said’ occurs frequently. While the
verb in initial position is not the norm in all forms of Middle Persian, and such
occurrences in Manichaean Middle Persian are often regarded as being the re-
sult of conservative translations from Aramaic originals with initial verbs, in fact
the position of the verb is fairly free in Middle Persian. It is, therefore, quite likely
that in the spoken language verbs were used in initial position. Answers to
questions, for example, will often have contained just a verb. If the sentence re-
ferred to the past and needed an enclitic pronoun, the only possible formation
was a sequence of verb + enclitic pronoun, such as kird-am, guft-am etc. ‘Yes, I
have done (it)’, ‘Yes, I have said (it)’. This also applies to the enclitic pronouns of
the second and third persons, kird-at and kird-aš and also to the plural kird-
amān, kird-atān and kird-ašān thereby establishing a short a vowel for the whole
paradigm. But only the first person kird-am is ambiguous in the sense of also
 Vowel Length in Middle Persian Verbal Endings 315

being the result of the development of kird hēm to kird-am. On the other hand,
the third singular of the past participle plus auxiliary verb in the expected form
ast ‘is’ regularly occurs without the ast in both intransitive and transitive sen-
tences (āmad ‘he/she/it came’ and not āmad ast; kird ‘he/she/it made’ and not
kird ast), so that this form cannot have been the origin of the developments. The
fact that this position seemed to be empty, kird ø, left it open to be filled with the
appropriate form of the auxiliary verb according to the agreement rules of
Mondern Persian, e.g. u-m …. kird ø to …. kird-am ‘I did it’. This is a spoken phe-
nomenon, showing the real replacement of the ergative construction by the non-
ergative one. But it is also a literary phenomenon showing interference of (Early)
Modern Persian in the transmission of Pahlavi texts. There the scribes often re-
place ø with -am (or hēm) but they usually do not delete the enclitic pronoun at
the beginning of the sentence and so produce a hybrid construction, u-m …. kird-
am.

BIBLIOGRAPHY
Hoffmann, K. / B. Forssman (1996), Avestische Laut- und Flexionslehre, Innsbrucker Beiträge zur
Sprachwissenschaft 84, Innsbruck.
Tsui, Chi (1943), “Mo Ni Chiao Hsia pu Tsan” (“The Lower (Second ?) Section of the Manichaean
Hymns”, with contributions by W. B. Henning), Bulletin of the School of Oriental and Afri-
can Studies 11: 174-219.
 Some Khwarezmian Names

Vladimir Livshits
Institute of Oriental Manuscripts, Russian Academy of Sciences, St. Petersburg

Abstract
The earliest evidence of the Khwarezmian onomastics dates back to the 5th century B.C. The arti-
cle presents the analysis and etymology of several old Khwarezmian names, mostly theophoric, at-
tested in the Aramaic documents from Elaphantine and on an ostracon found during the excava-
tion of the fortress of Kalaly-gyr 2.

Keywords
Khwarezmian, Khwarezmian Onomastics, Iranian Onomastics

Khwarezmian, one of the Eastern Iranian languages of the Middle Iranian pe-
riod, was spoken by the inhabitants of the land in the lower stream of the Amu
Darya river (Oxus of the Classical authors, Waxshu). There are no texts in a lan-
guage closely related to Khwarezmian, nor any of its direct descendants have
survived. However, to a certain extent, Khwarezmian reveals some historico-
phonetical and morphological characteristics similar to Sogdian.
It is hard to define the exact time period when Khwarezmian was a living
language. The earliest texts in this language are: inscriptions on two silver phials
found in 1989 during the excavations of the first burial place of Isakovskij, lo-
cated near Omsk; inscriptions on pottery and ostraca excavated from the ar-
chaeological sites of Koi-Krylgan-kala, Kalaly-gyr 2, Gyaur-kala 3, Burly-kala, and
Xumbuz-tepe, which can approximately be dated by the period between the 3rd
century B.C to the 1st-2nd cc. A.D. Other Khwarezmian documents, dating to the
2nd and 3rd centuries A.D., are inscriptions on wood and leather (currently kept
at the Institute of Oriental Manuscripts, RAS, St. Petersburg), as well as imprints
on clay that were found during the excavation of the palace of Toprak-kala. The
most interesting of these documents are the lists of men with Khwarezmian
names from extended patriarchal families, with a division between freemen and
slaves. These lists, most likely compiled for the conduction of a census of men fit
 Koninklijke Brill NV, Leiden, 2015
318 Vladimir Livshits

for military service, contain names of the householder, his sons, as well as names
of the male servants of the head of the families, his wife, sons, and sons-in-laws.
The names of males who are registered in this kind of census for the first time
are also indicated.
Two fragmented Khwarezmian documents on leather, excavated in the sur-
roundings of Yakke Parsan settlement, as well as inscriptions on six silver vessels
(found in Prikamye and currently kept in the Oriental Department, of the State
Hermitage Museum), can be dated to the 6th-7th centuries. Another vessel of
similar type is kept in the British Museum; two others with inscriptions dating to
the 8th century A.D. (probably, years 709 and 714 respectively), at the Pushkin
State Museum of Fine Arts. An attempt to read these inscriptions was made by
the present author in 1964 (Tolstov/Livshits 1964).
More than a hundred of inscriptions-epitaphs inscribed in ink on the walls
and lids of alabaster or stone vessels were found during the excavations of the
necropolis of Toprak-kala in the northern Khwarezm (14 km distance from Nu-
kus). Only the ten of these inscriptions have been published yet (more than two
dozen of them are currently kept in the Oriental Department of the State Her-
mitage Museum). Twenty four short inscriptions on the fragments from ossuar-
ies were found during the excavations in the necropolis of Mizdahkan.
Within the limits of my estimation, the earliest evidence of Khwarezmian
onomastics dates to the 5th century B.C. In the Aramaic documents, among the
colonialist soldiers mentioned in the Achaemenid garrison of Elephantine, in the
south of Egypt, there is an indication of a certain Khwarezmian, drgmn br ḥršyn
/Dargamanah puϑr (puhr) Xwaršaina/ ‘Dargamanah, son of Xwaršain’ (Cowley
1923: 6.2, 7, 8, 17, 22; 8.5); cf. late Khwarezmian pr, ’pr/pur/əәpur?/ «son» (Henning
1956: 432; Benzing 1983: 74, 523).1
The name Dargamanah has a transparent etymology (lit. ‘longminded’); cf.
Av. darəәga-, darəәγa-, OPers. darga- ‘long, extended’, MPers. dagr ‘long’, NPers.
dēr, Manich. MPers. dgr /dagr/, dyr /dēr/. H. H. Schaeder interpreted this name
as ‘The one who endures long’ (Schaeder 1930: 290), which does not seem con-
vincing.2

1
Khwarezmians are mentioned in Babylonian documents from the 595-570 B.C., registering
supply of provisions to foreigners from the imperial treasuries of the Achaemenids (cf. Dandamaev
2009: 312, 315б 343б 363), but I have not found any Khwarezmian names in these documents.
2
Cf. Δαργαµάνης, a river-name in Bactria (Ptol. VI. 11.2; see Markwart 1938: 25-30).
 Some Khwarezmian Names 319

N. Bogolyubov thinks that the name of Dargamanah’s father, ḥršyn, should be

read as Xwaršaina- (Bogolyubov 1967: 22, fn. 7), in the Elamite transmission,
Maršena, Pairšena (On.P. 8.998; 8.1299). R. Schmitt sees in this name a hypocoris-
ticon with suffix *-aina-, formed from *xwar- ‘son’. For the name Dargamanah-,
cf. Avestan adjective darəәgō.bāzauu- (lit. ‘with long hands’), Old Indian dīrgha-
bāhav-, Ossetic darğbazyg ‘id.’.
The second Khwarezmian name, Φαρασµάνης, attested from the Achaemenid
period, is known by the Greek transmission. F. Justi has defined the person to
whom the name belonged (Φαρασµάνης, the son of Φραταφέρνης) to be the Achae-
menid Satrap of the Khwarezmians (Justi 1895: 91; cf. Arrian, Anab. IV. 15.4; Curt.
Ruf VIII.1.8; variants in Arrian’s text: Φαρισµάνης, VI.27.3; Φαδασµένης, VII, 6.4). W.
Hinz interprets *Farrasmanah- as an OMedian name derived from OIran. *Far-
nasmana- (lit. ‘(Having) blessed reason’) (Hinz 1975: 95). S. P. Tolstov has pointed
that “according to the data provided by Arrian (IV, 15.4), confirmed in most of
the part, and by Curtius, Khwarezmian Shah Farasman visited Alexander the
Great with some 1500 horsemen during the latter’s winter stay in Bactra (329-
328) and offered him a union against the Kolkhs and Amazons” (Tolstov 1948:
17). F. Justi mentions Φαρασµάνης as the names of several kings of Iberia from the
1-4th centuries B.C. (Justi 1895: 91).3
It is reasonable to assume that the most probable literal meaning of the name
is ‘(Possessing) the arch of heaven’ or ‘(Standing) in front of the arch of heaven’;
cf. Avest., OPers. asan-, ašn- ‘stone’; Av., OPers. asman- ‘sky’; the OId Indian áśan-
, áśman- ‘stone; sky’; MPers., NPers. āsmān, Manich. MPers. and Parth. ’sm’’n,
’sm’n /āsmān/ ‘id.’; cf. the Avest. name Asmō.xvanuuant-.4
Here I present two more examples of theophoric Khwarezmian names. On an
ostracon, found at the placement 1/1 during the excavations of the fortress of
Kalaly-gyr 2 (see Livshits 2004: 189-191), the text of three lines was preserved on
the inner side of the ostracon. It goes:

3
Hübschmann (1897: 90) commented on this name in the following way: “Armenian P‘arsman,
the king of Georgia (Movses Khorenatsi 122); Greek Φαρασµάνης, the king of Iberia, contemporary of
the Emperor Adrianes; Φαρασµάνης, a Kolkh (Procop. Pers. 1.8), the father of Zannas”. He assumed
that this is a Persian name; Movses Khorenatsi knew it from Greek annals, Łazar Parpetsi from
Ełishe,; Georgian P‘arsman (ibid.).
4
Fr. Altheim suggested that Φαρασµάνης comes from OIran. *Xwārazmaina- ‘The land of the
sun (?)’; Av. zəәmaēna- ‘of earth’ (AWb. 1960) ( Altheim/Stiehl 1970: 188).
320 Vladimir Livshits

(1) ršnwδtky hr’γ(?) 3

(2) wḥšwδtky ḥr’γ 3
(3) ‘L KL βntk r(?)ḥ(’)γ 5
(1) For Rašn(u)δātak the tax(?) (makes) 3 (measures of a product or a monetary
item?).
(2) For Waxš(u)δātak the tax(?) (makes) 3 (measures of a product or a monetary
item?).
(3) For all slaves (or servants) (?) (the tax makes) 5 (measures of a product or a
monetary item?).
Only the proper names contained in the presented short document of regis-
try, as well as suffix-ideogrammess L, lit. ‘towards, in the direction of’, apparently
corresponding to the early Khwarezmian aβ, late Khwareamian f- (MacKenzie
1990: 111; Benzing 1983: 265-267), from OIran. *abi, Avest. aibī, aiuui, OPers. abiy,
Old Indian abhí, MPers. ō; and the ideogramme KL ‘everyone, each’, corre-
sponding to early Khwarezmian *wisp, late Kwarezmian wsp, ’wsp /ūsp/ (Mac-
Kenzie 1990: 104; Benzing 1983: 110, 654), Avest. vīspa-, vispa-, OPers. vispa-, Old
Indian víśva-, MPers., Manich., Parth., Sogd. wisp, can be clearly interpreted. Both
ideogrammes are attested in the Khwarezmian documents of the 2nd and 3rd
centuries found during the excavations in Toprak-kala (Livshits 1984: 260, 264).
The meanings of the rest of the words in the text are uncertain. Βntk can
likely have the meaning of ‘a slave, servant’; cf. early Khwarezmian *βandak
(with allophone /d/ in -nd-, like Sogdian βnt’k, βnty, Manich. Sogd. βndy(h), late
Khwarezmian βntk, βydk /βē(n)dek/) (MacKenzie 1970: 530; Benzing 1983: 200,
206), OPers. ba(n)daka-, MPers. bandag, NPers. banda, etc.; cf. the ideogramme
‘BDn’ ‘slaves’ /*βandakī/ina/ with -(i/ī)na, the formant of the direct case, known
from the documents on wood from the palace of Toprak-Kala.
In line 1 the last word before the numerals is hr’γ, in line 2, ḥr’γ, in 3, most
likely, a mistake of the scribe, rḥ’γ (instead of ḥr’γ ?); cf. MPers. harg ‘duty; debt;
tax; effort’, NPers. xarāg, in Talmudic Hebrew krg’. This word is widespread in its
Arabicised form, xarāǰ ‘land-tax’. W.B. Henning demonstrated that it comes from
Aramaic halāk; in OPers. we should have expected *haraka-. The ultimate source,
according to Henning (1935: 212), was the Akkadian ilku, lit. ‘duty, obligation; tax’.
The proper names mentioned in the document are theophoric, Rašn(u)δātak,
Waxš(u)δātak ‘Created by (the deity) Rašnu’ and ‘Created by (the deity) Waxšu’.
Presumably by the 1st century A.D., the period to which the document can be
related to, the final u in Rašnu- and Waxšu- was not already pronounced. It
should be noted that the early Khwarezmian orthography was historical, like the
 Some Khwarezmian Names 321

orthography of other alphabetic systems of Middle Iranian languages based on

the scripts of Aramaic origin (cf. Parthian names ršnw /Rašn/; ršnwdt /Rašndāt/;
ršnw(d)tk /Rašndātak/; ršnwmtr /Rašnmihr/; (r)šnyn /Rašnēn or Rašnīn/ in the
documents dating from the 1st c. B.C. to the 1st century A.D., preserved on the os-
traca excavated from Old Nisa (cf. Livshits 2010: 139-140, №№ 537-541).
God Rašnu is mentioned in the Avesta many times as the embodiment of jus-
tice, order and obedience. According to Zoroastrian beliefs, Rašnu, holding
scales in hands, was very influential during the posthumous trial over the souls.
A special hymn, Rašn-Yašt, was dedicated to this deity. In Middle Persian texts,
Rašn was the deity of justice and equity. It was also the name of the 18th day in
the Zoroastrian calendar.
The earliest stratum of Old Iranian proper names with the theonym Rašnu-
have reached us in the Elamite garb: Rašnubāra- ‘Brought by (the deity) Rašnu-
(?)’ (On.P. 8.1420; Hinz 1975: 200); Rašnudāta ‘Created by Rašnu-’ (Benveniste
1966: 91; On.P. 8.1421; Hinz 1975: 200); Rašnuka-, a hypocoristicon from a com-
pound name with Rašnu (Benveniste 1966: 91; On.P. 8.1422; Hinz 1975: 200);
Rašnuizza = OIran. *Rašnuča-, a hypocoristicon (Hinz 1975: 200); Rášnumauttiš =
OIran. *Rašnuwatī- (toponym, fem.; Hinz, ibid.) and the New Persian name Rašn-
vad in the Shahnama (‘Following the truth (?)’; Justi 1895: 259). Middle Persian
names Rašn, Rašnag (Gignoux 1986: 153, Nos. 804-805); Rašndād; Rašn-Mihr (id.
2003: 57, nos. 285-286); renderings of Pers. names in Syriac: ršndd; ršdyndwk
(Gignoux, et alia 2009: 118-119, nos. 358-359), the Sogdian name ršnδys /Rašnδēs/
(lit. ‘(Possessing) the guise of (deity) Rašnu (?)’)(Lurje 2010: 330, no. 1018). The
late Khwarezmian Ršn is the name of the 18th day in the Zoroastrian calendar
(Al-Biruni 1957: 62).
The second theophoric name in the current document is wḥšwδtky
/Waxšδātak/ (lit. ‘Created by Waxšu, with the genius of AmuDarya’). From the
Khwarezmian anthroponyms containing the name Waxšu in the document from
Toprak-kala on wood we have wḥšwβrk /Waxšβarak/ ‘The fruit/offspring of Wax-
šu’ (Livshits 1984: 258). Waxšu, the genius of the river, referred to by Al-Biruni, is
absent from the Avestan pantheon. However, this deity was popular not only in
Khwarezm, but also in Sogdiana and in Bactria.
During the excavations in the Temple of Oks (Takht-i-sangin, southern Ta-
jikistan), a small votive altar situated on a stone-base was found. On the frontal
surface of the base there was a 4-line inscription in Greek, written by a certain
322 Vladimir Livshits

Atrosōkēs and dedicated to the deity of Amu Darya (Litvinskij et alia 1985: 102-
104; Litvinskij/Pičikjan 2000: 304-307).
The ancient layer of Iranian names with Waxšu was preserved thanks to the
Classical sources. According to Arrian (Anab. III, 20, 3; IV, 18, 3) and Quintus Cur-
tius (IV, 2, 11; VIII, 3, 17), the name *Waxšudāta- (in Greek rendering ̉Οξυδάτης) be-
longed to a ‘noble Persian’ captured by Darius III in Susa and later appointed by
Alexander the Great as the Satrap of Medes (Grenet 1988: 377-378). F. Grenet
(ibid.) considers the name of the famous Oksiart to be a theophoric one and ex-
plains it as *Waxšuta-, “lit. ‘(Created by deities) Waxšu and Righteousness’, or as
*Waxšu-arti-, (lit. ‘(created by) Waxšu and the goddess of Destiny and Bliss’
(Grenet, ibid.). To the same category of theophoric names belongs *Miϑrawaxšu-,
lit. ‘(Created by) Mithra and Waxšu’, the father of a Bactrian, who sojourned in
Delos ca. 180-150 B.C. (Grenet, ibid.).
Bactrian names were also attested in the Greek inscriptions on clay pots
found during the excavations of the treasury of Ai-Khanoum (end of the IV or
the beginning of the III–mid. II c. B.C.). These findings, published by K. Rapin
(1983), contain references to Bactrians who served in the Greek polis as local fi-
nancial functionaries. Among them was Waxšuwazdah- (Ὀχήβοακης), lit. ‘Strong/
Powerful (thanks to) Waxšu’ or ‘The support of Waxšu» and, possibly, *Uxšēbu-
nak (Ὀχήβοακης), the etymology of the second component being obscure (Rapin
1983: 326; Grenet 1988: 376-377).
Bactrian names with Waxšu appear in documents (on leather and parch-
ment), dating from the IV-V centuries A.D., found in the vicinity of the settle-
ment Tochi (Northern Pakistan). In these documents, deciphered and published
by Sims-Williams (2000; 2007), god Waxšu, Bactr. bago i OαχÞo, ΟαχÞοβαγο, ΟαχÞ-
βαγο appears in the names ΟαχÞοβορδο /Waxšburd/ (lit. ‘Given (from the deity)
Waxšu’; ΟαχÞογολο /Waxšgul/, lit. ‘from the kin of Waxšu (?)’; cf. OInd. kulá- ‘kin,
family’; ΟαχÞοβοιαµÞο /Waxšyamš/, lit. ‘Dedicated to (the deities) Waxšu (and)
Yamš’; ΟαχÞοµαργο, ΟαχÞµαρηγο, ΟχÞµαρηγο /Waxšmareg/ ‘Slave/Servant of Wax-
šu’; ΟαχÞοοανινδο /Waxšwanind/ ‘Triumphant (thanks to) Waxšu’ (Sims-Williams
2010: 325; nos. 321-325; ср. Davary 1982: 243-244).
Many theophoric proper names with the component Waxšu- are attested in
the Sogdian documents from the mount Mug, in the manuscripts from Chinese
Turkestan, and on rock inscriptions from the upper Indus River valleys: wxš’βy’rt,
wxwšwβyrt /wUxušuəәvyart/ ‘Acquired (thanks to) Waxšu’; wxšmryk, ’xwšmryk,
’wxšmryk, ’xšwmryk, wxšmryk’ /wUxušmarīk/ ‘Servant of Waxšu’, wxwš’kk, wxwšk
 Some Khwarezmian Names 323

/wUxušak/, wxwšβntk, wxšβntk, wxwšwβntk /wUxušuvandē/ ‘Servant of Waxšu-’;

wxwšpnn /wUxušfann/ ‘Charisma/Fame of Waxšu-’; wxwšw’sk’n, wxwšw’sk’’n
/wUxušuəәskān/ ‘Design/Intention of Waxšu’ or ‘Painting of Waxšu- (?)»;
wxwšwδβ’’r, wxwšwδβ’r, wxwšδβ’r /wUxušJvār/ ‘Gift to Waxšu’; wxwšwk’n /wUxu-
šukān/ ‘Related to Waxšu’ or ‘Descendent from (?)Waxšu’, a patronimyc ?;
wxwšwprn, ’xwšprn /wUxušufarn/ ‘Fame/Charisma of Waxšu’ (Lurje 2010: 418-423,
nos. 1355-1356, 1363-1367; 1369-1370; Livshits 2008: 30; idem 2011: 163).

BIBLIOGRAPHY
Al-Bīrūnī, (1957), Pamjatniki minuvšix pokolenij, Translation and commentaries by M. A. Sal’e, Taš-
kent.
Altheim Fr. / R. Stiehl (1970), Geschichte Mittelasiens im Altertum, Berlin.
Benveniste, É. (1966), Titres et noms propres en iranien ancien (Travaux de l’Institute d’études irani-
ennes de l’Université de Paris, 1), Paris.
Boyce, M. (1977), Word-list of Manichaean Middle Persian and Parthian, (Acta Iranica 9), Téhéran-
Liège.
Benzing, J. (1983), Chwaresmischer Wortindex, mit einer Einleitung von H. Humbach, Wiesbaden.
Cowley, A. (1923), Aramaic Papyri of the Fifth Century B. C., Oxford.
Dandamaev, M. A. (2009), Mesopotamija i Iran v VII-IV vv. do n. é. Social'nye instituty i ideologija, S.
Peterburg.
Davary, G. D. (1982), Baktrisch, Ein Wörterbuch, Heidelberg.
Gignoux, Ph. (1986), Noms propres sassanides en moyen-perse epigraphique (Iranisches Personen-
namenbuch, hrsg. von M. Mayrhofer und R. Schmitt. Bd. II: Mitteliranische Personnenna-
men, Fasz. 2), Wien.
―― (2003), Noms propres sassanides en moyen-perse épigraphique, Supplément (1986-2001) (Irani-
sches Personennamenbuch, hrsg. von R. Schmitt, H. Eichner, und B. Fragner. Mittelirani-
sche Personennamen, Fasc. 3), Wien.
―― / Chr. Jullien / Fl. Jullien (2009), Noms propres syriaques d’origine iranienne (Österreichische
Academie der Wissenschaften. Philosophisch-historische Klasse. Sitzungsberichte, 789 Bd.
Iranisches Personnennamenbuch, hrsg. von R. Schmitt, H. Eichner, B. G. Fragner und V.
Sadovski. Bd. VII. Iranische Namen in semitischen Nebenüberlieferungen. Fasz. 5), Wien.
Grenet, F. (1988), “L’onomastique iranienne á Aï-Khanoum”, Bulletin de correspondence hellénique,
T. 107, Paris: 373-381.
Henning, W. B. (1935), “Arabisch harāğ”, Orientalia IV, Roma: 291-293.
―― (1956), “The Khwarezmian Language”, Z. V. Togan’a Armağan, Istanbul: 421-436.
Hinz, W. (1975), Altiranisches Sprachgut den Nebenüberlieferungen, Wiesbaden.
Hübschmann, H. (1897), Armenische Grammatik, I. Armenische Etymologie, Leipzig.
Justi, F. (1895), Iranisches Namenbuch, Marburg.
Litvinskij, B. A. / Ju. G. Vinogradov / I. G. Pičikjan (1985), “Votiv Atrosoka iz xrama Oksa v Severnoj
Baktrii”, Vestnik drevnej istorii 4: 84-110.
324 Vladimir Livshits

―― / I. R. Pičikjan (2000), Ellinističeskij xram Oksa, vol. 1, Moskva.

Livshits, V. A. (2008), Sogdijskaja epigrafika Srednej Azii i Semireč'ja, S. Peterburg.
―― (1984), “Dokumenty”, Toprak-kala. Dvorec, Moskva: 251-286.
―― (2010), Parfjanskaja onomastika, S. Peterburg.
―― (2011), “Ličnyje imena v xorezmijskom jazyke”, Vestnik drevnej istorii 4: 156-166.
Lurje, P. B. (2010), Personal Names in Sogdian Texts, (Österreichische Academie der Wissenschaf-
ten, Philosophich-historische Klasse. Sitzungberichte, 818 Bd. Iranische Onomastik, hrsg.
von B. C. Fragner, und V. Sadovscki. N 7. Iranisches Personennamenbuch, hrsg. von R.
Schmitt, H. Eichner, B. G. Fragner, und V. Sadovscki. Bd. II: Mitteliranische Personenna-
men, Fasz. 8), Wien.
MacKenzie, D. N. (1970), “The Khwarezmian glossary I”, BSOAS, vol. 33: 540-558.
―― (1990), The Khwarezmian Element in the Qunyat al-Munya, London.
Markwart, J. (1938), Wehrot und Arang, Leiden.
On.P. = M. Mayrhofer (1973), Onomastica Persepolitana, Das altiranische Namengut der Persepolis-
Täfelchen, unter Mitarbeit von J. Harmatta, W. Hinz, R. Schmitt und J. Geibert (Öster-
reichische Akademie der Wissenschaften. Philosophisch-historische Klasse. Sitzungsbe-
richte, 286 Band. Veröffentlichungen der iranischen Komission. Hrsg. von M. Mayrhofer,
Bd. I), Wien.
Rapin, C. (1983), “Les inscriptions économiques de la trésorerie hellénistique d’Aï-Khanoum (Af-
ghanistan)”, Bulletin correspondence hellénique 107, Paris: 315-372.
Schaeder, H. H. (1930), Iranische Beiträge I (Schriften der Königsberger Gesellschaft. 6 Jahr, Ht. 5),
Königsberg.
Sims-Williams, N. (2000), Bactrian Documents from Northern Afghanistan, I: Legal and Economic
Documents (Studies in the Khalili Collection, vol. III: Bactrian), Oxford.
―― (2007), Bactrian Documents from Northern Afghanistan, II: Letters and Buddhist Text, (Studies
in the Khalili Collection, vol. VI: Bactrian), London.
―― (2010), Bactrian Personal Names (Österreichische Academie der Wissenschaften. Philoso-
phisch-historische Klasse. Sitzungsberichte, 806 Band. Iranische Onomastik, hrsg. von B.
G. Fragner und V. Sadovski. Nr. 7. Iranisches Personennamenbuch, hrsg. von R. Schmitt, H.
Eichner, B. G. Fragner und V. Sadovski. Bd. II: Mitteliranische Personennamen, Fasz. 7),
Wien.
Tolstov, S. P. (1948), Drevnij Xorezm, Opyt istoriko-arxeologičeskogo issledovanija, Moskva.
―― / V. A. Livshits (1964), “Decipherement and interpretation of Khwarezmian inscriptions from
Tok-kala”, AAASH 12, Budapest: 231-251.
 Kurdish bažn, Persian bašn and Other Iranian Cognates

Ela Filippone
La Tuscia University, Viterbo, Italy

Abstract
The paper deals with the lexical set of cognate words to which Kurdish bažn, Persian bašn (gener-
ally recorded as ‘stature’) belong. These words are attested in Middle and Modern Western Iranian
and are dialectologically connoted, being nowadays mostly found in Central dialects, Kurdish, Go-
rani and Zazaki areas. By reviewing lexicographical and textual sources, the author picks out the
semantic peculiarities of the bašn-forms. Starting from the notion of HEIGHT and through meta-
phorical associations and shift of reference, bašn-forms have acquired different senses mostly in-
volving the body domain (‘body’, ‘waist’ etc.) but with extensions to other domains. They have also
undergone grammaticalization processes. As far as Western Middle Iranian is concerned, all the
occurrences of Pahlavi bašn, Manichaean Parthian bašnān and Middle Persian bašnāy have been
listed and discussed.

Keywords
Iranian Philology, Iranian Dialectology, Lexicography, Body-part terminology; Persian bašn, Kurd-
ish bažn

1. Almost all the Kurdish dictionaries record the word bažn, though with slight
differences in meaning and usage. Cognates to this word are widely attested in
Western Iranian. Generally, the main reference is to Persian bašn, which, ac-
cording to Asatrian/Livshits 1994: 92, could be considered as the source of Kur-
manji Kurd. bažn. 1
To Garnik Asatrian, who has been contributing so much to increase the
knowledge of Western Iranian dialectology and comparative linguistics and en-
hance the interest of scholars towards these important research fields, I dedicate
the following comments on the Iranian lexical group to which Kurd. bažn, Pers.

1
I am deeply grateful to cAbdonnabi Salāmi for his precious contribution to this paper. Many
thanks are also due to other friends, who kindly supported my work in different ways. I would like
to mention in particular Desmond Durkin-Meisterernst, Claudia Leurini, Enrico Morano and Adri-
ano V. Rossi.
 Koninklijke Brill NV, Leiden, 2015
326 Ela Filippone

bašn etc. belong, as a token of my great esteem for his scholarly work and my
deep admiration for his enthusiastic involvement in publishing and promoting
cultural activities.

2. Kurmanji bažn is generally given as ‘height, stature (of a person); waist; figure’
in dictionaries; cf. bejn in Kurdoev 1960, Rizgar 1993,2 Chyet 2003. Orbeli 2002 rec-
ords bejin ‘stature, appearance (generally when speaking of alluring figures)’ in
the Kurdish dialect of Moks (Van area), now disappeared. More or less the same
senses are recorded in lexicographical works centred on Kurdish dialects other
than Kurmanji;3 see for instance Sorani Kurd. bejn s.v. taille in Hakim/Gauthier
1993, (cAmādiye) bejn ‘corps, taille d’homme’, (Sinǰār) bêjin ‘corps’ in Blau 1975,
South. Kurd. bažn ‘stature [qadd-o-bālā]’ in Hažār 1990, ‘stature/figure, body’,
s.vv. andām and kālbud in Bābān 1982, ‘stature; slim figure; the length of a dress
or the like [qadd, qāmat, bālā, borz, bolandi-ye lebās va γeyre]’ in Safizāde 2001,
(Mahābādi) bažn ‘stature, figure’ s.v. qadd, andām in Kalbāsi 1983. Blau 1965,
which sometimes appears to be somehow artificial, being a collection of terms
drawn from literary texts, mostly Kurmanji, but also from other dialects, records
both beşen ‘corps/body’ (apparently the ‘Persian’ form) and bejin ‘stature/stature
(height)’.
To the senses listed in dictionaries, one may also add that of ‘unit of meas-
urement equivalent to the height of a person’; see for instance bejin na-dé ‘ce qui
n’arrive pas à la taille d’homme (en parlant de profondeur)’ in Jaba/Justi 1879: 50,
or řō du bažnā hē bilind bū ‘the sun was up at the height of two men’ in Cabolov
2001: 138. The extension to the measurement domain of terms belonging to the
body part lexicon is a universal and in this connection Kurd. bažn behaves like
other words for ‘bodily height’ in different Iranian varieties: cf. qad in dialects of
South Iran (Persian Gulf; Nurbaxš 2003: s.v. ghad) and bālā in Daštestāni

2
In Rizgar 1993, two different headwords bejn are recorded (bejn1 ‘stature, height’ and bejn2
‘waist’).
3
For convenience, the glosses defining Iranian words drawn from dictionaries whose exit lan-
guage is Persian or Russian have been translated into English; the original gloss (in transcription) is
added into square brackets when considered useful to avoid misunderstanding. As for transcrip-
tion, a tendentially phonemic transcription is used for Persian; for all the other Iranian languages, I
have mostly conformed with the systems used by the individual authors of the written sources
from which any single expression has been extrapolated (always mentioned into brackets). In
source references, the page number is not given when the work is (or contains a section which is)
alphabetically ordered.
 Kurdish bažn, Persian bašn and Other Iranian Cognates 327

(Borāzǰāni 2003) and Bušehri (Hamidi 2001), which are also used as a measure-
ment unit equivalent to the height of a person of middle size, generally to meas-
ure the depth of water or wells, or the height of walls and similar (cf. Bušehri in
čāh panj bālā bolandi dārad ‘this well has a depth equivalent to the height of five
persons’).
Kurd. bažn is very frequently found in copulative compounds, in particular in
bažnbāl, bažn-ū-bālā;4 cf. Kurmanji bejnû bal in Rizgar 1993, bejnbal, bejn û bal in
Kurdoev 1960 and Chyet 2003 ‘appearance, looks’. See also Sulemaniye bejn-u-
bala ‘stature, waist’ in Wahby/Edmonds 1966, Sorani bažn-u-bāłā ‘body, figure;
stature’ in Kurdoev/Jusupova 1983, bejnubala in Hakim/Gauthier 1993: s.v. taille,
South. Kurd. bažn-û-bâlâ ‘stature [qadd-o-qāmat, qadd-o-bālā]’ in Safizāde 2001,
etc.
Notwithstanding the apparent consistency of the lexicographical recordings,
there is a difference in usage of the word bažn among the different Kurdish vari-
eties. This dialectal peculiarity was also recognized by Soane (1913: 180), who
contrasted bezhen, marked as belonging to the Northern group («Northern
Hakkārī and Erzerūm and Bāyazid dialects»5) with a ‘common’ Kurdish form lesh
‘body’. On the other hand, he contrasted (1913: 269) a ‘common’ Kurdish form
bezhn, bazhm ‘stature’ with anām,6 marked as belonging to the Southern group
(«Southern Hakkārī and Mukrī (Sauj Bulaq), Bābān, Sulaimānia»7). In Central
and Southern Kurdish, bažn is generally confined to describe the trim and pro-
portioned body of alluring girls or beautiful boys, who, for this reason, may be
compared to poplars or similar tall, slim trees, and therefore it often appears in
collocation with adjectives meaning ‘slim’, ‘beautiful’ or the like. The same usage
of this word seems to imply an aesthetic (positive) judgement. In Kurmanji
Kurdish, besides being used in the sense of ‘body, physical aspect, etc.’, as in
çelengiya bejnê û çavane (Lescot 1942: 100 [1051]) ‘the beauty of his body and his
visage’,8 bažn may also refer to a definite, concrete body part, as illustrated by
single passages, like

4
As remarked by Asatrian/Livshits (1994: 8), bālā is unquestionably a Persian loanword. Mac-
Kenzie (1961: 78 fn.2) points out that Central Kurd. bālā ‘stature’ is only found in bažn-ū-bālā ‘fig-
ure’ and bālā-barz ‘tall’.
5
Cf. Soane 1913: xii.
6
Cf. Pers. andām ‘body; figure, stature; limb’.
7
Cf. ibid.
8
The reference is to the Kurdish hero Mam.
328 Ela Filippone

bejna xwe xemilandin bi silehane (Lescot 1942: 40 [421])

‘et se parèrent de leurs armes’ (ibid.: 41),
where bejn denotes the waist, i.e., the part of the body where weapons are se-
cured, or even by the existence of lexical units like barbažn: cf. ber bejn ‘amulet
or charm worn about the waist’ in Chyet 2003, berbejn ‘amulet carried on the
belt’ in Kurdoev 1960, berbejin ‘amulette; littér. (ce qui est porté) sur le corps’ in
Jaba/Justi 1879.9 The fact the bažn (‘stature, figure, etc.’) has acquired the addi-
tional sense of ‘waist’ is a peculiarity of Kurmanji, as opposed to other Kurdish
dialects10 (and other Iranian languages with cognates to Kurd. bažn). However,
the PHYSICAL CONSTITUTION > WAIST semantic development is not unparalleled.11
In Central and Southern Kurdish, another word whose semantic range partially
overlaps with that of bažn has undergone the same process: qad ‘stature, waist,
trunk of tree, central body of mountain’ (Hažār 1990); (Mahābādi) qat ‘stat-
ure/figure; waist’ (Kalbāsi 1983: s.vv. qad, andām; kamar); (Sulemaniye) qed ‘stat-
ure, waist’ (Wahby/Edmonds 1966).12 In this respect, Southern/Central Kurdish
agrees with the Lori dialects and dialects of Fārs (and also a couple of varieties
among the Central dialects), where qad has taken on the meaning of ‘waist’ and
‘central part of the body/loins’:13 cf. Baxtiāri qad, kad (Sālehi 1990), (Čahār Lang)
qad (Sarlak 2002), Lori kat (Unvala 1958), Kuhgiluye–Boyer-Ahmadi qad (Lamce
1970), Āvarzamāni qa: (Dehghan 1970), Borujerdi qay (Esfandiyāri 1999 s.v.
kamar), Lirāvi–Deylami, Mamasāni qad (Lirāvi 2001: 243), Širāzi qad (Behruzi
1969), Davāni qaδ (Salāmi 2002), Bušehri qad (Hamidi 2001), Daštestāni qad
(Borāzǰāni 2003), etc. all of them meaning ‘waist’. See also Xunsāri qad ‘waist’
(Tasbihi 1975), Sivandi qad ‘dos’ (Lecoq 1979).14

9
Cf. also berbejn ‘amulet, written charm’ in Rizgar 1993.
10
See however bejn-u-bala ‘stature, waist’ in Wahby/Edmonds 1966, quoted above.
11
See, for ex., Russian stan, Italian taglia¸ French taille, etc.
12
Sul. Kurdish be qeda k. ‘thrust into waist-band’ (Wahby/Edmonds 1966) parallels Kurm. Kurd-
ish bejna xwe girêdan ‘to put on a belt’ (Chyet 2003, Kurdoev 1960). See also nāvqad ‘waist; central
part of mountain or tree’ in Hažār 1990 and Sulemaniye nawqed, ‘waist’ in Wahby/Edmonds 1966:
s.v. naw.
13
Cf. Pers. qad ‘stature; size, etc.’ (loanword from Ar. qadd).
14
Similar in structure to Pers. kamarband ‘belt’, one also find qadband ‘belt’ in Baxtiāri (Čahār
Lang, qadvan, Sarlak 2002), Lirāvi–Deylami, Mamasāni (Lirāvi 2001: 243), Širāzi (Behruzi 1969),
Davāni (qaδ band, Salāmi 2002), Zarqāni (γad-band, Malekzāde 2001), Bušehri (Hamidi 2001),
Daštestāni (qadvand, Borāzǰāni 2003), Larestāni (Eqtedāri 1955), Boruǰerdi (qadban, Esfandiyāri
1999), Naqusāni (Dorudiān 1986), Hamādāni (Garusin 1991), Ardestāni (Lecoq 2002), etc.
 Kurdish bažn, Persian bašn and Other Iranian Cognates 329

In the Kurmanji area, we also find bažnak, a derivative of bažn used as a sort
of indefinite pronoun; cf. bejnak, bejnek in Chyet 2003, bejnek in Kurdoev 1960:
s.v. bejn (sense n. 3) ‘someone’:
me dīt bejnek ber bi malê tê (Kurdoev 1960: s.v. bejn)
we saw someone going home
Hakim hîna raneketîye, carkê nerî go ji alîyê çîyê ve bejnak tê (Lescot 1940: 136)
A peine le prince s’était-il couché, qu’il aperçut un individu venant de la mon-
tagne (ibid.: 137).

This is an instance of a grammaticalization process whereby a noun meaning

‘bodily figure’ behaves as an indefinite pronoun (‘an unknown person as seen in
his exteriority’).15
Among the dialects adjacent to the Kurdish-speaking areas, we may note Za-
zaki bejn ‘body (only for height reference)’ in Todd 1985. An illustration of the
usage of this word in (Siwerek) Zazaki is found in Hadank 1932:16
bä ́ zhnâ nä ́ î [bzw. nấî] dä ́ rgâ bä ́ zhnâ ǰä ́ î [bzw. ǰấî] kílmâ (Hadank
(Hadank 1932: 180 [232]) 1932: 180 [233])
‘sein [bzw. ihr] Wuchs ist lang’ jenes [Mannes bzw. jener Frau]
(ibid.: 190) Wuchs ist kurz (ibid.: 190)

In the glossary of the Gorāni version of the legend of Bīžan-u Manīǰa, Mokri
(1966) records bažn as ‘taille, corps’.17 This word occurs five times in the poem. In
four of its occurrences, bažn is part of the copulative compound bažn(u)bālā,
which is glossed ‘stature, taille’ in Glossaire, but in fact is also ‘body’ in some pas-
sages of the text, as in
[har] šaw pištān bažnbālā-šan (Mokri 1966: ١۱٠۰ [232])
at night their bodies (were) locked in a clinch.18

In the only passage where bažn occurs alone, it seems not to mean ‘taille’ or
‘corps’, but rather ‘face, aspect’:

15
Cf. also Russian figura, Colloq. English body ‘a human being, a person; an individual’. Pers.
tan ‘body’ is also used in the sense of ‘person’ (= kas).
16
In the relevant glossary (1932: 150), Hadank records a form bäzhnâ. Final -â, however, is with
all probability the feminine ezafe.
17
See also Mokri 2003: 132 (where bažn ‘corps’ is attributed to the Awromāni–Gorāni dialects),
2005: 247 («Pour désigner le «corps», le persan utilise le mot tan, qui équivaut à l’arabe badan et
aux termes bajn et bašn, relevés dans les anciens fahlaviyāt et encore employés aujourd’hui, dans
les dialectes kurdes, gourani et awrāmānī»).
18
Cf. Mokri 1966: 152 [232] «Toutes les nuits il l’étreint dans ses bras».
330 Ela Filippone

šād ba wa bažn susān-xālān (Mokri 1966: ٩۹ [216])

rejoice over the beautiful face.19

A line from the Dîwanî Seîdî, a collection of Gorāni poems of the 19th c., re-
produced and translated (into Russian) by Jusupova (2000), reads as follows:
şîrînen bejn-it, şîrînen baĺa-t
‘enchanting is your figure, enchanting is your body [[kak] prelestna tvoja figurka,
[kak] prelesten tvoj stan!]’ (Jusupova 2000: 46).

Here the copulative compound bažn(u)bālā has been split and the two
members are both described as being šīrīn ‘sweet, pleasant’: the recurrence of
the same adjective precludes us from further commenting on semantic differ-
ences between bažn and bālā.

3. Traditional monolingual and bilingual Persian dictionaries record bašn (also

bašan) with the meaning of ‘stature, height; body; top, bottom and sides of any-
thing’. See Mocin 1992, Steingass 1963, Desmaisons 1908 and Dehxodā (bašn,
bašan, baš; with previous references). This word is absent in bilingual dictionar-
ies of common usage in Europe, as Lazard 1990, Haim 1992, Rubinčik 1970 (where
however one finds bašobālā ‘stature [rost]’, labelled as ‘rare’), etc. This mere fact
does not mean that bašn is a non-existent or lost word in Persian. It only means
that it does not belong to the common lexicon of the standard literary language.
In fact, it does not belong to Standard Colloquial Persian, as well: Persian speak-
ers native to Tehran do not seem to know it.20 In Cabolov 2001: 138, Pers. bašn,
pointed out as the direct source for Kurd. bažn (-šn > Kurd. -žn regularly in
loanwords; otherwise -šn > Kurd. -n), is marked by the author as an ‘archaism’,
since in the Persian translation of the Kurdish version of the literary text Šeyx
Sanan by Huseyn Javid, the translator, while rendering Kurd. bažn with Pers.
bašn, felt the need to add a line by Sa’di as an explicative gloss to Persian readers.
Nevertheless, bašn is recorded in lexicographical works where special attention
is given to Colloquial Persian, as Naǰafi 1999 and Šāmlu 1998, with the sense of
‘appearance, look [surat-e zāher]’; in both, the only reference is to a passage (bā
bašn-o-qiyāfe-i žende ‘with a shabby look’) from the famous work on the twenti-

19
See susān-xālān ‘aux grains de beauté couleur d’une sorte de lis gris’. Cf. also Mokri 1966 : 151
[216] «Réjouis-toi de voir ces belles esclaves au visage éclatant».
20
I interviewed several people on the matter, all native to Tehran and with a high level of ed-
ucation. None has ever heard this word and was able to recover its meaning. Some of them even
failed to acknowledge bašn as a Persian word.
 Kurdish bažn, Persian bašn and Other Iranian Cognates 331

eth-century social history of Tehran (Tehrān-e qadim, 1979) by Jacfar Šahri (born
in one of Tehran’s oldest neighbourhoods), where bašn occurs as a part of a cop-
ulative compound.
How could we interpret such contrasting data? The problem arises from the
label ‘Persian’, which covers many different linguistic realities, both from a syn-
chronic and diachronic point of view. We should always account for this and try
to define, if possible, the linguistic reality we are referring to. In the case of bašn,
resorting to a category that implies a distance in time (‘archaism’) does not help
us to better understand the actual situation.
In the relevant entry, Dehxodā mostly provides lexicographical attestations,
though he also mentions a couple of passages from classical poetry. The men-
tioned line by Sa’di, quoted in Cabolov 2001: 137, sounds as follows:
agar sarv-i be bālā-ye to bāšad na čun bašn-e delārā-ye to bāšad
though a cypress could be of your it cannot be similar to your beloved
stature figure

Here the poet, playing with the semantic closeness of bālā and bašn, more of-
ten linked in the copulative compound bašn-o-bālā, enhances in this confronta-
tion the semantic differences: bālā only points to a vertical physical dimension,
bašn to a physique, a body, with its individual features and peculiarities.

4. S.v. bašn, Dehxodā quotes a sentence (drawn from Ānandarāj and Ānjomanā-
rā) said to be used time and again by the poor people of Shiraz, when they are in
lack of clothes and food:
na bašn-am pušide va na šekam-am sir ast
I have neither my body covered nor my stomach full
[NOT my body is covered NOT my stomach is full].

If the interpretation given here is the right one, Širāzi bašn clearly refers to
the human body or to that part of the body where one puts clothes on. However,
the same sentence could also be interpreted in a different way. We may infer this
from the fact that it has been quoted in a Širāzi dictionary (Xadiš 2000) where
the author gives bašn as ‘clothing, dress, any cloth with which people keep
themselves warm’, illustrating the point with the sentence bašn-aš kam ast ‘he is
not dressed appropriately with respect to the cold weather [lebās-o-pušāk-ast
nesbat-e sardi-ye havā kam ast]’. Being based on an easily predictable associa-
tion, the metonymical extension BODY > DRESS does not need further explana-
332 Ela Filippone

tions. In English, for instance, body is also used with reference to that part of
dress which covers the body, viz. the trunk. In the case of bašn, this association is
even more predictable, since bašn refers to the human body in its exteriority,
and dress contributes a great deal to characterize the human exterior appear-
ance. The extension of bašn to the clothing domain is also attested in other dia-
lects; see below, in particular Xunsāri bašn and bašnobâr, Hamadāni bašn-o-bār.
In the dialect of Zarqān, a village not far from Shiraz, bašn (also bar-o-bašn)
means ‘features, aspect [sar-o-rui, andām, qiyāfe]’. Malekzāde (2001: s.v.) pro-
vides us with interesting sentences illustrating the usage of this word:
biš az in, be fekr-e sar-o-bašn-e xod-at bāš
more than to this, think to your aspect,
where with ‘to your aspect’ one should probably intend ‘to yourself/to your busi-
nesses;21
ham mard-e dānā-yi ast ham sar-o-bašn-e xub-i dārad
he is a wise man and has a good-looking aspect, as well;
bašn-aš gošne ast
his bašn is hungry (?),
this latter sentence referred to a person who does not care of his own dress and
goes about in ragged clothes. How to interpret the single words in it is however
unclear to me.
In Kermāni, bašn is a word for ‘body’ (cf. ‘body, from the head to the feet [tan,
kālbod, sar tā pā]’ in Purhoseyni 1991); it has therefore a concrete sense, as can be
evidenced from the following passages:
xoš ān bārun ke var bašn-am bebāre na’ un bārun ke var qabr-am bebāre
(Purhoseyni 1991: 552)
good is that rain which falls on my body, not that rain which falls on my grave;
bārun hame-ye bašn-am-o xis kard (Sarrāfi 1996: 45)
the rain wet all my body;
setāre-ye sob bar bašn-am damide/hanu dar entezār-e yār budam (Vahman 1974:
75)22
the morning star breathed upon my body/(and) I was still waiting for (my)
friend’.

21
If my interpretation is correct, we may note here the loss of referentiality (and the beginning
of a sort of a grammaticalization process).
22
From a local dobeyti. In Perso-Arabic writing in the text; my transcription could be imper-
fect.
 Kurdish bažn, Persian bašn and Other Iranian Cognates 333

In the same dialectal area we find Rāvari bašn ‘body, trunk [badane, badan]’
(Karbāsi Rāvari 1987), Bardesiri bašn ‘body, figure (of human being and animal)
[andām-o-heykal-e ādam va heyvānāt]’ (Barumand Sacid 1991), Sirǰāni bašn ‘body
[tan-o-andām]’ (Saryazdi 2001), bašn-e bar ‘the whole frame of the body [tamām-
e heykal] (Mohseni 2002). See also Yazdi bašn ‘the whole body from head to feet
[sar tā pā]’ (Kešāvarz 1993), ‘height of a tree [bolandi-ye deraxt]’23 bašn o búlú ‘hu-
man figure [qadd-o-qavāre-ye ensān, heykal-e ensān]’ (Kešāvarz 1993), Nāini
bašn-o-bāra ‘stature [qadd-o-bālā]’ (Sotude 1986), Anāraki bašnobār ‘stature
[qadd-o-bālā]’ (Sohrābi Anāraki 1994), Xuri bašn-o-bār ‘countenance, figure,
stature [sar-o-surat, andām, qadd-o-bālā] (Šāyegān 2006), Esfahāni bašn ‘stature’
(usually in hendiadys bar o bašn; Tafazzoli 1991: 208), Vajguni bašn (also bašn-o-
bār) ‘body [badan]’,24 Tāri bašn ‘body [badan]’,25 Xunsāri bašn ‘body, frame of the
body; skirt [tan, heykal, dāman (lebās)]’, bašnobâr ‘body; skirt [tan-o-badan,
dāman] (Ašrafi Xānsāri 2004), Deliǰāni bašan ‘1. eye/sight [dide]; 2. face [čehre]’
(Safari 1994), Hamadāni bašn ‘features [sar-o-surat]’; bašn-o-bār ‘individual as-
pect; clothing’ [zāher-e šaxs, sar-o-lebās]; ser-o-bašn ‘features, individual aspect
[sar-o-surat, zāher-e šaxs]’ (Garusin 1991), Aligudarzi (Lorestān Province) bašn
‘aspect, figure [sar-o-andām]’.26
Words belonging to the body part lexicon may lose referentiality and, in case,
be used with relational implications, in particular, spatial implications. This is
what happens to Kermāni bašn, which, besides meaning ‘body’, also means ‘ex-
ternal surface [sath-e xāreǰi]’ (Sarrāfi 1996) and is used as ‘on, near [ruy(-e),
pahlu(-ye)]’: cf. bašn-e divār ‘on the wall’ (Purhoseyni 1991).27 See also Xuri bašn
‘in the nearby, around [atrāf]’ (Šāyegān 2006).
Another lexical domain involved in our reasoning is the one related to the
description of the natural environment. Consider Yazdi bašn-e kâh ‘mountain
top/mountain slope (?) [bašn- (o-bālā-y)e kuh]’ (Firuzbaxš 1997: 36), bašn o blεndi

23
Information kindly sent to me by cAbdonnabi Salāmi (e-mail dated 01/01/2013).
24
My source is http://khpanah.persiangig.com/page2.html.
25
See fn. 23 above..
26
See fn. 23 above.
27
Interestingly, also qad has undergone in many varieties similar grammaticalization pro-
cesses. I am not going to dwell here on the matter; I would only like to quote the (Čahār Lang)
Baxt. sentence bezan gad-e divâr ‘hang (it) on the wall [be divār āvizān kon]’ (Sarlak 2002). Note
that in Colloquial Persian, badane is also used with the sense of ‘side, direction [samt, taraf, rāste]’
(Naǰafi 1999) and describes spatial relations.
334 Ela Filippone

‘mountain top [bālā-o-bolandi]’ (Mazdāpur 1995),28 Xunsāri bašn ‘skirt (of plain)
[dāman (sahrā)]’ (Tasbihi 1975), possibly Tarei bāšna ‘slope of mountain
[dāmene-ye kuh]’ (Asatrian 2011: 116), if -ā- is from a secondary lengthening,29 Lori
bašn ‘upland pasture [čamanzār-e kuhestāni]’ (Adib Tusi 1963-1964). Interest-
ingly, we also find Semnāni bašm ‘mountain pass [gardane]’ (Sotude 1963) and
Gilaki bašm ‘snowy and frost area, snowy mountain pass [mantaqe-ye barfgir va
yaxbandān; gardane-ye barfgir]’ (Pāyande 1987: s.v. kuh), which recall the topo-
nym Bašm, given as the name of a very cold place (between Tabaristān and Ray)
in traditional Persian dictionaries (see Dehxodā, s.v.).

5. As Old and Middle Iranian cognates to Kurd. bažn, Pers. bašn etc., scholars
generally refer to Avestan barəšnu- ‘Erhebung, Höhe, culmen 1) von Bergen; 2)
vom Himmel; 3) vom menschlichen Kopf’ (Bartholomae 1904) and Middle Pers.
bašn ‘top, peak; stature; mane’ (MacKenzie 1971). It is worth noting, however,
that in no passage of the Pahlavi translations of the Avesta the word bašn paral-
lels Av. barəšnu-, whose regular counterpart appears to be Pahl. bālist ‘highest;
summit’. In this case, then, the translators seem not to have followed the com-
mon practice of giving preference to etymologically related items30 (at least, to
the apparently most directly related; i.e., barəšnu- > bašn, being this latter as-
sumedly more directly connected to barəšnu- than bālist31). Compare the follow-
ing passages drawn from Yasna 9, 10 and Vīdēvdād:
Avestan Pahlavi
Y 9.26 baršnuš paiti gairinąm pad bālist abar garān on the mountain tops32
Y 10.3 baršnuš paiti gairinąm pad bālist abar garān id.33
Y 10.17 baršnušuua gairinąm pad bālist abar garān id.34
Vd. 2.23 barəṣ̌ nuš paiti gairinąm pad bālist abar garān id.
Vd. 8.40 barəṣ̌ nūm hē vaγδanəm bālist ī waγdān the top of the (his) head
Vd. 8.41 barəṣ̌ nūm vaγδanəm bālist ī waγdān the top of the head.

28
Cf. bašn o blεndi-ye kah hma-š vapr on ‘the mountain top is all snow-covered’ (ibid.)
29
In alternative to Asatrian’s etymological connection (< pāšna ‘heel’).
30
On this preference in the Avestan-to-Pahlavi translation practice, cf. Josephson 1997: 159 and
Cantera 2004: 303, 314 ff.
31
In the Pahlavi translation of the Avesta, bālist usually renders Av. barəzišta- (see also Cantera
2004: 317). Exceptions are few, as in the case of Vd. 2.22, where Av. barəzišta- is paralleled by Pahl.
bālēn (Av. barəzištaēibiiō gairibiiō : Pahl. pad bālēn abar garān).
32
See Josephson 1997: 70.
33
See Josephson 1997: 82-83.
34
See Josephson 1997: 101.
 Kurdish bažn, Persian bašn and Other Iranian Cognates 335

One could assume at first sight that the translators of the Avesta did not have
a word bašn in their lexical repertoires, or if they had, this had a meaning differ-
ent from that required by the relevant contexts.
It was Antonino Pagliaro (1925: 566) who first recognized the equivalent of
the Avestan expression baršnušuua gairinąm etc. in the Pahlavi sequence pad
garān bašn (transcribed bušn by Pagliaro35), occurring once in the Ayādgār ī
Zarērān (AZ):
ud pas wištāsp-šāh ō zarēr ī brād framān dād kū pad garān bašn kōf ī borz ātaxš
framāy kardan (AZ 23)
and then King Wištāsp ordered to his brother Zarēr to light fires on the mountain
peaks (gloss: high mountain).
Pagliaro’s reading was rejected by Benveniste (1932: 258),36 Nyberg (1974: 42
s.v. bag)37 and Monchi-Zadeh (1981: 58).38 It was later supported by MacKenzie
(1984: 158), for whom «Pagliaro’s reading /pad garān bašn/ ‘on the peak(s) of the
mountains’ is certainly to be preferred» and Shaki (1986: 260), according to
whom «Pagliaro correctly understood the text». Nowadays, it represents the
most usual interpretation (see also Oriān 1992: 51 [‘on the tops of the mountains
[bar bālā-ye kuhhā]’], 203 [pad garān bašn]). Taking into consideration the dif-
ferent meanings carried on by modern cognates to Pahl. bašn as far as the land-
scape domain is concerned (see above, § 4), it still remains to understand if,
when referred to a mountain, this word simply refers to the peak or to another
morphological reality in the highest (or central) part of the mountain.
To a mountain configuration also points Pahl. bašn in the Bundahišn (Bd.),
where it occurs as a part of a mountain name in the paragraph dedicated to the
description of mountains:39

35
See also bušn ‘parte superiore, vetta’ in his Glossario (ibid.: 597).
36
According to Benveniste, the sequence should be read pa garān [ī] bagān, and interpreted
not as a general designation, but as the name of a particular mountain (“les monts des Dieux”). He
however agrees with Pagliaro on the fact that kōf ī borz is to be considered as a gloss.
37
Cf. Nyberg’s interpretation: ‘the mountain of the mountain gods’ (garān bagān kōf).
38
Monchi-Zadeh suggests the reading +grīv ‹ī› bagān kōf ī burz “Auf dem Nacken des hohen
Götterberges” (cf. 1981: 32, 42, 57-58).
39
See also Bailey 1989.I: 88 («bašn/bišn is from older baršn-, Av. barəšnu- ‘height’»). Anklesaria
(1956: 98; note that this is a posthumous publication: Anklesaria died in 1944 and, furthermore,
«everything, so far as text and translation were concerned, was ready since 1935» [ibid.: 2]) read
bagān the graphic sequence here given as bašn. In the Bundahišn, a form bašn was also used to
name a (swift) star (α Orionis or Betelgeuse) and the sixth lunar mansion; for Pers. bašn ‘id.’ see
336 Ela Filippone

rāwag bašn pad zarābad, ēn gyāg ast kē ēdōn gōwēnd zrawat, ast kē bašn, ast kē
kalād xwānēnd (Indian Bd. 10; Behzādi 1989: 23)
Rāwag-Bašn (Rāwag Peak) is in Zarābād. There are some who call this place so,
Zarābād, there are some who name it Bašn (‘Peak’) and there are who name it
Kalād (‘Castle’).
Rāwag-bašn pad Zarāwad ēn gyāg ast kē Zarāwad ast kē Rāwag-bašn ud ast kē
kalād xwānēd (Iranian Bd. 9.38;40 Pākzād 2005: 137)
Rāwag-Bašn (Rāwag Peak) is in Zarābād. There is who names this place Zarābād,
there is who names it Rāwag-Bašn (Rāwag Peak) and there is who names it Kalād
(‘Castle’).

Pahl. bašn is however a polysemic word and may also occur in contexts dif-
ferent from those we have seen above. Consider the following passage from the
Ayādgār ī Zarērān:
murw-iz nišēm nē windād bē ka ō aspān bašn nēzag-iz tēx ayāb ō kōf ī sar borz
nišīnēnd (AZ 31).
Pagliaro (1925: 568) translates: ‘gli uccelli non trovano riposo se non
posandosi sulle teste dei cavalli e sulle punte delle lance o sulle vette dei monti’,
making explicit his stance on the point as follows (ibid.: fn. 4) : «Leggendo bušn
da barəšnav- si guadagna il senso ‘parte superiore testa’ (N. 55: barəšnvō vā paiti
vaγδanahe ‘sulla parte superiore del capo’) assai plausibile. Questa lettura è più
ovvia che quella, pure possibile, di b u d a a n bušān ‘criniere’, paleograficamente
più distante».41
Monchi-Zadeh’s understanding of this passage (1981: 43, 60) differs for both
reading and syntactic analysis. He translates: ‘Die Vögel fanden keine (andere)
Sitzstelle, ausser auf den Spizten der Lanzen, <die> auf dem Schopf der Pferde
<ruhten>, oder auf dem des Har-burz-Berges’.
MacKenzie (1984:159) disapproves Monchi-Zadeh’s interpretation; in his
opinion, it would be better to interpret ō aspān bašn nēčakān tēx as ‘on the ma-
nes of the horses < and > the heads [lit. blade] of the lances’. Notwithstanding his
criticism, MacKenzie agrees with Monchi-Zadeh on the meaning which is to be
assigned to bašn in this passage, i.e. ‘mane’. This would be a mistake for Shaki

references in Dehxodā. I do not know whether we are facing here with a case of polysemy or ho-
monymy.
40
Cf. also Bailey 1989.II: 41 [= A80 (15)-A81 (1)]. Bailey (1989.I: 89-90) reads kalāk instead of
kalād.
41
See also below.
 Kurdish bažn, Persian bašn and Other Iranian Cognates 337

(1986: 261), since, according to him, «bašn ([...] Av. barəšnu-) is ‘head, peak, stat-
ure’ not ‘Schopf’ or ‘mane’ as proposed inadvertently by MacKenzie, for which
MP has buš (Av. barəša- ‘neck’)». In fact, Shaki had already exposed his position
on the meaning of Pahl. bašn in a previous paper (1975: 68), commenting on a
passage from another Pahlavi text, the Draxt ī asūrīg (DA), where a goat, in its
verbal context with a palm tree, addresses the latter in these terms:
burz hē dēw ī buland bašn-it mānēd (DA 22/33)

For this passage, different readings and interpretations have been advanced,
the differences being concerned with (1) syntactic analysis (in particular, the
identification of the head of the ezafe-construction ī buland), (2) meaning of
bašn, (3) interpretation of the sequence :
Navvābi 1967: 55 (in Persian) You are high, tall demon; your crown [Pers.
bašn (kākol)] resembles the hair of a demon
[dēvgēs]
Shaki 1975: 68 (in English High art thou, O tall demon, Your stature
and Persian) [Engl. stature, Pers. andām] is like unto the
devil [dēw]
Oriān 1992: 147, 329 (in Persian) You are tall, o Demon, your high stature [Pers.
bašn] recalls the Demon of the Demons
[dēwān-dēw]
Čunakova 2001: 89, 157-158 (in Russian) You are as high as a very high demon, your
crown [Russ. krona] recalls the demons
[ŠDY’ŠDY’Š]

In principle, both ‘stature’ and ‘crown’ may fit a context where the discourse
is about a palm tree, which is essentially a pillar, with a tuft of leaves on top. Ac-
cording to Shaki, «Navvabi, apparently following some other scholars, unwar-
rantedly equates MP bašn ‘head, stature’ with NP, MP buš ‘mane’. But NP ‫ﺑﺶ‬
Arabicized ‫ﻓﺶ‬, and MP buš are derived from Av. barəša- ‘neck, back of animals’,
whereas NP and MP bašn come from Av. barəšnu- ‘head, peak’, although both
forms etymologically derive from Av. barəz- ‘to rise’».
Though one may be critical with Shaki’s assumption of a linear, direct deriva-
tion of the mentioned Middle and New Persian words from their Avestan cog-
nates, one has to agree with him on the fact that two distinct lexical sets existed
and still exist in Iranian, which we may summarize as follows: (1) words for ‘stat-
ure, figure, body, etc.’ (human body domain) and/or ‘top, peak of mountain, etc.’
(landscape domain), among which Av. barəšnu-, MPers. bašn, etc. may be in-
338 Ela Filippone

cluded; (2) words for ‘mane; ringlet, tuft, fringe, etc.’ (human/animal body do-
main), ‘crown of tree (in particular the tuft of leaves on top of palm trees); awn,
beard (on the spikelet of grasses such as wheat, barley, etc.’ (plant domain),
among which Av. barəša-, MPers. buš, etc. may be included.42 These two sets may
be or not independent in origin, but we cannot exclude in any case some kind of
crossing and/or contamination, which could explain, among other inconsisten-
cies I am not mentioning here, the usage of MPers. bašn as ‘mane’ and ‘foliage of
palm tree’ in AZ and DA, as well as the (Buddhist) Sogdian form βnš ‘mane’ (bet-
ter explained by MacKenzie’s derivation [< *brš-na-, 1970: 556] than by the ad
hoc phonetic development [«In the place of OIr. r̥, we have an in S. βnš»], formu-
lated by Henning apud Gershevitch 1953: 52 (§ 345).
Shaki’s arguments cannot be considered as conclusive also in consideration
of the textual evidence. If we may accept that in the passage quoted above (DA
22/33) bašn could be (but not necessarily should be!) interpreted as ‘stature’, one
can hardly contest the sense ‘crown/top of tree’ for another occurrence of this
word in the same text, where the palm is praising his own qualities:
bašn-um ēst zargōn yad ō rōz yāwēd (DA 25)
my top is green forever,
especially if compared with the following passages, drawn from the goat’s
scornful answer:
draxt-ē hušk-iz dārū draxt <-ē> sar-aš būd zargōn (DA 36)
even the tree with dry wood its top became green
tō až ēd kirdagān sar-it ēst zargōn (DA 37)
as for you, for these actions your top is green.

The comparison shows that sar and bašn are used here to denote the same
referent: the crown of the palm tree. Moreover, the meaning ‘head (of horse)’,
which Shaki (1986: 261),43 and before him, Pagliaro (1925: 568) and Benveniste

42
I won’t dwell long on this point. I would like just to quote some of the relevant items belong-
ing to the plant lexical domain, which are generally less quoted in the literature. For ‘date-palm fo-
liage’ see Choresmian (‘)fš (MacKenzie 1970: 556) and probably also Xuri barašk (Šāyegān 2006); for
‘awn, beard (of cereals), etc.’ see Xorāsāni (Qāini) boš (Šālči 1991), Kurd. (Kurmanji) bijî (Chyet
2003), (Sulemaniye) bij, pij (Wahby/Edmonds 1966), (Southern) biž (Hažār 1990), Kermāni boš (also
‘lops of wild plants’) (Purhoseyni 1991), Rāvari boš (also ‘branches and foliage of trees’)’ (Karbāsi
Rāvari 1987), Bardesir boš (Barumand Sacid 1991), Jirofti–Kahnuǰi boš (Dehqāni 1998), etc.
43
‘Not even the birds can find a seat, except they perch upon the heads of the horses, the
points of the lances and/or the peaks of the mountains’ (ibid.).
 Kurdish bažn, Persian bašn and Other Iranian Cognates 339

(1932 : 262)44 assigned to bašn in AZ 31 cannot be accepted: the same Avestan

phrase quoted by Pagliaro («barəšnvō vā paiti vaγδanahe ‘sulla parte superiore
del capo’», ibid.: fn. 4) to support his interpretation proves in fact that, if any-
thing, bašn should be ‘top’ not ‘head’.
The only Pahlavi text that records bašn as ‘figure, stature, external appear-
ance, etc.’, i.e. with the senses we have seen above for Kurdish and other Modern
Iranian languages, is the Wizīdagīhā ī Zādspram (WZ). At the end of time, we are
told in the paragraphs devoted to eschatology, humans will regain their bodies
and be resurrected. In particular, when Sōšan, with the assistance of Ērman, will
arrive on the earth to bring about Renovation,
panǰ ek +widordān az zamīg ul uzēnd kālbodōmand ud gyānīg pad bašn <ud> čih-
rag čiyōn ka be widord hēnd (WZ 35.23, Gignoux/Tafazzoli 1993: 132)
un cinquième des morts se léveront de la terre, pourvus d’un corps et d’une âme
vitale, avec stature et visage comme lorsqu’ils sont morts (ibid. : 133).

The copulative compound bašn <ud> čihrag, recalling similar compounds in

Modern Iranian, and probably referring to the physical aspect as a whole, in-
cluding physical constitution and features,45 has another occurrence in WZ,
when the discourse is about the final phases of Renovation. Humans will obtain
immortal bodies, made of a new material substance; at that time,
pad bašn <ud> čihrag homānāg ō čehel-sālān (WZ 35.51, Gignoux/Tafazzoli 1993:
138)
pour la taille et il visage ils ressembleront aux (gens) de quarante ans (ibid. : 139).

Much more complicated is a passage of the Šahrestānīhā ī Ērān (ŠĒ), possibly

containing bašn,46 for which different interpretations have been advanced. In the
last edition of this text (Daryaee 2002), one finds:
pas gizistag frāsiyāk ī tūr har ēk nišēmag ī dēwān [ud] uzdēstzār <ī> [ud] *bašn
padiš kard (ŠĒ 7, Daryaee 2002: 13)
then the accursed Frāsiyāk, the Tūrānian, made seats for each of the demons, and
an idol temple and a heathen temple (ibid.: 17; Engl. transl.) – then the cursed

44
‘sur la tête des chevaux’. See also Oriān 1992 : 52 (‘on the head of horses’ [bar sar-e asbān]’),
204 (ō aspān bašn); Čunakova 2001: 140.
45
In WZ one also finds the adjective ham-bašn ‘de même stature’ (Gignoux/Tafazzoli 1993), in
collocation with other adjectives: 3.72 ham-bašn ham-+dēsag ‘de même stature et de même form’
(ibid.: 52-53), 35.52 ham bašn ī +ham-čihr ‘de même taille et de même visage’ (ibid.: 138-139). See also
ham-bašnīh ‘stature semblable’.
46
For different readings see Daryaee 2002: 34.
340 Ela Filippone

Frāsiyāk, the Tūrānian, in any of them built a seat for demons (and) an idol tem-
ple and a heathen temple [sepas Afrasiāb-e turāni-e maclun dar har yek az ānhā
neštgāh-i barā-ye divān (va) botkade va botxāne sāxt] (ibid.: 22; Persian transla-
tion).47

The proposed reading differs in many points from that of Oriān (1992), which
is the following:48
pas gizistag frāsiyāg ī tūr har ēk nišēmag ī šāhān uzdēs-zār-ē bašn padiš kard
(Oriān 1992: 221)
then the accursed Frāsiyāk, the Turanian, on the top of any seat of the kings built
an idol temple [pas gojaste Afrāsiyāb-e tur (bar) bālā-ye har yek (az) nešiman-
(hā-y)e šāhān botkade-i sāxt] (ibid.: 64).
According to Oriān, bašn has the meaning ‘top’; however, its detachment
from the phrase har ēk nišēmag ī šāhān, to which it would be syntactically linked
in this interpretation, is puzzling. But Daryaee’s syntactic analysis is just as puz-
zling, with har ēk detached from ī dēwān, at least according to his English trans-
lation (but not the Persian one!)
As stated by Daryaee (2002: 34), the reading bašn, and the meaning ‘heathen
temple’ have been proposed by Tafazzoli, who referred it to the first element in
the Manichaean Middle Persian compound bšnbyd ‘master of an idol temple’.
For this word, the standard etymological reference is to Henning (1936: 583; with
more details 1940: 42-43), who connects it to Sogd. βγnpt-, suggesting the fol-
lowing derivation: *baginapati- > *bažinapati- > bašnbeδ.
I am not in a position to advance a convincing argument to improve under-
standing of this passage, though I feel more inclined to think that har ēk nišēmag
‘anyone of the seats’ could be better understood as referring to the seven abodes
(āšyān) in Sogdiana which belong to seven lords (xwadāy), having been both the
abodes and the lords mentioned in the preceding lines. In any case, if bašn is re-
ally the object of the verb (together with uzdēst-zār), as in Daryaee’s (and Tafaz-
zoli’s) proposal, one could also consider an alternative solution to both the in-
terpretation of the word bašn in the ŠĒ, and the etymology of Man.MPers. bšn-
byd. What ‘the accursed Frāsiyāk’ might have been charged with, is the construc-
tion of idol temples and images (of the idols); in this case bašn would mean ‘fig-
ure’ in the sense of ‘image or representation of a body form’, while Man.MPers.

47
Note the differences between Dariaee’s English and Persian translations.
48
Adopted in Titus Project (address: http://titus.uni-frankfurt.de/texte/etcs/iran/miran/mpers
/jamasp/jamas.htm).
 Kurdish bažn, Persian bašn and Other Iranian Cognates 341

bšnbyd would denote the ‘heathen priest’, as the person ‘who takes care of the
idol image’. I have to admit, however, that this hypothesis remains highly spec-
ulative: it is not supported by evidence in textual documentation and does not
find confirmation in any other Iranian language.

6. In Manichean Middle Persian and Parthian texts, we find two derivatives of

bašn, i.e. Man.MPers. bašnāy49 and Man.Parth. bašnān,50 both given as ‘height,
stature’ in Durkin-Meisterernst 2004. We have more occurrences of Parth. baš-
nān than MPers. bašnāy. To the Parthian word we may probably assign different
senses, depending on contexts, most of which somehow related to a physical
body, and not simply referring to its height or size, as the English word stature
would imply. In some passages, therefore, rendering this Parthian word with
English body, figure could be preferable51 to other solutions adopted in text edi-
tions or other publications where English is used as the target language.
It could be profitable to review in the following all its occurrences, or at least
all of which I am aware of:
’wd ’z hym r’štyft cy kyšt pd tw hnd’m ’wd gryw bšn’n (A.R.VI 52b)
and I am the righteousness sown in thy limbs, and (in) the stature of (thy) soul
(Boyce 1954: 148-49)
u hw(’)[n p]d syzdyft ’w (h)wyn bšn’n p’z’h (H VII19b)
agony in overwhelming might (will hold) their statures prone (Henning apud
Boyce 1954: 107) [better: ‘their bodies prone’]
[h]ynz’wyrft bšn’n (A.R.IIIa 12b)
stature of might (Boyce 1954: 130-31) [or better ‘body/figure of might’?]
[’]wt cstwmynyc yzd pt bšn’n hynz’wrystr ’wyšt’d (MMiii a 91-93)
und wenn der Letzte Gott mit vollster Kraft in seinen Gliedern dasteht (An-
dreas/Henning 1934: 852);52 e quando l’Ultimo Dio si erge con forza piena in

49
On MPers. -āy (earlier -ād/δ) see Henning 1958: 103 fn. 3 (see also Filippone 1996: 323f. and fn.
124).
50
According to Henning (apud Boyce 1954: 185 s.v. bšn’n; 1958: 103 fn. 3), -ān should be consid-
ered as a Parth. derivational suffix for dimensional nouns equivalent to MPers. -āy. This fact ex-
cludes once and for all any interpretation of bšn’n as a plural («a singular bšn is not recorded»; see
also Boyce 1954: 107 fn. 2), and in this light one should emend Andreas/Henning’s translations of M
2 II V1 18 [= MMiii a (92)] and M6 Ri 12 [= MMiii e]. See also below.
51
Remarkably, in the Ayādgār ī Zarērān and Draxt ī asūrīg, both considered as Pahlavi texts of
Parthian origin, bašn never refers to a human body (see above).
52
At that time, Andreas/Henning analysed bšn’n as a plural form. See also above, fn. 50.
342 Ela Filippone

(tutta) la (sua) statura (Provasi 2008: 192) [perhaps better: ‘with full strength in
(his) body’53]
’wš hnd’m bšn’n bw(ynd) drdj(.)[1-2] (M 99 II 6b) (Colditz 2009: 76)
and the limbs of his stature become *pain[ful(?)]’ (Colditz 2009: 79) [better: ‘the
limbs of his body’]
’wn d’lwg wzrg ky bšn’n ’mšt bwd wnwhg ’w mwrg’n ky ’hy’ng wygnd (MMiii e [= M6
Ri 11-15])
O großer Baum, dessen Äste zerschmettert wurden: Verwirrung erfaßte die Vögel,
deren Nest zerstört wurde’ (Andreas/Henning 1934: 20)54 [better: ‘O big tree,
whose crown has been wrecked’]
’wt ’sm’n’n bšn’n frbst (‘)[s](kyy) (A.R.Ia 3b)
‘and the height of the heavens sank down above’ (Boyce 1954: 10-21)55
pd tw bšn’n pdmwxt hym’d (M 71 V 2-3)
we dressed in your stature (Durkin-Meisterernst/Morano 2010: 25 [§ 39 b])56 [per-
haps better: ‘we dressed with your garment’ or ‘we dressed with your body’].

As for the last sentence, the first interpretation here proposed as an alterna-
tive to the published one, rests on the BODY > DRESS conceptual association we
have seen above in connection with Šir. bašn, etc. It may also be supported by
[m]rd cy pd hwyn pnj pdmwcn [pw](’g)’n pdmwxt cyštyd ‘der Mann, der mit den
fünf [rei]nen Gewändern bekleidet ist’ (M 6032 V 23-24; Sundermann 1981: 114-
115), which shows a similar construction. However, the alternative ‘we dressed
with your body’ seems also possible, in view of the content of other passages in
Manichaean texts, such as Man. Sogdian [rty t](γ)w ’yš ptm’wkh ZKwy mn’
[γryw]y-h ’z-nh ‘[and th]ou art the garment of my own body’ (H 36.10 R 9-10; Sun-
dermann 1990: 27).

53
It could be useful to recall that the Last God (the Last Statue [andriás] in the Coptic sources),
into whom the salvageable Light will assemble in the end, is expected to assume the aspect of a
human body.
54
For the wrong plural interpretation, see above fn. 50.
55
Note that, apart from this case (which, incidentally, reminds Av. barəšnuuō auuaήhe ašnō
‘from the summit of heaven’ [Yašt 13.42]), Man.Parth. bašnān and Man.MPers. bašnāy are never
recorded as ‘height’. Other words are used instead (Parth. bwrz, bwrzyyft, bwrzw’r etc., MPers. b’l’y);
cf., for ex., Parth. ’c bwrz (’w) jfr ‘from the heights into the abyss’ (M 99 II R 6a; Colditz 2009: 76, 79);
MPers. b’l’y (ẅ) zwpryh ‘height and depth’ (Šahb. 280; MacKenzie 1979: 514, 515 and 532 note to 280).
56
In fact, the authors leave open other possibilities: «Lit. ‘(he?/you?) dressed us or possibly
‘(we) dressed us (= ourselves)’», ibid.: fn. 24.
 Kurdish bažn, Persian bašn and Other Iranian Cognates 343

Unfortunately, contexts are of no or little help in other cases: see [b](š)n’n

’rg’w fr’gwd (A.R. IIIc 3b; Boyce 1954: 134) ‘covered (?) the noble body (?)’;57 bšn’n
in Gloss.h M752c V i (33), having ms ‘also’ as its Sogdian counterpart (probably,
as suggested by Henning, in the sense ‘this word has the same meaning as the
preceding’), with the side column (and therefore the preceding word) missing
(for the edition and a possible reconstruction see Henning 1940: 42);’wt bš(n)’n
‘and your stature’ (M407 b A5; Durkin-Meisterernst/Morano 2010: 81 [§ 263b]);
pd bšn’n (CPar 141); [ 7 ]ft bšn’n jyryf(t) (x)[rw](s’)[m] (M377 I, 3). The editor of
the latter fragment (Reck 2007: 333) translates «[ ]heit, der Glieder Weisheit
(rufen wir an)», specifying however in a footnote (ibid.: fn. 43) that «bšn’n ist
sehr vieldeutig. Hier sind alle Bedeutungen möglich: „Glieder, Teile; Statue; Sta-
tur; Höhe“». In fact, the most unlikely of the mentioned possibilities is just
‘Glieder’.58 Parth. bašn ‘part of the body, branch’ only rests on a misanalysis of
bšn’n,59 but it does not exist or at least, it is not attested in any survived docu-
ment. Therefore, we are expected to definitely cross it off the list of the Parthian
words and related meanings.
Similarly ambiguous, since deprived of a larger context, is Man.MPers. bašnāy
in [...](’n k)y pd bšn’yy [...] ‘(d)er (?) von Statur’ (KPT 1569; Sundermann 1973: 78).
Things go a little better with
(’)wd hwcyhryh (cy m)[n] (bš)n’y wzrg (M114 II V8) (Morano 2014)
and the beauty of my great body,
and even more with bšny in M842 R 1. It was Henning (1958: 103f.) who first rec-
ognized that two Manichean fragments, one in Middle Persian (M 842) and one
in Parthian (M 215), were two versions of the same Manichaean hymn. Here
MPers. bašnāy parallels Parth. tan ‘body’. This multilingual text has been re-
cently re-edited in Durkin-Meisterernst/Morano 2010: 94-99:
(MPers.) M842 R 1-2 (Parth.) M 215 R 10-11
dryst wys’y bšnywm wzrg [drwd] ’br tw mn tn wzrg
welcome, my great height welcome to you, my great body

57
Very unclear. We cannot rely on a meaning certain for fr’gwd («Poss. pp. of fr’gwnd-?»,
Durkin-Meisterernst 2004). Desmond Durkin-Meisterernst informed me that this text could have
had a duplication. It is likely, though not certain, that M2206 cy hw b(š) was a fragment of this du-
plication. In this case we could reconstruct the text cy hw b(š)n’n ’rg’w fr’gwd (presumably, ‘which
covered the fine/noble body’).
58
What about ‘figure/body of wisdom’ (recalling ‘figure/body of might’ above)?
59
See above fn. 50.
344 Ela Filippone

(Durkin-Meisterernst/Morano 2010: (Durkin-Meisterernst/Morano 2010:

95 [323a]) 95 [323b])
‘Sei mir gegrüsst, meine grosse Statur’
(Henning 1958: 103)
By rendering with a single German word (i.e. Statur ‘stature, figure, phy-
sique’)60 both MPers. bašnāy and Parth. tan, Henning intended to stress the link
between the two versions and to remark at the same time that the two words
belong to the same semantic domain (though they are not quite identical in
meaning). I agree with him. Durkin-Meisterernst/Morano 2010 opted for another
solution.
Beside bašnāy, we also find a Man.MPers. form bašn. An occurrence is in MIK
7285 Vii 6 (bšn hwy(r)[...]), in a seriously damaged and therefore obscure pas-
sage.61 Another one could be in a fragment of a bilingual (Parth./MPers.–Sogd.)
glossary (Gloss.h 31), but nothing more can be said on it, in the absence of its
Sogdian counterpart. Henning, the editor of the fragment (1940: 38f.), expands on
other MPers./Parth. words present in the list and starting with bšn- (bšnwg, bšn’n,
bšnbyd), reconstructing other possible forms (*bšnyh’, *bšnsr), which in fact are
unattested in our documentation, and their Sogdian counterparts (ibid.: 42f.).
For MPers. (?) bšnwg and its Sogd. equivalent mδy’’n, a hapax that could
represent «an older form of myδ’’n “middle”, possibly with a specialized meaning:
“middle” > “means” > “remedy”», Henning suggests the meaning ‘medicine’;
bšnwg/mδy’’n would be in this case synonyms of the pair MPers. bš/Sogd. βycy’h
‘id.’, occurring three lines below. Henning’s suggestion has been accepted by
Durkin-Meisterernst (2004, s.v. bšnwg) and Sims-Williams/Durkin-Meisterernst
(2012, s.v. mδy’n). An alternative solution (a very tentative proposal, in fact, un-
supported by textual evidence) could be taking into consideration the attested
meaning of myδ’’n, i.e. ‘middle, waist’, and considering bšnwg as a derivative of
bašn, with the semantic extension of ‘waist’, already seen above for Kurmanji
Kurd. bažn.

7. Many scholars agree on the fact that Av. barəšnu-, MPers. bašn and cognates
originate from OIr. *barz- ‘to make high’, IE *bherǵh ‘to be high’. The single devel-

60
Note that the semantic range of Germ. Stature does not map with that of Engl. stature (and
It. statura).
61
The whole fragment is very broken. It is worth noting, however, that the preceding line of the
fragment (5) contains the sequence bwrzyy and the following (7) the sequence (tn)w’r ‘body’.
 Kurdish bažn, Persian bašn and Other Iranian Cognates 345

opments, however, may be a matter of discussion. Henning (1958: 103 fn. 3), for
example, observes that MPers. bašnāy is not a direct continuation of Av. barəš-
nu-, «eher bašn- aus baršn- (vgl. apers. baršnā) + -āδ».62 I am not going into de-
tails on the matter.
Different etymological connections have also been advanced: Cabolov (2001:
138) derives Pers. bašn (> Kurd. bažn) from *var- ‘to grow’ (?), *var-s-ana- (?),
connecting it to Skt. varṣmán- ‘height, top’; Gercenberg (quoted from Hasandust
2004: 204) derives Pers. bašn from *vaxšana- (< vaxš- ‘to grow up’).
From a cognitive perspective, an original connection with the notion of
HEIGHT accounts well for the semantic peculiarities of this lexical set. Some bašn-
forms belong to the body part domain, with possible secondary extensions to the
clothing domain. The range of meanings goes from ‘stature’ to ‘body’,63 from
‘physique, constitution’ to ‘aspect, image’, and ‘features’64 (and further, peculiar
meanings in specific areas, as ‘waist’ in Kurmanji Kurdish). Some bašn-forms
belong to the landscape domain. This fact may rest on semantic extensions
based on metaphorical associations which involve body parts as the source, and
elements of the landscape as the target (or vice versa), a well-known lexical
strategy commonly found in most languages. However, it could also be due to
independent developments from the same etymon, and this could explain why
in some areas bašn-foms are never used with reference to the human body.
Nowadays, bašn-forms, frequently occurring in hendiadys,65 are mostly found
in some dialects of the Central area, the Kurdish, Gorāni and Zazaki areas, and
only sporadically in Fārs (Širāz, Zarqān).66 Traces are also found in North Iran,
here with the marked meaning of ‘mountain pass’ or the like. The Old and in

62
See also Chyet 2003: s.v. bejn; Hasandust 2004: s.v. bašn.
63
For another case of STATURE, HEIGHT > BODY see Balochi bālād (Sayad Hashmi 2000) ‘body’
(and cognates with the same meaning).
64
The notion of BODY is particularly complex. Different words referring to the body may convey
only a portion of this many-sided notion. Bašn-forms are generally connected to the exterior ap-
pearance of human beings.
65
In passing, I would suggest that -bār we find in bašn(o)bār (see above § 4) is to be connected
to the element -bār/-wār we find in Man. Parth. tanbār ‘body’, Man. MPers. tanwār ‘body, trunk (of
tree)’ (Durkin-Meisterernst 2004), Pahl. tan-bahr ‘physique’ (MacKenzie 1971), Sogd. tambār, tam-
mār, etc. ‘body’. The etymological suggestion by Skalmowski 1991: 330 («une continuation d’un
composé nominal *tanū̆ -pāra- lit. « ce qui remplit le moule/la forme » (cf. parth. ’mb’r-, « remplir »,
’mb’r « amas » ») is in my opinion untenable.
66
In the e-mail mentioned above (see fn. 23), cA. Salāmi has also confirmed that bašn is not in
use in Fārs.
346 Ela Filippone

particular Middle Iranian texts also attest that this lexical set had a strong dialec-
tal connotation, though the dialectological stratification of the documentation
hinders a more in-depth inquiry. Diffusion and divergences in usage and mean-
ings of bašn-forms in modern languages could also help us understand a little
how the situation was in the past.

BIBLIOGRAPHY
Adib Tusi, M. A. (1963–1964), “Farhang-e loγāt-e bāzyāfte (mostadarak)”, Našrie-ye dāneškade-ye
adabiāt-e Tabriz, 15/3, 15/4 (1342/1963); 16/1, 16/2, 16/3 (1343/1964).
Andreas, F. C. / W. B. Henning (1934), “Mitteliranische Manichaica aus Chinesisch-Turkestan, III”,
SPAW: 846-912.
Anklesaria, B. T. (1956), Zand-Ākāsīh. Iranian or Greater Bundahišn, Bombay.
Asatrian, G. (2011), Hūmbābā yā farhang-e guyešhā-ye markazi-ye Irān, Tehrān 1390/2011.
―― / V. Livshits (1994), “Origine su Systéme Consonantique de la Langue Kurde”, Acta Kurdica 1:
81-108.
Ašrafi Xānsāri, M. (2004), Guyeš-e xānsāri, Tehrān 1383/2004.
Bābān, Š. (1982), Farhang-e fārsi-kordi, Tehrān 1361/1982.
Bailey, H. W. (1989), Bundahishn (Linguistic commentary supplemented by transliteration, transla-
tion and word-index). I-II. Unpublished manuscript [revised version of the unpublished D.
Phil. thesis, Oxford, 1933].
Barumand Sa‘id, J. (1991), Vāženāme-ye guyeš-e Bardesir, Tehrān 1370/1991.
Bartholomae, Chr. (1904), Altiranisches Wörterbuch, Strassburg 1904.
Behzādi, R. (1989), Bondaheš-e hendi. Matn-i be zabān-e pārsi-e miāne, Tehrān 1368/1989.
Behruzi, cA. N. (1969), Vāžehā va masalhā-ye širāzi va kāzeruni, Tehrān 1348/1969.
Benveniste, E. (1932), “Le Mémorial de Zarēr”, Journal Asiatique 220: 245-293.
Blau, J. (1965), Dictionnaire kurde-français-anglais, Paris.
―― (1975), Le kurde de cAmādiya et de Djabal Sindjār : analyse linguistique, textes folkloriques, glos-
saries, Paris.
Borāzǰāni, T. (2003), Sir-i dar guyeš-e daštestāni, Širāz 1382/2003.
Boyce, M. (1954), The Manichaean Hymn Cycles in Parthian, Oxford.
Safizāde, S. (Burakai) (2001), Farhang-e Burakai-e kordi-fārsi. 2 voll., Tehrān 1380/2001.
Cabolov, R. L. (2001), Ėtimologičeskij slovar' kurdskogo jazyka, Moskva.
Cantera, A. (2004), Studien zur Pahlavi-Übersetzung des Avesta, Wiesbaden.
Chyet, M. L. (2003), Kurdish-English dictionary / Ferhenga Kurmancî-Inglîzî, Yale.
Colditz, I. (2009)‚ “The Parthian “Sermon on happiness” (Hunsandīft wifrās)”, in W. Sundermann, A.
Hinzte, F. de Blois (eds.), Exegisti monumenta. Festschrift in Honour of Nicholas Sims-Wil-
liams, Wiesbaden: 59-93.
Čunakova, O. M. (2001), Pexlevijskaja božestvennaja komedja. Kniga o pravednom Viraze (Arda Viraz
namag) i drugie tekstj, Moskva.
 Kurdish bažn, Persian bašn and Other Iranian Cognates 347

Daryaee, T. (2002) Šahrestānīhā ī Ērānšahr. A Middle Persian Text on Late Antique Geography, Epic,
and History, Costa Mesa.
Dehghan, I. (1970), “The Persian dialect of Āvarzamān”, Journal of Near Eastern Studies 29: 259–272.
Dehqāni, E. N. (1998), Barresi-ye guyeš-e Jiroft va Kahnuǰ, Kermān 1377/1998.
Dehxodā, A. A. (1946-1981), Loγātnāme, Tehrān 1325/1946–1360/1981.
Desmaisons, J. J. P. (1908), Dictionnaire persan-français (Publié par ses neveux), Rome.
Dorudiān, V. (1986), “Guyeš-e Naqusān”, Farhang-e Irānzamin 26 (1365): 78-108.
Durkin-Meisterernst, D. (2004), Dictionary of Manichaean Texts.Vol. III, Part 1: Dictionary of Mani-
chaean Middle Persian and Parthian, Turnhout.
―― / E. Morano, E. (2010), Mani’s Psalms. Middle Persian, Parthian and Sogdian texts in the Turfan
Collection. Brepols.
Eqtedāri, A. (1955), Farhang-e lārestāni, Tehrān 1334/1955.
Esfandiyāri, A. (1999), Zarb-ol-masalhā-ye borujerdi, Esfashān 1378/1999.
Filippone, E. (1996), Spatial models and locative expressions in Baluchi, Naples.
Firuzbaxš, F. (1997), Barresi-ye sāxtemān-e dasturi-ye guyeš-e behdinān-e šahr-e Yazd, Tehrān 1376/
1997.
Garusin, H. (1991), Vāženāme-ye hamadāni, Hamadān 1370/1991.
Gershevitch, I. (1953), A Grammar of Manichean Sogdian, Oxford.
Gignoux, Ph. / A. Tafazzoli (1993), Anthologie de Zādspram. Édition critique du texte pehlevi traduit
et commenté, Paris.
Hadank, K. (1932), Mundarten der Zâzâ, hauptsächlich aus Siwerek and Kor. (Kurdisch-persische
Forschungen: Ergebnisse einer von 1901 bis 1903 und 1906 bis 1907 in Persien und der asia-
tische Türkei ausgeführten Forschungsreise von Oskar Mann. Abt. 3 (Nordwestiranisch),
4), Berlin.
Haim, S. (1992), Farhang-e bozorg-e fārsi-englisi, Tehrān 1371/1992.
Hakim, H. / G. Gautier (1993), Dictionnaire français-kurde, Paris.
Hamidi, J. (2001), Farhangnāme-ye Bušehr, Tehrān 1380/2001.
Hasandust, M. (2004), Farhang-e rišešenāxti-ye zabān-e fārsi, I, Tehrān 1383/2004.
Hažār (1990), Farhang-e kordi-fārsi, Tehrān 1369/1990.
Henning, W. B. (1936), “Soghdische Miszellen”, Bulletin of the School of Oriental Studies 8: 583-588.
―― (1940), Sogdica, James G Forlong Fund XXI, London.
―― (1958), “Mitteliranisch”, in B. Spuler (ed.), Handbuch der Orientalistik. I, IV, 1: Linguistik, Lei-
den-Köln: 20-130.
Jaba, A. / F. Justi (1879) Dictionnaire kurde-français, publié [...] par [...] F. Justi, St.-Pétersbourg.
Josephson, J. (1997), The Pahlavi Translation Technique as Illustrated by Hōm Yašt, Uppsala.
Jusupova, Z. A. (2000), Kurdskij dialekt avramani po materialam “Divana” Saidi, Moskva.
Kalbāsi, I. (1983), Guyeš-e kordi-e Mahābād, Tehrān 1362/1983.
Karbāsi Rāvari, cA. (1987), Farhang-e mardom-e Rāvar, Tehrān 1366/1987.
Kešāvarz, K. (1993), Farhang-e zartoštiān-e ostān-e Yazd. Baxš 2-3, Eskilstuna.
348 Ela Filippone

Kurdoev, K. K. (1960), Kurdsko-russkij slovar′. Moskva 1960.

―― / Z. A. Jusupova (1983), Kurdsko-russkij slovar' (Sorani). Moskva.
Lamce, M. (1970) Farhang-e ‘amiyāne-ye ‘ašāyer-e Boyer-Ahmadi va Kuhgiluye, Tehrān 1349/1970.
Lazard, G. (1990), Dictionnaire persan-français, Leiden/Téhéran.
Lecoq, P. (1979), Le dialect de Sivand, Wiesbaden.
―― (2002), Recherches sur les dialectes kermaniens (Iran central). Grammaire, textes, traductions
et glossaires, Lovanii.
Lescot, R. (1940), Textes kurdes. Premiére partie: Contes, proverbes et énigmes, Paris.
―― (1942), Textes kurdes. Deuxiéme partie: Mamé Alan, Beyrouth.
Lirāvi, A. (2001), Guyeš va adabiāt-e farhang-e mardom-e Lirāvi va Daylam, Tehrān 1380/2001.
Malekzāde, M. J. (2001), Farhang-e Zarqān, Tehrān 1380/2001. (2nd ed.).
Mohseni, M. M. (2002), Guyeš-e mardom-e Sirǰān, Kermān 1381/2002.
MacKenzie, D. N. (1961), “The origins of Kurdish”, Transactions of the Philological Society: 68-86.
―― (1970), “The Khwarezmian Glossary-I”, Bulletin of the School of Oriental and African Studies
33/3: 540-559.
―― (1971), A concise Pahlavi dictionary, London.
―― (1979), “Mani’s Šābuhragān-I”, Bulletin of the School of Oriental and African Studies 42/3: 500-
534.
―― (1984), rec. a Monchi-Zadeh, Die Geschichte Zarēr’s. Uppsala 1981, IIJ 27 (1984), 155-63.
Mazdāpur, K. (1995), Vāženāme-ye guyeš-e Behdinān-e šahr-e Yazd. Fārsi be guyeš bā mesāl. Jeld-e
avval (ā–p), Tehrān 1374/1995.
Mocin, M. (1992), Farhang-e fārsi (motavasset), 6 voll., Tehrān 1371/1992 (8th ed.).
Mokri, M. (1966), La légende de Bīžan-u Manīǰa, version populaire du Sud du Kurdistan en langue
gouranie, Paris.
―― (2003), Grammaire et lexique comparés des dialectes kurdes. Éléments de linguistique ira-
nienne, Paris.
―― (2005) “Esthétique et lexique du corps humain dans la littérature classique iranienne (Deu-
xième partie)”, Journal Asiatique 293.1 : 245-356.
Monchi-Zadeh, D. (1981), Die Geschichte Zarēr’s, Uppsala.
Morano, E. (2004), “A Manichaean Antiphonal Hymn of the Body and Soul in Middle Persian, Par-
thian and Sogdian”, in Sunderman Memorial Volume, Wiesbaden [in print].
Naǰafi, A. (1999), Farhang-e fārsi-e cāmiāne. 2 voll., Tehrān 1378/1999.
Navvābi, M. (1967), Manzume-ye draxt-e āsurīg. Matn-e pahlavi, āvānevešt, tarjome-ye fārsi, fehrest-e
vāžehā va yāddāšthā, Tehrān 1346/1967.
Nurbaxš, H. (2003), Farhang-e dariāi-ye xaliǰ-e Fārs (sayd, dariā, kešti), Tehrān 1381/2003.
Nyberg, H. S. (1974), A manual of Pahlavi. Part II: Glossary, Wiesbaden.
Orbeli, I. (2002), Kurdsko-russkij slovar', II, Erevan.
Oriān, S. (1992), Matun-e pahlavi (tarjome, āvānevešt). Gerdāvarande : Jāmasbji. Dastur Manucehrji
Jāmasb–Āsāna, Tehrān 1371/1992.
 Kurdish bažn, Persian bašn and Other Iranian Cognates 349

Pagliaro, A. (1925), “Il testo pahlavico Ayātkār-i-Zarērān edito in trascrizione, con introduzione,
note e glossario”, Rendiconti della R. Accademia nazionale dei Lincei, Classe di scienze mora-
li, storiche e filologiche, VI ser., I: 550-604.
Pākzād, F.A. (2005), Bundahišn. Jeld-e avval: Matn-e enteqādi, Tehrān 1384/2005.
Pāyande, M. (Langarudi) (1987), Farhang-e Gil-o-Deylam. Fārsi be gilaki, Tehrān 1366/1987.
Provasi, E. (2008), “Testi medio-iranici III”, in Gnoli, Gh. (ed.), Il Manicheismo. Vol. III. Il mito e la
dottrina. Testi manichei dell’Asia Centrale e della Cina. Milano: 109-225, (Commento) 374-
478.
Purhoseyni, A. (1991), Farhang-e loγāt va estelāhāt-e mardom-e Kermān, Tehrān 1370/1991.
Reck, C. (2007), ‘Tage der Barmherzigkeit. Nachträge zu den mitteliranischen manichäischen Mon-
tags- und Bemahymnen’ in M. Macuch, M. Maggi, W. Sundermann (eds.), Iranian Langu-
ages and Texts from Iran and Turan. Ronald E. Emmerick Memorial Volume, Wiesbaden:
317-342.
Rizgar, B. (1993), Kurdish-English, English-Kurdish (Kurmançî) dictionary / Ferheng Kurdî-Îngîlîzî,
Îngîlîzî-Kurdî, London.
Rubinčik, Ju. A. (1970), Farhang-e fārsi be rusi – Persidsko-russkij slovar’, 2 voll., Moskva.
Safari, H. (1994), Vāženāme-ye rāǰi. Guyeš-e šahrestān-e Deliǰān, Tehrān 1373/1994.
Salāmi, cA. (2002), Farhang-e guyeš-e davāni, Tehrān 1381/2002.
Sālehi, S. (1990), Il-e bozorg-e baxtiāri (farhang-e vāžegān-e baxtiāri), Tehrān 1369/1990.
Sarlak, R. (2202), Vāženāme-ye guyeš-e baxtiāri-e Čahār Lang, Tehrān 1381/2002.
Sarrāfi, M. (1996), Farhang-e guyeš-e kermāni, Tehrān 1375/1996.
Saryazdi, M. (2001), Nāme-ye Sirǰān, Tehrān 1380/2001.
Sayad Hashmi, (2000), Sayad Ganǰ, Karāčī.
Shaki, M. (1975), “Observations on the Draxt ī asūrīg”, Archív orientální 43: 64-75.
―― (1986), “Observations on the Ayādgār ī Zarērān”, Archív orientální 54: 257-271.
Sims-Williams, N. / D. Durkin-Meisternernst (2012), Dictionary of Manichaean Texts, Vol. III, Part 2:
Dictionary of Manichaean Sogdian and Bactrian, Turnhout.
Skalmowski, W. (1991), “Le moyen-iranien gryw/γr(‘)yw ‘âme’,” in A. van Tongerloo; S. Giversen
(eds.), Manichaica Selecta: Studies Presented to Prof. Julien Ries, Lovanii: 329-32.
Soane, E. B. (1913), Grammar of the Kurmanji or Kurdish Language, London.
Sohrābi Anāraki, A. (1994), Vāženāme-ye anāraki, Mašhad 1373/1994.
Sotude, M. (1963), Farhang-e semnāni, sorxei, lāsgardi, sangsari, šahmerzādi, Tehrān 1342/1963.
―― (1986), Farhang-e nāini, Tehrān 1375/1986.
Steingass, F. (1963), A comprehensive Persian-English dictionary, London5 [first ed. 1892].
Sundermann, W. (1973), Mittelpersiche und partische kosmogonische und Parabeltexte der Manichä-
er, Berlin.
―― (1981), Mitteliranische manichäische Texte kirchengeschichtlichen Inhalts, Berlin.
―― (1990), The Manichaean Hymn cycles Huyadagmān and Angad Rōšnān in Parthian and Sog-
dian, London.
350 Ela Filippone

Šālči, A. (1991), Farhang-e guyeš-e Xorāsān-e bozorg, Tehrān 1370/1991.

Šāmlu, A. (1999) Ketāb-e kuče II. Harf-e Be, daftar-e dovvom, Tehrān 1377/1999.
Šāyegān, M. (2006): Vāženāme-ye guyeš-e xuri, Tehrān 1385/2006.
Tafazzoli, A. (1991), “Some Isfahani words”, in R. Emmerick; D. Weber (eds.), Corolla iranica. Papers
on honour of Prof. Dr. David Neil MacKenzie on the occasion of his 65th birthday on April 8th,
1991, Frankfurt am Mein: 207-210.
Tasbihi, M. H. (1975), Guyeš-e Xānsāri, Rāwalpindi 1354/1975.
Todd, T. (1985), A Grammar of Dimilī (also known as Zaza). PhD diss., The University of Michigan,
Ann Arbor (UMI).
Unvala, J. M. (1958), “Contributions to modern Persian dialectology. The Lurī and Dizfūlī dialects.
II”, Indo-Iranica 11/4: 1–16.
Vahman, F. (1974), Farhang-e mardom-e Kermān (gerdāvarande D. L. Lorimer), Tehrān 1353/1974.
Wahby, T. / C. J. Edmonds (1966), A Kurdish-English dictionary, Oxford.
Xadiš, H. (2000), Farhang-e mardom-e Širāz, Širāz 1379/2000.
 Once again on Iranian *kund

Adriano V. Rossi
University of Naples L’Orientale

Abstract
The current paper is the third treatment by the author (previous ones: Rossi 2002, 2006) of a com-
plex set of terms widely attested in the Iranian area, but also present in Armenian, NW Semitic
(and from here Arabic), Indo-Aryan and Dravidian.
Differently from the conclusions of Asatrian/Arakelova paper of 2001, the author identifies three
major lexical families, with the following prototypes: (1) *kōnd-/kŏnd- ‘stump, stub’; (2) *kund-
/gund- ‘globular, spherical; thick, large, full-bodied’; (3) *kōnd-/kŏnd- (a) ‘stem of a tree, stump,
stock’; and secondarily ‘stock of gun, stocks for offenders’; (b) any anatomical articulation con-
ceived as a support (metaphorical projection on human anatomy of a support stick), as ‘kneecap,
elbow, knee’. All of the linguistic families mentioned show interactions between them for all of the
three lexical families, and while core semantics are clearly demonstrable for each of them, periph-
eral (both geographical and semantical) differentiations are widely attested. Areal etymologies en-
compassing Indo-Aryan, Iranian and Dravidian are also hinted.

Keywords
Iranian Philology, Iranian Dialectology, Lexicography, IA and Iranian areal etymologies, Iranian
borrowings in Armenian; *gund, *kund

I am very glad to have had, thanks to the kindness of Garnik Asatrian, the oppor-
tunity to present on the occasion of the International Conference on Zaza Stud-
ies (Yerevan, 28-30 October 2011) this paper, originally planned for the Fifteenth
anniversary of Iran and the Caucasus. During the Conference, and thanks to the
kindness and cooperation of many scholars and young people attending the
event, I had the opportunity to complete my research with the Zaza cognates of
the lexical set I will discuss in the following pages, which were missing in the
Balochi Etymological Dictionary Archive. There is no better occasion than a
Festschrift dedicated to Professor Asatrian to offer the printed version of this pa-
per.
1. When in 2000-2001 I wrote my paper on Middle Iranian gund and cognates
(Rossi 2002), I did not know that Garnik Asatrian and Victoria Arakelova were
 Koninklijke Brill NV, Leiden, 2015
352 Adriano V. Rossi

treating around the same time a strictly connected issue in Iran and the Cauca-
sus (Asatrian/Arakelova 2001). Asatrian and Arakelova had started from the late
classical Armenian hapax kund, attested in Oskipcorik, a clear Iranian borrowing
meaning ‘sorceress’, and investigated on four homonymous kund-forms: (1)
‘brave’ in Khorenatsi (A, 31), as an epithet to Ahura Mazda; (2) ‘stump, stub’,
widely attested in modem dialects; (3) ‘sluggish person, heavy man, fat’, also at-
tested in modern dialects mostly in the form k’unt’i, and (4) ‘bald, hairless’ at-
tested in the Armenian translation of the Bible and later literature, where it is
still continued in a series of denominative verbal forms (kndel ‘to shave one's
head’; kndvel, kndenal, passive of the same form; kndut'iwn ‘baldness’, etc.).*

1.1 I will start by the kund meaning ‘brave’.

In the Ninth Italian Meeting of Afro-Asiatic Linguistics held in Trieste in 1998,
the Italian Semitist Fabrizio A. Pennacchietti (1999) presented a paper on
Qundâqôr, a hapax appearing in the Syriac version of the Alexander Romance, a
Greek-Egyptian work of the III c. CE, traditionally but erroneously attributed to
Callisthenes. The word qundâqôr treated by Pennacchietti appears in a Syriac
passage missing in the Greek versions.1 In it we are told that before entering in
Sogdiana (Sōd in the Syriac version) Alexander entrusted someone in his retinue,
defined qundâqôr (written <qūndāqōr> in the ms.), with exploring the routes
leading to the capital of the kingdom. Thanks to the Neo-Aramaic gloss sardār
‘general’ (a clear borrowing from Persian) appearing in ms B of the Syriac Ro-
mance, we learn that this word should indicate an elevated rank in the army. As
a consequence, Brockelmann translated qundâqôr as dux exercitus, accepting a
suggestion by Fraenkel, who considered it a corrupted form of the obsolete Per-
sian title <knd’gr/knd’wr>, synonym with sipāh-sālār ‘commander of an army’.
There are three different terms in classical Persian2 which may be connected
to Syriac qundâqôr ‘general, commander’, viz.:

*
For the sake of space full references for many authors quoted for minor dialects and lan-
guages are not given in the Bibliography at the end of this article, and the reader is referred to
Rossi 2002 and Rossi 2006. Thanks are due to Bilal Zilan and Mesut Keskin for their kind help in
tracing remnants of the *kund family in Zaza. Prof. Martin Kümmel, Jena, kindly suggests a pos-
sibile parallel in the Uralic word for ‘tree stump’, reconstructed as PUr. *kïnta(-w), cf. Finnish
kanto, even if the vowel doesn’t really fit.
1
The edition of the Syriac text used by Pennacchietti is Budge 1889.
2
Pennacchietti, who quotes New Persian from Desmaisons (1908-1913), makes usage through-
out his paper of the modern Persian vocalisation (so, e.g., kondāgar, kondāvar, gondāvar, sepāh-
 Once again on Iranian *kund 353

(1) kundāgar: (a) ‘wise, learned’; (b) ‘brave, intrepid, hero’;

(2) kundāvar: (a) ‘wise, learned’; (b) ‘warrior, hero’; (c) ‘commander of an army’;
(3) gundāvar: (a) ‘brave, intrepid’; (b) ‘commander of an army, sipāh sālār’.
One may at once realise that while number (2) and (3) refer to ‘commander
of an army’, only (1) kundāgar (a) ‘wise, learned’; (b) ‘brave, intrepid, hero’ is
phonetically close to Syriac qundâqôr.
If one removes the NPrs. derivative morphemes -gar and -āvar, the lexical ba-
ses, which could match and which may be found as autonomous bases in any
Persian dictionary are:
(1) kund: (a) ‘obtuse, blunt’; (b) ‘brave, hero’; (c) ‘wise, philosopher’;
(2) kundā and gundā: (a) ‘wise, learned, philosopher, astrologer’; (b) ‘brave, in-
trepid, hero’;
(3) gund ‘testicle’.
This wide gamut of meanings immediately recalls the question raised by Asa-
trian/Arakelova: is our kund really a polysemantic lexeme, or do we face a group
of homonyms with different origin within Armenian and/or within Iranian? I
will come back shortly on this, but I would like to expound now Pennacchietti’s
ideas about the linguistic pre-history of Syriac qundâqôr.

1.2 To explain the variant gund-āvar, with initial voiced g-, Pennacchietti be-
lieves that because of its military meaning, kundāvar could have been reanalysed
as containing gund, a MPrs. and Parthian word meaning ‘troop, army’ (on which
I will come back shortly), producing the following metanalysis: *gund-āwar ‘he
who conducts the army’.
As to the origin of kundāvar, Pennacchietti proposes to link the word and
Syriac qundâqôr to NPrs. (and Turkish (cf. Doerfer 1967, no. 1533: 520-522))
qundāq, a term with two quite different series of denotata: (1) ‘stock of a cross-
bow; gun-stock, rifle-butt’; (2) ‘bundle of rags or clothes; swaddling-clothes of an
infant; incendiary bundle of combustibles; woman’s kerchief’. The Italian scholar
then suggests that an unattested M/NPrs. form *kundāk3 could be a loan from
Greek κοντάκιον, a diminutive formation linked to κοντός ‘pole, rod, perch’ and

sālār), which has been changed in the present paper (with a different choice if compared with my
previous one quoted at Rossi 2006: 615 fn. 7), according to the transcriptional standards used for
classical Persian, with the exception of literal quotations from modern dictionaries (Mo‘in, Lazard
etc.).
3
Pennacchietti writes *kondāk-, but if such loanword really existed in classical Persian, one
should imagine it as adapted to the classical Persian vocalisation, so *kondāk > *kundāk.
354 Adriano V. Rossi

κόνταξ ‘lance, spear, pike’. A way to reconcile this (apparently) rather diverging
series of meanings is to suppose that Greek κοντάκιον (or *κονδάκιον) meant, be-
side ‘stick, sceptre’, also the wooden stick around which a parchment scroll was
rolled up. The latter meaning would account for Byzantine Greek κονδάκιον ‘vol-
ume, manuscript; ecclesiastical hymn; missal’, Syriac qundâqâ ‘volume’ (but the
original meaning remains in pl. qundâqê ‘poles, pins, axes’), Arm. kondak ‘bulla,
breve, decree’ and Christian Arabic qindāq ‘prayer-book’ (for the relevant
sources cf. Pennacchietti 1999).
1.3 What Pennacchietti proposes is therefore that NPrs. kundāgar, kundāvar and
gundāvar, and Syriac qundâqôr, all derive from an unattested MPrs. compound
*kō/undāk-āwar or *kō/untāk-āwar ‘stick-holder’. The link between a stick/scep-
tre and high positions at the Persian Court is seen by Pennacchietti in the refer-
ence (made by Xenophon in his Cyropaedy and Anabasis and Aeschylus in his
tragedy The Persians) to eunuchs at the Achaemenid court as σκηπτοῦχοι ‘holders
of a sceptre’. In addition to that, the reliefs on the staircases of the Apadana in
Persepolis depict eunuchs in Persian or Median attire holding a stick or a sceptre
in their left hand. Therefore, Classical Greek σκηπτοῦχος ‘holder of a sceptre;
Achaemenid court eunuch’ could be a semantic calque on an unattested Old
Persian antecedent of MPrs. *kō/undāk-āwar ‘sceptre-holder’, on its turn, ac-
cording to Pennacchietti, borrowed from Late Greek κοντάκιον (or Late Greek
*κονδάκιον) ‘stick, sceptre’ and expanded with the MPrs. derivative suffix -āwar
‘holder’.
2. As far as the abstract notions of this lexical series are concerned, Pennacchietti
holds that NPrs. kundā and gundā, and also NPrs. kund in the meaning of ‘wise,
learned’, which he seems to consider as an apocopated form of kundā, should
derive from Phl. kundāg ‘sorcerer, fortune-teller, astrologuer’; to the same series
he also refers Phl. kundāgī ‘magic, astrology’ and MMPrs. qnd’yy ‘magic’.4 On the
derivation of NPrs. kund in the meaning of ‘wise, learned’ from the family of
MPrs. kundāg Pennacchietti and Asatrian/Arakelova seem to have parallel views,
even if the two last mentioned scholars also add a possible connection with the
Av. name of the demon Kunda-5 and propose a semantic path from ‘wise’ to ‘ma-

4
Note that a MPrs. *kund ‘magician’ (as stated in Pennacchietti 1999: 72; Asatrian/Arakelova
2001: 202) is attested nowhere.
5
On Kunda- see Rossi 2006: 619-623 (embodying exchange of opinions on Av. Kunda- held by
the author with Gherardo Gnoli, cf. fn * at p. 613).
 Once again on Iranian *kund 355

gician’ and not vice versa, as can be seen from the semantic graph reproduced
hereunder (Asatrian/Arakelova 2001: 204):

In my paper on MPrs. kundāg ‘fortune-teller’, I proposed a slightly different

interpretation. According to my views, the lexical family of M/NPrs. ‘fortune-
teller, sorcerer’ should be traced back to a prototype *kōnd-/kŏnd-, possibly ap-
pearing somewhere as *kū̆ nd-, reconstructed on the basis of the following docu-
mentation from the modern Iran. languages:
(1a) ‘stem of a tree, stump, stock’; and secondarily (1b) ‘stock of gun, stocks for
offenders’. (1b ← Turkish ← Ir.):6
Prs. kund ‘a butcher’s chopping-block; a shoemaker’s cutting-board; a potter’s
wheel’ (Steingass); kond- ‘fers’ (1b) in dar kond-o-zanjir nehādan ‘mettre aux fers’
(Lazard); konde ‘big piece of wood’ (Mo‘in), ‘lower part of a tree’ (Mo‘in); qondāq,
qondāqe ‘stock of gun’; Prs. konde (“kundeh”) ‘coulter’ (Lambton); Afg. Prs. kunda
‘big stump of wood’; Dem. konde ‘root of a large tree, large log of firewood’; Gazi
kunde ‘Baumstamm, Trumm’, kund ‘the stays at the side of a dam in an irrigation
channel (Gaz)’ (Lambton); Hamadani konde ‘a big piece of wood, large trunks and

6
Doerfer 1967: 520-522, no. 1533, according to whom Turk. -āq/āg ← MPrs. *-āk. NPrs. qundāq
‘swaddling clothes’ is a re-borrowing from Turkish, considered as originally Iranian by Doerfer; if
so, their denomination could derive from the stump-like or staff-like shape that the baby assumes
after being involved in them. Pennacchietti (1999: 73-75) points to an ultimate Greek origin of the
NPrs. word.
356 Adriano V. Rossi

branches’; Soi kondä ‘Brennholz’ (Andreas-Christensen-Barr); Sivandi kond

‘chaîne qui relie plusieurs prisonniers’ (1b) (Lecoq); Raji kondae ‘part of the plow’;
Sarvestani kondæ ‘low part and body of the tree’; Mazandarani konde ‘wood for
chopping foods’; Kurd. kund ‘stump’; Bal. kundā ‘block, stump (1a); butt of a gun’
(1b); Parači kunda ‘stocks for offenders’ (1b) (IIFL ii); Ormuri gōn, gōṇ ‘stick;
wood’; Pšt. kunda ‘stump, beam (1a), stock of gun’ (1b), γuṇḍ ‘beam’ (only Eastern
dial. Aslanov; crossed with the type *gū̆ nd-?); Sangleči kʋndε ‘stump’ (IIFL); Yagh-
nobi kunta id. (← Taj. kunda id. Andreev); Shughni kundā, kunda ‘stump, beam’
(Karamšoev); ouside Iranian cf. possibly Kalasha gun, loc. gonḍuna ‘handle; hilt,
stick, pole’; also gond ‘wood’ (one informant only), for which IIFL quotes CDIAL
3998 gaṇḍa-2 ‘joint of a plant’, gaṇḍi- ‘trunk of a tree from root to branches’,
where ← Drav. DED2 1946 is hinted (but EWA 3: 151 states for gaṇḍa-: (“Nicht ge-
nügend erklärt”); guṇḍik ‘(walking) stick’, diminutive of gun ibid.
(2) any anatomical articulation conceived as a support (metaphorical projection
on human anatomy of a support stick), as ‘kneecap, elbow, knee’:
Qāini konderešk ‘elbow’; Birjandi konderešk, kongereš id.; Anaraki kināronǰ
(Sahrābi Anāraki), kīnāronǰ (Lecoq) id.; Esfahani Prs. kināronǰ id. (pers. com. M.
Schwartz);7 kondey zānu ‘kneecap’, cf. eyneyzānu; Khorasani kund-i rišk ‘elbow’
(Šālči), kunda-yi zānu ‘kneecap, knee’ (Šālči); Khuri kunda-i zānū ’ kneecap’; Ker-
mani konde-ye zānu ‘knee, kneecap’; Mahallati kûndä ‘kneecap’ (Mann); (Judeo-
)Hamadani konde ‘knee’; Semnani kend ‘elbow’; konda-zōna ‘kneecap’; Šahrudi
konde zânu ‘kneecap’; Jugi (Gypsy argot of Tajikistan) kund (kund-i zoni) ‘knee’
(pers. com. M. Schwartz); Southern Baškardi (Garu) kund, kuond, kuend id. (pers.
com. G. Barbera); Bal. kōnḍ, kōnd, gunḍuk ‘knee’, also ‘elbow’ in Panjgur (IIFL) has
kuṇḍōšk (from Nuške) both ‘elbow’ and ‘hip’); Pšt. gonḍa ‘bent knee’; γuṇḍai ‘top
of the hip’ (Aslanov); Pšt. gonḍa (and Bal. gunḍuk) document the only forms of
this group with initial voiced velar, possibly derived from crossing with the type
*gū̆ nd-. Kurmanji Kurdish kōnd ‘beam; hill, height, mound’ (Kurdoev) may belong
here for its vocalisation, while being closer to group (1) above for its meaning.
The diffusion area of this lexical set is much wider, from Hamadan to Khorā-
sān, and Baloči, Pašto and the Pamir languages in the east, and the forms are ra-
ther regular in each sub-area. Both features point to a term lexically steady and
culturally central. The anatomical extension of the term may go back to ancient
Iranian heritage, considering its possible checks in a Greek base *kond-o-,8 e.g. in

7
The existence in Esf. Prs. of both kune pā ‘heel’ and kun-ārenǰ ‘elbow’ (according to M.
Schwartz’ informant) shows that also formations such as kināronǰ do contain a reduced form of
kund with vocalic reduction and loss of occlusion in °nd# at the junction in compounds.
8
Chantraine (1968: 562, s. v. κόνδυλος) remarks: “pas de rapprochement sûr hors du grec”, with
 Once again on Iranian *kund 357

classical Gk. κόνδυλος meaning ‘osseous articulation’, and in OIA kanda- ‘tuber’,
and/or OIA kāṇḍa- ‘section, joint (of a plant)’,9 for botanical joints. If the links to
Greek and/or Indo-Aryan (suggested among others by Belardi (1954: 619, 622,
624-25)) are correct, *kond with anatomical meaning and *kond transmitting a
more general notion of ‘stock’ also as ‘support in human body’ could prove to
have common origin, and Kurmanji Kurdish kond ‘beam’ (recorded by Kurdoev
but not by Chyet) could preserve the original vocalisation.10 Pennacchietti’s hy-
pothesis of a borrowing from Old Persian to Old Greek of an unattested form
*kontāk-āwar ‘stick-holder’, if true, could point to the existence already in Old
Persian of a form *kontā̆ k ‘stick’, ‘staff’, ‘lance’, parallel to Skt. kunta- ‘lance’.11

3. Before proceeding to treat the other kunds quoted by Asatrian/Arakelova, I

would like to introduce two other Iranian words that may have had some
amount of interaction with the prototype of Iran.(?) *kōnd-/kŏnd-, i.e. (1) Phl.
gund ‘spherical/circular object’; (2) MPrs./Parth. gund ‘military district, troop,
army, detachment’. Both of them point to an original initial voiced velar, but
with many cross-influences by initial voiceless forms of different origin, as will
be seen in what follows.
3.1 A meaning of both ‘group of men, host’ and ‘armed force’ has been associated
to Phl. gund at least since Spiegel’s (“Schaar, Schaaren” in Spiegel 1860: 392; cf.
Rossi 2002: 145 fn. 28) and De Lagarde’s (1868: 24) early treatments of the word in
mid-nineteenth century, and then passed into all the subsequent scientific liter-
ature.

reference to KEWA: 152, s.v. Skt. kanda- ‘tuber’; EWA III: 55 s. v. kanda- “Wurzelknolle”) considers
its etymology “nicht geklärt” (see also Frisk 1960: 821).
9
EWA III: 55 (kanda-); 1992: 336-337 (kāṇḍa-)
10
Note that the presence of long -ō- in group (2) above is rather widespread, so it is unneces-
sary to hypothesise for Kurdish the lengthening of an original short vowel, as Asatrian/Arakelova
2001: 202 do, if Kurmanji kōnd ‘beam’ does belong to this group.
11
Cf. CDIAL 3893 *kuṇṭa- ‘post, peg’ (since Prakrit on) and perhaps Turner 1966: 3267 *kuṇḍaka-
(since Pali on) ‘husks’; cf. also the annotation in Maryhofer 2001: 3.102: “wohin gehört °kuṇḍaka- in
Kauṭ. kaṇa-kuṇḍaka- n. ‘Kleie und Spelsen ..., pali kuṇḍaka- m. ‘der rote Staub unter der Schale des
Reises’, mth. kūḍā ‘bran’ u. a. (Tu[Add] 3267)?”. Turner 1966: 3289 considers Skt. (ep.) kunta- ‘spear’
(with a sideform *kōnta-, cf. Pali konta- m. ‘standard’) as poss. ← Greek, quoting KEWA 1.229; EWA
3.105: “Vielleicht LW aus gr. κοντός m. ‘Stange, Ruderstange’, auch ‘Speerschaft, Speer’... Nicht si-
cher; Ablehnung bei Wüst”; CDIAL 3295 Skt. kunda- (lex) ‘a turner’s lathe’ (EWA 3.106-107 “wohl
fremder Herkunft”) could also belong here.
358 Adriano V. Rossi

In Aramaic, besides Mandaic gunda, Judeo-Aramaic presents a gunda' with

the meaning of ‘military host’. In Syriac the form gūddā with the same meaning
is the probable result of a more recent assimilation of -nd- > -dd-;12 the question
of the direction of borrowing (Iranian → Semitic, or Semitic → Iranian) has been
discussed for more than a century13 (see below).
The early interpretation of <gwnd> in the Pahlavi papyri as the name of a
unit in the army (‘Regiment’)14 was certainly influenced by the widespread opin-
ion put forward (since 1907) (references in Rossi 2002: 150 and fn. 45) by Chris-
tensen that Phl. gund should indicate a military unit corresponding to that of a
modern regiment, based on the equation χιλίαρχος = Arm. gunsalar = MPrs.
hazārbed, and is accepted in the last treatments of the matter (Widengren 1976:
282; Shahbazi 1987: 498; Tafazzoli 2000: 4-5). Later Armenian documentation,
where the element gund- is a clear borrowing from Iranian, may point to an
evolution towards a territorial meaning15 (or vice versa, if we admit that our Mid-
dle Iranian documentation may be incomplete). The same gund also occurs in
many Pahlavi texts, particularly in the formulary expressions spāh u gund, spāh
gund, gund u spāh, gund spāh ‘army’ (Tafazzoli 2000: 5). The isolated Mani-
chaean Parthian occurrence of gund in a passage of the lamentation for the
death of Mar Ammo could be considered as evidence of a transfer from the mili-
tary to the poetic Middle Persian lexicon (or vice versa?).
Actually, it is not only the isolated position of gund ‘army’ in Semitic that in-
dicates that Iranian should be considered as the source for a loan (and not vice
versa); it is mainly its interconnection with a vast set of kindred semantic fami-
lies, with a linguistic documentation that extends from the Indian subcontinent
to Asia Minor that definitively points to an Iranian (or Indo-Iranian, or Dravido-
Indo-Iranian) origin of the lexical set to which gund belongs.
3.2 If we pass from the military usages to the overall documentation of the lexical
base of gund, the oldest Iranian occurrence seems to be Av. gunda- ‘lump of
dough’; OPrs. *gunda-, *gunda-ka- ‘fat man' as a personal name or nickname in

12
Sources and discussion, see Rossi 2002: 148-149.
13
Since Nöldeke to Szemerényi and further, see Rossi 2002: 140-144; Ciancaglini 2008: 135.
14
Hansen 1938: 34 ff.; but nothing would be against a territorial interpretation as in Armenian
(see below).
15
Cf. Garsoïan 1989: 565 (6th c. CE); Gyselen 2002: 66 is uncertain between the meaning ‘army’
and territory’ for an isolated MPrs. gund appearing on an administrative seal from the reign of
Khusrow I.
 Once again on Iranian *kund 359

Persepolis Elamite tablets and Late Babylonian documents;16 Γύνδης as a Greek

transcription of a name of a river in Mesopotamia; then in Middle Iranian
Khwar. γwndyk ‘ball’; Sogd. γōndē/γōndāk ‘tarantula’; and, apparently homo-
graphic with MPrs. gund in the military sense, MPrs. gund as translation (or tran-
scription?) of Av. gunda- ‘lump of dough’, and MPrs. gund ‘testicle’.17 A semantic
feature implying ‘roundness, globularity’, therefore, very clearly connects these
Iranian words to each other, and the Iranian documentation as a whole to a very
wide (Indo-)Iranian lexical series.
3.3 The central nucleus of this lexical family (possibly crossed at different times
with the Drav. *kund-/gund- variants of DED2 1695 and the IA forms of CDIAL
3266, 3267, 3268, 3893, for details cf. Rossi 2002: 162-163) is based on the notion of
‘globular, spherical; thick, large, full-bodied’, e.g.:
Prs. gonde ‘huge, large man’, ‘thick’; Pšt. γū̆ ṇḍ ‘round, globular’, gunda ‘fat’; Arm.
(← Iran.) gndakan ‘spherical’, gnd-em ‘to thicken’, gund-ul, gnd-ul, gnd-lik (<
*g(u)nd-ul-ik) ‘short, plump’.
Globular body forms are part of the same semantics:
MPrs. gund ‘testicle(s)’; Prs. gon, gond id., also qond, γond and γonde, Kurd. gun
and gundik id., many central and Western Iranian dialects with the same mean-
ing; Bal. gund, Brahui (← Bal., but cf. DED2 1695 with many Drav. types kuṇḍu-
and guṇḍu- meaning ‘anything globular’) gund ‘testicle(s)’; Pšt. γunḍǝi marai
‘Adam’s apple, the prominent part of the throat’ (Raverty), γuṇḍārai ‘tumor,
hump’ and (probably from these) the names of any objects, insects, or landscape
elements, salient for their globular form:
Av. gunda- ‘ball of dough’; Khwar. γwndyk ‘ball’; Sogd. γōndē/γōndāk ‘tarantula’;
Prs. gonde ‘lump of dough’; γonde ‘ball of cotton or dough’ (Haim), γonde ‘taran-
tula’ (Haim); Northern Taj. γundi ‘hill’; Taj. γǝndal ‘scorpion’; Afg. Prs. γundal
‘venomous spider’; many similar forms in central Iranian dialects;18 Kurd. gundik
‘lump, clod’, kundēk ‘pumpkin’ (Rhea, also Garzoni, k° by confusion with the
*kond- family, or by similarity of the shape, cf. Jaba/Justi koundik ‘poire à poudre
en citrouille’?), kunde ‘water-skin’ (k° from the previous?), gendû ‘bin (of earthen-

16
According to a possible interpretation of the Elamite spelling <ku-un-da> and the Bab. <gu-
un-da-ka-‘>/<gu-un-dak-ka-‘>, El. <ku-un-tuk-ka4>, on which cf. Tavernier 2007: 192 (nos 4.2.672;
4.2.674); Gershevitch (1969: 199) prefers a connection of the El. name with the name of the Av. de-
mon Kunda-; for a possible Sogd. equivalent (<kwnt>, <kwnt’kk>) Livšic’s interpretation oscillates
between the demon and MPrs. kundāg ‘magician’ (Lurje 2010: 218-219 nos. 587-588).
17
Documented not only as an isolated gloss in the Frahang-i Oim but also in the hendiadic
compound (Dk. 9.21.5) kēr ud gund ‘the whole male genital apparatus’ (Skjærvø 2012: 508).
18
For which see Rossi 2002.
360 Adriano V. Rossi

ware)’, qondere ‘small bubble’; Ormuri γuṇḍɔ ‘knot’; Pšt. γuṇḍarai ‘ball’, gundāla
‘roundish lump of coal’ (← Afg. Prs.?); γandal ‘bud, sprout’, γū̆ nḍǝi ‘a mound sepa-
rated from a higher range’; γunḍai ‘little fat woman; tarantula’ (Raverty); γū̆ ṇḍai
‘bag of goats’ hair’, gunḍai ‘ear, spike’ (if from the same family, prob. ← Afg. Prs.);
Shughni γindōl ‘snow-ball’ (lw. EVSG); Arm. (← Iran.) gund ‘sphere, globe’, gndak,
gntak ‘ball, lump’, Mod. Arm. xmorgund ‘lump of dough’, gunj-al ‘ball of dough or
clay’; Georg. (← Iran.) gunda ‘lump’, tovlis gunda ‘snow-ball’, p’uris gunda ‘ball of
dough’.

3.4 Metaphorical transfers based on the circular/spherical form may belong here
or to subsequent 3.5:
Prs. γond, γonde ‘gathering’; Prs. gandu, kandu, kundu ‘beehive’, kandu id. (La-
zard); Taj. γundāštan (<γund d.) ‘to gather’, also γun kardan; Taj. γuntola kardan
‘to put in the pile’; Khuri γon, qon ‘mass, people’; Sivandi γund, γun ‘gathering’;
Gazi gundulī rāz (also gumbulī, different base) ‘beehive’, känd rāz ‘beehive’; Sar-
vestani qond kardan ‘to gather’; Pšt. γuṇḍ- ‘to gather’; Arm. (← Iran.) gndak, gntak
‘swarm’; Georg. (← Iran.) gundi put’k’ris id. (putkari ‘bee’).

3.5 From the bidimensional notion of ‘circularity’ round geometrical shapes de-
rive:
Prs. kondole ‘anything surrounded and brought together’; Afg. Prs. kundalū ‘diag-
onal path on a mountain side’, Zor. Dari gondola, ‘round, spherical’; various cen-
tral Iranian dialects, then: Pšt. γū̆ ṇḍ ‘round, globular’; γuṇḍai ‘circular; top of hip
(anat.)’; γuṇḍa ‘round piece of leaven, large round stone’; γuṇḍārǝy ‘bread roll’;
Wan. Pšt. kaṇḍōlī-kasī ‘tortoise’; Arm. (← Iran.) gund ‘circular; disk of the sun and
the moon’, gnd-em ‘to round’, also ‘wheel’ (Ciakciak), gunj-ul ‘circular, round’;
Aram. Targ. (← Iran. ?) gūndlīth ‘spiral form, writing in spiral form’, and proba-
bly, with a contamination of the semantic families (3.4) ‘pile > gathering’ and
(3.5) ‘round’.

3.6 The meaning of ‘military detachment, troop’: Manichaean Parth. gund ‘army’;
MPrs. pap. gwnd ‘garrison’ (?); Phl. gund ‘army, troop; group, gathering’; Prs. gond
‘military term’ (only as historical term according to Moin); Kab. Prs. gōnd ‘politi-
cal party’; Kurd. gun, gund, gund ‘village’; Pšt. γuṇḍ ‘regiment’, gund ‘party; tribal
subdivision’, gundai ‘group leader’ etc.; Γund village in Afghanistan (already
EVSG 36); J̌ und i Šāpūr toponym in Iran (also written Kund i Šāpūr in the Šāh-
nāme); Ar. (← MIran.) jund ‘troop’; Arm. (← MIran.) gund ‘section of an army
(zaur)’;19 Georg. (← MIran.) gundi ‘body of a thousand men’.

19
But see also fn. 15 above.
 Once again on Iranian *kund 361

The semantic connections between gund in the sense of ‘military detach-

ment’ and gund in the sense of ‘pile’ have been treated by W. Eilers (1971: 603),
who points out that in German Schar in the military sense etymologically derives
from the meaning of Haufen as ‘pile’.
The development from ‘garrison’ to ‘town’ and ‘circular settlement’ to ‘town’
was also treated by Eilers (1956: 213-14), who remarked that the prototypical no-
tion of ‘town’ often implies a ‘plain in which the army makes camp’ (see, e.g., Prs.
urdūgāh, maidān, kū̆ ran etc.). Eilers quotes many examples from different Near-
eastern cultures having semantic traits of ‘circularity’ in their prototypical no-
tions of ‘town’, among which Heb. pelek, Aram. pilkā and Akkadian pilkum from
the word for ‘spindle’, Arab. diyār and madār, Akkadian dūrum from the root dwr
‘to be round’, Lat. urbs < orbis, Prs. -gerd in many toponyms < *-kṛta-, etc.

4.1 Therefore, while I agree with the main assumption of Asatrian/Arakelova,

implicit in the semantic graph reproduced above, according to which the proto-
typical semantics of (Indo-)Iran. *kōnd-/kŏnd- was ‘stump, stub’, I would sepa-
rate from all the other developments their subgroup ‘hill, mound’ (top-right in
the graph), and their subgroup ‘sluggish, fat’ (right low in the graph) for both of
which I would refer to the words collected at § 3.3 above, i.e. to the (Iran.?) pro-
totype *gund, not *kund.20

4.2 As to the series ‘blunt, obtuse, stupid’, Hübschmann’s (1895) comment to

Horn’s (1893, no. 868) treatment of Prs. kund21 permits to date with some
accuracy the first proposal to separate Prs. kond ‘obtuse’ from Prs. kond/kondâ
‘brave’: “Np. kund ‘stumpf’ = bal. kunt ‘stumpf, grob, einfältig’ (Geiger) = skr.
(prākr.) kuṇṭha- ‘stumpf’. Davon kund ‘vir strenuus’ zu trennen, vgl. kundā,
kundāvar” (Hübschmann 1895: 88, no. 868).
The documentation of this series is as follows:
Prs. kond ‘blunt; sluggish’ (Lazard); Taj. kund, künt ‘blunt’, Bal. kunt ‘dull, blunt;
crude’ (← Prs. according to Elfenbein), kunṭ ‘blunt, dull; lazy’ (← IA according to
Elfenbein), Br. kunṭ ‘blunt, dull, stupid’ (Rossi 1979: 27, no. A197 ← Bal., but see

20
Azari kunda ‘lump of dough’, Turkish kunda ‘large poisonous spider’ and Kurdish kunde ‘wa-
ter-skin’, semantically belonging to the same group of § 3.3, show in their initial k° some form of
contamination with the series with initial g° treated in § 2 above.
21
Horn 1893, no. 868: “kund ‘stumpf; tapfer’ skr. (bezw. prkr.) kuṇṭha- ‘stumpf’. Vergl. Rückert,
ZDMG. 8, 280. np. kundāver hat nach Nöldeke (mündliche Mitteilung) im Šāhnāme immer die
Bedeutung ‘stolz’ ”.
362 Adriano V. Rossi

below), Parači kõnt, kõnd ‘cross-eyed’; Kurd. kund ‘stupid (istukan),22 Azari kund
‘blunt, slow, stupid‘; Kalasha koṇḍa ‘hornless’ (IIFL iv; to be connected with
CDIAL 3508 *kōṇṭha- ‘defective’ according to Morgenstierne, but commonly
CDIAL 3260 *kuṇḍa1 ‘defective’ and 3261 kuṇṭha- ‘blunt’ are quoted in this
connection), Pashai koṇḍā́ ‘bald’. According to Asatrian (1988: 163), Arm. kond-al
‘without hair’, k(o)nd-ol, kndlik (< *kond-al-ik) ‘without horns’ < Ir. *kund-, in view
of Arm. knt/d-el ‘to cut hair’ wherefrom knt-l-oz (with -oz suffix) and kond-al
‘bald’, with an original Iran. semantics ‘to cut’ > ‘with cut-hair, with a cut-horn’).
Bailey and Mayrhofer have treated in different occasions Skt. kuṇṭha- ‘blunt’
without proposing any sure etymology, but also without being aware of the
possible connections with the gund and kund families here in discussion. The
existence of DED2 1688, a Dravidian lexical family well documented and centred
on the semantics of ‘lame’, is anyway in my view a strong evidence for an areal
etymology (Iranian, Indo-Aryan, Dravidian).
One could observe that this semantics could be an evolution from that of
‘stump’, like that which connects, for example, ‘stump’ with ‘blockhead, foolish’
in informal French; the metaphor would be admissible also in other European
languages, including, e.g., Italian (ciocco = ‘stupid’), English (wooden = ‘dull’) and
Russian (čurban = ‘dull’), but, of course, I am also ready to accept the semantic
development proposed by Asatrian/Arakelova, i.e. ‘fat’ > ‘sluggish’ > ‘stupid’, in
view of their remark that in Iran and Armenia ‘stump’ is a regular metaphor for
the head of a family or a clan, or for a mature, venerable and sage man. I would
anyway remain rather prudent about any further connections between ‘wise’
and ‘bald’.

BIBLIOGRAPHY
Asatrian, G. S. (1988), “Suffiksal'nyj ėlement -l- i nekotorye voprosy fonosemantiki v armjanskom
jazyke”, Istoriko-filologičeskij žurnal 121. 2: 160-178.
―― / V. Arakelova (2001), “Blunt, Bald and Wise. Iranian kund(-)”, Iran and the Caucasus 5: 201-
206.
Bailey, H. W. (1955), “Indo-Iranian studies-III”, TPS: 55-82.
Belardi, W. (1954), “Una nuova serie lessicale indomediterranea”, RANL, sc. mor., VIII, 9: 610-644.
Budge, E. A. W. (1889), The History of Alexander the Great, being the Syriac Version, Edited from Five
Manuscripts, of the Pseudo-Callistenes with an English Translation, Cambridge.
CDIAL = R. L. Turner (1966), A Comparative Dictionary of the Indo-Aryan Languages, London.
Chantraine, P. (1968), Dictionnaire étymologique de la langue grecque, Paris.

22
Metaphorical development from kund ‘stump’ according to Kurdoev 1960 (s.v. kund I).
 Once again on Iranian *kund 363

Ciancaglini, C. (2008) Iranian Loanwords in Syriac, Wiesbaden.

de Lagarde, P. (1868), Gesammelte Abhandlungen, Leipzig.
DED = T. Burrow, M. B. Emeneau (19842, 19611), A Dravidian etymological dictionary, Oxford.
Desmaisons, J. J. P. (1908-1913), Dictionnaire Persan-Français, 4 vols., Rome.
Doerfer, G. (1967), Türkische und mongolische Elemente im Neupersischen, III, Wiesbaden.
Eilers, W. (1956), “Der name Demawend. Zusatznoten”, Archiv Orientální 24: 183-224.
―― (1971), “Iranisches Lehngut im Arabischen”, Actas do 4 Congresso de estudos árabes e islâmicos,
Leiden: 581-660.
Elfenbein, J. (1990), An Anthology of Classical and Modern Balochi Literature, II: Glossary, Wiesba-
den.
EVSG = G. Morgenstierne (1974), Etymological vocabulary of the Shughni group, Wiesbaden.
EWA = M. Mayrhofer (1992-2001), Etymologisches Wörterbuch des Altindoiranischen I-III, Heidel-
berg.
Frisk, H. (1960), Griechisches etymologisches Wörterbuch, vol. 1, Heidelberg.
Garsoïan, N. (1989), The Epic Histories attributed to P‘awstos Buzand (Buzandaran Patmutʿiwnkʿ), tr.
and comm. N. G. Garsoïan, Cambridge.
Gershevitch, I. (1969), “Amber at Persepolis”, Studia classica et orientalia Antonino Pagliaro oblata,
II, Roma: 167-251.
Gyselen, R. (2002), “Le kadag-xwadāy sassanide: quelques réflexions à partir de nouvelles données
sigillographiques”, Studia Iranica 31/2: 61-69.
Hansen, O. (1938), Die mittelpersischen Papyri der Papyrussammlung der Staatlichen Museen zu Ber-
lin, Berlin.
Horn, P. (1893), Grundriss der neupersischen Etymologie, Strassburg.
Hübschmann, H. (1895), Persische Studien, Strassburg.
IIFL = G. Morgenstierne (1973), Indo-Iranian frontier languages, IV: The Kalasha language, Oslo.
KEWA = M. Mayrhofer (1956-1980), Kurzgefaßtes etymologisches Wörterbuch des Altindischen I-IV,
Heidelberg.
Kurdoev, K. K. (1960), Kurdsko-russkij slovar', Moskva.
Lurje, P. B. (2010), Personal Names in Sogdian Texts, II, 8, Wien.
Pennacchietti, F. A. (1999), “Qundâqôr: un hapax siriaco del Romanzo di Alessandro tra filologia e
archeologia”, M. Lamberti; L. Tonelli (eds.), Afroasiatica Tergestina. Papers from the 9th
Meeting of Afro-Asiatic (Hamito-Semitic) Linguistics, Trieste, April 23-24, 1998, Padova: 71-
82.
Rossi, A. V. (1979), Iranian Lexical Elements in Brāhūī, Naples.
―― (2002), “Middle Iranian gund between Aramaic and Indo-Iranian”, Jerusalem Studies in Arabic
and Islam 26 (= Studies in Honour of Shaul Shaked): 140-171.
―― (2006), “Mprs. kundāg ‘indovino’”, Loquentes linguis. Scritti linguistici e orientali in onore di Fa-
brizio A. Pennacchietti/Linguistic and Oriental Studies in honour of Fabrizio A. Pennacchietti,
a cura di (edited by) P. G. Borbone, A. Mengozzi, M. Tosco, Wiesbaden: 613-633.
364 Adriano V. Rossi

Shahbazi, A. Sh. (1987), “Army: i. Pre-Islamic Iran”, Encyclopaedia Iranica, II, London-New York:
489-499.
Skjærvø, P. O. (2012), “Jamšid: i. Myth of Jamšid”, Encyclopaedia Iranica XIV: 501-522.
Spiegel, F. (1860), Einleitung in die traditionellen Schriften der Parsen. 2. Theil: Die traditionelle Lite-
ratur, Wien.
Widengren, G. (1976), “Iran, der grosse Gegner Roms: Königsgewalt, Feudalismus, Militärwesen”, H.
Temporini; W. Haase (eds.), Aufstieg und Niedergang der römischen Welt, II, 9/1, Berlin-
New York 1976: 219-306.
Tafazzoli, A. (2000), Sasanian Society: i. Warriors ii. Scribes iii. Dehqāns, New York.
Tavernier, J. (2007), Iranica in the Achaemenid period (ca. 550-330 B.C.). Lexicon of Old Iranian
Proper Names and Loanwords, Attested in Non-Iranian Texts, Leuven-Paris-Dudley.
 A Note on Armenian hrmštk-el

James R. Russell
Harvard University

Abstract
The hapax *framaštaq in the Babylonian Talmud is a loan from a Middle Iranian slang word for the
penis; from its base comes the common Armenian verb hrmštkel, “to shove in”, which is not at-
tested in Classical texts and might have had an obscene connotation in ancient times that it no
longer possesses.

Keywords
Slang, Babylonian Talmud, Sasanian dynasty, Buzandaran, Köroghlu, framaštaq, marz-, mālīdan,
hrmštkel, Alecander Romance, Bušāsf, Paruyr Sevak

The Aramaic of the Babylonian Talmud is replete with Middle Iranian loan
words, some of which reflect the same dialect variations as the Middle Iranian
loans into Armenian of roughly the same period: hraman instead of framān for
“command”, navasard instead of nō sāl for “new year”, and so on. The narratives
in the text abound in Iranian themes, too; and in recent years a number of schol-
ars, notably Daniel Sperber, Isaiah Gafni, Shaul Shaked, Yaakov Elman, and
Geoffrey Herman have been exploring these. They are of interest to the study of
ancient Armenia in both expected and unpredictable ways. For instance, we
learn that in the fourth century the Sasanian monarch Šābuhr (Shapur II, r. AD
309-379) sat down to lunch with the Jewish Exilarch and the two august person-
ages dined upon a single ethrog―a sort of citrus fruit used by Jews during the
festival of Sukkot. This sounds rather silly on the face of it and it is probably an
abbreviation: presumably they ate, and most definitely drank, a great deal more.
The dessert of fruit is most likely to be understood as a kind of rhetorical short-
hand for the comfortable intimacy of friendship and leisure the two men en-
joyed. This was an enviable situation for the leader of a religious minority of
course. In a world of swords and sabers one checked at the entrance to the din-

 Koninklijke Brill NV, Leiden, 2015

366 James R. Russell

ing-tent or left with horse and groom, here they were concluding their meal,
wielding a little fruit knife of peace.
Knowing that this shorthand was current and understandable to contempo-
rary readers and listeners adds pathos to the already poignant scene of the death
of the Armenian king Aršak II imprisoned by the same Šābuhr. One recalls that
the epic narrative of P‘awstos, the Buzandaran, leads one through a series of
confrontations, each a set piece of oral topoi, at the climax of which the Arme-
nian Arsacid was cast into the Fortress of Oblivion. There, thanks to the inter-
ventions of his faithful eunuch Drastamat, he was allowed to enjoy a last feast in
the manner of kings: he reclined on pillows, watched dancing girls, and suddenly
reached for a fruit knife and ended his life. Once we know the register of sym-
bols, the little fruit knife as suicidal weapon becomes that much more tellingly
poignant, that much sadder. And one might add that free men at feasts after they
surrendered their heavier, longer blades were still allowed to retain in a scabbard
strapped to the upper leg a sort of pocket dagger, in Armenian nran, which one
explain from Iranian *ni-rāna- literally “on the thigh”. Aršak was lacking even
this token weapon allowed any nobleman at any occasion. One mentioned
pillows: in Parthian and Sasanian society, the more pillows one reclined upon at
a feast, the more elevated one’s rank was; and Armenian custom followed the
practice, even using the standard Sasanian term for “pillow”, bališ, to mean an
honor bestowed. In the Middle Ages, the khalat, or robe, was to take place of the
latter. So in the Talmud a Rabbi losing a disputation has one mat after another
yanked away from underneath him till he dies―yet another example of how an
image can be inverted to drive home a point in a standard tale.1
Many are familiar with a cycle of epic tales about a Robin Hood-like figure
called in Turkish Köroğlu, “Son of the Blind Man”: multiple versions exist in Ar-
menian, as well as the Turkic languages, Kurdish, and even Modern Greek. The
story begins with the father of the hero, who is master of the royal stables, being
blinded. In ancient Iran, the ākhwarrbed, “stable master”, was part of the court

1
The locus classicus for the investigation of this technique is Professor Nina Garsoian’s analysis
of the transformation of Trdat the Great in the Agathangelos into a pig for his imprisonment of St.
Gregory: here it is the royal varaz, the boar-totem of Verethraghna, that is inverted. I have pro-
posed, in a study of the bas-reliefs on the drum of the Cathedral of the Holy Apostles, an inversion
in the opposite direction in the same narrative cycle: the thirteenth “Apostle” is Gregory, beset by
two serpents. These torture him as they do the imprisoned Zahhak; but the Christian saint will
overcome them and emerge bringing salvation rather than apocalypse (see Russell 2004: 1165-1191).
 A Note on Armenian hrmštk-el 367

hierarchy but also the lowliest courtier. Of Rabbi Yehuda I the Babylonian Tal-
mud says, Ahōrērē də-Rabbī ‘atīr mi-Shābūr mālkā, “The Rabbi’s stable-master
[using a Persian loan word, the same as that found in Armenian] was richer than
king Shapur”. So the office is a marked term whose meaning is best understood
in its Parthian and early Sasanian context: it calls the attention of the listener to
an epic or folk tale to what I would term a switching point in the narrative plot.
That is, the son of the stable master can move either up or down in the social or-
der. In this case the man is blinded for a supposed offense of lèse majesté and his
son becomes a bandit-leader, an anti-king who dedicates his life to avenging his
father’s unjust humiliation. The mention of the stable master is not so much
Jewish hyperbole as Talmudic shorthand based upon the realia of Parthian and
Sasanian society in Mesopotamia; and we can see how the topos, when under-
stood in its Iranian context, serves as an anticipatory signal in the narrative from
the Armenian highland from which the Köroğlu cycle drew its core material.
Reuven Kipperwasser, a prominent Israeli Iranist and Talmudist, recently
called my attention to a passage in tractate Mo‘ed Qatan 18a of the Babylonian
Talmud, in which Avitul the Scribe (or maybe the Barber, there is some dispute
about how to read his epithet) says in the name of Rav Papa that “the Pharaoh
who lived in the time of Moses was one amah tall, his beard was one amah long,
and his *framaštaq was one anah and one zereth long”. This fulfills the prophet
Daniel’s observation (Dan. 4.14) that God appoints the lowest of men over the
kingdom of men, the text smugly observes. The average height of a man in the
Talmud is four amahs; so the king of Egypt is a priapic dwarf. The text adds that
he was a magus―a Zoroastrian. Prof. Martin Schwartz has proposed in corre-
spondence for framaštak, a hapax, a persuasive derivation from the base marz-
“rub”, with preverb fra-. The base gives us both the Armenian loan marz-em “ex-
ercise” and Persian māl-īdan “to rub” (so Persian-in-Turkish peshtimal, “back-
rubber”, i.e., “towel”).2 In the Armenian version of the Alexander Romance the

2
Jastrow explains Talmudic Aramaic parmašt-aq as contrectatus or extensus, membrum virile,
citing Pers. parmāštak. The latter New Persian form is not cited by Steingass, but one finds parmās-
īdan “touch, extend” (Steingass). (The suggestion is cautiously advanced here that Turkish parmak
“finger” may be the result of a loan from Iranian, digit and penis being typologically related.) This
loan-word from Middle Iranian into the Aramaic of the Jews of Parthian or Sasanian Mesopotamia
in the era of the compilation of the Babylonian Talmud may have been vocalized in fact as
*framaštak; and the Iran. form may be derived from OIr. *fra-mar-, with a base H. W. Bailey sug-
gested in his Dictionary of Khotan Saka could mean “mark, feel”. Kohut proposes an etymology
from the base vaxš- “grow, sprout” with preverb fra- “forth”. Such a shift from v to m is normal, as is
368 James R. Russell

Egyptian Pharaoh Nectanebos is a rather pathetic figure: he dies by falling into a

hole he has not noticed because he was too busy discoursing on the mysteries of
the stars above, as the young Alexander observes with prim indifference. The
hapless Egyptian is depicted as indecently lustful, too. When he meets the gulli-
ble Olympias, i t‘aguhin macuceal kaṙuc‘aw aṙanc‘ t‘ap‘eloy, k‘anzi ēr vawašamol i
kanays “at (the sight of) the queen he stiffened with an erection without ejacu-
lating, for he was crazed by lust for women” (Simonyan 1989: 73).3
The Iranian term for the membrum virile found in the Talmud is not attested
in modern Persian or Classical Armenian. But one recognized it at once as the
source of the common modern Armenian denominative verb hrmštk-el, “to push
or ram in hard”. The word escaped the notice of both the great etymological lexi-
cographers of the 20th century, Adjarian and Jahukyan; and it does not appear in
the 19th century Nor Baṙgirk‘ since it is not attested in Classical Armenian texts.
Yet it is self-evidently a loan into Armenian from northwestern Middle Iranian,
probably in the Arsacid or early Sasanian period, a slangy term for penis that ap-
pears to have been fastidiously bypassed by Christian writers, even though its
southwestern Middle Iranian form was judged suitable for the purposes of a sca-
brous, derogatory Talmudic yarn about the detestable, priapic Egyptians. In this
disregard for Egypt the Jews of Mesopotamia shared a sentiment with their Zo-
roastrian neighbors. The Pahlavi books call “Alexander the Accursed” (he is
never “the Great”) musrāyīg menišn, “the Egypt-dweller”—clearly, the son of a
fellow with endowments much like those of his ancestors. In Armenia, mean-
while, *hramaštak would have endured down the ages in unrecorded vernacular

the reduction of the cluster xš to š; and the form is possible. Moreover, one notes that the base with
prepositional frā- is indeed attested in Yašt 19.50, in a series of boastful physical threats against the
man-dragon tyrant Dahāka after the latter has threatened the sacred fire: frā θwąm zadangha uz-
uxšāne “I’ll mount both your buttocks!”. Such a passage would have lent itself in popular tradition
precisely to the kind of vulgar sexual jocularity one finds in the Talmudic passage. Prof. Martin
Schwartz finds scant foundation for Bailey’s understanding of the root as “feel” and argues that
mar- cannot produce *maštak in Parthian. (But one notes at least OIran. *fra-jar-> Arm., via Pth.,
hra-žar-em, “take one’s leave of, reject” and hražešt, “farewell”.) Schwartz instead proposes a deri-
vation from the well-known base marz- “rub”—this is the root that produces New Persian māl-
īdan. (One may recall the Arm. surname Peshtimaljian, from Pers.-in-Tk. “towel maker”.) The latter
is very well attested in Armenian of all periods, cf. marz-em “exercise”, marz-aran “stadium”, etc. In
Pahlavi the base can have a sexual overtone, as in kun-marz, “anal intercourse”, an activity of
which the Zoroastrian books universally and sternly disapprove (unless, one supposes, one is
Taxma Urupi or another Kayanian hero sticking it to Ažī Dahāka).
3
Albert Wolohojian’s English translation bowdlerizes the passage.
 A Note on Armenian hrmštk-el 369

usage, surfacing only when the latter began to be chronicled; and by then, only
the derivative verb remained, its meaning having become innocuous. Such de-
demonizations are fairly common, within the Iranian sphere: Asmussen long ago
noted that the Avestan demoness of sloth, Bušyąstā, a personified optative “let it
be (later)”, becomes a Judaeo-Persian common noun for sleep, bušāsf, which has
no connotation of evil. And for the tendency to render innocuous sexually ex-
plicit expressions one might cite the Modern English example of “suck”. “This
coffee sucks” means that the beverage tastes terrible, not that it has become mys-
teriously endowed with very unlikely erotic skills. (The lyric “where the bees
suck” of the Bard in The Tempest has been ruined for this modern ear, which
hears but a complaint about somehow inadequate insects.) Michael Adams, in
his book Slang: The People’s Poetry, points out that the derogatory use of the verb
actually had nothing to do with fellatio originally, though everybody polled
wants, it seems, to think it does anyhow. Fra-marz-, hrmštkel, “shove it in”: honi
soit qui mal y pense. We can recover something of the conversational language of
ancient Armenia, and imagine some of the topoi of its storytelling, too, by look-
ing at two kinds of sources: Irano-Talmudica and the modern Armenian lexicon;
and by applying some of the methods of analysis of the folktale. The hramaštak
signaled a lustful rascal of a king; an axoṙapet minding the royal mounts was a
low courtier about to get rich or rebel; one counted pillows and measured the
tension of position and prestige; and a knife at table, depending on its size, was
to be a symbol of amity or tragedy—Chekhov’s rifle on the wall that must be
fired by the end of the play.
But the story does not end here for hrmštk-el. The study of Pahlavi and Tal-
mudic Aramaic are enriched by a knowledge, not only of ancient Armenian, but
of the modern language, as here. Those other tongues are dead, but Armenian is
fertile and alive, and its literature evolves with old words and new imaginings; so
we find forms of this verbal base, for instance, in the poem “Nightfall” (Gišera-
mut) of the great poet Paruyr Sevak (1924-1971), in his cycle Ełic‘i luys, (“Let there
be light!”, Sevak 1969: 32):
The automobiles
That seemed until now to be blind,
Blind as a cat’s newborn brood
Now see one another and kindle their eyes.
And the silence already
Is thrusting all else aside
370 James R. Russell

To clear itself its proper place

And from that long-drawn thrust
The mountains appear to have moved afar.
‘U lṙut‘yunǝ
Arden amen inč‘ hrmštorum ē,
Or iren hamar kargin teł bac‘i,
Ev ayd tevakan hrmštoc‘ic‘ ē,
Or heṙac‘ac en t‘vum leṙnerǝ’.
Loneliness, like Khayyam, is getting drunk,
Getting drunk and cursing God.
And now the good dogs’ bark,
It seems, is not at all a barking but a prayer
Dispatched to blasphemed God
To ask Him that He pardon His abuser.
The dark becomes a wet sponge on the board
Erasing all the sky till heaven
Is slowly encrusted with hoar frost of stars.
And men begin to speak
Less, and more quietly,
Since the lights and the lamps
Tell more eloquent tales.
Every house sends forth
Its declaration to the sky
In the cry of a child,
A mother’s call,
The animals’,
The cars’
Telegraphic bleat and bark,
But above all else in these,
The lights that burn and blink, go out,
Unending, corresponding
To an alphabet, a new Morse code.
And then,
When all these voices are already still—
The lights, as well—
It seems the time has come
For the universe to respond:
In sleep all men receive their answers
In the shape of nightmares or good dreams.
 A Note on Armenian hrmštk-el 371

Good dreams alone to you, my darlings;

Let your nightmares only come to me, your friend.

The verb becomes a part now of an introspective meditation on the way si-
lence comes to a great city, Erevan, at dusk, the mountains in the darkness and
the quiet seeming to recede. It is not merely de-demonized, but ennobled in its
context. The poet, seemingly standing aside to perceive the esoteric language of
the nightfall, emerges at the end beside us, a friend ready to shoulder even the
weight of our nightmares. The scholarship of Professor Garnik Asatrian em-
braces all the provinces and ages touched upon in this philological note, and
many more; but those who have rejoiced in his hospitality know a Parthian par-
adise on the plain of Ararat where good wine and conversation mingle with the
magic night, and many are the burdens he has shouldered for us as a friend, like
a titan on Masis or a hero of Sasun, for which our gratitude will have no end.

BIBLIOGRAPHY
Russell, J. R. (2004), “A Bas-Relief on the Cathedral of the Holy Apostles, Kars, Armenia” (UCLA
Conference on Kars and Ani, Nov. 2001), Armenian and Iranian Studies, Armenian Heritage
Press and Harvard Armenian Texts and Studies 9, Cambridge: 1165-1191.
Simonyan, H. (ed.) (1989), Patmut‘iwn Ałek‘sandri Makedonac‘woy, Erevan.
Sevak, P. (1969), Ełic‘i luys, Erevan.
 Aspects of Udi-Iranian Language Contact

Wolfgang Schulze
University of Munich

Abstract
The paper examines some aspects of language contact in the regions of present-day Azerbaijan fo-
cusing on the Southeast Caucasian language Udi and its ancestor (Caucasian Albanian). Both lan-
guages are marked for a rather pronounced impact from languages that must have played a domi-
nant role since Classical antiquity and in Medieval times. Here, at least three layers can easily be
isolated: (a) Old Armenian, (b) both Northwest Iranian and Southwest Iranian, (c) Turkic in terms
of Oghuz Turkic (Azeri). Both (Early) Modern Persian and Azeri conditioned that Udi was later-on
at least partly integrated into the world of Oriental language despite of the fact that its speakers
remained Christians. The pre-Oriental impact from Armenian and (mainly northwest) Iranian lan-
guages resulted in significant shifts with respect to both the lexicon and the grammar of Caucasian
Albanian and Udi that set apart both languages from the world of Lezgian languages. The paper il-
lustrates the presence of different Iranian layers in Caucasian Albanian and Udi, addressing both
lexical and morphosyntactic issues. With respect to morphosyntax, two topics are discussed,
namely the emergence of Split-O strategies in Caucasian Albanian and Udi and the development of
a system of floating agreement clitics. Both patterns represent instances of structural borrowings
from local Iranian languages, which likewise testifies the former relevance of Iranian languages in
Central Azerbaijan.

Keywords
Language contact, Structural and lexical borrowings, Southeast Caucasian languages, Udi, Cauca-
sian Albanian, Iranian languages

1. INTRODUCTION
The world of languages spoken in the geographical regions of nowadays Azer-
baijan has been dominated by three language families since Classical and Medi-
eval times: Indo-European (Iranian languages and later-on Armenian), Turkic
(first Kipchak languages, then Western Oghuz languages in terms of Azeri),1 and

1
In this short paper, I cannot elaborate on the highly controversial question of dating the arri-
val (and stabilization) of Turkic speaking groups in Azerbaijan. In principle, we have to assume
that early Kipchak groups have settled in the region since 552 BE in the context of the so-called Sa-
 Koninklijke Brill NV, Leiden, 2015
374 Wolfgang Schulze

East Caucasian (mainly southeast Caucasian (Lezgian)). In addition, we have to

take into account Arabic as a superstrate language that, however, has never been
spoken in compact areas, but rather among the elites of built-up areas and urban
settlements.
The linguistic map of the Azerbaijan region is difficult to reconstruct if we
aim at periods prior to the 14th century. The corresponding sources stemming
from Greek, Latin, Armenian, and Islamic historians and geographers reflect ra-
ther heterogeneous, sometimes contradicting patterns that are at least in parts
grounded in idiosyncratic or even biased views or simply derived from rumors.2
Naturally, a major source for describing some aspects of the linguistic patterns in
the early Azerbaijan regions are the languages themselves (together with their
discernible history). Here, one can start from at least four observational points:
a. Local toponyms/hydronyms and their linguistic identification from an etymo-
logical point of view.
b. The reconstruction of basic botanic, zoological, social, economic, and cultural
terms that would be relatable to specific regional aspects.
c. Borrowings from identifiable languages (that must have been based on lan-
guage contact) together with their chronology.
d. Early datable and localizable (written) sources in the given language.

Results from these tasks should then be confronted with the non-linguistic
sources mentioned above. Given the fact that we have to start from (today) at
least fifteen languages in the Azerbaijan regions (ignoring their extremely rele-
vant local varieties), these tasks can be approached step by step only. Hence, any
assumption of the linguistic patterns at issue must always be very preliminary,
and researchers should be steadily ready to re-evaluate their claims pending on
new findings that stems from these tasks.
Nevertheless, some data allow depicting some rudimentary linguistic pat-
terns that can be related to the early Azerbaijan regions. Starting from what we
know about the history of the three language families at issue (Indo-European,

bir people and later in the context of the Khazars. Residues of their languages may be found in
some few Kipchakisms especially in Khinalug (see section 2 and Schulze (in press a)). However,
the speakers of these languages had soon become assimilated to adjacent languages.
2
In addition, we have to take into account the fact that the mentioning of just an ethnonym in
a Classical source hardly helps by itself to identify an ethnic unit. Even if we can relate such an
ethnonym to a specific language from an etymological point of view, we still cannot tell for sure
that we have to deal with an endonym. It may likewise reflect an exonym given to an ethnic group
by adjacent groups.
 Aspects of Udi-Iranian Language Contact 375

Turkic, and East Caucasian), we can assume that prior to the intrusion of Arme-
nian, the whole region was linguistically dominated by some kind of early
Northwest Iranian (Median) and by southern East Caucasian languages. The East
Caucasian languages are tentatively related to the Kura-Araxes (early trans-Cau-
casian culture) that ended before 2600 - 2000 BCE (see Edens 1995, Mallory 1997).
Towards the end of the second millennium BCE, the southern (and central?)
parts of the Azerbaijan regions were gradually taken over by Median settlers and
nomads, representing the first identifiable ethnic layer. By that time, the north-
ern parts of the area must have been inhabited by East Caucasian groups some of
which were later-on subsumed under the cover-term Ἀλβανία. This term may
have been known already in the 4th century BCE in case some of the toponyms
and ethnonyms occurring in the report on Alexander’s Anabasis by Lucius Fla-
vius Arrianus (ca. 95-175 CE) reflect older sources (related to the battle of Gau-
gamela, 331 BCE) (see Gippert et al. 2009 for details). For the time being, it is dif-
ficult to give the term Albanian a specific ethnic profile. The name itself is prob-
ably of Iranian origin. One argument is its mentioning in the trilingual inscrip-
tion of Šāhbuhr I (Naqš-i Rustam, Kaʽba-i Zardušt, ca. 245 A.D.), cf. graphic (1),
the text is taken from the edition by Huyse (1999: 22-23):

Graphic 1: The trilingual inscription of Šāhbuhr I (Gippert et. al. 2009)

Greek Ἀλβανία thus corresponds to Parthian ’rd’n, that is Ardān. Middle Per-
sian would give us graphical ’ld’n- or ’l’n-, which might be alān or arān. Given the
fact that later Persian and Arabic sources attest the form a(r)rān, ar-rān (also
compare Georgian ran-i) and that Greek Ἀλβανία and Armenian ałowan reflect *-
lb- rather than rd- suggests that the form ’rd’n (ardān) in Parthian inscription is a
borrowing from another Iranian language that was not marked for the shift -dw >
-b- characteristic for Parthian. It may well have been the case that in Parthian a
form ar/lban- < ardān was likewise common and that it ultimately served as the
376 Wolfgang Schulze

basis for the Greek and Armenian version of this toponym (see Gippert et al.
2009 for details and further arguments).
Most likely, Albania once denoted a toponym that was associated with a fed-
eration of different ethnic groups. One of these must have later-on become the
dominant unit within this federation. The language of this ethnic group has been
preserved in some minor inscriptions from the 6th to the 7th century CE), as well
as in the lower layer of two palimpsest manuscripts there had been found in the
Katharine Monastery (Sinai) and that had been identified as ‘Albanian’ by Zaza
Aleksidze in 2000 (Gippert et al. 2009). The palimpsest texts comprise parts of
the Gospel of John and of a Christian lectionary stemming from the 6th – 7th
century BE. The palimpsests include some 10.000 word tokens (reflecting roughly
800 basic lexical types) and hence represent both the oldest and the largest cor-
pus of an East Caucasian language in pre-modern times. Both the inscriptions
and the palimpsest texts are written in a specific alphabet traditionally related to
Mesrop Mashtots. Its decipherment is grounded in preliminary hypotheses by
Zaza Aleksidze and has been carried out by Jost Gippert and the author of the
present paper during the editorial work on the palimpsest texts. In order to illus-
trate this alphabet, the facsimile of three lines of the lectionary is given in
graphic (2) together with transliteration and translation:

Graphic 2: A sample of the Caucasian Albanian palimpsest texts (Act 13,35, Gippert et. al. 2009)

The fact that the language present in the Caucasian Albanian (CA) sources
mentioned above had been used to translate parts of the Bible or even the whole
Bible suggest that this language must have played a major role in the statehood
of Albania since the 5th century CE.3 By that time, Old Armenian must been rele-

3
This process is perhaps reflected in the famous passage from Strabo XI,4.6: διαφέρουσι δὲ καὶ οἱ
βασιλεῖς: νυνὶ µὲν οὖν εἷς ἁπάντων ἄρχει, πρότερον δὲ καὶ καθ’ ἑκάστην γλῶτταν ἰδίᾳ ἐβασιλεύοντο
ἕκαστοι. γλῶτται δ’ εἰσὶν ἓξ καὶ εἴκοσιν αὐτοῖς διὰ τὸ µὴ εὐεπίµικτον πρὸς ἀλλήλους.’Their kings differ
from one another; at present one king governs all the tribes. Formerly each tribe was governed by a
king, who spoke the peculiar language of each. They speak six and twenty languages from the want
of mutual intercourse and communication with one another.’ (A. Meineke, Strabonis geographica,
 Aspects of Udi-Iranian Language Contact 377

vant for at least the translators, if not for the region at issue. This is revealed by
numerous instances where the translators of the lectionary and of the Gospel of
John have undoubtedly referred to Armenian linguistic patterns. In addition, a
number of loans especially in the lectionary that was read aloud during the
church service presuppose their (rudimentary) understanding by the auditory.
From this it follows that Old Armenian loans are not just a technical solution (cf.
avel ‘many, much’, hać’eq ‘right (hand)’, k’ala ‘lame’, k’or- ‘bent, crooked’, garaz-
man ‘grave’, marmin ‘body’, and *žolovowrd (only in abbreviated form) ‘crowd,
people’, see Gippert at al. 2009: II,80). However, many other loans that seemingly
have parallels in Old Armenian ultimately stem from Iranian. Sometimes, there
are phonetic differences with respect to the Old Armenian correspondences,
which suggests that the terms may have been borrowed from Iranian inde-
pendently. Examples (all of them taken from Gippert et al. 2009: II,80) include:
powsak’ ‘crown’ vs. Arm. psak ‘id.’ < Early NW-MIran. *pusak-, cf. MP.Pth. pusag,
Sogd. pusak, afre-pesown vs. Arm. awhrnem / awrhnem ‘to praise’, bod’var ‘censer’
(< MIran. *bōδiβār-) vs. Arm. bowrvar̄ and Georg. bervar-i, vaˤamak’ ‘cerecloth,
napkin’ vs. Arm. varšamak, Georg. varšama(n)g-i and Sogd. wʾšʾmy, Modern Per-
sian bāšāma, xoˤak’ ‘heat’ vs. Arm. xoršak, Georg. xoršak’-i and MPT. hōšāg ‘hot
wind’, and mowˤak’ ‘worker’ vs. Arm. mšak, Georg. mušaḳ-i. In the last three ex-
amples, we can observe the reflex of *š/rš by the CA pharyngeal ˤ. Iranisms not
shared as such by Armenian are for instance bamgen ‘blessed’ (MIran. *bāmgēn,
lit. ‘ray-like, shining’), marġaven ‘prophet’, as opposed to Arm. margarē ‘id.’ (pre-
sent stem, not the past stem of the Iranian verb ‘to see’ (*marγa-wēn- vs. *marγa-
δē ‘augur’, lit. ‘bird-seer’), dowrowd (preserved in Udi durut’) ‘beam, wood (used
for the cross)’ (Pth. dārūβδag ‘crucified’), d’ip’ ‘scripture, book’ (OPers. dipī-’in-
scription’, Armenian only dpir ‘writer’ (< *dipī-βā̆ r-?).4
The ancient Iranian languages spoken in the Azerbaijan regions are difficult
to identify. As has been said above, we have to take into consideration early vari-
ants of Northwest-Iranian (Median), but we cannot exclude the presence of Old
Persian speaking peoples. By the time Albania was converted to Christianity,
Middle Persian (replacing Parthian) must have played a major role on a more
administrative or ‘official’ level (later-on continued by Dari resp. Modern Per-

vol. 2, Leipzig 1877 (repr. 1969). English translation: H. C. Hamilton and W. Falconer, The Geogra-
phy of Strabo. Literally translated, with notes, in three volumes, London 1903).
4
The identification of these terms as borrowings from Iranian as well as the Iranian data and
their analysis stem from the work by Jost Gippert (Frankfurt).
378 Wolfgang Schulze

sian). The Northwest-Iranian varieties seemed to continue functioning especially

in rural and urban (non-administrative) communication. These languages are
conventionally subsumed under the term Azari. They were present in many
parts of the Azerbaijan regions up to the 17th century, but lost importance after
the gradual intrusion of Oghuz Turkic. The only true remnants of this linguistic
tradition within the borders of what now is Azerbaijan proper seem to be
Northern Kurdish (Kurmancî)5 varieties and Northern Tolyshī. These languages
have to be distinguished from the layer of Tātī varieties in Northeast Azerbaijan
that belong to the Southeast Iranian dialect continuum.

2. THE EAST CAUCASIAN LANGUAGES OF AZERBAIJAN

Today, the Azerbaijan regions include a relevant number of ethnic groups the
languages of which belong to the family of East Caucasian languages. There is no
absolute proof, however, that groups such as the Bałasičkc, the Čiłbkc, the Lpinkc
(lat. Lupeni), the Głowarkc, the Maskcowtckc, the Pcoxkc, or the Vatkc mentioned in
Armenian sources (see Bais 2001) spoke languages of this family. None of these
names has a convincing etymology that would relate them to one of the overall
thirty East Caucasian languages. It may well have been the case that at least
some of these spoke a South Caucasian language related to Old Georgian or a
language without further descendants. Given the fact that full language shift has
been a common pattern in Azerbaijan regions since Antiquities, we cannot even
expect that these languages would have left their traces in terms of a substrate.
The only ancient language that we can clearly associate with East Caucasian in
fact is Caucasian Albanian. There is sufficient evidence that Caucasian Albanian
belongs to the eastern branch of the Samur languages that again form the bulk of
Southeast Caucasian (or Lezgian), cf. graphic (3) that gives a genealogical tree for
this language subfamily:6

5
Here I cannot turn to the question of whether the Azerbaijani Kurds can be regarded as im-
mediate descendants of the Azari tradition, or whether they have migrated to this region lately,
only. See Asatrian (2009) for an illuminating discussion of hypotheses concerning the ethnolin-
guistic patterns of Kurdish tradition from a historical point of view.
6
Those languages that are currently present in Azerbaijan are indicated by small caps.
 Aspects of Udi-Iranian Language Contact 379

Graphic 3: The Southeast Caucasian (Lezgian) languages in their East Caucasian setting
In addition, we have no evidence that speakers of the two Southern Samur
languages Kryts and Budukh (as well as now extinct relatives) have migrated
into their present locations in later times. Hence, it is rather likely that they
represent residues of a another, perhaps more compact linguistic subgroup that
had dwelt northeast of Caucasian Albanian speaking units in ancient times
(Shah-Dagh mountains and adjacent areas). The situation is perhaps different
with respect to Lezgi, Rutul, and Tsakhur. All three Lezgian languages are spoken
mainly north of the Great Caucasus range. It may well have been the case that
their speakers migrated later-on into the Azerbaijan regions in the context of
transhumance traditions. This probably also holds for the Avar speakers of the
Zaqatal region. The case of Khinalug is problematic. Most likely, its speakers
once penetrated a settlement called Khinalug (kätš in Khinalug) that was before
populated by Christian and Jewish groups (see Schulze 2008 for details).
Evidence from Oriental sources7 suggests that at least the settlement of Khinalug
had been known before 1200 BCE. However, this does not necessarily mean that
it had been inhabited by Khinalug speakers in the sense of modern Khinalug.
The ‘Lezgoid’ elements of Khinalug suggest that its speakers had been in
longstanding contact with Lezgi proper and Southern Samur languages, be it in
the village itself or in its surroundings. Nevertheless, we cannot exclude the pos-
sibility that early Khinalug speakers already formed a part of the East Caucasian
layer of pre-Turkic Azerbaijan.

7
Under this name, the village has first been mentioned by Yakūt al-Ḫamāwī, a 13th century
Turkish-Arabic geographer. The term also appears in the name of Muḥammad Ḫinalugī (1456) who
prepared a’family tree’ of local rulers.
380 Wolfgang Schulze

Another layer may have been represented (in the West) by ethnic groups lin-
guistically related to the Nakh language family (Chechen, Ingush, Bats). Most
likely, at least some Nakh speakers once dwelt in the eastern regions of Georgia
and northwestern regions of Azerbaijan (until the times the Georgian chronicle
kartlis cχovreba had been compiled (post 800 BE)) and later-on migrated to the
northern parts of the Caucasus8. As far as data go, however, there are no discerni-
ble traces of these languages in present-day languages of the Azerbaijan regions,
except for nevertheless highly significant isoglosses with Khinalug that hint at an
origin of the later language in the western parts of this region (see Schulze
2008).9
Udi seems to be the only Eastern Samur language that has been spoken in the
region since Antiquities. Currently, Udi is spoken by some 4.000 people, present
mainly in the village of Nij (niˁź) in Northwestern Azerbaijan. Some Udi speakers
still live in the adjacent village of Vartashen (now Oguz), although the bulk of
Vartashen Udis had left the village in 1989/90 and fled to places in Northern Ar-
menia, Russia, and Kazakhstan in the context of the Armenian-Azerbaijani con-
flict. Already in 1922, a group of Udis from Vartashen had migrated to Eastern
Georgia and had founded the settlement Zinobiani/Oktomberi. Historical evi-
dence suggests that before 1832, Udi had been spoken in a wider ensemble of
villages and settlements, including Vardanlu, Mirzabeily, Sultan-Nukha, Dzhurlu,
Malykh, Engikend, Ermanit, Mukhants, Oraban, Kungüz, Kutkashen (Qəbələ),
Kormukh, Gish, and Bum. Today, the village of Nij is marked for a rather com-
pact Udi population (plus two Azeri quarters).10 The situation on Vartashen had
been rather different from this pattern: Toward the end of the 19th century, the

8
This holds especially for the groups called Naxčcamatean, Dzurdžuk, and Gilgv (= Čiłbkc ?) in
medieval sources.
9
Two words seem to be key-terms in this respect: Khinalug mǝdá ∼ mda ‘mountain’ is difficult
to separate from Georgian mta ‘mountain’. In order to keep the ‘East Caucasian’ perspective, we
would have to propose a rather idiosyncratic sound change: m(ə)dá < *mlá < *mal-á. Second,
Khinalug k’inaz ‘wine grape’ has matches only in Chechen kems, Ingush koms, and Bats kaniz. The
term does not have further cognates in ECL. The segment -az is a typical Khinalug word formation
element, which suggests that the word had been borrowed into proto-Nakh (or showed up as a
wanderwort in the later Nakh languages). The only possible contact zone between Khinalug and
Nakh, however, can only have been the regions of present-day Eastern Georgia.
10
Of the 14 quarters (mahalla) (Falčilli, Mančili, Ağdaläkli, Melikli, Ferimli, Malbel, Vezirli,
Daläkli, Čirmählä, Daramählä, Darabağ, Godžibeyli, Yalgašli, Abdallı), only the last two are inhab-
ited by Azeris.
 Aspects of Udi-Iranian Language Contact 381

majority of Vartashen inhabitants was formed by Udis and Armenians. Addi-

tionally, there had been Christian Tātīs, Jewish Tātīs, Lezgis, and Azeris. The
number of Azeris gradually increased in the 20th century (Bezhanov 1892).
Udi’s closest relative is Caucasian Albanian (see Gippert & Schulze 2007,
Gippert et al. for details). However, we cannot claim that Udi is an immediate
descendant from Caucasian Albanian. Linguistic evidence from CA suggests that
Udi is an off-spring of a dialectal variant of the language documented in the
palimpsests (or even of a distinct language). Both CA and Udi are marked for in-
novations not common to both languages. From that it follows that both varie-
ties (Caucasian Albanian and ‘Early Udi’ must have developed out of a common
language (here called ‘Language X’) that again was an early off-spring of Eastern
Samur. Graphic (4) illustrates these relationships schematically:

Graphic 4: The Eastern Samur branch of Lezgian

Summing up what has been said so far, we can design a tentative linguistic
map of the regions of Azerbaijan that indicates the languages resp. language
families that are likely to have been present in the region at about 500 BCE to
300 CE:
382 Wolfgang Schulze

Graphic 5: A tentative linguistic map of the Azerbaijan regions (500 BCE – 300 CE)

3. UDI AND IRANIAN: THE LEXICON

Graphic (5) illustrates that the Azerbaijan regions must have once been marked
for two types of contact zones with respect to Caucasian Albanian / Early Udi:
Towards the southwest, we have to think of contact with Old Armenian at least
since the times the Albanian script had been introduced (~ 410 CE). Since then,
contact with Armenian has become a dominant factor in the formation of Udi
varieties (see Schulze 2011a for details). A second contact zone must have been
present in the context of Northwest Iranian languages, none of which, however,
today is in immediate contact with Udi. Instead, Southwest Iranian languages
have played a more or less pronounced role starting with Early Modern Persian
since the 16th century. This is especially true for the formation of the Udi lexicon.
Many terms have entered the Udi directly or via Azeri Turkish. The correspond-
ing layer can be easily illustrated with the help of the following examples:
aiz ~ ayiz ‘village’: Cf. Persian ize ‘village’.
andax-besun ‘think; remember’: The incorporated element andax is related to
Persian andiše ‘thought, idea, care’ (cp. andišmand ‘thoughtful’).
apči ‘lie (n.); wrong; false’: Most probably a metathesized reflex of Persian čāp ‘lie’.
azad ‘free’: Borrowed from Persian āzād ‘free’.
azar ‘ill; illness’: Based on Persian āzār ‘insult, hurting, torture’ (> āzāri ‘ill person;
ill’) > Azeri azar ‘ill’.
 Aspects of Udi-Iranian Language Contact 383

bačuk’-esun ‘be inflamed; be burnt’: The lexical base also reads bač‘uk’ and de-
notes ‘burning, inflamed’. It is borrowed from Armenian baccuk ‘[fire] stick’
and is also present in Colloquial Persian bačuq-ī ‘light [your cigarette]’ ~
bačup-ī [lit.: ‘[have [it] to the fire stick’ < čup ‘wood’].
bač’an ‘back; hip’: The term has no obvious correspondences in Lezgian. Perhaps
the lexeme is a corrupt from of Persian post ‘back’ (in a Northwest Iranian rep-
resentation (cp. Northern Talysh pešt)) to which the plural morpheme -ān has
been added [pešt-ān > *pešan > *bašan > bač‘an]?).
barabar ‘equal’: Borrowed from Persian bar-ā-bar ‘like what is opposed, equal’.
barun ‘wall’ (<‘what separates’ ?), cf. Persian bāru ~ bāre ‘wall, fence’.
baxt’avar ‘blessed; praised’: Borrowed from Persian baxtwar ‘happy’. The religious
connotation seems to be an Udi innovation. The medial -a- is probably taken
from the Armenian reflex baxtavor.
baxt’in ‘for; in favor of’: Ergative>instrumental (modal) of baxt’ ‘fate, luck, favor’
(< Persian baxt).
beivan ‘wild; wilderness’: Loan from Persian biyābān (lit.: ‘those without water[s]’)
’wilderness, desert’.
beś, beš, beˁś ‘in font of’: Postposition that basically denotes any kind of appeal
towards a landmark that is located ‘before’ the trajector (in time and space), cf.
Persian piš ‘front, forward, before, earlier’. The voiced stop may be motivated
through analogy from Northwest Iranian ba- ‘towards, in front’.
bia ‘evening’ [Vartashen begä]: Borrowed from Persian pagāh ‘early morning,
dawn’ which - in a number of East Caucasian languages - has been reanalyzed
as ‘dusk’ (probably through interference of Persian šāmgāh ‘evening, dusk’).
boš ‘in; into’: Obviously composed from *bo- and -š. Both segments are rather
opaque: bo- can tentatively be related to the preverb bai- ‘into’. There are no
acceptable proposals to relate the Udi preverb *bai- resp. adverb *bo to paral-
lels in other Lezgian languages. Rather, we should consider the possibility of a
loan from an earlier variant of Persian (or some Northwest Iranian) be ~ ba
‘into’.
boxov ‘ties’: Borrowed from Persian boxu ‘ties’.
butparaz ‘pagan’: Borrowed from Persian bot parast (lit.: ‘who adores an idol’).
därd ‘pain’: Borrowed from Persian dard ‘pain’.
därya ‘sea; lake’: Borrowed from Persian daryā ‘sea’.
däru ‘pain; medicine’: Based on Persian dāru ‘medicine’; the semantics ‘pain’
probably results from the contamination with därd ‘pain’.
384 Wolfgang Schulze

däst’ägä ‘sheath’: Borrowed from Persian daste ‘handle’ + gāh ‘place’, lit.: ‘place of
the handle’.
diźam ‘blasphemy; laughter’: Ultimately borrowed from Persian dežam ‘wild,
dark, exited, amazed’ ~ dižkāma ‘horrible, angry’.
ǯam ‘pot; cup’: Borrowed from Persian ğām.
ǯigär ‘liver; courage’: Based on Persian ğegar ‘liver, heart, innards’. The meta-
phorical use of ğegar to denote ‘courage’ has started in Persian.
färist’a ‘angel’: Borrowed from Persian freest ‘angel’.
gam-esun ‘become warm; warm oneself’: The lexical base is borrowed from Per-
sian garm ‘warm’. The fact that that medial -r- does not yield pharyngealiza-
tion suggests a younger borrowing.
gir-besun ‘collect; gather’: The lexical base gir ‘heap, quantity, collection’ is most
probably borrowed from Persian gerd ‘circle’, cp. gerd kardan ‘to collect’ which
parallels Udi girbesun just as Persian gerd šodan corresponds to Udigir-esun ‘to
gather (intr.)’.
girvänkä ‘pound (measure)’: Borrowed from Persian girvānke ‘pound’.
gölö ‘much, very’ [Nij: gele]: Cf. Kurdish (Sorani) gelêk ~ gele ‘much, many’. The
source may also be Persian gele (classifying intensifier), cf. yek gele mard ‘what
a man’.
hammaša ‘always’: Borrowed from Persian hamiše ‘always’.
harǯürä ‘any kind’: Borrowed from Persian har ğur ‘kinds, sort’ (Udi dative).
iräšp’är ‘overseer; steward’: The form seems to be related to Persian sar-parast
‘overseer’ (> *raspar > rašp’ar > iräšp’är?).
kam ‘few’: Borrowed from Persian kam ‘few, a little, rare’.
kar ‘deaf’: Borrowed from Persian kar) ‘deaf’.
kar-pesun ‘live’: Transitive derivation from kar ‘work, living’. The lexical base is
borrowed from Persian kār ‘work’.
kaś-esun ‘be broken up; be dug up’: The lexical base is probably derived from Per-
sian kāštan ‘to dig up the earth, to sow, to plant’.
ke ‘that (conj.)’: Borrowed from Persian ke ‘that’ (conjunction).
k’ic’k’e ‘little; small’: Most probably a corrupt form of Persian kuček ‘small’ (North-
ern Oriental küčük etc.).
k’oc’-baksun ‘lean on; be shaken; be bent’ [var. k’oć]: Most probably, the term is
borrowed from Persian kağ ‘bent, curved’ etc.
lägän ‘washing basin’, borrowed from Persian lagan (CA *laqen ‘basin, bowl’).
mandak’ ‘tired; exhausted’: A -Vk-variant of Persian mānde ‘tired, exhausted’ etc.
(< māndan ‘to stay > stand back’).
 Aspects of Udi-Iranian Language Contact 385

mandesun ‘stay; wait’: Borrowed from Persian māndan ‘to wait, to stay’. The ver-
bal root mand- is reanalyzed as man-d- in Udi (‘stem’ man- plus auxiliary -
desun, hence the endoclitic slot used for agreement clitics man_d-).
maran ‘sword’: Borrowed from Persian merān ~ marān ‘sword’.
mis ‘copper’: Borrowed from Persian mes ‘red copper’.
nägi, näyi- ‘not (cond.)’: Composed of the modal negative particle nä- (< Persian
nā-) and the conditional marker gi- (> -yi-).
namaz ‘temple; prayer’: Borrowed from Persian namāz ‘prayer’. The meaning
‘temple’ is an Udi innovation (cp. Persian namāzgāh ~ namāz-xāne ‘temple,
mosque, church’).
niśan ‘sign; mark (n.)’: Borrowed from Persian nešān ‘sign, mark’.
nökär ‘servant; slave’: Borrowed from Persian noukar ‘servant, slave’.
oğand ‘fitting; adequate; successful’ [ohand ~ ovand]: Most probably borrowed
from Persian āġande ‘full, fitting’.
ost’avar ‘strong; much’: Borrowed from Persian ostovār ‘strong, hard, durable’ etc.
puč’-baksun ‘be marred; be useless’: The lexical base is borrowed from Persian puč
‘empty, useless, senseless’.
q’ošč’ağ ‘courageous’: Composed from Persian xoš ‘good, strong’ and Udi č‘ağ ‘rip’
[‘having strong rips’ > ‘be courageous’].
q’ošin ‘army; soldiers’: Borrowed from Persian qošun ‘soldiers, army’.
q’uti ‘chest; coffin’: Borrowed from Persian qutī ‘chest, small box’ etc.
sövdäkär ‘merchant’: Borrowed from Persian soudāgar ‘merchant’.
subuk’ ‘light (of weight); light-minded’ [suvuk’ ~ subug ~ suhuk’]: Borrowed from
Persian sabok ‘light (of weight), light-minded’ etc.
šad-besun ‘loosen; free; untie’: The lexical base denotes ‘free, untied’ and perhaps
results from the metaphorization of Persian šād ‘happy, glad’.
širäng ‘pitcher of water’: Borrowed from Persian šarang ‘pumpkin’ (‘used to fetch
water’).
täg ‘twig’ [täy]: The Nij variant directly copies Persian tāī ‘twig’ whereas the
Vartashen variant täg is based on a Northwest Iranian form (as documented
e.g. by Baloči tāk, also cp. Pehlvei tāk ‘piece (of a tree))’.
t’ap’-pesun ‘hit (v.)’: Most probably from Persian tap ‘hit, clap’ etc.
xost’amaz-baksun ‘confess’: The lexical base is derived from Persian xostu kardan
‘to confess’ (suffix -āmiz > xostāmiz).
zad ‘time’: Numeral classifier borrowed from Persian zad ‘hit, kick’.
386 Wolfgang Schulze

The impact from (Early) Modern Persian on Udi cannot be isolated from cor-
responding processes given in other Southeast Caucasian languages. Although
we do not have at our disposal a comparative study on Persian loans in East
Caucasian, a preliminary inspection of the lexical data illustrates that with the
exception of Archi, in fact all Lezgian languages are marked for these Persian-
based Orientalisms11 (event though not to the same extent: The more a language
is located to the South, the more this impact becomes apparent). Nevertheless,
Udi seems to be affected by Early Modern Persian much before this happened to
the more northerly languages.
A typical Orientalism of this kind is also given by the pattern of light verbs
used in verb formation. We can safely assume that this pattern—even though
being a universal device to integrate both foreign and native lexical forms into
verb complexes—has emerged in the region at issue from Persian or even Ira-
nian patterns (see Karimi-Doostan 1997, 2001, 2005, Fritz 2009). In Udi, light verb
constructions have become the standard way of forming ‘new’ verbs: Today,
more than 75% of all Udi verbs are compounds based on light verbs (Schulze (in
press b)). This high frequency of light verb constructions is compatible with that
of Modern Persian, cf. table (1) that lists the most relevant light verbs of Persian
together with their correspondences (if given) in Udi. ‘Percentage’ relates to the
number of lexical units under consideration in each language:
Persian light verbs Udi light verbs
% (n = 1140) % (n = 1484)
DO 41,75 kardan 26,28 besun
BECOME 11,92 šodan 22,23 baksun
GIVE 6,66 dādan 12,33 -desun
HAVE 5,7 dāštan ---
HIT 4,73 zadan ---
TAKE 3,77 gereftan 1,9 biq’sun/aq’sun
BE 3,15 budan ---
BRING 2,1 āvardan 0,32 ečesun
EAT 2,1 xordan ---

11
I use the term ‘Orientalism’ to denote lexical and grammatical units that are present in two
or more of the three major ‘Oriental’ languages Arabic, Persian, and Western Oghuz-Turkish,
whereby one of these languages served as the donor language. Northern Oriental would encom-
pass Turkish/Persian commonalities, whereas Southern Oriental would include Arabic/Persian
parallels.
 Aspects of Udi-Iranian Language Contact 387

GO 2 raftan 0,51 taysun

COME 1,66 āmadan 40,70 e(y)sun12
DRAW 1,57 kešidan 0,30 Zapsun
PUT 1,4 Gozāštan 0,22 Ladesun
SAY ---- 8,82 Pesun
Table (1): Types and frequencies of light verb constructions in Persian and Udi

Data from CA illustrate that the Udi light verb constructions have started
from analoguous patterns in Early Udi /CA, cf. the following lights in CA:
ak’esown ‘see’, ap’esown ‘reach’, batkesown ‘turn’, biyesown ‘do’, biq’esown ‘take’,
bowresown ‘stand’, efesown ‘hold, have’, (e)pesown ‘speak’, heq’esown ‘take’, ihe-
sown ‘(be)come’. However, the dominance of this pattern in Modern Udi is un-
doubtedly conditioned by impact from Persian.
Prior to the times of Modern Persian influence, both local Northwestern Ira-
nian languages and ‘official (Arsacid) Parthian’ as well as Sassanid Middle Per-
sian seem to have contributed to the formation of both Udi and Caucasian Alba-
nian. In section 1 of this paper, I have already given some examples for Iranisms
in Caucasian Albanian. Some of the lexical forms have survived in Udi, such as
afre-pesun ‘to pray’, durut’ ‘beam’ etc. Although the Udi lexicon has not been
comprehensively monitored with respect to older Iranian loan layers, some
items clearly suggest that this layer must have been present in the language. Ex-
amples are for instance:
Udi aˁm ‘arm’ (‘hands’ excluded). It also means ‘pole, thills, door wing, assistant
(‘right hand’)’. Most probably, the word is related to Avesta arəma- ‘arm’ or an-
other Iranian reflex of PIE *H2ṛH-mó- (*’əṛm). A Northwest Iranian reflex such
as Northern Tolyshī am ‘shoulders’ should, however, be excluded out of se-
mantic reasons (the same is true for Armenian armukn ‘elbow’).
Udi kala ‘big; large; old’. The terms must have been borrowed from Northern Tāti
kala ~ kälä ‘large, big’.
Udi aper ‘father (honorific)’. The term has its best match in Mazandaran piyer ‘fa-
ther’ (also compare Northern Tolyshī piya), even though the initial a- remains
obscure.
Udi nana ‘mother’, cf. Tātī nana, Tolyshī nana (as opposed to Persian nāne).
Udi borǯ ‘work, duty, affair’ (CA borz-pesown ‘to be zealous’ (cp. MIran. warz
‘work’).

12
The high frequency of e(y)sun-constructions is due to the fact that this verb has developed
into a passive resp. mediopassive marker, see Schulze (in press c).
388 Wolfgang Schulze

Udi šad ‘opened, free’ (in CA -šad/t compound member corresponding to Arm. -
šat (< MIran. šāt / šād) meaning ‘glad, happy’ or ‘enough, much’.

4. UDI AND IRANIAN: MORPHOSYNTAX

Apart from lexical issues, the Iranian impact on Udi can also be morphosyntax.
Basically, two domains can be addressed: (a) The development of the so-called
‘differential object marking’ (DOM, see Bossong (1985, 1991)) strategy in both Udi
and Caucasian Albanian, and (b) the development of floating agreement mark-
ers.
In this paper, I do not want to dwell in details upon point (a). This is due to
the fact that we cannot exclude the possibility that this strategy has emerged
also from contact with Old Armenian and that it has been reinforced later-on by
the strategy of definitive object marking in Oghuz Turkic. Nevertheless, that fact
that both CA and Udi make use of a dative/directive case (-V(x)) suggests that at
least the morphosemantic realization of the DOM strategy is affine to the adja-
cent Iranian languages, compare the exmaples in (1):
(1) Old Armenian Preposition z-
Persian Postposition > Dative -rā
Tolyshī Oblique marker -i
Balochi Dative -ā < -rā
Bactrian Preposition av
Azeri Accusative Case Marker -I
Udi / CA Dative(Allative) -Vx

Admittedly, we have to bear in mind the fact that “[b]road typological com-
parison reveals that the DAT marker is by far the most important single source
for newly developed ACC' markers” (Bossong 1991: 157, also see Bossong 1985: 109-
121). Hence, we cannot exclude anindependent development of the DOM
strategy in Udi and CA. However, given the fact that the Udi/CA pattern itself is
unique within the world of East Caucasian, we can safely include both languages
in the areal pattern of DOM. The following examples illustrate DOM in Udi (2)
and CA (3):13
(2) adamar-en śum uk-al-le
man-ERG bread eat-FUT-3SG
‘The man will eat bread.’ (Udi, Vartashen)

13
It should be born in mind that both Udi and CA are dominated by an ergative case pattern.
 Aspects of Udi-Iranian Language Contact 389

adamar-en śum-ax uk-al-le

man-ERG bread-DAT2 eat-FUT-3SG
‘The man will eat the bread (you know/we’ve been talking about’). (Udi,
Vartashen))
(3) ič e xowˤ-n-owl hala biya-y-q’a-zow
REFL DEF.N rock-SA-SUPERESS1 on make:PAST-PAST-HORT-1SG
ek’lesi bezi
church my
‘On this very rock I shall build my church. (CA, Mt 16,18)
k˜s-en bowq’ana-biya-y hačink’e ek’lesi-ax ič-ey
Christ-ERG beloved.DO.PAST-PAST how church-DAT2 REFL-GEN
‘... (just) as Christ loved his church.’ (CA, Eph. 5,25)

With respect to Iranian, Bossong (1985: 9) wrongly describes DOM strategies

as a feature that—he claims—would be an exclusive feature of New Iranian lan-
guages. In fact, it is clearly present in Bactrian (see Sims-Williams (2011) for the
corresponding data) and may also be present in yet not fully surveyed instances
of the use of adpositions with noun phrases in objective function in e.g. Middle
Persian and Parthian (see Paul 2003, Haig 2008). As Sims-Williams (2011: 27) has
put it: “DOM is not an exclusively New Iranian development, but is already sol-
idly attested in at least one Middle Iranian language”. The functional dimension
of DOM in Udi is related to features of individuation, typicity, and definiteness.
This pattern comes close to what is conventionally described for DOM in the
Iranian languages, although it lacks the feature of animateness that may become
relevant in these languages (see Bossong 1985). DOM in CA, however, is less pre-
dictable (and not yet fully monitored). It does not match at full the ab-
sence/presence of the nota accusative z- in the corresponding passages from Old
Armenian and seems to be controlled by features different from that in Old Ar-
menian. Hence, the development of DOM conditions from CA to Udi comes
close to what we can observe for the corresponding processes from Middle Ira-
nian to Modern Iranian: In Middle Iranian times (roughly corresponding to CA
times) DOM gradually becomes stabilized. Starting with Early New Iranian,
DOM has developed into a more or less regularized strategy, matching the pe-
riod of Early Modern Udi. From this we can infer that DOM in Udi may have at
least become stabilized through impact from Early Modern Iranian languages.
There is, however, a major difference between DOM in many Northwest Ira-
nian languages and Udi: In these languages (as in Bactrian), DOM is usually con-
390 Wolfgang Schulze

fined to accusative alignment in imperfective (or: present-tense based) transitive

clauses, whereas the ergative pattern in perfective (or: past-tense based) transi-
tive clauses does not allow DOM, cf. the following phrases form Kurmancî
Kurdish:
(4) ez hesp-î di-bîn-im
I:NOM horse-ACC:DEF:M DUR-see:PRES-1SG
‘I see the horse.’ (Bedir Khan/Lescot 1986: 86)
şivên ez dît-im
shepherd:OBL:M I:NOM see:PAST-1SG:O
‘The shepherd saw me.’ (ibid.: 153)

Both CA and Udi, however, are controlled by an ergative case pattern. Look-
ing at case alone would give us a controversial picture: DOM is disallowed with
standard ergative patterns in Northwest Iranian, but allowed in CA and Udi, cf.
(5):
(5) še-t’-in eˁk-n-ux beˁ-ne-ğ-sa
DIST-SA.OBL-ERG horse-SA-DAT2 see-3SG-$-PRES
‘(S)he sees the horse.’ (Udi, Vartashen, field notes)
eˁk-en šo-t’-ux beˁ-ne-ğ-sa
horse-ERG DIST-SA.OBL-DAT2 see-3SG-$-PRES
‘The horse looks at him/her.’ (Udi, Vartashen, field notes)

However, blocking of DOM with ergative constructions in Iranian is strongly

coupled with agreement patterns that mark the NP in objective function on the
verb, but not the agentive NP, confer again example (4) above. Sims-Williams
(2011: 29-30), however, shows that at least in Bactrian, DOM may also occur with
ergative patterns, maintaining O-agreement, cf.:
(6) man yōbig av-āfag vrāmarz… āxasād-ēi
I.OBL Yobig PREP-you.SG Vramarz… fight.PAST-2SG
‘I, Yobig, fought you, Vramarz.’ (Bactrian, Sims-Williams 2011: 29)
In Northern Tolyshī, the extreme weakening of O-agreement conditions that
such a DOM pattern may go together with the rearrangement of agreement:
Now, it is the oblique (ergative) marking agentive that represents the major trig-
ger for agreement, not the objective. Sims-Williams (2011: 29) shows that this re-
arrangement may have been incidentally present as early as in Bactrian. An ex-
ample from Northern Tolyshī is:
 Aspects of Udi-Iranian Language Contact 391

(7) palang-i mə-nə-š gat-a

leopard-OBL I-ACC-3SG take-PERF
‘The leopard has taken me.’ (Northern Tolyshī, PA 44)

We have no evidence that O-agreement (usually carried out in East Cauca-

sian with the help of noun class markers, cf. Schulze 1988, 2000b) had been pre-
sent in Early CA/Udi. The corresponding strategy must have been lost in the con-
text of Early Aghul and Early Lezgi, both languages that lack agreement in gen-
eral. Graphic (6) illustrates the distribution of agreement patterns in Eastern
Samur:

Graphic 6: Agreement patterns in Eastern Samur

Accordingly, ‘Language X’ (the ancestor of CA and Early Udi) must have been
marked for having innovated the agreement pattern based on a S/A-trigger.14
Hence, it is not necessary to relate the CA/Udi pattern of DOM to younger stages
of Northwest Iranian that would allow DOM in ergative strategies. Rather, we
can assume that the ergative pattern of case marking in CA/Udi was less relevant
than the presence of the S/A-agreement strategy that relates the CA/Udi align-
ment pattern to the DOM-sensitive, imperfective patterns of Early Northwest
Iranian.
The second dimension is that of ‘floating agreement clitics’. Typically, East
Caucasian languages lack strategies of person marking on the verb. Some rele-
vant exceptions are Bats (Nakh), Lak/Dargi, Akhvakh (Andian), Tabasaran
(Lezgian), and CA/Udi.15 As has been said above, in both CA and Udi, the inser-

14
S/A-agreement in Tabasaran is an independent innovation that has probably been motivated
by contact with Dargi located in the north of the Tabasaran area, see Schulze (2011b) for details.
15
Personal agreement patterns can be found in other East Caucasian languages, too. However,
392 Wolfgang Schulze

tion of agreement clitics is based on an accusative strategy thus competing with

the ergative case pattern, cf. Udi:
(8) adamar burğ-ol cur-re-sa
man.ABS mountain-SUPERESS stand-3SG-LV.PRES
‘The man stands on a mountain.’
adamar-en burux beˁ-ne-ğ-sa
man-ERG mountain.ABS see-3SG-$-PRES
‘The man sees a mountain.’

The relevant patterns can be illustrated with the help of the correlations
given in table (2):16

Case Agreement Word order

S ABS yes First
A ERG yes First
O ABS/DAT no Second
Pattern Ergative (+ O-Split) Accusative Accusative
Table 2: The coding of basic grammatical relations in CA/Udi

Both in CA and Udi, the agreement markers clearly are clitics (except for cer-
tain conditions, see Harris 2002 for details). Here, I define clitics as atonic mor-
phemes that fuse with clausal constituents without being restricted to a specific
word class. Affixes differ from clitics in that they form paradigmatic units with
specific word classes. In Caucasian Albanian, agreement clitics are strictly verb-
oriented: The strong correlation between verbs and agreement clitics permits to
set up the category of finite verb forms for descriptive purposes (see Gippert et
al.2009: II,52-54 for details). An example is given in (5):
(9) bite-y-ne žak’-owr kowl-m-oxoc o-ya
fall:PAST-PAST-3 chain-PL.ABS hand-PL-ABL2 he-GEN
‘The chains fell (off) from his hands.’ (CA, Act. 12,7)

Nevertheless, verb-external hosts occasionally occur with agreement clitics in

CA, too. One of the rare examples in given in (6):

they are usually restricted patterns highlighting especially the first person singular. More recent
develop of agreement markers concern Aghul and Kryts, see Schulze (2011b) for details. Schulze
(2011c) discusses in details the morphological history of CA/Udi agreement clitics.
16
I use {S,A,O} to indicate the basic grammatical relations (S= intransitive agent, A = transitive
agent, O = transitive ‘goal‘). O-Split is another label for DOM, see above.
 Aspects of Udi-Iranian Language Contact 393

(10) zow-zow-ne pe-y b˜ġ abraham-i

I-1SG-3 say:PAST-PAST God Abraham-GEN
‘Me, he said, I am the God of Abraham.’ (CA, Mt. 22,32)
In (10), the agreement clitic is added to another clitic that functions as a first
person singular copula. Even though the number of examples given in the cor-
pus of CA texts is rather limited, we can nevertheless assume that they represent
first (and still marginal) instances of so-called floating clitics. A clitics is said to
be floating if it is not related to a specific type of host or position in the clause
(such as Wackernagel’s Position etc.). It can be thus hosted by lexical forms be-
longing to different word classes or exerting different functions. The functional
scope of a floating clitic is defined by the interaction of the host or functional
head of the clitic (constituent, clause, or pragmatic reference) and the function
exerted by the constituent to which the clitic is added (see Cysouw (2005) for the
general characteristics of such clitics). The pattern of floating agreement clitics is
typical for Udi (see Harris 2002 and Schulze 2011c) for details. The following
examples illustrate the use of agreement clitics with different types of hosts:
(11) Post-verbal:
t’e ği-mx-o K’esarev-i avgust’-axo
DIST day-PL-DAT1 emperor-GEN Augustus-ABL
č’er-i-ne buiruğ
go=out.PAST-PAST-3SG order
‘In those days of the emperor the order came out from Augustus’ (Udi, Lk
2,1)
Endoclitic (LEX-LV):
bixoğ-o färišt’i-n-en buiruğ-ne-b-i še-t’-u
God-GEN angel-SA-ERG order-3SG-DO-PAST DIST-OBL-DAT
‘God’ angel gave order to him’ (Udi. Mt 1,24)
Endoclitic (verb-internal):17
va ta-q´un-c-i bütün
and come-3PL-$.PAST-PAST all
‘And all came…’ (Udi, Lk 2,3)
Verb-external (noun, objective function)
pasčağ-en gädi-n-a sa eˁk-ne tast’a
king-ERG boy-SA-DAT one horse-3SG give.PRES
‘The king gives the boy a horse.’ (Udi (Vartashen, field notes).

17
The sign “$“ symbolizes the second part of a discontinuous lexeme.
394 Wolfgang Schulze

Verb-external (demonstrative):
bez kalba kalna t’iya-ne bak-sa.
my grandfather grandmother there-3SG be-pres
‘My grandfather and grandmother are there.’ (Udi (Nij), field notes).
Verb-external (noun, agentive function):
düšman adamar-en-ne b-e mo-t’-ux
hostile man-ERG-3SG make-PERF PROX-SA-DAT2
‘An enemy has done this.’ (Udi, Mt 13:28)

If we compare the overall frequency of verb-external and postverbal clitics in

the Nij and Vartashen dialects of Udi (1279 clitics for Nij, 1201 clitics for
Vartashen), the following picture emerges, see graphic (6):18

Graphic 6: Preverbal and postverbal clitics in Udi dialects

Obviously, Nij has developed a pronounced tendency towards preverbal clit-

ics (note, however, that the data in diagram 5 do not distinguish between clitics
in optional position and clitics that become suffixes after certain tense/aspect/
mood forms, see above). Still, the drift away from the postverbal position is pre-
sent already in the Vartashen dialect.
We can assume that this drift towards verb-external hosts has been moti-
vated by a Northwest Iranian language. It has its best parallels in Northern
Tolyshī (see Schulze (2000a) for details).19 Compare the following phrase:
(12) palang-i merd ba više kəšt-əš-e
tiger-OBL man in wood kill:PAST-3SG-AUX:3SG
‘In the wood, the tiger killed the/a man.’ (Northern Tolyshī, field notes)

18
Data are taken from Schulze (2011c).
19
Paul (2011) entails the discussion of corresponding strategies in Iranian Tolyshī.
 Aspects of Udi-Iranian Language Contact 395

This phrase allows the following variant that differ with respect to focus:
(13) a. palang-i merd-əš ba više kəšt-e
b. palang-i merd ba više-š kəšt-e
c. ?palang-i-š merd ba više kəšt-e

According to my informants, version (9c) is hardly acceptable, even though

at least one informant admitted that he might have heard it somewhere. In
Northern Tolyshī, floating agreement clitics are only allowed with perfective
transitive structures. This is due the origins of these clitics that stem from older
possessive markers (see Schulze (2000a)). Just as it is true for Udi, there is hardly
any constraint on the choice of preverbal host. (10) lists some examples:20
(14) Verb-external (noun in objective function):
mə tifang-əm tamiz kā
I:OBL rifle-1SG clean make.PAST.PERF
‘I cleaned my rifle.’ (PA 1)
Verb-external (pronoun in objective function):
palang-i-an (...) av-əš gat-e
leopard-OBL-FOC (...) he-3SG take.PAST-AOR.3SG
‘The leopard (...) took him.’ (PA 43)
Verb-external (LEX-LV):
ay ba čəmə riz nišon-əš do-e
he:OBL to I:pOSS trace sign-3SG give:PAST-AOR:3SG
‘He showed me the trace.’ (Field notes)
Verb-external (noun in locative phrase):
ba katto-ž šekayat kārd-e
to chairman=of=parish=council-3SG complaint make.PAST-AOR.3SG
‘(S)he complained to the chairman of the parish council.’ (Miller 1953: 168)
Verb-external (possessor):
čəmə-š glai müaxol bəri-e
I:POSS-3SG one plait cut.PAST-AOR.3SG
‘(S)he cut off one of my plaits.’ (Miller 1953: 168)
Verb-external (adverb):
albahal-əm tifang ba po pekərn-i
just=now-1SG rifle to down take=up.PAST-AOR.3SG
‘In this moment, I took up the rifle from below.’ (PA 56)

20
The abbreviation “PA” refers to the text palangi ahvolot given in Schulze (2000a).
396 Wolfgang Schulze

The original anaphoric function of these clitics (usually called set II or pos-
sessive clitics) has been reinterpreted as focus function because of the fact that
they were triggered by a prosodically marked argument focus: Set I suffixes (the
original person markers) did not participate in this functional shift because they
lacked an anaphoric function at least by the time the floating of clitics came into
use. In consequence, morphological focus marking became restricted to those
.

TAM forms that were based on residues of (oblique) anaphoric personal pro-
nouns. This restriction perhaps was strengthened by the fact that morphological
focus marking is more likely to appear with past tense/aspect structures than
with present/future tense or with those TAM forms that convey background in-
formation (past imperfectives). Present/future tense forms mainly appear in di-
alogues or in direct quotes. Here, focus structures are much easier to infer from
presuppositions or co(n)textual information. Hence, if a language system knows
a system of intonation focus, an additional morphological marking of argument
focus provokes some sort of redundant identification of focus that carries a high
pragmatic value. The intonation pattern of Tolyshī obviously has a rather dis-
tinctive function (which is clearly audible). In consequence, there is no func-
tional need to extend the technique of morphological focusing based on floating
set II clitics to present/future tense forms via analogy. Rather, the tendency in
Tolyshī seems to be the other way round: Set II clitics tend to be dropped if the A
trigger is overtly represented by a noun or (rarer) a pronoun. This is true espe-
cially for the pluperfect, but gains more and more ground even with the standard
domain of set II clitics, namely the perfect tense and the aorist:
(15) ğaz-i glai ğaz s-a
qadi-OBL one goose receive:PERF-PERF
‘The qadi (judge) bought (lit. received) a goose.’ (Miller 1953: 172)
ay hukumat kārd-a yo
he.OBL reign make.PAST-PERF here
‘Here, he reigns.’ (Miller 1953: 172)
mə ba tə čič kārd-a be?
I:OBL to you.SG what do.PAST-PERF AUX.PAST:3SG
‘What had I done to you?’ (Miller 1953: 174)

Functionally, floating clitics serve the same purpose as in Udi. In Udi, too, the
choice of host is strongly determined by features of constituent focus, although
constraints apply with certain tense/mood forms: In this case (-al-future, -a-
 Aspects of Udi-Iranian Language Contact 397

modal, ay(i)-conditional), the clitic has to follow the verb (see Harris 2002 for de-
tails). However, the underlying syntactic pattern differs from that of Northwest
Iranian: In Northwest Iranian, the floating ‘ergative’ agreement clitics stem from
oblique (probably possessive) pronouns, copying the oblique (dative/passive)
case function of the agentive noun, cf. table (3) that shows the history of the cor-
responding clitics of Northern Tolyshī:
Old Iranian / Possessive Early Tolyshī Northern Tolyshī
1SG *-maiy *-mai > *-mə -m
2SG *-taiy *-tai > *-(t)ə -ə
3SG *-šaiy *-šai > *-šə -š
1PL [Avesta -nə̄] *-m(ə)-ān -m-on
2PL [Avesta -bə̄/ -vō] *-(t)ə-ān -(ə)-on
3PL [Old Persian -šam] *-šə-ān -š-on
Table 3: The history of ergative agreement clitics in Northern Tolyshī

The general pattern must have be something like (16):

(16) *man-POSS>OBL dog-NOM-ANAPH.POSS>OBL see.PPP-COP-AGRO.
‘Of the man, dog is seen, of him.’

This pattern that is still recognizable in northern Tolyshī is structurally dif-

ferent from that in Udi. In Udi, agreement clitics stem from the merger of two
different paradigms: (a) the paradigm of pronominal echoing (esp. first person,
less transparent in the second person), (b) a paradigm of person-neutral focus
markers (here *-ni, see Schulze 2011c for details). Given the fact that Udi has an
ergative/accusative person split in the sense of Silverstein (1976), speech act par-
ticipants thus do not entail a specific ergative case form. Accordingly, those
agreement clitics that stem from personal pronouns (first person, second person
with modifications) simply echo the agent based on an accusative pattern. The
third person maintains the absolutive/ergative distinction. Hence, pronouns are
not used, but a simple, case-neutral focus marker, cf. table (4) that summarizes
the history of the Udi clitics:

Source Stage 1 Stage 2 CA Udi

1Sg *zu (Pro) *-zu -zu -zow -z(u)
2Sg *-ni (Foc)
*-ni *ni + -wun -nown -n(u)~ -(u)n
*wun (Udi: un) (Pro)
3Sg *-ni (Foc) *-ni *-ni -ne -n(e) ~ -e
398 Wolfgang Schulze

1Pl *-žan (Pro) *-žan *žan -žan -yan

2Pl *ni (Pro)
*-ni *-ni * waˁn -nan -nan
*wˁan (Udi vaˁn / efaˁn) (Pro)
3Pl (Udi: *-t’ğu- (part. Plural)
*-ni *-ni -ne *-tğu-ne > -t’un -/q’un
*-ni Foc)
Table 4: The history of agreement clitics in CA and Udi

(17) illustrates the Udi model with the help of Modern Udi data:
(17) zu śum-zu uk-sa
I.ABS bread.ABS-I.ABS eat-PRES
‘I, I eat bread.’
še-t’-in śum-ne uk-sa
DIST-SA.OBL-ERG bread.ABS-FOC eat-PRES
‘(S)he eats bread (you know).’

It comes clear that the Udi pattern of verb-external agreement hosting stems
from strategies different from those in Iranian. Accordingly, we cannot assume
that the technique of applying floating agreement clitics itself has been bor-
rowed from a late Medieval Northwest Iranian language into Udi. Rather, we
have to deal with some kind of functional borrowing that dwells upon given
strategies of agreement marking and upon the corresponding native material.

5. CONCLUDING REMARKS
In my paper, I have argued that both Caucasian Albanian and Udi are marked
for a rather pronounced impact from languages that must have played a domi-
nant role in Classical antiquity and in Medieval times. Here, at least three layers
can easily be isolated: (a) Old Armenian, (b) both Northwest Iranian and South-
west Iranian, (c) Turkic in terms of Oghuz Turkic (Azeri). Both (Early) Modern
Persian and Azeri conditioned that Udi was later-on at least partly integrated
into the world of Oriental language despite of the fact that its speakers remained
Christians. The pre-Oriental impact from Armenian and Iranian languages re-
sulted in significant shifts with respect to both the lexicon and the grammar of
CA/Udi that set both languages apart from the world of Lezgian languages. Given
the fact that Udi is not the immediate descendant of Caucasian Albanian
sharing, however, important innovations with CA suggests that these shifts must
have either started in CA times affecting in equal way over both CA and ‘Early
Udi’, or that they have commenced already by the time of the common ancestor
 Aspects of Udi-Iranian Language Contact 399

of CA and Early Udi (‘Language X’). If the latter is the case, Old and Early Middle
Iranian impact should predate the impact from Armenian. For the time being,
this question must remain unanswered. It has to be born in mind that nearly
forty percent of the ‘native’ Udi lexicon and much of the corresponding CA lexi-
con21 does not have recognizable East Caucasian or Lezgian etymologies. We can
hence expect that this part of the lexicon entails another loan layer not yet iden-
tified.22 Whether this is grounded in an Old Iranian language or in a language to-
tally unknown can only be told (if ever) once the layer has been surveyed in
more details. Maybe that a thorough analysis of the lexicon of other East Cauca-
sian languages in the Azerbaijan regions such as Khinalug, Kryts, and Budukh
will help to identify this layer. One of these layers perhaps goes back to late Indo-
European times. The following words suggest contact with speakers of a non-Sa-
temizing Indo-European language:
(18) Udi eˁk ‘horse’ ~ IE *ek`wo- ‘horse’
Udi boˁq’ ‘pig’ (< *borq’) ~ IE *porkho- ‘pig’
Udi fi ‘wine’ (OBL fin-) ~ IE (?) *u̯ oino- ‘wine’ [not fr. Arm. gini, Geo. γvino]
Udi ul ‘wolf’ (CA owl) ~ IE *u̯ l̯ku̯ os

‘Horse’-words are borrowed terms in all East Caucasian languages (there is no

common Proto-East Caucasian term for this animal), compare Lezgi balk’an, Ru-
tul balkan, Kryts barkan ~ balkän (Turkic), Tabasaran and Aghul ħaˁywan,
Tsakhur hiywan (Arabic), Khinalug pši (Iranian). Only Archi noˁš and Budukh
xila as yet lack a secured etymology. The same foreign impact also holds for
other East Caucasian languages, compare Avar ču and Lak čwu, perhaps taken
from Georgian ačua ‘horse’ (preserved in child language, obviously stemming
from a satemized variant of IE *ek`wo- ‘horse’), Karata ħwane, Akhwakh xwani,
Ghodoberi x:wani (all of them Arabic) etc. Hence, it is rather likely that Udi eˁk
has been borrowed, too. However note that at least boˁq’ ‘pig’ and ul ‘wolf’ may
likewise have Lezgian correlates, compare (for ‘pig’) Lezgi wak (obl. wak’-), Aghul
wak’, Rutul wok (only in yakdə wok ‘pork meat’), Tsakhur wok, Kryts wok, Budukh
wak, Archi boˁλ:’ < PL *bwerλ:’ ?); for ‘wolf’ we have Rutul ubul, Tsakhur umul,
Kryts eb, Archi yam, which may go back to PL *(y)əbw-ul > Early Udi *uwul (?).

21
By ‘native‘ I mean that layer of the Udi/CA lexicon that does not represent discernible bor-
rowings from Armenian, Persian, Arabic, or Iranian.
22
A typical Udi term that seems to represent this layer is bin ‘daughter-in-law, bride’. Most im-
portantly, Udi did not borrow the corresponding term from Iranian, contrary to nearly nearly all
East Caucasian languages and beyond (see Schulze 1999, Tuite and Schulze 1998).
400 Wolfgang Schulze

BIBLIOGRAPHY
Asatrian, G. (2009), “Prolegomena to the Study of the Kurds”, Iran and Caucasus 13: 1-57.
Bais, M. (2001), Albania caucasica. Ethnos, storia, territorio attraverso le fonti greche, latine e armene,
Milano.
Bedir Khan Emir Djeladet / R. Lescot (1986) Kurdische Grammatik, Kurmancî-Dialekt, Bonn.
Bezhanov, M. (1892), “Kratkie svedenija o sele Vartašen i ego žiteljax”, SMOMPK 14.
Bossong, G. (1985), Empirische Universalienforschung, Differentielle Objektmarkierung in den neuira-
nischen Sprachen. Tübingen.
―― (1991) “Differential Object Marking in Romance and beyond”, New Analyses in Romance Lin-
guistics, eds. D. Kibbee; D. Wanner, Amsterdam/Philadelphia: 143-170
Cysouw, M. (2005), “Morphology in the wrong place, A survey of preposed enclitics”, Morphology
and its Demarcations, eds. W. U. Dressler; D. Kastovsky; O. E. Pfeiffer; F. Rainer, Amster-
dam/Philadelphia: 17-38..
Edens, Ch. (1995), “Transcaucasia at the End of the Early Bronze Age”, Bulletin of the American
Schools of Oriental Research 299/300: 53-64.
Fritz, S. (2009), “Die ‘komplexen’ Verben des Neupersischen in Umgangssprache und Literatur –
eine areallinguistische und kulturhistorische Betrachtung”, Iranistik, Deutschsprachige
Zeitschrift für iranistische Studien 11: 5-55.
Gippert, J. / W. Schulze (2007) “Some remarks on the Caucasian Albanian Palimspests”, Iran and
the Caucasus 2007, 2: 201-211.
―― / Z. Aleksidze / J.-P. Mahé (2009), The Caucasian Albanian Palimpsests of Mt. Sinai, 2 Vols,
Turnhout.
Haig, G. (2008), Alignment change in Iranian languages: a construction grammar approach, Berlin.
Harris, A. (2002), Endoclitics and the Origins of Udi Morphosyntax, Oxford.
Huyse, Ph. (19990, Die dreisprachige Inschrift Šābuhrs I. an der Kaʽba-i Zardušt, Bd. 1 (Corpus In-
scriptionum Iranicarum, III/I/1), London.
Karimi-Doostan, Gh. (1997), Light Verb Constructions in Persian. Ph.D. Diss., University of Essex.
―― (2001), “N + V complex predicates in Persian”, Structural aspects of semantically complex verbs,
eds. N. Dehé; A. Wanner, Frankfurt am Main: 277-292.
―― (2005), “Light verbs and structural case”, Lingua 115: 1737-1756.
Mallory, J. P. (1997), “Kuro-Araxes Culture”, Encyclopedia of Indo-European Culture, eds. J. P. Mal-
lory; Q. D. Adams, London: 341-342.
Miller, B. V. (1953), Talyšskij jazyk, Moskva.
Paul, D. (2011), A comparative dialectal description of Iranian Taleshi. PhD-thesis University of Man-
chester, Faculty of Humanities.
Paul, L. (2003), “Early Judaeo-Persian in a Historical Perspective: The Case of the Prepositions be, u,
pa(d), and the suffix -rā”, Persian Origins – Early Judaeo-Perisan and the emergence of New
Persian, ed. L. Paul, Wiesbaden: 177-194.
Schulze, W. (1988), “Noun Classification and Ergative Construction in East Caucasian Languages”,
Studia Caucasologica 1: 251-274.
 Aspects of Udi-Iranian Language Contact 401

―― (1999), “Affinal Kin Terms in the East Caucasus: Inheritance vs. Borrowing?”, Diachronica
XVI.1: 97-122.
―― (2000a), Northern Talysh, Munich/Newcastle.
―― (2000b), “Towards a Typology of the Accusative Ergative Continuum: The case of East Cauca-
sian”, General Linguistics 37: 71-155.
―― (2008), “Towards a History of Khinalug”, Chomolangma, Demavend und Kasbek. Festschrift für
Roland Bielmeier zu seinem 65. Geburtstag, 2 Bände, eds. B. Huber; M. Volkart; P. Widmer,
Halle (Saale): 703-744.
―― (2011a), “A brief note on Udi-Armenian relations”, Cultural, Linguistic and Ethnological Inter-
relations In and Around Armenia, eds. J. Dum-Tragut; U. Bläsing, Newcastle upon Tyne: 151-
170.
―― (2011b), Personalität in den ostkaukasischen Sprachen, Banská Bystrica.
―― (2011c), “The Origins of Personal Agreement Clitics in Caucasian Albanian and Udi”, Folia
Caucasica: Festschrift für Jost Gippert zum 55. Geburtstag, eds. M. Tandashvili; Z. Pourtskh-
vanidze, Frankfurt am Main/Tbilisi: 119-168.
―― (in press a), “Word formation in Khinalug”, HSK Word-Formation - An International Handbook
of the Languages of Europe, eds. P. O. Müller; I. Ohnheiser; S. Olsen; F. Rainer, Berlin.
―― (in press b), “Word formation in Udi”, HSK Word-Formation - An International Handbook of
the Languages of Europe, eds. P. O. Müller; I. Ohnheiser; S. Olsen; F. Rainer, Berlin.
―― (in press c), “The emergence of diathesis markers from MOTION concepts”, Metaphor and
metonymy through time and cultures, ed. Javier E. Díaz Vera, Berlin.
Silverstein, M. (1976), “Hierarchy of features and ergativity”, Grammatical categories in Australian
languages, ed. Robert M. W. Dixon, New Jersey: 112-171.
Sims-Williams, N. (2011), “Differential Object Marking in Bactrian”, Topics in Iranian Linguistics,
eds. A. Korn; G. Haig; S. Karimi; P. Samvelian, Wiesbaden: 23-38.
Tuite, K. / W. Schulze (1998), “A Case of Taboo-Motivated Lexical Replacement in the Indigenous
Languages of the Caucasus” Anthropological Linguistics 40: 363-383.
 Armenian varkaparazi and its Iranian Background

Martin Schwartz
University of California, Berkeley

Abstract
The paper discusses critically the etymology of classical Armenian varkaparazi (“at random, by
chance, in a disorderly manner”) proposed by the late Anahit Périkhanian in 2002, and offer an al-
ternative etymology. It is thanks to Dr. Périkhanian’s calling attention to varkaparazi, and her pro-
viding a partial starting direction toward its etymology that, albeit with the rejection of her overall
interpretation, we have gained a lost Iranian compound whose members are of noteworthy inter-
est for the history of the Iranian vocabulary.

Keywords
Armenian Etymology, Iranian Etymology

It is a pleasure to offer this tribute, albeit humble, to my dear friend and colleague
Garnik Asatrian. I shall here discuss critically the etymology of Armenian varka-
parazi proposed by the late Anahit Périkhanian (Périkhanian 2002), and offer an
alternative etymology.
As Périkhanian detailed, varkaparazi occurs in translations of texts by diverse
Greek authors, and in two original Armenian texts, in a range of meanings: “at
random, by chance, uncontrolledly, willy-nilly, in a disorderly manner”. She ex-
plained varkaparazi from Middle Iranian (Parthian or Middle Persian) *varkapa-
raz, representing Old Iranian *vr̥ ka- ‘wolf’ + *aparazah- ‘fleeing, saving oneself by
escape’, cf. Christian Sogdian pryž-, Yaghnobi apirež ‘fled’, past stem Ancient
Letters Sogd. ’pr’št, and without *apa-, Manichean Middle Persian and Pers. rah-
(past stem rast- = Pahlavi Psalter lsty). Thereby, for Périkhanian, varkaparazi is
metaphoric from *‘dont la fuite/l’évasion est comme celle des loups’.
As a parallel to the original meaning of varkaparazi, she cites R̥ gVeda 2.30.4
vr̥ ́ kadvaras-, with interpretation of -dvaras- via the Iranian root dvar attested in
Avestan and Pahlavi for the movement of creatures of Ahriman. Périkhanian
viewed this specification in meaning as due to Zoroaster’s reform, before which,
 Koninklijke Brill NV, Leiden, 2015
404 Martin Schwartz

she asserted, the Indo-Iranian sense of the verb referred to swift rectilinear mo-
tion, as preserved in the Vedic compound: “Primitivement, en iranien aussi, il
désignait un mouvement rapide en ligne directe, sans qui apparaît dans le com-
posé védique dont le premier membre, vr̥ ka-, est identique à celui de iran. *vr̥ ka-
(apa-)razah-, et le second, un ancien neutre en -as- de indo-iran. *dvar-”. Vedic
vr̥ ́ kadvaras- would thus mean ‘whose impetuous course is that of wolves on the
attack; running with the impetuousness of wolves as they attack’.
Périkhanian accordingly posited two morphologically parallel but semanti-
cally opposed compounds involving the running of wolves: the aforementioned
vr̥ ́ kadvaras-, referring to wolves running to attack, and Old Iranian *vr̥ ka-apara-
zah-, referring to wolves rushing to escape, of which the latter compound yields
Arm. varkaparazi, used metaphorically.
A minor criticism of Périkhanian’s analysis: there is no reason to assume a -h-
stem *aparazah- just because vr̥ ́ kadvaras-, in Périkhanian’s opinion, parallels
such compounds as Vedic vāta-raṃhas- ‘quick as wind’, ū́ rṇa-mradas- ‘soft as
wool’, Av. gairi.masah- ‘big as a mountain’, and Av. zəm.fraϑah- ‘wide as the
earth’. Nouns of any shape could serve as second member of such semantically
functioning compounds. In the instance of the root raz (with preverb apa-), one
may at least as likely get a thematic formation *(apa)rāza- as *(apa)razah- for
the second member of the compound in question, cf. below on MIran. rāz vis-à-
vis Vedic ráhas- ‘secret’.
In fact, Périkhanian cites, in the midst of the aforementioned examples in
Ved. -as-/Iran. -ah-, a compound with the same semantic function, whose second
member has a different shape: vr̥ ́ kāyu- ‘living like a wolf’ = ‘rapacious, brig-
and(ly)’. I shall demonstrate an effectively similar meaning for vr̥ ́ kadvaras- ‘living
in a wolf’s den’, where *dváras- has nothing to do with running, but refers to a
cave or the like. Indeed, the attempt to relate the -dvaras of vr̥ ́ kadvaras- to Iran.
dvar was flawed at the outset, since there is no evidence for a Vedic verb dvar.
Rather, Iran. dvar is a creation which took place within the development of
Iranian. For these matters, an explanatory digression is necessary.
I focused at length on the meaning of both *dváras- in vr̥ ́ kadvaras-, and the
origin of the Iranian verb of motion, in a long article (Schwartz 1992) in which I
detailed the evidence for a Proto-Indo-European root *dʰwer(E) ‘to go apart’ and
other relevant words (ibid.: 391-406), and then focused on Vedic *dváras- ‘den’
and the Iranian verb dvar (ibid.: 406-410).
 Armenian varkaparazi and its Iranian Background 405

While the latter article should be consulted for details, the present simplified
and expanded version will serve the purposes at hand. Providing a priori and a
posteriori evidence for PIE *dʰw > Greek s-, and disproving alleged examples of
PIE *dʰw- > Gr. th-, I demonstrated that *dʰwr̥ (E) ‘to go apart bilaterally, to gape,
to be splayed’ gave the Gr. root sar attested in sésēra ‘has the lips retracted (with
teeth clenched); gapes (mouth, wounds, etc.)’; sarápous ‘splay-footed’, sárōn
‘vulva’, sárma ‘chasm in the earth’ and sḗragx (/sḗrang-/) ‘cave, orifice’.
Among the more straightforward derivatives of √dʰwer(E), which reflect the
basic verbal sense of ‘go apart bilaterally’ are: Lithuanian dvérti ‘go apart, be
stretched apart, go asunder’ (išdvérti ‘id.’, also ‘to gape, have the mouth open’;
dvėrà ‘someone who gapes’); Hittite duwarnai- (duwarne-) ‘to break something’;
with obj. aiš ‘to open one’s (shut) mouth, to tattle’ < PIE *dʰwer-ne-E ‘to cause to
go apart, to sunder’, cf. Hittite dudduwara- (with transitive reduplication) and
Hieroglyphic Luvian luwarryi- ‘to break’ (trans.), the latter probably demonstra-
tive from PIE *dʰworEo- ‘a gap formation’; and Lat. forus ( < PIE *dʰworE-o-) ‘a
passageway or free space bounded bilaterally’.
In addition, Lat. furcula ‘two-pronged fork, Y-shaped prop, bifurcation,
mountain pass’ is < *dʰwr̥ E-tl-eA; furcula yielded furca ‘fork’ by metanalysis. The
meaning ‘mountain pass’ (I now note) is also found in Young Avestan duuara-;
formally, cf. Lat. forus < *dʰworEo-. For ‘a Y-shaped gestalt’, the reconstruction of
a root-stem *dʰwr̥ E- ‘bifurcation, splay’ yields the word for the Y-shaped harness-
ing device consisting of a pole and yoke (or the apparatus coming bilaterally out
of the pole and onto the shoulders of the draft beasts), Vedic dhur-; cf. Tocharian
A tursko ‘draft ox’ (Pali dhorayha- < Skt. dhoreya- ‘having a dhur-’) < PIE *dʰwr̥ -
Hós gʷōu- ‘ox of the dʰwr̥ E-’, cf. RV 1.84.16 yuṅkte dhurí gā́ ‘yokes the oxen in/at
the dhur-’.
Our root provides the long sought verb etymon for door etc., PIE *dʰwor-
/dʰur-, properly ‘bivalvous door or gate’, named from its opening up bilaterally,
whence its earliest occurrences as dual, whose grammatical function was later
continued by plural forms. From this came PIE *dʰworo- ‘gated courtyard, court’
and expressions for ‘outside’.
One may also reconstruct a PIE neuter *dʰwerEtro- *‘that via which a gap is
formed’, ‘matrix for the two door-leaves, jamb’, pl. tant. ‘casing or frame of the
double door with the hingework’: (1) Gr. pl. thúretra ‘door-casing, door-frame’
(thúr- via influence of thúra(i) ‘door’), (2) Lat. Forculus ‘god of the door’s jambs or
frame’ (*forculum, pl. forcula; for- via influence of forēs ‘door’; -cul- < *-cl- < *-tl-
406 Martin Schwartz

dissimilated from *-tr-), and (3) Old Persian <dᵘvᵃrᵃϑᵃi-> = dvarϑi- (inscribed on
both jambs on each of two sets of opposed axial doorways at Persepolis) ‘door-
way, portice’ (Pers. dahlīz < *dvarϑiyači-); haplology from pre-Persic or non-Per-
sic *dvarϑri-.
Early Iranian had √dvar(H), which, among other remarkable Pashto reten-
tions, continued to be attested in the 19th and earlier 20th century as lwaṛedəl
and lwuṣ̌ təl (lwux̌ təl); here lwaṛ- < *dvr̥ ta- and *lwuṣ̌ t- (lwux̌ t-) < *dvr̥ sa- were re-
cast as pres. stem lwaṛ-, past stem lwuṣ̌ t- along the lines āwaṛ-: awuṣ̌ t- ‘to turn’,
etc.
In Avestan, √dvar/dvr̥ ‘go apart’ underwent influence of four other verbs,
which shared with √dvar and with each other both phonological and semantic
features: (1) √drav ‘to run’; (2) drav ‘not to go to the right place’, ‘to miss the goal,
go astray’; (3) zvar ‘to go crookedly, go wrong, be crooked’; and (4) ϑvar- ‘to hurry,
to rush’. I adduced the second verb from Av. drāuuaiia- ‘to prevent from going to
the right place, lead astray’ (subject aēšma- ‘Wrath’); Man. Parth. drāw- ‘to de-
ceive’, išmagān drāwanagān ‘deceptive Wrath-demons’, cf. Av. aešmō.drūta-,
‘lead astray by Wrath’; Chr. Sogd. ārδāw- ‘to seduce’; Khwar. ārδāw ‘misleading
(shape-shifting) desert demon(ess)’, etc. Formally, interaction of √dvar with
both roots √drav in the zero-grades (resp. dvr̥ and dru) was predetermined by
Indo-Iranian alternations of v(a)r and ru where, it must be noted, metathesis
took place in the zero-grade in words with preceding w or labiovelar in PIE.1
The vectors of phonic and semantic interaction between dvar and the four
other verbs yielded √dvar ‘to run perversely’, reflected in Av. duuara-, duuāra-,
MPers. dwāristan ‘to run or move, on the part of demonic creatures’. The re-
lationship of √dvar to √drav ‘run’, referring to the running of demonic creatures,
would be apt as iconic of perversity, and parallel to tabuistic metatheses.
As for Périkhanian’s assertion that the Indo-Iranian root dvar referred to
rapid movement in a straight line, but that the Avestan pejorization to the run-
ning of creatures of Ahriman “était étroitement liée avec la réforme religieuse de
Zoroastre”, it is now demonstrable that the dualization of the Avestan vocabu-
lary, whereby certain words referred only to evil entities, took place before the
composition of the Gathas. I have detailed elsewhere that a Yasna 32.10 is among
a series of stanzas in Yasna 32, which derive from a pre-Zarathushtrian form re-

1
*kʷetwor-, *kʷeturV-, *kʷetru- ‘four’; *swek̂ uro- ‘father-in-law’, *swek̂ ruA- ‘mother-in-law’;
*gʷreHwon-, *gʷr̥ nuA- ‘millstone’, etc.
 Armenian varkaparazi and its Iranian Background 407

arranges and recontexts vocabulary and themes of the “Hōm Yašt”, Yasnas 9 and
10. Here I shall only refer to Y32.10 vis-à-vis Y9.29-30. In both passages the words
ašibiiā/ašibiia ‘with the eyes of an evil creature’, vōižda- ‘raise’, vadarə/vadarə̄
‘weapon’ (which words do not occur in other Avestan texts) are collocated with
vaēna- ‘to see’ and aṣ̌ āunē/aṣ̌ āonē ‘toward/for the righteous one’, among other
correlations. This proves that the pejorized aši- existed long before the Gathas
and their religious innovations; the same is true for gauua- ‘hand of evil creature’
and zbaraϑa- ‘foot of evil creature’, which immediately precede aši- (again in the
instrumental case). It may be concluded that the dualization of the Avestan vo-
cabulary is an ancient pre-Zarathushtrian phenomenon (see further Schwartz
2006a: 216 et seq.; idem 2006b: 476-478, et seq.). This would include the verb dvar,
pejorative ‘to run’, which is a purely Iranian creation and of no relevance for Ve-
dic vr̥ ́ kadvaras-. I accounted for -dvaras- in the latter word as follows:
Proto-Indo-Aryan √dhvar(H) ‘to go apart’ (which was opposed in meaning to
√r̥ ‘to join, fit, be together’, whence Vedic r̥ tá- ‘cosmic coherence, Rightness’) un-
derwent interaction with √dhrav/dhru ‘to miss the goal’ (well reflected in Ira-
nian; see above) and √dhvarH also interacted with √hvar ( ~ √huHar) ‘be
crooked, off course, unstable’ (for *dhvarH vis-à-vis √hu(H)ar, note hūrcháti
‘goes off course, vacillates’ vis-à-vis OIran. *dvr̥ sa- ‘to go apart’ reflected by Pash-
to, discussed above). As a result, √dhvarH came to replace the verbal forms of
√dhrav (√dhru), leaving only the root-stem dhrut-, and the abstract noun dhrúti-
in related passages concerning the divine custodian of r̥ tá-: RV 7.60.9 varuṇa
dhrúti sā́ ‘O Varuṇa, deception/derailment it was’ (where the further reference to
liquor, anger, dice, and thoughtlessness is reminiscent of Av., Yt 1.18 aēšmō.drū-
tahe druxš.manaŋhō ‘of him led astray by Wrath, with mind under the sway of
Wrongness’).
These interactions resulted in Vedic dhvárati ‘injuries, deceives’ and dhváras-
(RV 2.23.5, 4.23.6-7), parallel to drúh- ‘harm, wrongness, deceit’ and vr̥ jiṇá- ‘crook-
edness’, cf. RV 10.25.1 satyadhvr̥ ́ t- ‘perverting or deceiving the truth’, parallel to
vr̥ jiṇāyánt- ‘crooked’. As a result, Vedic *dhvar- ‘leaf of a double door’ became
associated with forms pertaining to injury and unreliability, and underwent a
new association with dvā/duā ‘two’, becoming dvar-, dual dvā́ rā, the influence of
duā being reflected in the scansion duārā RV 4.51.2, Prakrit du(v)āra-, duāriā,
Panjabi and Nepali duār-, etc. Cf. also Pashai dūr and Nuristani Waigali dōr
‘door’.
408 Martin Schwartz

With OInd. *dhvar- > dvar- in the word for ‘door’, the same phonic change
was transferred to cognate forms from √dhvar ‘go apart, form a gap’. For the se-
mantic relationship, cf. inter alia Hom. Gr. thúrai Od. 9.423 ‘cave-mouth’ = ‘dou-
ble-door’. In Dardic Kashmiri has bar m. ‘door’ < *dvara-; boru m. ‘hole permitting
entrance, crack or fissure in a wall’ < dvara-ka-; and bürü f. ‘spy-hole, crack in
wood or stone’ < *dvari-kā. We may accordingly posit a meaning like ‘cave, lair’
for Av. duuara- in the epithet of the Vasī-fish, Av. paṇcāsaduuara- ‘having fifty
caverns or lairs’. These Kashmiri and Avestan data point to an interpretation of
RV 1.52.3 dvará dvaríṣu vavrá as ‘a door/cave-opening amidst fissures, an
engulfment...’ (see further Schwartz 1992: 407-408).
We thus arrive at the ferocious Asura’s name Vr̥ kadváras- RV 2.30.4 as ‘living
in the den of a wolf’. The *dváras would refer to the cave-like den of a wolf (or
even a cave itself, since she-wolves sometimes lactate in caves); *dváras- (via
*dhváras-) parallels Vedic hváras- ‘bent configuration’.
To return to Périkhanian’s thesis, RV 2.30.4 vr̥ ́ kadvaraso ásurasya vīrān means
‘the men of the Asura Wolfden’, object of Br̥ haspati’s assault; Périkhanian’s “les
hommes d’Asura, ceux qui courent comme des loups attaquants” must be re-
jected, and with it her support for varkaparazi as involving ‘wolves on the run’, in
this instance supposedly wolves on the escape. There is, moreover, nothing
inherently disorderly about wolves’ running in general, and specifically, in their
running away, whence she wished to explain varkaparazi metaphorically.
With the wolves gone from consideration, a second look at the vark(a)- of
varkaparazi is called for. From the outset of her article, Périkhanian, evidently
seduced by her lupine hypothesis, wrote: “Il n’y a pas eu de tentative sérieux
d’expliquer ce mot, car on ne saurait prendre au sérieux l’explication donnée
dans le Grand Dictionnaire de Venise qui le lie, par une fausse association bien
sûr, avec arm. vark ‘calcul, considération, comput’, mot qui n’a rien à faire ici”.
But shouldn’t vark, with such a meaning, be relevant for the essential meaning of
varkaparazi ‘without planning, without intent, haphazardly’? Arm. vark is now
to be derived from West Middle Iranian *vark, whence Persian barg ‘aim, design,
intent, care’ (the connection between the Arm. and Pers. words was made by
Oułourikean, as cited by Ačaṙean (vol. 4, 1935/1979: 322); non legi). Our WMIran.
*vark would be via OIran. *varka- from √var ‘to choose, *will’, cf. Khwarezmian
wrk ‘choice’, OAv vāra- ‘will, desire’. In formation, *varka- thus stands to √var as
Av. saoka- ‘benefit’ to √sav ‘to bring benefit’. The semantic development would
be ‘desired choice’ > ‘deliberate intent’ > ‘intentionality, calculation’. Accord-
 Armenian varkaparazi and its Iranian Background 409

ingly, in varkaparazi ‘unplannedly, unintendedly, haphazardly, carelessly’,

*(a)parazi would indicate an action’s lack of vark. Now, Cheung (2007: 322-3) has
brought together Old Persian ava-√rad ‘to leave, abandon’ with an abundance of
East and West Middle and New Iranian verbs, which denote fleeing, escape, be-
ing freed, etc. (some of which were mentioned by Périkhanian). With these
Cheung combines Khotanese rrāysaa- ‘empty’ and, drawing on Henning’s fur-
ther connections, OInd √rah ‘to be lost, be lonely’, ráhas- ‘secrecy, loneliness’,
Sogd. and Khwar. rāz f., MPers. and Parth. rāz ‘secret, mystery’. Cheung com-
ments: “The meanings ‘secret’ and ‘flee, escape’ are not necessarily incompatible
if we assume an older meaning *‘to escape one’s notice’”. However, all the words
involved are more essentially reconciled by the idea of absence: Khot. ‘empty,
abandon’ is ‘characterized by absence’; ‘flee, escape’ amounts to ‘absent oneself’;
while OInd. √rah ‘be lost, be lonely’ can be taken from ‘be absent/away from
people’, and ‘secret’ may readily be derived from ‘that which is away from
(absented from) the generality of people’.
In conclusion, we may now take Arm. *aparaz in varkaparazi as *‘the aban-
donment (*making absent)’, and varkaparazi from ‘with the abandonment of
any planning’, which is in effect the basic meaning of the Armenian word.
It is thanks to Dr. Périkhanian’s calling attention to varkaparazi, and her
providing a partial starting direction toward its etymology that, albeit with the
rejection of her overall interpretation, we have gained a lost Iranian compound
whose members are of noteworthy interest for the history of the Iranian vocab-
ulary.

BIBLIOGRAPHY
Ačaṙean, H. Y. (1926-1935), Hayerēn armatakan baṙaran, 1-4, Yerevan, reprinted, Yerevan 1971-1979.
Cheung, J. (2007), Etymological Dictionary of the Iranian Verb, Leiden-Boston.
Périkhanian, A. (2007), “Une métaphore póetique indo-iranienne en arménien classique: arm. var-
kaparazi”, Religious Texts in Iranian Languages: Symposium held in Copenhagen May 2002,
Copenhagen: 121-124.
Schwartz, M. (1992), “Relative Chronology in and Across Semantic Hierarchies: The History of
*dhwer(E)- ‘go apart’ in Indo-European”, Rekonstruktion und Relative Chronologie (Akten
der VIII. Fachtagung der Indogermanischen Gesellschaft, Leiden 31. August-4. September
1987), Innsbruck: 391-410.
―― (2006a), “On Harma and its Liturgy in the Gathas”, Proceedings of the 5th Conference of the So-
cietas Iranologica Europaea I, Milano: 215-224.
―― (2006b), “The Gathas and Other Old Avestan Poetry”, La Langue poétique indo-européenne,
Leuven-Paris: 459-498.
 The Polygenetic Origins of the Northern Talyshi Language

Donald Stilo
Max Planck Institute for Evolutionary Anthropology, Leipzig

Abstract
Within the study of Iranian languages, it has always been perplexing to see what striking differ-
ences “dialects” of supposedly one and the same language can have. Specifically in the case of
Northern and Southern Talyshi, the dialects are so different from each other in lexicon, historical
phonological developments, and grammatical domains and also share very low mutual intelligibil-
ity. The claim is made here that Northern Talyshi represents a case of polygenesis from at least two
different Iranian sources.
Talyshi and Tati (also called “Old Azeri”, i.e., the indigenous Iranian language of Azerbaijan, still
spoken there in remnant pockets) are closely related Northwestern Iranian subgroups, collectively
called Tatic. At some point after Talyshi and Tati diverged from each other, Northern Tati alone
underwent some very striking linguistic restructurings, especially in its verbal system. One group
of Northern Tati speakers subsequently moved eastwards off the plateau into the northern areas of
the Talyshi zone, fanning somewhat southwards, and meshed with the local Talysh populations,
affecting Northern Talyshi and much of Central Talyshi on all linguistic levels. Since the Southern
Talyshi dialects did not participate in this process, a wide rift came about between Northern and
Southern Talyshi. In addition, Southern Talyshi dialects probably also underwent their own paral-
lel mixture with neighbouring varieties of Central Tati on the plateau through regular and ongoing
contact and mutual borrowing.
The set of unusual features shared by Northern Tati, and Northern and Central Talyshi are quite
unique among Iranian languages. Most of these features are also shared by other languages in the
area: quite aberrant dialects of Armenian (the so-called -lis branch), Udi (Daghestanian), Neo-Ar-
amaic dialects, and Caucasian Tat (the Iranian linguistic enclave in Azerbaijan Republic and Da-
ghestan). This situation most likely developed as a result of a language shift to Northern Tati by
some pockets of one or more of the non-Iranian groups mentioned before Northern Tati groups
moved into the Talyshi zone.

Keywords
Talyshi, Taleshi, Tati, Old Azeri, Caucasian Tat, Azerbaijan, Armenian Dialects, Udi, Neo-Aramaic,
Language Mixing, Language Shift, Dialect Distance, Language Polygenesis

 Koninklijke Brill NV, Leiden, 2015

412 Donald Stilo

1.0 INTRODUCTION
Within the study of Northwest Iranian languages, it has always been perplexing
to see what striking differences “dialects” of supposedly one and the same lan-
guage can have. Specifically in the case of Northern and Southern Talyshi,1 the
dialects are so different from each other on all levels―in historical phonological
developments, in lexicon, and especially in grammatical structures―and share
only very low mutual intelligibility that Southern Talyshi and Northern Talyshi
should in fact be considered separate languages. While this debate has never
been broached for Talyshi, we find precedents in other Iranian languages, e.g.,
the status of Southern, Central and Northern Kurdish as dialects or separate lan-
guages (Haig/Yaron: 2002; Asatrian 2009: 51-57).2
Before delving into the substance of the issue, I should note here that I do not
use the terms “dialect” and “language” in the socio-cultural or political senses
commonly used in Iran and neighboring countries―although Talyshi does in
fact enjoy official status as a “language” in Azerbaijan Republic. I define my cate-
gories rather on purely linguistic criteria. While the linguistic terms themselves
are relative and no clear line can be drawn to demonstrate with absolute criteria
when a specific speech form should be considered a separate language from its
closest congeners, there are cases where the linguistic differences are very strik-
ing and too numerous to disregard. Such is the case of Northern Talyshi.
My research on Talyshi, begun in the mid-1970’s in Iran with Southern and
Central Talyshi but resumed more intensively between 2002 and 2006 in Azer-
baijan Republic with Northern Talyshi, in tandem with comparative work with
other Tatic languages (see §1.1 for this term), leads me to the unavoidable
conclusion that Northern and Southern Talyshi―although they see themselves
as one ethnic group, the Talysh, with one common identity—in fact differ more
than enough from each other to warrant a linguistic classification as clearly sep-
arate languages. More specifically, I will show below that Northern Talyshi and

1
My English spelling of the name of the language as Talyshi, which also serves as the general
adjectival form as well, follows the Russian and, more importantly, the spelling with -y- is more re-
flective of the original Talyshi (N. Tal.: tolïši, S. Tal. (Māsulei): talIši (with a front vowel similar to
English /i/ in hit, but higher than Persian /e/), thus shunning the purely Persian pronunciation,
which is spelled Taleshi. The term Talysh is never used for the language, as is occasionally encoun-
tered in the literature, but refers only to the people as an ethnonym.
2
See also Stilo (forthcoming) on this problem in speech forms usually considered dialects of
Mazanderani.
 The Polygenetic Origins of the Northern Talyshi Language 413

the dialects of Central Talyshi spoken slightly farther south in Iran―both of

which share more commonalities than either do with Southern Talyshi―in fact
exhibit a mixed language phenomenon in their history. That is, at least two dif-
ferent Iranian sources can be identified as having come together from different
directions to contribute to the formation of the present variants of Northern and
Central Talyshi.
The polygenesis of which I speak is not one of simple lexical borrowing, such
as the heavy presence of Arabic, Azerbaijani and Persian loanwords in Talyshi
(in addition to Russian in the Northern dialects), but rather one of language
mixing due to a population movement of a group of Northern Tati speak-
ers―that is, a different Tatic, but non-Talyshi, source–into the Northern and
Central Talyshi dialect areas at some point in the history of the present-day lan-
guage. The resulting population merger has led to a mixing or “intertwining” of
the languages of these two original populations on all linguistic levels, as already
mentioned. While it seems to be the case that the Northern Tati linguistic con-
tribution to Northern and Central Talyshi is slightly heavier all around than the
purely Talyshi contribution, the latter has contributed a substantial (but some-
what smaller) percentage to the grammatical and lexical composition of the lan-
guage.

1.1 BACKGROUND
The various dialects of Talyshi are closely related to the Tati3 group, hence my
coinage of the term “Tatic” (Stilo 1981) to include both groups. Tatic is one of the
seven branches of modern Northwest Iranian languages. The diversity within the
Tatic family in terms of the approximate degree of divergence from each other is,
roughly estimating, about that of Romance languages. Talyshi is spoken along
the southwest Caspian littoral, as well as in the somewhat higher altitudes of the

3
“Tati” as used here refers only to the group of Northwest Iranian (i.e., Median-derived) lan-
guages spoken on the Iranian plateau within modern Iran. This term has no direct connection to
the language variously called “Tat”, “Tati” or “Caucasian Tat” spoken by Sunni, Shi’i, Jewish and
Christian groups in Azerbaijan Republic and southern Daghestan. The latter is of Southwest Iranian
(“Persic”) origin and only coincidentally has a similar name to the NW Iranian group discussed
here. With a separation of almost 3000 years between NWI Tati of Iran and SWI Tat of the northern
areas of Azerbaijan Republic/Daghestan, these two language groups are mutually unintelligible.
The term “Tat(i)” is of Turkic origin and has also been applied by Turkic speakers to various non-
Turkic languages (and their speakers). In this paper, I refer to the SW Iranian language mentioned
here as “Caucasian Tat”.
414 Donald Stilo

mountain slopes that face the coast both in Iran and in the very southern areas
of the Azerbaijan Republic. As far as we know, with the exception of Talyshi
speakers in the Anbarān area to the north of Namin near the Iran-Azerbaijan
border, groups identifying as Talysh are not found indigenously beyond the
mountain passes that lead to the Iranian Plateau. The latter area is the domain of
the Tati group (see Map 1). The Talyshi speech communities of the coastal areas
and various Tati-speaking areas of the plateau, however, have not in fact re-
mained completely out of touch with each other due especially to seasonal work
usually available in the coastal areas in Iran.
Northern Talyshi of Azerbaijan Republic is classified into four closely related
dialect zones (Lerik, Masally, Lenkoran, and Astara). In spite of the fact that the
Astara zone is slightly divergent from the other three zones, all four dialect zones
of Northern Talyshi are completely mutually intelligible and when Talyshi speak-
ers from different zones in Azerbaijan Republic come in contact, they find no
difficulty in communicating with each other in their own dialects. Northern
Talyshi dialects in the immediately bordering Iranian areas, especially the Astara
and Anbarān areas on the Iranian side of the border, are also very close to the di-
alects of Azerbaijan, especially to the Astara dialect there. The language is spo-
ken by a population of some 80,000 in Azerbaijan according to official statistics,
but this is clearly a gross underestimate (or undercounting). Local estimates for
the Talysh in the Talyshi zones in fact put this figure closer to some 400,000, a
not implausible estimate considering that the Talysh region is shown as the most
densely populated area of the country (other than the capital and surrounding
areas) on the demographic maps of Azerbaijan4. In addition, a large number of
Talysh have also migrated out of the area to Baku, Sumgayit, Bina (east of Baku),
and beyond Azerbaijan to the larger cities of the more prosperous republics of
the former Soviet Union, especially Moscow, Minsk, Kiev, Odessa, etc. This out-
migration should not be underestimated in terms of the number of native speak-
ers of Talyshi: the number of Talysh resident in Sumgayit alone, an urban center
of some 400,000 inhabitants created during Soviet times 50 kilometers to the
northwest of Baku, is estimated to be at some 80,000 (Tiessen 17). Considering
the dense population in the Talysh region and the heavy settlement of Talysh in

4
See http://vamodc.wfp.org/geonetwork/srv/en/metadata.show?id=94&currTab =simple.
 The Polygenetic Origins of the Northern Talyshi Language 415

various urban centers beyond the Talyshi linguistic area, the figure of native
speakers of Talyshi in Azerbaijan may be as high as 500,000 or more.5

Map 1.

The estimates for the number of Talyshi speakers in Iran also hover around
75-80,000 but accurate figures are harder to determine in Iran due to the heavy
influx of non-Talysh (particularly Iranian Turkophones) in recent years, espe-
cially into the coastal cities of Iranian Tavālesh, reportedly relegating Talyshi
only to the villages of the surrounding areas. Māsāl and Māsule, located in
mountainous areas somewhat removed from the coast, have remained more
purely Talyshi speaking.
The Tati groups of the plateau in contrast to Talyshi, are both less populous
and more dispersed. There is only geographic continuity in Tati-speaking areas
for some segments of Central Tati, from southern Khalkhāl through Upper
Tārom north of Zanjān, but these areas are not very densely populated in any
case. All other Tati groups are found only in disparate pockets to the northwest,

5
According to recent calculations of non-biased foreign axperts, the number of the Talyshis in
Azarbaijan Republic can be estimated around one million (see Asatrain 2011: 25-28).
416 Donald Stilo

the south, and the southeast of this area, where speakers of these languages are
often totally unaware of the existence of other forms of Tati. While there are cer-
tainly some prosperous, populous areas inhabited by Tati speakers, e.g., the vil-
lage of Vafs6, many Tati villages are poor and can only support small popula-
tions―some were as small as 200 inhabitants or less, when I visited them in the
1970’s. The Tati dialects are also more endangered than the Talyshi dialects, at
least at present, since with the social upheaval of the 1970’s and later in Iran,
many villages were totally abandoned by their populations in favor of life in
large urban centers, particularly Tehran, Qazvin, and Rasht. In 1976, speakers of
Koluri in the Tehran bazaar informed me that they estimated their population to
be about 1,000 families, with half of them in Kolur and half in Tehran in cohesive
neighborhoods with their own extensive social network. On a visit to the Tati-
speaking village of Kabate (locally Kafti), speaking a Tāromi-type Tati (locally
Kafteji) within the borders of Gilan province, also in 1976, I found the village to
be almost completely deserted and was advised to look for speakers in Rasht and
the factory complexes outside Qazvin, both of which, in fact, yielded better re-
sults.
In addition to small numbers, another factor causing their endangerment is
that the vast majority of, if not all, Tati-speaking villages have been bilingual in
Azerbaijani for generations and the young people in these villages were already
beginning to speak the latter language more and more in the 1970’s. Also since
the 1970’s Persian has been gaining dominance even as the language of the home
in various areas of Iran that traditionally spoke Tatic, Caspian, and Central Plat-
eau languages. Thus, the fate of small Tati villages in the heart of Azerbaijan is
not at present known and does not seem very hopeful. Keringān, visited by re-
searchers in the mid-1990’s, still had speakers of the local Tati, although the pop-
ulation was greatly diminished and those left spoke Azerbaijani Turkish as their
predominant language.
Map One shows the locations of Tati, Talyshi and Caucasian Tat schemati-
cally. More accurate geographic locations of at least some of these villages and
language areas are found on Map Two.

6
The population of the village of Vafs is estimated at ±12,000―although local estimates claim
that at least 50% of this number is resident in Tehran, leaving houses in the village empty except
for long-term visits, particularly in the summers.
 The Polygenetic Origins of the Northern Talyshi Language 417

2.0 THE PRESENT PROPOSAL: THE MIXED ORIGINS OF NORTHERN AND CENTRAL
TALYSHI
The close relationship of Talyshi to various forms of Tati was recognized early in
the Iranist literature on the study of the structure and history of these language
groups. The Swiss linguist, Emil Baer, noted in 1938, “Harzandi and Shāhrudi of
Azerbaijan prove to be closely related to the Talyshi of the Caspian” (Baer 1938:
155). Henning (1954: 174) tells us that “Tālišī, like Harzanī, possesses a present
built on the preterit stem; the Tālišī forms ... can in fact be explained only with
the help of the Harzanī material”. Henning’s statement undoubtedly refers only
to Northern Talyshi, the most well-known form of Talyshi at the time, due to the
then-recent publication of Miller’s (1953) Talyshi grammar. Note the importance
this statement makes of Northern Tati, in this case Harzani, for the interpreta-
tion of Talyshi grammatical forms. Yarshater (2005: 269) also states: “Due to their
geographical location [Kalāsuri, Xoynarudi, and Keringāni] all share a number of
important features with Tāleši, to the extent that they could be equally consid-
ered a branch of northern Tāleši or a bridge between Tāti and Tāleši...”.
Regarding Southern Talyshi and Central Tati, Yarshater (1959: 53) states: “In
studying Shāhrudi one gains, from the first, the impression that this dialect
closely resembles southern Tāleši, especially the variety spoken in Shānderman
and Māsāl...”.
My own current view of the possible historical scenario that has led to the
unusual disparity among different forms of Talyshi is the following, presented
here first in the form of a brief synopsis and then expanded in stages in the re-
mainder of this article:
A) Pre-Talyshi seems to have split from Tati, moved off the plateau to the
Caspian littoral, and eventually fanned out up along the coast. It must have been
a relatively cohesive linguistic group, at least in the early stages of its expansion,
as seen from certain lexical and grammatical features that are only shared by all
versions of Talyshi but (generally) not shared by Tati.
B) Early forms of Tati―often also called “Old Azeri”, i.e., the indigenous Ira-
nian language of Azerbaijan having no relationship to the modern Turkic speech
of the area―were no doubt commonly spoken on the plateau throughout Azer-
baijan (Aturpātakān) at least as far north as the present Northern Tati areas both
south of the Iran-Azerbaijan border (Harzani, Keringāni, Kalāsuri, Xoynarudi,
Arazini and others), as well as just north of the border and the Araxes (cf. Kilit, a
Tati language, now most likely extinct, in the Ordubād area of Naxjavān/Nakhi-
418 Donald Stilo

chevan) province of the Republic of Azerbaijan (Gasanov 1966). Yarshater (2005:

269) also mentions having met a native speaker of Kalāsuri who states that “...he
personally had been in Darazin and Ḥaddādān in Northern (former Soviet)
Azerbaijan, close to the Iranian border and that their language was about the
same as Kalāsuri...”.
These Northern Tati dialects all have at the same time striking similarities to
each other, as well as a substantial set of unusual features that set them off dra-
matically from all other Tati relatives, i.e., Central and Southern Tati. Some of
their features are in fact quite unique among Iranian languages, except for North-
ern Talyshi. Not surprisingly, while all forms of Northern Tati share many fea-
tures with Northern Talyshi, the easternmost of the Northern Tati dialects still
extant today, e.g., Kalāsuri-Xoynarudi, seem to come closest to the similarities
shared with Northern Talyshi.
C) At some point, a group of Northern Tati speakers from the Araxes area
then evidently moved eastwards off the plateau into the Talyshi-speaking areas
of the Caspian littoral―most likely through the Namin-Anbarān area, the easiest
access from the plateau and the Ardabil area to the coastal regions of the Astara
area―and meshed with the local Talyshi speakers.
D) With the melding of the two populations, originally speaking related but
very different languages, a mixed language resulted. This form of the language
then eventually spread out over an area extending south roughly to the Asālem-
Hashtpar area of Iran (and possibly even farther south), as well as northwards to
the areas in the present Talyshi-language zones of Azerbaijan Republic. North-
ern/Central Talyshi is thus a language formed from two different sources: a vari-
ety of Talyshi from the coast and a variety of Northern Tati from the plateau.
This “intertwining” of two speech communities in the northern and central
zones of the Talyshi-speaking linguistic area caused a wide rift between these
two types of Talyshi on the one hand and Southern Talyshi on the other.
E) Southern Talyshi dialects may also possibly have their own mixture with
the nearest forms of Tati on the plateau. Such a contact situation resulting in
borrowing is on-going between Southern Talyshi and Central Tati. Māsāl/Māsule
from the Talyshi side and Shāl/Kolur from the Tati side are geographically quite
close. This proximity has facilitated constant interchange and hence linguistic
contact phenomena between Southern Talyshi and the local variants of Central
Talyshi, most likely flowing in both directions.
 The Polygenetic Origins of the Northern Talyshi Language 419

I must add that the above description in five stages is my current position,
but given my evolving research on these languages at present, I feel that the sce-
nario presented here still needs to be worked out in its details. The basic premise
of Northern Talyshi as a mixed language of two independent Tatic sources, how-
ever, will remain. The ideas laid out in the conclusion to this paper present some
potential areas where this proposal can be altered or enhanced.

2.1 THE FULLER DESCRIPTION OF THE PRESENT PROPOSAL

The proposal of the formation of the various varieties of Talyshi presented above
in synopsis form is now reexamined in further detail below.
A) Unity of all Three Types of Talyshi
Talyshi is generally divided into three basic types: 1) Southern Talyshi (Māsāli,
Māsulei, Shāndermani and others), 2) Central Talyshi (Asālemi, Hashtpari, and
others), and 3) Northern Talyshi (in four closely related dialect zones of Lerik,
Masally, Lenkoran, Astara in Azerbaijan Republic and in the dialects of Astara,
Sayyādlar, Vizane, as well as Anbarān and surrounding villages in Iran). Central
Talyshi is clearly more closely related to Northern Talyshi in both grammatical
features and lexical composition than it is to Southern Talyshi. There are also
transitional dialects between each of these groups, e.g., the dialect of Jowkandān,
transitional from Northern to Central Talyshi, or Tālesh-Dolābi, transitional be-
tween Central and Southern Talyshi.
Certain lexical and grammatical features unite all varieties of Talyshi but are
(generally) not found in any local forms of Tati. These isoglosses show a probable
original common source for all forms of Talyshi. An especially significant gram-
matical isogloss is the use of prefixed tense formant, usually called “the Aug-
ment” (added to the present stem), to form the Imperfect tense in most forms of
Talyshi, specifically Northern, Central and transitional dialects between Central
and Southern Talyshi, at least as far south as Tālesh-Dolāb (Yarshater 1959: 62),
but absent in Southern Talyshi and all forms of Tati. (See point (E´) below for a
discussion of the absence of the Augment in Southern Talyshi.) Table One shows
a sampling of lexical and structural isoglosses shared by the various forms of
Talyshi, and (generally) only Talyshi.
420 Donald Stilo

C.-N. Tati NORTHERN TATI

Lexical Features Kajali Harzani Keringani Kalasuri

bee ǽngæ, f. müz müz, müz- müz

ængin

each other ivæ-din ive-dun

fingernail dərnax nuta, notæ dernæx

large, big pillæ kælæ ketmæ yækæ

pound-past (lost in favor of kust, kuvest küsd küšt küs(t)

present stem)
run-past vrit veret, ære- veyrit, veret vrit
bərir
small, little villæ, ketæ vede, vedek güdæ, kicik güdæ

“stand” merges with “stay” vendær vendæ vende

GRAMMATICAL FEATURES
is the augment used in the - - - -
imperfect?
Table 1: Lexical Isoglosses uniting only Southern-Central-Northern Talyshi (part 1 ) →

The isoglosses (both lexical and grammatical) that are common to all forms
of Talyshi and unique to the Talyshi group are not numerous but are important
ones. There are many additional commonalities shared by all three basic types of
Talyshi that have been eliminated from the lists presented here because they are
also found in all Tatic languages, or are shared by Talyshi, Tatic and other
Northwest Iranian languages, and finally, features shared by all western Iranian
languages and thus cannot be considered as distinguishing features for our pre-
sent purposes.
There may have been at a previous time even more lexical and grammatical
isoglosses uniting all forms of Talyshi in contradistinction to Tati, but some of
these may have been lost due to diffusion of areal phenomena, particularly from
Gilaki, or as a result of borrowings from Azerbaijani. See section (B´) below for a
discussion of this phenomenon.
 The Polygenetic Origins of the Northern Talyshi Language 421

TALYSHI “DIALECTS”
N.Talyshi Lerik Vizænei Jow-kændani Asalemi Masali Masulei

muz, meš meš meš æsælæ-meš æsǽlæ-meš æsǽlæ-meš

yandï yandə bæyandə iví ~ eví

bəndí
nangïr nangər mængər mængə mængə
yol yol yaL, kæL1 pillæ pil(læ)

žæ, kü ku kua kust ku ku

tïl, vit vit vit vrit tilli, dæv, tellə

tajəni (caus)
gædæ(li), ruk, gædæ gəc, gec ruk ruk
hïrd,
jïk(ï)læ
(pe-) mand, vəndən mon mun, vənden mon/mænn
ïndïn

+ + + + - -

← Table 1: Lexical Isoglosses uniting only Southern-Central-Northern Talyshi (part 2) →

* Note:1. The symbol /L/ used in these charts represents a velarized /l/-sound, as the “dark” /l/ of
English, Dutch or Russian.

B) Close Origins of Southern Talyshi and Central Tati

Until further information on the origins of Talyshi―and Northwest Iranian lan-
guages in general―is uncovered, Talyshi seems most likely to have diverged
from the Tati group as defined in Stilo (1981: 138-141). Southern Talyshi is partic-
ularly close to the Central Tati of the Shāhrud type, especially Shāli-Koluri de-
scribed in Yarshater (1959), possibly to the point of being mutually intelligible
with them. Table Two shows the affinity of these two language types. Some of
these features extend into Central Talyshi (Asālemi) but not as far as Northern
Talyshi. See section (E´) below for a fuller discussion of the similarities of Shāh-
rud-type Tati and Southern Talyshi.
422 Donald Stilo

CENTRAL TATI
Koluri Kæræni Gandom-abi Nowkiani Hezar-rudi Kelasi Kafteji

æng ǽngæ æsǽlæ-mægæs,

æng, m., divæng

ivi- kæskæs-

næxen
pillæ, (kælæ), pillæ pillæ pillæ pillæ pillæ, pildæ pillæ
yækkæ

kuss kuss kušt kust

verit, tajenes virit, væšt dæ-væšt dævəst dævəst dævəst, vrit dævəst

velle/villæ, velleg xordæ xordæ

gec

vendær vendær išt

- - - - - - -

← Table 1: Lexical Isoglosses uniting only Southern-Central-Northern Talyshi (part 3)

Areal Interaction of Talyshi, Central Tati and Gilaki; Gilaki Diffusion

In addition to common inherited genetic features that unite Southern Talyshi
and Central Tati, both these groups also share commonalities that result from
the diffusion of many Gilaki features into the neighboring areas but not far be-
yond, that is, some features extend only to Southern Talyshi, and others to Cen-
tral Talyshi (e.g., “large”, Table Three). Still other Gilaki features have also dif-
fused to some Tati dialects. Such patterns of diffusion from Gilaki (or from Azer-
baijani in the case of Northern Talyshi) were only acquired later in the develop-
ment of these languages and would undoubtedly have interrupted other addi-
tional lexical isoglosses uniting all forms of Talyshi in contradistinction to Tati
and may have obscured the fact that the isoglosses common to all forms of Taly-
shi may have been even more numerous at a previous time. Table Three shows
some examples of probable Gilaki influence in Talyshi.
 The Polygenetic Origins of the Northern Talyshi Language 425

NORTHERN TATI TALYSHI

Lexical Item Harzani Keringāni Kalasuri N.Talyshi Vizanei Asālemi
egg ovæ üvæ, uyo ü(y)æ moγnæ, üvæ uvæ, kagǽ ɔqLæ, aγlæ

bone ösda astæ, æstæ, ostæ, æstæ, astæ astæ æstæ

osta astæ, asta
large, big kælæ ketmæ yækæ yol yol (/yal/) yaL, kæL
want höz/höšt pi/pist /pist, pi/pi pi/pi pi/pist
an_(h)ušte
belonging to (mal) Ø Ø, gəLe
call, invite-pres dæ-xan
Grammar: kərə adpos. + BE ka(r)
Table 3: Different Patterns of Gilaki Diffusion (part 1) →

mirzādi as an outlier to the south), is leading me to the conclusion that the earli-
est form of the family, i.e., “Proto-Caspian”, must have entered the Caspian litto-
ral from the plateau through the central mountain pass leading from the Taleqān
area to the Kelārdasht-Chālus area of the seacoast. Several linguistic factors
point to such a conclusion but their nature has not completely cry- stallized in
my research―I have yet to establish the details of this early population move-
ment, the original homeland area of “Pre-Proto-Caspian” on the plateau and the
Alborz areas, or what linguistic group(s) this group may have encountered that
were already existing along the Caspian littoral when they arrived―since I have
only been working on these issues very sporadically.
The relevant point for the present article to be made by introducing this brief
excursus into Caspian prehistory is simply this: As Proto-Caspian entered the
seacoast and spread both westwards and eastwards from the central Caspian
area, the group progressing westwards that would eventually become the various
dialects of Gilaki most likely encountered a Tatic language (early Talyshi? a form
of Tati? pre-Rudbari?) that was already established in the coastal areas of Gilan
to the west of the Sefidrud. Hence the dialects of Gilaki east of the Sefidrud share
certain linguistic features with all Caspian languages to the east extending even
as far as Gorgan, but not with Western Gilaki. The latter dialect cluster, however,
shares other features with Southern Talyshi, Rudbāri dialects (which I classify as
“Tatoid”), and the local forms of Central Tati.
In the process of language shift from their original Tatic speech to a Caspian
language (pre-Gilaki) west of the Sefidrud, the Tatic substratum left its imprint-
on the incoming Caspian speech form, i.e., the eventual Western Gilaki, thus dis-
426 Donald Stilo

TALYSHI Caspian CENTRAL TATI

Māsāli Māsulei Gilaki Koluri Shāli Karani Kelāsi Kafteji
murnæ, morγanə xa mərγanæ, xa morγone morγone
morγanæ xa
xaš xaš xaš xaš æssæ

pillæ pil(læ) pil(l)ə pillæ pillæ pillæ pillæ, pildæ pillæ

xa/xast xa/(xass?) xa/xast* xa/ xa/ gæ/gævess goy/gost go/

šen šen šin šen šen

dæ-xun dæ-xon du-xan dæ-xan dæ-xan de-xon de-xond
kərə kə́ ra kəra kær- kæræ kærdæ
← Table 3: Different Patterns of Gilaki Diffusion (part 2)
tinguishing it from Eastern Gilaki. This conjectured substratal influence has sur-
faced in Western Gilaki in the form of structural features and, as is typical of lan-
guage shift situations, lexical borrowings do not enter into the process. Lexical
features may have diffused later as a result of adstratal contact phenomena. One
common structural feature in the contiguous languages of this area is the lack of
Tense-Aspect-Mood formant in the Present tense, i.e., a Ø marker, found only in
Western Gilaki, Southern Talyshi and most of the Tati languages of the Shāhrud
area (e.g., Koluri, Shāli) and nearby areas (e.g., Karani), but not Eastern Gilaki or
other Caspian languages, Central or Northern Talyshi or other Central Tati lan-
guages. Another similar pattern is the suffixed -i- of the Imperfect,7 which is also
shared by Western Gilaki, some forms of Central Tati, and all Talyshi dialects as
opposed to other Imperfect formants in all other forms of Caspian and Tati lan-
guages. In addition, see Stilo (2001) for a discussion of Imperfect formations in
Eastern Gilaki, and Stilo (forthcoming) for Central Caspian and various dialects
of Mazanderani. For now, these ideas on the prehistory of Caspian languages and
the role a putative Tatic substratum may have played in the diffusion of features
into Western Gilaki will have to remain as a goal for future research into this
complex nexus of issues to determine the details of this history and the dia-
chrony of these languages.

7
Certain Central Tati dialects―Gandomābi, Karani, Karnaqi, Dizi, etc.―slightly farther afield
from Kolur and Shāl, mostly on the periphery of this area, have the imperfect suffix -æy- ~ -ey-,
which seems to be the more archaic, i.e., pre-monophthongized, version of -i-: Gandomābi: kætt-
æy-m “I used to fall”, Karani: hešt-ey-m “I used to get up”, Karnaqi: š-ey-m “I used to go”, Dizi: xærd-
ey-m-e “I used to eat”.
 The Polygenetic Origins of the Northern Talyshi Language 427

C) Disparity between Northern Tati and Central Tati

Sometime after Talyshi branched off from Tati―the chronology is not yet
clear―the northern areas of Tati located to the northeast and northwest of Ta-
briz, especially along the Araxes river where we find remnants of the original
Tati today in the form of Northern Tati dialects, became separated from the rest
of the Tati continuum, undoubtedly due to the expansion of Turkic Azerbaijani.
The Northern Tati dialects then underwent a substantial linguistic divergence
from their Tati congeners. The linguistic changes in this group could have arisen
through contact with, or language shift from, another language or languages―
whether Iranian, non-Iranian, or both―that were already in this area. The types
of features I have in mind, however, as we shall see below, are not those that can
feasibly be attributed to contact with or influence from Turkic.
The unusual linguistic features of Northern Tati and the location of its many
variants (formerly certainly more numerous) along the Araxes, on both sides of
the river, are now leading me to the conviction that there was heavy, on-going
contact between these languages and the unusual Armenian dialects located in
the same area. Shahdagh Armenian dialects until fairly recently were quite nu-
merous on the south side of the Araxes in present day Iran.8 These dialects ex-
hibit some rather unique structural features among Armenian dialects―many of
which are identical or very similar to the unique features of Northern Tati. My
current thoughts on this situation lead me to think that formerly not only was
there heavy contact between these forms of Northern Tati and dialectal Arme-
nian (if not with other languages as well, e.g., Udi/Caucasian Albanian, Kurdish,
and dialects of Neo-Aramaic, in the form of an earlier Sprachbund situation) but
that at least some groups of the Armenians underwent a language shift to North-
ern Tati bringing along with them the types of structural influences of their
original language into the target Tati language that we find in other language
shift situations in many parts of the world. Such language shift could account for
the unique outstanding features of Northern Tati that were ultimately also intro-
duced into Northern Talyshi. It is by no means implausible that at some point
there were significant conversions of Armenian-speaking groups to Islam in the
areas south of the Araxes―as we also see north of the Araxes with Caucasian
Albanian groups (Stilo in press A), probably contemporaneously―which most

8
I have recently had direct reports from Iranians visiting the area that there are still Armenian-
speaking villages scattered among the Northern Tati-speaking villages, although both groups are in
sharp decline.
428 Donald Stilo

likely would have involved a concomitant language shift to the Northern Tati
language of the local Islamic population.
If a switch from an Armenian dialect or from Udi to Tati variants occurred for
reasons, such as conversion to Islam, it most likely would have happened at a
time before Azerbaijani Turkish became prevalent in the area, otherwise the
language shift most certainly would have been to the latter language. This sce-
nario, of course, does not preclude additional conversions to Islam accompanied
by language shifts from Armenian and/or Udi to Azerbaijani at a later time (or
times) after the introduction of Turkic speech to the region. Another benefit of
factoring in a conversion to Islam and a concomitant language shift of certain
Christian populations of the area to Northern Tati would be that such a scenario
also provides a very plausible impetus for a population movement away from the
area due to possible hostile reactions to a group that would have then been seen
as apostates. Such a population movement could thus have been the reason for
an introduction of a form of Tati from the east into the Caspian littoral area
where the prototype of Northern Talyshi was already located, eventually leading
to the melding of an unusual type of Northern Tati with Talyshi ultimately
forming present-day Northern Talyshi. There is an significant number of linguis-
tic features, especially in the verbal systems, shared by Northern and Central
Talyshi, Northern Tati, Armenian dialects, Udi and even Aramaic dialects of the
area as well as Caucasian Tat somewhat to the north of the area that do not seem
to be simple effects of diffusion but rather results of language shift situations.
The full impact of such a language shift from Armenian dialects and other lan-
guages to Northern Tati will have to remain as a topic for a separate study I have
already planned for a future time (but see brief mentions of two such features
shared by all these languages―and only these languages―in Stilo 2008: 373,
383).
We are quite fortunate in having materials in a 14th-century version of
Northern Tati (Kasravi 1946) from the Ardabil area, consisting of four conversa-
tional sentences (two from Tabriz and two from Ardabil) and 14 quatrains―the
so-called Azari fahlavīyyāt―but these materials still need a full linguistic analy-
sis to have an accurate picture of their importance for the present hypothesis.
There is, I am happy to report, a currently on-going project conducted by Dr.
Hasmik Kirakosian of the Department of Iranian Studies, Yerevan State Univer-
sity, consisting of the compilation of an “Azari Glossary” of all word forms of
some several hundred items from the fahlavīyyāt, including comprehensive
 The Polygenetic Origins of the Northern Talyshi Language 429

comparative research. Upon preliminary investigation, however, the materials

do not exhibit anything dramatically different from what we find in Northern
Tati of today.
The end result of the possible language shift mentioned above is that North-
ern Tati is quite distinct from other forms of Tati. While there is a rather smooth
linguistic continuum of transitions (even factoring in some geographic disconti-
nuity) from Southern Tati varieties south and east of Qazvin to Tāromi-type Tati
(including Kafteji and Kelāsi) and thence to Shāhrudi Tati-type (Koluri/Shāli,
etc.) of Khalkhāl and also to Kajali-type Tati to the west, but still in Khalkhāl―I
consider Tāromi, Shāhrudi and Kajali all to be Central Tati―there seems to be a
rather dramatic break in linguistic type (both in lexicon and particularly in
grammatical structures) between all forms of Southern and Central Tati, on the
one hand, and Northern Tati, on the other.
Some of the substantial differences between Northern and Central Tati are
shown in Table Four (all forms of Talyshi are ignored here, but will be intro-
duced in the next section). This material demonstrates various types of iso-
glosses that distinguish Northern Tati from Central, including both Tati-internal
developments, as well as important borrowings from either Persian, e.g., “mill”,
“three (type I, se)” in Central Tati but not in Northern Tati (see also Table Two
for Gilaki diffusion and borrowings). Certain distinctive lexemes (“three (type II,
here)”, ”want”) are included here because, even though they also appear else-
where in NW Iranian, they do not appear anywhere else in the Tati group. Im-
portant grammatical isoglosses in the verbal systems have been omitted from
this Table but are planned as the special focus of Part II of this present article (to
appear) in which I intend to demonstrate and discuss the tense systems of
Northern Tati, Northern and Central Talyshi, and Caucasian Tat and contrast
them to Southern Talyshi and Central and Southern Tati.
D) The Contribution of the Northern Tati Element to the Formation of Northern/
Central Talyshi
The point explored in this section forms the crux of the proposal I wish to put
forth in this paper: As Northern Tati spread out across Azerbaijan, it eventually
diffused eastwards until it was only separated from the Talyshi speakers by the
mountains just inland from the Caspian coast. Some Tati group(s) then moved
through the mountain passes, particularly through the Namin area to the coastal
lowlands. The influx from the plateau may not have happened all at once. There
could have been an initial strong thrust that gave the incoming group a foothold
430 Donald Stilo

Gloss Harzani Keringāni Kalasuri- Kajali Koluri

Xoynārudi
arm alingæ, alænga, æl- ælangæ bálæ, f
ælungæ ængæ
ash ölünc, olünj ulinj ulenj xal
rooster kælæbör kælay kælæ(y)ö sukulǽ suglæ
from -ri bæb, -ku, - bæb, -ku, -ku -næn, -ku, -dæ,
ondæ -ær(e) (-ra)
mill orar araqa,oraqa asiav
return-pt ö-cöröst, u-cir, u-cur u-car a-gærdest a-gærdes
ceresd
three here heri heri, here se se ~ sə
throw-pres vört dæ, fi vur-dæ vur-dæ æ-druj dæ-ræn
to, towards -(r)e -ru, -ro -ku -ku
how næni, cijur celan celæn cezen
why qeyi cunæ cuné, cunan cera
wind (*d> r) vör vu, voy vur va
Table 4: Isoglosses that show differences between Northern Tati and Central Tati (part 1) →

in the area. This first group, due to size or some sociolinguistic factor, avoided
immediate assimilation to the local Talyshi-speakers and retained its own lan-
guage. Prior to language shift, one or both of the communities, now in some form
of symbiosis, must have borrowed some lexical items, either in one direction or
in both directions, since borrowings are typical in cases of adstrata, i.e. in lan-
guage maintenance situations. Subsequent population movements could then
have further reinforced the trend of Tati influx from the plateau. At some point
the original Northern Tati speakers must have then undergone a language shift
to the local Talysh language. It is clear that both languages must have been
present in the community since elements of both sources are still present in the
Northern Talyshi dialects of today. Eventually both groups blended into one
Talyshi ethnicity and their languages fused into the version of Northern Talyshi
which then also spread out and expanded northwards and even farther south-
wards into established southern Talyshi areas, forming Central Talyshi in the
process.
Table Five shows a representative sampling of the lexical contributions of
Northern Tati to Northern, and to some extent, Central Talyshi. Table Six shows
some of the sound changes shared by Northern Tati and Northern/Central (but
not Southern) Talyshi. Table Seven presents two types of areal patterns: a) a lex-
 The Polygenetic Origins of the Northern Talyshi Language 431

Karani Gandom-ābi Nowkiāni Kelāsi Kafteji

bálæ, f. bal, m

xak
suk sukulǽ xorus, m.
-næn, -ku, - -æ, -(æ)dæ, -ku -æde, -na, -ku/- (h)æmra, -jIn, - (h)æmra, -ræ, -
æder ko dər~der jæ
asia asia, m.
a-gærdes a-gærdæs a-gærdæs

so se se so so
jir-æn fer de tav-a-dæ tav-a-dæ
-ku

cera céra céræ

bad
← Table 4: Isoglosses that show differences between Northern Tati and Central Tati (part 2)

ico-grammatical isogloss, the preverb pe-, shared by all forms of Talyshi and
some local forms of Central Tati, and b) a grammatical isogloss―the mobility of
Set1 Person-Agreement Markers9 (essentially the copula), used in the formation
of the present (or present progressive), past progressive, future (where it exists),
and the perfect tenses. For specifics of the verbal system, particularly the present

9
For the convenience of the reader, the following table showing the differences between Set1
and Set2 in some Iranian languages that have them has been taken directly from Stilo 2008 (Set1
and Set2 Person-Agreement Markers (PAMs)):

Set1 Person Agreement Markers Set2 Person Agreement Markers

(+TR Present tenses, all - TR tenses; Enclitic = Copula or AUX) (Transitive Past system; Experiencer verbs)
(enclitic only, in all three languages)
Set1a Set1b Central
(suffix) (enclitic) Central Colloquial Colloquial
Talyshi Talyshi Kurdish Talyshi
Persian Kurdish Persian

1s -ïm =im -im -æm

=ïm =im =æm
2s -ïš =iš -ī(t) -i=e, =ï =it =et
3s -ï ~ =Ø =e ~ =Ø -ē(t) ~ =Ø -e=ïš
~ =Ø =ī =eš
1p -æmon =imon -īm -im
=ïmon =mān =emun
2p -on =(i)on -in -in
=ion =tān =etun
3p -ïn =in -in -æn(d)
=ïšon =yān =ešun
432 Donald Stilo

tense formation,10 see Part II of this article (to appear) ―shared by some (possi-
bly all, data unavailable) Northern Tati dialects and two dialects of Northern
Talyshi that are absent in (Central?) and Southern Talyshi. (N/A, “Not Applica-
ble”, in Table Seven implies that the durative tenses are not formed with the
copula.)
An additional morphological feature is introduced in Table Eight: the causa-
tive marking of the verbs. The great majority of causative verbs in Tatic lan-
guages―and, indeed, in Western Iranian in general―are built on intransitive
verbs and simply function as a valency-increasing strategy, i.e., intransitive to
transitive. There are very few causative formations on transitive verbs, but a
common example does in fact appear in Table Eight, the verb “eat”, e.g., Harzani
hord- > hord-own- and Māsulei xor- > xor-amən- “eat > make eat”.11
Note that in this table there are two rather unusual causative formants that
stand out: -ovn- ~ -own- ~ -öwn- and -amən- ~ -əmən- ~ -amen-. The important
point to make here is that the former morpheme and its variants are only shared
by Northern Talyshi and Northern Tati, while the latter formant and its variants
are only shared by Southern Talyshi (Māsulei, here) and two forms of Central
Tati, Dizi and Karani, but not by Koluri (at least not as far as is known about this
language at this point) – a curious point since the villages of Diz and Karan are
geographically farther from Māsule than Kolur is and yet this isogloss skips over
Kolur located between the other areas that have this morpheme.
The larger conclusion to be made from these two unusual causative formants
is the point that has been repeated throughout this paper: Northern Talyshi
shares more with Northern Tati while Southern Talyshi shares more with (rela-
tively) local varieties of Central Tati rather than with each other, i.e., Northern
with Southern Talyshi.

10
In the languages indicated in Table Seven, Set1 (the copula as an AUX) of the periphrastic
tenses that are formed analytically commonly moves leftwards away from the verb and encliticizes
to another element in the clause, usually to the element that bears the main stress of the clause.
That is, their leftward movement is controlled by issues of information structure. As alluded to in
section C´ above, the Iranian languages of this immediate area share this feature with Armenian,
Udi, and Neo-Aramaic dialects but it is not found in other Western Iranian languages.
11
The causative of the verb “eat” is used in most Western Iranian languages in the sense of “to
make (someone) eat, to get (someone) to eat, said especially of host feeding guests well or a lot”. It
is not used in a simple, neutral sense of “to feed”, such as “to feed a baby”.
 The Polygenetic Origins of the Northern Talyshi Language 433

Gloss Harzani Keringāni Kalasuri N.Talyshi Vizanei

bee müz müz, müz- müz muz, meš, kandïmüz meš
ængin
break-pres (trans) sənd send send sïndïn e-št

break-past (intrans) sisd sist sist si, pe-si e-(š)ši

buy-pres asda(n) ta, (i)štæn estæ stæn, hïr stæn,

hər
buy-past æstær, astar ti, tæ, te estær sæ~stæ sæ, həri

calf (names change ac- güg gügə, güj, quq, gug, poynæ
cording to age, sex of calf) jovno
cat kete keta, ketæ kete kïtï, pišik
clothes ölæt ulat, ulæt olat, olæt olæt olæt
early ræv ræv ræv Ræ ræv
go around-pres cor oy car var ~vur, car-u ca (v)o, gard, næv
næv
laugh-pres sər, sir ser ser sïr ser
lip dodax, lev löbüt löbüd lïpüt, lïv lev
moon, moonlight öšmæ, ušmæ ušma, ušmæ ušmæ ošum
mouse mora mura moræ
mouth gæv gæv, qæv gæv, qæv gæv
neck giri geri gïy
play, game hüyæ (N) tamašu (N) hüæ (N), hænæk (N)

pull-pres (various senses) kərn, keš keren ker(en) o-k(ï)rïn, dæ-k(ï)rïn, šek(ï)rïn,
o-vinjïn, o-vïškïn, o-rïndïn
pull-past kəre, kešd kerni ker(e)nir (dæ-) krïn~kïrni, šekïrni, dæ-
o-vïnjïni, o-vïškïni, o-rïndïni kərni
search for-pres næv næv næv
sleep (noun) hun hun han han
small, little vede, vedek güdæ, kicik güdæ gædæ(li), ruk, jïk(ï)læ gædæ
sneeze-pr šesn(en) šešnæv hešnev e-fešïn
snot vene tər, lemb lemb lïmb
vene övey
stomach gæd qæda, qædæ gædæ, lævæ, luz, lævæ-ruæ lævæ
sun hæræšï, ruz hæræši, rüz. rüz. hæši
want, must-past höšd pist pist Pi pi
year sor sor sar, sur, sor sor
sor

Table 5: Lexical Isoglosses shared by Northern Tati and Northern Talyshi (part 1) →
 436 Donald Stilo

C?N? Tati NORTHERN TATI

GLOSS Kajali Harzani Keringāni Xoynā-rudi Kalasuri
*æ > a (and > o in Harzani and sporadically in Keringani)
chicken kǽrgæ, f. kork kork, kærg kærg~k, karg,
korg
cook-pt (±Trans) pæt pot pat pæt pat
do, make-past kærd kord kard gad gad~god
eat-past hærd hord hard hard hard, hord
horse osb osb, asp, æsp asp asp, osp
I (1st sg. pronoun)
needle dorzan darzen, darzæn
dærzen
snow vohor vær vohor, vahar
wolf værg vorg varg, vorg, varg, vorg
værg

a > o (also u, ü, and ö)

come-pres a(y) boy (oy?) uy, öy u, a
for -ræ, -ra cün -ru -ro
good cak cök, cok cuk cuk
say-past vat ~ hat öt vut, vot vut, vot ut, ot
star astaræ ulduz usdura, ostro
wind vör vu, voy vur
*rd > r (and velarized /L/ in Asalemi)
spring (season) ævæsor ævesor ævæsor
Year sor sur, sor sur sur, sor
Table 6: Sound shifts shared by Northern Tati and Northern Talyshi (part 1) →

C.Tati Northern Tati Northern Talyshi

FEATURE Kajali Harzani Keringāni Xoynarudi Kalasuri Lerik Astara Lenkoran Masally Jowkandāni Vizænei
preverb æræ- pe- pe- pe- pe- pe- pe- pe- pe- pe- pe-
(sense:
“up”)
Set1 N/A* ? ? + + + + - - ? ?
fronted
in
Durative
Tenses
Table 7: Grammatical material shared by Northern Tati and Northern Talyshi (part 1) →
*N/A “Not Applicable” implies that the Set1 clitics as copula are not employed n the formation of the durative
tenses
 The Polygenetic Origins of the Northern Talyshi Language 437

TALYSHI “DIALECTS”
GLOSS Lerik Vizanei Asālemi Māsāli Māsulei
*æ > a (and > o in Harzani and sporadically in Keringani)
chicken kag kag kærg kærg

cook-pt (±Trans) pat pat pæt pæt pætt

do, make-past ka(rd) kard kærd kærd
eat-past ha(rd) ha(rd), hærd (h)ærd hærd hærd
horse asp asp æsp æsp
I (1st sg. pronoun) az az æz, mən æz, mə(n)
needle darzæn dærzæn dærzən

snow va, voæ vær vær

wolf neci; Astara værg værg værg
coastal Shiite
dialect: vag

a > o (also u, ü, and ö)

come-pres (v)o boy (oy?) a a a
for -ro -ro -ra -ra -ræ ~ -ra
good cok cok cok xub xub
say-past vot ~ hot vot vat vat
star astovæ əstaræši setaræ setare
wind vo va va va
*rd > r (and velarized /L/ in Asalemi)
spring (season) ævæsor ævæsor ævæsaL
Year sor sor sal sal
← Table 6: Sound shifts shared by Northern Tati and Northern Talyshi (part 2) →

C.Tati C. S. Talyshi C. S. Talyshi

FEATURE Kajali Asālemi Māsāli Māsulei Koluri Karani Gændomābi Kelāsi Kæfteji
preverb æræ- pe- pi- pe~i pi- a- u- u- u-
(sense:
“up”)
Set1 N/A* ? N/A N/A N/A N/A N/A N/A N/A
fronted
in
Durative
Tenses
← Table 7: Grammatical material shared by Northern Tati and Northern Talyshi (part 2)
438 Donald Stilo

CENTRAL TATI
GLOSS Koluri Karani Gandom-ābi Kelāsi Kafteji
*æ > a (and > o in Harzani and sporadically in Keringani)
chicken kærg kǽrgæ, f. kǽrgæ, f. kærg kǽrgæ, f.

cook-pt (±Trans) pæt pæjes pæjes

do, make-past kærd kæ(rd)
eat-past hær (pres) xwærd xwæ(rd) xa(rd) xard
horse æsb æsb æsp æsb
I (1st sg. pronoun) æz, mæn
needle dærzen

snow vær væhær

wolf værg værg værg

a > o (also u, ü, and ö)

come-pres a a a a a
for -ra -ra -ra -ra
good cæk, cak cök nazz xorom xorəm
say-past vat vat vat vat vat
star astaræ setaræ
wind va bad
*rd > r (and velarized /L/ in Asalemi)
spring (season) beháræ
Year sal
← Table 6: Sound shifts shared by Northern Tati and Northern Talyshi (part 3)

With one exception, all the other causative formants and processes men-
tioned for all the languages and dialects in Table Eight can essentially be ignored
for the purposes of this paper since either the formant -Vn- (and it variants) or
the lengthening of the root vowel in combination with the causative formant are
common elsewhere in Western Iranian and do not contribute any additional
supporting points of interest for the relationship of Tati to Talyshi. The only
other interesting causative formant is -enjen-, found (so far) only in Gandomābi
and Kajali but both these dialects are within Central Tati and likewise do not
contribute any additional supporting points of interest for the relationship of
Tati to Talyshi. It is interesting to note, however, that these two dialects are not
geographically very close to each other (at least in their present locations―there
 The Polygenetic Origins of the Northern Talyshi Language 439

may have been earlier population movements) and there are very many other
forms of Central Tati located between them such as Koluri and all the Shāhrudi
dialects. Gandomāb, in fact, is closer to Māsule than to Kolur12 is and yet the dia-
lects of the latter two locations do not (as far as we know) share this causative
morpheme.
Causative- Examples (Valency change: Intransitive>Transitive via Causativization):
marker
Harzani 1) -ovn- 1) toros->tors-ovn- “fear”; neš->neš-övn- “sit”; gel->gel-övn- “boil”;
sər->sər-own- “laugh”; səsd->səsd-own- “break”; hord->hord-own- “eat”;
yær->yær-öwn- “hit”; kešd->kešd-öwn- “pull”; self->self-öwn- “cough”;
dæšd->dæšd-öwn- “hurt, ache”; beræmesd->ber-öwn-esd “cry”; etc.
2) vowel 2) rasd->rösir- “arrive”; sisd->səsd- “break”; ceresd->corord- “turn”;
change
3) -n- + 3) levesd- > löv-n-esd- “move”; ræhisd- > röh-n-ösd- “finish”; rasd- >
vowel rösir-~ rös-n “arrive (for pres only)”;
change
Northern 1) -ovn-> 1) vit- > vit-ovn- “make run”; dæ-mand- > dæ-mand-ovn “stop,” gïl- >
Talyshi gïl-ovn- “boil”
2) length- 2) gard- > gord-ovn, (næv > nov-n “go around”);
ening of
root vowel
+ -ovn-
3) length- 3) -n- + lengthening of vowel: ræs > ros-n “arrived”; væš > voš-n “burn”;
ening of
vowel + -n-
Asālemi 1) length- 1) væz- > vaz-ən- “jump,” cæk- > cɔk-ən- “break,” beræm- > beram-en-
ening of “cry”
vowel + -ən-
~ -en-
2) -an- 2) xəs- > xəs-an- “sleep”; xur- > xur-an- “laugh”;
Kajali 1: a-ckan [a-s.kan] “stick!”, “cry” vræm > vræman, glan “boil”;
3: “run” bettæj > bettajen, “crack, burst” tæræk > t∂raken,
8: šk-i- > šk-enjen- “break”, “mount” æv-nes. > æv-sanjen (both comm
forms);

Gandom- 1) length- 1) dæv- > dav-en- “run”; tærs- > tars-en- “frighten”, væz- > vaz-en-
ābi ening of “jump”;
vowel + -en-
-enjen- Šeg- > šk-enjen- “break”, neš- > neš-enjen- “sit > seat”

12
See http://www.geographic.org/geographic_names/name.php?uni=-4299541&fid=2741&c=
iran.
440 Donald Stilo

Kafteji -In- ~ -ən- gol- > gol-In- “boil,” šg-y- > šg-ən- “break”
Kelāsi -on- ~ -In- væz- > væz-on- “jump”; suj- > suj-on- “burn”; tærs- > tærs-on- “frighten”;
juš > juš-on- “boil”; de-xos- > de-xos-ïn- “sleep”;
Koluri (optional Lengthening of vowel: cæk- > cak-en-es “break”; gærd- > gard-en-es “go
lengthening around; væš- > vaš-en-es “burn”; væz- > vaz-en-es “jump”
of vowel) + no length: xes- > xes-an-es- “sleep”
-ən-
Māsulei 1) -amən-, pe-gres-amən- ”kindle,” ez- > ez-amən- “rise,” dæ-cik- > dæ-cik-amən-
-əmən- “stick”; gIl- > gIl-amən- “boil”; xor- > xor-amən- “eat”; pær- > par-amən-
“fly”; pic- > pic-amən- “twist, wrap”; tell- > tell-amən- “run”; tæræk- >
tæræk-amən- “crack, chip (variant I)”; dæ-viær- > dæ-viær-amən- “pass,
get across”;
a-rəs- > a-rəs-əmən- “catch up with (variant I)”; tærs- > tærs-əmən- “fear >
frighten”); je-xəs- > je-xəs-əmən- “hide”;
2) length- cær- > car-əmən- “graze (variant I)”;
ening of væz- > vaz-amən- “jump”; ve-zer- > ve-zer-amən- “rip, tear (variant I)”;
vowel +
-ə/amən-
3) -ən- ~ a-rəs- > a-rəs-ən- “catch up with (variant II)”; ve-zer- > ve-zer-ən- “rip, tear
-an- (variant II)”;
tæræk- > tæræk-an- “crack, chip (variant II)”; ve-pilick- > ve-pilick-an- ~ ve-
pick- > ve-pick-an- ~ “wring, squeeze”;
4) length- cæk- > cak-en- “break”; cær- > car-ən- “graze (variant II)”; læk- > lak-ən-
ening of “fall”; je-væz- > je-vaz-ən- “separate”;
vowel + -ən-
~ -en-
Dizi -amen- ~ de-neš- > de-neš-amen- “mount”; cek- > cek-amen-~cek-an- “stick”
-an-
Karani lengthening cær- > car-en- “graze”
of vowel +
-amen-
Table 8: Grammatical material: The causative morpheme shared by Northern Tati and
Northern Talyshi

E) Mixed Nature of Southern Talyshi

The merger of two different sources in the formation of Northern Talyshi may
not be the only reason for the divergence between Northern/Central Talyshi and
Southern Talyshi. The latter cluster of dialects also seems to have a certain
amount of its own mixture (probably diachronically more recent) with the clos-
est forms of Tati. It is in fact quite difficult, at least at this point, to determine
whether the common features shared by Koluri-type Tati and the Talyshi of
 The Polygenetic Origins of the Northern Talyshi Language 441

Māsule, Māsāl, Shānderman, etc. as seen in Table Two derive from their com-
mon history as opposed to later mutual diffusion between the two groups.
Such an intertwining of Southern Talyshi and Koluri-type Tati―or at least a
certain amount of mutual borrowing and diffusion―is not at all implausible or
unlikely. According to my Māsulei informant, who provided me with extensive
materials on his dialect in 1976, Kolur and the Tati-speaking villages of Upper
Tārom are separated from Māsule and Māsāl only by a distance of a two-day trek
on foot through mountain passes (see Map Two) and that the villagers of Tārom,
specifically the Tati-speaking villages of Gandomāb, Siāvarud, and Hezārrud (but
not Gilavān or Nowkiān), were still transhumant spending the summer months
in tents in their summer camps (yeylāq), which were no more than a three- to
four-hour walk from the summer camps of Māsule and Māsāl on the slopes just
on the other side of the mountains. On this topic Yarshater (1959: 63) states: “In
general, there are important factors contributing towards the assimilation of
these two languages to each other. Many Shāhrudis spend more than half the
year working in southern Tālesh, especially in Shānderman and Māsāl, and a
number of Tāleshis use Shāhrud as their regular yeylāq [summer camp]. Tālesh
supplies Shāhrud with rice, wool, and dairy products, and buys fruit in return.
Girva, the most popular yeylāq of Shānderman and Māsāl, is almost within
Shāhrud. Indeed, most Shāhrudis can also speak Tāleši.
In considering the position of the Iranian dialects of Āzarbāijān, I believe one
should take Shāli, Kuluri (i.e. Shāhrudi) and Shāndermani, Māsāli, and Māsule’i
Tāleši as close members of the same group rather than separate dialects”.
A certain percentage of Shāhrudis over generations would probably have
permanently remained in Māsule and Māsāl. There must have been constant in-
terchange and probably intermarriage throughout generations between these
two groups, thus creating an open route for diffusion of linguistic features in
both directions.
442 Donald Stilo

Map 2.

In this way the absence of the Augment in the formation of the Imperfect
tense in Southern Talyshi dialects may possibly be attributed to Central Tati in-
fluence. Note the use of the Augment in the Imperfect form of the verb, e.g., “I
knew”, in Northern Talyshi (Lerik dialect) æ-zïn-i-m and Central Talyshi (Asāle-
mi) æ-zon-i-m vs. its absence in Southern Talyshi (Māsulei) zon-i-m and Koluri
(Central Tati) zan-i-m. It is too soon at this stage of this investigation to say for
sure why the Augment has been preserved on the one hand in most forms of
 The Polygenetic Origins of the Northern Talyshi Language 443

Talyshi―even in areas geographically close to Southern Talyshi (but probably

still structurally closer to the central dialects), e.g., Tālesh-Dolābi as mentioned
above ―but on the other hand has disappeared from most of Southern Talyshi.
The above explanation of contact with Central Tati, especially Koluri and the
Shāhrudi dialects, seems to be the most plausible reason.
On the other hand, the use of the present stem in the formation of the Im-
perfect in Koluri as shown in Table One may be a contact-induced structural
borrowing going in the opposite direction. That is, while this feature may be an
inherited feature common to both the various Talyshi dialects and Koluri, it is
also possible that through diffusion it later passed from local Southern Talyshi
into Koluri. Please note, however, that this formation is not found in Tati outside
this area restricted to Shāhrud, as we shall see next.
While the Imperfect formant -i- is found elsewhere in the area (see footnote 7
and the accompanying discussion above), it is only in Shāli/Koluri―at least as
far as we know at this point in the research of Tati languages―and all forms of
Talyshi that it is added to the present stem of the verb (in addition to the pre-
fixed Augment in Central and Northern Talyshi). I must stress, however, that the
suffix -i- as an Imperfect marker is a somewhat more widespread areal feature
throughout this area, whether used with the present stem as in Northern, Cen-
tral, and Southern Talyshi and in Koluri as just shown, or with the past stem as in
Western Gilaki (e.g., Rashti) danəst-i-m “I knew” and some other forms of Central
Tati. It is does not extend to many other languages in the area, cf. Hezārrudi
(Central Tati, Tāromi-type, Upper Tārom) mi-kæt-im “I used to fall”, Kabatei/Kaf-
teji (Central Tati, Tāromi-type, Gilan): me-zanəst-əm and Lahijani (Eastern Gila-
ki) Ø-donəst-əm “I knew (Imperfect)”. Note also that the Imperfect marker -i- is
similarly used in the past subjunctive in Talyshi and Koluri: Lerik æ-p-i-m-bæ
b-æ-š-i-m, Māsulei x-u-m b-ə́ -š-i-m,13 and Koluri: xa-i-m bé-š-i-m, vs. its absence in
Kelāsi: me-gost-em bÍ-š-im or Western Gilaki: xast-i-m bə́-š-əm “I wanted to go”.

13
The -u- in x-u-m “I wanted” is a regular Māsulei development occurring when the Imperfect
formant -i- directly follows a root vowel -a-. Thus xum “I wanted (durative, Fr. je voulais)” has an
underlying form of *xa-i-m (as found in other dialects, e.g., Koluri), cf. Māsulei ša-m “I can” > š-u-m
(< *ša-i-m) “I could (durative, Fr. je pouvais)”, va-m “I say” > v-u-m “I used to say”, a-m “I come” > u-
m “I used to come”. These Imperfect forms given here also have alternates with -i- > -y-: x-u-y-m “I
wanted”, š-u-y-m “I could”, etc. Note that the -u- in these Imperfect forms has nothing to do with
the past stem, which appears more clearly in the Preterit forms: (Māsulei) xast-em-æ “I wanted
(punctual, Fr. j’ai voulu)”, šas(s)-em-æ “I could (punctual, Fr. j’ai pu)”, vat-em-æ “I said”, ome-m-æ “I
came”, respectively.
444 Donald Stilo

Degree of Difference between Northern Tati and Northern Talyshi

While the contribution of Northern Tati to the lexical and grammatical composi-
tion of Northern Talyshi is substantial, one should not get the impression that
these two language groups are by any means identical or that Northern Talyshi is
merely a dialect in the Northern Tati continuum. In addition to similarities be-
tween the two groups, there are also many lexical and grammatical differences
between Northern Tati and Northern Talyshi. Table Nine shows some important
lexical isoglosses that distinguish these two groups:
GLOSS Harzani Keringani Kalasuri Lerik Asalemi Masali Masulei
arm alingæ, alænga, ælangæ bol, kæš baL bazu
ælungæ ælængæ
ash ölünc, olünj ulinj ulenj xok xak xakæbaš, xakæ
xakǽ
barefoot pörö-pö pure-pu gujli pa-vïrænæ
belonging yin, ki kin Ø, gəLe šen šen
to
blood vun vun vün xun xun xun xun
can-past æsd æst vïški, zïn šast ša šass
chick jüjæ jüjæ jüjæ kižæ, veræ kijæ gijæ, kijæ kiæ
cough selfest kekivest kekest e-kræ, e-tox xəs kærd
(past)
down (særi-) yær-kano, žæ:ru, beni bæžyo, saru, jir jir jer
pæræri, ære yær-sæ, žiy(ædæ)
pærifiš, pori
drink-past yenjir hard, hærd hur qa pe-šom hənt hint xənt
each other ivæ-din ive-dun yandï, yadï yandə diædə, iví
bæyandə,
bəndí
here ənta emgü, üngü, engü, éngu, íyo, í-vrædæ ía ya ya
ingü engö
itch-past tor kord viri hæye xarIš kær kəlaše
large, big kælæ ketmæ yækæ yol yaL, kæL pillæ pil
rooster kælæbör, kælay kælæ(y)ö sük, süklæ, suk suklæ suk
xöruz horüz
sell (past) höröt, horot hirišt, ferašt herat h(æ)vat xræt x(ə)ræt xorot
three here heri heri, here se se se
to-1 -(r)e -ru, -ro -ku, bæ_-ku -ka -ku
to-2 andæ bæ- bæ
weave- ömöšd e-tæn, boft væt væji, vætt vætt
past
 The Polygenetic Origins of the Northern Talyshi Language 445

how næni, cijur celan celæn cóknæ cïntǽ

why qeyi cunæ cuné, bó-ci(ro), co ce bæ, (céra) əcəra
cunan, cen
with-1 -(o)hun -anæ, -onæ, dï (bæ-_) - -næ -næn
-(y)an næ
with-2 -ra, ba
Table 9: Differences between Northern Tati and Northern Talyshi

In addition to lexical items, some of which are listed in Table Nine, gram-
matical differences between Northern Tati and Northern Talyshi include the im-
perfect formation―see, for example, the discussion above of the use of the
Augment and -i with the present stem in most forms of Talyshi to form the Im-
perfect tense vs. the absence of both features in Northern Tati. Note how differ-
ent the structures of the same Imperfect form is in these two languages:
(Northern Talyshi) æ-hæ-i-m
AUGMENT-eat:PRES-IMPERFECT-1S1

Kalāsuri (Northern Tati) be-hord-e=u=m

DURATIVE-eat:PASTINFIN=COP:PAST=1S1
“I used to eat”
The negative forms of the Imperfect tense in both languages show even
greater structural differences between the two languages, due to the floating na-
ture of the clitics in Northern Tati:
(Northern Talyshi) n-æ-hæ-i-m
NEG-AUGMENT-eat:PRES-IMPERFECT-1S1

Kalāsuri (Northern Tati) ne=u=m be-hord-e

NEG-=COP:PAST=1S1 DURATIVE-eat:PAST-INFIN
“I would not eat”
Among other morphological features that distinguish these two groups,
Northern Tati retains distinct direct and oblique nominal case forms in the plu-
ral, whereas there is only one form (the original oblique) for both cases in North-
ern Talyshi (but they are still distinguished in Southern Talyshi, e.g., Māsulei). In
addition, Northern Tati exclusively has postpositions while Northern Talyshi has
postpositions, prepositions and circumpositions, for which see Stilo 2009.

3.0 MUTUAL INTELLIGIBILITY

Aside from the clear lexical and structural differences between Northern, Central
and Southern Talyshi discussed throughout this paper, all of which add to im-
446 Donald Stilo

peding mutual intelligibility between these forms of Talyshi, it would be valuable

to have more formal testing that directly targets the issue of degree of intel-
ligibility to see if these results help us establish the status of separate languages
or subdialects of one language for Northern, Central and Southern Talyshi. Es-
tablishing absolutes in the degree of intelligibility between variants of the same
language or between closely related languages is an intricate and elusive topic.
And even if we can measure interintelligibility, at what point to we draw a cut-
off line between language and dialect?
Practically speaking, mutual intelligibility can really only be determined ten-
tatively even with extensive testing. Such formal testing is a complex, arduous
and multifaceted task that would require devising, implementing and evaluating
accurate testing procedures based on formal criteria and such testing has not
been an integral part of my research on Talyshi. On the other hand, however, I
can cite two concrete examples that fall into the realm of assessing mutual intel-
ligibility of Talyshi dialects: one informal in type and one anecdotal.
The informal testing conducted consisted of taking a simple recorded natural
narrative of one minute and 18 seconds in length, of excellent sound quality, that
I recorded in October, 2003 in a dialect of the Astara zone of Northern Talyshi
and playing it several times for a native speaker of the Māsāli dialect of Southern
Talyshi. The results were striking and the reaction of the listener fully confirmed
what I had suspected would happen all along: The Māsāli speaker could not even
determine the gist of the narrative and ultimately claimed that he could catch no
more than two two-word phrases (“my son”, “my mother”). His reaction was one
of shock. He had fully expected to be able to understand another “dialect” of “his
language” but in fact ultimately exclaimed, “How can you even consider this a
dialect of Talyshi?”.
The anecdotal information I have on this topic was related to me by the
scholar personally involved in the incident, which represented the opposite situ-
ation of the one just described. In this case, a speaker of a dialect of the Astara
dialect zone of Northern Talyshi―the late Novruz-Ali Mammadov, a linguist at
the Academy of Sciences in Baku, who was himself a Talysh working on the Taly-
shi language―traveled to Māsāl in Iran specifically to see how much of the
Māsāli dialect of Southern Talyshi he could understand. His reaction exactly mir-
rored the reaction of the Māsāli speaker mentioned above: “I could understand
nothing of the Māsāli dialect when people spoke naturally or told stories”.
 The Polygenetic Origins of the Northern Talyshi Language 447

Mention should be made in the middle of this discussion that I am basing my

evaluations of mutual intelligibility upon reactions of native speakers of various
types of Talyshi that have not been previously exposed to each other’s dialects.
On the other hand, however, since the collapse of the former Soviet Union there
is a strong drive among the Talysh on both sides of the present border to make
contact with and visit each other with the aim of furthering linguistic and cul-
tural ties, promoting language research and preserving the Talyshi language and
culture. While the Northern and Southern dialects are as widely diverging as I
have shown in this paper, there is, on the other hand, a strong tendency―but
practically speaking, among probably no more than a handful of people in to-
tal―on both sides to learn as much as possible from each other’s dialects and to
adapt, or at least accommodate, one’s own speech forms to those of the listeners
from other dialects. At the same time, however, those speakers who find them-
selves in these situations but who have had less contact with other’s dialects and
have not yet learned to follow such types of adaptive strategies, consistently
have to resort to a third language―Persian, Russian or English―in order to
communicate.
A few examples of simple sentences formed with everyday lexical material
will suffice to show the range of mutual intelligibility between Northern and
Southern Talyshi.
It can occasionally occur that a sentence in all forms of Talyshi are indeed
very close:
N.T. (Leriki): cævon kinæ bo cæmæ zoæ=ro dï gïlæ hïšk-æ šæy=š vardæ
S.T. (Māsulei): coun kinæ cæmæ zuæ=ra də gIlæ xəšk-æ šey=š voærdæ
gloss their daughter for our son-for 2 CLASS dry-CONN shirt-3S brought
English “Their daughter brought two dry shirts for our son.”
(bo_=ro “for” is a circumposition in Northern Talyshi)
N.T. (Leriki): vot=ïšun-æ ki n-æ-zon-i-n ïštï zuæ nom cic=e.
S.T. (Māsulei): vat=əšun-æ ki n-æ-zon-i-n əštə zoæ nom c=æ.
gloss said-3P that NG-AUG-know-IMPERF-3P your son name what-is
English “They said that they didn’t know what your son’s name was (lit: is).”
Even more sentences composed of average, basic words would not be intelli-
gible:
N.T. (Leriki): tï cic vot=dæ-š? Alternate: tï cic=ïš vot=dæ?
gloss you what say-TAM-2s you what-2s say-TAM
S.T. (Māsulei): tə əcə va-Ø-y?
gloss you what say-TAM-2s
English “What do you say?”
448 Donald Stilo

N.T. (Leriki): æv kovra še=dæ-Ø?

gloss he where go-TAM-3s
S.T. (Māsulei): æ kəræ ka šu-Ø?
gloss he PROG where go-TAM-3s
English “Where is he going?”
N.T. (Leriki): æγïl-í lævæ dæž-e=dæ-Ø
gloss child-OBL stomach hurt-INF-TAM-3s
S.T. (Māsulei): xIrdæn-í Iškæm xïs-Ø-I.
gloss child-OBL stomach hurt-TAM-3s
English “The child’s stomach hurts.”
N.T. (Leriki): maški æyo=š?
gloss tomorrow here-2s
S.T. (Māsulei): sæba ha mænn=ešæ?
gloss tomorrow here BE.LOCATION-2s
English “Will you be here tomorrow?”
N.T. (Leriki): pæs-í de ter-í bæ=pe o-kïrïn!
gloss sheep-OBL with rope-OBL to-up PVB-pull
S.T. (Māsulei): gusænd-é lafïn-é=næn kæfa=ku pir-ïnj!
gloss sheep-OBL rope-OBL-with up-to PVB-pull
English “Pull the sheep up with the rope!”
Most sentences would be somewhere between these two poles of intelligibil-
ity, but probably more towards the lower end. As we saw earlier, it is especially
the drastically different verbal systems that impede intelligibility between
Northern and Southern Talyshi, particularly the use of the past stem for the for-
mation of the present tense, the existence of the progressive tenses in Southern
Talyshi only and a formal future in Northern Talyshi only, the differences in, or
the impermissibility of, the leftward mobility of Set1 and Set2 PAM clitics, among
other features:
Māsulei Leriki Meaning
nam (= n-a-m) ome-dæ nimæ I don’t come
vam (= va-m) votédæ=m ~ vótdæ=m I say
vam (= va-m) bævoté=m I will say
kaške b-u-Ø kaške bæ-vo-i-Ø I wish he would come!
kaške bə́ -vuš! kaške bæ-vot-i-š! I wish you would say/
had said (something)!
šu(y)š, šu ævəškiš, ævəški you could, he could, etc.

4.0 LANGUAGE “MIXING” OR ”INTERTWINING”

The concept of a “mixed” language is not a new phenomenon in the linguistic lit-
erature, but until fairly recently was not clearly understood or well documented.
 The Polygenetic Origins of the Northern Talyshi Language 449

Various types and intensities of contact-induced language change situations

have been much better investigated by linguists in recent decades, especially be-
ginning with the publication of Thomason and Kaufman (1988). Then in very re-
cent years the concept of the most intensive stages of language contact resulting
in language “mixing” has been the center of great linguistic attention with the
appearance of more and more solidly documented literature on the phenome-
non with concrete case studies from many different parts of the world.
Publications, such as Matras and Bakker (2003), Bakker and Mous (1994),
Thomason (2003), Thomason and Terrence (1988), among others, have focused
on the phenomenon of highly mixed languages. It is clear that most languages on
the earth show some sort of influence, in varying degrees, from other speech
forms beyond their own language borders―take a representative sampling such
as: the high percentage of lexical borrowing from Italian/Sicilian into Maltese,
the even higher percentage of words of Anglo-Norman and Latinate origin in
English, the borrowing of new phonemes into German via loanwords from
French, English and other sources (Gelee, Dschungel, einchecken, Chance), the
problem of the existence of retroflex consonants in Sanskrit, the heavy cultural
component of Chinese loans in Japanese, Korean and Vietnamese, the usual
placement of the verb in sentence-final position in Central Asian Arabic and
most forms of Neo-Aramaic, to name a few.
The type of language mixing to which I refer in this paper, however, typically
has a much more deep-seated, more far-reaching effect on the entire language
than just the existence of loanwords, new phonemes or even an occasional ex-
otic word order pattern. At the level of Mixed Language (ML), languages usually
show “native”, as well as “borrowed” sources on all grammatical levels, including
the realms of function words, syntax, prosody (the least understood, most un-
derinvestigated and usually neglected domain) and even morphology, in addi-
tion to the usual presence of loanwords and new phonemes that often come
along with loanwords. Two languages may become so intimately intertwined in
the formation of an ML that the resulting hybrid confounds the linguist in the
genetic classification of the language.
These studies have also brought a certain sensitivity to the study of this phe-
nomenon by providing linguists with new terminology to help us obviate the
problem of inaccurate analogies, terms, such as “language intertwining” (Bakker/
Mous 1994), “code-copying” and “structural attractiveness”(Johanson 2002). My
intent in using the word “polygenesis” here is in the same vein.
450 Donald Stilo

The present paper approaches the topic from a slightly different perspective
than that examined in most ML studies cited above. While it introduces the type
of heavy language intertwining in the degree of intensity discussed above, it fo-
cuses on a case of close genealogical linguistic relatives participating in the inter-
twining. That is to say, in the case of Talyshi and Northern Tati, as was already
discussed above, the original speech forms that led to the blending as shown
here in Northern and Central Talyshi are on the one hand quite closely related in
that they both belong within the Tatic subgroup of Northwest Iranian but on the
other hand, as we have seen, are sufficiently different from one another in type
that they present no difficult task in identifying the individual elements that
went into making this particular resulting intertwined language. The only diffi-
culty that arises in looking at language intertwining with closely related lan-
guages, such as Tati and Talyshi, is having to set aside so many important aspects
of the languages, e.g., function words (ægæ “if”, the general subordinator ke, etc.)
that are shared by all or most Northwest Iranian languages or indeed even by
most West Iranian languages. As I have shown above, the elements from both
Talyshi and Northern Tati sources that have contributed to the formation of
Northern and Central Talyshi are in the domains of lexicon, morphology, syntax,
and tense formation typology, among others. In investigating the genesis of
Northern Talyshi, we can say that all levels have been affected.

5.0 FUTURE DIRECTIONS

Refinements on the proposal of the origins of Northern and Central Talyshi pre-
sented here are quite necessary and will indeed emerge. It is, of course, possible
that other historical scenarios may have led to the current Talyshi situation and I
am in the process of considering various refinements or alternatives to the pre-
sent proposal. Many issues still need to be investigated and many questions need
to be answered. Of the most likely alternate explanations or supplementary sce-
narios, the following will especially need attention.
1) Did a Tati group move off the plateau to the Caspian littoral where Talyshi is
presently located as posited here, or is the exact opposite also possible? That is,
are the Northern Tati groups, particularly Kalāsuri, Xoynarudi, Keringāni a result
of a migration of a group of Northern Talysh up onto the plateau? There are vari-
ous place names in north-central Iranian Azerbaijan that contain the word tāleš.
Such an alternate proposal is not unthinkable. In fact, Mortazavi (1962: 456-7)
states: “The old men say that the Harzandis (the people of Old Harzand), which is
the original center of the Harzani language and also the former center and the
 The Polygenetic Origins of the Northern Talyshi Language 451

source for the other villages of this settlement, came from Țavāleš and settled in
this area...”.
On the other hand, however, “ethnic remembrances” are not always com-
pletely accurate. If the Tati speakers indeed moved eastwards off the plateau
down to the coast as suggested in my proposal, they could easily have retained
their ties to their original villages on the plateau and been in constant contact
and movement back and forth, bringing families born in the Talyshi zone with
them, thus accounting for the Talyshi-type place names and origins legend there.
By the same token, I have heard many such “ethnic remembrances” among the
Northern Talysh in Azerbaijan that they came from Gilan.
2) Further investigation of the relationship of Southern Talyshi and Shāhrudi-
type Tati is needed to understand the genesis of those languages and the mutual
influence they have exerted on each other. What were the probable socio-linguis-
tic processes and the historical reality (including alternate scenarios) that led to
this situation?
3) Why is the Northern Tati group so different from other forms of Tati, viz. the
unusual verbal system? What is the relationship of this group to Zazaki/ Dimli, as
already pointed out in Henning (1954)? See also Vahman and Asatrian (1990) for a
discussion of the shared lexical features of Zazaki/Dimli and Northern Tati dia-
lects, establishing a special relationship between these two groups, as well as
other heretofore unrecognized connections to other West Iranian languages.
4) What role did language shift situations, say from Armenian (particularly Ar-
menian dialects) or Udi―which certainly occurred at some point in the history
of these languages―play in the grammatical restructuring of the Northern Tati
group?
5) Did Caucasian Tat play any role in the formation of Northern Talyshi? Are
there any contact phenomena or substratum issues involving Northern Tati and
Caucasian Tat, e.g., the similarities in the verbal systems and the rhotacism of *-d
in Caucasian Tat and Kalāsuri-Xoynarudi? What was the former geographic ex-
tent of both Caucasian Tat and Northern Tati (Old Azeri)? What is the source of
the copious Northwest Iranian elements in Caucasian Tat?
6) What is the contact relationship between Gilaki and Southern Talyshi? As dis-
cussed above in section 1.2 (B´) has there been only diffusion in one direction,
from Gilaki into Talyshi? Is there a Talyshi substratum to western Gilaki not
shared by eastern Gilaki? Or both?
7) What is the role of areal phenomena involving Iranian and non-Iranian lan-
guages of the southern Caucasus, the Caspian littoral, and the northern Iranian
Plateau? A related and possibly pivotal study would be to examine the correla-
452 Donald Stilo

tions of DNA markers defining (and possibly uniting) the populations of these ar-
eas historically, helping us to document potential cases of language shift.

6.0 CONCLUSION
The study of language intertwining, especially in the case of genealogically re-
lated languages is complex, the data is bewildering, and the histories are not
well known. At the same time, however, it is clear from the overwhelming num-
ber of lexical, and grammatical isoglosses shared by Northern Tati and North-
ern/Central Talyshi that some special relationship deviating from a typical
Stammbaum, tree-type model between these two groups exists. The proposals
presented here should be considered a first step in unraveling this multi-faceted
problem.

BIBLIOGRAPHY
Arends, J. / P. Muysken / N. Smith (eds.), (1995), Pidgins and Creoles: An Introduction, Amsterdam.
Asatrian, G. (2009), “Prolegomena to the Study of the Kurds”, Iran and the Caucasus 13.1: 1-58.
―― (ed.) (2011), Vvedenie v istoriyu I kul’turu talyšskogo naroda, Yerevan.
Baer, E. (1938), “Il metodo della geografia linguistica applicator all’investigazione dei dialetti irani-
ci”, Atti del xix Congresso internazionale degli orientalisti, Rome: 233-239.
Bakker, P. (1997), “A Language of Our Own”. The Genesis of Michif—The Mixed Cree-French Lan-
guage of the Canadian Métis, New York.
―― / M. Mous (eds.) (1994), Mixed languages. 15 Cases in Language Intertwining, Amsterdam.
―― / P. Muysken (1995), “Mixed languages and language intertwining”, Pidgins and Creoles: An In-
troduction, J. Arends; P. Muysken; N. Smith (eds.), Amsterdam: 41-52
Bazin, M. (1981), “Quelques échantillons des variations dialectales du Tāleši”, Studia Iranica 10: 111-
124, 269-277.
Gasanov, A. G. (1966), “O ‘taynom’ yazyke žitelej sela Kilit naxičevanskoj ASSR”, Voprosy Dialekto-
logii Tyurkskix Yazykov, Baku: 90-95.
Haig, G. / Y. Matras (2002), “Kurdish Linguistics: A Brief Overview”, STUF, 55.1, Berlin: 33-14.
Henning, W. B. (1954), “The Ancient Language of Azerbaijan”, TPS, 1954, Oxford. 157-177, http://
www.iranica.com/.
Hüseynova, G. (2002), Lahıc Tatlarının Dili, Baku.
Johanson, L. (2002), Structural Factors in Turkic Language Contacts, Richmond, Surrey.
Kārang, ‘A.-‘A. (1330/1952), Tāti va Keringāni, NDA, Tabriz.
―― (1333/1954), Tāti va Harzani, Tabriz.
Kasravi, A. (1946), Āzari yā zabān-e bāstān-e Āzarbāijān, Tehran.
Lazard, G. (1978), “Le dialecte tâleši de Mâsule”, Studia Iranica 7/2: 251-268.
―― (1979), “Glossaire Mâsulei”, Studia Iranica 8/2: 269-275.
 The Polygenetic Origins of the Northern Talyshi Language 453

Matras, Y. (2003), “Mixed languages: Re-examining the Structural Prototype”, Y. Matras; P. Bakker
(eds.), The Mixed Language Debate: Theoretical and Empirical Advances, Berlin.
―― / P. Bakker (2003), “The Study of Mixed Languages”, ibid.
Miller, B. V. (1953), Talyšskij Yazyk, Moscow.
Mortazavi, M. (1333/1954), “Noktei čand az zabān-e harzani”, NDA, Tabriz, 6: 304-314.
―― (1341/1962), “Fe‘l dar zabān-e harzani”, NDA, Tabriz, 14: 453-488.
―― (1342/1963), “Fe‘l dar zabān-e harzani”, NDA, Tabriz, 15: 61-97.
Mous, M. (2003), The Making of a Mixed Language: The Case of Ma’a/Mbugu, Creole Language Li-
brary, vol. 26, Amsterdam.
Nawata, T. (1982), “The Masal Dialect of Talishi”, Acta Iranica 22: 93-117.
Pirejko, L. (1976), Talyšsko-Russkij Slovar’, Moscow.
Stilo, D. (1981), “The Tati Language Group in the Sociolinguistic Context of Northwestern Iran and
Transcaucasia”, Iranian Studies 14/3-4: 137-187
―― (2001), “Gilan, x. Languages”, Encyclopaedia Iranica, Volume X, Fascicle 6, (Germany VI - Gin-
daros), New York: 660-668.
―― (2008), “Two Sets of Mobile Verbal Person Agreement Markers in the Northern Talyshi Lan-
guage”, Aspects of Iranian Linguistics: Papers in Honor of Mohammad Reza Bateni, Newcas-
tle-upon-Tyne: 363-390.
―― (2009), “Circumpositions as an Areal Response: The Case Study of the Iranian Zone”, Turkic
Languages 13.1, Wiesbaden: 3-33.
―― (in press A), “On the non-Persian Iranian Substratum of Azerbaijan”.
―― (in press B), “Further Notes on the Iranian Substratum of Azerbaijani Turkish”.
―― (forthcoming), “Mazanderani”, Encyclopaedia Iranica.
―― (in progress), An Atlas of Linguistic Structures of the Araxes Linguistic Area.
Thomason, S. G. (2003), “Social Factors and Linguistic Processes in the Emergence of Stable Mixed
Languages”, The Mixed Language Debate: Theoretical and Empirical Advances, Berlin.
―― / K. Terrence (1988), Language Contact, Creolization, and Genetic Linguistics, Berkeley.
Tiessen, C. F. (2003), Positive Orientation towards the Vernacular among the Talysh of Sumgayit,
Master’s of Arts thesis, University of North Dakota, Grand Forks, North Dakota.
Vahman, F. / G. S. Asatrian (1990), “Gleanings from Zāzā vocabulary”, Papers in Honor of Professor
Ehsan Yarshater, Acta Iranica 30, Leiden: 267-275.
Yarshater, E. (1959), “The Dialect of Shāhrud (Khalkhāl)”, BSOAS 22: 52- 68.
―― (1960), “The Tāti Dialect of Kajal”, BSOAS 23: 275-286.
―― (1970), “The Tāti Dialects of Țārom”, W. B. Henning Memorial Volume, London: 451-467.
―― (1996), “The Tāleshi of Asālem”, Studia Iranica 25/1: 83-113.
―― (2005), “Tāti Dialect of Kalāsuri”, Languages of Iran: Past and Present, Iranica 8: 269-284.
―― (to appear B), “Karingāni Dialect” (February 18, 2003 draft).
―― (to appear C), “Karani Dialect” (May 18, 2003 draft).
Zoka, Y. (1332/1954), Guyeš-e Keringān, Tehran.
―― (1336/1957), “Guyeš-e Galinqaya ya Harzandi”, Farhang-e Irānzamin 5, Tehran: 51-92.
 Not only in the Caucasus:
Ethno-linguistic Diversity on the Roof of the World

Matthias Weinreich
Yerevan State University

Abstract
The current paper is an invitation to a virtual journey to Gilgit-Baltistan (former Northern Areas), a
high mountain region in the north of Pakistan, endowed with an amazing variety of languages
spoken on its territory. The travel itinerary includes stops at Skardu (Baltistan), Gojal (upper Hun-
za valley), Karimabad (central Hunza valley), Taus (Yasin valley) and Gilgit town. At each destina-
tion the traveller is introduced to the languages used by its inhabitants: Balti, Wakhi, Burushaski,
Domaakí, Pashto and Shina. Local personalities, scholars and a foreign researcher share key infor-
mation about their language's geographical distribution, speaker numbers and dialectal division.
Special attention is given to expositions of the language attitude of the concerned speaker com-
munities, as well as to the description of local efforts directed at creating language-specific alpha-
bets and the promotion of mother tongue education. The interlocutors' narratives are comple-
mented by black-and-white photographs and references to recent academic publications dedi-
cated to the languages and peoples of Gilgit-Baltistan.

Keywords
Northern Pakistan, Gilgit-Baltistan, Northern Areas, Balti, Wakhi, Burushaski, Domaakí, Pashto,
Shina, Dardic languages, language-specific alphabets, mother tongue education

The Caucasus, this magnificent, blessed and in many respects extraordinary region to which
Prof. Asatrian has devoted a major part of his scholarly interest, is often compared to a
treasure trove of ethnic groups and languages. Another mountainous region similarly en-
dowed with an amazing variety of ethno-linguistic groups is Northern Pakistan. Ever since
our first encounter in January 2000, I have been trying to persuade Prof. Asatrian to pay a
visit to this fascinating place, because I am convinced that once there he would fall in love
with it as intensely and unconditionally as I did. Unfortunately, until now my efforts have
not been crowned with success.
Dear Garnik, since I know that your hesitation to act on my suggestion is linked to your
numerous academic commitments and because I fear that your formidable workload is not
likely to become less in the foreseeable future, I am taking advantage of this Festschrift to

 Koninklijke Brill NV, Leiden, 2015

456 Matthias Weinreich

present you with a virtual journey, hoping that one day we will be given the opportunity to
re-enact it in real life.1

Considering the fact that there are more than two dozen languages spoken as
mother tongues by communities in Northern Pakistan, we would hardly be able
to examine even half of them within a reasonable amount of time.2 Therefore I
would like to limit our journey in space, and to focus on Gilgit-Baltistan,3 one of

1
My congratulations on your 60th birthday are joined by Silvia Delogu, my wife, who is (if not
indicated otherwise) also the author of the accompanying photographs. The sources of the other
illustrations are: 1) Map of Northern Pakistan: Weinreich (2009: 120); 2) Photograph on page 481:
<http://www.concordiaexpeditions.com/pakistan/gilgit.html>.
2
For a comprehensive list of the languages in use in Northern Pakistan see Baart (2003). Many
of them are included in the “Sociolinguistic Survey of Northern Pakistan” (O'Leary 1992), which
contains separate investigation into: Balti, Bashghali (Eastern Kativiri), Bateri, Burushaski, Chilisso,
Dameli, Domaakí, Gawar-Bati, Gawri, Gowro, Gujari, Hindko, Indus Kohistani (Mayãn), Kalasha,
Kamviri, Khowar, Ormuri, Pashto, Palula, Torwali, Ushojo, Wakhi, Waneci, and Yidgha (Munji). For
additional literature see the bibliography compiled by Baart/Baart-Bremer (2001); as well as
Buddruss (2006) and Kreutzmann (2005b), who give an excellent overview of the region’s ethno-
linguistic composition, including speaker numbers and language distribution.
3
Until 2009, Gilgit-Baltistan was called the Northern Areas of Pakistan; the re-naming was ef-
fected on the basis of the “Gilgit-Baltistan Empowerment and Self-Governance Order”, § 96, which
also provides an outline of the territory’s current political and administrative set-up.
 Ethno-linguistic Diversity on the Roof of the World 457

the region’s administrative units and a place where I had the pleasure to work as
a researcher in the 1990s.
Gilgit-Baltistan is situated in the far north of the country. It encompasses a
territory of around 72.000 km2, covering the greater part of the Karakoram
mountain range, as well as the western fringes of the Himalayas and eastern
fringes of the Hindu Kush. Because of its dramatic landscape, dotted with an im-
pressive number of snow-capped mountain peaks, among them Nanga Parbat,
Rakaposhi and K2, this part of Northern Pakistan is often associated with the
legendary Roof of the World. Internally, Gilgit-Baltistan is divided into six ad-
ministrative sub-units: Gilgit, Ghizar, Diamer, Astor (until 2004 part of Diamer),
Skardu and Ganche. Together, these districts are inhabited by nearly 900.000
people, with approximately 200.000 of them residing in Gilgit, the region’s larg-
est town. The two other urban centres are Chilas and Skardu Town, headquar-
ters of the Diamer and Skardu Districts respectively. Although urbanisation is, as
in other regions of Pakistan, an ongoing process, the overwhelming majority of
the population is still found in rural areas, in villages and larger settlements scat-
tered oasis-like across the bottom of the bigger valleys and along the major
roads.4
It is by taking the most important of these roads, the world-famous Karako-
ram Highway,5 that I had thought to bring you up to Skardu, the capital of Bal-
tistan. However, due to a landslide, the highway is once again closed to all traffic,
and in order to reach our destination in time, we will have to change our plans
and take a plane.

1. BALTI
After a flight of just under two hours, passing over parts of the snow-clad moun-
tains of the western Himalayas, we reach Skardu Airport, situated at an altitude
of about 2500 meters above sea level. On the way into the town we follow the
mighty Indus, which in the Skardu valley meanders across a wide, sandy land-
scape. At the town’s main bazaar Mr. Yusuf Hussainabadi is waiting for us. He is
a highly regarded local scholar, who back in the 1990s translated the Koran into

4
For a comprehensive description of the area’s geography and recent history see Kreutzmann
(2005a).
5
The importance of the Karakorum Highway for the entire region and its impact on the social
and economic situation of Gilgit-Baltistan is discussed by Allan (1989) and Kreutzmann (1991).
458 Matthias Weinreich

the Balti language6 and is now engaged in promoting mother tongue education
among the youth of Baltistan.
Balti7 is a Western Tibetan language, spoken throughout Skardu and Ganche
districts, which is influenced by the neighbouring Dardic language Shina and,
like all other vernaculars of Northern Pakistan to a certain extent also by Urdu,
the country’s lingua franca and main medium of school instruction. Notwith-
standing considerable syntactic and lexical innovations, stimulated by contact
with its linguistic surroundings, Balti has preserved a number of archaic features,
reminiscent more of Classical Tibetan than of the modern variety now spoken in
Central Tibet.
Mr. Hussainabadi informs us that alt-
hough more than 300,000 people use
Balti in their daily communications, it is
not taught in any local school. On one
hand, parents prefer their children to
read and write in Urdu and English, be-
cause they perceive Balti as a language
of low social value, and therefore of little
importance for their offsprings’ life. On
the other hand, Mr. Hussainabadi is
keen to stress that one can’t seriously
hope to raise the prestige of Balti in the
eyes of its speakers unless they are pro-
vided with a unified alphabet, as clear
and as easy to employ as those used for
high-status languages like Urdu and
English. However, while the creation of
such an alphabet seems comparatively
easy, agreeing on a standard version of it appears to be a rather challenging task.
The problem is, that each of the handfull of language enthusiasts engaged in the
encouragement of Balti literacy is promoting his very own variety of Balti or-

6
Cf. Hussainabadi (1995).
7
On Balti grammar see Read (1934), Lobsang (1995) and Bielmeier (1985); the latter also con-
tains a long fairy tale as a specimen of the language. Balti’s linguistic affiliation is analysed in Biel-
meier (1998); on socio-linguistic issues see Backstrom (1992a). Balti oral literature is recorded and
discussed in Sagaster (1981, 1984, 1985, 1993).
 Ethno-linguistic Diversity on the Roof of the World 459

thography, one of them even arguing in favour of utilizing a modified Tibetan

script. In order to finally move forward on this long-standing, highly controver-
sial issue, Mr. Hussainabadi and friends have recently contacted the Forum for
Language Initiatives (FLI),8 a specialised, Islamabad-based NGO, which is not
only expected to reconcile the differing opinions on the script, but also to design
teaching material for Balti lessons at
primary school level, following a meth-
odology already successfully employed
with other minority languages of North-
ern Pakistan. One can only hope that in
this way Mr. Hussainabadi’s educa-
tional efforts will finally bear fruit and
Balti children will soon be able to read
and write in their mother tongue.

2. WAKHI
From Skardu we take the public bus in the direction of the Hunza Valley. Our fi-
nal destination is Gulmit, a large settlement in Gojal, a high altitude region
spread on both sides of the Hunza River’s upper reaches. Although the view from
the window does not allow for boredom, the journey seems never ending. First,
we follow for about six hours the course of the Indus, until the latter joins the
Gilgit River at Bunji. Changing the Indus Road for the Karakoram Highway and
our bus for a minivan, we then head northwards for another eight hours, first, in
the direction of Gilgit and then, shortly before reaching the town’s outskirts,
alongside the roaring Hunza River, up, up, up, into the night, almost until the
Pakistani-Chinese international border.
One of our co-passengers happens to be a Wakhi speaker, a teacher named
Didar Ali, who worked in the 1990s as a language consultant for the German-Pa-
kistani Culture Area Karakorum Project. He informs us that the ancestors of the
Wakhis now settled in Northern Pakistan arrived there during the 18th and early
19th centuries from Wakhan, a region in the south-eastern Pamirs, situated on
both sides of the River Panj. On Pakistani territory the Wakhi language9 is not

8
The FLI is an educational organisation based in Islamabad which supports language com-
munities in Northern Pakistan in their efforts to preserve and promote their mother tongue; cf.
<http://fli-online.org>.
9
On Gojal-Wakhi grammar and lexicon see Lorimer 1958, as well as Buddruss (1986b, 1998,
460 Matthias Weinreich

only spoken in Gojal, but also in parts of the Ishkoman and Yarkhun valleys, the
latter located in Chital, right across the western border of Gilgit-Baltistan. As the
term Wakhi is of foreign origin, Didar Ali, who is sensitive to all issues connected
with his language and culture, prefers to make use of the original self-appella-
tions, referring to himself as xik,10 and to his language as xikwor. He proudly
quotes the well-known Norwegian linguist Georg Morgenstierne, who charac-
terised Wakhi as “one of the most archaic, and at the same time most peculiar, of
living Iranian languages”.11
Most of the Wakhi studies conducted
by Morgenstierne and other scholars are
based on the Wakhan variety. In response
to our question if the latter differs from
the Wakhi spoken in Gojal, we are ensured
by Didar Ali that he does not experience
any difficulty in communicating with his
Wakhan brethren, who he regularly en-
counters on business trips in neighbouring
China. His observations are confirmed by
the distinguished German scholar Georg Buddruss, who established that the dis-
similarities between both Wakhi varieties are very slight, mostly pertaining to a
few morphological and lexical variations.12
Didar Ali draws our attention to the fact that Wakhi oral literature, especially
poetry and fairy tales, contains a significant number of Persian words and
phrases, many more than can be found in the speakers’ everyday language. Re-
ferring to Persian, the lingua franca of his ancestor’s Pamir homeland, our trav-
eller companion voices his appreciation by calling it “a sweet language”. This af-
fectionate attitude towards Farsi may also be linked to the fact that all Gojali
Wakhis are adherents of the Ismailia creed, the most prominent texts of which
are composed in this idiom.

2001), who also offers specimens of oral literature; socio-linguistic issues are discussed in Back-
strom (1992b), as well as in Reinhold (2006), which in addition contains a fine collection of texts.
10
Wakhi [x] is a voiceless palatal-velar fricative, pronounced slightly posterior to the German
ich-Laut [ç].
11
Cf. Morgenstierne (1938: 431).
12
Cf. Buddruss (2006: 240).
 Ethno-linguistic Diversity on the Roof of the World 461

As a teacher, Didar Ali is very concerned

that Wakhi children are not offered lessons
in their mother tongue. Similar to Mr. Hus-
sainabadi, he is in contact with the already
mentioned Islamabad NGO, but differently
from his Balti counterpart, Didar Ali does not
entertain any hope of receiving governmen-
tal funding for the elaboration of educational
material, because the number of Wakhi
Speakers living in Northern Pakistan is com-
paratively small, estimated to be only 8000.

3. BURUSHASKI
From Gulmit we move towards Karimabad, Hunza’s central settlement, set in
breathtakingly beautiful surroundings. In a coffee shop below the famous Ti-
betian style Baltit Fort, we are received by Susumu, a Japanese linguist engaged
in translating Burushaski legends and fairy tales. Susumu tells us that the Bu-
rushaski language13 is spoken by more than 90.000 people, not only in Central
Hunza, but also across the river in the Nager, as well as in remote Yasin, a valley
close to the Chitral border.
Because of its “polysynthetic” structure, Burushaski has the reputation for
being an extremely complicated language. To illustrate this, Susumu lists some
of its salient peculiar features: Burushaski has five primary cases and up to forty
different plural morphemes; some nouns, especially kinship terms and names for
body parts, take an obligatory pronominal prefix in three different grades, in-
dexing the possessor: accordingly one can’t say simply “heart”, but has to say as
“my heart”, or es “his heart” or mos “her heart”, etc. The verbal system is more
complex still. It is possible to derive hundreds of verb forms from a single root by
enlarging it with prefixes, infixes, and suffixes. Moreover, a single inflected verb
form often translates as a complete English sentence, e.g. eéumo “she did not give

13
The most authoritative researcher on Burushaski is Hermann Berger, who published exten-
sively about both of its dialects, cf. e.g. Berger (1974, 1998), who offers a sublime introduction into
Burushaski studies in Berger (1992); for other work on grammar, as well as texts related to Hunza
and Yasin see Lorimer (1935-38); Tiffou/Pesot (1989) and Tikkanen (1991). Text specimens from
Nager are also available in Skyhawk (1996, 2003); Burushaski’s possible role as a regional substra-
tum language is discussed in Tikkanen (1988).
462 Matthias Weinreich

it to him” (root -u-), atúkushoa “will she not bring you here?” (root shu-), atúko-
manuma “you were not born” (root -man-).14
All told, the most vexing
problem concerning Buru-
shaski seems to be its rela-
tionship with other tongues.
Beyond any doubt Burusha-
ski is an isolated language,
quite unrelated in origin to
its Indo-Aryan, Iranian, and
Tibetan neighbours. Moreo-
ver, there is still no conclu-
sive evidence relating it to
any other known language
family. Various theories have been put forward in this regard, but none of them
have been generally accepted. Historical connections have been suggested with
Yenisseian spoken in eastern Siberia, as well as with Munda, Nubian, Northern
Caucasian languages and even Proto-Macedonian (Phrygian). Although some
striking structural similarities with Basque exist, truly convincing etymological
equations are still missing.15
Asked about local efforts to put Burushaski into writing, Susumu informs us
that this issue has being discussed within the local communities since the 1980s.
However, until two to three years ago, it seemed that only a few Burushaski
speakers were interested in using
their mother tongue in any domain
other than oral communication. The
situation was further complicated by
the fact that the inhabitants of Hunza
and Yasin, who are almost exclusively
Ismailis, prefer a Latinised alphabet,
while the people of Nager, who are
mainstream Shiites, would only sup-
port an Urdu-based variety. But re-

14
The examples are taken from Tikkanen (2001: 186) and Berger (1992: 11).
15
For references to publications speculating about possible genetic ties of Burushaski with
other languages, see Buddruss (2006: 238).
 Ethno-linguistic Diversity on the Roof of the World 463

cently, and here Susumu smiles, things have taken a new turn, and this is thanks
to the Internet. Over the last few years a number of local sites have appeared, on
which young, educated Burushaski speakers from Hunza, Nager and Yasin alike
discus all kinds of topics in their mother tongue, using a simplified, non-
standardized Latin transcription.16 This, according to our Japanese friend, gives
hope for the future. But teaching Burushaski at school? No, Susumu admits, with
only a few parents interested in the education of their children in their mother
tongue, this is still a very distant dream.

4. DOMAAKÍ
In the evening we are invited to a wedding party in the neighbourhood. At
the height of the festivities, men in long woollen cloaks are leaping into the air
and whirling around to the tunes of a traditional orchestra. During a break we
are approached by one of these passionate dancers who introduces himself as
Shaban Ali Nageri. Answering our question about his mother tongue, he
promptly says “Burushaski”, but when he senses our genuine interest in the sub-
ject, he adds somehow shyly: “But I also speak Domaakí”. Like the musicians of
the local orchestra, who had brought him along as their technical assistant, Sha-
ban Ali belongs to the Hunza-Nager valleys’ tiny Dooma community. The Dooma
people are scattered in extended family units among larger host communities in
Gilgit-Baltistan and in Indus Kohistan (NWFP Province). Their comparatively
low position within society is strongly reminiscent of the humble status occu-
pied by the European and Asian Gypsies. Interestingly, the term Doom is con-
nected to the same Indo-Aryan root as the Gypsy self-appellation “Rom” or
“Lom”, although Domaakí itself does not show any special affinity with Romani
or Lomavren. According to local oral history, the Dooma’s ancestors came
through Baltistan from Kashmir, in separate groups and over an extended period
of time; others claim their forefathers arrived from Darel or even from Kashghar.
Traditionally, Dooma were only allowed to work as blacksmiths or as musicians,
but nowadays they are also engaged in a variety of other professions. So Shaban
Ali, who comes from a well-known blacksmith’s family, is now working as car
electrician. In almost all the places of their present settlement, the Dooma have
long since given up their original mother tongue in favour of the host languages.

16
See e.g. <http://hisamullahbeg.blogspot.com>; <http://gbtimes.wordpress.com>; <http://pa
mirtimes.net>.
464 Matthias Weinreich

Only in the Nager and Hunza valleys has Domaakí, in the form of two distinct di-
alects, survived to the present day.17
From a historical point of view,
Domaakí is a language of the North
Indian plains, and therefore affili-
ated to idioms like Urdu, Punjabi
and Rajastani. However, due to its
intense contacts with other lan-
guages Domaakí has lost or trans-
formed many of its original features.
This now places the language in a
number of important aspects much
closer to its Dardic neighbours than
to its Central New Indo-Aryan cous-
ins. Shaban Ali informs us that the Dooma people are typically bi- or multilin-
gual. He himself, besides Domaakí and Burushaski, also speaks Shina, Urdu and a
little English. Unfortunately, Domaakí, which is left with less than 350 speakers,
suffers from having a rather low prestige and is perceived by almost all Dooma as
an obstacle to their way out of a life of poverty and into greater social integra-
tion. In this context it is sad to hear, but at the same time absolutely under-
standable, that like many other Domaakí speakers, Shaban, too, does not teach
his mother tongue to his children but instead communicates with them and his
wife Nargis in Shina. The very limited size of the remaining Domaakí speakers’
community and almost everybody’s strongly negative attitude towards the utili-
zation of Domaakí in both the public and private domains makes it rather likely
that in one to two generations Shaban’s mother tongue will cease to exist as a
living language.
So, what about possible alphabetisation efforts? Well, when we ask Shaban
about it, he gives us a surprised look, shrugs and replies somehow sarcastically:
“None of our people really likes to speak this language, why should anybody be
interested in writing it?!”

17
On Hunza-Domaakí see Lorimer (1939); Buddruss (1983, 1984, 1986a); Tikkanen (2011, forth-
coming); on Nager-Domaakí see Weinreich (1999, 2008, 2011). Domaakí language loss is discussed
in Weinreich (2010); historical and ethnographic matters pertaining to the Dooma community in
Hunza are taken up in Schmid (1997).
 Ethno-linguistic Diversity on the Roof of the World 465

5. PASHTO
Our next destination is the Yasin valley, the home, as we already know, of one of
the Burushaski dialects. On our way—following the course of the Hunza and the
Gilgit rivers—we are informed that other idioms spoken by residents of the area
include Shina, Kirghiz, Wakhi, Khowar and Gujri. However, our interest is fo-
cused not on these, but rather on Pashto, an Eastern Iranian language current in
North Western Pakistan and in Afghanistan. Most of the Pashto speakers living
in Gilgi-Baltistan18 are temporary migrants, who come there for seasonal work,
mainly trade and construc-
tion, keeping their families
back home in the Pashtun
heartland. Other Pathan mi-
grants have settled on a per-
manent basis, sometimes in
connection with marriage
into a local family. One of
these permanent settlers is
Dilawar Khan, a shop-keeper
from the village of Taus in Ya-
sin. Dilawar Khan’s ancestors
arrived in the region at the beginning of the 20th century from Bajawur, a district
situated on the Afghan border. Although almost a century has passed since then,
our host, his Pashtun wife and his children still retain their original mother
tongue, while also being able to speak Shina and Khowar. For most Pashtuns
living in Gilgit-Baltistan speaking Pashto is a question of maintaining their social
standing vis-à-vis their local neighbours, as their mother tongue associates them
to the more than 20 million Pashtuns in the rest of Pakistan, who enjoy a reputa-
tion of being shrewd businessmen and uncompromising worriers.
The Pashto language is divided into a number of dialects and sub-varieties.
Dilawar Khan tells us, that similarly to his family, which still speaks the idiom
current in Bajawur, also other Pashtun migrants, permanent and temporary ones
alike, tend to stick to the dialect of their place of origin. As some of these varie-

18
In the mid 1990s there were around 5,000 to 5,500 Pashto speakers living in the Northern Ar-
eas (Gilgit-Baltistan), counting for about 0.8-0.9% of the territory’s total population (Weinreich
2009: 19). On the Pashto spoken by these migrants see Bauer 1998; Weinreich 2005, 2009, 2010; on
their migration history see Weinreich (2005).
466 Matthias Weinreich

ties are mutually unintelligible, it

can be quite a challenge for Pashto
speakers to converse with each
other in their common mother
tongue, especially since they can’t
fall back on a widely diffused nor-
mative language, functioning like,
for example, Modern Standard
English in the contemporary Eng-
lish-speaking context. As a trader
Dilawar Khan travels a great deal.
Many of his business partners are
Pashto speakers themselves, but by no means does everyone know the same di-
alect that he does. In order to make themselves understood, Dilawar Khan and
his Pashtun conversation partners follow a simple pattern: Each participant in
the dialogue uses his own dialect. If the person addressed indicates that he did
not understand something the other had said, the speaker provides a synonym
or a phrase, which is intended to illustrate the meaning of his words. In fact, very
often the speaker adds his explanation not just on request, but immediately after
a word or statement, identified by him as potentially creating difficulties in
comprehension. The relevant synonym can be taken from the speaker’s native
dialect or from any other variety, preferably the one spoken by the conversation
partner; the explanatory phrase is normally given in the speaker’s dialect. To il-
lustrate his words, Dilawar Khan provides us with a short example, in which a
shopkeeper from Mardan addresses his newly engaged assistant from the Dir
valley:
Za, bāzār-na mā-ta pey rāwra. pey. tāso če war-ta šode wāyey. se če čāy-kxe āčey.
Go, bring me pey from the bazaar. pey. What you call šode. What one adds to tea.

At the end of our conversation we ask Dilawar Khan about him and his com-
patriots’ relationship with the written word. Smiling, he responds that although
there exists a standardised alphabet and Pashto books and newspapers are freely
available in towns like Peshawar and Quetta, few of his family and friends have
ever bothered to learn it. The little written exchange that is necessary is normally
carried out in Urdu, which seems a perfectly convenient medium for these
communication purposes.
 Ethno-linguistic Diversity on the Roof of the World 467

6. SHINA
Our last destination is Gilgit town, Gilgit-Baltistan’s administrative capital and
its financial centre. Once a sleepy provincial backwater, over the last three dec-
ades the town has experienced an unprecedented economic boom, mostly due
to revenues generated by cross border trade with neighbouring China. Gilgit’s
population is divided into Sunnis, main-
stream Shias, and Ismailites. This religious
diversification, combined with a serious
imbalance in the distribution of wealth,
has created a high level of social tension,
which tends to manifest itself in frequent
inter-communal violence. The town’s so-
cial complexity is mirrored by its linguistic
situation. A survey conducted in the mid
1990s counted more than twenty-five lan-
guages spoken by Gilgit long-term resi-
dents. The town’s main medium of com-
munication is Shina, an Indo-Aryan lan-
guage affiliated to the Dardic group.
Shina19 is also the mother tongue of Mullah Shakil Ahmad Shakil, a local
Sunni preacher associated with the Tablighi movement, whom we meet at the
central mosque in Gilgit’s main bazaar. Following his religious vocation Mullah
Shakil travels all over Northern Pakistan, and, as he mainly preaches in his
mother tongue, he is not only well informed about the geographic distribution of
Shina speakers, but also acquainted with some of the language’s many dialects.
According to Mullah Shakil, Shina is by far the most important idiom of the area,
spoken by at least 300.000 people. Besides Gilgit town and surroundings, it is also
current in the lower Hunza valley, in Ghizar, Diamer, Astor and Skardu Districts,
as well as in Indus Kohistan. As might be expected from such a wide distribution,
the language is fragmented into a number of dialects, the main one being the

19
Shina, as spoken in Gilgit-Baltistan is analysed in Bailey 1924, Schmidt 2008, Hook 1996
(Grammar); Radloff 1992 (Socio-linguistics), id. 1999 (Phonematics); Shina literature is presented in
Buddruss 1987 (Proverbs), Degener 2008 (Proverbs; also contains an excellent grammatical outline
of Gilgiti Shina) and Buddruss/Degener 2012 (Radio features). Kohistani/Schmidt (2006) give a
general overview of Shina in Pakistan.
468 Matthias Weinreich

idiom current in Gilgit, which in and around the town also serves as a lingua
franca for a variety of other ethno-linguistic groups, including Wakhi, Domaakí
and Burushaski speakers. Further Shina dialects include varieties spoken in
Baltistan, Astor, Diamer, as well as in Indus Kohistan. From Mullah Shakil’s
subsequent description of his travels through Shina-country it appears that
linguistic borders are roughly paralleled by sectarian religious boundaries. In the
Hunza valley and to the west of Gilgit, the Ismailia sect predominates. In Gilgit
town and Skardu District, Shias are more numerous. Astor District has a mixed
Sunni-Shia population, and throughout the southern flank of the Shina area,
including Diamer District and Indus Kohistan, the Sunnis are greatly in the
majority.
In the same way as sectarian differ-
ences between Sunni, Shia and Ismai-
lia Shina speakers contribute to the
survival of Shina’s dialectal diversity,
they also intensify the social and ide-
ological separation between Gilgit and
its southern neighbours. In this con-
text Mullah Shakil observes that alt-
hough Gilgit is the largest Shina-
speaking town and the centre of Shina
religious and intellectual life, efforts to develop a standard written form of the
language are not only hindered by differences between the Gilgiti and other di-
alects, but also reinforced by the existing religious tensions. Nevertheless, Gilgit’s
Shina variety seems to be the only one that holds the potential for standardisa-
tion. It possesses a vast body of oral literature, and over the last few decades
there have been many efforts at creating a written tradition as well. Moreover,
Gilgiti is also used by Radio Pakistan for its daily Shina-language broadcasts,
which are highly popular throughout the entire region.
And what about education in the mother tongue? At least in theory, our in-
terlocutor strongly supports this idea. Recently, he remarks, some local volun-
teers have even undergone specialised teacher-training courses at the hands of
an Islamabad-based NGO, which, according to his and our knowledge is also en-
gaged in the Wakhi and Burushaski speaking areas. But as most of the Gilgit vol-
unteers were Shiites and the NGO itself is rumoured to be supported by Western
 Ethno-linguistic Diversity on the Roof of the World 469

donors, Mullah Shakil remains rather sceptical regarding the chances of wider
public support for this initiative.
The next morning we wave
goodbye to energetic Gilgit and
take the public bus going south.
On our way, via the Karakoram
Highway, we ride past the fa-
mous Nanga Parbat mountain,
traverse Diamer District’s high
altitude wasteland, pass through
the narrow gorges of Indus Ko-
histan and, sixteen hours later,
awake to the hustle and bustle of
Rawal Pindi. That’s it. The mountains have changed to the plains, the northern
languages have given way to Urdu and Punjabi, and I, dear Garnik, return you
back to your favourite Iran and the Caucasus. I hope you enjoyed the trip...

BIBLIOGRAPHY
Allan, J. R. N. (1989), “Kashgar to Islamabad: the impact of the Karakorum Highway on mountain
society and habitat”, Scottish Geographical Magazine, 105/3: 130-141.
Baart, J. L. G. (2003), “Sustainable Development and the Maintenance of Pakistan’s Indigenous
Languages”, Proceedings of the Conference on the State of the Social Sciences and Humani-
ties, Islamabad, 26–27. 09. 2003 (see <http://www.fli-online.org/documents/sociolinguis
tics/development_maintenance_of_pak_lgs.pdf>, accessed 28. 09. 2008).
Baart, J. / E. Baart-Bremer, E. (2001), Bibliography of Languages of Northern Pakistan, Islamabad.
Backstrom, P. C. (1992a), “Balti”, P. C. Backstrom, C. F. Radloff (eds.) Languages of Northern Areas,
Islamabad: 3-27, 219-242.
―― (1992b), “Wakhi”, P. C. Backstrom, C. F. Radloff (eds.) Languages of Northern Areas, Islama-
bad: 57-74.
Bailey, T. G. (1924), Grammar of the Shina Language, London.
Bauer, E. (1998), “Several Groups of Pashto-Speakers in Pakistan’s Northern Areas”, Culture Area
Karakorum. Scientific Studies 4/2. Köln: 627-640.
Berger, H. (1974), Das Yasin-Burushaski (Werchikwar), Wiesbaden.
―― (1992), Das Burushaski. Schicksale einer zentralasiatischen Restsprache, Heidelberg.
―― (1998), Die Burushashki-Sprache von Hunza und Nager, 3 volumes, Wiesbaden.
Bielmeier, R. (1985), Das Märchen vom Prinzen Čobzan. Eine tibetische Erzählung aus Baltistan, St.
Augustin.
470 Matthias Weinreich

―― (1998), “Balti Tibetan in its Historical Linguistic Context”, I. Stellrecht (ed.) Karakorum-Hin-
dukush-Himalaya: Dynamics of Change. Köln.
Buddruss, G. (1974), “Neuiranische Wortstudien”, Münchener Studien zur Sprachwissenschaft 32,
München: 9-40.
―― (1983), “Domáaki čhot ‘Ton’”, Münchener Studien zur Sprachwissenschaft 42, München: 5-21. ,
―― (1984), “Domaakí - Nachträge zum Atlas der Dardsprachen”, Münchener Studien zur Sprach-
wissenschaft 43, München: 9-24.
―― (1986a), “Hindi phūl, Domaakí phulé”, Münchener Studien zur Sprachwissenschaft 47, Mün-
chen: 71-77.
―― (1986b), “Wakhi-Sprichwörter aus Hunza”, R. Schmitt, P. O. Skjervoe (eds.) Studia Gramma-
tica Iranica. Festschrift für Helmut Humbach, München: 27-44.
―― (1987), “Zur ältesten Sammlung von Sprichwörtern und Rätseln in der Shina-Sprache”, Stu-
dien zur Indologie und Iranistik 13/14: 39-57.
―― (1998), “Eine einheimische Sammlungvon Sprichwörtern aus Hunza: Text, Übersetzung, Glos-
sar”, Kushev, Luzehtskaia, Rzehack, Steblin-Kamensky (eds.) Strany i narody vostoka 30,
Central’naja Azija, Vostochny Gindukuš, St. Petersburg: 30-45.
―― (2001), “Zwei Wakhi-Texte zum Hexenglauben in Hunza”, M. G. Schmidt, W. Bisang (eds.)
Philologica et Linguistica. Festschrift für Helmut Humbach, Trier: 189-206.
―― (2006), “Linguistic diversity in the Hunza valley”, H. Kreutzmann (ed.) Karakoram in Transi-
tion. Culture, Development, and Ecology in the Hunza Valley, Karachi: 236-245.
Buddruss G. / A. Degener (2012), The Meeting Place. Radio Features in the Shina Language of Gilgit
by Mohammad Amin Zia, Wiesbaden.
Degener, A. (2008), Shina-Texte aus Gilgit (Nord Pakistan), Wiesbaden.
Gilgit-Baltistan Empowerment and Self-Governance Order (see <http://gbtribune.files.wordpress.
com/2012/09/self-governance-order-2009.pdf>, accessed 20. 12. 2012).
Hook P. E. (1996), “Kesar of Layul: A Central. Asian Epic in the Shina of Gultari”, W. L. Hanaway, W.
Heston (eds.) Studies in Pakistani Popular Culture, Lahore: 121-183.
Hussainabadi, M. Y. (transl.) (1995), Al-Qur’an al-Karim: Balti tarjumah, Skardu.
Kohistani, R. / R. L. Schmidt (2006), “Shina in contemporary Pakistan”, A. Saxena, L. Borin (eds.),
Lesser-Known Languages of South Asia, Berlin: 137-160.
Kreutzmann, H. (1991), “The Karakoram Highway – Impact of Road Construction on Mountain So-
cieties”, Modern Asian Studies 25/4: 711-736.
―― (2005a), “The Karakoram Landscape and the Recent History of the Northern Areas”, S. Bianca
(ed.) Karakoram: Hidden Treasures in the Northern Areas of Pakistan, Torino: 41-76.
―― (2005b), “Linguistic diversity in space and time: A survey in the Eastern Hindukush and Ka-
rakoram. Himalayan Linguistics 4: 1-24 (see <http://www.linguistics.ucsb.edu/Himalayan
Linguistics/articles/2005/PDF/HLJ04_Kreutzmann.pdf>, accessed 14. 02. 12).
Lobsang, G. H. (1995), Short Sketch of Balti Grammar, Bern.
Lorimer, D. L. R. (1935-38), The Burushaski Language, 3 volumes, Oslo.
―― (1939), The Dumāki Language, Nijmegen.
―― (1958), The Wakhi Language, 2 volumes, London.
 Ethno-linguistic Diversity on the Roof of the World 471

Morgenstierne, G. (1938), Indo-Iranian frontier languages 2, The Iranian Pamir languages, Oslo: In-
stituttet for Sammenlignende Kulturforskning (2nd edition 1973).
O’Leary, C. F. (ed.) (1992), Sociolinguistic survey of Northern Pakistan. 5 volumes, Islamabad.
Radloff, C. F. (1992), “The Dialects of Shina”, P. C. Backstrom, C. F. Radloff (eds.) Languages of
Northern Areas, Islamabad: 89-203, 301-408.
―― (1998), Folktales in the Shina of Gilgit, Islamabad.
―― (1999), Aspects of the Sound System of Gilgiti Shina, Islamabad.
Read, A. F. C. (1934), Balti Grammar, London.
Reinhold, B. (2006), Neue Entwicklungen in der Wakhi-Sprache von Gojal (Nordpakistan), Wiesba-
den.
Sagaster, K. (1981), “Materialien zur Balti-Volksliteratur I”, Zentralasiatische Studien 15: 473-90.
―― (1984), “The Kings of Baltistan and Other Kings: Some Remarks on Balti Folk Literature”, Jour-
nal of Central Asia 7/2: 49-55.
―― (1985), “Materialien zur Balti-Volksliteratur II”, Zentralasiatische Studien 18: 247-91.
―― (1993), “Tales from Northern Pakistan: The Discovery of the Folk Literature of Baltistan”, S.
Zingel-Ave Lallemans, W. P. Zingel (eds.) Neuere deutsche Beiträge zu Geschichte und Kul-
tur Pakistans, Bonn: 83-92.
Schmid, A. (1997), Die Dom zwischen sozialer Ohnmacht und kultureller Macht. Interkulturelle Bezie-
hungen in Nordpakistan [Beiträge zur Südasienforschung 179], Stuttgart.
Schmidt, R. L. / R. Kohistani (2006), “Shina in contemporary Pakistan”, A. Saxena, L. Borin (eds.)
Trends in Linguistics. Lesser Known Languages of South Asia. Mouton de Gruyter: 137-160.
―― (2008), A Grammar of the Shina Language of Indus Kohistan, Wiesbaden.
Skyhawk, H. van (1996), Libi Kisar. Ein Volksepos im Burushaski von Nager. Mit Beitragen und Er-
gänzungen von H. Berger und K. Jettmar, Wiesbaden.
―― (2003), Burushaski-Texte aus Hispar. Wiesbaden.
Tiffou, E. / J. Pesot (1989), Contes du Yasin. Introduction au Bourouchaski du Yasin, Paris.
Tikkanen, B. (1988), “On Burushaski and other ancient substrata in Northwestern South Asia”, Stu–
dia Orientalia 64: 303-325.
―― (1991), “A Burushaski folktale, transcribed and translated: The frog as a bride”, Studia Orien-
talia 67: 65-125.
―― (2001), “Review of Berger 1998”, Kratylos 46: 185-189.
―― (2011), “Domaki Noun Inflection and case Syntax”, B. Tikkanen, A. M. Butters (eds.) Pūrvāpa-
raprajñābhinandanam: East and West, Past and Present [Studia Orientalia 110], Helsinki:
205-228.
―― (forthcoming), “Domaki Converbs”. Festschrift for Professor Boris Oguibenine.
Weinreich, M. (1999), “Der Domaakí-Dialekt von Nager”, Studien zur Indologie und Iranistik 22: 203-
214.
―― (2001), “Die Pashto Sprecher des Karakorum. Zur Migrationsgeschichte einer ethno- linguisti-
schen Minderheit”, Iran and the Caucasus 5: 285-302.
―― (2005), “Pashto im Karakorum. Zur Sprachsituation und Sprache einer ethno- linguistischen
Minderheit”, Iran and the Caucasus 9.2: 301-330.
472 Matthias Weinreich

―― (2008), “Two Varieties of Domaakí”, Zeitschrift der Deutschen Morgenländischen Gesellschaft

158: 299-316.
―― (2009), We Are Here to Stay. Pashtun Migrants in the Northern Areas of Pakistan, Berlin 2009.
―― (2010), “Language Shift in Northern Pakistan.The Case of Domaakí and Pashto”, Iran and the
Caucasus 14: 43-56.
―― (2011), “Domaakí Grammar Outline”, T. I. Oranskaja, Ju. B. Mazurova, A. A. Kubrik, L. I. Kuli-
kov, A. Ju. Rusakov (eds.) Jazyki Mira: Novye Indoarijskie Jazyki, Moskva: 165- 194.
 Hedayat’s Nationalism and his Concepts of Folklore

Anna Krasnowolska
Uniwersytet Jagielloński, Kraków, Poland

Abstract
The present paper is a hasty investigation of Sādeq Hedāyat’s concept of folklore as a source of
knowledge about the Iranians’ pre-Islamic customs and beliefs, however corrupted with the devas-
tating influence of the strangers. In the introduction to Hedāyat’s Nirangestān (1933), his approach
to the popular culture of Iran, viewed as a collection of “prejudices”, is sharply dualistic. He ideal-
izes its elements of allegedly “Aryan” origin and condemns those, which have been introduced by
the non-Iranian, in particular Semitic, peoples. Hedāyat’s understanding of folklore evolved in the
course of the years, that can be seen in his “General project for the study of the folklore of a region”
(1945), but even in this work the ethnic and linguistic diversity as a valuable element of Persian
culture seem to pass unnoticed.
Since the time when this short paper had been written, Ulrich Marzolph’s substantial entry on
Sādeq Hedāyat as folklorist appeared in Encyclopaedia Iranica, bringing much first-hand, well doc-
umented information on the subject.

Keywords
Sādeq Hedāyat, folklore, Iranian nationalism, anti-Semitism, prejudices, ethnic diversity of Iran

The promoters of Persian nation’s “Aryan” origins, discovered due to Western

scholars and politicians, looked for their roots in historical monuments, archae-
ological findings, in Zoroastrian holy scriptures and dead languages. An idealized
concept of “Iraneness”, which then became a part of the Pahlavi dynastic ideol-
ogy, included a fusion of simplified ideas on Achaemenian imperial past, Zoroas-
trian religion and epic tradition. Another field in which the enthusiasts of Iran’s
Aryan identity hoped to find the evidence of their pre-Islamic glory and cultural
superiority was popular culture, understood as “folklore”.
Sādeq Hedāyat (1903-1951) was a personality essential for the development.
Not only was he an author who influenced almost all Persian prose writers up to
our days, and one of the first in Iran to translate Middle Persian texts into New
Persian, but he also was a pioneer of the studies on folklore in his country.
Hedāyat’s best known work on Iranian folklore, i.e. Nirangestān (1933; here
 Koninklijke Brill NV, Leiden, 2015
474 Anna Krasnowolska

quoted after the edition of 1963), supplied to his successors a model, although
somewhat simplified, of research on popular culture. The introduction (Dibāče)
to Nirangestān, in which Hedāyat explains the basic notions and objectives of his
work, is especially revealing in this matter.
As we read in the Dibāče, under the term “folklore” (folklor) Hedāyat under-
stands the “beliefs of common people” (e’teqādāt-e avām), being nothing else but
“superstitions” (xorāfāt) (Hedāyat 1963: 9ff), coming from the remote, pre-his-
toric times.
Putting forward a question about the relationship between these beliefs and
the national character of the Iranians, Hedāyat states:
“That part of the customs and prejudices which is perceived by the modern soci-
ety as bad and reprehensible, obviously is not a product of Iranian mind; on the
contrary they have been imposed on them by strange races, and by the pressure
of foreign religions” (ibid.: 12).
Therefore, Hedāyat divides Iranian popular “superstitions” in two groups, ac-
cording to their origins: native (bumi), and foreign (bigāne).
The native (i.e. genuine Iranian) ideas and beliefs (afkār-o e’teqādāt) are
namely those, which originate from the everyday experiences and remote his-
torical memories of the Indo-Iranian race. They should be appreciated and pre-
served by the today Iranians, as their precious patrimony. As such Hedāyat
names the beliefs and stories related to the Moon and the Sun, to dragons,
speaking animals and plants etc. He considers such beliefs very old, probably
even pre-Zoroastrian. Similarly old and genuine are, in Hedāyat’s opinion, the
customs related to the old calendar festivals, wedding rites, respect for fire, ritu-
als of purification, some tales and anecdotes. In this group he also includes some
kinds of omens and predictions, but since, as he observes, early Zoroastrianism
was strongly opposing magic practices and witchcraft, their origins should be
looked for in a “Turanian influence” (ibid.: 14).
The strangers who were able to force their bad prejudices upon the Iranians,
either through a religious system, or through cultural influence were, according
to Hedāyat, the Scythians and Parthians (apparently not considered by him as
being of Iranian stock), Greeks and Romans, and most importantly the Semitic
peoples, i.e. the Chaldeans, Babylonians, Assyrians, Jews and Arabs. Hedāyat es-
timates the presence of the Magi in Zoroastrian religion already as the beginning
of foreign penetration because, as he claims, they recruited mainly from among
the strangers: Scythians, Parthians and Semites, specializing in astrology, horo-
 Hedayat’s Nationalism and his Concepts of Folklore 475

scopes and witchcraft. In particular he makes the ancient Semitic peoples re-
sponsible for such beliefs and customs adopted by the Iranians, as: bloody sacri-
fices (unknown to the Indo-Europeans, as Hedāyat believes), polygamy (also al-
legedly unknown to them before), beliefs in lucky and unlucky days and hours,
influence of stars on human life, fatalism etc. He blames the Greeks for the ex-
port of polytheism to Iran, Jews and Egyptians for the propagation of some kinds
of witchcraft, and finally the Arabs for a definite consolidation and legitimiza-
tion of all these ugly practices. Interestingly, in Hedāyat’s list of the destroyers of
the pure Iranian culture no Turks and Mongols are found (except for the above
mentioned mythological Turanians) (ibid.: 16-18).
From among all the strangers, according to Hedāyat, the Jews seem to be the
most harmful, and their methods of acting the most deceitful:
“The Jews, taking opportunity from their close kinship with the Arabs, played a
great role in diffusing the superstitions. The authors of the hadith and akhbar and
other prejudice makers joined them in the propagation of those rotten thoughts
among the common people” (ibid.: 17-18). 1
In Hedāyat’s opinion all those harmful ideas brought in from the outside of
Iran, “have made the life difficult and poisoned” (moškel-o zahr ālud) to the Ira-
nians (ibid.: 25).
Yet, Hedāyat sees some important reasons for bothering with the research on
folklore, understood as a collection of superstitions:
– Irrational ideas (owhām) and prejudices present in popular culture should be
recognized and described as such by the Iranians themselves, so that the
strangers (bigānegān) would not be able to present them in their writings as a
part of Iranian national patrimony (ibid.: 25);
– once written down and rationally analyzed, they will lose their paralyzing influ-
ence on the people’s minds (ibid.: 23-24);
– on the other hand the “good and praiseworthy customs and beliefs” (ādāb-o
rosum-e xub-o pasandide), which preserve the memory of Iran’s glorious past and
are now falling in oblivion, should be acknowledged and cultivated (ibid.: 25).
Yet another reason for collecting “folklore” is, that nowadays this is being
done all over the world, from civilized countries down to the wild African and
Australian tribes, while in Iran, except for the book of Kolsum-nane by Jamāl-e

1
Strong anti-Semitic stereotypes are also present in Hedāyat’s short stories such as, e.g. Talab-e
āmorzeš and, apparently, Gojaste-dež, which makes use of the myth about Jews producing gold of
the Muslim children’s blood.
476 Anna Krasnowolska

Xunsāri nobody had ever tried to write down popular superstitions so far, and
the task should not be left to western travelers, whose honesty is disputable
(ibid.: 25-26).
Thus, Hedāyat’s view on popular culture of Iran is dichotomous, the division
of its elements into “native” and “strange” being identical with their evaluation as
“good” and “bad”. Anything purely Iranian deserves appreciation and protection;
anything “foreign” (especially Semitic) should be eliminated. However, in the
text of Nirangestān itself no such division of described beliefs and customs has
been introduced, and the reader learns almost nothing neither about the origins
of these customs, nor about their ethnic affiliation and diversification. The work,
divided into the chapters on various customs, beliefs, magical practices, prov-
erbs, folk tales etc., reminds of a medieval Islamic book of marvels rather than a
modern scholarly study.
Hedāyat’s approach to popular culture, strongly biased by nationalist ideol-
ogy at the beginning, seems to evolve in the subsequent years of his life. His an-
other text about folklore, published twelve years later (1945), in Soxan (esfand
1323 – xordād 1324),2 shows some important differences with the previous one.
The author’s notion of folklore becomes more mature, he broadens it to the
knowledge of traditional culture in a larger sense (“anything learned by common
people outside the primary school”) (ibid.: 234). Hedāyat introduces to his writ-
ings the neologism nežād-šenāsi (“knowledge of the races”) as an equivalent of
“ethnography”, but rejects the term mardom-šenāsi (“knowledge of the peoples’),
which, in fact, is closer to “anthropology”. In this article Hedāyat underlines the
universality of popular cultures and their similarity to one another, all over the
world, thus arguing that “folklore reduces hostility towards the strangers and re-
veals the unity of mankind. Therefore, the life of simple people has common
ground, and an important point is, that it originates from pre-historical times”
(ibid.: 237).
At the same time Hedāyat notices the importance of popular art and litera-
ture for the development of the national culture. Thus, he considers the
knowledge of the “secrets of the language, songs, tales, beliefs, joys and sorrows
and, generally speaking, the whole of the material and spiritual life” (ibid.: 135) of
the common people, a duty of anybody who calls himself Iranian patriot.

2
Folklor yā farhang-e tude, recently published in: Sādeq Hedāyat, Farhang-e āmiyāne-ye mar-
dom-e Irān, ed. Jahāngir-e Hedāyat, Tehrān 1381, pp. 233-41.
 Hedayat’s Nationalism and his Concepts of Folklore 477

While criticizing the non-professional way in which the newly opened Eth-
nographical Museum in Tehran has been organized, Hedāyat states, that the ex-
position does not show local peculiarities in adequate ways: those who wrote the
descriptions of the objects, did not know local languages, and precious testimo-
nies in dialects have been rewritten with mistakes or even destroyed.
Also Hedāyat’s detailed questionnaire for field research (Tarh-e kolli barā-ye
kāveš-e yek manteqe, 1945) (ibid.: 242-273)3 shows an amount of professional
knowledge which Hedāyat did not have yet while writing his Nirangestān. The
range of subject matters covered by the questionnaire has been broadened to
such fields as material and economic aspects of the people’s life (food, dress and
architecture, work and its tools, means of transport and ways of traveling etc.),
popular art and knowledge, non-orthodox religious cults and practices; social life
and its institutions, etc. The topics are systematically arranged and the work is
provided with an instruction for a folklore collector.
But still the consciousness of ethnic complexity of Iranian society as re-
flected in this document seems low. The short passage on lahjehā va zabānhā-ye
bumi (“local dialects and languages”) (ibid.: 246) seems to ignore their belonging
to different language families and systems. The author encourages the investiga-
tors to ask about: “names of things and animals, abstract names, philosophical,
magical and religious concepts, idioms, proverbs, metaphors, technical termi-
nology and secret (zargari) languages, slang (zabān-e dāšhā), religious language
and that of the common people, popular tales explaining the origins of geo-
graphical names”. He does not advise them to write down the grammatical rules
of the languages in question. Further on, while describing the methods of field
work, Hedāyat again adds a passage on local language (Ašnāyi be zabān-e bumi)
(ibid.: 263) in which he underlines the need of knowing the local dialect of the
given region, as different from the literary language used, as a rule, by an edu-
cated folklore collector. So, he seems to perceive the language differentiation in
Iran in terms of an opposition of urban-rural, elitist-popular, rather than Per-
sian–non-Persian, (Persian-Turkish, Persian-Kurdish), etc.
In spite of its ideologically biased approach, Hedāyat’s pioneering role in re-
search on the folklore of Iran cannot be overestimated. His attempts at discov-

3
Later on supplemented and published by Enjavi Širāzi as “Tarh-e kolli barā-ye gerd-āvari-ye
farhang-e mardom-e yek ābādi”, in his Gozari va nazari… dar farhang-e mardom, Tehrān 1992, pp.
97-192; Marzolph (2014) explains the genesis of Hedāyat’s text and its dependence from Western
sources.
478 Anna Krasnowolska

ering the pure Aryan roots of Iranian culture finally led to the discovery and
professional studies of the multiethnic country’s cultural diversity by Hedāyat’s
successors.

BIBLIOGRAPHY
Hedāyat S. (1933), Nirangestān, Tehrān (quoted after the 3rd edition, Tehrān 1963).
Enjavi-Širāzi A. Q. (1992), Gozari va narazi… dar farhang-e mardom, Tehrān.
“Folklor yā farhang-e tude” (1381), Sādeq Hedāyat, Farhang-e āmiyāne-ye mardom-e Irān, (ed. Ja-
hāngir-e Hedāyat), Tehrān: 233-41.
Marzolph U. (2014), “Hedayat, Sadeq iii. Hedāyat and folklore studies”, Encyclopaedia Iranica,
<www.iranicaonline.org/articles/hedayat-sadeq-iii>.
 Early Specimens of Pashto Folklore

Mikhail Pelevin
St. Petersburg State University

Abstract
The article examines the early Pashto folklore texts included in The Khatak Chronicle which is the
original part of the historiographical compilation Tārīkh-i muraṣṣa‘ (The Ornamented History) by
the Pashtun tribal ruler Afḍal Khān Khatak (d. circa 1740/41). Among these texts are few Pashto
proverbs (matal) used as expressive stylistic means to answer various didactical purposes, a distich
landǝy representing a fragment of an improvised battlefield song, two short anecdotes demon-
strating how an allegorical story may have been a halfway towards pithy wise saying, and four ge-
nealogical legends, each having roots in Pashto oral tradition in contrast to those bookish accounts
which fabricated the “Biblical” origins of the Pashtuns. Literary material collected and analyzed in
the article may both contribute to the ethnological studies of the Pashtun tribes in pre-modern
times and prove the fact that the earliest recorded specimens of Pashto folklore should be traced in
the works of Pashtun literati written long before the first European researches on the subject were
published in the second half of the XIX century.

Keywords
Pashto Folklore and Literature, Proverbs, Popular Verses, Fables, Genealogies, Pashtun Tribes in
Pre-modern Times

It is a widely accepted opinion in Afghan studies that the first written recordings
of Pashto folklore texts appeared in the works of European researchers in the se-
cond half of the XIX century (Raverty 1860a; Thorburn 1876; Hughes 1873; Darme-
steter 1888-1890).1 With no intention to doubt the priority of these records as the
oldest systematized scholarly materials collected and published by foreigners it
should be noticed that folklore elements in various forms have been tangibly
present in the works of Afghan literati since the very beginning of Pashto written
literature on the edge of the XVI-XVII centuries.

1
For a detailed historical and bibliographical outline of Pashto oral literature see in the re-
cently published work by Heston (2010).

 Koninklijke Brill NV, Leiden, 2015

480 Mikhail Pelevin

In fact, Pashto written literature initially developed as a counterpoise against

Pashto folklore. It was a kind of spiritual competition between bookish culture of
newly-born Afghan intellectuals represented by a number of religious teachers
and educated tribal noblemen, on the one side, and popular oral tradition fos-
tered by countless as well as nameless illiterate Afghan bards and storytellers, on
the other. The eminent proponents of written literature explained their attacks
on “silly tales and songs” as being motivated by the purposes of people’s spiritual
enlightenment. They declaratively opposed their written works based on the
high standards of classical Persian poetry with its sophisticated religious and
philosophic ideas, deep lyric emotions, refined literary esthetics and well elabo-
rated formal instrumentalities to profane and plainly entertaining (as they
thought) oral literature composed in an "unripe language" (xāma žǝba), as Mīrzā
Anṣārī (d. after 1631), a gifted mystical poet, labeled his mother tongue (Mīrzā
1975: 190; Pelevin 2010: 10-12). Khūshḥāl Khān Khatak (1613-1689), a tribal ruler
and the leader among Pashtun men of letters, distinguished two types of poetry:
that of kings and that of pranksters, the latter being produced by lay people of
low morals who would sell their talents for a bowl of food (Khūshḥāl 1952: 539-
540). By “pranksters” (hazzāl) Khūshḥāl, obviously, meant popular bards, since
no other poetry for sale (e.g. court poetry) did exist in the Pashtun tribal society.
Despite this critical attitude towards oral tradition the same Khūshḥāl
Khatak was by no means immune to the strong and quite natural influence of
national folklore. On the contrary, he deliberately exploited folklore literary as-
sets and hence has become generally recognized as a true founder of Pashto na-
tional literature. Beside genres, originally popular but skillfully adapted to classi-
cal Arabic-Persian poetics (such as war ballad, love song, fable, lamentation,
mocking, or homily based on tribal ethics), Khūshḥāl also derived from folklore
some metrical schemes and less discernable on the surface, stylistic and figura-
tive tools which are still to be studied thoroughly.
Khūshḥāl’s numerous descendants who were engaged in literary activities
and created a large number of valuable poetic and prose works in Pashto kept on
exploring folklore tradition and introducing its accomplishments into written
literature. Romantic poems such as Dilay aw Šahǝy (1698/99) and, especially,
Ādam Khān aw Durkhānǝy (1706/07) by Ṣadr Khān Khatak (d. after 1712), Khūsh-
ḥāl’s son, were the most visible results of such endeavors. Both works are typical
examples of what may be called a literary version of folklore narrative. The very
plots, all key characters, themes and motives of his poems Ṣadr Khān took from
 Early Specimens of Pashto Folklore 481

well-known oral stories. In the preface to Ādam Khān aw Durkhānǝy (which en-
joyed among Pashtuns the same popularity as Madjnūn and Laylā in the Muslim
East or Romeo and Juliet in the Christian West) Ṣadr Khān tells that the plot of
this famed love story he drew from an oral account by a certain Tātay, a young
man from the Ḥādjīkhel clan of the Yūsufzay tribe. According to Ṣadr Khān, Tā-
tay, a great lover of poetry and good expert in local folklore, stated that his vari-
ant of the story was the most correct because he had heard it from the son of
Mīrogay, one of the drama’s real participants and its first narrator (Ṣadr 1989: 52-
53). Thus, Ṣadr Khān directly pointed at the folklore roots of his poem and even
listed the chain of the story's transmitters.
Another work of a Khatak writer where folklore elements permeate through
the text is Tārīkh-i muraṣṣa‘ (“The Ornamented History”) by Afḍal Khān (1664/
65-1740/41), Khūshḥāl’s grandson. Finished in 1724 this pioneering book on Af-
ghan historiography describes a vast range of political, military, social and
ethno-cultural developments pertaining to the legendary and real history of the
Pashtun tribes. Structurally, it is a compilation of different, sometimes badly ar-
ranged accounts—both originally composed or translated (or retold) from other,
mostly Persian, sources. In the text Afḍal Khān more than once calls himself
mu’allif dǝ kitāb (‘the compiler of the book’). However, taking into account the
great variety of written and oral sources of the book, its size (over 750 handwrit-
ten folios), and obvious stylistic dissimilarities in the language one may suppose
with certainty that Afḍal Khān had literary assistants.
Six late copies of Tārīkh-i muraṣṣa‘ have been catalogued up to nowadays; all
date back from the middle and the second half of the XIX century (Macken-
zie/Blumhardt 1965: 44-48; Hewādmal 1984: 155-157). The complete text of the
book was published once in 1974 in Peshawar by D. M. Kāmil with his preface
and commentaries (Afḍal 1974).2 This publication is based on a manuscript of
1867 from the Punjab Public library in Lahore. Additionally, D. M. Kāmil (1974: 6-
7, 58-59) used five copies more: two of them are known through the Mackenzie’s
catalogue and three unknown are from private libraries in Pakistan.
Few fragments from Tārīkh-i muraṣṣa‘ were published earlier by H. G. Raverty
in his Pashto chrestomathy “The Gulshan-i-Roh” (Raverty 1860b: 1-53). Partly the
same texts T. P. Hughes included in his anthology “The Kalīd-i Afghānī” (Hughes

2
The author expresses his thankfulness to M. Weinreich and W. Holzwarth for their kind assis-
tance in obtaining a copy of this rare edition.
482 Mikhail Pelevin

1873: 207-240); later translated into English by T. C. Plowden (1875: 167-208). Nev-
ertheless, outside the Pashto-speaking academic circles in Afghanistan and Paki-
stan Tārīkh-i muraṣṣa‘ has not yet become a well studied and familiar source on
Pashtun history and culture. What makes this book hard to particularly under-
stand is its fragmentary structure, in some places further complicated by a quite
mechanical mixture of diverse texts. Unfortunately, the Peshawar edition of
Tārīkh-i muraṣṣa‘ lacks primary textological research which would at least sepa-
rate logically unconnected passages and expose still undetected lacunas.
The original parts of Tārīkh-i muraṣṣa‘ which by the author’s intention consti-
tute the sixth and the largest chapter (lit. daftar ‘record’) of the book, more than
third of its total size (Afḍal 1974: 254-513), are of outmost importance as they rep-
resent unique documentary narratives and remarkable pieces of early Pashto
prose. Being a collection of genealogical data, historical accounts, family stories,
personal diaries and memoirs attributed to Afḍal and Khūshḥāl these texts on
the whole may be regarded as the tribal chronicle of the Khataks. In many sec-
tions of this chronicle Afḍal retells his grandfather’s diaries traditionally known
under the title Bayāḍ (“Notebook”). Although this work of Khūshḥāl’s prose work
is now considered to be lost, some of its original fragments are included in the
Tārīkh-i muraṣṣa‘ with due reference: “This is an account which the revered
Khān has written in his own hand…” (naql-dǝy čǝ xān ‘iliyyīn-makān pǝ xpǝl
dastxat kṣ̌ ǝlay wu)3 (ibid.: 298-327). For all these reasons Khūshḥāl Khān should
enjoy full rights as a co-author of the chronicle.
Below is a short survey of some folklore elements (excluding poetic meta-
phors) which can be traced in the Khatak chronicle as vivid markers of its ethno-
cultural background, on the one hand, and early examples of how Pashto oral
tradition penetrated into written literature, on the other.

PROVERBS
Resting at the very core of folklore, vocabulary proverbs are the most evident
traces of oral tradition in the chronicle, although their total number there does
not exceed a dozen of locutions.4 Together with maxims, aphorisms, wandering

3
All Pashto and Persian citations are given in contemporary phonetic transcription, i.e. with-
out transliteration characters for Arabic letters.
4
The first systematized selection of Pashto proverbs with English translations and commen-
taries was published in 1876 [Thorburn]. Other major collections belonged mostly to Afghan schol-
ars (Nūrī 1948; Benavā 1979a; Ṭāyir/ Edwards 1982). The most comprehensive work of this kind is a
 Early Specimens of Pashto Folklore 483

citations, and other phraseological units derived from written literature, prov-
erbs are regularly designated in the text by the common term matal (Arabic
mathal) or introduced by the verb wayǝl ‘to say’ (e.g. in passive perfect: wayǝli
šǝwi-di ‘it has been said’ (ibid.: 464). Afḍal Khān sometimes specifies the ethnical
attribution of the proverbs he quotes: dǝ paṣ̌ tano matal-dǝy ‘it is a proverb of
Pashtuns’ (ibid.: 425, 510), paṣ̌ tānǝ matal kā ‘Pashtuns use a proverb’ (ibid.: 504),
paṣ̌ tānǝ wāyi ‘Pashtuns say’ (ibid.: 445); also in his Persian letter to Ākhūnd
Qāsim: dar afγānān masal-i mašhūr-ast ‘among Afghans there is a well-known
proverb’ (ibid.: 472). By doing this the author obviously wishes to seperate gen-
uine Pashto proverbs from other literary material.
All the proverbs function in the text as expressive turns of speech, and the-
matically touch mostly upon social, political and military issues. Usually they are
quoted with no extra explanations, even if their semantic correlation with the
context is not very clear (what only confirms the fact that these proverbs were
widesperad and well comprehensible for local audiences):
…Also the Bolāqs came from Patyāla and Zīra. It was necessary. The hand when it
breaks, goes to the neck (lās čǝ māt ši γāṛe-lara drumi). Also the Muhmandays
came. Together we held a council… (ibid.: 309-310).
These lines Khūshḥāl Khān wrote in his diary in the summer of 1673 during
severe armed clashes with the detachments of his rebellious son Bahrām Khān
(1643-1712). Discords in his own family and henceforth within other Khatak clans
forced Khūshḥāl to ask for assistance from his remote Khatak relatives—the
Bolāqs. The literal meaning of the cited proverb—“the broken hand burdens the
neck”—arises from everyday practice when a broken or somehow injured hand
is tied (metaphorically “goes for help”) to the neck (cf. Shahrānī 2008: 129).
In other cases the context appears more verbose and lucid for reader to
properly understand the meaning of similar laconic sayings:
…The families of our people who with the coming of the enemy had been driven
away from Sarāy went to Mīla Shāhī. When the roar of the battle reached their
ears they went up through the Khāwrǝy pass and then down to Mīla Khāwrǝy. In
the evening we followed them. I recalled: “The Pashtuns have grabbed the don-
key’s tail” (paṣ̌ tano dā dǝ xrǝ lakǝy niwǝle-da). If we hold on to it too tightly, only
damage will happen out of this to all of us. So, it will be better if we take aside this

book by ‘I. Shahrānī (1975) recently reedited with corrections (Shahrānī 2008); it contains over
3000 entries (cf. G.-M. Nūrī’s classical collection with about 1900 proverbs).
484 Mikhail Pelevin

time. How long the people in each of our unit will survive without provisions?
(Afḍal 1974: 394-395).
…That year Amīr Khān who was fighting then with the Momands also came into
conflict with the Shinwārays. As it was already mentioned, he allowed the Shin-
wārays to have Daka. The goal of Amīr Khān was—as the Pashto proverb says
“One thing I am eating, another one I am cooking up” (yawa xwarǝm bǝla
paxawǝm) –to divide [these two tribes] by yielding Daka to them, and to make
them weak… (ibid.: 510).
The saying “Pashtuns have grabbed the donkey’s tail”, judging by the context,
means that a wrong way of conduct or action was taken, or that an improper
guide (lit. “donkey”) was chosen. Without context this saying might be under-
stood differently, in its direct meaning: “The Pashtuns seized the donkey’s tail”,
i.e. they stupidly satisfied themselves with only a small loot. Such interpretation
seems to have a parallel in Khūshḥāl’s verses: “A loaded camel went back home, /
while their [Pashtuns’] loot is only a ring-bell from the camel’s neck” (Khūshḥāl
1952: 340).
The literal meaning of the second proverb is to be interpreted as “One thing I
am openly doing, another one I am secretly plotting to do”.
Other Pashtun proverbs quoted by Afḍal Khān in the chronicle do not need
any comments:
- “Neither a stone turns into skin, nor an enemy into a friend” (na kāṇay post ši na
γalīm dost ši) (Afḍal 1974: 392);
- “Even if an enemy is a jackal or a fox, one should treat him like a tiger” (ka
duṣ̌ mǝn šaqāl gidaṛ wi ham ta‘bir joṛǝṣ̌ t warta dǝ mzari kǝṛay boya) (ibid.: 464;
Shahrānī 2008: 124);
- “If someone’s wattle-fence broke to pieces, whether it can be restored again?”
(pǝ haγo kasāno čǝ wadāna pakalǝy wrāna šwa os wrāna pakalǝy kǝla wadāneẓ̌ i)
(Afḍal 1974: 472).
Some scholars suggest to set apart two categories among proverbs—those
having both figurative and direct meaning and those having only literal meaning
(in Russian folklore studies these are distinguished as poslovitsa and pogovorka
(Zhukov 1991: 11)). All the examples given above would fall under the first cate-
gory. Some sayings mentioned by Afḍal Khān do not have metaphorical conno-
tation and thus belong to the second category:
- “Honor leaves the homeland and the home of the relatives as well” (nang dǝ wa-
tan aw dǝ xpǝlo xwešāno lǝ kora ham pāci) (Afḍal 1974: 425);
 Early Specimens of Pashto Folklore 485

- “I am after your good, and you are after my death” (zǝ dǝ tā pǝ barg tǝ dǝ mā pǝ
marg) (ibid.: 445);
- “Do not shoot or do it faster” (wištǝ-de ma ša lā-de gaṛanday kṛa) (ibid.: 504).

The first saying expresses the very essence of Pashtuns’ national unity and
their common historical and cultural heritage, for it means that every single
Pashtun family is responsible for the honor and dignity of the whole nation (here
watan ‘homeland’). The second one seems to express a routine complaint in the
sense: “I am doing good to you, and you are doing bad to me”. However this
complaint implies one of the major tenets of Pashtun customary law—the prin-
ciple of equal compensation (talio).
Of special interest is the case when a popular Pashto saying is cited not di-
rectly, but rather disguised. Such hint to a well-known proverb is placed into the
words of the Pashtun rebels pronounced on the eve of the grand battle in the
Khyber Pass which marked the beginning of the Mughal-Afghan war of 1672-76.
Relating this episode Afḍal follows the Khūshḥāl Khān’s diary:
…He [Kabul governor Muḥammad Amīn Khān] sent Mustadjāb and other Mo-
mands to them [the mutinous Pashtun tribes] for a meeting (jirga) [with a re-
quest]: “Leave the road free!” Those [who were sent] said to the Pashtuns some-
thing openly and something secretly. The Pashtuns replied to them: “Our hun-
dred years of life now have come to an end (zmunẓ̌ sǝl-kāla ‘umr os pura šǝway-
dǝy). You go away immediately! Those arrows which are in your quivers will help
you much”… (ibid.: 297).

The words “our hundred years of life now have come to an end” is a rephras-
ing of the famous proverb: “The Pashtun took his chance (i.e. he avenged) one
hundred years later and yet said: I hurried up” (paṣ̌ tun sǝl kāla pas xpǝl wār wāx-
ist ham wayǝl-ye čǝ jalti-me wukṛa) (cf. Shahrānī 2008: 31; Khādim 1952: 43). This
saying well reflects the Pashtuns’ attitude towards the rule of equal compensa-
tion and the duty of vengeance as a kea stipulation of this rule. In Khūshḥāl’s ac-
count Pashtuns make a hint to this saying in order to remind the reasons, at least
the outward ones, of their insurrection. According to the other report from the
chronicle, the Mughal-Afghan war was provoked by a local incident when some
Mughal officers abused a woman-trader from the Ṣāfī tribe and then were killed
in revenge by tribesmen who in turn were punished by the Mughal authorities
(Afḍal 1974: 295-297).
486 Mikhail Pelevin

LANDƏY
For many reasons the landǝy distich may be regarded as a corner-stone of Pashto
folklore poetry. Landǝys are widespread and highly popular all over the Pashtun
territories. Having a specific metrical structure5 and a genuine Pashtun designa-
tion (from adj. lanḍ ‘short’)6 these distiches belong to the oldest forms of national
oral poetry, unlike no less popular verses as ghazal or chārbayta, which
developed apparently under the influence of written literature. Besides, the
landǝy serves as a basic strophic element in other folklore verse forms, e. g. in the
sarokay (or nīmakǝy) where the main text is composed like a chain of landǝys.
Several published landǝy collections mark definite steps in scholarly gather-
ing, arranging and investigating these verses (Darmesteter 1888-1890; Nūrī 1944;
Zhwāk 1955; Benavā 1979b; Shahīn 1984)7. It is noteworthy that the earliest landǝy
selection is included in the work on Pashto grammar and lexicography Farhang-i
Irtiḍā’ī (1810) authored by Muḥammad Irtiḍā’ Khān ‘Umarkhel, a descendant of
the Pashtun rulers of Rampur (Hewādmal 1984: 189-190, 297-299; Kushev 1980:
74-75). The unique manuscript of this book, written in Delhi on the request of
“an Indian Raja” is now kept in the Public library in Patna. The last, 166th section
of the work, entitled “On the understanding of Pashto verses…” contains twenty
three landǝys with translations into Persian and Hindustani.
Some Afghan philologists believe that the oldest among known landǝys is a
distich allegedly composed in the times of Sultan Maḥmūd Ghaznavī (ruled 998-
1030) during the latter's military campaigns in India (see Rafī‘ 1970: 104). In this
verse a girl expresses her longing for her beloved who is returning from a ride.
The verse was recorded in the second half of the XXth century, and the view that
it has authentic roots in medieval antiquity is based only on oral tradition.

5
Each line in a landǝy has a fixed number of syllables and its own rhythmical pattren (see also
Mackenzie 1958): ххх-(х) / ххх-(х) // ххх-ххх-ххх-х. The metrical scheme of the first line—varied
in clausulas ten-syllable verse with a caesura—appears sporadically in the poems of mystical
Roshānī poets in the first half of the XVII century. About the same time Khūshḥāl Khān applied
this scheme as the regular meter for his numerous (over 1500) quatrains tsalorīdzī, thus clearly im-
plying the folklore origins of this verse form (Pelevin 2005: 139-152).
6
Other local names for these verses are tapa and tīkǝy which prevail mostly in the north-east-
ern areas of the Pashtun territories (particularly in the Peshawar Plain and the adjacent valleys of
the south-eastern Hindu-Kush) (Hewādmal 2000: 5-6).
7
On landǝy’s form, content, literary and social functions in detail see (Rafī‘ 1970; Ṭāyir 1980;
Lāyiq 1984).
 Early Specimens of Pashto Folklore 487

A historically authentic and genuinely antique landǝy is preserved in the

Khatak chronicle. Afḍal Khān quoted this distich in his diary in the summer of
1711 when on the new wave of a large-scale armed conflict between the Mughal
empire and the Pashtun tribes the fight for political power flared up again within
the Khatak ruling family. Khūshḥāl’s son Bahrām Khān and his supporters con-
cluded against Afḍal Khān an alliance with some Yūsufzay clans under the
command of Djamāl, the son of Ḥamza Khān Yūsufzay with whom Khūshḥāl had
been at odds several decades before. The detachments of Bahrām and Djamāl
marched victoriously through the northern Khatak lands adjoining the Kabul
river and attacked the town of Sarāy-Akora, the residence of the Khatak rulers.
Afḍal Khān was forced to leave his ancestral seat and escaped with his men to
the highlands in the south of the Khatak territories. Recounting these events
Afḍal made such a record:
…They [the Yūsufzay allies of Bahrām] were so delighted with all these robberies
and debaucheries that even sang a song (sandǝra) about it. In this song these
cowards praised Djamāl, the son of Ḥamza. Such things they sang:
Let Djamāl Khān be blessed by God!
He drove a certain person [i.e. Afḍal Khān] to the very-very big mountains.
(dǝ xdāy rahmat pǝ Djamāl Khān ša
flāna-ye wāṛāwǝ tǝr loyo loyo γruno) (Afḍal 1974: 395-396).
Although Afḍal Khān calls this couplet simply a song (sandǝra is a common
appellation for Pashto folklore verses), its metrics reveals that it is nothing else
than a typical landǝy. By genre attribution this Yūsufzay landǝy, or tapa in local
terminology, is a tūra (war, bravery) song. In this kind of verses eulogy to a mili-
tary commander or a tribal chief as well as derision of an enemy are standard
subject matters.8
Traditionally, landǝys are verses composed and sung by women. The bulk of
published landǝys are specimens of women’s poetry where love motives domi-
nate over all others.9 In this respect, the definition which Afḍal Khān employs for

8
These motives are well seen in modern Pashto tūra songs too, e. g. in those belonging to the
poetry of the Taliban. In these songs one may come across praise to Mullā Muḥammad ‘Umar, the
military and spiritual leader of the Taliban movement (“I am proud of him, Haydari, for he is a
symbol of national unity: / The leader of victorious men, the fighter for faith, Muhammad ‘Umar,
he is…”) or scornful remarks about the enemies of old (“The skull of an Englishman is hanging on
your saber of bravery, / You hacked to pieces every one of them. / The Russians fell down under
your feet, the Russians cannot stand up again…”) (Pelevin/Weinreich 2012: 55-56, 64-66).
9
The gender of the anonymous authors of these verses reveales itself not from their contents,
488 Mikhail Pelevin

the anonymous authors of the landǝy he cites (“cowards” in my translation) de-

serves special attention. The word is nar-ṣ̌ ǝdze (lit. ‘manlike women’ and, in
figurative sense, ‘courageous women’ or, pejoratively ‘cowardly men’), which is
an uncovered taunt with an allusion to the landǝy’s historical “gender origins”.
In conclusion it should be accentuated once more that the landǝy cited by
Afḍal Khān in the Khatak chronicle has every reason to be considered if not the
oldest so at least the earliest exactly dated (summer of 1711) piece of this folk
verse form.

FABLES AND GENEALOGICAL LEGENDS

Two short texts in the chronicle have clear stylistic features of a fable, a narrative
allegory with moral teachings. One is told by the Kabul governor Muḥammad
Amīn Khān during a conversation with Khūshḥāl Khān before the Khyber battle
in April 1672. The rather laconic text of this fable is repeated twice—in
Khūshḥāl’s diary and in Afḍal Khān’s rendering of the same story (Afḍal 1974:
297, 300). The fable consists of only three phrases:
There was a man. He put in his bag a snake dying of cold, in order to warm her
up. Later the snake stung the man…
The remarkable brevity of this fable directly corresponds to the definition
given to it by Khūshḥāl: dǝ rā-ta dā matal wukǝṛ (“he told me a fable”). Appar-
ently, in Khūshḥāl’s understanding, short fables and proverbs related to one type
of literary texts which were characterised by the presence of allegory in them
and, thus, could both be referred to by the term matal.
Terminological attribution of fables and proverbs to one genre category may
be viewed as an oblique prove of the fact that some proverbs developed from
allegorical stories. A halfway step in this process would be a short fable, such as
the example cited above. There are some proverbs in Pashto which have the
same meaning as the fable told by Amīn Khān: mār rāwǝṛay mār xoṛǝlay (“The
one who brought a snake [to his house is the one whom] this snake bit”), or mār-
ta lastuni-ke dzāy ma war-kawa (“Do not place a snake in your sleeve!”) (Shahrānī
2008: 137). In modern Pashto the latter proverb developed further into the verbal
idiom mār-ta pǝ lastuni-ke dzāy war-kawǝl ‘to place a snake in one’s sleeve’, i.e.
‘to warm a snake in one’s bosom’.

but also through the morphological gender markers. It is certainly not by chance that the book-
cover of one of the most popular landǝy collections is illustrated with the picture of a woman
(Momand 1984).
 Early Specimens of Pashto Folklore 489

The second short fable was told to Afḍal Khān by a man named Ghanī. In the
past Ghanī was an associate of Nāmdār Khān, Afḍal’s younger brother and his
tireless rival in quest for power. During an internal tribal conflict between Afḍal
Khān and the Khatak sheikhs in 1724, Nāmdār secretly sided with the sheikhs,
Khatak spiritual authorities, while his former friend Ghanī joined the lines of
tribal aristocracy under Afḍal’s leadership. In his conversation with Ghanī which
took place in February (or March?) 1724 Afḍal Khān made a rhetorical question
about the motives of Nāmdār’s treacherous behavior: “One day I asked [him]:
‘Ghanī, what do you think, why does he always do such indecent things and
never feels respect for us, his kin?’” In reply Ghanī recounted the following fable
which is very similar to an anecdote:
Two Hindus served as stewards somewhere. One of them was named Bakhū, an-
other one—Gūrūdith. Bakhū attained high standing, acquired honor and esteem.
He was always at work. People came to him [with their demands]. The another
one suffered disgrace. He became humiliated and had no dignity at all. Every-
thing he did was for nothing. When a chance turned up he approached Bakhū, sat
down beside him and emitted winds so that their bad smell would reach him.
Bakhū became tired of that, but whatever he told [Gūrūdith], it was to no avail.
When these things had happened several times someone asked [Gūrūdith]: “Why
are you doing that? What are you after?” [Gūrūdith] said: “I do not have any other
way. The grief he undergoes counts for a thousand benefits to me.” Bakhū learned
about that; someone told him about it. Then he stopped [these doings]. He said
to his men: “Do not let him approach me!” And that resembles exactly what [is
happening now]. He [Nāmdār Khān] has no other way; all these things are the
same winds of Gūrūdith… (Afḍal 1974: 461-462).

The characters of this fable suggest that the story has Indian roots. Presuma-
bly, it was current among Hindustani-speaking Muslims. It seems very probable
that the narrator of the story—Ghanī—came from this milieu too. Afḍal Khān
nowhere mentions the latter's tribal affiliation, although he is usually extremely
accurate about that as far as ethnical Pashtuns are concerned.
Alongside with these two fables, four short narratives in the chronicle di-
rectly stem as well from oral tradition. These are Khatak genealogical legends in-
cluded in the first section of the chronicle titled “An account of our lineage (an-
sāb) and the state of affairs of our tribe…”. Initially, Khūshḥāl Khān recorded
these texts in his Bayāḍ, and it is likely that Afḍal Khān reproduced them in the
Tārīkh-i muraṣṣa‘ with no significant textual changes.
490 Mikhail Pelevin

Khūshḥāl was a great expert in tribal history and well aware of his family tree.
There are some verses in his lyrics which prove his deep interest in genealogical
issues (see Pelevin 2001: 61-63). A whole paragraph in his didactical treatise Das-
tār-nāma (“A Book on Turban”) (1665) deals with Pashtun genealogy (Khūshḥāl
1966: 83-85). In the beginning of this paragraph Khūshḥāl writes: “On the lineage
of Pashtuns we heard and saw many stories. In studying this subject I made a lot
of efforts”. This introductory remark opens a short summary of some curious
facts related to Pashtun legendary genealogy. All these facts Khūshḥāl “saw” (i.e.
read) in written sources. Most of them were based on Muslim mythology and
cover pseudo-historic antiquity that speaks in favor of bookish, rather than
folklore origin of the Dastār-nāma's genealogical legends.10
On the contrary, all genealogical stories from the Tārīkh-i muraṣṣa‘ belong to
what Khūshḥāl “heard” from his ancestors, i.e. to Pashtun oral tradition. It is not
without reason that two of these stories begin with the words wāyi čǝ (“[they] say
that”, “it is said that”) whereas the Dastār-nāma's legends are introduced by the
phrase pǝ mu‘tabar tārix-ke-me lidǝl (“in a reliable history [book] I saw”). Ge-
nealogical legends which came from folklore lack Biblical-Quranic personages
and, notwithstanding their fabulous content, in some aspects reflect authentic
Pashtun lineages and historical realities.
Naturally, the degree of literary imagination in such a legend depends on the
interval which separates its presumed time of action from the moment of writ-
ten recording. The latest among the Tārīkh-i muraṣṣa‘s legends seems to echo
true historical facts. It narrates how the branch of the Tsalozays emerged on the
Khatak family tree:
…And the Tsalozay clan (tǝpa) which [originally] belongs to the Khalīl tribe of
the Ghoryakhel group is now a part of the Tarays. Here a short description of how
this came about: ‘Alī Khān and Malī Khān were two brothers from the Watars.11
They pitched their tents in a wasteland. A young man of wild appearances came
to their tents. Some years he spent with them. [Then] his own brothers went to
look for him. They found him and said: “We shall take him with us”. But he did
not agree to go [with them]. It is said that his name was Rashīd. He was a son of
Ya‘qūb Khalīl. He had a very big bow, and his brothers recognized him [exactly]
when he was drawing it. Then he ran away somewhere into the desert, to the

10
Khūshḥāl’s main written source was Tārīkh-i Khāndjahānī (1611/12), a work in Persian by
Ni‘matallāh Haravī (English translation of its abridged version see in Dorn 1829-1836). The Pashto
translation of this work represents the first parts of the Tārīkh-i muraṣṣa‘.
11
The Watars belong to the Tarays, a large subgroup within the Khatak tribe.
 Early Specimens of Pashto Folklore 491

mountains. In the end he went back again to the house of ‘Alī Khān. When
people learned of his lineage and that he was pure-blooded (aṣīl) [Pashtun] ‘Alī
Khān married his daughter to him. They gave birth to five sons. The descendants
of these sons prolong their kin nowadays… (Afḍal 1974: 263).
The events described in three other genealogical stories take place fifteen-six-
teen generations (i.e. approximately four centuries) before the times of
Khūshḥāl Khān. These stories which describe episodes from the legendary lives
of Karlān, the founder of the Karlānays—one of the four main Pashtun tribal
groups, Luqmān Khatak, the forefather and the eponym of the Khataks, and
‘Uthmān Afrīday, the common ancestor of the Afrīdays, contain clearely
imaginative and even fairytale elements:
It is said that Karlān was born from his mother on the lands of a nomadic camp
(zmǝka dǝ aylāq aw dǝ qišlāq). For certain reasons [people] went to their pastures
(lit. aḥšāmāt ‘live-stocks’), but he was forgotten [on the place]. When they re-
membered about him, Ormar who was the brother of Hūnay returned back after
his nephew and found him. There was also a cauldron left by them. [Ormar] laid
Karlān into the cauldron, put it on his head and brought [him in this way] to the
pastures. Ormar did not have his own son. He said to Hūnay: “I have brought your
son here with much effort. Do not take him from me. I shall bring him up. And
you, take this cauldron!” Hūnay agreed. Since he exchanged him for a cauldron,–
and Pashtuns call cauldron kaṛahay,–[the boy] was named Karlānay because of
that cauldron. When he reached full age Ormar married his daughter to him.
They had many children. There are several groups of people who are descendents
of Karlānay… (idem.: 255).
It is said that Luqmān became known under the name Khatak for the following
reason. [Once] Luqmān, Utmān, ‘Uthmān and Dzadrān–brothers all of them–
went for hunting. At the same time four girls from an Afghan tribe walked
through the steppe. They came across each other. [The brothers] said: “Let us cast
lots on these four [girls]. This is a great game!” Luqmān was the eldest among
them. He did not agree to cast lots and said: “I shall choose one of them [first],
then you cast lots on the three remaining!” They did in this way. From the four
girls Luqmān took the one who was dressed in beautiful garments. But when he
saw her [closely] her good looks did not match her clothing at all.
A verse:
[It happens] very often that a pretty face is [only] under a veil,
But when you open it there is a granny.
The other [girls] turned out to be better-looking than that one. The other broth-
ers cast lots on them and took them according to their lots. When they saw in
492 Mikhail Pelevin

what situation Luqmān found himself they said with a laugh: “Luqmān is stuck in
the mud (…pǝ xaṭa lāṛ).” And this is a saying (matal) among Pashtuns: If for some
reason someone finds himself in a difficult situation people say about him: “He is
stuck in the mud.” This is why Luqmān got the name Khatak…
Among the progeny from those four girls Luqmān had more descendants than
the others. That girl whom Luqmān had chosen and whom he owed his name
Khatak to was called Sabāka. She was dark-skinned, sturdily-built and clever. Two
of Luqmān’s sons were born from her—one Tormān, another Bolāq. Tormān
looked like his father, and Bolāq was dark-skinned like his mother… (ibid.: 256)
…And ‘Uthmān is called Afrīday for the following reason. [Once] guests came to
him. And he looked very miserable, ill-favoured and confused. It was winter time.
Because of the cold all people were sitting close to each other. The guests asked
about [‘Uthmān]: “Who is this one?” [Someone] said: “An āfarīda (Pers. ‘created’)
too”, that is “A creature of God”. This is why ‘Uthmān got the name Afrīday…
(idem.: 256)

Undoubtedly, the quoted legends reflect some real facts from the Pashtun
people’s national history, supporting the view of ethnical diversity in the genesis
of Pashtun tribes (see Morgenstierne 1986: 216-220). Thus, the legend about Kar-
lān proves that the Karlānay tribal group grew up through the assimilation of
neighboring peoples (including Ormurs, a small ethnic group speaking a West-
Iranian idiom) who inhabited the Eastern Hindu-Kush and the Indus valley re-
gions. The brothers Hūnay and Ormar (i.e. Ormur) mentioned in the legend—
Karlān’s consanguine father and foster-father accordingly—are reported to have
been foster-sons in the family of Sharkhbūn, the founder of the first branch of
the large Pashtun tribal group known as the Sarbanays (Caroe 1958: 20-21). The
name Ormar clearly indicates that both brothers and hence their offspring Kar-
lān were not “pure-blooded” Pashtuns.
Historically authentic in the legend about Luqmān Khatak seems to be the
description of the appearances of his “fortuitous” bride whom he met in the
steppe while hunting. The girl is portrayed as dark-skinned (lit. rang-ye tor wu
‘she was of black color’), and her son Bolāq “was dark-skinned (siyāhfām ‘of black
color’) like his mother”. Supposedly, it was the dark skin that made Luqmān’s
bride unattractive in the eyes of the other Pashtuns. The emphasis which the
story puts on dark color of skin, which in Pashtun perception of ethnicity is as-
sociated almost exclusively with Indo-Aryan origin, echoes the fact that Indian,
or Dardic, ethnical elements had a significant share in the genealogy of the
 Early Specimens of Pashto Folklore 493

Khataks who inhabit the eastern parts of the Pashtun territories situated on the
very border with the Indo-Aryan world.
It is also noteworthy that the plot of these folklore legends, as well as of two
genealogical legends from the Dastār-nāma (Khūshḥāl 1966: 85), are built
around the explanation of ethnic names. Such interest towards the etymology of
Pashtun ethnonyms which for the most part still do not have precise and un-
questionable interpretations reveals the level of national history studies in pop-
ular culture, which was scholarly by approach, but pseudoscientific in its results.
On the other hand, genealogical stories with focus on etymology correspond to
the principal meaning of the very term “folklore” as common knowledge in gen-
eral, and not just literary forms of expression of this knowledge.

BIBLIOGRAPHY
Afḍal, Khān Khatak (1974), Tārīkh-i muraṣṣa‘, ed. D. M. Kāmil Momand, Peshawar.
Benavā, ‘A. (1979a), Pashto matalūna, Kabul.
―― (1979b), Landǝy. Kabul.
Caroe, O. (1958), The Pathans 550 B. C.- A. D. 1957, London.
Darmesteter, J. (1888-1890), Chants populaires des Afghans, Paris.
Dorn, B. (1829-1836), History of the Afghans: translated from the Persian of Neamet Ullah. P. I, II.
London.
Heston, W. (2010), “Pashto Oral and Popular Literature”, Oral literature of Iranian languages: Kurd-
ish, Pashto, Balochi, Ossetic, Persian and Tajik, eds. Ph. G. Kreyenbroek; U. Marzolph, Lon-
don, New York: 135-166, 346-350.
Hewādmal, Z. (1984), Dǝ Hind dǝ kitābkhāno pashto khaṭṭī nuskhe, Kabul.
―― (2000), Dǝ Pashto adabiyāto tārīkh (The History of Pashto Literature), Peshawar.
Hughes, T. P. (1873), The Kalīd-i-Afghānī, Being Selections of Pushto Prose and Poetry for the Use of
Students, Lahore.
Kāmil, D. M. (1974), “Preface”, Tārīkh-i muraṣṣa‘ , ed. D. M. Kāmil Momand, Peshawar: 5-72.
Khādim, Q. (1952), Pashtūnwālay. Kabul.
Khūshḥāl, Khān Khatak (1952), Kulliyāt, ed. D. M. Kāmil Momand, Peshawar.
―― (1966), Dastār-nāma (A Book on Turban), 1966.
Kushev, V. V. (1980), Afganskaja rukopisnaja kniga (očerki afganskoj pis'mennoj kul'tury), Moskva.
Lāyiq, S. (1984), “Landǝy”, Pashto landǝy, ed. M. A. Momand, Kabul: I-LX.
Mackenzie, D. N. (1958), “Pashto verse”, Bulletin of School of Oriental and African Studies, xxi, 2: 319-
333.
―― / J . F . Blumhardt (1965), Catalogue of the Pashto manuscripts in the libraries of the British
Isles, London.
Mīrzā, Khān Anṣārī (1975), Dīvān, ed. Dost, Kabul.
494 Mikhail Pelevin

Morgenstierne, G. (1986) “Afghān”, The Encyclopaedia of Islam. New ed. I, Leiden : 216-221.
Nūrī, G.-M. (1944), Millī sandǝre (National Songs), Kabul.
―― (1948), Matalūna (The Proverbs), Kabul.
Momand, M. A. (ed.) (1984), Pashto Landǝy, preface by S. Lāyiq, Kabul.
Pelevin, M. S. (2001), Khushhal-khan Khatak (1613-1689): načalo afganskoj nacional'noj poezii, S.-Pe-
terburg.
―― (2005), Afganskaja poezija v pervoj polovine—seredine XVII vekai, S.-Peterburg.
―― (2010), Afganskaja literatura pozdnego srednevekov'ja, S.-Peterburg.
―― / M. Weinreich (2012): “The Songs of the Taliban: Continuity of Form and Thought in an Ever-
changing Environment”, Iran and the Caucasus 16. 1: 45-70.
Plowden, T. C. (1875), Translation of the The Kalīd-i-Afghānī, Lahore.
Rafī‘, Ḥ. (1970). Dǝ khalqo sandǝre aw dǝ folklor pǝ bāb landa tserǝna. Kabul.
Raverty, H. G. (1860a), A Grammar of the Puk'hto, Pus'hto, or Language of the Afghans, London.
―― (1860b), The Gulshan-i-Roh: Being Selections, Prose and Poetical, in the Pus'hto, or Afghān lan-
guage, London.
Ṣadr, Khān Khatak (1989), Ādam Khān aw Durkhānǝy, ed. M. N. Ṭāyir, Peshawar.
Shahīn, S. (1984), Rohī sandǝre (tape), Peshawar.
Shahrānī, ‘I. (2008), Pashto matalūna yā ḍarb al-amthāl, [s. l.].
Ṭāyir, M. N., (1980), Tapa aw zhwand, Peshawar.
―― / T. C. Edwards (1982), Rohī matalūna, Peshawar.
Thorburn, S. S. (1876), Bannu, or Our Afghan Frontier, London.
Zhukov, V. P. (1991), Slovar' russkix poslovic i pogovorok, Moskva.
Zhwāk, M. (1955), Pashtanī sandǝre, vol. I, Kabul.
 Activist, Professional, Family Man:
Masculinities in Marjane Satrapi’s Work

Nagihan Haliloğlu
Alliance of Civilizations Institute, Istanbul

Abstract
This study aims to explore the trajectory of the meanings of masculinity from pre-revolution to
post-revolution Iran, as represented in Marjane Satrapi’s work. Both Chicken with Plums and Per-
sepolis are autobiographical works that chart her relationship with male members of her middle
class family and country at large before, through and after the revolution. She provides a nuanced
understanding of masculinity that recognizes plurality and hierarchy- men as lay people, as mem-
bers of a family, as practioners of art, within a historical context. In both her memoirs men and
women suffer from ideals of manhood and womanhood presented to them, calling into question
whether men are beneficiaries of the patriarchal order of the pre or post-revolution Iran. Satrapi
demonstrates how the most important building block of masculinity has changed, from class and
professional membership to piety in post-revolution Iran. She also exposes masculinity’s core ele-
ment to be self-sacrifice. It is the regime, capitalist or Islamist, that seems to decide for what those
sacrifices are to be made. However, Satrapi tries to salvage an understanding of masculinity that
has honour and comradeship at its core and acquaints the reader with practitioners of this gender
model as she would have it be.

Keywords
Masculinity, Iran, gender roles, revolution, emancipation, westernization, piety

Marjane Satrapi’s work is expectedly very alert to gender politics, given that the
revolution that has led to her eventual exile expressed itself in terms of gendered
oppression. Using the Iranian revolution’s treatment of women as a starting
point, in her work Satrapi moves on to questions of gender relations both before
and after the revolution in Iran and also in Europe. I argue that Satrapi exposes
the reader to a multiplicity of masculinities as products of their own particular
cultural moment. While feminists have detailed the plight of Muslim women’s
lives under patriarchy, men, collectively, have been treated as mere beneficiaries
of the gendered order (Kandiyoti 1991). The genre of women’s memoirs that

 Koninklijke Brill NV, Leiden, 2015

496 Nagihan Haliloğlu

come from the Islamic world, and from Iran in particular, has become a
knowledge industry in which
[T]he “silenced” Iranian woman finally finds a voice with which to speak, these
memoirs reproduce reductive but familiar narratives which pin the constructed
"Third-world woman" against her male counterpart while setting the stage for what
is presumed to be her salvation.1
The concerns raised in the above quoted manifesto entitled ‘A Genre in the
Service of Empire’ go to the heart of the material that Satrapi deals with in her
graphic memoirs Persepolis and Chicken with Plums. While her memoirs seem to
follow the genre of the conventional female-emancipatory narrative coming
from Muslim countries, neither her memoirs, nor her medium fits comfortably
with the genre conventions or the political consequences that Akhavan, Bashi,
Kia and Shakhsari outline above. In these memoirs Satrapi recreates the lost
space of her Iranian childhood and youth through black and white comic book
figures that hark back to wood-carved figures.2 I read Satrapi’s work as an
oblique ‘writing back’ to the popular narratives of collective male brutality, an
attempt to expose different kinds of masculinities that are not necessarily an-
nexed to image of the male clerical elite. Satrapi attempts a nuanced depiction
that recognizes the plurality and hierarchy of masculinities- men as lay people,
as members of a family, as practioners of art. In her memoirs Satrapi exposes
gender relations in both pre-revolutionary and post-revolution Iran and lets the
reader explore the transformation of male role models within a historical con-
text. In both her memoirs men and women suffer from the ideals of manhood
and womanhood presented them: in Persepolis, the focus is on the effect of ‘hy-
permasculinity’3 promoted by the Islamic regime on gender relations, while

1
You can find a more extended version of this article entitled ‘A Genre in the Service of Em-
pire’ at http://www.zmag.org/content/showarticle.cfm?itemid=12010.
2
In the Muslim Other, through certain popular industries of knowledge mentioned above, the
European reader has come to see barbaric masculinity. Even in the women who wear the chador,
western critics can only see the male imposition, and not the women who are actually wearing it.
From a completely formal point of view, what the graphic narrative does, with ease, is to picture
Iranian women both veiled and non-veiled, which may facilitate a western reader’s understanding
of Iranian women as individuals. As emphasized by these images we understand that compulsory
or otherwise wearing of the chador may challenge but does not erase women’s individuality at
large. Through the use of the graphic genre Satrapi repudiates a way of seeing Iranian women only
as a function of a dress code imposed by men.
3
For an extended discussion see Gerami 2003: 257-274.
 Activist, Professional, Family Man: Masculinities in Marjane Satrapi’s Work 497

Chicken with Plums charts a male musician’s life, ruined due to traditional con-
cepts of masculinity- as an artist he is not considered ‘man enough’ to marry the
woman he loves.
The constellation of masculinities that Satrapi’s characters have to contend
with has been aptly described by Shahin Gerami, and her categorizations will
inform my readings of Satrapi’s characters’ engagement with post and pre-revo-
lution masculinities. This parallel reading enables us to see the differences and
continuities in gender relations between the periods before and after the revolu-
tion. While Satrapi, like Gerami, acknowledges the role that piety plays as the
basis of post-revolution Iranian masculinities, she exposes the hollowness of the
symbols at the heart of the regime’s constructs and tries to re-inscribe the role
that familial, professional and class alliances play in the construction of male
identities.
It is almost a truism to claim that maleness is as socially constructed as fe-
maleness- as indeed any other aspect of identity. It will serve us, however, to re-
member what ingredients go into constructions of gendered identity so that we
can understand how Satrapi’s view of masculinities are shaped. As Emma Sin-
clair-Webb (2006: 7) argues:
Factors of class, labour market relations, ethnicity and sexuality, as well as individ-
ual experience and relations with family and peers, are centrally implicated in the
formation of men’s identities, in patterns of association and in the categories men
find themselves occupying and sometimes also consciously seek to occupy.
In Satrapi’s work class and family relations become the important markers
through which masculinity is defined. Naturally, Satrapi’s ideals of masculinity
emerge from her own class, and in fact, from her immediate family. Masculinity
is, simply defined, behaviour and traits that are deemed appropriate for a man in
a given culture, and is very much associated with those who control and fully
participate in its dominant institutions.
While we have come to understand and experience masculinity as a tool of
patriarchal male hegemony, masculinity, as a basis on which men construct their
identity, it is conventionally associated with all things good, brave, loyal and be-
nevolent. In that sense, it covers a wide variety of virtues that ‘if needs be’ can be
taken on by women. Satrapi deals with this association of laudable qualities with
masculinity, this normativity of masculinity, head on in Persepolis by relating
that as a child she had always wanted to become a prophet. In the book there is a
frame in which she depicts various prophets and then underneath, in another
498 Nagihan Haliloğlu

frame, she, as a child, appears smiling at the line of prophets, her head shining
like the sun. The prophets knit their brows to ask “A Woman?” (Satrapi 2007: 7).
However, this does not affect her relations with God, and she continues to hold
council with him quite late into her teens. Marjane4 is happy acting out male role
models, playing Che Guevara to her male friends’ Fidel and Trotsky in their
child’s role-play (ibid.: 10). Indeed, Satrapi makes clear from the very start that
her family belong to the left leaning middle classes, and we realize that in their
ranks, as well, male role models are the norm.
It is generally believed and argued that the Islamic revolution discredited
some prerevolutionary masculinity types such as artists and certain kinds of
professionals. However, strict demarcations that disprivilged the ‘artistic’ were,
as Satrapi shows, already in place before the revolution. Gerami (2003: 236) iden-
tifies a list of pre-revolution male categories and hierarchies as follows:
In prerevolutionary culture, the central masculine prototypes were the merchants
(bazaaris), the professionals, engineers, doctors, and professors (doctor -o- mo-
handes), military officers (sarhang- va- afsar), civil service employees (karmands),
workers (kargar and amaleh), and peasants (dehati). Less significant peripheral
prototypes were urban cowboys (jahel) and misfits (lat-ha).
The influence of the labour market, and that of the professions on masculini-
ties in pre-revolutionary Iran as suggested by Gerami can be clearly seen in Sa-
trapi’s work as well. In Chicken with Plums, which is set in the pre-revolution pe-
riod, Satrapi tells the story of an uncle called Nasser Ali whose life is practically
ruined because the father of the woman he loves forbids their marriage. The fa-
ther believes working as an artist Nasser Ali won’t be able to take good care of his
daughter. He is reminiscent of another uncle, Anoosh, in Persepolis, who is a
leftist activist who ends up in prison. Anoosh speaks to Marjane about his ideals
and his lovelife with great openness and carves out swans from the pieces of
bread that they give him in prison. He becomes the type of masculinity that Sa-
trapi seems to favour the most, that of the poet warrior. We know that Satrapi
has little time for inactive men, or men who don’t fight for the values they be-
lieve in. In the section of the book that she speaks about her time in Europe she
portrays a conversation she has with her Austrian boyfriend who puts on a show
of being a socialist. She is going out to demonstrate against the Nazis, he refuses
to go along and continues to type at his desk and says:

4
When referring to narrated Marjane Satrapi as character I will use ‘Marjane’ and when refer-
ring to the narrating Marjane Satrapi, the author I will use ‘Satrapi’.
 Activist, Professional, Family Man: Masculinities in Marjane Satrapi’s Work 499

– I write. Culture and education are the lethal weapons against all kinds of funda-
mentalism. We have to educate the people so that they don’t vote for Nazis.
– Yeah, the intellectuals are too precious to waste their time shouting!
– Whatever…
– In any case, it is the cowardice of people like you who give dictators the chance to
install themselves (Satrapi 2007: 229).
After this exchange between the figures, in the narrator’s box Satrapi writes:
“These arguments marked the beginning of the end of our story” (ibid.). This is
an expected attitude from Marjane, who grows up reading Marx, in a house full
of political discussion and political activism. Indeed, she has been groomed, in a
way, as the perfect rafiq, mostly among male playmates. A rafiq as Germani
(2003: 236) explains:
was a young man with short hair, wearing regulation blue jeans and shirt, an army
jacket for winter, heavy shoes, and all together having a crumpled and unkempt ap-
pearance. A must was a heavy mustache. If a beard was adopted, it must be a goatee
for one not to be confused with the Islamic groups. Women comrades were expected
to emulate this appearance as much as possible.
The rafiq is at once a companion in whose company one can strive for a bet-
ter world and, as exposed by Gerami, a ‘flight from the feminine’.5 The flight from
the feminine in Iran right before/during the revolution is ideologically charged
as women, who were forced to take off their veil and encouraged to participate
in public life during the Shah’s regime became the most visible faces of westerni-
zation, and thus an easy figure for censure for the anti-imperialists. Thus, women
who wanted to be within anti-imperialist movements had to renounce West-im-
posed cultural norms and develop their own, and downplaying feminine aspects
of one’s appearance became part of resistance. As Gerami (2003: 262) observes,
informed by Afsanah Najmabadi,6 the westernized women dressed in European
clothes were at the receiving end of every political group’s censure.
The censure women get for ‘westernized behaviour’ is treated, albeit some-
what obliquely, in Chicken with Plums as well. Women are shown to be finding
their feet in the public space, and not always in a welcome manner, even within
‘enlightened’ circles. The scene in which Nasser Ali sees Irane for the first time is
very telling about the anxieties of the ‘westernized woman’ who seems to be de-
monized by all ranks of Iranian men- an anxiety couched, maybe, in a discourse

5
For a discussion of the idea of ‘the flight from the feminine’ see also Kimmel 2004.
6
For an extended discussion see Najmabadi 1991.
500 Nagihan Haliloğlu

of anti-imperialist and religious worries, but still revealing an anxiety about male
authority. Irane comes into her father’s shop and the following scene ensues:
– Irane! Where in the world are you going?
– I’m off to do some shopping [The graphics show Irane demurely looking at Nas-
ser Ali rather than her father]
– Be back in an hour [the father says this eyebrows raised with a wagging finger]
– Pfff [father shakes his head] I swear! Ever since they banned the veil, we’ve been
heading straight towards decadence (Satrapi 2006: 45).
To drive the point home Satrapi (ibid.: 34) provides us with the footnote: “In
January 1936, Reza Shah forbade the wearing of the veil in Iran”. The scene,
through the words and the expressions on the characters’ faces suggests that the
men don’t quite know how to react to the newly found freedom of the women to
be visible in public space. The father is concerned because his authority as father
seems to be contested, and Nasser Ali is quite unprepared to fight off Irane’s
charms. So the Islamic regime’s ethos that men’s piety somehow needs protec-
tion from the women (Gerami 2003: 260) is mirrored here in a pre-revolution
context, with secular male anxiety about women causing irregular behaviour in
men: both religious and secular men seem to be ‘threatened’ by increased female
visibility.
Chicken with Plums provides a good study of masculinities in Iran before the
revolution and allows us to assess them against the prototypes encouraged by
the Islamic regime. The book opens with Nasser Ali’s decision to starve himself
to death and then we learn the chain of events that has led to his decision. His
artistic sensibilities start to cause him trouble at school, and he grows in the
shadow of his academically much better performing brother. In his spats with
his brother who becomes a communist, we learn that the category of the ‘politi-
cally unengaged’ artist is one that is deemed ‘useless’ by leftist circles as well.
When he is not allowed to marry the woman he loves, he is further emasculated
as he gives into his mother’s plea to marry another woman, the woman who has
had ‘designs’ on him from very early on. Gender expectations, Satrapi thus lets us
know, affect the lives of men just as much as they effect women’s- in this case
driving Nasser Ali into marriage with a woman he does not love, a marriage that
later is unable to provide the emotional support he needs.
Satrapi makes it clear that relations with family and peers inflect greatly the
kind of ‘masculinity’ that a subject will experience. Nasser Ali’s masculinity is
defined very much through his (failing) performance as lover, husband and fa-
 Activist, Professional, Family Man: Masculinities in Marjane Satrapi’s Work 501

ther. He is emasculated because he is not seen fit to provide for a household and
not able to transfer character traits to his offspring: “Nasser Ali Khan didn’t like
Mozaffar for two very precise reasons: first, because his wife had decided alone
to bring this child to the world, and second, because the two of them had noth-
ing in common…” (Satrapi 2006: 51). To drive the point home Satrapi explains
what became of his children, making it clear that they did not inherit any of his
artistic sensibilities. Nasser Ali’s life is in many ways defined through his rela-
tionships with the women around him; Irane, his wife, and his mother who in
death is declared to be a saint. The ‘masculinity’ needed to run the house is taken
on by his wife who provides for the family and who thus has a right to criticize
his choices and indeed break his tar.
In many ways masculinity as a function of man’s position as head of the fam-
ily is reflected in the way masculinities are defined in post-revolutionary Iran.
For the mullahs, the women all become ‘sisters/daughters’ for whose honour the
men are responsible. They become the proper, authoritative fathers that men
before the revolution (like Irane’s father in failing to stop his daughter to have a
love affair in the first place) failed to be.7 I find Gerami’s categorization of the re-
gime’s prototypes useful for a reading of Satrapi’s understanding of masculinity:
The hypermasculine culture of the revolution created many prototypes. Three ver-
sions of masculinities in post-revolutionary Iran stand out, lasting well beyond the
revolutionary stage: the martyr as brave and innocent, the mullah as otherworldly
and pious, and ordinary men as sexual and dominant. Women were not discour-
aged from emulating the manly traits of the first two prototypes (Gerami 2003: 264).
The three categories that Gerami identifies are all interlinked through a hier-
archy, or a division of labour for piety. According to Gerami’s classification men
derive their authority, or claim masculinity, as a function of their relationship
with the figure of the mullah and the martyr. This system is sustained, as has
been discussed above, by very strong symbols, particularly the image of the mar-
tyr whose face is everywhere and who acts as the altarpiece to the mullahs’ con-
struction of masculinity. The martyr is of course a category that gains greater
currency after the Iran-Iraq war, which is depicted with very stark imagery by Sa-

7
In the discourse of the mullah, the Pahlavi system is very much associated with the feminine:
“vain, soft, superfluous, and corrupting. The prerevolutionary period is described by the word fet-
neh, or seductress. The pious and vengeful men of God battle the sultry femme fatale of consum-
erism and Westernism. The Republic has advocated a lifestyle of sparse, pious, and joyless devo-
tion to Islam” (Gerami 2003: 265).
502 Nagihan Haliloğlu

trapi. As Gerami explains: “The martyr is a young, unmarried (virgin, innocent)

man, fearless and strong. He is depicted with eyes cast forward to jihad and the
blessed state of martyrdom.”8 The mullahs as holders of religious knowledge use
the image of the martyrs, the embodiment of the ultimate sacrifice for the ‘true
religion’, to keep the behaviour of ordinary men (and women) in check. The im-
age of the martyr proves personally important for Satrapi: when she returns to
Iran from her studies in Austria, she sits the exam for the Fine Arts College in
Tehran and as she predicts, she has to produce a painting of a martyr. She draws
one that is based on Michelangelo’s La Pieta (Satrapi 2007: 281).
The regime, expectedly, evokes the image of the martyr for the further sup-
pression of women. One day, the whole department is gathered to listen to a
cautionary lecture by the bearded elite. The gist of the narrative is that men, or
rather their piousness, are continuously threatened by the presence of women,
and the foremost mission of the regime is to make sure that all obstacles should
be removed from righteous life that men should lead. For piousness, now, is the
most important characteristic that makes up hegemonic masculinity.
We can’t allow ourselves to behave loosely! It’s the blood of our martyrs which has
nourished the flowers of our republic. To allow oneself to behave indecently is to
trample on the blood of those who gave their lives for our freedom. Also I am asking
the young ladies present here to wear less-wide trousers and longer head-scarves.
You should cover your hair well, you should not wear makeup, you should… (ibid.:
296).
Marjane responds to the official rhetoric of women causing destructive desire
by saying “You don’t hesitate to comment on us, but our brothers present here
have all shapes and sizes of haircuts and clothes. Sometimes, they wear clothes
so tight that we can see everything” (ibid.: 297), raising the spectre of women’s
desire. Much earlier in the memoir, Satrapi contemplates on the martyrs, partic-
ularly in the chapter The Key. The young Marjane is still not fully aware of the
disciplining power of the regime and after a day of beating of breasts for
mourning at school, she starts making fun of the martyrs and her parents are
called to the school to be censured. The narrating Satrapi is reflecting on herself
as a child, recognizing the sad waste of young lives, by having her uncle explain
the situation to her father:

8
Gerami (2003: 267) continues: “Depicted as leading a group of women and older male mar-
tyrs, or […] depicted in the foreground of fully veiled women and young girls, protecting them and
the country’s honor”.
 Activist, Professional, Family Man: Masculinities in Marjane Satrapi’s Work 503

It’s awful… Every day I see buses full of kids arriving […] They come from the poor
areas, you can tell… First they convince them that afterlife is even better than Dis-
neyland, then they put them in a trance with all their songs… It’s nuts! They hypno-
tize them and just toss them into battle. Absolute carnage (ibid.: 101).
Thus Satrapi successfully subverts the idea of martyrdom as masculinity and
presents it as a boon offered falsely to young men from disprivileged back-
grounds. We see, once again, that being male does not necessarily mean one can
partake in the privileges of masculinity: masculinity is a location of power from
where authority and privilege are dished out unequally among the classes. The
current order of the mullahs in Iran that defines hegemonic masculinity is based
on a grid of relations as exposed by Satrapi’s various engagements with represen-
tations of the regime. In the last page of The Key chapter, through her black and
white pictures that offer a great sense of contrast, Satrapi juxtaposes the images
of the boys that are supposed to be the embodiment of Iranian masculinity being
blown to smithereens, with the image of Marjane dancing at a party with her
friends. The primary point made here is clearly the privileged status of Marjane
and a jab at the class system, but more importantly, the senselessness of the
waste of young lives for the survival of the regime, and the viciousness of the dis-
course of masculinity that has helped drive these young men into battle. The
caption for the image of the blown bodies reads: “The key to paradise was for
poor people. Thousands of young kids, promised a better life, exploded on the
minefields with their keys around their necks” (ibid.: 102). Masculinity, surely, is
everything to do with being a man, of having responsibility and agency. Here, by
emphasizing the fact that this image of masculinity is placed on the shoulders of
mere kids, Satrapi reveals the contradiction at the heart of this construction.
Men’s masculinities are monitored, not so much as in pre-revolutionary Iran
through class distinctions and ability to earn money, but as has been suggested
above, through signs of piety. As Germani (2003: 272) explains men:
Being in the public, they faced scrutiny in terms of adherence to the mandatory Is-
lamic codes. From their clothes, their demeanor, their smoking, or whether they par-
ticipated in the office noon prayer, men were scrutinized or questioned […] Among
the professional men, a clean-shaven face is a sign of protest emerging at some of-
fices. Add to this a tucked-in shirt, eau de cologne, and/or a three-piece suit, and the
signs of resistance are all there. Some professionals such as doctors go to the ex-
treme of wearing a tie to register their opposition to the mullahs’ codes.
Naturally, no man in Marjane’s family grows a beard. Through her illustra-
tions she shows the reader how men, too, revolted against the regime through
504 Nagihan Haliloğlu

small acts- like having their shirts tucked in, not growing beards, and rarely, by
wearing a tie. On their way home from a party cut short because of a possible
raid, Marjane’s family is stopped by the police and his father is accused of having
drunk alcohol (which previous pictures suggests he has). When he denies it, the
policeman (who interestingly has no beard or moustache) says: “You think I’m
stupid?!!! I can tell by your tie! Piece of Westernized trash” (Satrapi 2007: 108).
The accusation of being western seems, in the world that Satrapi depicts, one
that can easily lead to arrest. However, on this occasion, Marjane’s mother, mod-
estly dressed in a headscarf intervenes saying she is old enough to be the po-
liceman’s mother, that she is the mother of a twelve year old and asks for for-
giveness, and the policeman stops. “You are lucky to have this woman for your
wife, otherwise you’d be in hell!” (ibid.: 109) he says to Marjane’s father. It is an
instance in which we observe how masculinity expressed through a patronizing
championing of women’s rights, supposedly through Islam, may at times even
paralyze efforts of vigilantism.
When Marjane’s family tries to procure a passport for a sick uncle, they are
faced with the impossible bureaucracy of the regime that tries to discourage
people from traveling. In one of their many visits to various offices, this time for
the authorization to travel for medical purposes, they come across a man they
used to know:
All that creepy window washer had to do to become director of the hospital was to
grow a beard and put on a suit! The fate of my husband depends on a window
washer! Now he’s so religious that he won’t look a woman in the eye. The pathetic
fool! (ibid.: 121).
This observation of Marjane’s aunt chimes with Gerami’s above observation
that masculinity, and the authority stemming from it, is no longer connected to
class and professional associations as in the Shah era but to superficial, outward
signs of piety. This scene also makes clear that the restructuring of post-revolu-
tion masculinity is very much connected to pre-revolution resentment: all that is
to do with the Shah regime becomes suspect. The power relations expressed
through piety have replaced the power relations expressed through class and
professional training. Satrapi thus clearly shows that while the definition of heg-
emonic masculinity changes with the changing of the institutions of the current
order, it is very much informed by what has been going on before.
Satrapi demonstrates how the most important building block of masculinity
has changed, from class and professional membership to piety. She shows that
 Activist, Professional, Family Man: Masculinities in Marjane Satrapi’s Work 505

she does not have time for either element, and that masculinity for her, in the
sense of ‘good, loyal, brave, benevolent’ qualities, lies beyond these divisions. Sa-
trapi is very much aware of masculinity’s attendant sacrifices: they may change
form with changing regimes but sacrificing one’s self for family, for art, and for
society reveals itself to be a fundamental aspect of the construct of masculinity
under whatever auspices it occurs. For her, if masculinity is to be the norm, then
it has to be one that is made up of uprightness in whatever position of power or
dispossession one may find oneself in. That is the basis on which she sings
praises to the leftist men who brave (and die in) prisons, and that is why she
acknowledges the mullah who values her truthful answer (an answer which re-
veals less than the required amount of religious zeal) and offers her a place after
the university interview. Satrapi’s models of masculinity remain a benevolent
God with whom she keeps counsel, her uncle Anoosh who shares with her his
experiences without patronizing her and to a certain extent her uncle Nasser Ali
who rather dies than compromise his art. Her understanding of masculinity is
thus, one can argue, very much informed by the idea of the rafiq, the co-traveler
on the path to freedom and emancipation for all, despite the undertones of the
flight from the feminine.

BIBLIOGRAPHY
Akhavan, N. / G. Bashi; M. Kia; S. Shakhsari (2007), “A Genre in the Service of Empire”, Un-
published manuscript, <http://www.zcommunications.org/a-genre-in-the-service-of-empi
re-by-niki-akhavan-golbarg-bashi-mana-kia-and-sima-shakhsari-na-gb-and-mk-ss>.
Gerami, S. (2003), “Mullahs, Martyrs, and Men”, Men and Masculinities 5.3: 257-274.
Ghoussoub, M. / E. Sinclair-Webb (eds.) (2006), Imagined masculinities: Male Identities and Culture
in the Modern Middle East. London.Kandiyoti, D. (1991), Women, Islam and the State, Lon-
don.
Kimmel, M. S. (2004), “Masculinity as Homophobia: Fear, Shame, and Silence in the Construction
of Gender Identity”, Feminism and Masculinities, Oxford: 182-199.
Najmabadi, A. (1991), “Hazards of Modernity and Mortality: Women, State and Ideology in Con-
temporary Iran”, Women, Islam and State, Philadelphia: 257-274.
Satrapi, M, (2006), Chicken with Plums, New York.
―― (2007), Persepolis, New York.
 On the Interpretation of the Term “Futuwwa” in
Persian fotovvatnamehs*

Khachik Gevorgian
Yerevan State University

Abstract
This paper will attempt to present the reasons the term "futuwwa" interpreted in a different way in
medieval treatises. Alongside linguistically correct etymologies of the term, some treatises also
mention such etymologies which could only have religious and political background. In this paper
politicization of the futuwwa ideology during the Nasir’s reign will be presented and the political/
religious reasons according to which Suhrawardi introduces the etymology of futuwwa in relation
to “fatwa” will be demonstrated.

Keywords
Futuwwa, fotovvatnameh, Suhrawardi, Chaliph Nasir

Unlike its two synonyms ‘ayyārī and axī, for which scholars have proposed dif-
ferent etymologies,1 the term futuwwa (whith ǰavānmardī as an equivalent in Per-
sian) undoubtedly derives from the Arabic fatā, which means “youth” and
“manly”. The main bulk of the primary sources that contain the teachings of the
futuwwa, called kitāb al-futuwwa (fotovvat-name in Persian, and Fütüvvet-name
in Turkish), either do not discuss the origin of the Arabic term, or take its deriva-
tion from the Arabic fata/youthness for granted. It is only the famous theoreti-
cian of futuwwa Abu Hafs Umar Suhrawardi (d. 1235), who in his two fotovvat-
namehs discusses the etymology of the word, presenting a completely different

*
This paper was delivered during the International Conference “15 Years of Achievements:
Celebrating the 15th Anniversary of Iran and the Caucasus, BRILL Academic Publishers”, July 01-02,
2011, Yerevan, Armenia.
1
There are two different options for the etymology of the term ayyār. One of those favors the
Arabic origin of the word where it means “an active man who goes here and there”, and the other
version proposes an Iranian etymology to derive from Middle Persian yār “friend”. The term axī
also has two etymological variants, from Arabic axī “brother”, and from Uyghur axi “generous”.
 Koninklijke Brill NV, Leiden, 2015
508 Khachik Gevorgian

interpretation. This paper will discuss the mentioned futuwwat-namahs and will
try to show the reasons why Suhrawardi introduced such a different interpreta-
tion for the term.
Accordingly Šamsed-Din Muḥammad bin Maḥmud Āmuli in his short futuw-
wat-nama written between 1335-1341, mentions that the term “fata” means
“youth”, “being young” (Shams id-Din Muhammad ben Mahmud Amuli 1352/
1973: 60).
Najm al-Dīn abu Bakr Moḥammad b. Zarkub Tabrizi († 712/1313) in his exten-
sive futuwwat-nama particularly mentions that “fotovvat means javanmardi”
(Najm al-Din abu Bakr Muhammad ben Zarkub Tabrizi: 169).
In a futuwwat-nama by anonymous author from the 18th century, the author
states that “in Parsi fotovvat means javanmardi and the essence of javanmardi is
helping a friend” (Anonymous author 1382/2003: 31).
It is only Suhrawardi (539-632/1145-1235) to offer another interpretation for
futuwwa.
“(1) First one must know that futuwwat has been derived from fatwa, and the
meaning of fatwa is as follows. For any task that one performs or any occupation
or action in which one becomes engaged and in which there is a doubt regarding
its expediency some will say, “This task is expedient, do it”, but others will say, “It
is not expedient, don’t do it”. When a helpless person becomes confused about
(the expediency of that particular) task he should go to a judge who is just and
equitable, or to a wise man who is virtuous, or to a mufti with a formal legal opin-
ion and has attained perfection in knowledge, rules (adab) and practice (‘amal),
or to a master of futuwwat who is adorned with various kinds of knowledge and
wisdom. He should consult with such a person, (saying), “Such and such a prob-
lem has befallen me”, or “I have sworn an oath, and I (want) to carry out such and
such a task and business (but) the community is opposed (to it)”.
(2) (He says), “Now I request a formal, legal opinion. What command does the
Holy Law (shari’at) give? Can one perform (this task) or not?” If that task or busi-
ness is proper (salah) and recommended, then they write a fatwa and give it to
him. They say, one can perform (that task). If it is not a proper act (sawab), then
they do not write out the fatwa, and they say that one must not (carry out that
act). So it is clear that a fatwa for a task is good because nobody can complain
when the mufti writes out a fatwa. And fatwa and futuwwa have the same
meaning. He who is among the people of Futuwwat must also be good and pos-
sess justice, fairness and equity” (Suhrawardi: 104-105).2

2
The following translation is according Ridgeon 2011: 42.
 On the Interpretation of the Term “Futuwwa” 509

The above passage shows that according to Suhrawardi, the term futuwwa is
a derivative from fatwa and that the actual meaning of futuwwa is an act or be-
havior of a man, which completely corresponds to the shari’a, rule of God and is
confirmed by a fatwa given by a mufti. As was told so far, the scholarly literature
until recently has not paid any attention to this interpretation of the term and
only the futuwwa-fata- (rout) ftw meaning “youth” has been discussed. It is only
Lloyd Ridgeon to discuss the case in the foreword to the translation of Suhra-
wardi’s futuwwat-nama. He particularly states that “Suhrawadi’s attempt to link
fatwa with futuwwat are suggestive of his efforts to spread Sufism” (ibid.: 36).
This is true, but some other reasons may also be mentioned.
So why should Suhrawardi use a completely different interpretation for the
term.
The solution lies in the fact that this way of interpretation is in a complete
harmony of Suhrawardi’s devotion to Nasirian futuwwa. Chaliph Nasir solved
several problems related to futuwwa through Suhrawardi and announced him-
self the supreme leader of the futuwwa structure. To announce this and his role
in futuwa Nasir circulated a Manshur where the details of the futuwwa were an-
nounced.
In the manshur Nasir was particularly mentioning that hereafter he will be-
come the qutb of futuwwa and all futuwwa structures might follow him as a
Shaykh.
Nasirian ideas of futuwwa were well presented by Suhrawardi, who actually
was serving as an ambassador of Nasir in the Islamic world to promote futuwwa.
Suhrawardi composed 2 treatises, where he was presenting the futuwwa as a part
of Sufi knowledge and his futuwwat-namahs may be considered Sufi fotovvat-
namehs.
Suhrawardi’s fotovvatnamehs particularly stress the importance of the exist-
ence of a teacher-shaykh for anyone who wants to start futuwwa, moreover he
strictly refuses the existence of two or more shaykhs. Each tarbiye must have
only one shaykh. Thus indirectly Suhrawardi stresses the importance of Nasir in
futuwwa nominating him as the only shaykh of all possible futuwwa organiza-
tions.
Now let us return to the Suhrawardian interpretation of the term futuwwa.
He stresses that the futuwwa is an act which is acceptable by the society and is
agreed upon by a religious leader. When Nasir announced his futuwwa, the Is-
lamic world was full of many futuwwa groups which were confronting each
510 Khachik Gevorgian

other, fights and unrests between futuwwa groups were usual phenomena in
Baghdad in the end of the 12 and the beginning of the 13th centuries. Suhrawardi
shows to his readers, that afterwards all futuwwa groups are to be unified under
the Islamic law and be obedient to Nasir.
From the linguistic point of view Suhrawardi does not make any mistakes. As
mentioned before, this text offers a unique interpretation of the term futuwwa.
Both futuwwa and fatwa derive from the same rout FTW, and Suhrawardi plays
with this uncertainty of Arabic routs and convinces the readers that futuwwa
and fatwa are the same.
In conclusion, Suhrawardi was creating a certain political phenomenon from
futuwwa to serve the Nasirian policy and to this end, he was using different pos-
sibilities to reach his goal, and one of those is the manipulation of the etymology
of the word futuwwa.

BIBLIOGRAPHY
Shams id-Din Muhammad ben Mahmud Amuli (1352/1973), “Futuwwat-nama-yimustakhrijaznafa-
yis al-fonun”, ed. M. Sarraf, Rasail-i javanmardan mushtamil bar haft futuwwat-nama/
Traites des compagnons-chevaliers, with analytic introduction by H. Corbin, Tehran-Paris.
Najm al-Din abu Bakr Muhammad ben Zarkub Tabrizi, “Futuwwat-nama”, Sarraf, M. ed., Rasail-i
javanmardan.
Anonymous author (1382/2003), Futuwwat-nama, ed. Mehran Afshari, Futuwwat-nāma-hāva Ra-
sāil-i Khāksariya (sīrisala), Tehran.
Shihab id-Din Umar Suhrawardi, Futuwwat-nama, ed. M. Sarraf, Rasail-i javanmardan.
Ridgeon, L. (2011), Jawanmardi, A Sufi Code of Honour, Edinburgh University Press.
 The Alevi Discourse in Turkey

Çakır Ceyhan Suvari, Elif Kanca

Yüzüncü Yıl University, Van

Abstract
This paper is an attempt at exploring different interpretations of Alevism and Alevi identities,
having emerged as a result of rapid and large-wave migrations, particulalry from 1960 onwards,
from the countryside to the urban centres of Turkey. Those Alevis, who had become more and
more isolated from the larger Alevi community and each other, ended up divided into different re-
ligious and ideological sects. Emergence of various Alevi associations and foundations proved un-
able to prevent such disintegration. On the contrary, it was the newly established Alevi institu-
tions, emerging upon different bases, which actually heterogenised the Alevi phenomenon. Today,
each Alevi institution in fact promotes its own particular perception of Alevism; the latter may
even vary among family members. Therefore, it will be more accurate to speak of Alevi identities
rather than of a single, unified Alevi identity in today’s Turkey.

Keywords
Turkey, Alevism, Alevi Identity, Urbanism

There are numerous approaches in present-day Turkey to the roots and essence
of Alevism, many radically differing from each other. The Sunnis of Turkey have
two interpretations of Alevism. The first is based on a nationalist approach con-
sidering Alevism as the Turkish interpretation of Islam, and the second one, that
of orthodox Islamists, regarding Alevism a rafizi sect, which deviated from the
Muslim dogma. The Alevis themselves, in their turn, have various interpretations
of their faith: religious leaders and intellectuals have made declarations, which
are essentially in conflict with each other.1 The internal debates resulted in the
situation when, from time to time, Alevi leaders even declare each other düşkün
(“seceded”).2

1
See the previous bibliography on the issue in Varhoff 1998.
2
Being declared as düşkün (“seceded”) is a penalty in Alevism. The one who was declared düş-
kün, is not accepted to the congregation for some time, depending on thenature of the crime he
committed. At the end of the penalty period, the person is supposed to pass through the ceremony
 Koninklijke Brill NV, Leiden, 2015
512 Çakır Ceyhan Suvari / Elif Kanca

This difference of opinions among the Alevis separates them from each other
not only in terms of religious ideas, but also politically. While previously Alevis
were engaged in politics mainly in the Cumhuriyet Halk Partisi (CHP), today they
have settled in the Adalet ve Kalkınma Partisi (AKP), which comes to the fore-
ground with its Islamist stance, and in the Milliyetçi Hareket Partisi (MHP). How-
ever, even among the CHP politicians there is no unified opinion. As a matter of
fact, the CHP leader Kemal Kılıçdaroğlu, an Alevi himself, has demonstrated vi-
olent reactions to the declaration made by the Tunceli delegate Hüseyin Aygün,
who is also a CHP member and his own cousin by that. The CHP’s Alevi member
Sabahat Akkiraz,3 in reply to Aygün’s statement on Alavism being a separate reli-
gion, accused him of being a provoker: To call Alevism a religion is a big mistake
and provocation. Now, there is no difference between you and those who called the
Alevi people “nonbelievers” in the past.4
The CHP leader Kemal Kılıçdaroğlu also opposed the words of his cousin
Aygün with the following statement:
Alevism is a belief, it is a part of Islam. We believe in the same Allah, the same
prophet and the same holy book. Can it be a different religion? No. But I do not want
the subject of religion to be used in politics either. No one has scales in his hands to
measure other people’s beliefs. So, it is not right at all for a person, let alone a politi-
cian, to intervene in the spiritual relations between Allah and his servants. I do not
consider it right to use this pure emotion and belief as a tool for politics. Everyone
must be able to practice his beliefs freely and everyone must respect others’ beliefs.
Mysticism in Islam is the area of people who think of religion and who experience the
spiritual emotions in a more crystallised manner. We read and learn about that
area, we experience it in our spiritual world but we must never use it as a material
for politics. If we do, we will do ill to Mevlana, Hacı Bektaş-ı Veli, Sarı Saltuk, Aşık
Sümmani, Karacaoğlan and Yunus Emre. They never dealt with politics; they
kneaded their beliefs with the philosophy of humanity. They showed how a society
can be led to a bright future and enlightenment.5

of “removing the status of a seceded”. As a matter of fact, Reha Çamuroglu, a former Alevi delegate
of the Adalet ve Kalkınma Partisi, has been declared düşkün by the Alevi Araştırmaları Merkezi
(Center for Alevi Studies). Çamuroglu stated that he considered the people who declared him se-
ceded as inquisitors.
3
Sabahat Akkiraz is a popular singer of Alevi folk songs and currently a delegate having serious
support among the Alevis.
4
See <http://haber.mynet.com/alevilige-din-demek-provokasyondur-640646-politika/>.
5
See <http://www.radikal.com.tr/Radikal.aspx?aType=RadikalDetayV3&ArticleID=1094991>.
 The Alevi Discourse in Turkey 513

Alevis have always been among the main defenders of the republican regime,
its secular character being opposed to the shari‘a dogma as the fundament of the
Ottoman Empire (Aktay 1999; Okan 2004) However, the problem is that there is
a marked difference between the principle of secularism proclaimed in the Con-
stitution and in the reality. Alevis put the problem in the following way: could
the Republic of Turkey found a laic government the way Alevis see it?
Contrary to the expectations of Alevis, the Sunni Muslim identity has been
accepted as basic since the very first days of the Republic.
The official identity of the Republic was founded on Turkishness with the
particular stress on the Hanafi school associated with Turkishness more than
any other Muslim doctrine. The Hanafism became in fact the privileged mazhab
in the state.
The secularisation process of Turkey can be chronologically summarised as
follows: the Caliphate was abolished on March 3rd 1924, and Department of Reli-
gious Affairs was founded on the same day. A decision on the apparel of religious
officials has been made on September 2nd 1925. Dervish lodges and hermitages
were shut down on November 30th 1925. The Gregorian calendar was accepted
on December 26th 1925. The first Turkish khutbah was read on February 5th 1928.
The law stating that “Islam is the religion of the government” was removed from
the Constitution on April 10th 1928.6 Arabic and Persian lectures were removed
from the school curriculums on September 1st 1929. Department of Religious Af-
fairs prohibited the azan to be read in Arabic on July 10th 1931. Laicism as one of
the basic principles, was included into the Constitution on February 5th 1937.
Although all these steps were officially declared as the secularisation process
of Turkey, each of them, in fact, shaped the official religion of the Republic. De-
partment of Religious Affairs, which is an institution affiliated to the Prime Min-
ister, only spreads the teachings of the Sunni-Hanafi doctrine and gives the fat-
was accordingly. The rule stating that “only a follower of the Hanafi mazhab can
be chosen as chairman of Department of Religious Affairs”, brings the Republic
closer to a Hanafi-centric shari‘a state excluding the other three Sunni doctrines,
all the more—any other religious ideology, from political life. Similarly, there is
kind of ideological inconsistency in the situation when universities, emphasising
their secular system of education and claiming to be, together with the military,

6
Yet, Islam has been accepted as the official religion of the government in the first Constitu-
tion.
514 Çakır Ceyhan Suvari / Elif Kanca

the defenders of such government, simultaneously contradict to this idea

through the ideology of the faculties of Theology, the latter rather representing
the mere continuation of the madrasahs closed with the foundation of the Re-
public.
Another issue to be discussed in this regards is the concept of “minorities”.
Taking into consideration that the Turkish government insisted (the Treaty of
Lausanne) on the population classification according to their religious identities,
it will not be an exaggeration to say that the Ottoman Empire of the Tanzimat
Reform Era seems to have been more laic than the Republic.7 On the other hand,
it is the non-Muslims in the laic state who are often reminded, although indi-
rectly, that they are pseudo-citizens8 either by the government itself or repre-
sentatives of the so-called millat-i hakimi (“governing people”).9 Among the ex-
amples of such policy is the application of the Property Tax10 in 1942 and the

7
According to Oran (2005: 48), the Republic of Turkey is the continuation of the Ottoman Em-
pire in many aspects, and the core of the Ottoman social order is “the National System”, which is
based on religion and even its particular doctrine. This system emerged in a year after the con-
quest of Istanbul, that is in 1454, and existed officially until the Imperial Edict of Reorganisation in
1839.
8
Today, people with different political and ideological opinions officially have the same status
as representatives of different beliefs.
9
Millet-i hakime (“governing nation”) implies Sunni Muslim Turks who are considered as real
owners of the country.
10
The Property Tax is the extraordinary wealth tax that was imposed by law act number 4305
on November 11th 1942. The official justification of the Property Tax suggested that government
would impose taxes on “the high level of profitability created by extraordinary war conditions”.
However, the then Prime Minster Şükrü Saracoğlu proposed a different justification at the CHP
group meeting: “This law is also a reform law. It gives us the possibility to become economically in-
dependent. We have an opportunity that will make us get our economical independence. By re-
moving the foreigners controlling our markets we will hand over Turkish market to Turkish peo-
ple… This law will be enforced with all its power on the people who have become rich thanks to
the hospitality shown by this country but avoided to fulfill their duties during that fragile time”.
Three commissions established in Istanbul published the tax lists that had accrued by December
18th 1942: 87% of the accrued taxes were imposed on non-Muslim citizens, whereas 7%, on the
Muslim citizens. The remaining 6% of the tax were on other items and they were mainly imposed
on non-Muslims either. A huge amount of immovable property belonging to non-Muslims
changed hands in December 1942 and in January 1943. 67% of the sold properties were purchased
by Muslim Turks, whereas 30%, by official institutions and organszations (see <http://tr. wikipedia.
org/wiki/Varl%C4%B1k_Vergisi>).
 The Alevi Discourse in Turkey 515

Christian pogroms on September 6th and 7th in 195511—as a result of both many
non-Muslims left the country.
Another example materialising this approach in Turkey, which is a country
where religious belonging of the citizens are considered more important than all
the other affiliation, concerns population exchange. As a result, Orthodox Chris-
tians living in Anatolia were replaced with the Bosnians, Pomaks, Albanians and
Muslim Turks from the Balkans (Oran 2005: 57).
One of the factors, which made Alevis welcome the Republic with enthusi-
asm is the popular belief about the Alevi origin of the founder of the republic,
Kemal Ataturk. The following statement of the Alevi writer Cemal Şener summa-
rises this situation: “Ataturk was not a pious but laic and democratic person. This
is because he had an Alevi-Bektaşi spirit and he is of Turkmen origins. If there is
a laic republic in the Islamic geography, this is because Mustafa Kemal has such
an origin. He was not pious and he had a belief of God as laic people had.12 His
lifestyle was closer to Alevi-Bektaşis rather than to Sunnis.”13
Alevis even deify Ararurk as incarnation of Hacı Bektaş Veli (Işıklı 2006: 194).
This belief is so dominant that many Alevis hang his picture side by side with
those of Ali and Hacı Bektaş in their houses, mausoleums and in the cemevis.14
However, in fact, Ataturk and his regime delivered a major blow to Alevism,

11
During and after these pogroms, which started in the evening of September 6th and contin-
ued for 9 hours, from 13 to 16 Catholics and at least one Armenian lost their lives, and 32 Catholics
were heavily injured. Besides, 4.348 offices, 110 hotels, 27 pharmacies, 23 schools, 21 factories, 73
churches and graveyards, as well as more than 1000 houses belonging to Christians were damaged
(see <http://tr.wikipedia.org/wiki/67_Eyl%C3%BCl_Olaylar%C4%B1>).
12
Ataturk could hardly be of the Alevi origin. The fact that he was first sent to a local school
(mahalle mektebi), in which the Qur’an and Sunni Islam were taught, proves that at least his
mother was a Sunni Muslim. On the other hand, his statements he made both during and after the
foundation of the Republic, speaks of his Islamic identity or at least of his desire to demonstrate it.
As a matter of fact, he considered the religion as one of the virtues that shapes a nation. "Our na-
tion has two assets, the religion and the language. No power can erase these virtues from the heart
and spirits of our nation" (apud Özakıncı 2001: 15). However, Ataturk’s approach to the religion
was, of course, different from that of an ordinary Muslim. Being a representative of the generation
of enlightenment and, above all, the architect of a newly founded country, he tried to form an au-
ditable "official religion/Islam" as the previous governments did. His preference can be clearly un-
derstood from the speech he gave in Balıkesir on 07.02.1923 (See Işıklı 2006:188).
13
C. Şener, “Atatürk ne Dindardı ne de Ateist”, Radikal newspaper, October 8, 2006.
14
Although mausoleums and the cemevis (or cemxanahs) were officially prohibited, they func-
tioned under different institutional names in traditional Alevi areas, where the Alevis built a ma-
jority.
516 Çakır Ceyhan Suvari / Elif Kanca

which could have resulted in its complete elimination. The shrine of Hacı
Bektaşi Veli, the holiest point for the Alevis in Anatolia, was turned into a mu-
seum, and could be no longer visited as a place of worship and pilgrimage. On
the other hand, the Alevi Dedes (grandsires),15 the real bearers and masters of
Alevism, whose status had been recognised even during the Ottoman period and
whose position had been strengthened by the Sultan’s imperial orders, were offi-
cially deprived of their functions with the foundation of the Republic. Moreover,
Alevi intellectuals, who greeted the Republic with fervour, considered both some
Alevi essential religious elements and the dedelik institution as outdated and
thus set the Alevi identity at stake and seriously damaged it.
Since historically Alevism was mainly shaped as an oral tradition, an im-
portant part of its cultural memory is fixed in its ceremonies and rituals. Various
events, such as commemorations, sacrifice ceremonies and the cem ceremonies,
which have the power of adding value and meaning to the people who organised
them. All ceremonies are repeated, what makes us think the past continues nat-
urally without an interruption. Rituals, transmitting and transferring the
knowledge, protect the identity with regular repetition, reproduce the cultural
identity and thus guarantee the temporal and spatial unity of the group.
However, since Alevi mausoleums and shrines were shut down with the
foundation of the Republic, the organisation of ceremonies became problematic
as well, what created another gap in the Alevi identity and resulted in its cultural
dissolution. The process accelerated in the 1960s with the Alevi migration to ur-
ban centres. Alevis, who became more isolated from the community and each
other, have divided into different religious and ideological trends. Emergence of
various Alevi associations and foundations did not prevent such disintegra-
tion—on the contrary, the newly established Alevi institutions emerged upon
different bases what heterogenised the Alevi phenomenon. Each institution has
in fact its own particular perception of Alevism. For example, the definition of
Alevism on the web site of Cem Vakfı (“Cem Foundation”) brings it closer to
Sunni Islam.16
The Hacı Bektaş Veli Kültür Derneği, contrary to the Cem Vakfı, approaches
Alevism as a mixture of various religions, in particular Sufi ideas, ancient Anato-

15
The Institution of grandsires (dedelik) is one of the most important religious structures in
Alevism. The dedes who are believed to be of Ali’s kin, stay at the top of the Alevi religious hierar-
chy (See recently Thörne 2011).
16
See <http://www.aleviislamdinhizmetleri.com/?p=151>.
 The Alevi Discourse in Turkey 517

lian beliefs, Hittite and Mesopotamian beliefs, Shamanism, Mazdakism, Zoroas-

trianism, Buddhism, as well as Jewish, Christian and Islamic ideas. They charac-
terize Alevism as Hak Yolu (the Path of God).17
The Pir Sultan Abdal Derneği, another Alevi institution, states that Alevism is
an independent faith outside Islam, close rather to socialism than to a religion.18
Some Alevi leaders, while defining the Alevi identity, apply to history and so-
cial memory. Among the historical references are Caliph Ali, Haci Bektaş, Pir Sul-
tan Abdal, the Kerbela incident and the massacre of Sultan Yavuz, in addition to
the violent attacks against Alevis from the recent history, such as the Çorum,
Maraş, Sivas and Gazi incidents. However, the attempts to shape an Alevi iden-
tity referring to history, depend on different interpretations of the mentioned
events. For example, the execution of Pir Sultan is among the strongest markers,
which has turned into a myth as part of identity. However, in this particular case,
one Alevi institution can approach Pir Sultan as a bearer of an ideology similar to
socialism, while another, as a real representative of the Alevi faith, the former
giving Pir Sultan the status of an intermediary leftist leader, while the latter, that
of an Alevi religious hero who rebelled against Sunni Islam.
Separation among Alevis has grown so much that Alevi ideologists have
started to debate whether Alevism should be discussed “beyond Islam or within
it", each bringing their own arguments regarding the roots of Alevism and inter-
pretations of the Alevi narratives. This means that social memory as basis for
identity is as variable as the identity itself; it is constantly reinterpreted, rede-
fined and even reinvented in different situations according to new realities and
current needs.
The niche of Alevism in or outside Islam is the ranging mark of the whole
amplitude of opinions on the roots of Alevism among the Alevis themselves (see
Özel 2006). Çınar, for example, claims (ibid: 9-10) that Alevism has a two thou-
sand year history, and, thus, the term “Alevi” has nothing to do with the name of
Ali. He derives it from the Turk. alev (“flame”) with the suffix -i, interpreting the
term as “belonging to flame, light”, or “one who comes from the light" and refer-
ring to the name ışık and ışık taifesi (“tribe of the light”), the community itself
had allegedly used before the 16th century.

17
See <http://www.hacibektasdernegi.com/alevilik-bektasilik-nedir.html>.
18
See <http://www.pirsultan.net/kategori.asp?KID=4&ID=25>.
518 Çakır Ceyhan Suvari / Elif Kanca

Şener (ibid: 31-34) considers Alevism as part of Islam, associating it with Ca-
liph Ali, and relating some Alevi elements, especially in family structure, attitude
to women and the perception of hell and paradise to early Islam.
Erden (ibid: 38) dates the emergence of Alevism back to 2500 years ago,
making special stress on the authentic culture of the Turkmen Oghuz tribes, lo-
cal Anatolian substratum and other civilisational elements (e.g. Sumerian).
Bermek (ibid: 44-47) puts Alevism within Islam with a proviso that the Sunni
faith is not the only possible way of Islam. Moreover, he considers Islam as supe-
rior to other religions, “since Mohammad came together with such an analyst as
Caliph Ali, and Ali was able to explain the core of this faith to masses”.
Özel (ibid: 52-55) suggesting that Alevism should be regarded both outside
and inside Islam, specifies the phenomenon as the pure Turkish interpretation
of Islam. In his opinion, it is not Islam that contains Alevism, but it is Alevism
that contains Islam, alongside with its other constituents—elements of Mani-
chaeism, Zoroastrianism, Buddhism and Shamanism.
Atıcı (ibid: 61-66) believes, Alevism is the core of Islam, at the same time,
however, placing Alevism closer to Sunnism.
Genç (ibid: 71) approaches Alevism as a system of Anatolian beliefs influ-
enced by Islam.
Demir (ibid: 79-81) defines Alevism as heterodox Islam, also emphasising its
pre-Islamic elements.
Kaygusuz (ibid: 92-101) states that Alevism is a syncretic religious system
based on the esoteric interpretation of Islam.
Finally, Yıldırım (ibid: 113) suggests that Alevism is a “unique” religion, quite
different from Islam and typical exclusively to Anatolia.
This wide variety of opinions among the Alevi intellectuals virtually mani-
fests two extreme trends within the internal Alevi discourse in Turkey, one
melting Alevism inside Islam, and the other alienating it from the religious con-
text and turning it into an ideological system.
CONCLUSION
Despite the myth of Ataturk’s Alevi origin spread among Alevis, and the fact that
they even approached him as an incarnation of Hacı Bektaş, actually, Alevis were
among those who welcomed the Republic with the particular stress on its
secular character. A laic state, be it even atheistic, seemed to be less dangerous
to the Alevis, than an empire governed according to the shari‘a rules (Aktay 1999;
Okan 2004). Many Alevi intellectuals expected the laic regime of the Republic to
 The Alevi Discourse in Turkey 519

be a real chance for the promotion of their religion. Moreover, they almost con-
sidered themselves both the founders and owners of the Republic. However, in
reality, the laicism project has done the ill service to Alevism: Alevi milieu was
disintegrated, the communities deprived of their spiritual guidance, polarised in
religious and political terms. The Alevi identity varies sometimes even among
kins, so that the identity issue can be presently approached only as the diversity
of Alevi identities in Turkey.19

BIBLIOGRAPHY
Aktan, U. (2006), “Türk Devrimi, Laiklik ve İslam”, Radikal Gazetesi, 13 Ekim.
Aktay, Y. (1999), Türk Dininin Sosyolojik İmkânı, İstanbul.
Işıklı, A. (2006), Sosyalizm, Kemalizm ve Din, Ankara.
Okan, M. (2004), Türkiye’de Alevilik, Ankara.
Olsson, T. / E. Özdalga / C. Randvere (1998), Alevi Identity: Cultural, Religious and Social Perspec-
tives, Istanbul.
Oran, B. (2005), Türkiye’de Azınlıklar, İstanbul.
Özakıncı, C. (2001), Dünden Bugüne Türklerde Dil ve Din, İstanbul.
Özel, S. (2006), Aleviler Aleviliği Tartışıyor, İstanbul.
Thörne, A. (2011), “Dedes in Dersim, Narratives of Violence and Persecution”, Iran and the Cauca-
sus, 16.1: 71-95.
Vorhoff, K. (1998), “Academic and Journalistic Publications on the Alevi and Bektashi of Turkey”,
Olsson et al.: 23-50.

19
We leave aside the dogmatic aspect of the problem: the contamination of Alevism, a Shi‘a
trend, with the Bektashi order is another symptomatic sign of the erosion of Alevi milieu in Turkey
(see Olsson, et al.)
 Turkey’s Border with Armenia:
Obstacle and Chance for Turkish Politics

Pascal Kluge
Freie Universität, Berlin

Abstract
The strained relations between Turkey and Armenia are dominated by two issues: firstly, the
events of 1915 which are declared as genocide by Armenia while Turkey rejects that term; and sec-
ondly, the questions around the Armenia-supported declaration of independence of Nagorno-
Karabakh from Azerbaijan, a country with which Turkey feels the bonds of brotherhood, thus de-
manding the restitution of Nagorno-Karabakh to Azerbaijan. Other topics such as the course of the
border, last determined in the Treaty of Moscow and Kars in 1921, irritate their relations further as
Turkey sometimes suspects that Armenia might raise claims to those parts of Turkish territory that
were historically inhabited largely by Armenians.
With the signing of two protocols, or roadmap, in October 2009, however, hopes were high that
Armenia and Turkey would finally embark on the road to rapprochement and mutual under-
standing. Those hopes were soon deceived, though, as pressure on both governments finally led to
a de facto abandonment of the protocols. A resumption of the process of rapprochement receded
into the distance again. Still, many voices demanded a political solution to the pressing issues—
although perceiving politics in that regard generally as not being overly successful, but rather
caught between various pressures from the outside.
To understand the relations between the two nations and to identify the chances of a rap-
prochement, an analysis of the perception of the common border can be helpful, as in the concept
of borders two important dimensions, separation and contact, are inherent. The interplay between
these two dimensions and the possible emphasis of one of them inform us not only about the dif-
ferent approaches to the Turkish-Armenian border but can suggest possible paths for future rela-
tions. Even if today the border appears to be a clear separating line between the Turkish and the
Armenian populations of the region, the complex relations between the two nations cannot be
understood by consulting simple dichotomies as there exists a deeper layer of shared history and
collective memory, including long and fruitful periods of coexistence of the two peoples—not only
in today’s border zone but throughout large parts of the Ottoman Empire.
When analysing today’s perspective of the Turkish political establishment and of the ideologi-
cally rigid parts of society in general, the Turkish-Armenian border is often perceived as monolithic
or hard, as a border of separation, particularly if one of the issues of conflict is touched upon. On
the other hand, when looking at the non-official, less rigid views, conveyed by the sometimes in-
fluential columnists of newspapers, the border is at times rather perceived as porous, soft and a

 Koninklijke Brill NV, Leiden, 2015

522 Pascal Kluge

border of contact. If both, politics and the influential parts of society, would encourage a border of
contact, the border population could extend its identity beyond the borderline, and possibly de-
velop an enriched identity distinct from the common national Turkish identity. This would con-
tain the chance to (re-)create a common cultural basis with the Armenian side of the border,
which might eventually even have an effect on the country as a whole. One sign of this are the
projects and activities between the two countries—although few in number—which are generally
perceived in a positive manner.
It remains to clarify to what extent those projects and the constructive and soft approach to-
wards the border can influence the political establishment of Turkey in terms of a resumption of
the efforts concerning the Turkish-Armenian rapprochement.

Keywords
Border, frontier, genocide, Nagorno-Karabakh, nation-state, Turkish-Armenian relations

INTRODUCTION
The image of Turkey to the outside world has changed substantially over the last
decades: from bureaucratic and state-controlled to economically strong and
modern, with a new and upcoming elite. Economy, politics and culture are
tightly knit together in vital centres, whereas the main motor, capitalism, is
pushing the country still further on, with an ambitious government directing
it—a government which is, in recent times, often challenged by scandals of na-
tional and international scope, which, for many observers, put into question its
integraty and its claim to be fully democratic. While the world is looking at those
vibrant centres, Turkey’s border regions—only sometimes drawing international
attention—are out of the limelight. When researching borders, however, Tur-
key’s position is unique in many aspects. As Talat Sait Halman (1987: 3–4) points
out:
It stands at Asia’s westernmost edge; and, although its cultural roots are firmly im-
bedded in Central Asia, it dissociates itself from Asian realities. It is the northern-
most part of the Middle East, but has no kinship with the Arab nations or with Iran
[…]. Turkey is regarded as Europe’s south-eastern frontier although only a tiny por-
tion of it is in Europe.

Furthermore, in religious terms, Turkey, as a predominantly Muslim country,

shares borders with the predominantly Christian Bulgaria, Greece, Georgia, and
Armenia. Of those countries, arguably, the most challenging relations Turkey has
with Armenia with which it shares a 325 kilometre long border.
Looking at the Turkish-Armenian border region, it could be described as an
important cultural landscape, encompassing a territory of various geographical
 Turkey’s Border with Armenia 523

and historical characteristics, with a multireligious and multiethnic heritage.

However, in the course of time, it has turned into a region where common per-
sonal, social and cultural bonds are increasingly unmixed. When characterising
the region, today, there is a sense that Armenians on the one side and Turks (and
Kurds) on the other side have lost a cultural closeness to the ethnic groups with
whom they once interacted. Although juxtaposed against each other, the feeling
of belonging to that region as a whole has substantially changed or has perhaps
even been eliminated altogether (Kennard 2010: 14–15).
This paper attempts a discussion of the relationship between the main issues
dominating Turkey’s and Armenia’s relations—the disputed genocide of 1915
and the issues around Nagorno-Karabakh—and the perception of the common
border, taking into account the interplay between the heritage of the multieth-
nic and multireligious Ottoman Empire and the concept of the Turkish nation-
state. By discussing those contexts, we will not only shed light on Turkey’s differ-
ent approaches towards its borders with Armenia, but also learn about the po-
litical implications for the Turkish-Armenian relations—where the pitfalls are
and where the chances lie. To expect simple dichotomies, however, would not
do justice to the subject. The two nations are much interwoven in a complex sys-
tem of history and culture in which reducing the questions to simple dichoto-
mies would lead to an oversimplification, obscuring more than it would clarify.
To answer those questions, media research seems an appropriate tool. The
media reflect the various discourses of a society and act as a forum for discus-
sion. Often dominated by a few individuals and holdings, the media can also
shape the public view on certain topics.1 With regards to the topic of this paper,
the media play an important part in nationalising a country, the national dis-
course being even dependant in large part on the media (Carnevale 2005: 29). It
is also via the media that those parts of the population that do not live in a bor-
der area get a sense of the country’s borders and border territories, thereby re-
inforcing their appreciation of the nation as a single entity. Furthermore, the
media feed the collective memory and keep the heritage alive, “for they can be
regarded as a social framework for coherently organising past events and the
pasts of persons and places that have not been directly experienced by individu-
als or groups” (Holtom 2010: 275). However, this paper does not pursue an analy-

1
In a process of reciprocal interaction, also the general public and the readers influence the
discourses.
524 Pascal Kluge

sis of the degree of truth in the media that were researched. Relevant is rather the
function of the discourse and its implications for Turkish-Armenian relations.
We will look at three years of online-versions of publications (September
2009 to August 2012) of the Turkish daily newspapers Hürriyet and Sabah—both
distributed and read nationwide, with high circulation.2 During this three year
time range, Turkey was governed by the moderate Islamist AK Parti, its politics
focussing domestically on a conservative modernisation and economic progress
of the country and internationally on good relations, especially with Turkey’s di-
rect neighbours, as well as attaining a politically leading role in the region and
beyond (Buskjær Christensen 2009). Hürriyet is part of the Doğan Holding and
secular in its approach while Sabah is part of the Çalık Holding (Milliyet 2008)
which is said to be close to the ruling AK Parti, and has a more religious-con-
servative approach. In the two newspapers, besides general news, the views of
different players of society, mainly of the political and economic elite, and the
opinions of the, sometimes influential, columnists and analysts are conveyed.
Although there is a difference in the political-ideological orientation of the two
newspapers, one must not necessarily expect much of a difference or even oppo-
site views on all societal questions. According to Seufert, the public views of
Turkey are all within one worldview when not taking the political and moral cri-
teria of Europe as a basis of analysis (Seufert 1997: 48). Moreover, the notions na-
tion-state and Turkish identity, which this paper touches upon, are located in
many of their aspects beyond the ideological faultlines of Turkish society.

BORDER AND FRONTIER

In English, both terms border and frontier are used when talking about political
boundaries between two countries. However, both terms have different facets in
meaning, although sometimes blurred. In the course of time, as with “all histori-
cal phenomena, contemporary and subsequent views of the frontier [and bor-
der] often diverge widely” (Power 1999: 1). Since these terms are used throughout
the literature, we will clarify their differences for their use in the further analysis.
As for the term frontier, we can differentiate between an American concept and
a British concept (with its European cognates). The American term frontier sig-
nifies less a barrier but rather a zone, a zone of passage and a land of oppor-

2
In the period from 03 September 2012 to 09 September 2012, Hürriyet sold 409.909, Sabah
308.985 copies. See: “Gazete Net Satışları”, Medyatava, http://www.medyatava.com/tiraj.asp, ac-
cessed 17 September 2012.
 Turkey’s Border with Armenia 525

tunity, often involving conflict with the natural environment. The British use of
the term frontier and its European cognates signify “a political barrier between
states or peoples which is often militarised” (ibid.: 1–3).
The term border, in both British and American English, has a less figurative
connotation than the term frontier. In the broadest sense, it is a linear marker of
an entity, describing a definite space, and thus creating territory. Therefore, bor-
ders provide us with a certain structure concerning the world around us, which
is, when looking at it from a political angle, usually governed by a central au-
thority. As the central authority cannot go on indefinitely, borders are always
implied in the concept of state, constituting the administrative lines separating
two states from each other (Standen 1999b: 58). The very principle of the border
as the ultimate marker of national sovereignty was never put into question
(Medvedev 1999: 47).3 It is the actual border of a state which is seen by the aver-
age citizen as the significant dividing line (Kennard 2010: 7). Socially con-
structed, kept, reproduced, and memorised, the concept of border is the motor
for thinking and acting in the context of nation, constantly negotiated and re-
constructed (Buchowski/Chołuj 2001: 7).
In this paper, we will use the term border when discussing the actual political
and administrative line, separating two states from each other. The term frontier,
however, will be used when talking about a (border-)zone of passage and inter-
action, offering us opportunities for reinterpretation of the political border
(Power 1999: 12). The term frontier will not be used in its British notion as it
comes close to the term border, being basically just more figurative in the sense
of a barrier.
In Turkish, the terms for border are sınır and, synonymously used, hudut. The
two terms generally describe a line that separates two neighbouring states from
each other, but could also separate neighbouring provinces, districts, quarters,
villages, or lands—thus coming close to the meaning of term border. An older
term in Turkish is uç, meaning the last tip to what a thing can be widened to or
spread out to, thus coming close to the notion of the American meaning of the
term frontier (Püsküllüoğlu 1995: 1526).
Not only do the various terms differ in their respective meanings, also each
and every border is distinct. Although not acknowledging the uniqueness and

3
At least since the Peace of Westphalia in 1648, introducing a new system of political order in
central Europe, based upon the concept of a sovereign state.
526 Pascal Kluge

complexity of each border, Martinez (1994) developed a classification system

that categorises borders, taking as parameters the cross-border interaction be-
tween the nations and their quality. Martinez’ approach is insofar helpful as it
does not speak of borders as simple lines alone, but offers a more systemic ap-
proach, taking into account also the hinterland of borders, consequently speak-
ing of borderlands. Martinez differentiates in his classification scheme between
four categories: firstly, alienated borderlands with little or no cross-border inter-
action, resulting in the consequence that both sides grow apart from each other
which can lead to tension; secondly, coexistent borderlands with limited bi-na-
tional contacts, leading to some stability; thirdly, interdependent borderlands
with cooperative relations and a relative stability; fourthly, integrated border-
lands in which the inhabitants see themselves as part of a system with a solid
stability between the two nations. As useful as the typology of Martinez is “in
explaining cross-border functionality, it does not take into account at all the cul-
tural or ethnic specificity of those living in border regions, nor does it have any-
thing to say about a cultural overlap which exists to a greater or lesser degree”
throughout a border region (Kennard 2010: 14).
The border between Turkey and Armenia seems to be, as we will see further
on, in between the two categories Martinez classifies as alienated and coexistent
borderlands. In this context, there had been a serious attempt, through political
and diplomatic means, to bring the two countries closer together, leading in
2009 to the signing of two protocols (basically a roadmap) between Turkey and
Armenia, a layout of further steps with the objective of a rapprochement be-
tween the two countries.

THE FUNCTIONS OF BORDERS

Any rapprochement between two countries is insofar per se difficult as the pri-
mordial function of a border is to divide and separate. The national discourse
supports this function by developing a sense for its borders, strengthening the no-
tion of nationality, and sharpening the identity of a nation. In this regard, we can
distinguish four main functions of borders, namely order, protection, control, and
identity formation: firstly, borders are a metaphysical necessity and create a pat-
tern of order (Demandt 1993: 21). Human beings, for example, follow an inner
need to define their own living spaces when constructing an own territory, sur-
rounded by borders (Dittgen 2000: 56); secondly, borders have a military protec-
tive function, although in many cases today, only a symbolic protection remains
 Turkey’s Border with Armenia 527

(Blake 2000: 20); thirdly, from the perspective of a nation-state, the border is the
marker of space for territorial state sovereignty, and at the same time the limit of
a state’s influence (Haslinger 2000: 60); fourthly, there is a socio-psychological
function of borders, produced by defining and setting the limits of a territory, se-
curing it to the inside while delineating it from the outside, or, in other words,
including the familiar while excluding the unfamiliar (Stüver 2001: 157; Weich-
hart 1999: 22–23). Through the moment of exclusion and the creation of an other,
identity formation takes place. The socio-psychological function, thereby, deeply
touches the identity questions of society and is particularly difficult to influence
(Dürrschmidt 2002: 123–124). Even if the hardware of borders, such as border
controls, fences and cameras, are diminished or get abolished altogether, the so-
cio-psychological dimensions usually persist and remain as cultural and sym-
bolic borders, in some cases gaining even more importance (Krämer 1997: 14).
The functions of borders we looked at so far, stressed their separating dimen-
sions. However, borders and frontiers also have an integrative side (Mertels-
mann 2010: 215). Lord Curzon, a British statesman, distinguished in the year 1907
between borders of separation and borders of contact (Power 1999: 2). More re-
cently, borders were classified into hard and soft borders. The terms hard and
soft “have become a kind of shorthand to describe various aspects of the activi-
ties which are, or are not allowed to take place around state borders” (Kennard
2010: 12). De Bardeleben (2005: 11) refers to several authors when identifying hard
borders as closed, barriers, or exclusive, accompanied by strict visa regimes, ex-
tensive policing, and controls on the cross-border transport of goods. On the
other hand, soft borders are described as open, porous, communicative, inclusive
and bridges, with measures to loosen them, including relaxed visa requirements,
free flow of traffic and goods, and an uncomplicated exchange of human con-
tacts.
In case of a border of contact or a soft border, the identity of the population of
the border region is not necessarily a strict function of the political and adminis-
trative borderline of a nation-state, but can extend beyond the borderline, turn-
ing into a cross-border identity, its quality being closer to the American notion of
the term frontier as described before. The border, then, functions as an entity of
contact, a place, or rather a space, where one can connect to the other. In this in-
stance, regional transformation takes place, specific territorial units (re-)emerge
and become “established parts of the regional system in question” (Paasi 1995:
46), allowing for a cross-border identity to be developed or (re-)created, possibly
528 Pascal Kluge

even by (re-)discovering commonalities in heritage and collective identity. Com-

mon heritage and collective memory, important elements of a cross-border iden-
tity, are thereby “less ironclad attributes than tendencies and aspirations” (Hol-
tom 2010: 274).
Such processes, however, do not unfold smoothly, as opposing tendencies of-
ten operate in tension with each other (Holtom 2010: 274–275). A cross-border
identity has to compete with the national identity, especially with the socio-psy-
chological function it is challenging. Borders, thus, encourage on the one hand
the creation of a national identity and sharpen the sense of a commonly shared
territory, but can, on the other hand, also be a part of a distinct territorial system
which, by transcending the national border, encompasses a cross-border identity
(Hoorn 2005: 24). The approach of cross-border identities is not new, though. In a
similar way, but not yet influenced by the concept of nation-states, scholars until
World War II used concepts of culture and society that were not confined to the
borders of nation-states (Glick Schiller 2004: 450).
With regard to Armenia and Turkey, the common border might serve to
(re-)discover past or bring forth new, former unmaterialised and even unimag-
ined aspects of commonalities, and create a platform for a common identity, as
the region is full of distinct historical and cultural features shared by the people
inhabiting it. When delving into the history of the Ottoman Empire, we also see
that it was a common place for inhabitants of that empire to have cross-border
identities. The present Turkish-Armenian border region might possibly profit
from that heritage. However, the developments in the Turkish-Armenian rela-
tions in the 20th century could complicate the (re-)creation and development of
a cross-border identity (Riedel 1994: 39).

FROM FRONTIERS (OTTOMAN EMPIRE) TO BORDERS (TURKEY)

The search for new territory was a core principle of the Ottoman Empire, as new
conquered territories financed its centre, especially the court in Istanbul,
through the subsequent influx of tributes and taxes (Coles 1968: 77). The borders
of the Ottoman Empire were a “region of colonisation and settlement involving
both military means and proselytisation”, but still an area of passage, interaction,
and thus closer to the notion of the American term frontier (Heywood 1999: 233).
Border zones changed frequently and in a dynamic way: what was one day a
border region might loose this status the next day through new conquests (or
losses) of the Ottoman army. The history of new settlements, emigration and
 Turkey’s Border with Armenia 529

wars was so complicated throughout the Ottoman Empire, there were many
cases of nations dispersed across different borders at different times, finding
themselves with other nations in locations which they considered their own.
Rulers in the periphery were in many cases very powerful and often chose to
keep the ties to the central authorities at a low level. To be successful as a border
region they needed to adapt identity and values according to the region as a
whole, distinct from the identity of the cultural and political centre of the empire
while the border population often pertained not only economic but also its his-
torical-cultural ties to territories beyond the empire’s borders (Goffman 2002: 32,
45). To maintain economic relations to regions beyond the border, even in times
of war, was common practice in the Ottoman Empire (Goffman 2002: 163). Thus,
borders were often porous, the border population acquiring an iridescent iden-
tity.
With the rise of nationalism in the 19th century, however, the notions of na-
tion and, along with it, national identity gained force as a concept to organise and
rule countries. The rise of the modern nation-state thereby “reinforced the sepa-
rating function of state borders by nationalising the people on both sides of it”
(Moll 2010: 114), transforming the population of a territory into a nation (Brock
2000: 8), facilitated by a sense of togetherness and an imagined community. Even
where there had been a shared history and a shared identity, the borders of na-
tion-states—drawn as a result of military conflict or political change—created
inevitably an other as a consequence of the nationalising forces within the na-
tion-state (Buchowski/Chołuj 2001: 8). In some cases, constituting a nation de-
pended not only on separating from other nations by creating an other, in a se-
cond step the other was further defined as the enemy, as the concept of enemy can
reinforce a strong national identity (Jeismann 1992). This construction is inti-
mately connected with power and dominance as the other (or the enemy) does
not only receive an ascription, but is an element within the power structure, in-
volving a strong emotional commitment (Padis 2009: 2). Lundén (2010: 10) ex-
plains the outcome of this process as follows:
Criterion of a successful state is that its residents are ‘nationalised’, meaning that
they accept, through communication with ‘their’ state, that they belong to the [na-
tion-]state. [This process happens …] partly voluntarily, partly through coercion,
through the school curriculum, legislation, military service and the mass media […]
and is implemented through the building of an infrastructure (transport routes,
communication networks).
530 Pascal Kluge

The collective national identity is neither directly emerging nor is it an ab-

solute stable and everlasting phenomenon (Stüver 2001: 156). It is rather a social
process that is generated, modified or weakened through different social factors
(Hoorn 2005: 23). Today’s pattern of perceiving and thinking is strongly influ-
enced by nationalism, and is a result of the teleological attempt to explain all of
history as leading to the formation of nation-states (Standen 1999a: 26). Accord-
ing to Medijainen (2010: 190),
[t]he idea of division into nations is [for some] the most natural development, and
their state […] the key component of a patriotic way of thinking. Added to this is the
conviction that each nationality has its unique nature and singular features. […]
Another component of such a way of thinking is the view that a nation can survive
only in its own country and only in this manner is its security sufficiently protected.
The late 19th and early 20th century added to this the idea that any person’s highest
loyalty shall be devoted to the nation state […].
In the late Ottoman Empire, the nationalising of the country gained strong
impetus with the War of Independence (1919-1923), in which, under the leader-
ship of Mustafa Kemal (Atatürk), national forces fought militarily against the oc-
cupying powers of the country and politically against the weakened Ottoman
government in Istanbul. In this period of nationalising, Turkey asserted itself as a
self-conscious entity in response to the perceived threat of a total attack or a bit-
by-bit absorption by foreign nations. After the foundation of the Republic of
Turkey in 1923, Atatürk tried to catapult the country into the 20th century by ori-
enting it socially on Europe and giving it a secular face. The materialising of the
Turkish nation-state in 1923 meant, thus, at the same time a turning away from a
multireligious and a multiethnic empire to the image of a more homogeneous
nation with its dictum Ne mutlu Türküm diyene (‘How happy is the one who can
say ‘I am a Turk’’). At the same time, the different cultural spaces in Turkey were
deeply affected, as the strong nationalist narratives, constructed around the
Turkish nation-state, have overlapped or even replaced other, more complex,
narratives—such as the common history of Armenians and Turks (Tocci 2007:
27). The War of Independence and the subsequent formation of the Republic of
Turkey created also opportunities for the powerful imagery of rebirth (Kollis
2010: 32), the nationalist discourse in Turkey bases, for a substantial part, its rai-
son d’être on. The concept of rebirth was associated with both the territorial res-
cue of what was believed to be the core borderlines of the country, and the po-
litical distinction from the historical precedent of the Republic of Turkey, the Ot-
toman Empire. This period was, and still is to a fairly high extent, quite influen-
 Turkey’s Border with Armenia 531

tial on the different discourses on the concepts of border, nation-state and na-
tional identity in Turkey.
In recent times, due to the consequences of globalisation, more mobility and
migration, as well as the use of modern information technologies, both the exist-
ence of the nation-state and the functions of borders are put into question in the
ideological, cultural, political and economic discourses (Dittgen 2000; Kasaba/
Bozdogan 2000: 19). The concept of the nation-state, though still strong in many
discourses, is losing its uniting force, as it seems to be partially replaced, or, at
least supplemented with other concepts of identity. Turkey is acquiring a new
face as the ideology of the nation-state decreasingly reflects the reality of society,
with religious, ethnical and other groups coming up with own identities (Kasa-
ba/Bozdogan 2000: 2).4 Thus, when examining Turkey’s borders, the idea of a cul-
tural confinement to one’s nation-state is perhaps not overly appropriate since
there are various historical and cultural layers of identity to be taken into ac-
count.

THE TURKISH-ARMENIAN BORDER REGION

When looking at the Turkish-Armenian borderline today, it matches the ethnic,
linguistic and religious boundaries of the region, and appears to be a logical, al-
most natural, separation between the two—three, if we add the Kurdish popula-
tion5—different main ethnic entities populating the area.
When examining historical maps, it quickly becomes clear, that the area was
over time a very dynamic entity, with frequently changing territories and bor-
ders. In more recent historical terms, with the expansion of the Russian Empire,
the conflicts between the Russian and the Ottoman Empire changed the make-
up of the region and its borderline, being the site of numerous conflicts and bat-
tles (Allen/Muratoff 1953). The empires redrew the borders and relocated popu-
lations in the course of the events, creating thereby a more homogenous border
region. After a short period of an independent Armenian state between 1918 and
1920, the Caucasus, including Armenia, became part of the USSR, and the Arme-
nian SSR was founded as one of its constituent republics, while the struggle for
territory with the Ottoman Empire continued for some years. After the last battle

4
On the other hand, the fear of losing the nation as a reference for identity might, in some
cases, also reinforce its strength.
5
There is a substantial Kurdish population on the Turkish, but also a small Kurdish population
on the Armenian side of the border.
532 Pascal Kluge

between the Red Army and the Ottoman forces, however, the border attained
today’s course, as determined in the treaty of Moscow and Kars in 1921 (Tocci
2007: 7).6
The changes after World War II fall, for the most part, “into the category of
power politics” (Lundén 2010: 13). As Turkey was a member of NATO and the
USSR headed the Warsaw Pact, the Iron Curtain separated Turkey and the Ar-
menian SSR from each other. Both sides of the border were in different camps,
making cross-border activities on a regular and daily basis most difficult, if not
impossible. With the demise of the USSR, many of the former Soviet Republics
claimed their independence, including Armenia. The borderline between Turkey
and the then independent Armenia coincided with the former one between the
Armenian SSR and Turkey, and thus the one stated in the treaty of Moscow and
Kars. Meanwhile, new ambiguities arose from different approaches as how to in-
terpret the treaty.7

RELATIONS BETWEEN TURKS AND ARMENIANS

However, looking at these developments, to postulate constantly troublesome
relations between Armenians and Turks would do no justice to the complex,
and, over long periods, fruitful relations. Also in the area of today’s Turkish-Ar-
menian border we can find a shared history of Armenians and Turks as this re-
gion is a main area of historical settlement of Armenians and a region which the
Ottoman Empire, expanding eastwards from its first outlines in Western Anato-
lia, incorporated into its confines.
In the Ottoman Empire, the various non-Muslim groups of the Ottoman Em-
pire—Armenians, Greeks and Jews—were integrated into the complex societal
system (Goffman 2002: 75). Until the implementation of laws in the years 1839
and 1856 during the Tanzimat period of reformation (1839-1876) in which the
different religious entities of the Ottoman Empire were for the first time legally
treated as equal citizens, the millet system divided and organised people of the
Ottoman Empire according to their religion (Hanioğlu 2008). Within this system,
the non-Muslim millets had partly their own legislation and enjoyed, compared

6
Today, Turkey sometimes suspects that Armenia might raise claims to those parts of Turkish
territory that were historically inhabited largely by Armenians.
7
Tocci (2007: 7) explains: Turkey demands from Armenia to officially recognise the existing
borderline. Armenia, however, argues that there is no need for a new declaration as the treaty of
1921 has never been revoked by either side.
 Turkey’s Border with Armenia 533

to their co-religionists in Europe, substantial freedoms. This regulated hierar-

chical order created a “deep layer of everyday culture […] which permitted a
high degree of ethnic and religious diversity and an official tolerance or rather
indifference towards it” (Roth 2006: 20). Within this system, “Armenians were an
important and visible part of the empire’s economic and cultural life and they
prospered in the Ottoman Empire until the last decades of the 19th [c]entury”
(Tocci 2007: 27).
At the end of the 19th century and the beginning of the 20th century, espe-
cially during the reign of Sultan Abdul Hamid II (1876-1909), major massacres
against Armenians took place in the confines of the Ottoman Empire. After these
uneasy and often violent years, in 1915, the relations between the Ottoman Em-
pire and its Armenian subjects took an ever abrupt turn with the abduction and
murder of large parts of the Armenian population of Anatolia. In a state-induced
act of ethnic expulsion and expropriation Armenians were, with coercion and
force, expelled from their homes and forced to march towards far-away, forbid-
ding regions while, in this process, many if not most of them perished.

THE CONFLICT ISSUES GENOCIDE AND NAGORNO-KARABAKH

One of the main controversial issues between Turkey and Armenia are the
events of 1915, mainly considered as an act of genocide. While Turkey officially
rejects the term genocide altogether, some, mostly Western, countries, have offi-
cially recognised the events as genocide,8 pressuring Turkey’s government to
reconsider its rejecting position and to officially acknowledge the events as gen-
ocide, often initiated and supported by organisations of the Armenian diaspora.
Another issue, on a geopolitical level, influenced more recently the Turkish-
Armenian relations negatively, as Tocci (2007: 5–6) outlines: the declaration of
independence of Nagorno-Karabakh—populated mostly by Armenians and re-
garded by Armenia as historical Armenian land—from Azerbaijan in 1992 and
the creation of security zones around it, including a corridor connecting Na-
gorno-Karabakh with Armenia. The violent conflict was militarily supported by
Armenian troops, and caused a significant displacement of Azerbaijanis. Alt-
hough not involved directly, Turkey’s close cultural, linguistic and economic ties
with Azerbaijan had the country take a firm stand for Azerbaijan: Turkey offi-

8
For a list of the countries see Sabah newspaper, http://www.sabah.com.tr/fotohaber/dunya/
soykirimi_ taniyan_ulkeler?tc=23&albumId=15310&page=1, accessed 13 August 2012.
534 Pascal Kluge

cially closed its borders with Armenia and ruled out diplomatic relations alto-
gether.
To solve the pressing problems between Turkey and Armenia, negotiations
were undertaken, leading to the signing of two protocols by Armenia and Turkey
in 2009 in Switzerland. As the problematic issues between the two countries are
very complex, no precondition concerning the disputed genocide nor the ques-
tion of Nagorno-Karabakh was included in the original text. Azerbaijan, there-
upon, having “successfully stricken moral and nationalist chords” in Turkey, ex-
erted immediate pressure on Turkey’s government, making its position clear by
“warning of the devastating blow the protocols would have on Turkey’s kin and
ally”, demanding Turkey to stand politically and morally behind Azerbaijan’s
cause (Tocci 2007: 18). Prime Minister Erdoğan travelled to Baku to calm down
Azerbaijan’s government (Taşpınar 2009a), announcing later on that Turkey will
not bend down nor will it give in to Armenia’s claims (Ay, H. 2009). Turkey,
thereupon, added as a precondition the solution of Nagorno-Karabakh if further
steps in the rapprochement between Turkey and Armenia were to follow. Also
the Armenian government was put under substantial pressure, not least from
organisations of the Armenian diaspora, to abandon the road to rapprochement.
Subsequently, both countries, Turkey and Armenia, stepped out of the political
process of the protocols.

THE TURKISH-ARMENIAN RELATIONS AS DEPICTED IN THE TURKISH PRESS

The question of the acknowledgment of the events of 1915 as genocide, and the
declaration of independence of Nagorno-Karabakh from Azerbaijan receive wide
coverage in the media, especially when there are current events connected to
one of the topics. In the period researched, the planned criminalisation of the
denial of the events of 1915 as genocide in the French penal law as well as the
failure of the protocols were broadly discussed. Another widely discussed topic in
this period was the situation of the estimated 100,000 Armenians working ille-
gally in Turkey, and the remarks of Prime Minister Erdoğan, saying that those
100,000 Armenians might be expelled from Turkey as a consequence of their ille-
gal activities. The treaty of Moscow and Kars of 1921, with its possible implica-
tions for today’s course of the border, was only seldom discussed in the two
newspapers.
 Turkey’s Border with Armenia 535

The disputed genocide

Prime Minister Erdoğan—representing Turkey’s official position—evaluates the
question of 1915 as a very sensitive issue for Turkey (Sabah 2010b). Erdoğan sees
the situation as complex and the steps Turkey needed to take as manifold, while
the claims would put Turkey under immense pressure (ANKA 2010). It is re-
ported that Erdoğan might stay away from the Nuclear Security Summit due to
the discussions on the topic of genocide (Bostan 2010). Concerning the planned
genocide declaration in the French parliament, Erdoğan, while declaring all
claims mere accusations and lies, argues that the reasons for the planned resolu-
tions ultimately lay in a deeply rooted Islamophobia (AA 2009a). The claims
were groundless to such a degree that they could not put dirt on Turkey’s history
(AA 2010c). Furthermore, Erdoğan is of the opinion that no excuse should be ex-
tended to the Armenians (rather to other minorities in Turkey, for example the
Romani) (Sabah 2010c). In any case, the evaluation of the events in 1915 should
be left to historians (Sabah 2010d). The opposition parties, such as the secular
Kemalist party CHP (Cumhuriyet Halk Partisi) and the nationalist party MHP
(Milliyetçi Halk Partisi), also condemn the proposed bill of France’s government
(AA 2012, Şafak 2005).
Also columnists and commentators of the two newspapers analysed
generally do not approve of the developments in France. In some articles, French
politics in general are criticised sharply. The then French President Sarkozy is
depicted as a poor politician, as he lacked sincerity concerning the Armenian
case, mentioning in this regard also Erdoğan’s remark Sarkozy should first deal
with the questions of his own country before interfering in other countries’
issues (Sabah 2011a). In another article, it is stated that Sarkozy’s attempt to
criminalise the denial of the events of 1915 as genocide will not disturb the
Turkish public to a great extent as exaggeration was a common means of Turkish
politics to attract public attention (Barlas 2012), giving the reader the impression
the case is not really worth discussing. One columnist argues that, while Turkey
had expressed its sorrow for the events of 1915, Armenia occupied seven areas
around Nagorno-Karabakh, concluding that it was Armenia, and not Turkey,
which is acting not considerately (Ilıcak 2009a). The events of 1915, in another
article, are seen as a mistake; however, that mistake had to be seen in its
historical context (Uluç 2010). A British politician is cited, saying that such
events were to be expected in that kind of war situation (Cihan 2010).
536 Pascal Kluge

Organisations of the Armenian diaspora are accused of initiating and sup-

porting many of the actions against Turkey, describing their lobby as very deci-
sive and strong on the issue of the genocide (Çakır Morin 2012). The organisa-
tions are blamed with disturbing the process of normalisation and rapproche-
ment between the two countries (Çandar 2010). It is noted that Armenia’s Presi-
dent Serzh Sargsyan was also put under pressure by the Armenian diaspora as he
agreed not to add any preconditions to the protocols (Batur 2009).
Governments, organisations and individuals that are critical of the genocide
resolution are mentioned in a positive manner. It is noted that some govern-
ments of (predominantly Muslim) countries in the Balkans, the Middle East, and
Africa (Çelik/Can 2012), but also organisations in France supported Turkey’s
stance and rejected the term genocide for the issue (Sabah 2012b). It is remarked
that even some Armenians questioned whether the events of 1915 are to be in-
terpreted the way they were by France’s government (Baydar 2011), as well as the
French-Armenian singer and entertainer Charles Aznavour who is quoted as not
approving of the word genocide in this context (Sabah 2011b).
On the other side, when not directly touching the political developments in
France, there are also voices in the media that, without mentioning the term
genocide, describe the events of 1915 as something the government of the Otto-
man Empire was aware of, thus knowingly allowing the diminishing of the Ana-
tolian Armenian population from over one million at the beginning of the 20th
century to 50,000 inhabitants today9 by expulsion, by lack of provision, or by
murder. The public demonstrations in Istanbul in 2010 commemorating the
events of 1915, and, prior to that, the great attendance of the funeral of the Turk-
ish-Armenian journalist Hrant Dink in 2007, are, according to one columnist, at
least a first step on the way to a normalisation of the relations between the two
nations (Tınç 2010a). Covering the same demonstrations, another columnist
concludes that, if the politicians did little for that matter, at least the Turkish
public pushed it further with its actions (Çandar 2010).

Nagorno-Karabakh
The Armenian-Azerbaijani dispute over Nagorno-Karabakh is covered widely,
assessing it as an important issue for Turkey with ample implications for the re-
gion. It is reported of a militarily precarious situation between Armenia and

9
These figures are contested, yet it is agreed on a dramatic diminishment of the Anatolian
Armenian population.
 Turkey’s Border with Armenia 537

Azerbaijan which might possibly lead to a war between the two nations (Dış
Haberler 2011; Sabah 2011c). Turkey’s official position on this issue is clear, con-
demning the declaration of independence of Nagorno-Karabakh, the creation of
the security zone and the subsequent displacement of Azerbaijanis as a violent
act against international law and human rights. The situation is described with
respect to its consequences for the security of the region, but especially of Azer-
baijan, as Azerbaijan is seen as Turkey’s brother nation. After the failure of the
protocols, Erdoğan attests that Turkey would always be on Azerbaijan’s side and
the borders closed until Armenia found a solution for Nagorno-Karabakh which
fully took into account the interests of Azerbaijan (Şafak 2009a). The close eco-
nomic and cultural relationship with Azerbaijan leaves Turkey in the eyes of
many no other option than to support its cause. Erdoğan confirms this when
saying that the key to a Turkish-Armenian rapprochement rested on the rela-
tions between Armenia and Azerbaijan (AA 2009b); and that he would never
make politics to the detriment of Azerbaijan (Sabah 2009b). A columnist, with
obvious sympathies for the AK Parti government, is of the opinion that it was
now Armenia’s turn to make the next move and that the brotherhood with Azer-
baijan was more important than the economic situation of 1.8 million Armenians
(Kadak 2009). Also the former President Süleyman Demirel is quoted as being
against an opening of the borders as long as Armenia was involved in Nagorno-
Karabakh (Donat 2010). Deniz Baykal, from the secular Kemalist CHP, says that
there had been reasons to close the borders, so one had to respect those reasons
(Ilıcak 2009b), and, agreeing with Erdoğan on this issue, that the border should
not be opened before a solution to the problems regarding Nagorno-Karabakh
was found (Ajanslar 2009). The military protective function of the border was
emphasised when a member of the CHP fraction attracted (media) attention by
taking off his tie and his jacket during his address in a parliamentary session, un-
derlining his protest against what he calls a neglect of the border region from
state-side (Hürriyet 2012). He asks provocatively if the area was going to be ren-
dered eventually to Armenia, showing a poster depicting a soldier standing alone
on a hill next to a Turkish flag. Also in an act of underlining the military protec-
tive function of the border, the leader of the nationalist MHP, Devlet Bahçeli,
goes to the Turkish-Armenian border, speaking there of the unjust actions of
Armenia. He asks the attending members of parliament to stand next to each
other at the border to form a wall, thus emphasising his hostile attitude against
Turkey’s neighbour (AA 2010b). Bahçeli presents himself as combative against
538 Pascal Kluge

Armenia as he would, if necessary, reconquer the Armenian Church of the Holy

Cross on Akdamar Island in Turkey, close the Armenian border (Becerikli 2010).
When reporting on atrocities directed against Azerbaijanis at the Armenian-
Azerbaijan border or at the border with Nagorno-Karabakh (Sabah 2009a; Cihan
2009), Armenians are depicted in some articles as aggressive (Aksoy 2010), their
doings as brutal and devious (Hacıoğlu 2011).
Turkey as a haven of multicultural life
Turkey, on the other hand, is often described as a country which is unique and
peaceful, having a long-standing tradition of accommodating people of different
ethnic and religious backgrounds, a country where different religions could live
harmoniously together. The (then) main negotiator for Turkey’s accession to the
European Union, Egemen Barış, says that Turkey had opened up considerably to
its minorities, as, for example, Armenians could nowadays hold their mass in the
Church of the Holy Cross on Akdamar Island (İnsan 2011), something that had
not been possible a few years ago. Prime Minister Erdoğan describes Turkey as a
multicultural country where, next to Turks and Kurds, also Laz, Cirkassians, Ar-
abs, Abkhazians, Armenians, Greeks, Georgians and Romani could live together
under one flag (Sabah 2012a). Regarding the Armenians of Turkey, it is main-
tained officially that nowadays they experienced no problems living as a minor-
ity in the country (Hürriyet 2010b). Mardin, in a report on Turkey’s cultural herit-
age, is praised for its multiculturalism, with Turks, Kurds, Arabs, Assyrians, Ar-
menians and Yezidi living together amicably (Cihan 2012), while Malatya is de-
scribed as a haven for many religions and ethnicities, including Armenians (Hür-
riyet 2011). Antakya is characterised as a multireligious place where the cuisines
of Jews, Christians and Moslems could be enjoyed (Dağlar 2010).
The history of Armenians and Turks is described as interconnected and in-
terwoven in lineage, culture, music, dances, and food, with many Armenians
having contributed to Turkish arts (Uluç 2010). On a linguistic level, one article
writes about the influence of the Armenian and Turkish languages on each other
regarding their vocabularies, with especially many Turkish words having entered
Armenian (Kuşdemir 2010). When speaking of the border region, it is often de-
picted as a zone of interethnic and multicultural contact. Kars, due to its geo-
graphical location—situated on a plateau surrounded by mountains—is de-
scribed as having been influenced more by the region as a whole then by the cul-
tural and political centre of Turkey (Köker 2010a). When reporting on the house
in which Yeghishe Charents, an Armenian poet and political activist, was born, it
 Turkey’s Border with Armenia 539

becomes clear that also the Turkish side of the border constitutes an important
part of the Armenian heritage. Kars, as the land of their forefathers (Sabah
2010a), is said to be of great interest to many Armenians as they would like to
travel to the region, which some already had (Köker 2010b). Some voices idealise
the situation at the border in a naïve, somewhat nostalgic way. One parliamen-
tarian of the CHP from the province of Ardahan (adjacent to the Georgian and
Armenian borders), for example, claims that today it was even possible to freely
cross the border from Turkey to Armenia on a horse—and that there was only
friendship and brotherhood between the two neighbours (Aktemur 2012).

…and some critical voices

Some voices, though less frequent, challenge the (often official) sugar-coated im-
age of Turkey as a haven for multicultural coexistence. It is criticised that Turk-
ish citizens of Armenian origin were not as protected by the state as their Mos-
lem Turkish counterparts, citing the case of Hrant Dink who had to die due to
the neglecting attitude of the Turkish authorities (Birand 2011; Yılmaz 2010). Fur-
thermore, it is complained that the Armenians of Turkey were not treated as
equal citizens as they, for example, faced discrimination in the army (Yılmaz
2011). Also Armenians from Armenia are reported to experience discrimination
in Turkey, as, for instance, a group of Armenian journalists was held up for many
hours at the airport of Istanbul and threatened by the authorities to be sent back
to their country if they caused any trouble in Turkey (Erduran 2009). These inci-
dents are viewed with great disbelief as it is argued that Turks, considering their
history, should comport themselves very respectfully towards other nationalities,
but especially towards Armenians.
Criticised is also the use of the fate of the estimated 100,000 Armenians,
working illegally in Turkey, for political propaganda. Erdoğan’s remark to expel
the Armenians is chiefly met with incomprehension and seen as an act against
the spirit of human rights (Ateş 2010). Some columnists say that Erdoğan’s re-
mark did provoke thoughts of the genocide (Ilıcak 2010). Turkey should not for-
get what it did to its minorities, so Erdoğan should be careful when making re-
marks of that kind (Barlas 2010b). It is found more adequate to show sympathy
for the humanitarian tragedy behind these Armenians, often women working as
domestic help (Barlas 2010a). Basic needs, such as education and health care,
should be granted to them (Sabah 2011d). It is also proposed to open the schools
of the Armenians foundations for their children (Yılmaz 2011), not least because
540 Pascal Kluge

they could be good ambassadors for Turkey in their homeland by telling how
well they were received and treated (Oğhan 2010).

Political deficiencies
Before the failure of the protocols, the Turkish government was often praised for
its efforts concerning the Turkish-Armenian rapprochement, arguing that Tur-
key should comprehend the protocols as a positive step into the future (Aras
2010). After their failure, however, a high level of frustration can be observed,
many voices demanding a different approach of Turkish politics towards the
Armenian issue. In retrospect, Turkey is believed to have caused the locked situ-
ation it is in today by closing its borders with Armenia in 1993. If it had not
closed the borders, it is argued, Armenia would not have a motive for its anti-
Turkish attitudes (Oğhan 2010). Looking at more recent developments, Turkey’s
government is seen as having fumbled to add the solution of the issue of Na-
gorno-Karabakh as a precondition after the signing of the protocols; the issue
should have remained excluded for the time being (Taşpınar 2009b). If the gov-
ernment had been more courageous and not have added this precondition, then
the issue of genocide would not be on the agenda the way it is today (Taşpınar
2010b). The way the government announced the opening of the borders is as-
sessed as inappropriate as it provoked an instant negative reaction from Azer-
baijan (Ekşi 2010; Haberler 2010). The matter should have been dealt with in a
more subtle way, by trying not to raise ample attention to the possible opening
of the border. When it comes to the Armenian diaspora and the Azerbaijani gov-
ernment, Turkish politics are criticized of only reacting to them instead of acting
proactively. In this context, Turkey is believed to be taken hostage by Azerbai-
jan’s government (Çandar 2012), while it seemed that Turkey will remain further
on Azerbaijan’s side (Taşpınar 2010b). The responsible officials are thought to
not act decisively and swiftly enough (Tınç 2010a), and it is assumed that there is
a general lack of political will to solve the questions with Armenia (Babacan
2011). If politicians would be more foreseeing and flexible, however, politics
could be more influential, coming closer to a solution of the problems (Tınç
2010b).

Ways out
In the various columns and articles researched, there are quite a few solutions
proposed to revive the Turkish-Armenian rapprochement. In addition, readers of
 Turkey’s Border with Armenia 541

the newspapers are also informed about projects strengthening Turkish-Ar-

menian relations.
Generally viewed, a step-by-step approach is favoured in order to break the
vicious cycle the two countries are stuck in. Turkey should not be impeded by
the negative outcome of the protocols as it has to be taken into account that both
sides are tense in this matter (Barlas 2009). A columnist argues that the Turkish
government should stick further to the protocols and put the question of Na-
gorno-Karabakh aside. Then, a new era—socially, economically, and diplomati-
cally—between Turkey and Armenia could begin (Taşpınar 2011), proposing that
Turkey should just open the border (Taşpınar 2010a). It is also called for the abo-
lition of the visa obligation for Armenians (Aköz 2010). Another columnist be-
lieves that Turkey’s government should not categorically declare the solution of
the Nagorno-Karabakh issue as an imperative precondition for the opening of
the borders with Armenia and the establishment of diplomatic relations, but
could, in a first step, be content with Armenia’s withdrawal from the seven secu-
rity districts around Nagorno-Karabakh (Şafak 2010). In short, to get the relations
with Armenia started, many are of the opinion that politics should, at first, not
concentrate on the delicate and complicated issues. It is proposed that the first
contacts between the two countries should be established in the fields of trade,
sports and arts (Uluç 2010), or, as some locals of the Turkish border town of Iğdır
tell a journalist, there should be trade instead of enmity. (Övür 2009).
Furthermore, it is argued that the region would profit from open borders as
they would make the construction of a new land road from Turkey via Armenia
to Eurasia possible. This would not only enhance the economy of the region, but
the road would, at the same time, become a road of peace (Şafak 2009b). Moreo-
ver, as the opening of the border would cut travel time significantly, Armenians
could easily visit historical sites of interest in Turkey (Sabah 2010a), which in
turn would bring more Armenian tourism to the country (Köker 2010a). It is also
reported of a project to create an air link between Van and Yerevan (Hürriyet
2010a). Eventually, it is suggested that the Armenian diaspora also join the pro-
cess (Güvenç 2010), and believed that Turkey needed to share openly the pain of
the Armenians, preferably conveyed through a statement by the Turkish Head of
State (Taşpınar 2011).
Apart from the political analyses made by columnists, the newspapers report
also of different projects with the objective of establishing relations of various
kinds between Turkey and Armenia. Some organisations in Turkey tried to keep
542 Pascal Kluge

the delicate ties between Armenia and Turkey alive by arranging meetings on
both sides of the border. TESEV (Türkiye Ekonomik ve Sosyal Etütler Vakfı), for
example, hosted together with the Friedrich-Nauman-Stiftung, a German foun-
dation, a conference in Kars, discussing, among other topics, the opening of the
border, whereas previous conferences had been organised in Istanbul and Yere-
van (Köker 2010a). The political party GTP (Güçlü Türkiye Partisi), politically
close to the ruling AK Parti, is reported to have founded a friendship group, ar-
ranging a meeting of 500 young Armenians and Turks first in Yerevan, while an
invitation of the Armenians to Istanbul followed at a later date (Övür 2010). In a
smaller scope, twenty Turkish students from Ankara invited twenty students
from Armenia to watch the football match between the two countries and the
signing of the protocols on television (Ay, S. 2009), a meeting which was planned
to take place in the only remaining village in Turkey solely inhabited by Armeni-
ans, located near the Syrian border. It is also reported of an interchange in aca-
demia, as the language department of the University of Iğdır, a Turkish city close
to the Armenian border, announces its intention to engage a language instructor
from Armenia (AA 2010b).

RESULTS AND OUTLOOK

When analysing the discourse of the Turkish-Armenian relations with regard to
the Turkish-Armenian border in the newspapers Hürriyet and Sabah, two dis-
tinct approaches can be observed: firstly, an approach, in which the border is
depicted as a line of separation, with a potential enemy on the other side, sec-
ondly, an approach, in which the border is depicted as a zone of contact, rather in
the notion of the American term frontier, with various options for further rap-
prochement.
One perspective—often an official one, conveyed by the Turkish government
or the political establishment in general—depicts the border as a border of sepa-
ration. Especially when discussing the topics of the disputed genocide and of
Nagorno-Karabakh, the description turns harder, sometimes emphasising hostile
aspects. Armenia is then perceived as a threat to the Turkish nation-state, iso-
lating itself with its aggression towards its neighbour Azerbaijan and exploiting
history vis-à-vis Turkey. Also Armenian atrocities are a recurrent topic, and Ar-
menians described as aggressive and merciless. In this context, the border is em-
phasised in its separating functions order, protection, control and identity for-
mation. On the other side, when describing Turkey in respect to its minorities,
 Turkey’s Border with Armenia 543

the descriptions are usually positive and speak of Turkey as an open, multireli-
gious and multiethnic society. Those descriptions are in their character at times
somewhat musealised like a backwardly directed utopia, in a way indirectly ide-
alising the multicultural aspects of the Ottoman Empire. However, in doing so,
they also affirm indirectly negative perceptions of the other as they create a set-
ting of a positively connoted us—implying a dichotomously perceived them or
other.
The approaches towards the Turkish-Armenian border by voices outside the
government and the political establishment are often much softer and have a
constructive character, portraying the border as something to overcome, even as
an entity with connecting qualities, and thus rather in the figurative American
sense of a frontier. The desire for more trust and considerate and thoughtful ac-
tions are expressed. Sometimes, the border region is depicted as a distinct cul-
tural landscape, its identity transcending the borderline, and (re-)creating a
space of common heritage. This transculturality in which many forms of culture
exist simultaneously and overlap, however, is sometimes conveyed in a nostalgic
way. Nevertheless, also pragmatic approaches do exist, such as academic ex-
change among universities, or a planned direct air traffic link between Van and
Yerevan. Still, the unifying narratives, let alone projects, which could create a
base for future relations, are less frequent.
The question for the future would be how the two different approaches
within the national discourse on the Turkish-Armenian border interplay with
each other, and to what degree the soft and constructive approach can affect poli-
tics (and society at large) positively. The expectations from politics are high,
and—mostly in the columns of the newspapers—many suggestions to improve
the situation are made. In any case, hopes for a quick solution should not be in-
flated as there is a substantial discrepancy between vision (sometimes naïve and
idealising) and the reality of politics.
544 Pascal Kluge

BIBLIOGRAPHY
AA (2009a), “Erdoğan, ABD televizyonunda”, Sabah 09 December 2009, http://www.sabah.com.tr/
Dunya/2009/12/09/erdogan_abd_televizyonunda, accessed 09 August 2012.
―― (2009b), “Siyaset, Erdoğan: Türkiye eksen değiştirmedi”, Sabah 08 November 2009, http://
www.sabah.com.tr/Siyaset/2009/11/08/erdogan_turkiye_eksen_degistirmedi, accessed 09
August 2012.
―― (2010a), “Igdır [sic!] Üniversitesi’ne Ermenistan’dan hoca”, Sabah 13 December 2010, http://
www.sabah.com.tr/Gundem/2010/12/13/igdir_universitesine_ermenistandan_hoca,
accessed 09 August 2012.
―― (2010b), “Bahçeli’den Bağış’a sert tepki”, Sabah 02 October 2010, http://www.sabah.com.tr/
Gundem/2010/10/02/bahceliden_bagisa_sert_tepki, accessed 09 August 2012.
―― (2010c), “Erdoğan’dan önemli açıklamalar”, Sabah 19 March 2010, http://www.sabah.com.tr/
Gundem/2010/03/19/erdogandan_onemli_aciklamalar, accessed 09 August 2012.
―― (2012), “Bahçeli’den Fransa tepkisi”, Sabah 28 January 2012, http://www.sabah.com.tr/Gun
dem/2012/01/28/bahceliden-fransa-tepkisi, accessed 09 August 2012.
Ajanslar (2009), “Baykal: Askeri güçle sonuç alma döneminin geride kaldı”, Sabah 06 November
2009, http://www.sabah.com.tr/Siyaset/2009/11/06/baykal_askeri_ gucle_sonuc_alma_done
minin_geride_kaldi, accessed 09 August 2012.
Aköz, E. (2010), “Yalan mı?.. Yanlış mı?...”, Sabah 18 March 2010, http://www.sabah.com.tr/Yazarlar/
akoz/2010/03/18/yalan_mi_yanlis_mi, accessed 09 August 2012.
Aksoy, Y. (2010), “Nahçıvan’da İzmirli sanat elçimiz”, Hürriyet 19 December 2010, http://hurarsiv.
hurriyet.com.tr/goster/ShowNew.aspx?id=16570367, accessed 03.08.2012.
Aktemur, D. (2012), “CHP’li milletvekili güreşte”, Hürriyet 31 January 2012, http://hurarsiv.hurriyet.
com.tr/goster/ShowNew.aspx?id=19811005, accessed 05 August 2012.
Allen, W. E. D. / P. Muratoff (1953), Caucasian Battlefields: A History of the Wars on the Turco-
Caucasian Border, 1828-1921, Cambridge.
ANKA, “CHP’den Ermeni önergesi”, Sabah 22 March 2010, http://www.sabah.com.tr/Dunya/2010/
03/22/minsk_grubu_zombi_mi, accessed 09 August 2012.
Aras, B. (2010), “Ermenistan’la normalleşme”, Sabah 17 Mart 2010, http://www.sabah.com.tr/Yazar
lar/bulent_aras/2010/03/17/ermenistanla_normallesme, or: Setav 17 March 2010, http://
www.setav.org/public/HaberDetay.aspx?Dil=tr&hid=28822&q=ermenistan-la-normalles
me, accessed 09 August 2012.
Ateş, Ş. (2010), “‘Ermeni açıklaması vahim’”, Sabah 18 March 2010, http://www.sabah.com.tr/Siya
set/2010/03/18/ermeni_aciklamasi_vahim, accessed 09 August 2012.
Ay, H. (2009), “Erdoğan: eğilip bükülmeyiz”, Sabah 11 October 2009, http://www.sabah.com.tr/Gun
dem/2009/10/11/erdogan_egilip_bukulmeyiz, accessed 09 August 2012.
Ay, S. (2009), “Tek Ermeni köyünde imza ve maç heyecanı”, Sabah 09 October 2009, http://www.
sabah.com.tr/Gundem/2009/10/09/tek_ermeni_koyunde_imza_ve_mac_heyecani,
accessed 09.08.2012.
Babacan, N. (2010), “Vekillerin önceliği tezkere ve pantolon”, Hürriyet 29 September 2011, http://hur
arsiv.hurriyet.com.tr/goster/ShowNew.aspx?id=18856004, accessed 03 August 2012.
 Turkey’s Border with Armenia 545

Barlas, M. (2009), “Savaşmak barışmaktan daha ucuz değildir”, Hürriyet 03 September 2009, http://
www.sabah.com.tr/Yazarlar/barlas/2009/09/03/savasmak_barismaktan_daha_ucuz_degild
ir, accessed 14 August 2012.
―― (2010a), “Kaçak Ermenilerin yüzde 96’sı kadın işçilermiş...”, Sabah 22 March 2010, http://
www.sabah.com.tr/Yazarlar/barlas/2010/03/22/kacak_ermenilerin_yuzde_96si_kadin_iscil
ermis, accessed 09 August 2012.
―― (2010b) “Başbakan Erdoğan artık doğaçlama yapmamalıdır...”, Sabah 19 March 2010, http://
www.sabah.com.tr/Yazarlar/barlas/2010/03/19/basbakan_erdogan_artik_dogaclama_yap
mamalidir, accessed 09 August 2012.
―― (2012), “Sarkozy bizi sinirlendirmeyi asla başaramaz”, Sabah 10 January 2012, http://www.
sabah.com.tr/Yazarlar/barlas/2012/01/10/sarkozy-bizi-sinirlendirmeyi-asla-basaramaz,
accessed 09 August 2012.
Batur, N. (2009), “Kan davası bitecek mi?”, Sabah 11 October 2009, http://www.sabah.com.tr/Gun
dem/2009/10/11/kan_davasi_bitecek, accessed 09 August 2012.
Baydar, Y. (2011) “Oylamanın ardından”, Sabah 26 December 2011, http://www.sabah.com.tr/Yazar
lar/baydar/2011/12/26/oylamanin-ardindan, accessed 09 August 2012.
Becerikli, U. (2010), “Bahçeli, Ani’de cuma namazı kıldı”, Sabah 02 October 2010, http://www.
sabah.com.tr/Gundem/2010/10/02/bahceli_anide_cuma_namazi_kildi, accessed 09 August
2012.
Birand, M. A. (2011), “Hrant cinayetine göz yumanlar artık cezalandırılmalı…”, Hürriyet 19 January
2011, http://hurarsiv.hurriyet.com.tr/goster/ShowNew.aspx?id=16796930, accessed 03 Au-
gust 2012.
Blake, G. (2000), “Borderlands under Stress: Some Global Perspectives”, in: Pratt, Martin; Brown,
Janet Allison (eds.), Borderlands Under Stress, London: 1–16.
Brock, L. (2000), “Risse in der nationalstaatlichen Ordnung: Die Grenzen von Staat, Wirtschafts-
raum und Gemeinschaftsbewusstsein sind nicht mehr deckungsgleich”, in: Der Überblick 4:
6–10.
Buchowski, M. / B. Chołuj (2001), Die Konstruktion des Anderen in Mitteleuropa. Diskurse, politische
Strategien und soziale Praxis, Frankfurt (Oder), Słubice, Poznań.
Buskjær Christensen, M. (2009), EU-Turkey relations and the functioning of the EU, Mobilizing
Local Networks for a Better Informed Dialogue on Turkey’s Accession to the EU. Policy
Brief 1, http://accesstr.ces.metu.edu.tr/dosya/ christensen.pdf, accessed 03 August 2012.
Carnevale, R. / S. Ihrig / Ch. Weiß (2005), Europa am Bosporus (er-)finden? Die Diskussion um den
Beitritt der Türkei zur Europäischen Union in den britischen, deutschen, französischen und
italienischen Zeitungen. Eine Presseanalyse, Frankfurt am Main, Berlin, Bern et al.
Cihan (2009), “1’i asker 2 Azeri öldürüldü”, Sabah 29 December 2009, http://www.sabah.com.tr/
Dunya/2009/12/29/1i_asker_2_azeri_olduruldu, accessed 09 August 2012.
―― (2010), “Lordlar Kamarası soykırıma geçit vermedi”, Sabah 31 March 2010, http://www.sabah.
com.tr/Dunya/2010/03/31/lordlar_kamarasi_soykirima_gecit_vermedi, accessed 09 August
2012.
―― (2012), “Dünya basını Mardin‘e hayran kaldı”, Sabah 08 May 2012, http://www.sabah.com.tr/
Turizm/2012/05/08/dunya-basini-mardine-hayran-kaldi, accessed 09 August 2012.
546 Pascal Kluge

Coles, P. (1968), The Ottoman Impact on Europe, London.

Çakır Morin, A. (2012), “Bakanın bu hareketi Ermenileri kızdırdı”, Hürriyet 20 July 2012, http://hur
arsiv.hurriyet.com.tr/goster/ShowNew.aspx?id=21026493, accessed 03 August 2012.
Çandar, C. (2010), “Türkiye-Ermenistan: Derin dondurucuda...”, Hürriyet 23 April 2010, http://
hurarsiv.hurriyet.com.tr/goster/ShowNew.aspx?id=14507914, accessed 08 August 2012.
―― (2012), “Hocalı’yı anma değil; milliyetçi-ulusalcı başkaldırı”, Hürriyet 28 February 2012, http://
hurarsiv.hurriyet.com.tr/goster/ShowNew.aspx?id= 20014163, accessed 05 August 2012.
Çelik, P. / M. Can (2012), “İslam dünyasından boykot”, Sabah 12 January 2012, http://www.sabah.
com.tr/Ekonomi/2012/01/12/islam-dunyasindan-boykot, accessed 09 August 2012.
Dağlar, A. (2010), “Lezzet cumhuriyeti Antakya’nın 593 tadı”, Hürriyet 24 May 2010, http://hurarsiv.
hurriyet.com.tr/goster/ShowNew.aspx?id=14816077, accessed 08 August 2012.
DeBardeleben, J. (ed.) (2005), Soft or Hard Borders? Managing the Divide in an Enlarged Europe,
Aldershot, Burlington.
Demandt, A. (1993), Deutschlands Grenzen in der Geschichte, München.
Dittgen, H. (2000), “The End of the Nation-State? Borders in the Age of Globalisation”, Borderlands
under Stress, London: 49–68.
―― (2010), “Ermenistan’dan Türkiye’ye 'sınır kapısı’ cevabı”, Hürriyet 16 July 2010,, http://hurarsiv.
hurriyet.com.tr/goster/ShowNew.aspx?id=15337468#, accessed 08 August 2012.
Donat, Y. (2010), “Kıbrıs”, Sabah 28 July 2010, http://www.sabah.com.tr/Yazarlar/donat/2010/07/ 28/
kibris, accessed 09 August 2012.
Dürrschmidt, J. (2002), Globalisierung, Bielefeld.
Ekşi, O. (2010), “Kapı aralanıyor”, Hürriyet 15 July 2010, http://hurarsiv.hurriyet.com.tr/goster/Show
New.aspx?id=15318883, accessed 08 August 2012.
Erduran, R. (2010), “Çuvallarımız”, Sabah 17 September 2010, http://www.sabah.com.tr/Yazarlar/
refik_erduran/2010/09/17/cuvallarimiz, accessed 09 August 2012.
Glick Schiller, N. (2004), “Transnationality”, A Companion to the Anthropology of Politics, Oxford:
448–467.
Goffman, D. (2002), The Ottoman Empire and Early Modern Europe, Cambridge.
Güvenç, D. (2010), “Sınırlardan önce diyaspora açılımı”, Sabah 16 April 2010, http://www.sabah.
com.tr/Dunya/2010/04/16/sinirlardan_once_diyaspora_acilimi, accessed 09 August 2012.
Haberler, D. (2011), “‘Azerbaycan-Ermenistan arasında savaş çıkabilir’”, Hürriyet 08 February 2011,
http://hurarsiv.hurriyet.com.tr/goster/ShowNew.aspx?id= 16968339, accessed 03 August
2012.
Hacıoğlu, N. (2011), “Ermenistan sınırında kanlı ‘çocuk oyunu’”, Hürriyet 16 July 2011, http://hurarsiv.
hurriyet.com.tr/goster/ShowNew.aspx?id=18266270, accessed 03 August 2012.
Halman, T. S. (1987), “Life of Literature and Death of Ideologies in Turkey”, Translation 19: 3-6.
Hanioğlu, M. Ş. (2008), A Brief History of the Late Ottoman Empire, Princeton, Oxford.
Haslinger, P. (2000), “Funktionsprinzip Staatsgrenze. Aspekte seiner Anwendbarkeit im Bereich
der Osteuropaforschung”, Grenzen in Ostmitteleuropa im 19. und 20. Jahrhundert: Aktuelle
Forschungsprobleme, Marburg: 57–66.
 Turkey’s Border with Armenia 547

Heywood, C. (1999), “The Frontier in Ottoman Histor. Old Ideas and New Myths”, Frontiers in Ques-
tion. Eurasian Borderlands, 700 – 1700, Basingstoke/New York, London: 228–250.
Holtom, P. (2010), “From Königsberg to Kaliningrad: A Journey through the Politics of History and
Memory”, Border Changes in 20th Century Europe, Münster: 271–296.
Hoorn, A. (2005), Die deutsch-polnische Grenze Grenz(er)leben und Zukunftsperspektiven von Ju-
gendlichen in Guben–Gubin, Berlin.
Hürriyet (2010a), “Türk-Ermeni sınırı açılsa 50 bin kişiye iş imkanı sağlanır”, Hürriyet 20 September
2010, http://hurarsiv.hurriyet.com.tr/goster/ShowNew.aspx?id=15819987, accessed 08 Au-
gust 2012.
―― (2010b), “Bakan Çağlayan: Kaliforniya’ya güç katmaya geldik”, Hürriyet 21 July 2010, http://hur
arsiv.hurriyet.com.tr/goster/ShowNew.aspx?id=15368581, accessed 08 August 2012.
―― (2011), “Gözlerimiz farklı ama gözyaşlarımız aynı”, Hürriyet 19 May 2011, http://hurarsiv.hur
riyet.com.tr/goster/ShowNew.aspx?id=17822143, accessed 03 August 2012.
―― (2012), “‘Çıldır’ma protestosu”, Hürriyet 04 July 2012, http://hurarsiv.hurriyet.com.tr/goster/
ShowNew.aspx?id=20910048, accessed 03 August 2012.
İnsan, M. (2011), “Bakan Bağış’tan sert eleştiri”, Hürriyet 26 November 2011, http://hurarsiv.hurriyet.
com.tr/goster/ShowNew.aspx?id=19334921, accessed 05 August 2012.
Ilıcak, N. (2009a), “Gazze’den, Hocali’ye [sic!]”, Sabah 20 October 2009, http://www.sabah.com.tr/
Yazarlar/ilicak/2009/10/20/gazzeden_hocaliye, accessed 09 August 2012.
―― (2009b), “Cesur bir adım”, Sabah 03 September 2009, http://www.sabah.com.tr/Yazarlar/ili
cak/2009/09/03/cesur_bir_adim, accessed 14 August 2012.
―― (2010), “Kaçak Ermeniler meselesi”, Sabah 20 March 2010, http://www.sabah.com.tr/Yazarlar/
ilicak/2010/03/20/kacak_ermeniler_meselesi, accessed 09 August 2012.
Jeismann, M. (1992), Das Vaterland der Feinde: Studien zum nationalen Feindbegriff und Selbstver-
ständnis in Deutschland und Frankreich 1792-1918, Stuttgart.
Kadak, Ş. (2009), “Başbakan’ın sözü var, sınırı açacaklarına inanmıyorum”, Sabah 25 October 2009,
http://www.sabah.com.tr/Ekonomi/2009/10/25/basbakanin_sozu_var_siniri_acacaklarina_
inanmiyorum, accessed 09 August 2012.
Kasaba, R. / S. Bozdogan (2000), “Turkey at a Crossroad”, Journal of International Affairs 54.1, http://
www.ciaonet.org/olj/jia/kasaba_final.pdf, accessed 12 September 2009.
Kennard, A. (2010), Old Cultures, New Institutions. Around the New Eastern Border of the European
Union, Berlin.
Kollis, A. (2010), “National Rebirth and Territorial Utopia (1914-45): the Experience of Italy, Ger-
many and Greece”, Border Changes in 20th Century Europe, Münster: 31–66.
Köker, İ. (2010a), “Kazanan yok, kaybeden çok”, Hürriyet 06 November 2010, http://hurarsiv.hur
riyet.com.tr/goster/ShowNew.aspx?id=16233595, accessed 08 August 2012.
―― (2010b), “Kars’taki bu metruk evin değeri 250 bin TL”, Hürriyet 04 November 2010, http://
hurarsiv.hurriyet.com.tr/goster/ShowNew.aspx?id=16215938, accessed 08 August 2012.
Krämer, R. (1997), “Von Brücken und anderen Grenzüberschreitungen an der östlichen Peripherie”,
in: Krämer, Raimund (ed.) Grenzen der Europäischen Union, Potsdam: 67–88.
Kuşdemir, Y. (2010), “MHP’li Başkan’a TDK’dan onur belgesi”, Hürriyet 14 October 2010, http://hur
arsiv.hurriyet.com.tr/goster/ShowNew.aspx?id=16045254, accessed 08 August 2012.
548 Pascal Kluge

Lundén, T. (2010) “Language, Ethnicity and Boundary Changes in 20th Century Europe”, Border
Changes in 20th Century Europe, Münster: 9–20.
Martinez, O. J. (1994), “The Dynamics of Border Interaction. New Approaches to Border Analysis”,
Global Boundaries. World Boundaries 1: 1–15.
Medijainen, E. (2010), “The Tartu Peace Treaty and Permanent Neutrality for Estonia”, Border
Changes in 20th Century Europe, Münster: 183–202.
Medvedev, S. (1999), “Across the Line: Borders in Post-Westphalian Landscapes”, Curtains of Iron
and Gold: Reconstructing Borders and Scales of Interaction, Aldershot, Brookfield: 53–56.
Mertelsmann, O. (2010), “The Disappearance of a Border and Incorporation into the USSR”, Border
Changes in 20th Century Europe, Münster: 203–216.
Milliyet (2008), “Kamu bankalarından ortak Katar’dan geldi”, Milliyet 22 April 2008, http://www.mil
liyet.com.tr/default.aspx?aType=HaberDetay&ArticleID=520017, accessed 17 September
2012.
Moll, M. (2010), “The German-Slovene Border in Southern Austria: from Nationalist Quarrels to the
‘Bleeding Wound’ and a Friendly Coexistence”, Border Changes in 20th Century Europe,
Münster. 111–140.
Oğhan, Ş. (2010), “Tabu yıkan Varujan mektupları,” Hürriyet 26 September 2010, http://hurarsiv.hur
riyet.com.tr/goster/ShowNew.aspx?id=15868414, accessed 08 August 2012.
Övür, M. (2009), “Alışveriş yapmak 'düşman’ olmaktan iyidir”, Sabah 27 October 2009, http://www.
sabah.com.tr/Yazarlar/ovur/2009/10/27/alisveris_yapmak_dusman_olmaktan_iyidir,
accessed 09 August 2012.
―― (2010), “Küçük partinin büyük hedefi”, Sabah 23 October 2010, http://www.sabah.com.tr/
Yazarlar/ovur/2010/10/23/kucuk_partinin_buyuk_hedefi, accessed 09 August 2012.
Paasi, A. (1995), “Constructing Territories, Boundaries and Regional Identities”, Contested Territory:
Border Disputes at the Edge of the former Soviet Empire, Aldershot: 42–61.
Padis, M.-O. (2009), “France: National identity”, in: National identity, http://www.jpr.org.uk/com
mon-good-in-europe/downloads/respublica_national_identity.pdf, accessed 17 September
2012: 2–3.
Power, D. (1999), “Frontiers: Terms, Concepts, and the Historians of Medieval and Early Modern
Europe”, Frontiers in Question. Eurasian Borderlands, 700 – 1700, Basingstoke/New York,
London: 1–12.
Püsküllüoğlu, A. (2004), Türkçe Sözlük, Istanbul.
Riedel, H. (1994), Wahrnehmung von Grenzen und Grenzräumen. Eine kulturpsychologisch-geogra-
phische Untersuchung im saarländisch-lothringischen Raum, Saarbrücken.
Roth, K. (2006), “Living together or living side by side? Interethnic Coexistence in Multiethnic So-
cieties”, Negotiating Culture: Moving, Mixing and Memory in Contemporary Europe, Berlin.
Sabah (2009a), “Sınır kentinde Ermeni ateşi: İki Azeri öldü”, Sabah 30 December 2009, http://www.
sabah.com.tr/Dunya/2009/12/30/sinir_kentinde_ermeni_atesi_iki_azeri_oldu, accessed 09
August 2012.
―― (2009b), “Ankara’dan Bakü’ye: Hiçbir süreç Azerbaycan dostluğundan önemli değil”, Sabah 02
September 2009, http://www.sabah.com.tr/Siyaset/2009/09/02/ankaradan_bakuye_hicbir_
surec_azerbaycan_dostlugundan_onemli_degil, accessed 14 August 2012.
 Turkey’s Border with Armenia 549

―― (2010a), “45 dakikalık yolu 12 saatte geldiler”, Sabah 06 November 2010, http://www.sabah.
com.tr/Yasam/2010/11/06/45_dakikalik_yolu_12_saatte_geldiler, accessed 09 August 2012.
―― (2010b), “Başkanlık sistemi gündeme gelebilir”, Sabah 19 April 2010, http://www.sabah.com.
tr/Gundem/2010/04/19/baskanlik_sistemi_gundeme_gelebilir, accessed 09 August 2012.
―― (2010c), “Romanlara devlet özrü”, Sabah 20 March 2010, http://www.sabah.com.tr/Gundem/
2010/03/20/romanlara_devlet_ozru, accessed 09 August 2012.
―― (2010d), “Angela Merkel Ankara’da”, Sabah 29 March 2010, http://www.sabah.com.tr/Gun
dem/2010/03/29/angela_merkel_ankarada, accessed 09 August 2012.
―― (2011a), “Sarkozy’ye sert çıktı”, Sabah 11 October 2011, http://www.sabah.com.tr/Gundem/2011/
10/11/basbakan-erdogan-konusuyor, accessed 09 August 2012.
―― (2011b), “Ermeniler Aznavour’a kızdı”, Sabah 08 September 2011, http://www.sabah.com.tr/
Dunya/2011/09/08/ermeniler-aznavoura-kizdi, accessed 09 August 2012.
―― (2011c), “Savaş kapıda mı?”, Sabah 09 February 2011, http://www.sabah.com.tr/Dunya/2011/02/
09/savas_kapida_mi, accessed 09 August 2012.
―― (2011d), “Kaçakları ihbar etmemek etik mi?” Sabah 07 January 2011, http://www.sabah.com.tr/
Etik_Rehberi/2011/01/07/kacaklari_ihbar_etmemek_etik_mi, eingesehen 09 August 2012.
―― (2012a), “Başbakan Erdoğan konuştu”, Sabah 29 April 2012, http://www.sabah.com.tr/Gun
dem/2012/04/29/basbakan-erdogan-konusuyor, accessed 09 August 2012.
―― (2012b), “Hayırcı senatörlere mektup”, Sabah 27 January 2012, http://www.sabah.com.tr/Gun
dem/2012/01/27/hayirci-senatorlere-mektup, accessed 09 August 2012.
Seufert, G. (1997), Politischer Islam in der Türkei. Islamismus als symbolische Repräsentation einer
sich modernisierenden muslimischen Gesellschaft, Istanbul.
Standen, N. (1999a), “Nine Case Studies of Premodern Frontiers”, Frontiers in Question. Eurasian
Borderlands, 700 – 1700, Basingstoke/New York, London: 13–27.
―― (1999b), “(Re)Constructiong the Frontiers of tenth-century North China”, Frontiers in Ques-
tion. Eurasian Borderlands, 700 – 1700, Basingstoke/New York, London: 55–79.
Stüver, A. (2001), “Transnationale Arbeitsmärkte in der deutsch-niederländischen Grenzregion?
Hemmnisse der grenzüberschreitenden Integration”, Europäische Grenzregionen—Brü-
ckenköpfe für die Integration regionaler Arbeitsmärkte?, Aachen: 143–163.
Şafak, E. (2009a), “ Ermeni açılımına destek”, Sabah 19 October 2009, http://www.sabah.com.tr/
Gundem/2009/10/19/ermeni_acilimina_destek, accessed 09 August 2012.
―― (2009b), “Barış yolu”, Sabah 07 September 2009, http://www.sabah.com.tr/Gundem/2009/09/
07/baris_yolu, accessed 14 August 2012.
―― (2010), “Kısır döngüyü kırmak”, Sabah 04 March 2010, http://www.sabah.com.tr/Yazarlar/
safak/2010/03/04/kisir_donguyu_kirmak, accessed 09 August 2012.
―― (2012), “Prangadan kurtulmak”, Sabah 03 March 2005, http://www.sabah.com.tr/Yazarlar/
safak/2005/03/03/Prangadan_kurtulmak, accessed 09 August 2012.
Taşpınar, Ö. (2009a), “Cesur adımlar zamanı”, Sabah 07 September 2009, http://www.sabah.com.tr/
Yazarlar/taspinar/2009/09/07/cesur_adimlar_zamani, accessed 14 August 2012.
―― (2009b), “Protokolleri komadan çıkarma formülü”, Sabah 28 December 2009, http://www.
sabah.com.tr/Yazarlar/taspinar/2009/12/28/protokolleri_komadan_cikarma_formulu,
accessed 09 August 2012.
550 Pascal Kluge

―― (2010a), “Ermeni lobisi çok memnun”, Sabah 22 March 2010, http://www.sabah.com.tr/Yazar

lar/taspinar/2010/03/22/ermeni_lobisi_cok_memnun, accessed 09 August 2012.
―― (2010b), “Kazanan Azeri ve Ermeni lobisi oldu”, Sabah 08 March 2010, http://www.sabah.com.
tr/Yazarlar/taspinar/2010/03/08/kazanan_azeri_ve_ermeni_lobisi_oldu, accessed 09 Au-
gust 2012.
―― (2011), “2015’e doğru Ermeni meselesi”, Sabah 26 December 2011, http://www.sabah.com.tr/
Yazarlar/taspinar/2011/12/26/2015e-dogru-ermeni-meselesi, accessed 09 August 2012.
Tınç, F. (2010a), “24 Nisan’ı daha geride bırakamazken”, Hürriyet 26 April 2010, http://hurarsiv.hur
riyet.com.tr/goster/ShowNew.aspx?id=14528443, accessed 08 August 2012.
―― (2010b), “Doğru bir geri adım”, Hürriyet 05 April 2010, http://hurarsiv.hurriyet.com.tr/goster/
ShowNew.aspx?id=14317781, accessed 08 August 2012.
Tocci, N. (2007), The Closed Armenia-Turkey Border: Economic and Social Effects, Including those on
the People; and Implications for the Overall Situation of the Region, Brussels, 2007, http://
www.armenews.com/IMG/armenia-turkey.pdf, accessed 23 September 2012.
Uluç, H. (2010), “Türkiye’nin yolunu 45 Amerikalı mı çizecek?..”, Sabah 06 March 2010, http://www.
sabah.com.tr/Yazarlar/uluc/2010/03/06/turkiyenin_yolunu_45_amerikali_mi_cizecek,
accessed 09 August 2012.
Weichhart, P. (1999), “Territorialität, Identität und Grenzerfahrung”, Grenze im Kopf: Beiträge zur
Geschichte der Grenze in Ostmitteleuropa, Frankfurt, Main, Wien: 19–30.
Yılmaz, M. Y. (2010), “Endişeye gerek yok zaten öyle olacak!”, Hürriyet 15 December 2010, http://hur
arsiv.hurriyet.com.tr/goster/ShowNew.aspx?id=16535252, accessed 03 August 2012.
―― (2011), “Önemli ama yarım bir adım”, Hürriyet 03 September 2011, http://hurarsiv.hurriyet.
com.tr/goster/ShowNew.aspx?id=18637876, accessed 03 August 2012.
 On the Causes of Socialism’s Deconstruction:
Conventional Debates and Popular Rhetoric in
Contemporary Kazakhstan and Mongolia*

Irina Morozova
Humboldt University, Berlin

Abstract
The current article deals with the production of knowledge on late socialism and its dismantling in
Soviet Central Asia (Kazakhstan) and Mongolia. Starting with deconstructing conventional de-
bates on the nature of Soviet socialism in international and national historiographies it bridges
them with the popular narratives on the causes of socialism’s dismantling. Based on oral sources
mainly recorded in the genres of oral history and life story the research shows the gradual re-
placement of the quest for social equality by nation building and ethnicity narratives in Kazakh-
stan and Mongolia in the 1980-90s and after 2000. Pointing out the complexity of social composi-
tion in the two republics the article attempts to compare the Kazakhstani and Mongolia’s elite’s
status vis-à-vis Moscow and the population’s attitudes towards it that formed the course of the pe-
restroika reform. The analysis of the main popular narratives shows their long-lasting acuteness
and potential for reproduction not only in the republics under study, but also internationally. The
article underlines interconnectivity of the discursive fields in the West and in the second world’s
peripheries such as Central Asia, calling for multidisciplinary approach and contextualisation in
regional studies.

Keywords
Socialism, post-socialism, perestroika, comparative history, oral history, memory and narrative,
nomenklatura, intelligentsia, Kazakhstan, Mongolia

This research aims to illuminate the phenomenon of socio-political transfor-

mation under perestroika in a light of societal and cultural transformation. As
most of the scare research on perestroika in Central Asia was devoted to the stra-
tegic and tactical alliances and pacts between different groups striving for power
throughout the 1980s and reaffirmed after the disintegration of the USSR, practi-
cally nothing has been written on the societal background against which this

*
The author is grateful to Volkswagen Foundation for sponsoring this research.
 Koninklijke Brill NV, Leiden, 2015
552 Irina Morozova

political change occurred. The persistence of conventional approaches to the

history of socialist societies, as well as methodological difficulties of analysing
and combining various written and oral sources, have complicated the endeav-
ours to write the history of late socialism. This article consists of two parts: the
first discusses the scholarly production of knowledge, outlining which ap-
proaches to the studies of perestroika have become mainstreams in Western and
Soviet (as well as post-Soviet) historiographies and how they influenced each
other. The second part presents the analysis of different social groups’ percep-
tions about the causes for the fall of socialism. Thus, literature studies, discourse
analysis and the scrutiny of the interviews’ data are integral parts of the study,
which deems to show how certain conventional debates and rhetoric echoed
both in scholarship and in popular public domain and how these spheres have
been interconnected and developing in the shared historical space.
The comparative component of this study reflects the author’s aim to pursue
a multi-perspective view on society and reflect above all on the conceptual ap-
paratus. The cases of Kazakhstan and Mongolia were chosen for comparison,
since their histories have a number of distinctly intercrossing points in modern
times. The peoples of the region “intercrossed” through invasion, migration and
cultural exchange (there are still enclaves of Mongolian peoples in present Ka-
zakhstan, and there is the whole aimag [Mong. “province”] of ethnic Kazakh in-
habitants in Mongolia). The twenty first century socialist development of the so-
cieties under study forms the solid ground for a comprehensive comparison. Par-
ticularly illuminating is the comparison between the Soviet Republic and de-jure
independent Mongolia (the Mongolian People’s Republic, MPR), as it sheds light
on the correlation between international status, as well as the status of the elites
vis-à-vis Moscow, and social transformation. Such angle of compassion is missed
if comparison is narrowed to the realm of the Soviet Republics. The particulari-
ties of present developments in the societies under study also help to reflect on
and compare their perestroika histories.

1. APPROACHES TO THE STUDIES ON PERESTROIKA IN CENTRAL ASIA AND THE HALL OF

MIRRORS IN HISTORIOGRAPHIES
Most of literature on perestroika has been devoted to the political struggle in the
CPSU Politburo and especially to the personal contribution to perestroika of its
initiator Michael Gorbachev (Brown 2007). Quite a few works on property rela-
tions and co-operative enterprise-building in the USSR in the 1980s were pub-
 On the Causes of Socialism’s Deconstruction 553

lished; some of them convincingly depicting how the division and transfer of
state-owned property into private stocks had been carried out by a few interest
groups in the government and party bureaucracy, as well as the new economic
and political elite (Cox 1996; Nelson/Kuzes 1994). Most authors agree that the
radical change bringing reform at the end of the 1980s occurred as a reaction to
the perceived crisis and perestroika was a set of reforms planned by the Soviet
leaders in response to the accumulated corruption and ineffectiveness of the So-
viet economy and a desired shift towards more open relations with the Western
world. Albeit quite recent works on the post-socialist economic and social re-
form emphasise the connection between globalisation and the course of late so-
cialism, they regretfully lack comprehensive quantitative and qualitative data to
trace the socio-economic transformations of the Soviet Union in the 1980s (see,
e.g., Rudra 2011: 205-206).
In general, there have been three approaches to understanding the social na-
ture of the reforms under the perestroika (and later privatisation) slogan. All of
them are derived from the neo-classical, positivist definition of power and elite
groups as those who control or participate directly in wealth accumulation and
redistribution in the society. Two approaches were based on conceptualising
capitalism as a world system. One emphasised the malfunction of the Soviet cen-
tralised economy and consequently interpreted perestroika as a logical step in
the evolution of the socialist “deviation” towards a market economy (Kornai
1992), even if some specific “national” features of that newly arisen capitalist
form were recognised (Owen 1995). Another was developed by Western Marx-
ists, who focused on class conflicts in Soviet society, arguing that by the 1980s the
ruling Soviet elite faced systemic difficulty in appropriating surplus products
while providing for social benefits to the working class, so that, by the end of the
1980s, it had nothing to do but to find a “legal” solution to resume control of the
State property by the full-scale restoration of capitalism (Ticktin 1992; Filtzer
1994). Beyond those two opposite approaches, some scholars refuted that the so-
cio-political course of perestroika demonstrated any clear logic or a programme,
which the ruling elite might possibly have had and raised the problem of the
complex social composition of late Soviet society (Cox 1996: 136-137). Neverthe-
less, research on social networks and interconnections between party agents at
all levels, economic ministerial branches and other departments and the con-
flicts existing among them have remained underdeveloped, especially in the case
studies of the former Soviet Central Asian Republics.
554 Irina Morozova

While the majority of the former Soviet sociologists and historians, who used
to revert to the Marxist School to develop their theoretical approaches, formed
their view of perestroika as the commencement of the liberation of Soviet soci-
ety, its “humanisation” and first steps towards “normal development” of social-
ism, free from the conditions of “command-administrative system”1 and one-
party rule (Antonovich 1990); others, repeating (consciously or unconsciously)
their leftist colleagues from the other side of the Iron Curtain, underlined the
economic interests of the upper echelons of the nomenklatura class as the main
driving force behind perestroika reform (Fursov 2006). According to the latter,
perestroika was a systemic crisis in the Brezhnev model of socialism, the policy
planned by the nomenklatura to preserve its privileged position in society and
guarantee the existing system of exploitation by legalising its property rights.
However, such an interpretation was undermined by some quantitative socio-
logical research, demonstrating the relatively insignificant percentage of the
former nomenklatura bosses among the new large-scale private business owners
at the beginning of the 1990s (Kryshtanovskaya 1991). The recently published
works on the radical socio-economic reform of the early 1990s criticise the esti-
mates of the late perestroika economists and reformers by pointing out essential
gaps in the data, on which they relied to set up the economic programme, and
the numerous mistakes (see, e.g., Fridman/Belchuk 2007). In agreement with
some of their colleagues in the West, these authors in Russia claim that the late
perestroika reforms were carried out precipitately without sufficient planning or
vision of the future. They were executed out of necessity in efforts to find an
immediate response to the economic crisis.
The chronology and dynamics of the 1980s’ developments within the Soviet
society in relation to the international change is also understudied. Although the
events were obviously developing with the accelerated speed and culminated in
1989-1991, the actual shift from one social model to another can hardly be de-
tected for these years. At the same time, some works on the crucial year 1989 and
its international effect have appeared lately: some of them sum up the already
known open and journalist sources,2 others present more comprehensive collec-
tions of documents (Florath 2011), while a few have a more specific focus on a
particular public sphere that underwent fundamental change (Koschovke 2009).

1
On the construction of “the command-administrative system of governance”, see Kirchik
2007.
2
Editors of Time Magazine. TIME 1989: The Year that Defined Today's World (Times: 2009)
 On the Causes of Socialism’s Deconstruction 555

Other projects have given critique to the common vision of the importance of
historical dates and argued that the year 1990 was not less but, perhaps, even
more vital for the social transformation of the late socialist societies.3
In the recent studies on the twentieth-century history of Central Asia, little
attention has so far been paid to the late Soviet period (from 1982 to 1991), alt-
hough the crucial change, which was ushered into the Soviet Central Asian soci-
eties with the penetration of new trends from the Western world and the East in
that period has been widely acknowledged. Not just the official historiographers
of the newly independent states viewed the whole socialist period as a deviation
from the normal development of their nations, this vision prevailed in the writ-
ings by many prominent Western scholars and was inherited by the researchers
re-conceptualising and rewriting the Soviet history of Central Asia.
The debates on social and socio-economic developments of late socialism
have been practically completely overlooked by the researchers in Central Asian
studies. No solid works on economies of Central Asian Republics are known.
Basing their research on the available Soviet sources (central and local press),
the Western academia enthusiastically supported the accent on ethnicity in the
studies of the Soviet Central Asia. Any research on perestroika in Central Asia
has been grounded in discussions on nationalism. The idea generally accepted in
the 1990s, to which such renowned scholars as H. Carrere D’Encausse, A. A. Ben-
ningsen and S. E. Wimbush contributed, was that perestroika released the sup-
pressed national feelings and identities, allowing them to rise to the surface and
predominate in political and public life.
In media and political domains alarmist interpretations of the developments
in Soviet Central Asia were given by Western scholars at the last stage of pere-
stroika, many talked about “ethnic strife”, which would cause political instability
and long-term socio-economic consequences for Europe (Hyde-Price 1991: 67-
88), when it would face migrants from those parts of the continent. Those pan-
icky authors relied solely on Gorbachev’s speeches on the dangerous potential of
the nationalist contradictions within the Soviet Union and on the publications
by the supporters of Gorbachev’s course, who positioned the policies of pere-
stroika in such a way, as if its failure would jeopardise the hopes for peace in Eu-
rope and the world (see, e.g., Karaganov 1990).

3
See, for instance, the collection of articles in Novoe literaturnoe obozrenie 83, 2007.
556 Irina Morozova

The voices of Western scholars acquired among Central Asian intellectuals

special meanings of truth, previously unknown and hidden, but suddenly re-dis-
covered correct interpretations of the region’s history and contemporary devel-
opment.4 These scholars promoted the vision of the “formerly oppressed ethnic,
national and religious feelings of the Soviet Muslims” (Hostler 1993; Paksoy 1993;
Rywkin 1982). In most works by them, Islam (except for the official Islam) or
Buddhism were seen as brutally oppressed religions, and Muslim or Buddhist in-
tellectuals as potential rebels against the Soviet/socialist power. Those Cold War
times’ interpretations of Islam and Buddhism were at their peak during pere-
stroika and had an utterly deep effect on the local discourse in Central Asia, in-
cluding the academic discourse.

2. ORAL SOURCES AND AN APPROACH TO INTERPRETING THE DATA

This study is based upon the materials of twenty interviews (recorded as life sto-
ries narratives, sometimes with minimal interference of the interviewer, some-
times with provoking questions imposed by the interviewer) taken in each coun-
try and three focus-groups in Almaty and one in Shymkent in Kazakhstan and
two focus-groups in Mongod sum (Mong. “smaller administrative unit in aimag”)
of Bulgan aimag in Mongolia. The interviews were taken in Kazakh, Russian and
Mongolian languages by the author of this article and her project partners in Ka-
zakhstan and Mongolia.5 The long-term fieldwork enables contextualisation and
verification of the recorded narratives.
The question “why socialism was dismantled?” (very different from “how was
it done?”) is not a mere scholar’s construction, it was crystallised in the process
of working with oral data. The question resurfaced continuously and many peo-
ple formulated it themselves: why at certain point their individual life and the
path of their community, generation, group went according to one particular
way, whether or not that way was optimal and what it has brought to them at
present. The present dissatisfaction with life (in all possible social or personal
aspects) is an essential motivation for people’s formulating the question “why”
and “when” something went wrong. At the same time, the question on socialism

4
On the apologetic promotion of the liberal ideology of free market and democracy by Central
Asian scientists and intellectuals (on the example of Kyrgyzstan), see Amsler 2009.
5
The participants of the project in Kazakhstan were Dr. Tolganai Umbetalieva, Ms. Saltanat
Orazbekova and Dr. Andrei Chebotarev; in Mongolia, Prof. Jigjidijn Boldbaatar and Mr. Enkhbaatar
Munkhsaruul.
 On the Causes of Socialism’s Deconstruction 557

deconstruction has been so many times discussed and propagated in public de-
bates and recreated in people’s conversations and everyday communications,
that the gap between the pure narrative and the reflection ad hoc became ex-
tremely wide.
Our analysis of the collected data shows that no distinct lines existed among
population groups. There have been various small groups and sub-groups
formed along professional, cultural and generational lines in the late socialist so-
ciety. Although scholars’ attention has been given to the reinvented concepts of
ethnicity, nationalism and religion, other sub-cultures in Central Asia―like
criminal and prison societies or youth cultures, including musical sub-cul-
tures―have not been given attention in literature so far.
In determining social groups, while analysing the collected oral data, certain
oppositions were fixated. The first opposition underlines the difference in per-
ception by the intelligentsia (mainly urban), on one side, and peasants (common
members of collective farms), as well as workers employed in the industries, on
the other. The main distinction is exactly the perceptivity to the imposed ideo-
logical debates (as one about the identity of the Soviet people, for instance) by
well-educated individuals and the less educated and often depoliticised people
with stronger local communal loyalties. This dichotomy becomes especially clear
at Brezhnev’s times and sharpens during perestroika.
This opposition is important to bear in mind, since practically all the dis-
courses on perestroika change are a continuation of the intelligentsia’s narratives.
The Soviet official discourse on perestroika was elitist: it was determined and
shaped by a circle of representatives of the liberal wing of the nomenklatura in
co-operation with and due to legitimisation of their course by intellectuals and
the intelligentsia. Many of them belonged to the 1960s’ generation, the so-called
shestidesyatniki. Since they were in a certain way a privileged strata, since they
had some access to the outside world (via international socialist institutions,
communist parties’ networks and numerous socialist cultural festivals and other
international structures and events) and potentially were interested in broad-
ening contacts with the Western public, later, when that was achieved, many in
the West learnt to perceive the history of perestroika through their eyes. Our re-
spondents among the shestidesyatniki usually speak as if they always believed in
their mission to change the socio-political system. Despite the statements these
people make at present (as if they anticipated the dismantling of socialism and
the USSR’s disintegration), in the 1960-80s they did not constitute any kind of
558 Irina Morozova

counter-elite that fought against the Soviet state, but tried to broaden their in-
volvement into political and public life and suggest such socio-cultural ideas that
would anticipate the regular course of the CPSU by the prospective next group to
come to power within the Politburo.6 Moreover, the necessity to preserve the pro-
fessional environment called for a certain amount of mimicry (Firsov 1997: 138-
139) that remained as a norm of adaptation after the dismantling of socialism.
The second opposition is to a great extent a stereotype highlighted by the lib-
eral Soviet intelligentsia and especially by the shestidesyatniki: “liberal intellectu-
als versus the conservative Soviet nomenklatura”. However, a mere analysis of
biographies of political leaders of the Newly Independent States shows close and
complex interconnectivity between the Soviet intellectuals and the top nomen-
klatura. The contribution of academicians, especially in the field of humanities,
as well as writers and artists in defining perestroika “new thinking” and in deter-
mining the course of reforms was exceptionally noteworthy in Soviet Central
Asian Republics and the MPR.
Education and knowledge had been important among Central Asian elites
well before the socialist revolution; and under the Soviets university education
became a matter of social prestige for wider sections of the population. In the
KazSSR the rhetoric of national awakening had its roots in the 1924 delimitation
(and even earlier). In 1946, the establishment of the Academy of Science of the
KazSSR was accompanied with the launch of fundamental projects on writing
the history of the Republic. The national liberation rhetoric was used practically
unanimously by Kazakh historians, who had to follow the mainstream of the So-
viet social science. Academicians and specialists from the European part of the
country were dominant in hard science and technical spheres,7 while titular
scientists were predominantly employed in humanities departments.8 Lucrative

6
One of the principal manifestations of the new co-operative relationships of the perestroika
group in Moscow and intelligentsia in Central Asian Republics was the Issyk-Kul Forum held in Oc-
tober 1986 by Chingiz Aitmatov. The writer, encouraged by M. Gorbachev, invited Alexander King,
Arthur Miller, James Baldwin, Peter Ustinov, Claude Simon, Alvin Toffter, Federico Mayor and oth-
ers to participate in the event that had to set up the perestroika “new thinking”.
7
In the 1920s and 1930s, Russians and Jews constituted the overwhelming majority of the
technicians and trained professionals, such as agronomists, surveyors, veterinarians, and doctors;
they were in direct contact with the indigenous population (RGASPI. F. 62. Sch. 4. D. 1, 13, 243; see
also Martin 2001: 140-141).
8
On the specificities of the academia and educational system in the USSR Republics, see Lane
1971: 93-95; Il’in 1996: Chapter 8.
 On the Causes of Socialism’s Deconstruction 559

positions in the academia and the opportunity to work on national history or na-
tional literature and form their own research groups, albeit within the Soviet
mainstream, gave those people additional motivation to co-operate with the
state and party authorities, rather than to oppose them in an open or hidden
way. An academic career in the Academy of Science was considered to be elitist
and granted a person great social prestige comparable to the social cult of the
families of saints9 or belonging to a noble lineage.
Although the new Kazakhstani intelligentsia developed in unison with the
mainstream coming from research institutions in Moscow, there was some room
for deviation. As those intellectuals and the upper layers of nomenklatura cadres
often shared the same educational background and had been exposed to the
same ideological propaganda, transition from one group to another was rela-
tively easy and typical. Quite frequently the representatives of the intelligentsia
were appointed at high nomenklatura positions or officials joined the academia
after retirement.10
The social cult of knowledge, scriptural knowledge and education as a proof
of belonging to the upper strata of community and as a legitimisation of the sta-
tus of spiritual teacher was also noteworthy in the MPR. Perhaps due to the ex-
treme underpopulation, the ties between academia and nomenklatura in this
country were even closer than in Soviet Kazakhstan. Till recently it has been ar-
ticulated in history books that the progressive perestroika change (towards de-
mocracy) came to Mongolia via the young national intelligentsia of the 1980s:
sculptors, painters, writers and journalists, who formed their own social net-
works of urban intelligentsia and held informal gatherings,11 which in 1989
evolved into the first democratic units. The story of the young freethinking intel-
ligentsia’s evolution into the Democratic Party has become principally important
for the new political and intellectual elites in Mongolia.12

9
Some scholars describe how the descendants of the urban families of saints (among the Uz-
bek and Tajik) became recognised scientists, preserving due to that lineage some kind of intellec-
tual if not to say spiritual authority in the eyes of the people (see, e.g., Abashin 2007: 223-228; Mu-
minov 2011).
10
See, for instance, the research on the biography of the famous Russian politician and Orien-
talist E. Primokov: Roshchin 2011: 103-111.
11
Author’s interview with Dr. Hulan Hashbat, Ulaanbaatar, May 2008.
12
Many works by Mongolian and Western authors have developed this vision, see, for instance,
Rossabi 2005; Boldbaatar 2007.
560 Irina Morozova

In evaluating the impact of intelligentsia’s narrative on perestroika upon the

narratives of other groups of the population one should take into consideration
that the improvement of living conditions and the broadening of the sphere of
high education in the 1960-70s went in parallel with the development of the So-
viet state and the party institutions, which worked positively for the new social
identities and for the wider involvement of educated national cadres into the
public life of the Republics. This led to the situation, when during the perestroika
years the people, who were not professionally involved in any type of intellectual
work and education, but were provided with good secondary and high educa-
tion, also started to identify themselves as “intelligentsia” and later claimed that
they had been in opposition to the existing political regime and had been long-
ing to reforms in the Soviet state. Practically all interviewed by us Kazakh intel-
lectuals and politicians made the same statement about their reactions to the
December events in Almaty in 1986:13 they claimed that at that particular mo-
ment it had become clear to them that Soviet state would soon stop existing.
This claim contradicts the actual events that followed (Umbetalieva 2012: 14).
Such unanimity of reactions tells much more about the people’s desire to create
a myth about their missionary role for the purpose of the present, than about ac-
tual events in the past.
Another phenomenon that requires special attention, when depicting social
history of the 1980s, is the generational change and its role in social and ideo-
logical transformation. The generational composition of the MPR’s society, in
this regard, represents a very special case, for we see the country’s extreme un-
derpopulation and the very rapid generation change. As rightfully pointed out by
Udo Barkmann, in 1985-1990, the new young generation (people of the average
age between 15 and 39 accounted for 48,24% of the whole population) (Bark-
mann 2006: 315-317) filled in the middle layers of the nomenklatura and got the
momentum to appear at the forefront of the political arena at the end of 1989-
beginning of 1990. The quantitative data very well corresponds to the qualitative
data extrapolated from our interviews. “I was young that time and the goals of

13
The December events—demonstration of protest in Almaty in December 1986 with the fol-
lowed repressions performed by security organs in major Kazakhstani cities—had the pretext of
dismissing the long 25-years’ stay first secretary of the CC of KPKaz Dinmukhamed Kunayev from
his post and appointing a non-Kazakh and non-Kazakhstani representative of provincial Soviet
nomenklatura Gennady Kolbin.
 On the Causes of Socialism’s Deconstruction 561

those new democratic forces seemed good to me and I became their member”,14
tells a member of the Democratic Party in the city of Khovd in Western Mongo-
lia, while remembering the December of 1989, when he joined the first demon-
stration for democratic change at the city’s central square. That generation
claimed a wider political and public space for themselves to fill in positions in
the officialdom and since the end of 1988 promoted the idea of glasnost in Mon-
golia.
The first democrats’ generation in Mongolia capitalised on their generational
trauma by calling for reconceptualising collective memories of the recent so-
cialist past and emphasising their belonging to the families of the prosecuted
(Enkhbaatar 2011). Nevertheless, not only democrats, but the Mongolian People’s
Revolutionary Party (MPRP) members as well like to mention that they come
from the families of the prosecuted.15 Whether these genealogies are real or
imagined, the moment when they are spoken out in a conversation is never ran-
dom, as people tend to exploit their autobiographies.
At thickened historical times generational change happens with an acceler-
ated speed. The recorded interviews with the representatives of political and in-
tellectual elite in Kazakhstan and Mongolia show great variations in their politi-
cal views and social orientations, which depend on which year of perestroika
their formation as a public activist occurred: whether the person already finished
the University by 1982, when Brezhnev died and Andropov came to replace him,
whether he/she entered it before 1985, when Gorbachev first declared the pere-
stroika course, or whether he/she became a student at the second part of 1989 or
was writing his/her diploma “under the accompaniment of the First Congress of
People’s Deputies”.16 The pace of change during perestroika (especially after 1987)
makes it difficult to determine and categorize one generation as “perestroika
generation” with its typical collective loyalties and perceptions.
Another substantial methodological difficulty lies in evaluating the degree of
social polarisation in late socialist society. Our analysis of oral history interviews
and focus groups demonstrates that inequalities were perceived more vigorously
at local level, where communality was stronger and the opportunity for vertical
mobility lesser. Although the respondents of the focus-groups argue that people
were in principle equal and in everyday routine the neighbours, having different

14
Authors’ interviews. Khovd, Mongolia, June 2008.
15
Author’s interviews. Ulaanbaatar, Mongolia, May 2008.
16
Author’s interviews. Almaty, June 2011.
562 Irina Morozova

social status still dined together in one kitchen and let their children play to-
gether in the streets, in personal conversations they talk for social stratification
on family background. During two focus-groups in Mongod sum participants
were remembering how some poor pupils could not afford apples, which the rest
of the class eat, and the child of the head of the sum administration enjoyed
sweets, which no one else had.17 “We hated them and did not mix with them”,—
was a typical notion by people in Shymkent, who were students from rural non-
nomenklatura families, about their co-students of urban nomenklatura back-
ground. Especially sharp this opposition appears for women, as their sphere of
sociologisation was often narrower than the one of young men.
These notions reveal the already elaborated rural-urban social opposition
connected to post-WWII industrialisation projects and the migration of rural
population into the cities, which was not adequately met by the development of
the socialist economy and the social sector, causing the appearance of new social
inequalities and tensions. In case of Kazakhstan this polarisation seems to be
stronger18 than in Mongolia, where, on the one hand, a better vertical mobility
occurred to be possible since the 1960s, and on the other hand, greater collective
respect was given to a local community’s representative, who managed to be-
come famous in the capital or nationwide.19 Thus, in the above-mentioned Mon-
god sum, as well as in Myangad sum of the Khovd aimag, the portraits of well-
known figures (like sportsmen, for instance), originating from the sum, still
adorn the walls of secondary schools and local museums.20
The newly introduced system of leasing cattle (arendnyi podryad, Rus.) was
also practised in the MPR at the second part of the 1980s, although the aimag
leaders learnt about this principle in the Soviet Union, for instance, in Burya-
tiya.21 There was a distinction between more economically successful families of
herdsmen that next to the collective cattle accumulated more private livestock
and those families that could rely solely on the socialist social system, lacking
private herds and being sometimes forced to migrate from a poorer to a richer

17
Author’s field work in Mongod sum, Bulgan aimag, Mongolia, August 2012.
18
Author’s field work in aul near Taraz, Kazakhstan, July 2011.
19
Author’s field work in Myangad sum Khovd aimag, Mongolia, June 2008.
20
Ibid.
21
Interview with the former chairman of the cooperative (negdel) and the former second secre-
tary of the MPRP committee of Ölsijt sum of Dundgovi aimag, recorded by Enkhbaatar Munkhsa-
ruul, Ulaanbaatar, May 2011.
 On the Causes of Socialism’s Deconstruction 563

collective (negdel, Mong.) in another aimag. On the other hand, the socialist re-
distribution system balanced this type of polarisation of rural nomadic popula-
tion.
Certain stratification can be noted even within the ulama in Soviet Central
Asia. Despite of the well-known fact that Soviet Muslims were performing hadj
to Mecca since the Khrushchev Thaw, pilgrimage trips were only available for
carefully selected representatives of the official administration of the Soviet
Muslims of Central Asia, better known in its Russian abbreviation SADUM. At
local level paid pilgrimage trips could have been offered (after 1987, when reli-
gious ceremonies became officially recognised and allowed by the Soviet state),
which kept Muslims, who did not have the means to pay for their hadj rather dis-
satisfied.22
The most complicated and intriguing dilemma connected to the social po-
larisation in the late socialist society is the emergence of the new layer of middle
and high-scale entrepreneurs. This problematic was touched upon in the studies
of post-socialist migration (describing, e.g., the phenomenon of ‘chelnok’). How-
ever, studies on Central Asian economies which would be truly support our un-
derstanding of the social background and evolution of entrepreneurs are practi-
cally absent nowadays23 and the history of privatisation has not been given ade-
quate research attention. Data on the process of the entrepreneurs’ crystallisa-
tion as a social group is hard to obtain in Mongolian and particularly Kazakh-
stani context. Even if the interviewed people are open in their political commen-
tary, they rarely tell about their own or their acquaintances’ success (or misfor-
tunes) in business. It would be logical, nevertheless, to assume that those people,
either on individual or collective ground, must have supported the systemic
change in their countries. The respondents point (in most cases indirectly) at lo-
cal ministerial top officials and directors of collective farms and factories as
those who had accumulated administrative power and social capital (including
networks) and welcomed the privatisation that helped them to accumulate
quickly actual capital.

22
Author’s interviews with local mullahs. Shymkent, Turkistan, Kazakhstan, June-July 2011.
23
Only the works by Richard Pomfret and Leonid Friedman could be mentioned in this regard.
564 Irina Morozova

3. THE CAUSES FOR DISMANTLING SOCIALISM IN PEOPLE’S MEMORIES AND NOWADAYS’

PERCEPTIONS
3.1. The “International Assistance” Narrative
As people rarely distinguish between the processes of the USSR’s disintegration
and the socialism’s deconstruction, when reflecting upon their own life or when
confronted with the question “why do you think socialism stopped to exist and
the USSR was disintegrated?”, most respondents talk on the USSR’s collapse as a
cause for the fall of socialism. In both Mongolia and Kazakhstan people attempt
to contextualise the systemic change within wider international and historical
trends. If in Kazakhstan the majority of the interviewed common people said
that “the USSR had been assisted from outside to disintegrate”, in Mongolia, per-
haps due to its independent status at the international arena, more affirmative
concepts of Mongolia’s statehood in the twentieth century were pronounced.
The way the people in Kazakhstan tend to believe that despite any possible
systemic problems of socialism it was external force (the U.S.A. or Western in-
telligence agencies or a clique within the international capitalist system) that
brought the USSR to disintegration—might mean that they are still reluctant to
fully accept the fall of socialism. The respondents’ replies usually go within the
following lines: “the USSR was such a great power… and then disintegrated so
suddenly and quickly”, while no reason that they can possibly name seems to
them sufficient enough. People from the southern regions of Kazakhstan, where
socio-economic decline at the beginning of the 1990s was especially devastating,
realised that the Soviet Union totally had gone into the past only by the end of
the 1990s. At the end of the 1980s, many people had expected the perestroika
course to give way to another new course of the CC of the CPSU, and through the
1990s they had been waiting for the Union to be re-established.
When someone, nevertheless, tries to search for the explanation how the im-
agined external forces could have managed to disintegrate the country so seem-
ingly powerful, she or he either connects it with the betrayal of the ruling Soviet
elite and perestroika leaders or with the gradual spread of Western (predomi-
nantly American) culture through smuggled videos, rumours, radio (the Voice of
America), literature (samizdat) and fashion (the bizarre fashion for jeans, for ex-
ample), which seemed to be the more attractive the less it was accessible. Com-
mon people themselves contributed to and reproduced the image of the Ameri-
can life style, which substituted a sort of naïve dream for a better world that was
deliberately hidden from them. In this way “the American match fell upon the
 On the Causes of Socialism’s Deconstruction 565

lulled vigilance”, as Almaty-based psychologist Zhanat Smirnova allegorically

points out.24
“I ask myself now: why we respected them [the perceived Western capitalist
mass societies] so much those years…”―speculates Tatyana, a designer and
seamstress from Shymkent. Interestingly, Tatyana cannot acknowledge any real
influence those images of Western culture had upon her life, in fact, having three
children to take care of alone, she hardly had any time to read, to watch televi-
sion or to search for prohibited radio stations. She picked up the notion of “re-
spect” for Western culture in everyday conversations and gossips. Tatyana tends
to see a conscious effort behind this popular tendency rather than a reproduced
imaginary reality, while for the other interviewed people reminiscences about a
garment brought from abroad (iz-za granitsy, Rus.) still retain social actuality.
The “international factor” in the dismantling of the MPR’s socialism is peculi-
arly twisted round by the articulated intention to see the country’s independent
performance at the international arena. The nation-building policies of the pol-
icy-makers of the beginning of the 1990s in many ways hit their target: the rec-
orded life story narratives by Mongolian citizens remind more of a chronologi-
cally correct story of the development of the Mongolian nation with its roots in
Genghis Khan's efforts to unite the scatted Mongolian tribes. The ideas of Mon-
golia’s openness to international trends and eagerness “to develop according to
the world civilisation paths” are pronounced and interpreted as the main rea-
sons for the systemic change of 1989-1991.
When confronted with the question on the causes of socialism deconstruc-
tion in their country, Mongolian intellectuals tend to prefer to imagine the Mon-
gols of the MPR as victims. In their interpretations, socialism is sometimes asso-
ciated with repressions of the 1930-40s and believed to have happened not ac-
cording to their own choice, but because of the pressure of the Kremlin. At the
end of 1989 several campaigns of rewriting recent socialist history took a new
breath in the MPR (against the background of Gorbachev’s official recognitions
and apologies for the former USSR’s repressive and inhumane policies in Europe
before and after the WWII). History was rewritten in such a way that the USSR
and its communist leaders occurred to be guilty for the repressions of the 1920-
30s in Mongolia (Baabar 1990). In such a way the MPRP managed to avoid ac-
knowledging its own guilt in eliminating Buddhist monks and monasteries in the

24
Interview recorded by Tolganai Umbetalieva, Almaty, July 2011.
566 Irina Morozova

1930s. In general, the MPRP to a larger degree followed the example of the
Communist Party of China than the ideological policies of the CPSU (Brady
2005). In this regard, nation-building policies in Mongolia happened to be more
affirmative, which is reflected, above all, in the lower social deprivation index (if
compared with Russia) (Sablonnière de la; Tougas; Lortie-Lussier 2009: 327-348).
The second dominating narrative, next to the MPRP’s, is the Democratic
Party’s story (Morozova 2010). Its adherents have made their voices much better
heard internationally, and many in the West learnt to understand the systemic
change of 1989-1990 through their eyes. “Of course, the MPRP did nothing. They
only followed Russia in everything”,—says Khereed Urangua, a doctor in history
and lecturer at the National University of Mongolia.25 She tells about her eye-wit-
ness impressions of the events of 1989-1990, which are characteristic of her gen-
eration of the young MPR’s intellectuals, who came at the forefront of political
life at the end of the 1980s: “We formulated the idea that ours on the top do
nothing… and the members of the democratic movement united…” This inter-
pretation that definitely derives from her communication to students and young
colleagues at the University, who occurred to be active in anti-governmental
demonstrations in Ulaanbaatar at the end of 1989—beginning of 1990, not only
legitimises the revolutionary power transfer and the systemic change of 1990, but
provides meaning for her personal life, as she recreates the story of her personal
contribution to the reform. The selectiveness of such explanation by the young
perestroika democrats, with whom Urangua associates herself, reveals an at-
tempt to reconstruct historical knowledge and becomes even more evident,
when compared with the life story narratives by common people, who tell about
their socially secured future under socialism and positive experiences of com-
munication with the Soviet people. Against the background of the last ten-years’
neo-liberal reforms in Mongolia, which created and sharpened social polarisa-
tion, the nostalgia about socialism and “friendship with the Soviet Union” has
tended to increase.26
In the case of Mongolia, we could detect how the narrative about the reform
that belongs to a certain strata of the population, very elitist and representing
certain generational section, quickly finds itself reproduced in written from—in

25
Interview with Dr. Khereed J. Urangua taken by Enkhbaatar Munkhsaruul, Ulaanbaatar,
April 2011.
26
Author’s fieldwork. Ulaanbaatar, May 2012.
 On the Causes of Socialism’s Deconstruction 567

published memories, journalist essays, but also in new historical text-books.27

The Mongolian intellectuals occurred to be in a better position in relation to the
new elites (which came to power as a result of the 1990 change) than their coun-
terparts in Kazakhstan. (While in Kazakhstan where the call for writing a new
national history was also urgent in 1992, the intellectuals were “better off” than
their counterparts in the Russian Federation.)
3.2. The "Rotten System” Narrative
Another cause for the systemic change from socialism to capitalism, which peo-
ple note, is the constructed (and not only by elites, but by different people in
their everyday communications, as shown by Nancy Ries (1997) in “Russian
Talks”) image of “the rotten system”. Sometimes educated people in speculations
on the reasons for socialism's deconstruction start comparing the last difficult
years of the USSR’s existence, when they were loosing social and cultural orien-
tation, with the spreading of a cancer or with an old organism, which is gradually
and inevitably digressing to death. Projecting these physiological processes of
death onto social reality seems to provide for a convincing and logical explana-
tion for the deconstruction of socialism. When people are asked how they came
to the conclusion that the socialist system was that bad and non-functioning, it
turns out that they often found explanations in the newspapers already in the
1990s. The popular notions of “the command-administrative system of govern-
ance” and “totalitarian system”28 are reproduced mainly by the intelligentsia.
Among the internal reasons for the systemic change in their countries com-
mon people name careless attitudes to labour and numerous occasions of steal-
ing from the factories and collective farms. In discussions they frequently refer to
the drawbacks of socialist economies as they were publicised during the first
years of perestroika, such as “tuneyadstvo” (Rus., doing nothing and being egois-
tic), “beskhozyaistvennoe otnoshenie k sotzialisticheskoi sobstvennosti” (Rus.,
lack of the feeling of the owner in relation to the socialist property) and alcohol-
ism. The range of styles, in which the stories of stealing from the factories are
told (when elaborated specifically or mentioned randomly in the conversation)
varies from emphatic denial of such facts to cynical notions on stealing as the
real way the Soviet economy functioned. Two interviews recorded by us in
Shymkent appear to be particularly characteristic in their antagonism:

27
Khorduugaar zuuny Mongol. [The Mongols of the 20th century] (Ulaanbaatar, 1995)
28
See the memories of perestroika ideologists, for instance, Yakovlev 2005.
568 Irina Morozova

Tulgur29 (Kazakh by ethnic affiliation), who worked as a leading engineer at

the Moscow Filial of OrgProektZement (the central investigation and planning
department for the production of cement), insists that Soviet economy sup-
ported fair redistribution of social welfare and granted career opportunities to
hardworking and honest individuals. Svetlana (of mixed Kazakh and Russian
origin), who worked at Chimkentshina, the Moscow Filial of the Plant for the
Production of Tyres attached to the Ministry of NefteHimProm (oil chemical in-
dustry), tells sad stories of mass stealing from the plant. The plant’s directors, ac-
cording to Svetlana’s guess, were in cosy relationships with other directors of the
industry all over the country. The latter agreed that such a complex and techno-
logically advanced plant should be built in Shymkent and directed their profits
(the over-fulfilment of the plans) to its construction and maintenance only in
exchange for a share from the illegal distribution of tyres. Tulgur, on the con-
trary, considers the notion that the black market prevailed in the late socialist
economy to be totally misleading and views the Soviet economy as a complex in-
ter-dependent system based on scientifically approved norms that could not
have allowed huge disproportions and misbalances. “While “pripiski”, Rus. (up-
ward distortion of the fulfilled plans) did happen somewhere, I would never be-
lieve that they existed everywhere”,—says Tulgur, going into detailed explana-
tion how misbalance in one link would have immediately coursed the misbal-
ance in another link and, consequently, the malfunctioning of the whole eco-
nomic system. Svetlana, in her turn, emphasises the interconnectivity of fraud
within the centralised Soviet economy.
Svetlana’s personal motivation to belong to a higher class (pervye rukovoditeli,
Rus.,—the top executives) and occupy a higher position becomes apparent also
in her imagination how much better the plant could have been, if it had been
constructed in the Baltic Republics or Central Russia. Tulgur's life story reveals
his motivation for a better career path within the Soviet system, for instance, via
party membership, which he could not acquire. Svetlana sees social inequalities
within the Soviet system as they existed geographically, while Tulgur had
definitely stronger regional affiliations and desired career growth in Central Asia.
Tulgur’s rhetoric about the smart Soviet economy (probably learnt in his
younger years, when he, possessing a high education degree, was obliged to lec-
ture the workers) questions his actual openness to discuss the topic of “pripiski”,

29
The real names are replaced by fiction.
 On the Causes of Socialism’s Deconstruction 569

for which he has obviously more local communal rather than political con-
strains. Svetlana’s narrative that draws a picture of decay, in its turn, gives con-
cern to what extent the negative stories of mass stealing from the plant are real
or imagined post factum. The very informal atmosphere of the interview might
have encouraged her to depict the reality in the darkest way and her personal
performance within the Soviet system, although as nothing outstanding, but
nevertheless, remarkable as based on her life philosophy, of which she is very
proud. She explains her own involvement in stealing from the plant as “uncon-
scious non-resistance” to “the rotten system”, in which nothing could have been
changed, and points out that many other people stole from the plant on a much
bigger scale.
3.3. The “Weak Leaders” Narrative
Another raw of reasons for the systemic collapse includes reflections on poor
governing potential of the first secretaries of the communist parties in the USSR
and the MPR and is probably connected with the personalisation of different pe-
riods of Soviet/socialist history in people’s memories (the time of Stalin or Choi-
balsan, the time of Brezhnev or Tsedenbal and etc.) (Dadabaev 2010) “Gorbachev
was a weak leader”, “Andropov was a good chaban (herder, Kaz.)”,—those were
typical answers of provincial workers and peasants in Kazakhstani auls. Gorba-
chev is often believed to have been manipulated or even to have been a marion-
ette in the hands of external players. Sometimes, when people express sympathy
for him (usually with a reference to his young age and vitality in comparison
with the previous old and sick leaders), they attempt to justify his negative deeds
(the country’s disintegration followed by a rapid decline in living standards) as
an involuntary failure under very disadvantageous circumstances.
There is a bit lesser focus on personalities of the first leaders of the MPR, as
shown by the recorded histories. Although respondents like discussing the po-
litical struggle at the top of the Mongolian nomenklatura and in particular the
dismissal of Yu. Tsedenbal from the post of the first secretary in 1984 (with the
unhidden assistance of Moscow) and the appointment of Jambyn Batmönkh,
they tend to emphasise the collective nature of decision making (Boldbaatar
2012). Here it would be interesting to dwell upon the specifics of the MPRP in
comparison with the CPSU. The MPRP’s transformation through the whole
twentieth century in many ways determined the societal change. Especially after
1946, when the MPR acquired de-facto independence, the MPRP became a legal
actor in international politics, and its leaders played on the clashing interests of
570 Irina Morozova

the USSR and the PRC and later the COMECON countries. This important fact
distinguished the MPR’s elites’ experience from the Soviet Central Asian elites’
formation. The latter could not play at the international arena at all and in gen-
eral occurred to be localised, as they rarely received appointments at other parts
of the Soviet Union, outside their Republics.
Against the already mentioned “democratic narrative” on perestroika, various
publications witness that the MPRP adopted the perestroika rhetoric and pro-
moted it as a new way of thinking for all socialist countries (Pürevjav 1986; Dorj
1987) trying to search for alternative ways of co-operation and gaining interna-
tional aid and loans even before the economic ties with the former Soviet Union
were completely broken. In 1988, at the Fifth Congress of the MPRP CC, namyn
shinechlelt was the key topic for discussion: democratisation of public life was
believed to be achievable via party reform; it was suggested to reconsider the
Marxist-Leninist ideology and to integrate various social groups, particularly the
intelligentsia, into party-building, thus, making the party more open to the peo-
ple. The ideas and instructions to democratise public life and to rebuild the state
system were clearly pronounced and imposed from the top.30 In contrast with
the histories of other former ruling parties of post-socialist states, the MPRP took
a new breath from the Mongolian perestroika namyn shinechlelt and did not
loose all its former positions.
3.4. Do people believe in “ethnic conflict”?
For common people, whose voices are not that often heard as the intelligentsia’s
stories, it was social stratification within the late socialist societies but not the
imagined ethnic identities that became the main ground stone for memories
about perestroika. The bulk of stories, revealing extreme dissatisfaction by those,
who occurred to be socially declassified and marginalised, is immense. The ma-
jority of my respondents talked of ethnic conflict as a reason for the socialism’s
deconstruction only if I specifically asked them about. At the same time, in the
interviews taken by the Kazakh partners—respondents emotionally emphasised
social inequalities on the ground of ethnicity. In general, in the perception of
common people ethnicity is imagined and interpreted as a factor of accumulated
social deprivation and frustration that finally lead the socialist system to

30
Mongol Ardyn Khuvsgalt Namyn Töv Khoroony V Bügt Khural [The Fifth Congress of the Cen-
tral Committee of the Mongolian People’s Revolutionary Party] (Ulaanbaatar: Ulsyn hevlelijn ga-
zar, 1988)
 On the Causes of Socialism’s Deconstruction 571

collapse. More than any other narrative, these stories are closely connected with
the recent twenty years’ post-socialist history and nation-building policies of the
states. Public debates on national history and recreation of it through the prism
of ethnicisation, produced by the intelligentsia of these countries, has had an
outmost crucial effect on people’s perceptions.
The rhetoric of national independence, associating socialism with the exter-
nal order imposed top down by another nation, adds to the feeling of being ex-
ploited and manipulated. The post-imperialist rhetoric of contemporary Central
Asian historiographies depicts socialism as an alien system imposed upon local
communities from outside. The notions on acquiring independence in 1991 are
particularly characteristic for the younger generation of the titular nation (now-
adays in their thirties and forties).
At the same time, when common people speculate about ethnicity, they
demonstrate certain self-censorship: the manipulation and reproduction of eth-
nic conflicts in contemporary Central Asian states make people reluctant to talk
of ethnicity as the reason for dismantling socialism. Among the preconditions
for the USSR’s disintegration people rather see false policies or the lack of
political will in Moscow. Interestingly, many ethnic Russians in Central Asia also
perceive Russia in general and Moscow in particular as a world unfriendly to
Central Asian societies. Actual for the present, such perception is projected onto
the past. For the citizens of Mongolia, unpleasant personal experiences in the
Russian Federation format the complex of offence about the USSR’s quick with-
drawal from their country at the end of the 1980s.31
Some representatives of the Russian intelligentsia in Kazakhstan (thirty-forty
years old people) like to stress that their social perceptions go beyond any ethnic
identity; they imagine themselves as anti-communist or as passionate adherents
of the idea of world revolution or even as anarchists. Others connect social dep-
rivation with ethnicity. Many Russians in the provinces perceive the USSR’s dis-
integration (and consequently the fall of socialism) as personal tragedy.32 The
feeling of social exclusion they connect with ethnic exclusion, criticising the na-
tion-building policies of contemporary Central Asian governments.33
The highest degree of social deprivation is fixated for people, who do not re-
flect on social system or inequalities at all, but focus on ethnicity as the key rea-

31
Author’s interviews. Mongolia, Ulaanbaatar, May 2008.
32
This topic has been already researched. See, for instance, Kosmarskaya 2006.
33
Author’s interviews in Shymkent, June 2011.
572 Irina Morozova

son for their personal misfortunes. However, when confronted with the question
“how and when did you learn about your ethnicity and ethnic tensions?”, many
people say that they never thought of it during socialism and started recognising
it as a problem during perestroika.
In case of Mongolia, for which ethnic conflict has not been actual, the collec-
tive and individual identities, nevertheless, are very much based on the concept
of ethnicity and nation. The academic and public foci were given to national his-
tory of the Mongols and re-evaluation of the Mongolian civilisation’s legacies in
the world. The social cult of the writers of national history has become stronger
than in Kazakhstan, while recent international historiography trends of working
out a more critical approach to reading these histories are even less visible in
Mongolia.
For both Kazakhstan and Mongolia, the variability of individual and collec-
tive behavioural patterns of the respondents shows that nation and ethnic iden-
tities promoted by contemporary states call for collectivist rhetoric, marginalis-
ing individual reflections on current policies.

4. CONCLUSION
Oral histories and interviews reveal many contradictions of people’s reflections
on the causes of socialism’s deconstruction. On the one hand, the narratives
(when the “pure narrative” is recorded on the everyday life of an individual) do
not speak for any feeling or consideration of the soon-to-come systemic change,
on the other hand, the exact question about the causes of the change reveals
predominantly catching-all explanations learnt and perceived from public dis-
course rather than concluded based on personal observation. Questions on the
living standards at perestroika (before the second part of 1989) reveal the general
satisfaction of the population with the system, particularly social guarantees are
regretted to be lost.
People’s memories are interconnected with scholars’ encounters to a much
greater extent than usually imagined. Based on the same public discussions on
the causes of socialism’s deconstruction we have been reproducing the
knowledge of late socialism, forming, above all, the concept of Central Asia. The
studies of historical and social roots of those discussions are called for, as well as
writing of the history of socialist societies.
 On the Causes of Socialism’s Deconstruction 573

BIBLIOGRAPHY
Abashin, S. (2007), Natsionalizmy v Tsentral’noj Azii. V poiskax identichnosti, St. Petersburg.
Amsler, S. (2009), The Politics of Knowledge in Central Asia, London-New York.
Antonovich, I. I. (1990), “Sotsialism na iskhode dvattsatogo veka”, Sotsialogocheskie issledovanija 7:
3-11.
Baabar, B. (1990), Buu mart! Marval sönönö! [Do not foregt! Otherwise—we will perish!], Ulaanbaa-
tar.
Barkmann, U. B. (2006), “Politische Parteien, Parliament and demokratischer Prozess in der Mon-
golei”, Mongol uslyn shilzhiltijn üeijn uls törijn namuudyn üüreg [The role of state parties of
contemporary Mongolia], Ulaanbaatar: 309-402.
Boldbaatar, J. (2007), Mongolchuud Delhij Dahiny Tüühnee [Mongols in the World History], Ulaan-
baatar.
―― (2012), “Die demokratischen Veränderungen in der Mongolei an der Schwelle der 1980-er zu
den 1990-er Jahren und weitere Tendenzen”, Working paper within the project The History
of Perestroika in Central Asia, Ulaanbaatar.
Brady, A.-M. (2005), “From opening minds to regimenting minds: a comparison of propaganda
policy in Gorbachev’s Soviet Union and Jiang Zemin’s China”, Tragedija velikoj imperii: na-
tional'nyj vopros v raspade SSSR. [The Tragedy of the Great Power: national question and the
dissolution of the USSR], Moskva.
Brown, A. (2007), Seven Years that Changed the World. Perestroika in Perspective, Oxford.
Cox, T. (1996), From Perestroika to Privatization. The Politics of Property Change in Russian Society
1985-1991, Aldershot.
Dadabaev, T. (2010), “Power, Social Life and Public Memory in Uzbekistan and Kyrgyzstan”, Inner
Asia 12, Brill: 25-48.
Dorj, P. (1987), ZHUKN-yn tüühen turshkagyn zarim asuudal [Some questions of the SPSU historical
pattern], Ulaanbaatar.
Editors of Time Magazine, TIME 1989: The Year that Defined Today's World, Times, 2009.
Enkhbaatar, M. (2011), “The History of Perestroika in Mongolia”, MA thesis, under the supervision of
Prof. Boldbaatar, National University of Mongolia, Ulaanbaatar.
Filtzer, D. (1994), Soviet Workers and the Collapse of Perestroika, Cambridge.
Firsov B. M. (1997), “The Elite and the Power in Russia. Historical Dynamics of Relations”, Elites in
Transition. Elite Research in Central and Eastern Europe, Leske-Budrich: 133-148.
Florath, B. (2011), Das Revolutionsjahr 1989: die demokratische Revolution in Osteuropa als transna-
tionale Zäsur , Göttingen.
Fridman, L. / A. I. Belchuk (2007), Rossija v okruzhajushchem mire, Moskva.
Fursov, A. I. (2006), “Mify perestrojki i mify o perestrojke”, Sotsialogicheskie issledovanija 1: 31-36.
Hostler, C. W. (1993), The Turks of Central Asia, Westport.
Hyde-Price, A. G. V. (1991), European Security beyond the Cold War. Four Scenarios for the Year 2010,
London.
574 Irina Morozova

Il'in, V. (1996), Gosudarstvo i sotsial'naja stratifikatsija sovetskogo i postsovetskogo obshchestv (1917-

1996), Syktyvkar.
Karaganov, S. A. (1990), “The Year of Europe: A Soviet View”, Survival 32.2 (March-April): 121-128.
Kirchik, O. (2007), “Istorija kak ekonomika ili puteshestvie iz 1921 v 1906 cherez 1990”, Novoe Litera-
turnoe Obozrenie 83. (http://nlobooks.ru/sites/default/files/old/nlobooks.ru/rus/magazi
nes/nlo/196/329/351/index.html).
Khorduugaar zuuny Mongol, [The Mongols of the 20th century], Ulaanbaatar, 1995.
Kornai, J. (1992), The Socialist System: The Political Economy of Communism, Oxford.
Koschovke, K. (ed.) (2009), Falling Walls―the Year 1989/90 as a Turning Point in the History of
World Christianity, Wiesbaden.
Kosmarskaya, N. (2006), ‘Deti Imperii’ v postsovetskoi Tsentral'noj Azii: Adaptivnye praktiki, mental'-
nye sdvigi, Мoskva.
Kryshtanovskaya, O. (1991), “Soviet Millionaires”, Moscow News 29, 21 July.
Lane, D. (1971), The End of Inequality? Stratification under State Socialism, Harmondsworth.
Martin, T. (2001), The Affirmative Action Empire. Nations and Nationalism in the Soviet Union, 1923-
1939, Ithaca.
Mongol Ardyn Khuvsgalt Namyn Töv Khoroony V Bügt Khural [The Fifth Congress of the Central
Committee of the Mongolian People’s Revolutionary Party], Ulaanbaatar, 1988.
Morozova, I. (2010), “Political Parties against the Background of Neo-liberal Reform in Present
Mongolia”, The Journal of The Mongolian Society 32, Bloomington: 55-76.
Muminov, A. (2011), Rodoslovnoe drevo Mukhtara Auezova, Almaty.
Nelson, L. / I. Kuzes (1994), Property to the People: The Struggle for Radical Economic Reform in Rus-
sia, New York.
Novoe literaturnoe obozrenie 83, 2007.
Owen, T. C. (1995), Russian Corporate Capitalism from Peter the Great to Perestroika, New York-Ox-
ford.
Paksoy, H. B. (1993), “Basmachi Movement from Within: Account of Zeki Validi Togan”, Nationali-
ties Papers 23.2: 373-399.
Pürevjav, S. (ed.) (1986), Uls Törijn Namuud [State Parties: Digest], Ulaanbaatar.
Ries, N. (1997), Russian Talk. Culture and Conversation in Russia During Perestroika. Ithaca, New
York.
Roshchin, M. (2011), “Evgenii M. Primakov: Arabist and KGB Middleman, Director and Statesman”,
The Heritage of Soviet Oriental Studies, London-New York.
Rossabi, M. (2005), Modern Mongolia: From Khans to Commissars to Capitalists, Berkeley-Los An-
geles.
Rywkin, M. (1982), Moscow’s Muslim Challenge: Soviet Central Asia, London.
Sablonnière, R. de la / F. Tougas / M. Lortie-Lussier (2009), “Dramatic Social Change in Russia and
Mongolia. Connecting Relative Depravation to Social Identity”, Journal of Cross-Cultural
Psychology 40.3, May: 327-348.
 On the Causes of Socialism’s Deconstruction 575

Sil, R. (2011), “Privatization, Labor Politics, and the Firm in Post-Soviet Russia: Non-market Norms,
Market Institutions, and the Soviet Legacy”, The Politics of Labour in a Global Age Continu-
ity and Change in Late-Industrializing and Post-Socialist Economies, New York: 205-232.
Ticktin, H. H. (1992), Origins of the Crisis in the USSR: Essays on the Political Economy of a Disinte-
grating System, New York.
Umbetalieva, T. (2012), “Kazakhstan and Perestroika: was a chance at ‘Heroism’ Lost?”, IIAS News-
letter 60 (summer).
Yakovlev, A. N. (2005), Sumerki, Moskva.
 The Limitations of Military Psychology:
Combat-stress and Violence-values among the
Chechens and Albanians*

Caspar ten Dam

Leiden University

Abstract
This article examines the interplays between combat-stress and traditional violence-values—two of
the four variables of my Brutalisation theory—among the Chechens and Albanians in the latest
conflicts against the Russians and Serbs. I first discuss in considerable detail some major theories
and approaches on combat-stress in military psychology. I then point to some current shortfalls in
this field, including a serious dearth of research on combat-stress and traumas among armed non-
state actors generally and Chechen and Albanian insurgents specifically. As a partial compensation
to this lack of knowledge, I describe how violence-values affect combat-stresses (and vice versa)
among Chechens and Albanians. Finally, I suggest that stresses and traumas of Chechen and Alba-
nian combatants account for many of their brutalities (also in post-war settings). Such brutali-
ties—even if less common and systematic than Russian and Serb atrocities during and after the
wars—violate international and local norms, i.e. the very violence-values of martial valour and
honour that enhanced their combat-stress to begin with.

Keywords
Chechens, Albanians, Combat-stress, Violence-values, Stressors, Stress-responses, Reluctance and
Eagerness to Kill, Social and Group Pressures

I. INTRODUCTION
This article examines the presence, impact and role of combat-stress among
Chechen and Albanian rebels during the latest Russo-Chechen conflicts (1994-
1996, 1999-present) and the Serbo-Albanian conflict in Kosovo (1997-1999). Com-

*
This monograph is an extended version of the third article of my “How to Feud and Rebel” Se-
ries in Iran and the Caucasus (Brill): “3. Combat-stress and Violence-values among the Chechens
and Albanians”, vol. 16.2 (2012): 225-245. Particularly section II is much enlarged here, with an elabo-
rate analysis of fear, rage and other stress-responses, social pressures in the military (or combatants
generally), and the contrary human dispositions towards killing and not-killing.
 Koninklijke Brill NV, Leiden, 2015
578 Caspar ten Dam

bat-stress constitutes the third variable of my Brutalisation theory, the other

variables being violence-values, conflict-inducing motivations and conflict-induced
motivations (grievances, ‘greeds’ or avarices, interests, and ideologies), i.e.
motivations that cause or contribute to the conflict, and/or motivations that
occur during the conflict and exacerbate it.1 Brutalisation concerns an increasing
resort to terrorism, brigandry, gangsterism, war crimes and other forms of
violence that violate local and/or international norms.2 I also assess salient
violence-values (a composite term coined by the author)—that collectively
constitute the Brutalisation theory’s first variable—as stress-reducers and stress-
enhancers affecting the behaviour and resilience of Chechen and Albanian
fighters. I already have amply described Chechen and Albanian violence-values,
and compared these to international norms, in the first article of my “How to
Feud and Rebel” Series (Ten Dam 2010). Moreover, I have assessed the largely
valid saliency of long-term, historical grievances among Chechens and Albanians
in the second article of the same Series (Ten Dam 2011).3 Once more, I depart
from the post-constructivist proposition (expounded in the first article of the
Series) that the “acting-out” of norms, values and beliefs, irrespective their factual or
invented provenances, underlies one’s identity.
Section II of this article assesses the main theories on combat-stress in mili-
tary psychology. This discipline primarily aims to treat or ameliorate the effects
of (post-combat) post-traumatic stress disorder (PTSD) and other psychological

1
From 2005 till 2013, I have described my brutalisation theory, with some minor modifications,
as “a cycle of violence involving four main variables: “values on “good” and “bad” violence (variable
1); grievances leading to armed conflict (variable 2); combat stress leading to atrocities (variable 3);
and new conflict grievances emanating from such atrocities (variable 4), spawning counter-atroci-
ties and eventually hardening or debasing the original violence-values (the cycle returns to the first
variable)”: “1. Violence-values among the Chechens and Albanians”, Iran and the Caucasus, vol. 14.2
(2010): 332. Yet since then, I have widened and thereby reformulated the theory’s second and
fourth variables, so as to more equally represent different motivations as explanations of brutal
behaviour, taken from or inspired by diverse theories propounding particular kinds of motivations
as the primary causes of such behaviour.
2
For fuller descriptions of both the earlier version and later versions of the theory see C. ten
Dam (2010, 2011), Iran and the Caucasus, vol. 14/2: 332 and vol. 15.1-2: 236. See also “Brutalisation
theory”, with a new Diagram of the reformulated version, at <http://sites.google.com/site/tristanso
lutions>.
3
These historic grievances reinforce, and are reinforced by, violence-values of honour, blood
feud, raid, hospitality and mediation, and the societal values of martialism, resistance and egalitari-
anism (Ten Dam 2010: 333-334).
 The Limitations of Military Psychology 579

disorders, and prevent, by timely diagnosis and ‘hardening’ training-and-condi-

tioning techniques, it from (re)occurring or proliferating among ‘their’ soldiers.
Section III criticises the skewed attention in military psychology on state rather
than non-state armed actors. At present, I only can compensate the resultant gap
in knowledge by exploring the interactions between combat-stress and violence-
values among Chechen and (Kosovar-)Albanian insurgents. Section IV reveals
how traditional violence-values, and the consequent high expectations towards
honour, courage and sacrifice, have exacerbated rather than reduced combat-
stress among Chechen and Albanian combatants. This led to mental breakdowns
even or rather among the hardiest of them, and accounted for many brutalities,
particularly ‘berserk’ violence and terrorism i.e. lethal violence against civilians
or other practically defenceless individuals (even though generally these pale in
comparison to Russian and Serbian atrocities in severity or scale). Section V re-
capitulates the fateful dynamics between combat-stress(es) and violence-values
among Chechens and Albanians.

II. COMBAT-STRESS: CAUSES AND CONSEQUENCES

Long before the rise of military psychology as a distinct and recognised disci-
pline, participants and observers spoke of the horrors and (consequent) stresses
and traumas of war. Thus U.S. General William Sherman famously lamented in
1880 that “I am tired and sick of war. Its glory is all moonshine. It is only those
who have neither fired a shot nor heard the shrieks and groans of the wounded
who cry aloud for blood, more vengeance, more desolation. War is hell”. This
outburst occurred when he was telling military graduates of the horrors of the
1860-1864 American Civil War. With 60,000 men, marching “from Atlanta three
hundred miles to the sea”, Sherman destroyed “every town, rail yard, mansion,
and crop across a swath of sixty miles”; his own role in it “haunted Sherman for
the rest of his life” (Cooke 1974: 218). Sherman’s war-is-hell thesis came to include
the belief that combatants inevitably execute, massacre, rape and loot the de-
fenceless and vanquished; all are sucked into a brutality inseparable from or-
ganised violence; there is no decency in combat: that happens, at best, in excep-
tional circumstances by exceptional individuals. Even they cannot dispel the
nightmare of armed conflict with its corrosive effects on battlefield ethics; com-
batants are bound to brutalise, terrorise and criminalise, despite lofty ideals on
how they could or should behave. Easily-confirmed notions that “atrocity is in-
evitable in war” (Watson 1980: 175), and that “all armies in all wars do terrible
580 Caspar ten Dam

things” (Gall/De Waal 1997: xi) have gained currency ever since. Nevertheless,
‘war-is-hell’ is not the most unreservedly negative view of human behaviour; it at
least conveys the notion that people brutalise only under extreme stress. In con-
trast, Robert Cribb’s “fatal discoveries”, of feeling betrayed by the “other”, and the
realisation that one could get away with anything when violence breaks out,
suggest that people are naturally inclined to inflict pain.4 Indeed, my theory of
brutalisation departs from that most pessimistic viewpoint on human nature
(not that I share it; I actually hope to disprove the general validity of brutalisa-
tion, or some parts of it). A related yet distinct dimension is radicalisation, the
process of increasing militancy in one’s cultural, societal, political and ideologi-
cal viewpoints that tend to exclude or demonise the “other”.5 Brutalisation easily
leads to radicalisation—and vice versa. Still, combat-stress may turn out to be
the most determinate variable in this (self-)destructive process, relatively inde-
pendent from the particular grievances, avarices, interests, ideologies, societal
values and even violence-values of the participants.

MILITARY PSYCHOLOGY
On 6 April 1917, just after the United States declared war on Germany, Robert M.
Yerkes, the president of the American Psychological Association (APA) who de-
fined military psychology as the “application of psychological methods to mili-
tary problems”, urgently wrote that the psychologist’s duty was to work toward
the “increased efficiency of our Army and Navy” (Yerkes 1918: 86-87). Military
strategists, participants, observers and physicians always have exhibited psy-
chological insights, especially in deceiving and demoralising their enemies but
also in seeking to understand their own fears and weaknesses, from the Pelopon-
nesian Wars around 400 BC to the American Revolution of the 1770s and the
American Civil War of the 1860s. Still, military psychology, psychiatry and pa-
thology, military psychology in short, only arose as a science during the First
World War– though Russian psychiatrists were the first to diagnose battle stress-
es, traumas and breakdowns at the front during the 1905 Russo-Japanese War.6

4
R. Cribb, in: Dirk Vlasblom, “Het is de vonk, niet het droge gras (It is the spark, not the dry
grass)”, NRC Handelsblad (New Rotterdam Courier-General Trades’ Paper), 17 April 2008: 9; terms
translated from Dutch.
5
I will more amply analyse radicalisation in “Conflict-Induced Motivations among the
Chechens and Albanians” (Iran and the Caucasus, forthcoming).
6
Richard A. Gabriel emphasises the pioneering role of Russian military psychiatrists and psy-
 The Limitations of Military Psychology 581

Authorities sought to understand why so many soldiers, marines and aviators

suffered mental collapse, and how they could be treated (preferably near the
frontline) and redeployed. Also during World War II the military sought to
“make men fit for combat” including those temporarily incapacitated; “our func-
tion is not long-term care” (Porter, apud Sladen 1943: 240, 243). Military psychol-
ogy is broadly defined as the application of psychological principles to military
‘problems’, ‘needs’ or ‘environments’. Yet the discipline became specifically con-
cerned with how to select suitable recruits and bar those with severe mental dis-
orders (psychopathologies); choose the best for “responsible positions” (Yerkes
1919: 90); train and harden them against battlefield adversities; optimise (their
use of) weapons, vehicles and other machines (human factors engineering); and
take care of them if they buckle under the stress—though prioritising the “return
to duty of service members with manageable combat stress reactions” (Williams
et al, apud Kennedy/Zillmer 2006: 201-202). Some include efforts to “minimize
the enemies’ behavioural capabilities” through propaganda7 and other (ill-de-
fined) forms of psychological warfare (Walters 1968: 2 (quote), 19-25)—including
(better-defined) psychological operations (PSYOP) to “influence the emotions,
motives, decision making, and .. behavior of adversaries” (Williams et al, apud
Kennedy/Zillmer 2006: 195).
In World War II the American priority became to better select and train the
mentally fit; on that issue Samuel A. Stouffer and associates undertook one of the
largest investigative projects ever (Stouffer et al 1977 (1949): 3-4, 12-14, 20-21).
During the Korea and Vietnam wars, the United States and other countries per-
fected techniques to ‘brainwash’ soldiers into efficient, brutal killers. Military
psychologists thrived during such wars, but became less prominent during the
interbella due to diminished government priority and funding.8 Nevertheless, a
new ‘wolf pack’ led by Lt. Col. Dave Grossman appeared in the mid-1990s, biting

chologists, though he also argues that their Western counterparts eventually overtook them in the
sophistication and validity of diagnoses (Gabriel 1988: 9-10, 17, 22 (Russo-Japanese War); 67-92
(Chapter 4) ).
7
Word “propaganda” (propagation) is derived from the Catholic Congretatio de Propaganda
Fide or “Association for the Propagation of the Faith” formed in the 15th century (Kennedy, apud
Sladen 1943: 342).
8
WWII gave a “boost” to military psychology, though the US had ‘fortified’ it in “peacetime”
(Cronin 1998: 1-2,4). According to others, psychologists and military “parted ways” during 1918-1939,
though less so after 1945 (Driskell/Olmstead 1989: 43-45). Still, “research funding declined in the
late 1980s through the late 1990s” (Krueger, apud Cronin 1998: 24).
582 Caspar ten Dam

their teeth into the “taboo topic of killing in combat” (Grossman 1996: xxix),
which Br. Gen. L. S. A. Marshall had already broached in the late 1940s: “essen-
tially war is the business of killing” (Marshall 1978 (1947): 67). Ever since, Gross-
man and his colleagues teach killology (study of killing in combat) and the wider
discipline of combatology i.e. study of the “psychology and physiology of combat”
(Grossman/Christensen 2007: xxiii) to military and law enforcement students,
officers and veterans. These teachings are intended to improve their resilience,
understanding and after-the-event coping of the “toxic, corrosive, destructive
realm of combat” (Grossman/Christensen 2007: xiii).9
Some analysts have criticised the “inadequacy of psychological .. language in
describing” moral trauma (Marin 1981: 74). Thus ‘stress’ appears to have become
a “jargon word” that substitutes, and even dehumanises, the concept of fear
(Watson 1980: 37). Military-psychological jargon may indeed obscure, patholo-
gise and delegitimise fear, trauma (post-event distress), guilt and other emotive
states; yet I have come across no better overall term than stress to encapsulate
these phenomena.
Therefore, it seems that military-psychological theories can explain atrocities
emanating from combat-stress, the Brutalisation theory’s third variable. How-
ever, many (military) psychologists and sociologists (over)confidently posit that
stress-induced cruelty, and violence sui generis, occur irrespective of grievances,
convictions and other root causes. They argue that one should “put the [group-
behavioural] interaction in the center of .. analysis, not the individual, .. social
background, .. culture, or even .. motivation” (Collins 2008: 1).10 Such a view effec-
tively dismisses most motivational conflict theories I have succinctly discussed
elsewhere (Ten Dam 2011: esp. 237-241). Even so, some scholars seek to explain
the “vast array” of violence, ranging from domestic quarrels to foreign wars, by “a
relatively compact theory” reminiscent of frustration-aggression theory (Collins
2008: 1). In my research I am modestly interested in accounting for (rebel) bru-
tality, not violence per sé. Here I only deal with ‘narrow’ stress theories that posit

9
They prefer to call themselves “sheepdogs” (Grossman/Christensen 2007: 120-121). Unlike
“wolves”, sheepdogs work “under authority, .. as team players, to protect innocent lives” (ibid: 122
(quote); 180-190). Yet, army and police forces do not always practice this ideal. And many ‘illegiti-
mate’ wolves, like rebels, do act under (a) authority as a bonded group—and protect innocent
lives.
10
Yet Randal Collins appears to circumvent the fact that the individual remains the ultimate
unit of analysis, even if humans are social animals.
 The Limitations of Military Psychology 583

a single, few, many or all of the following factors as valid explanations of combat-
stress:
Stress-responses of shock, fear, fatigue, rage, trauma and other distresses i.e. “nega-
tive affective states” as reactions to stressors i.e. “stressful [combat] events” (Co-
hen/Williamson 1991: 5) like receiving or witnessing gruesome injury, or seeing a
friend killed.
Avoidance and overdrive behaviours, ranging from an apathetic, non-aggressive
reluctance to kill, injure, capture or otherwise incapacitate the enemy, to an
overly aggressive, excessive and (often) brutal eagerness to kill, injure, capture or
otherwise incapacitate the enemy.
Social pressures of or by military training, indoctrination, conditioning (imbibing
behaviour-patterns through psychological techniques), and group convictions,
group bondings and (consequent) group expectations in the civilian and military
spheres.

Shock
When tens if not hundreds of thousands of soldiers suffered physical-and-mental
breakdowns in WWI, authorities became desperate to ‘patch them up’ and send
them back to the front. Pathologists first noted, analysed and treated the ‘pecu-
liar’ numbness many soldiers displayed after shells had exploded near them. The
provenance of the term shell shock is contentious: possibly “medical officers
adopted .. a soldier’s phrase”, yet British Col. Frederick Mott reputedly coined it.
Though Mott wrongly diagnosed it as generally arising from minute brain hem-
orrhage, it became a famous term denoting the prevalent paralysis among sol-
diers ‘shaken’ by explosions and other shocks ‘breaking the camel’s back’. Earlier
terms like nostalgia (apathy due to stress, fatigue or homesickness) and later
terms like combat neurosis covered a wider range of nervous breakdowns (Butler
1943: 99 (quote); Weinberg 1946: 466 and his note 4; Gabriel 1998: 14-15).
Be as it may, American and British WWII studies confirmed that the “battle
incident most likely to ‘break’ a soldier is the explosion of a shell” near him, due
to the explosion’s noise and shock-wave; even stable personalities may succumb
to a single event overwhelming the senses, particularly that of hearing: the “im-
mediate impact of war is noise ... incredibly loud ..—small arms, bombs and air-
craft” (Watson 1980: 151, 153). Even “under normal peacetime conditions, noise is
a strain or stress, ranging from irritation to pain” (Dinter 1985: 37). Actually, the
“noise intensities of modern military equipment are much higher than those
found in most industrial jobs”; consequently thousands of US “ex-military per-
sonnel wear hearing aids” (Krueger, apud Cronin 1998: 98,99). Usually the soldier
584 Caspar ten Dam

is temporarily ‘concussed’ but recovers quickly, even in minutes—if there is no

serious damage to internal organs or long-lasting, blast-related traumatic brain
injury (TBI) (Ryan et al, apud Kennedy/Zillmer 2006: 108-111).11 He or she also
recovers “quite rapidly” (if superficially, as later research found) from other
“mental disturbances” like “acute schizophrenia” (a once typical diagnosis ac-
counting for rage) and “acute war neuroses” like anxiety, fatigue, depression and
(consequent) low morale (Porter, apud Sladen 1943: 245(quote)-246). Still, ex-
treme sounds, flashes, heat, cold, smoke, shock-waves (by explosions) and trem-
ors (ibid) may precipitate nervous collapse, even if unaccompanied by concus-
sions, other injuries, or other stresses like anxiety and exhaustion. The reverse
may also be true: rather than concussion or paralysis, “loud noises” can provoke
“fear and terror responses in adults”, as William James discovered in the late
nineteenth century (Bartone, apud Cronin 1998: 115).12 However, while “explo-
sions farther away can be extremely loud”, explosions “close enough to slam your
body” are not heard i.e. registered by your ears. The body reacts to the primary
danger signal: “when a creature .. lands on your back and roars .., your dominant
survival information would be the sensation of something on your back”
(Grossman/Christensen 2007: 61). If one does not rebound quickly from such
natural yet detrimental survival-responses like auditory exclusion, auditory en-
hancement, concussion or tunnel vision (see Grossman/Christensen 2007: 8-11, 14-
15, 30-49, 54-73, 94-99, 112-122), one may become paralysed, apathetic or unre-
sponsive—i.e. descend into shell-shock or a state very much like it.
Fear
The battlefield is one of the most lethal places to be. Many analysts “question our
ability to endure the horror of battle” given that “fear and madness [mental
breakdown] have been man’s companions in war since the beginning of rec-
orded history” (Gabriel 1988: 7). Even hardened combatants can succumb to fear,
a primal emotion and response to danger, if not the only one (see Maslow 1941,
1943, 1987 (1954) ). The classic frustration-aggressionists have recognised this: the
“dominant reaction among survivors” of the nuclear bombs on Hiroshima and
Nagasaki was “acute fear rather than anger” (Berkowitz 1962: 42). However, the

11
Combat-related TBI is “not well described in the literature”; most research has focused on
internal-organ, torso, limb and “penetrating head injuries” (Ryan et al, apud Kennedy/Zillmer 2006:
109). Perhaps F. Mott’s erroneous brain-hemorrhage diagnosis of shell-shock discouraged research
into TBI.
12
From William James (1890), Principles of Psychology, New York.
 The Limitations of Military Psychology 585

so-called relative deprivationists tend to underrate fear among civilians: “heavy

bombings .. first produced acute fear, not anger, but also generally led to in-
creased hostility” (Gurr 1970: 35).13 On the other hand, many scholars tend to
overrate fear, regarding anxiety and hysteria as the “most prevalent” of “trau-
matic or situational war neuroses”, relegating fatigue, depression and other con-
ditions as non-essential components of ‘combat neurosis’ (Weinberg 1946: 466-
467). Be as it may, anecdotal if not structural evidence suggests that, as part of
the “omnipresent fear of the unknown and the unexpected”, “fear of mutilation
is certainly greater than the fear of death itself”, whereby injuries which “do not
lead to mutilation do not frighten the soldier so much” (Dinter 1985: 20, 25).
Elmar Dinter believes that group expectations produce the most fundamental
anxiety of all: “the soldier’s fear of losing both his primary group and the love and
recognition of the people back home, exceeds his strong fears of the unknown,
the unexpected and mutilation” (Dinter 1985: 50).
In contrast, Lt. Col. Dave Grossman criticises the “simplistic” yet “widely ac-
cepted” proposition that the “cause of most trauma in war is the fear of death
and injury”; rather, “fear may be one of the least important factors” accounting
for a soldier’s breakdown: thus strategic WWII bombings caused surprisingly few
“psychiatric casualties” among civilians, active combatants and prisoners of war.
Especially well-trained commandos on reconnaissance and sabotage missions,
who are not obliged to kill unless detected, exhibit little anxiety. Danger and
“fear of death and injury” are “not the predominant cause of psychiatric casual-
ties in battle” (Grossman 1996: 52, 54, 60-62). Even the “universal human phobia”,
the “irrational, overwhelming, uncontrollable” fear of interpersonal human ag-
gression, rarely engenders severe mental breakdown (Grossman/Christensen
2007: 2). Yet it typically engenders a bodily survival-response (including sensory
distortions), panic, humiliation, anger and long-lasting trauma: “It is not a fear of
death. … We can accept the fact that we will die of old age, or that an “act of na-
ture” might take our lives .. . But we cannot accept .. someone “playing God”, and
… steal away our loved one’s lives” (ibid: 5-6). Then the response is extreme an-
ger. In short, fear generally leads to combat-stress, often to after-the-event com-
bat-trauma, but hardly ever to acute combat-breakdown. Thus Grossman distin-
guishes between fear as a natural stress-response (typical) and fear as a direct

13
Both Berkowitz and Gurr refer to I. L. Janis (1951), Air War and Emotional Stress: Psychological
Studies of Bombing and Civilian Defense, New York: 4-66. See on frustration-aggression, relative dep-
rivation and other conflict theories C. ten Dam, Iran and the Caucasus, vol. 15.1-2: esp. 237-241.
586 Caspar ten Dam

cause of mental collapse (atypical). Even Peter Watson acknowledges that

“bombing has less of a psychological impact than might be expected” (Watson
1980: 155). Nevertheless, to this day scholars place anxiety at the centre of their
theories of violence. Randall Collins defines his “micro-situational theory of vio-
lence” as a “set of pathways”—“find a weak victim to attack”, “focus .. on the au-
dience”, limit oneself to “bluster”—around “confrontational tension and fear”
(Collins 2008: 4, 8, 9, 10). Yet such tension and fear only lead to severe traumas
and pathologies among some people under extreme conditions of, say, war.
Through training one learns to suppress and control fear, when one must
move forward, fight, and kill. Researchers found that the social fear of “letting
others down” and simply “combat experience decreases fear of death or injury”
(Grossman 1996: 52).14 Even so, after each battle and adrenaline rush, fear needs a
mental-physical outlet: one gets the ‘shivers’. American military researchers in
the 1960s did find that intelligent and able soldiers, better aware of the risks,
went absent without leave (AWOL), inflicted wounds on themselves or pretended
physical and mental diseases in order to get medical discharge, or managed prior
to any war to “cluster in the safer jobs away from the frontline” (Watson 1980:
34(quote), 105, 160). These patterns specifically emerged during the Vietnam
War, when the draft induced capable men most unwilling to take risks or wage
war to dodge it. However, risk avoidance, as posited through rational-utilitarian
gain theory, hardly accounts for soldiers pretending illness or openly fleeing
defensive positions to medical stations and other places that actually “offered no
protection” from artillery fire (Grossman 1996: 57).15 Moreover, other intelligent,
stable and strong men are eager to join the special forces, or accept other chal-
lenging jobs in the army, air-force and navy. Unsurprisingly, high-risk-personnel
candidates receive high scores in aptitude tests on “stress resilience, adaptability,
cooperation with others, and .. physical fitness and stamina”—leading to the
tautological conclusion that “psychological hardiness and stress tolerance” are
“critical to successful performance” (Picano et al, apud Kennedy/Zillmer 2006:
359). These attributes, and high motivation coupled with extensive training,
practically ensure that these “special warriors” (Mountz, apud Kaslow 1993: 121-
129) can ‘conquer’ fear while on mission.

14
From Ben Shalit (1988), The Psychology of Conflict and Combat, New York.
15
Grossman seems to forget that panic can swamp cost-benefit rationality.
 The Limitations of Military Psychology 587

Fatigue
Exhaustion continuously accompanies the soldier: in training, drilling, patrol-
ling, deploying, camping, fighting, retreating, advancing, and endlessly cleaning
and repairing one’s kit, weaponry and other machinery. Especially the “para-
sympathetic backlash”—the body shuts down for maintenance after it has gone
through one or more adrenaline rushes to face a dangerous situation—accounts
for so-called “soldier fatigue”: psychiatrists accompanying an experienced US
combat unit in the Korean War found that after many an intense battle, “the
men had fallen into an exhausted sleep, though they knew they would soon be
attacked” (Grossman/Christensen 2007: 15, 16). In their classic WWII study of the
Allied landing in Normandy in 1944, R. L. Swank and W. E. Marchand found that
after sixty days of continuous combat, practically all soldiers succumb to physi-
cal and emotional exhaustion and become psychiatric casualties. Other studies
found that practically all combatants collapsed only after 90 to 120 days of con-
tinuous combat, but generally confirmed Swank and Marchand’s ‘bell-curve’
findings on combat efficiency over a period of time (Swank/Marchand 1946: 236-
247; Dinter 1985: esp. 66-67; Grossman 1996: esp. 43-44). Still, Grossman and
Christensen caution that situations in which soldiers are “trapped in continuous
combat for 60 to 90 days” without rotations or reinforcements are rare: “On the
beaches of Normandy … there were no rear lines, and for two months there was
no way to escape the horror of continuous fighting” (Grossman/Christensen
2007: 12(quotes)-13; Grossman 1996: 44-45).
One of the main challenges for the military psychologist has been to devise
optimal sleep rhythms, nutrition, clothing, rest, morale and relaxation, and if
necessary medicinal ‘boosts’ which “merely buy time” (Dinter 1985: 29), for the
soldiers even when on campaign or in battle. Mental concentration and adrena-
line peak either for too long—“in a slit trench a man tends to brace himself con-
tinuously when under fire and this is exhausting”—or drain too quickly away
when one ‘relaxes’ after dangerous action: the “parachutist feels tired .. a few
moments after .. a safe landing” (Watson 1980: 167). During the taking of Hill 440
in Korea, American officers had the greatest difficulty in “keeping their men
awake—in broad daylight under intense bullet and mortar fire” (Watson 1980:
167). Generally, post-adrenaline relaxation (if not exhaustion) after battle makes
combatants lax and thus vulnerable to counter-attack (Marshall 1978 (1947): 143-
44, 193-197). Similarly, guard duty—especially in confined trenches—produces
tiredness and lower performance the longer it lasts: “Any form of rest helps—
588 Caspar ten Dam

even being allowed to fidget improves performance” (Watson 1980: 70). Even “in
the most confined spaces he should at least shout or shoot, if necessary without
a target” (Dinter 1985: 39). If one must hide one’s location, then one only could
fidget to diminish tension, cramp and tiredness.
Different physical deprivations have different effects across different
timeframes. Corroborating Maslow’s hierarchy of basic (immediate) versus sup-
plementary (long-term) human needs (Maslow 1941, 1943, 1987 (1954) ), only
chronic or severe sleep deprivation deteriorates task performance, while lack of
water, food and even “additional noise, .. heat or cold” (Dinter 1985: 30) may do
so far more quickly and drastically. This shows how stress-responses and their
stressors can interrelate and accumulate; thus shock or its stressors can enhance
fatigue and vice versa. Likewise, fatigue can enhance fear and vice versa: thus if
“fatigue sets in”, it often “creates a fear of incapacity, fear of failure and a fear of
inferiority” (Reinartz, apud Sladen 1943: 276). J. F. Mackworth found almost no
task degradation, even improvement, if people could rest for thirty minutes after
every thirty minutes of work (Watson 1980: 70 (note 25)-71).16 Unfortunately,
such a work-leisure distribution is impracticable in most civilian occupations,
and utterly undoable in military ones (not counting the long waits filled with
boredom). When a soldier or any combatant needs to be vigilant incessantly, fa-
tigue easily trumps fear or any other stress-response. Many combatants fall vic-
tim to post-traumatic stress disorder (PTSD) due to sleep deprivation alone. This
reportedly happened to many US soldiers in Iraq and Afghanistan because they
were allowed to play video games, watch TV and do other escapist activities after
combat patrols—this in stark contrast to the strictly enforced sleeping regimes
in both world wars. Contemporary defence and law enforcement departments
need to relearn that “you can die from lack of sleep faster than you can die from
lack of food” (contrasting earlier findings) and that “throughout history sleep has
always been the soldier’s best medicine” (Grossman/Christensen 2007: 23-25).

Rage
The average combatant will at least once experience rage i.e. extreme, barely
controllable anger, in response to shocking, traumatic experiences: one loses a
comrade in combat, comes under continuous, ‘unfair’ enemy fire (leading to
other stress-responses like shock, fear and fatigue), or witnesses atrocities com-

16
On Mackworth: D. R. Davies; G. S. Tune (1970), Human Vigilance Performance, Staples Press:
138 ff.
 The Limitations of Military Psychology 589

mitted by the enemy—or by one’s own. All too often, such rage becomes uncon-
trollable: combatants go berserk, i.e. descend into a state of frenzy and become
extremely aggressive against enemy combatants, bystanders (eg. civilians) or
even their comrades. Either their last restraints break down under the stress,
leading to “desperate aggression” to “destroy the [enemy] object before he suc-
cumbs” (Weinberg 1946: 472),17 or these are deliberately broken down by their
comrades, superiors and (other) agitators. Anxiety often brings “resentment and
hostility”, not only ‘healthily’ towards the enemy but also ‘neurotically’ towards
oneself (repressed hostility) or one’s comrades (overt hostility), leading to “im-
pulsive outbursts of temper and violence” (Weinstein 1947: 310, 312).
Even the most romantic accounts in war literature acknowledge and depict
notable examples of rage and consequent brutality. However, military psycholo-
gists rarely grapple with this phenomenon, perhaps because it remains a taboo
topic among the military they work for—and because they remain “confounded
by the unpredictability with which aggressions sometimes explode, in a fury no
one sees coming”.18 They and the military themselves typically settle for the suc-
cinct, obvious observation that stress “can also increase the risk of soldier mis-
conduct” (Bartone, apud Cronin 1998: 119), without mentioning rage at all. The
US Army and Defence Department, at least until recently, did not even consider
misconduct (let alone rage) as a possible combat stress reaction (CSR), despite
recognising that “combat exposure can lead to misconduct” (Campise et al, apud
Kennedy/Zillmer 2006: 215, 216). Even Watson rarely and obliquely refers to rage
as an intermediating factor between stress and brutality: when “a high-risk sol-
dier finds himself in a .. position that is highly stressful for long periods of time, ..
the chances of atrocities are increased” (Watson 1980: 178).19
Some analysts refer to rage more directly, when speaking of “behavioral fac-
tors” contributing to or reflecting combat-stress: “Is the enemy treated with a
measure of humanity? Are bodies of the dead treated with respect? Has the ci-
vilian community become a target of anger?” (Campise et al, apud Kennedy/
Zillmer 2006: 227). Others imply that rage, indeed any stress-response, inevitably

17
Soldiers “losing self-control” (Weinberg 1946: 473) may perform “heroic deeds” that are not
necessarily atrocious; neither are all heroic deeds born of rage or directly precede mental break-
down (ibid and his note 22).
18
Erica Goode, “When Soldiers Snap”, New York Times, 8 November 2009.
19
See also Watson’s note 5 on the main source: Peter G. Bourne (1970), Men, Stress and Vi-
etnam, Little, Brown and Co.: esp. 63-102, 144-150.
590 Caspar ten Dam

arise from war because war itself is a horrific pressure-cooker, and wonder
“whether men can continue to plan and fight wars while remaining truly human”
(Gabriel 1988: 1). However, Dinter insists that hatred—a kindred emotion to
rage, but (usually) of a more simmering, controllable kind—“plays a much less
important role than assumed” (Dinter 1985: 38): “Feelings of hatred and revenge
are the result of fleeting frustrations. If for instance a good friend has just been
killed, these emotions increase, but as time passes, they diminish rapidly again.
The average soldier does not maintain constant feelings of hatred and revenge”
(ibid: 45). Nevertheless, absolute and even relative deprivationists20 would disa-
gree with that observation, and particularly with the notion that a traumatic
event like a ‘buddy’ being killed would constitute a ‘fleeting frustration’.
Arguably, witnessing the death of one’s comrades is the most traumatic and
enraging event of all: the “ranks are never hardened by death in their midst.
Losses are never a help”; indeed “death in the company is like death in the fam-
ily” (Marshall 1978: 118, 121).21 True, rage may simply be an overheated reaction to
concussion and exhaustion. A soldier subjected to too many “physical depriva-
tions or over-stimulations” may become manic-depressive: even “minor upsets
lead to outbreaks of fury and good news is received with exaggerated elation”
(Dinter 1985: 60, 66 (quotes); Swank/Marchand 1946).22 Still, rage emanating
from grievance, hatred and revenge appears to be quite common. Thus Chechen
and Albanian rebels harbour salient grievances, and exhibit strong and persistent
emotions of hate, rage and revenge. Indeed, “revenge killing during a burst of
rage has been a recurring theme throughout history” (Grossman 1996: 179).
Joanna Bourke acknowledges that combatants are “transformed by a range of
emotions—fear as well as empathy, rage as well as exhilaration” (Bourke 1999:
1).23 Yet she accords most significance to the last emotion (see sub-section Eager-

20
I am the first to use the terms “deprivationism” and “depredationism” in the greed vs.
grievance context: see Iran and the Caucasus, vol. 15.1-2: 237.
21
Marshall’s observations on the ‘dead-buddy-trauma’ are widely shared by other analysts be-
fore and after him, though not many (nor Marshall) link this phenomenon to (brutal) rage as a re-
sponse to such a traumatic event.
22
If “the soldier cannot be relieved .. [of] this state, his excited mood will change into apathy”
(Dinter 1985: 66)—i.e. (shell-)shock or combat neurosis.
23
When Bourke recounts how Eugene B. ‘Sledgehammer’ Sledge (1st US Marine Division) ex-
plains the “collecting [of] bodily parts” from slain Japanese during WWII, rage remains unac-
counted: “Death, fatigue and stress wore away even the ‘veneer of civilization’ ” (Bourke 1999: 38, 39
 The Limitations of Military Psychology 591

ness to kill below). Likewise, Collins emphasises the “intertwining of human

emotions of fear, anger, and excitement” lies at the centre of his micro-sociologi-
cal theory of violence—though accords most relevance to fear (see sub-section
Fear above): “even in situations of apparently uncontrollable anger, people are
tense and often fearful” (Collins 2008: 4). If anything, they and other analysts
deem frustration rather than more intense rage as the harbinger of atrocity (sug-
gesting a lingering influence of the Frustration-Aggression theory)—despite
their assertion that most combatants are not predisposed killers prior to training
and conditioning. Does this mean that ‘mere’ frustration already leads to brutali-
ties? A disconcerting thought. It is certainly vital to discern for every instance of
“severe misconduct” if it is “a function of combat stress” among “otherwise nor-
mal soldiers” or “a manifestation of previous psychopathology” (Campise et al,
apud Kennedy/Zillmer 2006: 216, 219). If the psychopathic type turns out to be
frequent or predominant in all the (reported) atrocities, and if it concerns a sig-
nificant part of all military personnel, there may be something seriously wrong
with the recruitment and selection processes—despite repeated efforts to im-
prove these since the beginning of the twentieth century.24 Yet even the best psy-
chological tests and selection procedures will, by themselves, do little to counter
brutalisation if even mentally stable personalities easily succumb to it through
mere frustration rather than uncontrollable rage. I am not convinced, however,
that moral and behavioural constraints are so brittle. My research on Chechen
and Albanian political violence does not support such an extreme Frustration-
Brutality thesis. I give more credence to a Rage-Brutality thesis accounting for
many of the atrocities—while moral and behavioural constraints still prevent or
curtail atrocities, at least in the initial stages of (any) armed conflict.
Trauma
Since 1994 the American Psychiatric Association (APA, not to be confused with
the American Psychological Association) defines a traumatic event as involving
“actual or threatened death or serious injury, or a threat to the physical integrity
of self or others” (Schnurr/APA 1994: 424). I define trauma as an anguished, self-
harming or debilitating response to such an event. APA circumscribes post-

and note 104). Yet US army sniper Dave Nelson, after trying to uphold a “professional ‘warrior ethic’
” (ibid: 50) in Vietnam, killed out of “hatred, revenge and frustration” (ibid: 51, note 4).
24
Though men with “clear-cut psychoses” were detected “rather easily”, the “psychopath and
the neurotic” show “very little to the examiner” and often are “bright looking and mentally alert”
(Porter, apud Sladen 1943: 253).
592 Caspar ten Dam

traumatic stress disorder (PTSD)—a term it first formally adopted in 1980—as a

condition that exhibits at least some symptoms of six diagnostic criteria
(Schnurr/Green 2004: 5; Schnurr/APA 1994).25 It is an excellent summary of the
disorder’s ‘cocktail’ of physical and mental causes and effects, based on Cohen
and Williamson’s “event-reaction distinction” (Cohen/Williamson 1991: 5). Over-
all, “PTSD plays a crucial role in mediating the relationship between traumatic
exposure and poor physical health” (Schnurr/Green 2004: 7; Cohen/Williamson
1991).
Nevertheless, I conceptualise PTSD as a stress-response, albeit a complex and
cumulative one composed of other stress-responses like rage, fatigue and fear.
Moreover, a response like rage could either be a cause or consequence of a
traumatic event (or both in a cyclical fashion). Most theories consider just one or
two responses, like fear, fatigue and rage, as the “primary” responses to traumatic
events like explosions, mutilations and massacres. However, all these responses
(can) account for combat-stress.
The Russian Army and its Soviet predecessor have been grappling with com-
bat-stress, and PTSD in particular, among its veterans from the Afghan and Che-
chen wars in the 1980s and 1990s. It is hampered by orthodox, out-dated and
prejudiced diagnoses and treatments, though some military psychologists op-
pose the “Stalinist concept of the individual … which considers trauma to be a
personal weakness” (Sieca-Kozlowski, Elisabeth 2009 (2007): par.13 and note 6;
Merridale 2000). The Russian political and military leadership also tend to
downplay or deny debilitating and aggression-inducing traumas among return-
ing “war heroes”. Thus Maj.Gen. V. S. Novikov (medical service) castigated the
poor psycho-physiological selection, training, preparation and support mecha-
nisms of or for military recruits during the First Chechen war of 1994-1996. Ap-
parently, these shortcomings and more intense urban warfare brought higher
rates of PTSD and other combat-stress disorders than during the Soviet occupa-
tion of Afghanistan in 1979-1989 (Norikov 1996; Thomas/O’Hara 2000).26

25
APA’s “Diagnostic Criteria for Posttraumatic Stress Disorder” (shown as Table 1.1 in Schnurr/
Green 2004: 5) is reproduced on my webpage “Trauma and Combat-stress” at <http://sites.google.
com/site/tristansolutions>.
26
In Novikov’s survey of 1, 312 troops, 72% exhibited PTSD or other disorders (Norikov 1996: 37-
38; Thomas/O’Hara, 2000: 49-50). In contrast, 65 of 105 Afghan veterans interviewed in the mid-
1990s exhibited little if any disorders, while 18 veterans exhibited full PTSD (Zelenova et al 2001
(1997): esp. 3, 7).
 The Limitations of Military Psychology 593

In contrast, the US Army has become aware that in most 20th century wars
“more [US] combatants were disabled by stress than were killed by the enemy”
(Grossman/Christensen 2007: 173; on PTSD and treatment: 261-348). Faced with
one-fifth of its Iraq and Afghanistan veterans breaking down, becoming aggres-
sive (murder, assault, drunken driving, etc.) or committing suicide, the US Army
introduces ever more refined programmes to improve the soldier’s performance
and resilience, and thereby reduce PTSD. Surprisingly, however, a mental train-
ing programme based on tested techniques in middle schools, planned to be in-
troduced to soldiers (from October 2009) through “weekly 90-minute classes”,
has only been implemented recently, while psychologists have recommended
such programmes for decades. Even then it will be difficult to “transform a mili-
tary culture that has generally considered talk of emotions to be .. a sign of
weakness”.27 Thus efforts initiated in the 1990s to “change the cultural norms and
beliefs that had discouraged help-seeking behavior in the past” among the US
military still lack definite success: a 2003 survey of US Army personnel in Iraq
shows that “there are still significant problems in .. applying what we know
about the treatment of traumatic stress”; thus “only one-third of soldiers who re-
ported that they wanted help actually got it”—and 63% of all respondents be-
lieved that the “leadership might treat differently those seeking mental
healthcare” (Ralph/Sammons, apud Kennedy/Zillmer 2006: 375, 379 incl. quotes;
see also Campise et al, apud Kennedy/Zillmer 2006: 235-236).28
Even the best programmes cannot prevent the nerve, morale and discipline
of the average and even strongest soldier ultimately to unravel in combat. At
“some level of pressure” even “the most firmly integrated individual will suc-
cumb” (Reinartz, apud Sladen 1943: 271). Yet, to conclude from this given and the
high psychiatric casualty figures that “human beings are very fragile psychic be-
ings” with “no significant personality traits that immunize a soldier against psy-

27
Benedict Carey, “Mental Stress Training Is Planned for U.S. Soldiers”, New York Times, 18 Au-
gust 2009 (quotes). Timothy Williams, “As a Brigade Returns Safe, Some Meet New Enemies”, NYT,
13 July 2010. Greg Jaffe, “Military reckons with the mental wounds of war”, Washington Post, 18 July
2010.
28
During April-December 2003, 23 US soldiers killed themselves in Iraq and Kuwait (Ralph/
Sammons, apud Kennedy/Zillmer 2006: 379). From C. W. Hoge et al (2004), “Combat duty in Iraq
and Afghanistan: mental health problems, and barriers to care”, New England Journal of Medicine,
vol. 351: 13-22.
594 Caspar ten Dam

chiatric breakdown” (Gabriel 1988: 2, 5) underestimates the resilience in the

coping process of combatants and humans in general.
Since the 1990s, proponents of positive psychology, making use of research on
coping by Richard Lazarus, Susan Folkman and others, focus on “human strengths
and virtues” emanating from trauma-recalling, trauma-retelling, trauma-ac-
ceptance, hope, adaptation and other positive coping strategies. They seek to
counterbalance the “traditional pathogenic approach” that exclusively looks at
trauma, PTSD and other pathological disorders with supposedly negative, debili-
tating coping strategies like denial and anger (Ai et al 2007: 55, 56 (quotes); Folk-
man 2011: 3-11).29 Given human resilience and coping, it is not typical that even a
“full range of psychiatric responses” (Gabriel 1988: 17), ranging from panic and
paralysing fear to apathy and suicide, lead to a complete disintegration of a
fighting force, like during the 1876 Battle of Little Big Horn (Gabriel 1988: 19-21).
Most battles are not so desperate or one-sided. Most soldiers recover from com-
bat-stress “in days with appropriate intervention” through both self-help without
the “illness label” and “natural social support” (family, partner, friends)—with-
out incurring debilitating trauma’s (Campise et al, apud Kennedy/Zillmer 2006:
223).
Eagerness to Kill
The given that “not all young men are violent” is an obvious truth. Yet Collins
wrongly criticises other theories on such presumed generalisations: most theo-
ries, including that of Frustration-Aggression, do not propose whole classes of
pathologically determined “violent individuals” (Collins 2008: 1-2). Most atro-
cious acts seem “a reaction to extremely stressful events by personalities unfor-
tunately unable to cope” (Watson 1980: 175) rather than by psychopathic person-
alities. Still, the military often fail to deselect the latter in screenings—or delib-
erately enlist them to commit atrocities. Then psychopaths, sociopaths and vio-
lent criminals could make up large proportions of combat-units. Nevertheless,
many analysts believe that all human beings—not just psychopaths—are inher-

29
Margaret Stroebe refutes that “social sharing .. and .. social support are [always] important”
in dealing with bereavement, and that “denial, repression, and avoidance” are invariably detri-
mental coping-processes (Folkman 2011: 6). I agree that anger, “denial and .. disengagement” do not
necessarily constitute “maladaptive coping” (Ai et al 2007: 58). Still, others speak of “evidence for
experiential avoidance being prominent in … anxiety and mood disorders”, which is “expected to
disrupt .. trauma recovery” (Kashdan et al 2009: 186, 187).
 The Limitations of Military Psychology 595

ently eager to kill or inflict pain on ‘fellow’ human beings: “Deep down in his
subconscious, man seems to enjoy killing” (Dinter 1985: 23).
As indicated earlier, Joanna Bourke perceives rage as a mere precursor or
trigger of exhilaration: “Did actual combat dent the pleasures of imaginative vi-
olence [as shown in literature and films]? For most combatants, the answer must
be ‘no’ ”. When a soldier in Vietnam “went berserk and massacred many of the
enemy, he remembered feeling suffused with joy: ‘I felt like a god, this power
flowing through me . . . I was untouchable.’ ” (Bourke 1999: 30, 31-32 (quotes), her
note 66).30 However, such ‘joy’ may be quite superficial if intense, more often
than not a by-product of the adrenaline rush. Significantly, Bourke inadvertently
provides numerous reluctance-or-guilt-in-killing counter-examples.31
Many kinds of battle ‘joy’ have nothing to do with sadism, but everything
with experiencing relief of having survived a deadly encounter, followed by ex-
hilaration of being alive. Alternatively, the ‘joy’ concerns a grim determination
by “level-headed warriors” (Grossman/Christensen 2007: 139) to do the job, meet
any deadly challenge and test one’s acquired skills in real combat (ibid: 138-139;
166-167 (survivor euphoria) ). Grossman and Christensen thus support an alter-
native eagerness-to-fight thesis: many “perfectly healthy, functional” veterans
“liked it” in Vietnam—even after many tours of duty, and without any apparent
PTSD (ibid 2007: 140).
More importantly, any ‘pleasure in killing’ typically is transient, quickly fol-
lowed by feelings of guilt and long-lasting trauma. Most veterans have at least
once experienced a “brief feeling of elation upon succeeding in killing the en-
emy” at close range, yet were almost immediately “overwhelmed by the guilt
stage” (Grossman 1996: 115 (quote), 234-237, 243-245).32 Many more combatants
may feel immediate remorse once they are pushed to kill or commit atrocities,
without experiencing exhilaration at all: “no one who speaks to many distressed
vets can doubt that their involvement in the excessive violence of Vietnam is a

30
Unnamed Vietnam veteran, quoted in Jonathan Shay (1994), Achilles in Vietnam: Combat
Trauma and the Undoing of Character, New York: 84.
31
For instance, “if they did not enjoy killing, they should not be in the infantry” (Bourke 1999: 3;
note 3); “Never have I forgotten the look on their [dead] faces” (ibid: 4; note 7); “Men were not pre-
pared for the horror of being unable to remove their bayonets from the body of their foe” (ibid: 64),
“only .. ‘most perverted’ could take pleasure in such .. ‘cold-blooded’ .. killing” (ibid: 66).
32
If combatants fail to rationalise the “backlash of remorse and nausea”, it “can become one of
the paths to PTSD” (Grossman/Christensen 2007: 167).
596 Caspar ten Dam

fundamental source of their inner turmoil, and that it expresses not just psy-
chological stress but moral pain” (Marin 1981: 71).33 There is little evidence that
most people can “kill casually and guiltlessly in combat” (Grossman 1996: 88)—
though Grossman later acknowledges that many older, mature combatants do
not experience the “backlash of remorse or nausea” or are better able to deal
with it (Grossman/Christensen 2007: 167(quote)-171). Generally, however, “it
would be surprising if the inevitable release of aggressive impulses in active
warfare failed to produce ... anxiety and guilt” (Rosenberg 1943: 32; from Clark
1946: 423 (quote; note 3)-424).
Nevertheless, I have issue with the reluctance-to-kill notions that: a) only the
two percent of “aggressive psychopaths” or those with “aggressive psychopathic
tendencies” among the armed forces (Swank/Marchand 1946) and their societies
(Grossman 1996) “will kill without regret or remorse”; b) only psychopaths and
sociopaths commit most or all atrocities; and c) war will “psychologically debili-
tate” all of the 98% mentally stable soldiers who “participate in it for any length
of time” (Grossman 1996: 50, 61, 180 (quotes) ). In that regard Bourke’s research
does plausibly show that, apart from psychopaths, a significant minority of men-
tally stable combatants do kill, injure and maim without compunction—often
cruelly so. Thus her eagerness-to-kill thesis seems frequently, if not generally,
valid.
Reluctance to Kill
The military recognise—yet rarely analyse in depth—the widespread phenom-
enon of soldiers refusing, avoiding or pretending to kill. Just “because a soldier
can fight, it does not follow that he will fight” (Appel, apud Sladen 1943: 294).
Remarkably, people generally seem reluctant to resort to any sort of lethal and
non-lethal violence. The dynamics of “confrontational tension and fear” usually
prevent, curtail or sidetrack actual outbreaks of violence; “people are still for the
most part not good at violence” (Collins 2008: 10).
Soldiers are much more willing to fight, i.e. defend positions and fire in the
general direction of the enemy, than directly kill. Grossman refers to original
findings by Ardant du Picq (1821-1870), supported by Paddy Griffith, Richard
Holmes, Richard Gabriel, F. A. Lord and other 20th century scholars, that 80 to 85
percent of soldiers did not fire their weapons during engagements in the Ameri-

33
Bourke unconvincingly claims that she gives the “guilt and .. pain which combatants some-
times felt … due weight” (Bourke 1999: 11; italics added).
 The Limitations of Military Psychology 597

can Civil War (1860-1864) and other 19th century conflicts, overcoming “powerful
conditioning (through drill) to fire”. The few who did, often had a “desperate
urge to fire their weapons” out of fear, nervousness or posture “even when (or
especially when) they cannot possibly do the enemy any harm”. Generally, com-
batants only are willing and able to kill in “kill-or-be-killed circumstances of self-
defense or the defense of one’s friends” (Grossman 1996: 10, 24, 173 (quotes); 5-11,
21-22, 25).34
In his seminal study Men Against Fire (1947) U.S. Army Brig.Gen. (then
Lt.Col.) S. L. A. Marshall estimated that “less than 25 percent of our infantry line
employed hand weapons effectively when under fire” in World War II. Indeed,
the average fire ratio was a mere 15 percent in most infantry companies. Those
with heavy and crew-served weapons performed better (especially in the more
‘aggressive’ companies), like those with officers present ordering them to fire. Yet
Lt.Col. Robert G. Cole observed in a June 1944 engagement that “the moment I
passed on, they quit [firing]”. Significantly, the “active weapons participation” i.e.
fire ratio of the US infantry had risen “beyond 55 percent” in the 1950-1953 Ko-
rean War partially due to Marshall’s findings and consequent training-and-con-
ditioning improvements (Marshall 1978: 9, 72 (quotes); 50-63).
Referring to the US Medical Corps finding that “fear of killing, rather than
fear of being killed, was the most common cause of battle failure”, Marshall fa-
mously propositions that the “average and normally healthy individual” has “an
inner .. resistance toward killing a fellow man” (Marshall 1978: 78, 79). This reluc-
tance is both instinctive (against killing one’s own species) and nurtured, as their
societies typically prohibit “the taking of life” in peacetime (ibid: 78). Thus Mar-
shall criticises the “mistaken doctrine which would have us believe that by re-
peating platitudes and by teaching men to snarl when going at a bayonet course,
we can train them into something other than what they are by nature” (ibid: 154).
A reluctance to kill does not necessarily equal ‘cowardice’ or unwillingness to
engage and take risks: some (or many) who “did not use their weapons .. did not
shirk the final risk of battle. They were not malingerers” (Marshall 1978: 59).
Moreover, they do not have a “demoralizing effect” on those who do fire at the
enemy; rather, the “presence of the former enable the latter to keep going” (ibid:
65).

34
After the battle of Gettysburg, nearly 90% of the 27,574 recovered muskets were loaded with
one, two or multiple non-fired bullets: from F. A. Lord (1976), Civil War Collector’s Encyclopedia,
Harrisburg, Pa..
598 Caspar ten Dam

Still, Marshall recommended realism training-and-conditioning to increase

the fire-ratio of soldiers in battle; even he believed that the innate resistance to
kill could be overcome. Thus the firing rate increased to 90-95 percent in the Vi-
etnam War according to Marshall’s later research, “independently verified” by
[Michael] Scott (Grossman 1996: 35, 181, 251 and his note 2; 344 (quote)).35
Not surprisingly, Bourke questions Marshall’s “extremely dubious” fire-ratio
findings and reluctance-to-kill presuppositions, referring to critiques of Mar-
shall’s research by Roger J. Spiller, Donald E. Graves and others. One reason why
it has taken so long before anybody decided to scrutinise Marshall’s claims is
that “we want to believe that it is difficult to kill other men” (Bourke 1999: 75-76
and her note 24, 112, 399 (quotes)).36 Spiller has been one the first to point out
that a) Marshall varied the number of infantry companies he purportedly inter-
viewed in the Pacific and Europe during WWII from 400 to 603; b) Marshall’s
wartime aide Captain (later Professor) John Westover did not recall Marshall
“ever asking” the question “who had fired his weapon and who had not”; and c)
“Marshall’s own personal correspondence leaves no hint that he was ever col-
lecting statistics” (Spiller 1988: 88 (and his notes 52, 53)). Even if “in earlier wars
there had never existed the opportunity for systematic collection of the data”
(Marshall 1978: 53 (quote); Spiller 1988: 68 and his note 50)—doubtful given Du
Picq’s questionnaires distributed to fellow French officers in the 1860s (Gross-
man 1996: 9-10)—Marshall fails to present any such data himself. Still, even if
Marshall’s fire-ratio estimates are intuitive i.e. based on “his own experiences
and observations” (Spiller 1988: 69), Du Picq, Griffith, Holmes, veterans Lt.
George Roupell (WWI) and Col. Milton Mater (WWII), and others corroborate
Marshall’s reluctance-to-kill thesis “if not his exact [fire-ratio] percentages”
(Grossman 1996: 3-4, 333 (note 1, quote); 12, 27, 34-35).37 Even if his methodology
“does not meet rigorous .. standards”, Marshall’s research has been “replicated
and validated” among soldiers, police and other security personnel (Grossman/
Christensen 2007: 78, 79 (quotes); 161-165, 200).

35
The early 1996 edition of On Killing fails to specify that Scott’s and Marshall’s fire-rate find-
ings can be found in R. W. Glenn (1989), “Men and fire in Vietnam”, Army: Journal of the Association
of the United States Army: 18-27. This is rectified in later editions (Grossman, email 28-09-2011).
36
Marshall “did not interview as many men as he said he did, and not one .. remembered being
asked whether or not he fired his weapon” (Bourke 1999: 76).
37
A 1986 British Defence study actually confirmed Marshall’s fire-ratios in 100 re-simulated
19th and 20th century battles and test trials (Grossman 1996: 16).
 The Limitations of Military Psychology 599

Be as it may, Marshall’s low fire-ratio estimates seem to concern all possible

killing ranges, or perhaps primarily the medium rifle and hand-grenade killing
ranges. Combatants seem most reluctant to kill, injure and maim in the intimate
hand-to-hand, knife and bayonet ranges, and least reluctant to do so in the more
impersonal sniper, missile, artillery and bombing ranges. Despite aggressive
training, most soldiers in close combat use their rifles as clubs, rather than bayo-
net or knife the enemy. Even elite soldiers and commandos hardly ever use the
well-practised yet gruesome kidney-strike and eye-gouge-and-brain-penetration
techniques; they prefer—if they must kill—to slit the enemy’s throat while not
having to see their faces. Even Bourke admits that during both World Wars “less
than one half of 1 per cent of [lethal and non-lethal] wounds were inflicted by
the bayonet” (Bourke 1999: 6 and her note 18). Therefore, overall fighting, firing
and killing rates are lowest on the minimum short-distance range, and highest
on the maximum long-distance range, as “the farther away you are the easier it is
to kill” (Grossman/Christensen 2007: 203). Certain classes of soldiers like com-
mandos secure higher kill rates, especially in hand-to-hand combat. Still, one
must keep in mind that for many combatants engaged in short-range combat,
the survival instinct and the you-or-me rationalisation kick in: arguably the “only
thing greater than the resistance to killing at close range is the resistance to be-
ing killed at close range” (ibid: 201-202, 203 (quote); Grossman 1996: 97-137).
Training
Military (and law-enforcement) training involves drills, physical exercises and
learning the routine handling and maintenance of weapons and other equip-
ment. Ideally, before training even begins, recruits have passed sophisticated
and reliable (pre)screening tests on general suitability for military service, and
suitability for specialist functions in the army, navy, air-force, commandos or
other services like bomb-disposal units (Walters 1968: 5-12). These drills and
skills must enhance strength, discipline and fighting capability, but above instil a
willingness to fight and kill, and a sense of duty, loyalty and identity geared to-
ward group bonding. Recruits already should possess, beyond sufficient intelli-
gence, alertness and dexterity, a paradoxical mix of “aggressiveness, fearlessness,
[and] calmness” (Ryan et al, apud Kennedy/Zillmer 2006: 115; Flanagan 1942).
However, even basic training brings social pressures, being “away from family
and friends” while “feeling command and peer pressures to conform” (Krueger,
apud Cronin 1998: 89). It brings stress-responses like exhaustion due to “adverse
working conditions”, including deliberately invoked noises and extreme temper-
600 Caspar ten Dam

atures (heat, cold) to toughen up particularly elite-commando recruits (ibid: 89-

90). So-called stress-inoculation techniques to toughen and prepare soldiers for
extreme battlefield conditions may become counterproductive if these create
stressors and stress-responses detrimental to physical and mental health. In the
1960s and 1970s, the Human Resources Research Office (HumRRO), the Special
Operations Research Office (SORO) and other Pentagon ‘think-tanks’ subjected
unsuspecting recruits to simulated and real dangers, like a fake ‘engine fault’ in
an airplane or unannounced life-fire in ground exercises, in misguided efforts to
‘inoculate’ them against fear. One should not expose recruits to simulated dan-
gers of which they are not informed about (violating a central criterion of ethical
research), and certainly should not expose them to “dangers equal to those of
battle itself” (Watson 1980: 21-30, 141(quote)-144). The US Army Ranger school
once submitted recruits to weeks, even months, of “food and sleep deprivation”,
yet curtailed this practice after “long-term studies” found that such deprivations
were causing “serious, long-term health effects” (Grossman/Christensen 2007:
26).
Even the best training focused on physical conditioning, with group benefits
like self-esteem and confidence due to acquired fighting skills like marksman-
ship (accurate and quick firing), do not by itself turn soldiers into ruthless, effi-
cient killers, let alone effective combatants: “Discipline is not the key. Perfection
in drill is not the key”; many of the few soldiers who do fight, shoot and kill i.e.
become “lions on the battlefield” are notorious for “laziness, unruliness, and dis-
orderliness” during training (Marshall 1978: 60). These unruly men “could fight
like hell but they couldn’t soldier” (ibid: 61).
More frequent, and more worrisome to the military, were the often highly in-
telligent men and women unable or unwilling to accept the rigours and purposes
of the training camp: “those who fall foul of military law too often have been
found to make poor fighters” (Watson 1980: 105, 140 (quote)). Malingerers,
deserters and draft-dodgers typically make bad soldiers and bad fighters, while
ex-convicts and (other) unruly, insubordinate malcontents often make bad sol-
diers but good fighters. Many malingerers, dodgers and conscientious objectors
had “psychiatric histories” indicating either a complete lack of aggressiveness or
an unconsciously repressed aggressiveness unsuited to military life (and leading
to wartime brutalities if still in active service). They did or could not answer to
expectations from civilian society that masculine-martial “heroes” kill without
hesitation and guilt, while honouring notions of honourable conduct. Early
 The Limitations of Military Psychology 601

studies of those who “adjusted poorly in training camps” focused on discharged

ex-soldiers who had become psychiatric patients, and thus tended to assume
that all ‘unsuitable’—either insufficiently or excessively aggressive—servicemen
were psychotic or ‘psychoneurotic’ (Clark 1946: 424, 425(quotes)-426).
Military establishments were, and still are, reluctant to admit, deal with and
resolve the paradoxical phenomenon of the frustrated, potentially aggressive
and effective, yet disobedient “malcontent”, as it challenges traditional notions
of hierarchy, discipline and command. Still, during WWII the British recruited
such undisciplined and intelligent individuals for elite units such as the Special
Air Services (SAS); to this day SAS recruiters and psychologists do “not want
people who are emotionally stable; instead they want forthright individuals who
are hard to fool and not dependent on others” (Watson 1980: 282).
Conditioning and Indoctrination
The most basic form of aggressive conditioning within training is the bayonet
course. While Bourke deems bayoneting the epitome of human aggressiveness
i.e. eagerness to hurt, injure, mutilate and kill, and bayonet training a mere
honing of this predisposition, Marshall deems the latter a quintessential yet
doomed attempt to instil an aggressiveness that is hardly there (see ‘Eagerness to
Kill’ and ‘Reluctance to Kill’). One only can instil and enhance aggressiveness,
with the greatest effort and expenditure, through transparent conditioning. One
should inform both recruits and (training) officers that a) fear, other stresses and
reluctance to kill are normal responses; and b) conditionings can overcome or
deal with these human responses (Marshall 1978: 36-43, 71, 81-84, 124).
Based on Marshall’s findings and recommendations, the US military, through
HumRRO, introduced a new conditioning technique: shooting in group for-
mations at shifting man-shaped targets during life-fire, simulation, paint-bullet
and other realistic exercises, replacing the old technique of shooting as isolated
individuals at bull’s-eye targets in static lying, crouching or standing positions.
Mainly this innovation is credited for the dramatic increase in the fire-rate (six-
fold) and fire-accuracy (fourfold) of American soldiers in combat missions after
World War II. Nowadays, many soldiers and law enforcement officers practice
their firing skills on “photorealistic” targets and on each other (paint bullets).
They also learn under “precise rules of engagement” not to shoot at a “life-size
photo of a man holding his wallet” and not to shoot wildly too many bullets at
any legitimate target in an ineffective “spray-and-pray” overkill. In contrast,
those who still practice on “blank, man-shaped silhouettes” are “conditioned to
602 Caspar ten Dam

shoot anyone who jumps up in front of them”—which apparently leads to many

unintended killings of civilians and surrendering combatants in combat mis-
sions (Grossman/Christensen 2007: 75, 80, 103-107, 209-212, incl. quotes).
Unfortunately, numerous psychologists have provided the knowledge and
techniques to condition uninformed soldiers into potentially brutal killers—
without providing moral and ethical counter-balances. Particularly stimulus-re-
sponse behaviourists (think of Ivan P. Pavlov and B. F. Skinner) reasoned and
(self-)justified their work as follows: “I earn my living by my ability to predict a
man’s behavior under a given set of stimuli and to influence a man’s behavior by
exposing him to various types of stimuli” (Appel, apud Sladen 1943: 293).38 Thus
Berkowitz and LePage allowed university students to administer electric shocks
to other ‘subjects’ when the latter apparently erred in their tasks—which the
former did more often and aggressively when a gun and a rifle were visible on a
table nearby (Berkowitz/LePage 1967). Grossman and his colleagues insist that
nowadays the “conditioning used to enable killing in soldiers and law-enforce-
ment officers” contain sufficient moral “safeguards” (Grossman 1996: xvi). How-
ever, the history of “modern training/conditioning techniques” (ibid: xvi) suggest
otherwise. Grossman does acknowledge that the conditioning enabling the 90-
95% firing rate in Vietnam came with the “hidden cost” of “psychological
trauma” due to the suppression of the innate resistance to killing (ibid: 250)—
with much of the trauma arising from guilt about atrocities like the My Lai mas-
sacre. American soldiers were desensitised (still including bayonet training) and
indoctrinated against the ‘geeks’. Thus “we may have enhanced the killing ability
of the average soldier through training (that is, conditioning), but at what price?”
(ibid: 291).39
During the Cold War, particularly American, British and Soviet military psy-
chologists developed “behaviour-modification techniques to make soldiers less
worried about killing”, without developing counter-balancing techniques to pre-
vent atrocities (Watson 1980: 23 and note 7: ‘The strange tale of Commander
Narut’, Sunday Times, 6 July 1975). With brutal experiments like forcing recruits
(by Lt.Col. Thomas Narut, US Navy, allegedly to make them assassins operating

38
Appel admits that “this knowledge .. of bringing stimuli to bear might be available to the
Nazi government” (Appel, apud Sladen 1943: 293).
39
See Grossman 1996: 160-164; 190-191 (My Lai); 251-256; 262-280. I have no room here to discuss
his alarming (and perhaps alarmist) views on violent movies and computer-games (ibid: xvi-xviii;
260-261; 299-332; Grossman/Christensen 2007: 81-91).
 The Limitations of Military Psychology 603

from US embassies) to watch horrific films with their eyelids clamped open, they
intended to “completely desensitize the men to pain or suffering, to remove any
emotion .. that might interfere with killing” (Watson 1980: 35-36, 37 (quote) ).
They indoctrinated soldiers to dislike, even hate, potential and actual enemies,
through propaganda movies ridiculing the local customs, beliefs and politics of,
for instance, the North Vietnamese. They developed, for psychological operations
(PSYOP) against these enemies, non-physical yet dehumanising interrogation
methods amounting to torture (and occasionally used them, as the British did
against IRA suspects). Thus HumRRO and other think-tanks in America con-
ducted experiments of sensory deprivation (SD) and sleep deprivation on their
own civilians and soldiers throughout the 1960s and 1970s. Numerous psycholo-
gists, like Thomas Myers at HumRRO with his social isolation and monotony re-
search, studied effects of mental discomfort. The gained insights allowed the US
military—and the British military in Northern Ireland—to ‘improve’ these tech-
niques (involving severe physical discomfort) against potential and actual ene-
mies—and harden their own men against such torture in the event they were
captured: “captivity is fare more disorientating than .. sensory deprivation, but
together they are lethal” (Watson 1980: 207). Watson also suspected that Sig-
mund Streufert and other psychologists studied atrocities like the My Lai Massa-
cre so as to “learn more about killing and to train people to be better at it”, rather
than “to prevent atrocities occurring in the future” (ibid: 183).40
Soviet, Chinese, North-Korean, North-Vietnamese and other Communist psy-
chologists researched and thereby encouraged equally brutal isolation and inter-
rogation methods against dissidents, insubordinates, and enemy prisoners.
These usually lasted shorter, but included food deprivation and intense indoc-
trination called brainwash (from Chinese hsi nao, “wash-brain”) or “mind con-
trol”, i.e. alteration of someone’s behaviour and beliefs through deliberate, ma-
nipulative deprivations (Watson 1980: 208, 211-213, 214-228; Hinkle/Wolff 1956;
Lifton 1962; Hunter 1971). During the Korea and Vietnam Wars, they applied both
subtle and harsh brainwash techniques on captured American and other West-
ern soldiers and civilians, to break their will, gain information, and alter their
loyalties and political convictions (extreme indoctrination), often successfully
so. Due to institutional conservatism, the American military kept teaching their
recruits how to “stand up to pain, or isolation, [rather] than .. up to persuasion”

40
See further Watson 1980: 179-183, 199-208; Grossman 1996: 257-261.
604 Caspar ten Dam

(Watson 1980: 225). In contrast, the CIA became as notorious for ‘peace-time’
brainwashing as its Communist counterparts, and allegedly became even more
sophisticated in it. Even though many CIA ‘brainwash’ missions were miscon-
ceived or ineffective, “the ability to do something almost invariably means that
an attempt will be made to actually do it” (ibid: 210). The thousands of convicts
the United States recruited in WWII were not submitted to such desensitising
and dehumanising experiments—and had not been “convicted of murder, rape,
kidnapping and treason” (ibid: 110); that time, the US military did not actively re-
cruit those who had committed grave, violent crimes. Yet Lt.Col. Narut’s recruits
included convicts from military prisons, even “convicted murderers” (ibid: 182).
The fact that Narut and his colleagues were allowed to recruits such criminals, at
least on a small scale in the US Navy during the early 1970s to condition them as
secret commandos into remorseless killers, points to institutional, authorised
brutalisation in the US armed forces—long before the post ‘9-11’ War on Terror.
Some military psychologists and officers do warn against overly aggressive
training, conditioning and consequent brutal behaviour in military missions;
they also warn against their profession’s involvement in these practices “without
clearly delineated ethical or legal parameters” (Williams et al, apud Kennedy/
Zillmer 2006: 206). Aggressive conditioning may contribute to or at least coexist
with moral laxity. Thus the abuse against Iraqi prisoners in the US-run Abu
Ghraib prison—shocking photo’s released in April-May 2004 produced a public
outcry and urgent US Congress and US Army investigations—were partially due
to a “laissez-faire attitude of leaders .., a lack of training of the prison guards, lack
of discipline, and the psychological stress of being in constant danger over an ex-
tended period of time” (ibid: 203). The “medical professionals' participation in
intelligence-gathering” (ibid: 207) at Abu Ghraib and Guantanamo Bay, Cuba
(and other places), is equally alarming (Bartone 2004; Bloche/Marks 2005).
Therefore, without professional training focused on moderation, self-discipline
and (self-)critical awareness of one’s duties, soldiers easily may commit brutali-
ties—especially if they are conditioned and indoctrinated into remorseless kill-
ers while being treated “as if they were children or serfs”, as mere cogs in the
military machine (Marshall 1978: 114-115).
Group Convictions, Bondings and Expectations
The group identities and bondings of the military unit (from platoon level up-
wards) lead to expectations of the individual soldier or any combatant towards
oneself and one’s fellow soldiers and combatants. These expectations typically
 The Limitations of Military Psychology 605

concern societal values of loyalty, courage and comradeship, all to do with main-
taining individual and collective self-respect, even under the severest dangers
and privations. Ideally, at least in the minds of political and military leaders,
these expectations have been present if dormant in the original group identities
and convictions of family, clan, tribe or nation that nurture notions of honour
and destiny. Then these should be awakened, solidified and enhanced through
ideo-political indoctrination and military training. Interestingly, Harold Gerard
et al found in their experiments that the “greatest amount of conformity oc-
curred in groups with five members” (Watson 1980: 95 and his notes 9-10; Gerard,
et al 1968); the US army constructed eight-member squads for logistical reasons,
but likewise discovered in Korea that four-member training produced the
strongest bonding and morale. Consequently, military leaders and psychologists
expect that the combat unit “increases the individual’s endurance and courage
by challenging him to uphold his self-esteem” (Weinberg 1946: 473).
The military regard obedience to command as a vital component of unit co-
hesion. However, it “may be extremely disruptive to appeal to a sense of duty to
the group (e.g. organization, unit) as a means of convincing the soldier/officer to
disregard other loyalties and obligations (e.g., to his dead buddies, his family)”
(Bartone, apud Cronin 1998: 135).
Scholars generally agree that the Homo Sapiens is a “social animal”. The basic
human need of acquiring group contact and approval accounts for the saliency
of group convictions, group bondings and (consequent) group expectations in the
civilian and military spheres, even in cultures where individualism is both
preached and practised. Indeed, morale and leadership play crucial roles in these
group processes.41 Especially group-bonding should be a stress reducer. The en-
hanced bonding through training, conditioning, indoctrination, and above all in-
tense comradeship, should make unit cohesion the “critical factor that moder-
ates or buffers the impact of combat stress on performance” (Bartone, apud Cro-
nin 1998: 118 (quote); Manning/Ingraham, apud Belenky 1987). Even if “army
units break up .. or flee, old comrades will try to remain together. This explains
why units assembled quickly in an emergency under a strong commander could
fight with considerable success” (Dinter 1985: 7).

41
I do not have room to analyse morale and leadership at length here, even though these con-
cepts are extensively discussed in the military literature.
606 Caspar ten Dam

Loyalty to and protection of one’s closest ‘buddies’, combat unit, and family
at home supersede those to one’s military command, political system or nation.
Therefore, ‘light’ political indoctrination unaccompanied by ‘heavy’ brainwash-
techniques hardly seems to increase a soldier’s willingness to fight, invalidating
“exaggerated notions of man’s capacity to endure and to sacrifice on behalf of
ideals alone” (Marshall 1978: 154). Remarkable instances of fanaticism like Na-
zism, Stalinism and religious fundamentalism aside, soldiers generally “do not
fight for a cause but because they do not want to let their comrades down” (ibid:
161; from Maj.Gen. T. de la Mesa Allen, WWII Sicily). In order to cope with the
stress, aftermath and possible guilt of killing, soldiers need the “praise and ac-
ceptance from peers and superiors. Warriors do not fight for medals, they do it
for their partners, buddies and friends” (Grossman/Christensen 2007: 170-171).
Even so, comradeship, unit cohesion and unit performance are strengthened if
soldiers not only are aware of and agree with the objectives of the war and the
campaign, but also are apprised of the strategic and tactical situations by their
superior officers. They should be encouraged to tell their superiors of their own
observations and assessments (“rearward flow of information to higher head-
quarters”): information is the “soul of morale in combat and the balancing force
in successful tactics” (Marshall 1978: 92).
However, group convictions, bondings and expectations also, at the same
time, can act as stress enhancers. These group pressures also may account for a
greater tendency toward cruelty and ‘berserk’ violence by groups as opposed to
isolated individuals, even without aggressive conditioning. West Point experi-
ments in the 1960s found that “people in groups get more worked up than people
alone. Given the opportunity, they are far more cruel and aggressive … and calm
down more slowly” than separated individuals (Watson 1980: 93-94 and his note
6 (some errors in source-reference); Baker/Shaie 1969: 80). Individuals in groups
either use the group-anonymity, group-approval and group-bonding dynamics to
commit cruelties (and killings per sé) they already were itching to do, or are thus
pressured by leaders and others to do so. If “an individual is bonded to the group,
then peer pressure interacts with group absolution in such a way as to almost
force atrocity participation” (Grossman 1996: 225).
In sum, group pressures constitute two-edged swords that could both en-
hance and reduce combat-stress and consequent brutality. Marshall acknowl-
edges that “social pressure, more than military training, is the base of battle dis-
cipline”, though he prefers the “simple statement that personal honor is the one
 The Limitations of Military Psychology 607

thing valued more than life itself by the majority of men” (Marshall 1978: 149).42
Be as it may, fear of losing contact with or approval of the group is a primary
stress-response. The challenge is to nurture the soothing and energising benefits
of belonging to a primary group, like the combat unit, without encouraging bru-
tality.
Concluding Observations
Combat-stress may not lead to breakdowns and ‘berserk atrocities’ as commonly
as commonly thought. American military researchers discovered in the 1960s
that “out of every three individuals predicted by tests as likely to break down in
battle, only one actually does so. Conversely, only one half of those who do break
down are recognizably predisposed on the tests” (Watson 1980: 140 and his note
5; Kern/McFann 1965). Even today’s most refined aptitude tests have limited
predictive power. Social pressures (unit loyalty, bonding), coupled with warrior-
ethos expectations, may account for the resilience of so many combatants de-
spite their ‘psychiatric histories’. Only the most recent assessment and selection
(A&S) tests begin to recognise and incorporate these ‘stress-inhibiters’; even
then these A&S tests tend to regard interpersonal attributes like teamwork and
team-adaptability as individual rather than group-induced traits (Picano et al,
apud Kennedy/Zillmer 2006: 357-359). On the other hand, these inhibiters do not
prevent or counteract traumas indefinitely, leading to post-traumatic stress dis-
orders long after the events. In that sense, a high-strung warrior cult, as evident
among Chechens and Albanians, enhances rather than reduces stress (see sec-
tion IV). Social pressures created through training, conditioning, indoctrination
and other mechanisms are intended to enhance combat performance and man-
age combat-stress; yet ironically, these very pressures often, sooner or later, un-
dermine that performance and contribute to that stress. Such pressures make
combatants more resilient in the short and medium terms, yet more high-strung
and eventually susceptible to (sudden) breakdown in the longer term.

III. LIMITATIONS IN MILITARY PSYCHOLOGY

Arguably, the field of military psychology contains some of the strongest pro-
(Western-)state biases known in the social sciences. True, due to government
prioritisation, funding and large test populations (soldiers), it innovated its sis-

42
Yet when “other men flee, the social pressure [of showing no fear] is lifted” (Marshall 1978:
150)—a major reason why panic can spread so quickly.
608 Caspar ten Dam

ter-discipline civilian psychology in education, industry and healthcare research

(Cronin 1998: 3-4; Yerkes 1919: 87, 92). What “we have learned about soldiers in
combat has opened the door for understanding violent situations in general”
(Collins 2008: 6). However, the field often lacks scientific independence and ob-
jectivity: most military psychologists work for a government, are employed by
them, or receive military ranks. Most of them perceive home-grown and foreign
non-state actors as enemies under the counter-insurgency rubric, whose mentali-
ties they dissect in order to defeat them. Few psychologists even-handedly study
rebels and other armed non-state actors, let alone willingly select and train their
fighters. Consequently, one hardly knows whether or to what extent the plethora
of findings on pro-regime combatant stress directly applies to anti-regime com-
batant stress. Do rebels behave like or unlike soldiers? Do typical or frequent (yet
not universal) circumstances of rebellion—waging lightly-armed guerrilla (hit-
and-run fighting) in urban and rural areas, living underground, being harassed
by government forces, etcetera—constitute intervening stressor-variables that
make insurgents behave differently from soldiers? Can non-state groups never
muster the organisation, training and discipline that all state armies supposedly
possess (Henriksen/Vinci 2008: 88-89)? And if insurgents do have more conven-
tional (territory-occupying) and heavy-armour capabilities (like General Fran-
co’s forces in the 1936-1939 Spanish Civil War), will they exhibit the same com-
bat-stresses as ‘normal’ soldiers? Military psychology as a discipline currently
lacks the body of research to confidently answer these questions.
Like their Western counterparts, Russian psychologists mainly investigate the
combat-stresses among their “own” soldiers, and hardly ever among the “alien”
rebel “enemies” (eg. Norikov 1996; Zelenova et al 2001; Sieca-Kozlowski 2007).
Thus (pro-)regime, state-employed psychologists often make prejudiced, unrelia-
ble and invalid statements about the behaviour of non-state fighters. Thus
American military researchers adopt the grossly-generalised assessment of their
Russian counterparts that “mutilations or torture were commonplace against
Russian prisoners” by Chechen rebels (Thomas/O’Hara 2000: 46-47).

COMBAT-RELATED STRESS AMONG NON-COMBATANTS

Nowadays many scholars do investigate, through meticulous surveying (inter-
views, questionnaires) and statistical analysis, the traumas, PTSD, depression
and other post-conflict stresses among Chechen, Albanian and other actual and
former refugees from the Russo-Chechen wars (eg. De Jong et al 2007; Maercker
 The Limitations of Military Psychology 609

et al 2009) and Serbo-Kosovar wars (eg. Eytan et al 2004; Roth et al 2009). The lat-
ter studies seem more numerous, perhaps because Kosovar refugees (research-
populations other than or including refugees: Gordon et al 2004; Ahern et al
2004; Wenzel et al 2009) are more accessible given the relatively short war in Ko-
sovo, with Serb authorities having fewer opportunities to obstruct research than
their Russian counterparts.
Some analysts suggest that some surveyed people may be former combatants
(Favaro et al 1999: 306-308; Ahern et al 2004: 769 (no “further information on
their roles in .. combat situations”); Wenzel et al 2009: 242 (“half of the sample
had been in combat situations”) ). Others exclude them from the surveys (Ai et al
2007: 56 (“Kosovars in this study were not soldiers”); Morina et al 2008, 2010, 2011;
Kashdan et al 2009). Yet none of them did investigate rebel-veterans as a distinct
group among Kosovar or Chechen refugees and other war-affected people (eg.
Eytan et al 2006; De Jong et al 2007; Maercker et al 2009)43, let alone compare
their stresses and traumas with those who (already) were civilians in the con-
flicts.44
Some values may hamper aggression, and thereby actually exacerbate (com-
bat-)stress and trauma among both combatants and non-combatants. In “espe-
cially traditional societies such as Kosovo, rape is a fundamental dishonour in
which the individual is shamed … and the family/community which has failed to
protect the woman is similarly shamed” (Kellezi et al 2009: 63). Victims usually
isolate themselves by not telling their partners, families or outsiders of their or-
deal—“to do so would broadcast their violation of group norms and compound
their isolation” (ibid: 63). Consequently, rape victims—just like combatants who
violate “group values (say in hiding or running away from the enemy)” (ibid:
63)—will become more traumatised. The “primary appraisal” of (ethnic) group

43
Thus Eytan et al neglect to identify ex-combatants among the 996 Kosovars interviewed in
late 2001, if only to verify whether combat roles created different traumas, coping processes and
healthcare use (Eytan et al 2006).
44
I have been able to ask half a dozen authors (eg. A. Maercker; K. de Jong; N. Morina) whether
one can distinguish (ex-)combatants from their data, and whether they or others are conducting
psychological-diagnostic research on armed non-state actors. They responded that their data were
either non-specific on former rebels in or from Chechnya or Kosovo, or that they had excluded the
latter from their data. Few if any of them were planning to analyse stresses and traumas among ex-
insurgents, or knew of anybody planning to do so. Only Dr. Thomas Wenzel and colleagues had
“finished a project on ex combatants .. and work on the publication” (email, 15-01-2012). Thus one
should undertake further research “to fill this .. gap in knowledge” (email to authors, 14-01-2012).
610 Caspar ten Dam

identity is negative (rape violates and endangers it), making the “secondary ap-
praisal” of social support negative (i.e. minimal or non-existent); only “positive
individual coping strategies” may prevent or curtail “negative mental health out-
comes” (Kellezi et al 2009: 59, 63 (quotes) ). Ironically, if Kosovar(-Albanian) so-
ciety really blames and stigmatises the rape victim, it prevents (male) partners
and family-members to openly take (blood-)revenge against the actual or per-
ceived perpetrators, and thereby restore their honour; such revenge can only be
done in secret, if at all.

COMBAT-STRESS AND VIOLENCE-VALUES AMONG NON-STATE COMBATANTS

Presently, one of the few avenues available to say something tangible about
combat-stress (the Brutalisation theory’s third variable) among Chechen and Al-
banians, is to look at how their violence-values (the Brutalisation theory’s first
variable) affect their stressors and stress-responses. Unfortunately, looking at vi-
olence-values as either stress-reducers or stress-enhancers (or both) can only
partially compensate for the astounding lack of research on combat-stress
among armed non-state actors. The horrors of war must have put Chechen and
Albanian insurgents under immense pressure, and probably led to PTSD and
other psychological disorders among even (or especially) the bravest of them.
Outside observers and volunteers rather than rebel veterans have been ready to
talk about this. Former Dutch Marine Marco van Eekeren, who as a thirty-three-
year-old ‘Major’ led the ‘Pitbulls’ elite-unit of the Kosovo Liberation Army (KLA;
or Ushtria Çlirimtare e Kosovës, UÇK): “Clean wars do not exist. War is always an
unorganised mess, yet Kosovo was extreme. ..... I was not a hero, as all heroes are
dead. I shit and peed in my pants from fear …. That had never happened to me
before in the other wars. If necessary, I went out with my bare hands and came
back with weapons; a normal soldier does not do that. When I realised this, I
knew that ... this were to be my last war” (Ruigrok 2000: 34 (translated from
Dutch)).45 Yet apart from heavy fighting in summer 1998 and spring 1999, Koso-
var-Albanian rebels experienced little attrition, exhaustion and other combat-

45
Van Eekeren claimed that he killed several Iraqi’s in the 1991 Gulf war, fifty-two Serbs in the
Croatian war, forty-nine Serbs in the Bosnian war and fourteen Serbs in Kosovo, apparently all
(para)militaries. He later neutralised mines in Prizren for the NATO-led Kosovo Force (KFOR): “I ..
cope with the war by doing good things” (Ruigrok 2000: 34). For a sceptical view on his claims, see
Arnold Karskens, “Heldenstatus zonder een schrammetje (Hero status without a scratch)”, DePers,
14 October 2009, archive <http://www.depers.nl>.
 The Limitations of Military Psychology 611

induced stresses over an extended time period, in stark contrast to their Che-
chen counterparts. Be as it may, little seems to be known about the stressors and
stress-responses (shock, fear, fatigue, trauma, etc.) among Chechen and Alba-
nian insurgents.

IV. CHECHEN AND ALBANIAN COMBAT-STRESSES AND VIOLENCE-VALUES

The martial (violence-)values among the still traditional, clan-based communi-
ties of the Chechens and Albanians may function as simultaneous stress-reduc-
ers and stress-enhancers. Violence-values are perhaps the most salient indicators
of group convictions, bondings and expectations. Here, I focus on the stress-en-
hancing properties of violence-values, if only because little is known about other
stress-responses among rebels. This is not to say that combat-related values,
pressures and traumas inevitably or invariably lead to brutalities, or lead to bru-
talisation among all Chechen and Albanian combatants all the time. For one
thing, their significant bouts of terrorism, banditry and (organised) crime gener-
ally pale, in severity or scale, in comparison with the systematic atrocities com-
mitted by Russian and Serb combatants. Still, combat-related values, pressures
and traumas can account for many of the rebel brutalities—and help to explain
the brutalisations that parts of the Chechen and Albanian communities did suc-
cumb to.

GROUP PRESSURES ON CHECHEN COMBATANTS

In traditional-patriarchal societies one has little room for personal freedom.
Chechen youngsters are pressured to obey their parents, kinsmen, elders and
councils, to unconditionally implement the “social and moral code” (Zelkina
2000: 16). Duties to friends are “as exacting” (Jaimoukha 2005: 92). A religious el-
der once remarked that as the teip (clan) is nowadays “too large” and “dispersed”,
what “matters is the family”, which decides “whether a youth goes to fight”
(Lieven 1998: 345). The family is the “raison d’être of the individual”, while a fa-
ther could show little public or private parental affection i.e. imitate “a woman”
(Luzbetak 1951: 44, 151).46 Such psychological deprivations may easily exacerbate
combat-stress.

46
The sub-clan or “joint-family” is paramount (Luzbetak 1951: 194), reflecting a tendency to
supplant “sib [clan, tribe]” solidarity (ibid: 51, note 58).
612 Caspar ten Dam

Brutalising honour-stress
Most suffocating are high jigit (brave-man) expectations, like the taboo on feel-
ing or showing fear. Harbouring fear just the ‘size of an ant’ in one’s heart makes
one an outcast. Even old-fashioned Chechens privately admit that their exploits
do not constitute natural courage, but a face-saving device. Beybulat Taymiev,
leader of the 1825-1827 Greater Chechnya revolt, to Russian poet Alexander
Pushkin: “I am afraid of shame and that is why I am always on guard. No, I am
not bold” (Souleimanov 2007: 27 and his note 22). Other leaders never allowed or
acknowledged fear. Imam Shamil, leader of the 1834-1859 rebellion in the North
Caucasus, punished his mother for defying his call that Chechens fight the Rus-
sians ‘to the last man’—yet took thirty-nine of forty whiplashes upon himself.
Chechens have a saying for this relentless pressure to self-sacrifice (yakh, ac-
ceptance of duty): “it is tough to be Chechen”. Self-restraint (özdangalla), ironi-
cally meant to ensure civility and prevent feuding, acts as a pressure-cooker,
bound to explode during endemic inter-group rivalry. Chechen society lacks hi-
erarchies i.e. classes; competition occurs among clans and other kinship groups
instead, and tends to be violent given the martial tradition (Souleimanov 2007:
31-32 (his notes 20, 31, 32) ); Luzbetak 1951: 146-147). Kin quality, i.e. kin ancestry,
size, prestige and honour, appears to determine one’s status, not class (though
Derluguian may disagree).
Kinship groups are not culturally equal. Clans certainly were treated differ-
ently in customary law; once, a killer could compensate a victim’s family of a
powerful clan with sixty-three cows, yet one of a weak clan with just twenty
cows (Jaimoukha 2005: 137). All this provides the rationale for violence against
weaker clans and ruthless attempts by the latter to rise in the pecking-order. This
coincided with weakening ancestral bonds and territorial ties, despite na-
tionalists trying to exclude ‘foreign’ and ‘impure’ lowland tribes and clans from
the nation (k’om).47 Still, the Soviet Union’s collapse drove Chechens, like Albani-
ans during Yugoslavia’s disintegration, to refortify their clans so as to preserve
order—even via blood-feuds.
Consequently, youngsters and adults are pressured to excel: in “their eager-
ness to secure fame” the “more ambitious .. pushed valour” to “their utmost lim-
its” (Baddeley 2005 (1903): xxxvi). Youths are constantly reminded that “honour

47
Term süli (foreign) was “expanded to include .. Dagestanis” and “label Chechen clans or fami-
lies .. not of Chechen origin” (Souleimanov 2007: 318, note 7; 319, note 28).
 The Limitations of Military Psychology 613

and toil make the man” and “idleness in summer” brings “torment in winter”;
they must meet contrary principles of sacrificial bravery (“Rather than live like a
chicken, it is better to die a cock”) and tactical caution (“Rashness is folly”) (Law-
son/Jaimoukha, apud Jaimoukha 2005: 242, 243, 244). Such group expectations
lead to potentially brutalising honour-stress among youngsters with fragile self-
esteem.
Frictions between personal freedoms and kinship-nokhchalla (honour-codes)
obligations of bravery, mutual help, assistance and work (belkhi), based on duty
(bekhk) and integrity (haenal), spawn rather than prevent violence. Analysts like
Rieks Smeets believe that inter-clan solidarity and eagerness to avoid blood-
feuds keep the number of intra-ethnic fatalities, such as those of pro- and anti-
Dudaev forces in 1994, relatively low (Smeets 1995). Yet many Chechens, espe-
cially those from minor or ‘impure’ clans (including those made up of former
slaves)48, felt compelled to prove their valour in spectacular acts as smertniki
(suicide fighters) against non-Chechen enemies, Russians in particular. Such vi-
olence reveals double brutalisation, i.e. discarding of both international and tra-
ditional norms: war-traumatised youngsters came to reject customs and adore
brute strength (Souleimanov 2007: 22 (his note 7), 24-26, 31 (his note 30); Jai-
moukha 2005: 90, 93, 134-135).
Another group-expectation has affected Chechen men, both young and old:
the safeguarding of one’s family-members—and avenging them if the latter are
killed, harmed or humiliated. The Chechen obligation to avenge cruelties, injus-
tices or ‘mere’ insults by “foreigners” is particularly painful and hazardous: “If a
man takes up arms and joins the separatists” he “leaves his wife, children and
family without a protector. .... If, however, he decides to .. give up revenge and
dedicate himself to his family, he ceases to be a “Chechen” ” (Souleimanov 2007:
273). That will bring utter shame; one already is ‘in’ shame so long as retribution
remains unfulfilled. The “custom of the blood feud will by no means resolve their
moral dilemma” (ibid: 273). Indeed, this custom brings it about; this is worsened
by the likelihood that he and his family will be victims of a Russian zachistka
(sweep operation), even after deciding to face the ridicule and stay with his fam-
ily. Russian brutalities and local customs steer many a Chechen into an impossi-
ble ‘Catch-22’ situation. This pressure-cooker, highly similar to that of honour-

48
Clans who originated from former slaves, who often stayed at the localities of their onetime
owners, probably had the same urge to prove their valour—and thereby upgrade their status—as
other weak clans (Inozemtseva 2010: 21-24).
614 Caspar ten Dam

stress in battle, may explode into orgiastic violence, the “commission of desper-
ate, even unfathomable acts” (ibid: 274) whenever Chechens see ‘appropriate’
targets to vent their rage.
Criminalising honour-stress
Brutalising war and crime, rather than ‘excessive’ individualism, undermined
age-old bonds of kinship, elder authority, hospitality and honour (Zelkina 2000:
16, 44-45). The “discipline of honour and shame” (Lieven 1998: 345) among Che-
chen fighters could not stem these effects, in a way germinated them, eroding in-
ter-clan solidarity. Families vouched for a mafioso’s recruits, as otherwise both
would be shamed: “you rarely find Chechens killing each other” (ibid: 345) as
even gangsters rarely kill those known to have families and clans ready to retali-
ate. Yet Ruslan Labazanov swore blood-revenge against Johar Dudaev when in
June 1994 three decapitated heads were displayed in Grozny, including that of his
cousin Arbi; some fighters at his mother’s house in Argun wore black bands to
signify this oath (remarkably, also an Albanian custom). Indeed, “clashes be-
tween criminal groups contributed to a major return of the blood-feud” (Lieven
1998: 351) in the 1990s. Elder Haji Mahomet: “Youth hit each other more easily,
use their weapons more easily” (ibid: 29). Natasha, a hospital doctor: youths are
“ready to fly off the handle, and too many leaders .. encourage them” (ibid: 78);
“when I see young kids ready to attack tanks almost with their bare hands, it
makes me cry” (ibid: 79). Despite “reckless acts of nobility and generosity” (Gall/
De Waal 1997: 198) early in the war to protect foreign journalists, kidnapping
(umykanie)—once a feared yet circumscribed practice to capture enemies or
prospective brides as opposed to ceremonial nuptial ‘abduction’ (Zelkina 2000:
45)—resurfaced as a criminal activity to earn ransom, whereby captors treated
captives far worse than traditionally allowed. This brutalisation surpasses the
ruthless-yet-regulated umykanie of earlier times: “in the eighteenth century ..
Chechens would mercilessly rob strangers and .. demand a ransom. If the victim
happened to be a fellow-Mohammedan, he .. could not become a slave if he
failed to be ransomed, and was .. murdered. .. if the captive was a non-Muslim
and .. of a poor family .. unable to ransom him, he had to remain .. a slave”
(Luzbetak 1951: 61 (and his note 128)).49 Indeed, throughout the Northern Cauca-
sus the “overwhelming majority of the slaves were .. non-Muslims or Muslim

49
J. Reineggs (1796-97), Allgemeine historisch-topografische Beschreibung des Kaukasus, vol. I,
Gotha: 40.
 The Limitations of Military Psychology 615

Shi‘ites, as, according to the precepts of Islam, co-believers cannot be taken into
servitude” (Inozemtseva 2006: 186-187). Be as it may, during the 1990s bridal kid-
napping degenerated into “lawlessness and rape” as “young males knew nothing
of the restraints” of “traditional courtship” and “possessed no notion of legality”
(Derluguian 2005: 45).
Religious honour-stress
Youths across the Caucasus are turning to orthodox Sunni Islam to deal with
war(-torn) and socio-economic deprivations, defying the folk-Islam syncretism
of their parents. So-called Young Muslims oppose the “complex mystical world-
view of “standard” Sufi Islam”50, adat (customary law) and other pre-Islamic cus-
toms, leading to confrontations with traditional elders and imams. Even though
“the youth and the elder generation managed to achieve peaceful coexistence” in
recent years, a sub-set of mostly young ‘Wahhabi’ or Salafi51 extremists “declare
all Muslims who do not share their point of view unfaithful” and consider a jihad
against them justified (Yarlykapov 2007-8 (2006): 10-11, 28, 29). Still, the “label
“Wahhabite” is undeservedly hung” on all New or Young Muslims (ibid: 11). Pray-
ers of young expatriate Chechens exhibit Hanafi and Shafi elements, while older
compatriots consistently fulfil Shafi rules (ibid: 15); it is unclear whether this sig-
nifies youthful ignorance or a Young-Muslim departure.
It remains to be seen whether peaceful Young Muslims are able or willing to
curb religious(ly inspired) violence—and counter the lure of the ‘Caucasus Emir-
ate’ encompassing all North-Caucasian republics and other parts of Russia, de-
clared in late 2007 by Dokka Umarov, then ‘President’ of the non-recognised
Chechen Republic of Ichkeria (Noxçiyn Respublika Noxçiyçö, NRN). Indeed,
“what originally had been mostly a Chechen nationalist movement finally trans-
formed itself into a multi-ethnic force where the liberation of Chechnya from
Russian rule became just one aspect of the struggle” (Shlapentokh 2010: 118
(quote)-121). Consequently, NRN’s secularist government-in-exile in London no
longer recognised ‘Emir’ Umarov as its President, but is generally considered too
weak to affect events on the ground. Umarov’s predominantly Jihadist fighters

50
E. Souleimanov, “Islam as a uniting and dividing force in Chechen society”, Prague Watch-
dog, 13 August 2005, <http://www.watchdog.cz/> (accessed 26-03-2006, revisited 07-04-2012; NB:
site not updated since June 2010).
51
Elsewhere, I discuss the similarities and differences between the purist-Orthodox Sunni
Wahhabiya and Salafiya schools, and explain why Jihadists in Chechnya prefer to call themselves
Salafis (Ten Dam 2011: 245-246).
616 Caspar ten Dam

seek to expand the war across the Northern Caucasus—and they may yet suc-
ceed, even after Russian forces or secret agents apparently have killed Umarov in
March 2014 or earlier (details about the time and circumstances of his death re-
main murky).52
Even so, the price of any Emirate ‘victory’ may be high: numerous eventual
PTSDs and nervous breakdowns among Umarov’s Jihadists due to the severe
demands of Salafism. Sheik Said Abu Saad Buryatskii (Aleksandr Tikhomirov),
the “driving force” behind the Riyadus Salikhin (“Gardens of the Righteous”) sui-
cide martyrs’ battalion revived by Umarov in April 2009, argues that these RS
martyrs “went on their death missions contentedly and with determination ra-
ther than confused, drugged or in fear” (Hahn 2009: 4).53 Yet behind this mar-
tyrisation lies extreme indoctrination, conditioning and overall group-pressure
on the ‘voluntary’ mujahideen, as implied by Buryatskii’s unwitting admission
that he “talked about this for hours with prospective suicide shakhids [martyrs]”
(ibid: 3).54 Such stressors would probably lead to PTSD and other disorders
among (m)any prospective and actual mujahideen who somehow survive their
suicide missions; perhaps some backed down due to fear and other stress-re-
sponses.

GROUP PRESSURES ON ALBANIAN COMBATANTS

In principle every Albanian is free to decide whether to take a besa (honour-
oath) to deny an accusation (in tribal court), seal a deal, avenge someone or
confront a common threat. Every man (increasingly also every woman) is also
free to decide whether to keep or break it: “Wash your dirty face” or “make it
blacker” i.e. “You are free to be faithful to your word; you are free to be faithless

52
Reportedly, Ali Abu-Mukhammad (Aliaskhab Kebekov) from Dagestan succeeded the de-
ceased Umarov later in March. Liz Fuller, “Latest Report Of Umarov's Death Leaves Details Un-
clear”, Radio Free Europe/Radio Liberty, 18 March 2014; “North Caucasus Fighters in Syria Pledge
Allegiance to Umarov's Successor”, RFE/RL, 31 March 2014, archive <http://www.rferl.org/> (all ac-
cessed 30-05-2014)..
53
Original source: Said Abu Saad Buryatskii, “Istishkhad mezhdu pravdoi i lozh’yu,”
Hunafa.com, 9 December 2009, 1:01, <http://hunafa.com/?p=2514>.
54
Buryatskii’s “assertion that he will do everything possible to get more mujahedin to join a
suicide martyrs’ unit .., contradicts his claim that he is not pushing mujahedin to suicide martyr-
dom operations” (Hahn 2009: 5). Buryatskii was killed by Russian forces on 2-3 March 2010 in In-
gushetia (Hahn 2010).
 The Limitations of Military Psychology 617

to it” (Gjeçov 1989 (1933): §518-519, 530, 537).55 Yet in practice few dared to
circumvent or violate such pledges, if only because one’s kin was a central object
of loyalty, as among Chechens. They felt the “mark of dishonor” more acutely
than any “fine for perjury” in Kanun (customary “mountain” law) cases of theft
and other minor crimes; only public ostracising as punishment for serious crime
brought similar shame (CLD §589; CLVI ‘The Ban’ (also eg. §1157)). Similarly,
those affronted have their honour violated—and remain themselves dishon-
oured unless they either pardon the perpetrator or (try to) kill him: “Forgive ..;
but if you prefer, wash your dirty face” (CLD §595; 598).56 However, many Albani-
ans, even among Northern-Albanian Gegs, forgot the Mountain laws during
Communist rule in Albania and Kosovo; nowadays ‘reinvented’ “kanun and the
elders become an important device for the younger generation only in times of
immediate need” (Schwandner-Sievers, apud Schmidt/Schröder 2001: 104).
Moreover, few Albanians ever politicised their Muslim beliefs for nation-build-
ing projects after World War I, let alone seek to establish a Caliphate fully under
Shari’a law. This is in contrast to Chechens who, often out of “ignorance or des-
peration” brought about by the latest wars (Ten Dam 2011: 244), came to follow
purist imams and their dogma’s from abroad. Due to their strong secularism
(Ten Dam 2010: 352-353), Albanians generally do not experience a ‘religious
honour-stress’ as observed among many or most Chechens.57

Brutalising honour-stress
Though Albanians have more leeway than Chechens to interpret besa (honour,
honour-oath) and kinship obligations independently from their elders, these still
sustain customs like blood-revenge. Unlike the average Chechen, the average Al-
banian does not lose his honour or prestige (ndera) as soon as he shows fear—as
long as he masters it, and does not commit suicide.58 Yet like the Chechen, the
Albanian “who has been dishonored is considered dead” (CLD §600) if he does

55
References to Gjeçov’s paragraphed (chapter) provisions I identify by CLD, the abbreviation
of the book’s English title, The Code of Lekë Dukagjini.
56
Yet these forgiveness-clauses contradict those stating that “offence to honor is never for-
given” (CLD §597), and that “blood is never unavenged” (CLD §917).
57
Thus there is no sub-section on “Religious honour-stress” among Albanians. There are tiny
yet vocal Salafi and other Islamist groups among Albanian communities in the Balkans and be-
yond, but this phenomenon is quite new and (as of yet) marginal. I will further study and report on
this phenomenon in due course.
58
If “someone kills himself, his blood remains unavenged” (CLD §958;959).
618 Caspar ten Dam

not rehabilitate himself through extraordinary bravery, forgiveness or remorse,

or avenges himself or others thus offended. While formal kanun revenge-obliga-
tions are limited to specific cases (like violation of hospitality), the Kanun allows
revenge for any offense to a ‘person’s honour’. Gjeçov: “he remains dishonored”;
even if “he is not required” to “avenge a murder resulting from violation of hospi-
tality, he considers it a matter of honor to do so” (CLD, note to §652).
Given their martial content, besa obligations effectively compel Albanian
adults and youngsters to excel as ‘warriors’ in times of crisis, war and survival.
Youngsters swore besa to the Kosova Liberation Army (KLA) when they joined
its ranks in the late 1990s. Even without the ‘contesting-status’ dynamic stimu-
lating “aggressive male behaviour”, “violent acts can be glorified, or excused, by
local ‘tradition’” by those already eager to do so (Schwandner-Sievers, apud
Schmidt/Schröder 2001: 104, 106, 107; Judah 2000: 99). Moreover, right after the
war, many Albanians, like those of the KLA-funded Kosovapress, argued that
war-traumas among both civilians and combatants make revenge ‘natural’ and
justifiable: “we cannot call it an organized crime when someone, whose family
was killed and his house burned, kills a Serb in revenge for those things that
Serbs committed during the war. All these things are natural and may happen to
anyone under such conditions”.59 Sadly, the pressures of traditional norms, and
their opportunist applications, have contributed to a frequent, if not endemic,
double brutalisation of traditional and international norms during and after the
war.
Criminalising honour-stress
Social obligations, combat traumas, other combat-stresses, and (consequent)
brutalities not just occur on battlefields but in any violent arena—like that of
organised crime. The widely noted low violence-threshold in Albanian culture
explains the viciousness and success of Albanian criminal clans known as fares,
making “the Italian Mafia look like a whist drive” (British Home Office report,
apud Cilluffo/Salmoiraghi 1999: 24). The patriarchal clan structure, which de-
mands absolute allegiance, perpetuates inat (spite), ndera (prestige) and blood-
revenge. An average (sub-)clan consists of sixty to hundred-and-sixty members,
an ideal basis for impenetrable fares. Crime bosses, especially those from North-
ern Albania, exploit this overriding sense of family loyalty, particularly since the

59
Kosovapress, “Hallucinations of Mr. Jiri Dienstbier”, 19 July 1999, <http://www.kosovapress.
com/english/korrik/19_7_99.htm>.
 The Limitations of Military Psychology 619

end of Albania’s Communist rule in 1990. This also led to inter-clan and intra-
community tensions: “Codes of honour exist with any Albanian brigands as they
do with youth gangs all over the world, but for northern Albanians who choose ..
violence along .. group inclusion or exclusion, kanun rhetoric confers an ad-
vantage …. Among their fellow Albanians … young men from northern Albania
are held responsible for most crimes. They are known as ‘hooligans’ or ‘Chech-
nians’ ” (Schwandner-Sievers, apud Schmidt/Schröder 2001: 104, 106).
Loyalty and a ‘culture of heroics’ put enormous pressure on young Albanians
whose fathers or uncles are crime bosses, to excel in ruthless violence. They up-
hold an ‘oath of silence’, refusing to divulge information to the police (Swiss eth-
nologist Pr. Christian Giordano, apud Barth et al 1999: 50, 51, 53).
Northern- and Kosovar-Albanian gangsters were not unique in exploiting
close-knit kin structures. Particularly since Albania’s anarchy of 1997 (sparked by
collapsing pyramid-funds, whereby thousands of Albanians lost their savings),
Southern-Albanian gangsters increased their influence by “sometimes brutalised
forms of violence”, legitimised by “ ‘modernised’ forms of kanun” (Schwandner-
Sievers, apud Schmidt/Schröder 2001: 111-113). Indeed, the destructive, state-
weakening impact of ‘1997’ led to a ‘retraditionalisation’ in Albania’s South (ibid).
Yet Kosovar-Albanian fares became the most notorious. Youngsters who fled Ko-
sovo to escape Serb repression were brutalised by that repression; they showed
even fewer inhibitions than older-generation gangsters from Albania. Through-
out the 1990s the typical Albanian gang in Italy consisted of a group of ill-edu-
cated, uprooted twenty-year-old members led by a better-educated Albanian in
his mid-thirties. The latter spoke several languages, knew the rules and loop-
holes in the host country, and had close contacts with elites among the Diaspora
and home communities. The youngster’s sacrifice for the clan supersedes all
moral and legal considerations—of which he is acutely aware. This fealty is di-
rected to the patriarchal leadership and line of the clan. This accounts, together
with criminalised brutalisation, for their cruel abuse of women—in contrast to
the old-fashioned, older generations among their Italian (mafia) counterparts.
Impoverished families sold their daughters for prostitution, despite protection of
one’s female relatives being a paramount Kanun norm (Barth et al 1999; Cilluffo/
Salmoiraghi 1999). Criminal activities and manipulations of traditional loyalties
corrode other traditional values geared toward the good of the entire clan or
community. Worse, these criminal-brutalising trends undermine even the best
customs of hospitality and reconciliation. As I will describe in future research
620 Caspar ten Dam

and publications, this corrosion of ‘indigenous ethics’ extends toward the politi-
cal violence by many Albanian—and Chechen—separatists.

V. CONCLUSION
Despite colonisation, industrialisation and urbanisation, Chechens and Albani-
ans retain many traditions. Paradoxically, their adherence to ‘old’ values—how-
ever much contemporary ignorance, manipulation and brutalisation disfigure
these—leads to the same “conflict of the aggressive impulses .. with the moral
norms” Talcott Parsons discerned in Western societies (Parsons 1947: 169). The
‘warrior’ and ‘carer’ roles of Western men and women, particularly in America
and Germany before WWII and the post-WWII gender emancipation, resemble
those of Chechens and Albanians. Despite differences between and within ‘tradi-
tional’ societies with large, integrated families-and-occupations and ‘modern’
societies with nuclear families segregated from the workplace, I—unlike Par-
sons—believe that males in both societies must live up to similar expectations.
Both types of societies celebrate “physical prowess, with an endemic tendency
toward violence” (ibid: 174)—even against authorities that impose these
norms.60 This celebration of prowess brought immense pressures on Chechen
and Albanian combatants to compete and achieve ‘incredible feats’, contributing
to combat-stresses beyond the ‘normal’ dangers and horrors of war.
At the same time, stress-inducing martialism has led some Chechens and Al-
banians to commit some of the worst violence against non-Chechens and non-
Albanians. Even so, intra-community and factional violence tend to increase as
soon as threats from outside ‘enemy’ communities diminish. Still, hostilities
among these and other peoples are semi-regulated feuds rather than no-holds-
barred wars.
Existing military-psychological theories tenuously underpin the Brutalisation
theory’s combat-stress variable at best: as applied on insurgents, it largely enters
uncharted territory. That is true in lesser degrees for the theory’s other variables
as well. In that sense the Brutalisation theory stands on its own, and must be
tested on more conflict cases to assess its (degree or lack of) validity. Most
urgent is pioneering fieldwork with structured and informal interviews of active
and former non-state combatants about their combat-stresses.

60
Violence can ensue between “relatively “emancipated” and .. traditional groups” (Parsons
1947: 178); no single society is purely traditional or modern.
 The Limitations of Military Psychology 621

Generally, ascertaining brutalisation or moral constraint from cross- and

counter-flowing codes, norms and values in political extremism, gangsterism,
blood-feuding and ‘warriorism’ remains challenging (Leach 1977: 25-26, 28, 33-
34). These are the “fateful interactions between deprivation, honour and re-
venge” (Ten Dam 2009: 269) among Chechens, Albanians and any other people
in turmoil.

BIBLIOGRAPHY
Ahern, J. / S. Galea / W. G. Fernandez / B. Koci / R. Waldham / D. Vlahov (2004), “Gender, Social
Support, and Posttraumatic Stress in Postwar Kosovo”, Journal of Nervous and Mental Dis-
ease 192.11: 762-770.
Ai, A. L. / T. N. Tice / D. D. Whitsett / T. Ishisaka / M. Chim (2007), “Posttraumatic symptoms and
growth of Kosovar war refugees: The influence of hope and cognitive coping”, Journal of
Positive Psychology 2.1: 55-65.
Anderson, B. (1991), Imagined Communities: Reflections on the Origin and Spread of Nationalism,
London/New York. (First published in 1983).
Aldwin, C. M. / L. A. Yancura, Chapter 5 in P. P. Schnurr; B. L. Green (eds.) (2004), Trauma and
Health: Physical Health Consequences of Exposure to Extreme Stress, American Psychologi-
cal Association, Washington DC.
Appel, J. W. (1942, 1943), “Psychiatry in National and International Relationships”, Psychiatry and
the War, ed. F. J. Sladen, Springfield/Baltimore/Detroit.
Baddeley, J. F. (2005, 1908), The Russian Conquest of the Caucasus, London/New York/Bombay/Cal-
cutta (<http://www.elibron.com>). (First published in 1903).
Baker II, J. W. / K. W. Schaie (1969), “Effects of Aggressing “Alone” or “With Another” on Physiologi-
cal and Psychological Arousal”, Journal of Personality and Social Psychology 12.1: 80-86.
Barth, A. / W. Bickerich / M. Grossekathöfer / O. Peter / H.-J. Schlamp (1999), ““Sprache der Mor-
de”—Familienclans aus dem Kosovo und Albanien haben über weite Teile Europas krimi-
nelle Netzwerke gespannt”, Der Spiegel 31, 2 August 1999: 42-59.
Bartone, P., (1998), “Stress in the Military Setting” (Chapter 7), Military Psychology, ed. C. Cronin,
Needham Heights, Ma.
―― (2004), “Understanding prisoner abuse at Abu Ghraib: Psychological considerations and
leadership implications”, Military Psychologist 20.2: 12-16.
Beleny, G. L. (ed.) (1987), Contemporary Studies in Combat Psychiatry, New York/Westport, CT/Lon-
don.
Berkowitz, L. (1962), Aggression: A Social Psychological Analysis, New York/Toronto/ London/San
Francisco.
―― / A. LePage (1967), “Weapons as aggression-eliciting stimuli”, Journal of Personality and Social
Psychology 7.2: 202-207.
622 Caspar ten Dam

Bloche, M. G / J. H. Marks (2005), “When Doctors Go to War”, New England Journal of Medicine
352.1: 3-6.
Bourke, J. (1999), An Intimate History of Killing: Face-to-Face Killing in Twentieth-Century Warfare,
London.
Butler, A. G. (1943), “Moral and Mental Disorders in the War of 1914-18”, Australian Army Medical
Services in the War 1, Canberra.
Campise, R. L. / Sch. K. Geller / M. E. Campise (2006), “Combat Stress” (Chapter 10), Military Psy-
chology: Clinical and Operational Applications, eds. C. H. Kennedy; E. A. Zillmer, New York/
London.
Cilluffo, F. / G. Salmoiraghi (1999), “And the Winner is ... the Albanian Mafia”, The Washington
Quarterly 22.4: 21-25.
Clark, R. A. (1946), “Aggressiveness and Military Training”, American Journal of Sociology 51.5: 423-
432.
Cohen, Sh. / G. M. Williamson (1991), “Stress and Infectious Disease in Humans”, Psychological
Bulletin 109.1: 5-24.
Collins, R. (2008), Violence: A Micro-sociological Theory, Princeton, NJ/Oxford.
Cooke, A. (1974), Alistair Cooke’s America, London (Copyright 1973).
Cronin, Ch. (ed.) (1998), Military Psychology: An Introduction, Needham Heights, Ma..
Davies, D. R. / G. S. Tune (1970), Human Vigilance Performance, Staples Press.
De Jong, K. / S. Van der Kam / N. Ford / S. Hargreaves / R. Van Oosten / D. Cunningham / G. Boots /
E. Andrault / R. Kleber (2007), “The trauma of ongoing conflict and displacement in
Chechnya: quantitative assessment of living conditions, and psychosocial and general
health status among war displaced in Chechnya and Ingushetia”, Conflict and Health 1.4:
<http://www. conflictandhealth.com/content/1/1/4> (accessed 7-6-2011).
Derluguian, G. M. (2005), Bourdieu’s Secret Admirer in the Caucasus: A World-System Biography,
Chicago/London.
Dinter, E. (1985), Hero or Coward: Pressures Facing the Soldier in Battle, London/Totowa, N.J..
Dollard, J. / N. E. Miller / L. W. Doob et al (1944), Frustration and Aggression, London. (First
published in 1939).
Driskell, J. E. / B. Olmstead (1989), “Psychology and the military”, American Psychologist 44.1: 43-54.
Eytan, A. / M. Gex-Fabry / L. Toscani / L. Deroo / L. Loutan / P. A. Bovier (2004), “Determinants of
Postconflict Symptoms in Albanian Kosovars”, Journal of Nervous and Mental Disease
192.10: 664-671.
―― / L. Toscani / L. Loutan / P. A. Bovier (2006), “Posttraumatic Stress Disorder and the Use of
General Health Services in Postwar Kosovo”, Journal of Traumatic Stress 19.1: 57-67.
Favaro, A. / M. Maiorani / G. Colombo / P. Santonastaso (1999), “Traumatic Experiences, Post-
traumatic Stress Disorder, and Dissociative Symptoms in a Group of Refugees from Former
Yugoslavia”, Journal of Nervous and Mental Disease 187/5: 306-308.
Flanagan, J. C. (1942), “The selection and classification program for aviation cadets”, Air Surgeon’s
Bulletin 6: 229-239.
Folkman, S. (ed.) (2011), The Oxford Handbook of Stress, Health, and Coping, Oxford/New York/etc.
 The Limitations of Military Psychology 623

Gabriel, R. A. (1988), The Painful Field: The Psychiatric Dimensions of Modern War, New York/West-
port, CT/London.
Gall, C. / Th. De Waal (1997), Chechnya: A Small Victorious War, London.
Gerard, H. B. / R. A. Wilhelmy / E. S. Conolley (1968), “Conformity and Group Size”, Journal of Per-
sonality and Social Psychology 8.1: 79-82.
Gjeçov, Sh. (1989), Kanuni i Lekë Dukagjini—The Code of Lekë Dukagjini (Transl. and Introd.
Leonard Fox), New York/Skoder (Shkodra, Albania). (First published in 1933).
Gordon, J. S. / J. K. Staples / A. Blyta / M. Bytyqi (2004), “Treatment of Posttraumatic Stress Dis-
order in Postwar Kosovo High School Students Using Mind-Body Skills Groups: A Pilot
Study”, Journal of Traumatic Stress 17.2: 143-147.
Grossman, D. Lt. Col. (1996), On Killing: The Psychological Cost of Learning to Kill in War and Society,
New York/Boston/London. (First published in 1995).
―― / L. W. Christensen (2007), On Combat: The Psychology and Physiology of Deadly Conflict in
War and in Peace, Warrior Science Publications (First published in 2004).
Gurr, T. R. (1970), Why men rebel, Princeton.
Hahn, G. M. (2009), “Buryatskii, Istishkhad, and the Riyadus-Salikhin Suicide Martyrs’ Battalion”,
Islam, Islamism, and Politics in Eurasia Report 5: 1-6.
―― (2010), “Said Abu Saad Buryatskii Killed, March 2, 2010”, Islam, Islamism, and Politics in Eura-
sia Report 10: 1-4; see <http://www.miis.edu/academics/faculty/ghahn/report>.
Henriksen, R. / A. Vinci (2008), “Combat Motivation in Non-State Armed Groups”, Terrorism and
Political Violence 20.1: 87-109.
Hinkle, L. E. / H. G. Wolff (1956), “Communist interrogation and indoctrination of “enemies of the
state” “, Archives of Neurology and Psychiatry 76: 115-174.
Hunter, E. (1971), Brainwashing in Red China, Vanguard Press (First published in 1951).
Inozemtseva, E. (2006), “On the History of the Slave-Trade in Dagestan”, Iran and the Caucasus 10.2:
181-189.
―― (2010), “On the Social and Legal Status of Slaves in the North Caucasus”, Iran and the Cauca-
sus 14.1: 19-25.
Jaimoukha, A. (2005), The Chechens: A handbook, London/New York.
Janis, I. L. (1951), Air War and Emotional Stress: Psychological Studies of Bombing and Civilian De-
fense, New York.
Judah, T. (2000), Kosovo: War and Revenge, New Haven/London.
Lifton, R. J. (1962), Thought Reform and the Psychology of Totalism: A Study of “Brainwashing” in
China, London (First published in 1961).
Kashdan, T. B. / N. Morina, N. / S. Priebe (2009), “Post-traumatic stress disorder, social anxiety dis-
order, and depression in survivors of the Kosovo War: Experiential avoidance as a contrib-
utor to distress and quality of life”, Journal of Anxiety Disorders 23.2: 185-196.
Kaslow, F. W. (ed.) (1993), The Military Family in Peace and War, New York.
Kellezi, B. / S. Reicher / C. Cassidy (2009), “Surviving the Kosovo Conflict: A Study of Social Iden-
tity, Appraisal of Extreme Events, and Mental Well-Being”, Applied Psychology 58.1: 59-83.
624 Caspar ten Dam

Kennedy, C. H. / E. A. Zillmer (eds.) (2006), Military Psychology: Clinical and Operational Applica-
tions, New York/London.
Kennedy, F. (1943), “Psychiatry and Propaganda”, ed. F. J. Sladen, Psychiatry and the War, First ed.,
Springfield/Baltimore/Detroit.
Kern, R. / H. McFann (1965), “Can our troops be battleproofed?”, Army Information Digest 20.2: 24-
27.
Krueger, G. P. (1998), “Psychological Research in the Military Setting” (Chapter 2), Military Psychol-
ogy, ed. C. Cronin, Needham Heights, Ma.
Leach, E. (1977), Custom, Law and Terrorist Violence, Edinburgh.
Lieven, A. (1998), Chechnya: Tombstone of Russian Power, New Haven, CT/London.
Luzbetak, L. J. (1951), Marriage and the Family in Caucasia: A Contribution to the Study of North Cau-
casian Ethnology and Customary Law (Studia Instituti Anthropos, vol. 3), Vienna/Mödling.
Maercker, A. / M. Povilonyte / R. Lianova / L. Pöhlmann (2009), “Is Acknowledgment of Trauma a
Protective Factor?—The Sample Case of Refugees from Chechnya”, European Psychologist
14.3: 249-254.
Manning, F. J. / L. H. Ingraham (1987), “An Investigation into the Value of Unit Cohesion in Peace-
time” (Chapter 4), Contemporary Studies in Combat Psychiatry, ed. G. L. Belenky, New
York/Westport, CT/London.
Marin, P. (1981), “Living in Moral Pain”, Psychology Today 15.11: 68-80.
Marshall, Brig. Gen. (then Lt. Col.), S. L. A. (1947), Men Against Fire: The Problem of Battle Command
in Future War, Gloucester, Mass. (ed. 1978).
Maslow, A. H. (1941), “Deprivation, Threat and Frustration”, Psychological Revie 48: 364-366.
―― (1943), “A Theory of Human Motivation”, Psychological Review 50: 370-396.
―― (1987), Motivation and Personality, New York/etc. (First published in 1954).
Merridale, C. (2000), “The Collective Mind: Trauma and Shell-shock in Twentieth-century Russia”,
Journal of Contemporary History 35.1: 39-55.
Molnar, A. / J. Tinkler / J. Leitoir (1966), “Human factors considerations of undergrounds in insur-
gencies”, US Special Operations Research Office (SORO), Centre for Research in Social Sys-
tems (CRESS).
Morina, N. / J. D. Ford (2008), “Complex Sequelae of Psychological Trauma Among Kosovar Civil-
ian War Victims”, International Journal of Social Psychiatry 54.5: 425-436.
―― / V. Rudari / G. Bleichhardt / H. G. Prigerson (2010), “Prolonged Grief Disorder, Depression,
and Posttraumatic Stress Disorder Among Bereaved Kosovar Civilian War Survivors: a
Preliminary Investigation”, International Journal of Social Psychiatry 56.3: 288-297.
―― / H. F. Böhme / L. Morina / G. J. G. Asmundson (2011), “The structure of post-traumatic stress
symptoms in young survivors of War”, Psychiatry Research 186.2: 306-309.
Mountz, T. (1993), “Special Warriors, Special Families and Special Concerns”, The Military Family in
Peace and War, ed. F. W. Kaslow, New York: 121-129.
Norikov, V. S. Maj. Gen. (1996), “Psycho-sociological Support of Combat Activities of Military Per-
sonnel”, Military Medical Journal 4: 37-40.
 The Limitations of Military Psychology 625

Parsons, T. (1947), “Certain Primary Sources and Patterns of Aggression in the Social Structure of
the Western World”, Psychiatry—Journal of the Biology and the Pathology of Interper-
sonal Relations 10.2: 167-181.
Picano, J. J. / Th. J. Williams / R. R. Roland (2006), “Assessment and Selection of High-Risk Opera-
tional Personnel” (Chapter 17), Military Psychology, eds. C. H. Kennedy; E. A. Zillmer, A.
Eric, New York/London.
Porter, Col. W. C. (1942, 1943), “Psychiatry in the Army”, Psychiatry and the War, ed. F. J. Sladen,
Springfield/Baltimore/Detroit.
Ralph, J. A. / M. T. Sammons (2006), “Future Directions of Military Psychology” (Chapter 18), Mili-
tary Psychology, eds. C. H. Kennedy; E. A. Zillmer, New York/London.
Reinartz, Brig. Gen. E. G. (1943), “Psychiatry in Aviation”, Psychiatry and the War, ed. F. J. Sladen,
Springfield/Baltimore/Detroit.
Reineggs, J. (1796-1797), Allgemeine historisch-topografische Beschreibung des Kaukasus, vol. I, Go-
tha.
Rosenberg, E. (1943), “A Clinical Contribution to the Psychology of War Neuroses”, International
Journal of Psychoanalysis 24: 32-41.
Roth, G. / S. Ekblad / H. Prochazka (2009), “A study of aggression among mass-evacuated Kosovo
Albanians”, Torture 19.3: 227-237.
Ruigrok, H. (2000), “Schone oorlog bestaat niet (Clean war does not exist)”, Wordt Vervolgd (To be
Continued), Monthly of Amnesty International Netherlands 32.12-1: 32-34.
Ryan, L. M. / Th. M. Zazeckis / L. M. French / S. Harvey (2006), “Neuropsychological Practice in the
Military” (Chapter 6), Military Psychology, eds. C. H. Kennedy; E. A. Zillmer, New
York/London.
Schmidt, B. E. / I. W. Schröder (eds.) (2001), Anthropology of Violence and Conflict, London/New
York.
Schnurr, P. P. / American Psychiatric Association (1994), Diagnostic and statistical manual for men-
tal disorders, Washington DC.
―― / Green, B. L. (eds.) (2004), Trauma and Health: Physical Health Consequences of Exposure to
Extreme Stress, American Psychological Association, Washington DC.
Schwandner-Sievers, S. (2001), “The enactment of ‘tradition’: Albanian constructions of identity, vi-
olence and power in times of crisis” (Chapter 1), Anthropology of Violence and Conflict, eds.
B. E. Schmidt; I. W. Schröder, London/New York.
Shlapentokh, D. (2010), “The Rise of the Russian Khalifat: The View from the Jihadist Side”, Iran and
the Caucasus 14.1: 117-142.
Sieca-Kozlowski, E. (2007), “Demos Center, Miliciya meždu Rossiej i Čečnej—Veterany konflikta v
rossijskom obščestve”, Moskva, 304 c.; Summary and review, English translation (2009),
Journal of Power Institutions in Post-Soviet Societies [online] 1.9, <http://pipss.revues.org/
index1960.html> (accessed 7-6-2011).
Sladen, F. J. (ed.) (1943), Psychiatry and the War: A Survey of the Significance of Psychiatry and its
Relation to Disturbances in Human Behavior to help provide for the Present War Effort and
for Post War Needs, “A Record of the Conference on Psychiatry held at Ann Arbor Michi-
gan, October 22, 23 And 24, 1942”, First ed., Springfield/Baltimore/Detroit.
626 Caspar ten Dam

Smeets, R. (1995), “De Tsjetsjenen zijn nog nooit opzij gegaan (The Chechens have never backed
down)”, Vrij Nederland (Free Netherlands magazine): 13-15.
Souleimanov, E. (2007), An Endless War: The Russian-Chechen Conflict in Perspective, Frankfurt am
Main/Berlin/New York/Oxford/etc..
Spiller, R. J. (1988), “S. L. A. Marshall and the Ratio of Fire”, R.U.S.I Journal Royal United Services In-
stitute (RUSI) for Defence Studies 133.4: 63-71.
Stouffer, S. A. E. A. Suchman / L. C. DeVinney et al (1949), The American Soldier: Adjustment During
Army Life, vol. I in four-volume series ‘Studies in Social Psychology in World War II’ (1977),
Princeton (First published in 1949).
Swank, R. L / W. E. Marchand (1946), “Combat Neuroses [Neurosis]: Development of Combat Ex-
haustion”, Archives of Neurology and Psychology 55: 236-247.
Ten Dam, C. (2009), “Book review on E. Souleimanov’s ‘An Endless War: The Russian–Chechen
Conflict in Perspective’ (Peter Lang, 2007)”, Ethnopolitics 8.2: 268–269.
―― (2010), “Violence-values among the Chechens and Albanians”, Iran and the Caucasus 14.2: 331-
365.
―― (2011), “Histories, Cultures and Grievances of the Chechens and Albanians”, Iran and the Cau-
casus 15.1-2: 234-273.
―― (2012), “Combat-stress and Violence-values among the Chechens and Albanians”, Iran and the
Caucasus 16.2: 225-245.
Thomas, T. L. / M. Ch. P. O’Hara (2000), “Combat Stress in Chechnya: “The Equal Opportunity Dis-
order”, Army Medical Department Journal, PB 8-00-1/2/3: 46-53; or see Foreign Military
Studies Office Publications, <http://fmso.leavenworth.army.mil/documents/stress.htm>
(accessed 7-6-2011).
Walters, H. C. (1968), Military psychology: its use in modern war and indirect conflict, United States
Military Academy, Dubuque, Iowa.
Watson, P. (1980), War on the Mind: The Military Uses and abuses of Psychology, Abridged and re-
vised ed., Harmondsworth (UK)/New York/etc. (First published in 1978).
Weinberg, S. K. (1946), “The Combat Neuroses”, American Journal of Sociology, Spec. issue Human
Behavior in Military Society 51.5: 465-478.
Weinstein, E. A. (1947) “The Function of Interpersonal Relations in the Neurosis of Combat”, Psy-
chiatry: Journal of the Biology and the Pathology of Interpersonal Relations 10.3: 307-314.
Wenzel, T. / F. Rushiti / F. Afghani / G. Diaconu / B. Maxhuni / W. Zitteri (2009), “Suicidal ideation,
post-traumatic stress and suicide statistics in Kosovo”, Torture 19.3: 238-247.
Williams, Th. J. / J. J. Picano / R. R. Roland / L. M. Banks (2006), “Introduction to Operational Psy-
chology” (Chapter 9), Military Psychology: Clinical and Operational Applications, eds. C. H.
Kennedy; E. A. Zillmer, New York/London.
Yarlykapov, A. A. (2007-2008), ““Folk Islam” and Muslim Youth of the Central and Northwest Cau-
casus”, Transl. J. E. Walker, Anthropology and Archeology of Eurasia 46.3: 9-35 (First pub-
lished in Ethnograficheskoe obozrenie, No. 2, 2006: 59-74).
Yerkes, R. M. (1918), “Psychology in relation to the war”, Psychological Review 25.2: 85-115.
―― (1919), “Report of the Psychology Committee of the National Research Council”, Psychological
Review 26.2: 83-149.
 The Limitations of Military Psychology 627

Zelenova, M. A. / E. O. Lazebnaia / N. V. Tarabrina (2001), “Psychological Characteristics of Post-

traumatic Stress States in Afghanistan War Veterans”, Journal of Russian and East European
Psychology 39.3: 3-28 (Russian original in Psikhologicheskii zhurmal 18.2, 1997: 34-49).
Zelkina, A. (2000), In Quest for God and Freedom: The Sufi Response to the Russian Advance in the
North Caucasus, London.
 The Ararat Factor:
Moral Basics in Western Political Theory from
Isaac Newton to John Stuart Mill

Garry W. Trompf
University of Sydney

Abstract
It is not well known that the great natural philosopher Sir Isaac Newton looked to the Biblical rev-
elations in the setting of Mount Ararat as the key to the solution of early modern Europe’s socio-
religious ills. When Jesus gave his twin commandments to love God and one’s fellow humans, in
Newton’s view he was distilling the Seven Precepts delivered to Noah after the Flood, regulations
accepted in Judaism as preparatory to the Ten Commandments. For Newton this primary ‘true re-
ligion’ had the power to heal the nations, and this paper explores how this platform and Newton’s
irenic commitments were taken up in the transition from the ‘Scientific Revolution’ to the eight-
eenth century’s ‘Enlightenment.’ His position connects with the agendas of Locke, Voltaire, Mon-
tesquieu, Rousseau, Priestly and Paine (among others), and, although the appeal to the twin loves
of God and neighbour were often eviscerated of their original religious purport, they persisted in
the development of modern political liberalism, lying behind John Stuart Mill’s dictum that we can
do what we like so long as we do not harm others.

Keywords
Noah, Noachian Precepts, Newtonians, Nonconformity, Enlightenment, Masonry, Political Liber-
alism, Utilitarianism, Marxism, Isaac Newton, John Locke, Voltaire, Montesquieu, Jean-Jacques
Rousseau, Joseph Priestly, Thomas Paine, Jeremy Bentham, James Mill, John Stuart Mill

A mark of consummate genius is to be able to sum up the nature of things.

E=mc2 marks such an impressive summation in our own age that Einstein’s brain
has been bottled for future admiration! For many thinkers of the eighteenth and
nineteenth centuries, the time period considered in this article, Sir Isaac Newton
was the extolled equivalent. His three basic laws of motion, auxiliary theory of
gravitational attraction, and renowned experiments with light veritably epito-
mized the achievements of the so-called ‘scientific revolution.’ When Alexander
Pope parodied Genesis in his poetic epitaph,
 Koninklijke Brill NV, Leiden, 2015
630 Garry W. Trompf

Nature and Nature’s laws lay hid in night,

God said, “Let Newton be!” and all was light;
he could be fairly rated an early standard-bearer of ‘enlightenment,’ of the con-
viction that through Newton humanity’s generally “untutored mind,” even the
best ancient philosophy, could be completely surpassed. By mounting “where
science guides,” we now proceeded with a Newtonian confidence in “gen’ral
laws” and “the Universal Cause” to
measure earth, weigh air, and state the tides;
Instruct the planets in what orbs they run,
Correct old time and regulate the sun.1
Few people are aware, though, that while Newton was framing physical laws
he was in search of social ones. He apparently wanted to grasp the “fundamental
principles” of everything. In one late-in-life conversation he distilled what un-
derlay the material world as “gravity” and the biological order as “instinct,” while
asserting the rational realm to be “perfectly free” (Stukeley 1836: 71). What more
aptly stated agenda was there to be for generations of scholars to come? Georg
Wilhelm Hegel would simplify the encapsulations further: the Prinzip of Nature
was gravity (Schwere) or necessity; of human history it was freedom2; and of
course it was the German philosophers of the Aufklärung who were going to do
their best to ensure a distance between ‘blind physical forces’ and the Freiheit of
mind (Kant 1963: 17; idem 1995: 162-170).3
Newton’s social theory will appear rather peculiar to many of us—for being
so affected by the sacred narrative of the Bible. Inspired by the Genesis account
of the rainbow in his optical work on the color spectrum, and by the proportions
of the Ark in his theory of measurement, Newton took the Biblical story of the
world-wide Deluge seriously as historical fact, with Halley’s comet (named after
his friend) then being large enough—as it passed near earth in 2347 BC—to pull
up all the subterranean waters according to his enunciated gravitational princi-
ples.4 The socio-political implications of the Deluge, however, that is, the role of

1
Alexander Pope, Epitaph for Newton, 1726; Essay on Man 1735, in 1906: 203, 371. Cf. also New-
ton 1782, vol. 4: 262.
2
Hegel 1837: 44. Germans, incidentally, might not think first of Newton as formulator of the
law of gravity, but of Galileo Galilei; cf. Crombie 1994: 563-566, or if nationalistically inclined of
Athanasius Kircher, see Rowland 2000: 12.
3
The quoted phrase derives from Joseph Görres (1958: 257), however, reminding us of German
‘vitalist Romantic’ thinkers who conceived the Divine to be alive in natural process.
4
Newton was already respectful towards the chronological calculations of Archbishop James
 The Ararat Factor 631

Noah as primal legislator, Newton kept close to his chest, and his ideas have only
come to light through manuscripts he was working on in the secrecy of his study
and from the tenor of their contents passed on by his friends. When the popu-
lating of the world began, from around Mount Ararat, he maintained, it was
Noah who had revealed to him the bases on which future societies should be
founded—in seven precepts (well ahead of the Ten Commandments for Israel).
When asked to write in honour of my learned friend Professor Garnik Asa-
trian, I soon recalled escaping Yerevan’s summer heat to reach the coolness of
the Ararat massif, there to wander through an old orchard, be fed generously
from a traditional oven, ‘drink of the wine’ and philosophize vociferously into
the night as if a giant bearded Noah, Newton’s first social legislator, was listening
to us from his primal domain in the dark. Relishing the hospitality and academic
mutuality of Garnik (and of course his dear wife Victoria) in an ancient land, I
now write in thankfulness and admiration from Australia, continent of timeless
Dreaming, about ‘Western’ quests for ultimate value in a morally relativizing
world. The history of ideas to be traced is a strangely neglected one: it is about
how projected ethical absolutes which have survived in the distinctly ‘religious
sphere’ get watered down in political theory. The tensions discussed here are
familiar enough for Armenians, in the pushes and pulls between national church
ideals and socio-political liberalisms in the present post-Ottoman, post-Soviet
context.
In Newton’s view, the basic ethical truths intended for all peoples on earth
were first relayed to Noah under the shadow of Ararat:
The seven precepts of the Noachides were originally the moral Law of all nations, &
the first of them was [1] to have but one supreme Lord God & not to alienate his wor-
ship [especially through idolaltry], [2] the second was not to profane his name, &
the rest was [3] to abstain from blood or homicide & [4] from fornication, (that is in-
cest, adultery and all unlawful lusts) & [5] from theft and all injuries, & [6] to be
merciful even to the bruit beasts, & [7] to set up magistrates for putting these laws
into execution.5

Ussher (1722: 3), then a cutting-edge researcher! The latter had the date of the Flood at 2346 BC,
which sits remarkably well with the calendar of Edmund Halley’s comet appearances. Both Halley
and Newton were favourable to William Whiston’s statement of the physics of the Flood (Whiston
1696: 305-307, 341-344).
5
Newton Ms schol. to Opticks 1717 in Manuel (1962): pl. X (my square brackets). Regarding kill-
ing and the treatment of animals and of blood in sacrifice, note background in the commands to
Noah in Gen. 9: 4, 6.
632 Garry W. Trompf

These injunctions had already attracted attention in Newton’s time. John

Selden, for instance, eminent English jurist and lawyer-parliamentarian (who
reluctantly conceded Charles I’s beheading) had decided that all known legal
systems had been expanded out of these rules,6 a claim met with some incredu-
lity by other masters of the complex jurisprudential game—one being Giambat-
tista Vico of Naples, who, probably acting on gossip that Newton sought to tap
the foundations of society, hurried off to him a copy of his own New Science of
principles (La scienza nuova) “concerning the common nature of nations” in
1725.7 Among others, Vico worried about a trampling on the providential
development and modification of centuries-old European laws over the centu-
ries by new Puritan urges to pluck up age-old accretions ‘root and branch’ and
reimpose some ideal Israelite Torah, perhaps available to people in “the bigness
of a pocket book” (Vico [1720] 1936: 277-279. 343-358; Woolrych 1982: 271-272).
Puritan-touched though Newtown was, however, what the great English cos-
mographer did with the Noachian admonitions was to the treat them as the
original “true religion” and by implication basic morality.
The far-distant textual source of the seven precepts did not lie so clearly in
the Bible (cf. Gen 9: 8-17) as in the Sanhedrin Tractate of the Jewish (Babylonian)
Talmud, which comments on the divine regulations (going back to Noah) that
every Gentile (goy) would be expected to accept or already follow if they were
thinking of converting to Judaism.8 Newton satisfied himself that the Noachian
seven laws lay behind the Mosaic Decalogue, and that the pristine “true religion
of Noah” was made more basic as a teaching of justice in the Old Testament
prophets, and distilled by Jesus into the two renowned ‘summarizing’ charges to
love God and your neighbour as yourself (which confirmed both Christ’s authority
and the albeit punctuated presence of the true religion through Biblical history,
see esp. Mk 12: 30-32, cf. Dt 6: 5; Lev 19: 18b).9 The clever scholarship of his
contemporary Cantabrian John Spencer, a founder-figure of comparative reli-

6
Selden 1640; idem 1679. His key source is Moses Maimonides 1683. See esp. vol. 17: Sefer
Shoftim, 89.
7
At the hands of a travelling Jewish rabbi Guiseppe Athias of Leghorn, see Vico 1947: 90, 184-
90. For Vico on Selden, see Vico 1725: 17.
8
Males to be circumcised, of course. See Sanhed. 56a-b, cf. Jub. 7: 20; Newton himself acknowl-
edging this Jewish understanding in ‘A Short of Scheme of the True Religion’ (McLachlan 1950: 52-
53).
9
Newton, esp. ‘Irenicum,’ and ‘Short Scheme’, McLachlan 1950: 28-9, 48; Yahuda Ms 41, fol. 7.
 The Ararat Factor 633

gion, allowed Newton to deduce that, when the early Christian leaders laid down
prerequisites for Gentiles joining the Church, they appealed to the Noachian
precepts (at least in part) in requiring abstention from idolatry, homicide
(“blood”), fornication and the flesh of cruelly killed animals (Acts 15: 20, 29).10
This was in conformity with early Jewish Christian lines of thought partially
known to Newton that these prescriptions prepared the soul for knowing what
the true love of God and one’s fellow human beings was all about.11 But beyond
this, Newton was on the look out for the percolation of Noachian truths through
other traditions. He spots them in Persia’s Zoroaster, Egypt’s Hermes (or Thoth),
Arabia’s Job and Canaan’s Melchizedek, in the Tyro-Phoenician Hiram, the Ish-
maelites and Ethiopians, in Confucius, the Brahmins, and he avers that from the
seven precepts “came the moral philosophy of the ancient Greeks” and the Lat-
ins—through Pythagoras to Socrates and Cicero.12 Although remaining only in
manuscript, he designed a work, the ‘Theologiae Gentilis Origines Philosophi-
cae,’ to plot the fate of the true religion at the hands of polytheisms, the arch-
source of these being Egypt (Trompf 1991: 216-33; Westfall 1982: 23-31).13 This pre-
cious tradition survived, and significantly because its essence was “profoundly
simple: to love God and to love one’s neighhbour as oneself,” twin fundamentals
going back to Noah himself and always lying somewhere behind the teachings of
those passing on his precepts (Christianson 1984: 566-7). The double commands,
interestingly, also constituted the major religious message of Newtown’s mentor
at Cambridge, the mathematician Isaac Barrow (Barrow 1680; Feingold 1990).

10
Spencer 1685: 442-8; and for Newton himself, esp. Keynes Ms 35: 35; Yahuda Ms 16.2, fol. 45v;
Ms 7.2p (surviving scrap of paper); ‘Irenicum,’ theses 6-7.
11
See esp. Clementine Homil. VII, 4, cf. 8; II, 6; XII, 32 and also Didache I, 2, cf. ii, 1-7, vi, 3 (noting
also the approaches to blood and animals in Clem. VII, 4; 8, and Did. vi, 3), cf. Daniélou 1964: 318-9.
For signs of Newton’s familiarity with Clementine literature, e.g., Harrison 1978: 120, but all the
texts were not published together until 1835 (by Albert Dressel, from the least defective codex, Ot-
tobon 443).
12
E.g., Yahuda Ms 41, fol. 4r; New College Mss III, fol. 65 (Persia); New College Mss II, fols. 108,
162, III, fol. 6 (Thoth); Newton 1728: 155-7 (Job and Arabia Petraea); New Coll. II, fol. 238; cf. ‘Ire-
nicum,’ posit. 2 (Melchizedek); Chronology: 255-7 (Hiram) Yahuda Ms 16.1, fols. 3ff.; 16.2, fols. 22v ff
(Ishmaelites and Ethiopians); Keynes MS 7 (= ‘Short Scheme,’ p. 52) (Confucius); Chronology: 351
(Brahmins, etymologically related to Abraham); Ms schol. to Opticks 1717 (Greeks, quotation); Ya-
huda Ms 41, fol. 4r (Trojans, Aeneas and Rome).
13
Newton has some knowledge of the philosophia perennis in the early modern European tradi-
tion that recognized divine truth revealed outside the Bible, cf. Walker 1972, but develops a ‘theo-
logically safer’ version of it.
634 Garry W. Trompf

Let us ponder these moves as a strategy of social philosophy. Newton read

the true way of life as a prisca theologia, a pure original forever potentially avail-
able but only clearly actualized in recurrent instances, like stepping-stones
through time (McGuire/Rattansi 1966: 108-38).14 It is a basic set of injunctions
about the right way of behaving towards God (no idols, no blaspheming) and
towards other humans and living creatures (no killing, ‘sleeping around,’ stealing
or hurt, and care for animals), plus the commitment to set up governments—
more precisely conciliar (or ‘parliamentary’) governments—to enact these reg-
ulations.15 Vera religio becomes a fusion of right spiritual outlook (Newton’s
“Righteousness”) and moral probity (his “Ethics” or “duty to man”) which at
depth is already “written in the hearts of the Gentiles” (cf. Rom 1:20-21; 2:1) and
(crucial for Enlightenment themes to come) “dictated to … all mankind by the
light of reason.” Moreover, it is the final underpinning of humanity’s socio-politi-
cal order, with the contingent commands of humans in government being kept
separate from divine rules, and God protected from mortal schemes so as to be
final arbiter.16
This positioning attempts to avoid the clamour of dogmatic dissensions
within European (and by then trans-Atlantic) affairs experienced during the six-
teenth and seventeenth centuries. It is ‘irenicist’ (McLachlan 1950: 27), a wilful
return to preconceived basic principles, in the spirit of a continuing and radi-
calizing Reformation that keeps doctrinal particularities out of state governance
(implicitly also international relations), prioritizes the Decalogue to exclude Ca-
tholicism’s Seven Deadly Sins, while falling back on love (or a “law of righteous-
ness [toward God] and charity [toward fellow humans]”) and thus also peacea-
bleness as ultimate principles in the exercises of serious thought on temporal
affairs. Components of his appeal were not new and already had some place

14
Note, however, that Newton did speculate a physical continuity in religious history, in the
keeping of the sacred fire—generalized by him as the prytenea—which from Noah onwards into
widening religious life represented a ‘lasting factor’ even while corruptions occurred. See esp. Mar-
kley 1999: 135-138.
15
For Newton on the chronological priority of councils over kings, see Newton 1693-1694. Re-
member that Newton MP was a signatory to the 1688-9 Resolutions of the ‘Glorious Revolution’
against James II “for breaking the original contact between king and people” (Commons Journals
[1688-9], 10: 14-15). Of related interest, Nelson 2007.
16
‘Short Scheme’, 52 (quotations; my emphasis); ‘Seven Statements on Religion,’ no. 3 (McLach-
lan 1950: 58).
 The Ararat Factor 635

among ‘tolerationist’ thinkers,17 most noticeably in his own lifetime within emer-
gent English Quakerism, which had its own way of putting “enlightenment” on
the agenda, placing peace before dogma, and cherishing inner knowledge
through ‘the true Light which lighteth every man that cometh into the world” (Jn
1: 9).18 But such a resting on straightforward principles seemed to have been lost
amid all those formulations of competing theological systems as the Catholic
unity of the West fell apart.19 Confronting the great dissonance, Newton dreamt
of a solution, and one not against his own demanding principles. He did not
want any religious tests for office-holding (for he himself questioned the tradi-
tional formulations of the Trinity—a big issue!); and he abhorred Catholicism
and all it had done to Christianity over a millennium, for it was corrupted by
idolatry and thus broke the very first prohibition among the rules he extolled.20
In any case, as the fame of his scientific discoveries spread, Newton developed a
secret sense of his own eschatological, almost cosmic role—of recovering the
Noachian precepts as humanity’s paradigm of Biblical virtue for the Last Days
(Manuel 1974: chs. 1, 4).
Admittedly he was not aware as specialists now are of the function these
basic ethical principles had at the time the earliest Christians were differentiat-
ing themselves from Judaism and encountering Hellenistic philosophy. To love
God and one’s neighbour (with the commands to Noah perhaps in the back-
ground) was one expected commitment in becoming a Christian; and some no-
tice was actually given as to how we know (‘naturally’) “through reason” that
“one ought not wrong others just as we do not wish being wronged by them”
(something like Jesus’ ‘golden’ injunction to “do unto others…” [Lk. 6: 31] but

17
For background, Bossy 1985: ch. 1; Lecler 1960, vol. 1: 237-41, 261-72; 2: 269-71, 319-23, 403-06,
not forgetting a somewhat jumpy story of pithy attempts to sum the Christian faith in terms of love
over the centuries, inter alia, from Piers Plowman’s “lerne to loue and leue of alle othre” to the Jes-
uit injunction to be “the man for others,” and including discussions of Jesus’ twin commands under
the theological category of Caritas (as in Raimon Llull’s Blanquerna).
18
The Quaker humanizing stress here on each person rather than the Light coming into the
world is based on the AV, cf. RSV; see Brinton 1973: 1-11. The he sixteenth-century German Spiritu-
alist Sebastian Franck is a long-term influence here.
19
A clash of rhetorical styles also comes into this story: consider O’Malley 2004: 63.
20
McLachlan 1950: 36-43; ‘Queries regarding the Word “Homoousios”’ (McLachlan, pp. 44-7);
Yahuda MS 1.4, fols. 67-8; cf. also Trompf 2005: 107. The replacement of Charles II’s Test Acts (1673,
1678)—pace the king’s special leniencies towards him—by the 1689 Toleration Act came as a relief
and boon to Newton personally.
636 Garry W. Trompf

stated negatively, as we can also be found in the great rabbi Hillel [flor. 10s CE]
and much earlier in a saying of Confucius).21 Whatever the reaches of his
knowledge, Newton sensed his own irenic approach would draw many
Protestant Christians and Jews into a common bond, at a time when Puritan
leaders (especially in Britain and the Netherlands) articulated the hope of Jewry
recovering its homeland, and when Masonry (with which Newton had shadowy
connections) brought Jew and Christian together in non-confessional ac-
ceptance of the Divine Architect’s “moral law,” to be kept as “a true Noachida”
(Berg/Wall 1988; Goldish 1998; Trompf 1991: 234-5, 248).22 One suspects, however,
that Newton was not ready to deal with Islam, despite his own anti-Trinitarian-
ism, and, following his friend Humphrey Prideaux, would have rated Muham-
mad an imposter.23 On Newton’s reckoning, secure human directions must go
back to a Biblical grounding; and would question even near-contemporary
Thomas Hobbes’ statement concerning “the law of nature,” that one must (in the
light of the first social compact)
lay down [the] right to all things; and be contented with so much liberty against
other men, as he would allow no other man against himself,
Newton would have taken such views to be tainted by ‘atheism,’ even though
Hobbes himself presented this principle as a version of ‘the Golden Rule’ (that
“whatever you require others should do to you, that do ye to them”) (Hobbes
1651: 104).24 If Newton might have conceded with Hobbes that we have to lose
something of ourselves for the benefit of the state, the great cosmologist sat

21
See above n. 10; cf. Tobit 4:15a; Aristides Athen., Apol. 15; Hillel, Bab. Talm. Sabb., 31a; Akiba,
Aboth. R. Nathan (Akiba himself being responsible for a ‘Noachide drive’ to convert Gentiles); with
the Mandaean Ginza Raba 1.22, and with various Graeco-Roman thinkers Plato, Leg. XI.913; Isocra-
tes, Ad Nicoc.; Epictetus, Frag., 38; Seneca, De benef. vii; Epist. Xl.vii.11; with [Pythagoras], Chrys. Ep.
1; 4-7; cf. Dihle 1962; and for China’s Confucius, Anal., Lun yü IV,1; cf. Chunyung, XIII,4b. Note also
Socrates, who, despite his prior military career, did not believe in reciprocating harm, apud Plato,
Crito 49C; cf. Scott 1927: 452-3; cf. Confucius, Chuny. X, 3; and Jesus, Matt 5: 21, 38-9.
22
Note James Anderson’s 1738 Constitution for the Freemasons (om. 1723 Const.), where
Talmudist Anderson apparently sought to bring back foundational principles (distilled into “the 3
great [unspecified] Articles of Noah” [from Jub. 7:20b?]) that were not being formally recognized, a
position also taken by Chevalier Ramsay. Cf. Chakmakjian 2008: 178.
23
Prideaux 1723, passim; although Newton would have to deal with Prideaux’s charge (pp. xi-
xii) that, in the English scene, Socinians (with non-Trinitarian views comparable to his), Quakers,
Deists, etc. had affinities with Islam.
24
For Newton on Hobbes, see Newton 1961: 280, cf. Vico 1744, ‘Conchl.’ listing Machiavelli,
Spinoza and Hobbes as either chance-believing Epicureans or fatalists.
 The Ararat Factor 637

much more behind the dicta of the earliest English ‘Commonwealth Men’ (of
Reformation times) that we were “born primarily for the service of God and of
the Commonwealth,” and that “it is not lawful for man to do what he lists with
his own,” without “utmost benefit” for others (Pollard 1907: 148).25 Even the idea
that morality, in its simplest form, was [a] “natural law,” Christ’s Golden Rule
(“Do unto others…”) being deemed so by such great seventeenth-century jurists
as Hugo Grotius and Samuel Pufendorf, ran against the grain of Newton’s Bibli-
cist thinking—as a hallmark of deist-cum-atheistic proclivities, or at the very
least as “theonomy” (the equating of natural with revealed law) (Tindal [1730]
1978: ch. 12).26
It is the contention of this paper that Newton’s socio-ethical vision, which
may be from here on be called the ethical ‘distillation of the Gospel’ and which is
quite distinct from any naturalistic justification of preferred behaviour, none-
theless provided a key grounding, impetus and legitimacy for so-called Enlight-
enment images of a basic religion and morality, thus helping to lay the founda-
tions of modern liberalism, as most famously formulated in John Stuart Mill’s es-
say On Liberty (1859). All the strands of the story cannot be followed at once, and
an exhaustive account can hardly be encompassed in one article, for various
‘borderline issues’—such as mercy, fairness, even-handedness, goodwill, the
‘idea of the good,’ ‘the good society,’ ‘public morality,’ let alone peace, preserva-
tion, justice, equality, mutual assistance and rights—merit their own historical
investigations. Yet we may advisedly ask at this point: if Newton worked secre-
tively on his social and historical ideas (with manuscripts central to his vision
only published by 1950), and considering his own self-estimate as the crucial
bearer of truth to the world, how did he actually have an impact on ethical the-
ory over and above his scientific repute?
Enter, first, the early English Newtonians, or the circle of Newton’s friends
and acquaintances. Those closely connected with him travelled internationally,
and had unexpected “honours paid” to them by Continental intellectuals, as
Martin Folkes’ period in Italy nicely illustrates. And then there were visitors to
Sir Isaac, before and after his death, let alone relevant correspondents (Stukeley
1837: 64). The learned (especially) wanted to know what the great man believed

25
Cf. the later cohort of (not dissimilarly-oriented) thinkers, Robbins 1961.
26
On Spinoza as ‘atheist,’ for example, see Corvinus 1702: 47. On the jurists and natural law,
start with Wolf 1944: 238 (for Grotius on respect for all who honour one God and our immortal
souls), cf. Edwards 1981: 37-38, 80-88
638 Garry W. Trompf

more generally, philosophically and of course theologically (including what he

made of chronology and the Bible’s sacred history).27 Naturally there were those
who wanted to paint over Newton’s more heretical opinions (his neo-Arian’ pro-
pensities) (McLachlan 1950: 14-17), so that “the Church of England” could “intirely
[claim] him as her son, in faith and practise” (Stukeley 1837: 71)28. That resort
became part of the agenda for consolidating the Royal Society in London, within
which many of Newton’s friends and admirers clustered, because patronage de-
pended on its members keeping their noses clean of quirkish religious notions
(as background, see Hunter 1995). The turn into the eighteenth century saw the
appearance of impressive scholarly histories of heresies up to modern times
(especially in Germany), with the English Civil War dissentership, not just
Spinozists and deists, being taken as a beehive of unorthodoxies (Corvinus 1702).
The forgotten Jakob Brucker, whose polymathic labours still remain the most
crucial yet unrecognized source of the Encyclopédie of Denis Diderot et al.,
divided knowledge into (orthodox) theology; any philosophy (including ‘natural
philosophy’) fruitful for Christianity; and everything left over, which thus
became vulnerable to suspicion or neglect (alchemy, astrology, old theories of
emblematic correspondence, let alone various forms of religious enthusiasm
[Schwärmerei]) (Brucker 1742-1767).29 Parallel to such developments, London’s
Royal Society was learning how to keep confessionalist diatribe and suspect
methods out of its forum, but not to exclude acknowledgement of divine
providence or to avoid upholding the most basic tenets of the Christian faith
from its members’ working premises (for background see Sprat 1959: 82; Gas-
coigne 1999). Its august members were hardly unaware of the advantage
Protestants would have by this relative separation, considering the Papacy was
locked into an anti-Copernican, anti-Galilean cosmology.30
Intimates of Newton and bearers of his attitudes played their intriguing part
in this atmosphere. Publicizing that his treatise ‘Irenicum’ on pivotal Noachian

27
On the last issue as indicative, see Newton 1725 (later translated by [M.] Reid (Geneva:
Henri-Albert Gosse, 1743; and into Dutch: Historische beschryving van de oudheden der Grieken,
Egipenaren, Assyriers, Babyloniers, Mederen en Persianen, etc. [Leiden: A. Kallewier, 1763]).
28
And note that Horsley, as first editor of Newton’s work, was an Anglican bishop.
29
For background to this development, see Lehmann-Brauns 2004: 399 et passim; Hanegraaff
2010: 91-111. Cf. Corvenius, E. G. Colberg, Gottfried Arnold, etc. as early modern ‘enyclopaedists’ of
religious enthusiasms.
30
For Copernicus’ and Galileo’s work on the Papal Index, start with Stimson 1917; Domingues
1996: ch. 4.
 The Ararat Factor 639

teaching was “his creed,” and that a search for international peace based on true
religion governed his project, was his close friend John Conduitt (Westfall 1980:
820; Christianson 1984: 567-575). To his negative account of Muhammad,
Prideaux attached a history of religions in Newtonian style—as a process of gen-
eral “decay” and occasional recovery of the original “divine Truths … deliver’d” to
Noah (Prideaux 1723: 228-231). Newton’s early biographer and renowned Anti-
quarian William Stukeley considered that the famed scientist, who had reckoned
all those great religious figures outside the Bible to be touched by the Noachian
heritage, ought to have included the British Druids (so honouring something vi-
tal in his own nation’s destiny) (Haycock 2002).31 In the Rev. Samuel Clarke,
moreover, as one of Newton’s personally connected scientific allies, we find a
deliberately non-sectarian (albeit theologically Arian) spirit—in dialoguing with
deism and in maintaining that ethical law is as constant as mathematical ones,
but that matured humans act with freedom, not through the “necessary action”
of clocks or animals or infants.32 And there was John Locke. Prominent among
Newton’s correspondents, even trusted to receive his anti-trinitarian manu-
scripts, Locke was best known among all the great natural philosopher’s friends
in grounding ethics and principles of behaviour in reason, to help distinguish
sound from unfounded dogma.33 But Locke was an older man, and if he was
happy to defer to Newton on Natural Philosophy, and also caught the general
tenor of Newton’s irenicist agenda, he already had his own theory about funda-
mental ethics.34
As is well known, even by 1690 Locke had introduced two new important in-
gredients into the equation, both of which became crucial for liberal political
theory: the idea of not being harmful; and the right to protection, both premised
on human equality in a state of nature. In his Second Treatise of Civil Government
he maintains that in the original “state of nature” there is a “law of nature to gov-
ern it” for the need of humans’ “bare preservation,” namely that, since we are

31
Stukeley was first in the systematic mapping and preliminary excavation of such sites as
Stonehenge and Avebury.
32
Clarke 1712: pts. 1-2; idem 1717: 5-7, 31 (against Anthony Collins’ Epicurean determinism). On
the Newton-Conduitt-Clarke family connections (including the Noachian principle of animal care)
see Brewster 1860: 408-409.
33
For background, Axtell 1965: 235. Note Locke [1696] 1958: 60, on distilling the Bible, and Je-
sus’ role in it, as conveying the knowledge of God and duty more clearly than anywhere else or any
other way. Cf. idem [1690] 1961, vol. 2: 170 (mathematical parallel).
34
Start with Rogers 1978: 217-32.
640 Garry W. Trompf

all equal and independent, no one ought to harm another in his life, health, liberty,
or possessions; for men being all the workmanship of one omnipotent and infinitely
wise maker … they are his property.
The integral connection between consideration for God and humanity is pa-
tent, and considering most of Locke’s quotations in this text are Biblical, the
purport is theologically Christian. But for the ‘post-Adamic,’ pre-Noachian, pre-
Mosaic context, Locke feels free to postulate a natural primal morality, that of
‘not doing harm unto each other’ (a kind of negative Golden Rule) because of a
shared ‘obligation’ for being God’s creatures and “servants of the one sovereign
Master.” The seed of what was to become a fundamental moral premise of liber-
alism began germinating, for it now suffices as the primary social (and eventually
political) regulation that “all men may be restrained from invading others’ rights,
and from doing hurt to one another,” including another’s property. The
implication was, moreover, that this rule came by “reason” over time, not by
innately human properties.35
Regarding basic morality, though, it was Newton’s rather than Locke’s legacy
that dominated the scene. Locke did not even put his name to the Two Treatises
of Government and readers did not connect them to the trajectory of moral
questing we have begun exploring (Thomas 1995: passim).
Now, to consider a second important pro-Newtonian impetus, an earnest
would-be visitor to Sir Isaac immediately comes to mind: the young François-
Marie Arouet (pseud. Voltaire). Released from a French prison and temporarily
exiled to Britain, Voltaire beat his path to the great natural philosopher’s door in
1726, only to be disappointed (because Newton lay ill and dying).36 Intriguingly,
Voltaire was staying in the house of his main host Francophile Viscount Boling-
broke the next year when the latter was penning his long essay Reflections con-
cerning Innate Moral Principles (pub. 1752), which maintained that the
vivifying power of religion, its influence in shaping and perfecting human conduct,
lies in its moral precepts, not in the dogmas and sanctions enforcing them.37

35
Locke 1689: Bk. II, ch. 2, sect. 7; with sects. 6-8 (though cf. 11 on “writ into the hearts of all
mankind”, ch. 5, sect. 26 (reason), and ch. 5 generally (property). Cf. Waldron 2002: ch. 3; Rapa-
czynski 1987: chs. 3-4.
36
Voltaire was to participate at the funeral solemnities for Newton in Dec. 1727.
37
Translated from French by the author (London: S. Bladon). Note Bolingbroke’s own exiles to
France for suspected Jacobinism! Consider Harrop 1884: 323-4 (quotation); Torrey 1963: ch. 6 (limi-
tations of Bolingbroke’s influence on Voltaire).
 The Ararat Factor 641

Written while Bolingbroke and Voltaire were fraternizing with the English in-
telligentsia (including Samuel Clarke) near the time of Newton’s death, this was
an argument that took a basic ethical religion to be inbuilt into human nature
and not, as per Locke, learnt by experience or through a slow imprinting of it
upon the tabula rasa of our ‘blank’ social beginnings (Barber 1979: 47-65; Locke
1690: 9-24). Voltaire was very eager to criticize the Lockean demolition of innate
ideas, but it was actually not Bolingbroke he was going to use as his sword—so
much as the mighty Newton. For Newton, in Voltaire’s assessment, was not “per-
suaded” by a mechanical (‘secular’) affect on us by the senses, but
that God, having given the same senses to all Men, there must result therefore the
same Wants, the same Sentiments, and consequently the same general Notions,
which are throughout the World the Foundations of Society.38

The light had to beam from Newton that, along with a divine order behind the
mysteries of the cosmos, was an underlying positive structure of human nature, a
morality ‘uniform’ through human space and time, leaning to peace and the
good.39 This was more than an instinctual need or ‘mechanically-set’ proclivity,
we should be clear; it was a mark of freedom.
Thus it was, then, that when embarking on an introduction and evaluation of
Newtonian optics and cosmology, in Élements de la philosophie de Neuton (really
co-authored with his mistress Émilie de Châtelet, 1740), Voltaire prefaces all
éclairissements (enlightenings, elucidations) of the master’s work with crucial
chapters on la liberté dans l’homme and on la religion naturelle (Voltaire 1992).40
Anomalous (and neglected) though these chapters may be for historians of sci-
ence, they are mediators of the ‘Enlightenment’ as an incipient formula or mind-
frame, and significantly attributable to the greatest natural philosopher of the
age. Newton becomes the champion of our human “spontaneity,” teaching that
our disposition to live in society (one of our idées innées) “is fundamentally the
natural law (la loi naturelle) that Christianity perfects (le christianisme perfec-
tionne)” (Voltaire 1992: 216, 220, 222). Voltaire apparently knows little of New-
ton’s ‘true religion of Noah,’ but he honours the latter’s seriously Christian out-

38
Voltaire 1740, using the early, beautifully written English translated by David E. Baker, (1747:
30). Cf. also Tillotson 1958: 15-7 for background.
39
For Enlightenment uniformitarianism, see Cassirer 1955: ch. 2.
40
On De Châtelet’ role, see Fara 2004: ch. 5; on her invitation to France the Venetian Newto-
nian Francesco Algaretti (affected by Folke’s visit south), Robert Walters and Barber, “Introduc-
tion” to Voltaire’s Oeuvres, (1992: vol. 15, p. 41).
642 Garry W. Trompf

look while developing his own agenda—the uncovering of a basic, common,

natural, rational and ‘inner’ sense of right and wrong among all people. Newton
might have been “persuaded with Locke” that all knowledge derives from the
senses, but if “truth, gratitude (la reconaissance) and friendship” are recognized
“from Siam over to Mexico,” why did not Locke see with Newton that the same
needs and sentiments to live in society are innate to us and “given by God to all
peoples”? This was by le dessein (Voltaire and de Châtelet accepting ‘intelligent
design’ from Newton), and it applied also to animals, although (as les newtoniens
rightly warned) with no suggestion of fatalism in human behaviour (ibid.: 196-7,
210, 216, 219-20; Walters 1954). Despite reports of human savagery, indeed, “natu-
ral religion is nothing else than that which is known in the whole world. ‘Do
what you would have someone do to you’,” so people will do terrible things by
custom accepting the same to be done to them when appropriate. Christianity
perfects this law because, as Newton taught, God gives humans “une disposition
à la compassion,” and that includes our compassion for the animals, that has in-
creased over time (Voltaire [1740] 1992: 220-23; [1734] 1937: ch. 8).
The approach is taken further in Voltaire’s account of the British and North
American Quakers, “a people so extraordinary” whom he rightly sensed to an-
ticipate his own éclairissements, for, despite their apparent strangeness, they had
dropped all corrupting Christian rituals (by implication idolatries) from the
times between themselves and Jesus (“le premier Quaker”), eschewed any words
that might take God’s name in vain, and chose simplicity, peace and toleration
over dogma.41 They were the teachers of the light of conscience, and for Voltaire
(sounding like a Newtonian and Quaker at once) “morality [as against human
doctrines] comes from God like light.”42 Thus, when it came to enunciating his
mature position in his 1764 Philosophical Dictionary, Voltaire admitted wanting
to repeat “everyday to everyone” how “morality is one and comes from God”
while dogmas are human products. “Jesus,” he insisted, “taught no metaphysical
doctrine at all,” and we ourselves, “if calm,” can be “enlightened” with the same
basic “principle of reason and morality” that he taught, as also could Socrates,
Epicurus, Cicero and Marcus Aurelius, indeed the Egyptian Amurath II, Persia’s

41
‘Sur les Quakers,’ being the first four letters of Voltaire’s Lettres philosophiques ([1733] 1964: 1,
3-6, 19, 21). Note also idem 1764: 93-120 (‘Catéchisme de l’honnête-homme’).
42
“Morale” in Voltaire [1764] 1972: 322.
 The Ararat Factor 643

Zoroaster and China’s Confucius, all outside the pages of the Bible.43 Even a good
word was now occasionally put in for the Turks or Muslims, that they might be
justified for alleging idolatry of (Catholic) Christians; and China rather than
Christendom might be the best place to look to as the world’s paragon of moral
virtues.44 After all, reason has now established that we are all brothers under
“God” as “universal father,” and as “brothers” we “must be good and just to one
another” (Voltaire 1749: 342-343).
The ‘distilled Gospel of loving God and Man’ still shone through, although
once in the hands of independent (by arrière-pensée ‘Enlightenment’) intellectu-
als there were signs of it undergoing a process of evisceration. A general ‘Newto-
nianism’ started to set more store by natural philosophy (often in Newton’s
name) so that science becomes more an arbiter than a listener toward the sacred
text. For we can hardly conclude that the deity of Voltaire’s “natural religion” is
‘just the same’ as “Newton’s biblical God.”45 Matters were moving on. Suffice it to
say that the journey of ideas we have begun to plot is like a moving stone gath-
ering up sticky leaves and then splintering along more interesting avenues than
can be covered adequately in this small space. We can but sketch out major tra-
jectories of relevant thought in our effort to plot the background ethical premise
of political theory lying behind ‘classic’ liberalism of John Stuart Mill. One can
nicely take off from the 1764 Grand Tour of Dr Johnson’s companion Boswell,
who was bent on finding out what Voltaire’s basic commitments had been.
Surely enough, on finding him at Ferney (in an ambiguous position on the Swiss-
French border), the old, brilliant savant offers what we might expect, that he
loved “the Supreme Being,” and “expressed his desire to resemble the Author of
Goodness by being good himself.” But Boswell had also had a complementary
visit in view—to Môtiers, Switzerland, and to Jean-Jacques Rousseau, that natu-
ralist of sensibility, who, upon some hard questioning, piqued himself for being a
Christian, confessed the Trinity in a somewhat uncharacteristic, unconvincing
rhetorical flourish, and summed up his personal creed thus:

43
“Du juste et de l’injuste” Voltaire 1972: 273. Note the Christian-sounding Ciceronian senti-
ment humani generis caritas, made much of in Fénelon 1699.
44
Note Voltaire 1972: 251; Pomeau 1956: ch. 2, pt. 4; cf., for background, Harrison 1990: 154 (on
China).
45
For general assessments, Barth 1947: 81; Dillenberger 1961: 173 (Newtonianism); Brooke 1991:
163-4 (Voltaire); cf. Kulklick 2004: 29, on popular Newtonianism in Franklin, Paine and Jefferson.
644 Garry W. Trompf

Begin your life anew. God is good, for he is just. Do good. You will cancel all the debt
of evil.46

Was not Rousseau’s renowned doctrine of le contrat social, the key teaching
in his political theory, that individuals had to alienate some of their rights to the
general will, implicitly based on the assumption that “a man may only demand
of his neighbour similarly (semblable) to what his neighbour has a right to de-
mand of him,” or that “nobody could do exactly as they liked, lest they injure
their neighbour?”47 The shadow of Locke, of course, falls over Rousseauan ‘con-
tract discourse’ of primal equality, rights, general will, preservation, and avoid-
ance of harm, and we also learn that for Rousseau’s “sturdy savage” the dictum
“Do good to yourself with as little evil as possible to others” would be more
“useful” than the Golden Rule in his “state of nature.” Still, the twin principle of
loving God and one’s fellow humans—what we may term Rousseau’s common
Gospel—remains more discernable in his expressions, even if one can sense its
vulnerability to permutation, indeed subjective interpretation, or to the reduc-
tion making pitié a sentiment naturel already sufficient for our collective “preser-
vation” (Rousseau 1971: 110).
Matters were now going to be cut two obvious ways. Diderot, on one side,
whom Rousseau came to spurn as an enemy, gave what he still harboured of the
‘distillation’ twin antinomian and naturalistic twists, making its background vir-
tually unrecognizable. “Strictly speaking,” he averred, “there are no rules for the
sage,” yet “Nature [setting aside the discourse of God] made good laws for all
eternity,” and “I am a good man” when I summon “this force” of “natural equity”
before “the tribunal of my heart,” so that before one’s conscience wickedness can
do nothing against good. The approach to equity is reminiscent of Gabriel Bon-
net de Mably, in whose work natural egalité has virtually been equated with
goodness (bonheur), even the truth always present in “my heart” for being or-
dained by “the Lord” or “Providence,” but in Diderot the position is now secular-
ized.48 Going further, ‘almost atheist’ revolutionary American politician Thomas

46
Boswell 1952: 231 (and n. 8), 294; cf. also Rousseau [1761] 1960: 730.
47
See Rousseau 1966: 50-52, 68-9; cf. idem 1915: 452-3 (first quotation); Birkhead 1917: 27 (se-
cond).
48
An elite view, however, which he did not believe should be spread to the masses: Vernière
1956: 430, 443; Blum 1974: 116-7; and see Rousseau [1782] 1953 Bk.9: 429-30, 442-53 on the Rousseau-
Diderot conflict; and Bk. 8: 365 on the ‘common Gospel’ of Christians. Cf. De Mably 1849: 1, 3, 5 et
passim.
 The Ararat Factor 645

Jefferson simply read the twin rules of love as the centerpiece of “The Life and
Morals of Jesus,” suitable as a basis for American civic society if detached from
confessional issues and superstition towards the miraculous.49 Others, per contra,
preserved the Christian vein. Kant, for one, was to turn the dual principle into
his “categorical imperative,” so that duty to act upon one’s conscience under God
was an act of freedom, not a mere reflection of Nature’s laws, and because he
found in Rousseau a “Newton of the moral world,” the twin basic Biblical com-
mands were not going to be lost from his sight as the prototype for philosophical
ethics.50 As his part-supporter Georg Lichtenberg was to put it, “in the eyes of
God there are only rules, [yet] perfectly speaking only one with no exceptions,”
that practical benevolence found in “the teaching of Christ,” that “perfect system
… for the promotion of peace and happiness” (Lichtenberg 1990: 46, 128, 178, 201).
Savants of quite different sensitivities were willingly playing the same tune.
Quirkish Swedish theologian Emanuel Swedenborg, a visionary hard for Kant to
bear and yet whose stress on the “two spiritual loves” suspiciously derived from
probing Newton’s views in London, resolved that entrance to heaven or hell fi-
nally depended on whether or not one’s life was “ruled by love of one’s neigh-
bour and for the Lord,” and was the first to publish a vision of Muslims high
enough in heaven, for providentially suppressing so much idolatry and correct-
ing common Christian “tri-theism.”51
One related tendency in eighteenth-century thought accepted concern for
love of God and one’s fellows as something of a jurisprudential “natural law,”
taking the cue from Grotius.52 We see this in the sharp-eyed Baron de Montes-
quieu and ‘stormy petrel’ Tom Paine, for two interesting cases. The former soon
shows us his own subtle methods of objectification. Under les lois de la nature,
Montesquieu took the human inclination towards God as Creator to be of pri-

49
Start with his cut-and-pasted Gospel verses under that title, in Adams 1983; Cullen 1986.
50
Kant 1789: ch. 20, sects. 58-9; Kant 1781: passim, esp. ch. 1 (finis); in Eng. Trans., idem 1963: 11.
This view of freedom qua duty was actually being more restrictive than Roussseau’s notion of
moral freedom as the distinctive quality of humanness, see Roussseau 1966, I, 2: 42; and on Kant
calling for self-sacrifice to better the lives of future humans who were not in our time ‘neighbours,’
as an act of the distributive [general] will of all,” see Kant 1995, vol. 6: 313. Cf. Kant’s “ethical cate-
chism” on the good-heartedness and goodwill for others’ happiness, in Kant 1995, vol. 5: 579, 581.
51
Swedenborg 1975: sects. 828-34, cf. 786; idem 1750: 7178-82, 7366-77, 8033-7. Cf. Hanegraaff
2007: 56.
52
For Montesquieu’s anxiety that Grotius’s natural law theory reads ‘is’ as ‘ought,’ however, see
Hulliung 1976: 176.
646 Garry W. Trompf

mary importance, and if he deftly dissociates this relationship from the mun-
dane realm of legalities, he proceeds on quickly to declare that the first (i.e.,
strictly human) “law of nature” is that of peace (not Hobbesian war), and that a
fear of not surviving (instead of Hobbes’ stark urge for self-preservation in a hos-
tile world) would lead to a reciprocal attraction towards those with the same
needs.53 His admiration for both Locke and Newton are known, and Locke’s
views on social beginnings definitely count here.54 His point is to ground a polit-
ico-legal discourse on basic universal truths independent of any religio-dogmatic
tradition, to establish not a distinctly “moral, nor a Christian, but a political vir-
tue,” and an early ‘cosmopolitan’ touch has been suspected, as if Montesquieu
weds suggestions of the Golden Rule (“the general virtue which comprehends
the love of all”) with the Stoic ideal of mortals being potentially ready for a citi-
zenship of the whole world. Indeed he then goes on to write of “the law of the
nations” (a late Stoic jurisprudential concept accepted in Christianized Roman
law), because “no society can subsist without a form of government,” this being
Montesquieu’s social extension of “natural law.”55 The idea of a ‘natural law’ in-
forming basic sociability, of course, has a long history. It is found most famously
in Aquinas, who detects a natural impulsion among humans in societas to help
and do good to their fellows and avoid evil (an approach haling from the Aristo-
telian image of “political animals” naturally tending to participate in the good of
the polis).56 Although it meant finding it hard to publish his magisterial Esprit des
lois, Montesquieu by-passed the Scholastic nexus between theology and Aristo-
tle, and joined other cutting-edge jurisprudentialists who had begun to keep
theology at bay.
Against the increasing tendency to naturalize ethics, admittedly, theologians
were able to retort that “to love God” completely covered everything, because

53
(De Secondat/Baron de) Montesquieu 1748: bk. I, 2.
54
Start with Postigliola 1999.
55
Esprit, préf. (first quotation); avert. (second); I, 3 (others). Cf. Ulpian, Instit.; Regul., apud [Jus-
tinian], Digest. seu Pandect., esp. I.i,3-9; and see Lettres Persanes (1721) (using Paris: Gallimard, 1973
edn.), nos. 12-13, on the ‘natural’ attention of Troglodytes to both the gods and their own relation-
ships). For useful philosophical comments on Montesquieu, note post-modernist Kristeva 1988:
128-9.
56
Aquinas [1274] 1485: I, quaest. 94, art. 2 (and I, q. 79, a. 12 on synderesis); cf. Aristotle, Eth. Ni-
com. I.2-3; III.4, etc.; cf. Horace, Carm. Saec., 15, etc. On the later radicalization of the idea that “eve-
ryman seeketh his own good” with the help of reason, experience and God (Leveller William Wal-
wyn), see Huehns 1951: 111-112.
 The Ararat Factor 647

the injunction inevitably included His fellow creatures (so meditated St.
François de Sales) and brought true liberty (thus Augustine’s paradoxical maxim
“love God and do what you like”).57 Yet Later Renaissance social theorists had
long accepted the importance of self-interest in human affairs (thus, as a matter
of political realism, Machiavelli-affected Francis Bacon) and of self-preservation
(especially Hobbes, in contemplating a war-ridden world). If negativity towards
human motives could be one implication (which certainly suited ‘pejoristic’ Cal-
vinists!),58 ‘concern for oneself’ could be alleviated as beneficial (thus Rousseau,
above), and even made a theological positive. For, had not Jesus enjoined us to
“love thy neighbour as thyself” (KJV Matt. 19:19 and paras; cf. Lev. 19: 18 etc.) in the
‘ethical encapsulation of the Gospel’? But because secularizing minds were bent
on interpreting human behaviour naturalistically, self-preservation could legiti-
mated in disengagement from Christian moral teaching as a collective “common
sense” (thus Anthony Cooper, Earl of Shaftesbury). Against the naturalistic-
looking challenges of Grotius, Pufendorf and their like, though, and their pre-
sumption that one could do jurisprudence as if God might not exist or come into
the hermeneutical picture, the great Vico reacted with his cunning “new sci-
ence,” deploying Bacon as an intriguing ally by arguing that humanity’s acts of
self-preservation were actually tokens of divine Providence through the ages,
slowly leading towards the recognition of such higher principii as justice and
love.59 Montesquieu’s formulations ought to be considered against this back-
ground debate, balancing out long-inured ecclesiastical insights with Stoic inter-
ests of Late Renaissance humanists into an unprecedented ‘cosmopolitan out-
look’ and objectivism: why, he can even imagine himself a Muslim, looking in
with bemusement at European ways!60

57
See Renaudin 1941: 114; Trompf 2010: ch. 3 (Francis de Sales). Cf. Augustine, In Ioan., vii, 8; cf.
Serm. CCCXLIX, 4, and see Burnaby 1938: esp. ch. 5.
58
For Hobbes playing on ‘Calvinist’ attitudes in his approach to human nature, note Gavre
1974: 1551-56; and on self-interest, Leviathan, I, 13.
59
For background, Trompf 1994. Cf. Haakonssen 2004: ch. 4; cf., however, Stumpf 2006 on Hugo
Grotius rehabilitated as concerned with a theology of ethics (his Socianism being comparable to
Newton’s anti-Trinitarianism).
60
Montesquieu develops his positions by rubbing off against the great Neapolitan jurist Joseph
V. Gravina (Esprit, I, 3), as had Vico, see Adams 1935: 106-8. Cf., on Stoicism, Shaw 1985; Trompf:
forthcoming, vol. 2, ch. 6, introd. (late Renaissance), cf. Montesquieu 1734: chs. 16 (Marcus Aure-
lius); Lettres Persanes (imagined Turk and Persian).
648 Garry W. Trompf

Thomas Paine, in his renowned answer to Edmund Burke’s conservativism,

takes us further down the track of legal reflection. Ex-Quaker, and the only ide-
ologue participant in the two great Revolutions of 1774-6 and 1789-99, he la-
mented Burke’s antagonism toward the French Revolution and countered his
traditionalist reverence for kings, nobles, priests and parliaments. With a Digger-
like appeal to the original nature of our social condition, Paine avers that what is
basic for humanity is not such “a wilderness of turnpike gates” as modern politics
was creating, but motivations much more “plain and simple,” the twofold “duty
to God, which every man must feel,” and “respect to his neighbour, to do as he
would be done by.” Jesus’ love becomes replaced here by duty and respect, ad-
mittedly a ‘radical’ (secularizing) move, yet without deliberately concealing the
Gospel origins of his point. These two truths then appear as the natural duty or
law pertaining to the “natural rights” that “all men are born equal”—of one social
degree and in unity—within Creation, and from this Paine enters into discussion
of the consequent “civil rights of man,” that allow for anyone to act “as an indi-
vidual for his own comfort and happiness,” when “not injurious to the natural
rights of others” (Paine [1791-2] 1961: 305-6). With these principles already begin-
ning to be realized in America, Paine sensed a situation arising that “hath not
happened since the days of Noah until now,”61 just to confirm that the Ararat fac-
tor had hardly disappeared from European imaginings of ethico-political begin-
nings. And now three significant principles of social theory had now become co-
attached to the ethical premise we continue tracing: rights; the experience of
comfort and happiness (which could also be taken as rights); as well as the
avoidance of harm as a minimum requirement in human relations (already in
Locke and Rousseau).
We certainly have to remind ourselves again of idea-clusters impinging on
our discussion that follow various trajectories deserving full treatment in their
own right. The claim of natural human rights, which in Paine’s specific context is
especially connected with Locke’s first Treatise and the American Declaration of
Independence, has its own background. Its story is also tied in to ideas of free-
dom and equality as primal (perhaps Edenic), or, in other common Enlighten-
ment terms, to notions of the “order of Nature, that common mother,” or to the
dictates of (a sometimes hypostatized) Reason62 As for happiness, felicitas as a

61
Common Sense (1776), in Passos: 1940: 84 [Append.].
62
See, first, Locke 1689: esp. Bk. 1, sects. 9-11; Congress Declaration, USA, 4 July 1776, p. [1]. The
quotation is from the Venetian playwright Carlo Goldoni, see Davies 1968: 16; cf. on the excess of
 The Ararat Factor 649

welcome end for all was obviously an old ideal for ethical thinkers,63 but the new
stress on the right to happiness amid comforting and more widely accessible ma-
terial goods owes itself before anyone else to Francis Hutcheson and his massive
System of Moral Philosophy, a book seminal for Scottish Enlightenment defences
of luxury (by David Hume) and the benefits of free trade for life improvement
(by Adam Smith).64 Sporadic European deferrals to the Epicurean (as against
Stoic) tradition now subtly begin to merge with mainstream strains of thought in
the second half of the eighteenth century, distinctively in discourse that set pri-
ority on experienced pleasure over pain.65
In Paine, indeed, what were defended previously as natural rights are not
only expressed as primal liberty and original equality (a position comparable
with Rousseau’s “all men are born free...”),66 but also as protection from injury
(Locke) and thus from the unecessary pain of possessing little opportunity to se-
cure “comforts and happiness.” This last manner of envisioning a bettered hu-
man condition had not yet been married or related to any theoretical sense of
collective utility, although the French revolutionary Declaration of the Rights of
Man (1789) defended natural liberté and égalité while also allowing that social
distinctions would still have to arise out of “common utility.” In the Declaration’s
article 2, citizens’ “liberty, property, security” are to be protected from “oppres-
sion;” and in article 4, most significantly, liberté “consists “in being able to do
whatever does not cause harm to another.”67 Paine’s own verbiage about care to-
wards God and neighbour was partly to echo these dictums, but he was clearly
less secular for accommodating “duty to God.”
The aforementioned Declaration’s art. 4, we note, is more truly prototypical
of the Millsian principle to come that the only “rightfully exercised power” by
individuals or groups “over any member of a civilized community” is “to prevent
harm to others,” and that we can be free “for any purposes not involving harm to
others” (Mill [1859] 1910: 73, 75) Locke, Rousseau and others amy have rational-

deifying Reason (and Liberty) in the French revolution Burleigh 2006: 87.
63
On happiness or felicity as ethical ‘end’ over the centuries (see Aristotle, Eth. Eudaim.; Aqui-
nas, Summ. contra Gent., III, 26; Boyle [1645] 1991: 3.
64
Hutcheson 1750: vol 2, Bk. II, chs. 4-5. On Hume and Smith, Trompf, Recurrence, vol. 2, ch. 7,
sect. 7.
65
Sir William Temple, a contemporary to Newton’s, is an interesting embodiment of late Re-
naissance neo-Epicureanism; see esp. Sieveking 1908.
66
Contrat social, I, 1.
67
Declar. 28 Aug., 1789, using Giacometti 1974: 1-2. Cf. also Talmon 1952: 92, cf. 41.
650 Garry W. Trompf

ized the necessity of government to be protective of rights and against harm, but
the Declaration implies more a minimum of political power than they do,
thereby rather paradoxically anticipating the decidedly laissez-faire liberalism to
come. ‘Basic morality’ was actually at this point officially disconnected from any
theological moorings, even if there were going to be pressures to retrieve the na-
tion from its new securality, and even though a range of avant-garde thinkers
would still defer back to a distilled Evangel or Golden Rule for their basic socio-
political principles.68 Looking askance at these tendencies, Germanics would fear
that true Freiheit, whether on a traditional spiritual understanding or philo-
sophically conceived as rooted in a divinely guided moral imperative à la Kant,
would be subverted by such revolutionism; and in Italy is is well known how
Mazzini lamented that our sense of duty (dovere) would be swamped by the new
clammering for rights (diritti).69 Certainly for religious believers the perennial
question would arise: will our love of God, or obedience to the divine will, be still
involved in political (not just private) life?
At least the more patently communal ideal of fraternité kept applications of
the ‘distillation’ alive. Enlightenment philosophical and revolutionary evocations
of “the Fatherhood of God and the Brotherhood of Man” might seem like Stoic
substitutes for the Gospel or a Biblically-grounded basic religion,70 but there
could be a mutual reinforcing of this catchcry with Christian discourse of loving
God and one’s fellows. We see this in emergent Universalist and Unitarian theo-
logical trajectories acquired growing group support (especially in America), and
also with Saint-Simonians and early French pre-Marxist exponents of communis-
me (with equal participation or communio of the Mass held up as the key para-
digm).71 Besides, the rhetorics of ‘ethical summarizing’ led traditional Christian
thinkers to see afresh how central the Bible’s love commandments were for the
Faith.72 But let us take stock; we have started with Newton’s appeal to the Noa-
chian precepts as the means by which one would arrive at true, final and Bibli-

68
Start with McIntire 1997: 248-54; Manuel 1962: 183.
69
See, on anti-Enlightenment tendencies in German religious thought, Epstein 1975: 58-83,
365-366, 448-454 (noting also in this context that Kant’s denial of knowing nature in itself implic-
itly pitched him against the American Declaration of Independence on “self-evident truths”); Maz-
zini 1907: chs. 1-3.
70
Take, e.g., on Voltaire, etc., Voegelin 1974.
71
See esp. Ballou 1856; Brow 1966: ch. 6 (Unitarians); Manuel 1962: 139, 183.
72
E.g., Vincent Pallotti’s 1835 petition for a new Society, Weber 1963: 168.
 The Ararat Factor 651

cally-grounded ethic principles, and seen how his appeal to Jesus’ twin com-
mandments became interwined with various positings of a basic morality. But
the Noachian regulations themselves seem to have disappeared: what happened
to them?
Noah’s laws à la Newton, expectedly, were too personal and too arcane a
component of his opus to be picked up by subsequent social theorists. Their ac-
ceptance remained crucial, however, for Jewry, as the mark of a civilized person,
ready to be a “righteous Gentile” (Clorfene/Rogalsky 1987). But Noah also held a
special place within Masonic movements. Commonly blamed for revolutionary
tendencies, fraternity under the divine Architect was a key element in Masonry,
the basic unit of all mesasurement (the cubit) going back to the Ark.73 Masonry’s
cultivation of a non-confessionalist forum allowed a link back Newton’s interest
in a shared morality across cultures deriving from a common source in the one
Deity. In a populist vogue of the perennial theology, Masons covertly taught a
moralitas perennis, thought by many to derive from Noah, with the Torah, Jesus
and Socrates enjoining the ‘positive’ love of neighbour as oneself, while Confu-
cius, Zoroaster and Hillel offered ‘negatively’-stated formulae.74 Starting from the
Saxon Baron Karl von Hundt’s lodges (from 1754), moreover, Muslims could be
welcomed as members.75 The story goes on, of course, on to that neglected side
to the history of modern Western empires, for, Masonry facilitated business
dealings and political compromises, especially between Jews, Christians, Mus-
lims, Sikhs, and Parsee/Zoroastrians under shared moral principles in colonial
contexts. The story includes the later use of these same shared Masonic princi-
ples, including “Love thy neighbour as thyself and God above all,” to fuel anti-
colonial sentiment amomg indigenous elites;76 and and it proceeds on to the
strange mixture of Masonic, Jewish, Puritan and early Enlightenment influences
in politics inspiring the joint Congressional Houses of the United States (2004) to
resolve that the nation take a lead in “returning the world to the ethical values
contained in the Seven Noachide Laws.”77

73
See Roberts 1972: chs. 4-6. As background, Jacob 1991.
74
Systemized eventually by Pike 1906: 169-70, 174 (Newton), 309 (Christ as expounder of new
Law of Love), 331 (morality as the basis of all religions), 334 (Noachites); cf. Jones 1950: 314-8.
75
See esp. McCalman 2003: 38, 70.
76
For a famous literary indication of such imperial transactions, see Kipling 1901: ch. 1; yet for
the quoted dictum in anti-colonial and founder Theosophist Blavtasky 2003, vol. 1: 446.
77
US House Joint Resol. 104, Public Law 102-14 (30 Sept. 2004: Pres.: George Bush, 2). We can go
652 Garry W. Trompf

Our focus, though, has to remain on the Biblically-related condensed ethical

Gospel that Newton proposed as a socio-political platform with Newton and on
how much of this ‘religious device’ for peace and tolerance lasted through “the
evolution of liberalism” (Girvetz 1966). For this the emergence of liberal political
theory recquire some addressing. As an ‘ism,’ one should realize, liberalism is in-
triguingly latecoming. That les idées liberales could be applied to certain classical
authors and those supporting them (e.g., by Lamothe le Vayer), to the ethos of
some former time of great creativity (Germaine de Staël on the the age of Louis
XIV), and to those professing ‘liberal views’ about religion and politics (François
de Chateaubriand) is well known enough, to take French cases (Littré 1954:
1568). The British Tories were accustomed to talk of the Whigs as liberales (from
the late 1820s), and by the first quarter of the nineteenth century liberal opinions
were associated with The Edinburgh Review, as purveyer of Whiggism and the
Scottish Enlightenment, and the London and Westminster Review, founded to
propagate utilitarian radicalism.78 Certainly France and Britain are places where
we must look above all for the first emergence of liberalism as systematically
formulated political theory.
Emergent modern liberalism after the French Revolution has consequently
become as hard to define as any ‘classical Enlightenment’ position. There are
many side-tracks to explore, such as the continuing rationalizing of morality, the
secularizing of retributive thinking, and the advocacy of certain political liberali-
zations (such as a widened franchise, full protection of private property, and a
clear separation of powers). Liberalism’s most arresting characteristics, though,
are the cultivation of the free market (Herbert Spencer’s laissez-faire liberal eco-
nomics), its insistence on minimal interference by government in the lives of in-
dividuals (and thus also business), and its considered reformist stances for par-
liamentary government to prevent “whatever crushes individuality” (John Stuart
Mill).79

beyond this to Pres. Barack Obama’s concluding evocation of the Golden Rule in his 2009 Cairo in-
ter-religious speech.
78
Thus following Chapman 1970: ch. 4; Christie, 2009: 182-185.
79
For a standard introduction, Sabine 1951: ch. 31; and for important comparative reflection,
Berlin 1969: ch. 3; although there is room to talk of ‘radical conservatism’ in characterizing liberal-
ism, e.g., regarding the opposition to state-run education, Trompf 1969: 267-80. Cf. also Spencer
1892; Mill 1910: 121.
 The Ararat Factor 653

These latter stresses make less obvious inroads on the religious domain and
could just as easily be accepted by ‘more traditional’ religionists as by ‘freer’
thinkers. In fact, an old suspicion has it that the long-term seedbed of liberalism
lay with “the Hebrew prophets” or “the Sermon on the Mount” (more accurately
our preconceived basic Biblical religion or ‘distilled Gospel’)—coupled with
Greek philosophical ideals which upheld “a liberation of the individual from
complete subservience to the group, and a relaxation of the tight hold of custom,
law, and authority.”80 That only provides further raison d’être to our history of
ideas here, while not gainsaying the clear drift of ideo-political secularization. If
the French liberals (who as a whole are historically prior to the British) could not
imagine a society being “stable without religion” and its powers of human “ame-
lioration” (thus de Tocqueville, Guizot), the overwhelming message they sent to
post-revolutionary conservatives, especially the ultra-royalists, that the old order
(l’ancien régime) had “given way irremediably” to democratic impetuses, and
that political society cannot get on its feet by artificial legislative programming
but only by responding to social changes that will “produce” its governmental
form.81 De Tocqueville, well known as a theorist of “enlightened self-interest,” re-
moulded our ‘ethical premise’ almost out of recognition. “It is the interest of
every man to be virtuous,” he pronounced, for “in the end it is commonly the
happiest and most useful track;” and although a “believing Christian” he ap-
pealed to an ostensibly neutral political language (combining Aristotle with
newer talk of collective happiness) in a concession to ‘worldly conceptions.’82
That was the way intellectual life was going. If in Britain the early liberal move-
ment had involved an intriguing “working alliance” between Evangelical (typi-
cally Dissenter) Christianity and non-religious radicalism,83 the secularizing phil-
osophical theorists (especially Jeremy Bentham and the two Mills, James and
John Stuart) were the ones conducting British liberalism to its ‘classic formula-
tion.’

80
Viereck 1987, 27: 471b, cf. also 471a on the liberal thinkers’ Enlightenment motif of tending to
“shun dogma”. Cf. Mill [1859] 1910: 120 on the Greek educational ideal.
81
The term constitution being avoided. For assistance, Siedentop 1979: 157, 161 (last two quota-
tions). Cf. de Tocqueville [1856] 1916: 159 (first quotation); Guizot 1911: 143 (second): cf. idem 1823;
other French liberals being Constant, de Staël, Roger-Collard and Barante.
82
De Tocqueville 1998: 230 (first two quotations); Lukacs 1959: 25 (last quotation, on de
Tocqueville’s Catholicism). On Aristotle, see above n. 53.
83
The still viable thesis of Wallas 1898. Cf. Cowherd 1956.
654 Garry W. Trompf

Bentham enunciated the principle of utility for moral and jurisprudential

analysis to avoid applying any external, imposed or abstracted (including religio-
theological) “standard of right and wrong,” as against physical experiences of
natural laws that allow units of “individual interests” to be aggregated as the in-
terests of the whole community (Bentham 1967: 140, cf. 126-7, 148-50). His maxi-
mization of individuals’ pleasure and the minimization of their pain, then, seems
mainly adapted from Hutchesonian and neo-Epicurean philosophical strands, to
forge the general maxim of governance that could secure “the greatest happiness
for the greatest number” or actualize the “greatest felicity principle” (as pro-
claimed by the 1820s).84 When it came to calculating appropriate penalties
against those acting to diminish the amount of others’ happiness, admittedly,
Bentham looks cold and conservative; yet he was distinctly reformist in his
commitment as a jurist of ‘protection,’ not just to protect society from typical
criminalities, but also from the “sinister interests” and “ill-begotten prejudices” of
persons possessing great social power, including the British aristocracy.85 Secular
though his “felicific calculus” seemed, moreover, he could hardly disconnect
happiness from the common expectation that people should do good to others.
The English Unitarian preacherman Joseph Priestly had already pinpointed the
main guideline of government was “to promote the public good,” and that what
“often escaped the notice of writers” was God’s background intention for indi-
viduals to be happy and “to communicate happiness to others.”86 In any case, the
tacit majority understanding continued that the Ten Commandments under-
girded the British legal system(s), that Dieu sat alongside Droit, and that, in the
minds of politicians, professionals and business leaders, something like the
Golden Rule should always inform the daily dealings of a fast progressing society.
And such a working assumption had long since crossed the North Atlantic.87
James Mill and his son John Stuart are two famous inheritors of Benthamite
utilitarianism, the former virtually elevating it as an ‘admired idol’ of all liberalist
ethical formulations. A child of the Scottish Enlightenment and former licensed
Presbyerian preacher having no truck with either religious establishmentarians

84
Consider the 1822 note to the first page of his Introduction, see Warnock 1962: 33.
85
Bentham 1843: 375-487. Also see Halévy 1934: 500 (quantifying the interests of society’s sepa-
rate members); Warren 1971: 333-47 (nature of Bentham’s reformism).
86
Priestly 1965: 198, 202. For long-term, interesting background, see also the ‘Westminster’ Con-
fession of Faith (1643): The Larger Catechism, 1.
87
Of relevance for the United States, Thomas 1983: 5 et passim.
 The Ararat Factor 655

or aristocrats, James often sounded rather Painean populist and politically pes-
simistic notes against prevailing authorities, and disturbed Bentham by openly
deploying his phrases about “sinister interests” against the nobility. (Hamburger
1965: 38-45; cf. Mazlish 1975: 95). When framing his fundamental ethical
principle of politics, then, James Mill expectedly had the key end of government
as protective, to guard individuals from the “universal tendency” of each man “to
take from any other man who is weaker than himself,” and thus to “protect the
people from injuries that can be inflicted” by the “small number” of powerful
people (necessarily) delegated to govern others (Mill 1955; Hamburger 1965: 43).
The shadow of the negative version of the Golden Rule remains; yet it is now
reduced to a grey spectre—the fear of our exertions of power, that we typically
do to others what they would not like us to do to preserve their happiness.
Hobbesian war seems to have echoed down the years, yet Mill was an analyst of
competing interests, and of privileged groups’ depreciation of labour time, and
was going to be of use to Marx in his formulations of class warfare and the labour
theory of value (Mill 1995: 55-7).88 Indeed, James Mill was among the first to
address ‘Empire’ in terms of its powerful interests, and despite his attack on prior
British colonial administrators in India, he secured an important official post
precisely because of his unabashed assessment that the British rulers were too
self-serving among the Indians and had forgotten the interests of their new
subjects (including their divergent religious traditions) (Mill 1817).89 Leading
Evangelicals, interestingly, had something in common with this stance, for, even
though their missionaries sought to convert the ‘heathen,’ they were at the
forefront of the radical campaign to protect abuses against indigenous peoples in
the colonies, and to abolish the slave trade (Reynolds 1992: chs. 4-5). The
interpretation of betterment as useful protection of others’ happiness and rights
chimed with long-inured religious ideals and the intellectual pursuit of high val-
ues divorced from residual theonomy.
James Mill’s son, however, fell increasingly disillusioned with the utilitarian
project as an ethico-political panacea. In his 1859 essay On Liberty, John Stuart
Mill apparently still regarded “utility as the ultimate appeal on all ethical ques-
tions,” but only very broadly, as “grounded in the permanent interests of a man
as a progressive being.” He had to take into account criticisms made against his

88
Note that James Mill perhaps wrote the first textbook on ‘political economy’: Mill 1821; Dun-
can 1973.
89
Working in the India Office, he became its head in 1839.
656 Garry W. Trompf

father’s opinion that humans were always more likely to prey on the vulnerable
in their own interests, when in fact people were more motivated by “basic moral
sentiments” to “benefit their neighbours,” affected by a kind of social “glue.”90 He
thus resolved that “the only freedom which deserves the name is that of pursuing
our own good in our own way;” and if this produces problems in relations and
the wider society, “the only purpose for which power [whether by authorities or
self-appointed actors] can be rightfully exercised, against [anyone’s] will, is to
prevent harm to others” (Mill [1859] 1910: 73-5, cf. ch. 3). Intriguingly, the per-
sonal crisis bringing about this shift involved difficulties in his mind regarding
Empire—when forced to reflect on British policy as Examiner in the East India
Company. Hence his translation of liberalism to social matters: that in India, for
instance, there should be no interference with indigenous religious practices
unless there are “abhorrent to humanity,” and that there was a danger in direct
rule to “deeply-rooted feelings” of India’s inhabitants. In matters of political
economy, concomitantly, laissez-faire was advocated with the caveat that the
government should act only to overcome harmful effects.91 Mill is distinctly a ‘lib-
eral’ rather than a ‘philosophic radical,’ moreover, for retreating away from
Benthamite calculus of pleasure and pain, especially of legal penalties; and yet
by the same token he thought religion’s basic Commands had been “unduly”
purveyed as “’thou shalt not’ ... over ‘thou shalt’.” This reaction away from the re-
tributive mentality, one might sense, keeps up an ‘Enlightenment trait,’ but Mill
was never going to go the way of a William Godwin, to oppose all punition and
trust that “every man” would eventually “seek, with ineffable ardour, the good of
all.” And while he could see nothing in Jesus’ teaching “irreconcilable with ... a
comprehensive morality,” it would have to be supplemented (for political the-
ory) from other sources, so that his own formulations rather than an encapsula-
tion of Gospel ethics better sufficed as socio-political principles.92

90
On the critiques of Lord Macaulay and Samuel Taylor Coleridge in this regard, see Wilson
1998: 209-221 (though centred around the fear that an enfranchised poor would legislate in their in-
terests to take property from the rich); Lively/Rees 1978.
91
Although still with individual freedoms in mind, not collective propensities dominant in Au-
guste Comte’s approach, cf. Mill [1865] 1961. On India and his crisis, esp. Zastoupil 1988: 31-54; on
economics, Wilson 1998: 237. On his crisis connected with his divorce, thus explaining the space
given to related problems in Mill 1910: 154-60), cf. Himmelfarb, 1970: 352-5; and note Mill/ Mill 1970.
92
Mill [1859] 1910: 108-109 (first and last quotations); for Godwin, see Godwin 1793, which had
already been attacked as utterly utopic by Thomas Malthus, see Flew 1982: 10. Note that Godwin’s
definition of justice as a “reciprocity” through which “each man ... should contribute to the cultiva-
 The Ararat Factor 657

Meanwhile, something of a parallel process seems to have been going on in

the Socialist camp, but there, intellectual competition was over whether to keep
a wide ethical vision or to narrow it to suit the victors in a projected Revolution.
Most early Socialists and communistes, of course, were inspired by the French
Revolutionary universals of liberty, equality and fraternity, and, as already
hinted, many saw these ideals formerly embodied in the first Christian move-
ment.93 But we can learn how, on Marxism’s road to victory, the more Christian
dictum of “universal fraternal love” and the catch-cry “All Men are Brothers!”
came to look petit-bourgeois, giving way under Marx and Engel’s influence to
“Working Men of the World Unite!” (Schapper 1948: 22; Oizermann 1981, pt. 2:
454). This was a route along which what was left of the distilled Gospel was an-
nounced to be the exclusive, non-hypocritical possession of one class, albeit the
“universal” one for being the most populous “lowest stratum,” which once arising
cannot but disrupt “the whole superincumbent strata” of society (Marx/Engels
[1848] 1951: pt. 1: 42-43). But then, had not leading protagonists for religion be-
come criticizable for commonly placing severe limitations on their own charity
(in favour of their own faith or denomination)? and had not the male half of our
species persistently read women out of the compass of maximum social possi-
bilities?
We have more properly come to the end of our ‘venture in a set of ideas,’
however, with Mill’s liberalism, for his principle that we can do what we like so
long as we do not harm anybody else seems the most archetypal outcome or ob-
vious end-point in the process of secularizing the ‘distilled moral Gospel.’ Mill
could not make up his mind of about the place of God, in any case (Mill 1924;
idem 1910; cf. Sell 2004), and his ethics start from individual voluntarism—not
only with the value of choice in view but working on the assumption that any-
body can test, and on adverse reaction, correct their will (which is autonomous
and not an intrinsic component of our ratiocinations).94 Here lies an extraordi-
nary concession to the independence of reason. The ethical fundament that

tion of the common harvest” can also be read as a type of “do unto others” formulation; see Wood-
cock 1963: 80-81.
93
See above ns. 57, 60. Of value, Pilbeam 2000. For particular references to God and brotherly
love in such distilled social theory, note, for two interest cases, C. H. Saint-Simon and Shepherd
Smith, see Manuel 1962: 139; Harrison 1969: 117.
94
For background, Schneewind 1997 on the history of relevant positionings. (‘Anybody’ for
Mill, significantly, did include women).
658 Garry W. Trompf

spoke ‘from above’ and ‘outside the human system,’ or at least by traditional so-
cio-religious inheritance held a more-than-natural status, was not enough. Put-
ting it more cautiously (to better explain the whole slow process away from an
overt espousal of Jesus’ basic ethics by our succession of modern avant-garde in-
tellectuals), philosophically adept thinkers were increasingly bent on addressing
“the generalization of values ... free from narrowly restrictive contexts” (such as
distinctly religious ones).95 This, after all, is how moral philosophy and political
theory secured working independence in the academic forum. It is not as if the
‘God’ in the injunction to love God and one’s neighbour had been killed. Pace
Nietzsche, it was rather that the issue as to how we deal with deity was rendered
steadily unnecessary in shifts towards methodological agnosticism (or a-theism)
in the framing of purportedly neutral, non-sectarian theory, and after that in the
making of most critical humanistic disciplines of the academy.96 The process in-
volves a subtle disembodiment, whereby abstract principles replace more di-
rectly graspable inter-subjective relations.97 At the commencement of our story,
Biblicist Newton accepted our key twinned injunctions as divinely and cosmi-
cally sanctioned, impelling us into a dynamic interaction with our Maker and
our fellows; at the end of it, John Stuart Mill takes a residue of these as minimal-
ist social demands. Whether such liberalism leaves us with a justifiable basis for
virtue, of course, will remain a perennial matter of debate.98 Certainly in Profes-
sor Asatrian’s Armenia, where liberalism has played a role in party politics since
1885, the debate will continue to be a serious one, because national identity has
been for long bound up with national self-preservation and the national Apos-
tolic Church, and many policy questions about freedom have to be sorted out af-
ter decades of Communist intervention (Khachikyan 2010: chs. 16-23).

95
Here using Habermas 1993 (while reflecting on Max Weber’s theory of secularization).
96
For foreground, Berger 1963: ch. 8.
97
For sidelight, Taussig 1993: 18.
98
Begin with Berkowitz 1999; Spragens 1999.
 The Ararat Factor 659

BIBLIOGRAPHY
Adams, D. W. (ed) (1983), Jefferson’s Extracts from the Gospels, Princeton.
Adams, H. P. (1935), The Life and Writings of Giambattista Vico, London.
Aquinas, Thomas ([1274] 1485), Summa Theologiae, Basel.
Axtell, J. L. (1965), “Locke, Newton, and the Elements of Natural Philosophy,” Paedagogia Europaea
1: 235-245.
Baker, D. E. (1747) (trans.) Voltaire, The Metaphysics of Sir Isaac Newton, etc., London.
Ballou, H. (1856), Sermons on Important Doctrinal Subjects, Boston.
Barber, W. (1979), “Voltaire and Samuel Clarke,” Studies in Voltaire and the Eighteenth Century 179:
47-65.
Barrow, I. (1680), Of the Love of God and our Neighbour, in Several Sermons, London.
Barth, K. (1947), Die protestantische Theologie im 19. Jahrhundert, Zürich.
Bentham, J. (1843), “The Book of Fallacies”, Works, ed. J. Bowring, Edinburgh.
—— (1967), An Introduction to the Principles of Morals and Legislation, ed. W. Harrison, Oxford.
Berg, J. van den / E. van der Wall (eds) (1988), Jewish-Christian Relations in the Seventeenth Century:
Studies and Documents, Dordrecht.
Berger, P. (1963), Invitation to Sociology: A Humanistic Perspective, Garden City, NY.
Berkowitz, P. (1999), Virtue and the Making of Modern Liberalism, Princeton.
Berlin, I. (1969), Four Essays on Liberty, Oxford.
Birkhead, A. (1917), The Story of the French Revolution, London.
Blavatsky, H. P. (2003), “Letters”, Collected Writings, vol. 15, Wheaton.
Blum, C. (1974), Diderot: The Virtue of a Philosopher, New York.
Bossy, J. (1985), Christianity in the West 1400-1700, Oxford.
Boswell, J. (1952), On the Grand Tour: Germany and Switzerland, New York.
Boyle, R. ([1645] 1991), “Aretology or Ethicall Elements”, The Early Essays and Ethics of Robert Boyle,
ed. J.T. Harwood, Carbondale.
Brewster, D. (1860), Memoirs of the Life, Writings, and Discoveries of Sir Isaac Newton, vol. 2, Edin-
burgh.
Brow, R. (1966), Religion: Origins and Ideas, London.
Brucker, P. (1742-1767), Historia Critica Philosophiae a Mundi Incunabulis ad nostram usque Aetatem
Deducta, 6 vols, Leipzig.
Brinton, H. (1973), The Religious Philosophy of Quakerism, Wallingford.
Brooke, J. H. (1991), Science and Religion: Some Historical Perspectives (Cambridge History of Sci-
ence Series 9), Cambridge.
Burleigh, M. (2006), Earthly Powers: Religion and Politics in Europe from the Enlightenment to the
Great War, London.
Burnaby, J. (1938), Amor Dei: A Study of St. Augustine’s Teaching on the Love of God as the Motive of
Christian Life, London.
660 Garry W. Trompf

Cassirer, E. (1955), Philosophy of the Enlightenment, trans. by F. C. A. Koellin; J. P. Pettegrove, Bos-

ton.
Chakmakjian, P. (2008), “Theological Lying and Religious Radicalism in Anderson’s Constitutions,”
Aries 8: 167-190.
Chapman, R. (1970), The Victorian Debate: English Literature and Society 1832-1901, London.
Christie, W. (2009), The Edinburgh Review in the Literary Culture of Romantic Britain: Mammoth and
Megalonyx (The Enlightenment World, 15), London.
Christianson, G. (1984), In the Presence of the Creator: Isaac Newton and his Times, New York.
Clarke, S. (1712), “The Scripture Doctrine of Unity”, Works, London, 1838.
―― (1717), Remarks upon a Book Entitled A Philosophical Enquiry concerning Human Liberty, Lon-
don.
Clorfene, Ch. / Y. Rogalsky (1987), The Path of the Righteous Gentile: An Introduction to the Seven
Laws of the Children of Noah, Southfield, Mich.
Corvinus, J. (1702), “Die verschmitze Welt-Mann und scheinheilige Tyranne[i] in England: Olivier
Cromwell, etc.”, Anabaptisticum et Enthusiasticum Pantheon, etc., Koethen.
Cowherd, R. G. (1956), The Politics of English Dissent, New York.
Crombie A. (1994), Styles of Scientific Thinking in the European Tradition: The History of Argument
and Explanation, etc., London.
Cullen, C. T. (ed) (1986) Papers of Thomas Jefferson, 2nd ser., Princeton.
Daniélou, J. (1964), The Theology of Jewish Christianity, London.
Davies, F. (1968), ‘Introduction,’ to Goldoni’s The Venetian Twins, etc., Harmondsworth.
Dihle, A. (1962), Die goldene Regel: Ein Einführung in die Geschichte der antiken und frühchristlichen
Vulkgärethik, Göttingen.
Dillenberger, J. (1961), Protestant Thought and Natural Science: A Historical Study, London.
Domingues, B. (1996), Tradiçao na modernidade e modernidade na tradiçao: A modernidade ibérica
e a revoluçâo copernicarna, Rio de Janeriro.
Duncan, G. (1973), Marx and Mill: Two Views of Social Conflict and Social Harmony, Cambridge.
Edwards, C. S. (1981), Hugo Grotius: The Miracle of Holland, Chicago.
Epstein, K. (1975), The Genesis of German Conservatism, Princeton.
Fara, P. (2004), Pandora’s Breeches: Women, Science and Power in the Enlightenment, London.
Feingold, M. (ed.) (1990), Before Newton: The Life and Times of Isaac Barrow, Cambridge.
Fénelon, François de S. La Mothe- (1699), Les aventures de Télémaque, fils de’Ulysse, 3 vols., The
Hague.
Flew, A. (1982), “Introduction”, Malthus, Th. (1798), An Essay on the Principle of Human Population,
Harmondsworth.
Gascoigne, J. (1999), “The Royal Society and the Emergence of Science as an Instrument of State
Policy,” British Journal of the History of Science 32: 171-184.
Gavre, M. (1974), “Hobbes and his Audience: The Dynamics of Theorizing”, American Political Sci-
ence Review 68.4: 1542-1556.
Giacometti, Z. (1974), Quellen zur Geschichte der Trennung von Staat und Kirche, Aalen.
 The Ararat Factor 661

Girvetz, H. (1966), The Evolution of Liberalism, New York.

Godwin, W. (1973), Enquiry concerning Political Justice and its Influence on General Virtue and Hap-
piness, 2 vols., Luke White.
Goldish, M. (1998), Judaism in the Theology of Sir Isaac Newton, Dordrecht.
Görres J. J. von (1958), “Über die Grundlage, Gliederung und Zeitenfolge der Weltgeschichte”, Ge-
sammelte Schriften, ed. Ernst Deuerlein, Cologne.
Guizot, F. (1823), Essais sur l’histoire de France, Paris.
―― (1911), The History of Civilization in Europe, Eng. trans., London.
Haakonssen, K. (2004), “Protestant Natural Law Theory: A General Interpretation”, New Essays on
the History of Autonomy: A Collection Honoring J. B. Schneewind, ed. Natalie Brendfer; Larry
Krasnoff, ch. 4, Cambridge.
Habermas, J. (1993), The Philosophical Discourse of Modernity: Twelve Lectures, translated by Fred-
erick Lawrence, Cambridge.
Halévy, E. (1934), The Growth of Philosophic Radicalism, translated by Mary Morris, London.
Hamburger, J. (1965), Intellectuals in Politics: John Stuart Mill and the Philosophic Radicals (Yale
Studies in Political Science, 14), Yale.
Hanegraaff, W. J. (2007), Swedenborg, Oetinger, Kant: Three Perspectives on the Secrets of Heaven,
West Chester.
―― (2010), “Western Esotericism in Enlightenment Historiography: The Importance of Jacob
Brucker”, Constructing Tradition: Means and Myths of Transmission in Western Esotericism,
ed. A. B. Kilcher, Aries Book Series 11, Leiden: 91-111.
Harrison, J. (1969), Quest for the New Moral World: Robert Owen and the Owenites in Britain and
America, New York.
Harrison, J. (1978), The Library of Isaac Newton, Cambridge.
Harrison, P. (1990),‘Religion’ and the Religions in the English Enlightenment, Cambridge.
Harrop, R. (1884), Bolingbroke: A Political Study and Criticism, London.
Haycock, D. A. B. (2002), William Stukeley: Religion, Science and Archaeology in the Eighteenth Cen-
tury, Woodbridge.
Hegel, G. W. F. (1837), “Vorlesungen über die Philosophie der Geschichte”, Hegel, Sämtliche Werke,
Jubiläumsausgabe, ed. H. Glockner, vol. 11, Stuttgart.
Himmelfarb, G. (1970), “The Other John Stuart Mill”, Perspectives on Political Philosophy 2: 352-355.
Hobbes, T. (1651), Leviathan, ed. M. Oakeshott, New York.
Huehns, G. (1951), Antinomianism in English History, London.
Hulliung, M. (1976), Montesquieu and the Old Regime, Berkeley.
Hunter, M. C. B. (1995), Science and the Shape of Orthodoxy: Intellectual Change in Late Seventeenth-
Century Britain, Woodbridge.
Hutcheson, F. (1750), A System of Moral Philosophy, Glasgow.
Jacob, C. M. (1991), Living the Enlightenment: Freemasonry and Politics in Eighteenth Century Europe,
Oxford.
Jones, B. E. (1950), The Freemason’s Guide and Compendium, London.
662 Garry W. Trompf

Kant, I. (1781),Vorlesungen über die philosophische Moral, Königsburg.

―― (1789), Kritik der reinen Vernunft, Königburg.
―― (1963), Lectures on Ethics, translated by L. Infield, New York.
―― (1995), Werke in sechs Bänden, ed. Rolf Toman, Cologne.
Khachikyan, A. (2010), History of Armenia: A Brief Review, Yerevan.
Kipling, R. (1901), Kim, London.
Kristeva, J. (1988), Strangers to Ourselves, translated by Leon S. Roudiez, New York.
Kulklick, B. (2004), A History of Philosophy in America, 1720-2000, Oxford.
Lecler, J. (1960), Toleration and the Reformation, 2 vols., London.
Lehmann-Brauns, S. (2004), Weisheit in der Weltgeschichte: Philosophiegeschichte zwischen Barock
und Aufklärung, Tübingen.
Lichtenberg, G. Ch. (1990), Sudelbücher (J: 1789-93), translated by Reginald J. Hollingdale, London.
Littré, É. (1954), Dictionnaire de la langue française, Paris.
Lively, J. / J. Rees (eds.) (1978), Utilitarian logic and politics: James Mill’s “Essay on Government,” Ma-
caulay's Critique, and the Ensuing Debate, Oxford.
Locke, J. (1689), Two Treatises of Government, London.
―― ([1690] 1961) An Essay Concerning Human Understanding, vol. 1, London.
―― ([1696]1958), Reasonableness of Christianity, ed. and abrdg., I. T. Ramsey, London.
Lukacs, J. (1959), “Introduction”, to his translation and editing of Tocqueville, “The European Revo-
lution” and Correspondence with Gobineau, Garden City, NY.
De Mably (1849), Théories sociales et politiques, etc., ed. P. Rochery, Paris.
Maimonides, (1683), Mishneh-Torah, London.
Manuel, F. E. (1962), The Prophets of Paris, New York.
―― (1974), The Religion of Isaac Newton, Oxford.
Markley, R. (1999), “Newton, Corruption, and the Tradition of Universal History”, Newton and Reli-
gion: Context, Nature and Influence, eds. Force J. E.; Popkin R. H. Dordrecht: 121-143.
Marx, K. / K. Engels ([1848] 1951), “The Manifesto of the Communist Party”, Selected Works, Mos-
cow.
Mazlish, B. (1975), James and John Stuart Mill: Father and Son in the Nineteenth Century, London.
Mazzini, G. (1907), The Duties of Man, ed. B. King, trans. Ella Noys, London.
McCalman, I. (2003), The Last Alchemist: Count Cagliostro, Master of Magic in the Age of Reason,
New York.
O’Malley, J. (2004), Four Cultures of the West, Cambridge.
McGuire, J. / P. Rattansi (1966), “Newton and the ‘Pipes of Pan’,” Royal Society (Great Britain) Notes
and Records 21: 108-143.
McIntire, C. T. (1997), “Religious Liberty in France, Part I”, Freedom and Religion in the Nineteenth
Century, ed. R. Helmstadter, Stanford: 248-54.
McLachlan, H. (1950), Theological Manuscripts, Liverpool.
―― (1961), The Correspondence of Isaac Newton, ed. H.W. Turnbull, Cambridge.
Mill, J. (1817), History of British India, London.
 The Ararat Factor 663

―― (1821), Elements of Political Economy, London.

Mill, J. S. (1859] 1910), On Liberty, London.
―― ([1865] 1961), Auguste Comte and Positivism, Ann Arbor.
―― (1924), Autobiography, Oxford.
―― (1955), Essay on Government, ed. C. V. Shields, Indianapolis.
―― (1970), Essays on Sex Equality, ed. A. S. Rossi, Chicago.
Montesquieu (1734) Considérations sur les causes de la grandeur des Romains et de leur decadence,
Amsterdam.
―― (1755), L’Esprit des lois, Amsterdam.
Nelson, E. (2007), “Talmudic Commonwealthmen and the Rise of Republican Exclusivism,” The His-
torical Journal 50: 809-835.
Newton, I. (1693-1694), “Original of Monarchies”, Isaac Newton, Historian, by F. Manuel (1963),
Cambridge: 198-221.
―― (1725), Abregé de la chronologie de M. le chevalier Isaac Newton, etc., translated by Nicolas
Fréret, Paris
―― (1728), Chronology of Ancient Kingdoms Amended, London, 1728.
―― (1782), “Opticks”, Opera quae exstant omnia, ed. S. Horsley, vol. 4, London.
Oizermann, T. I. (1981), The Making of Marxist Philosophy, translated by Yuri Sdobnikov, Moscow.
Paine, Th. ([1791-2] 1961), The Rights of Man, Garden City, NY.
Passos, J. D. (ed) (1940), The Living Thoughts of Thomas Paine, London.
Pike, A. (1906), Morals and Dogma of the Ancient and Accepted Scottish Rite of Freemasonry,
Charleston.
Pilbeam, P. (2000), “Dream Worlds? Religion and the Early Socialists of France”, Historical Journal
43.2: 499-515.
Pollard, A. F., (1907), Factors in Modern History, London.
Pomeau, R. (1956), La religion de Voltaire, Paris.
Pope, A. (1906), The Poetical Works, with Memoir, Explanatory Notes, etc., London.
Postigliola, A. (1999), “Montesquieu entre Descartes et Newton”, Montesquieu: Les années de for-
mation (1689-1720), ed. Catherine Volpilhac-Auger: Cahiers Montesquieu 5: 91-108.
Prideaux, H. (1723), The True Nature of Imposture fully display’d in the Life of Mahomet, etc., London.
Priestly, J. (1965), “An Essay on the First Principles of Government”, Priestly, J. Writings on Philoso-
phy, Science and Politics, ed. J. A. Passmore, New York.
Rapaczynski, A. (1987), Nature and Politics: Liberalism in the Philosophies of Hobbes, Locke, and
Rousseau, Ithaca.
Renaudin, P. (1941), Résurrection de la France, 1589-1640, Paris.
Reynolds, H. (1992), The Law of the Land, Melbourne.
Robbins, C. (1961), Eighteenth-century Commonwealth Men: Studies in the Transmission, Develop-
ment and Circumstance of English Liberal Thought, Cambridge.
Roberts, J. M. (1972), The Mythology of the Secret Societies, London.
664 Garry W. Trompf

Roberts, W. Jr. (1971), “Behavioural Factors in Bentham’s Conception of Political Change”, Perspec-
tives on Political Philosophy, ed. D. H. Hart; J. V. Downton, Jr, New York.
Rogers, G. A. G. (1978), “Locke’s Essay and Newton’s Principia,” Journal of the History of Ideas 39.2:
217-32.
Rousseau, J.-J. ([1761] 1960), Julie, ou la nouvelle Héloïse, XIV, lettr. 12, ed. Pomeau, Paris.
―― ([1755] 1971), Discours sur l’origine de l’inégalité, Paris.
―― ([1762] 1966), Du contrat social, Bk. I, 6; II, 4, ed. P. Burgelin, Paris.
―― ([1782] 1953), Confessions, translated by John M. Cohen, Harmondsworth.
―― (1915), The Political Writings of Jean-Jacques Rousseau, ed. C. E. Vaughan, vol. 1, Cambridge.
Rowland I. (2000), The Ecstatic Journey: Athanasius Kircher in Baroque Rome, Chicago.
Sabine, G. H. (1951), A History of Political Theory, London.
Schapper, K. (1948), “Brief an das ‘Kommunistische Korrespondenz-Komitee’ vom 17, Juli 1846”,
Fragen der Geschichten 2: 5-23.
Schneewind, J. B. (1997), The Invention of Autonomy: A History of Modern Moral Philosophy, Cam-
bridge.
Scott, J. (1927), “Socrates and the Golden Rule,” Classical Journal 22.6: 452-453.
Selden, J. (1640), De iure naturali et gentium, iuxta disciplinam Ebraeorum, London.
―― (1679), De synedriis et praefecturis veterum Hebraerorum, Amsterdam.
Sell, A. P. F. (2004), Mill on God: The Pervasiveness and Elusiveness of Mill’s Religious Thought, Al-
dergate.
Shaw, B. (1985), “The Divine Economy: Stoicism as Ideology”, Latomus 43: 16-54 (ancient);
Siedentop, L. (1979), “Two Liberal Ideologies”, The Idea of Freedom: Essays in Honour of Isaiah Ber-
lin, ed. A. Ryan, Oxford.
Sieveking, A. F. (1908), Sir William Temple upon the Gardens of Epicurus, etc., London.
Spencer, J. (1685), De legibus Hebraeorum, Cambridge.
―― (1892), Social Statics, together with Man versus the State, London.
Spragens, Th. A. (1999), Civic Liberalism: Reflections on our Democratic Ideals, Lanham.
Sprat, Th. (1959), History of the Royal Society, eds. J. I. Cone; H. W. Jones, St. Louis.
Stimson, D. (1917), The Gradual Acceptance of the Copernican Theory of the Universe, Doctoral
dissert., Columbia University.
Stukeley, W. (1836), Memoirs of Sir Isaac Newton’s Life, etc., ed. A.H. White, London.
Stumpf, Ch. (2006), “The Grotian Theology of International Law: Hugo Grotius and the Moral
Foundations of International Relations”, Berlin.
Swedenborg, E. (1750), Arcana Coelestia, translated by John Marchant, vol. 2, London.
―― (1975), The True Christian religion, translation by W. C. Dick, London.
Talmon, J. L. (1952), The Origins of Totalitarian Democracy, London.
Taussig, M. (1993), Mimesis and Alterity: A Particular History of the Senses, London.
Tillotson, G. (1958), Pope and Human Nature, Oxford.
Tindal, M. ([1730] 1978), Christianity as Old as Creation, or the Gospel as a Republication of the Reli-
gion of Nature, New York.
 The Ararat Factor 665

Thomas, J. L. (1983), Alternative America: Henry George, Edward Bellamy, Henry Dumarest and the
Adversary Tradition, Cambridge.
Tocqueville, A. de ([1856] 1916), L’Ancien régime et la révolution, ed. G. W. Headlam, Oxford.
―― (1998), Democracy in America, translated by Henry Reeve, Ware.
Torrey, L. N. (1963), Voltaire and the English Deists, Oxford.
Thomas, D. L. (1995), Routledge Philosophy Guidebook to Locke on Government, New York.
Trompf, G. W. (1969), “Radical Conservatism in Herbert Spencer’s Educational Thought”, British
Journal of Educational Studies 17: 267-80.
―― (1991), “On Newtonian History”, The Uses of Antiquity, ed. Gaukroger S., Amsterdam: 213-249.
―― (1994) “Vico’s Universe: La Provvedenza and la Poesia in the New Science of Giambattista
Vico”, British Journal for the History of Philosophy 2.: 55-86.
―― (2005), “Isaac Newton and the Kabbalistic Noah,” Aries N.S., 5, 1 (2005): 91-118
―― (2010), “On the Origins of the Western Meditation Movement”, Pathway to the Centre – Purity
and Mind, ed. Edward Crangle, Sydney.
―― (2015) The Idea of Recurrence in Western Thought, vol. 2, Berkeley.
Usher, J. (1722), Annales Veteris et Novi Testamenti, etc., Geneva.
Vernière, P. (ed) (1956), Oeuvres philosophiques, Paris.
Vico, G. (1725, 1744), Principij di [una] scienza nuova, Naples.
―― ([1720] 1936), De uno universi iuris principio et fine uno, ed. Fausto Nicolini, Bari.
―― (1947), Autobiografia, ed. F. Nicolini, Milan.
Viereck, P. R. E. (1987), “Socio-Economic Doctrines and Reform Movements”, Encyclopedia Britan-
nica, Chicago.
Voegelin, E. (1974), From Enlightenment to Revolution, ed. J. H. Hallowell, Durham.
Voltaire (1740), La Métaphysique de Neuton, ou parallèle des sentiments de Neuton et de Leibnitz,
Amsterdam.
―― (1749), “Sermon des cinquante”, Oeuvres completes, vol. 26, Paris, 1829.
―― (1765), L’Évangile de la raison, London.
―― ([1734] 1937), Traité de métaphysique, ed. H. T. Patterson, Manchester.
―― ([1733] 1964), Lettres philosophiques, ed. R. Naves, Paris.
―― ([1764] 1972), Dictionnaire philosophique, trans. by Th. Besterman, Harmondsworth.
―― ([1740] 1992), “Élements de la philosophie de Neuton, Contenant la métaphysique, la théorie
de la lumière, et celle du monde”, Oeuvres completes, vol. 15, Oxford.
Waldron, J. (2002), God, Locke, ad Equality: Christian Foundations of John Locke’s Political Thought,
Cambridge.
Walker, D. P. (1972), The Ancient Theology: Studies in Christian Platonism from the Fifteenth to the
Eighteenth Century , London.
Wallas, G. (1898), The Life of Francis Place, 1771-1854, London.
Walters, R. (1954), Voltaire and the Newtonian Universe, Doctoral Dissert., Princeton University.
Warnock, M. (ed.) (1962), Utilitarianism, London.
Weber, E. (1963), Vincent Pallotti: An Apostle and Mystic, Staten Is., NY.
Westfall, R. (1980), Never at Rest: A Biography of Isaac Newton, Cambridge.
666 Garry W. Trompf

―― (1982), “Isaac Newton’s Theologiae Gentiles Origines Philosophicae”, The Secular Mind:
Transformations of Faith in Modern Europe: Essays presented to Franklin L. Baumer, ed. W.
Warren Wagar, New York: 15-34.
Whiston, W. (1696), A New Theory of the Earth, from its Original to the Consummation of All Things,
London.
Wilson, F. (1998), “Mill on Psychology and the Moral Sciences”, The Cambridge Companion to Mill,
ed. J. Skorupski, Cambridge: 209-221
Wolf, E. (1944), Grosse Rechstdenker der deutschen Geistesgeschichte, Tübingen:
Woodcock, G. (1963), Anarchism, Harmondsworth.
Woolrych, A. (1982), Commonwealth to Protectorate, Oxford.
Zastoupil, L. (1988), “J. S. Mill and India”, Victorian Studies 32: 31-54;
 Communication and the Oral-Aural
Traditions of an East-Anatolian Ethnicity:
What us Stories tell!

Eberhard Werner
SIL International

Abstract
The Zaza people of East Anatolia have recently been moving from an oral-aural (the relationship of
a speaker to a hearer) society to a writing or literate society. In this way they are striving to stimu-
late the use of their language, so as to escape their language becoming endangered, to adapt to a
world based on writing (globalization), to enhance the education of children and the training of
adults (the process of cultural identification), and lastly to investigate their own oral-aural history.
Century-old traditions that are passed on from one generation to the other, from father to son,
from mother to daughter, from the grandparents to their grandchildren and from one social class
(religious or political) to the other (class-crossing) all representing the heritage of the ethnicity. In
addition to that, due to language endangerment, the preservation and the publication of those tra-
ditions are essential to the cultural and linguistic survival of the Zaza ethnos. The stories and tradi-
tions carry the knowledge and wisdom of the people, and reflect their recent and historical envi-
ronment as well as their anthropological-linguistic configuration. History, it seems, becomes the
key to the future. Key elements in handing down traditions, and cultural-linguistical reflections of
anecdotes, stories, songs and rituals published in mother-tongue newspapers, books and other
publications are under investigation here. The main reference is from “Mahmeşa”, a book which I
had the pleasure of taking part in publishing. The topic fits well with the excellent research that
Prof. Dr. Asatrian brought into the discussion about the Zaza ethnos through the magazine Iran
and the Caucasus.

Keywords
Zaza, Zazaki, Orality, Oral-aural, Tradition, Communication, Translation, Transmission, Culture,
Story, Text, Discourse

COMMUNICATION AND ORAL-AURAL TRADITIONS

“History”, as is often said, “is the key to the future.” Orality performs such a key
by opening treasures of anthropological and linguistic wisdom.

 Koninklijke Brill NV, Leiden, 2015

668 Eberhard Werner

The point of view taken here is from a communicational stance. The frame-
work for this study comprises communication sciences, linguistics, sociology, the
science of interpreting and translation, psychology and text discourse.
Communication as interaction demands a speaker who presents his infor-
mation, message or talk in an oral way to an audience. The hearer takes the in-
formation aurally up and processes it cognitively. In the passing on of traditions,
communication thus represents an oral-aural process. The basic principle of
communication is a combination of proceedings, such as encoding, or the act of
sending information, by a sender (the oral part) and a decoding, or the act of re-
ceiving by an audience (cognitive-aural part; Shannon/Weaver 1949). However,
in addition to this functional- and dynamic-equivalent process (Nida/Taber
1969) the communication participants of an oral-aural speech act assume a
commonly shared encyclopaedic lexicon and the mutual expectation of suc-
cessful communication (Sperber/Wilson 1995: 50, 161, 174; Gutt 2000: 32; Rele-
vance Theory). The communication theory states that a speaker follows the “low
cost—high benefit” rule. A communication or speech act is closed up and thus
“successful”, when following this Minimax principle. In oral-aural traditions the
main goal is to stimulate the audience with historical, cultural or mother-tongue
awareness and to bring them to the same cognitive level as the speaker (osten-
sive-inferential theory).
The subject under investigation is the Zaza ethnicity of East Anatolia. Much is
said about them, their language, their culture and their social environment
(Todd 2002; Bläsing 1995; Asatrian/Gevorgian 1988; Keskin 2008; Paul 1998; idem
2009; idem 2012; Selcan 1998a; idem 1998b; Asatrian 2009; Söylemez 2012). This
three to four million people group inherits the headwater of the Euphrates-Tigris
area (Zazaki: Ro, Fırat, Diçle), including the river beds of the Murat and the Mun-
zur river. The Zaza people are religiously split in an Alevism (Shi’a sect) practic-
ing Northern Group (NZ; Dersim area; Ovaçik; Mamekiye, Pulumoriye, Hozat,
Varto) and a Sunnism practicing Southern (SZ; Çermik, Siverek; Hanafi) and
Eastern Group (EZ; Palu, Bingöl, Hani; Hanafi and mainly Shafi’i). An approxima-
tion of the Eastern Group to the Kurmanji speaking ethnicity is amplified by
their shared Shafi’i orientation. A linguistic-dialectical split follows a religious
divide. One can speak about three main dialects referred to as the Northern Der-
sim-Varto Alevi Dialects, the Southern Çermik-Siverek dialects and the Eastern
Palu-Bingöl dialects (sometimes called Central dialects). They represent a collec-
tivistic and tribally segmented, patriarchal and endogamy practising culture. The
 Communication and the Oral-Aural Traditions of an East-Anatolian Ethnicity 669

dialect of the NZ is said to be endangered, because of assimilation processes into

the national language, Turkish (Todd/Werner 2012: 92). The Southern and East-
ern dialects are still actively used at home and in public. Lack of official mother
tongue education jeopardizes the whole language group (ibid.: 96). Mother
tongue inherits a sort of sacredness, because it provides ethnicities with ethno-
centrism, a powerful aspect of self-defence which is basic for language survival
(ibid.: 93, 99). Zazaki (Turkey; East Anatolia) resembles other “Kurdish lan-
guages”, like Kurmanji (Turkey; East Anatolia), Sorani and Bahdeni (Northern
Iraq) and the South and West Caspian languages (Iran; e.g. Gilaki, Talysh). How-
ever, text discursive features of Zazaki differ or are unique in some ways (see
below). The religious foundation is performed by Alevism of the Northern Group
and partially of the Eastern Group (Bingöl area), and Sunnism of the Southern
Group (Islamic school of Hanafi) and partially the Eastern Group (Islamic school
of Shafi’i). Also some influence from mystical orders contributes to this founda-
tion, such as the Sufism following Mevlana, the Haci Bektaş (Shi’a) and the
Naqşebendi (Sunni) orders.
In this first section we want to have a closer look on the term tradition(s) and
their passing on as an oral activity. In a second step we examine the phrase oral-
aural and its context.
In contrast to the peoples of the Near and Middle East, the western world
(from my western ethnocentric focus) forgot about the traditional oral-aural
ways of passing on their traditions. In the West traditions were passed on in writ-
ten form, in the modern age of literacy, especially since the use of paper (15th-
16th century AD). However, numerous people groups still function on an oral-
aural level, passing on their cultural and linguistic heritage by handing down
traditions.
The term tradition has a double meaning. First, as a culturally-social institu-
tion it is a general term for passing on the ethnicity's knowledge, as a cognitive
heritage, from generation to generation. Tradition becomes the description for
the methods and the principles used thereby (tradition of a people group). Se-
cond, it is an individual element of the overall process limited to one singular
genre (e.g. a story, a song or a poem). Tradition has also a double function. First, it
mirrors the core values, included in the worldview and conscience of the people
group, and thus describes the processes of social interaction, such as encultura-
tion or processing of traditions. Second, as an institution it brings a society to-
670 Eberhard Werner

gether and forms unity by shaping the individuals and the people group’s world-
view and conscience, both are basic to ethnocentrism.
Traditions are part of communicational, linguistically, discursive, interpreta-
tional, translational and social human interactions. The tradition of an ethnicity
reflects either a ritual or is presented in a ritualized way. It is a mode of thought
or behaviour that a people group follows continuously from one generation to
the next. As a result it generates an unwritten custom or practice (a ritual). A set
of such rituals is viewed as a coherent body of precedents that influence and con-
trol the present. A tradition also comprises a body of unwritten fixed religious
precepts or a time-honoured practice or set of such practices which are due to
culture change and which drive the ethnicity that are a people’s worldview and
behaviour.
Individual traditions are not generally open to everybody, some are for spe-
cific events and others are only for a particular group of the community. That is
why myths, saga and religious anecdotes do not fall under one category but need
to be differentiated (Kohl 1993: 75). Only slowly do we get a feeling about the
wealth of Zaza traditions as they are coming out of the shadow of their unwrit-
ten cultural heritage. Traditions not only tell us what has to be done (activities,
actions, profane), they also tell about the realm of the taboo and restriction in
public life (sacred and hidden areas). The former is passed on by imperatives,
derivations or practical implementation by the addressed audiences in daily life
(explicit communication). The latter is part of the implicit and inferable content
of speech, act and communication. For instance, take the story Lazê Axay: while
it is not explicitly stated in the story, it is inferable from the text that the father
(an ağa) yearns for his son to come back after their struggle. Also, although the
mother is only mentioned in the beginning, the agha’s whole family is repre-
sented by the father in the context of the patriarchal Zaza society (Hayıg 2007:
26-32). Yet, that loyalty and cohesion in family are important powers is the un-
stated essence of the story.
Traditions alone do not provide for a perception of the world and are not the
only way of enculturation, of building worldview or to form the conscience. They
are just one part of the on-going and complex processes of socialization and ac-
culturation. This has to be kept in mind when the effects of the oral-aural Zaza
traditions in the audience are considered.
What do the terms oral, orality as well as oral-aural include and describe?
The main difference of both concepts—oral vs. aural—comes from their align-
 Communication and the Oral-Aural Traditions of an East-Anatolian Ethnicity 671

ment. Orality is generated by a sender and directs towards the transmission of

information. The aural part of tradition goes with the audience or receiver. It is
through him/her that the information or message is deciphered and developed
into her worldview, particularly into the encyclopaedic storage and the con-
science. Orality is differentiated in primary and secondary orality. The former de-
scribes cultures like the Zaza ethnicity, which was always only passed in oral
forms for their cultural and linguistic heritage. The latter term refers to people
groups that move into the stage of orality (e.g. modern Western youth). In sec-
ondary orality, the oral transmission becomes a means to an end (Ong 1988: 31-
77). Orality describes the innate and dynamic process, the function and the af-
fect, of passing on ethnicity's cultural and linguistic information, traditions, and
instructions, as well as their mutually shared knowledge and history (Kohl 1993:
72-73; Maxey 2009: 84). In orality the basic moral and ethical meaning of the
“text” (content of narration) is stable and culturally fixed, but its circumstances
or emphasis on specific features are flexible. As a consequence, the representa-
tion of a tradition leads to a three-dimensional perception of the world, while
written traditions represent the world two-dimensionally. Orality is performed
in social gatherings, communicational interactions, media, arts, music (e.g. Neyzi
2002 investigates the Zaza brothers Metin and Kemal Kharaman who are both
Musicians), by language or by symbols (gesture, mimic, icons, and graphics). It is
essential to enculturation. The succeeding generation is introduced to the reli-
gious, social, psychological and linguistic characteristics of the ethnicity. Spe-
cialists, often religiously motivated (Zazaki: pir, şeikh, mıllah), are responsible for
the enculturation process and the initiation of community members. It is not
only these specialists that bring forth tradition but also the parents and individ-
uals of the society. The setting in which this happens are in the realm of the core
and extended family (Z: key, çê) and the community centres (Z: cem; cemaat).
Oral texts are “more involved interpersonally, whereas written texts are more de-
tached” (McCarthy 2001: 94).
The traditions under investigation are Mahmeşa (Hayıg 2007), Gulbahare:
Merselei “Narrations of Gulbahar”, Zonê gulu de qesei “Words in the Flower Lan-
guage” (Astare 2011), Sîyamed û Xeca (Berz 1993), Kole Nêba “[We are] No Slaves”
(Berz 1995), Ewro Şorî Meşti Bêrî “Today Go Tomorrow Come” (Berz 1996), Morı-
ber (Cengiz 2011), Dersim Fablları-I “Fables from Dersim” (Cengiz 2001), Hazar
Reng—Hazar Veng “Spring Colour—Spring Voice” (Dewran), Folklorê ma ra çend
numûney “Some Tellings from our Traditions” (Malmisanıj 2000), Budelaê Gırşi
672 Eberhard Werner

“Big Fools” (Mergarıji 1997; Çimen 2002), Geçmişten Günümüze Zazaca Dergiler
“Zaza Magazines from Past to Present” (Söylemez 2012), Külden Evler “Burning
Houses” (Taş 2007), Dersim Türküleri—Tayê Lawıkê Dersımi—Şiwari Ağıtlar
“Some Songs from Dersim” (Düzgün 1992) and the musical CD’s and Tapes
named Kilıtê Kou “Key of the Mountain” (Aslan 2003), Lauke Dersım “Songs from
Dersim” (Çapan 1999), Şêlê “Sele” (Zelê Melê 1999), Alemano “Germany” (Barihas
1999) and Ra u Rêçe “Way and Demand” (ibid.). Other works are also taken into
consideration and specifically mentioned are the publishing houses Tıj Yayınları,
Berdan, Vatê Yayınevi (all in İstanbul) and İremet Förlaget (Stockholm), as well as
the magazines Ayre, Ware and Vatê made a particular contribution to the
preservation of oral traditions. As a side effect they affect the ongoing standardi-
sation processes on a written alphabet and the version of a dialect chosen as
Standard-Zazaki.

GENERAL OBSERVATIONS
We will begin with the locations in which traditions are progressed. Traditionally
the Zaza people met at the cem or cemaat, a social gathering (Çimen 2002: 17-22).
The term cem describes both the location (house or room) as well as the social
institution. All members of the society took part in those meetings (women,
children, disabled and old people). Gender equality was practiced there, alt-
hough leading positions were filled by men only. Because of its central and im-
portant function, outsiders were suspicious of the cem of the Alevism-following
Northern Zaza group. Ignorance and envy brought forth accusations of religious
and sexual defilement during the meetings. Unfortunately the Sunnism-follow-
ing Zaza people did follow these indictments, by criticising the Alevis too,
although they also had met in cem-gatherings. Thus, there is strong evidence that
the Southern and Eastern Group was formerly also Alevi, or at least that the
whole group was formerly following one and the same religious, cultural and
linguist orientation (Sökefeld 2008: 7-15). Beside the cem/cemaat gatherings, the
core and extended family performed the other main space for passing on
traditions. Even during weddings or holidays, e.g. lent or period of fasting
(Ramazan; a unique four-/five-day fasting in March for Alevis; Kormışkan; Rozê
Gağan; etc.), there was always a formal or informal time of sharing traditions
(RH personal communication January 2012).
We are focussing now on the persons and social institutions that were re-
sponsible for the passing on of traditions. Every individual (man and woman),
 Communication and the Oral-Aural Traditions of an East-Anatolian Ethnicity 673

religious and political leaders as well as religious specialists (Z: pir, mıllah, şeikh,
ağa/aga) of the upper class could participate in the transmission of traditions.
The latter were mainly interested in the education and enculturation of the eth-
nicity, because of maintaining ethnocentrism and the flow of power play by re-
counting such traditions. Preferred story tellers were older and experienced
people of the ethnicity. During childhood the mother takes on the responsibility
for guaranteeing the enculturation of the children by transmission of and educa-
tion in these traditions. With the initiation, either by circumcision (boys) or sex-
ual maturity (girls), the father became responsible (e.g. figured in the struggle of
father vs. step-mother in Hayıg 2007: 3, 50). The girls of a family were incorpo-
rated in the service of the household and farming by the mother (e.g. ibid.: 21-22).
The grandparents, as the respected elders, always play an important role in the
enculturation and tradition process for both genders, too (e.g. ibid.: 15, 45).
The Zaza traditions are playing within the real and physical world, as well as
within an after- or counterworld that is a non-physical imaginary reality. Since
the main actors can move between both realities we do have a reflection of vi-
sionary and real life scenery. The social means of passing on traditions were part
of daily life activities, as well as special events in the course of life. These were
passed on as
• stories/anecdotes,
• fables,
• folktales,
• narratives/parables,
• songs,
• music/dance,
• poems,
• recitations or
• theatre.
We will split this research into: 1.) Main focuses of anthropology in Zaza tra-
ditions, 2.) Linguistical and text discourse features of Zaza traditions and the 3.)
Pedagogical effects of oral-aural Zaza traditions.

ANTHROPOLOGICAL FOCUS
In general, Zaza traditions are expressing the ethnicity's deeply anchored desires
or longings. Such core values of culture, echoed in these desires or longings are
framed by a description or observation of daily life. In between these frame-
works, one gets information about the rural and strenuous life of Zaza peasantry.
674 Eberhard Werner

The deeper lying morals and ethics become either explicitly or implicitly obvious
by sometimes expected and sometimes unpredictable and strange events or
state of affairs. However, the aim is mainly to pass on cultural and linguistic
information about the past to preserve and convey old customs and beliefs into
the future.
a. Main Focus
The main focuses in the Zaza traditions are on the desires and longings towards:
• the welat (homeland),
• the keye/çê or ezbet (relatives, family, tribal unit, close social segment).
Beside those, important themes are
• the êşk (Tk. aşk; love for opposite gender) between opposite gender, and
• the maintenance of reputation (namûs) by upholding the nome/name (name,
honour, prestige and pride)
The following social institutions are facilitators to transmit and preserve
these culturally highly valued customs and beliefs:
• hierarchical leadership
o pirdom,
o aghadom,
o sheikhdom and
• mullaharchy as well as social stratification
o core and extended family (pi, bawo, dayê, dapir, keko, dedo, xalo),
o relatives and musahıb/musayıb—chosen brotherhood,
o tribal and small scale societal groupings (ezbet, kuflet, şar),
• a three-class system
o upper class (leaders; pir, şeik, mıllah, axa)
o middle class (peasants and craftsmen),
o lower class (foreigners—non-Zaza, widows, handicapped; ‘Ereb; ‘siyak;
aşık etc.).
We will start with the strong desire towards the welat (homeland; love of
country). This topic is not just expressed by and for those in the Diaspora (either
outside Turkey or in Western Turkish cities), but plays also a huge role in narra-
tives, stories and songs coming from the homeland. The hero of the story Lazê
Axay (the Agha’s Son; Hayıg 2007: 26-32) is forced to leave his father’s (an agha)
house, due to struggle with his father. He leaves his house and seeks his fortune
far away. Nothing is said about his father’s longing, but the son goes to another
axa/ağa (agha), which conveys the strong class-affiliation of the Zaza people.
The shame-orientation of conscience, either individually or as people con-
 Communication and the Oral-Aural Traditions of an East-Anatolian Ethnicity 675

science, becomes obvious by the fact demonstrated here, that one will not lose
or miss his social status quo (Muller 2001; Müller 2009). Even, while the Agha’s
son lost or damaged his namûs (honour), he will hold his rank (Hayıg 2007: 27;
see also Mahmeşa: 43-44). The lesson given here is about strong group cohesion,
exemplified in the defence of the geographical area that one comes from. The
traditions and song ask the audience to do everything to support the homeland
and the Zaza people. It is interesting that the European Diaspora of the Zaza
people adheres to this desire by regular visits to the homeland and huge finan-
cial support of the ones left behind. A huge travel market lives based on this
need (e.g. banks, travel agencies, airlines like Onur Air, Sun Express). Also, Zaza
members living in the main Western cities of Turkey are oriented towards their
homeland area.
Another strong feature of Zaza folktales and traditions is the desire to make a
fortune in the distance and come back to show what one accomplished in life
(e.g. rich wife; riches; glory and fame). Gorma’hmed “Gormahmed”, ‘Heso Çıhar-
çım “Four-Eyed Hesen”, Kecel Ahmed “Bald Ahmed” and Lazê Axay “The Agha’s
Son” are telling us about this. Especially with Lazê Axay there is some parallel to
the Biblical narrative of “The Prodigal Son” (Luke 15: 11-30). Re-unification and
implicit forgiveness are issues in these narratives. The prodigal son asked to get
his legacy to make his fortune. He was aware that he thereby lost face and was
not a legal heir of his family or even accepted as member of the community any
more (Deut 21: 18-21; see 5). In the Zaza traditions the reasons for leaving home
are struggle with respected/honoured family members, rejection by close or
distant family (e.g. father, stepparents), and betrayal. At the same time, they are
welcomed back to the family and the society, presented in ‘Elicanek u Warda
Xoya (Hayıg 2007: 3, 7), Keçel Ahmed (ibid.: 21, 23), the Lazê Axay “The Agha’s
Son” (ibid.: 26, 32), Mahmeşa (ibid.: 37, 44). Obviously loyalty and social cohesion
within the core and extended family and the tribal segments of the Zaza ethnos
play an important role. The oral-aural forewarnings, as part of the stories, display
the moral and ethical framework of the community, to the audience (youth;
responsible community members). Yet, the lesson is about:
• respect, by following the parents, the elderly and responsible people,
• loyalty towards the social group that collectively provides for the individual,
• avoidance of struggle within one's own small scale society, but
• defence of one’s own honour (individual prestige) and group prestige.
676 Eberhard Werner

b. Main Settings—Origin, Geography and Actors

The oral-aural traditions of the Zaza people reflect their geographical setting in
East Anatolia. Huge forests––although nowadays they are enormously reduced––
(e.g. Lazê Axay), high mountains (Taurus mountain range in stories figuratively
seven mountains; ibid.: 14, 15, 50), huge rivers (Euphrates and Tigris; Ro; ibid.: 7,
53), the Sea (dengız; only mentioned in Gorma’hmed ibid.: 11-12; hint about the
Caspian or Mediterranean?) and plains (desta; ibid.: 12, 54), are forming the
scenery of the heroes’ adventures. Interestingly the desert (e.g. Syrian, Iranian,
Gobi or Saudi-Arabian desert) does not play a main role. In this they mirror the
homeland and give shallow hints toward a former homeland, although this is
disputed.1
There is an influence of Iranian folktales (e.g. One thousand and one night)
and anecdotes. For instance, the following story reflects direct Iranian influence:
Keçel Ahmed “The Bald Ahmed” (Hayıg 2007: 21-25) is similar to Hasan Kachal
“Hasan the Bald” (well-known Iranian character). Other stories show some influ-
ence but either lost the coherence of the Iranian counterpart or changed its nar-
rative string in such a way that only little resemblance is left .
Not only is the physical world reflected in the traditions but also the spiritual
counter- and afterworld. As a “spiritual doublet” or “double-spiritual reality” the
two areas of being are brought into relation (Käser 2004: 110-111). The entire con-
cept reveals the religious perception of the world, mirrored in worldview. He-
roes, main players and actors transgress from one world into the other to experi-
ence thrilling adventures. Giants, demons, dragons, evil and good characters, su-
pernatural powers and normal people are mixed together in both worlds. This
scenery of both worlds leads to the assumption that a soul is part of both realities
and can move from the one to the other (ibid.: 65, 214, 319). As Käser mentions,
such worldviews are typical in societies worldwide, but contrast to Western
reading (ibid.: 55). One would be slow to overemphasize the differences when
European folktales are in focus, but looking at the whole set of “tradition”, such
an assumption can be made.
In stories, folktales and narratives the main players are representing society
and reflecting core values. In folktales main players can take on roles and repre-

1
The theories in question are: the Daylam province in Iran (Daylam-Thesis), the Mongol area
of the Zazak Mountains (Zazak-Theory) or the area of the Dunbeli-Clan in Northern Iraq around
Suleymaniye (Dunbeli-Thesis).
 Communication and the Oral-Aural Traditions of an East-Anatolian Ethnicity 677

sent characters that are exaggerated. This discursive style is a key feature of
folktales. It is meaningful to identify main players and their function:
• male heroes,
• dew (giant, demon Farsi: div, daeva),
• animals (see below),
• abandoned children, and
• crippled or handicapped persons.
Main players and actors are identified and named by their significant mark
and seldom by a proper name.2 This reflects the Zaza tradition that a name given
during lifetime is related to the outstanding achievements or failures of people.
The given birth name becomes insignificant. In the same way physical abnor-
malities (e.g. baldness, beardless, crippled, black) contribute to the designation
of the hero: ‘Heso Çı‘harçım “Four-eyed Hesen” (ibid.: 15-20), Qolo Poto “The Mu-
tilated” (ibid.: 50-53), Mêrdeko Tersano “The Scaredy Cat” (ibid.: 59-62) or a Siyak
“Black” (ibid.: 52-53).3 Other descriptions refer to the position or the effort one
has put into practice for his society: Seydwani “Hunter” (Berz 1996: 99-102) and in
Lazeko Zerez “The Son of the Partridge/Partridge Boy” (Hayıg 2007: 63-64), and
Lazê Axay “The Son of the Agha” and Axayo Axay niyo? “Is he an agha or not?”
(ibid.: 26-32 and 65; person is called agha), Şêx Biyayena ... “There was a Sheikh
...” (ibid.: 54-58), Lûwa Towbedari “The repentant Fox” (Malmisanıj 2000: 211-214),
Miftî Silîvanî û Şikir “The Mufti Silvan and Sugar/Sikir” (ibid.: 234).
In general, Mediterranean shame and collectivistic cultures address people
by their social rank or their social relation (e.g. age, relatives, education, and fa-
miliarity). In Zazaki, an older man is addressed as bıra “brother” (e.g. Düzgün
1992: 212, 324), keko “older brother”, apo “uncle” (Turkish loan), dedo “uncle”,
hoca “teacher”. An old man is called pile mı “my respected old”. A woman is ad-
dressed as wayê “sister” (ibid.: 63), xalê “aunt”, dayê “mother” (182; 184), pirê “(re-
spected) old woman”.4 Having said this, an exception is given by songs and popu-

2
E.g. Gorma’hmed “Hell of a chap Mahmed” (Hayıg 2007: 8-14), ‘Heso Çı’harçım “Four-Eyed
Hesen” (ibid.: 15-20), Lu u Ardwaniya “The Fox and the Miller” (ibid.: 33-36), Qolo Poto “The Muti-
lated (Girl)” (ibid.: 50-53), Lazê Zerez “Partridge Boy” (63-64), etc. Often in Zazaki, until today, a pe-
culiar characteristic or feature becomes the nickname of a person which is much stronger then the
name given by birth. An exception is ‘Elicanek the main actor of the first story, who is given a name
(ibid.: 3-7).
3
Gule (rose) refers to a woman emphasizing her beauty and not representing her proper noun
(Hayıg 2007: 56)
4
Please note that these are the vocative forms.
678 Eberhard Werner

lar people. In songs the names Ali (ibid.: 168, 202, 215, 236), Hesen (Hasan; 141,
146, 261), Hewa (Eve; 197, 221, 261, 287), Sılêman (Salomon; 124, 251) and Uşen
(106, 276, 320, 342) are very popular. However, an attribute is added, such as bıra
“brother”, beg “mister”, imam “teacher”, xıdır “holy”, sey “rel. leader /seyyid/
sheikh”. Calling a person by his position is common, such as axa (Agha), mıfti/
mıllah (Mufti/Millah), seyd (Seyyid), şêx (Sheikh) and pir (Pir). Persons titled like
that are identified as holders of unique social positions inside the Zaza commu-
nity. Women are not named by their given name and mainly introduced in rela-
tion to men. Thus a woman is identified as ceniya (xo) “(his) wife”, pirê “old
woman”, maya xo/mar “mother”, a wayê “sister”, a keynek “daughter” (Hayıg
2007). The social rank that a woman holds in a story tells us a lot about the cul-
tural expectations that the Zaza society looks for in a female member. The fol-
lowing rough assignations can be made about women:
• maya xo—the mother conciliates family members and tries to prevent her fam-
ily against bad luck,
• wayê—the sister is charming, helpful and supportive (ibid.: 26, 83)
• pirê—the old woman possesses wisdom and often takes care of her grandchil-
dren (ibid.: 21, 24, 45),
• ceniya cı—the wife is sometimes neutral, sometimes nagging (ibid.: 3, 25, 52)
• dêmarri—the stepmother is mad at her stepchildren (ibid.: 24, 71)
• keynek—the daughter is inexperienced, in need of help (ibid.: 25, 33, 38, 58, 71)
To sum this paragraph up one finds also cin/çin “jinn” (demons), the hoca
“religious leader”, the qıral, the padişa or the paşa “sultan, king” as secondary
characters in the traditions of the Zaza ethnos.

c. Fables and Characters as Peculiar Tradition

Looking at a third anthropological feature of the Zaza traditions, the focus is on
figurative animal fables. Fables are focusing on animal beings, human beings or
animal- and human-like beings. They transfer morals or ethics, by interactions of
non-humans among themselves or with humans or human-like beings. A resem-
blance to Western and Near Eastern fables becomes obvious (e.g. Germ.: Fuchs
und Rabe / Engl.: Fox and Crow), although the consequences, argumentation
and conclusions that are taken out of the fables are sometimes different to
Western thinking. As an outsider the emic stance that the ethnographer holds is
sometimes misleading. Because of that only the effects, pedagogically and so-
cially, are presented in this study (see 2.4).
 Communication and the Oral-Aural Traditions of an East-Anatolian Ethnicity 679

The lüye/lûya (fox; vixen) represents the clever and cunning animal. The fox
is always female in Zaza tradition. She is often in rivalry with the stronger şêr
(lion) or verg (wolf), demonstrating cleverness over physical superiority:
• Leads humans to make a fortune (Lu u ardwan—Fox and Miller in Hayıg 2007:
33-36; 3, 7).
• Interacts with other animals and tricks them (Hes ve lüera “Bear and vixen” in
Cengiz 2001: 49-50; Bıraênia heş u verg u lüe “The brothers bear, wolf and vixen”
in: 65-66; Lüe ve tığtığanira in ibid.: 143-146; Lûwa towbedari “The remorseful
Vixen” in Malmîsanij 2000: 211-214; Luw û sêr “Vixen and Lion” ibid.: 231), as well
as being tricked (Lüe ve heremuşira “Vixen and mole” in ibid.: 136-137; Lüa sare-
huske “Stupid Vixen” in ibid.: 114).
• Helps other animals by tricking others (Pıtwezir “Catboss” in Çimen/Neupert
2011). For instance in Pıtwezir the vixen supports a cat’s survival by tricking a
şuwanê (shepherd), a verg (wolf), a heş (bear) and a xoz (wild boar).
The conclusion of these fables is mainly positive and maintains the idea of
tricking others by all sorts of means. This is in contrast to Western thinking, in
which the fox is considered as untruthful and negative because tricking is con-
sidered to be bad behaviour.
The şêr (lion) stands for the king of the animals. He represents authority,
power and wisdom. In Near Eastern stories, as in today’s Western fables and fairy
tales, the lion symbolizes the wise and unquestionable ruler. The most common
western figure is represented in King Richard I. the Lionheart (reigned 1189-1199).
In heraldry and arts the “lion” holds an extraordinary stance as the symbol of
strength and invincibility. The Zaza story of Mahmeşa contains elements of the
Greek mythology, namely the story of the slave Androklus (1st century AD) who
removed a thorn from a wounded lion’s foot pad, and the same lion saved his
live later when they meet in the Circus Maximus. Mahmeşa, on the other hand,
was tricked by his mother. She asked him to get her lion’s milk, so that she could
stay and marry a dew (giant; Hayıg 2007: 37-39). Mahmeşa found a lioness with
three babies, which had been wounded. He shot an arrow into the wound which
opened and healed. As thanks the lioness gave Mahmeşa her little ones and her
milk (ibid.: 39). Later they rescued the hero. Cengiz (2001: 153-155) presents the
story Pısıng ve Şêrira “About the Cat and Lion” and Malmisanıj (2000: 231) offers
the story Luw û Şêr “Fox and Lion”. The latter parallels the West-European fable
of the Lion, the Fox and the Wolf. The fable tells about the tricky attempts of the
Wolf and the Fox to gain prestige and honour before the lion. But at the end
both have lost their reputation for trustfulness, whereas the tricky fox often
680 Eberhard Werner

demonstrated his intellectual superiority over the physically stronger wolf. In all
these stories the lion reflects the steady wisdom and honour of a responsible
character in authority. The conclusions taken by the Zaza people are partially in
coherence with Western thinking, although a stronger emphasis is given to the
authority of the lion as a model for a leader.
The heş “bear” mirrors a phlegmatic character. He is the one who represents
the steady and conservative preservation of morals, ethics and rules. Malmisanıj
(2000: 215-216) gathered a story from Mutki about Heş in which a son gabbles
with his mother. The main actors are the şêr “lion” and the heş “bear”. Cengiz
(2001: 27-30) offers the stories Hes ve mordemeki ra “About the Bear and the Man”
and Hes ve Luera “From the Bear and the Fox” (ibid.: 49-50). In contrast to West-
ern thinking the bear embodies a truthful brother- or sister-relation which some-
times is even not for the good of the bear. Interestingly the real bear is feared in
daily life and people know about accidents with bears (RH personal communica-
tion 2012).
The mor/mar “snake” symbolizes a mix of a wise and influential being as well
a fateful and tricky character. In peasant societies the snake is often considered
to be a demon or goddess. The Zaza tradition follows this perspective. Cengiz
(2001: 103) presents Lazek ve Morira “The Boy and the Snake”, Koremore “Blind-
worm” and Mordemek ve Morira “The Man and the Snake” (ibid.: 151, 163). In
other traditions the snake is identified with a house god or idol that protects and
influences the inhabitants called wayırê çê/wayırê key (Asatrian 1995; Comerd
1996). The snake is able to transform its appearance into a man-like being that
sometimes protects a well or a hidden place or sometimes fights humans (RH
2012 in personal communication). A white snake is thus more protective,
whereas the black snake is not trustworthy. However, in contrast to the western
concept of the snake being evil, the Iranian religions consider the snake more as
a god/goddess which can be evil or good.
Lastly the verg (wolf; delverg = she-wolf) represents a mix of tricky and often
devious character. However he/she is taken in by other characters, thus losing
face and prestige. Miye ve Vergira “The Sheep and the Wolf” (Cengiz 2001: 34),
Verg ve Bızera “The Wolf and the Goat” (ibid.: 43), Vergo Vêsa “The hungry Wolf”
(ibid.: 112) and Naşibê Vergi “A share of the Wolf” (ibid.: 120). The verg symbolizes
strength and authority, however the negative connotation that he/she holds is
due to his/her jealous attitude towards others. Since jealousy and envy play an
important part in driving societies (Malina 2001: 113-114, 119), the verg is empha-
 Communication and the Oral-Aural Traditions of an East-Anatolian Ethnicity 681

sized as the ideal character to demonstrate the consequences and dangers of

such an attitude. Many negative effects of envy, on the individual and the society
level, are passed on in the nature of the wolf. Nonetheless the verg is not per se
the bad guy, but his character is understood to be a warning, if one goes too far.
Other characters are the helpful miriçik/mıriçık “sparrow” (e.g. Hayıg 2007: 4-
5; Cengiz 2001: 23), the clever pısınge/pısinge “cat” (e.g. Cengiz 2001: 53, 57, 86,
155), the mysterious mose/masa “fish” (Hayıg 2007: 51-52; Cengiz 2001: 110), and
the stupid her “donkey” (ibid.: 139). The latter being the most important tool of
mobility in the past made him a favourite target for blameworthiness.

LINGUISTICAL AND TEXT-DISCOURSE FEATURES

The structures of the oral-aural traditions tell us something about the perfor-
mance and reception as well as the interaction of the narrator with the audience.
Going beyond that, written standards need to take into consideration the discur-
sive text structure of the oral-aural traditions. Text is always subject to (textual)
discourse (Genette 1983: 33). Every cultural and linguistic system performs its
peculiar discursive text features (Paltridge 2006: 11-12). We will first identify gen-
eral features of text discourse.
Story telling as an oral-aural tradition follows the principle of building so
called “non-event” information (“flesh”) around a so called “event” (“the skele-
ton”; Hopper 1979: 213). The event follows an event line, a tension and a climax,
the non-event contains the setting, flashbacks and an explanatory coda (Cran-
dall 2002: 31; Werner 2012: 50-51). Thus, the skeleton is enriched by flesh on an
informational and an entertaining level.
Zaza traditions are told in a historical present (Hayıg 2007: XIII). The audience
is thus included in the course of the story. The historical present is chosen to pre-
sent progressively the actual time of speech. Individual stories that have taken
place in past realities are told in preterite (Werner 2012: 57). In translation (e.g.
German and English) the historical present is replaced by the past, which corre-
sponds to the historical present. The rising literacy of Zazaki shows a tendency
to replace the historical present with preterit. This traces back to the influence of
Western literature, which is increasingly translated. These Zaza translations are
not paying attention to the traditional way of passing forward traditions.
Typically, oral-aural traditions are marked by introductory phrases like cake
beno, cake nêbeno /a place was, wasn’t it/ “once upon a time” or wextê dı /time at/
“at a certain time”. The starting point of an event line is marked by rozê /the day/
682 Eberhard Werner

“one day” (e.g. Hayıg 2007: 3, 26, 33, 54). In the Northern dialect similar construc-
tions are used as starting phrases, often the shortened saying is beno, nêbeno /it
is, it is not/ “There was, wasn’t it” (Çelker cit. in Werner 2012: 52).
Oral-aural tradition in narration is sometimes closed up by the saying
Istanıka mına weş, 'hewt koya pey dı bi ze leş. /My story is nice, seven mountain
ranges behind it is like a carcass./ “Here is my charming tale: It has turned into a
carcass behind seven mountain ranges”, meaning “And they all lived happily
ever after” (e.g. Hayıg 2007: 14, 141). Another ending phrase is nê resenê mırazdê
xo. /they reached their wishes./ “all their wishes were fulfilled.” or better “And
they all lived happily ever after.” (ibid.: 25, 150). The final stage of a narrative is
introduced by hıni “finally; now” or wıni “in that manner” (ibid.: 32).
Concerning marked word order, any change from the general SOV-or-
der/SVGoal5 or SOVGoal/S-IO-O-V6, disturbs the natural flow of information
(natural information flow; Roberts 1997; idem 2003: 3, 8-13) and attracts the focus
of the audience. Change of word order is thus a familiar way of emphasizing and
highlighting (see also below). In the story Axayo, Axay niyo? (ibid.: 65) a couple is
visited by three men with a saz (sort of long-necked lute), a typical instrument
for a wandering bard or musician (pir; aşık) or a follower of a pir or agha. The
couple expects one to be an agha, but does not know which one. They offer them
a calf but before serving it, the husband addresses them one by one and asks
them if he is an agha. All three affirm it, so the husband understands that he is
tricked, and the visitors are just musicians (often Kurmanji speakers). They serve
them only the bones because every one of the musicians accuses the others of
being kutık (dogs). When each musician is asked one by one if he is an agha, he
answers axa ezo. /Agha I am./ “Sure I am the agha.” or Ya ezo. /Yes I am./ “Sure I
am the one.” Here, the change of order by using Scomp-S, signals a punch line for
the hearer. The same goes for the phrase ko goşt? /Meat where/ “Where is the
meat.”, instead of the expected phrase Goşt kotiyo? /Meat where-is?/ “Where is
the meat?”. Obviously, in these examples, a change of focus is used to get the au-
dience’s attention.

5
E.g. SOV = Zew Erebê esto. /One Arab is./ “There was an Arab.” (ibid.: 12; Werner/Werner 2007:
7); SVGoal = Nê şınê vera Erzurum. /They go in direction Erzurum./ “They went to Erzurum.” (Hayıg
2007: 66).
6
E.g. SOVGoal = Nê şekerê xo dekenê qedıxa miyan. /They sugar fill cups into./ “They put the
sugar into their cups.” (ibid.: 66); S-IO-O-V = … wıhêrê keyi ninarê çay ano, … /... the host them-to
tea brings…/ “the host brings them tea …” (ibid.: 66; Werner/Werner 2007: 8).
 Communication and the Oral-Aural Traditions of an East-Anatolian Ethnicity 683

During the course of a narrative text, discourse markers are signalling the
flow of information (information flow). Temporal, textual and emphatic points of
departure interrupt the progress and/or change the focus. Zazaki has a rich vari-
ety of markers functioning as point of departures. They facilitate the main un-
marked event line. The following are the major ones used in the book Mahmeşa:
ew “and, then”, na wuza dı; wuza dı “there; then”, nofın “then, thereafter”, zi “also,
then”, wexta “then”, bahde “then; after”.
A marker can also be used to signal change in the temporal course. Werner
(2012: 52) speaks about marked and unmarked forms to signal the event line.
This is necessary because in Zaza folktales, fables and anecdotes, words relating
to time are not used to indicate change of time but Zazaki continues in the his-
torical present. In personal narrative, the temporal aspect is used to distinguish
time events.
Out of this develops the discursive text setting of fore- and background struc-
tures. A good story teller in Zazaki switches from the main event line to fore-
ground events and throws in additional background information now and then
to keep the tension. Background facts are posted into the information flow by
markers such as: hıma “but, so” or çıkı “because”. They disrupt the event line and
the narrative is not carried forward. Information of this sort is given on the non-
event line (Crandall 2002: 34). With hıma and çıkı information which is necessary
to understand the ongoing main story is also passed to the audience. Background
details in this sense are fulfilling two functions. On the one hand it supports the
fluency of the story; on the other it feeds additional information to guarantee
successful communication (see Relevance Theory; 1). In the oral-aural tradition
the narrator uses the tool of background information to explain a problem or de-
scribe circumstances to keep the audience on track. Another tool of background
information is presented by flashbacks. This is an event line before the main
event line of the narrative, and is marked by adverbial phrases like wexto kı /time
that/ “at the time when” or oxmo kı /instance that/ “in the meantime” (Hayıg
2007: 55, 12; Werner 2012: 52-53).
Besides these linguistical and text-discursive features, Zazaki offers other
topics of text discourse, which are not discussed here. The following small list
represents those areas (Werner/Werner 2007):
• left-dislocated constituents (Hayıg 2007: 66),
• specificity (Lambrecht 1998: 81; Todd 2002: 38),
• repetition (‘Heso Çı’harçım; Hayıg 2007: 18),
684 Eberhard Werner

• the activation and re-activation of participants (see above) and

• the topic-comment articulation7. In Zazaki topic always precedes the comment.
Thus, it is in the range of language universals claimed by discourse analysts
(Dooley/Levinsohn 2000: 32).
Lambrecht (1998: 165) shaped the term “pragmatic accessibility” of topic ref-
erents in text discourse. He draws a line from the most to the least acceptable
topic referent, which describes the information structure of participants.
We will now take a look at some pedagogical effects, generated by oral-aural
traditions in the Zaza community.

PEDAGOGICAL AND SOCIAL EFFECTS, WORLDVIEW, MORALS AND ETHICS

One outstanding and recent topic for the people group in focus here is how to
continue their traditions in times of culture change. Assimilation leads to a loss
of the traditional tribal, religious, and social structures. Orality no longer works,
because this development, over the last thirty years, hurts the whole ethnos es-
sentially. The focus and function of orality and traditions in the Zaza community
has changed over time. In the past the oral-aural tradition worked to guarantee a
stable and functioning social life. The hierarchical system, the religious setting
and the linguistic analogy are reflected by the story-teller in his narrative. Simul-
taneously, the position of the narrator is strengthened, be it a pir, sheikh or an-
other professional. Nowadays traditions fulfil mainly a political function and of-
fer an effect of preservation. They are told to be recognized by following genera-
tions and are written down to demonstrate the richness and variation of the his-
tory of the Zaza people.
The Zaza narratives and songs are not meant as direct moralizers. The audi-
ence is asked to draw conclusions from the events for their daily life. In conse-
quence morals and ethics are an outflow of inferences drawn from these tradi-
tions. But not only the saying, also the way how, from whom, and in what cir-
cumstances it is said are important. The context that is the setting of the presen-
tation as well as the process of derivation is flexible. This flexibility is counter-
acted by a central and straightforward main event-line which represents the core
morals and ethics alive.

7
According to Dik (1978: 130), topic is the entity that the “utterance is primarily about”. Others
call this the Theme-Rheme Structure [German: Thema-Rhema-Gliederung]. Topic stands for the
known or available information, whereas comment stands for the unknown or new information
about the theme.
 Communication and the Oral-Aural Traditions of an East-Anatolian Ethnicity 685

The story Lazê Axay (the Aghas Son; Hayıg 2007: 26-32) shows some parallels
to the story The Prodigal Son in the New Testament (Lukas 15: 12-24; see 3.1). This
similarity provides insight in the reception of traditions within societies. Both
stories tell about a moral misbehaviour—extreme disobedience against the par-
ents—which in the case of the Prodigal Son is sanctioned by the Mosaic Law to
be punished by stoning (Deut 21: 18-21). It seems that the core family is given a
strong protection to guarantee cohesion and loyalty within this central social
unit. In the story Lazê Axay the main actor also risks his prestige and honour by
leaving his family. During his time abroad, in due course, he makes his fortune.
In comparison, the New Testament text was told to demonstrate the all forgiving
love of a father as a model for the absolute love of God. The whole setting of the
parable forced the audience to draw that conclusion. The repentance of the
prodigal son was secondary although important. The Zaza narrative, instead,
demonstrates the high value of family relationships and natural desire to return
home. One is rooted in his home region and even wanderlust is stilled at some
moment in life. The picture is completed, by
• a high value for hospitality,
• a strong social cohesion of tribal and microculture structures,
• a collectivistic orientation and
• effective sanction mechanisms (e.g. blood feud, honour killings, women abduc-
tion, girl exchange; Werner 2011).
The whole package of such social core values is transferred in tradition. It is
within the reflection of the audience, by adhering to the actors of such tradi-
tions, that a picture of the world and mankind is formed. At the same time the
worldview is presented to the audience as a necessary prerequisite for the tradi-
tions. This mutual interplay requires a story teller who needs a good reputation
and life experience. Thus he will be accepted and given influence.
In the past the cem/cemaat meetings were also legal gatherings, so it is im-
portant that misbehaviour, crime and offence were ruled directly in combination
with the passing on of morals and ethics in the traditions. As such, sanctions
have been carried out that guaranteed a stable social structure. The same hap-
pened in the family as an institution. The oral-aural tradition of the Zaza reflects
such stability with the counter- or afterworld. Mirroring the real world, the he-
roes trespass the borders of both worlds and deal with giants, dragons and good
and evil characters (supporters and destroyers), but always come back into the
real world. Beside a concept of a so called “double-spirit soul” and “spiritual dou-
686 Eberhard Werner

blet” perception of the world, this demonstrates that the traditions of the Zaza
include a non-physical reality in their worldview. This spiritual world is as realis-
tic and therefore not idealized as the physical reality. It is one outcome of the
oral-aural traditions that such religious awareness is kept alive. It effects the per-
ception of nature as a soul inhabiting essence.

SUMMARY
In this article the oral-aural traditions of the Zaza people came under investiga-
tion. Although the transition to a culture of writing is not finished yet, because
the older generation is not literate, it is time to research the former ways of
passing on worldview. Besides educational and environmental learning, the art
of tradition-submission plays an essential role for the Zaza ethnicity. Two prin-
ciples of communication are thereby foundational: The reflection of tradition by
an audience (the aural part) and the submission of tradition by a speaker (the
orality part). Both are main concepts of cultural survival. It opens the ground to
include new concepts and at the same time to protect the society from losing
strong core values. Both social processes are necessary to allow culture change
and at the same time to avert the danger of foreign concept intrusion. Healthy
and resistant ethnocentrism, preserved in worldview and conscience, is per-
formed in such a development.
General observations can be made about the Zaza traditions. The oral-aural
transmission of traditions happened in the core and extended family. A second-
ary social institution was the cem or cemaat, a social gathering which happened
only a few times a year and included a whole village or a local audience. A story-
teller with prestige and authority (e.g. musicians, older people, seyyid, pir, şeik)
was given the task of passing on traditions such as songs, narratives, stories, an-
ecdotes, parables, fables or folktales. The cem was an important social institution
which got lost during the last twenty years.
Traditions are only a part of the full process of enculturation. Besides educa-
tional and real life environmental learning they contribute to the perception and
development of a worldview and the conscience, both in the individual as well
as by a whole people group. This is mainly done by conscience, outlined in the
location of the emotions, the intellect and the character (LEIC; Werner 2011).
Story-telling is a central feature of the oral-aural tradition; it brings two-dimen-
sional written stories to three-dimensional life. The concept thereby is dynam-
ical and not static.
 Communication and the Oral-Aural Traditions of an East-Anatolian Ethnicity 687

The anthropological focus on oral-aural tradition reveals morals and ethics.

Their setting refers to daily life situations, adventures, supernatural develop-
ments or extraordinary states of affair. The main topics are desires or longings
for the welat “homeland”, towards the preservation of stable social units (core
and extended family, tribe), eşk “worldly love” and the name “one’s reputation as
honour and prestige”. The geographical setting of the traditions reflect the
homeland environment especially the huge mountains and rivers. Also the non-
physical realities and perceptions such as “spiritual double” of soul and after-
world are of interest. The Zaza traditions demonstrate a close relation between
the parallel concepts of realities in the physical world and the spiritual after-
world. Although in the narratives only heroes, main players or supernatural be-
ings cross the border back and forth, it becomes obvious that the actors are par-
ticipating in both realities.
Fables are often used as traditions. The characters are either animals, human
beings, or animal like human beings. They own human qualities and represent
tricky (lü “fox/vixen”), caring (şêr “lion”), devoted (mırıçık “sparrow”), sorrowful
(her “donkey”), clever (luyê “fox/vixen”), vicious or sneaky (verg “wolf”) charac-
ters. The pedagogical and social effects of fables are not similar to Western socie-
ties, although some fables are very close. However, the punch line is sometimes
quite the opposite of Western expectations. Zaza fables are taken more literally
and at the same time the reflection of a character is transferred to a secondary
implementation. For instance the story of Qolo Poto “The Mutilated (Girl)” is un-
derstood as an example of the suffering way of life and restoration. The main
player is the fish who offers wishes for free and changes fate. In case of Lazê Axay
“The son of the Agha” the hero is understood to help others. Morals are not
stated explicitly therefore the effect of Zaza tradition on the Zaza audience is not
predictable for outsiders.
Linguistical and text-discursive features of the oral-aural tradition are gram-
matical and stylistic peculiarities. The Zaza folktales follow a strict historical pre-
sent, whereas individual narrative (personal stories) is told in preterite. The
structure contains an introduction, a climax and a coda. The main actors are
named after specific features, marks or life achievements, but seldom given a
birth name. Women are never given a name but always described in relation to a
male, e.g. wa “sister”, xal “ant, dayê “mother”, pir “old woman”. This custom is
also used for males, but then a specific attribute specifies them. This custom
mirrors the perception of a patriarchal society. The unmarked main event-line is
688 Eberhard Werner

only disturbed by markers that are used to emphasize some features and to get
the attention of the audience. A good story-teller plays with such markers and
throws in flashbacks and background information to provide successful com-
munication. The narrator keeps the audience in the loop about the context of
the narrative (ostensive-inferential principle). The story-teller activates the au-
dience’s cultural-linguistic mutually shared and encyclopaedic knowledge store
by discursive tools.
The pedagogical and social effects that the oral-aural traditions arise in the
Zaza community, the reception and the cognitive processes to interfere differ
from Western expectation, experience and perception. Zaza traditions are firstly
trying to bring about social cohesion. By describing daily life situations figura-
tively, the audience is asked to draw out conclusions about their core values,
specifically in morals and ethics. Traditions have a twofold function: firstly they
express a value in themselves by demonstrating and reflecting how the society is
constituted in form of submission (e.g. values, hierarchy, social linking etc.; oral
part); secondly they transport the worldview, the perception of men and nature,
to the audience and generate reflection (aural part) by the individual and the
whole group. Processes like this, allow for culture change within an acceptable
range and at the same time build up ethnocentrism and self awareness. These
powers of social resistance are important to allow for a literal approach and po-
litical activity on a legal base, such as the proclamation of mother-tongue educa-
tion and official recognition as an ethnicity with an own language. Recently the
challenge is to come up with a standardized orthography in a reference dialect
which would work as a language of public education.
I hoped to stimulate research on oral-aural traditions in the area or East Ana-
tolia. This area of a rich history changes rapidly. Assimilation processes, migra-
tion movements, and loss of viable social structure lead to culture change in
such dimensions that language endangerment and merging into larger social
units is the consequence.
“History, as is often said, is the key to the future.”
 Communication and the Oral-Aural Traditions of an East-Anatolian Ethnicity 689

BIBLIOGRAPHY
Asatrian, G. S. (1995), “Dımlī”, Encyclopedia Iranica.
―― (1998), “Dımıli”, Ware 12: 182-187. Baiersbronn.
―― (2002), “The Lord of Cattle”, Iran and the Caucasus 6.1-2, 1-2, Brill-Leiden.
―― (2007), “The Foremother of the Yezīdīs”, Religious Texts in Iranian Languages, eds. F. Vahman;
C. V. Pedersen, Copenhagen: 323-328.
―― (2009), “Prolegomena to the Study of the Kurds”, Iran and the Caucasus 13-1: 1-58.
―― / N. Gevorgian (1988), “Zaza Miscellany: Notes on some religious customs and Institutions”,
Acta Iranica, Encyclopédie Permanente des Études Iraniennes. Publiée par le Centre Inter-
national d'Études Indo-Iraniennes. Hommages et Opera Minora. Volume XII. A Green
Leaf. Papers in Honour of Professor Jes P. Asmussen, Leiden: 500-508.
Aslan, M. (2003), Kilıtê Kou—The Key of the Mountains—Dağların Anahtarı, İstanbul.
Astare, K. (2011), Worte in der Blumensprache—Zonê gulu de qesei: Kırmoncki (Zaza)—Deutsch,
Bonn.
Barihas, Ş. (1999), Şervan 3: Alemano, Hildesheim.
―― (2003), Ra u Rêçe, Ethno Müzik Yapım.
Berz, K. (1993), Sîyamed û Xeca, Stockholm-Spanga.
―― (1995), Kole Nêba: Şeher ê Dimlî, Stockholm.
―― (1996), Ewro Şorî Meşti Bêrî, Stockholm.
Bläsing, U. (1995), “Kurdische und Zaza-Elemente im türkeitürkischen Dialektlexicon, Etymolo-
gische Betrachtungen ausgehend vom Nordwestiranischen”, Near Eastern Languages and
Literatures (NELL) 1/2, 173–218, Leiden.
Çapan, M. (1999), Lauke Dersım, Kornwestheim.
Cengiz, A. (2011), Morıber, İstanbul.
Cengiz, D. (2001), Dersim Fablları-I, Ankara.
Çimen, H. (2002), Große Narren. Erzählungen, Krefeld [Translation from H. Mergarıj (1996), Bude-
laê Gırşi, İstanbul].
―― / S. Neupert (2011), Pıtwezir—Zu sanıka Zazau.
Comerd, M. (1996), “Yitiqatê Dersımi de Wayırê Çêi (Asparê astorê kimenti): Sarê Çêi xıravunu ra
'be neweşiye ra sevekneno, karê dine raşt beno rızqê çêi dano, malê çêi 'be qısmetê çêi
keno jêde, nasivê çêi sevekneno”, [Turkish S. 69-74: Dersim İnancı'nda Ev ve Aile Tanrısı
(Wayırê Çêi): Ev halkını her tür kötülükten, kötü melklerden ve hastalıklardan koruyor,
İşlerinin İye gitmesini sağlıyor, evin rızkını veriyor, malını ve kısmetini artırıyor, nasibini
koruyor.], Ware 9: 10-14, Baiersbronn.
Crandall, M. (2002), Discourse Structures in Zazaki Narrative, Unpublished Master Thesis, Mainz.
Dewran, H. (o. J.), Hazar Reng—Hazar Veng: Lyrik & Musik in der Zaza Sprache, Frankfurt.
Dik, S. (1978), Functional Grammar, Amsterdam.
Dooley, R. A. / S. H. Levinsohn (2000), Analyzing Discourse: A Manual of basic concepts, Dallas.
Düzgün, M. (1992), Dersim Türküleri—Tayê Lawıkê Dersımi—Şiwari Ağıtlar, Ankara.
Genette, G. (1983), Narrative Discourse: An Essay in Method, New York.
690 Eberhard Werner

Hayıg, R. (2007) Mahmeşa: Vızêr ra Ewro Istanıkê Zazayan—Mahmescha: Zaza Volksmärchen von
Damals und Heute—Mahmesha: Zaza Folktales: Then and Now, İstanbul.
Hopper, P. (1979), Aspect and Foregrounding in Discourse, ed. Talmy Givón, Syntax and Semantics,
Discourse and Syntax 12: 213-241.
Käser, L. (2004), Animismus: Eine Einführung in die begrifflichen Grundlagen des Welt- und Men-
schenbildes traditionaler (ethnischer) Gesellschaften für Entwicklungshelfer und kirchliche
Mitarbeiter in Übersee, Bad Liebenzell.
Keskin, M. (2008), Zur dialektalen Gliederung des Zazaki, Masterthesis, Frankfurt.
Kohl, K.-H. (1993), Ethnologie—die Wissenschaft vom kulturell Fremden: Eine Einführung, München.
Lambrecht, K. (1998), Information structure and sentence form: Topic, focus, and the mental repre-
sentations of discourse referents, Cambridge.
Malina, B. J. (2001), The New Testament World—Insights from Cultural Anthropology, Louisville.
Malmîsanij, (2000) Folklorê ma ra çend numûney, İstanbul.
Maxey, J. A. (2009), From Orality to Orality: a New Paradigm for Contextual Translation of the Bible,
Eugene.
Mergarıji, H. (1996), Budelaê Gırşi, İstanbul.
Muller, R. (2001), Honor and Shame: Unlocking the Door, Philadelphia.
Müller, K. W. (2009), Das Gewissen in Kultur und Religion: Scham- und Schuldorientierung als em-
pirische Phänomen des Über-Ich/Ich-Ideal, Nürnberg.
Neyzi, L. (2002), Embodied Elders: Space and Subjectivity in the Music of Metin-Kemal Kahraman,
Middle Eastern Studies 83/1, Jan.: 89-109.
Ong, W. J. (2002), Orality and Literacy: The Technologizing of the Word. 4th ed., London.
Paltridge, B. (2006), Discourse Analysis: An Introduction, New York.
Paul, L. (1998), Zazaki: Grammatik und Versuch einer Dialektologie, Wiesbaden.
―― (2009), “Zazaki”, ed. G. Windfuhr, The Iranian Languages, New York: 545-586.
―― (2012), Die Herkunft und Stellung des Zazaki und das Verhältnis von Sprache zu Ethnie, eds.
M. Varol; Ö. F. Elaltuntaş, 1. Bingöl Symposium on the Zaza Language (13th-14th May 2011),
Bingöl: 13-18.
Roberts, J. R. (1997), The syntax of discourse structure. Notes On Translation 11/2, 15-34, Dallas.
―― (2003), Features of Persian Discourse Structure, Work Paper. Horsley Green, forthcoming.
Selcan, Z. (1998a), Die Entwicklung der Zaza-Sprache, Ware 12, Baiersbronn: 152-163.
―― (1998b), Grammatik der Zaza-Sprache—Nord-Dialekt (Dersim-Dialekt), Berlin.
Shannon, C. L. / W. Weaver (1949), The Mathematical Theory of Communication, Urbana.
Sökefeld, M. (2008), “Einleitung: Aleviten in Deutschland—von takiye zur alevitischen Bewegung”,
ed. Sökefeld, M., Aleviten in Deutschland: Identitätsprozesse einer Religionsgemeinschaft in
der Diaspora, Bielefeld: 7-36.
Sperber, D. / D. Wilson (1995), Relevance: Communication and Cognition, Oxford.
Söylemez, I. (2012), “Geçmişten Günümüze Zazaca Dergiler: Kronoloji, Sorunlar ve Çözümler”, eds.
M. Varol; Ö. F. Elaltuntaş. 1. Bingöl Symposium on the Zaza Language (13th-14th May 2011),
Bingöl: 175-191.
 Communication and the Oral-Aural Traditions of an East-Anatolian Ethnicity 691

Taş, C. (2007), Külden Evler, İstanbul.

Todd, T. L. (2002), A Grammar of Dimili (also known as Zaza), Stockholm.
―― / E. Werner (2012), “The Sacredness of Minority Language and Mother Tongue Based Multi
Lingual Education”, eds. M. Varol; Ö. F. Elaltuntaş. 1. Bingöl Symposium on the Zaza Lan-
guage (13th-14th May 2011), Bingöl: 89-99.
Werner, E. (2011), “The Struggle for Identity: Cognitive Anthropology: Aspects of the Zaza
Worldview”, Paper presented at the 15th Anniversary of the publication Iran and the Cauca-
sus, Leiden (forthcoming).
Werner, B. (2012), “Coding of Background Information in Zazaki Narrative”, eds. M. Varol; Ö. F.
Elaltuntaş. 1. Bingöl Symposium on the Zaza Language (13th-14th May 2011), Bingöl: 49-68.
―― / E. Werner (2007), Outline of Narrative Discourse Features in Dimli/Southern Zazaki,
Holzhausen (forthcoming).
Zelê Melê (1999), Şêlê, İstanbul.
 SELECT BIBLIOGRAPHY OF PROF. GARNIK S. ASATRIAN

1982
1. “Hayerenum iranakan mi k‘ani poxaṙutyunneri šurǰ”, Patma-banasirakan hands
(PBH), N2, 1982: 191-193.
1983
2. “Perfekt perexodnyx glagolov v srednepersidskom i parfyanskom”, K III mežduna-
rodnomu simpoziumu po teoretičeskim problemam yazykov Azii i Afriki (Tezisy dok-
ladov sovetskix yazykovedov), Moskva, 1983: 7-10.
3. “Obryady detstva i vospitane detej v tradicionnoj kul’ture persov”, Étnografiya
detstva. Tradicionnye formy vospitaniya detej i podrostkov u narodov Perednej i
Yužnoj Azii, Moskva, 1983: 68-88.
4. “Pričastiya na -ān v zapadnosredneiranskix yazykax (na materiale turfanskix tek-
stov)”, Vestnik obščestvennyx nauk Armenii, Erevan, N5, 1983: 53-56.
5. “Infinitivy v srednepersidskom I parfyanskom”, PBH, N1, 1983: 161-176.
1984
6. “K voprosu ob armyanskix leksičeskix zaimstvovaniyax v kurdskom”, III vsesoyuz-
naya škola molodyx vostokovedov, Tezisy, vol. II/2, Moskva, 1984: 124-126.
7. [Trans. to Armenian:] K. H. Schmidt, “Die Indogermanische Grundlagen des altar-
menischen Verbums”, PBH, N1, 1984: 17-37.
1985
8. “Grabari -ak anvanakan verǰacanc‘ə”, PBH, N3: 136-150. (with G. Muradian)
9. “O ‘brate i sestre zagrobnoj žizni’ v religioznyx verovaniyax ezidov”, Strany i naro-
dy Bližnego i Srednego Vostoka XIII, Erevan: 262-271.
1986
10. “Krder”, Haykakan Sovetakan Hanragitaran, vol. 12, Erevan: 491-492.
11. “O rannix armenizmax v kurdskom”, PBH, N2: 168-175.
1987
12. Materials on the Ethnography of the Baxtiaris. West Iranian Dialect Materials from
the Collection of D. L. Lorimer, vol. 1, Copenhagen, 1987, 160 pp. (with F. Vahman)
13. “Armyanskie zaimstvovaniya v kurdskom”, Vtoroj meždunarodnyj simpozium po
armyanskomu yazykoznaniyu. Tezisy dokladov, Erevan: 35-36.

 Koninklijke Brill NV, Leiden, 2015

694 Prof. Garnik S. Asatrian

14. “K voprosu o klassifikacii kurdskix dialketov”, Tezisy dokladov konferencii molodyx

naučnyx sotrudnikov na temu “Aktual’nye problemy filologii i istorii stran Bližnego i
Srednego Vostoka”, Tbilisi.
15. “Otglagol’nye imena na -aka- v srednepersidskom i parfyanskom”, Voprosy vosto-
kovedeniya 3-4, Erevan: 306-320.
16. “Yazyk zaza i armyanskij (Predvaritel’nye zametki)”, PBH, N1: 159-171.
17. [Trans. to Armenian of:] J. P. Asmussen, “Kamel und Nadelohr”, PBH, N4: 139-144.
1988
18. “Zaza Miscellany: Notes on some Religious Customs and Institutions”, A Green
Leaf: Papers in Honour of Prof. J. P. Asmussen, Leiden: 499-509. (with N. Gevor-
gian)
19. “Arevmtyan nor ašxatank‘ner nvirvac hayagitut‘yan ev dasakan iranagitut‘yan
xndirnerin”, HH GAA Lraber Hasarakakan Gitut‘yunneri (LHG), N10: 84-87.
20. [Review of:] A. Poladian, Kurdy v VII-X vv. po arabskim istočnikam, Erevan, LHG 1:
87-89.
21. “Suffiksal’nyj élement -l- i nekotorye voprosy fonosemantiki v armyanskom yazy-
ke”, IFŽ AN, N2: 160-178.
22. “Nekotorye materialy po étiketu u persov”, Étiket u narodov Perednej Azii, Moskva:
148-163.
1989
23. Otglagol’nye imena v srednepersidskom i parfyanskom (na materiale turfanskix
teksktov), Erevan, 128 pp.
24. “Zametki ob étimologičeskix figurax v iranskom (na materiale sogdijskoj versii
Vessantara džataki)”, PBH, N3: 152-163.
25. “Yezdineri davanank‘ə (himnakan astvacut‘yunner, surb grk‘erə)”, PBH, N4: 131-
150. (with A. Poladian)
26. “Suffiks množestvennogo čisla -gal v iranskix yazykax”, Strany i narody Bližnego i
Srednego Vostoka, vol. XV, Erevan: 300-304.
1990
27. Azerbajdžan: Princip prisvoeniya i iranskij mir, Erevan, 23 pp. (with N. Gevorgian)
28. “Gleanings from Zaza Vocabulary”, Iranica Varia: Papers in Honour of Prof. E. Yar-
shater, Leiden: 267-275. (with F. Vahman)
29. “Kuzey Iran Dilleri”, Piya 10, Stockholm, 1990: 14-15.
30. “Hay siyasal Dusunce sayfalarından”, Piya 13, Stockholm: 11-16.
31. “Kan ardyok‘ haykakan poxaṙut‘yunner parskerenum?”, PBH N5: 139-144.
32. “Ešče raz o meste zaza v sisteme iranskix yazykov (zametki po novoiranskoj dia-
lektologii)”, PBH, N4: 154-160.
 Select Bibliography 695

33. [Review of:] V. Bayburdian, Hay-k‘rdakan poxharaberut‘yunnerə Osmanyan kay-

srut‘yunum, Yerevan, 1989, HH GAA Lraber Hasarakakan Gitut‘yunneri N6: 89-92.
1991
34. “Zametki ob azari, isčeznuvšem yazyke Azerbajdžana”, K osveščeniyu problem is-
torii i kul’tury Kavkazskoj Albanii i vostočnyx provincij Armenii, Erevan: 484-492.
35. Short Stories of the Baxtiaris. West Iranian Dialect Materials from the Collection of
D. L. Lorimer, Copenhagen. (with F. Vahman)
1992
36. Kontakty armyanskogo s novoiranskimi yazykami, Erevan, 400 pp.
37. “Nekotorye voprosy tradicionnogo mirovozzreniya zaza”, Tradicionnoe mirovoz-
zrenie u narodov Perednej Azii, Moskva: 102-110, 210-212.
38. “Zazaların Ulusal Dünya Görüşü”, Raştiye, N5, Stockholm: 6-9.
39. “Zaza Dilinin İran Dilleri Sistemindeki Yeri”, Raştiye, N7, Stockholm: 12-18.
1993
40. “Sak‘ žołovurdə (gitakan t‘argmanut‘yun anglerenic ev aṙaǰabanə)”, Iran-Namē,
vol. 1: 23-24.
41. [Trans. with Introduction and Additions of:] Énajatolla Reza, Azerbajdžan i Arran,
Erevan, 130 pp.
42. “Miǰinasiakan ev Kovkasyan usumnasirut‘yunner (gitakan hratarakut‘yunneri te-
sut‘yun)”, Iran-Namē, vol. 1: 21.
43. “Serob Pašan k‘rdakan banahyusut‘yan meǰ”, Iran-Namē, N3: 20.
44. “K‘rdakan mamuli ēǰerum”, Iran-Namē, N4: 11.
45. “Tališner”, Iran-Namē, N4: 17-19.
46. [Trans. with Introduction and Additions of:] R. Grishman, “La civilisation Ache-
menide et l’Urartu”, Iran-Namē, N5: 16-17.
47. “Zāzāhā qoume nāšenāxte”, Maǰalleye tahqīiqāte tārīxī, N8, Tehran: 97-111.
1994
48. “Hayastani Gitut‘yunneri Azgayin Akademiayi 50 amyakə (Mtorumner Hayasta-
num gitakan vorakavorman harci šurǰ)”, Iran-Namē, N8: 1.
49. “From the Editor”, Acta Kurdica, London: 1-4.
50. “Origine du système consonantique de la langue Kurde”, Acta Kurdica, vol. 1: 81-
108. (with V. Livshits)
51. “Reviews”, Acta Kurdica, vol. 1: 222-225, 230-231.
1995
52. Poetry of the Baxtiaris, Copenhagen, 1995, 216 pp. (with F. Vahman)
53. “Guranner”, Iran-Namē, N11: 21-30.
54. “Azariner”, Iran-Namē, N12: 20-24.
696 Prof. Garnik S. Asatrian

55. “Asorinerə ev T‘urk‘ian”, Iran-Namē, N12: 27-29.

56. “Azgagrakan ētyud”, Iran-Namē, N12: 30.
57. “Zaza: People and Language”, Encyclopaedia Iranica, vol. VII/4, New York: 405-411.
58. “Nikołayos Adonc‘ə ev k‘rderi Hayastan nert‘apancman patmut‘yunə”, Iran-Namē,
N13: 3-7.
59. “Zradaštakanut‘yan Havato hanganakə”, Iran-Namē, N13: 16-17.
60. “Alek‘sandr Leonovič Gryunberg (Xosk‘ hišataki)”, Iran-Namē, N13: 23.
61. [Introduction:] V. P‘ap‘azeanc‘, “Hay bošaner”, Iran-Namē, N13: 30-31.
62. [Introduction:] V. P‘ap‘azean, “Hay yełapoxut‘yunə ev harevan žołovurdnerə”,
Iran-Namē, N14-15: 9.
63. “Divakank‘. Miscellanea Daemonologica”, Iran-Namē, N14-15: 21-35.
64. “Yerevani masin hnaguyn hišatakut‘yunə sołdakan mi jeṙagrum. K‘ristoneakan
hamaynk‘nerə naxak‘ristoneakan Hayastanum”, Iran-Namē, N14-15: 37.
1996
65. [Introduction:] Atrpet, “Ali-allahiner”, Iran-Namē, N16-17: 7, 10.
66. “Akademikos Mkrtič Nersisyan (Cnndyan 85 amyaki kapakc‘ut‘yamb)”, Iran-Na-
mē, N16-17: 13.
67. “Iranakan žamanakakic‘ arvest: Šarare Lorestani”, Iran-Namē, N16-17: 13.
68. “Zazaneri avandakan ašxarhayac‘k‘ə”, Iran-Namē, N16-17: 38-40.
69. “Reviews”, Iran-Namē, N16-17: 19, 21, 31, 41.
70. [Intoduction:] N. Mkrtčyan, Baṙeri kensagrut‘yunic‘, Erevan: 1-5.
71. “Hayagitut‘yunə porjut‘yunneri šemin”, Iran-Namē, N18-19: 1-2.
72. “K‘rdagitut‘yan himnadruyt‘ner (K‘rdagitut‘yan patmut‘yan meknman porj)”, Iran-
Namē, N18-19: 3-6.
73. “Nikołayos Maṙə ev nra gitakan žaṙangut‘yunə”, Iran-Namē, N18-19: 18.
74. “Kaxardnern u mogut‘yunə Hayastanum ev Iranum”, Iran-Namē, N18-19: 236-28.
75. “Akademikos Gagik Sargsyan (Cnndyan 70 amyaki kapakc‘ut‘yamb)”, Iran-Namē,
N18-19: 30-31.
76. “Baṙak‘nnakan-stugabanakan ētyudner — I, II”, Iran-Namē, N18-19: 32-34.
77. “Haykakan p‘oxaṙut‘yunnerə t‘urk‘erenum”, Iran-Namē, N18-19: 35.
78. “Sirołakanut‘yunə hasarakakan gitut‘yunnerum”, Iran-Namē, N20-21: 1-2.
79. “Iranagitut‘yunə Hayastanum”, Iran-Namē, N20-21: 18-19.
80. “Harem”, Iran-Namē, N20-21: 20-23.
81. “Iran, Rusastan, Hayastan (Azgayin anvtangut‘yan xndirner)”, Iran-Namē, N20-21:
27-28.
82. “Inčpes en aṙaǰac‘el šaxmatə ev nardin: šaxmati ev nardu meknut‘yunə (Pahlava-
kan bnagri t‘argmanut‘yunə aṙaǰabanov ev manramasn canot‘agrut‘yunnerov)”,
Iran-Namē, N20-21: 38-41.
83. “Reviews”, Iran-Namē, N20-21: 33, 59, 62.
 Select Bibliography 697

84. [Editing of:] Ałayed un-nisa. 17-rd dari parskakan ergicaxaṙn matyan nvirvac ka-
nanc‘, Erevan, 32 pp.
85. “Byzantium in New Iranian Classic Literature and Folklore”, Byzantium: Identify,
Image, Influence. Abstracts of the XIX International Congress of Byzantine Studies,
Copenhagen: 1244.
86. “Qoume gūrān”, Iranšenāxt, vol. 1, Tehran: 31-65:
87. “Suffiks -l- v kurmandži”, Strany i narody Bližnego i Srednego Vostoka XVI, Erevan:
193-197.
1997
88. “Georgi Ałabekovə, HYD-n ev k‘rderə”, Iran-Namē, N22-23: 9-10.
89. “Zaza žołovrdi mec zavakə. Ēbubek‘ir P‘amukču (1946-1991)”, Iran-Namē, N22-23:
11-14.
90. “Baṙak‘nnakan-stugabanakan ētyudner-III”, Iran-Namē, N22-23: 42-43.
91. “Zaza Halkı”, Raştiye, Paris, N4: 5-12:
92. “The Origins of the Kurds and the Early Armenian-Kurdish Contacts”, Iran and
the Caucasus, vol. 1: 1-16.
93. “Iranian Notes”, Iran and the Caucasus, vol. 1: 179-180.
94. Bāzsāzāī-ye vāžehāye fārsī va nāmhāye vīžeye īrānī, Erevan, 62 pp.
1998
95. Ētyudy po iranskoj ētnologii, Erevan, 120 pp.
96. [Editing:] Parsic lezvi ənt‘erc‘aran (in 2 vols.), Erevan, 148 pp. and 150 pp.
97. “Armenian xoygoƚowt‘wn: Tracing back an Old Animal-breeding Custom in An-
cient Armenia”, Iran and the Caucasus, vol. 2: 63-67.
1999
98. [Introduction:] Rossiya-Armeniya-Iran: Dialog civilizacii, Erevan: 1-2.
99. “The Holy Brotherhood”, Iran and the Caucasus, vol. 3-4: 79-97.
100. “Iranian Miscellanea”, Iran and the Caucasus, vol. 3-4: 203-209.
101. “Kurdish Etymologies I”, Iran and the Caucasus, vol. 3-4: 209-213.
102. “The Origin of the -ng- Suffix in Kurmanji”, Iran and the Caucasus, vol. 3-4: 213-215.
103. “The Mothers of Night: An Armenian-East Iranian Parallel”, Iran and the Cauca-
sus, vol. 3-4: 171-173.
104. “A Manual of Iranian Folk Magic”, Iran and the Caucasus, vol. 3-4: 239-243.
105. “The Delmiks (Translation and Commentaries)”, Iran and the Caucasus, vol. 3-4:
409-412.
2001
106. “Farhang-e rīšešenāsī-ye zabān-e fārsī”, Maǰalle-ye zabānšenasī, vol. 16, Tehran: 21-
34.
698 Prof. Garnik S. Asatrian

107. “Die Ethnogenese der Kurden und frühe kurdisch-armenische Kontakte”, Iran
and the Caucasus, vol. 5: 41-75.
108. “Al Reconsidered”, Iran and the Caucasus, vol. 5: 149-157.
109. “Blunt, Bald and Wise (Iranian kund)”, Iran and the Caucasus, vol. 5: 201-207. (with
V. Arakelova)
110. “D. L. Lorimer’s The Bakhtiari Tribal”, Iran and the Caucasus, vol. 5: 227-239. (with
F. Vahman)
111. [Reviews in:], Iran and the Caucasus, vol. 5: 239-261.
112. “Arūsīye gūsfandān”, Irānšenāsī, Bethesda, vol. 1: 450-462.
113. “Az tārīx-e vāžehā”, Nāme-ye Farhangestān (Journal of the Iranian Academy), vol.
1: 99-112.
2002
114. The Ethnic Minorities of Armenia, Erevan, 25 pp. (with V. Arakelova)
115. Notes on the Language and Ethnography of the Zoroastrians of Yazd, Copenhagen,
115 pp. (with F. Vahman)
116. “Armenia and Security Issues in South Caucasus”, Connections, Special Issue,
Paris: 13-27.
117. “Armenien und Sicherheitsfragen im Südkaukasus”, Connections: Partnerschaft
für den Frieden Konsortium der Verteidigungsakademien und Institute für sicher-
heitspolitische Studien 1 Jg./Nr 3: 27-39.
118. “The Lord of Cattle in Gilan”, Iran and The Caucasus, vol. 6. 1-2: 75-86.
2003
119. “Malak-Tawus: the Peacock Angel of the Yezidis”, Iran and the Caucasus, vol. 7. 1-
2, 2003: 1-36. (with V. Arakelova)
120. “Tālešhā”, Peymān (A Quanterly Journal), N24, Tehran: 103-110.
121. [Review of:] M. Lazarev, Š. Mgoi, E. Vasil’eva, M. Gasratyan, O. Žigalina, Istoriya
Kurdistana, Moskva, 517pp., Iran and the Caucasus, vol. 7: 326-331 (with S. Marga-
rian)
2004
122. “The Muslim Community of Tiflis (8th-19th Centuries)”, Iran and the Caucasus,
vol. 8.1: 29-52. (with H. Margarian)
123. “The Yezidi Pantheon”, Iran and the Caucasus, vol. 8.2: 231-280. (with V. Arakelo-
va)
124. “Iranian Notes”, Spirit of Wisdom: Essays in Memory of Ahmad Tafazzoli, Califor-
nia: 238-44.
125. “Talish and the Talishis (The State of Research)”, Iran and the Caucasus, vol. 9.1:
43-72. (with H. Borjian)
126. “Talyši”, Talyšskie narodnye predaniya i legendy, Erevan: 1-6.
 Select Bibliography 699

2006
127. “Kurdish Lō-lō”, Iran and the Caucasus, vol. 10.2: 239-242.
128. [Review of:] Farhang-e vāžegān-e Tabarī (A Dictionary of Tabari), in 5 volumes,
edited by Jahangir-e Nasri Ashrafi, Iran and the Caucasus, vol. 10.2: 303-305.
129. [Review of:] Dānešnāme-ye Jahān-e Eslām (Encyclopaedia of the World of Islam),
edited by G.-‘A. Haddad-e Adel, vol. 9, Iran and the Caucasus, vol. 10.2: 309-310.
2007
130. [Introduction:] K‘alila ev Dimna: Aṙakner (Trans. from Persian Gevorg Asatrian),
Erevan: 1-10.
2009
131. “Prolegomena to the Study of the Kurds”, Iran and the Caucasus, vol. 13.1: 1-57.
132. “Iranian Notes III”, Iran and the Caucasus, vol. 13.2: 319-330.
133. “Lorimer D. L.: Life and Career”, Encyclopaedia Iranica, Columbia, N.-Y., 2009.
2010
134. [Introduction:] “Inč ē hek‘iat‘ə”, Parskakan žołovrdakan hek‘iatner, Erevan: 1-24.
2011
135. “Bāft-e nejādi-e mardom-e Irān: Gūnāgūnī-e qoumī yā gūnāgūnī-e zabānī”, Far-
hang-e mardom, Tehran : 10-26.
136. [Editing and Introduction:] Parsic‘ lezvi dasagirk‘, Erevan: 286 pp.
137. “Editor’s Note”, Iran and the Caucasus, vol. 15.1-2: 1-6.
138. [Review of:] Claudia Ciancagnini, Iranian Loanwords in Syriac, Wiesbaden, 2008,
315 pp., Iran and the Caucasus, vol. 15.1-2: 322-325.
139. [Review of:] Ela Philippone. The Fingers and their Namēs in the Iranian Lan-
guages, Verlag der Oesterreichischen Akademie der Wissenschaften, Iran and the
Caucasus, 15.1-2: 326-333.
140. Comparative Dictionary of Central Iranian Dialects, Tehran, 1000 pp.
141. [Introduction, editing and Chapter 1], “O proisxoždenii talyšej: Kadusii i talyši”,
Vvedenie v istoriyu i kul’turu talyšskogo naroda, Erevan: 5-12.
142. “Kto byli predki talyšej?”, Vestnik Talyšskoj Nacional’noj Akademii, Minsk, N1: 64-
66.
2012
143. “Marginal Remarks on the History of Some Persian Words”, Iran and the Cau-
casus, vol. 16.1: 105-116.
144. “The Festival of Throwing Stones”, Iran and the Caucasus, vol. 16.2, 2012: 185-187.
145. Étničeskaya kompoziciya Irana: Ot “Arijskogo prostora” do Azerbajdžanskogo mifa,
Erevan, 130 pp.
700 Prof. Garnik S. Asatrian

2013
146. “Armenian Demonology: A Critical Overview”, Iran and the Caucasus, 17.1: 9-25.
147. “Talantlivyj vostokoved iz Luganska”, Sxidnij svit, N2-3, Kiev: 5-8.
148. [Editor’s Note:] G. G. Pogosian (Xaxbakian), Kul’turno-istoričeskoe nasledie Arme-
nii. Naxidževanskij kraj, Erevan, 2013, 355 pp.: 5-7.
149. [Editor’s Note:] G. G. Pogosian (Xaxbakian), Armyane Šamaxi v upominaniyax
nekotoryx zapadnoevropejcev, Erevan, 2013, 352 pp.: 5-10. (with S. A. Margarian)
2014
150. “Parfyanskoe gōsān”, Commentationes Iranicae. Sbornik statej k 90-letiyu V. A. Liv-
šica, St. Petersburg: 102-105.
151. “Middle Iranian Lexical Archaisms in Armenian Dialects”, Gherardo Gnoli Memo-
rial Volume, ed. by A. Rossi, E. Provasi, E. Morano, Wiesbaden (forthcoming).
152. “On the South Caspian Contact Zone: Some Talishi Folk Beliefs”, Iran and the
Caucasus, vol. 18.2: 135-146.
153. “Nose in Armenian”, Iran and the Caucasus, vol. 18.2: 147-152.
154. Armeniya, Iran i region, Erevan, 514pp.
155. The Religion of the Peacock Angel: The Yezidis and Their Spirit World, London, 160
pp. (with V. Arakelova)