Dossier: Was the Early Arabian ʿūd “fretted”?

Amine Beyhom

To cite this version:

Amine Beyhom. Dossier: Was the Early Arabian ʿūd “fretted”?. NEMO-Online, 2020. �hal-03680896�

HAL Id: hal-03680896

https://hal-cnrs.archives-ouvertes.fr/hal-03680896
Submitted on 29 May 2022

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est

archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents
entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non,
lished or not. The documents may come from émanant des établissements d’enseignement et de
teaching and research institutions in France or recherche français ou étrangers, des laboratoires
abroad, or from public or private research centers. publics ou privés.
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

“Al-Kindi oversaw one of the two main groups of

translators in the ninth century (the other group was
led by Hunayn ibn Ishaq). The ‘Kindi circle’ […]
translated numerous works of philosophy and science
from Greek into Arabic. […]
Al-Kindi seems to have been a mediator
Dossier: Was the Early between the patrons of these translators
and the scholars who actually did the translating,
Arabian ʿūd “fretted”? many of whom were Syrian Christians
or of Syrian extraction. His own writings might be
thought of as a sustained public relations campaign
intended to display and advertise the value of Greek
Amine Beyhom  thought for a contemporary ninth-century
Muslim audience”
[Peter Adamson, “Al-Kindi”]3
“I have long felt that the practical music of many
Asiatic peoples, ancient and modern, must have
been and must be a totally different thing from INTRODUCTION
the metaphysical or mathematical music of their phi-
losophers, which as pure speculation must always This dossier is a complement to the dossier on
have held itself apart from practice. We have erred Hellenistic Orientalism published in 2016 in NEMO-
in reasoning from the writings of theorists among Online4, in which I explained the main process of
these peoples to the nature of their art itself.
Orientalism in musicology and how it was based on
[…T]herefore, we should not say: the music
of the Chinese, of the Indians [Hindus], of the Arabs, Music theories beginning with Greek Neo-Pythago-
of the Persians etc., but: the musical system (or enig- rean theories and ending with the so-called “Reso-
mas) of the Chinese, the Indian, Arab, Persian philo- nance” theory.
sophers, of Master Chrysanthos, etc. – Maybe it was
no different for Ancient Greek music…” I also explained the need for most Orientalists to
confine Early Arabian theories of the scale to the Py-
[Raphael Georg Kiesewetter,
Über die Musik der neueren Griechen]1 thagorean tonal model based on the division of the
(Just) fourth in two whole-tones and one leimma.
This division – that I name “ditonic” to differentiate
“In the best cases, the theory [of music] inspires or
it from other “diatonic” divisions of the tetrachord5
enlightens the [musical] practice, […]. On the con-
trary, in the worst cases, theory comes to the aid (see Fig. 3:118)6 – translates in the equal-tempered
of ideology to impose one [particular] system 2 whole-tones and one half-tone commonly used to-
and erase subtle nuances” day. Apart from the use of biased, unfit for the anal-
[Jean During, “Introduction au Volume ysis of Arabian music, theories and notation(s) as
71 1/2 de la Revue de Musicologie”]2 explained earlier7, and besides the fact that the Py-
thagorean theory – and other theories – of the scale
fail at explaining the formation of the heptatonic

Amine Beyhom is currently the editor in chief of NEMO-Online 5
I use the terms “ditonic” and “ditonism” to differentiate the west-
and director of the CERMAA (Centre de Recherches sur les Mu- ern exclusive concept of “diatonism” (or “tense diatonic” which cor-
siques Arabes et Apparentées), a research centre affiliated to the responds to the Pythagorean two whole-tones + leimma division of
FOREDOFICO foundation in Lebanon. the tetrachord – or to the equal-tempered two tones + one half-
1
[Kiesewetter, 1858, p. 32]: partly translated and quoted in [Gil- tone division) from generalized diatonism (or “zalzalism”), exam-
man, 1892, p. 57]. ples of which are provided in Fig. 3:118. (See also [Beyhom, 2016,
2
[During, 1985, p. 7]. My heartfelt thanks to Jean During and Rich- Chapter 1] for a discussion of these concepts.)
ard Dumbrill who reviewed this dossier and proposed numerous
6
The second number for the figure is the page number.
improvements. 7
See aforementioned [Beyhom, 2016], and [Beyhom, 2018].
3
[Adamson, 2011].
4
[Beyhom, 2016], more specifically the section on “Musicological
Orientalism” p. 158-163.

113
NEMO-Online Vol. 5 No. 9 – November 2020

scale, 8 Orientalist musicologists – beginning with crystallization in the ditonic paradigm. Thus the Ar-
Henry George Farmer and (not) ending with Eck- abs would have – according to Orientalist musicolo-
hard Neubauer – promoted with great diligence the gists – merely copied their theory from their prede-
hypothesis of the fretting of the Early Arabian ʿūd, cessors, and their music would have further “re-
to the point that most musicological publications gressed” being influenced by Persian (or other) mu-
dealing today with Early Arabian music consider sic(s)15, i.e. musics supposedly outside the realm of
this “fretting” as an established fact. restricted Hellenism. In the meantime, European
(musical) culture retrieved its legitimate Greek leg-
Thus, in the Encyclopedia of Islam:
acy in its “purest” (ditonic) form, from which we
“Unlike the mediaeval lute, the modern lute is not fretted”,9 can conclude that Europe and the Occident became
or further widened such as in Poché’s assertion in effectively the only “legitimate heir” of Greek cul-
the New Grove: ture and civilization.

“The neck [of the ʿūd] rarely has frets (dasātīn), but some are In parallel to this demonstration – or “fairy tale”
found on the Tunisian ʿūd of Khumayyis Tarnān”,10 – and as I show further, all indications in the Early
which is all that Christian Poché had to say on the Arabian treatises on praxis16 at that time are deemed
matter,11 while we can read in the same dictionary: insignificant or simply avoided – as with Neubauer
for the latter process –, the role of ditonism is am-
“The ʿūd still survives over all the Arab world, where it is used
plified and Zalzalian 17 praxis minimized while ar-
as a solo instrument and for accompanying song,12 though it
no longer has frets”.13 cheological evidence is ignored for the sake of “con-
tinuity” and, when the evidence becomes too in-
While this myth has already been invalidated
sistent, Arabian music becomes promoted as formu-
elsewhere14 and is further invalidated here, very few
lary music with the scale playing a secondary role
contemporary researchers have put in doubt this
in its structure.
common-place belief. To understand fully the rea-
sons of the persistence of this fabrication against all As with the “ditonism of the origins” of Byzan-
indications of its invalidity, there needs only to re- tine chant – which ended up being a major fabrica-
member that the music of the Early Arabs, in the tion of Western Byzantinologists18 – but for opposite
eyes of Occidental musicologists, may explain the purposes19, the alleged fretting of the ʿūd served as
European music of the Middle Ages and its (later) the main vector of the historical forgery of music
history, mostly in the 19th and – mainly the first half

8
[Beyhom, 2016] and [Beyhom, 2003 ; 2004 ; 2010a ; Beyhom, 16
With regard to the scale and intervals used by performers.
2017]. 17
“Zalzalian”: non-tempered music, not based on semi-tonal scales,
9
[Chabrier et al., 2000], entry “ʿūd” (to which Farmer contributed and mainly relating to maqām music. The terms Zalzalian and zal-
originally). zalism are used after Manṣūr Zalzal a-ḍ-Ḍārib, an 8th-9th-centuries
10
[Poché, 2001, p. 27], entry “ʿūd”: as explained in [Beyhom, ʿūd player who was (supposedly) the first to introduce the fingerings
2016], all Tunisian colleagues and musicians that I could consult of the mujannab(s) – that is the so-called “neutral” seconds and
on the matter confirm that they never saw – or heard of – “frets” on thirds – on the fingerboard of the ʿūd. It refers more generally to
the ʿūd of Khumayyis Tarnān. intervals (or musical systems which use them) using subdivisions
11
I suspect that Poché deliberately avoided a subject he knew was other than the semi- (or “half-”) tone, noticeably all the varieties of
very controversial, precisely because of Farmer’s (posthumous, en- mujannab seconds spreading from the (exact or Pythagorean) half-
during) influence on the musicology of Arabian music. tone to the disjunctive (Pythagorean, or whole) tone. The same ap-
12
This is a very strange statement which restricts greatly the use of plies to intermediate intervals between the (exact or Pythagorean)
the ʿūd today as it is included in both large orchestras and small tone and the one-and-half-tone interval (either equal-tempered or
formations (sometimes few lutes playing together) in the Arab Pythagorean “augmented” second), etc.
world – as well as in Europe – and frequently today in jazz ensem-
18
See the “Appendix On the Origins and alleged ‘Diatonism’ of Byz-
bles (for fusion music). antine Chant” in [Beyhom, 2015, p. 429–478], and Chapter 4 in
13
[Wachsmann et al., 2007]: entry “Lute”. (Bold type mine.) [Beyhom, 2016].
14
See [Beyhom, 2010b, v. 1, p. 324–363 ; Beyhom and Makhlouf,
19
Excluding maqām music from the evolutionary scheme, and in-
2009]. cluding Byzantine chant in the European identity.
15
See for instance [Parisot, 1898, p. 10].

114
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

of the – 20th centuries, the sole purpose of which was the Carolingian Era (9th century – See Fig. 2.) and
the establishment of an evolutionary process of mu- abundantly copied since25 – purely on Pythagorean
sic20 culminating with the Western Classical music ground.
of the common-practice period21.
While trying to prove that European music is the
“Why?”, could – still not convinced – benevolent heir of (the music of) Ancient Greece, musicologists
musicologists ask, “what has the West to gain in de- were compelled to consider the missing link, which
fending the Pythagorean or ‘ditonic’ thesis”, “why is Arabian writings on music (theory). Therefore,
do they wish to retain their simplistic scale when- Early Arabian music must have been ditonic (as An-
ever this affects their music – and its perception – cient Greek music was mainly supposed to be), and
rather negatively?”, would they sustain?22 transmitted to the West on this ground.

As explained above, the answer is simple: as the

Ancient Greek legacy came to Europe mainly – at
least in the first centuries of the Islamic civilization
– from Arabian sources, Arabian music of the Early
Islamic times (7th-9th centuries) was considered by
European musicologists in the 18th and the (first half
of the) 19th centuries to be the missing link between
Ancient Greek music and the European music in the
(European) Middle Ages23. As European music was
allegedly a relic of the Early European music, it
could only be ditonic in its essence, as with Euro- Boethius’ tomb in San Pietro in Ciel d’Oro, Pavia.26
pean music in the common-practice period.
As I already wrote in “The ‘fretting’ of the Ara-
Furthermore, prominent researchers on Arabian bian ʿūd – or Sequencing Musicological Oriental-
music such as Henry George Farmer promoted the ism”27, as long as the main threat comes from the
existence of an embryonic form of polyphony – on a theories of the scale, and while early Arabian writ-
ditonic basis, evidently – in Early Arabian music ings about music theory 28 base the scale theory on
which confirmed – in their opinion – the role of the the ʿūd, this instrument had to be fretted because if
latter music as the missing link with Ancient Greek it were not, this would leave open perspectives for
music.24 all non-tempered musics to be performed on it,
which would directly contradict the evolutionary
Let us remember that the early theory of Western
thesis based on the ditonic dogma.
music was heavily influenced by Boethius’ (see Fig.
1) De institutione musica – which was rediscovered in

20
This procedure is explained in detail in [Beyhom, 2016], more 24
Byzantine Chant could have been a parallel link to Ancient
particularly in the “Preliminary Synthesis” [Beyhom, 2016, p. 175– Greece, but its “Oriental” nature deeply disturbed European spe-
176], the reading of which is recommended for a better understand- cialists who followed a similar scheme, however not to exclude but
ing of how the Orientalist scheme led to the necessity of the “fret- to integrate Early Byzantine Chant in (Western) Europe.
ting” of the ʿūd. 25
See [Boethius, 2004, p. 1]. See also [Wikipedia Contributors,
21
The terminology is borrowed from Ruth Solie’s “Melody and the 2020] and the preface of the more academic [Bower, 1989, p. xiii]
Historiography of Music” [1982, p. 297]. (written by Palisca), notably: “Beginning around the ninth century,
22
These points were effectively questioned by Jean During in a pri- De institutione musica became established as the foundation of West-
vate and virtual discussion about this dossier on September 19th ern music theory, and throughout the Middle Ages Boethius re-
2020. mained the authority most revered for music-theoretic matters.”
23
This time-period corresponded to the Golden Era of Islamic civi-
26
Retrieved from [Dall’Orto, 2009]: Boethius was a “martyr” of the
lization. Catholic cause.
27
[Beyhom, 2016, p. 159–162].
28
And until at least the 14th-15th centuries.

115
NEMO-Online Vol. 5 No. 9 – November 2020

treatise as well as a teaching method for the

instrument.

 In the only instance where Kindī describes the ʿūd

playing techniques and strings stopping, his
explanations are at some point inconsistent and
incompatible with an effective fretting of the
instrument.

 Kindī further described notes “used by singers”

The Carolingian Empire at its peak.29
from which it can be deduced that the effective
The procedure followed by Orientalist musicolo-
division of the scale was the seventeen (unequal,
gists was therefore to change the Zalzalian aspect of
Zalzalian) intervals division explicitely given by
theories using this instrument, namely: his famous successors Fārābī (the “Second
 Firstly, and from one, single (theoretical) description Master” – by reference to the “First Master”,
by the first major Arabian theoretician, Yaʿqūb Ibn Aristotelēs) and Sīnā.30
Isḥāq al-Kindī (“The Philosopher of the Arabs”), and  All subsequent authors who mention “ligatures”
by neglecting all indications about praxis given by on the neck of the ʿūd explain that the string must
the author, the “Early Arabian ʿūd” (of the “Middle be stopped at exactly the position of the “tie-
Ages”) is proclaimed “fretted” ditonically, thus: fret” 31 , (which is incoherent with the “ties”
 Early Arabian music was ditonic and tempered. having the function of effective frets – as shown
in Appendix B), and that most of them mention
 Secondly, from this first example, it is taken that all the possibility of stopping the strings between the
Arabian ʿūd(s) were “fretted”, not only in theory but ligatures, or to use hand shifts (towards the
also in praxis, not only at the time of Kindī, but from bridge) for higher notes, to positions where there
the very beginnings of this music until the post- are no ligatures (or marks).
Ṣafiyy-a-d-Dīn period (post-13th-Century),
“forgetting” that:  The second proven description of an effective
“fretting” of the ʿūd 32 is found in Muḥammad ibn
 Kindī was the first Pythagorean philosopher al-Ḥasan ibn a-ṭ-Ṭaḥḥān’s treatise. He was a
influenced by Plato who took over Ancient Greek Fatimid musician, singer and teacher who
theories for the purpose of theorizing the yet un- explained that a particular type of fretting was
theorized Arabian music of his time, and that it used for beginners.
was tempting to materialize the Pythagorean
division of the octave directly on the neck of the  All subsequent authors mentioning ligatures
ʿūd – the primary instrument of Arabian music at (dasātīn) on the neck of the ʿūd either do not
that time and up to the present. mention any material for those, or say that they
are marks 33 on the surface of the neck indicating
 Kindī’s epistle Risāla fī-l-Luḥūn wa-n-Nagham – in stopping positions of the strings.
which the description of the “frets” is given – was
All these facts are ignored – or brushed aside
dedicated the son of Caliph al-Muʿtaṣim (833-
when mentioned – by Orientalists and hence, de-
842), Aḥmad ibn al-Muʿtaṣim – an amateur
spite very few contradicting views, the myth of the
musician – and was meant as an informative
fretting of the early ʿūd promoted by a series of more
29
Source: https://3.bp.blogspot.com/-tn4kVZWdXos/Ulfqyedi_wI/ dastān (pl. dasātīn) used in the Arabic literature about music theory
AAAAAAAAInY/W29gY0Powto/w1200-h630-p-k-no-nu/Carolin- which avoids the recourse to the term “fret” – which is misleading.
gian-Dynasty.gif. The dasātīn (from the Persian “dast”, “hand”) were generally, as I
30
As explained in [Beyhom, 2010b, v. 1, p. 183-276 (Chapter II)]. show further, marks on the neck of the instrument.
31
I use the term “tie-fret” for frets made of gut or other material
32
Probably inspired by Kindī’s description.
which are winded around the neck of the ʿūd and knotted on the 33
Or possibly threads as explained further.
back side. “Ligature” is a more reliable translation of the term

116
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

or less renowned authors including Lachmann, erroneously fretted by (some) Western musicologists
Farmer, Manik, and finally Neubauer, is still taught – including Eckhard Neubauer’s attempts at reviving
in maqām musicology34 against all factual data35. the thesis of the fretting of the (early) Arabian ʿūd in
his article “Der Bau der Laute und ihre Besaitung nach
The main aim of this dossier is to assemble all
arabischen, persischen und türkischen Quellen des 9. bis
possible data about this alleged fretting of the in-
15. Jahrhunderts” 40 – then by autochthonous re-
strument, in order to draw worthwhile conclusions,
orientalists. It exemplifies the – willful or
set on a firm ground. unconcious – blindness of some (modern and
contemporary) Orientalists when it comes to the
PREFATORY REMARKS ditonic (or “tense diatonic”) dogma of Western
musicology.
This dossier is composed of three main parts and
 The third part consists in a series of four appendices:
accompanied by two videos:
 Appendix A (“The ʿūd, its components and its
 Part I features explanations about (al-) Kindī’s 36
proportions”) is a reminder about the proportions
division(s) of the fingerboard of the ʿūd. It then
of the ʿūd and its components in the early period
explains the partitioning of the tetrachord in seven
– and nowadays for its proportions.
divisions (and of the tone in three divisions) which
ends with the partitioning of the octave in 17  Appendix B (“Organological clarifications”) lists
unequal intervals (in this case with Zalzalian the organological problems raised by the fretting
intervals, or generalized diatonism37). This division of the ʿūd.
is present in Arabian specialized literature from the
 Appendix C reviews the contents of The Risāla fī-
very beginnings,38 and is rooted in music practice
l-Mūsīqā by (al-) Munajjim (856-912) and shows
since the Forerunners 39 . It was the main
that the Pythagorean division attributed to this
representation of the scale in the Golden Age of the
author cannot be sustained.
Arabian Civilization from (al-) Fārābī (9th Century)
to (al-) Urmawī (13th Century).

 Part II is a reflection about the theoretical use of the

Arabian ʿūd, and how this instrument was

34
I recently had to warn a colleague from publishing in an article ṣ-)}, 1991]; 5. The Moderns: beginning with Mashāqa (19th century)
that the Early ʿūd was fretted, despite his protests that this “fretting” and his mentor Farīd-a-d-Dīn al-ʿAttār and ending with the 1960s
was “an established fact”. (not forgetting [Khulaʿī (al-), 1904]); 6. The Contemporary Period:
35
In fact, a converging array of evidence contradicting the thesis of roughly since the 1970s and the predominance of the Conserva-
the “fretted” ʿūd. toires in the teaching of Arabian music. (Note that periods 3 and 4
36
The “Philosopher of the Arabs” and the first author whose works may overlap.) As for Arabian music per se, [Jargy and Chottin,
on Arabian music theory are (partly) extant. 2001, p. 527] identifies (for example – Guettat has another division
37
The term “generalized diatonism” is used to oppose the general still, as seen in Chapter V of [Beyhom, 2016]) five time periods
concept of diatonism in Ancient Greek theories to the particular (which correspond only partly to the aforementioned six, and dis-
tense (Western) diatonism. (See Fig. 3:118.) regard the post-Congrès du Caire period), namely: “1) Bedouin period,
from the Jāhiliyya [‘the time of ignorance’] till Early Islam (death of
38
Although not explicitly in the case of Kindī.
ʿAlī, 661); 2) Assimilation period, from the Umayyad dynasty till the
39
The term comes from my proposed (in [Beyhom, 2010b]) divi-
First Abbasid cycle (circa 830) ; 3) Period of Fulfilment and Dispersion,
sion of the history of maqām music (theory): 1. The Forerunners:
with the second Abbasid cycle and the establishment of the Umay-
mostly (al-) Kindī (9th century) and (al-) Munajjim (9th and begin-
yad in Spain; 4) Period of Decline, from the taking of Granada [note
ning of the 10th centuries); 2. The Golden Age: from (al-) Fārābī (lat-
here that Jargy does not term this as ‘the Fall’ of Granada] (1492)
inized “Alfarabius” – 10th century) to ibn Zayla (d. 1048), not for-
till the end of the 18th century; 5) Renaissance: from the Nahḍa
getting the mentor of the latter, ibn Sīnā – or Avicenna – (980-
[hence the term “Renaissance”] in the 19th century, beginning with
1037); 3. The Systematists: beginning with (al-) Urmawī (13th cen-
the expedition of Bonaparte in Egypt end of the 18th century, until
tury), with followers such as (al-) Lādhiqī or (al-) Marāghī; 4. The
the [C]ongrès du Caire (1932)”.
Intermediate Period: with writings such as the Anonymous A-sh- 40
[Neubauer, 1993], which is, as a matter of fact, a dossier of nearly
Shajara dhāt al-Akmām ‘published as [Anonyme, 1983]), or from
80 pages.
[Ṣaydāwī (a-ṣ-), XVe siècle] (translated to French in [Ṣaydāwī (a-ṣ-)
and Antar, 2001]) or the pseudo Ṣafadī published as [{Ṣafadī (a-

117
NEMO-Online Vol. 5 No. 9 – November 2020

Ancient Greek tetrachords with equivalents in the writings of (al-) Fārābī (9th-10th centuries – see [Wright, 2001a]) and (ibn)
Sīnā (10th-11th centuries – see [Wright, 2001b]), the two major Arabian music theoreticians of the Golden Age. Arabian tetrachords are
taken from [Fārābī (al-), 1930 ; Fārābī (al-) et al., 1935 ; Yūsuf, 1956 ; 1998 ,‫ الفارابي‬1967 ,‫ ;]أبو نصر محمد بن محمد بن ترخان الفارابي‬Greek tetra-
chords from [Mathiesen, 1999]; the enharmonic tetrachord in its 2nd form in the lower table (Ptolemaeos – last column to the right) is
taken from the Appendix of [Erlanger, 1930]. First published (in French) in [Beyhom, 2010b].

Urmawī’s two divisions of the tone (L L C and alternative L’ C’ L’’ – from right to left, top to bottom, in the figure) in the Kitāb al-
Adwār [Urmawī (d. 1294), 2001] and corresponding ratios and intervals in cents.

118
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

 Appendix D (“Original texts”) is a collection of A CONCLUSION AS FOREWORD47

the original texts the translation of which is
Between the 7th and the 9th centuries, the expansion
proposed in the main text.41
of Islam48 resulted in an Arabic-speaking empire extend-
 The two complementary videos were published with ing from Persia to Spain, including North Africa and
the article on the ʿūd – in French – by the author with parts of Central Asia. The major confrontation of this
Hamdi Makhlouf42, with subtitles by Beyhom: Arabian-based empire was, at that time, not with the
West but with the Byzantine Empire which predomi-
 The first video is entitled Fretting of the ʿūd
nated on its North-Western front. The music of the king-
according to (al-) Kindī,43 and shows the stringing
doms of the Arabian Peninsula could compete with dif-
and positioning of the frets as explained by (al-)
ficulty with the music of some of the conquered peoples,
Kindī, for both a “Harmonic” and a Pythagorean
as with the music of Bilād a-sh-Shām49 being part of the
tunings.
former Roman Empire,50 and with the music of Persia.
 The second video is entitled Fretting of the ʿūd Claiming a purity of the Arabian (“Bedouin”) musical
according to Ibn a-ṭ-Ṭaḥḥān44 and shows the same art would be, with such facts at hand, an aberration.
procedure but with one set of strings described by The process of acculturation of the Bedouin Arabs,
(ibn a-ṭ-) Ṭaḥḥān. which became the rulers of an empire extending far be-
While most of the material proposed to the yond their original habitat, is unfortunately not docu-
reader is based on my first book (in French) about mented for their music.51 All the writings on music of
Arabian music theory and praxis to the 13th century the period, anterior to (al-) Kindī’s, have been lost. Later
chroniclers such as Masʿūdī and (ibn) Salma52 have tried
and on the article published with Hamdi Ma-
to retrace the evolution of the Arabian society towards
khlouf45, new data is provided in this dossier which
a Pan-Islamic society, starting with the small kingdoms
complements my earlier writings on the subject.46
of the Arabian Peninsula and ending with the Abbasid
caliphate, and recreated thus the illusion of a continuity
of the original “purity”, an ongoing and exclusive filia-
* tion of the Arabian Empire.
* *
It is, however, self-evident that Arabian music in
the 9th century could only be the hybrid result of
the Islamic melting pot, with influences as diverse as
Persian, Byzantine, Mediterranean music and music
from Central Asia (Fig. 5).53
41
This is for clarity of the main text, while philological differences Arabian writings”, may be of use for the reader and is cited where
impose the reproduction of the original texts for verification pur- deemed necessary.
poses. 47
This section was originally the “Synthesis” of the first chapter of
42
[Beyhom and Makhlouf, 2009]. [Beyhom, 2010b] regarding the theories of the Forerunners, which
43
Available at https://youtu.be/d7TTlnH_pKM. seemed to me best suited as a foreword to this dossier.
44
Available at https://youtu.be/demT-hpcX1s.
48
Both religion and civilization.
45
[Beyhom, 2010b] and aforementioned [Beyhom and Makhlouf,
49
Syria and Lebanon and, by extension, Jordan and Palestine.
2009]. Most of the translated text from [Beyhom, 2010b] is an 50
Then of the Byzantine Empire (or the Eastern Roman Empire).
adapted, emendated and shortened version the purpose of which is 51
The process of acculturation was not one-sided: Arabian culture
to expound the results of the research undergone, while more de- (poetry, language, rhythms, music) influenced also the conquered
tailed explanations (and translations) are proposed for the alleged peoples, whatever influence the culture of the latter had on the Ara-
fretting of the ʿūd. Likewise, other (Pythagorean) divisions (such as bian rulers.
by Ikhwān a-ṣ-Ṣafāʾ and al-Khawārizmī) are not incorporated in the 52
See [Masʿūdī, 1987 ; Salma (a-n-Naḥawī al-Lughawī), 1984], and
dossier, as Kindī’s and Munajjim’s propositions (expounded further) [Khalidi and Masʿūdī, 1974] for the importance of Masʿūdī as a his-
are a significant enough sample of the theoretical speculations of torian.
the “Forerunners”. 53
Two short (and available) references on, respectively, the con-
46
Note that some of the material included in [Beyhom, 2016], quest of Egypt and the conquest of Central Asia are [Butler, 1902]
mostly from the last section of Chapter I entitled “Greek theories in and [Gibb, 1970].

119
NEMO-Online Vol. 5 No. 9 – November 2020

More than two centuries after the beginning of the (and still is today) very different from the modern West-
expansion of Islam54, Arabian scholars and philosophers ern concept,57 and that this procedure was concurrently
had to get on with the heavy task of characterizing this transposed to translations from Greek Masters, it be-
music and to establish a unified presentation of it in- comes less surprising that many of the most early writ-
tended, above all, for the Abbasid Caliphs and for other, ings on Arabian music are much alike, and use mainly
lesser, contemporary potentates. Pythagorean ditonism as the basis of their theoretical
This procedure took place concurrently with the as- explanations.
similation of the vast scientific and cultural corpus of This theoretical handling, although already
Ancient Greece from which these scholars quickly tried breached in Kindī’s epistle (“Risāla”) fī-l-Luḥūn wa-n-
to establish an “Arabian” 55 music theory with pretense Nagham, changes radically with (al-) Fārābī 58 in his
to universality. (Examples of the appropriation of An- Great Book of Music in which we find, finally(!), the ex-
cient Greek music theories by Arabian theoreticians are pression of a powerful (and critical) mind exploring mu-
proposed in Fig. 3:118 and Fig. 4:118.) sic and music theory of his time. This theoretician’s ap-
proach is respectful of “The Masters of the Art” – Ahl a-
ṣ-Ṣināʿat to which he refers when practical details are
needed – and of practical music, which ended up in him
being the first to explicitly include Zalzalism in his the-
oretical descriptions of the Arabian scale.

*
* *

Expansion of the Caliphate until 750:  Expansion It is worthwhile, even at this early stage in this dos-
under the Prophet Muḥammad, 622-632;  Expansion during sier, wondering about the social and intellectual con-
the Patriarchal Caliphate (Rāshidūn), 632-661;  Expansion
texts which resulted in the exclusion – for many decades
during the Umayyad Caliphate, 661-750. (From [DieBuche,
2010] based on [Anon. “Age_of_Caliphs.png (Image PNG,
– of an already existing, even characteristic phenome-
684 × 347 pixels)”]: conquered territories included all or part of non as Zalzalism (or non-temperalism), from the theori-
the Levant, Mesopotamia, Persia, North Africa, Iberia, Gaul, zation of Arabian (maqām) music; this context is explic-
Transoxania, Sindh and Caucasus – see also [Wikipedia Contrib- itly scrutinized in the following pages.
utors, 2017b].)
Let us note that the already signaled (al-) Kindī59 –
It is important to remember that the first “theoreti-
surnamed the “Philosopher of the Arabs” – seems to
cians” of the Arabian Empire56 were neither simple mu-
have well earned his surname in music theory60 as he
sicians trying to codify and transmit their art, nor “mu-
was the first to include the ditonic division of the (Neo)
sicologists” in the contemporary sense of the word –
Pythagoreans in his theoretical reasoning.
meaning by that Music historians or analysts. Music
We should however also note from the outset that
“science” was therefore originally confined, through the
the principal aim of this philosopher was to incorporate
influence of Pythagoreanism and neo-Pythagoreanism,
Greek “science” in Arabian nascent philosophy – to “ad-
to the mathematical and cosmogonic domains, as the
vertise” it as Adamson writes in the epigraph. In such a
near totality of Early (and extant) works shows. Adding
context, the concordance between Greek theories and
to this fact that the Arabian concept of plagiarism was
Arabian (or even “Greek”) musical praxis becomes of
54
The civilization, here differentiated from the religion. for manuscripts in the libraries of Egypt, Morocco, Russia, Tunisia
55
I use “Arabian” for Arabian-Persian-Turkic – and later Ottoman and others seem to be also available (as noted in “Seminararbeit
– music. von Silja Geisler-Baum, Sommersemester 2004, Betreuung: Prof.
56
The fact that the Caliphate was an empire exonerates me from Dr. Ursula Georgy”).
specifying whether the authors were Arabs, Persian or Turkic (or
60
Unlike Aristoxenos – who, as reminded in [Beyhom, 2016] and
Armenian, Jew and other nationalities – or religions). even though he was also a philosopher, approached nevertheless
57
See [Grunebaum, 1944]. music from a practical point of view – Pythagoreans and Neo-Py-
58
The “Second Master” (to Aristotle). thagoreans had a strictly philosophical, if not dogmatic approach to
music.
59
For a comprehensive reviews of Arabian sources, see [Farmer,
1965 ; Shiloah, 1979] – a second volume of RISM by Shiloah, 2003,

120
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

secondary importance.61 Let us also note that Kindī was as Ikhwān a-ṣ-Ṣafāʾ and (al-) Khawārizmī – adopted the
more of a translator than a “transmitter” of Ancient Pythagorean premises of this philosopher, forgetting
Greek tradition.62 He was however the first to describe however about his mentions of praxis which stand far
the Arabian musical system through the division of the from intervallic mensuration and from arithmetical han-
fingerboard of the ʿūd, even if we do not know for cer- dlings. The latter – practical – approach, which would
tain if his use of the ditonic division of the fourth corre- have probably been welcomed by the musicians of his
sponds at all to praxis63 at that time. The ditonic divi- time, was unfortunately an exception.
sion, which is probably justified by Kindī’s pretense in Mentions of practice are rare – if not inexistent – in
his epistle fī-l-Luḥūn wa-n-Nagham64 to a “simplicity” of the literature until the advent of Fārābī, and while the
music – as a “science” and inherited from Platonic inter- latter tried to reform music theory he had, however re-
pretations – contradicts somewhat the description of the luctantly, to contend with earlier writings whatever la-
genē in his epistle fī Khubr Ṣināʿat a-t-Taʾlīf which in- cunae he may have found in them. Whenever Kindī
cludes, notably, an enharmonic genos with two quarter- avoided67 introducing new ratios to describe the Zal-
tones.65 zalian intervals used in praxis, (al-) Fārābī and later
Whenever the question of the adequacy of the sim- (ibn) Sīnā and (al-) Urmawī68, while keeping the ditonic
plistic ditonic division with the music of that time is norm imposed by their predecessor(s),69 integrated new
clearly raised by Kindī’s description of praxis (singing – and old divisions based on string-length equal-divisions,
ghināʾ – as opposed to “musical science”) in the Risāla fī- or recalling non-ditonic ratios used by these predeces-
l-Luḥūn wa-n-Nagham, his interest in the “science” of sors.70
music is undeniable, as testify the numerous epistles he More generally, the question that is raised concerns
devoted to the subject. the adequacy of the theoretical systems which were de-
In a very Arabian-like approach about the transmis- scribed by Arabian theoreticians, with praxis.71
sion of knowledge,66 several of Kindī’s successors – such
61
Plausibly, the same phenomenon took place in Western theory of compelled to modify it by introducing explicit zalzalism in his sec-
music beginning with the so-called “Middle-Ages”. ond major (and comprehensive) work, the Risāla a-sh-Sharafiyya
62
Notably for ethos theory and numerical correspondences with [Urmawī (d. 1294) and [Jurjānī (al-)], 1938].
the four elements, nature, etc. To “transmit” is here used in the 69
This includes Ancient Greek theoreticians that Fārābī would re-
sense of a living tradition which is handed down – modified and luctantly criticize, while preferring (see [Beyhom, 2016], p. 79,
augmented but still traditional – to others (see [During, 1994]). fn. 197) to ascribe their imperfections to the translators of their
63
Or to the extent of this practice. works (notably Kindī and his group of translators?).
64
Reviewed further.
70
See for example Appendix 3 in [Beyhom, 2016] and Fig. 3:118
65
Knowing that the translation of Ancient Greek sources was an as well as Fig. 4:118.
ongoing process in the time of Kindī, it is very possible that his suc-
71
This topic is seldom addressed for example by Sawa in his article
cessive epistles on music – for which we do not have a precise chro- [Sawa, 1981] or in his book [Sawa, 1989], although the author in-
nology – were based on different translations from different, and sists on the practical aspect of the music he researches, as in [Sawa,
more or less complete, Ancient Greek sources. 1981, p. 85–86]: “Obviously, even for ethnomusicologists inter-
66
See footnote no. 57:120 above. ested in modern musical practices and musical life, historical eth-
67
Maybe because of the lack of intellectual audacity, or capacity to nomusicology can be a lively and extremely useful subject of re-
conceive them: Arabian music “science” was still to be founded at search in at least two ways. First, it can clarify the reasons behind
his time and, while Kindī was a pioneer at introducing Ancient many modern concepts and practices. Second, ethnomusicologists
Greek theories to the Arabs, he would reluctantly “alter” them (or with an intimate knowledge of modern practices can clarify ambi-
the part of it he had access to). In his Risāla fī-l-Luḥūn wa-n-Nagham, guities in the historical sources. Finally, for present-day native Mid-
intended as a manual for the son of the Caliph (see further), Kindī dle Eastern musicians, the study of the past offers the necessary
had to resolve the obvious discrepancies between (Greek) theory methods and terminology for the study of their own music. This is
and (Arabian) praxis, which he did by signaling approximate posi- a much more suitable and fruitful procedure than borrowing irrel-
tions for Zalzalian (Arabian) notes between the notes of the Canoni- evant, if not damaging, concepts from 18th- and 19th-century Euro-
cal (ditonic, Pythagorean) division – as is explained further. pean art music”. While I agree with the conclusion of Sawa, I could
not help but note that the author’s descriptions of the Early Arabian
68
Although Ṣafiyy-a-d-Dīn al-Urmawī used a refined Pythagorean
theoretical systems are but a little too… theoretical, as he does not
adaptation of zalzalism in his Early Book of Cycles [Urmawī (d.
even address Kindī’s writings and neglects comparisons with praxis
1294), 1984 ; 2001] (see explanations and comments in [Wright,
with (al-) Fārābī and others, a steady attitude with re-orientalizing
1969], [Beyhom, 2010a ; Beyhom, 2018], and Fig. 4:118), he was
musicologists of the maqām (as explained in Chapter V of [Beyhom,
2016]).

121
NEMO-Online Vol. 5 No. 9 – November 2020

Other questions remain unanswered, concerning no- 17-intervals division of the octave (both in unequal in-
tably the relation of Arabian music praxis at the time – tervals) is a constant feature of these theories, beginning
ascribed to the court of the Caliph and to the high soci- with Kindī and extending to the late Systematists.
ety and characterized by the use of seven subdivisions
within the tetrachord (just fourth) –72 with the music of
the peoples of this vast empire. *
* *
Is it possible that Court music followed the same
rules and system as with shepherds, artisans, farmers,
city ruffians and prostitutes of both sexes scattered in
PART I. FIRST THEORETICAL AND PRACTICAL DE-
such disparate regions as the Arabian Rabʿ al-Khālī,
Post-Byzantine Syria, Egypt, Iraq, Central Asia and Per- SCRIPTIONS
sia, not to mention North-Africa and Spain?
The theoretical treatment of the scale in the period
Maybe not, and maybe yes, as popular musics today, of the Forerunners is characterized by the recourse to
in the Arab world, follow the same principles as with Pythagorean ditonism. It must be remembered that,
Art music,73 while the main question can still not be an- during Kindī’s time, the large-scale translation of An-
swered definitely as sources on the subject are unavail- cient Greek texts was still in its infancy.76 Arabian phi-
able, or maybe never existed.74 losophers hurried to use these texts and adapt them to
The second question which is (inevitably) raised Arabian music, whatever differences with praxis.
concerns the adequacy of the Pythagorean ditonic The first theoretical procedure of which we are
model with Court music per se: does the Pythagorean aware with Arabian theoreticians about the modeling of
substrate, which is contradicted by Zalzalian inclusions, the melodic vertical space is the division of the strings
coincide even loosely with the praxis of Art music at the on the neck of the ʿūd,77 mostly limited for each string
time? to its first acoustical characteristic interval, the fourth.78
Here again the lack of sources compels us to delay While music was assimilated by these philosophers to a
the answer to this question.75 What is today clear is that theoretical science, 79 and whenever the ʿūd was the
the 7-intervals per just fourth division, extended to the main (and very versatile) instrument for performance, it

72
This has been determined for Arabian music in [Beyhom, Kindī died in 866. The latter wrote numerous epistles on music that
2010b], beginning with the first Arabian Philosopher (and theore- we are unable to date precisely. Note also that, contrarily to previ-
tician), (al-) Kindī, and is expounded in Part I of this dossier. ous assimilation of Arabian music theory in this period, which be-
73
Till the Modern period and excluding school syllabuses. gins with (al-) Munajjim’s extant epistle on the subject, I begin in
74
Extant sources deal only with caliphal – or Art – music, with few my book (and in this dossier) this review with Kindī. This is justified
exceptions (such as Fārābī’s and Kātib’s descriptions of the ṭunbūr by the simple reason that Munajjim was born in 856 (and died in
Baghdādī – see [Beyhom, 2010b, v. 1, p. 311, 320]) – which are not 912 according to Farmer). (Note that Munajjim’s epistle includes a
conclusive. few references to Aṣfahānī – a music chronicler who gives no indi-
75
Sources are scarce or unavailable for the period before Kindī, and cations about the composition of the Arabian scale – and to Isḥāq
the research on Arabian scale theory must begin with works dating al-Mawṣilī – a well-known singer of the Abbasid period from which
two centuries after the advent of Islam (the religion) – in the Ab- we have no extant works although some of his scattered quotes can
basid period – with Arabian theoreticians finally addressing Ancient be found in later works –, cf. [Farmer, 1966a, p. 1146], [Farmer,
Greek theories and some of them trying to adapt these theories to 1966b, p. 99], [Maalouf, 2002 (Chapter 2) ; Manik, 1969, p. 22 ;
the musics practiced in the vast countries dominated by the Caliphs. Shiloah, 1981, p. 29 ; Wright, Poché, and Shiloah, 2001, p. 800 (iv)
The craze for these theories (which reminds of the Philhellenic Early theory – written by Wright].)
trend in Europe in the 18th-19th centuries) has perhaps determined
77
A review of the main divisions of the fingerboard of the ʿūd is
a de facto inclusion of Pythagorean ditonism in music practice at the proposed in [Beyhom, 2016, p. 79–80], in the section entitled “The
court of the Caliphate in Baghdad and, by extension and impregna- ʿūd as the ‘Monochord’ of the Arabs”.
tion, in other population segments and other regions of the Arabian
78
Some descriptions – as expounded later – include hand-shifts be-
empire. {Richard Dumbrill reminds here – personal communication yond the (just) fourth, sometimes for theoretical purposes (such as
– that ditonism seems to have been known since the middle of the complementing the second octave of the scale). (See also footnote
first millennium BCE, as is shown in the tablet CBS 1766 dating from no. 418:184.)
the Neo-Babylonian Period historically known as the Chaldean Em- 79
And “singing” (ghināʾ) being ascribed to music practice – see for
pire (626 -539 BCE).} example [Farmer, 2011].
76
Bayt al-Ḥikma (“The House of Wisdom”), in which this large-scale
operation started, was founded by Caliph al-Maʾmūn in 830, when

122
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

was only natural that this instrument became the pre- (786-809), stemmed from the South-Arabian tribe of
ferred tool for theoretical explanations. Kindā (hence the origin of his second surname).84
It is worth noting that even Kindī – the philosopher On the Philosophical and religious front Kindī was
who was probably the most influenced by Plato 80 – an adept of mu[ʿ]tazilism,85 a theological school (and
could not contend himself with the precise, but never- political party) which contributed notably in introduc-
theless arbitrary explanations of the Pythagoreans and ing Greek elements into Islamic thought. He was the pro-
neo-Pythagoreans, and was compelled to include addi- tégé in Baghdād of al-Maʾmūn86 and of al-Muʿtaṣim87 ,
tional positions on the fingerboard of the instrument to then fell in disgrace in 84888. His library was then con-
reflect effective (Zalzalian) praxis. fiscated but was given back to him sometime before his
death. Ehwany underlines one aspect of Kindī’s works
which reconciles Hellenistic legacy with Islam:
“It was due to al-Kindi [Kindī] that philosophy came to be
acknowledged as a part of Islamic culture. The early Arab his-
torians called him ‘the Philosopher of the Arabs’ for this reason.
It is true that he borrowed his ideas from Neo-Platonic Aristo-
An artist’s view of a futuristic guitar.81
telianism, but it is also true that he put those ideas in a new
context. By conciliating Hellenistic heritage with Islam he laid
the foundations of a new philosophy. Indeed, this conciliation
*
* * remained for a long time the chief feature of this philosophy.
Furthermore, al-Kindi, specializing in all the sciences known at
A. First description of the ʿūd and of the divi- his time – of which his writings give sufficient evidence – made
sion of the fingerboard by (al-) Kindī (c. 802- philosophy a comprehensive study embracing all sciences […].
Ibn Nabata, quoting […] al-Kindi, mentions […] the theoreti-
c. 866)82
cal divisions. The philosophical sciences are of three kinds: the
Yūsuf Abū Yūsuf Yaʿqūb ibn Isḥāq ibn a-ṣ-Ṣabbāḥ first in teaching (taʿlīm) is mathematics which is intermediate
ibn Ismāʿīl ibn al-Ashʿath ibn Qays al-Kindī, whose fa- in nature; the second is physics, which is the last in nature; the
ther Isḥāq was the Governor of Kūfā83 under the reign third is theology which is the highest in nature. The priority of
mathematics goes back to Aristotle but the final sequence of
of Abbasid Caliphs al-Mahdī (775-785) and a-r-Rashīd
the three sciences beginning with physics came from the later
80
And probably Ptolemaeos – See footnote 89:124. the birth in Basra as one possibility, concurrently with Kūfā. Kindī
81
From the cover of [Sterling and Bear, 1996]. was also an algebraist in line with Muḥammad ibn Mūsā al-
82
The dates of birth and death of Kindī are taken from [Guettat, Khawārizmī (‫ محدن ح حى او مح ح حح ح ح ح محي‬c. 780 – c. 850). The latter,
2004, p. 116]; these dates are controversial, as Farmer gives for ex- whose name was Latinized as Algoritmi – from which comes “Algo-
ample other dates (see fn. 84 below), [Ehwany (El-), 1966, p. 421] rithm” –, was a mathematician, an astronomer, and a geographer
approximates them as (c. 185/801-c. 260/873) and Yūsuf, in [Kindī during the Abbasid Caliphate, also a scholar in the House of Wis-
(al-), 1962a, p. 6], advocates for the approximate (801-866) as with dom in Baghdad. {See also [Wikipedia Contributors, 2017a] and,
the Encyclopaedia of Islam (see https://referenceworks. more generally on Arabian mathematics and astronomy, [Siddiqi,
brillonline.com/entries/encyclopaedia-of-islam-2/al-kindi-SIM_ 1966] – a domain which, however and according to [Colebrooke,
4380?s.num=0&s.f.s2_parent=s.f.book.encyclopaedia-of-islam- 1817, p. lxxix–lxxx] and [Rosen, 1831, p. ix–x], owes more to In-
2&s.q=al+kindi). [Adamson, 2011] (in the Stanford Encyclopedia dian than to Greek science. One of the questions which is also (and
of Philosophy) states: “We know that al-Kind[ī] died after 866 CE, still) raised today concerns the relation between Indian and Arabian
and his death date is usually placed in the early 870s. His birth date musics at that time (and after), and cross-influence.}
is harder to pin down, but he is said to have served as a scholar 85
See also the re-evaluation of Kindī’s connection with the Mutazi-
under caliph al-Maʾm[ū]n, whose reign ended in 833, and he was lites (adepts of a school of Islamic theology) in [Ivry, 1976], but
certainly associated with the court of the next caliph, al-Muʿta[ṣ]im also [Walzer, 1957, p. 15 sq.].
(reigned 833–842). He is thus usually reckoned to have been born 86
Abbasid Caliph (813-833).
around 800 CE”. {See also [Qifṭī (Ibn al-), Müller, and Lippert, 1903, 87
Abbasid Caliph (833-842).
p. 366–378].} The Fihrist of (ibn a-n-) Nadīm [s.d., p. 315] confirms 88
Under the reign of al-Mutawakkil (847-861) because of a con-
the surname and mentions 7 writings on music by this philosopher. spiracy due to the jealousy of two of the Banū Mūsā, the brothers
{About the importance of (ibn a-n-) Nadīm and his Fihrist see Aḥmad and Muḥammad – according to [Ehwany (El-), 1966,
[Neubauer, 2001a ; Stewart, 2007].} p. 422] citing (ibn abī) Uṣaybiʿa (Ṭabaqāt al-Aṭibbā’, Cairo, Vol. 1,
83
Which is the probable birthplace of Kindī. p. 207 – in fact ʿUyūn al-Anbāʾ fī Ṭabaqāt al-Aṭibbāʾ, with numerous
84
[Ehwany (El-), 1966, p. 421] and [Wright, 2001c]. For [Farmer, editions of which [Uṣaybiʿa (ibn abī), 1882]). (For Muḥammad ibn
1929, p. 127], Kindī would be born in “al-Bāṣra” (Basra – Iraq) Mūsā, see [Hassaan, 2004].)
c. 790 and died in 874. Yūsuf, in [Kindī (al-), 1962b, p. 7], mentions

123
NEMO-Online Vol. 5 No. 9 – November 2020

Peripatetics. Most probably al-Kindi was following Ptolemy, such as cosmology, numerology, Ethos theory and a
who gave a division of sciences in the beginning of Almagest description of the rhythmic system),93
[...]. Mathematics was known to the Arabs from that time as
 clearly and explicitly described the tuning of the
the ‘first study’”.89
strings of the ʿūd in successive fourths,94
It is to be noted, most interestingly, that Kindī was  mentioned some points regarding music practice,95
also a translator and a propagator of Ancient Greek writ-  explained what where the ties used on the neck of
ings from Syriac (and perhaps from Ancient Greek) lan- the ʿūd,96
guage(s).90 Furthermore, he was a theoretician of music  gave a (nearly) complete organological description
and possibly a(n amateur) musician.91 of the latter instrument, including a detailed
Most importantly for us, he wrote a few epistles on description of the material and precise proportions
music – of which four are extant –92 which greatly influ- for the strings,97
enced his successors.  introduced the fifth (theoretical) string of the ʿūd,98
For different reasons – maybe because of his practi-  considered a sixth hypothetical string while
cal discourse on music – his epistles are almost system- explaining the acoustical and organological reasons
atically seen as a kind of auxiliary, a later additional conflicting with this addition,99
documentation to the epistle of (al-) Munajjim (856-912  and, finally, described a practical system for the
– See Appendix C for the epistle of this author). mounting of the ties when applying them onto the
neck of the ʿūd, with an alternative system to ditonic
However, mere chronology shows that the last as-
Pythagoreanism coupled with indirect mentions of
sertion is false. Let us note that Kindī is the first who:
Zalzalian inclusions (cf. infra) to reflect musical
 wrote a series of epistles and treatises on music, practice.100
 integrated some theoretical procedures from
Ancient Greeks (while also integrating other aspects
89
[Ehwany (El-), 1966, p. 424]. Antissa (which Shiloah corrected from “Anusa” or “Anisa” – plead-
90
[Ehwany (El-), 1966, p. 421]. ing for Terpander to be born in the previous) become “Laris” (to
91
[Ehwany (El-), 1966, p. 421] mentions an anecdote in (al-) Qifṭī’s understand as “Larissa”?) and “Anusa” (possibly “Anisa” – today on
“Tārīkh [sic] al-Ḥukamāʾ, Cairo edition, p. 241”, (the corresponding the site of Kültepe in Turkey). (I couldn’t get a hold on the original
– and correct – citation would be Taʾrīkh al-Ḥukamāʾ [Qifṭī (Ibn al-), manuscript – copy? – of Kātib’s Kamāl Adab al-Ghināʾ which is sup-
Müller, and Lippert, 1903, p. 376–377]) relating the healing, by posed to be in Dār al-Kutub in Cairo.) Finally, note in [Kātib, 1973,
Kindī and through music, of a paralyzed boy. A-t-Tīfāshī – a 12th- p. 112] – which is Zakariyyā Yūsuf’s edition – Therpidros” (?) and
13th-Centuries author who wrote on music – describes, in the Faṣl “Arton” in “Laris” and “Anusta” (with question marks by the edi-
al-khiṭāb fī madārik al-ḥawāss al-khams li-ulī al-albāb (Manuscrit 118- tor).}
06 Ennajma Ezzahra – Chapter 6; a printed edition [Tīfāshī, 2019]
92
From a total of about 270 writings which are ascribed to him
is also – recently – available) the hypothetical use by Kindī of music [Ehwany (El-), 1966, p. 422]. There exist differing opinions about
to cure otherwise terminal diseases with patients, while Fārābī the number of his extant works on music (possibly 13), a discussion
would have used a musical instrument to make people laugh, cry which exceeds the needs of the current exposé, but which is detailed
or sleep at will. Such anecdotes can be traced back to the Ancient in fn. 403 in [Beyhom, 2010b, v. 1, p. 122].
Greek sources as found notably in [Grame, 1972, p. 26]: “Plato […] 93
Which are of lesser importance for our purpose.
was described as a brilliant performer who was able, by playing 94
Notably in the Risāla fī-l-Luḥūn wa-n-Nagham [Kindī (al-), 1965]
appropriate music, to affect his auditors so strongly that he could examined infra.
first calm them, then put them to sleep, and finally to awaken them. 95
Which interest us in particular as for their incidence on the scale.
They tell us further that Aristotle, who attempted to emulate Plato 96
See [Beyhom, 2010b].
in this respect, was able to send his listeners to sleep, but unable to 97
See Appendix A, notably FHT 2:158.
awaken them! For this reason, according to the tale, he became the 98
The ḥād (a denomination which was adopted by his successors),
disciple of Plato.” See also in Shiloah’s translation [Kātib (al-), 1972,
which he also named “second zīr” or “lower zīr”.
p. 45–46] (here translated from this French language annotated
version): “It is also well known that Terpander [Terpandros] and
99
cf. infra the Risāla fī-l-Luḥūn wa-n-Nagham.
Arion the musicians delivered the people of Lesbos and Antissa[?]
100
This last point alone explains why early commentators such as
from a plague that fell upon them, with melodies that they devised Farmer ignored this author, since his division could jeopardize the
which relieved [the sick] from this pestilence.” {Note that there is admirable structure elaborated around the “linear” evolution of the
generally much confusion in Greek names in the Early Arabic writ- Arabian scale, originally devoid of Zalzalism – which allegedly
ings (at least those which I have consulted in my research) with – came later to Arabian music. A second reason could be that Kindī’s
for example and in the Arabic language version published by Ḥifnī
and Khashaba in Egypt [Kātib (al-), 1975, p. 23] – Terpander and
Arion becoming “Therpidoros” (?) and “Odeon”, while Lesbos and

124
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

To conclude on Munajjim’s “precedence” in Arabian 1. Kitāb al-Muṣawwitāt al-Watariyya min dhāt al-
music theory: while Munajjim is presented as “reveal- Watar al-Wāḥid ilā dhāt al-ʿAshr[at] Awtār: to
ing” the theory of Isḥāq al-Mawṣilī (767-850) 101 and Caliph al-Muʿtaṣim (833-842).105
considered to be the First Arabian theoretician of music, 2. Risāla fī-l-Luḥūn wa-n-Nagham: to Aḥmad ibn al-
this persistent “mistreating” of Kindī the Philosopher102 Muʿtaṣim (son of al-Muʿtaṣim).106
(which Munajjim was not)103 is totally unjustified and 3. Risāla fī Ajzāʾ Khubariyya fī-l-Mūsīqā: as above.107
chronological exposés on Arabian music theory should 4. Risāla fī Khubr [knowledge] Ṣināʿat a-t-Taʾlīf:
clearly give precedence to Kindī. dedicated to one of Kindī’s late students.108
The four epistles which were undoubtedly written Out of these, two, the Kitāb al-Muṣawwitāt … and the
by this author are, chronologically104 and followed by Risāla fī Ajzāʾ Khubariyya fī-l-Mūsīqā do not relate di-
the name of their dedicatees: rectly to the scale of Arabian music. The two other epis-
tles, the Risāla fī Khubr Ṣināʿat a-t-Taʾlīf and the Risāla fī-
l-Luḥūn wa-n-Nagham, are analyzed below.

master epistle, the Risāla fī-l-Luḥūn wa-n-Nagham, was known in Eu- [in today’s Iraq]. He worked out music in Mosul [Iraq] for one year
rope from an incomplete copy only, until the discovery by Zakar- – hence his surname – then went to Ray [in Iran] to work with
iyyā Yūsuf of a complete copy in India. Javâniyeh, a zoroastrian originating from Abū ʿAlī [a harbour situ-
101
Isḥāq al-Mawṣilī was the most famous singer of the Abbasid era. ated in the Western part of the Persian Gulf – today in Saudi Ara-
His conflicting relation with his nemesis Ibrāhīm al-Mahdī is ex- bia]. Ibrāhīm marries Shāhak, an Iranian, and moves to Baghdad
pounded in [Neubauer, 2001b ; Meynard, 1869], with Neubauer were he becomes the recognized entertainer known to us through
explaining that Isḥāq was “a court musician and companion literature on music of that time. Isḥāq was born in Ray, learned the
(nadīm) under every caliph from Hārūn al-Rashīd (786–809) to al- music trade from his father then afterwards with Zalzal (who – ac-
Mutawakkil (847–61). As an upholder of the classical Arab music cording to Mashhun was also of Persian origin).” In conclusion:
style, he stood in opposition to the innovator Ibrāhīm ibn al-Mahdī knowing that the “Persian wusṭā” (Persian third) of the Early Ara-
and his followers”. See also the New Grove [Wright, Poché, and Shi- bian theoreticians was Pythagorean, and that Zalzal was suppos-
loah, 2001, p. 800] in which Wright comments the relation be- edly the first to have introduced the “Zalzalian” (i.e. “Arabian”, i.e.
tween “Traditionalists” and “Modernists” at that time, notably: non-Pythagorean) positions on the fingerboard, it seems impossible
“[T]he extent to which variation might be either cultivated or – at least today and especially as long as the trend in Oriental mu-
avoided was also coloured by attitudes to tradition, and in parallel sicology is to incorporate all known characters of music in the na-
with the literary debate on the respective merits of the ancients and tionalistic struggle between Arabs and Persians – to disentangle the
moderns, we find advocates of faithful musical transmission op- knot of the relations between Arabian and Persian music(s) at that
posed to innovators. Chief among the latter was Isḥāq al-Mawsilī’s time.}
great rival, the princely amateur Ibrāhīm ibn al-Mahdī (779–839). 102
See footnote no. 76:122. Note however that Owen Wright, alt-
Renowned for the quality and reputed four-octave range of his hough he begins his explanations, in the second edition of the New
voice, he was portrayed as a champion of greater freedom of ex- Grove and for the aforementioned entry [Wright, Poché, and Shi-
pression. The innovations espoused appear to have involved a fur- loah, 2001, p. 800] (article “Arab Music”), with the Tonal system
ther injection of Persian elements, but exactly what these might of Munajjim, specifies a little further: “Of particular importance are
have been is by no means clear, for again we encounter curt indi- several short treatises of the philosopher al-Kindī (c801–c866), the
cations of stylistic contrast rather than analysis. When used in rela- first major theorist whose works are extant”.
tion to Umayyad musicians, the distinction between ‘heavy’ (thaqīl) 103
And neither was he a musician.
and ‘light’ (khafīf) appears to have implied a contrast between a 104
According to Shawqī in [Kindī (al-) and Shawqī, 1996, p. 231].
more complex and serious style and a simpler, gayer one, the for- 105
This is the second epistle in the compendium [Kindī (al-),
mer commanding more prestige, the latter greater popularity. In its 1962a] published by Zakariyyā Yūsuf.
Abbasid manifestation, however, it appears that the lighter, more 106
This is the fifth epistle out of five (of which one – the fourth,
persian style involved an association of freedom of interpretation
probably a translated excerpt from Euclid’s writings – is incorrectly
with greater melodic elaboration, in contrast to the sobriety of the
attributed to him) in the compendium [Kindī (al-), 1962a] pub-
traditionalists”. {Note here that while the relations between the Per-
lished by Zakariyyā Yūsuf, also published as a standalone booklet
sian and Arabian musics have been argumented by many commen-
[Kindī (al-), 1965]. Another version was published by Yūsuf Shawqī
tators, but with no conclusive indications about which one con-
in 1996. (See footnote no. 104.)
sisted in what exactly, Jean During adds to this discussion – in a 107
This is the third epistle in the compendium [Kindī (al-), 1962a]
private conversation on the 9th of September 2020 – the following
published by Zakariyyā Yūsuf.
details which only confirm the “Oriental melting pot” at that time:
“Isḥāq al-Mawṣilī’s (Persian) style would have been sober, in con-
108
This is the first epistle in the compendium [Kindī (al-), 1962a]
trast to his nemesis Ibrāhīm al-Mahdī the (Arabian) style of which published by Zakariyyā Yūsuf. It should be entitled, following the
would have been exuberant. According to Mashhun (as retrieved in MS. British Library Or. 2361, fo 168 (see [Wright, 2006, p. 1, fn. 3]),
in the Târikh-e musiqi-e iran, vol I. 1994, p. 119-121), Ibrāhīm al- Risāla fī Khubr Taʾlīf al-Alḥān.
Mawṣilī [the father of Isḥāq], son of Mahān son of Bahman son of
Pashank originated from the Fars [Persia] and emigrated to Kūfā

125
NEMO-Online Vol. 5 No. 9 – November 2020

THE RISĀLA FĪ KHUBR ṢINĀʿAT A-T-TAʾLĪF109

This epistle is the “typical” of Kindī, not because
of its informative value, but because a copy existed
relatively early in Europe at the British Museum un-
der the reference Or. 2361 (Fig. 7). Its subject is
doubtless about scale theory and melodic composi-
tion [taʾlīf], a theoretical writing par excellence.

The manuscript, a copy dated 1622, is incom-

plete and fraught with errors.110 The description of
the positioning of the fingers on the fingerboard of
the ʿūd is also incomplete. However, it allows for the
reproduction of the possible division(s) as shown in
Fig. 8:127.111

Evidently, different commentators propose dif-

ferent choices among the partitions shown on Fig.
8, where each may be disputable,112 but possible. In
the Risāla fī-l-Luḥūn wa-n-Nagham, Kindī brings a di-
vision resulting from equal-divisions of the string in
opposition, immediately followed, however, by a
theoretical Pythagorean partition as detailed in the
next section.

B. Praxis – or Zalzal versus Pythagoras

The first descriptions of the theoretical sys- First page of the Risāla fī Khubr Ṣināʿat a-t-Taʾlīf by Kindī
(as shown in the compendium [Kindī (al-), 1962a, p. 46]. The
tem(s?) of Arabian music seem strangely familiar to
image is treated for clarity.
researchers in the field of Greek music (or of “some”
While Kindī’s and Munajjim’s divisions of the finger-
Greek music – especially with Pythagorean arith-
board of the ʿūd mostly lack precision, connoisseurs of
metical speculations).
Arabian music may legitimately wonder about the ab-
sence of Zalzalism in these descriptions.
In opposition to these rare descriptions,113 numerous
(later) accounts of praxis exist, notably with chroniclers
such as Masʿūdī and Aṣfahānī, whose narrations remain,
however, mostly anecdotal, or are limited to descrip-
tions of the progress of the performance, or of its socio-
logical dimension.114

109
I base myself for the following on [Kindī (al-), 1962a] and [Kindī 112
Because the choice of some positions and not others will remain
(al-) and Shawqī, 1996] to which must be added [Wright, 2006] arbitrary as long as another, complete copy, is not discovered (if
with a critical evaluation of some of the aspects of the epistle. ever).
110
According to the copyist (as reported in [Kindī (al-), 1962a, 113
Very few other theoretical descriptions are extant, for example
p. 66]), who notes that he copied from a version which is “defective from Khawārizmī, the Ikhwān a-ṣ-Ṣafāʾ, Naṣīr a-d-Dīn a-ṭ-Ṭūsī, but
and unauthenticated” [saqīma wa ghayr muʿtamada]. these are mainly copies of Kindī’s or Munajjim’s divisions. None of
111
I do not give here details of the multiple, sometimes contradict- these later writers – as far as we know – was a musician.
ing interpretations of this division which are explained in [Beyhom, 114
Note that Aṣfahānī mentions rhythms, and “modes” (aṣwāt –
2010b]. sing. ṣawt) which correspond to “courses” (as with Munajjim – See

126
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

Knowing that later authors such as Fārābī and account of the neutral 3rd fret said to have been introduced by
Sīnā,115 who wrote voluminous books (or book chap- Zalzal (d after 842), the ʿūd teacher of Is[ḥ]āq al-Maw[ṣ]ilī117
himself, and named after him (wusṭā zalzal)”.118
ters) on this subject and who had a more respectful atti-
tude towards the “people of the Art” (Ahl a-ṣ-Ṣināʿa in While these descriptions are accepted by most re-
Arabic), included explicit Zalzalism in their theoretical searchers, the reader may imagine my astonishment
description, one cannot help but wonder at the fact that, when I found mentions of practice in the very heart of
as Wright wrote: Early Arabian Pythagoreanism, in the Risāla fī-l-Luḥūn
“Al-Munajjim’s116 neat 2 x 4 scheme probably also tidies up a wa-n-Nagham by Kindī which I examine below.
more complex reality. One evident anomaly is that it takes no

Division of the (stylized) fingerboard of the ʿūd in the Risāla fī Khubr Ṣināʿat a-t-Taʾlīf by Kindī. The “ligatures” (positions of the
dasātīn – “ties” in the figure) are mostly hypothetical due to the missing information in the epistle, but they are all possible. Positions are
identified within the octave by the first 12 abjad (Syriac alphabet) letters a b j d h w z ḥ ṭ y k l. Kindī mentions a series of correspondences
between octaves, fifths and (whole) tones, which explains the presence of numerous alternatives between leimma and apotome in the
division. As a reminder: in the ʿūd with 4 (ranks of) strings, the string which is lowest acoustically (and placed above on the figure for the
instrument performed by a right-hand musician) is called the bamm, then consecutively (and respectively higher acoustically, while placed
lower on the figure) the mathlath, the mathnā and the zīr. The fifth string (ḥād or “low. zīr” in the figure) is theoretical. “Ligatures” (vertical
markers, “ties” in the figure) are generally attributed positions corresponding to the four fingers (excluding the thumb), beginning with
the nut (a), then evidently the sabbāba (b – index), the wusṭā (j – middle finger), the binṣir (d – annular) and the khinṣir (h – auricular).
Notes are thus identified by the string name and by the finger name. This is only practical when the division is limited to four ligatures
(vertical markers): in more complex cases (such as with Fārābī – see « Impracticality of the performance with dense division » in Appendix
B – Also see Fārābī’s division in FHT 17 in [Beyhom, 2016, p. 188] and, at the end of Part II, FHT 28:181 sq. – For “Figure Hors Texte”
or “Plate” 28, p. 181), terms such as “the [first, second] neighbor” (which he named mujannab – pl. mujannabāt) of the sabbāba [or of the
wusṭā]” are used to indicate the position of the vertical marker, although the wusṭā(s) can be differentiated as “Persian wusṭā”, “Arabian
wusṭā”, etc.

Appendix C). He also frequently mentions the wusṭā Zalzal, the so- 116
See Appendix C – This division could be similar to the one by
called “3rd fret” in the quote from Wright below in the text. Kindī reproduced in Fig. 8:127, but more simple.
115
Which belong to the second period, the “Golden Age”. 117
And also his uncle: see [Farmer, 1929, p. 124].
118
Wright in [Wright, Poché, and Shiloah, 2001, p. 802].

127
NEMO-Online Vol. 5 No. 9 – November 2020

The reader may also wonder, as I wondered for Note that Kindī mentions three different tunings for
some years after this discovery, how and why these in- the strings, the first (and most used) being successive
dications by the leading theoretician at that time had just fourths,123 while the two other tunings are varia-
been dismissed by successive generations of Orientalist tions with different resulting notes for the (unstopped)
musicologists,119 and this for more than two centuries. lower (acoustically) string (the bamm)124 in order to un-
As for local musicians (or musicologists), the reason derline particular tonic notes125.
for not questioning Orientalist writings is evident: the
“Harmonic” and Pythagorean divisions
grand names of the “science” of musicology (Western
and local) having spoken,120 it becomes difficult to bring As mentioned in Appendix A, in this epistle Kindī
their “findings” to the test… provides the dimensions of the ʿūd in “full fingers”126
(“ff” from this point on), a unit roughly corresponding
THE RISĀLA FĪ-L-LUḤŪN WA-N-NAGHAM BY KINDĪ to 2 cm (today). The vibrating string is 30 ff long with
10 ff (which is the third of the total vibrating length –
The contents of the epistle (Risāla) fī-l-Luḥūn wa-n-
see Fig. 9:131) over the fingerboard until the junction of
Nagham are described in details in Appendix A.3 of my
the neck with the soundboard and the body. Tie-frets
(first) book on Arabian music.121 The description of the
must not be mounted further as the fourth of the total
proportions of the ʿūd, the first one known to us, is re-
speaking length (from the nut)127 and are placed at the
produced in Appendix A in the current dossier, while
successive distances of 3 ff, 2 ff, 1 ff and 1½ ff, forming
the description of the “tie-frets” on the neck is included
(with the strings) a “harmonic” division of the finger-
in Part II.
board (cf. Fig. 9). The reason for this positioning, which
The epistle, which is written as a teaching manual for
is a little far from the simple Pythagorean ditonic posi-
the ʿūd, has many useful indications on the practical as-
tioning, is given as practical, the author justifying his
pects of the fabrication of the ʿūd and performance. When
point by the necessity to use superparticular ratios (in
compared to the Risāla fī Khubr Ṣināʿat a-t-Taʾlīf reviewed
the form [n + 1]/n) beginning with the “tenth of the
above, it has a decisive advantage as it is complete, well
string” and ending with its “half”.128
written and is vocalized which helps understanding
The actual, physical ties must be “firmly tied at the
words or phrases that would be otherwise unclear. Its im-
back of the neck to avoid the possibility, due to their
portance is crucial for our research as it contains the first
tension, of lateral displacement”129. While this is an in-
description122 of actual “tie-frets” on the neck of the ʿūd,
dication of Kindī’s practical concerns, tying the “frets”
with precise and detailed explanations about their
firmly is, however, premature, as further equivalences
mounting and proportions.

119
And by their Arab (Persian, Turkic, etc.) students: these and their 125
There is a contemporary example with maqām Sīkā where the
masters are so many that I do not bother mentioning them, alt- acoustically lowest string is frequently retuned to the note SĪKĀ
hough their writings are expounded in [Beyhom, 2010b]. (𝑒 − ).
120
The problem with local musicologists is clearly a re-Orientalist 126
This is a conventional reduction of the literal translation of
matter: if they have recourse to Western musicological “science”, (“‫)”أصابع منتلئة حسنة للحم‬: “full fingers with good flesh”.
they are compelled to learn under the supervision of Western mu- 127
[Kindī (al-), 1965, p. 11].
sicologists and as a result of these studies use the same biased tools 128
The half of 1 ff, which the author seems to consider as a lower
Orientalist musicologists use for their descriptions of Arabian mu-
limit in this measuring method, prevents him from using some ra-
sic; mostly, however, they are so overwhelmed by this “science”
tios such as the 1/9 of the string ([30 ff]/9=3,3333… ff), the eighth
that they take the statements of their mentors for granted.
([30 ff]/8=3,75 ff) and the seventh ([30 ff]/7=4,2857… ff), while
121
[Beyhom, 2010b]: other organological aspects, such as the the sixth (5 ff – position of the wusṭā), the fifth (6 ff – position binṣir),
thickness and material of the strings, are detailed in this reference, the fourth (7½ ff – position of the khinṣir), the third (when extrap-
with the corresponding texts reproduced (in Arabic) in Appendix D. olating, to get the fifth – 10 ff, which equates to the vibrating length
122
And the only one, if not for a-ṭ-Ṭaḥḥān’s description explained over the fingerboard), including the half (octave at 15 ff) of the
below. string are all compatible with this method (this line of reasoning is
123
[Kindī (al-), 1965, p. 16–17]: this is also the tuning used for the- deduced from [Kindī (al-), 1965, p. 13]).
oretical descriptions (scale system). 129
[Kindī (al-), 1965, p. 12].
124
[Kindī (al-), 1965, p. 17–18]. In today’s practice, and depending
on which mode is used, the bamm can also be tuned differently to
suit the needs of the performer. (See next footnote.)

128
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

between octaves and fifths mentioned by the author some explanations about praxis, with regard notably the
compel to reconsider this initial division.130 accompaniment of singers, and provides in the last part
Equivalences of octaves mentioned by the author re- an exercise in form of tablature.133 While explanations
sult in a modified placement of the tie-frets, shown on about singing practice are given, seemingly, with reluc-
Fig. 10:131. The modifications make the first measure- tance, the information about the “additional notes [used
ment procedure obsolete, as the new positions do not by singers] outside the tie-frets [dasātīn]” stands:
comply with a superparticular division of the string. “It may be that singers use also a note [naghma] which lies out-
The result of further equivalences between notes a side of all the ligatures, that they name maḥṣūra [“compressed,
limited”]. It lies outside of the ligature [dastān] of the khinṣir by
fifth apart (Fig. 11:132) is similar and implies a Pythag-
extending the auricular [khinṣir], and behind this one also – at
orean division of the fingerboard (practically)131 equiv- the same distance as the ligature of the khinṣir – except that
alent to the division expounded in the Risāla fī Khubr they move the sabbāba [index] to the ligature of the wusṭā
Ṣināʿat a-t-Taʾlīf reviewed above (Fig. 8:127). [middle finger] or of the binṣir [annular]”.134
This result is not compatible with the initial (“Har- A thorough review of this quote shows that the au-
monic”) description, although it could be hypothesized thor gives in fact indications for three new ligatures, or
that the differences of one comma between different po- series of notes a fourth apart, the placement of which
sitions of the ligatures would be considered as insignifi- can be deduced in two steps.
cant by Kindī.132
1st additional ligature (series of notes): “It may be that sing-
However, the mere fact that Kindī explores octave ers use also a note [naghma] which lies outside of all the liga-
and fifth correspondences compels us to consider both tures, that they name maḥṣūra [“compressed, limited”]. It lies
divisions, “Harmonic” and Pythagorean, as possible. outside of the ligature [dastān] of the khinṣir by extending the
Consequently, the placement of the “additional notes for auricular [khinṣir]”.135

singers” explained below is undertaken for both divi- Kindī mentions no string for the maḥṣūra: he consid-
sions. ers hence, by default, that it applies to the four strings
of the ʿūd: this is indeed an additional ligature.136 As for
Additional notes used by singers its positioning on the fingerboard, and knowing that the
At some point in his epistle after the description of author does not mention a hand-shift for it – neither
the division of the fingerboard of the ʿūd, Kindī adds

130
If we follow the author’s indications, “frets” will have to be tied 134
[Kindī (al-), 1965, p. 19]. These indications are remarkable as
and untied repeatedly, which is unpractical. they underline a crucial difference between the practical modal sys-
131
The division in the Risāla fī Khubr Ṣināʿat a-t-Taʾlīf can be inter- tem (by singers) and the theoretical system inspired by Ancient
preted in multiple ways as explained in Part A. Greek theories. It suffices to remind the reader that Kindī consid-
132
This would be plausible knowing the general propensity of the ered music, as the Pythagoreans did, as a science: “The soul has an
author to mimic Plato by despising “mixtures” of notes. Thus, in affinity with music – That is [the science of] the composition of
[Kindī (al-), 1965, p. 19] the philosopher “cites” (pseudo)-Plato melodies”, which is a quote from Plato in [Kindī (al-), 1965, p. 19]:
who would have complained about the “aimless and endless tarākīb “‫” لنفس تنكفي مع مل سحقى – أي تألقف أل حان‬. This is possibly the first
[combinations] of mixed notes”. Compare to: “Then, I said, if these differentiation between ghināʾ [“singing”] and mūsīqā [“music”],
[Dorian and Phrygian harmonies] and these only are to be used in the criteria differentiating praxis (“singing) and theory (“composi-
our songs and melodies, we shall not want multiplicity of notes or tion on the instrument”) being in this epistle clearly stated.
a panharmonic scale? / I suppose not. / Then we shall not maintain 135
Excerpt from the previous quote from Kindī.
the artificers of lyres with three corners and complex scales, or the 136
Virtual, evidently and for many reasons which will become clear
makers of any other many-stringed curiously harmonised instru- in Part II (where the process of the mounting of the ties in the Risāla
ments? / Certainly not” – in [Plato, 1908, p. 399 C-D (Book III)]. fī-l-Luḥūn wa-n-Nagham is reviewed). Let us note for the time being
While Kindī chooses here austerity (some would write “simplicity”) that these additional “ties” would create, if they were to be materi-
in music and endorses Plato’s complaints about praxis, the realities alized as “frets”, incommensurable problems within this process,
of music both in Ancient Greece and in the countries of the Arabian which is already very debatable. Further, and while Kindī wishes in
Empire at his time seem to be far different from the (too) simple no way to step out of the framework of Pythagoreanism in his the-
Pythagorean scheme he adopts in theory. oretical expounding of the “Arabian” system, this is another reason
133
Besides being the first known literal (using intersections of for him not to mention an exact position for the maḥṣūra, a position
strings and tie-frets – or tablature) notation of Arabian music, this that he cannot quantify by giving a Pythagorean ratio for its inter-
musical exercise features two simultaneous and differentiated me- val, or that he simply did not bother to examine more thoroughly
lodic lines. (as this is praxis, not theory). Only with Fārābī and (ibn) Sīnā would

129
NEMO-Online Vol. 5 No. 9 – November 2020

does he give its precise position – it should reasonably If we sequence Kindī’s proposals, we can determine
be situated somewhere between the just fourth (khinṣir) that these two series of notes can be found:
and the just fifth, which could complement octave equiv- 1. After the maḥṣūra,
alences missing in his division (see Fig. 12:132).137 2. at a distance which is equivalent to the distance
However, the use of the term maḥṣūra (“compressed, between the sabbāba and the khinṣir,
limited”) by the author compels us to consider other op- 3. with this distance being measured
tions that will become clear when we place the remain-  from the wusṭā,
ing “additional notes”.  or from the binṣir.
As for the other two positions, we shall note that The first term of the sequence above is clear, show-
moving the sabbāba (the finger) towards the position of ing that two additional series of notes are to be placed
the wusṭā or of the binṣir (the ligatures) corresponds to a between the maḥṣūra and the bridge (Fig. 12:132).
hand-shift, or lateral displacement of the (left, for right- The second term must be understood as a distance,
handed performers) hand towards the bridge in order to because if it must be the interval between the sabbāba
reach locations for the fingers which cannot be reached and the khinṣir, a Pythagorean “augmented second”,
using the traditional hand position (Fig. 13:134).138 such “new” notes could be found on the lower string ei-
The reason for this positioning outside the range of ther on the estimated position of the maḥṣūra (if the in-
the ligatures (of the fourth) is that while the description itial starting point is the wusṭā), or on the sabbāba (if the
of their mounting shows that they are material ties initial starting point is the binṣir – Fig. 12:132). Assum-
made of gut, positioning the finger between two (con- ing Kindī had in mind octave correspondences, the sec-
secutive) tie-frets is unfunctional139. In order to maintain ond series of notes would be superfluous, because it
the consistency of his demonstration, Kindī had to posi- would already be delimitated by an existing tie-fret. We
tion these additional notes outside of the fretted zone, after are dealing with distances with this second term.
the khinṣir (and towards the bridge).
As for the remaining (two) series of notes:
2nd and 3rd additional ligatures (series of notes): “and be-
hind this one also – at the same distance as the tie-fret of the
khinṣir – except that they move the sabbāba [index] to the tie-
fret of the wusṭā [middle finger] or of the binṣir [annular]”.

these “additional” positions be given “rational” quantification. (For p. 352], where images 2 and 4 show the left hand position de-
the latter authors, please refer to Chapter II in [Beyhom, 2010b] for scribed by (ibn) Sīnā, while images 1 and 3 show the left hand in
a complete review.) As we shall see in Part II, this mounting proce- shift position. Ligatures a and b are reached by moving the sabbāba
dure is adapted only for the final aim of this epistle, teaching the (the – index – finger) towards the position of the wusṭā or of the
rudiments of the technique on the ʿūd. binṣir (the ties), which corresponds to a hand-shift, or lateral dis-
137
The position of the maḥṣūra cannot exceed the fifth as Kindī ex- placement of the (left, for right-handed performers) hand towards
plains that one of the other additional notes is positioned behind the the bridge in order to reach locations for the fingers which cannot
maḥṣūra (further towards the bridge), and before the fifth. be reached using the traditional hand position. (Reminder and com-
138
Note that hand shifting is – relatively – seldom mentioned in plement:) Hand shifting means moving the thumb towards the
early Arabian writings, whenever today this has become a standard wusṭā or further, in which way further positions for stopping the
procedure in ʿūd technique as it has been for centuries for the Euro- strings (and further towards the bridge) can be reached by the other
pean lute. Note also in [Sīnā (Ibn) or Avicenna (980?-1037), 1956, fingers, mostly the auricular for the further positions towards the
p. 47–48]: “When it came to the insertion of melodic intervals […] bridge – see also http://www.lutesociety.org/pages/beginners-les-
only three were inserted within the fourth […]. The reason is the son-3, notably:
absolute necessity to appreciate the location of the fingers for the “The easiest and most efficient way to achieve [hand-shifts] is to
stopping of the strings on the ligatures [with Sīnā these are vertical simply pull the whole hand and forearm towards you to shift up
markers as shown in Part II]. There was a difficulty for moving the (towards the bridge), and to push the hand and forearm away from
hand at the same time as moving the fingers. It was then agreed to you to shift down (towards the nut). Be careful not to twist the hand
keep the hand in a fixed position and to move the fingers only. The during shifts; common faults include moving the fingers but leaving
optimal position allowing for this movement was reached within the thumb behind, leaving the wrist sticking out awkwardly after
the fourth of the string, on which was mounted the khinṣir. With the the shift, and making excessive movements of the upper arm which
thumb holding the instrument, the four [other] fingers could move leave the elbow sticking out”. (See also Fig. 13:134.)
within this fourth [of the string]”. See for example [Spencer, 1975, 139
This is further explained in Part II.C.

130
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

Division of the fingerboard of the ʿūd in Kindī’s epistle (Risāla) fī-l-Luḥūn wa-n-Nagham, in “full fingers” (= “ff” in the figure); the
note corresponding to the non-stopped mathnā (g2 in the figure) is called the yatīma140. In the literal notation in the figure the g2 is taken
as the central note; the diesis “#” raises the note by one approximate apotome while the “b” lowers the note by the same amount; “+” and
“-” signs are “corrections” with approximate value equal to one comma (or 1/8th to 1/9th of the tone) respectively higher or lower. Note
that the ḥād string is hypothetical: Kindī further considers (see [Kindī (al-), 1965, p. 21]) adding one string still below the ḥād for the sake
of demonstration of octave equivalences. (“Vibrating string” in the figure = speaking length of the string.)

Octave equivalences (in double-arrowed curves) as expounded in Kindī’s Risāla fī-l-Luḥūn wa-n-Nagham, with sequential num-
bering (Arabic numbers) and consequent modifications (Roman numbers) of the positions of the tie-frets. The resulting system is Pythag-
orean ascending (two whole-tones and one half-tone from the nut to the just fourth) then descending one tone (from the Khinṣir to the
Wusṭā).

140
The (she) “orphan”: so called because it does not have in practice a corresponding (higher or lower) octave, except for the approximate
g3- on the hypothetical ḥād string (Fig. 9 and Fig. 10).

131
NEMO-Online Vol. 5 No. 9 – November 2020

Positioning of the left hand on contemporary ʿūd(s) as shown in the opening pages of [Rūḥānā and ‫روحانا‬, 2001]. From top left
to right then bottom left to bottom right: 1st position, 2nd position, 1st “half” position and 4th position. The two positions to the left are for
traditional performance. The two positions to the right correspond to hand-shifts.

Calculating the position of ligature a and including (then fitting) it in Kindī’s Pythagorean division in the Risāla fī-l-Luḥūn wa-n-
Nagham.146 (“Tie” in the figure = “ligature”.)

146
The calculation of the distance between a and the nut was made by calculating the length of the string part which results from the
addition of the equivalent of a fourth to section Lc-a = 203L0/288 (by subtracting a fourth from the interval delineated by ligature a): it
suffices then to multiply 203L0/288 by 4/3 (being 812L0/864), and to simplify the result by dividing both numerator and denominator by
4, which gives 203L0/216.

134
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

Positioning ligature b with inclusion of the three new “tie-frets” (a’, b’ and maḥṣūra) within the Pythagorean division of Kindī in
the Risāla fī-l-Luḥūn wa-n-Nagham. (“Tie” in the figure = “ligature”.)

“Harmonic” division of the fingerboard of the ʿūd with Kindī’s indications on the “supplementary notes”, an alternative (based
on praxis) division described by Kindī and totally overlooked by Orientalist (and re-Orientalist) musicologists – iṣbaʿ (pl. asābiʿ, dual
iṣbaʿayn) means “finger(s)”, here the “full fingers” (“ff”) of Kindī.

135
NEMO-Online Vol. 5 No. 9 – November 2020

PART II. ON “FRETS” AND “TIES” ON THE NECK A. Different opinions about the “fretting” of
OF THE ʿŪD the ʿūd
Although many Western “specialists” in Arabian
In this second part we shall examine one of the most
music state(d) that the Early Arabian ʿūd was
debated questions in the history of (Arabian) music, the
fretted, this question was never really settled. 149 In
alleged fretting of the Early ʿūd.147 Opinions have been
1969, Liberty Manik reviewed the arguments brought
historically far apart on this subject, and changing. We
forward against this thesis:
have also seen in the first part that theoretical divisions
“With regard the tie-frets150 of the ʿūd, which theoreticians of
of the fingerboard, although described as “purely” Py-
the Middle Ages151 have described in fine details to explain
thagorean in the literature are, when it comes to facts their [musical] system, Berner actually argues that these tie-
(to praxis), approximate and Zalzalian. The problem frets never existed as this is, as he literally says, ‘pure fiction’.152
that arises is that, given material “frets” as those advo- Berner refers for this to Geiringer who, after having determined
cated by Farmer and others, Zalzalian (maqām) music that a lute with ties was not to be found in the iconographical
as described in the Kitāb al-Aghānī148 and many other context at that time, came to the conclusion that tie-frets were
only used with the aim of measurement and research, and that
sources would be impossible: how can then this contra-
these tie-frets could have no significance at all in [musical]
diction be resolved?
practice.153 Before that, Curt Sachs came to a similar conclu-
I provide here a foretaste of the final answer pro- sion”.154
posed in this dossier: there is no solution to this prob-
Manik takes sides against these opinions by arguing
lem, unless early ʿūd(s) were never fretted – except for
that a lack of proof (corroborating the mounting of tie-
teaching or theoretical purposes.
frets on the neck of the ʿūd) is not a proof of the absence
of tie-frets because (1) an image (or a sketch) is not a
* photography and (2) lack of proof per se is not
* *
enough.155
Part II is divided in two main sections: To defend his thesis, Manik cites in a footnote
A. A review of different opinions about the “fret- Farmer’s article “Was the Arabian Lute fretted?”156 to-
ting” thesis gether with Lachmann’s157 and settles for the authority
B. A historical clarification, and conclusions of these authors to conclude that “frets” were indeed
mounted on ʿūd(s) during the Golden Age158 (the West-
ern Middle Ages according to Manik).

147
For the influence of the ʿūd on Arabian music theory and praxis, 154
[Manik, 1969, p. 12]. Note that Manik’s reference for the last
see the entry for the instrument [Chabrier, 1982] in the Larousse de assertion corresponds to [Sachs, 1940, p. 254]: “Lutes seem to have
la musique, notably: “With the Abbasid Caliphs in Iraq, [the ʿūd] no frets, either in older times or today, in spite of the constant use
becomes the conceptor of genē and modes of Meso-Islamic musics by the theorists of the word dāsatīn [sic – ‘dasātīn’], plural of Persian
and the creator of melodies, a role maintained till today in both dast or ‘hand’, which is used to indicate frets. And it would have
Arabian Popular and Art Musics”. been difficult to string them securely around the sloping end of a
148
See footnotes nos. 424 and 425:184. pear-shaped lute [see Appendix B: Organological Clarification].
149
More precisely not before the publishing of my book [Beyhom, Very probably, the frets existed only theoretically to symbolize the
2010b] (preceded by [Beyhom and Makhlouf, 2009]) in which this positions of the stopping fingers”.
question was comprehensively examined and in which I concluded
155
Such arguments are totally acceptable per se, but the proof of the
that (tie-)frets were not used unless for teaching or for theoretical contrary was not provided either, as I explain further.
purposes. To this day, I have not read, or heard of, a refutation of 156
[Farmer, 1937, p. 458].
my demonstration. 157
Reference to Lachmann, Robert, 1934, “Die Vīnā und das in-
150
Manik uses the term Lautenbünde (“Lute-frets”) in German – see dische Tonsystem bei Bharata”, Zeitschrift für vergleichende Musikwis-
also the original text in Appendix D. senchaft, II, s. 64 (which I could not find).
151
Once again: whose “Middle Ages”, and is such a time period per- 158
Manik’s argumentation is reduced to showing the inexistence of
tinent even for Western History? proofs for the presence of tie-frets on the ʿūd, stating that the lack
152
Manik cites here [Berner, 1937, p. 19]. of evidence in the iconography does not confirm their absence,
153
Manik cites here likewise K. Geiringer’s “Vorgeschichte und Ges- which he uses as an argument in favour of the “fretting” thesis. This
chichte der europäischen Laute bis zum Beginn der Neuzeit”, ZMw, is an arbitrary procedure as it is in every science. Furthermore, he
x (1927–28), p. 570 (which I could not find).

136
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

The thesis of the “fretting” of the ʿūd is, conse- B. Historical clarification
quently, based on already “old” (at the time of Manik)
arguments from Farmer, Lachmann having simply fol- FARMER’S “WAS THE ARABIAN LUTE FRETTED?” REVISITED
lowed Farmer in his argumentation as explained next. We saw that Liberty Manik, the 1969 author of a
However, and due to the fact that Farmer’s writings re- concise treatise on the theories of the scale of Early Ar-
main an easily accessible reference for researchers,159 abs, supports the thesis of the “fretting” of the ʿūd at that
this interpretation resurfaces regularly in the musicolog- time, referring in so doing to Farmer. Returning to
ical literature160 as explained in the introduction to this Farmer’s article, 163 we find from the outset Manik’s
dossier for the New Grove and the Encyclopedia of Islam, problematic stated in the introduction, in which Farmer
to which should be added this statement by Farmer in cites an encounter with Lachman who (according to
his History of Arabian Music…: Farmer), was influenced by Geiringer’s164 opinion about
“The ʿūd qadīm or classical lute of four strings still continued to the absence of frets on the early ʿūd(s), and asked
be favored,161 in spite of the introduction of the ʿūd kāmil or Farmer if proofs of such fretting(s) existed.165 A few lines
perfect lute of five strings, which was fretted according to the further down Farmer cites Curt Sachs and his standpoint
‘systematist’ scale”.162
against the fretting thesis, while revealing concurrently
In his book, as in the Encyclopedia of Islam to which Arabian musicologists’ dependence on Western musico-
he contributed, Farmer considered the fretting of early logical science:
– and less early – ʿūd(s) as fact. However, there are many “In 1932, whilst I was at Cairo at the Congress of Arabian Mu-
indications which contradict Farmer and other propo- sic, the question arose officially. At the plenary session of the
nents of the “fretting” thesis and confirm Sachs’, Commission of Musical Instruments the well-known Egyptian
Berner’s and Geiringer’s opinion, especially for the “Sys- musicologist Ahmad Amīn al-Dīk Efendi […] suggested that
frets should be adopted on the modern Egyptian lute (ʿūd) as in
tematist” scale as Farmer wrote.
days of old. Dr. Curt Sachs, […], who was President of the
Commission, replied that the Arabian lute in days of old was
not fretted. Several Egyptian savants and musicians questioned
me privately at the time about Dr. Sachs’s statement. […] I
promised that I would deal with the question at length not only
*
* * for its own sake, but in defence of my own thesis that Europe
was influenced by the introduction of musical instruments with
frets during the early Arabian culture contact”.166

Having stressed, at the end of his introduction, his

own intention to prove the existence of frets,167 and hav-
ing explained the issue and his own agenda, Farmer
refers to Farmer and Lachmann without quoting them, and con- article “Der Bau der Laute und Ihre Besaitung…” [Neubauer, 1993]
cludes in favour of the “fretting” thesis which is, to say the least, a which is examined further.
flawed argumentation. The reader may compare this argumenta- 161
Farmer states here in a footnote: “It was still in use in the
tion to Raasted’s conclusion on the ditonicity of the Byzantine chant 15th century. Bodleian MS., Marsh, 282, fol. 77”.
“of the origins” in the section entitled The “unambiguous supporting 162
[Farmer, 1929, p. 208–209].
testimony” for the ditonicity of “Medieval” Byzantine chant in 163
[Farmer, 1937].
[Beyhom, 2016, p. 235–236]. 164
Cited by Farmer in the same page mentioned by Manik.
159
Farmer contributed to many articles, for example, in the Ency- 165
[Farmer, 1937, p. 453]: later, Lachman changed his mind (ac-
clopedia of Islam.
cording to Farmer) and distanced himself from Geiringer on this
160
For example Bouterse’s article, entitled “Reconstructing the Me- matter.
dieval Arabic Lute: A Reconsideration of Farmer’s ‘Structure of the 166
[Farmer, 1937, p. 453–454].
Arabic and Persian Lute’” [Bouterse, 1979], and his critique of 167
To defend his thesis about the introduction of fretted lutes in
Farmer’s description of Early ʿūd(s). In spite of his critical attitude
Europe at the time of the first interactions with the Arabs as quoted
towards Farmer, Bouterse adopts the latter’s opinion: “As Sachs has
above. The purpose of this demonstration was the justification of a
pointed out, {citing [Sachs, 1940, p. 254]} frets would have been
wider thesis by Farmer on the influence of Arabian music on Euro-
difficult to tie on the sloping neck, but […] Farmer has conclusively
pean music, advocated in [Farmer, 1930] with, notably in pages
proved that the Arabs did use frets on their lutes in the Middle Ages
104, 108, 112 and 363-364, the claim that Arabian music intro-
{citing ‘Was the Arabian and Persian Lute Fretted?’ examined fur-
duced harmony in Europe. The “fretting” thesis allows for steady
ther}” – [Bouterse, 1979, p. 2–3]. More recently, the advocates of
the “fretting” had the strong support of Eckhard Neubauer in his

137
NEMO-Online Vol. 5 No. 9 – November 2020

starts168 by analyzing the term dastān used in the trea- The second main question which comes is
tises for the “tie-frets”. He first quotes (al-) Khawārizmī’s mention of the stopping of the string on
Khawārizmī169 in his Mafātīḥ al-ʿUlūm:170 the “tied places” which, as I show further, is incompati-
Quote 1: “dasātīn171 are the tied places (ribāṭāt) upon which ble with physical frets.176
the fingers are placed”.172 In the second part of his article,177 Farmer informed
Farmer immediately concludes: the reader that “Frets (dasātīn) are frequently mentioned
in the Kitāb al-Aghānī” by Aṣfahānī,178 and further refers
“This definition is in itself quite sufficient to settle the question
at issue. These ‘tied places’ were made by means of gut or string to “the Arabic theorists”:
tied around the neck of the instrument”. “their treatises prove conclusively that the lute (ʿūd) as well
as the pandore (ṭunbūr) had these frets or dasātīn tied around
Many questions arise here. Firstly, why would these
the neck of the instrument”.179
“tied places”173, if these are tie-frets, need to be made
“by means of” gut or string? Why are these tie-frets not He also adds that Kindī, speaking of the dasātīn of
simply “made of” any material, be it gut or anything the ʿūd in one of his epistles on music,180 “shows […]
else? Would Farmer have only given a reference for this that they must have been frets”.181
material from which the tie-frets were (allegedly) made, We must note here, firstly, that the ṭunbūr and the
then we would not have to ask the question. By return- ʿūd are different instruments, 182 this difference lying
ing to the original source,174 we find that there is no sin- mainly in the (relative) length183 of the neck184 but also
gle reference by Khawārizmī to the material from which in the playing techniques which, frequently, stem from
the dasātīn (“ligatures”?) were made, and that the page the fretting, or from the non-fretting of these instru-
cited175 by Farmer contains only the description of the ments.185 Secondly, I have shown elsewhere186 – and ex-
tuning of the ʿūd and the positions of the dasātīn. plain further in the following pages – that the ṭunbūr

intervals on the ʿūd which is in favour of polyphony (in Western 177

The remaining section of the first part of the article (§1) consists
mainstream thought); the ditonic thesis allows further for harmony in a digression by Farmer on the terms ʿataba and dastān which is
in the Arabian Empire as early as Kindī (his “exercise for the ʿūd” in not relevant to our discussion (more information on this digression
the Risāla fī-l-Luḥūn wa-n-Nagham is explained by Farmer in p. 104 in [Beyhom, 2010b]).
of the aforementioned reference). 178
Which, in itself, does not inform us on the material from which
168
[Farmer, 1937, p. 454]. these dasātīn (Farmer calls them “frets”) were made.
169
Khawārizmī’s encyclopedia [Khawārizmī (al-), 1895] Mafātīḥ al- 179
[Farmer, 1937, p. 456]. (Bold type in quotes is mine, unless oth-
ʿUlūm [The keys to sciences] – edited by Van Vloten and which is erwise stated.)
most probably Farmer’s reference, contains 12 pages [235-246] ex- 180
“British Museum MS. Or. 2361, fo 165vo”, see [Kindī (al-),
pounding music, in Chapter VII of Book II. Guettat [Guettat, 2004, 1962a, p. 51–53].
p. 115] cites an 1978 edition – I rely here on an edition from Beirut 181
[Farmer, 1937, p. 456].
[Khawārizmī (al-), 1991], and on Van Vloten’s for confirmation. 182
By equating ʿūd and ṭunbūr, and knowing that long-necked lutes are
170
Further quotes of Arabian sources by Farmer and other authors frequently fretted (see next footnotes), Farmer tries to reinforce his the-
are numbered from this point on. sis for which he has, in fact, no solid arguments as shown further.
171
(Reminder:) Plural of dastān. 183
The length is relative to the soundboard: short-necked lutes have
172
[Farmer, 1937, p. 454]. [Khawārizmī (al-), 1991, p. 210] has: (roughly) a neck which is shorter than half of the speaking length
.“‫”و حىها َدستان والدساتين هي لرباطات لتي ت ضع ألصابع عليها‬ of the string, while long-necked lutes have a neck which is longer
than half of the speaking length of the string.
173
This is effectively the first question as the terms “tied places” are
a circumvolution for the word “ligature” (in French also “ligature”,
184
The ʿūd is short-necked while the ṭunbūr is long-necked.
in German “Bund”) which is the correct translation of the Arabic
185
Long-necked lutes are mostly fretted (with exceptions, notably
ribāṭ (sing. of ribāṭāt). for some African long-necked lutes – see for example [Charry, 1996,
174
[Khawārizmī (al-), 1991, p. 207–214]. p. 5–6]) while short-necked lutes are not fretted – with one notable
exception, the European lute. In the latter case, however, it may be
175
[p. 210] in [Khawārizmī (al-), 1991].
argued that frets (in fact “tie-frets”) were mounted to make the per-
176
Unless these “frets” are so thin (for example made of one or few silk
formance easier, notably for polyphonic purposes or to make it eas-
strands the sole purpose of which being to materialize the positioning
ier for non-professionals – for example nobles or high-classes repre-
of the fingers, which means their purpose is indicative (showing the
sentatives who wished to play easily on the instruments.
stopping points) and not effective (stopping the strings). While this prob- 186
See Appendix A.2 in [Beyhom, 2010b] for the descriptions of
lematic is explained further, it is worth mentioning that these tied
the ṭunbūr by Fārābī and Kātib.
threads would hardly hold on to the neck of the ʿūd (even less than ties
made of gut, as explained in Appendix B).

138
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

(here “of Baghdād”) might as well be considered as hav- Quote 4: “the dastān next to the nut (anf) was to be placed
ing no frets at all. (mawḍūʿ) on the fingerboard at one-ninth of the vibrating
string-length”.195
Further: when returning to Kindī’s manuscript at the
page cited by Farmer (fo 165vo), we find absolutely no Masʿūdī, however, wrote in the Arabic version:
indication that the dasātīn may have been “frets” (or tie- Quote 5: “wa-d-dastabān 196 al-ladhī yalī al-anf mawdūʿ ʿalā
frets). Kindī discusses in fact, in this folio (recto-verso), khaṭṭ a-t-tusuʿ min jumlat al-watar”,197
the positioning and the locations187 of the anghām (notes) which can be translated as:
on the fingerboard of the ʿūd, and mentions three times
Quote 6: “the dastān which [immediately] follows the nut is
a dastān which coincides with one or another of these positioned on the line of the ninth of the whole string”.
locations, while failing to inform us if these dasātīn, that
he had not previously defined,188 do exist physically.189 While still no material for the “frets” is mentioned
by any of the authors quoted or mentioned by Farmer,
Farmer continues his argumentation citing Munaj-
he asserts further, without providing references to the
jim who explains:
190

reader, that the Ikhwān a-ṣ-Ṣafāʾ, (ibn) Sīnā, (ibn)

Quote 2: “the place of every note (naghma) upon every fret
Zayla198 and Ṣafiyy-a-d-Dīn al-Urmawī and others,
(dastān)”,191
Quote 7: “all confirm the view that dasātīn were gut or string
then Fārābī who, while describing the ʿūd, would have frets tied on the neck of the lute”,199
written192:
which is absolutely unsubstantiated, as we shall see, and
Quote 3: “that the dasātīn (frets) were tied (shadda) on the
while still no precise references are provided describing
neck (mu[s]tadaqq) of the instrument, and that they were fixed
parallel with the bridge-tailpiece”,193
the material (except for Farmer’s own assertions) and no
indications about this material can be found in his arti-
which corresponds to the original.194 cle until this point.
It seems here that, for the first time, Farmer’s asser- As there is no other way to be sure of the material
tions may come true. Still, we do not know what was existence of “frets” on the neck of the ʿūd, we must
the material used for the making of these “frets”… ex- therefore try to find them in the works of the four au-
cept for Farmer’s unsubstantiated statement at the be- thors mentioned by Farmer, but not referenced by him.
ginning of his article. The “Brethren of Purity” (Ikhwān a-ṣ-Ṣafāʾ) do men-
Farmer further quotes (incompletely) Masʿūdī: tion, 200 in their fifth epistle entitled On Music (Fī-l-
ُ
Mūsīqā), dasātīn which would be tied (tushadd ‫ )تشحى‬on
the neck of the ʿūd although they do not mention the

187
8 mentions at least of the words mawḍaʿ (“location”) or mawāḍiʿ
ُ
‫”وتشى على ملكان ملستىق منها دساتين تدت ألوتا تدىد أقسامها‬
(plural of mawḍaʿ). ُ
‫ وتجعل م ية‬، ‫لتي تسنع منها لنغم فتى م لها تلك مىام ح مل ألوتا‬
188
It is noteworthy that the first sheets of this manuscript were (and
still are) missing in the copy cited by Farmer. These will be ad-
.“"‫ لتي تسم " ملشط‬،‫لىاعىة آللة‬
dressed further.
195
In a footnote (no. 2) on this page, Farmer gives as a reference for this
189
While Farmer’s argumentation seems already, at this point, description Les prairies d’or, viii, 99, which is flawed because Masʿūdī is
heavily flawed, intellectual integrity compels us to go along with otherwise not referenced in his article... I could nevertheless find the
his reasoning. corresponding Arabic quote which is reproduced below.
190
“British Museum MS. Or. 2361, fo 236vo”, see [Munajjim (al-),
196
Masʿūdī uses the word dastabān in place of dastān. The editor of
1976, p. 189–209]. the Arabic version mentions (cf. fn. 1 in the reference of the next
footnote) another version still of this term in one of the manuscripts
191
[Farmer, 1937, p. 456]: note that these are still indications about
he consulted for his edition, which is rasān.
the locations of the notes on the dastān; note also that Farmer insists
heavily on the word “fret” in his quotes. In the original [Munajjim
197
[Masʿūdī, 1987, p. 225]:
(al-), 1976, p. 189] we find “mawḍaʿ kull naghma min kull dastān”: .“‫”و لىستبان لذي يلي ألنف م ض ع على خط لتسع مو جنلة ل تر‬
.“]...[ ‫] م ضع كل نغنة مو كل دستان‬...[” 198
(Ibn) Zayla studied under the supervision of (ibn) Sīnā – see
192
So far in this discussion (and in his article), all of Farmer’s asser- more in [Beyhom, 2010b] and [Wright, 2001d].
tions turned out to be imprecise or, worse, flawed. Therefore, I am
199
Still [p. 457].
compelled to use the conditional mood. 200
[Ikhwān a-ṣ-Ṣafā’, 1983, v. 1, p. 203–204].
193
[Farmer, 1937, p. 457].
194
[Fārābī (al-), 1967, p. 498–499]:

139
NEMO-Online Vol. 5 No. 9 – November 2020

material from which these are made. (Ibn) Sīnā does Quote 9: “[…] and the dasātīn are marks put on the necks of
mention,201 in the second part202 of his last discourse203 stringed instruments to localize the positions dedicated to de-
termined notes, and they are used for the appropriate compo-
that the dasātīn must be tied (shadda), but nowhere in
sition [of music]”.209
his whole book-chapter on music is there a mention of
the material of those dasātīn to be found. As for his stu- Note also here two (later) indications by Shirwānī:210
dent (ibn) Zayla, he also says that the dasātīn must be Quote 10: “and [the dasātīn] are marks put on the necks of
tied (shadda)204 on the fingerboard, but, although he, as stringed instruments to localize the positions dedicated to the
did (ibn) Sīnā,205 explains that in order to complete the sounding of specific string-parts”,211

second octave musicians must perform a hand-shift,206 and Lādhiqī:212

he still does not mention the material from which the Quote 11: “[there are] dasātīn in some instruments and these
dasātīn were made. are marks which are put on the neck of stringed instruments
As for Ṣafiyy-a-d-Dīn al-Urmawī, and while not to localize the positions dedicated to the sounding of specific
knowing 207 which of his two treatises was read by notes in the course of melodies”,213

Farmer, I read both the Kitāb al-Adwār and the which confirm, with practically the same words,
Sharafiyya looking for the dasātīn. At the very beginning Urmawī’s descriptions.
of the 2nd chapter of the Kitāb al-Adwār, Urmawī ex- As a consequence of the last references it can be con-
plains:
cluded, at this point, that the dasātīn do not compel the
Quote 8: “The dasātīn are marks which are put on the neck musician to play the notes at their exact position (they
of stringed instruments following precise proportions to serve
do not constitute a compelling temperament), but that
as locators for the emission of notes from parts of the
their main function is indicative: they simply show the
string”.208
locations of the “ideal” (theoretical) notes which “com-
As for the Sharafiyya, in the fourth discourse (in pose the melodies”.
which the division of the fingerboard of the ʿūd is ex- While these references show that Farmer’s argumen-
plained), Urmawī likewise writes:
tation, in this article, is at least unconventional, if not

201
In a-sh-Shifāʾ, an encyclopedic work the 13th chapter of which is 207
Which also applies to Ikhwān a-ṣ-Ṣafāʾ, (ibn) Sīnā and (ibn)
dedicated to music and was translated in French by Rodolphe d’Er- Zayla but, for these authors, I could find corresponding editions in
langer, cf. [Fārābī (al-) and Sīnā (Ibn) or Avicenna (980?-1037), Arabic; as for Urmawī, here is the reference provided by Farmer:
2001, v. 2, p. 234–235] and Chapter II in [Beyhom, 2010b]. “British Museum MS. Or. 136, fo 235vo”.
202
Dedicated to musical instruments. 208
[Urmawī (d. 1294), 1986, p. 93]:
203
[Sīnā (Ibn) or Avicenna (980?-1037), 1956, p. 144–145]. ‫” لىساتين هي عالمات ت ضع على س عى آلالت ذو ت ألوتا على‬
[Zayla (ibn), 1964, p. 74–75]; note also [p. 73] – “among [the .“‫نسب مخص صة ُليستىل بها على مخا ج لنغم مو أجز ء ل تر‬
204

instruments]: those with strings and dasātīn which are tied 209
[Urmawī (d. 1294), 1984, p. 141]:
(mashdūda) at the locations of the notes, such as the ʿūd and the
ṭunbūr”: ‫”و لىساتين هي عالمات ت ضع على س عى آلالت ذو ت ألوتا‬
‫ لتنتىل‬،‫ ذو ت أوتا ودساتين مشىودة على م ضع لنغم‬:‫”منها‬ ‫ليستىل بها على مخا ج نغم معل مة في أماكو مخص صة ليستعان بها‬
“. ‫ كالع د و لطنب‬،‫ألصابع عليها في تخاذ لنغم‬ .“‫على لتألقف ملالئم‬
205
[Sīnā (Ibn) or Avicenna (980?-1037), 1956, p. 144]. See [Wright, 2001e] for more details on this author.
210

206
In the description of the two authors, the musician must perform a [Shirwānī (al-), 1986, p. 70] (French version available in [Shir-
211

hand shift on the zīr (see for example Fig. 15:135) to reach the two notes wānī (al-) and Lādhiqī (al-), 1939, v. 4, p. 70]):
g3 and a3 – these have correspondences on the theoretical string ḥād, but ‫” فهذه سائر أمكنة لىساتين وهي عالمات ت ضع على س عى آلالت‬
Sīnā and Zayla are describing here the praxis – which means that, at .“‫ذو ت ألوتا ليستىل بها على مخا ج لنغم مو أجز ء ل تر‬
least for the a3, the location of the “tie-fret” must be on the soundboard,
which is incompatible with the “tying” of the dastān. In other words, the See [Wright, 2001f] for more details on this author.
212

dastān either does not exist physically (its location can be simply marked [Lādhiqī (al-), 1986a, p. 59] (French version available in [Shir-
213

on the soundboard) or the ligature is effectively a fret which is inserted wānī (al-) and Lādhiqī (al-), 2001, v. 4, p. 292]):
on the soundboard. No such fret is however described by either of the ‫”دساتين في بعض آلالت وهي عالمات ت ضع على س عى آالت ذو ت‬
two authors. Another possibility is that the ʿūd have a slender body at .“‫الوتا ليستىل بها على مخا ج نغم مى ال حان‬
the junction with the neck, which is unlikely because Kindī’s and
Ṭaḥḥān’s descriptions of the ʿūd (see Appendix A and [Beyhom, 2010b])
do not confirm such a shape.

140
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

flawed, his next argument 214 seems to be more con- 3. Ligatures are not necessary when the locations of
sistent: notes are known to the performer.
Quote 12: “If further proof were necessary one might quote 4. If physical tie-frets are needed, gut strings can be
from the Ḥ[ā]wī al-funūn wa salwat al-maḥzūn of Abū’l- used for this purpose.
Ḥ[usay]n Muḥammad ibn al-Ḥasan, better known as Ibn al-
From which it is easy to deduce that:
Ṭaḥ[ḥ]ān215 (fourteenth century ?)216, the only copy of which
is in the Dār al-kutub at Cairo.217 Ibn al-Ṭaḥ[ḥ]ān,218 himself a 1. Mounting physical tie-frets is superfluous for
musician, recommends the use of a pair of compasses219 when confirmed musicians.
fixing the places of the dasātīn on the neck of the lute. He tells 2. As a consequence of the previous point, ligatures
us, however, that he did not need dasātīn on his lute because he were used only for beginners.
knew the place of every note on the fingerboard without
dasātīn. He says, further, that four rolls of gut string were re- While remembering earlier quotes from Urmawī
quired to ‘fret’ a lute, and he recommends that several thick- and other early authors, we may add two supplemen-
nesses ought to be used”.220 tary inferences:
Finally we have here a substantial indication by 1. Ligatures, wether physical (material) or not (for
Farmer of the material existence of frets “made of gut example markers on the upper side of the neck –
strings”. Let us note, for future reference, that the pro- which are still in use nowadays), serve primarily
vided pieces of information in the last quote are numer- as locators of notes.
ous and can be sequenced thus: 2. Ligatures are mainly used, in theoretical writings,
to mark these positions.
1. Locations of ligatures are marked, then these are
mounted (it remains unclear however how) on the Before going any further in our reasoning, it is time
neck of the ʿūḍ. to examine more thoroughly the particular case of the
2. Theses ligatures are not of common usage, as (ibn ligatures on the ṭunbūr and the iconography of the ʿūd
a-ṭ-) Ṭaḥḥān does not use them. (First and Second Interlude thereafter).

214
[Farmer, 1937, p. 457]; this description reminds of Kindī’s de- both his edition of the same [Ṭaḥḥān (ibn a-ṭ-~ al-Mūsīqī), 1990,
scription in the Risāla fī-l-Luḥūn wa-n-Nagham, reviewed further for p. iii] and in his article [Neubauer, 1993, p. 285]) and [Wright,
the mounting of the dasātīn. 2001g] place Ṭaḥḥān’s active period in the 11th century. As a fur-
215
“Ṭaḥḥān” means “miller” (see [Beg, 2000]); see also [Wright, ther indication about the persistent influence of Farmer’s erroneous
2001g] about Abū-l-Ḥusayn Muḥammad ibn al-Ḥasan ibn a-ṭ- assertions on contemporary musicology of the maqām, Poché, in the
Ṭaḥḥān al-Mūsīqī, and more about his and Kātib’s books (men- entry ʿūd of the New Grove [Poché, 2001, p. 27], mentions Ṭaḥḥān
tioned in the following quote) in [Wright, 1999, p. 545], notably: as active in the 14th century and refers for that to another of
“Together with the kamāl adab al-ghināʾ {by (al-) Kātib, available in Farmer’s articles in Studies in Oriental Musical Instruments [Farmer,
French as [Kātib (al-), 1972] and in Arabic as [Kātib (al-), 1973; 1939a, p. 30], while concurrently citing Neubauer’s article of 1993
1975] – also examined in [Beyhom, 2010b]} Ibn al-Ṭaḥḥān’s work in which the latter corrected Farmer(!). (See also next footnote.)
provides an invaluable insight into the conceptual and analytical 217
There currently exists (October 2017) three different copies of
categories familiar to practising musicians in the major cultural cen- Ṭaḥḥān’s Ḥāwī al-Funūn wa Salwat al-Maḥzūn in Dār al-Kutub at
tres in both Egypt and the Fertile Crescent during the first half of Cairo: (1) Funūn Jamīla 32, and (2) Ṭalʿat 84 while the Funūn
the period between the great theoretical syntheses of al-Fārābī in Jamīla 539 published by Neubauer is seemingly lost. According to
the tenth century and Ṣafī al-Dīn al-Urmawī in the thirteenth. Less Rosy Beyhom (personal communication) another version, (3) the
systematic and more eclectic, with a greater interest in human be- M 1362, is certainly a photographed copy of the Funūn Jamīla 539.
haviour and wit than mathematical abstraction, they embody a ma- 218
Farmer’s repeated errors with Arabic (and Persian) names and
jor strand in sophisticated urban culture, happily combining the words may be an indication of his insufficient knowledge of the Ara-
presentation of specialist knowledge with a participation in the bic language. Bouterse’s article [Bouterse, 1979] explains some of the
more general world of ideas by drawing upon and prolonging an deficiencies in his translations (see also [Beyhom, 2011]).
already well-established literary tradition concerned with musical 219
With the help of which marks can be made (lines, or segments
origins and the achievements of outstanding performers”. of circles).
216
This is one additional example of Farmer’s hasty assertions as, 220
See [Ṭaḥḥān (ibn a-ṭ-~ al-Mūsīqī), 1990, p. 175-176 (89–90)]:
in the article entitled “The structure of the Arabian and Persian Lute Arabic original in Appendix D; additional explanations provided in
in the Middle Ages”, he confirms [Farmer, 1939b, p. 46–47] this [Beyhom, 2010b], notably in p. 520-521.
information about (ibn) a-ṭ-Ṭaḥḥān living in the 14th century, while
it is here tempered by a question mark. All of a-r-Rajab (cited in
Yūsuf’s edition of Ṭaḥḥān’s Ḥāwī al-Funūn wa Salwat al-Maḥzūn
[Ṭaḥḥān (ibn a-ṭ-~ al-Mūsīqī), 1976, p. 2, fn. 5]), Neubauer (in

141
NEMO-Online Vol. 5 No. 9 – November 2020

FIRST INTERLUDE: LIGATURES ON THE ṬUNBŪR IN EARLY above the fourth of the length of the total string. As for the
ARABIAN DESCRIPTIONS wasaṭiyyāt [pl. of wusṭā = middle finger], they make them be-
tween S-ʿA and the locations of their banāṣir [pl. of binṣir = an-
Farmer pretends that all the treatises of (some
nular]. Most of them make the distances between their fingers
mentioned by him) early Arabian theoreticians equal, or close to the distances between the dasātīn except for
“prove conclusively that the lute (ʿūd) as well as the the sabbāba, for which they use the last dastān of the Jāhiliyya
pandore (ṭunbūr) had these frets or dasātīn tied which is dastān S-ʿA”.227
around the neck of the instrument”221. Therefore, and according to the greatest theore-
The only extant early (till the 12 century) de-th tician of the Arabian Golden Age, the ʿūd and the
scriptions of the ṭunbūr, to my knowledge, are by ṭunbūr had dasātīn (“ligatures”). However, these
Fārābī (9th century) and by Kātib (probably end of dasātīn did not prevent performers to play between
the 10th/beginning of the 11th centuries). None of the ligatures, above them or below them, in which
the authors mentions any material for the ligatures case the sounded pitches are modified.
or mentions that ligatures have a physical exist- Which makes me wonder if Farmer really read
ence,222 either for the ṭunbūr or for the ʿūd, while the authors he cites, or if he even wished to under-
Fārābī specifies: stand what they wrote.228
“It is possible that an uneven placement of the dasātīn on the
ṭunbūr of Baghdād alters the consonance of notes, in which
SECOND INTERLUDE: ICONOGRAPHICAL ARGUMENTS
case it is necessary, in the course of performance, to use
evenly disposed places between the existing ligatures, as One of the major arguments against the thesis of
described above for the ʿūd223”.224 the “fretting” of the ʿūd was the lack of iconograph-
Using intermediate positions, which change the ical evidence. Farmer himself acknowledges this
pitch of the produced sound, is equivalent to say fact 229 and mentions “hundreds of illustrations of
that the ligatures have no physical existence or that the lute which reveal no trace of frets”, while repro-
they are so thin that they do not have the function ducing230, as a contribution to his thesis, an illustra-
of tie-frets, but are markers for the positions of the tion credited to Riẓ[ḍ]ā ʿAbbāsī and dated from the
fingers. A little further in his treatise, Fārābī ex- 1630s.
plains (see Fig. 17:143)225:
“In our days, most Arabian users of the [ṭunbūr of Baghdād] ne-
glect the dasātīn of the ‘Pagan times’226. They use the part of
the fingerboard below dastān S-ʿA and make of it the [new]
sabbāba [index]. They put the binṣir [annular] below it in the
direction of J, and follow up with the khinṣir [auricular]. They
place their khanāṣir [pl. of khinṣir = auricular] farthest just

221
Italics in the quote are mine. (d. 1294) and [Jurjānī (al-)], 1938, v. 3, p. 110] “two-stringed in-
222
See the description of the dasātīn in [Kātib (al-), 1972, p. 89–91] struments” and refers to his Kitāb al-Adwār ([Urmawī (d. 1294),
and [Kātib (al-), 1975, p. 54]. 1984, p. 44–45] or [Urmawī (d. 1294), 1986, p. 229–230]) in
223
See Quote 13:148. which Chapter 7 (in fact a long paragraph) is dedicated to stringed
224
[Fārābī (al-), 1967, p. 655]. The French version [Fārābī (al-), instruments, and where “ligatures” (dasātīn) are mentioned but not
1930, v. 1, p. 224] is different from my translation, but does not described. Likewise the “Brethren of Purity” mention [Ikhwān a-ṣ-
contradict it. Ṣafā’, 1983, v. 1, p. 202] the ṭunbūr among a dozen other instru-
ments but restrict themselves, in the following pages, to a descrip-
225
See also [Hassan, 1982, p. 10 sq.] for contemporary Iraqi ṭun-
tion of the ʿūd and of its tuning.
būr(s) with different divisions of the fingerboards. 229
[Farmer, 1937, p. 457–458]: “Although it is quite clear from lit-
226
Reminder (see footnote no. 45:119): Jāhiliyya (“Age of igno-
erary sources that the lute of the Arabs and Persians was fretted in
rance”) in Arabic.
the early Middle Ages, it has to be admitted that our iconographical
227
[Fārābī (al-), 1967, p. 663–664]. The French version [Fārābī
sources do not support this”, adding [p. 459]: “Clearly, iconography
(al-), 1930, v. 1, p. 227] is (also) different from my translation, but
is an uncertain guide”.
does not contradict it. 230
Insert (Plate I) between [Farmer, 1937, p. 452–453], with the
228
Note that Urmawī does not even mention the ṭunbūr in his Risāla
following acknowledgment: “(Reproduced by the courtesy of
a-sh-Sharafiyya but mentions, in a very concise paragraph [Urmawī
Messrs. Bernard Quaritch, Ltd.)”.

142
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

The system of the ṭunbūr of Baghdād as explained by Fārābī.

143
NEMO-Online Vol. 5 No. 9 – November 2020

It shows a lute-type instrument the fingerboard of his thesis, as well as he did not bother consulting Er-
which clearly bears marks perpendicular to the strings langer’s book to verify if his thesis could be confirmed
(Fig. 18:144). The drawing allows, however, no identi- or infirmed.
fication of the type of “tie-frets” (or to know if they have This is even more disturbing when we know that
any physical consistency).231 Manik’s understanding of Arabian sources could have
helped him substantiate his thesis, as the Risāla fī-l-
Luḥūn wa-n-Nagham of Kindī, reviewed above for the de-
scription of the tie-frets, was published seven, then four
years before the publication of his book.236
However, before reviewing this description, Fig. 19
is a remarkable example of “chimerical forms” 237 for
music instruments in the literature on Arabian music.

“A Lute Player at the Court of Shāh Ṣafī (1629-1642), by

Riẓā ʿAbbāsī (?)”.232
Farmer was, there is no doubt about this, one of the
best-placed Orientalists to dissert on the iconography of
the ʿūd, and he was notably the editor, in 1966, of a
compendium of illustrations on “Islamic music”.233 The
mere fact that these illustrations were not used as po-
tential “proofs” for Farmer’s thesis is somewhat disturb- Sketch of a ʿūd (duplicate from the original) from a Turk-
ing. ish manuscript from the Bibliothèque Nationale de France
(BNF).238
On the other hand Liberty Manik, who cites no less
than 9 of Farmer’s works along with at least 15 other
references in French (including La musique arabe by Er- Few other representations of the instrument are pro-
langer234), did not bother to consult Farmer’s aforemen- posed in Fig. 20 to Fig. 24.
tioned book235 published in Germany three years before

231
And, this is no early ʿūd. 235
[Farmer et al., 1966]. (Either Manik knew about this book and
232
[Farmer, 1937, Plate facing p. 453]. did not want to cite it, or he simply did not do a thorough research
233
Farmer has published no less than 821 books, articles and Ency- for the relevant literature.)
clopedia entries, of which 334 are dedicated to Arabian music and
236
[Kindī (al-), 1962a ; 1965].
musicians; 121 additional works were still unpublished in 1999 (ac- 237
As Farmer himself describes them in [Farmer, 1937, p. 460].
cording to [Cowl and Craik, 1999]). 238
From [Shiloah, 2002, p. 207]: this sketch is made by Rosy Azar
234
[Erlanger, 1930]. (The six volumes were published between Beyhom.
1930 and 1959).

144
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

Lute-type vīnā found in representations from Amaravati,

Nāgārjunakoṇḍa and Pawaya (India – Gupta period239).240

Lute-type vīnā – Pattaḍakkal (India, c 700 CE).241

Pipa-type 4-strings lute, mural painting no. 288 in the

Mogao grottos (today in the Cansu province – China), North-Wei
epoch (北魏), 4th century.244

Lute player and 6-stringed non-fretted ʿūd-type lute,

Nāgārjunakoṇḍa. (India, 2nd-3rd centuries, bas-relief.)242

“In short”, I agree with Farmer that:

“whilst iconography has an undoubted value in recording the
existence of classes of musical instruments of which no literary
evidence has come down to us, we must always be critical be-
fore accepting the forms and details of such instruments”.243

Two sketches (duplicates from the originals) of so-called

*
* * “fretted” ʿūd(s), allegedly with gut (or silk) strings tied around the
neck.245

239
Between 320 and 480 CE. Narayan, 1990], and kindly provided by François Picard. (This fig-
240
Carbon copy by Rosy Azar Beyhom, from [Subramanian, 1985, ure was previously published in [Beyhom, 2010b].)
p. 12, Fig. 8]. (This figure was previously published in [Beyhom, 243
[Farmer, 1937, p. 460].
2010b].) 244
From [Liu et al., 1988, ill. II-86], kindly provided by François
241
As above, from [Subramanian, 1985, p. 12, Fig. 9]. Picard. (This figure was previously published in [Beyhom, 2010b].)
242
This is to this day the oldest representation of a ʿūd-type lute that 245
(These sketches were drawn by Rosy Azar Beyhom, and were
I could identify, taken from the booklet of [Zakir Hussain and Brij previously published in [Beyhom, 2010b] and [Beyhom, 2016]).

145
NEMO-Online Vol. 5 No. 9 – November 2020

THE TIE-FRETS (DASĀTĪN) IN KINDĪ’S RISĀLA FĪ-L-LUḤŪN The fourth tie-fret is mounted with a zīr string
WA-N-NAGHAM 1½ ff after the third tie-fret. The zīr string is made
In the Risāla fī-l-Luḥūn wa-n-Nagham, unlike in of twisted silk strands the section of which corre-
other references cited by Farmer, 246 the author sponds to the section of one gut strand.
(Kindī) thoroughly describes the material(s) used
for the tie-frets, as well as their thickness propor-
tions. Furthermore, it seems that Farmer had access
to a copy of this epistle247 which, as he writes in a
later article,248 he had consulted in 1926, and men-
tioned that the first folios were missing.249

The missing folios happen to be those where

Kindī describes the tie-frets. 250 The total speaking
length of the strings is 30 “full fingers” (“ff”), which
roughly equates to 60 cm. The first tie-fret, the
sabbāba (index), is positioned (see Fig. 9:131) at a
distance of 3 ff from the nut, and is made of a bamm The first (sabbāba) tie-fret (“tie” in the figure) described
by Kindī and tangency point (“Bend”) with the bamm string. This
string (the thickest and, acoustically, the lowest
is a side view (length section) in which d is the diameter of the
string) winded twice around the neck (Fig. 25). It is bamm string, d is (twice) the undetermined error estimation –
firmly tied251 to avoid lateral displacements.252 due to the elasticity of the composing gut-material – between the
theoretical position of the tie-fret and the effective contact point
The bamm string is made of four strands of ho- of the mounted (and stopped) bamm string. The tie-fret, which is
mogeneous gut of constant cross section thoroughly also made of a bamm string, is winded twice around the neck.
twisted together (Fig. 26 and Fig. 27).253 All these explanations by Kindī are clear and
The second tie-fret, the wusṭā (middle finger), is consistent, and show that, at least for this author,
made of mathlath string, and mounted 2 ff away tie-frets made of strings were effectively mounted
from the sabbāba. The mathlath string is made of on the neck of the ʿūd, contradicting thus Urmawī’s
three strands of twisted gut. The third tie-fret is indications (Quote 8:140 and Quote 9:140) while
winded 1 ff apart from the wusṭā, with a mathnā confirming (a priori) the description of Ṭaḥḥān
string made of twisted silk strands (Fig. 27) the sec- (Quote 12:141).
tion of which is equal to the section of two gut One undeniable contradiction is however raised
strands.254 by Kindī himself as he mentions in the same epistle
notes performed outside the tie-frets, between the

Such sketches (above Fig. 84 in [Farmer et al., 1966] – taken from 251
This contradicts Kindī’s further indications for octaves and fifths
Kanz a-t-Tuḥaf, unknown author, Iran, mid-14th-century, British correspondences as the knots would have to be undone and done
Museum MS. Or. 2361, fo 260vo; below Fig. 81 in [Farmer et al., repeatedly.
1966] –from the Kitāb al-Adwār by Ṣafiyy-a-d-Dīn al-Urmawī, Bod- 252
[Kindī (al-), 1965, p. 12]: the other tie-frets are described on this
leian Library Oxford, MS. Marsh 521, fo 157vo, 1333-1334) served page and the following.
as “proof” that ʿūd(s) from the Early Islamic Era were, like Occi- 253
[Kindī (al-), 1965] [p. 15]: descriptions for the material of other
dental lutes in the Baroque and Renaissance periods, “fretted”. strings are given on this page.
246
Except for (ibn a-ṭ-) Ṭaḥḥān. 254
The reasons invoked by Kindī [p. 16] for this change of material
247
Which probably corresponds to the Berlin MS. Ahlwart, 5530, are firstly that the sound of silk strings is “purer” for higher notes
fos 25ro – 31ro, published as the fifth epistle in [Kindī (al-), 1962a]. such as on the zīr, and secondly that the mathnā and the zīr need to
248
[Farmer, 1939b]. be tensioned to such an extent as to possibly rupture were they
249
[Farmer, 1939b, p. 43, fn. 2]. made of (one or two) gut strands, whenever silk strings would not.
250
The missing folios correspond to pages 9–14 in [Kindī (al-), The complete original quote for [Kindī (al-), 1965, p. 12–16] is
1965], and the incomplete manuscript consulted by Farmer begins available in Appendix D.
at the end of the first line of page 14.

146
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

tie-fret of the khinṣir (auricular) and the bridge, shall seek evidently in Kindī’s works, but first in
maybe even on the soundboard of the instrument.255 other author’s works.

Twisting and knotting of the silk threads which make the

zīr in Kindī’s Risāla fī-l-Luḥūn wa-n-Nagham. Above: bunch of un-
twisted and unstretched “silk” threads. Below: stretched string
made of twisted silk threads and knotted at the two ends.257

REFERENCES TO LIGATURES (DASĀTĪN) IN FĀRĀBĪ’S KITĀB

AL-MŪSĪQĪ AL-KABĪR AND (IBN) SĪNĀ’S KITĀB A-SH-SHIFĀʾ

The Kitāb al-Mūsīqī al-Kabīr of Fārābī is considered

to be the most complete work on Arabian music of the
Result of the (thorough) twisting of the 4 gut strands to
make the bamm string on the ʿūd (the figure shows cross sections Golden Age of Islam.258 It is therefore somewhat disturb-
of the gut strands). If the gut is 1 mm wide (diameter = 1 mm), ing that this author was not more quoted by Farmer
the resulting string will have a diameter D of approx. 2 mm.256 with regard to “frets” on the neck of the ʿūd; note that
We need therefore additional information in or- the Kitāb al-Mūsīqī al-Kabīr as well as (ibn) Sīnā’s book-
der to be able to conclude on this subject, which we chapter on music were already translated at the time
Farmer wrote his article.259 The excerpt which was indi-
rectly quoted by Farmer (Quote 3: 139) contains (in the
255
As seen in Part I.B of this dossier. oughly washed before twisting; e) was garlic/alum used as an anti-
256
Given 𝑑1 as the diameter of one gut strand (assuming it was al- septic prior to twisting, as this would change the structure of the
ready treated as to have a homogeneous density), the cross-section collagen; f) how many twists per cm; g) what was the tension of
of one strand is  × (𝑑1 )2. A good twisting – depending also if the guts during the twisting process; h) were the guts split prior to twist-
gut is wet or dry – discards the empty spaces between strands in ing. Many other factors such as hygrometry at the time of the twist-
such a way as to form one gut string of circular cross-section the ing must also be taken in consideration: in practice, I would expect
diameter of which we name 𝑑2 . The cross-section of the resulting that the consideration of the few parameters I gave you would lead
string is equal to the cross-section of the 4 strands of gut twisted to a + or - 25% either way. The only way to set a proper formula-
together to form the string, i.e. × (𝑑2 )2 = 4 ×  × (𝑑1 )2 which, tion would be empirically to emulate ancient methods, and with
after simplification and reduction, gives 𝑑2 = 2 × 𝑑1. Four consid- proper microscopic examination”. Other considerations must be
erations are important for this reasoning: firstly, in the case of not taken into account (such as the angle of the twisting, the polishing
sufficient twisting, there will remain tiny empty spaces between the of the string) – see [Abbott and Segerman, 1976 ; Bonta, 1999] for
strands, which results in a slightly greater diameter of the string. more details.
Secondly, while a gut strand is initially hollow (tubular), and while 257
Note that silk strings are not made as with gut strings. To make
Kindī does not mention a prior twisting of the gut strands, we can a silk string the silk threads must be twisted over double the length
assume that 𝑑2 will be smaller than 2 × 𝑑1 . The same applies for required and then folded in two (and knotted again at both ends)
the third consideration, which is that the process of twisting will to make one string where the strands will not unwind. Further: “an
elongate the gut (but reduce the total length of the resulting string), old cloth of linen is [soaked in glue], and silk strings should be
reducing in so doing the diameters 𝑑1 of the strands, in which case, rubbed with it until they are infiltrated thoroughly with the ingre-
it is better to assume that the final cross-section will be less than dient” – in [Tsuge, 2013, p. 178]. (See also the complete description
2 × 𝑑1 . However, the material (the gut) plays a major role and of the “making of” gut and silk strings in Appendix A.)
makes the final result more complicated, as Richard Dumbrill ex- 258
See for example Carra de Vaux’s appreciation of Fārābī’s work
plains (private communication): “You must consider the nature of in the foreword to Erlanger’s translation of [Fārābī (al-), 1930, v. 1,
the gut (collagen) and depending on many factors, such as (a) the p. vii].
animal from which the gut comes; (b) the age of the animal when 259
Respectively in 1930 and 1935 – see previous footnote and
it was ‘gutted’; (c) the time between the slaughtering of the animal [Fārābī (al-) and Sīnā (Ibn) or Avicenna (980?-1037), 1935].
and when the guts were used (never longer than 30 hours); (d) were
the guts salted as soon as extracted from the animal and then thor-

147
NEMO-Online Vol. 5 No. 9 – November 2020

original version) the phrase “these dasātīn play the role close to them. They consist in plucking the string stopped at the
of bridges” that Farmer omitted, although it could have location of one ligature and moving then the finger to another
ligature located above or below it, with no disruption [on the
helped him defend his thesis.
string]. The aim is to modify the sound continuously from low
Whenever this indication remains inconclusive for to high or from high to low”.266
our purpose,260 other indications in Fārābī’s book may
While for the two major authors of the Golden Age
help us better understand the role of the ties for the per-
of Arabian music the “tie-frets” on the neck (finger-
former:
board) of the ʿūd are just visual markers for the notes,
Quote 13: “The leimma is close to a quarter-tone, which is the
their explanations seem to contradict fully the indica-
reason why its consonance may be found close to the conso-
nance of the quarter-tone. The reason is that the finger does not tions of Kindī and Ṭaḥḥān.
always reach the exact location of the intended note and may As all indications in the literature and the iconogra-
[stop the string] a little further or closer. If the quarter-tone was phy about the organology of the instrument seem to
intended and the finger went a small amount further, it becomes
converge towards the use of fretless ʿūd(s) in the Golden
a leimma which is not originally consonant. If the interval of
Age of the Arabian Empire, it may be concluded that
leimma was intended and [the finger] went a small amount
closer, then the leimma becomes closer to a quarter-tone. […]
either (1) Kindī did not know much about the matter
This is why it is difficult to conclude on the consonance of the (and Ṭaḥḥān copied him without experimenting with
leimma performed on the ʿūd”.261 tie-frets on the neck of the instrument)267 or that, effec-
tively, (2) tie-frets had existed historically over a short
While this completely contradicts Kindī’s “solid” ties
time period (around the 9th century) but were limited in
(and Farmer’s assertions), the following excerpt from
number on the fingerboard and were used for teaching
Fārābī’s book is even more explicit on the subject:
or theoretical purposes exclusively.
Quote 14: “It is however useless to multiply the dasātīn. Many
musicians [“persons”] use other notes than these [the ones lo-
However, Kindī’s descriptions of the proportions of
cated by ligatures] which have no predetermined locations, de- the ʿūd and of the tie-frets being very precise in compar-
pending on the needs of the composition of their melodies. ison to other authors, it is possible that further explora-
Some of these notes are sounded from between the ligatures tions of his and other writings may reveal other, com-
(dasātīn) and others below the ligature of the khinṣir [auricu- plementary details. Eckhard Neubauer’s 1993 article on
lar]262 while others [still] are found over the ligature of the
the ʿūd is such an attempt which is typical in its Orien-
sabbāba (index).263 These notes are used to enrich the melody.
talist handling of the sources.
If somebody wishes to determine these notes, he must search
for the corresponding location on the dasātīn or between
them”.264 *
* *
As for (ibn) Sīnā, the following explanations can be
NEUBAUER’S “BAU DER LAUTE…”
found in the book-chapter dedicated to music in his
Kitāb a-sh-Shifāʾ, in the section concerning the tuning of In “Der Bau der Laute und ihre Besaitung nach
the ʿūd and the division of the fingerboard: arabischen, persischen und türkischen Quellen des 9. bis 15.
Jahrhunderts”,268 Neubauer examines the problematic of
Quote 15: “Concerning the tawṣīlāt265 [pl. of tawṣīla – “link,
connection”] – these are of the same type as the ‘mixtures’, or
the “Bünde” (“frets”) on the fingerboard of the ʿūd

260
Because “ligatures” can still play the role of “intermediate” 267
Let us remind ourselves about the well-known anecdote (based
bridges if they are simply marks on the fingerboards as quoted for on Nicomachus and) mentioned in [Chailley, 1985, p. 7–14], in the
Urmawī (Quote 9: 140). chapter entitled “The Harmonious Blacksmith”, which explains
261
[Fārābī (al-), 1967, p. 580–583]. French translation in [Fārābī how an error stemming from a “fanciful experience […] that five
(al-), 1930, v. 1, p. 201]. minutes and a piece of string would have been enough to rectify”,
262
Between the ligature of the auricular and the bridge. lasted 22 centuries in the musicological literature. To the like of the
263
Between the ligature of the index and the nut. story of the Silesian child of Rousseau, or to the tale of the fish of
King James (see the quote and corresponding footnote at the begin-
264
[Fārābī (al-), 1967, p. 516], and [Fārābī (al-), 1930, v. 1, p. 174]
ning of Chapter III in [Beyhom, 2016]), the desire to enhance one’s
for the French translation.
writings is often enough to lose one’s critical sense.
265
Otherwise called “portamentos”. 268
[Neubauer, 1993] – “The construction of the lute and its string-
266
[Sīnā (Ibn) or Avicenna (980?-1037), 1956, p. 140], and [Fārābī
ing according to Arabian, Persian and Turkish sources from the 9th
(al-) and Sīnā (Ibn) or Avicenna (980?-1037), 1935, v. 2, p. 231]
to the 15th centuries”.
for the French translation.

148
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

within four pages269 supported by various quotes in the ʿūd”) – with four strings – and the two others, later
second part of his article. He states from the outset270 ʿūd(s), and knowing that in his first indication
that Farmer’s 1937 article271 – published fifty-six years (Quote 11) he mentions only marks on the fingerboard
before his own article – is “clear” on this question de- to be used as dasātīn, it is difficult, in this case, to be sure
spite of the increasing skepticism which, according to about which instrument or epoch this (new) indication
him, reached the “secondary literature”, and endeavors (Quote 16) is.
to “correct” the “erroneous opinion”272 of authors who We cannot decide either if the multiplication of the
do not believe that Early Arabian ʿūd(s) were fretted. strings created specific constraints (for example for oc-
In order to support his thesis, he relies on a series of tave and fifth correspondences) which compelled some
quotes, either already known to the reader – from “modern performers” to add marks or ligatures (dasātīn)
Farmer’s 1937 aforementioned article – or “new”, from to show the new positionings for some notes, and nei-
sources that he read. ther can we conclude if Lādhiqī, in this excerpt (Quote
16:149) simply retells the history of the dasātīn by im-
Neubauer’s references to Lādhiqī, Fārābī and other au- plicitly quoting Kindī – a sort of a tribute to his prede-
thors cessor – on the matter. Let us simply note that, accord-
Neubauer’s first quote “in favour” of the fretting the- ing to this quote, the dasātīn can either be tie-frets (made
sis is from Lādhiqī (the first phrase in bold type, in the of gut or of another material) or marks (lines) drawn on
original and which Neubauer leaves out, is added to ex- the neck.
plain the context)273: Next, Neubauer mentions Kanz a-t-Tuḥaf277:
Quote 16: “Some modern performers mount a sixth string Quote 17: “The dasātīn consist in a series of marks (nešānī-ye
on this instrument and call it the ʿūd akmal [(even) “more čand) 278 affixed (waḍʿ karde) on the necks (sawāʿed) 279 of
complete” (than the “complete” ʿūd with 5 strings)274 ], stringed instruments (ālāt-e dawāt-e outār) for a firm and pre-
and markings are put on the neck of these instruments to show cise positioning (tašaddod) of the fingers on the string and for
the [places] for the emission of the notes of the melodies the production of the notes (esteḫrāǧ-e naġamāt) on it”.280
[madār al-alḥān] from the neck, and these markings are called
dasātīn, be they from tied strings, marked lines or others In itself, this quote confirms that the dasātīn were
[still]”.275 marks 281 “affixed” 282 on the fingerboard of the ʿūd or
other stringed (and probably lute-like) instruments.
This excerpt comes late in Lādhiqī’s work and com-
After quoting Khawārizmī at the beginning of the
plements Quote 11:140 276 , while restricted to either
next page of his article, 283 Neubauer, while asserting
“modern performers” or to the ʿūd al-akmal. Knowing
that the “usual material for the frets [was], according to
that in the previous two pages Lādhiqī’s discourse re-
Ancient sources, pieces of string” 284 quotes then (di-
lates to the differences between the ʿūd qadīm (“the Old
rectly) Fārābī (corresponding to Quote 3:139) 285 and
269
[Neubauer, 1993, p. 328–331]. 280
Translated from [Neubauer, 1993, p. 328]: the German original
270
[Neubauer, 1993, p. 328]. quote, as for all quotes from Neubauer, is reproduced in Appendix D.
271
The aforementioned [Farmer, 1937], which Neubauer errone-
281
“Zeichen” in the original German.
ously dates from “1939”. 282
“die man auf den Hälsen […] anzubringen pflegt” in the original
272
“Fehlmeinung”. German. “Anbringen” can be translated as either “affixed” or
273
This is one further indication, if needed, that Neubauer picks out “mounted” (or other possibilities), which does not help in determin-
in the literature what comforts his thesis specifically, and discards ing the material from which these “marks” are made.
whatever information or facts that can put it in doubt.
283
Translated from [Neubauer, 1993, p. 329] – Neubauer’s quote
274
Expounded in [Lādhiqī (al-), 1986b, p. 178]. corresponds to Quote 1:138 : “dasātīn are the tied places (ribāṭāt)
275
Translated from the Arabic version [Lādhiqī (al-), 1986b, upon which the fingers are placed”.
p. 179].
284
Which, as we have seen, is false as the majority of references pro-
276
“[There are] dasātīn in some instruments to localize the positions vided in this dossier concern markers on the fingerboard. Note that in
dedicated to the sounding of specific notes in the course of melo- Neubauer’s article the only mention of “pieces of string” till this point
dies” – [Lādhiqī (al-), 1986b, p. 59]. is in Lādhiqī’s reference, which is far from being all the “Ancient
sources”, and even further from being the “usual material” for the ties.
277
Which I could not find (and I have no knowledge of Persian). 285
Translated from [Fārābī (al-), 1967, p. 498–499], in which I re-
278
All transliterated Persian terms in this quote are in Bold type.
instate here [between brackets] the phrase at the beginning: “[And
279
“Sawāʿid” in Arabic transliteration.
this instrument is one in which the notes are emitted according to

149
NEMO-Online Vol. 5 No. 9 – November 2020

then indirectly (and with no reference to a page num- if the finger stops the string on the corresponding posi-
ber): tion, although it may not be marked by a dastān.292 In
“Elsewhere he [Fārābī] says that notes which are positioned other terms, notes on the ʿūd that he describes can be
above the ties can be played without additional ties only by produced whether there are ligatures or marks (dasātīn)
Masters of the corporation [of musicians]”.286 affixed to the neck, or not.
Searching for this unreferenced (and indirect) quote Neubauer’s does not stop, however, at these trun-
in Fārābī’s Great Book of Music, the only corresponding cated or tampered quotes, but quotes as well Kindī in
quote to be found is the aforementioned Quote the Risāla fī-l-Luḥūn wa-n-Nagham and (ibn a-ṭ) Ṭaḥḥān
14:148287 in which, however, Fārābī does not mention in his Ḥāwī-l-Funūn wa Salwat al-Maḥzūn.
“Virtuoso performers” (or “Masters of the corporation”)
but “many persons” who, furthermore, can play between Neubauer’s interpretation of Kindī’s Risāla fī-l-Luḥūn
the ligatures, over them or below them, a fact that wa-n-Nagham
Neubauer (very) lightly overlooks(!).288 Further quotes from Kindī’s Risāla fī-l-Luḥūn wa-n-
Immediately after, Neubauer quotes Fārābī (both in- Nagham are provided in Neubauer’s article, explaining
directly and directly) a second time: the tuning, and the mounting and tying of the tie-
frets293, while concluding (see FHT 9:166)294:
“In one equivalence of fifth, the fifths can be for example only
produced ‘if a fret stands there, otherwise not. Unless [the per- “According to the indications [of Kindī] the proportions of the
former] succeeds in positioning the finger [correctly]’”.289 strings from the zīr to the bamm strings are 1:2:3:4. In an anal-
ogous way, the sizes of the frets, from the fret of the index to
Here is the complete translation of the excerpt:290 the tie of the auricular, should diminish in the proportion
Quote 18: “in this tuning [to the fifth between the bamm and 4:3:2:1. Both are unrealistic”.295
the following string], the notes produced by the three strings
Let us firstly note that these proportions are not nec-
below the first string [the bamm] are displaced when compared
to the same notes in the usual tuning [in integral successive essarily unrealistic as similar increasing thicknesses of
fourths] one whole tone above [towards the lower tones]. If a ties – but not necessarily similar dimensions – were used
dastān happens to be there, they will be produced, if not they in European lutes296 and, secondly, that this quote sug-
will not, or it may happen that the finger stops [the string] on gests that Kindī may have given, in Neubauer’s opinion,
the corresponding [location of the] dastān”. false indications for these proportions which would in
Fārābī clearly says in the Arabic original291 that the turn make him unreliable as regards the organology of
notes, if they are not found on one of the dasātīn (pl. of the ʿūd.297
dastān) of the previous tuning, could still be performed

the division of the strings with which it is strung]. And ligatures are 289
Translated from [Neubauer, 1993, p. 329].
winded on the neck of the instrument under the strings and deline- 290
In both Erlanger’s translation in [Fārābī (al-), 1930, v. 1, p. 208]
ate on each of them the string-parts from which the notes are and in the Arabic original [Fārābī (al-), 1967, p. 600]. French and
sounded, and as a result they play the role of a string-holder and Arabic texts are reproduced in Appendix D.
are placed parallel to the bridge”. 291
And in the French translation.
286
Translated from [Neubauer, 1993, p. 329]. 292
In any lute-type instrument a change in tuning compels the per-
287
[Fārābī (al-), 1967, p. 516], reproduced here for convenience: former to adapt his technique to the new positionings of the fingers
“It is however useless to multiply the dasātīn. Many musicians on the strings – this is common-knowledge among performers.
[“persons”] use other notes than these [the ones already located 293
These descriptions are provided in full in “Annexe II.3” of
by ligatures] depending on the needs of the composition of their [Beyhom, 2010b].
melodies, which have no predetermined locations. Some of 294
For “Figure Hors Texte 9, p. 166”.
these notes are sounded from between the ligatures (dasātīn) and 295
Translated from [Neubauer, 1993, p. 330]: bold type mine.
others below the ligature of the khinṣir (auricular) while others 296
[Abbott and Segerman, 1976, p. 431]: “the grading of frets for
[still] are found over the ligature of the sabbāba (index). These notes
fine adjustment of the action made them remarkably thick at low
are used to enrich the melody. If somebody wishes to determine
positions (near the nut)”.
these notes, he must search for the corresponding location on the
dasātīn or between them”.
297
This supposed unreliability of Kindī is an important element in
the following discussion.
288
This indirect quote by Neubauer is clearly biased in order to in-
fluence the reader in favour of his thesis as the direct quote clearly
mentions performance between the ties, which is the most probable
reason why Neubauer avoided quoting Fārābī directly.

150
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

However, in the second part 298 of his article Neubauer’s interpretation of (ibn a-ṭ) Ṭaḥḥān’s Ḥāwī-l-
Neubauer dedicates eight full pages299 to Kindī’s epistle Funūn wa Salwat al-Maḥzūn
while explaining in Section 20 of his article entitled “Die Further, in the 19th section of Neubauer’s article, in
Stellung der Finger auf den Bünden beim Greifen der which the “frets” are explained, the author mentions the
Saiten”300 Kindī’s description of the position of the fin- description of the tie-frets by (ibn a-ṭ-) Ṭaḥḥān (FHT
gers on the neck.301 11:167) which follow similar proportions to Kindī’s (de-
The description (see FHT 17:172) is compatible with gressive from the nut).
gut tie-frets as he already described them, and the posi- The author concludes this section by a further quote
tion of the finger must not change, in either direction (nut of Ṭaḥḥān and commentary:
or bridge) otherwise the sound will be muffled (Taubheit) (if “… ‘There is still a fret which lies between the *ring finger*304
the position changes towards of the bridge) or will be accom- and the *auricular* frets, [but]305 it is also [normally] not used.
panied with “chirping[?]” (Zirpen)302 when the finger stops This [?] are frets, which fall out of the original number [6].
the strings between two ties (see FHT 18:172). Neubauer They were used by the Persians for their modes. I use them also
then concludes: and reach their [correct] places [on the fingerboard], because
I know them, also without [additional] frets. For students this
“[T]he description of the correct position of the fingers of the
is however difficult. To leave them [(Sie fortzulassen)] is
left hand still applies today and is a remarkable testimony for
[therefore] better and more appropriate (richtiger)’.306 From
Kindī’s precise observation and formulation. Thus the last
this follows that an Egyptian Court musician of the 5th/11th cen-
doubt on the practical use of frets must be here dismissed”.303
tury performed also Persian music and that he played it on his
Neubauer has no more doubts here, whatsoever, local lute with or without additional frets. The fact that he
avoided the additional frets and with that, the Persian reper-
about Kindī’s reliability for organological matters, in an
toire for beginners is understandable.”.307
assertion which totally contradicts his opinion in the
previous quote about the “unrealistic” description of the Neubauer’s translation above would have been ac-
proportions of Kindī’s tie-frets. curate were it not for the terms in bold type (by me) in
Let us note, for the record, that the whole “Section the quote. The “also” is added by the author in the trans-
20” is dedicated to this description, and that all the other lation, while the second expression “To leave them (the
authors mentioning the precise stopping of the strings on the ligatures or tie-frets) is better” (in Arabic “‫”فتركه أولى وأحق‬
ties or that the dasātīn are marks affixed to the fingerboard or “to leave it”) should have been “to leave it to [or for]
of the ʿūd are disregarded. Let us also note that Kindī ad- them is better” 308 (in Arabic “‫)”فحتححركححه ل ه أولى وأححق‬. Fig.
vises against stopping the strings between the ties, and 28:152 shows an excerpt from Ṭaḥḥān’s manuscript ed-
against reaching farther than (just before) the needed tie- ited and published by Neubauer in 1990 with a frame
fret in direction of the bridge, to preserve sound quality (line below) encompassing the phrase in Arabic “ ‫فتركه‬
(see FHT 16 and FHT 17:172). ‫”ل أولى وأحق‬.

298
Dedicated to translations of Early authors. 305
The words between brackets were added by Neubauer.
299
[Neubauer, 1993, p. 334–342]. 306
The original Arabic version from the manuscript of Ṭaḥḥān pub-
300
“The position of the finger on the frets when stopping the lished by Neubauer [Ṭaḥḥān (ibn a-ṭ-~ al-Mūsīqī), 1990, p. 175] is
strings”. reproduced in Appendix D.
301
The complete text of Section 20 [Neubauer, 1993, p. 331–332],
307
Translated from [Neubauer, 1993, p. 331].
is reproduced in Appendix D. The finger stops the string near the 308
My translation converges towards Farmer’s narration of
tie-fret, immediately behind it as shown in FHT 16:172 (2nd position Ṭaḥḥān’s manuscript in [Farmer, 1937, p. 457], notably: “Ibn al-
– in dotted lines). Ṭaḥ[ḥ]ān […] tells us, however, that he did not need dasātīn on his
302
‫ صرررارا‬in Arabic, which is different from “chirpen” (or the Ger- lute because he knew the place of every note on the fingerboard
man “Zirpen”) and should be translated as “squeak”. without dasātīn. He says, further, that four rolls of gut string were
303
Translated from [Neubauer, 1993, p. 331–332]. required to ‘fret’ a lute, and he recommends that several thicknesses
304
I use the following three levels for quotes and sub-quotes ought to be used”.
(namely stars inside simple quotes inside double quotes): “a ‘b *c*
b’ a”.

151
NEMO-Online Vol. 5 No. 9 – November 2020

The excerpt clearly proves that Neubauer ignored Arabian-Islamic music history solely used for theoretical pur-
pose[s] but not in praxis 310 must henceforth belong to the
the word “‫ ”لهم‬which means “to them”, or “for them”,
past”.311
excluding thus the possibility for beginners to make a
choice between keeping the dasātīn, or performing Per- Strangely enough, Neubauer’s argumentation is that
sian music without them.309 The elision of one single this excerpt from Ṭaḥḥān shows that it was possible to
word by a competent philologist allows for the reversal play notes between the tie-frets, and he uses this possi-
of the meaning of the phrase, imposing thus no other bility of playing between the “frets” as a further argu-
choice as special tie-frets “for beginners” to perform ment for his thesis – which is even more astonishing as,
“Foreign” music. in accordance with Kindī’s explanations mentioned by
Neubauer one page after, stopping the strings on a dif-
ferent position than the one shown in FHT 17:172
would cause the sound to become “muted” (“muffled”)
and the string to sound “squeaks”.
Unless Neubauer, through his negative appreciation
Excerpt from fo 89 ro of the manuscript [Ṭaḥḥān (ibn a-
of Kindī’s string proportions above, considers that the
ṭ-~ al-Mūsīqī), 1990, p. 175] reproducing lines nine and ten. The
10th line (below in the excerpt) contains the (framed) expression “frets” were so thin that they would not hinder the per-
“‫ ”فتركه لهم أولى وأحق‬with the possessive term “‫ ”لهم‬meaning “to [or formance between “frets”.
for] them”. However, this would mean that these dasātīn did not
Neubauer concludes that Ṭaḥḥān “avoided the addi- have the role of frets, which would contradict once
tional tie-frets and with that, the Persian repertoire for again his praise of Kindī’s “precise observation and for-
beginners”, which is contradicted by the fact that the mulation” of the position of the stopping finger imme-
latter simply states that he “uses” two supplementary diately behind the (physical) tie-fret.
ligatures (dasātīn) without “marking” or tying them on Whichever way we may try to understand
the neck, which is a clear indication that dasātīn did not Neubauer’s astounding statement, its inconsistency re-
prevent the performance between the ligatures. mains obvious.
Let us also note that, while according to Neubauer
Ṭaḥḥān does not use supplementary ligatures for special Conclusions on Neubauer’s “new facts” on the fretting
notes, he would need them even less for usual, much of the Early Arabian ʿūd
better known to him places for the dasātīn.
In Neubauer’s argumentation on the “fretting” of the
These contradictions do not stop Neubauer from Arabian ʿūd we can single-out one quote from Lādhiqī
concluding: which gives alternative possibilities for the material of
“However, the argument that is today to hear, that it was gen- the ties on the neck of the ʿūd – including simple mark-
erally not possible to play intermediate notes on a lute with ings – that may apply, with the latter author, restric-
frets, and that this was the reason why frets were, with time,
tively to some musicians or to one particular type of ʿūd,
removed is in this exclusiveness (Ausschließlichkeit) not accu-
the “ʿūd akmal”. The ʿūd akmal holds, according to
rate. Similarly, the persisting representation that frets be in the
Lādhiqī, 6 strings tuned in successive (just) fourths, a
fact which complicates the identification of the stopping

309
Did he ignore it or forget it in his translation? In both cases, the II, 59-68’.] He relied for this conclusion on the frequent use in the
meaning was changed in favour of the thesis of the fretting of the ʿūd. sources of the Persian word dasatin [dasātīn] (hands; frets); further,
310
The author inserts here a footnote (no. 119): “as still with Theo- it seems unreasonable to suppose that the lute when used for acous-
dore Grame, The Symbolism of the ʿūd, in: Asian Music (New York), tical experiments would have been unfretted. Whatever may be the
Bd. 3,1 (1972), S. 25-34, hier S. 32”. Neubauer is probably reacting truth of his thesis–and it is possible to suppose that the ʿūd, like the
to the following passage [Grame, 1972, p. 32]: “As to whether the viola da gamba, was played both with and without frets–the evi-
medieval ʿūd was fretted, there has been much controversy. Most dence is quite incontrovertible that the present-day traditional ʿūd
scholars, who have relied on iconographical evidence, have con- is not fretted. Perhaps, as we have suggested, frets were used for
cluded that the lute was not fretted, for there is no known delinea- investigations into the physics of sound, but were abandoned when
tion of a fretted ʿūd, though many illustrations are extant. Farmer, virtuoso musicians performed”.
however, adamantly maintained that the instrument was fretted. 311
[Neubauer, 1993, p. 331].
[citation here of ‘H. Farmer, Studies in Oriental Musical Instruments

152
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

points on the strings for the performer especially for “to sketch [rasm] a summary of the instrument of the Wise Men
note correspondences from one octave to another. fitted with four strings and called [the] ʿūd, allowing for one to
be aware of its structure and compose on it, and all that is
The persisting inconsistency in Neubauer’s argu-
needed to know about it”.313
ments, who dismisses organological facts whenever
they contradict his thesis, then uses the same facts to re- This epistle is then, above all, written as a method
inforce his thesis, is obvious in this review. for the ʿūd with a preliminary description of the struc-
ture of the instrument [p. 11-12], of the mounting of the
As a result, no consistent additional proof for solid
tie-frets [p. 12-13], and a justification for the propor-
tie-frets used in performance by trained musicians is
tions used in this description [p. 14].
provided by the author, on the contrary as we can infer
from both Kindī’s and Ṭaḥḥān’s descriptions that tie- In the second part of his epistle Kindī thoroughly de-
frets were used, if any, for beginners only. scribes the material of which the strings are made and
their specifications (homogeneity, constant cross-sec-
Moreover, Neubauer provides solely (and mostly
tion, etc.), then explains the tuning of the instrument
failed) arguments in favour of the “fretting” thesis and
and lists the consecutive notes and their matches at the
deliberately disregards the substantial, precise and even
octave, with an exercise consisting in humming the suc-
detailed arguments against this thesis.
cessive notes while playing them on the ʿūd.314
Thus, Neubauer’s “New-Orientalist” approach be-
He proceeds then, after a digression on the relation
comes clear in its endeavor to impose forcibly the “fret-
between the instrument and the celestial bodies, with a
ting thesis” in maqām musicology. This, in turn, allows
second exercise for which he describes note after note
for the final conclusions on the “fretting” of the ʿūd
the fingering (tablature) to be used, with detailed indi-
which follow.
cations (on three successive pages)315 on the pace of per-
formance and on the fingers of the right hand used to
Conclusion of Part II pluck each string. He concludes this part by advising the
reader (the apprentice) to repeat the exercise while
TIE-FRETS, IF THEY EVER EXISTED, WERE SOLELY USED FOR
gradually accelerating the pace, which will help him
BEGINNERS OR FOR THEORETICAL PURPOSES
master the instrument.
Whenever all other authors state or explain that lig-
As a conclusion to his epistle Kindī explains finally
atures on the neck of the ʿūd are equivalent to visual
that there existed at his time many schools for the per-
locators of notes used in the composition of songs and
formance of the ʿūd including the Arabian, the Persian
melodies,312 two authors, Kindī and Ṭaḥḥān describe ex-
and the Byzantine [rūmiyya] schools, and apolo-
plicitly the mounting of tie-frets on the neck of the in-
gizes to the reader not to be able to expound them all
strument.
due to the volume of explanations this would require,
The principal explanation for this (monumental) dis- and because these explanations would be understood in
crepancy lies firstly in the nature of Kindī’s Risāla fī-l- writing only by the “Wisest and the Most Open” of peo-
Luḥūn wa-n-Nagham which, opposed to the voluminous ple, whilst these “arts of teaching” can be transmitted,
treatises written by Fārābī, (ibn) Sīnā and Urmawī, is an better and faster than in a book, directly by the profes-
epistle dedicated to the ʿūd and to its apprenticeship. sional musicians (Ahl a-ṣ-Ṣināʿa).316
In his introduction, Kindī explains to the reader that While this epistle is clearly a method for beginners,
his aim is: tie-frets on the neck of the ʿūd are also intended for
them, 317 which resolves the contradiction between

312
Or mention the dasātīn without specifying the material of which 317
The most acute problem in the apprenticeship of fretless lutes
they are assumed to be made of. such as the ʿūd, the violin, etc. is the constant sounding of false notes
313
[Kindī (al-), 1965, p. 9] – As a reminder, the complete text of the in the first years of the apprenticeship. It is therefore totally accepta-
epistle is available in [Beyhom, 2010b, v. 1, p. 496–504]. ble to think about either fretting the fingerboard or marking the
314
[Kindī (al-), 1965, p. 15–22]. positions of the main notes for beginners. Knowing, however, that
315
[Kindī (al-), 1965, p. 27–29]. the technique of the ʿūd relies on the possibility of constant modu-
lation and interval modifications, no professional musician, were it
316
[Kindī (al-), 1965, p. 29–30].

153
NEMO-Online Vol. 5 No. 9 – November 2020

Kindī’s (and Ṭaḥḥān’s)318 explanations and the explana- musically speaking – vertical markers on the finger-
tions in Fārābī’s Kitāb al-Mūsīqī al-Kabīr and (ibn) Sīnā’s board of the ʿūd.
Kitāb a-sh-Shifāʾ as well as in the later works of Urmawī, Even the pretense to the existence of “tie-frets” for
Shirwānī and Lādhiqī, as the latter described techniques beginners is doubtful as Kindī was such an uncondi-
and divisions which had a wider, theoretical and practi- tional admirer of the Pythagorean “science” that he
cal, scope. could well have invented these tie-frets for beginners
Furthermore, with no indication in the extant litera- (and Ṭaḥḥān would have espoused this statement), or
ture for Kindī being a professional musician or a music used them for himself for learning how to play, while he
teacher, he would have had the usual difficulties in iden- and other theoreticians may have also used ligatures,
tifying the correct notes to play on the fretless ʿūd (and made of solid material or not, to materialize the stop-
to explain their locations to his patron), and may have ping points of the strings on the fingerboard of the in-
experimented these tie-frets as an original way to teach strument.
how to play correctly the instrument. Adding to this that the organology of the ʿūd creates
Kindī’s fretting and his location of pitches outside specific problems for these alleged frettings (as shown
the fretting zone also become coherent as, knowing that in Appendix B), no doubt remains possible about the
the practical system of Arabian music was more com- fact that the ʿūd was never fretted for performance pur-
plex than the simple Pythagorean division that he ex- poses – or that we have no indications ever mentioning
plained, he was compelled, out of intellectual honesty, such a use of tie-frets.
to show practical ways for their sounding. All in all, Sachs, Berner and Geringer were right in
While this problematic is further examined in Ap- their opposition to the “fretting thesis”. However, biases
pendix B about the organological particularities of the – as with the “Byzantine Church organ”321 – die hard322
instrument, we can conclude that the use of the dasātīn and myths will not be forgotten but are ever renewed
by early theoreticians and performers amounted to ma- because of the reputation of their authors, and because
terializing visual markings319 on the fingerboard, to en- of the wide distribution of their works.
sure a correct pitch for the most frequent notes on the Researchers in musicology have generally had, no-
ʿūd ,320 improving the precision of the performance and tably in maqām musicology, a simple pattern which was
of the composition. Physical tie-frets may have been the uncritical use of past research concurrently with the
used for beginners, or even with (beginner) theoreti- conscious or unconscious need to preserve these myths
cians wishing to experiment on their own the adequacy in order to ensure the supremacy of Western music over
of their descriptions – but lacking the ability to do so other musics.
correctly without tie-frets. It is evident that the silence of today’s musicology of
Farmer’s initial assertions about the fretting of the the maqām on this subject, and the perpetuation of the
ʿūd are not only unjustified, but clearly wrong for most myth of the fretting of the ʿūd is in the interest of West-
of them. To the very few sources stating the use of phys- ern music. While this is perfectly understandable – but
ical tie-frets (all in all Kindī and Ṭaḥḥān) we can oppose not acceptable – politically and socially, mere intellec-
multiple assessments by the same, or other authors, tual honesty compels to dismantle these myths in such
clearly showing that the dasātīn in question are but – a way as to avoid their further utilization.

today or in the Early days of Islam, would contend himself with one 320
As frequently observed on the marquetry of modern and con-
type of fretting. temporary ʿūd(s).
318
With both authors, tie-frets are intended for beginners, explicitly 321
Rosy Beyhom, in a private conversation, brings to my attention
with Ṭaḥḥān and implicitly with Kindī. Note that Ṭaḥḥān has prob- that (a-t-) Tīfāshī mentions the “organ” in his 34th chapter of Mutʿat
ably his source of inspiration from Kindī’s – and other writers of al-Asmāʿ… [Tīfāshī, 2019, p. 197-200] and mentions its use by the
which probably the Ikhwān a-ṣ-Ṣafāʾ and maybe a-s-Sarākhsī – Rūm (Byzantines) for big ceremonies and during prayer; this should
whose works are today lost but which Ṭaḥḥān copied at least indi- be further investigated hopefully in an upcoming publication.
rectly from al-Ḥasan al-Kātib. 322
Since this problematic often resurfaces in discussions among (or
319
Lines drawn on the fingerboard, thin threads of silk or other ma- with) musicologists, were they Western or local, influenced by
terials – which do not intervene in the performance as they do not Farmer’s (or Manik’s and, today, Neubauer’s) thesis on the subject.
help stopping the strings, but only show the positions for finger-
stopping them, etc.

154
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

GENERAL CONCLUSIONS Further descriptions by Fārābī and (ibn) Sīnā, the

two greatest theoreticians of the Golden Age of this civ-
Among all the instruments of the Arabian instru- ilization, confirm the Zalzalian model which is still in
mentarium, the ʿūd is the one which provoked the most use today, and for which the ʿūd, with its melodic versa-
controversy, because of its origins, its proportions, its tility and multiplicity of techniques, is (still) a perfect
tuning or its “fretting”, or concerning its part in Arabian receptacle as well as an inspiring theoretical tool for this
music323 and, by extension, its influence on European music.
music:
“Scholarship has tended to concentrate on the nature and ex- Such writers as Farmer and his gigantic ʿūd(s) which
tent of Arab musical influences on Europe, an area where pau- would have been described by Kindī and Ṭaḥḥān326 re-
city of evidence allows conflicting interpretations. But one ceived no criticism for decades. This seems to be com-
thing is clear: European interest in the Arab intellectual herit- monplace with the musicology of maqām music (includ-
age did not extend to music theory, and none of the major texts ing Byzantine chant), probably because for these musics
was translated. Turning to practice, however, a very different
what is said is less important than the moral authority
picture emerges. There is abundant lexical and iconographic
evidence for the European acquisition of a wide range of instru-
of researchers in this domain, as Neubauer in the article
ments, the lute (ʿūd), rebec (rabāb) and nakers (naqqāra) being reviewed in Part II of this dossier.
only the most obvious. […] In short, although the music of the
Arab courts must have provided a cultural model to be emu- Evidently other, ideological and societal factors in-
lated, musical influences were probably not unidirectional”.324 terfere with the needs of “science” which, in musicol-
ogy, seems to be an overrated characteristic.
Arabian music being the possible “missing link” be-
tween Ancient Greek music and European music, musi- As stated by my illustrious predecessor Abū-n-Naṣr
cologists such as Farmer endeavored to prove that the Muḥammad ibn Muḥammad ibn Tarkhān ibn Uzlagh al-
Early ʿūd was fretted, relying on a causal link between Fārābī:
the fretting of the instrument and the establishment of “To be an accomplished theoretician, whatever science is in-
volved, there are three conditions:
its “Pythagorean temperament” which would have been
then transmitted, through multiple contacts in Southern  To know all the principles of the given science.
France and Spain,325 to Europe and justifying thus the  To have the capacity to deduce the necessary consequences
of these principles in the beings (the data) which belong to
use of ditonism – if not of harmony with Farmer – in
this science.
their music.  To know how to answer erroneous theories and analyze
what is true from what is false and correct the errors”.327
As has been shown in this dossier, neither the fret-
ting of the ʿūd nor the Pythagorean division of Kindī and I would add: knowing that (one of) the burden(s) for
other Early philosophers apply for performance prac- the future generations of scientists will be to correct our
tice. Tie-frets may have been used for beginners (Kindī, errors today.
Ṭaḥḥān), or for theoretical purposes. The Pythagorean
division was inherited from the Greeks, and was the
only template the Arabs had to test their Early theoreti-
cal representations in string-length ratios. Kindī, the first
Arabian theoretician on music whose writings are ex- *
* *
tant, describes however “notes performed by singers”
which testify that, already in the Early phases of the Ara-
bian Civilization, praxis departed from this simplistic
model.

323
All these topics are explored in the “Annexes” of [Beyhom, 325
Notwithstanding the Byzantine influence on the Eastern – while
2010b], namely and respectively in appendices II.5 and II.6, appen- often changing – border, and its interaction with Arabian influence.
dix II.4 and appendix II.2.B. 326
See [Beyhom, 2011 ; Bouterse, 1979].
324
[Wright, Poché, and Shiloah, 2001, p. 805 (Arab music, §I, 3, 327
Translated from [Fārābī (al-), 1930, v. 1, p. 2].
IV)].

155
NEMO-Online Vol. 5 No. 9 – November 2020

APPENDIX A: THE ʿŪD, ITS COMPONENTS AND ITS Kindī adds complementary information further be-
PROPORTIONS 328 low in his text:
“'Then they adopted (sayyarū) the ratio which is after the third
I have explained elsewhere 329 that most, if not all,330 [of the length of the strings] – and it is the half – for the width
Early Islamic speculations on music theory used the ʿūd and it is the largest width it must be, and its position on the ʿūd
as the main vector for their explanations. In turn, as in- must be three fingers away from the end of the bridge in the
direction of the [‘following the’ – ilā mā yalī al-] strings [width
heritors of the Greek tradition through the translation
of the bridge = 3 – 7.5 + 6 = 1.5 fingers], and the reason for
enterprise set by Caliph al-Manṣūr in the 9th century,
this [is] that it is placed along [bi-muḥādhāt = at the proximity
Arabian philosophers and theoreticians adapted Greek of] the place where the strings are plucked, and this because
theories for this instrument (notably used as a “poly- this emplacement [on the ʿūd] is the widest and the most per-
chord” – as compared to a “monochord” – with strings fectly sounding. With regard the plucking of the strings, it is at
tuned in successive fourths), which became thus the three fingers from the [front of the] bridge [6 + 3 = 9 fingers
main vector for the maqām genos – and mode – theory. from the bottom] because it is the position of one of the parts
of the strings and it is its tenth”.335

First detailed descriptions of the ʿūd by Kindī To summarize, Kindī’s proportions for the ʿūd in this
epistle are (FHT 2) as follows (fractions are given in re-
The first known complete description of the ʿūd and lation to the total length L, the unit is “ff” (or “full fin-
its construction is found in the epistle Risāla fī-l-Luḥūn gers”):
wa-n-Nagham by 9th-century “Philosopher of the Arabs”
 Total length: 36 ff = L
Yaʿqūb ibn Isḥāq al-Kindī. 331 Kindī’s description says
 Total width: 15 ff = 10L/24 = 5L/12
(FHT 2:158):
 Total depth: 7.5 ff = 5L/24
“[and the] length [of the ʿūd] will be: thirty-six joint fingers –
 Length: 10 ff = 5L/18
with good thick [‘full’] fingers332 – and the total will amount to
 Soundbox length: 26 ff = 13L/18
three ashbār.333 And its width: fifteen fingers. And its depth
seven and a half fingers. And the measurement of the width of  Position of the bridge: 6 ff from the lower end =
the bridge with the remainder behind: six fingers. Remains the 4L/24 = L/6
length of the strings: thirty fingers and on these strings take  Total speaking length: 30 ff = 20L/24 = 5L/6
place the division and the partition, because it is the sounding  Speaking length above soundboard: 20 ff = 5L/9
[or ‘the speaking’] length. This is why the width must be [of]  Optimal plucking point (from the lower end): 9 ff =
fifteen fingers as it is the half of this length. Similarly for the
L/4
depth, seven fingers and a half and this is the half of the width
and the quarter of the length [of the strings]. And the neck must
 Soundbox: width/length = 15/26, or around 3/5;
be one third of the length [of the speaking strings] and it is: ten depth/width = 1/2
fingers. Remains the vibrating body: twenty fingers. And that
the back (sound box) be well rounded and its ‘thinning’ (kharṭ) Note, however, that the proportions of the total
[must be done] towards the neck, as if it had been a round body depth to the total width, then to the total speaking
drawn with a compass which was cut in two in order to extract length is 1:2:3, or the two first tetradic ratios based on
two ʿūd(s)”.334 the first three elements of the tetrad.

328
This appendix relies on [Beyhom, 2011]. members of the caliphal family, depended heavily on these transla-
329
In [Beyhom, 2016]. tions” – in [Adamson, 2011]. More information on Kindī is pro-
330
Very few theoretical descriptions were, in Early Islam (the civi- vided in the main text.
lization), undertaken using the neck of the ṭunbūr, mostly for music
332
(Reminder:) Literally “full fingers with good flesh”.
of particular areas and periods – see the appendix on the ʿūd and 333
The shibr (singular of ashbār, “span” in English) is a measure-
the ṭunbūr in [Beyhom, 2010b] and the “First Interlude” in the main ment unit which equals roughly 20 cm. It equates to the length be-
text of this dossier. tween the tip of the thumb and the tip of the auricular finger when
331
“Abū Yūsuf Yaʿqūb ibn Isḥāq Al-Kindī (ca. 800–870 CE) was the stretched flat and in opposite directions. The shibr otherwise
first self-identified philosopher in the Arabic tradition. He worked measures 12 fingers (which equates to 36:3 in Kindī’s description):
with a group of translators who rendered works of Aristotle, the a “full” finger should be about 2 cm in width.
Neoplatonists, and Greek mathematicians and scientists into Ara- 334
Translated from the original Arabic [Kindī (al-), 1965, p. 11].
bic. Al-Kindī’s own treatises, many of them epistles addressed to 335
Translated from the original Arabic [Kindī (al-), 1965, p. 15].

156
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

Description of the “Modern” ʿūd by Ṭaḥḥān  the two Bīṭār ʿūd(s) made by the Lebanese luthier
for, respectively, Saad Saab (FHT 8:164) and Amine
Whenever Kindī’s ʿūd appears to be a monoxyle lute-
Beyhom (FHT 7:163) (the latter being an electro-
type instrument,336 the first extant detailed description
acoustic instrument).
of the “modern” ʿūd337 is Abū-l-Ḥasan ibn a-ṭ-Ṭaḥḥān’s
Such “modern” instruments may have even more
(11th century):338
“inharmonic” proportions as Ṭaḥḥān’s, with a resulting
“The dimensions of the lute should be as follows [see FHT
quality of sound345 which is probably different, but not
3:159]: its length should be 40 aṣābiʿ339 maḍmūma340. Its width
should be 16 aṣābiʿ maḍmūma. Its depth should be 12 aṣābiʿ
necessary less pleasant than with (Pythagorean influ-
maḍmūma. The bridge should be placed at about 2 aṣābiʿ enced) ʿūd(s) with “harmonic” proportions. I show else-
[“iṣbaʿayn” – the flexion of iṣbaʿ for the dual case] odd from the where346 that this evolution from the purely theoretical
bottom. The neck should be 1 shibr + 1 ʿaqd341 in length. The application of Pythagorean mathematics to more prac-
pegs should be eight unless there is a zīr ḥād string342 when tice-oriented methods and proportions applies also to
there will be ten strings, 343 but this is not known in our
Arabian music theory.347
times”.344

If we compare the proportions of Ṭaḥḥān’s ʿūd to

Kindī’s, we note that the ratio (Total)
depth/width/speaking length of the string is no more
1:2:3 but (FHT 3:159) 12:16:38, which is equivalent to
3:4:8, slightly further from the “ideal” Pythagorean pro-
portions.
This also applies to modern ʿūd(s) with proportions
shown on FHT 4:160 to FHT 8:164 namely:
 The ʿūd described by Khulaʿī (beginning of the 20th
century – FHT 4:160 and FHT 5:161), FHT 1 “Padauk/Walnut Body with Cedar top [ʿūd with]
 the ʿūd of the well-known Munīr Bashīr (2nd half of Amazing wood in wood inlay (Maker: Farouk Shehata,
the 20th century – FHT 6:162), 1993)”.348

336
And most probably a forerunner of the barbaṭ. first four strings were called (from the lowest to the highest – acousti-
337
In Ṭaḥḥān’s description of the ʿūd, as in the modern instrument and cally and conventionally) the bamm, the mathlath, the mathnā and the
unlike Kindī’s description, the back (or the shell) is assembled from thin zīr. Whenever today’s ʿūd(s) incorporate six (or seven) (double – except
strips (ribs) of hardwood, joined (with glue) edge to edge to form a deep generally for the lowest, acoustically) courses of strings (FHT 5:161 and
rounded body, and is at a later stage of its construction joined to the FHT 7:163), it seems that, in Fatimid Egypt at the time of Ṭaḥḥān, this
monoxyle neck. fifth string was still not in use, or came to be in disuse, which may seem
338
Ṭaḥḥān was a musician of high repute during the Egyptian Fatimid less likely but is possible; note that iconographic sources show five
Period, who died sometime after 1057. He was mainly a singer and an strings as early as the 10th-11th centuries – see [Beyhom, 2010b, v. 1,
instrumentalist, and is with Kindī one of the very few having described p. 92] and [Farmer, 1966b, p. 49], the latter showing six courses. The
the ʿūd and its facture. His work entitled Ḥāwī al-Funūn wa Salwat al- need for the ḥād string was mostly theoretical in the time period of the
Maḥzūn is in two parts, the second of which being about praxis. Forerunners (see footnote 45:119 for time periods for Arabian music
339
Plural of iṣbaʿ, Arabic for “finger”. theory), to complete the double-octave. We find a mention of five
courses of strings in the practice of the instrument in Urmawī’s epistle
340
The verb ḍamma means “to join”, maḍmūm, or munḍamm meaning
A-r-Risāla a-sh-Sharafiyya [Urmawī (d. 1294) and [Jurjānī (al-)], 1938,
“joined” or “tightened”. Farmer’s notable error was the confusion be-
v. 3, p. 110] (reedited as [Urmawī (d. 1294) and [Jurjānī (al-)], 2001]),
tween “joined” and “doubled”, which made him double the sizes of the
in the 13th century.
ʿūd(s) he described in his “The structure of the Arabian and Persian lute
in the Middle Ages” [Farmer, 1939b]. (This is detailed in [Beyhom,
343
In fact, five courses with two identical strings each.
2011].)
344
[Ṭaḥḥān (ibn a-ṭ-~ al-Mūsīqī), 1990, p. 172].
341
The ʿaqd is a particular Arabian value which in context equates to a
345
The tone-color (or timbre) for example, although this characteristic of
“unit” (1) or to “ten” (10): in this context it is equivalent to “10 joined sound depends on other, organological and environmental factors as well.
fingers”.
346
Mainly in [Beyhom, 2010b], and partly in this dossier.
342
The (theoretical) 5th string of the ʿūd, the zīr ḥād (or simply ḥād –
347
See also [Hilarian, 2005] for a comparative study of the Malay-Lutes
“sharp” – or “2nd zīr” for some authors) is already cited by Kindī in his (Gambus) with the Arabian lutes, which gives an insight into the variety
Kitāb al-Muṣawwitāt al-Watariyya min dhāt al-Watar al-Wāḥid of shapes of short-necked lutes together with [Hellwig, 1974] (for West-
ilā dhāt al-ʿAshr[at] Awtār [Kindī (al-), 1962b, p. 78]. As a reminder: the ern lutes).
348
Retrieved 20/10/15 from http://www.mikeouds.com/oudpics.php.

157
NEMO-Online Vol. 5 No. 9 – November 2020

FHT 2 (Al-) Kindī’s description of the ʿūd, in “full finger (iṣbaʿ – pl. aṣābiʿ) thickness” measurements, and deduced (calculated)
proportions.349 The same procedure is used for the “Harmonic division” shown on Fig. 9:131. (“Vibrating string” = speaking length
of the string.)

349
First published in [Beyhom, 2011].

158
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

FHT 3 Drawing of the ʿūd described by Ṭaḥḥān.350 (Reminder: “The dimensions […] should be as follows: its length should be
40 aṣābiʿ351 maḍmūma. Its width should be 16 aṣābiʿ maḍmūma. Its depth should be 12 aṣābiʿ maḍmūma. The bridge should be placed
at about 2 aṣābiʿ odd from the bottom. The neck should be 1 shibr + 1 ʿaqd in length. The pegs should be eight unless there is a zīr
ḥād [double] string and ten strings [in all], but this is not known in our times.” Note also that “vibrating string” = speaking length
of the string.)

350
First published in [Beyhom, 2011].
351
Arabic grammar is complicated: the plural for more than 10 aṣābiʿ (or anything or anyone) is like the singular form, iṣbaʿ. Hence: 40
iṣbaʿ, 16 iṣbaʿ, 12 iṣbaʿ, etc., but also imraʾa (a – or one – woman), imraʾatayn (two women), three (to ten) nisāʾ and 11 (and more) imraʾa!

159
NEMO-Online Vol. 5 No. 9 – November 2020

FHT 4 Revision of the ʿūd described by Khulaʿī in his Book of Oriental Music [Khulaʿī (al-), 1904]. The measurements are those
taken from the original drawing (next figure).352 (“Vibrating string” = speaking length of the string.)

352
First published in [Beyhom, 2011].

160
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

FHT 5 Depiction of a ʿūd in [Khulaʿī (al-), 1904, p. 52].

161
NEMO-Online Vol. 5 No. 9 – November 2020

FHT 6 Drawing of the ʿūd of Munīr Bashīr (1957 – described in [Rashīd, 1999]). (“Vibrating string” = speaking length of the
string.)

162
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

FHT 7 Front and side views of the Bīṭār 2001 electro-acoustic ʿūd with thin soundbox, engineered and re-designed by the author
and crafted by Lebanese luthier [string instrument maker] Georges Bīṭār in 2001. This instrument is a straightforward adaptation
of the physical elements of which the Bīṭār-Saab ʿūd (FHT 8) is made. No Pythagorean proportions can be seen were it for this
instrument or for Khulaʿī’s in FHT 4 and for Bashīr’s in FHT 6.

163
NEMO-Online Vol. 5 No. 9 – November 2020

FHT 8 Drawing of the Bīṭār-Saab ʿūd. The original instrument was made by the Lebanese luthier Georges Bīṭār in 2001-2002
following the specifications of ʿūd teacher Saad Saab for teaching purposes at the Lebanese National Conservatory. The transverse-
slice view is from the electro-acoustic Bīṭār 2001 ʿūd shown in FHT 7.

164
NEMO-Online Vol. 5 No. 9 – November 2020

FHT 9 Proportions of the strings of the ʿūd according to Kindī in the Risāla fī-l-Luḥūn wa-n-Nagham357: s1 to s4 are the cross-
sections, d1 to d4 are the diameters of the strings from zīr to bamm. The proportions of the sections from zīr (right) to bamm (left)
stand as 1:2:3:4. The intermediate strings are called the mathlath (s3) and the mathnā (s2).358

FHT 10 Proportions of the strings of the ʿūd according to Ikhwān a-ṣ-Ṣafāʾ in their Fifth epistle (“On Music”)359: s1 to s4 are the
cross-sections, d1 to d4 are the diameters of the strings from zīr to bamm. (See figure above for the names of the strings.)360

357
Originally published in [Beyhom and Makhlouf, 2009].
358
Following the hypothesis that the diameters of the twisted strands of guts remain unchanged after the twisting – see Fig. 26:147 and
corresponding footnote.
359
See [‫الصفاء‬, S.D.] or [Dieterici, ‫اخوان الصفاء‬, and Ih̲wān al-Ṣafāʼ, 1865, p. 117–118]; for the Ikhwān note [Wright, 2001h]: “A 10th-century
group of Islamic encyclopedists of Ismaili tendencies centred on Baṣra, one of whose epistles (Rasā’il) deals with music. Unlike most other
music theorists of the 10th and 11th centuries, the Ikhwān al-Ṣafā’ were chiefly concerned with the neo-Platonic and Hermetic aspects of the
Greek heritage. Their work is of some interest for its scientific aspects (in particular the theory of the spherical propagation of sound) and
for its treatment of musical practice: for example, following al-Kindī, the discussion of the lute gives, in addition to a (simple Pythagorean)
fretting, details of proportions and construction. But the most characteristic features of their work, again following al-Kindī, are to be found
in their study of cosmology, where the notion of cosmic harmony (based on the Pythagorean concept of the primacy of number and
numerical relationships) is the unifying principle in the discussion of such topics as the music of the spheres, the moral and medical effects
of music, and the sets of natural phenomena (including the elements, winds, humours, colours and perfumes) to which the rhythms and
the four strings of the lute could be related.”
360
Following the hypothesis that the diameters of the twisted strands of guts (or silk) remain unchanged after the twisting – see Fig. 26:147
and corresponding footnote.

166
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

FHT 11 Proportions of the strings of the ʿūd according to (ibn a-ṭ-) Ṭaḥḥān’s Ḥāwī-l-Funūn wa Salwat al-Maḥzūn361: s1 to s4 are the
cross-sections, d1 to d4 are the diameters of the string from zīr to bamm. The proportion are originally given by weight of the string
by this theoretician, which corresponds to proportions by the section (with the weight – if the material of the gut is homogeneous
– being proportional to the section of the string and to its length as the product of the multiplication of the two values equals the
volume of the string). Ṭaḥḥān also proposes the same proportion “by sight” – meaning by their thickness or diameter. The corre-
sponding diameters and thicknesses are shown as “Ṭaḥḥān II” in THT 2:177 while the set shown in this figure – which is more
realistic with regard to a possible fretting of the ʿūd – corresponds to “Ṭaḥḥān I” in the same THT 2, and in THT 3:177.

FHT 12 Proportions of the (silk) strings of the ʿūd according to Kanz a-t-Tuḥaf: s1 to s4 are the cross-sections, d1 to d4 are the
diameters of the string from zīr to bamm.362 (Not including the thinnest string – the ḥād.)

361
Originally published in [Beyhom and Makhlouf, 2009].
362
Following the hypothesis that the diameters of the twisted threads of silk remain unchanged after the twisting – see Fig. 26:147 and the
corresponding footnote.

167
NEMO-Online Vol. 5 No. 9 – November 2020

A digression: When theory contradicts prac- shall simplify the problem by positing that the con-
tice (and facts) tact points between the string and the fingerboard
as well as between the string and the nut or the
The lengthening of the strings as a result of con-
bridge are ideal (points).368
current stopping of the strings and of the presence
of solid frets such as those described by Kindī is ex- Provided (see FHT 14:169) the total vibrating
amined here. length of the string is L0, and that the string is
stopped somewhere on the fingerboard at a contact
Fārābī, as well as (ibn) Sīnā and his student (ibn)
point dividing it in 2 parts LS0’ (Length of the –
Zayla 363 , all three state that there is a significant
lengthened – string in direction of the nut) and LC0’
modification of the tension in the strings of the ʿūd
(Length of the – lengthened – string in direction of
when these are stopped: I show that this modifica-
the bridge), and the projections of these string-parts
tion is in fact insignificant when the instrument is
on the fingerboard of the ʿūd be LS0 and LC0 (which
not fretted.364
are the corresponding lengths of LS0’ and LC0’ when
The arguments of the three authors are similar, these are not lengthened).369
of those below (ibn) Zayla’s (see FHT 14:169 and
FHT 15:169):
“If the mushṭ [bridge] – or the anf [nut] – is so high that the
strings would be far from the fingerboard, 365 stopping the
string will lengthen it because, instead of forming a straight line
it would form 2 lines delineating the unstopped string. Thus,
and the sum of the lengths of two sides of a triangle being
greater than the length of the third side, the string can but
lengthen, and lengthening modifies the register [a-ṭ-ṭabaqa]
and produces a higher sound366”.367

FHT 15:169 shows a ʿūd in cross-section (the

missing parts are shown in dashed lines), with
raised bridge and nut for clarity.

While the height of the bridge does not exceed 8

mm on modern ʿūd(s), and the height of the nut does
not exceed 1 mm (see an example of Modern ʿūd in
FHT 13:168 and compare with the neck of Hamdi FHT 13 The neck of a Modern ʿūd370
Makhlouf’s ʿūd with a raised nut in FHT 22:177), we
363
Among other authors. CIM09 were still available (on the 21st of October 2020) at (respec-
364
For realistic proportions of ʿūd(s): very high nuts or bridges can tively) http://www.hamdi-makhlouf.com/cim09/video-1-kindi.
affect the results shown below in the text. mp4 and http://www.hamdi-makhlouf.com/cim09/video-2-tahhan.
365
Although (ibn) Zayla’s statement seems coherent, it lacks of pre- mp4 – and referenced as [Makhlouf, 2009a ; 2009b]. Two other cop-
cision about the exact height(s) for bridge, nut and “ties”: for mod- ies, subtitled in English by Amine Beyhom, have been made avail-
ern ʿūd(s) and as shown below, the exact height of the bridge (or able on YouTube at https://youtu.be/d7TTlnH_pKM and
the nut) plays a major role for the perception of the difference be- https://youtu.be/demT-hpcX1s. These videos are practical demon-
tween two pitches. strations of some organological problems raised by the fretting of
366
Note that a lengthened string, strictly speaking – and theoreti- the ʿūd. For both Kindī and (ibn a-ṭ-) Ṭaḥḥān, two divisions of the
cally – and with all other variables (except the frequency) being fingerboard, “Harmonic” and Pythagorean with two different sets
equal, would sound “lower” acoustically as shown by Taylor’s for- of strings, are experimented.
mula expounded further; this would occur only if the tension of the
369
We will also contend that the effective length of the unstopped
string remains unchanged, in which case the frequency would drop string is (nearly) equal to its projection L0 in FHT 14.
in order to compensate the lengthening of the string (in the for-
370
Retrieved 20/10/15 from [Anon. “Mike Ouds - My Ouds Page”]:
mula). “This oud was made in 1925 by the oudmaker, Mohamad el-Hif-
367
[Zayla (ibn), 1964, p. 76]. nawi. It was owned by Mohamad el-Qasabji […].”
368
For this and other organological procedures about the fretting of
the ʿūd, the two videos originally made by Hamdi Makhlouf for the

168
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

FHT 14 Lengthening of a ʿūd string when stopped on the fingerboard. Above: length-section of the ʿūd with unstopped string.
Below: same as above, with a stopped string. The bridge and the nut are oversized in height. The circled part is magnified in the
next figure. (“Vibrating string” = speaking length of the string.)

FHT 15 Lengthening of a ʿūd string when stopped on the fingerboard – magnified length section of the fingerboard. The thickness
of the stopping finger is approx. 2 cm, the tip of the finger is approx. 1 cm. Lengths of the total speaking length of the string (L0=60
cm) and of the string-part over the soundboard (LC0=2 L0/3=40 cm) and over the neck (LS0=L0/3=20 cm) are coherent with
Kindī’s description, and with the proportions of modern ʿūd(s)

169
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

ligatures to obtain the correct pitch. Yet this formulation would be much more difficult (he would have to be
is inconsistent with “physical” tie-frets the thickness of much more precise in his performance and exert much
which is not negligible – such as Kindī’s and Ṭaḥḥān’s. more pressure on the string to be able to stop it cor-
As all performers on fretted lutes (such as Western rectly). Furthermore, modification in pitch would occur
lutes, guitars and mandolins, etc.) know, quality emis- in such case due to the lengthening of the string.
sion of notes on these instruments means stopping the (See also Appendix B for more details on “ties”, “lig-
string just before the tie-fret or fret, as close as possible atures” or “frets”.)
to it without compromising the quality of the sound.
FHT 16:172 illustrates these two specific cases as *
* *
length sections of the fingerboard, on a ditonic division
(Pythagorean) of the fingerboard materialized as ties of
homogeneous thickness = 1 mm.
The main reason for the stopping of the string before
the tie is acoustical and organological: fingertips376 have
incompressible thicknesses. When stopping the string
directly on the tie the borders of the fingertips will inev-
itably exceed this point by a few millimeters377 which
creates an unpleasant buzzing sound. The best sound is
obtained when the string is stopped a few millimeters
before the ligature.378 Therefore, an indication for stop-
ping the strings on the ligature is an indication that the
“tie-frets” are line markers drawn on the neck, or would
be very thins tie-frets.
As for playing between ligatures, FHT 17:172 shows
that, even with “thin” tie-frets only 1 mm thick,379 the
string will effectively be stopped on the tie below it.380 In
Detail from “Two men having fun with music” (c. 1300)
the figure, this would be the wusṭā when pressing the from Staatsbibliothek zu Berlin, Diez A, f.71, S11-2; (copyright
string between the sabbāba (index) and the wusṭā (mid- Staatsbibliothek zu Berlin, Preussischer Kulturbesitz, Orientabtei-
dle finger).381 lung).382

Note that there exists a possibility, in the case of a

(very) high bridge, that the string stopped closer to the
sabbāba (index) in FHT 18:172 would not even make
junction with the wusṭā (middle finger). This would be
an exceptional case {for ʿūd(s)} and inconsistent for the
organology of the instrument as the performer’s task

376
Which can be estimated as 1 cm, for an estimated 2 cm for the pitch (note) of the upper tie-fret (the wusṭā in the figure). However,
finger. stopping the string near the sabbāba (and after it, in the space be-
377
Due to the thickness of the fingertips. tween sabbāba and the wusṭā) will (1) produce an unpleasant sound
378
This observation (which is commonplace among performers on and (2) can in extreme cases (see below in the text) fail at stopping
fretted lutes) comes from my experience as a guitarist, but also from the string on the wusṭā.
the aforementioned reconstruction of the frettings of Kindī and
381
This is in fact the main reason for mounting frets on a lute, as
Ṭaḥḥān with Hamdi Makhlouf. the performance will be much easier, although limited melodically,
379
Compare this to ties 8 mm thick as the ones advocated by because the performer needs no more be (so) precise in his stopping
Maalouf for Kindī’s “fretting” in the section of Appendix B below of the string. An approximate stopping precision is enough to emit
entitled “Impracticality of the performance with dense divisions”. an acceptable sound. Note also that, in the case of stopping nearer
380
To the left in the figure. On fretted instruments, to sound the to the “higher” fret (to the right – the sabbāba in FHT 18), what
desired note, the string must be stopped just before the “fret” (liga- changes is mostly the quality of the emitted sound (which becomes
ture) corresponding to it; this means that whatever the position be- worse – with regard to traditional performance).
tween the sabbāba and the wusṭā, the note sounded would be the
382
From [Tsuge, 2013, p. 258].

171
NEMO-Online Vol. 5 No. 9 – November 2020

FHT 16 Two positions for the stopping finger on the tie of the sabābba (heights of bridge and nut are realistic = correspond
loosely to the measurements of ʿūd(s) nowadays). The 1st position (theoretical and to the left, mentioned by all early authors) is not
advised if the ʿūd is mounted with solid tie-frets (“deafness” of the sound occurs), but is coherent with the use of fretless instruments.
The 2nd position (to the right in dotted lines) is the (approximate) correct position for a fretted instrument (such as a guitar). The
thickness of the fret (tie-fret, ligature) is 1 mm.383

FHT 17 Advised position to make

the string sound at the length of the
wusṭā (at 27 L0/32), between the tie-
frets of the sabbāba (index) and the
wusṭā (middle finger). Organological
configuration (proportions) is “nor-
mal”: the string is mounted the closest
possible to the tie-frets (and to the fin-
gerboard) for a better quality of the
performance; the stopping occurs just
before the wusṭā. (“Tie” in the figure =
“tie-fret”.)

FHT 18 Not recommended stopping

of the string at the length of the wusṭā
(27 L0/32), between the tie-frets (“ties”
in the figure) of the sabbāba (index) and
the wusṭā (middle finger). The stopped
string will “sizzle” or “crackle” (will be
accompanied by “squeaks” according to
Kindī – see Part II). If the bridge is over-
sized in height, it is possible that the
string will not even touch the tie of the
wusṭā.

383
The tie-frets are not, in these figures, winded twice (as described by Kindī – see Fig. 25: 146, and as attested for example for the sāz) as
to avoid additional complexity of the graphic representation.

172
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

APPENDIX B: ORGANOLOGICAL CLARIFICATIONS by eminent musicologists such as Farmer and

Neubauer, led to the broadly accepted assertion that
The fretting of the ʿūd is one of the most contro- the early “mediaeval” – in the Western acceptation
versial issues in Arabian musicology. Many studies – ʿūd was fretted.
published in the 20th century have conflicting opin-
ions on a subject the essence of which appearing to Nonetheless, descriptions of early practice con-
stretch beyond organological matters. tradict this assertion. Consequently, the main ques-
tion explored in this appendix is: How would the in-
Among manuscripts of the Arabian Golden Age strument respond should it be fretted as described
from the 8th to the 11th centuries, only two describe by Kindī and Ṭaḥḥān?
a “fret” system made from ligatures tied at specific
places on the fingerboard of the ʿūd. As seen in To answer this question, practical organological
questions – notably Sachs’ remark on the shape of
the main text, the first description is from one of
(al-) Kindī’s epistles and dates to the 9th century. the neck of the ʿūd – 385 are examined.
The second is the 11th-Century description from (ibn
*
a-ṭ-) Ṭaḥḥān. * *

Both authors give relatively complete descrip- The fretting of the instrument was undertaken in
tions of ʿūd tie-frets contradicting significant asser- 2008-2009 by Hamdi Makhlouf. Two videos386 were
tions of philosophers and theoreticians such as (al-) produced, showing the making of four different fret-
Fārābī, in the 9th and 10th centuries – who was tings, with two sets of strings (for Kindī and Ṭaḥḥān
known as the ‘Second Master’, Aristotelēs being the respectively) and two tunings – Pythagorean and
first – and (ibn) Sīnā, known to the West as “Harmonic” – described by Kindī in his Risāla fī-l
Avicenna, and nick-named ‘the Commentator’ (of Luḥūn wa-n-Nagham.
Aristotelēs), and also with other later writers such
as Urmawī, a musician and theoretician of the 13th The following two sections address general or-
century, and Shirwānī in the 15th century. Further- ganological problems concerning the fretting pro-
more, only few contemporary authors have studied cess, which should clarify, in the third section, the
the possibility of the ʿūd fretted according to ancient next examination of yet another difficulty arising
descriptions. from the multiplication of tie-frets on the neck of
the instrument.
In Early Arabian writings about music, both the-
ory and practice use the instrument as a common These clarifications are most needed for the pur-
denominator. Recent research384 has also shown the pose of this dossier and are justified, notably, by the
antecedence of the ʿūd and its influence on the con- zeal of Re-Orientalist musicologists who, while con-
temporary musical repertoire. currently adopting the myth of the fretting of the
ʿūd, demonstrate that the Arabian divisions of the
Significant peculiarities of the modern instru- fingerboard were “perfect”.
ment, such as the semi truncated conical shape of
the neck, possibly a smaller gap between strings and *
* *
fingerboard, but also practice of subtle variations of
intonation, different from any temperament-based
systems, all contradict the premise that frets, or ac-
tual physical (consistent, thick) ligatures were used.
However, the fretting thesis, which was promoted
384
[Beyhom, 2005]. 386
See footnote no. 368:168 for the two videos made by Hamdi
385
See footnote no. 154:136. Makhlouf for the CIM09. These videos are practical demonstrations
of some organological problems raised by the fretting of the ʿūd.

173
NEMO-Online Vol. 5 No. 9 – November 2020

About organological difficulties arising from ON THE GLOBAL ORGANOLOGICAL IMPRACTICALITY

the use of frets on the ʿūd OF TIES ON THE SHORT, SEMI TRUNCATED CONICAL NECK
OF THE ʿŪD
Ties (“tie-frets” or “ligatures”) are commonly used on
Let’s begin with common sense reasoning: today, the
lute-type instruments, be it on Western lutes or others. As
ʿūd is not fretted (or mounted with tie-frets) and the best
a general observation: frets are used mainly on long-
Conservatoire technicians are precisely very proud that
necked lutes with a neck the two upper and lower sides
they can perform chords or arpeggios with an accurate
(the edges of the neck) of which are (almost) parallel. This
(i.e. as when playing on a well-tuned guitar) pitch and a
feature is also necessary for the Western exception, the
“clean” (with no crackling or sizzling) sound on a fretless
fretted lute. When the neck is in the form of a truncated
instrument.
semi-cone (FHT 20:175), typical difficulties arise for the
fretting procedure. Furthermore, some fretted long- As already explained, the ʿūd has a short neck390. This
necked lutes (for example the Iranian tār, setār and the implies that:
dotār, exception made for the Turkish sāz) have a groove  The span (ambitus) on a string is generally reduced,
at the back of the neck to make it easier to tie the nots of in traditional performance, to a just fourth (or
the tie-frets on the instrument, while this procedure is not fifth).391
described anywhere for ties on the neck of the early  This in turn implies that there is not much place for
ʿūd.387 tie-frets – much less as with a ṭunbūr – if these have
Moreover: we have learned from Fārābī that some substantial dimensions.
notes on the ʿūd are sounded when the strings are stopped  This also implies that a non-equal-temperament
between the “usual” dasātīn (ligatures), while others are division with vertical markers (and with a regular
sounded when the strings are stopped on the ligatures388 tuning in successive just fourths) can not as easily
– not to mention (ibn) Sīnā’s explanations about porta- provide octave and fifth correspondences and, more
menti with the strings of the ʿūd389. generally, equivalences between the notes of one
All in all, using tie-frets restricts the performance to octave and another octave.
predetermined series of notes which narrow the possibil-  This in turn implies that, in order to obtain these
ities of melodic expression for the performer. To solve – equivalences a multiplication of the tie-frets is
partly – this technical limitation, more tie-frets can be necessary as for example in Kindī’s division in the
added. However, this raises new organological problems Risāla fī Khubr Ṣināʿat a-t-Taʾlīf shown in Fig. 8:127.
not thought through by Orientalist (and here, mainly, by  Furthermore, the neck of the ʿūd has a semi
Re-Orientalist) maqām musicologists, more concerned by truncated conical form, slightly flattened and
their theoretical demonstration than by practical “de- limited by the nut on one side and by the body on
tails”. the other side (FHT 19:175 and Fig. 9:131). In the
course of performance, it is most probable that (even)
These questions are mainly about the size and the
a firmly knotted tie-fret will not remain in its original
numbers of the “tie-frets”. To show these difficulties, I will
position – because of the lateral friction on the tie-frets
try to apply indications about the tie-frets as provided by
– (FHT 20:175), which will make it inoperative.392
Kindī, in conjunction with the emplacement of the stop-
(This applies even more when the hygrometry is
ping finger(s).
high.)

387
The tying of the tie-fret is always a delicate operation: see for exam- neck (see for example the two plates inserted in [Hellwig, 1970, p. 64–
ple the YouTube videos [Zapico, 2015 ; PaololiutaioPD, 2010 ; Shep- 65]) which prevents the tie-fret from losing its adherence to the neck as
herd, 2016 ; Carey, 2017 ; Espinoza, 2015]. in FHT 20:175.
388
See Quote 1:138, Quote 2:139, Quote 11:140, Quote 13:148 and 391
It is hence more interesting to explore the infinite possibilities of “micro-
Quote 14:148. modulation” within the span of one string, i.e. one fifth or a little more.
389
Quote 15:148. 392
See for example the Video no. 2 (Ṭaḥḥān – see footnote no. 394)
390
Shorter than the neck of the Western lute. Note that while Western between 5:36 (mm:ss) and the end, especially the third fret from
lutes have frets, we do not know if this was the case from the beginning the right, and more precisely around 6:19. The “tie-fret” of the binṣir
(which would be surprising). Moreover, Western lutes have – unlike the (third from the right) moves constantly while Hamdi Makhlouf tries
ʿūd – wide, almost (semi-)cylindrical necks, with nearly no sloping of the to play a melody on his fretted instrument.

174
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

FHT 19 Neck of a ʿūd (based on Khulaʿī’s ʿūd

shown in FHT 5:161) mounted with a single tie-fret
(“tie” in the figure); A and Aʾ delineate a cross sec-
tion shown in (free) perspective view in FHT 20.393

FHT 20 Free perspective view of section A-A’ in the previous figure. To the left: firmly knotted tie(-fret) on the neck. To the right:
tie-fret displaced towards the nut because of wanted or unwanted (which may happen during the performance) lateral thrust: in
this case, the tie-fret adheres no more to the surface and fingerboard of the neck, and becomes inoperative.

393
In the case of slightly conical necks, the fret can be tied on the thinner part, just before its intended position, then displaced towards
(and on) the intended position, which will ensure a better fixation, but the problem of displacement remains (as, for example, with the
Video no. 2 in the previous footnote); this could, furthermore, result in scratches on the neck which is highly not recommended.

175
NEMO-Online Vol. 5 No. 9 – November 2020

PRACTICAL DIFFICULTIES FOR FRETTING THE ʿŪD – AN EXPE- The two videos are explicit about the whole fret-
RIMENT394 ting procedure, and about the difficulties met by
Other problems arise when trying to reconstruct Makhlouf during this process.
the frettings as proposed in the Arabian literature,
such as with Kindī and Ṭaḥḥān – a task that no Ori-
entalist (or Re-Orientalist) musicologist seems to
have undertaken before our experiment with Hamdi
Makhlouf in 2009.395 This experiment aimed to re-
create (or simply create) the frettings of the two
early authors on a modern ʿūd – here the instrument
of Makhlouf shown in FHT 21.

Two sets of strings (and “tie-frets”) were used

(FHT 26:180 & FHT 27:180) which were the closest
in diameter, from the available gut strings from Sa-
varez,396 to the proportions proposed by Kindī and
Ṭaḥḥān (respectively FHT 9:166 and FHT 10:167)
with, for the latter, the two proportions per weight
(= per section) or per diameter (THT 2 & THT
3:177).

Two videos were produced about the fretting

process and its results. The first video is named from
here on “Video no. 1”. In this video, Makhlouf uses
the set of strings “Ṭaḥḥān II” (THT 2: – also named
“Kindī II”). The ūd is stringed,397 then fretted first
according to the “Harmonic” system of Kindī, then
according to the Pythagorean system of the same au-
thor. After mounting each set, Makhlouf plays an
improvised melody by positioning the fingertips of
his left hand firstly directly on the frets, and secondly
by positioning them before the frets.

The same procedure is applied in the second FHT 21 ʿūd used by Hamdi Makhlouf to test the frettings.
video (“Video no. 2”), using the set of strings
“Ṭaḥḥān I” (THT 3:177).

394
(Reminder:) This section relies on the two subtitled in English made for Early period instruments, do not result from the same pro-
videos available on YouTube at https://youtu.be/d7TTlnH_pKM cedure as earlier gut strings; specifically, they are made according
to a process originating in 16th-Century Italy, which is: gut strands
(for Kindī) and https://youtu.be/demT-hpcX1s (for Ṭaḥḥān).
are sliced (sometimes twice) in the direction of the length and hung
395
By Makhlouf with advice from the author. This experiment was to dry. A small rock of given weight is attached to the sliced strands
part of a wider research undertaken with musicologist and ʿūd(ist) at the bottom, after which the small rock is rotated till the strands
Hamdi Makhlouf for the CIM09 (Cinquième Congrès Interdiscipli- are shortened for a given length. This procedure ensures that there
naire de Musicologie, Paris, Octobre 2009) conference, and the pur- are no gaps between the (sliced) strands of gut. The diameters of
pose of which was precisely a better understanding of organological the resulting strings are then evened with a special tool to make
specificities with regard the fretting of the ʿūd. The research is doc- them homogeneous all long, then oiled or varnished. Strings made
umented in [Beyhom and Makhlouf, 2009] and in the aforemen- following this procedure are generally more resistant and sound
tioned videos. better than gut strings made in the traditional way.
396
Note that Richard Dumbrill, who has an extensive experience in 397
And tuned in successive fourths.
the making of gut strings, explained to me very recently and in a
private communication that the Savarez strings, although they are

176
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

This created definite difficulties for playing the in-

strument, whether directly on the tie-frets (more dif-
ficult with an unpleasant sound) or before them
(less difficult but still with an unpleasant sound).

Moreover, the tie-frets did not adhere well to the

fingerboard (FHT 22:177) and moved laterally dur-
Diameters of the strings expressed as a function of
the (weight or the diameter d of the) thinnest string, the zīr, ing the attempted performance of an improvised
as given (as proportions) by Kindī and Ṭaḥḥān. In the second melody.
possibility for Ṭaḥḥān (“Ṭaḥḥān II”), diameters of the strings
are nearer to those of Kindī I – apart from the bamm, clearly
thicker in Ṭaḥḥān II. The set of strings “Ṭaḥḥān II” is renamed
“Kindī II”.

Closest string-diameters to the strings proportions of

Kindī and Ṭaḥḥān with Savarez gut strings, taking the zīr
string to be 0.53 mm in diameter (equivalent to d). (For the
bamm string, values between brackets are the “ideal” – i.e.
computed according to the two theoretician’s explanations –
values, while the values outside the brackets are the effective
values of the closest – in diameter – Savarez gut strings.)

One of the difficulties which arose while fretting

the instrument was the exaggerated diameters of the
bamm and mathnā strings, mostly for the set
“Ṭaḥḥān II” (or “Kindī II”), for which the tying of
the frets was very difficult due to their thicknesses. FHT 22 Specific difficulties arise in the process of mount-
ing the ties on the neck of the ʿūd following the indications of
This is also illustrated in the photographs of FHT Kindī in the Risāla fī-l Luḥūn wa-n-Nagham. Here, the tie-frets
22:177 and FHT 25:180, in which the non-adher- do not adhere to the surface of the fingerboard because of
ence of the tie-frets to the surface of the finger- undue rigidity (due to the thickness) of the material of the
first two tie-frets. (Note the wooden layer beneath the nut
board398 is evident when using string thicknesses as piece which was added to raise the strings.)401
advocated by this early author (FHT 26:180), which
in turn creates problems.399 While the second set of strings (“Ṭaḥḥān I”) was
easier to install and allowed as well for an easier
Secondly, the thickness of the tie-frets compelled
tying of the nots of the tie-frets, the difficulties did
Makhlouf to insert a wooden piece beneath the nut
not disappear, with the same (but less) unpleasant
(FHT 22:177) 400 to raise it in such a way as for the
resulting sound and lateral displacements of the tie-
strings not to be in permanent contact with the frets.

398
This phenomenon is also due to the fact that the fingerboard is 400
See Video no. 1, 01:40 (mm:ss) to 02:00.
completely flat, with relatively sharp edges which (1) results in the 401
Note that today in Iran the thickest tie is 0.8 mm (diameter) for
non-adherence of dick ties as the ones shown in this figure and (2) the tār with relatively finer (slimmer) ties towards and after the
creates additional tension of the ties at the edges which can lead fifth. In Central Asia, the thicknesses can reach up to 1 mm. (Private
them, eventually, to sever. communication from Jean During.)
399
See Video no. 1, beginning 04:33 (mm:ss).

177
NEMO-Online Vol. 5 No. 9 – November 2020

FHT 25 Reconstructed Harmonic division of Kindī in the Risāla fī-l Luḥūn wa-n-Nagham, using tie-frets of the set “Kindī II”.

FHT 26 Set of gut strings “Kindī II” used in the reconstruction of the “fretting” of Kindī’s Risāla fī-l Luḥūn wa-n-Nagham with
diameters, from zīr (thinnest string) to bamm (thickest string): 0.53 mm, 0.71 mm, 0.94 mm and 1.27 mm.412

FHT 27 Set of strings “Ṭaḥḥan I” with similar proportions as given by Ṭaḥḥān (proportional weights/sections) with diameters,
from zīr to bamm: 0.53 mm, 0.61 mm, 0.71 mm et 0.81 mm.

412
Detailed information for the procurement of thicknesses of strings for Kindī’s and Ṭaḥḥān’s ʿūd(s) is available in [Beyhom and Makhlouf,
2009].

180
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

FHT 28 Impractical areas appear when including the octave equivalences for the scale of (al-) Fārābī as described by Maalouf.413
This figure is adapted and translated from [Beyhom, 2010c, v. 1, p. 175, 357]: virtual fingers reproduced in the figure are approx.
2 cm wide.

FHT 29 Computer re-created copy of the upper part of figure no. 3.5 in [Maalouf, 2002, p. 94], showing the proposed thicknesses
of Kindī’s tie-frets described in the Risāla fī-l-Luḥūn wa-n-Nagham.

413
[Maalouf, 2002, p. 126]. Many musicologists maintain that the Early Arabian ʿūd was “fretted” notwithstanding the complex divisions
described by Arabian theoreticians, and forgetting (or overlooking) the fact that some of the positions for the notes are alternative position-
ings, as here for the wusṭā(s). Moreover: “tying frets” on only half of the neck (as for the first six positions beginning from the right) is a
practical impossibility. Note that this description is espoused in [Abou Mrad, 2005, p. 773–774] with “full frets” on the fingerboard. In the
same reference, Abou Mrad cites [p. 784] Maalouf’s book and asserts that “frets were associated to the fingers of the left hand and placed
on the fingerboard of [the ʿūd] till the end of the Middle Ages” (“des frettes associées aux doigts de la main gauche sont disposées sur la touche
[du ʿūd] et ce, jusqu’à la fin du Moyen Âge”). Note also that Shireen Maalouf is a pianist, while Abou Mrad is a violinist, (both being Ph.D.
holders from Université du Saint-Esprit – Kaslik in Lebanon) which would explain their non-familiarity with the specificities of the fretting
of lute-type instruments.

181
NEMO-Online Vol. 5 No. 9 – November 2020

FHT 30 Tangent point of the string with the tie in the case of a lowered bridge, and in the case (as advocated by Maalouf) where
gut strands are superposed. (“Tie” in the figure = “tie-fret”; Each gut is considered as homogeneous and cylindrical, in accordance
with Maalouf’s indications: Richard Dumbrill – personal communication – reminds that it would not be possible to have exactly
superimposed guts in the manner in which they are described in this figure. The upper row of guts would force its way in between
the guts of the lower register. But moreover the guts would not be of circular section surface due to the fact that they would have
needed to be wet when affixed and would be of ovoid section surface.)

FHT 31 Tangent point of the string with the tie: as above but in the case of a heightened bridge. (“Tie” in the figure = “tie-fret”.)

182
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

FHT 32 Fārābī’s division as described by Maalouf (and advocated by Abou Mrad) with overlapping (or very close one to another)
“frets” in case tie-frets are winded around the neck of the ʿūd following Kindī’s indications. Performance is practically impossible
in this case.414

FHT 33 Fārābī’s division as described by Maalouf (and advocated by Abou Mrad) with overlapping (or very close one to another)
“frets” in case “realistic” tie-frets (thickness is taken as equal to 2 × 1𝑚𝑚) are mounted on the neck of the ʿūd. More impractical
areas appear while one zone of (nearly) impossible performance remains for the plain (27/32) and Persian (68/81) wusṭā(s).415

414
Translated and adapted from [Beyhom, 2010b, v. 1, p. 358].
415
Adapted and translated from [Beyhom, 2010b, v. 1, p. 358].

183
NEMO-Online Vol. 5 No. 9 – November 2020

APPENDIX C: THE RISĀLA FĪ-L-MŪSĪQĀ BY (AL-) through one epistle on astronomy and one other on as-
MUNAJJIM (856-912) trology, and would have written (at least) two works on
music, one of which – one about singing (ghināʾ) – is
The first extant theoretical (and historical) divisions lost.422
of the neck of the ʿūd are, as explained in the main text, The other epistle, the Risāla fī-l-Mūsīqā, is considered
by Kindī and Munajjim.416 While previous reviews of by some commentators as the key for the comprehen-
Arabian theories assert, with a little haste it seems, that sion of a voluminous compendium of anecdotes and
these divisions are Pythagorean and ditonic, 417 and songs of the 10th century,423 the Kitāb al-Aghānī 424 by
based on the tuning of the strings of the instrument in Abū-l-Faraj ʿAlī al-Aṣfahānī (or Iṣfahānī).425
successive fourths,418 things do not stand however in This epistle brought numerous analyses and inter-
such a simple fashion. pretations. 426 Munajjim claims in the introduction 427
The manuscripts of these authors are not explicit that he would explain the teaching of Isḥāq al-
about this information and, while Kindī proposes an al- Mawṣilī,428 but this task is not really fulfilled as not only
ternate harmonic division – which is far from Pythago- later (contemporary) commentators would not agree on
reanism –, Munajjim’s alleged “Pythagorean” division, the structure of the modes mentioned by him, but also
even if it were possible – if not probable – must still be because even the structure of his division cannot be
sustained.419 proven with the extant data.
The naming system (literal notation) is alphabetic,
*
* * and uses the same Syriac alphabet as with Kindī (abjad)
for the ten named notes, beginning from the unstopped
Yaḥyā ibn ʿAlī ibn Yaḥyā ibn abī Manṣūr al-Munaj- mathnā string (FHT 35:186) and ending on the zīr string
jim comes from a family of astrologists ,420 of poets and for the last one (produced by a shift of the hand posi-
of historians. He was close to al-Muwaffaq, the brother tion). The exact placement of the “last note” remains
of Caliph al-Muʿtamid (870-892),421 and he is known conjectural.429
416
Please note here that by “division” I mean a theoretical mesh of Era] of ignorance” – see footnotes no. 45:119 and 226:142) till the
the neck which could have been materialized by the strings, on one 10th century.
side, and by – perpendicular to the strings – drawn lines, or by 424
See a short description in [Sawa, 2001]. Most other writings of
threads tied on the fingerboard of the instrument, on the other side. Sawa relate to this period of Arabian music theory and practice and
The controversial – and very rare – descriptions of physical “liga- could be relevant for the reader seeking additional comments, for
tures” (or “tie-frets”) are examined in Part II of this dossier. example [Sawa, 1981 ; 1985 ; 1989 ; 2002].
417
This is the Pythagorean ascending-descending division shown in 425
The title of Yūsuf Shawqī’s 1976 edition, Risālat ibn al-Munajjim
FHT 13 in [Beyhom, 2016, p. 186]. fī-l-Mūsīqā wa Kashf Rumūz Kitāb al-Aghānī [The epistle of ibn al-Mu-
418
(Reminder:) The ʿūd at that time had 4 strings, named consecu- najjim on music and the unveiling of the symbols of Kitāb al-Aghānī], is
tively (from top to bottom – for a ʿūd played by a right-handed per- for example explicit about this matter. Abū-l-Faraj al-Aṣfahānī (or
former and seen from the front side – but “lowest” to “highest” Iṣfahānī, 897–967) a.k.a. Abulfaraj, “was an historian of Arab-
acoustically) bamm, mathlath, mathnā and zīr ; the ḥād (an addi- Quraysh origin who is noted for collecting and preserving ancient
tional string situated lowest – and acoustically “highest”) is cited in Arabic lyrics and poems in his major work, the Kitāb al-Aghānī. [He]
Urmawī’s a-sh-Sharafiyya, while several earlier authors (including was born in Isfahan, but spent his youth and had his early studies
Kindī – who names it the “lower zīr” in the Risāla fī Khubr Ṣināʿat a- in Baghdad. He was a direct descendant of the last of the Umayyad
t-Taʾlīf) mention this 5th string although they specify that its use(full- caliphs, Marwan II, and was thus connected with the Umayyad rul-
ness) was merely theoretical. ers in al-Andalus, and seems to have kept up a correspondence with
419
The following section contains a few, simple algebraic formulae them and to have sent them some of his works. He became famous
for Munajjim’s division of the fingerboard of the ʿūd. An accessible for his knowledge of early Arabian antiquities” – [Wikipedia Con-
review of algebra is available in [Pratt, 2007]. tributors, 2017c] (see also [Neubauer, 2001c]).
420
Besides [Beyhom, 2010b], Owen Wright’s articles [1966 ;
426
For a review of these interpretations, see [Sawa, 1989, p. 74–
2001i] can be consulted for additional information about Munajjim 78].
and his epistle. “Munajjim” (root: n[a]jm – “planet”, “celestial 427
[Munajjim (al-), 1976, p. 189].
body”) in Arabic means “astrologist”. 428
See footnote no. 101:101.
421
[Farmer, 1929, p. 167]. 429
There are contradictory statements in the epistle about the “10th
422
[Farmer, 1929, p. 168] and Erlanger in [Fārābī (al-), 1930, v. 1, [last] note” – see [Beyhom, 2010b] and [Wright, 1966] for more
p. xxii]. details.
423
Including anecdotes, stories and poetry from the time of the
Jāhiliyya (the period before Islam – the religion – or “the time [or

184
NEMO-Online Vol. 5 No. 9 – November 2020

FHT 35 Stylized fingerboard of a ʿūd showing the sequential assignment of the 10 notes by Munajjim in his Risāla fī-l-Mūsīqā.
The division and the position of the 10th note are still undetermined.

FHT 36 Stylized fingerboard of a ʿūd with unspecified intervals corresponding to algebraic formulae deduced from the epistle of
Munajjim fī-l-Mūsīqā. The double sided arrows show (sequentially numbered) equivalences between octaves (bold) or unisons (bold
italics).

186
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

FHT 37 Calculated boundaries (in cents) of the intervals between the vertical markers for the general case, in the division of
Munajjim. D is the interval between the nut and the “10th note”.

FHT 38 Stylized fingerboard of a ʿūd with, assuming a tuning of the strings in fourths, with Munajjim’s resulting Pythagorean
“ascending then descending one tone” division. (This corresponds to two ascending whole-tones then a leimma from the nut, to
which we add a whole-tone descending from the Khinṣir which completes the division.)

187
NEMO-Online Vol. 5 No. 9 – November 2020

APPENDIX D: ORIGINAL TEXTS ،‫ و ضطرت إلى أن يستعنل عليها ألصابع‬،‫في تىىير لنغم إلى لىساتين‬
‫ ففرض على لكف‬،‫معا‬ ً ‫وعسر في اتى ء ألمر أن يدرك لكف و ألصابع‬
Due to the necessity of including multiple quotes in ً
‫ساكنا‬ ‫ وكان لىى لذي يلزمه لكف‬،‫لسك ن وعلى ألصابع حركة‬
the dossier, and to the importance of their translation ُ
for a better understanding of sometimes subtle (but ef- ‫ فشى على‬،‫وتتصرف علقه ألصابع متدركة مو ط ل آللة ملعتىلة ه بعه‬
fective) differences between interpretations, the original ‫ وبىي‬،‫ وشغلت إلبهام االضبط‬،‫لربع أول لىساتين منس با إلى نصر‬
texts are included in this appendix, except for small .“‫للتصرف فقنا اين حىي ذلك لربع أصابع أ بعة‬
quotes which are kept for quick reference in the main [Manik, 1969, p. 12]:
text.
“In bezug auf die Lautenbünde, die die mittelalterlichen
Musiktheoretiker zur Darstellung ihrer Tonsysteme ausführlich
* beschrieben haben, vertritt nun Berner die Meinung, daβ diese
* *
Bünde niemals bestanden haben, weil es sich hier, wie er
[Kiesewetter, 1858, p. 32]: wörtlich sagt, nur um eine “bloβe Fiktion” handele. Dabei
beruft sich Berner auf Geiringer, der, nachdem er festgestellt
“Ueberhaupt kann ich mich schon lange des Gedankens nicht
hatte, daβ eine Laute mit Bünden in dem ikonographischen
erwehren, dass die ausübende Musik verschiedener älterer und
Befund der Zeit nirgends anzutreffen war, zu dem Schluβ
neuerer asiatischer Völker ein ganz anderes Ding gewesen sein
gelangt, daβ Bünde nur für Messungs– und
oder noch sein müsse, als jene metaphysische oder
Untersuchungszwecke verwendet wurden, so daβ sie für die
mathematische Musik ihrer Philosophen, deren Theorien, ein
Musikpraxis keinerlei Bedeutung haben konnten. Zu ähnlicher
Werk bloser Spekulazion, sich von der Praxis immer entfernt
Folgerung war auch Curt Sachs schon früher gekommen”.
gehalten haben mussten. Ich meine, […] dass man demzufolge
nicht sagen sollte: die Musik der Chinesen, der lndier, der [Ṭaḥḥān (ibn a-ṭ-~ al-Mūsīqī), 1990, p. 175-176 (89–90)]:
Araber, der Perser u. s. w., sondern: die musikalischen Systeme
‫”بعض لناس يظو أن لنغم لتي في لع د مختلفة لعىد ختالفهم في شى‬
(oder Mysterien) der chinesischen, der indischen, arabischen,
persischen Philosophen, des Meisters Chrysanthos, u. s. w. – ‫ وتجري‬.‫ وهي قان ن لغناء ملتبع‬.‫لىساتين وندو نذكر مو ذلك ما يتفق‬
Vielleicht dass es in der Musik der alten Griechen eben auch ‫مجرى شى نف س َحىي وه مو أجل هذ لشأن و لىساتين حىود لنغم‬
nicht anders gewesen”.
‫و لسنه الوتا ومنها مخا ج لنغم مو لع د وم ضع حروف مو حلق‬
[Jargy and Chottin, 2001, p. 527]: ً
‫صافقا وكذلك لنغنة‬ ‫فإذ خرج حرف مو حلق مو م ضعه حىقىي خرج‬
“1) Période bédouine, depuis la djâhilîya jusqu’aux premiers
‫إذ خرجت على دستان صحقح خرجت صافقة وجنقع لىساتين لتي‬
temps de l’Islam (mort d’Ali, 661) ; 2) Période d’assimilation, de ُ ُ
la dynastie omeyyade au premier cycle Abbaside (vers 830) ; ‫تستخرج فيها لنغم لطبقعقة لالنسان وتستعنل في جنقع ال حان ستة‬
3) Période d’épanouissement et de dispersion, avec le second cycle ‫دساتين أولها دستان ملجنب ودستان لسبااة ودستان وسط لفرس‬
Abbaside et l’établissement des Omeyyades en Espagne ;
‫ودستان وسط لعرب ودستان لبنصر ودستان نصر وبين دستاني‬
4) Période de repli, de la prise de Grenade (1492) à la fin du
XVIIIe siècle ; 5) Renaissance : la Nahda, du XIXe siècle, à partir de ‫وسط لعرب ودستان لبنصر دستان آخر يسم دستان لزل وأكثر لناس‬
l’expédition de Bonaparte en Égypte, jusqu’au congrès du Caire ً ‫ُيهنله ودستان آخر يىع اين دستان لبنصر ودستان نصر ُيهنل‬
‫أيضا‬
(1932)”.
‫وهذه لىساتين ا جة عو لعىد ألول فهي منا يستعنله لفرس في‬
[Chabrier, 1982]: ‫طر ئىهم وأنا أستعنل ذلك وأطرق م ضعه ملعرفتي اه بغير دساتين وذلك‬
“Avec les Califes Abbasides de l’Iraq, [le ʿūd] va devenir le luth
‫ وشى لىساتين يدتاج إلى علم‬.‫يصعب على ملتعلنين فتركه لهم أولي وأحق‬
concepteur des genres et modes des musiques méso-islamiques ً ‫ مدتاج لذي يريى شىها على لع د أن يأخذ اركا ً فقفتده ف‬.‫بها‬
et créateur des mélodies, rôle qu’il conservera jusqu’à nos jours ‫تدا اىى‬
dans les musiques arabes savantes et populaires”. ً
‫صحقدا اقنها و ملطب ع ملرتاض لعا ف‬ ً ‫ما يريى شىها ويىيس اه‬
‫ققاسا‬
[Kindī (al-), 1965, p. 19]: ‫ملرتاض يعرف أقى ها وم قعها اال اركا ال اا حس ُومىاالة بعض لنغم‬
‫أيضا نغنة خا جة مو جنقع لىساتين يسن نها‬ ً ‫”وقى يستعنل ملغن ن‬ ‫ابعض وبالعادة و لى بة ثم يشىها فإذ كنلت على ما ذكرناه صحت لنغم‬
ً
‫ وخلف‬،‫" ملحص ة" وهي خا ج مو دستان نصر ينىون إليها نصر‬ ‫وصفت وهذه حنلة كافقة وال يدتاج في عرض لىساتين أكثر مو أ بع‬
ً ‫هذه‬
‫ غير أنهم ينىل ن‬،‫أيضا – انثل مسافة دستان نصر – نغنة أخرى‬ ‫طاقات مو الوتا لبقض ملصا يو ويجب أن يك ن على تى يج في أن يك ن‬
ً
.“‫لسبااة إلى دستان ل سط أو لبنصر‬ ‫ألول غلقظا و لثاني دون غلظه و لثالث دونه كذ إلى أخرها على هذ ملثال‬
ُ ‫وإن لم يعتبر االعين‬
.“‫فلقعتبر اال ن فإنه أصح‬
[Sīnā (Ibn) or Avicenna (980?-1037), 1956, p. 47–48]:
‫] فلنا حاول يى عه للحنقات ]…[ ودع ثالثة أبعاد للسبب لذي‬...[” [Fārābī (al-) and ‫ الفارابي‬,1967, p. 655]:
‫ أن حاجة مست‬: ‫ وقى أعان هذ لسبب سبب مو جهة آللة وه‬.‫ذكرناه‬

188
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

Les points fournissant certaines de ces notes coïncident avec Laute mit Bünden auch Zwischentöne darzustellen, und daß
des ligatures sur lesquelles on les produit. D’autres ne man aus diesem Grund die Bünde im Laufe der Zeit abgeschafft
coïncident pas avec une ligature et ne peuvent être produites, habe, ist in dieser Ausschließlichkeit nicht zutreffend. Auch
à moins qu’on ait la chance de placer le doigt au point juste”. sollte die zählebige Vorstellung, Bünde seien in der arabisch-
(Bold type mine.) islamischen Musikgeschichte lediglich zu theoretischem
Gebrauch, nicht aber in der Praxis verwendet worden,
[Neubauer, 1993, p. 330]:
nunmehr der Vergangenheit angehören”.
“Nach den von ihm tradierten Anweisungen verhalten sich die
Saitenstärken von der zir- zur bamm-Saite wie 1:2:3:4. Analog [Ṭaḥḥān (ibn a-ṭ-~ al-Mūsīqī), 1990, p. 175]:
müßte der Umfang der Bünde vom Zeigefinger- bis zum ‫”وبين دستاني وسط لعرب ودستان لبنصر دستان آخر يسم دستان‬
Kleinfinger- bund im Verhältnis 4:3:2:1 abnehmen. Beides ist
‫لزل وأكثر لناس ُيهنله ودستان آخر يىع اين دستان لبنصر ودستان‬
unrealistisch”. (Bold type mine.) ً ‫نصر ُيهنل‬
‫أيضا وهذه لىساتين ا جة عو لعىد ألول فهي منا‬
[Neubauer, 1993, p. 331–332]:
‫يستعنله لفرس في طر ئىهم وأنا أستعنل ذلك وأطرق م ضعه ملعرفتي اه‬
“Im Zusammenhang mit dem Stimmen der Saiten gibt al-Kindī
eine Anweisung für die richtige Haltung der Finger auf den, wie
‫بغير دساتين وذلك يصعب على ملتعلنين فتركه لهم أولي وأحق‬
al-Hindī (6./12. Jh.) sagen wird, ‘Bünden für die [Fārābī (al-), 1930, v. 1, p. 2]:
Fingerkuppen’: ‘Der kleine Finger wird auf die bamm-Saite
“Pour être un parfait théoricien, quelle que soit la science dont
gelegt und mit festem Griff gegen den *Kleinfinger*-Bund
il s’agit, il faut trois conditions : En bien connaître tous les
gedrückt, ohne sich von der Stelle zu bewegen, für die er
principes. Avoir la faculté de déduire les conséquences
vorgesehen ist, und indem er nach einer Seite die Saite frei
nécessaires de ces principes dans les êtres (les données) qui
[schwingen] läßt, denn [sonst] ergibt sich zwangsläufig eine
appartiennent à cette science. Savoir répondre aux théories
Trübung der Töne. Der kleine Finger soll jeweils am Anfang der
erronées, et analyser le vrai du faux et redresser les erreurs”.
Bünde, direkt hinter ihnen, liegen, während die übrigen [Finger
sich] in der Luft zwischen dem Kleinfinger- und dem
Ringfinger-Bund [befinden]. [Diesen Punkt] überschreite [d]er
[kleine Finger] nicht und bleibe auch nicht hinter ihm zurück,
dem wenn er ihn [nur] ein wenig überschreitet [und auf den
Bund gerät], entsteht im Ton eine Taubheit, und wenn er hinter
ihm zurückbleibt und zwischen die beiden Bünde zu liegen
kommt, entsteht ein *Zirpen*. Dies ist ein allgemein gültiges
Gesetz, das für alle Finger gilt bei ihrer Bewegung über die
Saiten hin und bei allen Bünden für denjenigen, der der Sache
auf den Grund geht.’ Dies ‘allgemein gültige Gesetz’ und die
Beschreibung der korrekten Position der Finger der linken
Hand gilt bis heute und stellt der präzisen Beobachtung und
Formulierung al-Kindī’s bzw. seiner Quelle ein hervorragendes
Zeugnis aus. Auch der letzte Zweifel am praktischen Gebrauch
der Bünde dürfte hiermit ausgeräumt sein”.

[Neubauer, 1993, p. 331]:

“… ‘Es gibt noch einen Bund, der zwischen dem ‘Ringfinger’-
und dem ‘Kleinfiger’-Bund liegt, [aber] der wird
[normalerweise] auch nicht gebraucht. Dieses [?] sind Bünde,
die aus der ursprünglichen Zahl [sechs] herausfallen. Sie
werden von den Persern in ihren Modi verwendet. Ich benutze
sie auch und treffe ihre [richtigen] Stellen [auf dem Griffbrett],
da ich sie kenne, auch ohne [zusätzliche] Bünde. Den Schülern
fällt das aber schwer. Sie fortzulassen ist [daher] besser und
richtiger.’ Hieraus folgt, daß ein ägyptischer Hofmusiker des
5./11. Jahrhunderts auch persische Musik zu Gehör brachte FHT 39 First page of the Risāla fī-l-Luḥūn wa-n-Nagham
und daß er sie auf seiner einheimischen Laute mit oder ohne (Mukhtaṣar al-Mūsīqā fī Taʾlīf a-n-Nagham wa Ṣinʿat al-ʿūd) by
zusätzliche Bünde spielte. Daß er die zusätzlichen Bünde und Yaʿqūb ibn Isḥāq al-Kindī, taken from [Kindī (al-), 1965]. 436
damit das persische Repertoire für Anfänger des Lautenspiels
lieber vermied, ist verständlich. Das heute zu hörende
Argument jedoch, daß es generell nicht möglich war, auf einer
436
The image was edited and cropped for clarity.

191
NEMO-Online Vol. 5 No. 9 – November 2020

References ICONEA Publications |University of London – Senate House,

2011-12-3| [url: http://foredofico.org/CERMAA/wp-content/
Please note that “Grove Music Online” stands for “The New Grove uploads/2017/01/Amine-Beyhom-Two-common-errors-about-
Dictionary of Music and Musicians”, 2nd Edition |2001|, online the-‛ŪD-from-iconea2011web.pdf] p. 81–110.
version |2001-2020|, originally McMillan then |2004+| Oxford 15. BEYHOM, Amine : Théories et pratiques de l’échelle dans le chant by-
University Press. [https://www.oxfordmusiconline.com/grove zantin arabe : Une approche comparative et analytique proposant une
solution inédite pour le système théorique de Chrysanthos le Madyte,
music/]
par l’auteur |Broummana (Liban), 2015-9|.
1. ABBOTT, D. and E. SEGERMAN : “Gut strings”, Early Music 4 4 |1976| 16. BEYHOM, Amine : “Dossier: Hellenism as an Analytical tool for Oc-
p. 430–437. cicentrism (in musicology) (V2)”, Near Eastern Musicology Online 3
2. ABOU MRAD, Nidaa : “Échelles mélodiques et identité culturelle en 5 |2016-11| [url: http://nemo-online.org/wp-content/uploads/
Orient arabe”, Musiques et Cultures Vol. III/V, ed. Jean-Jacques Nattiez, 2017/02/(V2-INTERNET)%205-04.%20Article%20NEMO%20
Musiques : une encyclopédie pour le XXIe siècle, Actes Sud / Cité de la mu- n%C2%B05%20Amine%20Beyhom%20170219S.pdf] p. 53–275.
sique |Arles (Bouches-du-Rhône) ; [Paris], 2005| p. 756–795. 17. BEYHOM, Amine : “A Hypothesis for the Elaboration of Heptatonic
3. ADAMSON, Peter : “Al-Kindi”, Stanford Encyclopedia of Philosophy Scales”, Near Eastern Musicology Online 4 6 |2017-11|
|2011| [url: http://plato.stanford.edu/entries/al-kindi/]. [url: http://nemo-online.org/wp-content/uploads/2017/05/IN-
4. Anon. “Age_of_Caliphs.png (Image PNG, 684 × 347 pixels)” TERNET-6-01.-Article-NEMO-no.-4-6-1-Amine-Beyhom-Hypo-
[url: https://upload.wikimedia.org/wikipedia/commons/2/20/ thesis-170501.pdf] p. 5–89.
Age_of_Caliphs.png]. 18. BEYHOM, Amine : “MAT for the VIAMAP. Maqām Analysis Tools
5. Anon. “Mike Ouds - My Ouds Page” [url: http://www.mikeouds. for the Video-Animated Music Analysis Project”, Near Eastern Mu-
com/oudpics.php]. sicology Online 4 7 |2018-11| [url: http://nemo-online.org/]
6. ANONYME : A-sh-Shajara Dhāt al-Akmām al-Ḥāwiya li-Uṣūl al- p. 145–258.
Anghām [The tree with calyxes which contains the origins of the melo- 19. BEYHOM, Amine and Hamdi MAKHLOUF : “Frettage du ʿūd (luth
dies], editors Ghaṭṭās ꜤAbd-al-Malik Khashaba (‫)غطاس عبد الملك خشبة‬ arabe) dans la théorie musicale arabe et influence sur la pratique
and Isis Fatḥ-al-Lāh (‫ أازاس فتح هللا‬.‫)د‬, Al-Hayʾa al-Miṣriyya al-ꜤĀmma [The fretting of the ʿūd in Arabian music theory and its interaction
li-l-Kitāb (‫| )الهيئة المصراة العامة للكتاب‬Le Caire, 1983-xive siècle|. with practice]”, 5ème Congrès de Musicologie Interdisciplinaire
7. BEG, M. A. J. : “Ṭaḥḥān (a.): Encyclopaedia of Islam, Second Edi- (CIM09), La musique et ses instruments, Fifth Conference on Interdis-
tion : Brill Online”, [Print Version: Volume X, page 102, column 1] ciplinary Musicology |Paris, 2009-10-26| [url: http://cim09.
|2000| [url: http://www.brillonline.nl/]. lam.jussieu.fr/CIM09-en/Proceedings.html, http://cim09.lam.
8. BERNER, Alfred : Studien zur arabischen Musik auf Grund der gegen- jussieu.fr/CIM09-en/Proceedings_files/Beyhom-Makhlouf.pdf].
wärtigen Theorie und Praxis in Ägypten. - Leipzig: Kistner & Siegel 20. BOETHIUS : Traité de la musique, editor Christian Meyer, Brepols
1937. VI, 124 S. 4°(8°) Zugleich Berlin, phil. Diss. (Schriftenreihe d. |Turnhout, 2004|.
Staatl. Inst. f. Deutsche Musikforschung. 2.), Fr. Kistner & C. F.W. Sie- 21. BONTA, Stephen : “The Making of Gut Strings in 18th-Century Pa-
gel |Leipzig, 1937|. ris”, The Galpin Society Journal 52 |1999-4-1| [doi: 10.2307/
9. BEYHOM, Amine : Systématique modale – En 3 volumes, PhD Thesis, 842553. url: http://www.jstor.org/stable/842553] p. 376–386.
Université Paris Sorbonne |2003-9-17| [url: http://www.sudoc.fr/ 22. BOUTERSE, Curtis : “Reconstructing the Medieval Arabic Lute: A
075767538, http://foredofico.org/CERMAA/publications/ Reconsideration of Farmer’s ‘Structure of the Arabic and Persian
publications-on-the-site/publications-amine-beyhom]. Lute’”, The Galpin Society Journal 32 |1979-5-1| [doi: 10.2307/
10. BEYHOM, Amine : “Systématique modale : génération et classe- 841532. url: http://www.jstor.org/stable/841532] p. 2–9.
ment d’échelles modales”, Musurgia XI 4 |2004-12| 23. BOWER, Calvin M. : Bower 1989 Boethius’ Fundamentals Of Music,
[url: http://foredofico.org/CERMAA/wp-content/uploads/2017/ editor Claude V. Palisca, Music Theory Translation Series, Yale Uni-
11/Beyhom-2004-Systématique-modale-intro-appendices-Mu- versity Press |New Haven & London, 1989| [url: http://ar-
surgia-XI_4-2004_OCR.pdf] p. 55–68. chive.org/details/bower-1989-fundamentals-of-music].
11. BEYHOM, Amine : “Approche systématique de la musique arabe : 24. BUTLER, Alfred Joshua : The Arab conquest of Egypt and the last thirty
genres et degrés système”, De la théorie à l’Art de l’improvisation : years of the Roman dominion, Oxford, Clarendon Press |1902|
analyse de performances et modélisation musicale, ed. Mondher Ayari, [url: http://archive.org/details/arabconquestegy00butlgoog].
Delatour |Paris, 2005-12| [url: http://foredofico.org/CERMAA/ 25. CAREY, Travis : Fret Tying with Travis Carey |2017-4-19|
wp-content/uploads/2017/11/Beyhom_2005_Genres_et_ [url: https://www.youtube.com/watch?v=6T5MgtQSofI].
echelle_systeme_Delatour.pdf] p. 65–114. 26. CHABRIER, Jean-Claude Ch. : “ʿūd”, editors Antoine Goléa and Marc
12. BEYHOM, Amine : “A new hypothesis on the elaboration of hepta- Vignal, Larousse de la musique (2) : Kreuzspiel - Zyklus, Librairie La-
tonic scales and their origins”, ICONEA 2008, Lulu on behalf of rousse |Paris, 1982| p. 1853.
ICONEA PUBLICATIONS, LONDON-UK |London, 2010a| 27. CHABRIER, Jean-Claude Ch., A. DIETRICH, Clifford Edmund BOS-
[url: http://foredofico.org/CERMAA/wp-content/uploads/2017/ WORTH and Henry George FARMER : “ʿŪd: Encyclopaedia of Islam,
11/ICONEA-2008-2010-Hypothesis-Beyhom-complete.pdf] Second Edition : Brill Online”, editors Peari Bearman, Th. Bian-
p. 151–209. quis, C. E. Bosworth, et al., [Print Version: Volume X, page 767,
13. BEYHOM, Amine : Théories de l’échelle et pratiques mélodiques chez les column 2] |2000| [url: http://www.brillonline.nl/].
Arabes – Volume 1 : L’échelle générale et les genres – Tome 1 : Théories 28. CHAILLEY, Jacques : Éléments de philologie musicale [Recherche des
gréco-arabes de Kindī (IXe siècle) à Ṭūsī (XIIIe siècle) 1/4 (vols.), Li- principes - Intervalles et Échelles], Alphonse Leduc |Paris, 1985|.
brairie orientaliste Paul Geuthner |Paris, 2010b-11|. 29. CHARRY, Eric : “Plucked Lutes in West Africa: An Historical Over-
14. BEYHOM, Amine : “Two persistent misapprehensions about the view”, The Galpin Society Journal 49 |1996-3-1| [doi: 10.2307/
ʿūd”, The Oud from its Sumerian Origins to Modern Times ICONEA 842390. url: http://www.jstor.org/stable/842390] p. 3–37.
2011 – 1-3 December 2011, ed. Richard Dumbrill and Irving Finkel,

192
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

30. COLEBROOKE, Henry Thomas : Algebra, with Arithmetic and Mensu- 44. FĀRĀBĪ (AL-), Abū-n-Naṣr Muḥammad ibn Muḥammad ibn
ration, from the Sanscrit of Brahmegupta and Bhascara, John Murray Tarkhān and al-Ḥusayn ibn ʿAbd-al-Lāh SĪNĀ (IBN) OU AVICENNE
|London, 1817|. (980?-1037) : La musique arabe (2) – Al-Fārābī [Grand traité de la
31. COWL, Carl and Sheila M CRAIK : Henry George Farmer : a bibliography, musique] Livre III du Kitāb al-Musīqī al-Kabīr ; Sīnā (ibn) ou Avi-
Glasgow University Library Studies, GLUS |Scotland, 1999|. cenne, Kitāb a-sh-Shifāʾ – Mathématiques, editor Christian Poché,
32. DALL’ORTO, Giovanni : “File:0372 - Pavia - S. Pietro - Cripta - translator Rodolphe (d’) Erlanger, ed. 2e en fac-similé, Les Geuthner
Tomba Boezio - Foto Giovanni Dall’Orto, Oct 17 2009.jpg”, (Paris), ISSN 1294-6656 2/6 (vols.), Librairie Orientaliste Paul
Wikipedia |Pavia - Italy, 2009-10-17| [url: https://en.wikipedia. Geuthner |Paris, 2001|.
org/wiki/File:0372_-_Pavia_-_S._Pietro_-_Cripta_-_Tomba_Boezio_- 45. FARMER, Henry George : A History of Arabian Music to the XIIIth Cen-
_Foto_Giovanni_Dall%27Orto,_Oct_17_2009.jpg]. tury, Luzac |Bristol, Great-Britain, 1929|.
33. DIEBUCHE : “Map of expansion of Caliphate” |2010-7-3| [Adapted 46. FARMER, Henry George : Historical facts for the Arabian musical in-
from http://guides.library.iit.edu/content.php?pid=27903 fluence, Ayer Publishing |1930|.
&sid=322018 (via Image:Age_of_Caliphs.png), traced on high re- 47. FARMER, Henry George : “Was the Arabian and Persian Lute
solution version of Image:BlankMap-World6.svg. Information Fretted?”, Journal of the Royal Asiatic Society of Great Britain and
from The Times Concise Atlas of World History ed. by Geoffrey Bar- Ireland 3 |1937-7-1| [url: http://www.jstor.org/stable/25201549]
raclough published by Times Books Ltd. Isbn 0-7230-0274-6 p. p. 453–460.
40-41. url: https://commons.wikimedia.org/wiki/File:Map_of_ 48. FARMER, Henry George : Studies in oriental musical instruments. With
expansion_of_Caliphate.svg]. four plates and other illustrations. Second series |1939a|.
34. DIETERICI, Friedrich Heinrich, ‫ اخوان الصفاء‬and IHW ̲ ĀN AL-ṢAFĀʼ : Die 49. FARMER, Henry George : “The Structure of the Arabian and Persian
Propaedeutik der Araber im zehnten Jahrhundert ; die Logik und Psy- Lute in the Middle Ages”, Journal of the Royal Asiatic Society of
chologie der Araber im zehnten Jahrhundert n. Chr : being a German Great Britain and Ireland 1 |1939b-1-1| [url: http://www.jstor.
translation of the Rasāʼil Ikhwān aṣ-Ṣafāʼ. Ch. 1-13, ‫منشورات معهد تاراخ‬ org/stable/25201835] p. 41–51.
22 ; ‫ فلسفة االسالمية ؛‬.‫ العلوم العربية واالسالمية‬Manšūrāt Maʻhad tārīh̲ al-ʻulūm 50. FARMER, Henry George : The sources of Arabian music: an annotated
al-ʻarabiyyaẗ wa-al-islāmiyyaẗ. Falsafaẗ al-islāmiyyaẗ ; 22, Mittler u. bibliography of Arabic manuscripts which deal with the theory, practice,
Sohn |Berlin, 1865|. and history of Arabian music from the eighth to the seventeenth century,
35. DURING, Jean : “Introduction au Volume 71 1/2 de la Revue de E. J. Brill |1965|.
Musicologie(1985)”, Revue de musicologie 71 1/2 |1985| 51. FARMER, Henry George : “Music”, A history of Muslim philosophy:
[doi: 10.2307/928593. url: https://www.jstor.org/stable/928593? With short accounts of other disciplines and the modern renaissance in
origin=crossref] p. 7–10. Muslim lands, ed. Mian Mohammad Sharif, Harrassowitz |Alle-
36. DURING, Jean : Quelque chose se passe : le sens de la tradition dans magne, 1966a| [url: http://www.muslimphilosophy.
l’Orient musical, Verdier |Lagrasse, 1994|. com/hmp/] p. 1124–1177.
37. EHWANY (EL-), Ahmed Fouad : “Al-Kindi”, A history of Muslim phi- 52. FARMER, Henry George : Geschichte der arabisch-islamischen Musik,
losophy: With short accounts of other disciplines and the modern re- VEB deutscher Verl. für Musik |1966b|.
naissance in Muslim lands, ed. Mian Mohammad Sharif, Harrasso- 53. FARMER, Henry George : “G̲H̲inā”, Encyclopaedia of Islam, [Print
witz |Allemagne, 1966| [url: http://www.muslim Version: Volume II, page 1072, column 2] |2011| [url: http://
philosophy.com/hmp/] p. 421–434. www.brillonline.nl/] p. 1072.
38. ERLANGER, Rodolphe (d’) : La musique arabe, 6 tomes – I. [Al-Fārābī] 54. FARMER, Henry George, Werner BACHMANN, Heinrich BESSELER
(1930) ; II. [Al-Fārābī et Ibn Sīnā] (1935) ; III. [Commentaires sur Le and Max SCHNEIDER : Musikgeschichte in Bildern: Islam, [Band III:
Livre des Cycles de Safiyy-a-d-Dīn al-Urmawī] (1938) ; IV. [Anon. – Musik des Mittelalters und der Renaissance / Lieferung 2], Veb Deut-
a-sh-Shirwānī – et al-Lādhiqī] (1939) ; V. [Échelles et modes] (1949) ; scher Verlag Fur Musik |Leipzig, 1966|.
VI. [Rythmes et formes] (1959) 6 (vols.), Librairie orientaliste Paul 55. GIBB, H. A. R : The Arab Conquest in Central Asia, Reprint from the
Geuthner |Paris, 1930-1959|. (1923) edition, AMS Edition |New York -EU, 1970|.
39. ESPINOZA, Santiago : Vihuela Frets/Trastes Part 1 |2015-3-25| 56. GILMAN, Benjamin Ives : “On Some Psychological Aspects of the
[url: https://www.youtube.com/watch?v=NYaXlKRfC2E]. Chinese Musical System”, The Philosophical Review 1 1 |1892-1-1|
40. FĀRĀBĪ (AL-), Abū-n-Naṣr Muḥammad ibn Muḥammad ibn [doi: 10.2307/2175529. url: http://www.jstor.org/stable/21755
Tarkhān : La musique arabe (1) – Al-Fārābī, Grand traité de la mu- 29] p. 54–78.
sique [Kitāb al-Mūsīqī al-Kabīr] (Livres I et II), editor Rodolphe d’Er- 57. GRAME, Theodore : “The Symbolism of the ’Ūd”, Asian Music, Asian
langer, translator Rodolphe (d’) Erlanger 1/6 (vols.), Librairie Music 3 1 |1972-1-1| [doi: 10.2307/834102. url: http://www.
Orientaliste Paul Geuthner |Paris, France, 1930|. jstor.org/stable/834102] p. 25–34.
41. FĀRĀBĪ (AL-), Abū-n-Naṣr Muḥammad ibn Muḥammad ibn 58. GRUNEBAUM, Gustave E. von : “The Concept of Plagiarism in Arabic
Tarkhān : ‫كتاب الموسيقى الكبير‬, editors Maḥmūd Aḥmad al-Ḥifnī and Theory”, Journal of Near Eastern Studies 3 4 |1944-10| p. 234–253.
‫ تراثنا‬,‫ محمود أحمد الحفني‬Turāthuna, 1967 ,‫|دار الكاتب العربي للطباعة والنشر |القاهرة‬. 59. GUETTAT, Mahmoud : Musiques du monde arabo-musulman : Guide bi-
42. FĀRĀBĪ (AL-), Abū-n-Naṣr Muḥammad ibn Muḥammad ibn bliographique et discographique : Approche analytique et critique, Chants
Tarkhān : ‫ كتاب الموسيقى الكبير‬Manuscrit/, ‫معهد تاراخ العلوم العربية واإلسالمية في‬ et poésies du Maghreb, Dar al-Uns [El Ouns] |Paris, France, 2004|.
1998 ,‫ جمهوراة ألمانيا االتحاداة‬،[‫|إطار جامعة فرانكفورت |]فرانكفورت‬. 60. HASSAAN, G. A. : “Banu Musa, the founders of automatic control in
43. FĀRĀBĪ (AL-), Abū-n-Naṣr Muḥammad ibn Muḥammad ibn the 9th century”, Proceedings of the MDP-8, Cairo University Confe-
Tarkhān and al-Ḥusayn ibn ʿAbd-al-Lāh SĪNĀ (IBN) OU AVICENNE rence on Mechanical Design and Production, Cairo, Egypt |2004|
(980?-1037) : La musique arabe (2) – Al-Fārābī [Grand traité de la [url: http://scholar.cu.edu.eg/sites/default/files/galal/files/banu-
musique] Livre III du Kitāb al-Musīqī al-Kabīr ; Sīnā (ibn) ou Avi- musa_control.pdf] p. 1–6.
cenne, Kitāb a-sh-Shifāʾ – Mathématiques, translator Rodolphe (d’) 61. HASSAN, Scheherazade Qassim : “The Long Necked Lute in Iraq”,
Erlanger, 1st ed. 2/6 (vols.), Librairie Orientaliste Paul Geuthner Asian Music 13 2 |1982-1-1| [doi: 10.2307/833875. url: http://
|Paris, 1935|. www.jstor.org/stable/833875] p. 1–18.

193
NEMO-Online Vol. 5 No. 9 – November 2020

62. HELLWIG, Friedemann : “An Example of Lute Restoration”, The music], ed. Zakariyyā Yūsuf, Vol. 2, MaṭbaꜤat Shafīq |Baghdād,
Galpin Society Journal 23 |1970| [doi: 10.2307/842064. 1962b|.
url: http://www.jstor.org/stable/842064] p. 64–68. 77. KINDĪ (AL-), Yaʿqūb ibn Isḥāq Abū Yūsuf (0801?-0867?) : Risāla fī-
63. HELLWIG, Friedemann : “Lute Construction in the Renaissance and l-Luḥūn wa-n-Nagham (Mukhtaṣar al-Mūsīqā fī Taʾlīf a-n-Nagham
the Baroque”, The Galpin Society Journal 27 |1974-5-1| wa Ṣinʿat al-ʿūd) [from Manisa (Turquie), MS. 1705, fos 110vo-123ro]
[doi: 10.2307/841751. url: http://www.jstor.org/stable/841751] ‫رسالة الكندي في اللحون و النغم‬, MaṭbaꜤat Shafīq |Baghdād, 1965|.
p. 21–30. 78. KINDĪ (AL-), Yaʿqūb ibn Isḥāq Abū Yūsuf (0801?-0867?) and Yūsuf
64. HILARIAN, Larry Francis : “The Structure and Development of the SHAWQĪ : Risālat al-Kindī fī Khubr ṣināʻat al-taʼlīf, Dār al-Kutub al-
Gambus (Malay-Lutes)”, The Galpin Society Journal 58 |2005-5-1| Miṣrīya |Al-Qāhira, 1996|.
[url: http://www.jstor.org/stable/25163827] p. 66–216. 79. LĀDHIQĪ (AL-), Muḥammad ibn ꜤAbd-al-Ḥamīd :
65. IKHWĀN A-Ṣ-ṢAFĀ’ : Rasāʾil Ikhwān a-ṣ-Ṣafā’, Turāth al-ꜤArab 1/4 (‫محمد بن عبد الحميد الالذقي الرسالة الفتحية )في الموسيقا‬, editors ‫الحاج هاشم محمد الرجب‬
(vols.), Dār Bayrūt li-ṭ-ṬibāꜤa wa-n-Nashr |Beirut - Lebanon, 1983|. and Hāshim Muḥammad Rajab (a-r-) |Kuwait, 1986a|.
66. IVRY, Alfred L. : “al-Kindī and the Muʿtazila: A Philosophical and 80. LĀDHIQĪ (AL-), Muḥammad ibn ꜤAbd-al-Ḥamīd : A-r-Risāla al-
Political Reevaluation”, Oriens 25/26 |1976-1-1| [doi: 10.2307/ Fatḥiyya (fī-l-Mūsīqā), A-s-Silsila a-t-Turāthiyya 16, Al-Majlis al-
1580657. url: http://www.jstor.org/stable/1580657] p. 69–85. ꜤArabiyy li-t-Turāth wa-l-Funūn wa-l-Ādāb |Kuwait, 1986b|.
67. JARGY, Simon and Alexis CHOTTIN : “La Musique Arabe”, editor Ro- 81. LEONI, Stefano A. E. : “Kanz al-Tuhaf (Al-Musiqi). The Casket of
land-Manuel, Histoire de la musique Tome I - Des Origines à Jean- (Music) Rarities: Ars Musica and Musica Practica between Islam
Sébastien Bach, Encyclopédie de la Pléiade 1/2 (vols.), Gallimard |Pa- and Christianity”, International Review of the Aesthetics and Sociology
ris, 2001| p. 526–571. of Music 27 2 |1996-12-1| [doi: 10.2307/3108345. url: http://
68. KĀTIB (AL-), Al-Ḥasan ibn Aḥmad ibn ʿAlī : La Perfection des connais- www.jstor.org/stable/3108345] p. 167–183.
sances musicales : un traité de musique arabe du XIe siècle [Kitāb 82. LIU, Donsheng, 劉東升, Quanyou YUAN and 袁荃猷 : Zhongguo
Kamāl Adab al-Ghināʾ], translator Amnon Shiloah, Bibliothèque yinyue shi tujian [An illustrated guide for Chinese Music history], Ren-
d’études islamiques, ISSN 0151-1750 ; 5, Librairie Orientaliste Paul min yinyue |Beijing, 1988|.
Geuthner |Paris, 1972|. 83. MAALOUF, Shireen : History of Arabic Music Theory, change and con-
69. KĀTIB (AL-), Al-Ḥasan ibn Aḥmad ibn ʿAlī : ‫[ كتاب كمال أدب الغناء‬Kitāb tinuity in the Tone Systems, Genres, and Scaled., Université Saint-Es-
Kamāl Adab al-Ghināʾ], editors Ghaṭṭās ꜤAbd-al-Malik Khashaba, prit de Kaslik |Kaslik - Lebanon, 2002|.
Maḥmūd Aḥmad Ḥifnī (al-), and ‫ الهيئة المصراة‬,‫محمود أحمد الحفني‬ 84. MAKHLOUF, Hamdi : Orientalism in Musicology - frets-Kindī for CIM
‫ العامة للكتاب‬Al-Hayʾa al-Miṣriyya al-ꜤĀmma li-l-Kitāb |1975|. 09 Hamdi Makhlouf, .MOV, .MP4, Documentary |2009a|
70. KHALIDI, Tarif and Abī-l-Ḥasan ʿAlī ibn al-Ḥusayn ibn ʿAlī al- [url: http://www.hamdi-makhlouf.com/library/biblio/
MASʿŪDĪ : “Masʿūdī’s Lost Works: A Reconstruction of Their Con- makhlouf/cim09/video-1-kindi.mp4, http://cermaa.foredofico.
tent”, Journal of the American Oriental Society 94 1 |1974-1-1| org/wp-content/uploads/2017/10/Orientalism%20in%20
[doi: 10.2307/599728. url: http://www.jstor.org/stable/599728] Musicology%20-%20frets-kindi%20for%20CIM%2009%20
p. 35–41. Hamdi
71. KHAWĀRIZMĪ (AL-), Abū ꜤAbd-al-Lāh Muḥammad ibn Aḥmad ibn %20Makhlouf.mp4].
Yūsuf al-Kātib : Liber Mafâtîh al-Olûm [Mafātīḥ al-Ꜥulūm], editor G. 85. MAKHLOUF, Hamdi : Orientalism in Musicology - frets-Ṭaḥḥān for CIM
Vloten (van), E. J. Brill |[Leiden ?], 1895|. 09 Hamdi Makhlouf, .MOV, .MP4, Documentary |2009b|
72. KHAWĀRIZMĪ (AL-), Abū ꜤAbd-al-Lāh Muḥammad ibn Aḥmad ibn [url: http://www.hamdi-makhlouf.com/library/biblio/
Yūsuf al-Kātib : Mafātīḥ al-ꜤUlūm, editor Jawdat Fakhr-a-d-Dīn, makhlouf/cim09/video-2-tahhan.mp4, http://cermaa.foredofico.
Dār al-Manāhil |Beyrouth, 1991|. org/wp-content/uploads/2017/10/Orientalism%20in%20
73. KHULAʿĪ (AL-), Muḥammad Kāmil ‫ كتاب الموسيقي الشرقي‬: ‫محمد كامل الخلعي‬ Musicology%20-%20frets-Tahhan%20for%20CIM%2009%20
Kitāb al-Mūsīqī a-sh-Sharqiyy [The book of oriental music / Le livre de Hamdi%20Makhlouf.mp4].
la musique orientale] |1904|. 86. MAKHLOUF, Hamdi : Fretting of the ʿūd according to (al-) Kindī, mp4
74. KIESEWETTER, Raphael Georg : Über die Musik der neueren Griechen, |Paris (France), 2009c/2020| [url: https://www.youtube.com/
Breitkopf und Härtel |1858|. watch?v=d7TTlnH_pKM&feature=youtu.be].
75. KINDĪ (AL-), Yaʿqūb ibn Isḥāq Abū Yūsuf (0801?-0867?) : ‫مؤلفات‬ 87. MAKHLOUF, Hamdi : Fretting of the ʿūd according to Ibn a-ṭ-Ṭaḥḥān,
‫ الكندي الموسيقية‬Muʾallafāt al-Kindī al-Mūsīqiyya: 1. Risāla fī Khubr mp4 |Paris (France), 2009d/2020| [url: https://www.youtube.
Ṣināʿat at-Taʾlīf {British Museum, MS. Or, 2361, fos 165ro-168ro, pos- com/watch?v=demT-hpcX1s&feature=youtu.be].
sibly 160ro-163ro; 2. Kitāb al-Muṣawwitāt al-Watariyya min dhāt 88. MANIK, Liberty : Das arabische Tonsystem im Mittelater, Tuta Sub Ae-
al-Watar al-Wāḥid ilā dhāt al-ʿAsharat Awtār [Bodleian Library gide Pallas, E.J. Brill |Leiden, 1969|.
(Oxford), MS. Marsh, 663, p. 246-265], 3. Risāla fī Ajzāʾ Khuba- 89. MASʿŪDĪ, Abī-l-Ḥasan ʿAlī ibn al-Ḥusayn ibn ʿAlī al- : Murūj A-dh-
riyya fī-l-Mūsīqā [Berlin, MS. Ahlwart, 5503, fos 31vo-35vo], (4. Dhahab wa Maʿādin al-Jawhar, 10th century 4 (vols.), Al-Maktaba
Mukhtaṣar al-Mūsīqā fī Taʾlīf a-n-Nagham wa Ṣinʿat al-ʿŪd [Berlin, al-ʿAṣriyya |Sidon - Beirut, 1987|.
MS. Ahlwart, 5531, fos 22 ro-24vo, supposedly from Euclidos]), 5. A- 90. MATHIESEN, Thomas J : Apollo’s lyre: Greek music and music theory
r-Risāla al-Kubrā [aw al-Kitāb al-Aʿẓam] fī-t-Taʾlīf [Berlin, MS. Ahl- in antiquity and the Middle Ages, University of Nebraska Press |Ne-
wart, 5530, fos 25 ro-31 ro, in fact the Risāla fī-l-Luḥūn wa-n-Nag- braska - EU, 1999|.
ham], 9th century, editors Zakariyyā Yūsuf and ‫زكراا اوسف‬, Maṭbaʿat 91. MEYNARD, M. C. Barrier de : “Ibrahim, fils de Mehdi, Fragments
Shafīq (‫| )المجمع العلمي العراقي‬Baghdād, 1962a|. historiques. Scènes de la vie d’artistes au IIIe siècle de l’Hégire (778-
76. KINDĪ (AL-), Yaʿqūb ibn Isḥāq Abū Yūsuf (0801?-0867?) : “Kitāb al- 839 de notre ère)”, Journal Asiatique ou recueil de mémoires d’ex-
Muṣawwitāt al-Watariyya min dhāt al-Watar al-Wāḥid ilā dhāt al- traits et de notices relatifs à l’histoire, à la philosophie, aux langues et à
ʿAsharat Awtār [Bodleian Library (Oxford), MS. Marsh, 663, p. la littérature des peuples orientaux 6 XIII |1869-1| p. 201–342.
246-265]”, Muʾallafāt al-Kindī al-Mūsīqiyya [Kindī's writings on

194
 Amine Beyhom Was the Early Arabian ʿūd “fretted”?

92. MUNAJJIM (AL-), Yaḥyā ibn ʿAli ibn Yaḥyā : Risālat ibn al-Munajjim Mediaeval Studies. Studies and texts, ISSN 0082-5328 ; 92, Pontifical
fī-l-Mūsīqā wa Kashf Rumūz Kitāb al-Aghānī, editor Yūsuf Shawqī, Institute of Mediaeval Studies |Belgique, 1989|.
Dār al-Kutub |Cairo, Egypt, 1976|. 114. SAWA, George Dimitri : “The Kitāb al-Aghānī”, Routledge |United
93. NADĪM (IBN A-N-), Muḥammad ibn Isḥāq al-Warrāq : Kitāb al-Fihrist, Kingdom, 2001-8-1| [url: http://glnd.alexanderstreet.com/View/
Librairie Khayyāt (rue Bliss) |Beirut, s.d.|. 330237#page351] p. 351–356.
94. NEUBAUER, Eckhard : “Der Bau der Laute und ihre Besaitung nach 115. ṢAYDĀWĪ (A-Ṣ-), ‫ شمس الدان الصيداوي الدمشقي‬Shams-a-d-Dīn (a-d-Di-
arabischen, persischen und türkischen Quellen des 9. bis 15. mashqī) : BNF Ms. or-2480 Kitāb al-InꜤām fī MaꜤrifat al-Anghām wa
Jahrhunderts”, Zeitschrift für Geschichte der Arabisch– Islamischen Sharḥihā = Kitāb fī MaꜤrifat al-Anghām wa Sharḥihā
Wissenschaften, Institut für Geschichte der Arabisch-Islamischen (‫| )كتاب في معرفة األنغام وشرحها‬15th century|.
Wissenschaften an der Johann Wolfgang Goethe-Universität 116. ṢAYDĀWĪ (A-Ṣ-), ‫ شمس الدان الصيداوي الدمشقي‬Shams-a-d-Dīn (a-d-Di-
|Frankfurt am Main, 1993| p. 279–378. mashqī) and Thérèse (soeur Berthe) ANTAR : Kitāb al-InꜤām fī MaꜤri-
95. NEUBAUER, Eckhard : “Ibn al-Nadīm”, Grove Music Online |2001a|. fat al-Anghām wa Sharḥihā = Kitāb fī MaꜤrifat al-Anghām wa
96. NEUBAUER, Eckhard : “Mawṣilī, al-: (2) Isḥāq al-Mawṣilī”, Grove Sharḥihā (‫[ )كتاب في معرفة األنغام وشرحها‬traduction et commentaire par
Music Online |2001b| [. Thérèse B. Antar du manuscrit BNF Ms. or-2480] (Exploitation de la
97. NEUBAUER, Eckhard : “Iṣfahānī, al-”, Grove Music Online |2001c|. couleur en musique : Livre de la connaissance des tons et leur explica-
98. PAOLOLIUTAIOPD : Fastening gut frets |2010-12-5| [url: https:// tion), Chemaly and Chemaly |Beyrouth, 2001-15th century|.
www.youtube.com/watch?v=6aItqGZ0zJs]. 117. SHEPHERD, Martin : Luteshop’s Top Tips - Tying Frets |2016-10-23|
99. PARISOT, Jean (1862-) : Musique orientale : Conférence prononcée [url: https://www.youtube.com/watch?v=Qs_pXOoBVLU].
dans la salle de la Société Saint-Jean, le 28 février 1898, Schola Can- 118. SHILOAH, Amnon : RISM: the theory of music in Arabic writings (c.
torum |Paris, 1898|. 900-1900) : descriptive catalogue of manuscripts in Libraries of Euroe
100. PLATO : The Republic Of Plato, translator Benjamin Jowett |1908| and the U.S.A., Henle |1979|.
[url: http://archive.org/details/in.ernet.dli.2015.223393]. 119. SHILOAH, Amnon : “The Arabic Concept of Mode”, Journal of the
101. POCHÉ, Christian : “ꜤŪd”, Grove Music Online |2001|. American Musicological Society 34 1 |1981-4-1|
102. PRATT, Vaughan : “Algebra”, Stanford Encyclopedia of Philosophy [doi: 10.2307/831033. url: http://jams.ucpress.edu/cgi/doi/10.
|2007| [url: http://plato.stanford.edu/entries/algebra/]. 2307/831033] p. 19–42.
103. QIFṬĪ (IBN AL-), Jamāl-a-d-Dīn Abū-l-Ḥasan Alī Ibn Yūsuf, August 120. SHILOAH, Amnon : La musique dans le monde de l’islam : une étude
MÜLLER and Julius LIPPERT : Taʾrīkh al-Ḥukamāʾ, Dieterich |1903| socio-culturelle, translator Christian Poché, Les Chemins de la mu-
[url: http://archive.org/details/TarikhAlHukama]. sique (Paris. 1992), ISSN 1242-5540 |2002|.
104. RASHĪD, Ṣubḥī Anwar : Tārīkh al-ʿ Ūd, Dār ꜤAlāʾu-d-Dīn |Damas - 121. SHIRWĀNĪ (AL-), Fatḥ-al-Lāh : Majalla fī-l-Mūsīqā [‫ مجلة في الموسيقى‬MS.
Syrie, 1999|. 3449, Topkapi (Istanbul), 95 fos], editor Eckhard Neubauer, Institut
105. ROSEN, Frederic : The Algebra of Mohammed Ben Musa, The Orien- für Geschichte der Arabisch-Islamischen Wissenschaften |Frank-
tal Translation Fund |London, 1831|. furt am Main, 1986|.
106. RŪḤĀNĀ, Sharbil ‫ منهج العود‬: ‫شربل روحانا‬, CNSML (Conservatoire Natio- 122. SHIRWĀNĪ (AL-), Fatḥ-al-Lāh and Muḥammad ibn ꜤAbd-al-Ḥamīd
nal Supérieur de Musique - Liban) |Beirut, 2001|. LĀDHIQĪ (AL-) : La musique arabe (4) – I. Traité anonyme [attribué à
107. SACHS, Curt : The History of Musical Instruments |1940| Shirwānī (a-sh-) – Majalla fī-l-Mūsīqā‫ ]مجلة في الموسيقى‬dédié au Sultan
[url: http://archive.org/details/the_history_of_musical_instru- Osmānlī Muḥammad II (XVe siècle). II. Al-Lādhiqī, Traité al-Fatḥiyya
ments_curt_sachs]. (XVIe siècle), translator Rodolphe (d’) Erlanger 4/6 (vols.), Librairie
108. {ṢAFADĪ (A-Ṣ-)}, Khalīl ibn Aybak Ṣalāḥ-a-d-Dīn : ‫خليل بن أابك الصفدي‬ Orientaliste Paul Geuthner |Paris, 1939|.
‫ رسالة في علم الموسيقا‬Risāla fī ʿIlm al-Mūsīqā, editors ʿAbdul-Majīd Dyāb, 123. SHIRWĀNĪ (AL-), Fatḥ-al-Lāh and Muḥammad ibn ꜤAbd-al-Ḥamīd
Ghaṭṭās ʿAbd-al-Malik Khashaba, ‫األستاذ غطاس عبد الملك خشبة‬, and ‫الدكتور‬ LĀDHIQĪ (AL-) : La musique arabe (4) – I. Traité anonyme [attribué à
‫ تراثنا‬,‫ عبد المجيد دااب‬Turāthuna, ‫ الهيئة المصراة العامة للكتاب‬Al-Hayʾa al- Shirwānī (a-sh-) ‫ ]مجلة في الموسيقى‬dédié au Sultan Osmānlī Muḥammad
Miṣriyya al-ʿĀmma li-l-Kitāb |Le Caire, 1991-1297?-1363) XIVe II (XVe siècle). II. Al-Lādhiqī, Traité Al-Fatḥiyya (XVIe siècle), editor
siècle|. Christian Poché, translator Rodolphe (d’) Erlanger, ed. 2e en fac-
109. SALMA (A-N-NAḤAWĪ AL-LUGHAWĪ), Abū Ṭālib al-Mufaḍḍal (ibn) : similé, Les Geuthner (Paris), ISSN 1294-6656 4/6 (vols.), Librairie
Kitāb al-Malāhī wa Asmāʾihā min Qibal al-Mūsīqā, editor ʿAbd-al- Orientaliste Paul Geuthner |Paris, 2001|.
Malik Khashaba Ghaṭṭās, 9th century, Al-Hayʾa al-Miṣriyya al- 124. SIDDIQI, M. R. : “Mathematics and Astronomy”, A history of Muslim
ʿĀmma li-l-Kitāb |Cairo, 1984|. philosophy: With short accounts of other disciplines and the modern
110. SAWA, George D. : “The Survival of Some Aspects of Medieval Ara- renaissance in Muslim lands, ed. Mian Mohammad Sharif, Harras-
bic Performance Practice”, Ethnomusicology 25 1 |1981-1-1| sowitz |Allemagne, 1966| [url: http://www.
[doi: 10.2307/850975. url: http://www.jstor.org/stable/850975] muslimphilosophy.com/hmp/] p. 1277–1292.
p. 73–86. 125. SĪNĀ (IBN) OR AVICENNA (980?-1037) : “Kitāb a-sh-Shifā’ -
111. SAWA, George D. : “Oral transmission in arabic music, past and Riyādiyyāt 3 - Jawāmiʿ ʿIlm al-Mūsīqā”, Wizārat a-t-Tarbiya wa-t-
present”, Oral tradition 1989 |2002| [url: http://www.journal. Taʿlīm, Al-Maţbaʿa al-Amīriyya |Cairo, 1956|.
oraltradition.org/files/articles/4i-ii/4_1-2_complete.pdf 126. SOLIE, Ruth A. : “Melody and the Historiography of Music”, Journal
#page=265] p. 254–65. of the History of Ideas 43 2 |1982-4-1| [doi: 10.2307/2709205.
112. SAWA, George Dimitri : “The Status and Roles of the Secular Musi- url: http://www.jstor.org/stable/2709205] p. 297–308.
cians in the Kitāb al-Aghānī (Book of Songs) of Abu al-Faraj al- 127. SPENCER, Robert : “How to Hold a Lute: Historical Evidence from
Iṣbahānī (D. 356 A. H./967 A. D.)”, Asian Music 17 1 |1985-10-1| Paintings”, Early Music 3 4 |1975-10-1| [url: http://www.jstor.
[doi: 10.2307/833741. url: http://www.jstor.org/stable/833741] org/stable/3125403] p. 353–354.
p. 69–82. 128. STERLING, Bruce and Greg BEAR : Mozart en verres miroirs, Denoël
113. SAWA, George Dimitri : Music performance practice in the early |Paris, 1996|.
ʿAbbāsid era : 132-320 AH/750-932 AD, Pontifical Institute of

195
NEMO-Online Vol. 5 No. 9 – November 2020

129. STEWART, Devin : “The Structure of the Fihrist: Ibn al-Nadim as his- 142. WACHSMANN, Klaus, James W. MCKINNON, Robert ANDERSON, et
torian of Islamic legal and Theological Schools”, International Jour- al. : “Lute”, Grove Music Online |2007-2012|.
nal of Middle East Studies 39 3 |2007| p. 369–387. 143. WALZER, Richard : “New Studies on Al-Kindī”, Oriens 10 2 |1957-
130. SUBRAMANIAN, Karaikudi S. : “An Introduction to the Vina”, Asian 12-31| [doi: 10.2307/1579634. url: http://www.jstor.org/stable/
Music 16 2 |1985-4-1| [doi: 10.2307/833772. url: http://www. 1579634] p. 203–232.
jstor.org/stable/833772] p. 7–82. 144. WIKIPEDIA CONTRIBUTORS : “Muhammad ibn Musa al-Khwarizmi”,
131. ṬAḤḤĀN (IBN A-Ṭ-~ AL-MŪSĪQĪ), Abū-l-Ḥusayn Muḥammad ibn al- Wikipedia |2017a-9-26| [url: https://en.wikipedia.org/w/index.
Ḥasan : Ḥāwī al-Funūn wa-Salwat al-Maḥzūn [transcription dactylo- php?title=Muhammad_ibn_Musa_al-Khwarizmi&oldid=8024
graphiée de Zakariyyā Yūsuf], editor Zakariyyā Yūsuf, Al-Majmaʿ al- 87345].
ʿArabiyy li-l-Mūsīqā |Baghdad, 1976|. 145. WIKIPEDIA CONTRIBUTORS : “Early Muslim conquests”, Wikipedia
132. ṬAḤḤĀN (IBN A-Ṭ-~ AL-MŪSĪQĪ), Abū-l-Ḥusayn Muḥammad ibn al- |2017b-10-1| [url: https://en.wikipedia.org/w/index.php?title=
Ḥasan : Compendium of a Fatimid court musician – Ḥāwī al-Funūn Early_Muslim_conquests&oldid=803209553].
wa-Salwat al-Maḥzūn [Reproduction du manuscrit Funūn Jamīla 539 146. WIKIPEDIA CONTRIBUTORS : “Abu al-Faraj al-Isfahani”, Wikipedia
de la Bibliothèque nationale, Dār al-Kutub, du Caire], editor Eckhard |2017c-10-5| [url: https://en.wikipedia.org/w/index.php?title=
Neubauer, Facsimile Editions ; 52, Institut für Geschichte der Ara- Abu_al-Faraj_al-Isfahani&oldid=803974536].
bisch-Islamischen Wissenschaften |Frankfurt am Main, 1990|. 147. WIKIPEDIA CONTRIBUTORS : “Boethius”, Wikipedia |2020-9-20|
133. THE BIRYANI BOYS : Amazing Sitar Player |New-York?, 2008| [url: https://en.wikipedia.org/w/index.php?title=Boethius&ol-
[url: https://www.youtube.com/watch?v=hTPxqUtlLdo]. did=979470413].
134. TĪFĀSHĪ, Aḥmad ibn Yūsuf : Mutʻat al-asmāʻ fī ʻilm al-samāʻ, editor 148. WRIGHT, Owen : “Ibn al-Munajjim and the Early Arabian Modes”,
Rachid Sellami, ed. al-Ṭabʻah al-ūlá, al-Majmaʻ al-Tūnisī lil-ʻUlūm The Galpin Society Journal 19 |1966-4-1| [doi: 10.2307/841911.
wa-al-Adāb wa-al-Funūn |Tūnis, 2019|. url: http://www.jstor.org/stable/841911] p. 27–48.
135. TSUGE, Gen’Ichi : “Musical Instruments Described in a Fourteenth- 149. WRIGHT, Owen : The modal system of Arab and Persian music, 1250-
Century Persian Treatise” |2013| p. 25. 1300 : an interpretation of contemporary texts. |1969|.
136. URMAWĪ (D. 1294), Ṣafiyy-a-d-Dīn ꜤAbd-al-Muʾmin ibn Yūsuf ibn 150. WRIGHT, Owen : “Music at the Fatimid court: the evidence of the
(ab-ī-l-Ma)Fākhir (al-) 1 : ‫صفي الدان االرموي‬. Kitāb al-Adwār fī-l-Mūsīqā Ibn al-Ṭaḥḥān manuscript”, L’Égypte fatimide: son art et son histoire
[Livre des cycles musicaux – MS. Istanbul, Nuruosmaniye 3653, fos 1a- (Actes du Colloque organisé à Paris les 28, 29 et 30 mai 1998), ed. M.
49a (1ro-49ro), possiblement autographe] 2. A-r-Risāla a-sh-Shara- Barrucand, Presses de l’Université de Paris-Sorbonne |Paris, 1999|
fiyya fī-n-Nisab a-t-Taʾlīfiyya [L’épître à Sharaf-a-d-Dīn sur les propor- p. 537–545.
tions musicales – MS. Ahmet III 3460, fos 1a-68a (1ro-68ro), copie da- 151. WRIGHT, Owen : “Fārābī, al-”, Grove Music Online |2001a|.
tant de 827H/1424C], editor Eckhard Neubauer, ( ‫معهد تاراخ العلوم‬ 152. WRIGHT, Owen : “Ibn Sīnā”, Grove Music Online |2001b|.
‫ )العربية واإلسالمية‬Institut für Geschichte der Arabisch-Islamischen 153. WRIGHT, Owen : “Kindī, al-”, Grove Music Online |2001c|.
Wissenschaften |Frankfurt am Main, 1984|. 154. WRIGHT, Owen : “Ibn Zayla”, Grove Music Online |2001d|.
137. URMAWĪ (D. 1294), Ṣafiyy-a-d-Dīn ꜤAbd-al-Muʾmin ibn Yūsuf ibn 155. WRIGHT, Owen : “Shirwānī, al-”, Grove Music Online |2001e|.
(ab-ī-l-Ma)Fākhir (al-) ‫ كتاب األدوار في الموسيقى‬: ‫[ صفي الدان االرموي‬Kitāb al- 156. WRIGHT, Owen : “Lādhiqī, al-”, Grove Music Online |2001f|.
Adwār (fī-l Mūsīqā)], editors Maḥmūd Aḥmad Ḥifnī (al-), Ghaṭṭās 157. WRIGHT, Owen : “Ibn al-Ṭaḥḥān”, Grove Music Online |2001g|.
ꜤAbd-al-Malik Khashaba, ‫غطاس عبد الملك خشبة‬, and ‫ الهيئة‬,‫محمود أحمد الحفني‬ 158. WRIGHT, Owen : “Ikhwān al-Ṣafāʾ”, Grove Music Online |2001h|.
‫ المصراة العامة للكتاب‬Al-Hayʾa al-Miṣriyya al-ꜤĀmma li-l-Kitāb |Le Caire, 159. WRIGHT, Owen : “Munajjim, al-”, Grove Music Online |2001i|.
1986|. 160. WRIGHT, Owen : “Al-Kindi’s Braid”, Bulletin of the School of Oriental
138. URMAWĪ (D. 1294), Ṣafiyy-a-d-Dīn ꜤAbd-al-Muʾmin ibn Yūsuf ibn
and African Studies 69 01 |2006| p. 1–32.
(ab-ī-l-Ma)Fākhir (al-) ‫ صفي الدان االرموي‬: “Kitāb al-Adwār – ljs235”,
161. WRIGHT, Owen, Christian POCHÉ and Amnon SHILOAH : “Arab mu-
[D. 1293, XIIIe siècle] |2001| [formerly available at
sic”, New Grove Dictionary of Music and Musicians 1/29 (vols.), Ox-
url: http://sceti.library.upenn.edu/ljscollection/index.cfm,
ford University Press |2001| p. 797–833.
http://dewey.library.upenn.edu/sceti/ljs/PageLevel/index.cfm?
162. ZAKIR HUSSAIN and BRIJ NARAYAN : Raga Lalit / Raga Bairagi Bhairav
ManID=ljs235].
- Brij Narayan, Sarod / Zakir Hussain, Tabla, CD, Nimbus Records
139. URMAWĪ (D. 1294), Ṣafiyy-a-d-Dīn ꜤAbd-al-Muʾmin ibn Yūsuf ibn
|1990|.
(ab-ī-l-Ma)Fākhir (al-) ‫ صفي الدان االرموي‬and ʿAlī ibn Muḥammad
163. ZAPICO, Pablo : Cómo poner un traste de tripa / How to put a gut fret |2015-
[JURJĀNĪ (AL-)] : La musique arabe (3) – I. A-sh-Sharafiyya ou Épître
3-25| [url: https://www.youtube.com/watch?v=IDVvXFEWiTE].
à Sharaf a-d-Dīn. II. (Commentaire anonyme du) Kitāb al-Adwār ou
164. ZAYLA (IBN), al-Ḥusayn ibn Muḥammad : ‫الكافي في الموسيقي‬, editor Za-
Livre des Cycles musicaux, translator Rodolphe (d’) Erlanger, ed.
karīyyā Yūsuf, ‫( دار القلم‬Dār al-Qalam) |1964|.
orig., La Musique Arabe 3/6 (vols.), Librairie Orientaliste Paul
Geuthner |Paris, 1938|.
140. URMAWĪ (D. 1294), Ṣafiyy-a-d-Dīn ꜤAbd-al-Muʾmin ibn Yūsuf ibn
(ab-ī-l-Ma)Fākhir (al-) ‫ صفي الدان االرموي‬and ʿAlī ibn Muḥammad
[JURJĀNĪ (AL-)] : La musique arabe (3) – I. A-sh-Sharafiyya ou Épître
à Sharaf a-d-Dīn. II. (Commentaire anonyme du) Kitāb al-Adwār ou
Livre des Cycles musicaux, editor Christian Poché, translator Ro-
dolphe (d’) Erlanger, ed. 2e en fac-similé, La Musique Arabe 3/6
(vols.), Librairie Orientaliste Paul Geuthner |Paris, 2001|.
141. UṢAYBIʿA (known as IBN ABĪ ~ ), Muwaffaq a-d-Dīn abī-l-ʿAbbs
Aḥmad ibn al-Qsim ibn Khalīfa ibn Yūnis a-s-Saʿdī al-Khazarjī:
ʿUyūn al-Anbāʾ fī Ṭabaqāt al-Aṭibbāʾ, 12th century, Al-Maktaba al-
Wahbiyya |s.l., 1882|.

196