The modal system of Arabian and Persian music,

1250 - 1500:
An interpretation of contemporary texts.

Owen Wright
 ProQuest N um ber: 10752705

All rights reserved

INFORMATION TO ALL USERS

The quality of this reproduction is d e p e n d e n t upon the quality of the copy subm itted.

In the unlikely e v e n t that the a u thor did not send a c o m p le te m anuscript

and there are missing pages, these will be noted. Also, if m aterial had to be rem oved,
a n o te will ind ica te the deletion.

uest
ProQuest 10752705

Published by ProQuest LLC(2018). C opyright of the Dissertation is held by the Author.

All rights reserved.

This work is protected against unauthorized copying under Title 17, United States C o d e
M icroform Edition © ProQuest LLC.

ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 4 8 1 0 6 - 1346
 \cL)

abstract

The period 1250-1300 witnessed the emergence of a corpus

of musical theory which was to provide a framework for most
of the important treatises composed during the following
centui'ies. One aspect of the theory is examined here: the
analysis of scale (intervals and combinations of intervals);
and on the basis of this examination an attempt is made to
present a coherent picture of one aspect of musical practice:
the modal system.

The introduction discusses the relationship between

Arabian and Persian music, and describes briefly the attitude
towards music prevalent in Islamic society and their effect
upon the literature.

Part 1 presents a, critique of the preconceptions

involved in the analysis of intervals, and an annotated
transcription of the species and fixed scales recorded during
the period.

Part 2 begins by examining the various groups to which

the modes were assigned in order to ascertain whether they
can be justified on purely structural grounds. Consideration
of the criteria employed in determining grades of consonance
is followed by a discussion of the modal system in terms of
basic units and the ways they combine. Comparison between
two accounts of the system suggests a pattern of historical
development; and also that the earlier account misrepresents
certain features.

rart 3 continues the examination of the later account,

further material from this is given, revealing a second set
of unit combinations which is compared and contrasted with
the first. finally, the examples of notation are transcribed
and briefly discussed.

The relevant texts are presented in part 4.

 \ ' /

contents

acknowledgements 8
note on transliteration; abbreviations 6
introduction 7
part 1 34
chapter 1 35
chapter 2 71
part 2 129
chapter 3 130
chapter 4 150

chapter 5 195
part 3 216
chapter 6 217
chapter 7 242
chapter 8 316
part 4 361

kitab al-adwar 362

durrat al-taj 453
index of technical terms 471
bibliography 490
 ackno wl edg emen t s

I am happy to he able to take this opportunity of

expressing my gratitude to those who have helped me at
various stages in the writing of this thesis: to Professor
0. F. Beckingham, for his encouragement, and for casting
a critical eye over the introduction; to Dr. P. 0. Gandjei
and Mr. A. Haidari, who patiently corrected many mistakes
in the preparation and translation of the Persian text; and
to Miss S. Khammash, for her kind assistance in checking
the Arabic text.

Above all I should like to record my indebtedness to

Mr. N. A. Jairazbhoy, who has generously spent much time
and energy in reading and discussing this work with me, and
whose criticisms and suggestions have invariably proved both
stimulating and fruitful.
 note on transliteration

The transliteration from Persian is unsystematic.

Arabic loan-worcls are transliterated as if from Arabic.
Persian words used frequently in Arabic texts are sometimes
given in their Arabicized form (e.g. awaz, kardaniya, '
zankula), but more frequently the Persian consonantal values
and occasionally the vowels also, are. retained (e.g. buzurg,
chahargah, dugah, panjgah, parda; kuchek, segah).

In the song transcriptions, initial hamza is indicated.

abbreviations

Only one abbreviation is used:

BM - British Museum
introduction
 (8)

In the field of musical theory the period from 1250 to

1500 is dominated by the Systematist school, founded by $afx

1
al-Dxn ‘Abd al-Mu}min al-Urmawx. The present work is based

upon an examination of two treatises by him, and a section

of an encyclopaedia by the first of his many followers, Qufb

2
al-Dxn al-Shxrazx. The latter is especially valuable for

the amount of material it contains on practice, with which

we are here primarily concerned. Nevertheless, in all three

texts this aspect is of secondary importance, and a

consideration of their analytical procedures is an essential

prerequisite to any interpretation of the modal system.

While it would be foolish to deny the presence of empirical

elements, it is clear that their statements about the

intervals and scales in common use result from passing the

raw material of practice through a filter of theoretical

presuppositions, mainly about the nature of consonance*

Denied access to the raw material itself, we must enquire int<

1 One of the most famous musicians of his day, gafx al-Dxn

was a prominent figure at the court of the last Abbasid
caliph, al-Musta*gim (1243-58)* His life was spared when
Baghdad was sacked, and the Mongol conquei'ors took him__into
their service. Of his two treatises on music, the kitab
al-adwar was probably written in 1 2 5 2 , and the risala al~
sharafiyya, composed for his pupil Sharaf al-Dxn Harun,
about 1267* He died in 1294.

2 The polymath Qu^b al-Dxn (1236-1311) was the most brillis

of Nagxr al-Dxn al-Jusi’s pupils, excelling in medicine,
astronomy, and optics. He also wrote on theology and
philosophy, producing an important commentary on the frikmat
al-ishraq of Suhrawardx. The date of the encyclopaedia, th
durrat al-tatj , is unknown, but may reasonably be assumed to
be c1300.
 (9)

the nature of the filter before sifting the residue.

Spaced out over a mere 50 years, the treatises of gaf1

al-Din and Qutb al-DIn differ but little in content. Thus

in general one can deal with them en bloc, and most of the

conclusions to be derived from them may be considered valid

for the whole period. Where they do not coincide, however,

it is sometimes possible to detect a pattern of change which

may be viewed diachronically, as part of an evolutionary

process. In such cases references are made to the evidence

of 14th and 1 5 th century works.

Unfortunately there is little point in trying to enlarge

the perspective by an appeal to authors writing before 1 2 5 0 .

It is true that Ibn Sina furnishes a valuable (if incomplete)

account of the modes used in his day; but if certain lines

of development are discernible in the years 1 2 5 0 -1 5 0 0 one

should be wary of positing connections between a mode used

then and one described two centuries before. Caution is

reinforced by the regrettable absence of material in the

intervening period. This, unless one insists on talcing into

account one or two encyclopaedias reproducing in abbreviated

form opinions better expressed elsewhere, is quite barren of

 (10)

1
theoretical works on music- Nor can recourse to writings of

a non-specialist nature do much to dispel our ignorance of

the fundamentals of musical practice during the 12th century.

There are no later works comparable to the kitab al-aghanj

for the wealth of its technical, as well as social and

biographical, information.

An historical introduction to the study of the 13th

century modal system is therefore hardly possible- Instead,

it may be worthwhile discussing briefly a number of points

which relate to the hypothesis that in the late 13th century

Arabian and Persian art-music shared the same modal system.

However surprising this might appear, it is the obvious

conclusion to be drawn from the texts themselves. In effect,

the available evidence suggests that although Arabian and

2
Persian music may have differed In other respects, they not

only employed the same intervals and combinations of interval

but were hardly to be distinguished in the way they put them

1 The one theoretical treatise of any stature to be compos

after the kitab al-shifa? and before the works of gafi al-
Dln is the kitab al-kafl fi al-musiqa by Ibn Zayla, a pupil
of Ibn Sina. This in any case dates from the first half of
the 11th century, and its originality resides mainly in the
treatment of rhythm: it contains no new information about
the modes.
Details on other texts of the period may be found in
Parmer, The Sources of Arabian Music, d3~4-8.

2 Notably in the relative popularity of certain rhythms

(see the kitab al-adwar, ch- 13)*
 (11)

1
to melodic use. It was perhaps only to be expected that

the majority of post-13th century treatises, whether Arabic

or Persian, should adopt gafx al-Dxn’s terminology, and that

their analyses should be developed within an uniform

theoretical framework. The period is, after all, one in

which independence of thought and enquiry was on the decline,

at least in the sciences. Nevertheless, some of these

treatises do provide an account of the modal system which is

more than a mere repetition of what gafx al-Dxn had to say on

the subject. The fact that they make not the slightest

mention of differences between the Arabian and Persian modal

systems cannot therefore be attributed to their being

predominantly derivative in nature: nor can it be claimed

that they show a total lack of concern for the realities of

musical practice. In addition we may note that there are

one or two 14th and 15th century texts in which Persian views

on a certain mode are criticised by an Arab author, or vice

2
versa; yet the disagreements relate not to the structure of

the mode, but to such subsidiary matters as its classificatioi

1 The names of the modes themselves are a mixture of Arabi<

and Persian. An even more positive indication of identity
or at least near identity is furnished (admittedlyover a
century later) by ‘Abd al-Qadir b. Ghaybx who states e.g.
that a particular section of a song is differently termed
by Arabs and Persians (maqagid al-alban, 102) - a remark
that only has point if the structure and content of that
section were common to both.

2 E.g. in the kitab ghayat al-matlub fx *ilm al-angham

wa ’1-durub of Muhammad b. ‘Isa al-Baghdadx.
 (12)

within the whole repertoire. This last indication is all the

more valuable in that these particular texts do not adhere to

the theory of the Systematist school, and thus constitute

an independent source of information.

It might be objected, however, that this evidence is less

than convincing: that silence need not confirm similarities

and could equally well mask dissimilarities. Further, that

the basic proposition flies in the face of historical

probability, for prior to the 10th century Arabian and Persiar

music were certainly not identical, and in present-day

practice there is also a marked difference.

With regard to the contrast between the unity postulated

for the 13th century and the divergencies observable now,

two general remarks may be made. The first is that if one

considers the two modal systems in their present form, the

similarities are far more striking than the dissimilarities.

In both melodic organization is centred chiefly on the

tetx^achord, and there is further a considerable degree of

identity with respect to the various possible arrangements of

1
intervals within the tetrachord. The second is that in view

of this it is not unreasonable to suppose that at an earlier

1 Comparative studies of melodic structure have not yet

been undertaken. But although these would establish
certain lines of demarcation, it would appear that the
major distinction is to be sought in the field of formal
organization.
 (13)

stage the relationship was far closer; and that the two

began gradually to drift apart in the 16th century when the

military and religious confrontation between Persia and the

Ottoman empire, which by that time had subjugated the greater

part of the Arabic-speaking world, imperilled the process of

cross-fertilization brought about by personal contact between

1
court musicians.

It is rather more difficult to put forward a plausible

defence for the proposition that unity succeeded the

diversity of the 10th and earlier centuries. As we have

seen, there is a dearth of reliable material for the period

1050-1250. But matters are made worse by the fact that

although the indisputable differences between Arabian and

Persian practice were recognized by a number of earlier

2
writers, none of them do more than baldly state that the two

are not the same.

However regrettable, it is perhaps not surprising that

this should be so. The general difficulty of adequately

verbalizing musical experiences precludes the likelihood of

any valuable contribution being found in literature at large;

and works of a specialist nature unfortunately tend to embark

1 In other fields mutual exchanges and cultural influences

may have continued. Musical practice, on the other hand,
could not be transmitted by the written word.

2. Including al-Kindi, Ibn Khurdadhbih and al-Igbahanl.

 (14)

on two courses in neither of which could problems of a

descriptive and comparative nature be dealt with.

Scientific enquiry, on the one hand, touches on musical

practice - albeit at a tangent - when defining the size and

relationship of intervals. But it cannot go beyond this

within the limits imposed by its predominantly mathematical

outlook: music was, after all, part of the quadrivium. We

are provided, as it were, with the precise dimensions of

individual (and ideally formed) pieces of masonry; but from

these we are unable to infer the shape and effect of the

whole edifice.

Gosmological speculation, on the other hand, while

emphasizing the position of music as an integral part of the

macrocosmic order, leaves even less room for the development

of an analytic or aesthetic terminology with which two

musical systems might be contrasted or evaluated. In these

respects music suffered much more than literature, for there

was no parallel to the impulse, furnished by a causally

related complex of factors, which led to the creation of an

1
adequate vocabulary of literary criticism.

1 The 14th century sharh mubarak shah bar adwar asserts

that in a work of his on poetics (presumably based on
Aristotle) al-Farabi discussed the relationship between
certain types of music and poetic and dramatic genres.
No later writer seems to have pursued this line of enquiry.
 (15)

We have therefore no means of determining in exactly what

respects Arabian and Persian music differed. Hence it is

impossible to suggest how, in the course of time, they could

have shed these differences. We can do little more than

put forward the view that a process of interaction culminating

in the establishment of a common idiom would be in accordance

with the general pattern of cultural and social development.

Also, we may note the possibility that the area over which

this presumed common idiom may be thought to prevail was

relatively modest in the period 1250-1300, and was gradually

extended during the following two centuries. It is safe to

assume that gafl al-Din7s comments on practice relate

primarily to the art-music of Baghdad, where he spent much of

his life, while Qu^b al-BIn was no doubt more familiar with

the traditions of Shiraz, his native city, and Tabriz, where

he lived for many years. Besides Iraq and Western Persia,

on the other hand, later theorists of the Systematist school

come from as far apart as Khurasan and Transoxiana to the

East, Syria and Anatolia to the West. This may, it is true,

be no more than a coincidence; nevertheless it seems

reasonable to suppose that any initial fusion should take

place where contact between Arabs and Persians was closest.

In any case, North Africa and Spain must be excluded from the

regions for which the hypothesis of unity may be entertained.

As a result of the destruction wrought by both Christian

conquest and Muslim intolerance, the Arab West provides but

 (16)

few texts dealing with music, and as far as the modal system

is concerned they are generally uninformative.

Musical contact between Persia and Arabia, however, is

1
well documented as far back as the Jahiliyya* In the early

years of Islam we hear of direct borrowings, and during the

Umayyad period some of the most distinguished Arab singers

spent part of their musical apprenticeship absorbing Persian

2
styles. Por this to have been a meaningful experience, and

for borrowings to have taken place, there must have existed

beneath the differences a considerable degree of similarity

in both modal organization and forms of melodic expression.

After the consolidation of Arab rule in Persia the interaotior

no doubt increased, both in pace and extent. By the 9th

century the structure of urban society in the Middle East

seems to have been fairly uniform, and cultural attainments

and aspirations also. In almost every sphere of human activil

Persian ideas and methods were being assimilated by the Arab

aristocracy in the major cities of Iraq as well as in Persia

itself. It would thus be quite natural to find that a musical

rapprochement took place which furthered the evolution of a

common fund of modal forms. If so, we may assume that the

1 The contact is of course millenary. But most references

are in works of the 3rd and later centuries of Islam, and
hence are rarely to be trusted if concerned with events
prior to the 6th century A,D.

2 Including Ibn Misjaft, Ibn Surayj, and Juways. The Persic

musician Nashl$ was immensely popular in Medina.
 (17)

process was substantially complete by the end of the 10th

1
century* But whatever uniformity resulted should not be

thought of in terms of a rigid system. It is far more likely

that there should have emerged a somewhat flexible lingua

franca to which a number of regional traits could be added

without undue disturbance. It is probably variations of this

kind that the Ikhwan al-gafa’ (10th century) are referring to

when they distinguish the music of Arabs, Persians (presumably

2
the inhabitants of Fars), and Daylamites. The last two

would certainly not have had independent musical languages,

but rather local and characteristic inflections of the same

tongue. The kitab al-shifa* of Ibn Sina, the last important

source prior to the treatises of gafi al-Din, supplies a few

concrete details about this language. But any conclusions

we may draw from them would again have to be tentative, for

although Ibn Sina adds to our knowledge of the way tonal

space was divided by detailing the intervallic organization

3
of several modes, he still affords us no insight into the

main elements of melodic syntax through which, no doubt,

1 The justification for this dating is partly of a negative

nature: about this time references to the contrast between
Arabian and Persian practice become few. More important,
the major theorists, al-Farabi and Ibn Sina, make no mentior
of it. (The individuality of the tara* iq of Khurasan, notec
by al-Farabi, is a matter of form rather than substance.)
Further, al-Farabi5s discussion of the tunbur baghdadi indi­
cates the gradual dissemination of a common scale structure.

2 rasa’i l , i, 196.

3 BM. MS. Or. 11190, ff. 204v-205.

 (18)

regional differences could at least in part be defined*

Lacking more precise information, we are entitled to say

no more than that by the end of the 10th century certain modes

and scales appear to have gained general acceptance. One is

therefore tempted to conclude that the synthesis to which

texts of the 15th to 15th centuries point is the result of a

developmental process whereby those features forming as it

were a common denominator expand and at the same time are

welded together into a coherent system, the organization of

which may be shown to be governed by a number of underlying

structural principles. But this presupposes that the

intervening period was one of regular growth and consolidatior

which is open to doubt. Having argued that in the 16th

century musical unity could be jeopardized as a result of

political conflict, it might equally well be objected that

to all appearances conditions in the 12th and 15th centuries

should have impeded, rather than furthered, progress towards

unity. The final decline of the Abbasid caliphate was marked

by ever increasing political and economic fragmention, the

effect of which was to retard, if not to reverse, the movemenl

towards cultural levelling. Thus, far from being conducive

to the establishing of musical homogeneity, one might have

expected the fissiparous tendencies which were so strong at

this time to foster rather regional autonomy, the affirmation

of local characteristics, and a gradually increasing

divergence of both style and substance - all the more so

 (19)

since there was in music no parallel to the fixed ideals

obtaining for the written word, and thus no means of

imposing uniformity at a time when the cultural ambitions

which were a concomitant of the desire for political

independence tended to promote an upsurge in artistic

activity.

While due attention must be paid to such factors, it

should be borne in mind that the ebb and flow of dynastic

fortunes did not bring about any decisive changes in the

fabric of society. Although the peasantry suffered from

extortionate and often arbitrarily imposed taxation, the

general pattern of urban life was but little affected by the

substitution of a land-based economy for a monetary one. The

music with which gafl al-Din and Qutb al-Dxn are concerned

evolved in conformity with the canons of taste prevalent

among the upper and middle-class town-dwellers, and there

can be little doubt that the stability and continuity which

characterized the cultural outlook of these classes were in

the long run far more significant than the passing whims of

rulers, despite the lavish rewards which the latter could

bestow. As we have seen, post-Abbasid treatises indicate the

general, or at least ever-widening acceptance of a unified

modal system, the regularity and complexity of which would be

inexplicable had it not been preceded by a considerable period

of development, experiment, and consolidation. Seen against

the background of urban conditions, rather than that of

 (20)

military and political reversals, such a process becomes

more readily explicable. Also, it may be noted that except

during the brief periods of violent upheaval communications

were affected but rarely by the rivalries between various

factions. In consequence there was little to prevent

musicians moving about freely. It is true that artists of

the first rank, once attached to a court, did not often

experience the urge or necessity to seek their fortune

elsewhere, but the slave-girls who showed sufficient talent

to become their pupils were merchandise to be exported as

demand arose. They too may have excercised a levelling

influence on the musical taste of the upper classes which

they served, but at the same time had the opportunity to

modify. In the event it is more than possible that by the

mere fact of their encouraging music as a vital art-form the

virtually independent dynasties of the Abbasid decline may

have helped to maintain the composite tradition and, indeed,

to promote a further integration. One might compare the

connection between the pretensions of the numerous German and

Italian principalities in the 18th and early 19th centuries

and the strength of the operatic tradition in those countries.

Among further factors contributing positively to the

growth of musical conformity one might mention the military

band. This had, apart from its primary function, an important

r6le to play as an essential adjunct to court ceremonial.

It became in effect an expensive status-symbol with,

 (21)

interestingly enough, the added possibility of expressing

1
distinctions of rank in the way it was used* In view of

this it is natural to suppose that the accoutrements, and

music, of the military nawba in the capital of the empire

were slavishly copied by independent, and even would-be

independent, rulers. One may cite as supporting evidence the

survival of the nawba tradition at princely courts in India

almost down to the present day - one of the most striking

examples of Islamic musical influence in the sub-continent.

It is true that for the military band sheer quantity of sound

was often more important than quality, but on the whole the

innate conservatism of imitated pomp should if anything have

helped reinforce the general stability of the musical scene.

Of considerably greater importance is religious music.

But here we are faced with a wide range of forms, used in a

variety of contexts. Only in some of these is it likely that

there existed a certain degree of uniformity over wide areas.

The main reason for this is the fact that most religious
2
authorities viewed music with grave suspicion, and in

1 See Farmer, A History of Arabian music, 206-208, and his

article "Tabl khana", Encyclopaedia of Islam, suppl., 1938,
217- 222 .
2 There are a considerable number of treatises which deal
specifically with the legal status of music. While some
speak in defence, the majority condemn. The viewpoint of the
four main law-schools lies somewhere between disapproval and
outright condemnation.
The Sufi attitude is discussed briefly below.
 K.22)

consequence no effort was made to impose any canonic standards

upon the music that was used to accompany ceremonies of a

non-official nature. As these often included elements of

local customs and beliefs (many of considerable antiquity)

tolerated by Islam, it is likely that apart from containing

archaic features this music was closely linked to the

prevailing styles of folk-music, and thus differed somewhat

from province to province. In addition the strong antipathy

on the part of the legists ensured that music was all but

banished from official ceremonies. Hence it could never

become an integral part of ritual, as in the Christian church.

Hence also the lack of any parallel in Islam, despite the

prominence of the similarly orientated doctrine of ta*thir,

to the hermeneutic use of the Jewish prayer-modes. In spite

of this, however, the cantillation of the Q u r ’an became an

art in itself, and one which could not have been divorced

entirely from the modes and melodic characteristics of seculai

1
art-music.

A much closer connection may be observed between secular

art-music and that of the Sufis. Already by the 11th century

(at the latest) the place of music within the liturgy derived

1 The connection was never admitted by the legists. Indeec

the legal fiction of total separateness was buttressed by
the use of a different technical vocabulary for cantillatior
However, there are cases - just as in mediaeval church
music - of popular songs being used in religious contexts
(see Ibn Qutayba, al-ma6arif, 232).
from the o r i g i n a l l y spontaneous dhikr was assured, and it is

possible that this may have provided a further unifying

factor. At a later date Sufi ideas were often disseminated

through poetry, and so the songs associated with particular

poems would in all probability gain wide currency. Also, the

Sufi fraternities began to develop their ov/n ceremonies, and

these frequently included the performance of music. However,

although the period between Ibn Sina and gafl al-Din saw the

founding or institutionalizing of some of the more important

orders, those which laid particular emphasis on the use of

music and dancing as a means towards ecstatic communion date

from somewhat later. In view of this it would be imprudent

to suggest that the growth of the fraternities contributed,

however obliquely, to the development of a standard idiom

during that time. Nevertheless, it is likely that the music

heard at the more intimate Sufi gatherings was, considered

solely as an organization of sonorities, substantially the

same as that to which the discussions of the philosophers and

1
theorists relate. If so, it may be assumed that this near­

identity persisted even when the fraternities had developed

their own musical forms, for during the 15th century, if not

before, some of the ternary and aksak rhythms characteristic

of Sufi dancing began to find their way into the secular

1 That heard at large open-air gatherings would probably

have differed in both vocal and instrumental style (and
choice of instruments) from court music with its
predominantly chamber-music textures.
 (24)

1
repertoire.

Despite these links it would he idle to expect Sufi

writings to contribute to our knowledge of musical practice*

For the Sufis music, with its fleeting evocation of divine

harmony, provided a means whereby the adept could progress

towards a comprehension of ultimate reality. Preoccupied

with its transcendental significance, they naturally omitted

all mention of the purely physical factors involved.

This symbolic approach was by no means exclusive to

Sufism. The influence of the esoteric philosophies current

in the Near East during the early centuries of the Christian

era not only contributed to the development of Islamic

mysticism, but also played a significant part in encouraging

metaphysical and cosmological speculation. During the Abbasic

caliphate the majority of thinkers were less concerned with

analysis than with synthesis - elucidating the essential

unicity of Nature. Thus for them music could hardly become

an object of study in itself: rather it formed part of an

uninterrupted chain of phenomena between the various links of

which correspondences might be established. Already in the

1 Their origin is indicated by the appelation "sama*I” .

Such rhythms are mentioned in al-Ladhiqi’s kitab al-fatfriyyi
a later work (c 1500) in the Systematist tradition. The
"classical" characteristics of at least some present-day
Sufi music (as instanced in H. Ritter, "Der Reigen der
tanzenden Derwische", _2Le.it:schr if t fur vergl eicliende
Huaikwiase.nscJaaft, Jg. 1, Nr. 2, 1933) suggests that the
two have remained closely related.
 (25)

9th century al-Kindi related the strings of the lute and the

rhythmic modes to the humours, elements, to perfumes and

1
character-traits. Later similar series were to he establish.ee

for the melodic modes, and even al-Din, who otherwise

ignores this type of speculation, devotes a chapter to their

emotional content.

When seeking material on musical practice it is therefore

to the Peripatetics and the theorists who followed them that

we must turn. Por the philosophers imbued with the objective

rationalism of the Aristotelian tradition music was one of

2
the mathematical sciences. Thus their work utilizes and in

a few cases, notably in the kitab al-musiqa al-kabir of

al-Parabi, extends the methods of analysis evolved by the

Greek theorists. Unfortunately much of the resulting matter

was also adopted, so that all too often we are regaled with

an exposition of tetrachord species the majority of which were

completely foreign to Arabian and Persian practice, followed b

various arrangements of these within the systema teleion. On

occasion, indeed, the series of hypothetical tetrachords is

even enlarged, from which it is evident that the study of the

numerical relationships of intervals was being pursued for

its own sake, and not for its value as a descriptive tool to

1 See Farmer, Sa*adya Gaon on the influence of music, 5-9*

2 Cosmological speculation is openly rejected by Ibn SIna

in the introduction to the section on music in the kitab
al-shifa? (D’Erlanger, la musique arabe, ii, 106).
 (26)

be applied in a fresh context. Where al-Farabl and Ibn SIna

do provide information about the intervals used in Arabian

music is in their detailed accounts of the main melody

instruments, especially the *u d , a short-necked lute.

Nevertheless, they afford us little insight into the way

these intervals were organized in practice. Interest in their

functional r6le was the exception, the rule being the concern

taken over the manipulation of the numerical relationships

to which they can be reduced. In the last resort, the

philosopher’s approach to music was prescriptive rather than

descriptive.

With the theorists of the Systematist school the balance

is only partially restored. While retaining the analytical

method and some of the definitions of their predecessors,

they frequently jettison much of the Greek ballast, with

correspondingly greater emphasis being placed upon the

tetrachord and pentachord species actually in use. Neverthe­

less, for all the wealth of detail on intervals and scale,

their treatises are as a rule hardly forthcoming about the

melodic characteristics of the modes. Once abstracted into

a simple succession of intervals they become mere objects to

be juggled with, and much ingenuity is expended in detailing

the ways in which they can be derived from each other and

transposed throughout the theoretical gamut. Needless to say,

this is of little relevance to what performers were actually

doing. Furthermore, the one chapter in £>afl al-Din’s kitab

 (27)

al-adwar ostensibly devoted to practice is perhaps the most

superficial in the whole work, however valuable the examples

1
of notation it contains* No attempt is made to remedy its

deficiences in the several commentaries on the kitab al-adwar:

none of them devote to it a fraction of the attention lavished

on the more abstruse sections, and in some the examples of

notation are even reduced in number, as indeed they are in

gafi al-DIn’s second (and otherwise more extensive) treatise

on music. What was notated was evidently less important than

the method of notation.

The almost total disregard for those aspects of musical

practice not amenable to definition in terms of number is

somewhat puzzling. At a much earlier period considerable

interest had been shown in such matters as formal procedures,

the relationship of melodic structure to prosodic pattern,

vocal and instrumental technique, expression and dynamics.

Yet none of them is dealt with by gafi al-Din, and with rare

exceptions the same is true of his successors. That this line

of enquiry, which might have resulted in an aesthetic, was

not sustained during the latter part of the Abbasid period

may be attributed in part to the increasing importance

attached to Neo-Platonism, Hermeticism, and allied currents

of thought which stressed symbolic and esoteric aspects to

the detriment of exact observation. But this is not in itself

1 The final chapter (fl mubasharat al-*ama,l).

a sufficient cause - and in any case the triumph of one

particular philosophical approach is unlikely to stem solely

from its intellectual cogency: the intellectual temper of

society at large also has an important r61e to play in

determining the acceptability of any given set of ideas, and

it is likely that where music is concerned social factors

were instrumental in establishing the predominance of the

emblematic, non-analytical approach.

In effect, the position of music within Islamic society

has always been ambiguous, and attitudes towards it have

varied considerably. Poetry was able to overcome initial

hostility and gain early acceptance as one of the "Arab

sciences"; but music, although an equally indispensable

means of emotional expression, and one moreover not explicitly

condemned in the Qur’an, remained suspended between acceptance

and rejection. Patronized by the court, it was frowned upon

by the law-schools; a means of spiritual liberation for the

Sufi, it was excluded from most religious ceremonies; an

encouragement to the faithful in battle, it was widely

considered conducive to immorality. However facile such

contrasts, they do reflect an inability to establish for

music a generally accepted social function.

This uncertainty also affected the status of the musician.

A cursory glance at the kitab al-aghanl and similar works

will reveal that the virtuosi were held in high regard at

court, and that their artistry was more than generously

 (29)

rewarded. (They also suffered on occasion from the

capriciousness of their masters, hut not noticeably more so

than others.) Their fame is attested by the abundance of

biographical material available, and is in sharp contrast to

the anonymity of the craftsman and architect. The picture

of musical life at court presented in the kitab al~aghani

may be considered valid for the following centuries also:

the fabulous sums lavished on the best artists during the

heyday of Abbasid rule may have become a thing of the past,

but the degree to which extravagance could go was restricted

not by a decline in appreciation but by worsening economic

circumstances. Nevertheless, apart from acquiring positions

of rank, outstanding musicians were still able to amass

considerable fortunes - gafl al-DIn himself is a case in

point.

For all this music had never been, and was not to become,

a completely respectable calling. The musician’s rights

within the Muslim community were in certain respects

circumscribed: positions of religious authority were denied

him, and a (free) musician’s evidence might be disregarded in

1
a court of law. Further, one finds cases, even among
2
pampered courtiers, of artists exhibiting feelings of guilt.

1 See Farmer, A History of Arabian music, 31.

2 See for example kitab al-aghani, x, 163*

(Outside the court hostility to music could no doubt be more

forcibly expressed, although the social implications of this

can only be guessed at.) A number of philosophers may, it is

true, have prided themselves on their musical skill, but they

constituted within the total context of Islamic society a

peripheral group which was regarded with considerable

suspicion by the orthodox majority. Nor should it be

forgotten that the fulminations of the more narrow-minded

legists were perhaps not so unrepresentative and ineffectual

as one might like to believe. Although their attempts to

proscribe music entirely were, of course, unrealizeable, they

should be regarded not only as products of excessive puritan

zeal, but also - or rather - as indications of a profound

ambivalence of attitude common to the whole community.

Already in Umayyad times the association of music with

frivolity and licentiousness had been reinforced by the

growing importance among the wealthier classes of the female

slave-musicians (qiyan), and more especially by the unsavoury

reputation of the mukhannathun, a group of effeminates who

were variously drunken, dissolute, homosexual, agents for

prostitutes and, very frequently, musicians. By the end of

the Abbasid caliphate the factors that had facilitated the

emergence of this latter class no longer existed; however,

1 Further material on these two groups may be found in

Akel, Studies in the social history of the Umayyad period
as revealed in the Kitab""al-aghani.
the opposition they provoked merely reinforced opinions

already held, and did not disappear with them: music was to

remain a part of human activity too essential to be ignored,

but sufficiently dangerous to have to be banished as far as

possible into the shadows of social opprobrium.

The danger is a specifically moral one. There is no

trace of the ancient notion that the integrity of the state

could be damaged by changes in the musical system, since this

was condemned root and branch. The attitude of the Muslim

legists is rather an extension of that expressed in "The

Republic": Plato wished to see certain modes forbidden

because of their deleterious effect, while for the legists

all music corrupted. The aura of impiety and frivolity

surrounding music is established by its perennial association

with wine-drinking - in entertainment-literature the two are

virtually inseparable. Less obvious, but of far greater

significance, is the connection with sexuality. This aspect

was underlined by the frequently erotic nature of the verses

set to music, to which on occasion may have been added a

degree of sexual attraction or tension between listener and

performer. (For the qiyan, as for courtesans in other

cultures, music was sometimes no more than a useful

accomplishment to be cultivated for utalitarian purposes.)

As might be expected, the erotic potential of music was

emphasized by those bent on attacking it, but in general it

may be said to have been less a formally recognized factor

 (32)

than one which, because of the irrational elements resulting

from the numerous taboos involved in sexual relationships,

1
operated mainly at a subconscious level* It is perhaps for

this reason that the excessive emotional reactions to music

always seem so curiously unfocussed.

In the event, it is hardly surprising' that those who

wrote about music should have treated it so circumspectly.

The most drastic solution, and at the same time the simplest,

was that adopted by the Sufis: as with poetry, the emotional

language is accepted in full, but understood to be symbolic.

For those thinkers concerned with cosmological schemes and

implications the dangers of music are also exorcised by

stressing its esoteric aspects, and in particular by pinning

it in a web of numerical sets and associations, so that it

again becomes symbolic, but this time providing a link with

the abstract symmetries of the universe, rather than a means

of liberation and mystical insight as with the Sufis. The

Peripatetic philosophers and theorists, too, escape into the

realm of pure number, although physical and physiological

aspects are also dealt with. However, the basic procedures

of musical communication - the ways in which the raw material

of sound is organized into an expressive whole, and the

conventions which govern and safeguard comprehensibility -

1 Of. the interesting comments of J. Berque in The Arabs:

their history and future, London, 1964 (ch. 11).
 (33)

are only treated indirectly, so that we can form only the

most shadowy picture of the musical system as a whole.

Nevertheless, the factual nature of their approach to, e.g.,

instruments and the intervals produced on them, does furnish

sufficient material for us to be able to undertake a fairly

detailed examination both of the modal system and of the

surviving examples of notation in which its practical

application may most clearly be discerned.

part 1
 (35)

chapter 1

The kitab al-adwar, first of ^afi al-DinJs two treatises

on music, consists of fifteen fairly brief chapters. At

first sight there appears to be little logic in their

arrangement, apart from the fact that chapters 1 to 11 deal

with different aspects of the same subject-matter. Within this

group, however, one may detect a pattern (not necessarily a

conscious one on the part of the author) consisting of an

alternating presentation of original and non-original

material. Beginning in the manner of earlier treatises with

definitions, gafx al-Din rushes abruptly (chapter 2 ) into an

account of his innovatory 17-note octave gamut. Previous

models are again followed in the ensuing discussion of

interval relationships, dissonance and consonance (chapters

3 to 5 )? although we may note that within this section the

approach to-dissonance is untraditional in that the criteria

are couched in terms of non-permissible sequences of intervals^

rather than in terms of single intervals. Similar alternations

are discernible in chapters 6 to 1 1 , which deal with possible

scales; how to produce one of these on a stringed instrument

(chapters 7 8 ); the most common modes; their similarities

of structure; and how to transpose them throughout the gamut.

 (56)

The remaining chapters are a rather mixed bag, containing an

academic digression and, in chapter 1 3 » a relatively extensive

discussion of rhythm including an account of contemporary

practice. It seems reasonable to regard the two final

chapters as addenda. They deal with the emotional impact of

1
certain modes, and with notation, the latter being presumably
2
a novel contribution.

By virtue of this process of alternation, gafl al-DIn

successfully integrates material derived from practice into

an inherited framework of theory. -&s a result the reasonably

clear distinction maintained between the two in earlier works

is blurred. There is, for instance, no trace of the empirical

procedures for siting frets described by al-Farabi and Ibn

5
Sma. This, it may be remarked, is not a criticism, but

merely an indication of one of the major difficulties with

which the present study is faced. In fact, given the brevity

of the kitab al~adwar, the theory embodied in chapters 1 to

11 is to a remarkable degree both fully developed and

coherent. Compared with previous treatises in the same

1 As noted in the introduction, such aspects were felt to

be of considerable importance, although generally disregardec
by writers of the Peripatetic tradition. It is possible
that in this chapter gafi al-DIn is merely reflecting
received opinions.

2 Notation had been used, although not for the same precise
purpose, by both al-Kindl and Ibn Sina.

3 Al-ParabI, kitab al-muslqa al-kablr, 516-515; D ’Erlanger

la musique arabe, ii, 2 3 5 .
 (37)

ultimately Greek-derived tradition it also evinces consider­

able originality.

The same cannot be said of his second treatise, the

risala al-sharafiyya. This may, not unjustly, be described

as a much expanded and in some respects more finished version

of the previous work. The expansion results in part from

the inclusion of material which had been presented by earlier

writers, but was of dubious relevance to the structure of

1
Arabian and Persian music in the 13th century.

The section on music in Qutb al-DIn9s encyclopaedia, the

durrat al-taj, is made up of five discourses (maqalat)

prefaced by a short introduction. The first three deal

respectively with the definition of basic elements and the

physical propagation of sound; with the numerical relation­

ships of intervals; and with the addition and subtraction of

intervals. Thus far there is nothing worthy of comment,

since we are presented with what is little more than an

adaptation-cum-translation of the corresponding sections in

gafl al-DIn 9s risala al-sharafiy,ya. This does not, however,

mean that Qutb al-DIn is just another unblushing plagiarist,

1 It may be doubted whether much of this material

(especially that concerned with the analysis of chromatic
and enharmonic tetrachords) had ever been of relevance to
Arabian practice. It is possible that in this work - quite
naturally so in view of the circumstances under which it
was written - gafi al-DIn was tempted to demonstrate his
grasp of ancient theory as much as to restate his own.
 (38)

for apart from later passages which hear witness to his

critical acumen, there is evidence here too of independent

judgement. Furthermore, he readily acknowledges his deht to

gafi al-DIn, and where he is not explicitly quoting or

expounding his teaching he refers to him in terms of the

utmost respect. Attention need only be drawn here to the

testimony these passages provide of the high esteem in which

gafl al-DIn5s works were held so soon after his death.

Dependence on gafl al-DIn naturally does not stop at this

point, and the last two discourses are also squarely based

on the text of the risala al-sharafi.yf

ya. However, interest

is quickened by the fact that in these two discourses, which

deal to a certain extent with the realities of the practical

art, Qu^b al-DIn gives an account of the modes and rhythms

in current use which both enlarges upon, and in some respects

1
differs from, that of gafi al-DIn. fhese differences are

all the more curious for the general lack of explanation or

justification. But Qutb al-DIn5s great competence in musical

theory is in evidence throughout these chapters, despite the

preponderantly derivative nature of the material, and in

1 Although later works confirm much of the material pre­

sented by Qutb al-Dxn, the description of the two major
groups of modes, shudud and awazat, found in the kitab al-
adwar, is generally reproduced unchanged in treatises of
the 14th and 15th centuries. fhis veneration of gafl al-DIn
seems to have affected European scholars, for it is rare to
find any mention of later accounts of the modal system.
Certainly the considerably larger and more detailed list in
^ ie hurrat al-taj has been universally ignored.
 (39)

1
addition he must have had a thorough grasp of practice*

Thus there is no possible reason, on the occasion when he

diverges from the version of his mentor, for doubting the

validity of the modal structures he presents* Lastly, it

should be noted that he concludes this section of the durrat

al-ta(
j with a composition far longer, more complex, and more

exactly notated than those in the kitab al-adwar.

* *

Before proceeding to a discussion of the modes as

described by these two authors, a number of points concerning

the nature of the octave division established by gafl al-DIn

need to be clarified. When one considers the painstaking

thoroughness with which Qu£b al-DIn and later theorists,

especially the author of the late 14th century sharb mubarak

2
shah bar adwar, classified and notated the modes in

accordance with the 17-note gamut evolved by gafl al-DIn, it

is a great temptation to suppose that this gamut accurately

1 Bor all that he was later to be criticized by ‘Abd al-

Qadir as being an amateur in such matters. However, ‘Abd
al-Qadir seems to have misunderstood, or misrepresented,
some of the points made by Qu£b al-DIn. (See the comments
on the passage translated in chapter 7 *)

2 This important treatise will be referred to henceforth

as the sharh. It was completed in 1375* and has been
attributed? by Farmer to al-Jurjam (d. 1413)-
 (4-0)

reflects the intervallic relationships then obtaining.

However, there is reason to believe that the picture thus

presented is an oversimplification, in some ways a formalized

abstraction, and that in fact §>afl al-Din attempted to

squeeze the raw material of the practioner’s art into the

straitjacket of a mathematically coherent (and therefore

intellectually respectable) system, and in so doing distorted

it.

1
The scalar system itself is, for all its sophistication,

a basically straightforward development from the simple

Pythagorean scale of whole-tones and limrnas. In it each

whole-tone is divided into two limmas and a comma, and the

]

full octave arranged thus:

2
G A B e d e f g

L L 0, L L 0, L, L L G, L L G, L, L L G

The following indicates the order in which the 17 notes are

3 ‘
established on the monochord, the first note of each pair

being the one from which the second is derived:

1 Parry described it as "the most perfect scale ever

devised 11 (The art of music, London, 1896, 29)*

2 The choice of G is purely arbitrary, and it should not

be taken to refer in any way to a fixed pitch. It will be
used throughout the present work as the starting point or
lowest note, equivalent to alif (the Systematist notation
being an alphabetic one based upon the abe
jad sequence).

3 kitab al-adwar, chapter 2 (BM. MS. Or. 136, ff. 4-4-v).

This apparently random arrangement consists in fact of three

clearly differentiated sections. gafi al-Din provides first

the octave, the fifth, and the notes limiting the conjunct

tetrachords into which he later analyses the majority of the

scales he lists. He then goes on to define, according to the

1
procedure found in earlier works, the notes within one

Pythagorean tetrachord (G- - c) of whole-tones and limmas.

Finally the remaining notes are added, beginning with those

derived from the last note to be established within the

lower tetrachord (Ab). The notes in each of the three

sections are therefore:

----- S--------
------- i— J---- <>’ t>« — ^ —
------------ b« b.
-<

This method of presentation embodies, as it were, a historical

introduction to his novel scalar system. Elsewhere this is

given in a different order, and it would appear that gafi

al-Din thought of the 17 notes as forming a cycle of ascending

fourths (or descending fifths), although the uninterrupted

1 E.g. al-Khwarizmi*s m afatip a l ^ u l u m and the rasa*il of

the Ikhwan al-gafa*, in which the notes are established on
the *u d , so that the fretting for the one tetrachord
provides the remaining notes of the octave on the other
strings. The order in which the notes of the tetrachord
are presented is the same: G, c, A, B, B b . Ab is a later
addition. As neither of these works delves deeply into
theory, it is likely that this procedure for siting frets
was fairly widely known.
 (42)

sequence begins from B, not G:

This series of fourths is used in the presentation of

1
transposition scales (frabaqat), although as it begins perforce

from G there is one break (g~c - B).

The arrangement of intervals within the whole-tone (L L

G) is somewhat unusual. It may have been suggested, as

2
D 5Erlanger has pointed out, by the identical layout of the

whole-tones and limmas within the tetrachord, and, in most

cases, of the tetrachords and disjunctive whole-tone within

the octave, thus providing an admirable symmetry for seekers

of extra-musical associations and symbolic references. A

more mundane but perhaps more likely postulate is that the

adoption of this order for all the whole-tones of the octave

may have been due to the causes suggested by D ’Erlanger,

while the original acceptance of the L L C division should

be ascribed rather to the influence of the frunbur khurasani

scale, the first tetrachord of which is, as a theoretical

construction, identical with the first tetrachord of the

3
Systematist scale. A further reason which may possibly have

^ kitab al-adwar, chapter 11 (BM. MS. Or. 136, ff. 25-29v).

2 Op. cit., iii, 595*

5 The fretting of the junbur khurasanij, a long-necked lute,

is described by al-Earabx, kitab al-musiqa al-kabir, 6 9 9 -
721.
 (43)

contributed to the choice of the L L 0 , rather than the

L 0 L (or even the 0 L L) division, is that it gives,

within two cents, the major third ( 5 : 4) and minor whole-

tone (1 0 : 9) against the more exact minor third ( 6 : 5)

of the other two. This, however, presupposes that the comma

distinction between major and minor whole-tone, or between

the Just Intonation major third and its Pythagorean equivalent,

was se functionally significant - a crucial point which

will be discussed later with reference to the wusfra zalzal.

Whatever the reason, or reasons, which lay behind gafi

al-DIn5s adoption of the L L C division, the application

of it to every whole-tone in the octave incurs one serious

disadvantage when the scale is considered as a referent for

notation, i.e. is thought of in relation to actual practice.

Given the adherence, for theoretical purposes, to a scale

made up exclusively of limmas and commas, the L L C division

might seem the best fitted to express, with the least distor­

tion, the intervals used within the whole-tone considered as

an integral part of the tetrachord. However, to divide the

disjunctive whole-tone in this way seems quite artificial,

since it is difficult to imagine the melodically meaningful

use of a note one comma below the octave or tonic. In the

song transcribed (in chapter 8 ) from the durrat al-taj, for

instance, there occurs a passage which in the original

notation (with G as tonic) is as follows:

Apart from the formidable difficulties of intonation involved,
—c
it is hard to discern any significance in having g rather

than, say, f#, for the note in question is extraneous to the

1
basic structure of the mode. Similar objections might be

raised against the validity of a note one comma below the

fifth when dealing with a pentachord, or an octave scale in

which the whole-tone from fourth to fifth may assume the

disjunctive r 6 le-

Such a judgement is based on general considerations and

may be thought subjective. However, even if we wish to ignore

as untrustworthy the corroboration supplied by the interval

ratios in the durrat al-taj, sufficient evidence can be found

both in the lcitab al-adwar and in the writings of later theorist

to justify the assertion that in the context of the Systematise

scale of limmas and commas it is the enharmonic L C L (and

not the L L 0) division which provides the closest

approximation to reality for the disjunctive whole-tone,

The mode in question is the upper

-c
pentachord of which is notated as c d e-^ f g. The note
between f and g is therefore either an ornament or melodic
embellishment or, as will be suggested in chapter 8 , part
of a sub-section in another mode which requires fit (and not
-C \
S )•
 (45)

1
whatever its position within the octave* F i r s t l y we may note

the possibility that gafi al-Din himself may have given the

L C L division as an alternative for the upper disjunctive

whole-tone when establishing his octave gamut it the kitab

2
al-adwar. A further and more conclusive indication is

supplied by the same work in the chapter on the causes of

dissonance (fi asbab al-tanafur). One of these is stated to

be the juxtaposition of three whole-tones, and the example

G A B c# is given - except that the upper note is written

-c 5
as d . (This particular instance is paralleled in the sharb,
-c 4
which notates the mode mahuri a s G A B c d e g g,

and elsewhere describes the upper pentachord of the same mode

as containing a whole-tone between e and g"*c , an accurate

5
notation of which would therefore be e - f #•

1 Both D ’Erlanger and Barkeshli come to similar conclusions.

r
fhe former does not discuss the matter, but often gives e. g.
fit as an alternative to g c in his transcriptions (la musique
arabe, iii, passim). Barkeshli ("la musique iranienne",
Encyclopedie dd la Pl&iade: Histoire de la musique, i) goes
so far as to suggest that a better approximation to the
intervallic values obtaining in practice would be gained by
substituting L G L for L L C "dans la plupart des cas",
i.e. not only in the disjunctive whole-tone. The arguments
adduced are not wholly convincing.

2 It is given in one MS. (of four consulted) - BM. Or.

2 5 6 1 , fol. 7 8 .

5 BM. MS. Or. 156, fol. 8 v. 4- BM. MS. Or. 2561, fol. 105v

5 Ibid., fol. 95* In later treatises this mode is notated

as G A B c d e gb g. Despite the fact that the^
modern Persian mahur has the major scale, this change in
notation should probably be regarded not as a correction,
but as an indication of a change in scale structure, gb
standing for ft (mid-way between f and fit).
 (46)

In order to provide a closer approximation to what

obtained in practice, the arrangement of intervals within

gafi al-DIn’s theoretical scale may therefore be amended to:

G A B e d e f g

L L C)
L L 0, L L C, L, LL C, L, L C L
L G Lj

The order of intervals within the whole-tone from fourth to

fifth will be L G L if it is disjunctive or the upper whole-

tone of a pentachord, and otherwise L L G. Given that all

scales may be considered to begin from G there are no cases

which, when analysed in terms of the Systematist scale of

limmas and commas, would require the division L L C in the

1
whole-tone f - g. The whole-tone G - A is never disjunctive.

It has already been noted that the lower tetrachord of

gafl al-Din’s scale is identical with that defined by al-

Farabi for the frmibur lchurasanl. According to al-Barabi the

upper pentachord of the funbur khurasanl scale was:

c d e f g

L C L, L G L, L, C L L

i +C
However, he adds that the frets producing the notes eb and f

were not used, being merely theoretical presences inserted in

1 This is true for the scales listed by gafi al -Dm. Two

short-lived exceptions are cited in the dux^rat al-taj.
 (47)

1
order to facilitate the placing of other frets. The notes

which remain are therefore:

*

c db cit d, dif e, f, f# g

L C L, L+G L, L, C+L L

If we disregard as insignificant the omission of the

enharmonic alternative in the whole-tone f - g, it will be

seen that this differs from the amended version of gafl al-

Din *s scale proposed above only in the whole-tone d - e.

Admittedly, we possess no account of the tunbur khurasanj

other than that of al-larabi, but it is quite reasonable to

suppose that the fretting to which these theoretical values

relate may have survived with but little modification during

the period under consideration. One may compare the equally

theoretical values assigned by Belaiev to the fretting of

2
the modern Azerbaijani and Transcaucasian t a r :

G A B e d e f g

L L G, L-C G L G, L, L G L, L L 0, L, L L+C

Here the order in the whole-tone d - e may perhaps be

attributed to the influence of the Systematist scale.

1 kitab al-musiqa al-kabir, 720-721. As this suggests,

"fret 11 (dastan) may" indicate no more than the position at
which a note is stopped. The use of the term, in what
follows, need not therefore imply a physical presence or-
mark on the fingerboard of an instrument.

2 In "The formation of folk mode systems", Journal of the

International Folk Music Gouncil, xv, 196J.
 (48)

Otherwise there is little difference: an extra note has been

added, and in the upper whole-tone gb takes the place of fS.

This type of octave division would be somewhat cumbersome

to define on the 4u d . In general the player did not exceed

the limit of a single tetrachord on any one string, so that

a fret required on that string might produce unwanted notes

on the others. In this respect the 4ud differs markedly from

long-necked lutes such as the tar and the funbur khurasani,

where a full octave and more may be played on a single string,

which for theoretical purposes could therefore be thought of

as a monochord. Of course, gafi al-Din also established his

octave division on the monocliord, but it may be assumed that

he was thinking throughout in terms of the 4u d , for which the

rigorously symmetrical arrangement favoured by him, and

enthusiastically adopted by later theorists, was extremely

well suited since only two notes are missing in the upper

octave (b and e', taking G as the open note of the lowest of

_c' )
the five strings) and only two are added (c and f .

Had he included both the L L C and the L G L divisions on

the lines suggested above, the number of extra notes would be

increased to eight. It may well be that considerations of

this nature combined with the motives examined by D ’Erlanger

to persuade him when establishing his scale to accept an order

which, although perhaps admirable from a theoretical

standpoint, nevertheless failed to correspond exactly to the

realities of musical practice*

 (49)

There is in addition a further and more important aspect

of the Systematist scale to he discussed: the treatment of

the wusfa zalzal and related notes. The wusfa zalzal,

situated between the second (wusfa) and third (bingir) finger

frets, takes its name from Mangur Zalzal, an 8 th century

lutenist who is credited with having successfully incorporated

it into the hitherto purely diatonic modal system of court

music. Both al-Farabx and Ibn Sina, although they differ

slightly in their definitions of the placing of the wusfa

zalzal on the strings of the Lu d , agree that it produces a

neutral third (approximately 35 0 cents) in relation to the

2
note of the open string. In the Systematist scale the

empirical methods they used to define its position are

ignored, and it is placed resolutely at 384 cents, not a

5
neutral third but a Just; Intonation major third. gafi al-Din

1 Farmer, A History of Arabian music, 118.

2 A1-Farabi gives 354 cents and Ibn Sina 344. The method
of derivation is however the same in both cases: al-Farabi
sites the wusfa zalzal mid-way between the wusfa al-furs
and the third finger fret (kitab al-musiqa al-kabir, 5 1 1 ),
Ibn Sina mid-way between the first and fourth finger frets
(D’Erlanger, la musique arabe, ii, 2 3 5 )* In all probability
the difference results simply from the fact that Ibn Sina
does not recognize the wusfa al-furs fret as defined by
al-Farabi.

Throughout the present work the arbitrary value of 350

cents has been assigned to the wusfa zalzal. This is neither
a correction nor an absolute, but merely a convenient figure.

3 Which is strictly 386 cents. The 2 cents inaccuracy

occurring within the Systematist scale for this and other
intervals will be ignored.
 (50)

concedes , however, that this adjustment is a theoretical

expedient, for after establishing his scale on the *ud he

adds a paragraph which, as far as musical practice is concer-

ned, reinstates the neutral third. In this passage,

interestingly enough, he follows al-Farabi and Ibn Sina,

2
adopting the latter5s placing of the fret. But there is one

small and rather revealing amendment - an odd kind of verbal

disguise drawn over the obvious discrepancy between the

theoretical value of the interval and that occurring in

practice. Having rationalized the neutral third as a Just

Intonation major third the name wusta zalzal is retained for

5
the latter, while the neutral third is termed wusta al-furs.

It is as if, having exorcised the neutral third by regularizing

it within his extension of the Pythagorean system, gafi al-Din

then permitted it to return under an assumed name - the truth

is admitted while appearances are in some measure preserved.

He was also honest enough to admit that the neutral third

fret was frequently used in the music of his day, while the
Just Intonation major third, together with the za 5id, minor

1 This is found only in the risala al-sharafiyya (Bodleian

MS. Marsh 521, fol. 8 8 v ) .

2 gafi al-Din also does not recognize the wusta al-furs

of al-Parabi. His choice is therefore of no significance
with regard to the intonation of the wusta zalzal.

5 This is the name given in earlier works to the Just

Intonation minor third and notes approximating to it, and
£>afl al-DIn does use it elsewhere in this sense. In the
present work the name wusta zalzal will be applied exclusively
to the neutral third, and will be the only name given to it.
second, fret, occurred only rarely. It may be assumed that

when the Just Intonation major third did occur it did not
replace the neutral third but, as in the time of al-Parabl,

was no more than a non-functional variant of the Pythagorean

major third.

The realignment of names and frets may be seen in the

following diagram:

al-Parabi Ibn S m a gafi al-Din

(rnutlaq)
O

90 za 5id
o
vO

98 112 r a ’s a l -dasatm

al-sabbaba muj annab 146

145 .
wusta zalzal \|mu j annab
168 180;

sabbaba 204 204 sabbaba 204 sabbaba

mujannab
al-wusta 294- 294 wusja_ al-furs 294

wusta al- 505 Iwusta al-furs

furs 554 __ 544 wusta zalzal 344]
wusta zalzal
384 wusta zalzal
bingir 408 408 bingir
408 bingir

khinsir 498 498 khingir 498 khingir

1 This statement of gafl al-Din’s is valuable not for

objective accuracy, but for the contrast it draws between
theory and practice.
 (52)

£>afl al-Din also quotes two of the positions given by

al-Farabi for the muj annab fret (mujannab al-sabbaba in al-

Farabi’s terminology). But he adds that for his contemporaries

the most common procedure was to place it half-way between

za?id fret (at 9 0 cents) and the first finger fret,

sabbaba (at 204 cents), and it is the resulting position

which has been entered above. The mujannab fret is thus now

almost exactly a whole-tone below the neutral third, and

this is obviously the value intended and that to which

musicians approximated in practice.

These adjustments and concessions lead one inescapably

to the conclusion that the Systematist scale of limmas and

commas does not provide an accurate account of the intervals

then in use. What it does provide is an ingenious solution

to the text-book problem of how to integrate the Mirrational”

neutral intervals into a system dominated by simple numerical

relationships. In fact, the notes employed within the

tetrachord would appear to have changed remarkably little

since the time of al-Farabi. They may be transcribed as

follows:

1 The extent of the deviation is negligible, for in this

position the mu;jannab fret is 208 cents from the wusta
zalgal of al-Farabi at 554 cents, and 198 cents from that
o~f Ibh STiia at 54-4 cents.
 (53)

1
G Ab At A Bb B$ B c
2
0 90 14-6 204 294 550 408 498

Similarly, in a conjunct tetrachord scale, the notes in the

second tetrachord from c to f were:

c db d1? e eb e1? e f

It has been suggested that gafi al-Dinss original order

of intervals within the whole-tone f - g (which should

correspond to f gb g1? g in practice) is unrealistic, and

should be amended in the first instance to f gb fjf g.

However, there are no two scales among those listed by gafi

al-Din and Qufb al-Din which could be distinguished solely by

the contrast between gb and f# (notated as g"~C). Thus if we

accept the argument, set out below, that the difference of

a comma was of no functional significance, so that both fjf

1 The neutral intervals will henceforth be indicated by

the accidentals I? and $. These respectively lower and
raise the pitch of the note to which they are attached by
approximately a quartertone.

2 The mu jannab (A*!?) is a whole-tone (204 cents) below the

wusta zalzal at 350 cents. It should be stressed that these
two figures are arbitrary, and there might be considerable
fluctuation in the intonation of these two notes. The
other values, which relate to the Systematist scale division,
ignore such possible intonational variants as the Just
Intonation major and minor thirds for the Pythagorean major
and minor thirds.

D 9Erlanger (op. cit., iii, 5 9 0 - 5 9 7 ) reaches similar

conclusions about the nature of the tetrachord in practice.
Instead of assigning an arbitrary value to the wusta zalzal,
however, he adopts the value given by Ibn Sina and taken
over by gafi al-DIn (344 cents).
 (54)

and gb may be considered to lie within the zone of intonation

acceptable for one note the precise pitch of which will be

influence by the melodic context, and which we may write

conventionally as fif, the notes within an octave consisting

of tetrachord, tetrachord, and whole-tone become:

Or Ab Ab A Bb Bb B c db d eb db e^ f fjf g

In the icitab al-adwar all octave scales are held to

consist of tetrachord (below) and pentachord. In most cases,

however, the pentachord may be related to a tetrachord species,

adding a whole-tone above, so that one may distinguish here

two types of scale:

(1 ) tetrachord, tetrachord, whole-tone; and

(2 ) tetrachord, pentachord when f does not occur.

In the latter type there are scales containing, in gafl al-

Dln’s notation, the sequence e“c gb g, and the interval

- c i ‘ 1
e - gb is stated to be a whole-tone. Accepting that the

note represented as e”c may be more realistically notated as

eb, we may add the note ft (a whole-tone above eb) to the

above gamut. In the risala al-sharafiyya, apparently, gafl

al-Din also allows the division

1 BM. MS. Or. 136, fol. 10v.

 (55)

1
(3 ) pentachord, tetrachord.

This is the o n l y one possible - ignoring the potential

division of the pentachord into whole-tone (below) and

tetrachord - for the two scales in the durrat al-taj which

omit c. In addition, it may be suggested that a number of

scales are best thought of as

(4) tetrachord, whole-tone, tetrachord,

despite the fact that this division is not recognized by

theorists of the Systematist school. It should however be

remarked that if such scales contain c and f they could

equally well be placed in category (1 ), so that a subjective

element may be involved in their classification. Particular

2
cases will be discussed in part 2 .

Taking into account all the octave scales listed by gafl

al-Din and Qu£b al-Din, the following may be established as

the notes occurring in each of these categories:

1 This divisionis given for one scale in the translation

by D ’Erlanger (la musique arabe, iii, 1 3 6 ) • correspon­
ding figure in the MS. consulted (Bodleian Marsh 521 * fol.
9 7 ) gives tetrachord, pentachord, but is incomplete in
other respects and therefore not wholly reliable.

2 If (2) has only scales omitting f, it would appear

justifiable to place in (3) only scales omitting A and to
class the remained as a separate category: whole-tone,
tetrachord, tetrachord. This would however contain only
two scales.
 (56)

(1) tetrachord, tetrachord, whole-tone

j _ _ l]-------------------------------- 1! ,

G Ab Ab A Bb Bb B c db db d eb eb e f fit g
(2) tetrachord, pentachord

t \\— - - - - - - - - - - 1
- - - - - - - - T — I
G Ab Bb Bb c db eb eb ft f# g

(5) pentachord, tetrachord

I i-i---------- ; i
G Ab A Bb B c c if d eb e f ft fl g

(4) tetrachord, whole-tone, tetrachord

G Ab Ab A Bb Bb B c c# d eb eb e f ft fit g

It will be seen that of the two possible orders suggested in

the amended theoretical scale for the whole-tone c - d,

L L G and L G L, the former relates to (1) c db db d

and (2) cdb in actual practice, while the latter relates to

c cit d in (3) and (4-)* where cif, as with the fit in the whole
tone f - g, stands for both cit and d b .

When listed separately in the durrat al-tat

j , pentachords

are also notated from alif, G. Thus one could add a further

note, c t , corresponding to the ft within the c - g pentachord.

However, the ct is nowhere found in an octave scale, which

means simply that the pentachords in which it occurs are

never placed below the tetrachord when the two are combined

to form an octave scale. It is likely that if these

1 It is possible that in this class ft and ftt are merely

notational (and intonational) variants of the same note.
 (57)

pentachords functioned as independent entities they would

1
tend to have c (or possibly f) as tonic rather than G. For

all practical purposes the ci may therefore be disregarded.

In gafi al-Din?s treatises scales are given in a simple

alphabetic (abjadi) notation which relates to the 17 notes of

the Systematist octave division, and it would be a straitfor-

ward matter to amend the transcription of this notation in the

light of the above remarks in order to provide what one might

2
hope to be a more precise delineation of the modes as known

to the practitioners. Before attempting to do so, however,

it is necessary to examine the other, diagrammatic method of

representing the scales and species. In the durrat al-taj

the two methods are used conjointly: above the letters of the

notation (written in a horizontal line) are inscribed a number

of semicircles which link points spanning intervals that can

be expressed in simple ratios. These are marked in. Below,

figures are given indicating the relevant string length for

1 gafl al-Din*s notation of all octave scales from the same

note is probably a deliberate simplification for theoretical
purposes. We may assume that convention assigned different
tonics (relative to each other, not to any absoliite pitch
standard) to various modes, although lack of evidence renders
us unable to do more than surmise in which cases this was so.
The one striking exception is pi*ovided by the fact that at a
later period, when the rast scale had assumed the form G A
Bt c d e f^ g, ewas called busayni, from the tonic of
the mode of that name (e ft g a b? c' d' e^ notated by
gafi al-Din as G Ab Bb c d? eb f g).

2 Although the attempt will be made to discriminate between

the terms “scale” and “mode” there may be some confusion and
overlapping, arising from the fact that in the sources modes
are often notated as scales and nothing more.
 (58)

each note, i.e. the ratios are multiplied up to the lowest

common denominator. Such diagrams are also found in the

treatises of gafl al-DIn and elsewhere, but generally they

are employed in the more theoretical sections, so that except

in the durrat al-taj it is only rarely (and incidentally) that

the modes in actual use are defined by this method. The

1
following is a simple example taken from the durrat al~tat
j:

rast

approx.\7 * 6

-e

180 160

In more complex cases not all the possible semicircles are

inscribed, and it would be therefore more correct to speak of

the scale being defined by listing the string lengths from

which its notes are produced. The string lengths are

nevertheless always derived from interval ratios.

1 BM. MS. Add. 769-4-, fol. 225-

2 These letters stand for classes of intervals determined

by size. They will be discussed below.
 (59)

In many instances the definition of a scale by ratios in

durrat al~taj differs from the notation. It is clear

from this that the two are arrived at independently, and in

view of the much greater work involved in multiplying out

(let alone evaluating) the ratios, It is extremely likely that

Qufb al-Din considered them more accurate than the notation.

He certainly realized quite clearly that the intervals of the

Systematist scale were often inadequate and misleading

1
approximations when considered in relation to practice. But

this does not in itself furnish sufficient grounds for

attaching more credence to the ratios in the durrat al-taj.

in certain cases,, notably in connection with the values

assigned to the intervals within the disjunctive whole-tone,

they confirm the amendments to the theoretical gamut proposed

above, but in many others they do not. There are, for example,

very few scales in which the ratios for the third can be

equated with the neutral wusta zalzal: the pentachord

isfahan-i agl, with a third of 343 cents; buzurg, 359 cents;

and nirizi, 354- cents. However, in the vast majority of

scales which do not have a Pythagorean major or minor third,

that given by the ratios is, as in the tetrachord rast

reproduced above, the lust Intonation major third of 386

cents.

1 E.g. in the following passage (BM. MS* Add. 7694, fol.

241) :
 (60)

It might be suggested that this consistency, coupled with

Ou^h al-Din*s evident faith in the accuracy of the ratio

values, constitutes an indication of their correctness, on

this particular point at least. From this it would follow

that contrary to the general argument proposed in the

introduction, gafl al-Din5s remarks on the wusta- zalzal need

not necessarily be considered valid outside the confines of

the Baghdad tradition of which he was a leading exponent,

and further that we are here presented with one example of

a clear difference between Arabian and Persian practice not

masked by uniformity of theoretical approach. It is possible

to adduce one or two broad comparisons from the modern period

In support of such an assumption* Nowadays, for instance,

rast has a neutral third in Arabian music while the Persian

rast o panjgah (panjgah being originally a segment of the

rast scale) has a major third. In fact, despite its prominence

in segah the neutral third is relatively infrequent in present-

1
day Persian music. Further, one may note that according to

Barkeshli Persian music distinguishes the Just Intonation

2
major third from the Pythagorean. Ihis, it might be felt,

is reasonable enough; but unfortunately corroborative

1 Other neutral intervals, notably the second, are however

of frequent occurrence.

2 Loc. cit. Apart from any queries as to the accuracy of

the measuring processes involved, it is unfortunate that no
attempt is made to discern, on a synchronic basis, the
functional significance of the distinction, or to determine
in which contexts one interval is preferred to the other.
Other writers on modern Persian practice (Farhat and,
implicitly, Safvate) have questioned these findings.
 (61)

evidence from texts of the 14th and 15th centuries, of

however general a nature, is nowhere to be found. At most

one might mention that the sharfc, dedicated to a Persian ruler

and possibly written in Persia, also gives the ratio 5 : 4

1
for the third in rast. But the same is true of at least two

later treatises written, like the sharfr, in Arabic, and which

cannot have purported to deal with a specifically Persian

tradition.

Besides being speculative to a degree, this assumption

also leads to unacceptable conclusions. If it is admitted

that there was a difference between Persian and Arabian

practice on this point, and that the l^atios listed in the

durrat al-taj are substantially correct, it follows that in

Persian music a considerable number of modes were, in terms

of intervallic structure, distinguished by no more than a

single comma in each tetrachord. As just one pair among many

one may cite the example of rast, the basic tetrachord of

which is stated to be 204, 182, 112 cents, and 4ushshaq, with

a tetrachord of 204, 204, and 90 cents. It has often been

alleged that oriental musicians, their hearing unaffected by

the artificialities of equal temperament, have preserved the

capability of recognizing and of applying to subtle aesthetic

effect minute shades of intonation which the Western musician

no longer perceives as meaningful. But this pleasant fancy

1 BM. MS. Or. 2561, fol. 150v.

 (62)

has never been proved* It is true that two intervals, one

20 cents larger than the other, can be clearly told apart

under laboratory conditions, or when attention is particularly

directed to them. Nevertheless, in the hurly-burly of actual

performance such a distinction would be as difficult for the

1
musician to maintain as for the audience to recognize.

Therefore, just as in North Indian classical music today,

where a rag seems to acquire its individual character by

means of phrasing and the prominence accorded certain notes

rather than, as has been maintained, through strict adherence

2
to minute distinctions of pitch, we may expect that it

needed more than a one comma difference in one or two inter­

vals for these modes to be distinguished one from the other.

It might be suggested that there were in addition other

criteria of identification: that the modes were also defined

in terms of melodic pattern, emphasis on certain notes, and

limitations of range. But if these features were sufficient

in themselves to distinguish one mode from another the

intonational difference of a comma would be otiose; and if not

1 This is perhaps less true for the fourth and fifth. A

slight, deliberate alteration of these would be readily
perceived, especially if the tonic were frequently sounded.
The intonation of thirds, with which we are chiefly concerned
here, may however fluctuate considerably and pass unremarked.

2 A view put forward in A. Dani^lou, North Indian music,

London, 194*9- That intonation is not the primary factor of
identification is maintained in N. A. Jairazbhoy and A. W.
Stone, "Intonation in present-day North Indian classical
music", Bulletin of the School of Oriental and African
Studi e s ,~xxvn (1), 1963.
 (63)

they may, in the present context, be disregarded. In any

case, Qu£b al-Din makes it quite clear that it is only the

secondary or "branch" modes (shu* ab) which can be defined in

this way, as they constitute particular types of melodic

movement based on the notes of the primary modes (or parent

scales, as one is inclined to call them), to which category

both rast and cushshaq belong.

The argument that a comma difference would have passed

unnoticed, and hence could not have provided a criterion of

modal identification is substantiated by earlier theorists

who were as familiar with Persian as with Arabian musical

practice. Al-Farabi states that a minor whole-tone (182

cents) is indistinguishable from a major whole-tone (204

1
cents). Ibn Slna goes so far as to say that musicians,

although some of them recognize the difference, often confuse

limma ( 9 0 cents) and quartertone (c. 5 0 cents) and substitute

2
one for the other during a performance. Since it is unlikely

that in the course of the two centuries between Ibn Sina and

Qu^b al-Din musicians and their audiences should have refined

their sense of relative pitch considerably, this testimony

may be thought sufficient to show that the ratios cannot be

accepted here as an accurate guide to the intonation actually

1 kitab al-musiqa al-kabir, 166. Cf. also 580-583 and

627- 628.
2 D ’Erlanger, la musique arabe, ii, 1 5 0 .
 (64)

used, and that a meaningful contrast was maintained in the

Persian as well as in the Arabian music of this time not

between the Pythagorean and the 5 : 4 major thirds but between

a major third, which could vary within certain limits, and a

neutral third.

Indeed, a rejection of certain ratio values is implicit

in a later part of Qu-fcb al-Din9s section on music. He remarks

that if the range of dugah, a mode relating to part of the

rast scale, is extended to g, giving (in the Systematist

notation) (c) d e“c f g, the ratio values for which are

(204), 182, 112, and 204 cents, the effect produced will be

that of nawruz, d e”c f g, for which the ratios are 112,

1
182, and 204 cents. Now, it is obvious that in this case -

and there are others of a similar nature - the discrepancy

in the ratio values for the two sequences of intervals is

considerable. Not even the musicians quoted by Ibn Sina as

confusing quartertone and limma would be likely to fall into

the same trap here. If, however, the wusj;a zalzal is taken

to be a neutral third with the muj annab a whole-tone below,

the intervals of d e1? f g in both rast (or dug ah) and

nawruz will be of the order of 146, 148, and 204 cents.

One possible objection to this interpretation might be

1 BM. MS. Add. 7694, ff. 233-233v. Only the names of the
modes are given here. The notation and ratio values are
taken from the list of species and modes transcribed in
chapter 2.
 (65)

raised, Theorists of the Systematist school classify

intervals up to and including a whole-tone into three

categories: J1 (whole-tone, 204 cents); J (minor whole-tone,

180 cents, or apotome, 114 cents); and B (limma, 90 cents,

or comma, 24 cents). It might therefore be thought that

Qu£b al-Din is here saying no more than that the sequence

J J T (d e“G f g) in dugah could be considered equivalent

to the J J f
j? sequence in nawruz. Unfortunately, these are

two of the best-known modes, and the substitution of minor

whole-tone (180 cents) for apotome (114 cents) or vice versa

would surely have been noticed. Furthermore, Qu£b al-Din*s

1
wording makes it clear that he is speaking about practice,

and not abstract sequences and categories that can happily

remain unheard, so that this objection may be dismissed as

2
both unlikely and far-fetched. Any lingering doubts about

this Judgement should be dispelled by the evidence of an

3
analogous case in the kitab al-fatb-i.yya of al-Ladhiqi.

1 Ibid. The passage is translated in chapter 7*

2 Especially when it is borne in mind that the J and B

categories are extremely strange, and could be cited in
support of the criticism voiced above to the effect that
the theoretical analysis they relate to is inadequate and
arbitrary when considered in relation to practice. Given the
existence of intervals of 204, 180, 114, 90 and 24 cents
the groupings whole-tone (major and minor); J: apotome
and limma; and B: comma, would be much more natural. Even
simpler, of course, are the correspondences J: whole-tone;
J: three-quartertone; and B: semitone or (rarely) quarter
tone. The few contexts in which a quartertone is found
exclude the possible occiirrence of a semitone; hence there
is no ambiguity.

3 BM. MB, Or. 6629, fol. 6 6 v.

 (66)

Again a 5 : 4- ratio is given for the third, on either side of

which is placed a J 3? sequence:

r “ ii---------- 1
G A-c B-c c cL

J ? J T

In theory therefore one consists of 182 and 204 cents, the

other of 112 and 204, the difference being the same as that

assigned above by Qu-fb al-Din to dugah and nawruz. Here,

however, al-Ladhiql is not drawing parallels between modes,

but is discussing one mode, or modal nucleus, giving it two

1
distinct pitch settings* This only makes sense if we have

a neutral third, giving

I--------------- II---------------1
G A 1? Bt c d

where both segments consist of a three-quartertone and a

whole-tone. It thus appears certain that he and Qufb al-Din

are not simply juggling with modes having markedly different

intervals which happen to belong to the same category, and

that in consequence their analysis of the neutral third as a

5 : 4 interval is incorrect.

Since then the ratios corroborate the arguments brought

against the theoretical disposition of intervals within the

disjunctive whole-tone, but fail to provide any correction to

1 In relation to an open string note or the lowest note

in notation (alif), and not in terms of absolute pitch.
 (67)

the notation where the neutral third and related intervals

are concerned, it may be asked in what respects and to what

extent they can be held to constitute an accurate representa­

tion of the species and modes.

One basic feature of the analysis of intervals by this

means is the predominance accorded to simple numerical

relationships. Since the ratios for the major consonant

intervals form the series 2 : 1, 3 : 2, 4 : 3* •*•, there

emerged the belief that only further members of the series

2
could be thought consonant. But because for the Systematist

school consonance was a property not only of the interval as

such but also of groups of intervals, i.e. of tetrachords and

3
octave scales, there followed naturally a complementary

tendency, namely to fit intervals found in scales considered

consonant into this series. Thus a hypothetical interval of

2 5 0 cents, which would be heard to fall between whole-tone

(9:8) and minor third (6 : 5 )9 would automatically be

classified as either 7 • 6 , 267 cents, or 8 : 7 ? 231 cents.

1 This is of considerable antiquity in the Near East.

However for Arab and Persian scholars the immediate source
was the Hellenic legacy.

2 The limma, for instance, which has no simple ratio (being

256 : 2 4 3 )7 is considered dissonant by theorists of the
Systematist school, while the semitone, which has (being
16 : 1 5 )7 is considered consonant.

3 In the kitab al-adwar chapter 4 considers dissonance in

relation to the tetrachord, while chapter 6 discusses the
consonance of octave scales in terms of the number of
fourths and fifths they contain.
 (68)

Even allowing for a reasonably degree of latitude in the

intonation acceptable for the major and minor thirds it may

be assumed that within the tetrachord these and the whole-

tone could be established as norms in relation to which the

calculation of ratios for the other intervals would operate.

A further factor which may have encouraged the tendency

to analyse intervals as simple ratios was the mathematical

difficulty involved in calculating square roots. Theorists

knew well enough that in order to find e.g. half an interval

1
it was necessary to determine the geometric mean. But as

this depended on the square root with its attendant

complications, the arithmetic mean was adopted instead. Thus

7 : 6 (267 cents) and 8 : 7 (231 cents) were considered halves

of 4 : 3 (4-98 cents), instead of 2 : J~5 (249 cents). When

halving intervals smaller than the fourth the error resulting

from this process is minimal, being 11 cents for the 5 : 4-

major third and diminishing progressively with the interval

to be halved.

It now becomes easier to see why the neutral third is

described as a 5 : 4- major third. Firstly, as it was a

component of a number of much used and "consonant" modes,

there was a natural impulse to assign to it a simple ratio.

Secondly, it is in fact half (i.e. geometric mean) of the

1 See for example Ibn SIna’s remarks (D’Erlanger, la

musique arabe, ii, 1 3 6 ).
 (69)

fifth, so that there is nothing startling, within the context

of interval division as practised "by the Systematist school,

is seeing it classed as one of the two members (5 : 4 and

6:5) derived from a division of the fifth by the arithmetic

mean. In addition, and this is perhaps the most important

factor, the 5 • 4 ratio was available. In view of the

importance attached to the diatonic tetrachord the major

third, whatever its shade of intonation in practice, was

always analysed as 81 : 64. This, although not a simple

ratio of the form x + 1 : x (where x is an integer), could

nevertheless be considered quasi-consonant as it was twice

the consonant whole-tone ( 9 : 8 ) . It was perhaps only to be

expected that the 5 : 4 ratio, known to be slightly less than

81 : 64, should be assigned to an interval which figured

prominently In a number of "consonant" scales and was

somewhat flatter than the major third.

Any attempt to test these values experimentally would

probably have failed to give conclusive results. This could

only have been done on one or more monochords, and it is

doubtful whether the conditions necessary for reasonably

accurate use of the instrument - rigid base, precise

measurement of the bridge positions, unvarying tension and

constant mass of string - were ever fulfilled. In any case,

although the monochord may have been employed as an adjunct

to the theoretical demonstration of intervals, there is no

evidence to show that it was ever used as a control.

 (70)

The ratios, then,are worked out with reference to certain

presuppositions about consonance, and also relate to the

interval being considered in the context, generally, of the

tetrachord* The slightly differing values found for a

particular interval must be examined therefore as bound forms

determined by the disposition of the intervals within the

tetrachord species of which they are an integral part. For

instance, the major third will be placed at 408 cents in a

Pythagorean diatonic tetrachord and at 417 cents in a

tetrachord the central interval of which is analysed as

7 : 6 (2G7 cents)* However, despite the caution with which

they must be approached, the ratios constitute a valuable

check on the notation of the species and modes, and in certain

respects are closer to the amended version of the Systematist

scale proposed above*
 (71)

chapter 2

In the kitab al-adwar gafl al-DIn defines 7 tetrachord

species (aqsam b u ‘d dhi al-arba*) and 12 pentachord species

(aqsam b u ‘d dhi al-khams), 11 of which may he related to the

tetrachords. From them, the tetrachord invariably being

placed below, 84 octave scales (adwar) are derived by

1
permutation. This might be thought a rather unwieldly and

artificial attempt to define the total of possible octave

scales within a system permitting novel junctures of

established units. But it is far more likely that the 84

octave scales as such have no relevance whatever to practice

and that the series of combinations giving rise to them are

presented primarily for their own sake. gafi al-Din records

7 octave scales not among the 84; and according to the

criteria laid down in the kitab al-adwar 24 of the 84 should

2
be classed as dissonant (mutanafir, or gahir al-tanafur).

1 The species are defined in chapter 5 (BM. MS. Or. 136?

ff* 9-*10v), the octave scales are listed in chapter 6
(ibid., ff. 14-17)*

2 Ibid., ff. 8-8v, 11-12, and 13v, In which the categorie

of consonance are discussed. gafl al-Din does not himself
apply these systematically to all the 84 scales.
 (72)

Nevertheless, it should he pointed out that gafl al-Din goes

on to say that occasionally compositions were based on some

of these (he does not state which), although care had to be

taken to avoid the more obviously dissonant combinations of

1
intervals. In consequence - however little the importance

which should be attached to the 84 as a set - one might

suppose that all the 60 non-dissonant scales were in use, for

they presented no such difficulties. fhe actual number of

octave scales in use, however, was certainly much smaller,

and probably not more than 30. gafl al-Din mentions 17?
2
adding that they are the most frequently performed modes,

while Qutb al-Din, whose account of the modal system is in

all respects far more comprehensive, lists 2 7 , and two of

these are stated to be little known. In view of this it may

be felt that little purpose would be served in transcribing

all 84 scales, and the list given below confines itself to

material relevant to practice.

3
I'he bulk of it is taken from the durrat al-tat
j , and Qufb

al-Dln5s order of presentation, which proceeds logically from

species of third through to a few examples of scales exceeding

the octave, has been retained. Scales given in gafi al-Din5s

1 Ibid., fol. 20.

2 16 are mentioned in the kitab al-adwar (ibid., ff. 1 9 v -

22v) and a further one in the ris~ala al-sharafiyya (D5Erlan-
ger, la musique arabe, iii, 1367".

3 BM. MS. Add. 7694, ff. 223-231*

 (73)

treatises and not mentioned in the durrat al-taj have been

included, their position in the list being determined by

identity of name as well as by similarity of scale-structure.

All species and scales are from the durrat al-taj unless

otherwise stated; and whenever a species or scale is also one

of those cited by gafl al-DIn this is indicated. Further

material on the modal system from the durrat al-taj will be

presented and discussed in chapter 7 -

In accordance with the findings of chapter 1 the following

principles govern the transcription:
G- is throughout taken as the equivalent of the alif

of the Arabic notation;

g~c is transcribed as fit, gb as ft (of. pp. 53-54-);

d~c is transcribed as d1> or as cS according to
context (cf. p. 56);
A~c , B~c , and e~*c are transcribed as A*!?, B1?, and et

respectively.

Any deviations will be discussed as they occur.

The ratios (found, unless otherwise indicated, only in

the durrat al~taj) have been expressed in cents and written

out below the transcription of the notation. In some

instances a discrepancy between the notation and the ratio

values has not been commented on, notably where the disjunctive
 (74)

1
whole-tone is concerned* Reasons fox* amending certain of

the latter have been presented in chapter 1 , and the

commentary therefore only deals with cases requiring further

explanation. But wherever the ratios have been thought

inaccurate a second line has been added below in order to give

what is hoped to be a more realistic appraisal of the nature

of the intervals concerned. In this respect we may repeat the

contention that an acceptable intonation of a given interval

may fluctuate within certain limits. The position of these

limits will also vary, being influenced by several factors, in

eluding the nature of the performance; whether the performer

is a singer or an instrumentalist (in which case the type of

instrument will be of significance); the size and position of

the interval; the melodic context; tempo; and dynamics. Hence

simply to "correct" a ratio value of 3 8 6 cents to 3 5 0 might

give a misleading impression of an alternative absolute, an

immutable standard intonation which could be incorporated into

a second theoretical gamut just as rigid as that of al-

Din, even if more practice-based. In view of this the second

line will be presented in diagrammatic form in order to show

a zone of acceptable intonation, rather than a single point.

The line will be marked off in 100 cent divisions, and the

position of the relevant Pythagorean and Just Intonation

intervals will be shown. The convention will be adopted of

1 One may find for instance 112, 90, and 85 cents for g~c
(i.e. fif) - g; 11 2 and 7 6 for d~c (i.e. cJf) - d.
 (75)

not indicating fluctuations in intonation for the tonic; the

fourth in tetrachords; the fifth in pentachords, and so on.

Nor will they "be indicated for non-neutral prominent notes.

Prominence is a feature discussed by Qu£b al-Din in the

following passage: "In practice musicians do not pursue the

same aim in all modes: in some their aim is to give prominence

1
to the interval limiting the mode; in others to give

prominence to a lesser interval; and in others to give

prominence to a single note. This too may vary, being in

some cases the tonic and in others the fourth from the tonic.

In our list we have marked in the interval or note concerned

2
above each species". It may be assumed that according

prominence to a "lesser interval" refers to the relative

importance of the segment of the scale that interval comprises.

We shall speak throughout of prominent note(s), whether it is

a case of a single note or of an interval. Such notes will

be given in the transcription as semibreves. Those marking

a "lesser interval" will be shown thus:

1 I.e. tonic and fourth in a tetrachord, tonic and octave

in an octave scale, etc.

2 BM. MS. Add. 7694, fol. 240v:

I; b j^

4A p 0^3 J.VL1 JJ 3 aS j b j 1 \ (j; J

J j.%— ,0 j^ 1 3 j £ JJ 3 3J y J i

* 1 d J ^3 j l-5 Ur ~ J.. ^ JL ^ ^ ^ ^ L h «5
 (76)

9 ---r---------
o
k\ A *

while alternatives will he indicated by minims, thus:

" , -----1-----'-----*-----‘ 1
-Jh - o ’

Section headings are from the durrat al~tac

j , as are the

headings for each item, unless otherwise stated.

The list has been numbered for ease of reference.

* *
 (77)

species of minor third ( 6 : 5)

1 zirafkand-i kuchek. May he termed simply zirafkand, or

1
kuchek. xt is also called mukhalifak.

I/ u
139 128 49

]___________ p ™ 1 ___I________jJ l p]JIBIBI~ i

267 316

Safi al-Din mentions this species in the risala al-

sharafiyya only, calling it zirafkand. The notation is the

same, hut there is a minor difference in the ratios which

reverse the first two intervals (giving 128, 139? and 4*9
2
cents). There is an ohvious discrepancy between the ratios,

which indicate a compass of a minor third, and the notation

/ c
ending on the major (B , interpreted as signifying a neutral

third). The latter value is undoubtedly correct. It is

confirmed by Qufb al-Din himself when he refers to mukhalifak

in a passage discussing the inadequacy of the (fixed

Systematist) fretting for reproducing certain modes, and

states that its highest note should be stopped half-way

-c , 5
between the B and Bb frets, i.e. should be a neutral third.

1 BM. MS. Add. 7694- has mukhalif, India Office MSS. 2219
and 2 2 2 0 mukhalif ale.

2 Bodleian MS. Marsh 521, ff. 65-65v.

3 BM. MS. Add. 7694, fol. 222.
 (78)

In accordance with the conclusions reached above one

should therefore expect the ratios to add up to a Just

Intonation major third rather than a minor third, The

presence of the latter is to be explained by the fact that

the ratios for zirafkand in the risala al~sharafiyya are

taken from a pentachord species of which they constitute the

lower section. It would appear that when defining a neutral

third by dividing the fifth (3 : 2 ) it was possible in

certain cases to place 6 : 5 "below 5 : 4-* Although Qutb al-

Din does not refer to this pentachord it is likely that his

values are taken from the risala al-sharafiyya, despite the

inversion of the first two intervals. Had he arrived at them

independently we may assume that he would in fact have given

5 : 4- for the neutral third here as well as in the following

1
species.

1 The compass of zirafkand is in fact stated to be 5 : 4-

by aJ-Ladhiql (BM. MS. Or. 6629* fol. 67)* Ho reference
to a pentachord species is made.
 (79)

1
species of major third (5 : 4)

2 *iraq

f.

182 204

T-— -T----- T ^ V
586
For gaf l al-Din the name *iraq denotes a tetrachord

species (no. 1 0 , clearly related to the above) and an octave

2
scale. The existence of 2 tiraq is confirmed by al-Ladhiqi.

zawli

204- 182

1 T -----r— p
204- 586

The final note is stated elsewhere to be A or G. For

4
later writers zawli is characterized by a pronounced vibrato

on the Bt amounting to an oscillation between Bl? and B.

1 In 2 and 3 both the notation and theratios give a 5 : ^

major third. In 4, however, the notation has a Pythagorean
major third (81 : 64).

2 BM. MS. Or. 6629, fol. 6 6 v.

3 BM. MS. Add. 7694, fol. 233v.

4 ‘A M al-Qadir (Bodleian MS. Marsh 282, fol. 51v) and,

following him, Jami and al-Ladhiqi.
 (80)

_1
4 rahawi

(' --
wA - P# n*

139 128 119

i i mm i |jb^ b *fr
26? 386

Both the ratios and the notation are identical with

those given in the risala al-sharafiyya, where we are told

that the intervals are more or less interchangeable, i.e. of

2
equal size. According to Qutb al-Din, however, this
3
particular arrangement is the most consonant, which suggests

that the notation, with a semitone between Bb and B, may be

more accurate. Further reasons for rejecting the ratio

values may be found in the comments on 11 igfahan.

1 An alternative form, rahawi, is also frequently found.

2 Bodleian MS, Marsh 521, ff. 61-63V, 90v. gafi al-Din

states that rahawi was previously called mazmum.

3 BM. MS. Add. 7694, fol. 21.

 (81)

A
tetrachord species

5 cushshaq

4> o * — —

204 204 90

6 busallk

90 204 204

7 nawa

a
• ""I*.
r 9
ZJi-- *---

204 90 204

Neither here nor in the octave nawa does Qutb al-Din

mention which note, if any, was prominent.

1 Following £>afi al-Din5s definition, the term tetrachord

will be used in the present work to refer to any species
having the compass of a fourth, irrespective of the number
of intervals involved. Similarly with the term pentachord.
 (82)

8 rast

-------------- 5-------- v-*------

^ Q --

204- 182 112

386 4-98

The notation of the above four tetrachords is given in

1
the kitab al-adwar. In the risala al-sharafiyya the same
2
names are also added.

9 nawruz

* ..... 0
9 #
- 4 )— 0----

112 182 204-

I------r — n ---- T 1------ 1

294- 4-98

From the notation, which is the same as that listed by

3
£>afi a l-Dm, one would expect rather 182, 112 , 204-. However,

where this does occur (in 60 mufrayyir husayni) it is clearly

the result of the assimilative influence exerted by the rast

scale in which these intervals appear from A to d, and d to g.

In other scales the 112, 182, 204- order is adhered to.

1 BM. MS. Or. 136, fol. 9v.

2 Bodleian MS. Marsh 521, fol. 90v.

3 BM. MS. Or. 136, fol. 9v.

 (83)

10 ♦iraq (Safi al-Din)

ru-yi 1iraq (Qupb al-Din)

I *p* <-.. 0
-4- — © —

182 204 112

386 498

11 isfahan

139 128 119 112

h i -pT—
| --- J—
267 386 498

The same notation and ratios are given for igfahan by

1
gafi al - D m . The ratios may be taken to represent intervals

of equal size, and are arrived by a twofold division of the

fourth by the arithmetic mean (4 : 3 giving 8 : 7 and 7 : 65

8 : 7 giving 16 : 15 and 15 : 14; and 7 : 6 giving 14 : 13

and 13 : 12). 'With the exception of 4 rahawi there is no

species at all similar, for a tetrachord of four intervals of

equal size nowhere contains a major third, a minor third, or

a whole-tone. These intervals were certainly recognized as

1 Ibid* (notation), Bodleian MS. Marsh 521, ff- 61-63v

(ratios).
 (84)

Important in practice as well as in theory. Hence it is

unlikely that 11 igfahan and the related 4 rahawi would have

1
been acceptable in the form suggested by the ratios- There

can be no doubt that the notation is more accurate, and that

11 igfahan corresponded to 9 nawruz with a major third Included.

This is confirmed by a later statement in the durrat al-taj

to the effect that the intercalation of an fl in the segment

d e1? f g of the rast scale (in theory d e“c f g, 182,

2
112, 204 cents) produces the effect of (11) Isfahan.

15 0 267 81

The intonation of the major third (given by Qu£b al-DIn

as 417 cents) was in all probability not consistently sharper

than that of major thirds in other species, which we have

indicated as lying within the zone from c- 375 to c. 420 cents.

1 Although it is conceivable that they may have originated

in approximately this form, being borrowings from folk-music.
According to P. V. Olsen ("Enregistrements faits A Kuwait
et a Bahrain", Les Collogues de W&gimont, iv, 1958-1960
(Li£ge, 1964)) a tetrachord with four virtually equal
intervals is used at the present day among the Sulaib, a
tribe (of apparently non-Arab origin) in Kuwait.
The igfahan listed by Ibn Sina is quite different.

2 BM. MS. Add.7694, fol. 233-

3 Qutb a l - D m generally writes frij az , gafi al-Din frijazi.
The latter form will be used throughout the present work.
 (85)

This minor adjustment apart, 12 frinazi. is one of the rare

cases in which the ratios for a theoretical non-diatonic

1
genus would seem to correspond exactly to intervals used in

practice. Qutb al-Din also mentions variant forms of this

tetrachord including one in which the central interval is

even larger (a full augmented second), but concludes by

saying that the above division is the one most commonly

2
found. The fact that these variants are cited at all may

be considered a recognition of fluctuations in intonation.

gafi al-Dxn does not list this tetrachord, and in none of

the octave scales listed by him is it notated. However the

name bijazi is used at one point in the kitab al-adwar to

designate one of the constituent parts (tetrachord and

5
pentachord) of an octave scale. The significance of this,

and the relationship of 12 fcijazi to the octave scales of

the same name listed by gafi al-Din, will be discussed in

chapter 5 *

1 Qufb a l - D m (BM. MS. Add. 7694* fol. 222) describes it

as i
j^ ^3 “* g j\ \,:»

i.e. 12 : 11, 7 : 6 , 22 : 21.

2 Ibid.

5 BM. MS. Or. 136, fol. 22v.

 (86)

pentachord species

13 *ushshaq

b 0

204 204 90 204

14 busallk

90 204 204 204

Both 13 and 14 are among the pentachord species listed by

1
gafi a l - D m . He does not, however, name them. They

correspond to the tetrachord species 5 *ushshaq and 6 busallk

respectively, adding a whole-tone above, and constitute the

upper pentachords of the octave scales 40 1ushshaq and 41

busallk.

15 (not mentioned by Qu£b al-Din)

2
This species is found only in the kitab al-adwar. It

may be formed by adding a whole-tone above 7 nawa, with which

it combines to form the octave scale 42 nawa.

1 Ibid., fol. 10

2 Ibid.
 (87)

16 rast

**
» ->* .
—
__________

204 182 112 204

I
,
— t i— n— T ----------1
------ 1
-

386 498 702

Similarly, 16 rast relates to 8 rast and the octave scale

1
^5 rast. It is also listed, unnamed, in the kitab al-adwar.

17 igfahan-i agl. Some people also call it mukhalif-i rast

*
0-

204 159 128 119 112

543 471 498 702

This species may be derived from 11 igfahan, but with

the whole-tone added below. It is also given, unnamed, by

2
gafi al-Din, and constitutes the upper pentachord of 6 6

Isfahan, which is however an octave scale not listed by Qutb

3
al-Dm, despite being referred to elsewhere by him.

The other name, mukhalif-i rast, may be thought to

indicate the similarity and difference between this species

1 Ibid.

2 Ibid.

3 BM. MS. Add. 7694, fol. 232v.

 <s88j>

and 16 rast, and hence incidentally the falsity of the ratio

values for the intervals from A to d.

18 husayni

i- U.: ... -
ft.. - .. h i
V " '

112 182 204 204

|-------------------- (
T ----- 1------ J

-
294 498 702

Here the previous pattern is followed: 18 fcusayni may he

arrived at by adding a whole-tone above 9 nawruz. It is also

found in the octave scale 48 frusayni, and is listed by gafi

1
al-Din.

19 zirkesh fausayni

i 139 128 119 112 204

t-
t - Tr— r
267 386 498 702

Similarly this species relates to 11 Isfahan and the

2
octave scale 67 Isfahan. It is again listed by gafi al-Dm,

although 67 igfahan is not.

1 BM. MS. Or. 156, fol. 10.

2 Ibid.
 (89)

The name suggests however a different origin. One might

contrast zirkesh (indicating the inclusion of the open zir

string note, c) with zirafkand (omitting the zir) and posit

the development 1 zirafkand G A1? Bb B1?--- > G At Bb Bt

c ^ ---- > q At Bb B c d, with Bt becoming B by analogy

with 11 igfahan. But even if this is rejected as implausible,

the inclusion of feusayni in the name cannot be accepted as an

indication that this species may have been derived from 18

fcusaynl, since in that case the term zirkesh must refer to

the inclusion of B and hence imply a transposition from Ab

A:l: B (c) db eb, which is far more implausible. The

reference to frusayni is pn all probability analogous to the

reference to rast in 17 mukhalif-i rast,

20 buzurg (gafi al-Din)

buzurg-i agl (Qufcb al-Din)

-fr [?). * = g

128 231 139 128 76

h — r T — T— "H— |
359 498 626 702

21 variant of buzu:

_S't*L
t

150 267 81 128 76

These values fall within the zones of intonation suggested

 (90)

for 20.

21 is peculiar to Qu£b al-Din. 20 buzurg is analyzed by

gafi al-Din also in terms of ratios, and he offers the

1
following four combinations:

14 : 1 5 , 8 : 7, 13 : 1 2 , 13 : 1 2 , 27 : 27 (128, 231, 139,

1 3 9 , and 6 5 cents)

13 : 1 2 , 8 : 7, 14 : 13, 13 : 1 2 , 27 : 26 (139, 231, 128,

1 3 9 , and 6 5 cents)

14 : 13, 8 :7, 13 :1 2 , 14 : 13, 117 : 112 (the definition

given above from the durrat al-tat

j* This division is also
2
found in the shark *)

13 : 1 2 , 8 :7, 14 :13, 14 : 13, 117 : 112 (139, 231,

128, 114, and 90 cents).

What is significant - the differences are not - is that in

each of the four we find the ratio 8 : 7 corresponding to the

notation A”° - B~c (given here as A1? - B(1?)), whereas whole-

tones elsewhere are invariably accorded the ratio 9 : 8 . It

would appear from this that the interval thus represented was

in fact larger than a whole-tone and, as in the variant 2 1 ,

may well have been of the same order as the central interval

in 12 tiijazl • There would be nothing unexpected in gafi al-

Din choosing 8 : 7 rather than the 7 : 0 preferred by Qu^b al-

Dln for such an interval since the latter was held by him to

1 Bodleian MS. Marsh 521, ff. 6 6 -6 6 v. The notation is

given in the kitab al-adwar (BM. MS. Or. 136, fol. 10).

2 BM. MS. Or. 2361, fol. 95v.

 (91)

1
be unacceptable as an undivided unit within a scale. In the

notation of B(7) given above, the 1? corresponding to the

. . —c
original is placed in brackets to indicate that it is

suspect, but retained in order to differentiate the notated

forms of 2 0 buzurg and 2 1 .

It may be asked that if 20 and 21 both represent the same

thing, why include the latter? The answer to this is probably

that Qu£b al-Din wished to reproduce the notation and ratios

established by gafi al-Din but, realizing that these were

inconsistent, with the notation giving a distorted picture,

added a further definition which was for him in closer

conformity with practice. The opposite tendency is observable

in the shark? which in addition to adhering to the notation

“G ™“C
A - B and the ratio 8 : 7 later indulges in a hyper-
2
correction and provides the ratio 9 : 8 for this interval.

3
22a (gafi al-Din)

1 Bodleian MS. Marsh 521 fol. 68 . This in spite of the

fact that 2 5 a maya contains an even larger interval.

2 BM. MS. Or. 2561, fol. 153* gafi al-Din nowhere mentions
9 : 8 in connection with buzurg.

5 BM. MS. Or. 156, fol. 10.

 (92)

1
22b 4uzzal (Qufb al-Din)

150 267 81 204

The former consists of 10 *iraq with a whole-tone above,

the latter of 12 b-ilazi with a whole-tone above. There are

a number of such pairs in which a whole-tone notated by gafi

al-Din (in this case Afr - Bt) corx^esponds to a larger than

whole-tone interval in Qu£b al-Din’s equivalent species. The

relationship between them will be discussed in chapter 5 *

_ —c
In the shark 4uzzal is an octave scale notated as G A
—c —c 1 *
B c d e gb g, and it is possible that here, as with

20 buzurg, the notation A~c - B~c is a disguise for an interval

larger than a whole-tone (At? - B). When dealing with scales

not described by gafi al-Din the shark must however be

approached with caution, and in later treatises 4uzzal is

2 3
stated to be a pentachord. Al-Ladhiqi}s version is identical

with that in the durrat al-tat

j.

23a maya (gafi a l - D m )

4
• — — - — ■

1 Sometimes written with ghayn. 4uzzal is however the

only form found in vocalized texts.

2 E.g. in the treatises of 4Abd al-Qadir and Ibn JamI.

3 BM. MS. Or. 6629, fol. 69v.

 (93)

23b maya (Qu£b al-Dxn)

316 182 204

gafx al-Dxn states that maya is characterized by a

1
particular ascending and descending melodic movement. Qu£b
2
a l - D m quotes this description, but adds that the best-known

compositions in maya, including those of gafx al-Dxn, confine

themselves to the notes of 23b. This may be considered the

upper pentachord of a hypothetical extension of 2 3 a to the

octave.

maya is the only example of a gapped or (in gafx al-Dxn5s

version) pentatonic scale in the modal system of the period.

It Is therefore possible that It was an impor*tation either

from folk-music or from outside the specifically Arab and

Persian cultural areas. A. A. Saygun suggests that its origin

3
was "Turco-asiatic” : if so it had certainly been fully

integrated into the modal system of art-music by the mid- 1 3 th

century, for gafx al-Dxn includes it in one of the two most

important groups of modes, the awazat *

1 BM. MS. Or. 136, fol. 22v.

2 BM. MS. Add. 7694, fol. 233v.

3 "la musique turque" in Bncyclopbdie de la Pl^iade:

Histoire de la musique, i l^Parxs7 l9 6 0 j7 " 57 4*•
 1
24 a shahnaz (gafi al-Din)

y • y0 v*
— •— — # -----------------

24b shahnaz (Qu£b a l - D m )

.... h
H aU ...
.... * " 0
1>9 K*
w w

139 128 49 139 128 49

\- i i— r T r J,BLi— Y — 1" M
267 316 4-98 702

gafi al-Din states that the type of melodic contour

2
found in 23a maya also serves to characterize 24a shahnaz

as, indeed, the notation suggests* It Is a specific melodic

form of 1 zirafkand*

Qu£b al-Din’s version reduplicates 1 zirafkand. Thus

despite the prominence of tonic and fifth it is possible

that the neutral third functioned as a melodic fulcrum,

rendering the consonance of the fourth Imperceptible as such.

As a pentachord species 24b shahnaz is also unique in the

number of Intervals it contains. For these reasons it may

be suggested, in explanation of Qu£b al-Din5s later comment

3
on the paucity of compositions in shahnaz, that the type of

1 BM* MS. Or. 136, fol* 21.

2 Ibid*, fol. 22v.

3 BM* MS. Add. 7694, fol. 234.

 (95)

melodic movement specified by gafi al-Din was too restricted

to serve as a basis for extended composition, and that

subsequent developments had not yet produced a stable and

acceptable scale-structure. In support of this view one may

cite the conflicting accounts of this mode to be found in

1
later treatises.

The ratios of 24b shahnaz present an extreme example of

the distortion that arbitrary theoretical analysis can bring

about. The initial error of classifying 1 zirafkand as a

species of minor third is made worse by repetition, so that

while 24b is rightly listed as a pentachord, the sum of the

ratios is an augmented foiirth, and the natural fourth of the

notation is placed at 455 cents.

Because of the relationship between 24b shahnaz and 1

_ __(2
zirafkand the d of the notation has been transcribed as

dt> and not c$.

2
25a (gafi a l - D m )

1 Some reproduce gafi al-Din5s version. The t

jami* al-alfran
of ‘Abd al-Qadir gives both 24a and 24b. The kanz al-tufaiaf
has 24b omitting d1? (a variant omits d instead")^ and the
kitab al-fatpiyya of al-Ladhiqi 24b omitting c * These
changes may be viewed as attempts to normalize shahnaz as a
pentachord (ignoring the variant in the kanz al-tuhafj by
reducing the number of intervals to 5 * the maximum elsewhere.

2 BM. MS. Or. 156, ff. 10-10v.

 (96)

25b bisap (Qu^b a l - D m )

— -------------------- — --------------- - ---------

1 H? in
V" * > ____
?*
---

159 128 49 267 119

I 1 "" i
— |T| i 1 i
267 316 583 702

These may both be described as extensions of 1 zirafkand.

The latter constitutes the upper pentachord of 58b \iigar, the

former of 58a zirafkand. The relationship between the two

is underlined by gafi al-Din, for in the risala al-sharafiyya

he defines by ratios a species with the intervals 128, 1 5 9 ?

49? 2 5 1 ? and 1 5 5 cents, and then adds that it is preferable

to have in place of the 251 and 1 5 5 cent intervals a major

whole-tone (204 cents) followed by a minor whole-tone (182

1
cents). However, the mere fact that the former is first

defined as a larger interval would seem to indicate that

Qufb al-Din5s version is the more accurate.

Concerning the phrasing of fri.gar Qu^b al-Din elsewhere

states that after descending to the lowest note (the final

note in most modes), it is permissible to return to the upper

note and to make slight use of the notes flanking it, thus

1 Bodleian MS. Marsh 521, ff. 65-65v.

 (97)

1
affirming the particular character of this mode- This

implies that the range was more that a fifth, and the added

et has been taken from the description of Uigar in the kitab

2
al-fatfriyya.

26 kardaniya

139 65 204 182 112

H " T — i If-— im m |i {
408 590 702

This species, as gafi al-Din points out, relates to the

3
tetrachord 8 rast, adding a whole-tone below. But in no

other pentachord is this whole-tone divided, and it may be

assumed that 26 kardaniya has been simply abstracted from the

octave scale 50 kardaniya (of which it forms the upper

pentachord) for theoretical purposes. There is nothing to

suggest that it was ever found except as part of this octave

4
scale. Hence the notation of A as Gt, and not Ab - The

1 BM. MS. Add. 7694, fol. 240:

U? I Jai Cwv 2 jO jl(£C>- Jb IJJaj I C5 g?

I I I I O' JJ CJlpXSA Uu ^ 2 d— *JLrO t

. ijJ ^ ^ O jjJUca—<» ^ *?J J
2 BM. MS. Or. 6629* fol. 6 8 v. Al-LadhiqI’s account (G Ab
Bb B1?c# d e1? d c# d) corresponds exactly to Qu^b
al-Din’s description.

3 BM. MS. Or. 136, f f . 10-11.

4 Of.also the notes to the end of chapter 5 in the text of

the kitab al-adwar.
 (98)

ratio values for the intervals within the lower whole-tone

should be contrasted with the corresponding ones in 50

kardaniya.

1
27a (gafi a l - D m )

2
27b ninzi (Qutb a l - D m )

204- 150 267 81

These add a whole-tone below 10 1iraq and 12 hic

jazi

respectively. It is interesting to note that the descriptions

3
of 2 7 a found in later texts generally refer to fric
j azi.

28 has no name.

204- 204- 182 112

T i i """ 11
4-08 590

1 BM. MS. 136, fol. 10v.

2 Besides nyryzy other texts have nyrzy, nyrz,and nyrwz.

3 E.g. in the jami* al-alfran and the anonymous treatise

dedicated to Muframmad b. Murad.
 (99)

This adds an undivided whole-tone below 8 rast and is

the upper pentachord of 4-9 kardaniya.

29 also unnamed: similar to the above.

I ....... , .. — -------- *—
V./ w ---

204 204 204 90

In the same way 29 relates to the tetrachord 5 1ushshaq

and the octave scale 52 variant of kardaniya.

 (100)

species of major sixth ( 5 : 3)

30 has no name: similar to (2 2 b) 1uzzal; it may be called

segah wa frijazi.

¥
150 267 81 204 182

I i i ^ Y > T 1
417 498 702 884

is one of a set of terms (the others being dugah,

chahargah and pant

j gah) which receive their first mention in

the durrat al-taj and are frequently used, with a variety of

acceptations, by later theorists. They may denote the first

five notes of 40 rast (and sometimes, by extension, the first

five notes of any given scale) or they may relate to segments

of the rast scale starting, generally, from the fourth and

1
comprising the number of notes indicated by the name. Thus

segah designates here the notes c d et . These segments

also form the bases of independent modes (shu* bas) and it is

to them that these names are most commonly applied, denoting

at the same time the most prominent note. In segah, therefore,

this will be e1?. This aspect is discussed more fully in the

section on the shu* bas translated in chapter 7 - hi«inzi here

refers to the tetrachord G - c, identical with 12

du - 2; se - 3; chahar - 4; panj - 5
 (101)

There are further instances below of the descriptive

terminology exemplified by segah wa fcijazi. The first term

consistently relates to the upper section of the scale.

31 has no name: it may be called segah wa husayni

112 182 204- 204- 182

T
294- 4-98 702 884-

is the pentachord 18 jiusaynl (G - d), which here

overlaps segah.

30 and 31 have a range of a neutral, rather than a major,

sixth.

1
32 rahawi (gafi al-Din)

a
t

33 rahawi-yi tamam (Qutb a l - D m )

150 26? 81 128 119

417 4-98 637 765 884

1 BM. MS. Or. 136, fol. 22.

 (102)

The prominent notes are either c alone, c and e, or G

1
and e . 32 consists of 10 1iraq below 18 frusaym (c - g ) ,

and 33 of 12 bijazi below 4 rahawi (c - e). As the former

is mentioned by nearly all later writers of the Systematist

school it is rather surprising that Qu£b al-Din should have

failed to include it among the octave scales. However, the

appearance of 4 rahawi in 33 rather than 32 would seem to

suggest that it is Qujb al-Din5s version that represents the

2
earlier stage in the historical development of this mode.

Against this it may be argued that 33 also contains 12 bijazi

which is not recognized by gafl al-Din and might therefore

be considered a development from, or substitute for, the

1 0 j.iraq tetrachord of 3 2 rahawi .

1 Qu'tb al-Din adds that the last pair is "khilaf-i agl11,

presumably referring to the omission of c, which links the
two elements.

2 Later treatises state that 4 rahawi occurs in 32 rahawi

from Bt to eb, but this is. clearly incorrect - the specious
justification is that these intervals are of the same class
(J J J) as those in 4 rahawi.
 (103)

species of minor sixth (8 : 5)

34 ifffahanak. It is also called kawasht and basta.

182 204 112 139 128 49

386 498 637 765 814

This places 10 1iraq below the neutral third species

1 zirafkand. gafi al-Din nowhere mentions igfahanak or basta,

but lists an octave scale 5 9 a kawasht which adds ft and g to

2
the above notes. The sharh states that "(59a) kawasht is
3
igfahanak 1 and it is not until the early 1 5 th century that

all three are distinguished. According to 1Abd al-Qadir 34,

beginning on the upper note, is igfahanak; kawasht is the

octave scale described by pafl al-Din; and the notes c dt

eb e'b form the melodic nucleus of basta (nigar).B1?, At?,

4
and Ct may be added, but c remains the final note.

1 One would expect to find this section before the

"species of major sixth". Oddly enough Qufb al-Din’s
heading is, in full, ’the remainder (baqx) of the 8 : 5
species".

2 BM. MS. Or. 136, fol. 21.

3 BM. MS. Or. 2361, fol. 104v.

4 Bodleian MS, Marsh 282, fol.46v.

 (104)

35 variant of igf ah an ale: similar to the above.

15 0 267 81 139 128 49

I i B™ i i ^ (------ 1 ™ i— iT| ™ i
417 zi-98 657 765 814

This differs from the above in having 12 hi.jazl in place

of 10 *iraq. Hence the relationship between the two is the

same as that between 2 2 a and 2 2 b, except that in this case

both are listed by Qu^b al-Din. There are further examples

below of Qufb a l - D m notating a larger than whole-tone interval

where gafi al-Din has a whole-tone, but nowhere else does

he give both himself.

36 kuchek-i tamam

112 182 139 128

\ i 1— 1 i— tp— i----- 1 1 i
— | T| ™ i
294 498 657 765 814

kuchek is 1 zlrafkand(-i kuchek), placed here above

9 nawruz (G - c). This mode appears to be peculiar to the

durrat al-tat
j.

Again, the range of 34-> 35* and 38 is a neutral sixth,

rather than a minor sixth.

 (105)

species of minor seventh

1
57 nawruz (gafi al-DIn)
nawruz-i tamam (Qubb al-Din)

112 182 204- 112 182 204

" ■ “ i— T i— T— i----- 1
294 4-98 610 792 996
990

57 consists of conjunct tetrachords, both being 9 nawruz.

gafi al-Din also mentions a form of nawruz in which the

2
octave note is added, but this may be disregarded since it

coincides with 48 frusayni; furthermore, no later writer

recognizes it. The sharfc states, quite sensibly, that

^ 8 husayni (despite belonging to the most important group

of modes) is derived from 57 nawruz, and not the other way

5
round- However, the disposition of the prominent notes

according to Qutb al-Din suggests that the relationship may

be somewhat less straightforward.

1 BM. MS. Or. 156, fol. 21.

2 D sErlanger, la musique arabe, iii, 1 5 5 *

5 BM. M S . Or. 2561, fol. 115v.

 (106)

38 has no name: similar to (2 2 b) *uzzal; may be called

chah arg ah wa frit

jaz1 .

------------- 0----------- — 5.— ----------- 0

—

—I —
— ------•--

150 267 81 204 182 112

i ™ i
----- r — ^ f ------ 1 T i ™ | i |
41 7 498 702 884 99 6

This consists of chahargah (see the comments on 50)

above 12 foi,jazl (G - c).

39 has no name; may be called chahargah wa igfahan.

—- "• *~
4 0 _h 1 #

139 128 119 11 2 204 182 112

t — fr— r— i— f
267 386 498 702 884 996

The lower tetrachord is 11 igfahan. No other text lists

38 and 39*
 (107)

octave species

4-0 *ushshaq

204- 204- 90 204- 204- 90 204-

5 *ushshaq below 13 *ushshaq (c - g).

4-1 busalik

--- #---- m
V*
p#
I■ Kt
99
O
I\ * 99

90 204- 204- 90 204- 204- 204-

6 bus all k below 14- bus all k (c - g).

4-2 nawa

--- -— 0--------- ----------------— --------- -

•
' IJ *... . • — ■• r “
W

204- 90 204- 204- 90 204- 204-

7 nawa below 15 (c - g).

4-3 rast

& #------

■ » ■ ■ ^ "
204 182 112 204 182 112 204

— i---- T— r™*p--- 1
---- 1
----T----■
““"p T r
586 498 702 884 996

8 rast below 16 rast (c - g).

The above four are identical with their namesakes as

1
notated by gafi al-Dm.

^ bit
jazi (Safi a l - D m )
‘iraq (Qufb a l - D m )

182 204- 112 182 204- 112 204-

I r“ p ( r“ p r™p J i I
386 498 680 884 996

Safi al-Din’s notation of frijazi may be misleading,

especially with regard to the et. If this note was present,

it was almost certainly not prominent, as in *iraq.

Safi al-Din5s statements on frijazi are a little confusing.

2
He gives first 46a jp.ija.zi G A1? Bb c d1? e1? f g. Later,

1 BM. MS. Or. 136, fol. 22.

2 Ibia., fol. 20. This scale is also the one given later
-*-or b-ijo-zi when the modes are defined in terms of lute
tablature.
as a kind of postscript, he states: "For some fcit
jazl is

[62a nihuft al-frijazij G At Bt c d et f g; but what

we have called trijazi is ‘iraq if fS is added fi.e. to

1
hijaziI". But according to gafi al-Din *iraq is G At Bt
2
c dt et f fit g: thus for this remark to make sense

there must be an alternative form either of ‘iraq (G At Bi?

c dt et f fif g) or of fr.ic
jazi (G At Bt c dt et f

g). The existence of the latter is confirmed in the risala

3 —

4
^5 ciraq (gafi a l - D m )

variant of ‘iraq (Qutb al-Din)

--- •---- #4-- #

y*
— ^ —
* ft o
-

182 204 112 182 204 112 119 85

b— r,l,ILi----r“ p---- 1
----- r™n-----r**I1 T T I
386 498 680 884 996 1115

Some later theorists call this scale *iraq and ignore 44.

‘Abd al-Qadir lists both, calling 44 ‘iraq and 45 *iraq ma‘

5
al-baqiyya.

1 Ibid., fol. 20v.

2 Ibid., fol. 22.

3 D ’Erlanger, la musique arabe, iii, 1 3 5 *

4 BM. MS. Or. 136, fol. 22.

5 Bodleian MS. Marsh 282, fol. 40v. His follower Jaml

also gives the same two scales as ‘iraq.
 (110)

1
^a bi,jazx (5>afx al-Dxn)

... — 0— *

/ Jr K. *
sy
Op *
9
y ....

it
jazi (Qu£b al-DTn)

0 O -- ---
■X. I-j- 0 1r*
... U . t r*

112 182 204- 150 267 81 204-

I t“ t--f — i--1 rT~i -- H— I-- 1

1 — I
294 498 648 915 996

46b m a y be divided into 9 nawruz (G - c) and 22b *uzzal.

The position of the prominent notes (which will be used as

guides in other cases also) suggests, however, that it would

be more meaningful to divide the upper pentachord into 12

frinazi and a whole-tone rather than to consider it as a unit.

As expected, one finds in 46a 10 *iraq corresponding to

12 bit
jazi in 4 6 b .

47 frusayni

---- ^ .. ^ k ---- -.------- «—

4, f- = ^ =

112 182 204 204 90 204 204

I— r-*i— T — I— T— I— I— r I T— I
— l
294- 498 702 792 996

1 BM. MS. Or. 136, fol. 22.

 18 fcusayni below 6 busalik (d - g). Qufb a l - D m is
1
alone in considering this scale to be busayni.

2
4-8 frusayni (gafi a l - D m )

variant of busayni (Qufb al-Din)

o
. - V#.... - ■------- #-------
-------K 0
-■jjp------- 1-------

112 182 204- 112 182 204- 204-

I i“ i y i
— (— rrJ,l,BI—i
— f— i
— y i l
294 498 6-10 792 996

^8 busayni above 9 nawruz (G - c). This is the only form

°£ b^ssiyni mentioned in treatises of the 14-th and 1 5 th

centuries.

4-9 kardaniya

---- 9--------
•
{V A •
— tp----------

204 182 112 204- 204- 182 112

1— T — r*p— f— i— |— 1 T[— i “ |i
386 498 702 906 1088

1 In the sharb (BM. MS. Or. 2361, fol. 104-) 4-7 is called
tj anfaza. But this does not necessarily mean that it was in
use at the time, for the name is probably personal to the
author, being one of a set of fanciful descriptive terms
applied to several consonant scales rarely, if ever, found
in practice. However, the same work (ibid., fol. 115v) does
state that 4-7 is the scale from which rakbi (see chapter 7)
is derived.

2 BM. MS. Or. 136, fol. 22. 4-8 constitutes the one case in
which Qu£b al-Din’s order has been disturbed.
 (112)

8 rast below 28 or, preferably, 8 rast plus whole-tone

plus 8 rast.

1
50 kardaniya (£>afl a l - D m )

variant of kardaniya (Quj:b a l - D m )

3'-

204 182 112 92 112 204 182 112

m T F T » T 1 T
386 498 590 702 906 1088

8 rast below 26 kardaniya (q.v.), equivalent to 8 rast

plus whole-tone (divided) plus 8 rast.

51 variant of kardaniya

........ -.. ......... -.... - .. -... -..i . *•--------- -

l>—

204 204 90 204 204 182 112

T T I T ' T
408 498 702 906 1088

5 *ushshaq plus whole-tone plus 8 rast.

1 Ibid., fol. 21.

52 variant of ka:

204 182 112 204 204 204 90

1 T 1--- f T
586 498 702 906 1110

8 rast plus whole-tone plus 5 1ushshaq.

55 variant of kardaniya

t
204 204 204 90 204 204 90

29 below 5 ‘ (cL - g).

50 is the only form of kardaniya recognised by gafi al-

Din, and the only one to include the pentachord 26 kardaniya*

It is likely that the other four derive from it. Each of them

avoids the juxtaposition of c and c$, a feature otherwise

found only in the pentachord 20 busurg and the modes relating

to it. The scale which results from omitting c in 50 is

listed by Qu£b al-Din as 65 kardaniya nirlsi.

1
A number of later theorists, headed by 1 Abd al-Qadir,

assign the name kardaniya to 49? 50 being called

1 Bodleian MS. Marsh 282, fol. 49v.

 U14-;

1
za’id. By the 15th century, if not earlier, 51 was known
2
as mahur. 52 and 53 appear only in the durrat al-taf
j.

54* buzurg

128 231 139 128 76 204 182 112

I r-h 1 |T T I ' T ■- P 1
359 498 626 702 906 1088

^0 buzurg (q.v.) below 5 *ushshaq (d - g).

55 variant of buzurg

-i it
1*- J6f *

128 251 139 128 76 139 128 119 112

T — T— Ti [ T — IT
359 498 626- 7 0 2 841 969 1088

20 buzurg below 11 igfahan (d - g ) .

1 The term id clearly refers to the presence of the oil,

even though this may not have been produced on the zaVid
fret.

2 Ibid., fol. 51v. mahurl in the sharh (BM. MS. Or. 2561,
fol. 105v) has f# in place of ft. The same work calls 51
farab (ibid., fol. 118) and, equally capriciously, terms
49 bay (La* (ibid., fol. 105v). In the anonymous treatise
dedicated to Muhammad b. Murad 51 is called nihawand (ibid.,
fol. 200v), a name usually applied to another scale.
 mb;

56 variant of buzurg

& i 4r :fe

128 231 139 128 76 150 267 81

1— r T — T— — f — r~
359 498 626 7 0 2 852 1119

20 buzurg below 12 azl (d - g) .

No other text mentions 55 &nd 50. gafi a l - D m lists 54

1
buzurg, and later theorists follow him in notating this mode

with B”c (B1? ).

As with 50 kardaniya and its variants, it may be suggested

that 55 &n& 56 represent attempts to obviate certain features

°£ 54 buzurg. The nature of the problem and the reasons

leading to the particular solutions adopted will be discussed

in chapter 4.

57a zankula (gafi a l - D m )

-£==&

1 BM. MS. Or. 136, fol. 21.

2 zankula is given without fit in three places (BM. MS. Or.

136, ff. 20, 22, and 27v) and with fit in one (ibid., fol.
2.1v) .
 (116)

57 b zankula (Qu£b al-Din): some people also call it

nihawand,

------- ----------------- ------------------- ^ -------- #--------.-------4 * ------ *—

t "fi, 0 r’
(
V/ — — ..... ....................... ■

204 182 112 150 267 81 119 85

— i— T — i — (--- 1 ^ i---- 1 1) T If 1
586 498 648 915 996 1115

57a consists of 8 rast plus 10 ‘iraq plus (divided)

whole-tone. In Qufb al-Din5s version (in the notation of

which the original gb has been rendered by ft) 10 ciraq is,

as one might expect, replaced by 12 frie

j azl. With regard to

the potential omission of ft in 5 7 it may be noted that it

is not included in one of the two versions of zankula in the

1
kanz al-tu^iaf. This treatise also mentions nihawand as an
2
alternative name: in most other texts of the I4thand 15th

centuries the name nihawand is however applied to a totally

3
unrelated scale.

Elsewhere Qujb al-Din gives a further description of

zankula which suggests that the above are misleading

4.

alterations, and hence that 57b does not represent the mode

1 BM. MB. Or. 2361, fol, 257.

2 Ibid., fol. 258, where it is stated that this usage is

found in Bukhara. It may therefore have been confined to
East Persia and fransoxiana.

3 Al-Ladliiqi does however list a nihawand gaghxr (BM. MS.

Or. 6629? fol. 70) containing 12 frijazi.

4 BM. MS. Add. 7694, fol. 254 (translated in chapter 7)*

as he knew it, "being no more than a slightly modified

borrowing from gafl al-Din. This alternative description

will be referred to in chapter 4.

1
53a zirafkand (gafi a l - D m )

= = j Z - — z— •fin d s fc— f‘ '

ft 9*

9 nawruz below 2 5 a (c - g).

58b higar (Qu^b al-Din)

112 182 204 159 128 49 26? 119

I----- r-* -r— T --- 1 tT i r^ i 1---F — 1

294 498 637 765 814 1081

9 nawruz below 2 5 b hiff&P (c - g).

For ‘Abd al-Qadir hih&r consists of the notes c dt eb

2
et fi g at bt, while for al-Ladhiql it includes only c
5
dt eb et fit g at. Because of the absence in later texts

of any reference to 58b as hi gar it may be suggested that

Qutb al-DIn?s comments on the phrasing of this mode, which

stress the prominence of the upper note, relate primarily to

1 BM. MS. Or. 156, fol. 22.

2 Bodleian MS. Marsh 282, fol. 51*

5 BM. MS. Or. 6629, fol. 68v.

 uio;

the pentachord 2 5 b frigar, from which the later forms quoted

are probably derived.

1
59a kawasht (gafi al-Din)

— b* 1>» f*
f

19 4iraq below 2 5 a (c - g)

59b has no n a m e : may be called higar wa igfahanak (Qutb

al-Din).

182 204 112 139 128 49 267 119

1 r t r
— r-n— nr*-
386 498 637 765 814 1081

10 *iraq below 25b fai^ar (c - g).

Qut?b al-Din5s descriptive terms overlap, 3^ i

comprising the notes from G to et, 2 5 b higar those from c to

The position of the prominent notes suggests, however, that

this mode had little in common with 3^ igfahanak.

1 BM. MS. Or. 136, fol. 21.

 (119)

60

— e- 5s.

182 112 204 204 182 112 204

i " ™ i— IT— i
— T— 1— I " r — T f
294 498 702 884 996

Safi al-Din mentions this mode only in the risala al-

1
sharafiyya. Erom the name, and the prominence of the d, it

would seem that this scale consisted of 18 busayni below

9 nawruz (d - g), i.e. of parallel disjunct tetrachords.

Alternatively, the order of interval sizes given in the

ratios implies the analysis 9 nawruz below 16 rast (c - g).

The example of notation in mubayyir busayni (transcribed in

chapter 8)suggests that both divisions are valid. The fact

that only the latter is given by theorists of the Systematist

school is without significance, for they do not admit the

disjunct octave scale. It is possible that the epithet

mubayyir was applied to this mode precisely because of this

ambivalence, on the one hand leading to 48 busayni (D - d)

by way of the lower pentachord, and on the other to 45 rast

(c - c') by way of the upper pentachord.

1 D 5Erlanger, la musique arabe, iii, 156.

 (1 2 0 ;

61 variant (of mubayyir busayni): may be called mubayyir

zlrkesh.

(>__* <P' b.
p#.... b*
t #

182 11 2 92 112 204 182 112 204

I r"n f f T--- 1 T r— p-T-- 1

-- 1
294 386 498 ?02 884 996

There are again two possible divisions: 19 zirkesh busayni

below 9 nawruz (d - g), or 11 igfahan below 16 rast (c - g),

Whichever is preferred, there is a difference in the ratio

values which is indicative of the extent to which these

could be determined by theoretical and/or analytical

preconceptions: the values for the first four intervals of

19 zxrkesh frusayni are identical with those of 11 igfahan;

the corresponding intervals in 61 are however assigned quite

different values relating to those already established for

60 mubayyir busayni■ In spite of this, it is doubtful

whether 61 is derived from 60.

1
62a nihuft (al-bit
jazi) (gafi al-Din)

1 Ibid. In the kitab al-adwar (BM- MS, Or. 136, fol, 20v)
62a is referred to simply as .
62b nihuft-i bi.jazl (Qu£b al-DIn)

•
---- 1-----
1%
v/ * ..
v
P# .. — ----- -

150 267 81 204 182 112 204

I
--- r_ n 1— i■ > — t — i
— H
41 7 498 702 884 996

62a is 2 2 a below 9 nawruz (d - g), or 10 *Iraq below

^ 8 rast (c - g), while 62b is 2 2 b 4uzzal below 9 nawruz or

frijazi below 16 rast.

The name is explained as meaning that this mode is derived

from frijaal in the 9 th tabaqa (degree of transposition), i.e.

Bt - b t . The reference must therefore be to 44 failazl which,

in this position, is Bt c d et f g at bt. This

derivation is however quite artificial, and there exists in

any case a much simplex' relationship between 62a and b and

46a and b respectively (46 being also frijazi), the former

pair emerging when the prominent fourth of the latter assumes

the r 6 le of tonic.

1 2
62a is listed by ‘Abd al-Qadir, 62b by al-Ladhiql.

1 Bodleian MB. Marsh 282, fol. 51*

2 BM. MS. Or. 6629, fol. 70.

63 little known: may be called kardaniya nirizi

204 150 26? 81 204 182 112

T r jn — ?•
354 621 7 0 2 906 1088

27b nxrlzi below 8 rast (d - g). Al-Ladhiql calls this

1
mode niriz kabir.

64 little known.

204 182 112 204 150 267 81

-i— T- l rr T * T -i
586 498 702 852 1119

8 rast (G - c) below 2 7 b nirizi: the tetrachord and

pentachord of the previous scale are reversed.

65 little known.

90 204 204 204 90 204 204

6 busallk below 15 (c - g) or, preferably, 14 busallk

below 6 busalik (d - g).

1 Ibid.
 C125)

In the sharh this mode is called wipal and is stated to

1
he in use among the lurks. No other text lists the scale,

or mentions the name.

66 i (gain a l - D m )

-fr -fr 1
• -- 0--
-Jn--- .---- ?---- ------------------------------------

67 i (Qufcb a l - D m )

112 182 204 139 128 119 112 204

b ■ ^ "“ i— p T — p""
294 498 637 765 884 996

66 and 67 appear at first sight to he quite unrelated.

Xet elsewhere Qu^h a l - D m makes it clear that he also

3
recognizes 66 as igfahan, while *Ahd al-Qadir and, following
4
him, Jam! list hoth forms. 67 may he derived from 66

(or vice versa) through the sequence:

66 G A Bt c d e1? f f$ g ------------ >

61 GAt Bb B c d et f g ------------ >

1 BM. MS, Or. 2361, fol. 103-

2 BM. MS. Or. 136, fol. 22.

3 Bodleian MS. Marsh 282, fol. 40.

4 risala-yi musxqa, fol. 442v -

 (124)

67 G Ap Bb c dt eb e f g

in which the fifth (beginning from 66) or the fourth (beginning

from 67) provides the tonic of each successive scale. But
such a relationship would appear to he no less theoretical
than those propounded by writers of the Systematist school,
for 61, the middle term, is nowhere associated with igfahan.
A much simpler and more convincing explanation is suggested
hy the placing of the prominent notes in 67* These embrace
11 igfahan from c to f, evidently the most important and
characteristic segment of the mode. It would he reasonable
to assume that in 66 too 11 igfahan (d - g) was prominent,
and that therefore the two forms were differentiated simply
hy the direction in which they extended to the octave from
an original form consisting of just two conjunct tetrachords
(A - g in 66, G - f in 67)* Hence we maydescribe 67 as
9 nawruz plus 11 igfahan plus whole-tone, and 66 as whole-tone
plus 9 nawruz plus 11 igfahan.
 (125)

species of octave plus whole-tone

68 has no name

is

1
150 267 81 204 112 182 204 204

2 2 b 6uzzal plus 18 frusayni (d - a), resolvable into

9 nawruz (d - g) plus whole-tone.

All later theorists define modal structures in terms of

the octave or some lesser interval, as does gafi al-DIn.

From this one naturally assumes that, as in 70 - 72, the

same notes would be repeated at the octave if a mode were

extended beyond that limit. It is therefore interesting to

find in 68 evidence that this did not necessarily always

happen. Non-repetition of notes at the octave, or more

correctly substitution of a different species (usually

tetrachord), is a common feature of the modern Arab maqam.

1 For the diagrammatic presentation of the intervals add

a whole-tone above 62b.
 (126)

species of octave plus tetrachord

69 nihuft-i kamil

1
182 204 112 150 267 31 204 182 11 2 204

*Iraq below 62b nihuft-i bilazi (c - c' )*

Under the heading n ihuft-i kainil Quj-,b al-DIn lists two

scales: the notation (with which the designation of interval

2
sizes by the symbols B J J* H agrees) is given below as 70?

while the ratios are those of 6 9 * Ihe notation of 69 has

therefore been derived from the ratios. It is not clear to

which scale the prominent notes indicated belong and so they

have been shown in both.

Because of this confusion one might question whether both

these scales existed. This is unfortunate in that 69 would

provide evidence even more conclusive than that of 68 with

regard to the possibility of extending the range of modes by

adding different species as well as by simple repetition at

the octave.

1 Bor the diagrammatic presentation of the intervals

combine 10 and 62b.

2 H designates a larger than whole-tone interval.

 (127)

70 nihuft-i kamil

62b nihuft-i biqazi below 12 (s - c' ) .

It may be of interest to compare these conflicting

accounts (? or variants) of the same mode with the modern

Egyptian maqam nihuft, in which the following notes occur:

D Et E G Ab B c d et f g a? bl? c' d'

(69 D Et Ft- G At B c d et f S
4-
70 G At B c d ei? f g at b c')

71 type of buzurg-i kamil

<$• *

128 251 159 128 76 159 128 119 11 2 119 267 112

55 variant of buzurg below 12 frijazl (g - c' ). The

difference between the ratios for the upper tetrachord and

1 Variation is possible (indeed necessary) in the tetrachord

G - c- If nihuft is on D the note corresponding to yahah
will be 0, not G. See Sami al-Shawwa, al-qawa?id al-fanniyya
fi al-muslqa al-sharqiyya w a ’1-gharbiyya, 29 i and I)’Erl anger,
la musique arabe, v, 142.

2 For the diagrammatic presentation of the intervals

combine 55 and 1 2 .
 (128)

those for 12 tLic

jazi probably results from an attempt to make

the values of the latter conform more closely to the ratio

values assigned to the corresponding intervals in the lower

octave. The heading implies that the other forms of buzurg

might be similarly extended.

72 busallk

---------------------
x ? # ■
— # --- ----- +-----
.... ..........
t 1 / - \ * y •

X ' V * r '
y y

90 204 204 90 204 204 204 90 204 204

41 busallk below 6 busalxk (g - c' ).

It is probable that no especial importance attaches to

the inclusion in this list of 71 and 7 2 , and that most other

modes could be extended beyond the octave if the performer

so desired.

One mode mentioned by §afl al-DIn, salmak, has been

omitted from the above list. It will be referred to in parts

2 and 3 .
part 2
 (150)

chapter 3

It is evident that there are a number of differences

between gafi al-DIn’s account of the modal system and that

presented in the durrat al-taj. Apart from any historical

or geographical considerations, this fact alone renders it

imperative to deal with the two separately, at least in the

initial stages of enquiry. We may begin with that of gafl al

Din which is, conveniently enough, the smaller as well as the

earlier.

In all, £>afi al-Din mentions 20 mode-names, 16 of which

relate to octave scales (in the case of bijazl one name is

appended to two scales). Of the others three ((23a) may a ,

(24a) shahnaz, and salmak) designate modes characterized by

particular melodic features. These will be discussed in

chapter 7* Here we shall be concerned primarily with the

octave scales plus the remaining mode, 37 nawruz, and an

attempt will be made to ascertain what, if anything, they

have in common and hence to define the principles underlying

the structure of that part of the modal system they represent

In addition, gafi al-DIn mentions 21 of the 29 species of

third, fourth, and fifth. As the comments on the octave

 (131)

scales show, these species may be thought of as the component

parts from which more extended scales are formed. But here

a distinction must be maintained between those which owe

their existence to the requirements of a sometimes arbitrary

method of scale analysis and are mere theoretical abstractions,

and those which may be saidto exist independently of the

scales in which they occur. It is the latter which form, in

effect, the raw material from which the system is constructed,

and as such they demand separate consideration.

We may begin, however, by examining the particular

groupings of modes recognized by gafl al-JJln. The two main

groups are the shudud and the awazat, comprising between them

18 modes. We are further informed that a number of scales

are called murakkabat, although only one example is given.

Finally, two of the modes he lists are not assigned to any

of the above and may be held, provisionally, to belong to a

fourth category. (Qu£b al-Din discards these categoides, at

least in the form established by his predecessor, and attempts

to classify scales along different lines. Most later

theorists, however, adhere to them and reproduce gafi al-

Din 5s version of the shudud and the awazat, which were

evidently thought to embody the most significant part of the

 C 152J)

1
modal repertoire. )

Of these categories that of the shudud is by far the

largest, comprising 12 modes and, since for one of these two

forms are mentioned, 13 of the 18 scales under consideration.

Ihey are divided by gafl al-BIn into three sub-groups, each

being associated with a particular emotion or sot of emotions

which the modes in it are thought to evoke. Ihe sub-groups

are:

40 1ushshaq G A. B c d e f 6
41 busallk G Ab Bb c db eb f e
42 nawa G A Bb c d eb f e

said to inspire strength (quwwa), courage (shaja* a ),

and pleasure (bast)

(b) 43 rast G A B1? c d et f S

45 *iraq G At Bt c dt et f f if g

66 isfahan G A Bt c d et f fI g

said to inspire refined pleasure (bast ladhidh

latlf)

1 It is likely that their continuing importance two

centuries later may be ascribed to other factors as well,
notably the tradition of extra-musical associations to which
these groups were particularly subject, and the almost
canonical stature that gafi al-Din5s writings acquired.
Although he was criticized on certain points, his versions
of these modes were reproduced unquestioningly down to the
end of the 1 5 th century, by which time they had long since
ceased to accord fully with actual practice.
 3 2 rahawl G At Bt c at eb f g
46a b-ijazl G At Bb c at et f g
?and/or 44 frijai51 G At Bt c at et f g
48 busaynl G At Bb c at eb f g
It
54 buzurg G At B(t)c Ctt d e ft g
5 7 a zankula G A Bt c et f (f II) g

5 8 a zlrafkand G At Bb c dt eb et ft g
1
said to inspire sadness (fruzn) and languor (futur)

In(b) gafi a l - D m also includes 37 nawruz G At? Bb c

dt eb f, although it is not oneof the shudud.

One might be tempted to dismiss this classification as

subjective. It is a simple reflection of the doctrine of

ethos which had previously been presented with complex

ramifications in the treatises ofal-Kindi and the rasa* il

of the Ikhwan al-gafa*, and which reappears with equal

profusion in a number of post- 1 3 th century musical texts.

However, it is clear that these sub-groups may also be related

to certain aspects of scale-structure. For instance, (a)

consists of diatonic scales, (b) and (c) of non-diatonic or

2
Zalzalian scales. Further, if one disregards the division

1 Given in the kitab al-adwar, chapter 14, fI ta*thir al-

nagham (BM. MS. Or. 136, ff. 37^-38)• Compare the three­
fold classification in Mas 4udl, muruj al-dhahab, viii, 96,
and a further such division, although based on different
principles, in Kayka’us, qabus-nama, 1 1 1 .

2 V/e shall term all species and scales containing neutral

intervals Zalzalian, whether or not the wusta zalzal itself
would have been used to produce them.
 (134)

1
of the upper whole-tone in 45 4 iraq and 6 6 Isfahan, it will

he seen that all the scales in (a) and (b) can be analysed in

terms of parallel conjunct tetrachords while most of those in

(c) cannot* We may expand these statements and say that

(a) contains only scales with parallel conjunct diatonic

tetrachords

(b) contains only scales with parallel conjunct Zalzalian

tetrachords (plus, in two cases, a divided disjunctive

whole-tone). 37 nawruz is also of this type.

(c) however comprises

(i) scales with parallel conjunct Zalzalian tetrachords

( ^ bijazi and 48 jiusayni)

(ii) scales with non-parallel conjunct Zalzalian

tetrachords (32 rahawi, 46a hijazi and 57a zankula)

(iii) scales with a Zalzalian tetrachord and pentachord

(54 buzurg and 5 8 a zirafkand)

Ihus the major difficulty in attempting to equate gafi al-

Din 5s divisions with a structural distinction relating to the

nature and distribution of the species in these scales is

1 In the present chapter, which is primarily a discussion

of the modal categories according to gafl al-Din, the analysis
of Isfahan given by him has been assumed to be valid since,
clearly, it may have influenced the choice of the sub-group
to which this mode is assigned. Although there can be
little doubt that his analysis obscures the relationship
between 66 and 6 7 , which was surely crucial to the identity
of the mode, it may nevertheless reflect the assimilative
pressure exerted by 43 rast*
 (135)

that there is no apparent reason to justify the inclusion of

(i) in (c) rather than (b). However, it is likely that when

defining these sub-groups (in which the mode-names are listed

without any notation) gafi al-Din used the name fcijazi to

refer specifically to 46a td-jaai, which has non-parallel

tetrachords and is the only form of this mode cited by later

authorities as in current use- Also it could be suggested

that the placing of 48 fcusaynl in (c) is attributable to a

desire to differentiate it from the almost identical 37 nawruz

in (b)- It is interesting to note here that the shark makes

certain reservations with regard to the emotional content of

kusaynl and k-ijazl (although this presumably means 46a

_1
frijazi ), stating that the sadness snd languor they inspire
2
is tempered by a certain gaiety- Thus they could be thought

of as a separate class between (b) and (c)* The sub-groups

according to the shark therefore become

(a) 40 4ushshaq, 41 busallk, 42 nawa

(k) 43 rast, 45 4 iraq, 66 igfahan (and 37 nawruz)

intermediate: 46a kij&zl, 48 kusaynl

(c) 32 rahawi, 54 buzurg, 5 7 a zankula, 5 8 a zirafkand

The shark also draws attention to the structural

relationship between the modes in each sub-group, but

1 Since the shark calls 44 "the kijazi of the ancients"

(EM. MS. Or. 2361, fol. 104v).

2 Ibid., fol. 149v.

 (136)

expresses this in a slightly different way. Mention is made

not of the parallel/non-parallel distinction drawn above, but

solely of the type of species involved. Each sub-group is

characterized by one particular type occurring in all of its

modes, the types being in

(a) diatonic species

(b) strong non-diatonic species, i.e. Zalzalian

tetrachords containing not more than three intervals

(c) with the exception of kusaynl and kdjjizl: isolate

species (ajnas mufrada), viz. 4 rahawi G At Bb B,

20 buzurg Ct At B(t) c cil d, and 25a G At Bb

1
Bt ct d

But this classification breaks down at the same point:

kusaynl and have to be excluded from (c), and according

to the criteria given could be included in (b). In addition,

as has been noted, the identification of 4 rahawi with a

segment of 32 rahawi G At (Bt c dt eb) f g is quite

unconvincing and is certainly of no importance as far as

practice is concerned. As a result, the definition offered

of the common factor in (c) may be dismissed as an example of

over-simplification as well as of ingenuity, and the problem

of elucidating the nature of the differences between the modes

in (b) and those in (c) remains unresolved.

1 Ibid.
 (137)

If, however, we accept that kusaynl and hinazi form an

intermediate category, it becomes possible to reformulate the

first set of definitions and provide a clear-cut distinction

between the two, for we may now say that while

(b) contains Zalzalian scales with parallel conjunct

tetrachords,

(c) contains Zalzalian scales which either do not have

parallel tetrachords, or cannot be divided into

conjunct tetrachords.

With only these criteria it is however Impossible to give

a satisfactory explanation for the separate status of 48

kusaynl and (presumably 46a) kij&zl* For 48 kusaynl one must

revert to the argument, proposed above, that it was felt

undesirable to include it in the same sub-group as the very

similar 37 nawruz. If however 46a ki j azl is examined from

a different standpoint, namely that of the number of intervallic

consonances (fourths and fifths) it contains, it transpires

that it does occupy an intermediate position between the

modes in (b) and the modes with conjunct tetrachords in (c).

This aspect will be dealt with more fully below.

Finally It may be pointed out that while gafl al-Dln*s

classification is based primarily on the type of species and

the arrangement of the species within the octave, no importance

is attached to the arrangement of the intervals within the

species. Thus the modes in (a) show that the variations in

 (138)

emotional content which in the West would he associated with

the contrast of major and minor thirds (and sixths) were

either not perceived, or were regarded as insignificant: it

is the size of the intervals within the species that matters,

not their disposition.

Given that the sub-groups are determined by features of

intervallic organization, it may be asked what this tells us

about the shudud as a whole: notably, is there any reason

why these particular modes should be classed together and

separated from the other four octave scales mentioned by gafi

al-Din? The similarity between the modes in (a) and those in

(b) is obvious: all are analysed as having parallel conjunct

tetrachords. But what do they have in common with those in

(c)? If, following gafi al-Din3s own approach, we continue

tothink in terms of conjunct tetrachords,affinities may be

found between the modes in (a) and (b) and four of theothers

32 rahawi G At Bt c dt eb f g

46a hijazl G At Bb c dt ef f g

(?and/ or 44 hi jazl G At Bp c dt et f g)

busayni G it Bb c d t e b f g

57a zankula G A Bt c dt et f (ftt) g

although in 32 rahawi, 46a hi jazl, and 57 a zankula the

tetrachords are non-parallel. The remaining two shudud,

however,
 54 buzurg G At B(t) c c# d e ft g

58a zirafkand. G At Bb c dt eb et ft g,

do not fit into this pattern. With the latter one can see a

certain similarity, in the partial parallelism at the fourth

(G A1? Bb/c d? ei?), to the modes of (a) and (b), but 5^

£ has neither parallelism nor conjunct tetrachords.

At this stage, therefore, it would appear that there are

no sufficiently distinctive common features to justify

isolating these particular 12 (or 13) scales. Any general

statement of the type: the shudud are octave scales which

may be resolved into conjunct tetrachord plus whole-tone or

into tetraclioi'd and pentachord is far too vague, being in

fact valid for all the octave scales listed by both gafi al-

Din and Qu^b al-Din. In addition, to anticipate once more,

there do not seem to be any features of consonance shared by

the shudud which could be added to the above definitions to

form the basis of a structural definition precise enough to

exclude other existing octave scales. According to the

_ 1
criteria laid down by J>afi al-Dm, 60 mufciayyir- busaynn is

as consonant as any of the Zalzalian scales and considerably

more consonant than 5 ^ buzurg.

The contrast between the structure of 5^ buzurg (and to

a lesser extent of 5 S& zirafkand) and the other scales provokes

BM. MB. Or. 136, ff. 11v-12.

 (1 4 0 )

a further question: why should there have been 1 2 , and not

10 or 11 shudud? Here it is likely that extra-musical factors

were involved* In all probability a group of 12 was chosen

as a significant entity because of the wide range of

macrocosmic associations that had crystallized around the

1
signs of the Zodiac. It may also be noted that 12 is a
2
multiple of 4, the number of the strings of the lute which,

from al-Kindi onwards, had been linked with several sets of

phenomena or concepts a m a n g e d in fours (e.g. the humours,

seasons, elements). The choice of these pai*ticular 12 modes

3
should probably not be ascribed to gafi al - L m , but rather

to a gradual selection process of an instinctive, not conscious,

nature. By the mid-13th century, if not before, the process

had been completed, and it is evident from the way they are

presented in the kitab al-adwar that the shudud were regarded

as the nucleus and most essential part of the modal repertoire.

1 It is true that neither gafi al-Din nor Qu^b al-Din

mention this aspect, but this is to be attributed to the
primarily scientific nature of their approach (cf. the
relevant passages in the Introduction). Several later
writers, especially those not strictly concerned v/ith
expounding the doctrines of the Systematist school, delight
in listing whole series of such correspondences, and it is
certain that they were felt to be of importance during the
1 3 th century too.

2 Long before the 1 3 th centuries the number had been

increased to five, but the significance of the four as a set
remained just as strong.

3 Had he chosen them, one would not expect to find the

hesitation over the form of frijazl which the kitab al-adwar
betrays.
 (141)

Hence it may be assumed that all were well-known and frequently

performed* But there were others apx^arently equally popular,

and so it is still not clear how these were chosen - which

brings us back to the original problem of finding distinctive

common features among the shudud or, less positively, criteria

of differentiation with regard to the other modal categories.

Of these the most important is the class of awazat, made

up of the following six modes:

50 kardaniya G A. Bt c c# d e It s
59a kawaslit G At Bt c dt eb et ft g

nawruz Gr At Bb c dt eb f

23a may a G Bt c eb f

24 a shahnaz G At Bb Bt* At G
j\
1
salmak G A B ct d et

Here one is faced by a multiplicity of scale-types, a fact

which in itself could be considered a justification of the

attempt to detect some kind of structural unity among the

shudud* Indeed, the only reason for forming these six modes

into a set - apart from any extra-musical factors that might

be involved - would appear to be the totally negative one

that in contrast to the shudud no common features are in

evidence.

1 BM. MS. Or. 136, fol. 21 (transposed from c).

 (142)

With the exception of 37 nawxfiz the non-octave awazat

will he discussed in chapter 7 * It may however he noted that

in comparison with the other scales listed by £&fi al-DIn

they are all unique structures: 2 3 a maya is the only example

1 2
of a transilient scale; saim ale apparently omits the fourth;

and 24a shahnaz has a range of a mere neutral third, being in

effect a melodically limited variation of 1 zirafkand. Such

anomalies, given that they may be viewed as such within the

context of an overall system, would tend to be eliminated,

and it is therefore of interest to find Qu^b al-DIn giving a

substantially different account of these three modes.

In contrast, the other three awazat appear to have much in

common with the shudud. But paradoxically it is probably

57 nawruz G At Bb c dt eb f, the non-octave scale, which

stands the closest, although it is by no means just sn

incomplete form of 48 dusayni G At Bb c dt eb f g.

It is true that an equal, if not greater, degree of similarity

would appear to obtain between 59& kawasht G At Bt c dt

eb et ft g and 56a zirafkand G At Bb c dt eb et ft

g. However, the evidence provided by Qu*£b al-Din in 34

igfahanak (or kawasht) G At Bt c dt eb et suggests that

1 Oddly enough the G - Bb interval in maya w as later

increased to G - B t . (This may perhaps be explained as an
assimilation to the rast scale, but it is neverthless
extraordinary that the third (whether minor or neutral)
should have remained undivided.

2 Taking Qu£b al-Din5s description of salmak into account

it is difficult to determine if any one note” should be
regarded as the tonic,and if so which.
 (143)

basically this mode was not an octave but a neutral sixth in

range, and that the form given by gafi al-DIn is a theoretical

levelling by analogy with 583- zirafkand* Although not

conclusive support, it is interesting to note that the melody

in kawasht presented in the kitab al-adwar contains only the

1
notes from At to et (inclusive)* 50 kardaniya G- A Bt c

c# d e f$ g on the other hand in undoubtedly an octave

scale: but it is also unique, for it has disjunct tetrachords

with between them a divided whole-tone (whereas in 54 buzurg

G At B(t) c cfi d e ft g the intervals c - cif emd cS - d

2
are integral parts of a pentachord). The large number of

variant forms listed by Qu^b al-Din for kardaniya (49? 51 - 53)

suggest that this was felt to be unusual and tended to be

avoided.

It would be unwise to ascribe a source to a particular

modal form on the evidence of its name alone. Nevertheless

it should be noted that both the term awaz and the names of

the six modes in the group are of Persian derivation.

Unfortunately only one of these, nawruz, is to be found in

earlier lists of Persian melody- or mode-names, and even here

there is no guarantee of continuity with regard to the modal

3
form itself. Hence there is no way of telling whether or not

1 Transcription in chapter 8 .

2 See also the comments on 20 buzurg and 26 kardaniya .

3 Of. also the suggestion of a Turkish rather than Persian

origin for 2 3 a maya (q.v.).
 (144)

the awaz group was formed before that of the shudud, which

have mixed Arabic and Persian names. The only thing that can

be stated with any certainty is that the numerical relationship

of the 6 awazat to the 12 shudud was of some significance,

although as was to be expected neither §>af! al-Din nor Qu^b

al-Din refer to this. It was left to later writers to

expatiate on the extra-musical correspondences between the

two sets. The awazat were sometimes held to be derived from

the shudud, and sometimes to be their origin, the former view

being the more widely accepted. The various groupings made

relate not to intervallic similarities or the way the species

are arranged but to the already established cosmological

affiliations of each mode.

The third category of modes according to §>afi al-Din, or

rather potential category since only one example is given,

1
is that of the murakkabat. This is a technical term by which

scales are defined according to their constituent tetrachords

and pentachords, and in theory could be applied to any of the

octave scales discussed above: thus 43 rast G A Bt c d

et f g is a murakkab of the tetrachord 8 rast and the

pentachord 16 rast, while 58a zirafkand G At Bb c dt eb

et ft g is a murakkab of 9 nawruz and the pentachord 2 5 a.

Treatises sometimes devote a certain amount of space to

1 BM. MS. Or. 156, fol. 22v.

 (14-5)

discussing the propriety of using this method as an alternative

description for modes which already have a name. At a later

period the term murakkab, or rather the related form tarkib,

is used to designate a number of modes outside the previously

established groups of shudud and awazat. For gaff al-Din,

however, murakkab is merely an adjunct to the vocabulary of

theoretical analysis, and in the 1 Jth century there was no

such class of modes. The example given to illustrate this

category is according to some manuscripts

G At Bt c dt eb e f g

and according to others

G At Bt cd et f f# g.

The accompanying description ("a combination of igfahan and

hijazlM) could be applied to either, but the former is the

more likely. In any case it is probably no more than a

hypothetical example, gafl al-Din attaches no name to it

and saysnothing to suggest that It might have been used in

practice. Neither scale is included in the lists of modes

2
found in later texts, and in the sharfr the former is dismissed

1 An example (which also quotes gafi a l - D m }s remarks on

the subject) may be found in the passage from the durrat
al-taj translated in chapter 7 •

2 Including £>afi al-Din*s own risala al-sharafiyya.

 (146)

■1
as dissonant. For these reasons they have not heen included

in the list in the previous chapter, and will be disregarded

in what follows.

gafi al-Din mentions two further octave scales, both

named, but fails to assign them to a particular modal

■2
category. They are

60 mufrayyir frusaynl G- A1> Bb c d et f g

62a nihuf t (al-bijazi) G At? Bt? c d e1? f g.

If the octave awazat can be differentiated from the shudud

by structural features these two cannot, for, like 32 rahawi,

46a bijazi and 5 7 a zankula, they may be divided into non-

3
parallel conjunct tetrachords. The reason for their being

excluded from the shudud (instead of, say, buzurg and zirafkand)

is probably that they were considered to be derived from two

modes already included in that group, and hence, presumably,

were felt to be of lesser importance. Theorists of the

Systematist school, as we have seen, posit a relationship

between 62a nihuft and 44 fcij azi, stating that the former

1 BM. MS. Or. 2361, fol. 104v. The causes of dissonance

according to gafl al-Din are discussed in chapter 4.

2 For Qufb al-Din, mubayyir busayni is an awaz.

3 Or alternatively into disjunct tetrachords. There is no

particular advantage to be gained by preferring this division
in the case of 62a nihuft, but in 60 mubayyir frusayni the
disjunct tetrachords are parallel. The composition in this
mode which concludes the section on music in the durrat al-
taj accords almost equal prominence to fourth and fifth.
 (147)

occurs when the latter is transposed to B1? with G remaining

as tonic:

4-4 bijazj G At? Bt c dt et f g --------- >

Bt c d et f g at bt (c' d' et' f' g' )

Similarly 60 mubayyir busaynx was held to he 48 busaynl

transposed to the 1 7 th frabaqa:

48 ha saynr G At Bb c dt eb f g ---------- >

i----------------------------- »
d etf g at bb c' d' (et' f' g')

The process of derivation in the second case is {just as

artificial as in the first, but nevertheless much closer to

the truth. Bor if 62a nihuft is related rather to 46a bi^azi

1 L ) I■I1 ■ IT I ~ ----

G Bb c dt et f g, occurring when the prominent

fourth of the latter is taken as the tonic, 60 muhayyir busayni

may be derived in exactly the same way from 48 husayni which,

according to Qu£b al-Din, also has a prominent fourth. In

the risala al-sharafiyya these two modes are listed together

1
with the shudud and are stated to be frequently played, but

this does not in itself invalidate the suggestion that they

were considered secondary forms. It may be added that apart

from a single oblique reference to nihuft (called in any

case bija.zi) they receive no mention at all In the kitab

1 D ’Erlanger, ia musique arabe, iii, 13zt-“176.

 (148)

al-adwar.

The above remarks on the non-shudud octave scales present

various reasons why they may not have been considered suitable

for inclusion in that category. If these are accepted, it

would appear that since the structural affinities between

54 buzurg and 5 8 a zirafkand and the remainder of the shudud

appear insufficient justification for their being grouped

together, the inclusion of these two modes should be

attributed to a purely negative cause - the absence of any

non-derived octave scales with a greater degree of affinity.

The modes as described by gafl al-Din could then be allocated

to the categories he established on the following basis:

The shudud comprise all those octave scales in common

use which are divisible into conjunct tetrachords (but

excluding derived forms of these) or into tetrachord and

pentachord where the pentachord is an indivisible unit.

The remaining modes, again with the exception of the

two derived from shudud, form the awaz group.

But in one respect at least this formulation is

misleading. There are grounds for supposing that the

presentation of (5^) buzurg and (57&) zankula as octave scales

is suspect, and that as with 59a kawasht gafi al-Din may have

been less concerned with providing an accurate account of

musical practice than with accomodating the form of these

 (149)

two modes to the prevailing pattern of the other shudud.

This statement, which will be amplified in the following

chapter, implies that even if the sub-groups within the

shudud may still be related to broad distinctions in

structure, the incidence of common features among the shudud

as a whole is reduced still further.

 (150)

chapter 4

Ihus far the discussion has centred round the categories

established by gafi al-DIn, and some Justification has been

sought for the particular distribution of scales within them.

In turning to a consideration of the 18 scales as a set it

may be advisable to begin by following a different line of

analysis, although again one adumbrated in the id,tab al-adwar.

This work contains what is in effect a theory of consonance

in relation to octave scales, expressed at first negatively

in the chapter "On the causes of dissonance" (fi al-asbab

1
al-mujiba li *1-tanafur) and then positively when criteria

for distinguishing consonant from non-consonant scales are

2
set forth. As there is a certain amount of confusion in

the terminology employed, while in addition it is not

immediately clear whether certain remarks refer to octave or

tetrachord, it will be necessary to examine this material

briefly.

The causes of dissonance are stated to be four in number,

1 BM. MS. Or. 156, ff. 8-8v.

2 Ibid., ff. 11-13v.

and are couched in the following terms:

(1) exceeding the limit of the lower tetrachord, c, e.g.

with three whole-tones (1) - G A B elf, or four

intervals of category J - G A“c Bb B db in the

Systematist notation, equivalent to G At? Bb B c$

or d b .

(2) combining the three interval categories J, J and B

within a tetrachord.

(3) placing a category B interval below a category J

interval.

(A) having a category B intei*val follow another category

B interval.

2
In the next chapter gafi a l - D m goes on to say that if these

dissonant combinations of intervals are avoided only 7

tetrachord species are possible; and further that only 9

pentachord species are possible if causes (1) and (2) are

3
avoided with respect to the section c - f of the pentachord

c - g, while otherwise 13 pentachord species are permissible,

(gafi al-Din lists the 7 tetrachords and 12 of the 13

pentachords, and it is from these that the theoretical corpus

1 Bor gafi al-Din, as we have seen, J has two values (minor

whole-tone and apotome), but in this context the distinction
is of no importance. Nor is it with B (limma or comma).

2 BM. MS. Or. 136, ff. 8v-10v.

3 I.e. the lower tetrachord-miles of (1) are applied to

the upper tetrachord (c - f).
 (152)

1
of 84- octave scales is compounded.) It would appear from

this that the four causes do not necessarily always incur

dissonance, and that they were formulated primarily as a

negative aid towards the definition of permissible tetrachords,

2
i.e. those occurring in practice. But apart from this

function y>af1 al-Din evidently considered them to be relevant

5
to the octave, for elsewhere he states that the dissonance

of any octave scale is the result of its containing (at least)

one of the combinations of intervals they refer to. The

implication is that any one of the four will suffice to render

the scale dissonant, but this is not consistent with the

admission that a number of pentachords fail to avoid causes

(1) and (2). In fact it would appear that of the combinations

mentioned (5) B J and (4-) B B were thought dissonant

wherever they occurred, as was (1) J J J J, despite the

fact that (4) is concerned ostensibly with the lower tetrachord

only, gafl al-DIn dismisses G A Bt c dt eb f g as

dissonant because the sequence J J J J occurs between A

Zj.
and eb, and G A1? Bb B c d1? eb e f g because (5) B J
■5
is included between B and d v. There is unfortunately no case

1 All 19 species will be found listed in chapter 2.

2 One may contrastthe risala al-sharafiyya, in which a

whole armoury of chromatic and enharmonic species is paraded*

3 BM. MS. Or. 136, fol. 11v.

4- Ibid., fol. 13vo

5 Ibid., fol. 11v •

in gafx al-DIn’s works in which a scale is rejected because

of the occurrence of (4-) B B, but when the 84- octave scales

are listed in the shark all those containing this particular

1
sequence are classed as dissonant. Of the remaining two

combinations of Intervals alluded to (1) 3? 3? J is evidently

dissonant not per se, since it is found in the consonant 4-1

busalxk Ct Ab Bb c db eb f ;g, but only when it results

in the omission of c; similarly the juxtaposition of J J B

(whatever the order) referred to in (2) is impermissible

within the tetrachord G - c but not necessarily so elsewhere,

for although it may not occur within the tetrachords c - f

and d - g it may within the pentachord c - g if f is omitted.

Like (1) T T 3? it may also appear in an octave scale partly

in one tetrachord and partly in another without incurring

dissonance.

It is apparent then that these combinations of intervals

are only partially relevant to octave scales. In effect,

since they are by definition excluded from the tetrachord

species recognized by £>afx al-Dxn, and with two exceptions

(J 3? f and 3- 4 B) from the pentachords, they can only

occur among the 84 octave scales in strictly definable

circumstances, so that we may state quite simply that with

reference to the 84 those scales are dissonant which contain

BM. MS. Or. 2361, ff. 102-105*

 (1540

1
the sequences B B, B <J, or J J J J.

The species from which the 84 octave scales are derived

are all consonant (or at least non-dissonant). The octave

scales however are considered to fall into three categories

which we may term consonant, intermediate, and dissonant.

Here, unfortunately, gafi al-DIn uses two sets of terms and,

what is worse, two sets of definitions. However, in hoth

cases the degree of consonance depends in the main upon the

number of consonant intervals (octave, fifth, and fourth) the

scale contains* In determining this number all possible

fourths and fifths are counted, including those which might

be considered inversions, e*g. c - g as well as G - c ; d - g

as well as G - d. The notes G c f g are termed fixed

(thabit), although §afi al-DIn recognizes that f, unlike the

others, is not found in all the 84 scales.

The first set of categories consists of

(a) consonant (mula?im) : scales containing as many

consonant intervals as notes (octave Included)

(b) dissonant (mutanafir) : scales containing one of the

above-mentioned causes of dissonance

1 In relation to practice one ought perhaps to draw

attention once more to the ambiguity of the designations B
and J. In certain contexts two successive semitones,
generally written J B, could also be given as B J (and in
at least one case were by Qu£b al-DIn), but remained
consonant nevertheless.
 (c) intermediate (khafi al-tanafur): scales containing
1
fewer consonant intervals than notes.

Stated in this way (b) and (c) are not necessarily exclusive

of one another, and it would be simpler to define (c) as

comprising all scales not included in (a) and (b). The

parallelism required for a scale to belong to (a) precludes

the possibility of it containing one of the dissonant

2
combinations referred to in (b). Curiously enough gafi al-

Din almost immediately proceeds to disregard the dividing line

between (a) and (c), for he presents as consonant (mula*im)

a number of parallel conjunct tetrachord scales one of which

3
has fewer consonant intervals than notes:

ir
G At? Bb c d1? eb
i______________ i
i______________ i
L___________________

The second set consists of

(b' ) dissonant (giahir al-tanafur) : scales with consonant

intervals occurring between the fixed notes only

1 BM. MB. Or. 136, ff. 11-11v.

2 The one exception to this rule is provided by the conjunct

repetition of 11 Isfahan, as in G At? Bb B c d1? eb e
f g referred to above, where B J occurs between B and dt?.

3 Ibid., fol. 12.

 (o' ) intermediate (mutala/im) : scales with more consonant

intervals than those in (b') (but with less than

those in)
1
(a') consonant (kamil fi al-tala*um) : as (a) above.

Ignoring the obvious disparity between the terms used in (c)

and (c') the two sets can be reconciled easily enough,

producing a scheme whereby scales are considered consonant

if they have as many consonant intervals as notes; dissonant

if they contain one of the sequences B B, B J, and J J J J

and/or if they have no consonances apart from those between

2
the fixed notes G c f g; and otherwise intermediate.

Hence 40 4ushshaq

ir
G A B c d e
i________________ 11_____

with 9 consonant intervals to 8 notes, is consonant, while

f-------------------- \
f---------------it---------------1
I----------------------------------1
G At B1? c db el? f g

with consonances between the fixed notes only, and

G At? Bl? B c d1? eb f

i__________________ i
j
J

1 Ibid.

2 Of. al-Ladhiqi, BM. MS. Or. 6629, ff. 52-52v.

which contains 7 consonant intervals to 9 notes but also the

sequence B J between B and d1?, are dissonant. An example of

$
an intermediate scale is 32 rahawi

ir

G At Bt c dt eb £ g
1 ___ 1

with 5 consonant intervals to 8 notes.

The only obscure point in this scheme arises from the fact

that one of the fixed notes, f, is not always present. A scale

which has consonances only between G, c, and g is evidently

dissonant, but $afi al-Din says nothing about the status of,

say,
i 1}------------------------ 1
} 1
G A Bb c d? eb et ft g
! 1

in which the total number of consonances, although no more

than the 4- produced by the full complement of fixed notes,

includes one not between these. However, this scale is

1
stated to be dissonant in the sharh, and it may therefore

be concluded that a scale must always have more than 4*

2
consonant intervals in order to avoid dissonance.

1 BM. MS. Or. 2361, fol. 103v.

2 f is omitted in three pentachords; 2 5 a c <dt eb e1? ft

g, 26 c c$ d e ft g, and 27a c d et ft g. This
particular situation can only arise in scales containing
2 5 a, since the presence of d in the other two automatically
supplies two consonances.
 (158)

We have seen that by terming consonant a scale with fewer

consonances than notes gafi al-Din does not adhere strictly

to definition (a). After giving the second set of definitions

he further obscures the picture by stating that the intermediate

(khafl al-tanafur) scales among the 84 are those containing

not more than 5 consonances (those with 4 remaining dissonant

1
(%ahir al-tanafur)). This of course implies that a scale

with 6 should be in the consonant category, a definition

markedly different from (a). Thus it would appear that

despite the terminology of the first set the essential

distinction is that maintained between on the one hand

consonant and intermediate scales and on the other dissonant

scales. Although the further distinction between consonant

and intermediate is no doubt a useful one, gafi al-Din5s

hesitation as to where the dividing line should be may be

taken as an indication of the subjective element involved.

The upper limit of the dissonant category can be established

on the basis of the scales recognized by him, for given that

no frequently performed mode can be considered dissonant the

fact that 52 rahawi G At Bt c dt e\> f g has only 5

consonant intervals precludes the possibility of scales with

2
more than 4 such intervals being dissonant. However, in

1 BM. MS. Or. 136, fol. 13v.

2 Assuming that no dissonant sequence (B B, B J, J J J

J) was present.

It should perhaps be pointed out that the consonance

categories were not formulated in order to distinguish the
scales occurring in practice from the remainder of the 84.
 V W )

spite of the wide gulf between 32 rahawi and 40 4ushshaq

G - A B c d e f g (with 9 consonant intervals), there is

no similar objective basis for separate classification. The

shark for instance attacks definition (a) as being too

rigorous, and suggests that a scale with one or two fewer

1
consonant intervals than notes should be considered consonant.

As the boundary between consonant and intermediate is a matter

of convenience and nothing more the proposal is an eminently

sensible one, for it divides the corpus of non-dissonant

scales into nearly equal parts. A definition of the two

categories might then be: an 8-note scale (octave included)

is intermediate if it has 5 consonances, consonant if it has

2
more; a 9-note scale is intermediate if it has 5 or 6

consonances, consonant if it has more.

According to these criteria 12 of the octaye scales

mentioned by gafx al-Din as occurring in practice are

consonant and 5 ( 5 2 rahawi, 5^ buzurg, 57si zankula, 58si

5
zirafkand and 59a kawasht) are intermediate. Although it

would be possible to make some tentative comparisons between

this division and the structural alignments suggested in the

discussion of the modal groups, it is essentially nothing more

1 BM. MS. Or. 2361, fol. 99.

2 For 8-note scales this scheme therefore corresponds to

gafi al-Din5s last statement, given above.

3 Assuming for present purposes that zankula and kawasht

are octave scales. In evaluating the grading of the latter
the interval B1? - ft is not considered consonant, for reasons
that will be given below.
 (160)

than an empty statistic, gafi al-Din may have sensed that

this kind of difficulty might arise, for he makes virtually
no use of the categories except to illustrate the proposition
that dissonant scales are not generally found in practice.

It might however he possible to utilize these categories

in an attempt to determine why it is the 17 scales mentioned
by gafi al-Din that occur and not others. For this purpose
the above formulations are imprecise and in one important
respect misleading: the fact that both G - d and d - g are
counted distorts the evaluation of a scale containing d when
compared with one that does not. In effect, any 8-note scale
containing G, c, d, f and g will have 6 consonant Intervals
and should therefore be consonant irrespective of what the
other notes are. Thus if one compares
I---------------- 1
< ji------------*
i------------------------------- i
G A1> B1? c & e f g
1 11 1

with
j------------ ----- 1
\ 11 *
f--------------------------------- 1
G A B1? c at eb f g
I_______________ I

it will be seen that although the latter has only 5 consonant

intervals, it contains two notes, A and dt, that do not stand
in a consonant relationship with one or more other notes while
the former contains three, At, Bt, and e. We shall term these
unattached notes. Thus in certain cases gafi al-DIn5s method
of evaluation may judge a scale with a fewer number of
 (161)

unattached notes than another to he the less consonant. In

other words the anomaly may arise of the scale with the

greater degree of parallelism between the upper and lower

tetrachords being considered the less consonant. It is rather

surprising, in view of the obviously normative nature of

conjunct tetrachord structure, implicit in the way scales are

analysed and presented by theorists of the Systematist school,

that none of them should have remarked on this weakness.

Nevertheless it would appear to have been recognized, even if

only tacitly, for in the sharfr- the scale

j-----------------
,--------------- , i----------------- , 1
* 1
G A B c d eb f g,
i-------------- ri_____________
i— ---------- 1
\
its 7 consonant intervals notwithstanding, is classed as
1
intermediate. The only reason that can be suggested for this

is that the presence of the two unattached notes B and eb was

felt to detract from the consonance of the scale. This, it

may be added, is by no means an isolated case.

It will be apparent that a more accurate estimate of the

comparative degrees of consonance of various scales may be

achieved by combining a count of the unattached notes with

that of the consonant intervals. With regard to the latter

it will be necessary not to include both G - d and d - g;

and for the sake of simplicity we may at the same time ignore

1 BM. MS. Or. 2J61, fol. 102.

 (162)

the ever-present octave note and one of the pair G - c and

c - g. The scales given above, which we shall call 7-^ote
rather than 8-note scales, may therefore be said to have:
, ,

\ 1
------------ 11------------- 1
i
G A B c d e l ? T g

4 consonant intervals and 2 unattached notes (4 : 2);

t------------- 1
( — il " i
G A B1? c d»1IM? e1? f g
—

3 consonant intervals and 2 unattached notes (3 : 2);

I-------------------1
}---------— -----1f--------- 1
G At Bp cd e f g

3 consonant intervalsand 3 unattached notes (3 : 3)- Hence

the first scale may be considered the most consonant end the
third the least consonant* These examples also show that a
7-note scale with two unattached notes will have partial
parallelism, i.e. one note will be different in each tetrachord,
while a scale with three (or more) unattached notes will be
devoid of parallelism.

But used in isolation the consonance classification as

it now stands is still of little help in determining why
these particular 17 scales should be popular. If 32 rahawi
(3 :2) represents the lowest level of consonance acceptable
in a 7-note scale, it will be found that there are among the
 (163)

84 octave scales a further 18 with an equal or better

consonance ratio. As for 8-note scales, even if we discard

5^ buzurg (3 : 3) on the grounds that it is an artificial

structure and take as the lowest permissible ratio that of

58a zlrafkand (4 : 1), there are still among the 84 another

5 scales with an acceptable degree of consonance. Hence the

potential scales number 23 as against the 17 in existence.

This amended classification possesses however the advantage

of directing attention to a factor of crucial importance,

namely that the pattern of interrelated notes is as significant

as their number. This brings us back once again to a

consideration of the way tetrachords and pentachords ax^e

deployed within existing scales. To say that a scale has no

unattached notes, and to say that it has parallel conjunct

(or disjunct) teti*achords, are no doubt two ways of expressing

the same phenomenon, and in dealing with a pui'ely diatonic

modal system more than one approach might be superfluous.

But when considering the Arabian and Persian system of this

period the difference of approach represents more than just

a change of perspective, for the tetrachords and pentachords

are not vague entities abstracted from the octave scales,

but constitute in themselves the basic units or melodic cells

which may exist independently of the compounds in which they

are generally found. Thus in one sense it would be legitimate

to consider the consonant intervals present in the octave

scales to be secondary phenomena resulting from the

 (164)

juxtaposition of these units. In addition, then, to positing

a number of potential scales on the basis of relative degrees

of consonance, we should reexamine the way units are combined

within existing scales in order to detect any patterns of

distribution which might constitute further criteria of

acceptability.

In accordance with Qu£b al-Din’s notation all the units

listed in chapter 2 were transcribed from G. gafl al-DIn,

however, generally places tetrachords on G and pentachords on

c, and in addition some tetrachords may bethought tobegin

on d. It may therefore be clearer toadopt asymbolism which

remains constant whatever the point of departure, and for e.g.

/G A B1? c/ , / c d e? f/, and /d e ft g/ we shall

substitute /1 2 3 V\ the numbers (other than 1) relating

to the major intervals with the distribution of tones and

semitones between them being in consequence that found in the

c mode.

As has been seen, gafl al-Din isolates 19 units: 7

tetrachords and 12 pentachords. Of the latter 7 may

considered extensions of the tetrachords, adding an undivided

whole-tone /T/ above, so that we may symbolize

c d e f g a s / 1 2 3 Vl/
c d eb f gas/1 2 3b 4/T/

c db eb f gas/1 2b 3b 4/T/

c d et f g as /1 2 3"b 4/T/
 (165)

c dt e1> f 8> as /'\ 2^ 3"b 4/T/

c dt? eb f s as /1 2t 3b 4/T/ and

c d*b eb e f S as /1 2t 3b 3 4/T /

A further 4 can be similarly related to the tetrachords,

but with the whole-tone either below

(c d e1? ft g, symbolized as /T/1 2t? 3"b4-A and

c d e1? f ft g, symbolized as /T/1 21? 3b 3 1-/);

divided into semitones (/SS/)

1
(c d£ e1? f ft g, symbolized as /1 2t 3^ 4/SS/);

or both

(c c# d e ft g, symbolized as /SS/1 2 3^ 4-/)*

The remaining pentachord, c d1? eb et ft g, cannot be

so related. Thus if we ignore the disjunctive whole-tone the
19 units identified by gafi al-Din can be reduced for purposes
of analysis to the following 8:

5/1 2 3 V 6 /1 2b 3b V
7 /i 2 3b 4/ 8 / 1 2 3^ 4/

9 /'\ 2V 3b 4/ 10 /1 2"b 3t> 4/

11 /1 2t 3b 3 4/ 25a /1 2^ 3b 3^ 4* 5/

1 c d et f £# g could, alternatively, be symbolized

as /1 2 3"b 4/SS/, a division which has little to
recommend it but is sanctioned by gafi al-DIn’s own
approach.
 nbb;

In one respect this is however an oversimplification: what

gafi al-Din notates as c d“c e“G f g“c g may, as in

45 4iraq, have the expected value c dt et f ft g (and

hence be reducible to /1 2t 3t 4/SS/), but it can also

represent 20 buzurg G At B(t) c oil

d, which must be
1
considered an independent and indivisible unit* To the

above 8 we shall therefore add /1 2t 3 4 4 if 5A omitting

the fictitious t on B (3)*

Octave scales containing both d and f may be represented

either as conjunct (/tetrachord/tetrachord/T/) or as disjunct

(/tetrachord/T/tetrachord/). Since gafi al-Din lays evident

stress on the conjunct pattern both possibilities will be

listed only when the conjunct tetrachords are not parallel*

Hence the 17 octave scales assume the following form:

32 rahawi G At Bp c dt eb f g /1 2t 3^ 4/1 2t 3 b 4/T/

40 4ushshaq G A B c d e f g /12 3 4/1 2 3 4/T/

41 busallk G Ab Bb c db eb fg /1 2b 3 b 4/1 2b 3 b 4/T/

42 nawa G A. Bb c d eb f g /1 2 3b 4/1 2 3b 4/T/

43 rast G A Bf c d et f g /1 2 3t 4/1 2 3t 4/T/

44 faijazi G At Bt c dt et fg /1 2t 3$ 4/1 2t 3 ^ 4/T/

45 4iraq G At Bt c dt et f f# g /1 2t 3t 4/1 2t 3$ 4/SS/

1 We may remember that the analysis of 54 buzurg into 10

4iraq G At Bt c and 26 kardaniya c cf d e fl g, given in the
kitab al-adwar (BM. MS. Or. 136," fol. 16v), is apparently
replaced in the risala al-sharafiyya (D’Erlanger, la musique
arabe, iii, 136) by a division into 20 buzurg and 8 rast ■
There Is no other example of a pentachord/tetrachord
division in gafi al-Din’s treatises.
 (167)

St-
G At Bb c /1 2t 3 b4/1 2t 3t 4/T/

CD
46a biiiazr f g
48 busayni G At Bb c dt eb f g /1 2t 3 b4/1 2t 3 b 4/T/
50 kardaniya G A Bt c cfi d e ft S /1 2 3t 4/SS/1 2 3t 4/
54 buzurg G A? B(t) c cl! d e ft g /1 2t 3 4 4# 5/1 2 3t 4/

57a zankula G A Bt c dt et f g /1 2 3t 4/1 2t 3% 4/T/1

58 a zirafkand G At Bb c dt eb et ft g /1 2t 3^ 4/1 2t 3b3t 4$ 5/
59a kawasht G At Bt c dt eb et ft g /1 2t 3t4/1 2t 3b3t 4$ 5/
60 mufcayyir G At Bb c d et f S /1 2t 3 b 4/1 2 3t 4/T/
busayni
or /1 2t 3 b 4/T/1 2t 3 b 4/
62a nihuft G At Bt c d et f g /1 2t 3t 4/1 2 3P 4/T/
or /1 2t 3t 4/T/1 2t 3b 4/

66 igfahan G A Bt c d et f f If g /1 2 3"b 4/1 2 3P 4/SS/

2
or /1 2 3P 4/T/1 2t 3b 3 4/

Taking the alternatives into account it will be seen from

this table that

/1 2 3 4/, /1 2 3 b 4-/? and / 1 2 b 3 b 4/ do not combine with

any other unit 5

/1 2 3^ V combineswith /12t 3 t 4/, /1 2t 3b 4-/,

/1 2t 3b 3 4/, and /1 2t 3 4- 4# 5/;

1 Or G A Bt c dt et f fit g /1 2 3t 4/1 2 t 3t 4/SS/.

2 Here the disjunct version is included because it includes

11 igfahan /1 2? 3b 3 4/. These two divisions and the one
proposed in chapter 2, /T / 1 2t 3b 4/1 2t 3b 3 4/, receive
further discussion below.
 (168)

/1 2t 3t 4/ combines with /1 2 3^ 4/, /1 2t 3 b 4/, and

/1 2t 3b 3t 44 5/;

/1 2t 3b 4/ combines with /1 2 3”b 4/, /1 2t 3"b 4/, and

/1 2t 3b 31? 4t 5/;

/1 2t 3b 3 4/ combines with /1 2 3^ 4-/;

/1 2t 3b 3t 44 5/ combines with /1 2t 3t 4-/ and /1 2t 3b 4-/;

and /1 2t 3 4 48 5/ combines with /1 2 3t 4/.

Of the units which combine with others the two pentachords

must do so within an octave scale. Hence the four Zalzalian

tetrachords are the only units used in combination for which

this is not a necessary condition. The scales made up of

three of them, /1 2 3^ 4/, /1 21? 3"i? 4/, and /1 2t 3b 4/, may

be related in diagrammatic form thus:

 hi up;

3P 4/1 2 3b 4/
/1 2b 3 P 4/'f/i 2b 3b 4/- 3b 4/1 2 3b h/t /

/1 2b 3b 4/T/1 2b 3 b 4/

/1 2b 3 b 4/1 2b 3b 4/22/
Z lJ ? b _ 3jLJi;/i_._2jL3b 4 / 1 / .

(Broken lines link alternative representations of the same scale.)

As it combines v/itli only one other unit, the position of /I 2b 3b 3 1

cannot be similarly fixed. By assigning an arbitra.ry position to this in
and the two pentachords, and assuming a mirror symmetry between the diat<
units and the three Zadzolian units shown above, this basic scheme may b<
expanded to include all 17 octave scales:
 1/1 2 5b 1-/1 2 3b y 5 7

(170)
I7 "'
2 b 2b 1-/1 2.9 3b 4/2/1 •■-/I 2 3 1-/1 2 3 4 / 1 /

il
!/1 22 3b V I 22 3 b V'2/jj ! / 2 32 4 / 1 2 3? 4/2/

l!/1 22 3b 4/T/1 22 3 b 4/j -j/1 21? 3 b 4 / 1 2 52 4/1/ i/i 2 32 V 1 2 3'b 4 / S S /

A
|/i 2 31? V B S / 1 2 32 V

/I 2 51? 4/2/1 21? 3 b 3 4/

v
/I 2? 3 b V 1 21? 5 b 5 p V _ 5 /

j/1 21? 3 b 5 4 /j|

-p-
CVI

/1 22 31/ V i 3? 4/2/

/1 22 32 V r /1 22 5 1? 4 / /1 21? 33 V i 2 51? 4/S/

/I 22 32 V i 23 3b 4/2/ /I 2 31? 1-/1 21? 32 1-/2/

1/1 22 32 V 1 22 3 b 32 V 5/
 (171)

(With regard to 59a. kawasht the alternative form suggested

above, G At Bt c dt el? e t , can be entered at the same point

on the diagram as the octave form G At Bt c dt eb et ft g if

we accept that /1 2t 3b 3"b At 5/ is closely related to, if not

an extension of, /1 2t 3b 3^ A and that the two couldtherefor

be placed in the same box.)

Clearly, such a diagram is to a certain extent

impressionistic. Bor instance, everything placed within the

triangle formed from /1 2 3t A/, /1 2t 3"b A/ and /1 2t 3b A/

could with equal validity be placed outside it. But the

difference is no more than that between describing these

three units as a nucleus or as a frame, and is equally

unimportant since what i’s shown here is primarily a

distributional relationship, not an organic one (in which case

the difference between the arrangements - and the metaphors -

would be significant).

We have seen that by taking the consonance levels 3 : 2

and A : 1 as the lowest admissible for 7“ and 8-note octave

scales respectively there are a further 2A possibles among

the 8A. Of these only four could be included in the diagram

as it stands: all the others would require additional boxes

and lines linking them. But to disregard them because of

this would be tantamount to saying that the diagram, which is

meant to do no more than compress a number of descriptive

statements, constitutes a definition of the limits beyond

which combinations of units must be rejected, and on a purely

synchronic level this is evidently unjustified. However, by
considering the modal system as described by gafi al-DIn in
historical perspective it becomes possible to regard certain
features of the diagram as virtually immutable, while others
may be liable to alteration and expansion.

The most obviously normative of these features is the

strict segregation of diatonic and Zalzalian units. It will
be remembered that gafx al-Din assigns the diatonic modes
to a separate sub-class within the shudud with its own
emotional connotations. In later texts the distinction is
maintained, and although mention is made of one or two scales
combining diatonic and Zalzalian units such juxtapositions
appear to have been of an exceptional and transitory nature:
from the 13th to the 16th century (and perhaps beyond) the
two types formed quite distinct groups, coexisting but not
1
interpenetrating. Thus we may with some confidence reject
as potential additions all scales in which both occur.

The 2A possibles are thereby reduced to 13* Bive of the

1 One of the most striking features of the diatonic group

is the fact that it contains only three scales, giving the
Zalzalian group an overwhelming numerical superiority.
Diatonicism, the foundation of the early Arabian modal
system of the 8th and 9th centuries, is now, four hundred
years later, apparently of only marginal importance. One
can do no more than guess at the reasons for this drastic
shift in emphasis, for in the 11th century diatonic and
Zalzalian units seem to have been fairly evenly balanced,
so that the shift presumably occurred for the most part
during the 12th century, a period for which information on
such matters is sadly lacking.
 (172)

13 are diatonic, and with regard to these it should be

observed that because of the larger number of whole-tones

they contain, diatonic combinations have, other things being

equal, a higher consonance ratio than Zalzalian combinations.

That of the three diatonic scales listed by gafl al-Din

ushshaq, 4-1 busallk, 4*2 nawa), 6 : 0, is found in only

two of the five:

G A Bb c d e f g, and G Ab Bb c d eb f g.

The others all have the much inferior ratio A : 2, and may for

this reason be discounted, thus bringing the total down to 10.

This number is made up of the above two diatonic scales,

three 7-note and five 8-note Zalzalian scales. Three of the

8-note scales,

G At B\> c d et f £# g

G A? Bt c dt et f ft g and.

G At B1? c d e1> f ft g,

differ from 60 mufrayyir frusayni, A6a bijazi , and 62a nihuft

respectively only by virtue of including f$, i.e. having a

divided upper whole-tone. Previously, when considering the

relationship between the tetrachords and pentachords described

by gafl al-Din, no particular attention was drawn to the fact

that some pentachords contain /SS/ while others have /T/.

There are however two pairs of pentachords in which this is

the only distinguishing feature:

 (1740

/1 2 3t 4/T/ and /'\ 2 3t 4/SS/;

/1 2t 4/T/ and /-\ 2t 3t 4/SS/.

Correspondingly, a contrast may be made between the following:

A3 rast G A B1? c d e1? f g and

66 igf ah an G A Bt c d et f fit g;
AA bijazi G At Bt c dt et f g and

A 5 6iraq G At Bt c dt et f ft g.

gafi al-Din5s analysis of 66 igfahan has already been

commented on briefly, and this mode will be further discussed

below. We may recall that the upper pentachord may be viewed

either as /1 23t A/SS/ or as /T/1 2tjb 3 A/. In all

likelihood emphasis would be placed in performance on the

tetrachord 11 igfahan /1 2t pb 3 A/ (d - g), thus

providing a modal characteristic sufficient to distinguish

66 igf ah an from A3 rast with the fit added as an ornament. In

other words it may be suggested that the presence or absence
of the ft was not in itself a crucial differentiating factor.

The same may well be true with regard to the other pair. Bor

Qutb al-Bin 6iraq is AA, of which A 5 is no more than a

1
variant. It is in fact conceivable (the point will be argued

in the following chapter) that gafi al-Din included the non-

critical ft in A3 *iraq simply in order to distinguish the

two in notation, whereas in practice they were identified by

1 Later texts also list *iraq both with and without the ft.
 (175)

other features which his methods of scale analysis either

ignored or distorted. More positively, it may be pointed out

that gafi al-Dln himself provides variant forms of 573- zankula,

4
one with the fit and one without. Hence it seems reasonable

to regard these three 8-note scales as implicit in already

existing scales, or as non-significant variants of them,

especially when the one corresponding to 60 mubayyir frusayni

does in fact appear in the durrat al-taj, where it is not

2
considered to be anything but that mode.

We are thus left in the final analysis with 7 possibles

out of the 24 originally postulated. They are:

G A Bb c d e f S
G Ab Bb c d eb f S
G A Bt c d el? ft s

1 We may also note that the shark (BM. MS. Or. 2361, fol.
105v) calls G Ab Bb c db eb f" f i g a shutb a , derived form,
^'1 husalik G. Ab Bb c. db eb f g. It would appear from
this that f?T~was often an optional extra which could be
added, presumably as a grace-note, without affecting the
identity of the mode in question. Apart from the two pairs
of modes discussed here, the addition of this note would
nowhere cause ambiguity.

2 It appears in a composition stated to be by §afi al-Din.

As a counter-argument it could be argued that in this
scale, as in G At B^ c d et f ft g, the upper pentachord
could be analysed as /T/1 21? 3b 3 4/, again drawing
attention to 11 igfahan. But this does not in itself
disprove the thesis that they were non-significant variants,
and it is significant in this respect that in the composition
referred to the fit is added in a passage stressing the
tetrachord d - g, but while a modulation into 11 igfahan
is noted elsewhere, this is not stated to be one.
 (176)

G At Bt c d et ft S
G At Bb c d et ft g
G At Bb c at eb e f S
G At Bb B c d et f

All appear to share features of consonance and unit

compatibility observable in existing modes. But there is

one interesting difference: with the sole exception of

G Ab Bb c dt eb e f g they are all either necessarily

disjunct, omitting f, or preferably so, having parallel

1
disjunct tetrachords* Now we have seen that of the 17

octave scales recognized by gafi al-Din only four may be

classed as disjunct:

5° kardaniya /<\ 2 3^ 4/SS/1 2 3$ V

&6 Isfahan /1 2 3^ 4/T/1 2t 3^ 3 V

60 muhay.yir husayni /1 2t 4-/T/1 2t 3^ 4-/

62a nihuft /'\ 3^ VT/1 2t> 3^ V

and it has been suggested that the latter two were derived

from conjunct modes in which the prominent fourth tended to

usurp the role of tonic. In view of this it would seem

reasonable to regard at least some of the 7 possible scales

as similar potential derivations, especially when we remember

that according to Qu'£b al-DIn a large number of the then

existing modes were characterized (at least in part) by the

1 Ignoring the extra note in the lower tetrachord of

G Ab Bb B c d et f g. G Ab Bt> c d et ft g both omits f and
has parallel disjunct tetrachords.
 (177)

prominence of the fourth. They would thus he further

manifestations of a process of historical development which

had already resulted in the emergence as independent modes

of 60 muhayyir husayni and 62a nihuft.

We may first of all establish the following correlations:

(48 fr-usayni /1 2b 3b 4/1 2b 3b 4/1/-

/1 2b 3b 4/T/1 2b 3b 4/ 60
46a fri.jazl /1 2b 3b 4/1 2b 3b 4/1/-

/1 2b 3b 4/1/1 2b 3b 4/ 62a)

41 busalxk /1 2b 3b 4/1 2b 3b 4/1/-

/1 2i> 3b 4/1/1 2b 3b 4/
42 nawa /1 2 3b 4/1 2 3b 4/1/ —

/1 2 3b 4/1/1 2 3b V
/1 2b 3b 4/1 2b 3b 4/1/'

/1 2b 3b 4/1/1 2b 3b 4/
32 rahawi /1 2b 31 4/1 2b 3b 4/1/-

/1 2b 3b 4/1/1 2b 3b 4/

Here four of the possibles (first, second, fourth and fifth)

are related to modes listed by gafi al-DTn, and if we accept

the thesis of modal development adumbrated above we should

expect to find these scales cited in later treatises. In

fact only two of them, those relating to 41 busallk and

^ fcitiuzl (G Ab Bb c d et f g and G At Bt c d e1?

ft g), are attested; the non-appearance of the other two,

however, can hardly be considered a refutation, for the same

texts reveal that the modes to which they relate, 42 nawa

and 32 rahawl, failed to survive in the form described by

gafi al-Din. 32 rahawi and its derivate (G A1? Bl? c d et

ft g) have the lowest consonance ratio of all the 7 “*10'ke

scales he recognizes, and it would appear that this was

beginning to be felt insufficient: it is exceeded by all the

octave scales known to have been in existence during the 14th

and 15th centuries. The reason for the subsequent changes in

the structure of 42 nawa, and hence for the non-appearance of

its disjunct correlate (G A Bb c d e f g) is not clear:

from later treatises we can however see that 7 /1 2 3^ 4/

was the least used of the basic units.

gafi al-Din mentions a further three conjunct scales from

1
which similar derivations may be made:

4° 4ushshaq /1 2 3 4/1 2 3 W --------->

/1 2 3 4/T/1 2 3 4/
43 rast /1 2 3^ 4/1 2 3*t?--- 4/1/------- »

/1 2 3^ 4/1/1 2 3t 4/

57^ zankula /1 2 3^ 4/1 3I? 4/1/------ >

/1 2$ 3^ 4/1/1 2 4/

The upper pentachords of the derivates are not recognized by

gafi al-Din, and therefore these scales do not form part of

his theoretical corpus of 84 scales, from which the 7 possibles

1 That is if we consider, for present purposes, 4 5 4iraq

to coincide with 44 b-iqazi.
are taken. Nevertheless, they conform to the criteria

governing the final selection of these, and two of them occur

later. The third again has the minimum consonance ratio

3 : 2, as does the mode to which it relates, 57a- zankula;

its non-appearance is thus no more surprising than the

1
changes that subsequently affect zankula itself. This

evidence seems to confirm that under given conditions relating

to consonance new modes were being added in the way suggested:

14th and 15th century texts attest the existence of all but

one of the scales derived from the conjunct tetrachord modes

known to gafx al-Din which have a consonance ratio in excess

of 3 : 2.

The scales discussed above are made up of the following

units:

5 / 1 2 3 4/ 6 /'\ 2b 3b 4/ 7 / 1 2 3b 4/

8 /1 2 3b v 9 /1 3b 4/ 10 /1 2b 3-b 4/

In the previous diagram the distributional relationships

obtaining between these units are presented synchronically,

so that the historical development postulated here is not

apparent, even with reference to 60 mukayyir busayni and

62a nihuft. In order to show it we may change the format to

a three-dimensional one:

1 In fact, as has been stated above, gafx al-DIn’s notation

of zankula may well be misleading. The structure of this
mode will be further discussed below.
 03

C o n ju n c t d i a t o n i c

s c a le s lis te d by

/1 2 5 4 / 1 2 3 4/ §> afi a l - D i n

D is ju n c t d ia t o n ic

s c a le s lis te d by
[/iTb 5 b 4/-J/1 2 b 3 b 4/j- •|/1 2 3 4 / 1 / 1 2 5 4 /1 la te r th e o r is ts

1/1 2? 3fr 4/1 2i? 5b4/T-/l (b )

C o n ju n ct Z a lz a lia n

/1 21? 3i> 4/ 1 2? 31? V ® / /1 2 31? 4/ 1 21? 31? 4 / 1 / s c a le s lis te d by

/1 21? 3? 4 / 1 2 31? 4 / 1 / £> afi a l - D i n

pi 21? 31? 4 /1 21? 31? V ’S S /j

|/1 2? 31? 4 / 1 21? 35 a / T /

l/l 2? 3 b 4/1/1 21? 3 b 4/S D is ju n c t Z a lz a lia n

s c a le s lis te d by

§ a f i a l-D in

/1 21? 31? 4 / 2 / 1 21? 3? 4 /

if1 2 b 3l> 4/D/I 2? 3? 4/ij-

D is ju n c t z

scales listed by

2 31? 4/1/1 21? 31? 4/1 0crists

1 -The relationshi o e Jv/een

50 k a r d a n v a and 65
will be discussed be
The modes represented on the upper planes of (a) and (b) are

not, with one possible exception, supplanted by those they

engender, but continue to exist alongside them.

If we wish to consider this form of modal development In

the wider context of the whole repertoire of octave scales

listed by gafi al-DIn as occurring in practice, plus the 7

possibles, in order to see if it has any further validity, we

must examine the following three questions:

(1) Can the remaining three possible scales be related

to existing modes in the same, or a similar, way, and are

they attested later?

(2) How can we explain the presence of two apparently

non-derived disjunct modes (5 0 kardaniya and 66 igfahan) in

gafi al-Din9s list?

(5) Did this development also result in the derivation

of modes from those (5zl- buzurg, 58a zlrafkand, and 59a kawasht)

to which the conjunct tetrachord pattern does not apply?

(1)
One of the three, G A c d e1? ft g, may be arrived

at by this process if we assume that 62a nihuft tended to

be assimilated to the prevailing conjunct pattern. We would

then have:

62a nihuft /1 2t 3^ V l / 1 2t 3b 4/ becomes

» /1 3t 4/1 2 3t 4/T/— »/1 2 3t 4/T/1 2t 3? 4/

G A Bt c d et ft g
 (,182;

(This scale would thus result from a second cycle of change

which in this case, however, would be contemporary with

several instances of the first, since 62a nihuft is itself

one of the earliest disjunct derivates to appear. G- A B1?

c d e1? ft g is cited in one later treatise, the sharjg.,

1
where it is called m ajlis afruz.

The remaining two may be related not to conjunct scales

listed by gafi al-Din but to each other:

/1 21? 3b 4/1 21? 3 1? 3 4 / 1 / ---------- >

/1 2$ 3b 3 4/T/1 21? 3b 4/

However, it may be noted that both contain 11 igfahan

/1 2*& 3b 3 4/? and they may be best discussed in conjunction

with the one mode in gafi al-Din5s list to contain this unit,

66 ifffahan.

(1) and (2)

If we make the following assumptions: that in the mid-

13th century the standard pattern for octave scales was

conjunct; that disjunct forms derive from already existing

conjunct scales; and thatdiatonic and Zalzalian units do

not combine, then itfollows that the first scale containing

1 Such names have previously been dismissed as fanciful

and as no guarantee that the scale in question was known in
practice, on the grounds that this work names all consonant
scales among the 84, and that few of them are reported
elsewhere. In this case, however, one can set against the
silence of other treatises the fact that majlis afruz is
the name of a modern Arab maqam
■III rtlI!!■I and a modern Persian gushe.
^ igfahan we should expect to find is not the apparently

disjunct 66 igfahan, but one of these two scales,

-

S it Bl> c 4t eb e f g /1 2t 3b 4/1 2t 3b 3 4/T/

From this, as we have seen, we may derive the other possible

scale, but not 66 igfahan* However, we need not immediately

abandon the concept of conjunct to disjunct development as

irrelevant with regard to 66 igfahan, for if we postulate a

conjunct scale from which this mode could be derived, and

extend the sequence in both directions, this will include

not only both possibles but the other disjunct mode, 50

kardaniya, as well:

* ST
m
- $T

/'\ 2t 3b 4/1 3 b 3 4/1/

,— ft— •
•___

/'\ 2$ 3b 3 4/T/1 2t 3b 4/

b ..

/T/1 2t 3b 4/1 2t 3b 3 4/
= /1 2 31? 4/T/1 2t 3b 3 4/ 66 igfahan-- >

1 But not however G At Bb B c dt eb f g, since

this contains the dissonant sequence B J between B and dt.
 (184)

, » *> * *

/T/1 3b 3 4/1 2 3$ 4/

= /1 2 3^ 4/SS/1 2 3"b 4/ 50 kardanlya

The existence of the two possible scales is attested by Qutb

al-Din, for whom A Bt C D El? F F# G is 6? igfahan

and D Et F F$ G A B1? c d 61 mubayyir zirke sh . But

while this suggests that the above scheme need not be

considered a total abstraction based on purely hypothetical

forms, it only reinforces the impression of a theoretical

reconstruction of an historical development that is both

fanciful and spurious: not only are the "earlier" forms

mentioned by the later writer, but there is an intermediate

stage, with a quite different name, between the two versions

of igfahan. However, even if the scheme must be rejected as

a whole, it may be maintained that the relationships posited

between the first two modes and between the last two is

substantially correct. A more convincing picture results if

we combine them with the suggestion, put forward in chapter 2

that the two versions of igfahan (66 and 67) differ in that

they extend in opposite directions to form octave scales

from an original form made up of just two tetrachords,

/1 2t 3 1? 4/1 2t 3b 3 4/. We would then have:

es A > ... • : > *

50 kardaniya
igfahan basic form
*{>_»

61 mufcayyir zirkesh

'This indicates that 66 igf ahan and 50 kardaniya might after

all be conjunct in origin, but with the disjunctive whole“tone

below, a scale-type never mentioned by gafi al-Din when

dealing with practice. However, neither the previous analysis

of these two modes as medial disjunct, nor even the Systematist

conjunct division, need be summarily rejected. The original

structure of each would tend to be gradually assimilated to

the prevailing pattern in which the focal points were tonic

and fourth. Thus the importance of the tetrachord between

second and fifth would be undermined, the process being no

doubt reinforced by the emergence of a second set of parallel

tetrachords when the fourth is emphasized. A more accurate

account of the way the two are related might therefore be:

66 igfahan

50 kardaniya
It may be doubted whether the last stage given of 66 igfahan,

to which gafi al-Dxn’s analysis corresponds, ever replaced

the previous one. Had it done so, it is likely that igfahan

would have lost its own individual character to become no

more than a hardly distinguishable variant of 43 rast G A B1?

c d e1? f g.

There is one other disjunct scale containing 1 1 / 1 21? 3b

3 4/ that might be mentioned:

G At! Bb B c d et f ft g /1 3b 3 4/T/1 2t 3b 3 V

Despite being included in gafi al-Din5s corpus of 84 scales

it has not been discussed previously because it is somewhat

unusual in having 9 notes (all those stated by him to occur

in practice having 7 or 8). In the sharb it is said to be

1
a shu* b a , derived form, of igfahan, but it receives no

mention elsewhere. If it did occur in practice we may assume

that it did not result from any conjunct to disjunct

development, for its conjunct correlate, G Ai? Bb B c d?

eb e f g, is dissonant, containing the sequence B J from

2
B to dt. No further 9-note scales from the 84 require

discussion.

1 BM. MB. Or. 2361, fol. 105-

2 We may suppose that by the late 14th century, when the

disjunct pattern had become common, new disjunct scales
could be added which were unrelated to conjunct ones. The
igfahan referred to is almost certainly not one of the two
octave scales of that name, but the tetrachord 11 igfahan.
 (187)

(5)
We now come to the three modes made up of a tetrachord

and an indivisible pentachord. Two of them contain the

1
pentachord /1 2p jb 51? 44 5A and the rules laid down

by gafi al-Din preclude the possibility of this unit occurring

as the lower segment of a scale, the reason being that any

such scale would omit the fourth. However, the omission of

the fourth in certain modes is recognized by writers of the

14-th and 15th centuries, so that it would appear not

unreasonable to posit, parallel to the form of derivation

examined above, the following developments:

58a zlrafkand /1 2t 3b 4/1 2fr 3b 3I5 4$ 5/--------- ►

/1 2i? 3 b 3t 4$ 5/1 2t 3 b 4/
59a kawasht /1 21; 3^ 4/1 2fr 3b 3? 4$ 5/--------- »

/1 2$ 3 b 3"b 4t 5/1 2 t 3"b 4/

The resulting scales are not themselves attested, but may be

compared with, respectively,

2
sunbula G At Bb B1? d e1? f g (listed by al-La&hiqi)

bigar G At? Bb B*b ct d et ft (as defined by *Abd

5
al-Qadir)

1 Assuming for present purposes that 59& kawasht is an

octave scale.

2 IPErlanger, la musique arabe, iv, 4-4-1-2.

3 Bodleian MS. Marsh 282, fol. 51- It is possible that

this form is prefigured in the definition supplied by Qu£b
al-DIn (see the remarks on 25b fcigar).
 (188)

There are no further consonant combinations of /1 2t 3b 3^

44 5/ with a tetrachord unit above which could be regarded

as additional possible scales.

The remaining mode, 54- buzurg G At B(t) c cff d e ft

g, is the only one to contain the other pentachord unit,

1
/1 2t 3 A 4-if 5/* Since this occurs as the lower segment

we may postulate a scale from which 54- buaurg might be derived:

/1 2 3t4-/1 2t 3 4- 4-if 5/ ------------------- »

/1 2t 3 4- 4-if 5/1 2 3^ 4/ 54- buaurg

This would be notated by gafi al-Din as G A B”c c d“°

e~° f g”c g, i.e. in exactly the same way as the 8-note

version of 57a zankula. Although, as we shall see, this may

not be entirely fortuitous, it is nevertheless unlikely that

there was any such developmental link between zankula and

buzurg. But the two do have something in common that sets

them apart from all the other modes recognized by gafi al-DIn:

they both have the surprisingly (and suspiciously) low

2
consonance ratio 3 ^ 3 * Not only is this much lower than

that of the other 8-note scales (4- : 1 or better), but it also

compares unfavourably with that of 32 rahawl (3 : 2), a mode

1 As distinct from /1 21? 3^ 4-/SS/, identical in gafi

al-DIn5s notation.

2 This is true of the 8-note version of zankula, not the

7-note. In neither case will the consonance ratio be
affected by whether we consider buzurg to contain Bt? or B,
zankula e'b or e.
 (189)

which seems to have hardly outlived gafi al-Din in the form

he knew precisely because of the consonance factor* In view

of this it would not appear unreasonable to suggest that the

way in which gafi al-Din notates buzurg and zankula is in

certain respects a disguise imposed in order to create the

impression of uniformity of scale structure throughout the

shudud group.

The process through which the notated form of 54- buzurg

arose may have been one of elimination. Given that a mode

consisting essentially of the pentachord 20 buzurg /1 2t 3

4- 4-8 5/ bas to be extended upwards to form an octave scale,

all the tetrachord units available for this purpose are to

a greater or lesser degree unsatisfactory: the best

consonance ratio would result from the addition of /1 2 3

4-/ or /1 2t 3b 3 4-/, but the former would involve the

combination of diatonic and Zalzalian units, while the latter

would produce the dissonant sequence B J from ct to e t , as

would /1 2t 3t 4-/; /1 2 3b 4-/ and /1 2t 3b 4-/,

which produce an acceptable consonance ratio, also suffer from

one or other of these disadvantages, while /1 2b 3b 4-/

combines them. We are thus left with /1 2 3t 4-/, which

neither infringes the norms of unit distribution nor produces

a dissonant sequence of intervals, however unsatisfactory

the resulting scale may be from the point of view of

consonance.
 The above argument may be thought conjectural. However,

it should be realized that gafi al-Din5s notation of 54 buzurg

is suspect not merely because of its low consonance ratio,

and that in addition he provides an obviously artificial

division of its scale. In the risala al-sharafiyya he lists

the pentachord 20 buzurg among the isolate species (ajnas

mufrada) and is at pains to define, by means of ratios, the

™c ' —c
interval notated as A - B as being larger than a whole-

tone. No mention is made of a possible division into

tetrachord and divided whole-tone, and the very name implies

that this would be erroneous, buzurg being an abbreviation of

zirafkand (-i) buzurg, i.e. "the major species omitting the

1
zlr-string note", which presumably means that the original

form was /1 21? 3 4# 5/* -And yet this is precisely the

division gafi al-Din adopts, along with the misleading

notation, when presenting the octave scale 54 buzurg in the

kitab al-adwar: /G ATC B~c c/c d”° d e gb g/. 'This

is an evident theoretical fiction which does little to bolster

one’s confidence in the accuracy of his account.

Later treatises provide little evidence either to confirm

or to deny these suspicions. As has been noted, most of them

simply reproduce the account of the shudud and awazat found

in the kitab al-adwar. An excex>tion, in this respect at

1 The corresponding “minor species” being 1 zirafkand

(~i kuchek) /1 3b 3^/*
least, is provided by the durrat al-taj, which includes 54

buzurg, but lists in addition two variants:

55 Ct At B(1?) c cif d et f flf g

56 G At B(t) c c# d et ft g

These are quite unlike the other variant forms mentioned by

Qu^b al-Din, which generally differ in only a single note from

their parent scales, and should more properly be termed

substitutes. They replace the unsatisfactory /1 2 4/

with Zalzalian tetrachords producing a far larger number of

consonant intervals but at the same time the dissonant

sequence B J, and in the case of 55 sin exceptional 9-note

structure. The fact that 55 &nd 56 are found only in the

durrat al-taj suggests that because of this they were unable

to establish themselves as viable entities, or that they embody

no more than an alternative way of solving an intractable

(but purely academic) problem that was rejected by later

theorists. Nearly all of these adopt gafi al-Din5s version,

but al-Ladhiqi states that in his day this mode consisted of

1
no more than the pentachord 20 buzurg. His evidence is

valuable, although unfortunately not decisive, as he was

writing two centuries later.

The above discussion has dealt to some extent with the

primarily theoretical aspects of notation, scale division,

1 BM. MS. Or. 6629, fol. 67.

and the quest for structural uniformity among the shudud.

The problems of consonance levels and dissonant sequences, on

the other hand, are clearly relevant to practice* As chapter

7 makes clear, a composition would seldom be based on just

one of the units we have isolated: usually it would be a

case of combining them, either in a fixed arrangement, i.e.

in one of the modes presented in chapter 2, or freely,

according to the artist’s taste and skill. But this freedom,

like most others, would be limited, in this instance by

considerations which found expression in the theorists’

statements on consonance and dissonance. Hence the seemingly

abstract formulation of the problem in the terms used above:

what unit, given such-and-such rules, can be combined with

20 buzurg to form an octave scale?, needs little alteration

to express the executant’s difficulty: what unit, given the

norms of consonance observed in such sequences, can precede

or follow 20 buzurg, and at what pitch?

That gafi al-Din’s notation of zankula as an octave scale

might also be an artificial extension is suggested by the

description of this mode in the durrat al-taj. From this it

would appear that a more correct representation (with a quaver

indicating a weak or unimportant note) would be:

<»
— --------- t ------- — ----- ?-----f... "
P O ' " V
"■"V
1 9 V* V
J 9

c is stated to be the centre around which the melody

revolves.

In view of the greater detail in Qufcb al-Din’s definition

(including comments on melodic movement), it would seem

reasonable to consider his version the more reliable. And

since he ascribes to gafi al-DIn the composition transcribed

in chapter 8, in the course of which zankula occurs in the

above form, we may assume that his is also a fairly accurate

account of this mode as it was known to gafi al-DIn some 50

years before. If so, one would expect gafi al-DIn to notate

it not with notes on either side of a tonic G, but as here,

from G to f, since the function of c as tonic would not then

be radically different from the pivotal rSle of the prominent

fourth in other modes. As he indicates neither prominent nor

weak notes, zankula would in that case conform in appearance

to the conjunct tetrachord pattern of the majority of the

shudud: it would therefore be natural to complete the

disguise by adding the pseudo-octave note g.

But why then introduce a further version with a

fictitious f# as well? The most likely answer to this is

suggested by the lack of agreement between gafi al-Din and

Qu£b al-Din as to whether zankula has e1? or e. If the above

1 BM. MS# Add. 7694? fol. 254. The passage in question

(which may be found translated in chapter 7) mentions
neither G nor f. Their presence may however be inferred
from an earlier remark, and they do in fact occur in a
section of a notated composition in zankula.
 (194)

argument is correct, gafi al-Din would analyse (G) A. Bt c

dt e (f) as /I 2 jt? 4/1 2t 3 4/ and wish to extend

this for the sake of conformity to /1 2 jt 4/1 2t 3 4/

T/. But when it comes to actually notating this a further

difficulty arises, for he does not recognize the unit /1 2t

3 4/ (12 fcijazi ) . One way of resolving the difficulty would

be to begin by adopting in place of an already artificial

pentachord /1 2t 3 4/T/ the one unit in which he does

recognize a larger than whole-tone interval, although of

course one masked in the notation: /1 2t 3 4 4$ 5/.

(G) A Bt c dt e (f) would thus appear as G A Bt c

dt e(t) f fit g, from which the redundant fit could be

discarded to provide the other and more realistic form. This

we may accordingly amend, in conformity with our notation of

20 buzurg, from G A Bt c dt et f g to G A Bt c dt

e(t) f g. That Qu£b al-Din, in the list transcribed in

chapter 2, should also include the redundant fit (which he

gives in fact as gb), may be considered no more than an

obeisance to the authority of gafi al-Din. If the fit were

ever used in practice it would almost certainly represent

a change from zankula to buzurg.

 chapter 5

In chapter 2 it was suggested that 21 buzurg G At B c

cff d, Qu^b al-Din7s addition., may not be a variant form but

simply a way of notating the buzurg pentachord more accurately

that in 20 buzurg G At B(t) c eft d. In chapter 4 it has

been contended that the differences in the notation of zankula

provide a parallel case. But if, in these two instances, we

would consider it valid to replace gafi al-Din5s /1 2t 3t

4.../ with Qu£b al-DIn’s /1 2t 3 4.../, it follows that

in other modes, too, what appears in gafi al-DIn’s version

as /1 2t 3t 4/ might be a theoretical adjustment of /1 21?

3 4/ if the latter occurs in the corresponding version given

in the durrat al-taj. In this respect we may contrast the

following pairs:

gafi a l - D m Qutb al-Din

I »
22a G Al o d 22b ‘uzzal G At B c d

2? a G A Bt c$ d 27b niri zx G A B1? cil d

46a fri.jazi G At Bt c dt et f g 46b frl.jazi G At Bt c dt e f g

62a nlhuft G At Bt c d et f g 62b nifauft G J t B c d e t f g

 (196)

with which one may also compare:

I----------\ j-- --- 1
32 hahawi G- At Bt c dt eb f g 33 rah awl G At B c dt eb e

The possibility that in these instances too, Qu£b al-Din5s

notation may be simply the more accurate, and thus (except

for the upper segment of rahawl) faithfully represents these

modes as they were also known to his predecessor, is reinforced

by a passage in the kitab al-adwar concerned with quite a

different matter. In the course of a discussion of the terra

murakkab, combination, gafi al-Din raises the following

objection to the descriptive formula it involves: "If one

subscribes to it [i.e. calls a certain scale a combination

of such-and-such units] , why not call rahawl a combination

nawruz and fcij azi; zankula a combination of £iit

jazl and

rast; and igfahan a combination of igfahan and rast?" What

is significant here is not the argument, which amounts to no

more than saying why use a compound name referring to the

constituent units when the scale in question already has a

simple name, but the fact that gafi al-Din twice uses the

term frijazi to designate one such unit. Both the examples

in which it occurs, (32) rahawl and (57a ) zankula, are cited

above; and it will be seen that in each case the tetrachord

referred to is given by Qu$b al-DIn as /1 2t 3 4/, 12

t-ijazi, but by gafi al-Din as /1 2'b 3"b 4/, 10 tiraq . Mot

1 BM. MS. Or. 136, fol. 22v.

only does gafi al~DIn not recognize /1 21? 3 4-/? but he

also omits all mention of the name hijazi when defining and

listing the tetrachord species. This makes the occurrence

of the name here all the more interesting, especially as it

is used in conjunction with zankula, with regard to which we

have already suggested, in the previous chapter, that Qutb

al-Din5s version with /1 2t> 3 4-/ should be considered the

more accurate.

We cannot of course be sure that this passage is an

inadvertent admission of the presence of 12 hi j azi /1 21? 3

4/ in the modal system of gafi al-DIn5s day. But as all the

other tetrachord names he uses correspond to those found in

the durrat al-taj it would be reasonable to assume that this

is so in the present case too. If not, hi.1azi would have to

be explained either as a mere synonym of 10 4iraq /1 21? 3"b

4/, or as one of an alternative set of terms known to gafi

al-DIn in which It replaced 10 *iraq. Neither explanation

convinces. Again, any attempt to relate the correspondences

between the two lists to regional variations or differences

of tradition would be faced with the same difficulty -

explaining why gafi al-DIn does not refer to *iraq rather

than to hi 3 azi.

If we accept, provisionally, the identification of gafi

al-DIn5s hi j azi with 12 h-ijazl /1 21? 3 4/ it follows that

rah awl and zankula become 0 A'b B c d1> eb f g and

G A B1? c e f g respectively, the latter being

identical with the version proposed in the previous chapter.

Further, if we admit that these two modes contain 12 fcilazl,

we may expect to find it also in 46a foie

1azi and the mode

derived from it, 62a nihuft. Again, the resulting forms

coincide with those listed by Qufb al-Din. In the case of

44 frit
jazi G At Bt c dt et f g, however, there is no

corresponding form in the durrat al-tat

j: nevertheless, it

is reasonable to assume that /1 2t 3 4/ would occur here

too, and if we accept, by analogy with 46b bit

j azi, that c

and f were prominent, the following two forms are possible;

or

^ hi(
jazi would thus be differentiated from 44 *iraq (Qu£b

al-Din)

with regard to both the position of the prominent notes and

the nature of the constituent units* I'his suggests that in

1**5 6iraq G- At c d1? el? f fit g the ft may have been

added to avoid any confusion between this mode and the notated

form Jpafi al-Din gives to 44 frit

jazi: for Qu£b al-DIn, it

may be remembered, 45 8iraq is no more than a variant of

1 Ignoring for the moment the status of the octave note.

 (199)

44 1 iraq.

These modifications are based on the assumption that

gafi al-Din5s references to a tetrachord named pijazi

constitute an "inadvertent admission", further, that other

explanations of the presence of this name are unsatisfactory.

But this type of argument fails in one important respect:

no solution is offered to the fundamental problem of why

Safi al-Din should neither list nor define 12 pijazi /1 2t

3 4/ if it did in fact occur in a number of well-known modes.

The answer to this - if there is one - can only be pieced

together from evidence scattered through gafi al-Din9s two

treatises, none of it concerned directly with the question,

much of it therefore open to argument.

Firstly, we may repeat that we already have in 20 buzurg

a clear example of misleading notation, a larger than whole-

tone interval being given as a whole-tone. But the risala

al-sharafiyya contains a further and even more explicit

example of this process of falsification. The intervals of

25a
the pentachord/are analysed in the first instance as 128,

139s 4*9> 231 s and 155 cents, but gafi al-Din comments that

it would be preferable to effect a division in which the

interval of 231 cents is replaced by a whole-tone, giving

1
128, 139* 4-9? 204 and 182 cents, and it is to the latter

1 Bodleian MS. Marsh 521, ff. 65~65v.

 ^■uu;

form that the notation (c d~c eb e”° gb g, listed

elsewhere) would appear to relate, r

fhe obvious conclusion

to be drawn from this is that a larger than whole-tone

interval was used in practice (otherwise the first analysis

would be entirely unnecessary), but that for theoretical

purposes it was more convenient to consider it, and notate

it, as a whole-tone•

What these theoretical purposes were, however, is not

entirely clear. It is possible that it was rather a question

of a theoretical compromise. In the ki t ab al-adwar gafi al-

Din distinguishes three classes of intervals according to

size:

kubra, large: double octave, octave plus fifth,

octave plus fourth, octave.

(ii) wusfa, medium: fifth, fourth.

(iii) gughra, small: I J B; T being defined as 9 * 3,

J as approximately 16 : 15? B as approximately

1
50 : 29.

(It will be noted that intervals between whole-tone (3?) and

fourth are not mentioned.) In the risala al-sharafiyya (i)

and (ii) are the same, but the situation with regard to (iii)

is somewhat more complex. This class comprises all intervals

of the form x + 1 : x (x being an integer) smaller that the

fourth (4 : 5)? and is subdivided into (p), large; (q), medium;

and (r), small, the definitions being:

1 BM. MS. Or. 156, ff. 6-7-

 (.201;

(p) 2p > 4-: 5* I*e. 5 : 4, 6 : 5, and 7 : 6.

(d) 4* : 3 ~ 2q ^ d * I.e. 8 : 7? 9 : 8, and 10 : 9*

1
(r) smaller intervals.

Later gafi al-Din states that, in a melody, group (p) intervals

2
are undesirable as undivided entities. Hence it is

understandable to find him defining as 8 : 7 (group (d)) the

interval defined by Qu£b al-Din in 21buzurg G At Bc eft d

and 25b higar G At Bb Bt elf d as 7 :6 (group(p)). But

if 8 : 7 is admissible there seems to be little reason to

substitute 9 : 8 in 25a G At Bb Bt ct d (both in the

ratios and in the notation) and 20 buzurg GAt B(t) c eft d

(in the notation). It is here that the compromise may come

in, 8 : 7 being mentioned in the analysis of these units, but

9 : 8 being used instead in the notation in order to make

them conform to the rather more rough and ready categories

1 J and B arrived at in the earlier work. Safi al-Din5s

remark to the effect that musicians employ 9 • 8 for all

3
three intervals in group (q) may perhaps be understood simply

as a justification of this compromise, for if taken at face

1 Bodleian MS. Marsh 521, ff* 4-7v-d8.

2 Ibid., fol. 68 . There would seem to be no objective

basis for this curious restriction. 23 a rnaya contains an
undivided minor third; the composition notated by Qu-fcb al-
Din and ascribed to Safi al-Din contains leaps of a fourth
and, in all probability, a minor sixth; and one of the
examples of notation given by gafi al-Din himself even
contains leaps of a minor seventh.

3 ibid., fol. d8v.

value it would mean that the distinction maintained between

major and minor whole-tone in his theoretical gamut is quite

pointless when considered in relation to practice, since it

would follow that certain pairs of modes could not be told

apart by intervallic structure alone.

Even if the above argument is found unconvincing, it is

at least certain that the process whereby £>afi al-Din might

arrive at the notated form G A""c B~~c c for b.ijaz,I has a

parallel in his treatment of 20 buzurg and 25a. It may also

be noted in this respect that in all probability gafi al-Din

would analyse bijazl as 15 : 12, 8 ; 7 ? 14- * 1 5 “ exactly the

ratios (in two of the four versions) for the three lower

intervals of 20 buzurg. But this is not all: if we compare

these ratios with the ratios and the notation given for 10
1

10 : 9 9 : 8 16 : 15
G A”c B“c c

it will be seen that two ways of notating bijazl suggest

themselves: G Ab B~c c and G A“c B c (that given by

Qutb al-Din). The former would be unacceptable, however,

since the interval Ab - B“c is 294- cents, larger than the

already undesirably large 7 : 0 . G A-c B c is therefore

1 The ratios are given only by Qu£b al-Din, but There can
be little doubt that these are the values gafi al-Din would
have adopted too.
 (203)

the obvious choice; but it will be noted that, if adopted,

the ratio value 16 : 15 in ciraq would be notated as a larger

interval than the 14- : 13 in frij azi. 1‘his anomaly may be

—c
added to the considerations which dictated the notation A -

B“° (in place of A“c - B) in 20 buzurg as a further impetus

m — c c
towards the notation of bddnzi as G A B c. A more

concrete disadvantage of G A”c B c is that it would give

rise in 32 rah awl (in amended notation G Ab B c d1? eb f

g) to the dissonant sequence B J between B and d1?.

fhere is yet another possible factor. It has been pointed

out that despite being in the mainstream of the Peripatetic

tradition, £>afi al-Din was perhaps not entirely unaffected

by the doctrine of t a ?thir, ethos; and there is one highly

relevant indication that betrays the influence of the

symbolic importance the doctrine attached to certain numbers.

When defining the tetrachord and pentachord species he remarks

that by making certain concessions with regard to proscribed

(dissonant) series of intervals, 13 pentachord species are

possible; yet he deliberately restricts himself to listing

1
only 12 - a number of some significance, as we have seen in .

connection with the shudud. Now, the inclusion of 12 fciqazi

would have meant on the one hand an increase in the number of

tetrachords to 8, which would probably have been quite

1 It is true that the pentachord not listed, /1 2 3 44

5 A does not occur in any of the scales which, according to
gaff al-Din, are found in practice - but then nor does
/1 2 3^ 44 5 /9 which is one of the 12 listed.
acceptable, 8 being conveniently related to 4- and 12, both

important, but on the other hand an increase in the number

of pentachords, which would appear to have been unacceptable.

Again, It may be noted that the addition of 12 hit

jazi would

have upset the symmetrical balance between the three diatonic

tetrachords (/1 2 3 4-/, /1 2 3b 4-/, /1 2b 3b 4-/) and

the three 4*-note Zalzalian tetrachords (/1 2 3^ 4-/, /1 2"b

3t 4-/, /1 2t 3b V).

The above arguments will, it is hoped, go some way

towards explaining why and how 12 frijazi / 1 2*b 3 4-/ might

have come to be notated in the same form as 10 *iraq /1

3^ 4-/. But they do not provide the answer to the problem

of why, in that case, gafl al-DIn failed to state, when

defining and listing the tetrachord species, that the notation

G A"° B""0 c stood for both, To do so would, no doubt, have

constituted an admission that there were certain shortcomings

or artificialities in his system of interval and unit analysis

But if this defect had to be covered up by mentioning only

one of them, the obvious choice would have been to omit any

reference to *iraq, since 12 hijazi would appear to occur in

far more modes.

The most likely answer to this problem may be found in

the hypothesis that 12 tiiqazi is not fortuitously related to

40 ii3?aq by the Imperfections of a particular notation, but

Is derived from it. This would mean that It is less a

question of frijazi "becoming" *iraq in notation than of *iraq

becoming tiijazi in practice. It would be quite natural to

find Safi al-Din giving the anterior form only (being as it

were both *iraq in esse and frijazi in posse), especially if

the latter was in his day a relatively new development

frowned upon by the stricter upholders of tradition.

Support for this hypothesis is to be drawn not from the

texts, which are silent on the matter, but from an examination

of the structure of these units. 10 6iraq /1 21? 4/ has

been frequently placed alongside 8 rast /1 2 4/ and

9 nawruz /1 2t? 5i? 4/, as if the three formed a set. In

some ways, of course, they do; but /1 21> ji? 4/ differs

from the others in one important respect: it does not, like

them, contain a whole-tone (and a minor third) from the

framing or, to borrow Safi al-Din’s term, fixed notes 1 and

4. Admittedly, the degree, and hence the importance, of the

consonance of whole-tone and minor third is less than that of

the fourth, both generally, within the series of overtones,

and specifically, within the modal system under discussion.

But the contrast becomes significant nevertheless when these

units are reduplicated to form parallel conjunct tetrachord

scales. Thus in

d stands in a consonant relationship with G (and g), one of

the fixed notes, and A with d. A similar pattern is

observable in

In both cases only one note in each tetrachord cannot be so

related, directly or indirectly, to a fixed note. The same

situation arises (except in the soon to be discarded 32

rahawi) when these two combine with other tetrachord units to

form consonant octave scales. However, wdien /1 2t 3^ 4/

is reduplicated (or combined with another unit) the interior

notes cannot be so related:

— ................... N
o
1 ** 42 o y# ^
£ o 99 'I ■_

If it is accepted that in the first two examples the stability

of the whole-tones G - A, c - d, Bb - c , and eb - f is to

some extent dependent on consonant relationships with the

fixed noted G c f g (themselves established by consonance),

then it may be suggested that in contrast the size of the

At - Bt and dt - et intervals in the last example would tend

to vary over a much wider range since they can only be defined

precisely as whole-tones by consonances with et and At

respectively, neither of which are fixed notes.

It follows from this that within the tetrachord /1 2t

3t A-/ both 2t and 3t should be potentially unstable notes.

By altering each of them separately four tetrachords may be

 (207)

obtained:

2 3^ V

/-I 2ti 3t 4/

/1 2l> 3^ V /"I 2ft 3b V

Two of these are already in existence - 8 rast /1 2 3ft 4/

an(^ 9 nawruz /1 2ft 3b 4/. A further four tetrachords may

be produced by altering both notes simultaneously:

/1 2 3 4/

/1 2 3b 4/<- /1 2*b 3t 4/ >/1 2b 3 4/

/1 2b 3b 4/

Of these three are already in existence - 5 ^ushshaq / 1 2 3

4/, 6 bus all k /1 2b 3b 4/, and 7 nawa /1 2 3b 4/. We

are thus left with three possible transformations which

preserve the independence of the unit: /1 21? 3 4/,

/1 2b 3 4/, and /1 2b 3^ 4/. fhe first two of these

are definitely no more than alternative forms of 12

for Qu£b al-DIn notes that the central interval of this unit,
normally 2t » 3, was in practice sometimes enlarged to 2b -
1
3. It is quite reasonable to assume that the third tetrachord

may also have occurred as a variant of 12 hijazi, for the

interior notes of this unit can be no more defined by reference

to the fixed notes than those of /1 2t 3t 4/, It is

feasible to suppose that the same musician may have produced

in ascent and descent varying forms, e.g. /1 2t 3 4/ and

/4 3"b 2b 1/. The interior notes would in effect tend to

2
function partly as leading notes of 1 and 4, according to

the melodic context, and the emergence of 12 hijazi may be

explained, especially if we take the variant /1 2b 3 4*/

into consideration, in terms of the attraction exerted on

the unstable notes 2t and 3^ in 10 1iraq by their respective

adjacent fixed notes. In fact, the impulse to alter the

structure of this unit would appear to have been sufficiently

powerful to render acceptable scale forms with a greater

number of unattached notes: one may compare e.g. 46b hijazi

G At Bb c dt e f g (consonance ratio 4 : 1 ) with the

scale it presumably replaced, 46a foijazi G At Bb c dt et

f g (consonance ratio 5 : 0 ) * In none of the octave scales,

however, does the replacement of 10 *iraq by 12 dijazi bring

the consonance ratio down to the critical level - 3 : 2 .

1 BM. MS. Add. 7694, fol. 222.

2 I.e. considering 2b to function as aleading note in

relation to 1, 3 in relation to 4.

3 The change of unit does not alter the consonance ratio

in 32 rahawi and 57 a zankula.
 Presumably because of this development, /1 21? 4/

seems hardly to have survived as an independent entity- The

-1
only form of *iraq recognized by al-nadhiqi is (Jutb a l - D m ’s

other version, 2 6iraq /1 21? 3^/* This is of course just

as unstable, and would appear to owe its continuing existence

to the use of 43 rast as a parent scale from which further

modes were derived by giving prominence to successive degrees,

usually starting from the fourth, which tends to become the

tonic in rast as in other modes:

Gr A Bt c a et f S (a bl? c' )

. 2 3t V

/1 2t 3b V

/V 2t 3*/ 2 6iraq

lack of a fourth in this form of 1iraq may thus relate

2

1 BM. MS. Or. 6629? fol. 66v.

2 A similar situation would appear to obtain in modern

Arab px*actice. If we take G as the lowest note, yak ah, then
B1? is called *iraq. One may also compare the ascending line
of the modern maqam ciraq, which adheres to the notes of
the rast scale, and is divided into units as follows
(D’Erlanger, la musique arabe, v, 148):

The use of 43 rast as a parent scale may also explain

the position of the prominent segment in 44 and 45 *iraq
(according to Qutb ai-'Dln):
 KJJ

It follows from the conclusions reached above, and in

previous chapters, that gafx al-Din5s notation of a number

of modes may be considered of questionable accuracy, and the

doubtful parts will henceforth be placed in brackets, as with

20 buzurg.
L__ J__ ______ II,1,u^1 I
There is one further amendment: it has been

argued that £>afi al-Din 5s first ratio analysis of the

pentachord 2 5 a is more realistic than his notation (c dt el?

et ft g), and for this we may substitute Qu£b al-Din?s

version, 25b / 1 2t 3b 3t 4$ 5/. The change will affect

5 8 a zlrafkand and 5 9 a kawasht.

The following are therefore the final notated forms to

be adopted for all the octave or quasi-octave scales stated

by gafx al-Din to occur in practice:

52 rahawx G At B(t) c dt el? f g

40 1ushshaq G A B c d e f g

41 bus allk G Ab Bb c db eb f g

42 nawa G A B b c d e b f g

G A Bt c d et f g (a bt c')
,— . •— ^
»iraq 1 2t 3t 4 5t 6t
the completion with at being by. analogy with both the lower
unit and the prevailing octave structure.
 43 rast G A Bt c d et f g

44 hijazl G A Bt or B(t) c dt e(t) f S

45 4iraq G At Bt c dt et f (f #) g

46a tdjazi G At Bb c dt e(t) f g

48 busayni G At Bb c dt eb f g

50 kardaniya G A Bt c cif d e ft g

54 buzurg G At B(t) c c3 d (e ft g)

5 7 a zankula G A Bt c dt e(t) f ((fit) s)

5 8 a zirafkand G At Bb c dt eb et f It g
I
5 9 a kawasht G At Bt c dt eb et (fit g)

60 mubayyir tmsayni G At Bb c d et f g

62a nihuft G At B(t) c d et f g

6 6 igfahan G A Bt c d et f f3 g

The diagrammatic representation of the distribution of

Zalzalian units in these scales (plus 37 nawruz G- At Bb

dt eb f) needs little adjustment in order to incorporate

'1 2 bijazi / 1 2t 3 A/ and the other amendments:

1 See the remarks on pp. 142-143.

 (212)

\ -4-
a
\ '4^
d- K\
4K\
^ ■CO
s-
CO \
m
'C“ CO
\ \
d* d*
K\ 4^
K\
CO CO
V V-
"N

V
Lf\
a \
-»-+-
\ a \
d- d* \ a
N— / d* \
4^ --j
4^ K\ K>
rC\
K\
-KS>. 4^
_Cv CO OJ
K\ CO
V" V V f
-a V
Mil 1 1
\ ,^—
CO d- d* \
^ r- d-
LT\ ,v“ -a 4c\
rc \ \ KN f^\
Lf\ \ d*
d- 4^ 4Es‘ -a 42>
K\ 4^ CO CO CO
d" 4^ rc\
K\ _Ci. X V“ V V
4^ K\ a d- a 4^ X \
4' i K\ ...rs^ \ \ CO
i rc \ 4^ d* d-
i CO 7o^
i
i 4=^ rA \
i CO '\- 4^ K\
1 \ 4^ CO
X X. CO '^— CO 4Ei.
E-f d" V CO
\ V V \
d" -£^ d" \ \ a V"
rc\ d* d* \ \
_E*. _r\ ■
“-J d-
rc\ rW
CO rc\ rc\ rc\
^ •
CO V CO 4^ 4^ -a
\ CO CO CO
T“ a T~
\ \ V V v~
d- d- d* N \ \
_E>. fV _£u
to* K\ K\

4^' 4^ 4^-
CO CO CO
r~ T" V"
\ \
 To conclude, we may formulate a number of brief

descriptive statements based on the examination, in this and

the preceding chapters, of gafi al-Din7s account of the

modal system;

■ With the apparent exception of salmak, the modes may be

discussed in terms of the following units:

1
? /1 21? 3V /1 2t 5b 3*/
/1 2, 3 4/ 2 3b 4/ /1 2b 3b 4/

/1 2 31; 4/ /1 21> 3"b 4/ /1 21? 3 4/

2t 3b 4/ 3b 3 4/ /1 3b 4/

/'\ 2t 3 4 48 5/ /1 2t 5b 3t 4# 5/

One mode (shahnaz) is certainly based on a single unit

(/1 21? 3^ 3 ^/)? and another (54 buzurg) may be (on

/1 2t 3 4 4$ 5/)* The remainder consist of a combination

of two units, or of the same unit repeated. We may distinguisi

non-octave scales:

parallel conjunct tetrachords (23a maya - /1 3^ 4/;

37 nawruz - / 1 4/)

non-parallel conjunct units (57a zankula - / 1 2 3^ 4/,

/'1 2? 3 4/; ? 59& kawasht - / 1 2^ 3p 4/*

/1 21? 3i> 3 ^ /)

1 Only if we wish to separate this unit from / 1 2? 3^ 4/

in the light of Qu|;b al-Din 7s notation of 44 and 45 1 iraq.
It has not been given below.
 (214)

octave scales:

parallel conjunct tetrachords:

diatonic (40 ;ushshaq - /1 2 3 4/; 41 busallk -

/1 2b 3b 4/; 42 nawa - /1 2 3b 4/)

Zalzalian (43 rast - /1 2 3*b 4/; ? 44 foijazi - /1 2'b

3 4/; 45 ciraq - /1 21? 31? 4/; 48 husayni -

/1 2t 3b 4/)

non-parallel conjunct tetrachords ( 3 2 rahawi - /1 21? 3

4/, /I 21? 3b 4/j ? 44 frijazj - /1 21? 4/,

/1 2^ 3 4/; 46a frljazi - /1 2t 3b 4/, /1 2t 3

^0 mufrayyir frusayni - /1 2‘b 3b 4/, /1 2 3^ 4/)

parallel disjunct tetrachords (5 0 kardaniya - / 1 2 3^

4/; 60 mutiayyir fousayni - /1 2t> 3b 4/)

non-parallel disjunct tetrachords (62a nihuft - / 1 2"b

3 4/, / 1 21? 3b 4/; 66 igfahan - / 1 2 3 b 4/,

/1 2b” 3b 3 4/)

tetrachord and pentachord units (? 54 buzurg - / 1 2 'i? 3

4 41 5/, /1 2 3V 4/; 58a zirafkand - /^ 2? 3b

4/, / 1 2t 3b 3t 4i
f 5/; ?5 9 a kawasht - / 1 21? 5^

4/, / 1 21? 3b 31? 4# 5/)

Of the disjunct scales 66 i$fahan appears to relate to

an original conjunct form, while the other three may be

derived from existing scales of conjunct form, the fourth

becoming the tonic. During the 14th century, if not beyond, the
 <.215;

system continued to be enlarged by this process of modal

development.

Diatonic units do not combine with any other unit. The

distributional relationships of the Zalzalian units have

been shown in the above diagram.

Among the octave scales, according to the criteria

discussed, both possible consonant combinations with / 1 2t

3 I? 3 I? 4-j} 5 / occur, while there is no satisfactory

combination including / 1 2 t? 3 4- 5 /* If / 1 y1> 4/

and / 1 2t 3 V are classed together (as representing stages

in the evolution of the one unit), then in conjunct tetrachord

octave scales all possible consonant combinations of / 1 2 3 "^

4/, /1 2t h 4/, /1 3b 4-/ and. /'\ 21> 3b 3 4/

occur, with the sole exception of (6 7 ifffahan) / 1 2t 3^ 4-/

1
1 3t 3 V T / .

1 Of the alternatives /1 21? 3 4-/1 21? 3^ 3 4-/T/ and

/1 21? 3*^ 4-/1 2t> 31? 3 4-/T/ the latter is unacceptable,
having three unattached notes. However, with the admission
of 12 frijazi elsewhere, the dissonant status of the B J
sequence, wEicli occurs in the former between B and d1?,
becomes open to question.
part 3
 chapter 6

In considering the account of the modal system presented

in the durrat al-ta,j, it may be remarked first of all that

Qu£b al-Din does not simply take over the shudud/awazat

classification of his predecessor*. He does i*efer to these

categories; but his views on which modes are to be assigned

to them differ in some instances at least from those of gafi

1
al-Dm , for he classes 60 mubayyir husayni as an awaz, while

maya and shahnaz (awazat for £>afl al-Dxn) are stated to belong
2
to another group, the shu* bas ■ However*, as he fails to

supply a full list of the modes he considers to be awazat,

there is little point in reopening the debate on the

structural principles, if any, underlying these two

categories.

Bor* mode and scale Qutb al-DIn uses the generic term
3
parda, which therefore comprises the shudud, the awazat, and

shu;bas , an important group not mentioned by £>afl al-I)in

which the durrat al-taj discusses in some detail. The shu* bas,

1 BM. MS. Add. 7694, fol. 233*

2 Ibid.
3 The durrat al-tafj appears to be the first text to use
MmaqamM as a generic term for mode. It occurs infr*equently.
 (218)

which are stated to he modal forms character'ized by particular

1
melodic features and derived from already existing pardas ,

will be discussed in chapter 7 ? as will the following, the

names of which suggest a combination of a shu* ba and another

unit:

30 (segah wa fric
jazi) G Ab B c d et*

31 (segah wa fausayni) G At Bb c d et

38 (chahargah wa fail a 2 1 ) G At B c d et f

39 (chahargah wa Isfahan) G At Bb B c d et f

In the preceding examination of the scales mentioned by

gafi al-DIn it was suggested that in certain cases his

notation should be amended. The resulting forms coincide

with the versions put foi'ward by Qufb al-DIn, thus reducing

the differences between their respective accounts. If we

accept these amendments, the following are the scales listed

in the durrat al-tan which gafi al-Din fails to cite:

33 rahawi-yi tarnam G Ab B c d"b eb e

34 i^fahanak G Ab B'b c dt? eb e"b
35 variant of igfahanak G Ab B c d? eb e"b
36 kuchek-i tamam G Ab Bb c dt» eb et?
4? frusayni G Ab Bb c d eb f g
49 kardaniya G A B'b c d e ft g

51 variant of kardaniya G A B c d e f$ g

Ibid., fol. 233V.

 ^ IV J

52 variant of kardaniya S A Bt cd e f# g

53 variant of kardaniya G A B cifd e fit g

55 variant of buzur_g G At B(t) c cif d et f fif g

56 variant of buzurg G At B(t) c oil d et fif g

61 muhayyir zirkesh G At Bb B c d et f g

63 k ardaniya nirizl G A Bt cSf d e ft g

64 G A Bt c d et f# g

65 G Ab Bb c d eb f g

^7 ifffahan G At Bb c dt eb e f g

68 G At B c d et f g a

69 nihuft-i kamil G At Bt cdt e f g at bb c'

70 nihuft-i kamil G At B cd et f g at b c'

7^ buzurg-i kamil G At B(t) c cif d et f fif g at b c'

72 busalik G Ab Bb c db eb f g ab bb c'

If 59a kawasht is amended to coincide with 34igf ahjlnak, then

59t GAt Bt c dt eb et fif g may be added.

Bor §>afi al-DTn a number of these scales would be

dissonant. The proscribed sequence B J occurs in 55 an <3. 56

(between ct and et), 33 and 35 (between B and dt), while in

71 it appears twice (from cif to et and from fit to at). In

a further two scales, 53 and 63? the fourth Is omitted.

However, some of these "dissonances" are no more than the

result of Qufb al-Din*s recognition of 12 hijazi /1 2t 3 4-/*

and occur also when this unit is substituted for 10 4 iraq

/1 2t 3t 4/ in certain modes notated by gafi al-DIn. In

effect, the sequence B J may be rejected as a valid criterion

of dissonance with reference to the corpus of scales in the

durrat al-taj. It should be noted that all these scales

would appear to be satisfactory structures when viewed in

terms of consonant intervals and unattached notes. It is true

that the two non-octave modes listed by gafi al-DIn which do

not just consist of one unit (25a maya G Bb c eb f and

37 nawruz G Ab Bb c dl? eb f) have better consonance

ratios (4 : 0 and 5 : 0 respectively) than most of the non­

octave combinations given above (33 rahawi-yi tamam, 3 • 1?

34 igfahanak, 4 : 1 ; 35 variant of igfahanak, 3 * 2, 36

kuchek-i tamam, 4 : 0), but only one of these has the same

low ratio as the octave scale 32 rahawi.

Of the two non-tetrachord units which figure in these

four combinations, 1 zirafkand-i kuchek /1 2? 3b 3^/ is

found in conjunction with /1 2t> 3b 4/, /1 21? 3^ 4/ and

/1 21> 3 4-/, while 4 rahawi /1 21? 3b 3/ combines only

with /1 2*b 3 4/. If one regards these two units as being

in some way related to, respectively, 25 /1 2*b 3 b 3"b 4 if 5/

and 1 1 / 1 2t? 3b 3 4/, it will be seen that the combinations

in which they are found closely resemble those in which the

latter appear. The diagram representing the distributional

relationships obtaining among the Zalzalian units described

by gafi al-Din plus 12 hijazi /1 2? 3 4/ (p. 212) would

thus need hardly any alteration in order to accomodate them.

In fact, it is likely that at least some of them wex^e known

to glafi al-Din: it may be recalled that although / 1 2V 3b

3t/ appears only once, and then in a specific melodic form

(24a shahnaz) , while /1 2t 3^ 5/ does not occur in any of

the scales he lists, he does go to the trouble of defining

both, by ratios as well as by notation, in the risala al-

sharafiyya. It would therefore be surprising if they were

not in common use during his lifetime also, and the four

non-octave modes listed here by Qu£b al-Dln are among the

most consonant conjunct combinations possible with these two

units.

The octave scales, all of which have a consonance ratio

better than the critical 3 : 2, may for the sake of

convenience be dealt with in two groups, the first comprising

49 kardaniya Ct A Bt c d e ft g

51 variant of kardaniya G A B c d e ft g

52 variant of kardaniya G A Bt c d e f if g

55 variant of kardaniya G A B cif d e ft g

S3 kardaniya n i n z i G A Bt ct d e ft i

64 G A Bt c d et f if i

We may safely assume, contr ary to the headings in

durrat al-taj, that the original form of kardaniya, to which

most of these scales relate, or were thought to relate, was

5>afi al-Din’s version, 50 kardaniya G A Bt c ct d e ft

g. It has been suggested that this mode was derived from

igfahan, appearing first as

 (222)

but that this was subsequently assimilated to the standard

octave-scale pattern which laid emphasis not on the tetrachord

A - d, but 011 the tetrachord G - c. The resulting form

— — e=v--z— r > — *—
\jyt
/
j v 9 P * --
is however unique in that the medial disjunctive whole-tone

is divided. Most of the above six scales may be explained in

terms of an attempt, on the part of practising musicians, to

normalize the structure of this mode by avoiding this unusual

feature.

The most obvious course of action would be simply to omit

the major source of difficulty, the cif, thus producing 49

kardaniya G A Bt c d e ft g. But the existence of the

other variants suggests that this change was felt at the time

to be insufficient or somehow unsatisfactory. The reason is

probably that the same scale was incipient elsewhere in the

modal system, for Qu£b al-Din gives c as the prominent note

4*5 rast G A Bt c d et f g, and it is the starting

point for the shu* bas derived from this scale. Already at

this period, therefore, it is likely that the modal development

1 Although 49 does survive at least until the end of the

15th century, being recognized by both 6Abd al-Qadir and
al-Ladhiql.
whereby the fourth becomes the tonic was also being applied

to 43 rast:

/1 2 V 1 2 31? 4 / ® / ------- >/1 2 4-/T/1 2 31? 4-/

The resulting scale is identical with 49 kardaniya.

Thus for kardaniya to preserve its individuality at the

same time as regularizing its structure, either a further

change is desirable in 49, or an alternative change in 5 0 ,

the original form. The former process may be seen in 51 0 A-

B c d e ft- g and 52 G A Bt c d e fif g, the latter

in 6J G A Bt cif d e ft g. In 51 lower tetrachord

°£ 49 kardaniya, /1 2 3t 4/, is replaced by /1 2 3 4/,

and in 52 the upper. However, both scales are again unusual,

for they combine diatonic and Zalzalian units, and the non-

appearance of 52 in later accounts of the modal system may

1
perhaps be attributed to this factor. In 63 the sequence

c cif d in 50 kardaniya is obviated by omitting the fourth

c. That this expedient was also felt to beunsatisfactory is

suggested by the presence of 64 G A Bt c d et fif g

which may be derived from 63, the fifth becoming the tonic:

63 G A Bt cif d e ft g

/T/1 2t 3 4/1 2 3t 4/--- >/1 2 3t 4/T/1 2t 3 4/

G A Bt c d et fif g 64

1 51 is however mentioned in later texts, where it is

generally called mahur.
 {ddH-)

64, with its prominent fourth, would thus be in effect an

analogical back-formation.

(The status of 55 G A B cif d e fif g as a variant of

kardaniya is less easily explained. As it is the only other
octave scale listed by Qu^b al-Din in which the fourth is
omitted it would seem logical to link it with 63 G- A B1? cif
d e ft f. The differences between the two are however
considerable, and there would certainly seem to be no good
reason for drastically altering the latter while leaving the
main irregularity, the omission of the fourth, undisturbed.
But if the appelation of 53 is difficult to account for, the
actual scale could be arrived at quite easily by assuming
a process of development parallel to that postulated for the
two forms of igfahan:

40 *ushshaq

But this still does not explain why a scale with such an
unusual structure should come to be adopted at all. The
fact that 53 is only recorded in the durrat al~tat
j suggests
that, whatever its origin, it was an experimental form quickly
abandoned.

The other group consists of:

 47 busayni G At Bb c d eb f g

55 variant of buzurg G At B(t) c c8 d et f fit

56 variant of buzurg G At B(t) c cit d et ft g
61 mubayyir* zirkesh G Ab Bb B c d et f e
65 G Ab Bb c d eb f g
67 igfahan G At Bb c dt eb e J* g

These are simply the remainder of the octave scales, placed

together fox5 convenience. Some of them have been discussed

in previous chapters. 67 igfahan was considered in

conjunction with £>afx al-Din’s version, 66 igfahan G- A

c d et f ft g, and it was suggested that 61 mubayyir

zirkesh derives from it:

67 G At Bb c dt eb e f g

/1 2? 3 b 4/1 2t 3b 3 4/T/— »/l 2t 3b 3 4/T/1 2t 3 b 4/

61 G IP Bb B e d et f g

Similarly it has been suggested that 65 may be derived from

41 busallk:

41 G Ab Bb c db eb f g

/1 2b 3b 4/1 2b 3b 4/!/--- >/*\ 2b 3b 4/T/1 2b 3b 4/

65 G Ab Bb c d eb f g

The two variants of buzurg, 55 56? have been discussed

in conjunction with 54* buzurg G At? B(t) c ct d e ft g.

Although much more satisfactory than 54 buzurg when considered

in terms of consonant intervals and parallelism (at the

fifth), both appear to have been short-lived.

The origins of the remaining mode, 47 fansaynl G At Bb

c d eb f g, are obscure. In the other form of busayni,

that recognized by both authors, 48 G At Bb c dt eb f

g, the prominent unit is the c - g pentachord, 18 busayni,

and it would therefore be natural to find c becoming the

tonic. As we have seen, this had in fact already happened

in gafi al-DIn5s lifetime or earlier, the resulting mode

being 60 mubayyir busayni G At Bb c d et f g. However,

there appears to be no particular reason for deriving a

further scale from this by substituting eb for et, and such

a change would hardly have occurred unmotivated. It is

perhaps moi'e likely that 47 should be related to 48, arising

indirectly out of the performer5s desire for unit symmetry,

i.e. in this case reproducing the pentachord 18 busayni

/1 2t 3b 4/T/ from G as well as from c:

48

The dt> would tend to he discarded in favour of d if the

lower pentachord then assumed greater importance, and 47

would result.

Among the scales exceeding the octave

68 G A1> B c d et f g a

with the contrast of Ab and a, can perhaps be similarly

 (227)

explained as resulting from an impulse on the part of the

performer to balance the lower pentachord (G - d) with another

(d - a), although in this case the two are not parallel. The

guest for consonance and symmetry might again be seen as the

motive dictating the choice of the upper tetrachords (g - o')

in

i — .— i r~ ..\
r
7 0 nihuft-* i kamil G At B c d et f g at b 0
i 1 V "
71 buzurg-i kamil G At B(15) c cfi d et f f if S at b
r M \ *
72 busallk G Ab Bb c db eb f s ab bb o'

The other form of nihuft--i kamil, 69 G At Bt c dt e f

g at bb c' , does not conform to this pattern, although it

may be pointed out that considered separately the octaves

G - g and c - o' are both satisfactory structures. It must

be remembered, however, that 69 and 7 0 are alternative

renderings of the same entry in Qu£b al-Din’s list, and it

is by no means certain that both were found in practice.

If 69 did occur, it might conceivably have resulted from a

fusion, or deliberate overlapping, of 92 rahawl (G - g) end

62 nihuft (c - c').

To return to the octave scales, it may be noted that

Qutb al-DIn’s additions include four of the seven possible

scales postulated in chapter 4- if one allows the substitution

of 12 fri.jazl /1 2t 5 V for 10 ‘Iraq /'\ 2t j

7t V in

the upper tetrachord of G A B1> c d et ft g. They also

 (228)

include a number of scales with sequences of intervals

considered dissonant by £>afi al-DIn, and two in which the

disjunctive whole-tone is placed below. These therefore

provide further patterns of unit distribution in conformity

with which another group of possible scales might be posited.

However, there would be little point in examining such scales

in order to determine why they did or did not subsequently

occur, for if the evidence of the durrat al-taj indicates

that the period 1250-1300 saw an increase in the repertoire

of octave scales, later texts suggest that despite the

continued addition of new derived forms the trend was rather

towards a reduction in the total number of octave scales in

current use, so that by the late 15bh century it would appear

that they were outweighed in importance by modes based

generally on a single unit (tetrachord or pentachord in most

cases), and characterized by particular melodic features,

i.e. modes comparable to the shu* bas described by Qu£b al-Din.

S|< S{C

The following is a schematic summary of the developmental

relationships between octave scales examined in this and the

preceding chapters.

Hypothetical forms (given on two pitch levels to avoid

 Kdd>3)

confusion or transposition elsewhere) are linked to the

relevant attested forms by a broken line. Derivation by the

1
fourth becoming the tonic is indicated by an arrow, derivation

by dissimilation or assimilation by a wavy arrow. In doubtful

cases these signs are doubled. The two entries for a scale
2
listed by both gafi al-Dxn and QutJb al-Dln are linked thus:

(the clef sign being omitted for reasons of space).

The last column includes only scales not mentioned by

either' of them: the non-appearance of any given scale in

it should not therefore be taken to indicate that it did not

occur later.

1 Or, In one case, the fifth.

2 66 igfahan has been included among those entered twice:

although not listed by Qu'fcb al-Din, its existence Is
recognized by him.
h y p o th etica l sc a les liste d by g a fi a l-D in • sc a le s liste d by Qut;b a l-D in sc a le s liste d by

^ fit*
 0
o
)
CD
l->
CD
CO

H*
00 ro
ct* v>
j
CD ro

CJ1
V)
P
i
-*}
HI

CD
HJ
Io
t
HI
3

H-
CO
c+
C
D
P'

O'
<<!

■cl*
o'
•v
p
I—1
1
t.;
HI
P

H CD
P O
ct*CD
CD H*
CD
hi
00
hi
H-
M
H-
ct*CD
CD ct
hi CD
CO fb
oJ
 (.233;

If this table is at all accurate as a statement of

diachronic relationships, it is valid only with regard to

the sequence of events through which the modal system expanded,

It should not be thought to give an accurate indication of

the dating of these events* The fact that Qufcb al-Dxn records

more than twice as many modes as gafi al-Dxn cannot be taken

to mean that the modal system doubled in extent in a mere

half-century or less: gafx al-DIn himself makes it clear in

both of his treatises that the modes he incliides are only the

best-known. Thus although It would seem likely that some of

those listed later were already in use during his lifetime,

we have no means of determining how many and which; nor of

course can we tell whether there were further modes, familiar

to him but soon to be discarded. It is possible, as has been

hinted in connection with his non-recognition of 12 foit

jazx

/1 23 3 4-/, that £>afx al-Dxn regarded certain changes

being adopted or even already firmly established in his day

as unacceptable innovations: it would certainly be consistent

with his position as an influential and highly respected

court musician for him to adopt a steadfastly conservative

attitude, especially towards the novelties of younger rivals.

It is perhaps worth mentioning in this respect that although

his second treatise, the risala al-sharafiyya, was composed

after the sack of Baghdad and while he was in the service

of the Mongol II-Khans, it betrays no outside influence,

Mongol or Turkish. Even the E mode, listed in the durrat

al-taj and the sharji, where It is stated to be in common use

 ^254J

1
among the Turks, is absent from it: yet the Turks had long-

exerted a cultural, as well as a military, influence in

Baghdad; and it is clear that this influence must have

extended to the music of the court, resulting in the addition

of new elements and, through the demands of taste, the

favouring of certain existing elements at the expense of

2
others.

A corresponding synchronic statement of the patterns of

unit distribution obtaining in the corpus of fixed scales

listed by Qufb al-Dln may again be presented diagrammatically.

1/1 2t? 3b 3^/ &nd 4 /1 2t> 3b 3/ are entex'ed at

the. same points as the units to which they may be related -

25 /1 2t 5b 5$ zl-il 5/ and 11/1 2t 5b 5 4/

respectively. 10 /1 2'b 3^ 4-/ and 12 /1 21> 3 4/ are

held to constitute one composite entry, andin addition no

distinction is di^awn between /1 2't? 3^/ &nd /1 21? 3^ 4-/.

66 igfahan G A Bt c d e’fe f fit g is included, v/hile

45 6iraq G At? BT c d? e? f (fit) g is subsumed under

^ *iraq G At? B'frc dt? e^ f g. The non-octave scales

(with the exception of 30, 31? 38 and 39? to be discussed in

the following chapter) and those exceeding the octave are

also included.

1 BM. MS. Or. 2361, fol. 103-

2 The earliest example of a new addition from this source

is in fact furnished by the durrat al~tarj , in which one of
the rhythmic modes is called turkl Cb M. MS. Add. 7694, fol.
239). ~ ~
 /

/1 2 5b 4/1 2 ygj/l...

/1 2b 5b 4/1 2b 5b 4/T/
/1 2 5 4/1 2 5 4/T/
/1 -2b 5b 4/1 2b 5b 4/T/1 2b 5b V
/T/1 2 5 4 / 1 2 5 4/
/1 2b 5b 4/T/1 2b 5b 4/________
> <

iV1 2 5b 4/
/1 2 5 4/r
/1 2b 5b 4/T/1 2b 5b V
/1 2 5b 4/'T/1 2 5> 4/
i \

/1 2b 5b 4/1 2b 5b 4/T/ /1 2 3b 4/1 2 3b 4/T/

/1 2b 5b 4/T/1 2b 5b 4/ /1 2b 5b 4/3 2 3b 4/T/ /1 2 5t 4/T/1 2 3b 4/
/1 2b 5b 4/1 2b 5b 4/ /1 2- 5~b 4/SS/1 2 3b 4/
CM

/1 3b 4/1 2b 5b 3b/
/1 2b 5b 4/1 2b 5b 3b 4# 5/ /1 2b 5 4 4# 5/(1 2 3b 4/)

/1.2b 5b 5 4/T/1 2b 5b V /1 2 5b 4/T/1 2b 5b 3 4/

/1 2b 5b 4/1 2b 5b 5 4/T/

/1 2b 5b 3b/
/1 2 M 4 4# 5/
/1 2b 5b 5b 4ft 5/

/1 2b 5 4 4# 5/1 2b 5b 5 4/

/1 2b 3b 3/
/1 2b 3b 3 V

>fc-
/1 2b 5 4 4# 5/1 2b 5b 5 V1 2b 5 4/
/1 2b 3b 4/1 2b 3b 3b/
/1 2b 3 4/1 2b 3b 3b/ /1 2b 5 4 4# 5/1 2b 5 4/f

/1 2b 3b 4/1 2b 3b 3b 4# 5/

/1 2b5 4/T/1 2b 5b 4/
/1 2b5 b 4/1 2b 5 V T / /1 2 3b 4/1 2b 3 4/(T/)
/1 2b 5 4/T/1 2b 5b 4/1 2b 54/ /T/1 2b 3 4/1 2 3b 4/
/1 2b 5b 4/1 2b 5 4/T/1 2b 5b 4/ /1 2b 5 4/ /1 2 3b 4/T/1 2b 3 4/
/1 2b 5 4/T/1■■I2b I5b I■4/T/ /1 2b 5b 4/1 2b 5b 4/T/
 Given that /1 2t 3 4/ and /1 2t 3t4/ aredifferent

manifestations of the same unit, it will heseen that the

only asymmetrical feature of this diagram is occasioned by

the placing of 55 variant of buzurg G At B(t) c cif d et

f f8 g (the only 9-note octave scale) and the related 7"1

buzurg-i kamil G AtB(t) c c$ d et f f8 g at b c' •

In contrast to the previous diagrams, which could be

variously arranged, the position of the outer triangle of

Zalzalian units is now fixed, and is mirrored by the triangle

of diatonic units- 'They are connected by a regular

progression:

/1 2 31? V

k'
/'\ 2b 3b 4/ /1 2 3 4/

/1 21? 3b 4/ /1 2 3t 4/

/1 2$
2$

(/I 215 3 4/)

The broken arrows show a progression by a semitone

 (237)

lowering of one note, the others by a quartertone lowering.

The particular direction chosen (up or, as here, down) is

without significance.

One may also note the balance within the Zalzalian

triangle, where the one 5’“no‘te tetrachord is flanked by the

two indivisible pentachords.

It is clear that an arrangement of this type is to be

expected in a system where the combination of units is

governed by principles of consonance, for the above progressio

juxtaposes units which show the greatest degree of similarity.

Nevertheless, it should be remarked that the lines joining

the diatonic and Zalzalian groups are to some extent

exceptional when viewed in the context of the historical

development of the modal system. As has been seen, §afi al-

Din makes no reference to the scales they represent; and

later treatises mention only one or two combining /1 2 3 4/

and /1 2 3^ 4/• In effect, the line linking these two

units marks an escape route from the presumably unsatisfactory

50 kardaniya G A Bt c c$ d e ft g. On the other side

of the diagram 47 jiusayni G At Bb c d eb f g is found

only in the durrat al-taj, and if the diagram were adapted

in order to represent the modes in current use during the

following tv/o centuries the link between /1 2b 3b 4/ and

2t 3b 4/ would disappear.

Thus the connection between the two diatonic units

/1 2b 3b 4/ and /1 2 3b 4/ and the others is somewhat

tenuous, and it is interesting to note that by the time of

al-Ladhiqi the names attached to them by gafi al-Din and

Qufb al-Din are used to denote other units, busallk referring

to /1 2 3b 4 5/ an<i nawa to /1 21? 3b 4/. The change

in the latter, certainly, is hardly surprising, for to judge

by Qu£b al-DIn’s list /1 2 3b 4/ was the least used

tetrachord: it is found in just one octave scale, and its

lack of importance is underlined by the fact that it is the

only unit for which no indication is given of the presence

of a prominent note or notes. Such indication is also

lacking in the octave scale 42 nawa, and the alteration in

the order of presentation of the tetrachord species may be

taken as further evidence: for gafi al-Din the order is the

logical sequence /1 2 3 4/, /1 2 jl? 4/, /1 2b 3b 4/,

but in the durrat al-taj /1 2 jb 4/ is placed after the

other two, and in addition the catalogue of pentachords omits

/^\ 2 3b 4/.

In conclusion we may present a summary of *scale types

akin to that given at the end of the preceding chapter.

Where a scale is identical with one given there the unit(s)

it contains will not be mentioned. 3 0 , 31, 38 and 39 are

 (239)

again excluded from consideration*

To the units found in the modes recognized by gafi al-Din

/1 2t 3 1? 3/ may be added.

We may distinguish:

modes based on a single unit (23b may a - /1 3b 4-/1/;

23b friffar ~ /1 2t> 3b 3^ 4-# 5/; ? 54- buzurg)

modes based on a combination of units:

non-octave scales:
1 2
parallel conjunct units (24b shahnaz - /1 21> 3b 3^/;

33 rahawi-yi tamam - /1 2^ 3 4-/, /1 21? 3b 3/;

34- ifffahanak - /1 21? 31? 4/, /1 2^ 3b 3^/; 36

kuchek-i tamam - /1 21? 3b 4/, /1 21? 3b 3^/;

37 nawruz)

11011-parallel conjunct units (3 5 variant of ifffahanak -

/1 21? 3 4-/, / 1 21? 3b 3^/; 57 zankula)

octave scales:
1
parallel conjunct tetrachords:

diatonic (40 4ushshaq; 41 busalik; 42 nawa. With

1 r
fhe term parallel is used here to refer also to those
cases where all the notes of a four-note unit are reproduced
in a five-note unit, e,g. /1 2t> 3 4/ in /1 2l? 3b 3 4/
When e.g. /1 21? 3b 4/ is placed below /1 2 p 3 b 3^/ in
a non-octave combination the same situation may be held to
arise since the uiiper note of the lower unit coincides with
the lower note of the upper unit.
2 Parallelism in this instance is at the neutral third.
 lower disjunction: 53 variant of kardaniya - /1 2

3 V)

Zalzalian (45 rast; 44 (? and 45) 4iraq; 48 \iusayni;

67 igfahan - /1 21? 3b 4/, /1 21? 3b 5 4/)

non-parallel conjunct tetrachords (46 bijazl; 60 mubayyir

busayni. With lower disjunction: 63 kardaniya

nlrizl -r /1 2i? 5 4/, /1 2 3t> 4/)

parallel disjunct tetrachords:

diatonic (65 - /1 2b 3b 4/)

Zalzalian (49 kardaniya - /1 2 3? 4/; 50

kardaniya; 60 mubayyir busayni; 61 mubayyir

zirkesh - /1 21? 3b 3 4/, /1 2? 3b 4/)

non-parallel disjunct tetrachords ;

Zalzalian (62 nihuft; 64 - /I 2 5^ 4/, /1 21? 5

4/; 66 igfahan)

diatonic and Zalzalian (47 busayni - /1 2t 5b 4/,

/1 2b 5b 4/; 51 variant of kardaniya - /1 2 3

4/, /1 2 3”b 4/; 52 variant of kardaniya -

/1 2 4/, /1 2 5 4/)

tetrachord and pentachord units (? 5^ buzurg; 55 variant

of buzurg - /1 2"b 5 4 4$ 5/, /1 2^ 5b 5 4/;

56 variant of buzurg - /1 21? 5 4 4 ft 5/, /1 2*b

3 4/; 5 8 b bigar - as 5 8 a zlrafkand; 59b (bigar

wa i^f ahanak) - as '59a kawasht)

scales exceeding the octave

add a whole-tone to an octave scale (68 - to 62 nihuft)

add a tetrachord to an octave scale

above (7 0 nihuft-i kamil - /I 21? 3 4-/ to 62

nihuft; 71 buzurg-i kamil - /1 2'b 3 4/ to 55

variant of buzurg; 72 busalik - /1 2b 3b 4/ to

41 busalik)

below (69 nihuft-i kamil - /1 2“b 3^ 4/ to 62

nihuft)
 (242)

chapter 7

As has been pointed out, the information Qu£b al-DIn

provides about the modal system of his time is not confined

to the catalogue of scales transcribed in chapter 2. Much

of the remaining material is compressed in two sub-sections

(mabbath) of the concluding section of the fourth discourse

(maqala) . Neither of these is based directly on gafi al-

Din’s treatises, although references to these are made. In

the first mabbath we are presented with a short commentary

on the modal categories established by gafi al-Din and, far

more important for the present study, a considerable amount

of extremely valuable material on the shu{bas, a group of

modes ignored by gafi al-Din. In the second Qu^b al-Din

goes on to detail a large number of further possible unit

combinations or sequences, from which it is clear that in

practice musicians were by no means limited to the combination;

found in the scales discussed above.

These two mabbaths are of sufficient interest to warrant

giving in full. Quite apart from the intrinsic value of the

contents, which greatly extend our knowledge of the musical

practice of the period in relation to the development of the

modal system, they constitute a representative example of

the sometimes sophisticated, sometimes logically naive types

of argument found in the more interesting treatises of the

Systematist school.

In the following translation, additions to the text are

placed within square brackets. Notes and explanations are

given on the pages opposite.

 (244)

An explanation of the terms

parda, aw az, tarkib and shu4b a .

A parda, as used by practising musicians, can be shown

by induction to consist of a limited series of notes usually

1
bounded by a major consonant interval, and is therefore
2
synonymous with jam4, group. Some groups, however, such as

kardaniya, nawruz, mufcayyir and igfahanak, are called

5
awazes, and others tarkibs, combina/bions, such as the
4
second buzurg octave, said to be a combination of ifffahan
5 6
buzurg, and the third, said to be a combination of
7
frijazi and buzurg.

In the kitab al-adw ar Safi al-Din objected to the term

tarkib. With reference to the first example [i.e. the second

buzurg octave], the objection should be voiced thus: "If

that octave is called a tarkib because it combines the

5
tetrachord igfahan with the pentachord buzurg, v/hy is

zankula not called a combination of 4uzzal and the tetrachord

8
rast, or igfahan-i agl a combination of the tetrachord
9
ifffahan and the pentachord rast?" - and not In the way he

put it, which v/as to ask why rahawi is not described as a

1U
combination of nawruz and tiijazi, zankula [as a combination!
11
of ftijazi and rast, or igfahan-i agl [as a combination! of

ifffahan and rast.12 r[This isj

i because the first two examples

[he givesj are erroneous: rahawi is not a combination of

nawruz and frijazl, nor zankula [a combination] of bijazi and
 (245)

1 Usually octave, fifth, or fourth. But from later

examples it appears that Qu£b al-Din also includes the third

2 gafl al-Din defines jam4 as the association of three or

more notes, the association of two constituting an interval.

3 58 kardaniya G A Bt c c# d e ft g (or 49? omittin

the ct); 37 nawruz G At Bb c dt eb f; 60 mubayyir
busayni G At Bb c d et f g; 54 igfahanak G At Bt
c dt eb et. igf ahanak may be equated with kawasht, one
of the awazat listed by al-Din: these do not however
include 60 mubayyir busayni.

4 55 G At B(1>) c ct d et f ft g.

5 11 igfahan /1 2t 5b 5 4/; 20 buzurg /1 2t 3 4

4t 5/ •

6 56 G At B(t) c ct d et ft g.

7 ^2 bijnzl /1 2t 5 4/; 20buzurg /1 2t 5 4 4 if 5/ •

8 57 zankula G A Bt cdt e f (g); 22b4uzzal

/1 2t 3 4/1/; 8 rast /1 2 3^ 4/.

9 66 igfahan(-I agl) G A Bt c d et f ft g; 11
igfahan /1 2t 3b 3 4/; 16 rast /1 2 3t 4/T/.

10 32 rahawi G At B(t) c dt eb f g; 9 nawruz

/1 2t 3b 4/; 12 bijazi /1 2t 3 4/.

11 See note 8. 12 bijaAI /12t 3 4/;8 rast /1 2 3t 4/

12 See note 9- 11 igfahan /1 2t 3b 3 4/; 8 rast

/1 2 3t 4/.
rast. Similarly with the third example, unless by “igfahan11

he means the tetrachord igfahan and by "rast" the pentachord

1
rast.

V/e shall now state the truth of the matter.

Although technical terms are not subject to controversy,

for since [the individual’s] desires are not at variance with

each other This use of3 terminology will be consistent -

2
indeed, he may establish his own terminology - yet there is no

doubt that to observe a proportion among terms commensurate

with that obtaining among the objects they designate is the

rule obeyed by the wise. One of the requirements of this

rule is that when a convenient word is used generically to

designate a set of items, and a further item shares in this

generic sense with the set, then the term is also applicable

to that item. To use the term specifically for the set, to

the exclusion of the particular item, is to restrict [its

applicability] improperly. Therefore since these groups

_ 3 ___
(jurau4), some of which are called awazes and others tarkibs

(the term p arda not being applied to them), have in common

with the pardas the fact that they are all limited series of

notes bounded by a major consonant interval, there is no

distinction known which could justify not applying the term

if
parda to them.

Even if a distinction is drawn from the fact that the

notes of igfahanak and nawruz are not bounded by a major

5
consonant interval, this objection may be refuted by the
 (247)

1 It could be argued that Qufb al-DIn rejects £>afl al-Din3£

analysis of rahawi simply because he does not recognise the
latter3s version ox this mode. In the other two cases,
however, it is clear that his objection is to £>afl al-DIn3s
(perfectly straitforward) ellipsis whereby only the two
constituent tetrachord units are mentioned, the reader being
left to supply the position of the disjunctive whole-tone
himself.

2 The rendering of the beginning of this paragraph is

somewhat free: Qu^b al-Din resorts to Arabic phrases and
cliches which do not greatly further his argument.

3 This remark is a clear indication that in ordinary usage

the term parda was not used indiscriminately of any mode
or scale, but corresponded to §afi al-Dln3s term shadd (pi.
shudud).

6Abd al-Qadir confirms this, remarking that what the

Arabs call shudud are known as pardas or m a gains in Persia.

4 Commenting on this argument, *Abd al-Qadir observes that

’^^ie pandas are all consonant combinations of intervals,
whex*eas according to Qu£b al-DIn3s definition a dissonant
scale could equally be termed a parda (Bodleian MS. Marsh
282, ff. 45v-46).

5 3^ igf ah an ak G At? c dl? eb et?; 37 nawruz

Ct A1> Bb c d1? eb f.

The former has a range of a neutral sixth, the latter of

a minor seventh.
 1
example of rahawi-yi tamam, which contradicts it, and on

the other hand by the example of kardaniya and mubayyir

2
among the awazes.

If it is said that kardaniya, for instance, is one of

the forms taken by igfahan when transposed, as has been

3
stated by 5a-fi al-Din, v/e may reply that bus all k is also

one of the transposed forms of *ushshaq, and therefore it

4
too should not be a parda. The truth of this proposition

has been demonstrated previously, [when it was shown] that

even if a transposed form of one particular scale coincides

with another the two do not necessarily produce an identical

effect on the listener; nor is one necessarily derived from

5
the other.

If it is said that in these awazes it is standard

practice to begin the composition on the highest note, this

[statement] is not valid for igfahanak, nor for mubayyir,

as may be seen from some of gafi al-DIn5s compositions. In

refutation one may cite busayni and the pentachord bdgar,

 (249)

^ 53 rahawi-yi tamam G At B c d? eb e.

2 50 kardaniya G A Bt c cif d e ft g; 60 mubayyir

busaynl G At Bb c d et f g.
This paragraph contains a damaging admission, for once
Qufb al-Din concedes that some of the scales he wishes to
class as pardas (i.e. igfahanak and nawruz) do not fulfil the
condition of being "bounded by a major consonant interval"
his case is seriously weakened. To get round this difficulty
one can hardly speak of a refutation - he first cites a parda
(r ailawl -y i t am am) with a range of a major sixth, not a "major
consonant interval", and then two non-pandas which are octave
sc al e s .

5 In the kitab al-a&war (BM. MS. Or. 136, fol. 22v).

4 In theory a scale could be transposed to any one of the

seventeen notes of the Systematist gamut. If, as a result,
its notes coincided with those of another scale (not
transposed), the latter might be thought to be derived from
it despite the fact thatthe tonic would not be the same.
Thus 66 Isfahan /1 2 4/T/1 2‘b 3 b 3 4/ in the 17th
fabaqa is d e f-t g a b1? c' c#' d' , the notes being the
same as those of kardaniya (see note 2). In the scale "derivec
in this way G (or g) remains the tonic. 41 busalik G Ab Bb
c db eb f g can be similarly'derived" from 40 6ushshaq /1
3 4/1 2 3 4-/T/ in the 5 th fabaqa: eb f g ab bb c'db'eb'

5 In the passage alluded to (BM. MS. Add. 7694, fol. 232)

Qufb al-Din approaches the question from a different angle,
observing that by taking the second note of 40 6ushshaq G A
c d e f g as tonic 42 nawa /1 2 3b 4/1 2 3b 4/T/
results, from which it follows that within a given sequence of
notes modal identity is determined by the position of the tonic

6 34 igf ahanak G A1? B1? c d1? eb e1?; 60 mubayyir

busaynl G AP Bb c ‘ d 'et f 'g*
 which disprove [the statement^.

If various other objections are made, the verdict on

them will become clear after careful investigation. The

i first, indeed essential, step is that they should all be

called pardas . gafx al-Dxn’s objections to this are in

fact invalid, since we have previously demonstrated that a

number of groups may share the same name while differing in

range: thus Safi al-Din designates both a tetrachord and

2
an octave scale by the name rast. Similarly with the rest
5 4
of the ten species v/hich he first of all listed thus:

(1) 1ushshaq 7 7 B G A B c

(2) nawa 7 B 7 G A Bl? c

(3) busallk B 7 I G Ab Bb c

(40 rast 7 J J G A Bt c

(5) nawruz J J 7 G At Bb c
5
(6) ciraq J 7 J G At Bt c

(7) Isfahan J J J B G At Bb B c

(8) buzurg J 7 J J B G At B(t) c

(9) zirafkand J J B G Ab Bb Bt

(10) rahawi J J J G At Bb B

3
He further completed these species with a pentachord or

tetrachord end entered them under the same names in the list
6
of pardas .

7
From the objection voiced above it is known that

according to Safi al-Din the upper pentachord in zankula

 C251)

1 48 busayni G At Bb c dteb f g; 25b frigar G At

Bb Bt ct d. Qu^h al-DIn?s rebuttal again takes the form
of citing awazes which do not conform to the statement, and
pardas which do. That 25b higar
*»HIIP.L................................... rw».—,.p| should be considered a |
. I n| parda
fr-n l.lll.lIIJ■II
in this context is somewhat surprising.

2 rast denotes /1 2 3t 4/, /I 2 3t 4/1 2 jt 4/T/,

and also the pentachord /1 2 ?t 4/1/.

These strictures on gafi al-DIn are difficult to justify,

gafi al-Din nov/here discusses whether all scales (and groups)
should or should not be called shadd (= parda), and hence
makes no objection to the proposal. The objection quoted
above is concerned solely with the use of the term tarklb.

3 aqsam, lit. "divisions".

4 This list is taken from the risala al-sharafiyya

(Bodleian MS. Marsh 521, fol. 90v). It will be noted that
bijazi is not mentioned.

5 Bor Qutb al-Din this is ru-yi *iraq, *iraq being 2 /1

^ 3t/.

6 This statement is inaccurate. zirafkand G At Bb Bt

and rahawi G At Bb B do not in fact occur in the list of
pardas (i.e. shudud) supplied by gall al-Din. The scales
resulting from the addition of a tetrachord to these two units
are, respectively, 36 kuchek-i tainam G At Bb c dt eb et
and 33 rahawl-yi tarnam G At B c dt eb e, and not the
octave scales zirafkand and! rahawl given by gafi al-Din. ’

7 On p. 244.
 1
is frinazi, that the lower pentachord in igfahan is
*2
rast. Since the names are used like this, why may one not

describe the scales both ways, so that the octave zankula

is called both zankula and a combination of ‘uzzal and the

3
tetrachord rast, and the octave igfahan-i agl is called

igfahan-i agl and a combination of the tetrachord Isfahan

and the pentachord rast?

As for the question put by gafi al-Din in the kitab al-

5 r *i
ad wa r : "Why does he who subscribes to this [viewj not say

that such-and-such a mode is a combination of such-and-such

6
{units]?", he answers it by discussing combinations in the

manner reported above: however, his remarks do not accord

7
with what v/e have stated.

[the shu* basj

As for the shu* bas, for practicing musicians there are

nine of these, as is v/e11 known:

dugah, segah, chahargah, p ant

jgah, zawlx, ru-yi ‘iraq,
8
mubarqa* , maya, and shahnaz.

^ shu* ba can also be defined by induction as a specific melodic

movement upon the notes of a [given] parda, scale. This

statement may be clarified thus: the masters of the practical

art progress through the notes of the parda in such a way as

to place emphasis on a certain note. (This has been alluded to

9
in the sixth mabbath.) As a result this note predominates,
 (253)

1 57 zankula G A Bt c d1? e f (g).The upper

pentachord is 12 frijazx /1 2t J 4/ (c - f) plus whole-
tone, i.e. 22b *uzzal.

2 66 Isfahan G A c d et ff$ g; 16 rast

/1 2 3t 4/1/ •

3The lower tetrachord of 57zankula (see note 1) is

8 rast /1 2 3^ 4/.

4 The upper tetrachord (d - g) of 66 ifffahan(-i agl)

(see note 2) is 11 ifff ah an /1 21? 3^ 3 4/.

5 BM. MS. Or. 136, fol. 22v.

6 Qutb al-Dxn here repeats in abstract form the passage

from the kitab al-adwar quoted on p. 244 together* with the
preceding ptoase.

7 The text of the last phrase is corrupt, and the

translation therefore conjectural.

8 Bor £>afi al-Dxn may a and shahnaz ax%eawazes. According

to Qu£b al-Dxn5s critexda the terms awaz and shu* ba are
not necessarily mutually exclusive, but the possibility of
his assigning these two modes to both categories may be
discounted.

9 Bee note 5> P« 249*

 j

together with those that are most consonant with it* If

1
this note is the lowest note (or tonic]! (mafru<ja) of the

parda, or the fourth upwards from the tonic, the melody is

said to be in that parda. For instance if, in the parda

2
rast, the prominent note is c, then the melody is in rast,

absolutely.

If the second note from c, i.e. d, Is prominent, then

it is said to be in dugah-i rast. Further if, in progressing

upwards, the note g is reached, or if A is reached downwards,

3
the melody will indubitably produce the effect of nawruz;

if fit is also used It becomes Isfahan; and if one goes

beyond g to a it becomes busayni.

In this case the name dugah is used when the melody both

begins and ends on d, the second note [in relation to c]|.

Originally, indeed, the [melodic] progression in dugah may

only have been on the notes close to d, i.e. without reaching

the fourth from d, so that the impression proper to rast was

not transformed into that created by nawruz. Subsequently,

however, artistic licence and daring led musicians to extend

its range, with the result that all the melodies composed in

dugah-i rast have the effect of the above-mentioned pardas,

the only difference being that in those pardas it is

customary to begin on the highest note and to end on the

lowest, whereas in dugah d is both the initial and the final

G
note.
1 Equivalent to thaqil alnmafruflat, the term used for
proslambanomenos of the systema teleion.

2 A fourth above the lowest note, or tonic, G, rast

being 4$ G A Bt c d et f g.

5 Both d et f g and A Bt c d are 9 nawruz /1 2t 5b

4/, while combined they form 57 nawruz /1 23 5b 4/1 2t
5b 4/.

4 d et f f'if g is 11 ifff ah an /1 2t 5b 5 4/,

5 d et f g a is 18 feusaynl /1 2t 5b 4/1/, while

A Bt c d et f g a is 48 jyusayni /1 2t 5b 4/1 2t
5b 4/1/.

6 In accordance with this description v/e may represent

dugah as

75 dug~ah

=•■■!> f,- , o t T =
^^ *»* _ '_ ^ ” '

In this and a number of further transcriptions the

position of the initial and final notes (her*e d) is not
indicated. It must be borne in mind that dugah does not
have a fixed range: the dots therefore indicate that
further notes from the parent scale 45 rast G A Bt c d
et f g might be added. The dots fulfil the same function
below.
 The case of segah is analogous. Thus when its range is

extended it sometimes gives the impression of 6iraq if f# is

1 -
not used; sometimes of igfahanalc if fi} [or g p \ is used,
2
together with d, c, and 331?; and sometimes of ^zrrafkand-i]

kuchek if these three notes are omitted.

chahargah is also analogous, except that the impression

it evokes is similar to that of rast. This is because of

the prominence accorded f, a fourth from c. As a result of

this consonant relationship c is swiftly established [as an

important note! when chahargah is heard, from which it follows

»■— in i i 1.1 - * n ^ . *

that other notes are perceived in relation to it, the cause

being nevertheless £the prominence of] f. The effect produced

is similax' to that of rast, especially when c often recurs.

However, should c seldom recur, and d often (because of the

[consonant] minor third relationship between f and d ) , then

the d will impress itself upon the listener when he hears

chahargah, and for this reason its notes will create an

effect corresponding to that of dugah, especially when d is

4
made the final note.

panjgah is again similar, except that the effect it

produces is even closer to that of rast (the consonance of

1 el? f g is 2 /1 2t 3V\ 1iraq according to Qufb al-
Din. The form of the statement implies that fl? was
frequently added, at least in 43 rast.

2 Bt c d et f g is 34 igf ahanak /1 21? 3"^ 4/

/1 2? 3^ 5^/* Considered in relation to a mode the
units of which begin on B*l? and et, the value of the
theoretical notation g“c need not necessarily have been fS,
which relates strictly to the octave G - g. gt is a minor
third above ei?.

3 e1? f g1? g is 1 zlrafkand-i kuchek /1 3^ 3"^/

(see note 2).

From these remarks we may derive the notation

74 segah

4 In 43 rast G A Bt c d eV f g, c is prominent.
Hence the 'feffect produced" by chahargah, based on the upper
tetrachord (c - f) and according prominence to f, will tend
to coincide with that of the lower tetrachord of rast. It
might be thought that d was sometimes made the final note in
order to avoid this resemblance; but if so one would expect
something similar to occur in pant
j gah.

We may notate chahargah as

75 chahargah
g and c being more perfect than that of other notes),

especially when c is made the final note, as is the custom

1
among present-day musicians.

It will be apparent from this discussion that in the

usage of a number of musicians "absolute11 rast is also one

of the particular melodic movements in the scale (parda) of

2
rast. The same approach can be aioplied to other scales,

so that for each parda there may be a dugah, a segah, a

chahargah and a pan{

jgah. This, however, is not common
3
usage at the present time.

It is also clear that if this approach is applied to

the notes A, Bt?, c and d considered in relation to G the

same forms, with the above names, will occur. However,

since it is customary in the parda rast to accord prominence

to c - as will subsequently become apparent - Q;he occurrence

of these forms in] the positions [described] above renderjs

their occurrence] here superfluous.

Another point is that in, for instance, the scale of

nawruz the shu* ba dugah gives the impression of ciraq; the

shu* ba segah that of "absolute" rast; the shu6ba chahargah

that of nawruz itself $ while the shu‘ba paipjgah is not used

if G is made the final note, since the interval G - dt is

5
not consonant. Thus the shu6bas of rast enable us to

dispense with the shu* bas of nawruz, and similarly with

1 From this v/e may deduce

76 panjgah

■. " n .. . p y
n ... r 1 i *
// H ’ V
K
--- -

panjgah thus corresponds to the upper pentachord of 43

rast, giving prominence to the octave note, g.

2 Qu£b al-Dxn here makes a distinction between mode end

scale. Having pointed out that rast has in effect become
a parent scale from which various shu*bas are derived by giviz
prominence to successive degrees, he now underlines the fact
that the mode rast also has characteristic features (at
least of prominence) which distinguish it from a mere abstract
succession of notes*
mode
3 A purely theoretical digression. Another/could
presumably be turned into a parent scale provided that the
derived scales conformed in intervallic structure to the
principles underlying the whole system (thus 54 buzurg G At
B(t) c cfi d e ft g would not be a possible parent scale)
However, as Qutb al-DIn makes clear in the next passage, the
only scale thus used was 43 rast G A Bt c d et f g.

4 I.e. it would be possible to base dugah , segah,

chahargah and pan(
jgab on the lower tetrachord also.

5 In 37 nawruz G At? Bb c dt eb f dugah will be ...At

Bb c ..., equivalent to 2 tiraq /1 2t 5t/, or ... At Bb
c d t ..., equivalent to 10 *iraq /1 2t 3t 4/; segah will
be ... Bb c dt e b ..., equivalent to 8 rast /1 2 3t 4/;
and chahargah will be ... c dt eb f , equivalent to 9 nawruz
/1 2t 3b 4/. In panjgah dt'would be prominent.
 Ubo;

1
those of *iraq.

[zawli]

-^n SQ^Q-h the final note may he d, lightly touched, or c,

in v/hich case it is called zawli; and since c to et

constitutes a group (jam*) according to the conditions which

define a group, it is not an error if v/e term zawli a parda,

2
in the same way as v/e termed the awazes pardas .

[ru-yi ‘iraq]

If Bt is made the final note [of segah] it is called

ru-yi *iraq, and there is even greater reason for considering

5
ru-yi *iraq a

[rabbi*]

rakbi is a particular melodic movement on the notes of

zirkesh fcusayni such that one begins from B and ends on G.

4-
c is little used while B is prominent.

jmubarqa*]

mubarqal is derived from igfahanak. It is characterized

by the frequent rectirrence of dt, and by the fact that the

range rarely extends beyond one or two notes on either side

of it. c is the final note, or B t , briefly touched -

although practising musicians may be daring enough to draw

5
out this note.
1 In 44 1Iraq G At Bt c dt et f g, dugah will
coincide with 8 rast / T 2 3^ 4*/; segah with 9 nawruz
/1 2t 3b 4-/; chahargah with' 10 *iraq /1 2t 3t 4-/;
while dt is again prominent in panjgah.

2 This mode has already been listed as

3 zawli a

It is here thought of as c d et. Presumably, as with

segah, the range covild be extended.

3 Again, this mode has been listed as

10 ru-yi 4iraq
.. fr

Here it is thought of as Bt c d et. Qufcb al-Din’s remarks

underline the relationship between 10 *iraq /1 2t 3t 4-/
and the rast scale discussed in chapter 5-

^ zipkesh frusaynl is 19 G At Bb B c d. One may

therefore represent rakbi as

77 rakbi

b— ■ - i>— v
?*- JT

According to later writers rakbi consists essentially of the

notes G At B b , to which others could be added for embellish-
men^ (tazyln). It is not among the nine shu{bas listed at
the beginning of this section.

5 igf ahanak is 34- G At Bt c dt eb et . Hence mubarqa*

may be represented as

78 mubarqa*

■« 1 “ ■' — '-4 e - ^ r o

Although somewhat different, later accounts agree on the

importance of dt and c in this mode.
 As for salmak, gafi al-Din has suggested that it is

[derived fromj zankula, in spite of the fact that he counts

„ 2
it among the awazes, v/hich he does not consider to be pardas*

What he says is quite self-contradictory, and the example of

3
salmak given in his list bears no relation to zankula. In

fact, as may be ascertained from his own compositions and

from those of present-day artists, it is a particular melodic

movement in that parda [i.e. zankulaj such that c often

recurs while the [type and range ofj movement is similar to

that in mubarqa* . One then progresses from c through

pant
j gah, the final note being the final note of rast - i.e.

the c of zankula is assumed to be the g of rast, so that one

L\.
ends on the c of rast.

[maya and shahnazj

With regard to maya and shahnaz, gafi al-Din has stated

that they are specific melodic movements. From his remarks

5
it is evident that they are not in any particular parda,

for he states in the kitab al-adwar: "As for maya, it is an

ascending and descending movement; similarly shahnaz11.

Also, in that the tv/o are counted among the awazes, what he

says here is self-contradictory, for the awazes are included

by him among the octave scales (adwar). Ibis is stated in

the kitab al-adwar: "Some of the octave scales are called

7
awaz". Ihus maya and shahnaz should both be octave scales,
8
and as such would not be limited melodic movements. Nor is
 I2b3)

1 In the kit ah al-adwar (BM. MS. Or*. 136, fol. 22v).

2 zankula is a parda (or shadd). This is a curious

objection: Qu£b al-Din says nothing against the previously
quoted identification of kardaniya (an awaz) with igfahan
(a shadd) .

3 Not so: salmak is given by gafi al-Din as c d e ft

g al?, notes present in 57 zankula when transposed to the
^7th j^abaqa: d e ft g at b(t) c' (d/ ). What lies
behind Qufb al-Din’s objection here is probably the fact that
salmak is not an octave scale, i.e. that the notes of zankula,
when transposed, are not all to be found in it.

4- We may therefore notate salmak as

79 salmak

f5.... ** ) v* o k-,.. X ....... .

1 7 V* P#
’ ” 9

salmak, like rakbi, is not one of the nine shucbas listed at

the beginning of this section. For gafl al-Din it is an awaz.

5 I.e. they cannot be related to (nor be considered derived

from) other modes in the way that dugah is related to rast
or r akbi to zxrkesh fasayni.

6 BM. MS. Or. 136, fol. 22v.

7 Ibid.

8 However interesting Qu^b al-DIn’s arguments may be, they

are by no means logically sound. From "some octave scales
are called awaz11 it does not follow that all awazes are octave
scales. A limited melodic movement might also relate to an
octave scale, although one may presume from what Qu^b al-Din
says that modes thus characterized did not in fact occur in
practice.
the real nature of this melodic movement apparent from the
1
way he represents them in his list: this is contradicted

by all the well-known compositions in maya., by g>afx al-Din

as well as by others, for they are without excex^tion melodies

based on the notes of a pentachord species as shown in our

2
list. The same is true of shahnaz, although compositions
5
in this mode are few, for the reason alluded to. On the

basis of the above remarks we may count both of them among

the groups Cjumu6 ) and call them pardas.

There is much disagreement among practising musicians

over the shu*bas and tarkibs because of their lack of

discernment. It may often happen, in accordance with usage,

that a particular melodic movement develops out of a [given]

scale (parda). If this should become a fixed form, they say

4-
that it is a parda. They are confused about its true nature,

in spite of the fact that the tonic snd the other notes are

the same [as in the parent scale]] - all this because they
5
fail to distinguish essences from accidents. The ordinary

musician, who has but little aptitude for differentiating

concepts, often makes mistakes of this kind. Investigation

6
[into such matters] falls to another group of people: hence

if a practising musician should entertain doubts about some

of the ideas expressed in this mabbath he ought not to voice

his disagreement without first reflecting carefully, for it

may lack coherence. Each person has a role for v/hich he is

fitted.
 (265)

1 This is rather unfair: §>afi al-Din at least suggests

a melodic contour in the way he notates shahnaz - G At Bb
Bt At G. Qutb al-Din provides no such information, and in
addition does no more than quote gafi al-Din5s terse verbal
definition of the nature of the melodic movement involved,
and then only in order to criticize.

2 I.e. 25b maya G Bb c d. gafi al-Din gives maya as

25a G Bb c eb f.

3 Qutb al-Din lists shahnaz as 24b G At Bb Bt c dt d

Unfortunately there does not appear to be any previous
remark giving a reason for the paucity of compositions in
this mode.

4 Whereas Qutb al-Din would term it a shu* b a . It would

seem that he maintains a distinction in general between
on the one hand pardas, awazes, and tarkibs, all of which he
insists should be termed pardas, and on the other shucbas,
melodically limited derived forms. The four shu* bas he also
terms pardas are zawli and ruyi 1iraq, with regard to which
no particular melodic features are mentioned; and maya and
shahnaz, which cannot be derived from already existing pardas

5 The philosophical terminology is conventional. It

nevertheless brings out clearly the assumption that the
shufcbas, modes characterized by, in most cases, a specific
melodic movement, are secondary phenomena - incomplete
manifestations of the abstract scale embodying the notes
used in them.

6 I.e. philosopher-theorists such as Qu£b al-DIn himself.

 It should he realized that each mode possesses a

particular form that presents itself to the mind. In

certain circumstances this form is not attained by [using]]

all the notes of the scale, although it may depend on this

for its perfect realization. The mode zankula is a case

in point: when c often recurs and there is an ascending and

descending melodic progression on the notes c dt Bt A

the characteristic form of zankula becomes manifest,

2
especially when e is itself added. gafi a l - D m ’s statement

identifying salmak with zankula relates only to this, [a

fact! which the above-mentioned objections do however take

3
into account.
 This last remark may be no more than a quibble - an
oblique reference to another philosophical commonplace,
the distinction between potential (bi ?1-quwwa) and
actual (bi *1-fi*1).

This description is well illustrated in the composition

from the durrat al-taj transcribed in the following chapter,
a short section of which is in zankula.

See notes 2 and 3» P* 263*

This grudging admission means in fact that the only

valid objection (from Qutb al-Din9s point of view) that
may still be levelled against the identification is the
rather curious one about the difference in modal category.
The definition of zankula given here is in effect virtually
identical with that of the first section of salmak (before
the. modulation into panjgah).
 On the mixing of modes.
1
Further remarks on well-known modes (maqamat).

It should be known that these groups (jumui ) and shu* bas

are interrelated, and that to pass from each one to another

related to it in the course of a composition is a source of

greater splendour and freshness. The relationship may be

[between modes] on the same base (markaz), i.e. the tonics

(mafruda) are at the same pitch level (frabaqa) in both, or,

as will be indicated below, it may be [between modes] on two

bases, i.e. there is an interval between the prominent notes

2
of each. In the latter case the appropriate position of

each, in relation to the other, may be above or below. One

example is the major third species rahawl with the pentachord

tiusayni, another the major third species rah aw! with the
3
tetrachord nawruz: between the tonics of each is a fourth,
4
with rahawi above. Likewise with the combined groups
„ - 5
(jurnu*-irnurakkaba) established in the list, when an interval

is postulated between the tonics [of each member] of these

and their [respective] positions above and below are defined.

For the sake of example we shall now indicate some of

these relationships. The others may be entrusted to the

ingenuity and insight of those who have delved into the

practical art.
1 The use of the term magam here is of no particular
significance. It should not be considered merely another
equivalent of shadd, a narrower meaning which it acquires in
later treatises.

2 and 3 It would appear that in this passage no distinction

is maintained between the expressions "tonic" and
"prominent note".

4 4 rahawx /1 2t> 3b 3/; 18 jiusaynl /1 21? 3b d/T/;

9 nawruz /1 3b 4/.

Assuming the tonic of the lower unit to be (a conventional)

G, we may obtain from this description

80 pusayni//rahawl

- „
V
— H'
3>— a fe- ^ * -.

81 nawruz//rahawi

„
V f r ..Jp.....
A --- :---

3 The reference could be either forward, to the list given

in the following pages, or backward, to the list transcribed
in chapter 2. If the latter, Qutb al-Din could presumably
include in the term "combined groups" all scales exceeding a
fifth (3 0 -7 2 ), since they may be analysed as unit combinations,
although it is more likely that the reference would be
specifically to 30 (segah wa bitiazl)? 31 (segah wa jiusayni) ,
38 (chahargah wa bijazi) ? and 39 (chahargah wa ifffallan) •
 List of the relationships of pardas and shu4bas

1
Relationship on the same base:

busayni rahawi

busayni kuchek

bus ayni fri.j azi

nawruz
1 It may be assumed from the above reference to
combinations of species of third, fourth, and fifth that
the names which relate to both a unit and a combination of
units (usually an octave scale) denote in this and the
following list only the former. The assumption is confirmed
below by Qutb al-Din.

Entries marked by an asterisk are those in which the

relative position of the constituent units is not absolutely
certain. They will be discussed below.

82 (18A)
A

83 (18/1)

84 (18/12) J -fri 1>»-

TT ■
W .. ■
85 (18/9) '. T t -p. -i><

L • 0
J >-e-- i*--------------- 0 w .- E -------- o-J* f>

87 (18/11)
4 — 1 v * 0 — -
busayni 1uzzal

busayni rakbi

busayni dugah

Isfahan dugah

bips-zi kuchek

dugah

shahnaz bigar

shahnaz buzurg

buzurg maya
 (273)

6 -
88 (18/22b) j, 0 -y. >^ ==5: Jtr

89 (18/77 )* -|- 0 •?. — L ¥ = £

— •-
. 4-
90 (18/73)* qfc;

If* • » -jf*~
> o
91 (11/73)* :±

92 (12/1 ) fc=±

— 5— * — *•— 6--------
.— ( - . ..... ... -. ----ji.«
T>t— ,
•----- -— ---- £— ——
93 (12/73)* i*
-4>----------------------------- t«•

94- ( 2 W 2 5 b ) ^ |>» -4« * [),— $« iff* -

95 ( 2 W 2 0 ) ~^r ..0 -5. I>« * ^ * 1 ^ 1 ----- _Z 3fi._ (.^ 7 = « r ^

96 (20/ 23b) > x (fl* • “sf5*”." b* • 0

— ---- o----- --------------- --------
buzurg rahawi

buzurg

rahawi dugah

rakbi dugah

rakbi igfahan

rakbi nawruz

nawruz isfahan

kuchek segah
 K^(?)

97 (20/4)

98 (20/2 or 1 0 ) o * jz ft* ftp of - ft. ft1

■_,---- ---------- L _ 4 ! -------- —

---- \d * — •------------
. o *• - i ty.—y y;—
.----
99 (4/73)* 4}-----------------------------

Hra f *** * 4ko h« -IT < a (#» ^ • T *"

---- — $— o ---- --- -1> p ---
100 (77/73)* = ---- j »----------- k— ?L_i----
11i

101 (77/11)* X . 0 fr. fr ft. >» fr

102 (77/9)* ---- 1>*----- 8-

3 >_g-4 ~ V . ------- 0----

103 (9/11) ^ -fi \h -ft.- l>» " f

4 *— •--- & — 'P .1.

104 (1/74)

. ft*
* 6 - 0 ft
105 (2 or 10/74)*3 ______cr — yf . « "' V ^
5
 bi,jazi segah

busayni

1ushshaq rast

kuchek dug ah

nawruz dugah

tushshaq chahargah

igf ahan segah

or

igffahanak segah

zankula rahawi

It should be realized that these pardas also are the

1
smaller species (numu6-! gighar).
 V W ( J

-fa 51
106 (12/74)* ± ==?;

107 (l8/23b) 0 ft. ^ *

-r-*-
108 ( 5 or 13 /

8 or 16)

4*--- •— *»-
109 (1/73)* ±

4r
T— * 2f±
110 (9/73 )* ±
■X.-

. ♦ fc* . •*la-£ * w * '».>*■ r

•fl--------err *------- 1 ---- h ; •;.-ftn? ^ |<> \> H
111 (5 or 13/75)* 5j- =53;

112 (1 1 / 7 4 )*

or
. i

(3 4 /7 4 )* -aft* * f *i* F

igxahan and igfahanak are MS. variants.

11 3 (5 7 /4 )* r^-i-fr t -6. 6 >• X ft. h *E

i I.e. species of third, fourth, and fifth, rather than

octave.
Relationship on two bases:

below interval between the above

two bases

nawruz fourth rahawl

nawruz fourth igfahan

fifth kuchek

tusayni three-quartertone

may a fifth dug ah

may a neutral third ciraq

major third busalik

81 (9/A) ------ e— r. >*— -M-
b —
7. 0 ..*'....^ = - g..... —

--------fr, ft----M --- 2-

114- C 9 / / 1 1\)“X;..... .L tf—............ .
I n ---- ft:... ^ 6 . _

w I:---...~£~ — 0---- - v#. * Tw

115 (18//1)'j
; ..

--------^—
or 1 0 ) 7 — — — ft— !--------- — ft .-ft -
*•------

1 Qutb a l - D m gives here the ratio 10 : 9? i*e. a minor

whole-tone.

117 (2387/73)* * ¥• .0
1> o — r-

118 (23b//2 or 1 0 ) * X Z Z = f . ♦

2 Given as 6 : 5i a minor third. This will be discussed

below.

........ . ..*.... * *
119 (5 or 15// • * w A • * 0 *
/!/
> 'W
0 * o • v or W
V V -- —
6 or 14)
 frusaym fifth busalik

nawruz three-quartertone cIraq

mhawi three-quartertone chahargah

1
These two relationships are approximate. The progressions

are common ones.

tA
iraq neutral third kuchek

tiraq neutral third

Now that these relationships have been made clear, it

should be realized that some of the best-known modes (maqamat)

3
consist of such combinations (tarkrbs ) : dugah wa frijazl

for example, which begins and ends with dugah and has a

central section in hhj^azl; dugah wa rahawi, which begins

in dugah, has a central section in buzurg, and ends in

5
rahawr; and humayun, which begins in zankula and ends in
6
rahawi.

The well-known modes (maqamat), in the usage of present-

day musicians, are confines to the pardas, shu* bas, and

tarklbs which v/e have set forth.

 1 - 0 1 - 0
A
W ^a “ fif' “* l>8 v
« ...
120 (18//6 or 14) 1___
/k * J*0
. Hm
t;c
0

-4^0—
.121 (9//2 or 10) d' , h-t “B ?l-S 5 <?? 75 >
fv.®
0'

0-
122 (V/75)*: K—

1 I.e. the three-quartertone is given as 16 : 15» the

ratio equivalent established by gafi al-Bin for category
J intervals.

1 c h^
41
125 (2 or 10//1) -/h ... 44 r> *K 44 - 0 t A
a
w **?>_. 4 ° e* n
w i7* tr"

yj— V-°-d£
124 (2 or 10//18)Zl='"" k1a %* 70-

2 Given in both cases as 5 •

3 The meaning of tarkib here evidently differs somewhat

from the particular sense accorded the related term
murakkab in the kitab al-adwar.

4 This may be notated as

<y U-
125 (75/12/73)* :p g ? =3>«- n

5 This may be notated as

126 (73/20/4)* -Q-

6 From this brief description humayun cannot be

distinguished from a combination given above, 113 (57/4)**
One may also note the similarity between 125 said. 53 (12/73)**
 C282)

The first part of the first mabfrath, an examination of

the criteria according to which the modes are assigned to

the various categories, is in the main self-explanatory. As

various points of difficulty have been dealt with in the

notes, no further discussion of this section seems necessary

here.

Of far greater importance for the present study is the

following section. In conjunction with the description of

shuMgas it contains we may also examine gafi al-Din’s

definitions of maya, shahnaz and salmak. As the name implies

the shu* bas were held by Qu£b al-Din to be derived from

ali’eady existing modes, scales and units. Further, they are

1
stated to be characterized by particular melodic features.

They may therefore be taken to represent a second type of

modal development, quite different to that already examined,

and which we may for the moment term development by

functional specialization. However, It is clear that the

degrees and methods of specialization are by no means uniform

and a distinction may be di’awn at once between a gi’oup

1 As has been seen, exceptions to_these statements,_in

varying, degrees, are provided by maya, shahnaz, zawli and
ru-yi *Iraq.
consisting of dugah, segah, chahargah and panjgah, together

with the apparently related shuchas zawll and ru-yi ‘Iraq,

and the rest.

73 dugah, 74 segah, 7 5 chahargah and 76 panjgah all

relate to 43 rast G A B't? c d el? f g, and are derived

by according prominence to successive degrees of that scale,

starting from the fourth. But, as Qufb al-Drn makes apparent,

the feature of prominence alone is not sufficient to

distinguish them from other modes, especially when their

range is extended beyond the original nucleus of four or

five notes. Their autonomy is ensured by defining the final

note in addition to the prominent note and, in the case of

1
73 dugah, the initial note as well.

They are not, however, unique in possessing such modal

characteristics. In argiiing against the separation of awazes

and par das Qu£h al-Din states that in the awazes, with the

exception of 34* igfahanak and 60 mufcayyir husayni, the

highest note is Initial, while in the pardas, with the

exception of 48 husayni and 25b bipar, it is not. Further,

when discussing 73 dugah, he remarks that in 37 nawruz, 66

Isfahan, and pusayni (all three being termed pardas), the

highest note is initial and the lowest note final. Ignoring

1 In chahargah and panj gab even this would appear to have

been hardly sufficient: the latter Is barely differentiated
from rast itself, while the former is stated^to be either
like rast or, with d as final note, like dugah.
the contradiction involved, it will be seen that we have

here explicit statements on the position ox the initial note

(the highest) for a number of non-shu4bas , and on the position

of the final note (the lowest) for three; in addition, it

would appear from the first remark that in most pardas the

highest note was not initial, and as it would be curious to

find in practice a situation where only one specific note was

thus excluded, it may reasonably be assumed that in these

pardas too the initial note was defined positively, being

presumably either the lowest or some other prominent note.

Thus it would appear that the degree of melodic specialization

dft 73 dugah, 74- segah, 75 chahargah and 76 panjgah was not

significantly greater than that found in most non-shu4bas .

With reference to these four, therefore, no particular

significance should be attached to Qu£b al-DIn’s introductory

definition in terms of "specific melodic movement": we may

remember that in the course of discussing them he concedes

that 4-3 rast itself could be described in exactly the same

way - in other words that there is no significant variation

in the extent of melodic specialization between 43 rast and

the shu4bas derived from it.

These may also include 3 zawli and 10 ru-yi *iraq. Du^b

al-DXn relates both to 74- segah, and they are similarly

defined by the position of the final note. Although both

are included in the list of species, being considered species

of third and fourth respectively, It is likely that, as with

 1
the others, the range could be extended, in which case

a distinction should probably be drawn between the shu4ba

ru-yi 4iraq and the unit 10 ru-yi 4iraq /1 21? 31? 4/

2
considered as an element of combinations.

From Qutb al-I)In5s remarks it would appear that 73 dugah

and 74 segah had acquired a greater degree of independence

from the parent scale 43 rast than 75 chahargah and 76

panjgah. The suggestion of a progressive development is

borne out by the relative frequencies of occurrence of these

shu4bas in the combinations listed in the second mahhath:

dugah appears 11 times, segah 4, chahargah twice, while

panjgah does not appear at all. In Qu£b al-Din’s day,

therefore, panqgah may have been a relatively recent addition.

A century later it would seem to have become reasonably well

3
established, for two forms are reported.

The remaining shu4b a s are 77 rakbi, 78 mubarqa4 , 79

salmak, 23 maya and 24 shahnaz. Qufb al-Din makes specific

statements about melodic movement only for the first three

of these; and if we except the second part of79 salmak (the

1 Although this is not mentioned in later treatises.

^ ru-yi 4iraq appears in the composition transcribed from

cturrat al-taj in the following chapter. Unfortunately
it is restricted to one brief phrase, eP d c B1?, from
which nothing can be inferred as to any characteristic
features it may have possessed (apart from the position of
Bfe as final note).
“V
2 By ‘Abd al-Qadir (Bodleian MS. Karsh 282, fol. 51).
 ^bb )

modulation into panjgah) they show a considerable degree of

similarity in structure: in each case a prominent note

furnishes a centre of gravity around which the melody

revolves, and from which it is rarely more than two notes

distant. It may be added that his suppositions about the

original form of 73 dugah suggest that the melodic movement

was of the same type in this shu4ba t oo. If

four together

77 rakbl G) A1> 3b B c d

78 mubarqa4 G At) B1? c eb et

79 salmak G) A Bt? C aft e (£

7 3 dug ah •• ) Bt c d et f

it will be seen that certain common features emerge: neutral

intervals preponderate within the central cluster; taking

the pivotal prominent note as a juncture between hypothetical

units, in no case would these be parallel; in three of the

four modes the intervals flanking the pivotal note are the

same. These features may be related. The lack of parallelism

(resulting from Zalzalian combinations or pseudo-combinations

where no whole-tone adjoins the pivotal note) may have

prompted the search for another kind of balance, and the

partial mirror symmetry furnished by the identical flanking

intervals could have provided such balance. Such considera­

tions may help to explain how the particular melodic shape

of these modes emerged, although they offer no answer to the

question of why, in rakbr and mubarqa* , B and dt respectively
became prominent.

Thus far we have taken only the first section of 79

salmak into account. Viewed as a whole, however, this mode

is of a somewhat different pattern: it is a fixed combination

of two elements at the same pitch level, Further, it may

be considered a secondary derived form, for it is based in

part on another shu4b a , 76 panjgah, while in addition the

first section relates to a mode already possessing "particular

melodic features". Neither of these aspects is mentioned by

£>afl al-Din, nor could they be inferred from his notation of

79 salmak as c d e ft g at; and were it not for Qu-fcb

al-Din’s description one would beinclined to dismiss his

identification of salmak with 57 zankula as an irritating

academic digression having no bearing whatever on practice.

If, for ease of comparison, we transpose gafi al-DIn5s

version down a fifth (F G A Bt c d ? ), it will however

be seen that it contains all but one of the notes in the

central cluster (A - e) in the first section of Qufb al-Din5s

version

79 -6. o ' V * g— •t -— -I”

The presence of the F in the earlier form suggests the

possibility of development by assimilation (F G A B1? c

d1?------- -> F G A Bt> c d1? e ------ -» G A Bl? c d? e

>57 zankula), with the ending in panjgah being added

-

later, perhaps in order to safeguard the threatened indepen­

dence of the mode. It may be added that bhe descent through
panjgah lorovides partial parallelism, the octave f - F

coinciding^ in intervallic structure with 64 G A Bl? c d

el? ft g.

In discussing maya and shahnaz Qu£b a l - D m unfortunately

seems more concerned with criticizing his predecessor’s

versions than with saying anything positive about their

melodic structure. It has already been suggested In the

comments on 2d shahnaz that the specific melodic form

represented by gafl al-Din as G- At? Bb B1? At? G gradually

came to be felt too restricted, and that there emerged in

consequence the more extended form known to Qu£b al-Din,

2db G At? Bb Bl? c d1? d. It should however be admitted

that, viewed against the structural norms found elsewhere in

the modal system, such a development Is somewhat unexpected:

parallelism occurs not at the fourth but at the neutral

third, and there is in addition no other example of a

functionally specialized form being expanded by repetition

at another pitch level. One would expect, rather, that if

an extension of the range was felt desirable this would be

effected by incorporating additional notes from the related

unit 25 /1 3l> 3$ 4-if 5/> or by a process of combination

with another unit. In fact, the former type of development

1
is attested later, while the form known to Qu^b a l - D m

appears to have been swiftly abandoned.

1 By al-Ladhiqx (BM. MS. Or. 6 6 2 % fol. 5 6 ).

 gafi al-Din*s remark about "ascending and descending"

melodic movement is the only hint we have as to the nature

of the specialization characteristic of ma y a . It is

interesting to note, however, that while gafl al-Din lists

maya as G Bl? c el? f, Qu^b al-Din Is quite adamant in

maintaining that the range is no more than a fifth. The

change may perhaps be viewed as an assimilation, the range

of the other shu* bas (or the melodically most significant

parts of them) being a fifth or less- be may remember that

Qutb al-Din stresses the difference between octave scales

and modes definable primarily in terms of functional

specialization, and there may well have been a direct

correlation between range and the degree of melodic

definition. It is thus possible that gafi al-Din has given

us a maximal version of maya (analogous to his account of

57 zankula) containing more notes than the performer

habitually employed.
 With regard to the material contained in the second

mabhath, it may be advisable to consider first the list of

"relationships on two bases" (pp. 278-281), for the resulting

juxtapositions of units are similar in type to those

occurring in the fixed octave scales; indeed, they are

exactly comparable in the five combinations the constituent

units of which are a fourth or a fifth apart. Qu^b al-Din9s

use of the term tarkib, combination, suggests in this context

something less unified or integrated than the octave scales:

it might be thought that in practice the two units would not

form an entity distinguishable from the sum of its parts,

and that by comparison with the octave scales there would be

greater emphasis on developing separately the melodic

possibilities of each unit, thus rendering intervallic

relationships between the two units of correspondingly

lesser significance. It would follow from this that one

might not expect to find in these combinations strict

conformity with the patterns of unit distribution and the

norms of consonance observable in the octave scales. In the

event, however tenuous the link between the two units may

have been in terms of melodic utilization, these combinations

criteria
fail to present us with what, according to the/previously

adopted, would be considered dissonant unit juxtapositions.

Considered purely as a sequence of Intervals, 120 busaynl//

(6) busallk is identical with a previously established octave

scale, 47 fcusayni G- At Bb c d eb f g, while 114

nawruz// igfahan similarly coincides with the section G - f,

and 81 nawruz//rahawi with, the section G - e, of the octave
1
scale 67 igfahan G Ai? Bl? c d1? eb e f g. 115
busayni//kuchek contains virtually the same units as 36
2
leuch ek- i t am am G At Bb c dt eb et?, but with parallelism

at the fifth, not the fourth, (117 maya//dugah will be

discussed latex', as will all the other combinations marked

as doubtful.) If we take into account only consonances

between the units, i.e. in 115 hasayni//kuchek

---- — ---I
-------- -------
1 —
--- ----- IT---1 »
0
tr..*&' ...* -
.1
>r y
a +P* bi • _J |
p
t 94 "

1— = ----------

G - d, At - et, Bb - f, c - f, but not G - c, it v/ill be

seen that the consonance ratios in the above four combinations

are; 120 busayni//6 busalik, 6 : 1 ; 114 nawruz//igfahan,

5 ■ 1; 81 nawruz//rahawi, 3 : 1 ; and 1 1 5 busayni//kuchek,

4:1. The alternative form of 120, busayni//14 busalik,

3
has a 7 : 1 consonance ratio. It will therefore be seen

that all four are on a level of consonance comparable to

that observable in the (mainly octave) fixed scales previously

examined.

1 These statements should not be taken to imply any

Identity of melodic potential. In this respect 81 nawruz//
rahawi will obviously be quite different from 67 igfahan.

2 The former has 18 busayni /1 21? 3b 4/T/, the latter

the related 9 nawruz 7T~~2l? 3b 4/.

3 The comparatively high ratios such as 6 : 1 and 7 : 1

result from the Inclusion of consonances (e.g. c - g as
well as G - c) not counted previously, but which cannot
here be taken for granted.
 A similar picture is presented by the other live non­

doubtful combinations !,on two bases". As the intervening

intervals in these are all smaller than in the above

combinations, the units overlap to a greater or lesser

extent- Where they do, they may perhaps be better thought

of as being superimposed, rather than as occupying discrete

positions in tonal space. In addition to degrees of

consonance we may therefore speak of degrees of identity and

in assessing the consonance/identity ratio of these combina­

tions, take into account the number of notes they have in

Thus in 116 fousaynl/A 10) *iraq

1----- -.*
... t /s
i)
/ Q bj • -¥0
r b
Vt* *
n* v .P*
i V* -- -- i— .. ..... ""
H i— — :...— " » _i

the units have three notes in common while the G, which has

no corresponding note, stands in a consonant relationship

with c in the other unit- To d, however, corresponds d1?,

a (horizontally) "dissonant" relationship which may be

considered the equivalent of the (vertically) '’dissonant11

unattached note, although two notes are in fact involved.

Thus according to these criteria the consonance/identity

ratio of this combination is A : 1. In other cases a note

which cannot be related in either of the above ways to a

note in the other unit may be considered neutral (and remain

uncounted) if it stands in a consonant relationship with a

note in the same unit, and unattached if it does not- The

consonance/identity ratios of these five combinations are

in consequence:
 frusaym//2 4iraq, 4 : 0; fcusayni//10 1iraq,4 :1

^9 4ushshaq//bus alxlc - 5//^» 5 - 0; 5//14*, 6 :0;

13//6, 6 : 0 ; 13//14-, 7 : 0

121 nawruz// 'iraq, 4 : 0

^23 2 *iraq//kuchek, 1- : 0; 10 4irag//kuc h ek , 5 : 0

124 2 4iraq//fcusayni, 5^0; 10 4iraq//husayni,6 :0

These figures are not exactly comparable to those arrived

at in evaluating the relative degrees of consonance among

the octave scales, but if we consider identity the equivalent

of parallelism the situation they reflect evidently is.

This emerges quite clearly if we superimpose the units

concerned, giving

116 G- At Bb c d or 0 At? Bb c d/dt

119 Ox A B c d e or G A B c d e ft

121 G At Bb c or G At Bb c dt

123 G At Bt c dt d

124 G At Bt c dt et ft

Features such as the sequence dt el? ft in 124 only appear

unusual because of the arbitrary notation from G (in practice

one would px^obably find Bt c d et f g a), and becaxise

no reference has been made to the position of the pi*ominent

notes. In 121 and the first alternatives of 116 and 119

these provide the only means of distinguishing the combination

from part of an existing octave scale-

Within the larger class of combinations "on one base"

 K d.'yH -)

there are 16 (82-88, 92, 94-98* 1 0 3 , 107 and 108) to be '

considered if we again disregard those marked doubtful.

Here it will be simpler to speak only ofdegrees of identity,

i.e. of the extent to which the notes of one unit are the

same as, or clash with, those of the other. The following

categories may, for the sake of oonvenience, be distinguished

1 the unit with the lesser range coincides with the

corresponding section of the other

85 /G At Bb c d/, /G At Bb c/

2a one unit omits one note occurring in the other but

otherwise coincides with it

103 /G At Bb c/, /G At Bb B c/

107 /G At Bb c d/, /G Bb c d/

2b the unit with the lesser range contains one note not

found in the corresponding section of the other but

otherwise coincides with it

82 /G At Bt c a/, /G At Bb B/
83 /G At Bb c a/, /G At Bb Bt/
87 /G At Bb c a/, /G At Bb B c/

97 /G At B(t) c c it a/, /C- At Bb B/

2c within the area common to both the units differ in one

note (there is one "dissonant11 relationship)

84- /G At Bb c d/, /G At B c/
 86 /G At Bb e d/, /G Ab Bb c/ or /G Ab Bb
c d/
88 /G At Bb c d/, /G At B e d /
98 /G At B(t)o cl d/, /G At Bt/ or /G At
Bt c/
108 /G A B c/ or /G A B c d/, /G A Bt c/
or /G A Bt c d/

3within the area common to both the units differ in one

note and one unit contains a note not matched in the other

92 /G At B c/, /G At Bb Bt/
94 /G At Bb Bt c dt d/, /G At Bb Bt c# d/

4- combinations with a lesser degree of identity

95 /G At Bb Bt c dt d/, /G At B(t) e cl d/
96 /G At B(t) c cl d/, /G Bb c a/

Here a combination in which one unit contains, say, B and

the other Bb and B1? is held to comprise one "dissonant"

relationship and one note not common to both units;

alternatively it would be possible to speak of two "dissonant"

relationships: the category sequence would not be affected.

In comparing this sequence with the degi'ees of consonance

established for the fixed scales, it would appear reasonable

to consider category 1 roughly equivalent to scales

containing no unattached notes; 2 to scales containing one;

3 to scales containing two; and 4- to scales containing more

 (296)

1
than two unattached notes. It therefore follows that the

type of unit combination characteristic ox the consonant

octave scales also seems to prevail in all but two (those of

group 4) of these combinations, so that they appear to

provide yet anothex1 illustration of the preference for

parallel structures (identity being parallelism at the

unison). This evidence is of particular value, for it is

above all in the combinations "on one base” that one might

have expected to encounter a number of more unusual unit

juxtapositions: owing to the fact that it is not possible

to reduplicate the same unit, as is done in parallel

tetrachord scales, complete identity is in any case excluded

and it would further be natural to find this necessary

differentiation of the two units reinforced through contrast

On occasion, no doubt, the desire for contrast was felt,

leading to the creation of somewhat "dissonant" examples:

Qu£b al-Din suggests as much when he speaks of leaving other

2
combinations to the experts. But his wording suggests

equally that the "consonant" combinations he lists were the

3
ones most commonly employed. Furthermore, we may note that

1 There is no equivalent to the grading achieved by the

consonant interval count for fixed scales since the
corresponding feature here, the number of common notes, or
unisons, will also vary with the range of the units
concerned.

2 p. 268-

9 Of. also the final remark (p.. 280): "the well-known

modes ... are confined to the pardas , shu*bas and tarkxbs
which we have set forth".
 \<^y ( /

the combinatory pattern most frequently found among them is

one in which the units coincide in all but one note within

the area common to both, so that the extent of the

1
differentiation is reduced to a minimum.

With this in mind we may turn to the combinations marked

doubtful. The reason for so marking them is that, as a

result of Qufb al-Din*s failing to distinguish between the

terms "tonic" and "prominent note" in the introduction to the

2
second mabfcath, two or three versions of each are in theory

possible. In the units occurring in the combinations already

examined the tonic (or lowest note) is always a prominent

note and sometimes the only one, so that it provides the

obvious base; but if we take, say, 73 dugah, it will be

seen that there are three notes that could furnish the "base":

G, the tonic of 43 rast, the mode from which dugah is derived;

c, the note on which it properly begins; and d, its prominent

note. However, since the previous discussion has established

the validity of known patterns of unit distribution for this

section of the modal system, we may accordingly expect that

combinations containing dugah should also conform to them,

and this can only occur if the third possibility, the

prominent note d, is taken as the "base". Two examples may

1 The combination of G At Bb c (d) and G At Bb B (c

occurs three times, while the note series /1 2t 3b 3 4-
(5) or (4if 5)/ arises from superimposing the units, in no
less than seven combinations.

2 p . 268.
 (29s;

suffice in demonstration- With G as "base" 91 igf ahan/dugah

would be /G At Bl? B c / ? /(G A) Bt c d et f .../,

which with two dissonant relationships and one note not

common to both units belongs to category 4. With c as "base"

it would be /c dt eb e f/ ? /...Bt c d et f .../,

which also belongs to category 4. With d as "base", however,

we have

/a et f ft g/, /... c a eU f ((ft) g ...)/

which belongs to category 1.

Similarly 93 hid azr/dugah is with G as nbasen /G At B

g/ , /(G A) Bt c d et f - -./, category 3; with c as

"base" /c dt e f/, /... Bt c d et f .../, also

category 3; and with d as "base”

/d et f$ g/, /... Bt c d eb f ((f$) g • **) A

category 2.

With d as "base11 the distribution among the four

categories of the combinations with 73 dugah (100 ralibi/

dugah excepted) is:

90 husayni/dugah /d et f g a/, /. .Bt c d et f ((fS)g a)/

91 i gf ah an/dug all /d et f sA /--Bt c d et f ((fS) g a)/
99 rahawi/dugah /d et f fit/, /..Bt c d et f C(f8) g..)/
110 n awru z/dugah /d et f g/, /. .Bt c d et f ((f #) g ..)/
'st

.hid azi/ dugah g/, /. . Bt c d et f ((f 8) g ..)/

CD

93 /d
 109 kuc hek/dugah /d et f f :i
:/ o /..Bt c d et i ((fif) g ..)/

3 117 maya//dugah /G Bb c d/, /(G A) Bt c d et f ../

4- no example I

If we accept that d is the "base" for the combinations

with dugah, it follows that et will be the nbasen for those

with s eg ah, and f for those with chahargah ■ We will therefore

have, arranged according to category:

1 104- kuchek/ s eg ah / et f gt g A /**C a et f (gt) g../

^ 0 5 2 *iraq/ segah /et f g/, /..c d et f (fS/gt) B-•/
1
112 igfafranak/segah /'Bt c a et f gt g/,
/(Bt) c a et £ (gt) g../

2 1 0 5 10 *iraq/segah /et f g at/, /..c a et f (fif/gt)g (a)/

^ ^ *ushshaq/ chahargah /f g a bb (c)/,
/.. c d et f (g a bt c)/

3 "106 bijazi/segah. /et f gt at/, /..c d et f (ffl/gt) g (a)/

112 igfahan/segah /et f gt gt at/,/..c d et f (gt) g (a)/
^2 rahawi//chahax^gah /et f gt gt/,/..c a et f (ga)/

4- no example

Both possibilities have been included for 105 and 112.

Box1 112, igf ahan ak/ s eg ah (category 1) seems the far more

likely alternative. If igfahan/segah did in fact occur it

is justconceivable that the note sequence et f fit g at?

(/1 2t 3bt 3t V ) was used as an approximation to

 (300 )

^ igfahan /1 2t 3b 3 A/, in which case this combination

would belong to categoi*y 2. Again, if gt was admissible as
a variant of the optional ft in chahargah
iVifu im.iU'j as well as in rsegal
•

122 rahawi//chahargah would also belong to category 2 rather

than 3*

Before passing on to the remaining combinations marked,

doubtful we may consider here the four scales previously
listed by Qu^b al-Din and given compound names in which
segah and chahargah figure. Qutb al-Din notates them as:

30 (segah wa fcijazi) G At B c d et
51 (segah wa frusayni) G At Bb c d et
38 (chahargah wa jiijazi) G At B c d et f
39 (chahargah wa igfahan) G At Bb B c d et f

The names suggest the possibility that these forms might be

compressed notations of combinations "on two bases".
According to the method used above these would apuear as:
 (3 0 1 )

But if they were in fact combinations of this nature it is

not at all clear why they were listed together with the fixed
scales. Fi*equencyof occurrence might be put forward as a
reason,but in that caseone would have expected to find in
the same list a number of combinations with dugah, which
appears to have been far more popular than either segah or
chahargah. Further, as combinations "on two bases" the above
are unusual in the size of the intervals (neutral sixth and
minor seventh) between the "bases", and, consequently, in
the negligible degree of overlap between the two units. It
would thex'efore appear likely that they do properly belong
to the corpus of fixed scales: all have partial or Incipient
parallelism at the fifth, and may perhaps be considered
restricted forms related to already existing octave scales
(60 mufcayyir frusayni G A1? Bb c d e1? f g, 61 mufrayyir
zirkesh G At Bb B cd et f g, and 62 nihuft G At B
c d et f g). We may remember also that the form of the
heading ("has no name ... it may be called ...") is the same
in all four cases, and - apart from making it virtually
certain that these modes were not of frequent occurrence -
suggests that the titles may be no more than convenient
cax^tions giving some indication of the intervallic structure
and at the same time of the positionof the prominent notes.

Of the other combinations to be considered four include

77 rakbi. In 73 dugah it was possible to establish that it
is d, the prominent (and final) note, that furnishes the
"base". By the same criteria we may establish that in rakbl
the note in question is G, the final note and tonic of the
species from which rakbl is derived, rather than B, the
prominent note. (Taking B as "base" and considering for
present purposes the notes occurring in rakbl, G At Bb B
c d, as a simple sequence without particular melodic
properties, it will be seen that rakbl/dugah (with rakbl
transposed up a minor third to make the two "bases" coincide)
becomes /Bb ct db d eb f/, /... Bt c d et f .../,
belonging to category 4-. With G as "base", however, this
combination (with rakbl transposed up a fifth) becomes
/d et f f# g a/, /... Bt c d et f ((fit) g a)/,
belonging to category 1.)

We have therefore according to category

^ ^90 rakbl/dugah /d et f ft g a/,/..Bt c d et f ((ft)g a)/

101 rakbi/ igfah an /G At Bb B c d/, /G At Bb B c/

8 89 bnsaynl/rakbi /G At Bb c d/, /GAt Bb B c d/

102 rakbl/nawruz /G At Bb B c d/, /G At Bb c/

3 and 4 no examples

With regal'd to 113 .zanku Ia/ rahawl it should be remarked

that the lack of further combinations containing zankula
renders the choice more problematical. Nevertheless, in this
case it is evidently preferable to take the pivotal prominent

note c as the "base" for zankula, rather than G. The former

results in
 113 aanku 1a/ rah awl /(G) A Bb c db e (f)/,
/c db eb e/

belonging to category 2, and the latter in /(G) A Bb c db

e (f)/, /G At Bb B/, category 4* Eurther evidence in
support of this interpretation is provided by the fact that
there is no example elsewhere of the juxtaposition of /1 2
3b 4/ and /1 2 p 3b 3A while /1 2b 3 z!-/ combines with
/I 2b 3b 3/ in 33 rahawi-yi tarnam G At B c dt eb e.

The one remaining doubtful combination presents a -rather

different problem. In 118 maya//4iraq the interval between
the twounits is stated by Qutb al-Din to be 6 : 5? and one’s
first reaction is to accept this figure at face value.
However, with the units a minor third apart the combination
becomes /G- Bb c d/, /Bb cl? db (el?)/, which belongs to

category 3 butnevertheless appears rather odd: cb occurs

nowhere else, and in addition one would expect 4iraq to
1
begin on Bt? rather than Bb * Now, we have seen that the
equal parts into which the neutral third divides the fifth
were interpreted as 5 : 4- &nd 6 : 5? and that while the
interval of a neutral third was usually given the former
2
ratio, the latter also occurs on occasion. If therefore

1 If 4iraq is taken from Bb. with the units kept a minor

third apart, we will have /GI Bb cl dt/^ /Bb c d (eb)/,
wrhich looks even odder, since Gb, cb and dl occur nowhere
else.
2 E.g. in the ratio definition of 1 zlrafkand-i kuchek.
 (3 0 4 )

we interpret the 6 : 5 ratio here as designating a neutral

and not a minor third we will have the far more satisfactory
/G Bb c d/, /Bt c d (et)/ (category 2).

In the concluding passage of the second mabfrath Qutb

al-Din mentions another three combinations which are stated
to be well-known. Since it is not made clear whether they
are on one "base" or two (or in one case three), these should
also be considered doubtful. It would nevertheless seem
reasonable to suppose them, to be on one V'base", for the
resulting unit juxtapositions are similar to those examined
above. One indeed appears to be identical: humayun is said
to “begin in zankula and end in rahawl11, a description that
could equally well be applied to 113 zankula/rahawl. Whether
or not the two did in fact coincide in practice cannot
unfortunately be determined from this one brief remark.
(Later accounts of humayun also relate it to zankula and
rahawl, but it is difficult to draw any positive conclusions
1
from them.) It may be noted that for Qutb al-Dxn the
structure of humayun would appear to have been of the same
type as that of 79 salmak, both consisting of fixed
combinations (presumably on one ubaseM), and both having in
addition 57 zankula as their first element.

1 1Abd al-Qadir gives the notes G A Bt c dt eb f,

which coincide with zankula from G to dt, but can only be
associated with 32 rlSiawi^ which Qutb al-Din does not recog-
nize. The form listed by al-Ladhiql is c dt c dt f e
dt c d db (? or cif) c B c, which relates to the upper
tetrachord of zankula and a later development of rahawl.
 Of the others 125 dugah wa frijazi, being in effect
dugaJh/fril azi/dugah, seems to be no more than an extension
of 95 b-it
jaai/dugah. 126 dugah/buzurg/rahawl on the other
hand is of considerable interest, for it demonstrates that
more than two units could be combined in a standard sequence.
Although it is impossible to generalize from an isolated
example, it may be remarked that this particular sequence
(given that all three units are on the one "base") does not
show a progressive departure from the first unit. In fact,
rahawl is closer to dugah in terms of identity than is
buzurg, and thus may have given the impression of a partial
return.

If we accept all the above versions of the "doubtful11

combinations it follows that 80-126 constitute a group in
which, mutatis mutandis, the same standards of consonance
(or consonance/identity) seem to obtain as in the fixed
scales 50-72. Hence one might expect to find among them
similar patterns of unit distribution also.

For the sake of comparison, such patterns ma.y be shown

graphically, as before. The following diagram represents
all the combinations at the fourth, fifth, and unison. In
 K 3 V O J

addition to units shown in previous diagrams it therefore

includes the shu4has 73 dugah, 74 segah, 75 chahargah,
77 rakbl, 25b may a and 24b shahnaz. As Qu£b al-Dln points
out, the first three of these tend to coincide with 9 nawruz
/1 2t 3b 4/, 2 4iraq /1 2t 3t/, and 8 rast /1 2 3t 4/
1 '

respectively: thus they may be entered at the samepoints

as these units while being distinguished by name. 77rakbl,
which similarly aligns itself with 19 zlrkesh jiusayni /1 2t
3b 3 Vf/, is in consequence entered at the same point as
^ ifffa-fran /1 2t 3b 3 4/, and is also distinguished by
name. 24b shahnaz is given as a reduplication of /1 2t 3b 3 V •

Combinations on one "base” are separated by a single

2
stroke, and where confusion could result by a comma;Those
on two "bases" are separated by a double stroke.

1 In dugah we have on either side of the melodic fulcrum

/1 2t 3b 4/: A Bt c d and d et f g. /1 2t 3t/
occurs above the corresponding note in segah (et f .g),.and
the related /1 2t ^t 4/ below (Bt c d et). There is
no evidence that the area above the. prominent.note in
chahargah was of comparable importance to that below
(c T et^ f, /1 2 3t 4/).
2 In 112 igfahanak/segah and 113 it is the upper unit of
the first element, not tlie lower, that has been taken to
be on the same "base" as the second element.
 1/1 2b 5b 4 / /1 21? 5b 4 / T / / 1 21? 3 b 3 / /1 2 5 V
/1 2b 5b 4 / / 1 2b 3b 3 /

/1 21? 5b 4 / 0 / , / 1 2b 3 b 3 /

/1 2b 3b V T / , / 1 2b 5b 4/(ny) /1 2b 5b 4 /T /,/1 2b 5b 5 4 / /1 2 '5 4 /(0 ? /),/ I 2 5 f 4 /(T /)

/1 2b 5b 4 / T / / 1 2b. 5b 4 / ( T / ) /1 2b 5b 4 / T / , / r a k b T / /1 2 5 4 / ( T / ) , /c h a h a rg a h /

/1 21? 5b 3 4 /d u g a h /

/1 21? 3b 3 /d u g a h /
ro

/1 5b 4 / T / , / 1 21? 5b. 4 / / r a k b i/d u g a h / /1 2 51? 4 /

/1 21? 5b 4 / T / / d u g a h / / r a k b i/ 1 21? 3b 4 / /chahargah/]
/ T 2b 5b 4 / / d u g a h / /1 2b 5b 4 /1 21? 3b 3 4 /

/1 21? 3b 4 / / 1 21? 3b 3 4 /

/1 2b 3b 4 /0 / , / 1 3 b 4 /T /
/dngah/1 21? 5 4 4# 5/1
/1 3b 4 / T / / d u g a h /

/1 2b 5b 4 /T //1 2b 5b 3b/
/1 2b 5b 5 "b /d u g a h /

/1 2b 5b 4 /T //1 2b 5b 5b/ /1 5 b 4/0?/

/r a k b i//l 21? 5 b 5 4 /
/1 2b 5b 5V /1 2b 5b 5b/ , / 1 2b 5b 5b 48 5/
/1 2t? 5b 5 /

/1 21? 5 4 4 If 5 /1 5b 4 / T /

/1 2t? 5b 5b//1 2b 5b 5b/,/1 2b 3 4 4if 5 / /1 2b 5 4 4 ft 5/1. 21? 5b 5 /

/1 2b 3 4 /1 2b 3 b 3 V / ^ 2b 5 4 4 if 5 /
/1 2b 3b 3b / s e g a h / /1 21? 5b 5 4 / s e g a h /
V
/1 21? 3b 4 / / 1 21? 3b 3 b / , ] /1 2b 3 4 4# 5/1 2b 5b (4 Y
t*
, /s e g a h /J

|/1 2b 3b ( 4 ) / s e g a h /

/1 21? 5b 4 / T / , /1 2b 5 4 /
... . ,
/1 2b 5b 4 / T / , / 1 21? 5 4 /T / \ /1 2b 3 4 - /s e g a h / /1 2 51? 4 / / 1 21? 5 4 / /
\ />
/1 21? 5 4 / d u g a h / /1 2b 5b 5 /

/ d u g a h /1 2 t 5 4 / d u g a h / /1 21? 5 4 / humayun
 Apart from the omission of /1 2 3b 4/ and the addition
of /1 31? 4/T/, the one obvious difference between the
patterns of distribution in this and the preceding diagram
results from the alteration in the position of 20 buzurg
/I 21? 3 4 4 If 5/* 4s the combinations of this unit with

73 (® /1 2t> 3b 3 4 5/), 23b maya (/1 3b 4/T/)

an(^ 4 (/1 21? 3 I? (4)/) are evidently more satisfactory

in a system where like units attract than that with 8 rast

(/1 2 3^ 4/), which appears in 84 buzurg, this change can
1
hardly be considered unexpected.

Of perhaps greater importance, however, are the

differences in density. In the diagram for the fixed scales
the units /1 3b 4/, /1 3t> 4/ (with /1 2t 3 4/),
and /1 2 3"b 4/ form a triangle within which, and along
the sides of which, the great majority of combinations are to
be found; in addition, they are fairly evenly spread over
this area. The present diagram is by comparison unbalanced:
the area of greatest density is concentrated within the
inner triangle formed by the nodes in which /1 2t? 3b 4/,
/-x 2$ 3b 5 V, and /1 2'b 31? (4-)/ (with /1 2t 3 V)

appear, end even here there is no even spread, a relatively

large number of combinations being situated on the line
joining the first two of these nodes.

1 In the fixed scales, the combinations of 20 buzurg with

11 i$fahan /1 2t 3b 3 4/ and 12 frijazx /1 2?””3 4/
(in 55 and 56 respectively) are comparable to the
combination 97 buzurs/rahawi (4- rahawl /1 2"b 3b 3/).
 Such changes indicate a significant variation in the
frequency of occurrence of certain units. The extent of
the variation may be seen more clearly from the following
table.

The columns denote

1
1 The number of occurrences in the fixed scales
2 The percentage for each entry in 1
3 The number of occurrences in the combinationslisted in
2
the present chapter
4- Percentages derived from 3 hy omittingall units not
occurring in 1

For present purposes a number of related tetrachords and

pentachords (e.g. /1 21? 3b 4-/ and /1 2t 3b 4-/T/)
are placed together to form a single entry.

1 Excluding 32 rahawl. In 69 and JO, which are uncertain,

only hijazi (one occurrence) and nawruz (one occurrence)
have been counted.
2 segah is only counted once in 112 igfahan/segah or
igfahanalc/segah. humayun is ignored.”
The figures in 3 are also percentages, the sum of the
column being, fortuitously, 101.
 1 2 3 h

nawruz /1 2$ 3 b 4/, Unsay 111 / 1 2i? 3 b 4/T/ 17 20 22 29

rast /1 2“ 3^ 4/, /1 2 3^ 4/T/ 14 17 2 3
hijazi /1 2‘l? 3 4/, *uzzal /1 21? 3 4/T/ 13 16 7 9
busallk /1 2b 3 b 4/, / 1 2b 3b 4/T/ 8 10 3 4
*iraq /1 21> 3^ (4)/, /1 21? 3^ 4-/1/ 7 8 8 11
4 ushshaq /1 2 3 4/, /1 2 3 4/T/ 6 7 3 4
igfahan / 1 2 'b yb 3 4-/ 6 7 6 8
buzurg / 1 2*b 3 4 4# 5 / 4 5 5
rj
1
kuchek / 1 21? 3b 3V 3 4 11 14
hi gar / 1 2'b 3 b 3’b 4$ 5/ 2 2 1 1
nawa /1 2 3b 4/, /1 2 3 b 4/T/ 2 2 0 0
rahawl / 1 2? 3b 3/ 1 1 8 11
dugah q'l

rakbl 4

segah 4

chahargah 2

33 99 101 101

1 If shahnaz is listed separately from kuchek, the figures

for both will be
kuchek 3 4 7 9
shahnaz 0 0 2 3
If hi gar is then listed together with kuchek, we obtain
kuchek, hi gar 5 6 8 11
shahnaz 0 0 2 3
 These figures should be approached with some caution.

The catalogue of fixed scales may be exhaustive, but the

lists of combinations, on Qujb al-DIn5s own admission, are

incomplete, furthermore, too great a reliance should not

be placed on the figures in column 4, for the frequencies

they indicate are to some extent dependent upon the incidence

of units omitted from' it. But however wary we must be of

generalizing from this table, there are nevertheless a

number of features too striking to be dismissed as merely

coincidental. Comparing columns 2 and 4, it will be seen

that there are considerable differences in the figures for

nawiuiz/tLUsayni (20, 29); rast (17? 3); hit

i azi/‘uzzal (16,
1
9)> busallk (1 0 , 4); kuchek (4, 14); and rahawi (1 , 1 1 ).

The discrepancy in the figures for nawruz/husayni can

be explained to some extent by the inherent defect in column

4 mentioned above. In this particular case, the relatively

high frequency of occurrence in the combinations is partly

attributable to the presence of the structurally similar

75 dugah. An additional factor is the increase In the number

of occurrencies of 1 kuchek / 1 23 3b 5^/ and 4 rahawi

/1 2p 3b 3 /? with which nawruz and fousayni form consonant

combinations.

The relative infrequency of busallk in the combinations

1 Even if kuchek is taken with Igigar instead of shahnaz

the difference remains considerable C6 , 1 1 ).
 (312)

may be considered symptomatic of the position of the diatonic

species as a whole. Taken together they form 19 per cent

of the total for the fixed scales, but only 6 per cent of

the combinations. Admittedly if, as seems likely, Zalzalian

and diatonic combinations were generally avoided, the fact

that there only occur three diatonic units clearly reduces

the possible numberj nevertheless, the possibilities were

by no means exhausted, for nawa /1 2 3 1? 4-/ fails to

appear at all. In effect, the evidence of the lists of

combinations only serves to underline the lack of importance

of the diatonic units vis-d-vis the Salzalian, and suggests

that at this period they may have been losing further ground
to them.

That rast was also losing ground is on the other hand

unlikely. There are as it happens relatively few units with

which it can combine to produce acceptable results in terms

1
of consonance/identity, and the importance of the rast

scale (4-3 G A B1? c d et f g) is amply demonstrated

by the vitality of the shu* bas generated from it, especially

The variations in the percentages for 1 kuchek / 1 2t

1 Principally, in addition to those found, /1 2V 3b 4-/

and / 1 2? pv 3 4-/ at the fifth, / 1 2t 3"b 4*/ an&
/I 2 pb. 4*/. at the fifth, fourth, or unison. iViost of
these reproduce already existing structures, while combina­
tions with the diatonic /I 2 3 b 4-/ would be unusual.
These units could be described as having a high density

relative to the majority of tetrachord species, i.e. they

contain as many notes (and are presumably as rich in melodic

possibilities) but compressed within the confines of a

smaller interval. The evidence of the lists of combinations

would appear to suggest that their importance within the

modal system as a whole was comparable to that of the

Salzalian tetrachord species, but that for structural

reasons they were generally excluded from the fixed scales:

because of their higher density they would tend to unbalance

a (vertically) extended modal structure, in particular by

reducing the significance of parallelism at the fourth or

1
fifth through the prominence of the third. In all probability

it is this factor which best explains the anomaly in gafl

al-DIn5s treatises whereby 1 kuchek and 4 rahawl are

laboriously defined, but with the exception of one melodically

specialized form relating to kuchek fail to appear in the

corpus of modes discussed by him. In fact, both are by

nature excluded from the type of formal organization

exemplified by the octave scales (with which gafi al-Din

and the majority of his followers were primarily concerned).

1 One may compare in this respect the coexistence of the

"^wo igbahan octave scales, outwardly so dissimilar (6 6 G A
Bt c d_ et f f# g, 67 G At Bb c dt eb e f g).
^ igi'&hsm /1 2 t 3 4-/ is. also a "high, density" unit,
and because of this may have provided such a strong centre
of gravity that the position of the nominal tonic was
relatively unimportant.
 \y i ~r j

Hence their true importance can only he gauged, however

approximately, from their frequency and context of occurrence

in the lists of combinations supplied by Qufb al-Din.

The difference in the figures for hit

jazi may, as with

rast, be explained in part by the fact that there are

relatively few units with which it might form consonant

combinations not already found in the fixed scales. It does

however seem rather curious that while bit

jazi occurs less

frequently in the combinations the opposite appears to be

true of the related 6 iraq, so that taken together the two

sets of figures suggest that the development 10 *iraq / 1 21?

3 I5 4 /------ » 12 fri.iazl /1 Zt 3 4/ is paradoxically more

clearly revealed in the fixed (and hence presumably more

stable) scales than in the combinations. The paradox may

be resolved if we assume that the *iraq referred to in the

combinations is in most cases not 10 /I 21? 3*1? 4/ but

2/1 2t 3*fr/« It appears from Qu£b al-Din’s indications

that in / 1 2t 3$/ the third was prominent, a feature which

would tend, in an octave scale, to undermine, and be under­

mined by, the tonic - fourth and fourth - minor seventh

relationships. Thus it is possible that /1 2? jb/ may

have been employed quite often (perhaps more often than

/1 21? 3 I? 4-/), but that because it differed in structure

from the units which combine to form octave scales, it

appeared almost exclusively in the contexts defined in the

above lists.
 (515)

In conclusion, it may be suggested that the evidence

provided by Qu^b al-Din demonstrates quite conclusively that

his predecessor’s account of the modal system is incomplete

and therefore to a certain extent misleading. The material

contained in the two mabbaths translated above indicates

that the melodic utilization of the units basic to the

system was not confined within the one type of compound

1
organization discussed by §afi a l -Dm. It further reveals

the context of occurrence of the non-tetrachord units and

shu* bas without an examination of which a balanced view of

the modal system as a whole is impossible. It is safe to

assume that the majority of the shu*bas were equally

prominent in Safi al-DIn*s day; his failure to discuss them

may be attributed in part, perhaps, to conservatism, but

above all to his overriding preoccupation with a theoretical

(and inflexible) analysis of the octave and Its parts. The

resulting scheme, together with its attendant disadvantages,

is taken over virtually intact by Qu^b al-Din: but by

supplying such a wealth of supplementary information he

enables us to realize the extent to which elements important

in practice might be disguised or even disregarded for the

sake of the theory.

1 The kitab al-adwar contains just one isolated remark

("every octave scale has a fundamental part (agl) on which
it is based" - BM. MS. Or. 156, fol. 19v) that can be
construed as Indicating the melodic primacy of the
individual unit.
 U Io;

chapter 8

The last chapter in the kitah al-a&war (entitled f1

mubasharat al-*amal) is the only one to be devoted exclusively

to practice. It presents, with brief comments, four examples

of notation, two of which are reproduced in the risala al-

sharafiyya. $afi al-Dxn uses the alphabetic notation

previously employed in the chapters defining the octave

gamut, and hence also in listing the 8d octave scales (adwar)

and the transpositions (fabaqat) of the twelve shudud. To

it are added numbers indicating the relative duration of

each note, and a statement of the type of composition being

notated.

Since rhythm and form have not been discussed, the

following brief remarks may be px^efaced to the transcriptions.

rhythm

gall al-DIn’s approach to rhythmic analysis is traditional:

much of the terminology used to desci*ibe the various rhythmic

cycles is derived from pi’osody. Each cycle is divided into

(watad, sabab, fagila), these being defined in terms

of long and short syllables. A short syllable is equivalent

 (517)

to one time unit, along syllable to two. For example, the

rhythmic cycle al-thaqil al-thani is represented as

tanan tanan tan tanan tanan tan

i.e./ — / v —■ / —* / v *— / v ’
— /■"/

The time units are therefore to be divided:

/1 2 3/4 5 6/7 8/9 10 11/12 13 14/15 16/

In each foot the first time unit is marked by a percussion,

the last in general not, while any othei's may or may not be

so marked, the performer being in some cases bounded by

convention and in others able to choose. Thus if we

symbolize the first time unit of a foot as X (an obligatory

percussion), the final one as . (generally omitted), and

others as o (generally an optional percussion), this cycle

becomes:

/ X o . X o . X . X o . X o . X . /

It will be seen from this that the rhythmic cycles are

distinguished not only by the number of time units but,

equally important, by an accentual pattern also which the

performer may embroider within certain limits, but in theory

not alter. A further means of identification according to

$afi al-Din is the flarb al-a$l or fundamental pattern in

which just two of the possible percussion are used. If we

underline these, and in addition symbolize as x (instead of

o or *) those time units conventionally marked by a

percussion which are not initial in a foot, the rhythmic

cycles described by Safi al-Din may be represented as

follows:

al-thaqil al-awwal / X o . X o . X o o . X . X o o . /
(or) - ~

al-thaqil al-thanl /X o . X o . X . X o . X £ .X ./
khafif al-thaqil /X. X x X . XxX . XxX. Xx/
thaqil al-ramal

/ X o o . X o o . X . X . X . X . X . X . X o o . /

al-ramal /X . X . X . X . X o o . /

khafif al-ramal / X . X o . X o X o . /

al-hazaj /X o o . X o . X £ . X . /
1
al-fakhiti /X . . . X . X ... X • . . X . X . . . /

It is unlikely that the flarb al-agl was ever used on its own

to define a cycle in practice, for unless there was a strictly

maintained gradation of tempo between cycles the patterns

provided by the flarb al-agl for al-thaqil al-thani (12+4),

khafif al-thaqil (12+4), thaqil al-ramal (18 + 6), and, in

the second version, al-thaqil al-awwal (12 + 4), would all

coincide.

1 No £Larb al-agl is given for this cycle. Safi al-Din

statesTEai only the initial time unit in each foot is
marked by a percussion.
The principal variant forms are, according to Safi al-Din
al-thaqil al-thani /X o .X o .X . /
al-hazat j /X o o .X ./
khafif al-thaqil /X .X x /
al-ramal / X . X . X o o . X o o . /
'chalTf’"'al-ramal /X .X o o ./
 (319)

form

^either gafi al-Din nor Qujb al-Din discuss formal

procedures. The sharp contains a few brief comments on the

1
nawba and its movements, but it is not until the '15th century

that we encounter, in the works of *Abd al-Qadir, a presumably

comprehensive account of the forms of composition then in

2
use. In general, distinctions between forms ere expressed

in terms of the language and structure of the verse sung, so

that an overtly nmsical analysis of the various song-types

is not given. In certain cases, however, the composition is

defined technically; in the kull al-purub, for example, one

progresses through a number of rhythmic cycles, while a

performance of the kull al-nagham either passes through all

the pardas, awazes, and shu*bas , or combines various modes

so as to include all the notes of the gamut.

The examples of notation given by gafi al-Din are stated

to be in two forms, tariqa and gawt. Of these, unfortunately,

little is known: the former is an instrumental piece,

employed chiefly as a prelude to a vocal composition; the

3
latter appears to be one of the less complex vocal forms.

As the first example demonstrates, a tariqa might be

transformed into a pawt by being sung to words.

1 BM. MS. Or. 2361, fol. 151v.

2 E.g. maqagid al~alpsn, 103-106.
3 The sharp, states (loc. cit.) that it is qalll al-at
j zav.
 In addition to the various MSS. of the kitab al-adwar,

some of these examples of notation may be found in the

risala al-sharafiyya, the durrat al-taj, and various

commentaries on the kit ab al-adwar. In establishing the

transcriptions the following sources have been used:

A Bodleian MS. Marsh 521, ff.32-33v (kitab al-adwar)

B Bodleian MS. Marsh 521, ff. 156-157 (kitab al-adwar)

C Bodleian MS. Marsh 161, ff. 41v 42v (kitab

- al-adwar)

D Bodleian MS* Marsh 161, ff. 80v~81v (kitab al-adwar)

E Bodleian MS. Marsh 115 (risala al-sharafiyya)

E Bodleian MS. Marsh 521 (risala al-sharafiyya)

G BM. MS.Add. 7694 (durrat al-taj)

H BM. MS. Or. 2361, fol. 67v (a commentary on the kitab

al-adwar)

I BM. MS. Add. 7471 (a commentary on the kitab al-adwar)

J BM. MS. Or. 136 (kitab al-adwar)

K BM. MS. Or. 2361 (kitab al-adwar)

L BM. MS. Or. 2361 (the text of thekitab al-adwar

reproduced in the sharb)

M A photographic reproduction of a MS. of the kitab al-adwar

introduced by busayn 1All Mabfu#, Baghdad, 1961.

 (321)

Example 1a

Sources: A B C D J K L M

Form: $ arI q a.

Mode; nawruz

Rhythm : ramal

The notation is given thus:

f eb d-c eb d_c c
6 6 12 6 6 12

Taking the accentual pattern of the rhythmiccycle into

account, this may he transcribed in the following (albeit

more cumbersome) form:

If however we take the flarb al-agl to represent the main

accentual division within the cycle, this may be simplified

to:

Example 1b

The above is immediately repeated and enlarged in the

form of a gawt, each shafrr of the verse repeating the melodic

pattern.

Sources: A B C D H I J K L M
 K D d .< ± )

A.number of MS. variants occur in 1b. In transcribing

these the verse has been omitted, since the alignment of

syllables and notes is in most cases imprecise. Eurther,

because of the alterations in the durational values no

attempt has been made to indicate the structure of the

rhythmic cycle. The variants are:

B O D
\ ^ fr .ii. \ r ¥- ft*.
J K M iv

Vrr
3"* 5 £-»- V -M f ft-f- hr V
A et
m

— b.-r-=p— t.v
O* •
h &

V- *.• ^ .y.-—
-(>■
H I

I then continues

Except for Ii and I (and the last shafr of A, where there

is a possibility of scribal omission) all MSS. agree on a

melodic outline which coincides with that of 1a and may be

unhesitatingly accepted as correct:

i-i-- ^ s . -- h
i— _ —:— ii— — .» «
 (5 2 3 )

'The rhythmic values, on the other hand, present some

difficulty. Even if we again discard H and I, it will be

found that the number of time units in each shafrr is in B 0

D J K and M, 50, 50, 50 and 50; in A, 50, 50, 46 and 58;

and in L, 5^? 5^? 5^ sn-cL 5^> B0 that in no case do we

encounter a compound of the 12 time units of the ramal cycle.

The shark, commenting on text L, states that f and el? extend

over two rhythmic cycles (24 time units), while the remaining

notes of the shafcr, dt eb dt c, take up two and a half

. 1 ■
cycles (5 0 time units). However, the shark concedes, in
2
remarks on the later examples, that text l» is unreliable,

and it is in fact the only one to which this interpretation

could be applied. The 50 time units upon which the great

majority of MSS. agree cannot be divided in this way, and

the repetition of an entity consisting of four complete

cycles plus two time units seems unlikely in the extreme.

The most reasonable solution to this problem may perhaps be

found in the fact that gafi al-Din nowhere employs a rest

sign, for a pause at the end of each sha^r would hardly be

unexpected, especially as a long high note follows. If, in

conformity with the shark (and, it may be added, gafl al-Din

himself, who gives a variant of ramal of six time units in

risala al-sharafiyya) we allow the division of the cycle into

halves, we may assign the value of four time units to the

1 BM. MS. Or. 2561, fol. 152.

2 Ibid., fol. 152v.

 (324)

hypothetical pause and transcribe thus:

I-.---- #— **--- 1—
v 1-- H — g t \ :V T> V f r ' k ;V -h-t-..r:—
' 1 Y. 1 r t'.t ;.. 1"^ 0 - 4 4 ^ -ft-4--- —
-------5---
4 ala £abbikum ya Jia--kimi— na tara f f aqu

—
*i—*
8+4— S— f-- !-+— ifd 1--p*r—;•— b#
,—
— f > •>,»• V. R V
--- p_:— i
4 >— ft--- '
-- :--- - t M ---b ■ 1 ----i—

wa mln wa glikum yawirtan 4alayhi t a ^ iddaqu

-... | | t » b
9T~r
f I *L
— b—|
v ' r. | - - ;—1
1-- “ ft . p#
t \4>4r ■ S. R 7 . (f f —
J >— It------ 1--- -K-J— H — 1— tn— . v f ■ 1 V\ :i ---- *—

wa la tut lifu— hu bi’1 -gududi f a ’in — nahu

1 I1 t *
* W ■ b*
rfr- H-t l : b ir- t-4*— h*; V I P ? ----- r—
- W ---- H- -' -tM ---- f.h 1
4 y-&— -— - P '-1 -fr-p— L
yu^ia— dhiru 5an yashku dlaykum fatashfaqu

In the text the note length of the individual syllable

is only given at the beginning of the fourth cycle of each

shafr, where the dt corresponds to a short syllable. This

arises simply from the fact that the duration of the dt is

two time units. Elsewhere one finds e.g. the value 18 (time

units) entered under 4ala ^abbikum - only the total duration

of the note.is given irrespective of the syllables sung to it.

Example 2a

Sources: A B C D R L M

Form: tariqa

Mod e : kawasht

Rhythm: ramal
 The following variants occur ( — -— indicating a note

the durational value of which is illegible or omitted):

K i r - n i q==±

~4>— yr •I y-jy
L M
£

B D ~n— ^ — k— £ ------- =— -u-— =— ->*r>—

> I _ .— Ij=-I— p -
1~- V —

A tr !»■ ■-y — ; H q
$— V B d Zl-L-J!

In B D L and M the total number of time units is 48,

four cycles of the ramal rhythm. As there seem to be no

grounds for preferring L M to B D or vice versa, we may

transcribe thus:

-T— &-H— V fr» V ^ V :.

— ■}--1 ** ~\— r-f-— }-- j

Example 2b

The relationship between 2b and 2a is exactly the same

as that between 1b and 1a: the above is repeated in extended

form as a gawt. In this case, however, the melodic line of

2a is not strictly adhered to.

Sources: B D J K L M

For the first line of verse the following variants

occur (a durational value for which no pitch indication is

given being rendered by the relevant numeral):

.l.ri> j J. t.(
'---,,p....
k* n h -------
t ^ l *1
' ,-------
-----
X .1. ' y 1 2 * V H tfz

K j.\... If,:.
, o' i |j "fy • •jsj’ >i "}>/'» 1/ * •>*— *
'rV 1
i r V V
p *, r' r\
U >f
f •
p

Jr 1 1 .. *■ 1 . L
i o- yr .t — >fi>i j.
— 1— V V \, J l----- V, -I r ■r g f r h . I 7f r

-i— — h\— V-V.-h-p^i *

t .~V.„yu t- ^ :rk
D ■N
it
—f—
f* * v
— v f
,
rfr-vj— v— i
— y * -
i— > ?»-%-?> p1— p-p

J
-V ft- 1 ».--s-rfl r~~1-- r— i 1TfU *. i L»
■?, f- 1 -^ f.V Wh I»f r-^4^- 1 V> -- !-■'(> I>

■t . ■■ ----
■>
V1>f
L.
ji|>' j
*.—t' > t----r
— »t-1 V^'-'
’irf--
_
Ti
’W 1 -•■9.--p*— i-

In one MS., J, notation is given for the second shajr

of the second line:

 From the resemblance of this to the second shafr of the

first line and, especially, the absence of notation in the

other MSS., it may be presumed that the melody of the first

line was repeated unchanged in the second.

As with 1b, there is more agreement in the melodic line

than in the durational values. The total number of time

units for the first shafr in K is 48 (four cycles of the

ramal rhythm), and this may be accepted as correct. The

last three notes assigned to the first shafr in B and D may

therefore be considered to belong properly to the second.

In this, however, the number of time units is nowhere 48,

and it would appear reasonable, as in 1b, to posit a pause

(here of two time units) at the end of each line. We have

in consequence:

.» V h . Jr-jV y\-- +T>: . V, h----- ;

0 —.1u
-I — I--- '-J— b— -J .... > {, 1-----M —

4a— la al-ha.i ri la watL-lahi ma ’a— na gabi—— ru

M \ .. ^ i
juft., , * ->-f>■l.\--yft-TT"
4 *• :p 1 K- V \i f. ; -t- iq=t

wa ghayrl 4a— la f aqdij il~?af.ibbati qa-— diru

+ __ 1 ____ __ — — -
*— #7— ;—<r— ;— 1 W . '9 m "
4 lb M — H J. .
.. y V y . hn^;-i— :-p— t* 4----- H ---

katamtu ha— wa---kum khifatan min 4awadhilx

 V )

4r- b. , 1>»/TF» -y.'-vo

m «'1m"
••
±
— r
rzti . $£
H*- t ± t±± 4 £= 3:
wa li wa la— lmm 1 inda al-liqa— ’i sara ’iru

Example 5

Sources: A B C D E F J K L M
Form: farlqa

Mode ; ? mut
jannab

Rhythm: ramal

The following variants occur:

ir
F

4r
V
I

A B C 4r
D K L 5
; <L f-r ^ - v p ^ y Vt~^t=*%7
M -» 1 1 t- ^ ~
t>.:.vkv
" .
z> v r -
- ' 1 1 1- Tj--M4 f
-

4r S— *----- L f
j 7«--^ 1»i
V 1 I # f
* » Jr.
i p » r i'~ P t

The version in A B C D K L may be accepted without

hesitation. It comprises four cycles of the ramal rhythm:
 --- ^ ^ ---
...raj* U* ' • f "t" "f; ■ " “T “.■-------
— I— : |* -jfi— £— j— :— £—
..' k 4 7Z - P - - 1..; J |; f p

Example 4

Sources: A B C D E E G J Ii L M

Mo d e : ? mufrlaq

Rhythm:

This composition comprises nine rhythmic cycles, each

heing notated separately* The variants for the first two

cycles are:

B D E F — |?V 4 ,
M > » J .. J/R v t v v .... n , *..ky
G J M —».-t v - -. r

— h-4.— r *• fr-k * ^ --i— C-- i----r*--- i--

K V t— u.f ?— M“
—^ H — — 1— -+--- fr

L I?,..-.- >•0 u, i ¥? • • f,
■o 'fe-M — - -M > - _>»— If H -------^-------V—k-

•>-ti L •■, 9— \— §----- 4#----i----r—

..H M r fr -V 1/ n - V f JL— y, y ^.....i.jif-----------------
p ' f— k...-y-
r

------------
-i—r— =— iy..t>i"£------
-h— p» ■trr-’
Tfc
..?—b
» — -----------------------------------
r*; fj,------ Jt ----------
• > *-i—r—i------— ;— t—
..\ ^ I h I— ll—k—Lfy— Vi V—M t w / ^ k i/

The version in B D E E G J M i s given for cycles

 U3o;

3 and 4 also in A B C D E F G K L M. It is clear,

therefore, that cycles 3 and 4 repeat cycles 1 and 2 exactly.

For the remaining cycles the variants are:

4r-h %
E T D

B I) B E F \rr^L

F K J L

-,.. h
J
"!
— CT>ft fTP
i > , 1». j*w* *•
#"--- j--n—
^ r k »* K C -b p. ;If—— ff— t— fa—--- j V v—

fc~4r
G A

4r-W-
L
 g
o, g
7

ft 1V-w— r
A B O E *~y V j
t>- - - ^ b /-
p. . . 1» <6.
D E E

J IC

i1 I' i* y
i»-P W ~ h f. .1
M ^ p> ^ b |//# \> Yzp.

B J
___ d t j z / A .. 3.1 ^ ^
K D
G y Vr—K*--jr,
I,: y.r ,

(I . K . U* W
A
^ *.* n .f
4.. J . ji. f a n
L
if

o ~f >- ^ 'li "#"

y: . t H . . p -
T
• ,•
t
^
.. ^
f~v b V

----- -■-*»..,.V ---- 1

K-l:. *« ,-
M b » 1r- »b t ' f \ b.
. ./._.(

■..1 |----------- - —
.. -- ----- (# .fc
- ■■»— ■-* if m -- j--- V* t
7
G l y f t m W 1 k = 4 _ - — •>— ¥-ir\.^ h

— - ................ w"Mfr .....* " k 'ft—b...

L 1- . J
+
r. . J-Ti- f->. * r
i-----,----------- —»--------------------------------

Talcing the accentual pattern of the rhythmic cycle into

account, the following version may he tentatively proposed:

 K0>£)

■fe-H tA -H- V >« j. -iqy *&

h.•* M-
V -T I£ ==£
14---- ^ Lk.> *

p=j—I • /*> -r- . < .-b. ' viv 1 • fcr-f. •V*

1 ],y -----------r
-p n q— n *fi' l ' rw*
- rUJ
• *

■1># • t* •_ ; >-. iv l — r f c - m
-*P
_• = P J — K— ;— A-4

'The above four examples may best be discussed in

conjunction with the more extended composition notated in

durrat al-tat
j■ This, according to Qufcb al-Din, is a

qawl; unfortunately it does not correspond exactly to the

earliest description of the form, that contained in the

1
shark, and there are no later notated examples with which

its structure might be compared. The mode is 60 muhayyir

frusayni (with others being briefly introduced), and the

1 BM. MS. Or. 2561, fol. 151v.

 (555)

rhythm khafif. gafl al-Din does not mention the khafif

cycle, which according to Qu£b al-Din consists of 16 time

units with the following accentual pattern:

X . X . X x o . X . X . X o o .

The method of notation, although related to that used

by gafi al-DIn, is far more complex and detailed. It consists

of 6 horizontal lines marked off by 17 vertical lines

encompassing the 16 time units of the rhythmic cycle. The

notes employed are entered between the top two horizontal

lines. Below them dots (from one to five to the time unit)

indicate percussions (naqarat) to be produced, presumably,

on an accompanying instrument. The layout suggests that a

tambourine rather than a drum might be employed. The dots

are on two levels: the higher will be transcribed — J— ,

the lower f . Between the third and fourth horizontal

lines are given the changes of mode. Each time the basic

mode (60 mupayyir husayni, not itself mentioned in the

notation) is temporarily replaced the name of the new mode is

entered, the writing being extended horizontally to coincide

with its duration. Several, but not all, of the next set

of entries correspond to expression marks in Western

notation: the terms used will be given below. Finally, the

syllables to be sung are entered between the fifth and sixth

lines.

The notation comprises 15 rhythmic cycles, but certain

sections of the whole are repeated. At the end of the tenth
cycle a repeat from the first is indicated, and at the end

of the 15th a repeat of the fifth to tenth cycles is called

for. There are therefore 51 cycles in all: 1-10, 1-15? and

5-10. In the first ten cycles certain notes and other signs

are in red, and it has been assumed in the transcription that

these relate to the repeats.

The terms in the notation referred to above have been

rendered as follows:

madd: indicates the prolongation of a note beyond the time

unit box in which it is entered. It is incorporated in

the time value assigned to that note in the transcription,

wagf: indicates a pause, and is rendered by a rest.

-lahjT. is transcribed by f .
mushaddad is transcribed by sf.

jahr and mushaddad, considered as a composite entry, are

rendered by f f .

khufut is transcribed by p ^ — —» .

mufakhkham occurs once, above the syllable la, and may only

indicate a velarized pronunciation. It is rendered by rfz.

mukhtalas is transcribed by a dot ( • ).

Unless otherwise indicated, the above apply only to the note

they are placed under. Any additional signs are placed in

brackets.

Three MSS. have been consulted: BM. Add. 7694-> and India

Office Ethb 2219 and 2220. Most of the variants have been
included in the transcription, which is as follows:
 (.335;

« 0 0 1 # * ♦ 1 • t » , •

n --------------- * — 1--------- 2— 1--------!--------u.— 1---------- 1-------1 — 1 ^ s_ | _ /

16
ya mall---- kan bihi ya — 4-T
f -L
8

rD-E:.
u]
fu u bu u y vu
— -------------- 2 ---- n — •----1— «-- »— 0— 0 -- » 0-9 -- 1— |— «— 1-- 0----1-- *--- (

M -f **»
[> y b y
2
: f. f -gr
••
% - i- y
-k=ag=sj
5j“ ”$F

m ., . y m m m rm m m m .T.
trrr+
u8 y u b b

3
£
V b Ig ¥
F F p j
I
bu za— ma-

-*_«— i--- 9 g 1---- «--- f— «— <----«-----•— i— i------ 0— #-- •--- «--- »----- » ♦ 1--- *-
y l> i> l, b b U U l> U b U U b U U
4
0 0 0 0
t'— f ± ± t
i w
ni dum mada al-dah-

3
1— I— «— I— I— ► -»-1—4-
b U b b b b bb u yj y u p u uj
eiSta; —«—*-
 (536)

e— >
A__ I
‘ ~^i--^ \ ty|* ^ %» f~1?t , >n "
7z ~1>« t%• |
* 1>a f
¥ L ’•c z' qv *z z n I 7 r 7 «T T 7 --a
-»-7

# S2a) s^j— -J^si 3^1

ri ra---- fi— la

-#—»— i— •— f ♦... f f i j T T B j m m ffn m r a r a m i m a u n

u j u uj u $ $
owuv: Si

9 * ■■ . i* .
r-------- ’ ---^•--- 1-
t

pn ffn in m
W-«— »—♦—<—i—a—♦—#— »-»-«?■ f— I— a---*— i— > a a a a a a • a-
U l' U i> i* v v v \ v vv
ouifri
7 omo/
•" T r P T 5 " y
------- r-i—
7 H 7----- 1 ■ A, L fc-V----- — t>f----------
i ;— V------------ — , - U - J --------------1--------------

---- a1a a=f\

a •— •— — ■— •------ 3 T (-1
h-if m
-a— tN tr — m -4--a-
ffi
a a-- a— a-- 0----»----*--- a-- -a---a--
• U U J t
c^st-cv; i«■#— 4
8 u

•*
-1fr- — f \
& T % t ) V i ' ^ Lf ^ 3

ffffl... ,
a U U Lj £>
mniarn ro ra m |.> ffnffi.n
OSSw.
Jti—an•--a—a— a-
LI V
 (.337 )

9
■W
Uq '# 1 '' I t&liI* * ■■■■Mb»
.j,>■■■■■HP ■■■?
■I.W *■
■ b> <J
II

.
n fi al-’ama-

jSL.fgjBia.fli
u 3

10
ir
-£> Lj L3

■ni tililililara dir

■»■■■> •— •— *— 1--- *---*-- #--- 1-- *--1- * Fn jd ffflro

LU U U ]f V V I/ V V I
/

11
t «

H I
la ba—riji' ta al-za— ma----na fi

_! f 1----- 1--- *- 4-1-- m -.m .j t u .

U b U U P U b U U l '
ossuv: -*
vuvu i/
12

^ m r r^ .
t ? m m y
u u O
 (338)

13
4s * 4r — f
1? tr
.Li. . p

gil- li 4ay-

u m i' m u y i
/ u i* u u f u u
14

•____I
ft
4--- F —

*
shin a— — mi— nan mi n
3
-*■-- >-- *---f-- «— *-1—*— »
U UJ U U l> U LU

u
15
4r~$
liu — jfc#-------- ^ .bji f fc * j?# t 1?» 1
3c ■V =sL aBal—i^ L _ tizfc
^=r
ta— wa------ri— qi-----------------------------

3 3 fr^mrBnifn J4. m j p l
I I I JL5SSj IwSSI
5>* M 1 (_J
SnwZ 3 3

16
V »r’
.t
2>. +

— ■ ■"*»V I- .———I
rn.m j h m ■»— ■«— «■
U F U F F F 1/1/1/ F F F
M&la,: —
 (359)

17
—- .!— >#— ^ jl# —?■»---—7#— {?#
i i* {?# ^t'
■& w y u 'if y 'v u

O T .
y

■■ ^ *5. ly t, * f t*. >*i*-------1f----ri

flASut-: kj y , y P S j j— y j
m m m . . m . m
LJ l> U 1/

18
; I--—
3.— yst-- --- ]— — r
K* >» f. 1" >« | IrJ— fJal > y--
ct —f
. J — tg|-.\ggi— .fcj--—

h rrl
«i■>■»■■> «— «— »— *— i— *— « fTfr)FT!ff) fTr)f7) Jf! . ffnm.jn *
5 t=J LJ LJ b ^
PP
u v
19
■L* ~w Vr
4 --fr - S ? P' v .
: £ = = = = £
al~ Ip.a— — — datha*

m m ffw r n m
u
-Mi*; S .
20
---------- ------------- ---- fa— i---
<r#— j-.-------- i— — - .. ■ - = * — u
css
f j - r j t : -
’fey ^
V' --

-— --- ni till lili lara dir

3
---1—«— i— i— i— #-- --- ----1--- --- . . P.... m m . - f f n m
 <^40;

21

na tililililaradir tan tan ta dir dir na

22

ta na taradiri t al analadi111in alaradirit al analadililili

23

^ F W r ; ft-;. T W 7 = * T 7 ^

(>=*- ?=sm f-
lilililila na tala dililaradiri tililililana na na na na

mm m m mnm an mam m aun

o$sl&,i— »Frrf
r ft

24-

)+-
£
f>^- f>£=»---- r;f*— —
ta na tara diri na tilililila

ffn fh m .m
u vu
 V’"1 t5”-^ . 4
■;Vsf •
, J-v‘:.\,
••■•••
'VkvV*;;
••. :.

4
-v ’f r * . ■’ "■ . . .'; --

.... .■ v...V ...

r:v •Y'>y‘' y '/& •.- V. ■£*-*.
.'.r>
t, - :: .*"
=.-- - ■**.'•, v‘
V’r
v?ve; -5*-T M .r,;:.P.
"1” .. ,k,. 5f V f- I-•«. x‘-'; Vy •> =
■*• * IT'". •. r:*.'i
I--- bs ■■■■■■ '-v-.^------ 41-- 1 --------C-■'...,.'—-- : -’T^TT--- r”^ ...."" 'sr -,h
-'
t-ttt— —— |r--.' ~ ~ ~
----- r— — -rr------- .'..>— !
— —— !
— --:
--:----- :— 7 ... ..... ■
---- ‘ ' - ■’•
C'V ,.fT
'-. . - - T -;V 1 1 <fc£: ,44: - • mi— nan-- “.;■riiin- .XV -
...7.4 ' .v, y
•••4. • ' V: ’4 :~7
7' . y
«*• # *■
*
;4...
•
.•::/•7.4 . . . - y " u ^ U v . u u - y ^ y
-yr "■■■„:I!,::y v ^ ■.•;■.,-. :;:-./•>'■
> ■-■- <• ■■..■■•ggy
, ; ■- ■ <— ', . ' - V - ,’ ; ■
y . , : ; - . . y . .-■ v
' , . ' -v . •■ . „ xxx -x , ■ . . ! i . ■,

V
' ■ ■ '.. .4' '
7
'-I- * „V *, f|-v
.rJiv.C^' . . V ' . ,777/
V-5. ;v4'*
' .v.
*. ...v 4:.^ tt- .4J* * ’■ • JT- >• U' *' -rJ

fv^V'.3

^ -. ■■• v- .’:w' •■ •
■ •■:.>

;
i;.;.'. •. . ^ x. v
.
■.■ '
ty y uj u ■■'■ ■' '■ .v
.
,• •• •
..........
^ V.
ussii'.i!®.
-.x. ■- ■ .-^jy■::,x--i .r ;■
■; t>...
7.";-“? : : i-„'.
x'v-r • ; V ; ' v: X &
v-/ •
.v-\'> ■.
.j
. .. ”, > t' i/
v«ty | ^ ^1,A.i.s'iX-A' 5.-.vTjr ? -4.^ >rrr . *> ■ : •- ’
/ »•l^‘I' .,1^557""* 1,1^ I 1 1 0 ;"
* • .... X. . . . -V_ ••i
ir ■* s
j ^.1 »*
.
.
.—.-^.'^,
saa I * ■■■ i --- “!;.... ..■:.,*■ 1
1 -•
,•' ? 1 ..
■■*■.■■■■m: ' L . ■■ ; ; w ------ •’-4—
■-V. -x •. . y 'y.
\ « w* , • • -3.". > L 'it
V '• ..

y k y k \/ v v v yyvF i
;
mvc,\

.asv

X-T.V-xxx■v.y ' m
£*■ •.;•• .

SSSt" |l" «< - ' VI" i-t".

'
V y 'U P V^'W;
:-.V v - ; _

» ))>4 ^ -.»..fr,x

"■'i'
4
y : t' ^v
 (3^2) I;

■'-''I••'X’’

*:29

-0
•,■al— .a a

_v
u y u u^ y □ •
/u u
6SS«VV ^ vr...f
■■.;■ -v ’
•XX .-7
■\
■5-il
-'•■*vx
*'* ' «.
I -rrr

- ' '••• V ' • v '

-r- nx v •.
V>

UJ LJ "k: V ' V !\> V V y V

•J. V

iT**'

-7 >
\y:'
.
■ \r '2;: i , . v' ;*: * ;';'X
:_iysS i‘x • ' ■' ' X... - X J
- ,; ►. •
■ ’
;^X\ j-'!'•"
-' JV *?'-*'
.?«
r yy’
V ,

y'V
'• X ..:.
*.
'•2 --V '2 ’.
-.. ••' > % ' ■■ ■
'•■?-;Vx-x . .v:
 When compared with the above composition, the four

examples of notation supplied by gafi al-Din appear to be

no more than simplified and abbreviated text-book ilInstigations.

Certainly gafl al-Din implies, at the end of chapter 14 in

kitab al-adwar, that other compositions might be far

more complex, for he states that his intention is to give

some examples of the tarxqa and gawt forms that are easy to
1
grasp (sahlat al-1anawu1 ).

The simplest of the four is the first, consisting of

a descending 3-note phrase which is then repeated one step

lower. The whole does not exceed the limit of the one unit,

9 nawruz /1 21?3 b 4/. After presenting 1b, gafx al-DIn

states that the mode of the piece might be changed to foijazi

(12 /1 21? 3 4-/) by substituting e(t>) for eb ; to rast

(8 /1 2 3*b 4/) by substituting d and e1? for dt and eb

respectively; and to zxrafkand ( 1 / 1 2t 3b 3i?/) by

2
substituting e1? for f. This suggests that example 1 might

not in fact be an actual composition - or even the bare

outlines of one - but an elementary illustration of the

technique of notation and nothing more.

The second exampleis somewhat more interesting. While

1a and 1b coincide in melodic shape, 2a and 2b do not- The

1 BM. MS. Or. 136, fol. 38.

2 Ibid., fol. 39- Thus we have here another instance of

the use of the term jiijazi,which d o e s ^ p p e a r in the list
of species, to designate one such unit.
reason for this is not clear; however the melodic movement

in both is of the same type and differs from that of 1, being

not a descending sequence but an up and down movement pivoting

on c, which is both the initial and the final note, Ihus

the outline of 2b may be schematically represented as:

Except for a leap of a neutral third and another of a minor

third the melodic progression, as in example 1, is stepwise

throughout.

It has already been noted that the compass of this example,

which comprises the notes from A*b to e1> of the scale 59st

kawasht G A1? B1? c dl? eb fS g, might be cited in

support of the suggestion that 59a could be a theoretical

extension of a mode listed by Qu'frb al-Din, 34- i gfahanak

(or kawasht) G At? B? c d1? eb eV . But to whichever

scale this example properly relates, the units of which it

is formed are not developed separately: there is no

suggestion of a movement away from c to dwell, however

temporarily, on another prominent (unit boundary) note. It

may also be pointed out that the up and down melodic motion

around the pivotal c corresponds exactly to Qu£b al-Dxn5s

account of the modal characteristics of a number of shujgbas

(73 dugah, 77 rakbl, 78 mubarqa*, and the first segment of

 W 5 )

79 salmak), and of 57 zankula. One of these shu* has, 78

mubarqac , does in fact relate to the igfahanak scale,
although with d1? as the pivotal prominent note* Conceivably,
therefore, its melodic properties may be derived from those
dgfadanak, given that example 2 is representative,
especially when we consider that the final note in mubarqa*
is not the prominent dl?, but generally c. We would thus have

Example 5 is stated to be in an ancient (qadim) mode

called mujannab, a term otherwise used by gafi al-Din only
to designate a fret on the lute and the note it produces.
Two such notes, Al? and dl?, occur in this piece if we assume
G, c, and f to relate to the open strings.According to
1
*Abd al-Qadir, however, muj annab was also one of the terms
used in connection with the six tetrachords known collectively
2
as "kde agabi* , fingerings. These tetrachords are in fact
the familiar units /1 2 3 4-A A 2 3^ A/, /1 2 3i? A/\
/1 21? 3^ A/, /1 2t 3b 4/ and /1 2b jb A/, so that
agabi* as such would appear to be little more than,
indications of how to produce them on the fingerboard of

1 maqagid al-alban, 96-97•

2 Smother term employed to designate the set is mawa,jib.
 two

1
the lute, starting always from the open string. The

history of these terms is nevertheless obscure, and at an

earlier period they may have implied specific modal properties.

Also, it would appear that at one stage they were compounded

with six of the rhythmic cycles (one of them being ramal)

to produce a corpus of 36 fared i q , and it is conceivable

that example 2 is a survival from this group. Certainly the

heading, muc
jannab al-ramal, is of the same type as those
2
cited by *Abd al-Qadir, and unless there were some particular

association of this nature there would seem to be no good

reason for failing to assign this example to tiraq G A1? B1?

c dt et f (fit) g*

The fourth example has a similar heading: fariqa fi

al-qadim thaqil awwal muflaq, muflag again designating one

of the agabi) . But what is meant by this term in the context

of the present example is not clear, and in conformity with

£>afi al-Dln* s usual method of analysis one might suggest

that it should be considered a compound of the units 9 nawruz

/1 2f 4/ (G - c) and 1 zirafkand /1 2t 3^ 3^/ (c - et)

in cycles 5 8, while cycles 6, 7 t and 9 consist of 37

nawruz G A? Bb c d1? eb f. Cycles 1 - 4 could belong

either to 37 nawruz or to the compound of 9 nawruz and 1

1 There may even have been a pedagogic element in this

since all the finger positions (frets) are thereby brought
into play.

2 loc. cit.
 VPH-/ ;

1 2
zirafkand, for neither f nor et occur in them. However,

it should be pointed out that the units are not developed

separately, especially in cycles 6, 7? sincL 9? where wide

leaps occur.

In addition to exhibiting the predominantly stepwise

structure found in examples 1 - 3 , cycles 1 - 5 snd 8 evince

certain similarities with the up and down movement

characteristic of example 2. Further, cycles 1 - 4 reflect

the structure of example 1 in the exact repetition of a

phrase one step lower in the scale. The remaining cycles,

however, 6, 7» and 9? would appear to contain none of these

features, and the leaps of a fourth and a minor seventh,

using the notes G, c, and f especially, suggest that the

melodic articulation in these cycles may have been influenced

by instrumental techniques, in particular the use of the open

strings of the lute. Since the open string note is termed

mutlag there may even be some connection between this feature

and the heading.

Certain aspects of the structure of these examples are

given in the following tables, which take into account 1b,

2a, 2b, 3 and all cycles of 4. It must however be stressed

1 Which coincides with the mode (listed only by Qupb al-

Din) 36 kuchek-i t am am G At Bb c dt eb et.

2 sharfr contains a rather curious analysis of this

piece (BM. IIS. Or. 2361, ff. 152v-153) in which the suggestion
is made that cycles 5 nnd 8 may considered to belong to
nawruz with an extra note added.
 V J

that the four examples constitute far too small a sample

for firm conclusions to be drawn-

1 1b 2 3 4
melodic ambitus fourth fifth octave minor seventh
initial note f c e1? Bb
final note c c G c

direction of movement between contiguous notes

1b 2a 2b 5 4 total °/ 0

r--
falling; 16 28 11

o
42. 9

o
o
5
rising 7 5 20 5 39 76 32. 6
level 48 0 9 0 0 57 24. 5

$ incidence of interval types

falling °'/0 rising °//0 total °//0
r~f
quartertone 7 0 0 7 4
thr ee“quart ert one 53 58 33 50 96 54.5
whole-tone 17 17 17 22.4 34 19-3
minor third 6 6 5 6 .6 11 6.3
neutral third 5 5 7 9*2 12 6.8
fourth 5 3 4 5-3 7 4
fifth less quartertone 2 2 1 1.3 3 1.7
fifth 1 1 1 1.3 2 1.1
minor sixth 0 0 1 1.3 1 0.6
sixth less quartertone 1 1 0 0 1 0.6
minor seventh 0 0 2 2.6 2 1.1

100 76 176
 In the transcription of the composition notated in the

durrat al-taj one of the sets of indications given by Qu£b

al-DIn has not been included. This shows the changes of

mode, and is incorporated in the following summary which is

a conflation of cycles 1 - 20, followed by cycles 21 - 2 5 ,

intended to show the melodic contour in abstract, without

reference to durational values. Repeated notes are ignored,

and phrase endings are marked by half bars.

CD u. (ib uo (iii) oa)

&

(
V)
£ U) . (vri) (ll)
£ :[ 1 4 ..•••(>)-*: T~t» . »-+)

. <? ,0 V
(X) (*»>
bir * {f
-#---- — r- IS: -k * f-ferV

(yiO *
> (
A,)

a ® 55
frigar
fcvfi)A3 (xix)
,» -fr,
r •...—-fcfKt ... #
I.., t , 5 I—# .7"Z5~
T
■t J fc
ipfahan zankula
(xx) 2? (v)* S’

(joining up with cycle 5)

phrase (I) (cycles 1,2) revolves round the upper octave note,

g, which it serves to establish. By ending abruptly on d

it also serves to identify the limits of the upper unit

/1 21? 3b 4-/ (d - g). (The cif may be considered a grace-

note: it is not intrinsic to the mode, which is 60

mufrayyir frusayni G- Ai? Bb c d et f g.)

phrase (ii) (cycle 2) fills out this tetrachord, beginning

with et, the note missing from (i). The cadence figure c -

f - c, however, draws attention away from the d - g unit

to another tetrachord, thereby underlining the duality

inherent in the upper pentachord of mufcayyir husayni (/1 2

3^ 4-/1/ or /T/1 3b 4-/), of which it completes the

exposition.
1
phrase (iii) (cycle 3) contains the first modulation to

occur. It consists of a simple descending phrase

identifiable as 10 ru-yi tiraq /1 21? 3^ 4-/ only by the

final note, B1?.

phrase (iv) (cycles 3*4-) returns with an abrupt and dissonant

jump (B1? - f) to reestablish the upper octave note,

phrase (v) (cycles 4-, 5) pursues the exploration of the upper

unit (d - g) with a florid passage in which f# is added.

No indication of a change of mode is given, even though the

fif cannot be considered a grace-note (like the c# in (i))

or a leading note (fii - g does not occur). In spite of

1 Used here to mean change of mode: no reference to

alteration of pitch is implied.
 Qufcb al-Din’s silence 011 the matter the prominence accorded

the fli, together with the fact that this long phrase remains

strictly within the limits of the d - g unit, renders it

nevertheless likely that (v) should be identified with

^ ifff&ban: presumably the juxtaposition of 11 i$fahan

/1 21? 5b 5 4*/ and /1 21? pb 4-/ was so common that it

1
was hardly considered a modulation,

phrase (vi) (cycle 6) reverts to "pure” mufrayyir fousayni,

again stressing the d - g unit, but completing the uppei*

pentachord by ending 011 a long-held c.

phrase (vii) (cycle 7 ) again emphasizes d and c, and at the

same time continues the descending movement into the lower

tetrachord (G - c), although without introducing the tonic G.

phrase (viii) (cycle 8) begins as a repeat of (vii) one step

lower in the scale, being thus reminiscent of the sequential

patterns which occur in the examples of notation given by

gafl al-bin. The x'epeat only continues for the fix^st few

notes, however, and after bidefly touching the tonic G the

phrase ibises into the c - f tetrachoi’d before terminating

on the same note as (vii), Bb .

phrases (ix) and (x) (cycle 9) should almost certainly be

taken together. fhey constitute the most puzzling passage

in the whole piece: not only is there 110 indication of a

modulation, despite the prominence of the intrusive eb, but

1 Of. the large number of combinations derived from the

juxtaposition of these and related units (presented
schematically on p. 3 0 7 ) •
 there is also no parallel to the leaps of a minor sixth,

a feature which contrasts markedly with the almost

1
exclusively stepwise movement found elsewhere. The mode

could be the not unrelated 47 frusayni G Ab Bb c d eb

f g, although there are other possibilities. It might

also be conjectured that (ix) and (x), which terminate a

section extending over cycles 5 ~ 10, are designed as a

bravura climax before the more sober delivery of the last

word(s) in the line of verse. If so, the wide leaps might

form a deliberate contrast with what follows,

phrase (xi) (cycles 9,10) is linked to (x) in the repeat. It

forms a cadential figure descending through the lower

tetrachord to end on the tonic G, thus concluding both lines

of verse (cycles 10, 20) and the whole composition (cycle

31). Ab is briefly introduced,

phrase (xii) (cycle 10) is a short rising link passage

spanning the c - f tetrachord which leads in to the repeat

of phrases (i) to (xi) for the second line of verse. Its

function is simply to take the melody back up from the

tonic G to the upper octave note.

The following phrases in cycles 21 - 25 constitute a

separate section before the return of the last hemistich,

like phrase (xii), they are sung to nonsense syllables.

1 So much so that one might be tempted to consider I (G)

a scribal error for U (d). However, all three MSS. are
in agreement.
phrase (xiii) (cycle 21) repeats phrase (xii) one step higher

in the scale, thus emphasising the arrival at the upper

octave note (and hence incidentally the d - g tetrachord).

phrase (xiv) (cycle 22) should perhaps be considered a part

of (xiii), which in a sense it summarizes,

phrase (xv) (cycle 22) is the second modulation to be marked

in by Qufb al-Din. The mode is 2 5 b frigar /1 21 3b 3^

4ic 5/ (c - g), particular attention being paid to the two

thirds (eb and e1?).

phrase (xvi) (cycle 22) presents a further modulation, into

^ igf3-han /1 2"b 3b 3 4/ (c - f) with the fifth, g,

added. It consists of a simple descent and ends, like

(xv), on the tonic c.

phrases (xvii) and (xviii) (cycles 23,24) comprise the final

modulation, a more extended presentation of 57 zankula

G A Bt c d'b e f. If we disregard the presence of G,

(xvii) could be considered a model illustration of Qufb

al-Din’s description of this mode (given in chapter 7):

"When c often recurs and there is an ascending and descending

melodic progression on the notes c dt Bb A the

characteristic form of zankula becomes manifest, especially

when e is itself added". The notes are introduced in

exactly this order, with e only appearing as the last note

of the phrase. (xviii) completes the upper tetrachord by

adding f, and ends on the pivotal prominent c.

phrase (xix) (cycle 24) may be considered a reassertion of G

as tonic after a series of modulations in which c is

 prominent.

phrase (xx) (cycle 2 5 ) leaps from this straight to the upper

octave note, reinforcing the effect of (xix) and preparing

the reintroduction of the last hemistich which begins in

the upper unit (d - g).

The structure of the whole may be summarized thus:

(cycle) range

(i) 1,2 tetrachord d-g cil-at exposition of the

tetrachord d-g
(ii) 2 c - g upper pentachord c-g
tetrachord c-f
(iii) 3 modulation:ru- Bt-et

(iv) 3,4- return to tetrachord d-g et- k first hemistich

(?inherent) modu­
(v) 4-,5 lation :igfahjn
d - g development of the
return:restatement
(vi) 6 of c-g pentachord c -at? upp er p entachord c-g

(vii) 7 entry of pentachord G-d At- d exposition of the

(viii) 8 completion G -et> lower pentachord G-d

(ix) modulation:
9 G - f
?4? jrusayni
(x)
restatement of the
(xi) 9,10 return, cadence G - c lower tetrachord G-c
second hemistich
(xii) 10 bridge passage > reintroduction of
repeat of (i )- (xi),then c - f
(xiii) 21, 2c
continuation d-g upper pentachord
(xiv)
(xv) 22 m odul at i o n :frigar c - S
modes with
(xvi) 22 modulation:igfah an c “ g
c as tonic
(xvii)
23,24 modulation:zankula G - f
(xviii)

(xix) return:bridge passage G G reestablished

24,25
(xx) — ►repeat of (v)-(xi) f - as tonic
 This may be further reduced to:

(iv)
(i) (vii) (ix) (xiii)
(iii) (v) (xi) (xii) ;
(ii) (viii) (x) (xiv)
(vi)
a b a' c d c' a' '-- --- a' '
u I i t
I
A B I' '

(xvii) (xix) (v) (vii) (ix)

(xv) (xvi) (xi)
(xviii) (xx) (vi) (viii) (x)

e f 6 h a/ c d c

T
0 A' B

In addition to the balanced structure resulting from this

arrangement, a further pattern of alternation may be discerned

in the division of the verse:

cycle first hemistich cycle second hemistich

dum mad a al-dah-

4 y a malikan bihi yafri- 5 ri rafilan ---- — ■■■—

2 7 -------------------------
Q ---------------- -----------------

3 ■bu zama- 9 fi al-ama--

4- — ni- ■// 10 — ni- //

In each hemistich the style is relatively simple and

declamatory, presumably in the interests of comprehensibility,

for the sections

 /ya malnkan bihi ya/ /bu zamani/

/dum mada al-dahri rafi/ /fi al-amani/.

The medial syllables /fi/ and /lan/, however, are stretched

out over two and.four cycles respectively, and it is in these

that the ornate passages exploring the melodic potential of

the various units occur* A concomitant differentiation of

range is also found. For /ya malikan bihi ya/ the range is

f - g; for /bu zamani/ ef - g; for /dum mada al-dahri rafi/

et - g; and for /fi al-amani/ G - c. The range for /fi/, on

the other hand, is B1? - a? , and for /lan/ G - a1?.

With regard to the relationship between the melodic

phrases end the accentual pattern of the rhythmic cycle, it

may be recalled that the latter was symbolized as

/X . X . X x o . X . X . X o o . /

which might be rendered by

2+ 2 +d + 2+ 2 + 4

Since time units 5 and 11 are seldom marked we may for present

purposes simplify this, at the same time incorporating the

darb al-agl, to

12 (4 + 4+ 4) + A

But even this pattern is not strictly adhered to: time unit
 (557;

5 is not marked by a separate note in 16 of the 31 cycles,

nor time unit 9 in 13; and time units 1 and 13? for all that

they represent the (Larb al-agl, fail to be marked in 6 and 4

cycles respectively. With regard to the present composition,

the normative patterns within the rhythmic cycle appear

rather'to be of the following order:

1 !
time units 1 - 4-:
r> (occurs in 20 cycles)
H *

j J
time units 5 - S ; n
* »
»—
#
i
* (occurs in 25 cycles)

rI j0 jT l J i j tI

time units 9 - 1 2 : the pattern given fox’ time units 1 - 4

occurs in 11 cycles, while that given for time units

5 - 8 occurs in 14- cycles.

time units 13 - 16: the pattern given for time units 5 ™ 8

occurs in 21 cycles.

<j ; ? h g —
Thus in place of ^
8 1 5 9 13

one might suggest as a more accurate abstract in the present

instance the following:

12+4 P
-0---- 9-
\ P'\ I
■ \

5 13
 (358)

The percussions indicated accompany the melodic phrases

closely. The lower level, in which one finds * , * # and

v LJ
3 b
• , relates strictly to time values of * and above in the

melody, while the upper line, consisting of JbH , ,

I r r i , Jt~hn! and tn'TJ , generally corresponds to J5*! .

3 ^

FfT) is almost always followed by , which is the most

frequent pattern in the upper level. Where both levels

occur in the same cycle the change generally, although by no

means always, occurs at time units 5> 9? or 13*

The various features of intervallic organization in the

examples of notation from the k itab al-adwar which were

presented in tables 2 and J may be compared with the following

tables relating to the above composition. No account has

been taken of the variants-

direction of movement between contiguous notes

total °/
'o

falling 219 48-9

rising 180 40.9

level 49 10.9
incidence of interval types

o
Xailing 7'o rising %
total 7'o

quartertone, 5 2-3 1 0.6 6 1-5

semitone 26 11.9 15 8.2 41 10.25
three-quartertone 92 42 72 40 164 41
whole-tone 73 33-3 59 32.8 132 53
whole-tone plus
5 2.3 2 1.1 7 1-75
quartertone
minor third 1 0.5 1 0.6 2 0.5
neutral third 2 0.9 5 2.8 7 1-75
major third 1 0.5 5 2.8 6 1.5
fourth 5 2.3 6 3-3 11 2.75
fifth less quartertone 0 0 2 1.1 2 0.5
fifth 0 0 2 1.1 2 0.5
minor sixth 9 4.1 6 3.3 15 5-75
minor seventh 0 0 3 1.7 3 0.75
octave 0 0 1 0.6 1 0.25

219 180 399

 (,pbo;

'The relatively low figure for the level category here

relates to the high incidence of melismatic passages in which

there are few repeated notes. Such passages are lacking in

the examples of notation supplies by gafi al-Din, a fact

which might be cited in support of the contention that these

examples (especially the first) were deliberately chosen for

their simplicity, and thus can hardly be considered

representative of the melodic style of the period.

A higher degree of correlation is observable in the other

table. In both cases the three-quartertone and the whole-tone

account between them for over 70°/ of the total. What

' /o
variations there are may be attributed simply to the fact

that within such a small sample not all units can occur:

thus although it may be taken as a further illustration of

the relative importance of Zalzalian and diatonic units, no

especial significance attaches to the absence of the semitone

in the table for gafi al-DIn*s examples.

The units occurring in these, taking into account the

substitutions mentioned for the first, are:

/■\ 2t 3 b 3 $/, /1 2 3*9 4/, /1 21? 3t 4/, /1 2t 3 4/,

and 2*9 3 b 4/.

In the composition in the durrat al-taj the units occurring

are:

/*\ 2 3 I5 4 /, /'\ 2? 3*9 4/, /'\ 2t 3 4/, /1 2 'b 3b 4/,

/^\ 2t 3b 3 V,/l Zb 3b 31 4?! 5/, ?/l 2b 3b 4 / or ?/1 2 3b 4/.
part 4
 The following text of the kitab al-adwar is based on

four K B S . :

( 1 ) - BM. Or. 136. Dated 7 9 2 /1 3 9 0 . Olear naskhi. The

folio numbers given relate to this MS.

( <7* ) - BM. Or. 2361. Dated 1 0 7 3 /1 6 6 2 . Small, occasionally

careless, but generally clear n asta*liq.

( >) - BM. Or. 2361: the text of the kitab al-adwar

incorporated in the sharju Dated 1 0 7 4 /1 6 6 3 * Same

hand as ( ^ ).

( -> ) - Photographic reproduction of a MS., published bj

yusayn 4All Mahfug (Baghdad, 1 9 6 1 ). Dated 8 7 0 /1 4 6 5 *

The pointing is incomplete.

Considerable similarities exist between ( I ) and ( -> ),

the latter being distinguished chiefly by what it omits. Do

reference has been made to these omissions, and as a general

rule ( > ) is mentioned in the notes only when it contains

material not found in the other three MSS.. All the variants

in ( t ), ( *r ) ^ and ( > ) , as well as several marginalia, are

included, with the sole exception of the variants for the

musical examples in the last chapter: these, together with

the versions from several other MSS., have been presented in

chapter 8. Any additions to the text are given in brackets.

 There are a large number of MSS. of the kitab al-adw ar

in existence. The above four were chosen not for any textual

reasons, but merely on grounds of convenience. Although it

is hoped that the text as given here is both accurate and

comprehensible, no attempt has been made to purify it of all

later accretions: the occasional explanatory insertion may

even be found useful.

 u r q— i * v ^oJ 1 <i.lfi ^9 j 5 J 'i i V—»i_«i *5"

(jjO 1 t-U,C- fU^ ^■''■uliilj1 J ,^-viJLI

gf—-j.oJL-~eJ1 j L «aa1 (L-J1j 2_<X ^J1—Xi aJJ1 d

dJl_^oj 3— ^ ('■■■AijO j\^j'i1y jvjXuI1 3-?jJW ijp j^ 'it I—*I" 0

• {jj ^ 1 (jft<J-D1 'X.lJ'' ijJ1 d-J^ ^ ^^ 3
♦ ^^ j^9 JyI ^
 ^ ^j I dwLl I *-

(*) e _ _ (1 )
>A»»0ii%.^ L— 'i1 js~»J1> ijfli' A■
.-*1fJkLsi3J) ^JUn/iJ\_JlJ1 L>j dU wL«4J$J1

* a:; I dJ 1

j ^ ^^ ^ ^ ^ ^| ti""'n \ t—^ a i j ■n,o 1 *4^9 LmO 1

{i ) ......... _ (r )
*.t" ^ ^ j^sv-nO i^p ^ dJ ^v{51 1 <1 1 (J^^ tilJ

* \^ ^ lA.y^Pji d™ ^|^ \3 ^)l 1 ^ I tj w) 1^ L^i J ^J) ^ ^j) ^ 1‘

<n ) (° )
13U £ &— L&JJ ^ j-*.v Id CU-».>jlv jj yl d^ j —»d 1 Id ^jJ1 C^jJ Lt5

(Y )
|(ji^almiJi ^jO '
f
c
-
d ^ ^ ^ hiftj^ 'lAJ ild (ti I A Lij1 ^Ja^iiild^

^~sJ 13 ^1 3 ^Xr* ^ 3 ^ dj1 cuJUv-^*3 ♦ 3^-^XXt^J1 &*-L& ^5 dJ j <3^

^^w Ili-^J1 ^^ ^ ^*iJt*J3 ^ 1 i^5^ji«*^0fJi ^ -

(A )
t ^jJ1 ^iCLa^s^ ^ 3J\ j*i^.j\3 ^ 3 L#3^11 V^t^p\ jj^Jb ^y u^ )

* b u Ia a v ♦ 1 * d ^i>!iJ * »>• * ^
• T

♦ IwWAV ♦ 1 * 1 *d C^-tl^iLtv •
• 1

1J 1 1— -d * «S>1 '— j ♦ ( L a j ^ J VjlS • «s* » 0

* 1

* ^pIwS 1 * ^>> *y

4 ^jj—JLl 1 ihA^ 1J»nS3 * O 1 * \ *\

3 ^ ^~<* 1 * ^ ^I^1»JI <3Lvv j 3 Lfc—4J$ ^ ^Ay ^ 3 —^ (I3*^3
* ^£rLuJ 1 J Lm^ * O 131!^^ ^ U-J 3 L?V d^Xl 1 15?

* (uOa-J^ffU I J fcA«Svj I fj j tfi IS—^icJ 1 i

1 Viii <31. ii.tii.ij-' <yX~r ^.J_™.S\ ^1 jjn» JtJ d-—vi? L <| .0 ^1-^iS5* tei^ijOi J}

• f)11■if»JM»Ptin■■d (JmnfciWAJ i
 \ j

(')..,...............
» ^1—
~^ii5 ^-w-tu-C- A +4J-Ij ^

(V )
♦ <JJUJJ_9 5J„:>JI Ly t) ^..Jvij 1 (*-£jj-^ {jp tj j) ^ 1 1

* O— ~tP^ ^ ^---------- fjP (jj n ] \ 1

* ^ Lt^W 1 LaArlUJ <y UjJ J^oiH

♦ ^9 <J—i-t,^*- 1 t—*Luv 'i 1 jj.9 1 jJ 1 (J-iiafiJ 1

(Y )
.^ y^j\ <uj bji ^ Q«—5UJ I 1

(*—* V )*
■ ^ j \jj j H \

* O i J~* ^ ^r—--Us- LauJJ ^J»siasJ \

Ai. i
.a j Ijjj'tM ^ ^j-**—ja» 1jj 1 ^ U /j 1 A—j ^9 (_J—■* UJJ ^vftsJ 1

* d J ^"-'■'WoJ J J t ^ J *jM 'I o-^1^ qjp ^±-W L J1 J.^asj 1

* J 1jjJ'jl1 p-Tl-» (iJj \~u*J ^fS J~^v UJ ^ ^

. 0 1 SuUJ 1 J I jj J \ ^9 j----- **& (_5 J U J i (J^aij 1

* 1 ^J—^OsJ 1 L—1 1 ^jj) J , . CLlS- LJ \ J

(T u3) * £ 1— ®i ^ I^ ^i>JI <j^ cUJ UJI J.j'xil1

* ^r—_^JlJJ ^ ijp j~^-$J> JUaiJJ

4^»IW*'~*AI^ I^ ^ It^cj1 ^

^ — <J1 <^iJ 13 * toS=- • y* * (J—~4 J4** * *

>—* ♦ Y * J c^iOS ( \J u 5 ) rt-^1.1; J 2 ; Ly . y
 J-AJI

♦^ I^ d J fjIw^ I C—^ (_fP

Jjj 6wA»>J\ L.-0 J— LiL»Q j ^ ^,}i> it-fcO-SvJJ

(1),
^$ jU t J— all1jj 3J-5pJJ (^>*i 2 cU-»4.J7>j JS i s » ^_*JaJLi A>aJ1 jjJ^ii!Tr^rfj

(V)
O^ ^ L? (j^^ <3— Lf 1 2^L>* J)1 L^J1 d—*3_JwjU J1 JIj

(3 fi ^ ^wwOi^ju!I J Hj><<AiJLi 3jLr** jZ \ 2s \ d— 1

( r )
^U ^ {jO 1 i^i" 1 1 ^<M^UVA»4iiL ^ dSbli^lnHp^

i (J1*0.~i~lt*tl1 I (\»m L (1^ 1 ^ J 1

(1 )
* ^_J"I ^ ^_fjlI I*&* ^ ^jJ 1 3—y-v.ttJ Li 3—uuUsJJ ^ ■Y J i rt ■‘
j.i'XJ

(1) ^ (o )
j)ib jj ^ ,jiw v J Lf-^iuvW 3J— JJ 1 CL-1—.<j.^j.i J d vJ.jEb 3 »-Ls'. 1 j u>—J Jt

♦ Li
lI J,y5 f^ J |1J9I UIIl
if
l d ^ ^|i».
^

(Y )
J ^J-..C111 V->1ii"!| * Ijl^uI JhI^J1n j..fljI1q

^¥*LStiXmw^ij1
** ^^ l.T^l ^ ^VbrtlJitrL^l ■
l^iJ Up*^“^ i ^^P
.
Li.I*t w. l. fU^n4V^ ^ )^ <3 ^ i

(^ ) ^ _,
j ^ (J c^j c^ J jj j-j \ j i 5 dU J L^ij (JJ^J 1 U-iLjdwijj

(n ) ...... () • )
.9LjJ J $ jjjd Ljo ^9 jj
*c

* i^ i i •^ t ^ *i §^)ji ^vj **^!

*d »j
* J fl ■ 1 Ijj ^ 4 wi^* L—
I *^ 4 ^ol J 4 vi1* 4 Y

4 p, yjlj 4 J • "| * <5~EW ^ JJ 4 ty 4 £>

4 diMtilJp* ^ d ^ ijN J I 4 4^ 4 ^J w i1*!.J 1 ^ d 4 !—

^*“ 4 Y
4 CLl ^ 4 4^>» ^—* 4 ^ 4 4 (jAJ jj 4 ^ 4 ^

4 ^—,9 LJJ ^9 4 dU-^-VUJ J jj^S j(J 4 L_J 1 Ll aJJ1J "L??.— J ^0 4 \ 4 ^^

 ^ UN Jtail\

♦ ^j-,— uj->iLui fctU1 ^Lujf1 us9

(V)
J b .^1 O 1^ *J O J_c 1^ v ^ir- jjj' o 1— a y[& (j-oj LewjJ 1

lm-» 1—wUt-tJ5^ 1 ^1 Qj HIMj i J L?k-«t Ljv iiJ a ^-»

L^°

• jj JJ ^9 3w> OL-J^l j iJ„a

CT"

(° ) + 4*
J , ; ( < l ...,, .u
/f ’/

• / ^j ^JLi''lJ1 I—u U- {} ^ j ^ j \ U o l^>. o - A ^ i>

• / e- / (t u*) 1It

<Vi 1,,-Q J ^ l j) ^bn?! sji >b .p A'

‘A /

(*l ) + 4*
♦ ^—da/^ 1 j »—.^ 1 O j(k n ^ ^— .9i^J ^ 1

rii..a j/m ^*w^J 1 d**"^ c 3-'w j \ n* *

• I d

d i. n/> j ^ i ^.ujij1 4—hjl-^-J (_5^ ^JLaJ j ^ j' / | ^vAt^a-' ^

•A-/ f c/

j ^ ji I^auaJ1 i—-_>L^aJ {J?i& ..JLew t| LuvS1 J-'p' ^u.aj

•/-V

• /3 / j^i jS-iuJiJ1 i}— -> tjA* ^-ku 5 ^Lu±9J dJWu J

f J/

1.1.*'V ■ ^ja j-i"~ 1 1— f^’

-vS iLuiSi 1 (_5”1 1
— 'i-i.'dJ^ a ~ ujji™*-' j ^ m SU
r c/

I^ui? ^j^JLtJ\ t—rwL>< L—*.5^ \ IfJ \ j) ^Lt*e9\ J— aj / f* 6 j pii

„B *
* ||- ^ * |

• 6-J.O tyJAJ?*~~A t^vtu-f ^,Lc> * 1 & ■>-'-»l^ ' (JjjO 3“•*ij J VlA ♦Y

* vV^ ^Ju^WMv * ^ *Y
* I^L 1 x>Q jJ^iJ! j^*A/ 1 >4J^ • |Jju <}* * * ^—hi«dJ.9 > 1 ^
♦ 1 (^6 J >J b j • «*^ * l—i (jj d J 1j j • -j. • * ^- *o
 (369)

* * J ^ ^j dm'»n.U ^

^J^ 1 jytuiJ1 d~~J ^tjlfi A

/Vi/ <w-^ jp V* / p*vJ£

./J./ d»&—«£ jj^ 1 ^rf.Jlj1 &—*Jl-^J ^ pLvj»5\ {L±wj 1 j *—/j ^«iij A

A
J/N jsi Ltu51 ^ / ( V 1 u3 )

♦ j j o

^^ 1 \rvfc» ^ ^ / Ji—
^itaj j j

* j^ j | |5»LaJ^ ^^|5«**a1J 1-5^ ^jfl 1

4* y_
*a -^"£'*^ L»4Att5 ^ ^ I 1 ^ nVtuiJ! ( i . >Lt^j j^ ^ j ^Mijij ^

( ' ) +
njl* j ^ ’1KliJ lr-^W ^l ^j 3 1 u -*J l.s*.
• /-

A
IlliJ l^yjj 1 5~jvi j 1 jyVAiij
U •*■/ f'

(V)x A
X
" *
l *-».J 1J pUutSl ^J(U j 1
/t ^ / r

(V) *» 1 ♦
^ lc ^ L j\»> j) jiiv iS 1 6^bJ j \ / f 3/

»* \| « A
drn^J Vr^%J is?io^w ^ ( t ) f uJ . o-j ->U/
/r J / r

-» ^} ^ i* i**&) ^ £zls-K *i “*i ^ ^ ~ ^ ^ ^ * -> '?'£r

1)
----- 1_1
_4_j_I_|
_I_—L_1_1,,„l 1_J_*1 I « ■_I_L_1_1_^ *> . t t I I 1 I I 1 1 1 -1
isi*'

* wr^f^»* Lj*^*I^AaV *^ * f * ^ d 1^ ^ * -J- * * * —J- * j

* j J A»i^i ^ I 1 |>L^P j|J L*>u^i <M /(* J j ♦ W?1* SrJ ^ * Y^

1 ^ * ^* 1i*— ^ ^ ^ ^ I ^ I j |j J IjNkl i |^ ^

1 AoW1&j £) {_j~4io^J1 tkA-viw <LtJ> jj iaJ I tJ j) ^ ^i) j -fi. / * * * ^ ^■y"'1^ M j\ /^ (j/

jiLiu^1 ^I/^^^ ^Lwt51 ^^ J* 1 ♦^ I ^jwIaw 1 1 liti1^
 (370)

j j O^ \j La-vjJ 1 ;_jr^ ^

b-~*Q ^-™*W'a5 jC
J ^ J dJ j &A*L& j) ^ ^|*-taU ^ ^«j 1 Lm^

^ ^ ^vvJ j \ ■'*'^^i!^ j dJ ! ^j iijhiJ

fcA.5^5 j 1j "■■^■^W lwaQ\j? *d 1

(V ) ^wViw^ri^jj \

( O .„ . .., ('“ ) . ,< ............. ...

^—n<^J fi 1 ^*y■
*d j (jr^1 J~"3 tj—*3 LJ*vi<i,lj—Ji Jl ^ / 5/

(6 )
Jtj-l IhJ i (3fcS^S'’' ^y/ IaJ 1 I l#t.i I I [y C
^^m
J fc-J f... ^JI,

(1 )
/■-.J^ 1 J d j 1—dJJ d-d liJ 1 tj ( j I—-* / tj 3 :;.j j.i't dj 1 J j J1 t-ilisijj 1 fjn

* \ ^ >-L«m.J j 6 Aj \

(Y ) . , • . .
^ 1
11 niit^ulfil ln«^S \ 1 ^j^.1d ^ ^ Iw
Awa^Q lj„lj ijj ^

* \Jfo<J\3^-^ CL* \ (jr^ ^^-C.^v9 6 J \i— \ \Jj& (3“^’ ^ (J^ ^

^ j ^ d b J tfJ vS* Jl^j L_#> ^iJ \J

Iw^sj ^mJ dj ^ trr^CiJ WiJ b ^ fc*J^

(A)
, •J dJ ^*jJ toArJ JJ

et •'-?'* V / r./5 /*J/ : ' • 1

* I.'Juli­ > : ’ •1 ■4*9 • Y
* |^l ^ t l *^ Cj 3^t * l~r* ^ ^
r\ ^ i^ J b dJJ \ jj ♦ \ ^jo <J J b j * ^ • • *—■> *y
 tiU J 111\ \

* J L .f u»
'i 1 • ,j
ifS
r

(l )
(3iJH3 'jk

Sr ) 1J,^.L jj 1 L^™»o ^ 6 2 j>

, <(v )
;uax-5 0^ 1—.<5 J Lfis.

a(1 )
f*j ^J«*5 i |V.
u^^^ J |)«LmiJ^1 jj-1 t1 1 I■■« l j|1i
nI.uj ^ 1

< .i
-^Liij»J\ ^ lj^I. *!J1 (^y if tjS^1 ***■ ^1 til—JlJU ^ ^ I—»ab L**wk*» ItJ1

♦ ^2 tUL-^iJ \ J_SV)J\ 3~f^ \^/ j h \-1j 3

I m 'U^ ^ l^J ^1 1 d' i%

JIp j j j1 ^ m^J ^ il I

^ I |||I1
^ |il^^ J 1^ ^■ifc^W^J1IH^1 11^ l-S^4

* j _U l (jjdj1 wLajJ1 1J_.<b t) t / ^ ^/ <J>^**~<

*-^“*
>'
1"■‘
^‘ ^a-sJU^ <^JL_**JIaJ1

J 1Jj!—.0 L^u{J3 U J j JJ j I*-A*?—.Q / ^/ d-=»4-SW (j'J LjLSv'iaJ i d—*M«J j j 1 j ^j

(r) (o)
{j7 I pLf).. .o i ....-i,,^J*A,^--1i 'il ^ Vjv~~o Is—Jllil' ^ L—o^j) £ j j j3 J

(n ) x ^x
^ b^JI ^ m ,11^ I Ll ($*A»i1 1 * f'j1b)«5W L*4p}jp 6 1 Lli1
 $ j L/ J2— / '/

^r ^1L* (_£ «iJ1 »*Xi I\ 1—-Q1^ I -A-glJ

.
,, ('f ) _
^ ^j-u™!i / ^/ ^__*»CW ^»~jj O *^ ^ <v--»Juj1 jr} Jj— oJ1 (JilliniVViiI* j ^WJ

t (i ) . (r ) ^
^3 *-£ ^ 1— o-&^tAjsA C-uajwJ 1J \ ^ ^j£i^ \ l™a ^ C j ^ i

{ ty i ^ j ta?[» 1j ^ ij iwmi ^ ^ \j |j^ ji n.^ t Ijv q Vaaw*^* IJ I

(*) ^ L^w A ^ j

,{1 )
j ^J d •■*t'1^ ^ ^ ^ j 1j j^>il ■Q_r^Jc^ ^ ^^r^\>Q 1 hO * j Uj \j

^ ^ ^j»*—A ^ ^y dwmflhJV ^ ^ ^ l ^ ^j«ln 1 J tat j ^ j ( ^^

(A ) ^ (Y )
c> > 3 Ji-Jl 3— ^uJ— Xs 4 /_>./ Jl / 1/ I?--"[in'*-'J*I-J 1km«41^ ^y/ ^ j

•<
() OU-tu-l—O KS^C^
• ^ j *"7*j {J^ '
„ f
,t ^—'*^~'VJi~i
, , (’" }^ t—O jJLJ
1— 4 •_5
-H\~»(<l J ^jO

J^U 1 i Sjj « tl... > iJ_!W

{n ) _^ (n ) U u j S."J1j j-auS* &J\i^»" CJi--“'- ^

Or)
^*J I^ J.JV jv'-t'-tJ.i* «d«~J'lu,4!aJ1 dUJb l—'L 1—,.<v'-v1 dJ id»>-1 jjis j-
•-’
r“ !j

• ! U_J j »d».f%J j l_-< l j iA^W t) j j isLlVi j 1j v-LjW j)

(U) j vj^
Lj <J_AJ

♦ •1 ♦ Y *1 »|
♦ l_-o,.^ \ \^1 * ki- \ ,£ »Is—a!Li . kj>» *y'
• j d \j I,iiAi I—»oVs » «»si>*i *i Vs iSuj • 1 *a
* 1*^1 3 <J— (^J—
~4>JJJ♦ ♦ ♦ ♦ I—* • j iJ j (J— j)kJ * ^ * Y
CaJj J) J 1*•1*^ ^ * I-' i> ^9 j) >>■ 1 • ^ « ^UutiA-i—5is *^> |juM>vii>* jj » ^utuu9 • -A
* A rlU n ^ ,p0 ( j .Q ^ I*
^.AJ 1 \ 6-i-AJJ &JLM.VU ^ # J (J wdV/ j • ^«

* ( J .M I h I a S I 1 J t k W l^ k i) J -* J t£ - (L w 1 . y \ Q j ^ V ftP £)uA-*4b-l L jjJ » id ^ j4 <3 t J L» J * ^ |

♦ ^.Ltfcjds • Jn. » ^ yJ • j) • -•>’ l-j ^ 1V

. / j/ :\. \i
 KO'(O)

^./bJ ^ Limft CLL)*XJ

u h*>'
t

(i)
j ^mJ ^ tj?^^ l^/vW-<.V i j dJ j \ j \j (U^UVM 1—^0^jj | V—* J

(r) ('f)
-Pi JU_ot

(• ) (t)
^j ^)l—,>XtJu V^*»<»5jk> Cf’iP _^“*c ^ i£ ^ *p„tu*J^ ^-jj-mu t) ij ^ ^ j

+
^1 J^ J { ^ j ^j ti1nw^i^tU 1 itQ1^ * Ijb^ 1 ^ ^i/Iiiiii^Bk^1 Ltt*

. (A ) (Y ) (1 )+
J IJ &|| iJ.^llAJ l^ovQI ^ ■+ ^ ^ ^^ ^j.nnwtl 1^ ^ t*U 1 1 ^ iJntii ^^

(A ) ()- ) (<1 )+ +
^| i^V1\*J *^L»l1 ^h^aJ1 (^j3iIIft ^ ^Jlilm'.'
M^^11 {^ ^ ti■mfc ^*^5 J j Igj^

(")
L^nat) L±«Jim4 ^ |Jm3 \liimlMW 1>J1 LfUi 4™iJ 1 iuAij^P • 1^

(^ ^ )
(n*)
^ Iwo dLJ jJ £-—iXlli 4^J jjN^ a—j ^L*j\ jf^-S-5*. •J I Ia9—.9.
"i

I d-.*J"W
• # ta^'» Wf • V1 * + fcj>« * j

» 1 ^ / f/ ,* -£ ♦ !—5 1* 1 d— j) • (JL—^51 * 1 * 'j'

i^—
' ***^ V.3 * - } **»♦ -|. **^ ♦ • k5* ^ ^ ♦ij
1 ^
. J„£uJL » \ ♦^ * ^j"PI
B^W■*/» ^ *
♦ W-1* t ^^1■ft1i* “ * *

* t™=/ f*^ a *JL»j * _ ] . . * • «|. »^

♦ „* ♦ i
4 jj-«,4rt(U Jf) 4 4 4 I 4^4

lu^4rlMH^ f j Lww4t*5W t 4^I 4 4=^^ * | |

4 jftMjaiJ () 4 wfc- 4 } Y*
 Wf'+J

\<jjb
(1 ):\j j)

cx

(Y )***Asw'i^ 3 l? ^ *■>L-sw^f 1 V-^-vl 3-J 6 Lrw^

^ <3— XvtJ1jj 6 -j t— t !j j) j V
* ~^"
^ V^W>'il\ 1 “*-U _9 (_£vJ ^J— jj ^_£jjSJ \ j\^w^!\ (Jr^*

(^ ) (^ ) . (y ) +
■— LJ1jl |[iiipJ 1 J j |gj 1 j ^j f)—.i
^-iV L«4JL*S IJ 1jl * 1

o )
^■■■'^^ j ^ j ^ ^ / f^ ^ J ta!5*j W ^ ^1J ^ Vr^^/1 J 'Jl3j 1 j 1*^J d. W *.m.j9■

(Y )
u?-^ / c/3 /.^ / ^ /V-5 / ^ / J' h/b f'-+t/ uH / d/3 /^/

ur~J ^ / S^i / .9 / £“i / / ^d / 5 / Ji / / i£ f3 / 3i / <j?H f f 3 / *i /

(A)
(jH j 3t j 3 / / (jH / H / 3 / ^/ <jr^ / ‘^ / i> / ^ / (jH / /3 / ‘Hi /

(1)
(jjoJ ^ j^ivJ1 L^j— 4 4 / <6 / y ^ y^ y^tSy ^J\ f y„ f 3 / /

♦ 1 1 jjj£> ft ^

♦ L—
^ ^ i*^1^ ^ * »|" * # • *J“

• /*■/
• /e/
• / 1?/
 V> ( ^ J

• m ji L

<1 ) , . ,

(V)
Ll vlJ^ i»LjvJLi j^ Lt (_£*jJI <-A.JVjJ^ djJLC' (LjILkO\ £ \v-LlJ-Sj '(I Ij9JJ\J

(r )
^p*OkW J }*" ^ 1 ^aj \ ^ 'j^ \ du a J c .i ^loJ\ 'IiJ 1

(6) (i)
(^>aJ5 1 j j \ j i <1 o>.mui (^)1 Ci_«ai& l™a\ ♦ dLu-wN I

/ c-/ r5 / c/ /V
(n)
•lill ft—5 Cf ^ l_—
a Y tjJ<->^ 1 (J?^ 1 ^rj9 J ilj'i! 1 p tAS-l 1*-J1 tlU v-t . 5 **l JJ

» j^....5 I '—uM.uJLi £»»jj'i!L» ijJ1 w*v^jtJL> L JiJ1 *4_a>aJJ <iJx* <xjuL«u La-*1

* <Vl—5i1 ^isi C_J>jS 5 \ wM ^ ^9 V,"^aJi j—.x& l~a1^

(j >^» f 3 t ^ L^u Lj j ilU vLi>^9 jj.uv?J 1 | ^ iJ J>JV ^ L \J 1 -5 IJ 1^

. ^ (Y )
• (^5 1 J1/U'i1 lA-UkL—.)^ *(—*LllfJ1 1 ^ylp jjj9Li'I jJ.nS(^ & j kU <Jj 2

j\— rut f j f / J /$ /i>/ J' /^/3 C f 6 / (J^ /v/

i d .JVkuJlJ\ <JV^W^I 1 (^lo ^J_—aJi^uLj j dJ \j d\3^J? (Vj— jd J-il Li (_J?ijJ1 *J—mJ L?

( ()t-I-^1-*.
v ^ ^ 1^ I^ ^ (}n—A*l!■—OJ J]^1—% ^^ ^ L ^

*
MA
d*UUktU
*
• w^i. I— -» * Y ♦
1 - *
U ^ — (JUUAul • — ^-' • ^

♦ ljis (3^^ ^ ^ *v
* l~-*—a-Lc1 L-a d L'.i-kv'^l 1 V—
x»vui j * ^2>» *—* » d-to-Jjj 1—a 1 ^^-kukAj 1 d Lut^y'^l 1 t—aj-«.w ^ ju-J ^ * ^ * L

• (jjkjJ 1 ^jj9 ♦ L-» 1 • ^ -J ^ * c>

*•—

t ^ 3" ~ a j L k « j * 2
w >» U -* ^ ^ • d ..^ 1£• jjj 3U 1 (_j--kO ^ » 1 « Y
 L?/o;

^ ^^ ^^ 1^ ^ ^ii* ^j*,i 0b53J Ij i^y^uJ\jj ^ t?-,i-«.n ^3\s> I^

^a*^! ^ ijj-I*^ J j *^**5^*1^ ^ ^|wnjapu \ j w*»j ^ ^ ^. .^ L L?U ^ j — .L^taiI ^

(1 )
* J V^w^ \

j t^1*y/ ^icAJ 1J 1? * 1 t<A/0 () vAaj L—a5 Ip j=5 1 J ju (J—^3

^ ^+ - I \ * 1 )J k^* j
*( ^ i b-*p / ^ / I f O® 0' 5 * / V /
-
L_aP fO*j

X
Ailw<ilh^ ^ >*^ |JI^
U* 5 ^ / S-* / CS^*i I—
-a-P ^v,ts> -Jo j j ,W (j'i>

U )>< « a''
t> • /v / cy;:Lu 1—,aP j ->■ j *A„sw d-~*Juv.-5 ;>b

(1 ) (»)
^,L> i ^ (3 j J..%j (jjLu L-ai y 'i j 1 ^J ^^ ) -L jw

s (i> ) (°>
^ 1 Ip ^ I C^ *J j ^^ I*11 l-»aP j 5J

<S *J
■* fLaj J-5J\ ^5“* “*»sy (jja ^ j \—*aP 1

x (Y } -f
KS j ‘J-^w <_>a * ij-taY-J \ tj:,"1 “)-aj, Li*—
aP ^*1 ^ Nl
/"^/ c3 j>

(1) * >'
<Ja (jl^ t

0 *)
U* 0^3 * /-k/ jjj.iLw L .aP ^-*j jN ^ JJxM 3 *J LX80”^ ^ 5 t3~*'^ ^ t,5

1—^4 (jj

(n )* i ftjhb
. ^J.J-C' 3

♦ • i

* j^T*i / -^/ o® * *V ♦ V—
> » 3"
*%^ * ^
O aClj.]:iLuj *X * * *X ♦1 (jja *-}-» » *4" *Y

* l—h» ^ja d*JLtj ."^ »—

r»» (j^a d^.-Lvfkk » £)

I (jo C^JiiLw «y * • * ^ ^ * 1~tj' 0>° *+ * * •4* *Y

* (JJ^i * ^ * ) * • *—* (jja Cj.J-'VLku .

* |Jpl dU Ijj < I (jo J JL j * j^

 ^w3^J3 3

Vi— xL! S— ^.!}J1 c^»l^uf^ ' «y

^4 I't^j\ j l—E-W^\ I
^pjw/i-»^A,9w?n]i jl-«*j ^ I ^nl |fij

__(';) _ _ .{i )
^ $}' ^ ^L«Ch»ti Vh^iwiaAj j j ^ ^ ^J wLiw j*p

(Y )
^v«JU ^gff |^3 ' |J 1 t) ^V ^ J ^ /y iJ i ^ . jS _ J # . j (jjJi I j I

(1) - (o ) ^ .U ) .* .
>b 3J 3 d U J l JS ^ ^ li^nJ

..(A).
1*^W (jji>|I»J( Y ) ^JimjQII Lwm^Jrw^- iXj—-JAi^ (JJ&iV1Wi»t9 *1^}£&\l^ 1jS ^5LlJ

d—— ^—*aJ 3 L—>Luw ^ 3 (JT^

^ ) . £j ^ ij^ ( t—' ^ ^ ^5 ^Lj 1^*j Is ^3— SLj 3 LJLtS

(3 ♦ ) . _ :
* d'~W^I ^ Lm.V\ ^ 41^ Lf.niafaU

(n )
diI J ^'i1 j'i1 J kLsw ^jo uL>'iI O ^JsiJ3 (jo (^jt^-fuJI J I

Or) t - _ ,t () \ )
^^SV4-!^ LZ*»1j£)Id—*^o (^J\_i.-aJ1 (JT^” j*»S5j J3_fu3 d—j-jJ)w (jJ Is * / ^/

(31)
1^1 ^—•Jt-'.uj J La»f 3 3^ w j 3 d U *4—5 2 4 j <_£ f J«5o^1 d— 9 *— 3*U 3 O^ ■

• •■>- (jj* C^jJsULvw * 3

* ( j o ^ wL^paJ 3 ^ In io O 111)***,a **i y js> j 3 wdAJ • l5~-m*h-WI * - (^S jj *d>' * t
♦ d U (A) * V*r*# * ^ * ( j l^j + —— *r
* l *^ 3 i< 3 (jo ti*« j * 3 * ft.
d lA i^ l j (Im^ v w ^ j # d* ibsi^ <AJ3^ 1 3 j gS ^ I# J ^i*** j 6 **i m j (jj'^ * ^“-'f * j l*""‘ j (Jr^ ^ * 3 * i

* / V /
♦ —J3 * -j - 3 *Y
* 3-fio 3s « 3 * 3* * C_J3__JL< * 3 * 1
* i ry^ ^ 3 '3? 3 * i™* 3 • 3Y * d*s * 3 * 3 3

* 1^>5 d- »«*1U 3 ♦ .s-* » 1^*9 jh * ♦ 3 * 31 * **2*. [ ^"^l*HLlX*tAj * 3 r

 J

♦j j fjliii l l ^^ l 1.P 1

^3>oJ 3 j bw ^ 3

0 )
* / *J: ' f •*4^w 3 3s j b j t-> j j,a / (jj<j 3 d s jl%\ 3 d -J IIS 3

* CLd>j&(Y) J£-JJ L

(t) (r)
♦ LtAiU fi wiT^1 3 ^ 3 d

* *4si ♦ 3 * T ♦ j *^» j J 3 * 93
♦ M dJLJ3 jj * 3 6*4 L j * £ ♦ J 3 ^ L ^ *Yj
 j*- 1 CJLaJbJ1

.(l)
^ 1 ('l CJ0 ) J ^ ^ C? *”* '«J~5w ^juva/b .«•> CoLou 'il1 <JwAJb CojP ffj 1J 1^

(r ) . _ (v )
rt.-..t.9 vip 'j**-*-* 1 £}\ i pLvaS1 6 -M >WaU ''J1 1J ^ JiJ1 .AsuJ 1J t aCu31 t" ■**■.-.

+ (i )
^ kAj 1 ^^ I |^ ^ AaJW ^9 1 |Cn>I*^Lj 1 laAj ^ I |*jAJ ^IMfc^l%i ^ (*j 1

(*1) (o) +
^Ls ^ ^^ 1 «J i-Lj^W xAj’-'il1 C.9 1 'jlJ) 1 (jjlJ

(A ) _ ^ ^ (Y )
l—o 1^ Co1«J / j jju.qSb j Aj j jji "'ii 1 j ^ j 3—-o»~o 1 ^Jii—JLL>

..^ )
^rv™.o J dj j LlS m.t.9 1 i-Ls>»l ^1 fl—oJ? -b j ^Cmj 1^1

(1 *)
d—o Ltuilj 1 ><y ^ d—aot-.U 1 J.. . »^l_tj 1 L rw 1 ^ . . . d - . o i* ^s—o-A'dl

CoJ l*J 1 1—01 j) ♦ lib j 1j) J I lL>u j) \_o-vj5 j -£iS> (J3**1 \a^o_o IXw >A,5j) • L»4>W)

• (^5-ttAAil—x) 1 ^9 <1—0 bp 1 CIaA^S

<* I CijJp
(n )Ajfe.1^S^tiAU1 O^ _J-v*P

♦ <J i.i^\muJ1 linlJ ^1 1 j t^JI ^ ^ 1^*1 ^ 'A9 ^

(lr) (it )
pa ( ^] 1 2—JlaJ^(J1 J— ■V'n.iJ^ J ^^ 1 £"'__j''J! .-* Juu a 'j1 ~mA)U£

0°) • 11_* j 1 3 1^ j —':*j I—*0 1^ j J? j 1—.a | J b u J1

00 'J
V -^ O ^ c*>° ^

* <1 '\Ati/j * Co * y . 1 CoJ^iLur 1

^ V)} 1 ^rp * ■~S5' • 4a9 4aJL?'>U1• » • * Co t ^ c5 ^ J - puJ 1 * 1 r
* 1 * 'X * ^ C 3 *J L J * •♦**(- &
•
• J s —W ^ 1 (^JALOajJ 1 (J? vAwtU ^Jj9 ^ 1 k vAjDO 1^0 * ^ ^ Y
*
^A? ^ll^J 1 JsS-1!5! «i—>- * b ib iA>»1 * Co > ^ ♦ 1. 0 1 J) H / » * • * J A
* J1 C5 ^ *A»JCJ jO-vm5oX « *—«• * ) * . J 5U ^ V
* f frl ^ ^ 1 * wT^v * ^ Y ♦ l^w^111 1 * Co 1 * 1 1
» ( ji j) • JN • ^jOjA< * ^ 1 *1 1 1 (WuuJ t 1 • ^Y
(S'*1 ^ / (T* / iS '^-E^ 1—«^ .-U*_J\ ^ 1— 4,"1 lj—a J1a9 j Joj (jisjS L-S

&— aJJ 1
^ )Vi_«
.
j} j vj3--j tJ 1 J t) I LUo j$ 1 <) * jjs* ^ j j jjI

(V)
■-0—■***£ (Jj^ 'J^ j V—Jj d\.^,,«CJj$ 1j} * jC_-» j ^1 j ».fo! jjI jJo^r---j L—4^

(O (r)
. i>us j j)I ! ,JS>.Joj d—Jljol ^ Ji^ ^ j^ / (_£“Lavjw «4_AjJi

(JJ2— l—«4 d—™$ I (jjJ1 j»3—— d— 1 ^jjGj J d-— J 0 )j _£».Joj 1—./)1

(A )+ + (Y) (1 )
♦jh L*-»eJJ LvO-^f)—A (Jf* lJ^ I—^LAuJ1 ^5 J i ( j^Jj (^Jbjj

(1 • ) , , h )
* 1^jJ12 *JlJ U 1t) U \ t■<■■>■uJ^ ^J»SJjJfcj J ^iu2»,^Ijs j Jo ^S'-j 1—41J)■

J V. L,o^rjj j ,Jo Joj J^' ^—*-WlJ\

/c 3

/ c * J '/ l_^J jJo L,_* Jo j cy \

VmJ ' / 1—w.^i.-T’

-J
i jJ.o Jo Lj j ej Ull hji»wJiJ1
/c d

'/ j . Jo j ^ \ ^ \
/c 5

/C 4 ' / ^-W iw^^rW jJo tO-. j I .Y111 ^

1 w ^. jlO* Jo ^Sla J {)*/jIaJ1

/c 5 '/

'/ L_4-JiJ j Lj ^ ^ »>■ j LtwJ\ 1

/z '
J 6

*
* l^NI>tU • *y * l,«.iiw^ * V *V
*
* 1— *£ *j JU j ^^ j^ *&*j * “?■ * r
*V (jP 5 ^ ^ ^ *J j # ^ •£>
.j ^ U M i s : v .j i i - ^ - ^ > : .y
♦ 1 (Jm»4.^tV|I U i^1*ftlll.fcQ Ja? * wO* t ^—■> (jn 3 ljJ • 4*** * + *A
*J i* ^* * j Joj * ^
. : # 11
^ ^ ^ 1,.j1 ^ii
.^yj5 J«d3 |j* J- jT> \ ^Ai4^v <JiA^J7

^ ^*J ^ JvAfcJ1 jjp^VU k#iJfc*\Jj) + i*A jj,ibJj \±X*2**\^ *^f1 i J Ley|

. O') ' ’ . , . ..
—*atu£ I^)'
1 iJ1 ^J—SJ1 ^ ^ uLfW p\—0r.l ^5 l^J1 .fj 1 *J »-Ow i
^.Mi■.
*1l.l^

: i_*j mi

4*i %
»*Li# Sj y 1*^'TV# l ■»^«fW y/ vi^ LwJ vi^ **J
Js^j j ^ i ^ l-*.«sj1
—

/ G dj U-u 1? j L*£o'L—ft^ssJ ! *Js>»iu o Jp

vll>j HJl r ^ iil

d»> ^—<->Js> \—- j LiJ 11)1 ^11 1

/G

is / Cv^ ■<fcailj1

»JSn j U L^fcl1
BMti-0 |[fabUdi i

j LJ1 |j»i ,.aj<’.lAi

+ /

->/ ^ L.J1 ^ -- 1

, Jaj { j — $ Vatj^ i

j £*.* j j dj v-Ssa* (^£ Xi / ^’f'"' ^”"^=sv^ j *-"J *> w~ J j j ijl ^ — s-ut^J1

/ (^£ iH *^d ^ (J* £ / ^—■A-sw j .>■ -^» ^ ^U)l p-awwijl

1w~»J j*vj£* J 1 ^__uuciL!^

j *£w 1j j U'-^ L —
a^svl j »Sn. -j^> ^s>-- J
vJ=?j
-> LJ1 ^ - * -^tj *1- ^1
 (382)

^J ( 1 ) im^ *^jAJII|..^J-1^J\ \*4dw9I l^v <JlAkf^5

3^ j j ^)\ j LiJ^Lr 3^ / \j 3™o-^w jj * 1

(v ) ^ (^ ) "*■
^ * j dj J d*"*vQ."W I ^hlj 1 U? •AS&j j) • l^ J j) ^i_L^3 ^\ 3 jj*9 jj-jfb jj

,(1 )
* 1 03“* 1 ^9 tJjs*tjfi tJI d/^it»Li." .aJI ijji1t)ojJ1j> CUj1 1 d*L^aJiaJ1

t d>I—.*o_Th»lJ1 jb ^ Cj\j 1 l«~C£w o' J 15 o' J ^ 3

^^'KlU ^l*iWWfc|AP ^ ^^)ll f

c* o$J Jjw Is^Jp <xi-vj j «o ^j Jl

1 0^ L~*J dJ1 *._-»1 \ if y'Js w^*j <^3 1 vJLfW^ ^ &.

(o )
\jL~4^lui> i jv„t>j (^Jb J-'XJ3 £*..3' l5J'
-s^ ^4 (J^Jlj >aJ1

H )
*A1/w^i1 ^jj»w 9 o' jj&> ^a / C" L/ / ^- 1/

Y) ^J»^
)iiiiQ li^W^ L(J1 ^ f
tw 1 *-L.OjJt ^JL11 3™*3 >un
^3 ^ 3"»^-a,

t VL^i1^ ^j{)IjJ1 1 u? O^-*41' 1—jS (} } (J2j j j (J?^~^iJ

(1* > (1) (A )

^d.»■* J1| fl—^ i iJl^fyj''J'
I U*** ^iiiihi^S^J1 ^ i \ <J jiaJ\

• 1 1kA^)

Ix^L.^ d lnn»A^ i lj£3 ^j<2JW 1«JI ^l^'iS'i1 dJ.ib^

* (3^ d i** *-}-♦ +*-J- * V *1

•
*r • (^1 d ♦ dw. ^
* L^ ^ dJ^Uv *o * »Altk * v^- 1 * %
* * te^1
* ^ \ •Y +^sniOJ^W *
*t* •n
♦ 1 lj? i i u 5 : ^ . ..d<j*5»* l»|<-WLMub ^rLV^ '3 * ' *A
*o ^ cJ^iW * )• *(J^vif * L-# *1
 (383)

tfe j i j Vr^JjjI 3'jS1J i.ft 3 ^ Jilil (_5*A »A«swJ t|kJ,

P ^ ■>."«* a

jj3^93 / '/ 'rt*'*"* *kjU*l 1<J' 4 4uUi <3^I «4 JS ^)i ^.Lhjl**'^ j

(r) (t )
^Va*4Lw' d *jj3 (j^3 ^ 3 olwi^uJl v.”.
<3-<* L>4 j j ji»>j tjJji

y? V*aIfc A !j 6 j**»IaJ1 3j51*aJ1 4 *J.U . <ii)j ^

(ft)♦ ylyt illi uu.^jCLw ! cJJ*aJU ^ L^."„.jl»Ip

b(J5» CU^d^ •d* • lu^iSl3 I j3 * I *y

t*j\ : ^ -i .»_*_: * ,.T

• dJL)!^ * 1 4«4j «ft

*
4Lw '^ jb it L'j ^4 j*\v ♦ Uli3mi«4 i'Op i^JLu3 * dt**—'Tt 4 'i1 fei. I
pjjj d#L>oJ\ J3 ^> 3 4 -U _)* j1ji •^bii ^ j\^ti\
^ 1) J 3MI j i4„o.ii Jl^jt O^ O' * IfeliHwl3 ^i d llk ib ^ Ji
t *4L w 'i 1 lis ii (3 * a J 1 | C >1 i« n * i j I t..w J j5 i) .j U j ^ •* J Iff 1." V~ j l )

i 4.+m\ Jj i ^ m L J I ^ i ^ySLu ^ ^ A - kd c J U 1 j V f w Y ' ^ ft.*L~ W

M « j ! O ^ o ' 3 *******
»,^i 1 ^<u4 ^jil Jj (Jrf
^ ji IftiJ &» Jjxj )b ^j L>w^l' 5— (JLs— )
» o l * i i.i.» ) \ l i ^1 ^ J i ^ lk *

^)i o* jtAtflj 4c ^ mi^4 <vJ^ 4Lm»«*')I1 o' O^ ^ 3

1 tct 3jvwUJl d^JliAJi La <u ^Ji-.ii (3 *AJi * ^**iJ'
^ L n iI | » M & iI 1 O i » U I ^y u > o ' 3 *,* 3 * ' A- j f i l L jw J ? >

l»j_& 0" 3 4 . l*.f 3 .v i^ ..* li Vn<>i.B>(a j \j jfcti,\ i 4 L.L< ; ^

3 *»d>»v»- vis m ~»* .> j> <1> * 4JLsu13 (^ dftiiJj5 >> 3J4 j j-UI *JaUoI i
I0 .» ,l> g i 1 .4 >>. ^ J _ * w o* t^'^i
I---------1
w>- Jp V_v ^ .Jp

# InAoJJl 2 ■*)ilil Jlsw^ll Oi** ^>—ft->«J1 ^ Aa5 iA>32 J j*lliuJ1 iJJSi (^Ifr3
viU-U 3 Oi O*^ 4—^ ^iai ^ (JUaiJ'5ll O^.i-w "ll 3 *Adw ^

J J Ipw J • |» > ^». L«» J * j ^ w IaJI jtiMtftJi 4 i ift (3 * ** 1 ^ -fa il-W C?"

<uJ\J jS » £& aJ' _ « <q..JUk> _^jx- ^ » d_Sl ft}* o'< ^

O' 4— * o* 3 - / V /3 / '/ O " ^ < / ^/3 / £/ OM

4a4 Q A.ftJb.1\ oX—Vf ^3 j3 ' ^Iu4^6 JlJLui-rf *4 itli ^w&Jl iJLft
3 "VI
 v.y w r J

P LuJ1 1

* j) J 1jjtJ ^ 1

4 3 ' 4—k3«JJ1 C>i(il_»{y^-> ^J1 d—*aJIuJ1 4—JLiLJ3 d-lpL^tA^' Liilii' \iJi 3

^ d ^^ mh^tij|{))
[j 1i ^l ^ 1«AV ^j^A^35* ^ ^|J ^ ^1 1 yW

^*3 ^^ ^ ^nc*>/^vt!N |^»nV^ | ^ ^ ^ j)^9 1U *0 ^ tf ^■Mfti^)

(i ) <. (V) .(V )

• ^j5J LuJ1 k_.Luur 31 v^t J l.'lw1 I—V-'i- »1Uki ^ ^ llo ^jj''" *^1

(y )_ (i) . . . (®>.
^iiM^j1 i—.Q'3 * i— k)k*Ap L^kfAkUU ^'Jlld^«kiLii^ LuJi ^i3

♦ ii^kJ lana^^W k^ kA^W ^ i|nlt^1jj I *A 3^

(A ) ^ m „
O® 4—«v<Mi> tj|ji.ao 3 ^11 3—mJaJ 3 ^LvJ51 o^> ^*i.i 1 N _tj'

(\\
' tJ3 ^ ' 4—5Lj4>3 ^jjJ3 'is' 4 j'ujJ
5 ♦^j3'3 ^ J d*rfi? {Jf^13 • 4 kJ-*Lj1 4.
...‘
LikJ'

0 0 00
3—^\,(pl LJ 1 3 j S \k»*J\ 4^sLJ\ jjjii jjjjLJ 1 3-~-5Lp 3 IL) 1 3j S 3***J 3 ^tJ 3 '

3 j S 3jJ\ it 33 ^jjJ1 1j i 1 3— s V.{3>3 4— 4^ 3 3«01 6 <juJLJ 3 ^jJ3 cuJLJ\

j^J1 (_yj k
,AU J ' 4—*9W?' 4-ujJ LuJ 3 4 3vjJ1 £ j^y-mL?J ' ^3 3 {3^0LsJ ' 3—*5L*o3 4-~*-«iQLuJ 3

kaLpk.^1 W3 ' ^J»sv^J <j Vi"-^a Cw LuJ\ 3 LuJ J LIAksJ Ok *(3«*ALuJ1

•4 j § 3.i..i.n^ LibkAjsv L«o LI ^ * >-*•” *4 j $ Lu-<t4 ♦ 3 ♦ y ♦ d—uajj 3 * ■P’- o i » ^

• La j L-uJ3 dj kA-J3 m LkUi'iI • I •^ ^■
<.*~^ I *Y
* 4—«S««LuJ3 4j j^SuA^J3 m*1.M.U^ * La J$j 1»aaJ1 (j^ 1 Vui^ ^ I • L»iA
o * JS ♦^ \ + 1 * £>

# ' i* 1 *^ 1 l ^
♦ ^ J^ l ^f|il jgj^^ l * *JSP^k-* l * ]^ * <3.iW!
^d,V ^ f W5*Lwii^ l
uji 31 4— LuJ3 .• • 3-— Ll!3 jjjJ1 J—pJl^J3 . u^» * ^ 1
(r)+ + (v ) i)
•3jS\juoJ1 J1j} M 1 ^9 LjJ—-ft J .j jJ ^ ^5jk> jl_-»j vJ_EW (Jj'l

(a) ( i )
o jjV?.» i
j h<>j 1 ^t^J^ 1 (^o 1a\ Lmq ij ci*-a«JI j51^jJ\ dh"Lfb^

<J■ 1^ ^|i1imiwi%11 ^ji^WAU ^ 1\ J i^l ^ ^ 1 ^ \ + L^iiMAU ^ 'Ap

L_,fl\jj •ViSSjJ tSJ1 <1 J»AJ1 dUwL^i5j) if uLiJUtJtj) ^J-.i—.qJ^ l—tv.wj ^jojj

%oJ^ ^ ^ JlJr^Lj I^ 1 ^UM^hWJ L(|t^ Ij\^ ^iHin41\*^iI 1 1

(Y )
1 C*^hUU ^ L»^|S^ I J1^ I 1 ^ 1wLl1 1IiQI^ * ^wrni^W ^ 1 ^^

(A )
^^ I ^1 L J ^ 1^ ^ ^1
J ^^AiW |^ ^ »b
l■frnf1 1^

(1 ) _
C_»<_-wv/ ^ LjwA.9 Q lp (jii^MtQlaJ 1 d 1uJI 1—41j) ♦ LS^SU t) J»i—*« d cp

(A)
3 I—*\j

CJLsvi'iv]1 jj <1#)♦ <J—-*-'<«-$UJJ V^a-vuJI tJiA& ^9 <t™*»»jjLwq IjAj^ (3'(‘

*"> <J™*avtJLuJ1

(n) .
* 54 J-J ' (J^

0 T)
* VjLw 1 L|^=jwuj j*j nAi* ^1L)^ d ^ 1>
1 »l jjxl ^

+ . . „. A ( ) r )
^ ^^ Lt,j1 d Crt#tJI^vjl ^jII
mW I«^VWJ liwilP

j k^wdw ^*jij I♦ >J J J tA^W [f^Q | |Aj 1 * 1 * ^ * '-J^1’ L— <,v * |

* d .^»t.lAtlJ I • I * ^ • ijt' * p|- + ♦ ♦ -j- *

♦^ilJ S LiiJj1 d LlJ1 dJJ1wlJ1^ i *Y * ^ ^J ^ *0

*^ 1 ^ | * d1k^^Q L^i1 d 1»*^tjl ^ljV^naal^1^ * l d"*iA^i(&» | ^
I^J\ 2j^JlM L~*I3 1 \*^ * o 1^ 1 * ^ V *A
* ^3kAj S 1 J ^^ I L^kftiiiWa/ ^*J l^ * ♦ * * taft* < d 1 ^mt | j j J) ^ 1 (^J^1 ^3*tw»Li5
UJJ Li^MAU iJ f1 ^ JJ * IfcU (i“'*jJ^ • 1 * ^*
*i*^» * |^ * *jj

.Iwjlj) *t1* )f
 V J

(1) . _ ())

(v ) + .
L ^ jeL ^ O 1 j) jS \ jj *AJ 1 ^ J <5 j)
(v )
* ^^
. 1 <3 ■*J D d* 1

J/sll Jl J^H

"(I )

(o )
^>rlU^Jl <JjS 1«-Ul dJL& 7_?l.:^"^'t 3_.i^-w1<j*j ^gdt3- *732 * l»JLswdd1 3-^^7 2

(Y) (1 )
^ ^ lijiJ i |L-> JY (jdJ iy«dJ1J ^ j j ^11 l^j (j® ^ *> 1

* / -4/ U>^ I JI

* ^ Mi)I! u I ,\ , 4_^i ^ t-, I . ^

* flu ^ Vw* I♦ ^ # w?» 54
L # A ,L l ii V t •{• ♦ * * * < ^«i"-v!J 1 ^ ^

* &vnm3 Vj£?^^I <3.... ^^ssf fcAj*^ * fca^1 ^ * *&

* ♦ Y * / ^ ^ / * ^ ^
 (387)

UU1

(1 )
CJLU1^ (JjL_vaJI

(V).

(r)
* / h / u^ / J / ^ liJ 1 |*w ij 1 & j 5 IjlH d Jlib 3—r9Vi'p'j 1

H
/ L/ ^j&j) t*i-JL*J1 M( 0 , 1Cf

(1 )x X {to ) +
* / ^/ J' /£/ */&/

• t-lfci^£LAV *^ *f) mIiP • tal*' 1 *j

•v^>. <J2-i..L^.vj * £ * <y*3 * **- .y
3 i;J *X * ♦ ♦X *1 4ij\jj •-{-• •*-}- *£
 \ ^UU J

t jU\ J\ ***■.J LJU1 I 4"~~£

^ ii^inii»atj^ ^na^irUA^ iwQ ^ ^l ^m^jLnn^FiJ^ ^^jiji ^ ^takt^»4U^ j^ ^

(\ ) H- 4*
*-t9 * CjLe,—vtsJ\ ^—,4 L-^qJ #^iLsa™^^1 ILfw-*v V-j-& -a-^2 ( S—3 yl

(T) .
J jj —-*^..jjj1 t)iJ \.i£?'(I^ ijiiAw 1 t^j ^ ^1 —*4 {) J“0 ^ <)<3.J5i ^9 ^Jkt^a=5»-

♦ / t
^*i/ LS^l /«/ t> 3^3 jjrJ UJ15 6 / Js*/, ^ 1 /v/ ''^1 ^

• 1 f)_i^-VJ \js»jj O' J^-i-uO £ \jJ » ^

. dU j I i ♦x
 j

^ j- . I I 1 J^ ^ | j ^ ^ y& j V l-f I 1rini I JJ i • (1)! ■

5 i *3J 1 d taUi ^9 JfIAJs- jj ^ UJ 1I1J ij luinj) ^ 1 ■tJ1 C^f^uJ 1 JJ

J jt L[5 'i I
(T)cJwlA
.
(<J^UI i ,^w*
r
(r)
^J \ / L j (j* (j'*-^ ^ J j ^ ^ j 3/ ua 3"^ 3 O*'

/c?/-

* dii>*>Vii * 0=*“ >i* * ^

{jjQ tj_f& a {^4™<1 u j i ^ ^ 4:5 : -^ -r

 j

& Ls>J1 ^li-J1 1__5_il*p^

CAiiuJ1j) 1 mj jj A ricj^1 <*i*i»jJ1jj 0L_^J1 j)

(V )
l_yj UJ1 1 d—3 ^ 1jjj 1'Aj1 6 j) *0 ^

* /Vi/ /*/ t> 3-^.9 L / IS / J' /-v/ c> 3-^3

♦ * ,„•° jj * 1 ^
^ * llk i. I j j j ) jj ^ * 5»_JtJ H U d—Uhj) L*U—lA J^r^5 ♦ 1 (5^ L j ♦Y
 <s

l^ Ji 'aJ1 ('^ ^ j ^ ^lLJIjj | Li ^

r'A&J1 <3~-5lio'i^ J— Ll*w!(y J^ 1tJ^ di-l-fij ^9 i-l5jj • <J. ■ l^J1) ^ ^ Lu1

/~>i/ <_?^ /^/ O'1 3*^3 A /^/ L?^ /“^/ 0° 3-^3 £^3-^

(n
t> l$^ jjI^v (_^<i a....jv^j ^.jt& \ L^;—.o '^S ljA+s)l ^ 13

(1 ) (r )
>u

(1 ) _ ...... <° )
<jM ^^»m.-jJ «L>^3 ^ jjjXJI fjf£sjj £ ^ LcJ\ A—.j-2^>* I.^(£vu^^ IA—3.9 J^ ^

(Y )
^ I fy^
J I 3 ^ ^
» 9Lm J1 J tf J I*

* ^ « 4»h^aJ ^ • w^> 1 * j

£——0 {_£1 * l™-' • d—£•jj3 1 LJ1La—*4-!I *1 .^v

C<5^ nAj?j) * >*<1 ■-i9 )_»qS t I •^ «d■-..sw*j1 * 1 •£
* Ilwm^ * *y *wU ♦ I*'i
 ^ ^ (t S .■
J ^ ^ wwJ ^ \ 3 *^135 CXt;\ ^ ^Jj 6 $—-i^ttt*i ..^i\

()Qj ^ ^ ^ ^ bJ1(1K - ^ L^lI ^ulsJI ^■■■*.^\-jd31 4— 1^11

(T )
^ l-SW1 d-i»'?W J 1 <V^3Uo'i1 <)bUi d>»bi%VJ 1 L^<X4 ^jufjfXS^jL? • | 1 AlJ^J1

'd j j$ 1 —<JLu 1 j »> j j Lru 1 j . f\ j ^ 1

111 1 t"-i...,<>1ft *jj) jj & j js j j„,^

. (V* )
• ^'j!I *J jJp O^ ^--^-vJJb^sJlJ

(1 )
<j 1iJ \j>$ 1 »4 3 3 ^
^j—>4LJ 'il*-. (}£ (j!?J LuU9^| bA5J

{0} .»j_jv 1 tvAJb (jr^ ^ ^9Via—*qJ1 ^jj C)

db»jj9 • ji>>" O • I ^4 C<-i%Uv • y ♦ wit" jjo £>31/ j * J

* ^J—»0 LlJ.1 ♦ wit. 1 » ^ • J*ib I • O |j3 • Y*

• <)jji&^ ^5Li„i uqJ1 jtylfi U13j^ • O «J Vjw-w<>J1 1"-L/& ^ ^j5jj) • 1 15

 (393)

<J,9^ ^1 *^i *^i L; C *"* Ja>

I
^
&1L jij
(j*
fa
“
V—*jL
• •*
«
.. j
C J
«J Ja«

euj L? fa l—-
lj vis j J Ja«

t'.'-’ (5 *i ->; >i C J J Ja'

Va>. ^j> 4j <— tj -- j J Ja ’

( j. v - * U ^ v js w (jS £ j J
Ja'

Lw (b »b i— u ^ j vJ
Ja1

(_j—vQ I? £ia jj 4j L j J Ja'

^ 3d *i ^ ^ XL J J Ja'

Ip wtj L ^ J >J Ja'

jvUi«£ iJ JVJ V—*j j J Ja'

j-Sv_c us3 ^ g 5i ’ -** k £ j J Ja'

 (V It u°)
01 j i 31 4>
j Us w?w l~-tjj b p
_ j g 1
__'

4j d kJ <j '—*
>' <?

L>J ^ 0 4j d J

ej u d_t d fcd L?— '

c-

ei ’j e- 4/ h^ti L> d j
C

d »>
LS-^
5 1— 5

J Uv dw d J LSr—'5 1
—5

04 Lw £iJ dj -ij (jf d J 1— 5

c

O-* ^ 01 3d 4/
c d J LS—;' 1— 5

\j 0 / j-> 4» {^5 ^» d J
l?— 5’ LJ

i; l5 t & u

J-'*—*- ij cr­ ib: *T" C? c

d -» L5— 5'1— 5

0 2d d u L5— '1— 5
 (395)

(1 * .up )

jJ' <y» <Xj *-L# **«■> Lu .L Jo r ; 3 0 J

**

c- &J >-"Lf L r
C
a ..1Ij

df ^ Ju
c- c

i «JV? di Jb c U-J e^Jtt

s

C- <b *tw if CLv^J Vj

c

»0 d> iS c. o.
Cr

L>v•J itu jaf

C“
dj W^f (Jid-JL5

Lav . df vLf Lw £ 0 tflj .Iii

0—0ii <Lf iiitj «.!if

<L> v^vf cL^JU

c

iU L? C 1A1 1IJ

«-S- tj ^nf
c

*\k
c
(396;
(397)
Cu It?
 (V 1Y o3)

+ ^^ |iLtoS^1I l^J ^131^ Li I^ijJ1 <3%>L^5

+
^ tjSJ*J \ <3— -t-*.iJt ^51 c) ~U 1 (^1 y A— -*u^j l 1 2 j3 \jJ 1 (^ja d j> 1 » (Jj j)JU9

(1 ) (IT ) +
♦ <1...,<V i j ''if J <3*J_& j)_ & t J fj f A ^ iJ 1 ^ J jiJ

^ * jj # t * 1

(V ). _ (1 ) . U ) (° )
*kS ^ <i—-rwj'*!\ axljS) l^J—»a ^jyll/JJ 12

L • ♦ ^ ^ • j

(Y ).* (,J> ,^ d....^ ^ ^ *'^-&

* (O^-to \-^.i I nO t-iiJ U I ^SNhfcj 1^

«£W ♦ L • • ^

(Y). ^ (1) (A)

♦ (_£ \ 3 ...n-> 1 <JiJwfb ^J3i L^il....a \ 1 ^r^jJ 1 j

4j * • L • • j-

(1* ). . (^ )
* c£ ^ 4— *w \ <3 <-),A Lj^ ^ t L . 0 1y

ml
C" t • <Xj * »5W • L

3^ Cjr* ^ J ^
( ) 'j £:*.k} \i» J I
(n ) (jr^ <3>-L& (_j 1 jtJjiii

* * v5*» •Y •^5=' (Jja Cld^Lw « ^

(_t>^-0 **^~ _j) • 1 •£ . |J (jjQ *3“*V.-3 *V*
»
u>-
,ej LiJ 1 jym.il 1
j^uJiJ 1 <>_C& Jq LJ1
U!\ jA aJ i
t^taJ I jj
^ •* I
1 * _2s»
* ^ f A J ll-» J*,
3 t. O
.\fi
s (ja
♦ **■ j^5^n>'iI •Y * (jo L j »Y
• • iji- ^ *Y * ^ (jo k j •A
' ^ (-5 1 <)...—?u jSj!1 jj # *j>* (_jQ (jj 1 ^ c^t,lriiJ-Av * 1*
♦ \ * )Y * V-^wLj.jw <>J~fti *1 * ) 1
 (r )
(ja \j-\1 d— 1 (ja £j1jS 1j 6 (_j'jJ luJ1 ^vjJkJ1 <1— »««a^

(* ) ^ (i)
♦^
Lv.ts'il (jj<s O'"1'
3\riiJ 1 jjJb 1 (ja (j.y.a lid1(j /^VvJj^1

(A) ^ (Y ) (1 )
^j
i 1 ,JJI—jjj ^j1 yS Vj^ dd i I O^1 ^j ^L^up^l1 ciuLdS 1*)1y

(Y ) + + (^ ) "*" **"
( A-~*~> dd ^jj.lp d^ 1 d—wiuqjf* (3—"J d-— rtj1^J1 A j3 1>aJ1

(1• )
Sw&,-^*v/ (3 1 L>'Atd3 ^D'^ l-*d^ 1A LJ° ) L?’ ^3 ^-d^ d 3 ^ ^

|)J Lli3 I Oj i.d IiJI I • Jl*£»'l<!1 A Mpjd^ JJI& Lm4>^»J»s»»1 (j1 'i1

()) )
(j1 1tAJb (ja ^ ^ U» ^jP L^*j1 a & ^Leifs'/!1 6wLTb ij^

» * w^> »Y * J—-jJ-ii?l_Jl * ^a. t™* 1 • ^

« ^jo p Lws'j1 ^jo dJsiA u • ^ * L^*w ju • «?» • y
* ^yfi) * O (ji A>*■3lij •Y * di■uju « O (js iJ»3b j ♦O
. .Usll <jj LiP 'i1 iji^>^ 1j I^ ♦y

(jo () 1 dMv!>• (jo ( J^Lj t ^- t t t - ] * * w8

' (jO 3-dliUv • ^

4—uOLaJI^ ^il^Jl^ lJ 2y}\ O*^ * ^ * 1* * <!U—eQ O^i'txw d^J i—* (Jjj—-a

» a bitio U J 13 <j~fc/»jLJ1jj j»-,tu 1 15 d ^ (j*i ♦ w» » (j^ioi d ^
A j\j& 1 Df i 1»
Jl (_
£ 1 A^m^J-lo {jP 1
— %'iJ
biJvJixl ( jl ^1 *1

* (_£j^'1 A~—
&fb {jp 0*J (jj1jj ^Lv.1^1 AtXJii (jjft
* L? 3—Jiji^ ^9 L^J1 ^-—.a • 1 A— L>* (j^Pj)
1»J— 5 ^ ^ir»l 3—sub (jjP 3— j) 3— wtaL'jJ! (jl 'il1 .1 A— .£-«»\-s*- (jp 3 I S'*
(j 1 (^ J"»1 ^ (j (j 1 d^ ^MV 1 (j ^ 1 A................^d ^
• ^ ij5w L^J (^|

(jll 1 'i1 (^ 1 a 1■i.oJb '3>v^'1 (j1 N 1 # U-/ V?- ( jj

(j 1 ''i 1 * ( Q ^3**1 A ■■ L(^j 1 ^

o jft 1 A »«A
jo c>° 3 J IaC' _^6»jJ1

IauP*(!1 AJJb (jjjs ^ J W' j»5w'l!1 (j)^ 1fcJ1 ^11 1 i— Api-to (jj9 |tLwSi
r
♦ ti5 1 3—iL-UJ1 j,<S' (jLj (jl^
 (402;

■( r > V, O')
kd d_FW ^ J (_J I - ^ j -ZjSs A- " P

(JZ*j — Lj 3j L& O^" ■■djSi l

(1) (^) \l“

3Jj) o LJ^i^ J 1 » e-.j^'

(A) ^ (Y ) ^
rJJl ^pJ L» j-jw jp p° o* A j $ L^va

(1 )
• IA t JIjj ^ ^ d

d^ W ^ s%j^ I ^ 'i l 'i 1 jjjS L i.ft jj 1 'if 1 * »>

* (J£ J , ^1 <)—juJ^U ^ JXS' Jjy9 O ^j) ^Luif}''^ 1 d fcAJb (3^

( jt d ^ Li< ! ^ \j \ {j; jS . 'i \ 3__iuis jjS- ^ L—oJb \ N1 ♦ J
» ^ d—
-JLfi^xJ1 ^P d-w ^ 1 ^ LviS''if I d'•J—b

c jJ*"ijjt^JLU\ L,Jo aJ1 <_9 J wLijj

. S— , $L»( L—vjjJ j) li& ( j j *dJ 1 ^ - ’-^ d^ tld
4f„ \jLvi>J ^ $ 4* £r'-'d —=O.Li (Jj 1 _^JbVihJ 1^

J ^Jds!—JfeijT 1 £»,J '(f i (JJP “A_IW j)J i ^ j)-J dJ 1 • (j£

jJj 1—3 d-<1 ^ —<.i)J1 jj • 1 |JLuJ 1 CL>U . J ^ ! — 0 1—>9j faj 1 ^

d1n^ff.. 1 |j^ (4jj^Aaj1 f|Ip \ *0-^*0 j) |fj■ i■q ^ P^J

^ X^ij l^j ^ ^ 1^ kd JJtdirluj 1 I 1 ( n»4»4V 1^ ^ ^ *d J ^

j \j llji> jb \j \ * ^J^ - - *^ 1 "'i ^ J\J \

c
♦ j i —* y ^J^uIaujU \ ^jj~j\>j(JJ wCLuj \ j 6-^4 J jj ^ \ \ Jp

* Js> ^ J U V 1 1 *^ ^ L-/ *|
* O ^i ^ *% * * ^ *Y
4 ♦
♦ < jij \ 2J ^ * "1 * (J*9 * *T* * ^
* fcjit* L*-1 czJ^L« . y

» dJ- . I (.yJ^Xui ^ • Lfb L ^1 (| j') ! 1 d >A^J • O ‘I

 UJ 1 I

* O—i£ '

())
»4j J V*i 'jl 1 i- J > ^jS 6—S>l-tynJ I d*-Lib J —Jh I—,'' 1^ d—w»I

(^): ..
if 3j —$ 1jj ( ^-' ^ ^ q s 'j
.
a— ^
< .
y jj—a j j)
. ., j
ti— d—J
. (Y), .
uo (j*&*>* t
. . i
<j—>*u» w

I ^™SUK?j) l-U <AJi? * >j„>"\ j) J) yy? yj....^.".-,Qjwj y|—JU I OJ ^J jj

(^ )
* ^Lh-^Ai*J liU tJ yjft 1j) 3—
<Vj J 'l j) £—3 Ci* 1jj lj r) y jj ^Jj 2 ^ I j} ^ O'"^ I

ii iA ^ 1 jjjiXi'X—
»o f *j ^bL^VTt^i I? I 1 L«4 1^

(Y )_
y|_iSJ VLu—
JUJ i j yj^~«&j l kkfcr^JJrnJ J ■iwAf^ tfi l I j j d i.Q*^Wl-3 J.ihJ ^ LuA

(1• )_ _(-l ) . ^ (A )
d—a-rviJ ii—mjj) Lvja yJiLut'i I yjJ.Lt>—a d—a,^w CuJLtjh^=. 1>J1 0—kj')} f> l-'JtJL lit—.a <y

(n )
^ ( j I™* j y^l&^l I yjQ yjJ U I C I _5 i j d** j (_£■* ^ S jJ^*9 t f

. „ . (n )
♦ y~—9I j)toJ 1 lit] brL^S j) * lS~~a yyP yy.^J jj t ^ (^4—^1 j «JOj y^ l j U~^ j

t> (u ) {)r ) '

yj^csw \J Ip ♦ ^iJL—a J 3 3 ^ J 3^ ^ jJ ^ ~ w I \J J ^ I I *3Is

C dj—»I j rj t <J^\j \^v ^ t d—

"vj 1; _T, i jA -tv^ I ^?iJ,i-"i_.a ^ I ! \ j U v5
d— t d5 I y~^ ^ lJ 9 ^ y jis 1
'^ l yf)J.l^ „.a

d I?" J^S J) l~> * )

* Vi5> ♦ -r * V
d J — PI j) J4 ^ J j) i)— . ^ ! 1^ dJ y^O——a J ^ w l. d—-^ U k yjPj) ‘ L_> * f

♦ yj—*J f^J I * * C) J 1—oJ ^ : i * 1

♦ L-< \_sJst!£>l . i *1
4 1 L^vu-f * *
JJfiGUj} * i
^I *^ • Vm II •* ♦
*Y
* I
4 ^juA-^’* * *p^* L-> l t ^ ►
a * 1 6 " *^-w »-5** *A
* y^a d*l^iVw • ) *
^ y u ii ; I . I I
. oJjl i \ • \y ^ 3 : 1 • n

• u
* (J 3 ^ *'
 4 j UJI

(I )
lJ >0.1I^ dj — y) ^"'■>
■*JjJI yfr™..AJ1-tvsJM.,P

» d-1—J-
 yj— (JHi \ (J-yaaJ\

♦ I ^ 1^ j VI ^■l^ikO-tvJj) <JJ Vl^ I dJ u3 j) 1-t ItJI -pj O5

|^»L^*3v»j) ^1 ^**i^ d-J \ \ JJ ^L^ JtrldI \

(t ) .
j ft. . i P - i.k o J I y j — wj=» L t V ta ^ j I O ' J I™—^ £ 3 5 * d— S ^ 5 \ >0 J I d " .» * ^LaJ If j} L v j— * a j)

♦ L^~>kA^ss.jj tV—a.-ik-i J-S

I I.O-XV 1 ^ k^J ^ ^ V .J. . —a V^kJ I . ■■.1l p + ^ v tv L f y j L t^ V *3 ^J , ^ 1 J j) ^

Lv/**l!I j jj'i \

v.1— a y jA J -^ JJ J wJl*»-
f~!
\
C <U J£

yj U-WtJ
I—>
H jS J
L - lI^ J l y)\jLui J
.J*- c5 Ji
a5
1 y) Lm* iJ
J L oS vJ
c-
(r)
yjv ^Sjl yJ^-Ui ^ yjtUvfJ
<J l—U -ijj 3^ k^J

J j 3 0^J
3 kSW
T 4*
eJ 3 jj
J^lft txl I y j \jLtv J
3 k^4 6 dJ
i?

Aj J£ >1

• (JJ I • w5» • d^it— ■1

tP ■k^-lI * ^— k** Y . &\ ^ I ♦ )
L
•l>v^ cy^w 3 ** *?■ ijr • \ .Y'
 * H"~*| I I C Luu d “J-S1*^ A II d1n»*L-UI ♦ A—JJkjs. ^ Ik j 11 y ij ^ 1 -n I | j

—.a jj ff ^ ^ j'l I U-A-^a * ^—>»>JI j ;Jx* j 2 t) if y j3 —-aJ 1 jis J u

u—Li—aJ I jj if *t-'U.M
.J .i,,>aJ I <i Lt-tv jt) A ^ 1 3 >-jIjlav * *A* J* , .a ' 1

jj&.li * ^M
MwhgJItLAU jj i \ A^w«*Vixu< I II ■iptol 1 ^ 1^ ,^ ^ ^ |^x«] \ LA1-?^

.(V )
^J—x j I y£ «J u)—
TUI d—ktkiu kJ J 1 J.l.O*Kf yjJ I p —) I I yjvLtl—Q d—jLtv™lP » tJ I ji J I

^4
j1 J

^ ii»ii\ jjaXt2
 \~TKJf J

tJI

* a j 1 J \ j) j ^ \ * L ~ iu v \ jjj9

J.kpI j $j J-JxJ^ (, ^iJwW ^1f}tJ'i!1

^ (J? ^ (3 * J—**'*£•(v )1,j

.
—kJ1 ^Jbw^tP 1j)«JY 1jj « 4a1&0 ) ^g.i-...*LtJ

£ d__J
<■ .
i~ U j Ir di j t
<•
>Xi™A51 tJ j
. (
•
(2) 'Lf'A~»s^ 1 ^ l3 ^ J '“ ' ^ ^ Cl^-u/ 1 j
(Y )/ tJlj.lw ^»i I

(t>
(o )
* t»S^ b?1 d 1*aJ1 j31—*~Cs* l_t4Is

(1 )
• ^—■i.vuJ& 1 1 d ^51jJ1 ^Jb jjj jj

(Y )
* ^ ^™^-fcuBJ3}eJ1JJ LvtJI ^S1ulJ1 ^fb dlvlw jy1^

(Y )
* ^-Tw* j 'tf 1 <j ^51 j J 1 (jr^* d**« 1 j 2

(A )^
♦^ 1fi lxl1 d^51iJ.11 {jr^ (V * "j (31 s>

* O J ^ ^ jJ d»sw1 1 6 1«xj1 ^ji.^£_v'£>1jj

* ( j ^<U1(^J 1^ d■I Lj1 |) 1»xj1 d, I 1 1^-™*J j jj

• 1 d 1fcXj1 ^Jb <iJ ^ Jj-» j)

* ^ ^^1 ^ J l^kj1 d 1Xl1(

lAaJ ^jb dwtl^ ■!*^

* *\* ♦*
♦ 1c-J--*J T ; 1
*1
i * t^ikLv t-_^ *r
CSrf^ vt • Js^ L-> *0

* LS-^ O*- *Y
diA-wIrJ1 /j ~ ^>mwJl^j iUvJ LuJ 1 : 1 *A
 (408)

• (Jj 13 ^ d j 3 ^* jJ \ L? ^ ^ $

♦ ^J ^ ^ L- *J11 J I I I lllllln^l ■ w j^ ^

* ^w O , 3J 1 3 d -A j 1 3 J 1 J ^ 51 tAJ i (J | Jj 3

♦ Lft( jS lu J V » ^ .J \ -A L L ^f \.^ - J — a 3 j$ \JuL—* J \ \.9 1 3 iJ \ ^ 21 ry j 1 jj 1 3

(T )
jJ j) ^ lJ 1— 5—-™w*^1 1 j ^.9 C -jiJsiJLd \ if , U \J j L j^ jjv.ji? 1 — s 1 C fjl , wugi V _ J 3

•(t)-
3 <»’
-L— 3 3 $ «oJ 1 j i 3 J ^1 1
. .f^j U
.j 1 j ^3 J ^ 1_js 1 3
(r )* A j^ - j -*m

(& ) ^
^ tjV « u L ^ O 3 /j ^ auP (iJS ^U v d jJ 1J | J —A J j j j I? • i J 1 1 jps

. d l) a J > 3 L a jl? > i l ^ c l^ J u is ^ .0 1 3 LjA jL L —vs \ j 1J Ip f li*s ^3 u L L w 3 C j L _ U a J \

{• ■ ■- |J L w J I iJ 1 4 I? a . .9 O 1 O u 1 ^ J n . L ^ 1 l-^ « * ■1 ^ “ 1^9

£ ff^na.."-*- \ ^ ^'
■
—j ^ . ^ y
-o \ ('■■ ■--" V’iiJ 1 ^ t (■, kj U J \---- j-'L/Ji'iJ1 ^9 O. 9 . \ jJj ^ajuaaua-f *1^

(1) (A) _ (Y)

y I ^ J O J 1 ^ ( ' - *I O I /) - cL j 4 -11 ^jS 3 O L^JW <0 ( V < lj I

( I* )

(n )
L lL -^ s 1»o 1—*q \ ^ * ^ jJ ^ ^ 1 3 — iL / t t J \ ijJ ? ^vj.'uaJ \ ^ c^w 1 j } \ 6jS \ 1 tj? ^

* * i ♦ ^ ^ ^ |

* &Jljp 1 dj \ *1 ^Ay J * vJ* *|y

* (3^ f ^ ^ j + ^ ^ * £>

* * *>- *^ ^ (jo 3jL ^ *y

* J ^ \ £ l^ r tJ ^ l^ * £ jJ ? ^ * f^ J ^ - ^ W - V 4 J \ ^ L w J \ * ^

^.u.C' ^},jTyLa<J\ A ■,^>J'^vJ\ ^ V^ JJ ^ V-tJ1 d1 i.^fJ^J^ ^ * |*

'i ^»1 jJ jj^i-itJIJJ d^AvJ1 1* * « * ^* J ) *^ ^ (J^^ {JT^3
c
• 1vA,,e. >*z j \j j ^ ^ it.^/j^tJ \
 VTU J J

^LlF—MI^WI ^ j jh 1aJ Is * I ^ aJ 1 > 'j ' 1j & ^^3 i^ ^ 1.

/>./ j

• d—*0 VSj \ d~. 1 |^S 3 i£ j 1 fcAw^.

<•). s „ ( i ) p< -.-0 3 ( r ) M3 3 J ^

^LJq^s) 1 jjj-U- J^3 J 31 d 4r^J 5

d»JLuv dUi Va-iv JdSXM -vQ Aj Lt-iy

vji.tJL.iaJ 1 d^JLtd 1 d»J*<aJ 1 ^yJUJl ^yW(J 1 jt 3-^ Ji 3^

n &e>LaV.U u* y ^ . iQ 4J Lj-«v L>v ^ ^Ui I bM'V iH

dtJLioJ 1 d*j5aJ J “-lu-LifjJ 1 u rU j1 ^ J l 1 jt 3^ A ? '

{ i■.a •451 j o* y .451 ^ U " J3 vJjJiv 1

CtdLlaJ 1 l± J u J \ d-JLSaJ 1 ^iSaJ 1 j aJ1 ■*.Qwn1
^^#W Ji 3^ J i^ '

Aa Iaa-j d>J J A*>L»jt« d 3-* 3 CU>» 1 j

vAtXtai ^ d*Ju<yJ 1 1 l^i~u.oJ 1 jt 3^ jij>*

* c5 3 t, : i .
• ^nifT *~\ 1*1 1 * 1^ ^ d l^,yj 1 # L>/ ’ • v

. H ^ dv^ \ V^tr^-J :^ . d^ 14^» d 4v|J : s . v

* d * I s^ Id • d * (3* dJ b J • £

(^ t> dJaiLi> J j 1,j 5 1 (j.d j 13 j ^\ ^1 1 L™a 1 3 * U-* 1 ^j.4 tf jl j • &

dui-i^w ^ *d 1 1 ^ *1 J (jr^ L^jlsj.io j5 Lu» ->

* j—kvL U 1 ^AilJ1 ^jj5 d...i,.l.i.ip, 1 <J»4_fb jwj *■>■• l—' ^ • d»* ^ d»..il-^J 1
d^il AJ13 ^ 3 t4.-uj1 d J) hK s* 3 ♦ Cl*1 1^'ill d .9 Lc& I 3
* LaJ1 JitiisJ1 ^j^ ^ J 3‘^* VJt
 iw j

>■& j ^ «w i^? v 5 J ^ iJ -X # h O ( V 1> L a v ^ I jX 1

tAclioJ\ \ Vjl^JUoJ1 \ \ \ JJ jJ\ JJ \ Jj JfJ\

ni ^ ^ ^J ^ *i ^ ■■h ^ jiu U, ^ J ^ Li 1
ui*uLj<>J\ Li*JL£&J1 dtXlftJ\ ^UJoJ1 \ ^Ju^J\ jj jjJ\ jj ^J\ jj y}\

^ ^ 1*5 innfrfl ^..w

ViA.j!^ | | | ^ 1^ (p Lt^V i ^|J ^

i L^tlioJ\ ^LaoJ1 1 jjjJuaJ1 ^JL£&J\ jj jJ\ jj j}i\

p * w «p? ^ ^ jjl iMl>JL>tV d j b

cu^LiqJ1 Li*JLoJ\ cu,jJJj>J1 jt j Jt

J 3JJ J)jJa—a
0)
^ J X l ^ ^p L u u cl3^J (3-1—a L_AA»sa SL>1j -w

O-iJLj^J1 tJUJu^J1 duUJ 1 1 1 jjj-ioJ1 j j 'j]\

(^11^1*Iij « L|P»I«U>X^ ^J ^ ^i"Vk?5^ (JAIJ$ jjad^ ^i^*l1^

d -tll< *J 1 liU -L ^J 1 i-liJLi&J 1 1 ^jJU^J 1 ^_w.o,J 1 jj 311 j j 311

IwtJLSiia (JiSf J$ 3 j Luv

Ci-JL+aJ 1 i~—*11<>i1 '.it.LuyJ 1 Ujl Sjl > 1 J' \\ jt

^JLl^ 1ft ( ^ j.y ^5 1 J 3!J ^^J-1

'^*""^ |VLtAV (_£jV-SSfc*1*
,
^<JLSoJ1 cUJ.*»J1 \ ^jJUJ\ ur ^ 0 ^>11 Jts>\

(r) (^ ) ~
<J 1 " il l^ i£ L S W j J 1 j 1 d j js ^ v j J 1 j »nJ \ £ jiS lfV J ^ '•—-< Y Y )
J-:

, 0 ) . ( ° ) U )
f)....,0 j $ ^.£2 ♦ j j i ^JL, I? . fjj ^ ,"j>J 1 ^ A-J3*j LvtJ 1 6 jj 1 ^ L«5j a dJ j& j-tij'

*■ ♦ * * jj 1 (jjJig*"*""1* jj j^1 jj j^ 1 d»»"*j l^ v # dJ j ♦ * j

• L^J • ♦y' •y
*

* ^ ju UJ1 * L^4^«j j * <}.' .a LJ1 « '—j 1 *& * 1*£ ♦

. Ljji : 1 *y
 (V ) () )
^ 3™^*^ 3 J ^ jj ^ ^ j , j VEJ \ £ * (J? j j) JLji? \

i \j j lS j ^%:^ * £— l-J \ dJ cy^jj jj & {£ j 3 j 3 J jH O^ 3—^

* 1 j jj ^lw^-^j£)I L^jI L-^«s.!i? 1 i*j^*3

$^ J<*^W^waJ\^ iU J LlS^]I J ^>4i*-AJI j^Js) J Lt>w.y1 tutAti L^ Ij 1^ ^)l1 ^ ^\J

(a ) (i ) (y* )
^ ^ ^ ^ JJ ™— J»-JV j'wlI^ £^-WJ LtyJ1 d 1tjJ1 JJ dAJ1J ^p ^

Cuu U»1 ^ f j-viS' j..,/w L iJ ! {,....

j^j1^^-J1
^ )1^9 4 w 1J j S 1 ^ 1$ « o ' '
■
'<JaJ 1 t~iQ

(A ) (Y )
^ J jjJ J ^ drjjjX^ J J^illlJJ t^lijLiVJ ™t— Cl»L_JU \

(n ) __ .„„('•). ('i)

(1 r )+ _ + (iv)
+ ij^-j^ ^tiHfw ^tj^sj ^ Lu^Ari/j ^u nj—^*sjj i ^jrjy i ^^

(U )
« (1^| 1 <)4fi) Lvt> * Lt&AV 1 LJ JJ '(IJ)j«> jtxaOriulj

jJjVsJ jj * * jtjJ d— _a.J 1«A^»> jJjIsJAs * A ♦ y 4 ■w*«itjtf3j0jJ 1 4 j j-Q d *4 L j *A

*Li/!4 *y 4 A«A.^<,/

* ^ (3a ^4\jj ♦£ * *.>■ |J^ A‘lj-L~i-£Li-u * i

*tj?i^ Lj14 1i.ftli>• <>■ * ^cS ^tyt^S (Lu Aw4 4 1 \ ■■j«iQ 1 J 4 1 • 1

* <U JJk^ j ( ^ LZ*^lS*tf ♦^ * * V' ^ O a 34 Aj j *Y
J 5^ U ^ Aj A . J £yi i4 L j

A • d-1—.0 * js t / A * n
j ii« .fv iv d ll ^ I ^ • i— .» • ^ V • j ^ . l J 1 JJ ^ J d J 1 ♦ .-IT - ■ j |

4j-«i!>• ^ jk jju d<4Aj^j> ^. • » t . \ Y' * [j*tsJA ^ J L»/*^,u dLJ j _ s 5 : a

* lib L^iMv 'I t 1-4 * ^ jj*i^pJI <JJJb C-u^J \ (jLiSx)JJ lib l^v It * * * 1 t | ^
 (412)

( n u*)

(J J t> t— ^)*j j$ j >>j -tfl3 fa L uv CU-^l,S

i iJLlLi&J1 cuiU J {S*X^'
^J\ ^*4>J1 x.J] jt j
J\

^JJ-Uwk* faLw/ J j (j^JLk>w..<i U-^JL^na faIaav jj&Xj JjlJ faVxav dj>j I j ^S

Ji 3 ^ jt jM

fa V
wA'.' J j} J 3j Lvav J jj J (Jjjjss-*^ fa L
a^a
u JSJjy

Jt 3 ^

faLa>c.v jxizCtj toi51jj faLh-aw J >1j t2—4Lvsf

^ 1 ^ ^ Jj ^ J^ ^ J^ J^ ^

fa faIjuav &JLa
I ^juoJ \ Ji. jt 3-^

J^t^
Ji 3 ^ Ji. 3 ^ Ji 3 ^ Ji 3-^ Ji ^ Ji^
 4 ^J
(') ^ ^ ^ Lv4& 3JJ \
( OW
^J
>»J j l ^ 1 L
!L-(J
L»*
41j lw)| CuJi ^

. < . ^ . . < ' ) . . .

_)—2>1 Lft^-*-^^j& CU&SJj) L?jH J j)^ / **I c J \ fc 3**Lsa-Jjj 3j51J

() ) (Y ) , ^
1*■) \ ^*1J ^ ^ ^ j ^ J jj I {^ Ltv ^4? jJ? !J I ifrVJ w L j ^^ 1

1J 1
■ ( * )J.yl !"■■‘ rW^ t[ ^ 1^ llU ■J----jj ^
(Y)
• jj I Iit.u& J ^1 I^JjXw ^ 1 ^JU(Ji«r jjOlJ

^ } . \. . . . 1^
^ ) {^ j»Si 1j ^ Uj ^ j ^jitJ ^ \ '>
■
■^i ^ * ^^ifmiHiifc c^\ L.j^/j 1j tj^ ^

(<l \ ■!>.. ! < . . , , , .,. ( A ) , , <■ . 0 <Y)

tfju UftwJ^ dj jjXi j L] J
J1 (^5jjJb 1^ Li lJ I UJ^->j
< j— ^ J4 Oa

(n ) _ _ _ _ () • )
j^ - 'i 1 j via I J_t_j p-wti* (j^ 1— o. 1 j ‘^ j j 3—A^ajii tj ( J Jj j 5—> u w

( ir ) ( i r ) + +
t d ■.. * j V.J 1 3 ^^ ll-- i-.'i.'w AJ ^^3 Jj u ij L'-Ji 1 ^ • / 1—» «->- j >JjUJ

(U )
\j JJ ^^ lJ J j j ^ 1J j J J * ^J J ^ ^ J 1j£* j j l<|J ^
1

0* )
• ^ J >. ^1 l^ljin I ^ Iyl L 'l l»
^1 tl V^a1
*1 ^ Lvn^j i

« CUj3 • fc
^« * y * ^ JH * ^ 1

* ^ L teU U lJ 1 j Cf \ ♦ U j • ^ Vi liV U tJ ^ 1 » f

(_ bj
£jj l.C • V
—# *£
> ♦ 1 <
£^g.,L. ^ Cttnu1 ^ CltJz^ui • £

* ^ C aI^ULui •y « ^jv91 i «5»i *

♦ V^J LrW ♦ O *^ ^
♦ 'if 1 • \ ^^ » J ^a dj,'-'^LA« » ^ •

•ud Lw • .>■ I • |^ *^ 1 ^js ()JL^ *-{-•♦ 4* * 1T

JJua> ♦ 1 Jj * 1 ft
 V "i- IJ J

j— [j?Jl-j’
sJI 1 ^yy yo ^

* j l ^ j'i1 01.I...S.I iM
>J^>

1*j1 d^JLi— • Ctiil d— ^1^ 'jM cj^ Cj ^ •&* d*JIj

(t£) J <) LltoJ^\

.
^fad\ jj U ^lwt__»1J j O j ^ 1jj »J^ t t j J) ^
0 )^ i j tJ

(O (r)
/h / f /^ / f^ /'r*i/ f 1^1 f / ) / f5 / * / f* /v/ kls

/J*/
f
(«> )
Lii) Li ^-11-' 1— \ () * d-* I--*! iJ^lJ 1 JJ.4-WWj 1JJ.Q ^1n ,lii ^9 \ Jj fad'J I <3tA.^9

I j uU~mJ LtJ I y ! <^-j LiJ 1 f) / ^/ L—

.iiJ^J J (J ^ Is * c^»L»o-.e>jJ 1 j 1-Ljv tip

j tyi j LU J ^ ^ Jt? j LttJ ^ ^ ^ *■—■»j j UJ t ^ j tw

—<~jj V>J \ ^ 1^1 ^ ^ ^3

v-U—J Ltl Jy j i£ j j —^ » p \U ! jj j _ss- j j~~.*S' ^ j LsJ ^ ej j vAtoj j j Jm L_nJ \ ^ 1^1 (^ J ^ -5

j j j j —^ (j-v j LuJ 1j) / Oj / \_s*J I ^ f j f *t ->^ ' |^ i! jJ ! ^ j ^ j

0 )
• / l-/ u LuJ11}

(A ) (Y )
^L^ilLwAa!«^ LtV L^UhIiiUbH !J ^ ^Wul^ ^^ Cul! ^|—I^BU»iI1
^^

» L .^11 ^ <3wLCi ^ \ j j i-J *^11 ^9 \ e^i c,LvijL>i ^

* C v w ^j ♦ 1 d ™* A -w \j>» ^9 ^ * ^fj LrtitE1 * ] . y * •e*S=* * ^

♦ /^/ * ' ’ /^/ r5 / c/ r5 / W ' -r

w^1 * ^

• *j L— !• • ji—S*dj♦* ja_~J (L#j• • !& j ijpjU ^^ j ^ 'Ij IsjJ^iJ1 jjL& ^

j (Jrj* * ^ lil y.. / J "^'*i/ ** / 3j* * / /r^'-^’/** f tT‘/
* j *£ L y • * y 1^ J ^• • / *4jj ** j *jj* *
* o Lwju Id1 * i^t § ^ ^ ^ *y
 I J

(v x r u°)
I I i 1 i
uijtf O'*3Us«. jL
w e
nu J i' '
Cuv1j
Jt3 ejl3 a
* ,v J Lu/ Lv J
i' <
--
e
i

(jr;113 U3J13 C- J U u-yj Lw aL

w J
i'

cuJ13 & J U *0 O '-

'J( j Lvt
e
*
-u J
i' ;\3
(JT1 ^

ulJ13 (w>'« lyjay U" j l»«,v C

-U J5' <
y
. t*
O'^ 1> U" J lw Criu J
*' L?-J1LJ liJU

U"Aa jL
(.1?w U
v ;13 e J L 3

L>. Jlw ,Jl3

tjr «±J13 cJj c

u
* u
'J lu> lui J
i' ?u
u <JUJ1? Ab
&
I
l»s
£* O" «JLm» J
i' cujJl3 tS

<j j J Law & u J

i' <
y J
tt & j LK*° l
y
. L

aw1J iui £u j
i' Ur' 315 13 tJj U'i0 hpi.

J
i' y313 ui*j13 Lrjt>ba. U° yJLuf

I J
i' y;L3 Otl
c
v-* U "jL
a w

~
u
?/\3 eJ\3 'jL
U a w L
C1
. au <!L>

J
i' U7J13 cuJl3 £
i-> U U"1j L l
w H

J
i' Lj ulj\3
C-*J I y
"»U '
.
Wi
Jl*
v eL
u
1 ^ 1 1 1
I 1
 VT 1( /
4*
{£* J jj L*y ^9 <J 1 j L^\ j ^£ j ''i 1 j J£*ami 1 1—C&1jj ( Y£ )

0 )+ ,
*J L*«iO-l^ 1JJH*

J L-K-'.iP Jl Ji

c? *i *4dd J J 3 ol&JaJl J j*
c
£ d-ti 6 d» L O IdJaJ 1 c?'^’
c? c
<-J u
£
4? S d. oLuid^ ilJ tt

d. L^i I d e^ldJdJ 1
e C e ;'-'
J£ £ dS £ H oU dd\ Ls*»

Ji d* d<_l J* 6 LS o l^ttJ I J Lw
C
iS 4 >i dj L^j jp d.. liLiiaJ \ qj Lev

.j&J J £ VS jt O&JaJl O'1^

d J.3rf H ta?W 2 O IL id \ ^M) L

J JS 5S •JfcS d Jqj *; o L u id l L&

til jt •^teJ us J2 J o IddJ! j^uiS’ <JLs»

j5 dl ■jaw d* IdtJaJ 1 j~vi& .I*

$ Ji H cJS

UJ H J jS jS d Ji O LaJhJ \ jjCj> eJU

tf «s! Ji *-U JW l? J c ;’ j

S jS £ d Ji *.L» Cj Isd-lJ 1 j*v L>*

c
fi Ji hJ,J L iS J J O lfldd 1 ^jv J Lew
jjuoS* i
c?

£
<£ fi
t? Ji ~b L O iSjid 1 jJLo & u

^o <3 1^ j * ™
|* * * * ^ * |
 io j

jV ' (Y& l^}

ptJ ^*“*t«J (J 1 Lut^ U L.U J 1 J y 1

\t- Vt *

<3^ L f^ bi^i yttj (U L^u Jj y~ O \«iL**i^?J^ ih

\J* * *" C ^

c-J Ja? «5 d? <— <5 O ULtJaJ 1 cUJ U

-
/ L“

U-S J^ji y> <Xj Otj .J^ 4 O' VsL/iaJ 1 ^w ij

Ja> y> o5 ,iaj j)j dj Ou O loi^'J \ (j^ta Li-

^1"4j cy> i_~*j fcja jj 0 I / O' liL*JaJ 1 ^_j.gvJ Lto

dJ wLtj ya .£*> Uf -U oUlJaJj ^ j U

J jjS y i dl *Jaj yj O ILi-laJ 1 L

uS^ dJ Jj yj ^Z-«J Jjj .ja ^ O \5jJoJ 1 ^~VJ

J y> u43 vSui dJ y> yj *£?*> O \SuJaJ 1 j ~*jj U

dJ yj »*0 ^?tti j y -i-- O liLi-IaJ 1 y*-.‘j£* *J Vi^

^ ^ dJ y) »b »>2y Ov IsjJaJ 1 y*<tSf> L

t*U ^ ^ y! tjS dJ yf Oj WjJaJ ^ ^ '"'■•3b

if^ b ^tV yj L y d-' litJ.

la3 1 y v i& ^y

^ ^ 4$* *J$ li ^ y> i'Li iO ViL»JaJ1 y.vi& \js>»

fi 4(i *Aj L 7' j ^ o bijiaJ \ ^jjj Iav

o* o *

£ O i> e 4j 1-U L olfijiaJ]

c?
 v ^ iy j

t*1...«aLu> jJ j 1^ j 1 ^ y ^ ^

Js ^1 <J o I o 1 J ij 1

d5 Uo Ji3j jjj (Li t—M Js> ™ O lIL/JaJ 1

u-J -ia? ^ ~^S *>~J& Ja^ jy-i &i olLiaJI LiJiS

cAvls-H ^#1J

*^^
4 d vj^ ^ ^K-tj ^—'Isla'I^iJ1 ^)j4^Ls^
1

>1^) k^aJ ^ ^ k^> LH-' \it»Jj^J 1 JU

^3 _y«3 tl3 jj .Jj> cu* LILaI^J 1 V*v

J «*0 tii ji. 3^ ^ (jjo Vj

kS»i '-,J w4j {j^ J $ '“■*■' ^ ^

J jS -6" € d jj _>,iJ o lL .kJ ] j^X U

v!.j JJ «-L» Li (j5 J *3 «JS- lSa*<l^J 1 JM1& (JJ*J Li***

jj!3 d>3 1} ^ ,4j L» (j£ OJ l&xi&J 1 j - vjS’ ^jpi Li

jj 'i ^ <3 jo € jj O l fL»i'3-l \ jMiS> Li

i<aS i3 4j <Aj Li £- j JJ> liu k J 1 ^ Ij

"^| &3 t) ^&i <L> *»ti 0J> 1 j ^kj-S- {jj-'*-'* L&>

IS *^J &J I—<J l d kA LIj Ls«»l-^J 1 ^AH& ^JAt %J Liy

d) Vkii3 kil^i di Li *k-VL^ji^gi I Lw

 lj (V U u°)

dj ta
S t
is Ij «- j} b 1 C b& Jdl J 31

<
.—o dJ ^ 3b cb ULjJsl) 3 Lt

L_d JoS" j£ di t- d dJ «*> d.) bib Is d d 3 L id L>

L-d* jj dj L-u ^ C d cd& JaJl J

J* ^ tp^ Lp^ jj dj O l&dld3

tpb ^jj pi? j d Lj L^l^ji?J3 [ bIav

^ ^ d^> vb L«y *J^ b-1 3 3av

,^J 1^ pJq^ *libidl 3

b—

tad* *d ^ bfw ij pi? ^ b-* li/jd 3 Vj

t| J^1 vil ^3J wW bj bbidi 3 l^

dJ jj 4j C bU ddl JM1$> b L>*

k.6i bdi dj <L *^vw Lib Isd -d 3 j-^£‘ l->

jj ^ j5 ^ s^ ^ k ^

tdS JJ «dj b^O J bpp* V&jJ-d 3

H fS 5$ «& S Jaj y Uj Cb&Jdl JrUtS* u~

L J* ^u£- U^'
o*

di «5«S tb ^ "b< Jj bib U i d d \ JJJtEr £bj L w

(-4-2-1)
 \ k'-lr£LdL J

Lr
dj b j J \ O Ib -b J L J JJ 1

di ^ ^ dj L O \ib--bJ 1 |q-JL*

«_J ^ JaS $ <1> u -5 dl pu <b id U b b J L < d J 12

. C"

u b> I> Jsj j j &j L_^i { jj (J t d bbJaJ 1 1j

JctS" jt! ^ t—
*5" -.1"4j jj 4< (vU i—
>LibixJ 1 jj«4 L^»

^ jj d/ C-4J hi? j d L-»* Lj 1 ^jv Lw

^5 di S' kb * jj >Jo Id Lbb J 1 b„

^s-J I—4 J (J ^5 j) ^5' b jj i d Ib i^ J \ fjK i b

^2^ L«-t) dJ 11 »
. ^ Lb ^*11 L^

^ ^ g tt/ Jj w2SW I.,—' I b J - l il L b

eJ dj «£*&* ^ b ^ %^1* L b-laJ 1 i ,£ *** b^-

j? jS l_ £ d] J^i ib k5j ( jj i d U u biJ 1 jjm & Ij

jJ M JaJ jj jS u-5" c) o lib k J l d J li

j5 ^ Jd j id VbJcJ\ JJM& \J
Ji, t*rt£

J ^ Li J& j jj 4j i d Is jJ a J L j m & ( j r ia L i=

dJ £&» jj *^j b Jo j J O Ib -b J 1 jM t& b bv

j> d> *s S b ^ JJn Uj L O LdJal \ bv

-A.j 5*qI J i J
 \ '— ■ /

(V V A u°)

cr- <S c
o
lLk
Jl J31

<b
cr- Ji c o iLiJd \ ijf 15

u -j J Ji <b ■IL.bd \ v±J 15

u-5 d «b
C? *k Vi IS

±4 ’$ £ <S b J=b Ji >-Li L_^ O lbd I ^j.«sW

Ji dj l J: Lblb-dI J Iw

£ .6 Vw-5" Ji kbj Jr^ 0 1 ^Uv

a-

s ^»i *)=» LbUdhJl ^ 15

S Jo* >d <b oVUdl

J £ ^~£ d
s
<b «s\*> Lb l i J d l

eJ Ji ^ Lb I s d d \ JM&* j lit*
S c

j
£
3
d ..tjj 4™
i v-o (j£ ’l b d I jmiS' I j

jj l_
j
J J Jsu J J O Is ** J~ lJ I L_d L

j S d Ji Ji Uj J :P j L b U L t id I jm& 1 j

'i wU> Ji -b Lb IdJ'd1 jj.ua b*

S J* Ji Uj «sw J Lb W vJ b

L Lb l i d d I j-i.iiS' b*i
<? ^ pi Ji "v“d
 vTL-y/

j\$ j! (v VY u°)

^ ^ ^ u
J I k , J~^J1 ^J ^ !

16 Ci <?} jj Oj L £- o liJ a J l wp

i_J J£ j S . t f . i J d 0 4j o li,JaJ\ 'JUJ12

Jj ^
‘ij <
J O l1
^ I j

■.lit} £ ‘^S
" L tJ^j dj c~<j ci lijji^J 1 ck-^

Uj L J d O O LLj-I^J1 j^<;J Lui

<£) -25" tf ^ t^i ijj C»u Jr> Cl/liL/ld^ ^rwLty

j g £ *l£ ^ J^J jjj Cl- ILii^J \ Qz I?

*^£ dj t>^«j 1/ vjs> J) Cl t \j

J Jfv dD c^ti tiJ ijss&i O Idjl^J \ Iff

ifJ ^j ib C^j ^ ^ Cl 1 j-w& J1

^-*

vA? C^ J^b> j-> dj hs&j Cl I'Li.i.d ^ j-*£' Li

jJ *“
51 J^S" cfi ~£ cj£ dl ^ o Wi^J \ mJ L

t) t^J wO tJ
t— J O l&jJ^J 1 1^

^ ^ li O 1 j^vS* 1>*

^ wj jp L j J O kjti^il 1 J Ltv
V#i

cJ k'tjj
 -rc..kj j

(n^)

& lS £ $ -*- 1 o IL-UJ\ J

<£} C-ib J^j, ^ 4j £• ci/lil*-kJ1 y \*

*Tt^ '■‘^ ^7^ d jj ^U Cl/ 1 d j b?

ur^ 3i. "^i <3 j <

5 cvlL>l^J 1 \j

'■ ^ b “i j j JJ Uj C-<rf Cl/ 1 y 4 l>>

.AiJ L "l^ j J I—' o bv/.-b_j1 (_ J bv/

yj

^ ^ ^.j »4j L vt d/Vi/J--tJ1 bw

7*J J ^ w5^ b *»/ j^j Cl/\AsaJ^JI jjab

^ ^ <4j j^
w L ^ ^ d/ltai,l^j1 b

J j£ jS uS d ^ dj ci/buJ^Jj ytlp'

^ •'O ^o {J? jf- d wtf* Cl/IviaJl J^JiSr (j£J b=>

*4>b lb d^ *Ttd Cl/ Vsj-btJ 1 ^ y bi

4i! 'i ^ ^ Cu Jj til/ \c* J^J 1 yuS c j b

^ *b Cu ijp ^ ^/liv^i^J1 ^tuj/ \^

'/J <sS lb »J^5j y »*Li d - b s j-tiJ 1 j^ H s - bis*

l5" dj b J? ^ *4 Cl/IajIaJ1 jvuJ/ C^ ^ bv

^ d5 C^ tii yi 1/ Cl/VciJ—i.J1 yuA ^?Wbu

 IL—W J

L?j J^3J^

pi> dj v2-ii ^5 p d w°- 1 CL/IsuiaJl J j*l

<5 i—5 dJ jj Cu <—<y (_5 £ O idj-ioJ\

i—J J5 j3 jS v—5 J^j jj <Jj d/ IsjJaJ1 clJ U

w5 <JjIj Jj ■
—<j Jj» J d d>\3jJ3J1 0itj

tJs5 jS t) <J^i jjj Jj d-4J d->VSjJ-iJ1 b>*

Jiii Jj b j J l_i (jj \ JLw

^S —S 5 0» «£w b -b1 d-» i £**>Llw

£ 5 w5 tiJ 0> j)j Cj\iL*i^J1 b>

^s5 cJ 0 dj ^ 0 j) CJl.'U-kJ1 0av\j

J $ <6 (—5 *il JJ 4j Cjbh-UJ^ j^jt Ip

Ct) jj <L/ l*T<i ^ () taIN o\ib-bJ\ j-<u& jl>>

jS jS <
—5 J.'2j jj Jj d*j (jj o ULjisJ^ (jj*b>

jj ^ jjS -5 b «-k?j jj O Iaj.LJ\ j^S' ClJL

JlS 5 jp Jj b j d Isbbd1 j-v&> 0j1j

^ <> *s& 5 0 jj Jj Cl/lib-I'd^ j-±u£- ^joli»-

5 01 3-1 b C ^ J O UlJaJ\ j^uS> f^yj JLu/

-5 <aS sj5 d) jeJ dj —aV b o UbJkJ\

C
 J J

1111 I (f-. ^

O )
• fcj 1 ^ ^ 1 jy

(V )
4 1 (JJ^^ / J / A—*0«AJL) L j}l^sfcO OcLl-*oJ1 1j)

'i j Uju ft j1
j^is''l Jji*1^ (j3 (t j L f) 1 ^ 1 t^j ellJn.^?

^W-5 dj\]9 *^ I ^ ^I^ ^!1 ^1^*1" 1 di f]i fcj.

(r)
djAnlfi ^|1/^I ^ 1 jfj>9 I (jI»LilII 1mi
I*i^^ I 1 Hb^1

lw*0 ^ ^l^AU lj^ ^ {^]S \ 1 Vi!>*Lu3 fji * dj

(° 5♦ j_j.u J •» I ^ ,(°
(} 1 d ...9 -•

0)
^jJcw> jj t—-> 1 c?^ ^3 ^ ^ O a Jl5 *** ^ ^jnXu/ 1 A— aSjS ..jmA.1j)

(A) (Y)
^ I*?^ i CLwvIj J J)J ^1 jAiw I L«J ^jl IJl J j^JLi—J? *jj.v^sJ1

(n )
^ ^s-aJ1 Jci_™a ^<C»J \ j L? ( rt ..9J^9 l_.<y~-™.Q ^j,oJ1 jjJ j)1<-vQ ^

(Jjjj.^_™q joJ ( 1 i—jJLsta ^ I d .^ i Llao jj cLiJLL—cJ 1 tA3\j ^ t d—.»Xf \j-^/

• ( 1 *)
* ^ ^ O1
^ ^ cr^&j> * “* ^ “^5\ J ^ t d—**irv3 1

*(j^ d-»1 ftiLl'l

.>■<1 J ^ 3 1 . I . U -Ju ^ 1 J .y'
I ♦'I *U" 1 .*
**“* j *A * 't* *-r* Oa ^^ j *Y
j ^ j d 1L^v ^ / ^/ 1***^^ jJjJJ-i—4 * I # ^ ^jJ^u'jjJ *1—1* ♦£|
^J>1,'1ihi44 |V^ j ^yj J ||J^ hud l |iJ j ill j Awm^U lu<<V ^ j j CluL_Jl ^r>

♦j j IjsJ^ vJ»5^j <pJ j j d—^uLs-o j U.-5j

 (V ) (1 )
( S"* r* ) Li j j 1j) t 1 ^ y} y} L $ &iUi:3_™.Q jZ ^ ^ 1— S ^ Is

•— 4J j j) j) a,w Imv pj |>jJ I (JLl-^ - .«<> Vj Is & d u « 1j j ^J j.1 1

^ <■1.i.aJ1 ^jviJ^wO ^ 4 ^—*,Li__^J1 (.^l^a — J jJ1 V->£>^£* 11 „qJ1 ^\

* Jt ^vEUw jj ^ ^fj

(i ) (Y*)
^ 1 j~> 1 'i I utS.i.«aC>ii9 t-iJJ f1”^ ^ J 1^

(Y ) o ) (t> )
(J-iusisuuls 3«Ajn^j S—
jLii^ <y q \~S 0 \s *

. <A > i^UU-ls d..—■sjxsp'l Cu L5 O ^.9 ^ 1 jjJ1 ^L^>- L'*_jvL>-

• V-^Rvto^^ Lucw <n)^|s—JLLkS cl>1

— 1 dcisLXj1

0 -)
\ ^ J.^1 ^^[aJ^ nlJ^ ^ ^Qs) L/l^»lp,j£?1 iil=luij ^^Luw&ulI

I(Jjio—^yj1 (_yjjSt] J.Jy\1 j} (fCUjjL™ 1 J ySy} —<J1 1 J-&ZlJ L ^l^uta

^.p<JJ1j) (1r} ,
1—«
4 J JJvJ , .. ,
1 \ZJ j\ (,1).
U,’3 • JJ1 Lu-^tJ j UII Jik- C|

dU±L_J1 JLSlj i ^J^J3 ^ivAifiiJHiiV |Jr*

J I\
* uu / Cmv 1J

Or)
iJjj-aj ^Z I jj 1 i—^_L^i ^JZ 6 <1 ..^iLuv3 (j^ii-J^ Jj1j ^ <X«^v S> 3

. \ ^ U jis jJJllU dL) . (u} - J1j) *

• j)1>UIQ : i . v * 3 1 * WuJ^” * :V • )
♦ *— : i .t * CL*— * V *Y
* 1 :i • \ •Q
. ^kxllta : i .a «
• : \ *y
♦ t> 1 * \ .,. * Js—Ilis ii »<\
♦
* j3*-’ 4 •*• • 1 T * w*'*L-* 4 ^)
A♦
* jJj^ * '•11 * r *^ 4 ir
 IJ

Otj\jJ1 1 (Y 1 ij°)

0 )
• t1-5^ 1 <J

(V )
J l tj J 1 1-(
'J JJ 3 1 d *J »J^O0 fO«.j 0J 1 O'\ J&3 3-- r<X- TJy,M» ^U ~?j 1

(i ) ^ (r )
A ■■,»■—a 1 dLL 3 Ltuj «iJ j J j d— .i£> 3 *A»£w> ^V;i) 3 1 L?^1 <j—»*aX) 1 <*■£ 3 Xw.~-.-.<>

(1 ) {* )
1 U°3 j13 J 1 (_j1 1—fllS3 * 1 ^dLd1 ^/Xd\ 1 j13 J 13

(3 3 ^ \3J 1 f |yX4(J1 1 jM >

,4^J ^1 f*)1j ^X 1 d—Q-111 ^1,^^ 1 d -1^

^jXiJ1 j P ml«*L> ^51 1*X

IJ
,wLh3
• <■ (A )6 .3 1 1^ V■■niQ 1 !- '■ 1
1,4 (v )^ ^|S J IJJJ 1

()• ) ^ (<1 )
(J^j13'i
1“,
“q 1 ^ ^ 3 ^3 *“^ O4 3 3 “^ tX^ ioj—aj1 3 L«w d 1j<31 £j»S ^j_Lv.J1

dXb^ * ^ 1«J1 ^Jai\

(i t )
^ <> '•i
4,
p • 1-j■li^l^ i iAi)J ( ..Bn.I1 A aJ ^Ka$wJJ d Id jdJ1

11 ^^ XLlJl wwn ^j5*l1 1dji (Jr*^,

9 ^^JaMJ L/ Lw d»5^

(U )
d—*d d-j.—a-L .a ij t 1^ .j-jd .1y I ij .A.|~ j

(IS )
d 3 I I II ij^ u f jy d ^ i d—X X w O ^ ™ » J ||' 4 ^ 3 di ^ 1 ^J 1 1 ^ t**J Y | ^

♦ 1—— .(
j.-T
^- ^ i*-JLi ^ 1 f Oj * Y * ^ 1— JL# 'tM 3 1 3 1 (_3 ^ 4 1 * }

♦ dJJ j * 1 * ^ - « O'L 3 Xt'-tA * w>- 1 «y

* ^vAS——‘''■td 1 ^wXd 1 * l~.J * 1 * d—J^W \jd 1 3 * ^jJluJ 1< ^wXlJ 1 * 1 * g

» ^ da OtLis-w *y
* 3 J-d 1 (_j 1 j^t),l1 U^*j,9 pair'd 1 ^VjdJ 1 j PL~31Sj d * ->• •

* 1 3i
iitJ^i1 Xiiip • 1 1 %4j 3 * 3^' 11 X MilJf
cnJn *y
*4j d3 uLj * L** ♦ If.ilJ 4bI i~J^ ii^r 1 '* n • J9J j3 J ,^
1 4 v 4 \*

♦ 1^3-SSrJ/* < _£=• * \ 3 _Ll3d> * 1- ' * 1V 4 d-Sju) 3 « <B>- 4 ) Y

* J 1 L.ic- 13 13 13 3 t _s*. * 3i,^.JL,'-'d 1 3j1^ ^=^»dLfi* >A^dL>-13 13 J—■>.3 « <^ ^

+ l-lj l3^.1 j|J * u^l f |^

 * ^ ^ ^J^ ^ ^ 1 (T) 4 &.
—*&/£kA»JI ^^tI^VcJU lw4^1& d1

^Ji* jj I&j I™-*Luv L« vi^isJj (Jj1 CL<.J1 J ^A*—*A * 1.9^J^> l^L*^ J j jvJlJi?

K—j 1 (JjQ ji»*3 ^.,'S'. ti——-^jJ>- (Jjj£U ^ i $_jj.^3 ^"')!1 (J? ,91*«aILJ 1.1^59U- loIlJ1

* "" 0° )
^ * [|) I'wJ ) J ^*w«J U^Ll ^ lm-~kO l»^w vI^aLjJ 1 ^J W . ^ jiU J ^jA»*

(1 )
VI [*j ^ [ g^VtAwlnl1 jI^l^ Ij^ur1 kl^Jj ("j1 1

(1 ) .. _ («)...............
• (Jj«*j ( j ” 1"J LA-"^ O*- ’' njj—5 t!— LmJ 1 U^tJ jj=j (Jj jj >J d ^jiij 2 ^jAj V.j. J --^!

*AJJJ) (j5' SL»X» (J

j $ (jJU’
J1 (jjif-
(Y )J tf ^ 1 -
2—
(1 )vJ^sJj (j1 La; 1

(1 ) _ _ . - * . -(A 5

^ J} ^ (J ^)Aj Lrj— <) 5 J ^^ Q J) ixxj 1 ^vlc- j£> ^ ^

(V"*T J ^ »-*-- -j ill]^JL—5 f ! ob 1

(i•) _ ('*),...

^ ^ ^ Lwv^ 1 ^"-^1 ^jijuj t^iiHiLpJ■ ^1 ^j^ ^*n-01^

jliijj'^M Cjt^ j—k-1—^j 19 . <kJUiJl i^U.«l^I O 1jJJj (jw L»fl J1

J^\ 3 rtJ 3-AL- o JJU£U j) * |

CjJ^Ltu «Y « taU **>* 1♦ y

♦ ti) i ♦1
^^ • fa *■ *£
• t?—- & JL ^ I * I *y LvJ \ • „> ^
f*jiV 1 V* ^JS1
* Lm^^AAJ * ^ {*jQ dtiJJj^ *Lu^K^AUMfo) ^3 * 1 ^3 43^IjJJ *
* 5 ^ *> j jj l< u I j —5 \ jj A-iS i« J I i j j^-aJ i jj-j j i y 'i

ihAiii&'il\ 1iiA^jj * J d ^ * ]* *
* \ ^RfcA-bJU cfjjwU
 (j— 1— o vvl—.a_)1 (jo (J 3 13 < (J-ifSi13AJ 1 0 1 (j— j u I— o 3-1— .oj 1 (jo 3**091

(J1*^*J1 L-w'"i!1 ^1 (j——w (_j?1*^1 d.i'>o ^wuuJi? * (__9 1 i_j Ijl- u ; 1 13AJ

(V ) (')
i>j j J 1“ 3^11 0 13^ ^■j■fc-U l
L*3_ (° - JiifcQ^M1 J I
t* j i_ . j < / ! / o '—■= j

(1 ) _ a ,(° ) .. (O
•^ *, * ^ is^v 1 111k^v1 <L—*x9 1■ ^ V. iQ j jj « J j ^{51 jvilJ \ j

(A) (Y)
I—oJl> t ( jlo jl (3 . 0 jS 3 J .J 1 ^ 3 L* d—jl9 JosJj 1
— o jjl—o3 J 3 L
<ah
i

(3 i—^Jo 9 L?»t Vj^ »o j —aj>~> t (Ji*p 1 3“ASL> 3 j>-'<! 1 3

f1 )J ls-J 1 L-> Law ' i I j J a a 'i

• V*svj (jjJ 1 V—%©w-^kj 3 3 *A Lj (. >-ca 5 3 ^ 3 3 ^ 1 d A Lni^l 1<9 1 3 1 3 ^ Ij ^ T.■0

1 3 1

. 0* ) ...
( jl— 0 3J 3 V*«o 3 w>v>*l 3 3-t'O I® 3 <-* Vj 3 1 <A?W 3 \ j ^thAlj 1— *0 ( j 1M *a 3J
u ctlJ i.J _ j 3

♦ o«
-■Lutv1 d—*jj1—<>-!> 31 (_j,o13A 3L L*

0—_» 3 3*5—j j 1 d—-*j 1 3 JO (J L5»*J 1 U-» Luiv ''il L/ ki^li? I ^ t—f y Y ^jp J 1kA1 3

3 <-ioJ 1 & jit-d 1 ( j 1— 03 j 1j di t *-0 3 1 t> j 1—Q3 J S 3VtiO <1—*j Lo'U) 1 o l i 3*1_>

(—9 l& id 1 t—* Ijowf'i! L/ kl^A ^ lj 1 ^£3 tj 1 1kJ 1 t!.U kJ_ S 3 ♦ j t_ -* j ( jl «0 3 J I; 3 LviO 1 3

(IV)
.O 3 jSU d j Bj J-S L>*

* O I—**j * 1 •t * ( j 1—.Q 3 * 1 (jo d L 3 ♦^

* 13 1~**0 jS V *1 • &—& ij4^a « 1 jo <) J L 3 »Y
♦ jj Iaj Jlil* »y * JoaLl * t_> • £,

♦ *ih4i.w ~~ *A * —*03—^
(j1 » <->» ♦ y

: 1 * 1* • •*—' (jo 5 1J J » VJ
10 cuJai^t/ . ]y A-L . V —* 1 . } }

* (J. *!—< * *• *-T’ * 1Y

 J

0 >
♦ j f ( j L—o 3I 3Lwio 1 3 3*-^“ j 1 (i j^ -J 1 ( j l—o y j '—*»j 3-1 -o j 1

d— / 1 3 ^ (3 ^ ( )m- *■J 3 ^ |J | ) 'ipAa1 3 A J ( J— JJ (jo J Ij 3 ^ it k j-i 9 > *U 1 3 *9 1 \ t—^1 ttU kLkS

3

d->* 3*»ioJ 1 3 3>SjJ1 ( j 1—.03 ^jvo j *j- j d-1 .o 31 (jo ( j -—0 I—«3 (J—J)* 3\—;5> (T )3 3AJ 3 3&I

+
Jh&Lj *-A9 ( j jS j d j'jl £ j o j j j Ln-o 3 31iAS—o>j j j ( j l —o 3 1fcAj 13 1vJw^ \ 3

d 1 * ^ 1—J-l 1 3 ^ 1^ IaV ( J .■1*^ 9 3 ^ * - 3 J 1^ 3 % 4 0 Cm? 3 ——Ikl jjn II HIJ J*)11—»t 11 3 A j .1l 1 ( j — *W Inw 0jk 9

(r) +
♦ j <—• j 3j>*Sj 1 (j> UwjJ1 ( j 1~*0 3 ^/ *** / ly>l l—o 3 (j^ 1*«J 1 £*0

(1 )
L j Lj 31 kA*9 *j 3 1 '-j 3 3 -idJ 1 3 13 J 1 (j 1 |jl& \5 (iU J Cd 3^ J*S «J 1 3

(Y ) (1 ) _ (® ) .
4 (jk j I J 1 1 3 C (J 9 ^ 1 (J jA -J 1 3 t tl—.j^lv 3 _SvJ 1 (jo <J— J^liedJ 1 d *4.A

* 3 -^J 1 3 £ (J — 0 3J 1 C-Aa5j>- 3 £ ( j-- kO 3 J 1 3 £ 1 t-JLtA?5*

(A )
djA JAjsJLj l—o c 13L1 d 313 1 (jo 3 3J (J—S ( jl—o j? (J 3^ 1 (JjkEiJ 1 1—4 19

^ (UhnMv ^Ihl>1 I yO J ('•> j •. y <•. ^ J„Uj

.. , uv1
UjL >.f
)HaM.,1—^*.0

^ jj 1 ^ ( j vi—
*0 £j ^ ^

2—AJ1—4-*J1 l-j Lt^u^ 1 (j£ \-;i)j.P •f 1j d1?'"0 ^^

(n*)
* ^kid3J1 1*-L& (j^is’ d3A; d3A1 L|tJ1pi*tQ ^J*-* J-5 ^ ^ 1SaA>*

dy1 1 *Y * J^ J * ^ * ]
* J ♦ -|* * * * *[* ♦Y t dy ^0 4 O

*^ •ji t1*^1 * 1 # ^
* jjwMtWUW +Y * 1 * j {j *\
* (Up-%kj t 1 * ^ ^ * A»

m
JG
* ^ 4 tJ^^II 4- w£^~* I # ^ ^ J ^#

u >,0j : 1 . )f
 (j— ,1— 1— / (j— - (j— I— J— , (j— . (j— 1— J

f r r r r

d_no d ( j kAJ1 J 3 >»i ,qJ13 (j^ LwJ 1 3 (■ f ^f ^ 3 ^g*j.-..oJ 1 j —,o 3

(r }
t^J3 ^ ^—vJj^* O— l“a O I—o 3 3 ♦ 2— d ])lc d"■,*o 1 tfj 3 ^ J1 £ I) 3 3 —*— 0

13 ( J 1" * ' ^ 1 ^ Ij^ 1 3 1 /J 1 d,..■,/ 3 O (3 ... u 1 ,0 ( j 1 i0 3 J 3 I ujo d 11 1L J 1 3

3 1 3 *11 3 U 1 d..,.../ 3 */ (3 ..\jj l^ o ( j 1—^*43 3 * j w"^--' j ( j L^o 3 1„,,q-.<,,j. ,q

1— 9'-- 5< **J 1"

^*
*3 3 »jJ 1 J Iffl ( j —,oJ 3 ^ 13 <V**tOVo>J 1 d... ." j Sj ( j —JmI V—o ( j 1—*0 3 J

(n ) ^ u (o ) m (t) +
( j l—o 3 3^ d«4.uo l»sJ 1 3 d_r\j I 3 J I S jA lJ l o—v. 0 o'— *3 j •/ J/ o'—“3

(A ) __ (Y )
A ^ IjJI dJjJl L ^ JjW ♦ ^LulO dJ j jj~U1 |jjP ^ ^ O j j

Mill

• 3 3 A/ d 3 AJ 4 (j^ 1 3-*J 1 1**'-£• l— o ^ (J-S-^31 3 A.! 1 3 I—' IjwAi'.f 1 3 tJ Ij 3 'i! 1 0 1—b 3 ^«

2^1 „» * ouSjJi 3^1 ^ (n} (vrr o») 3-^

%Q j) * ^J 1 CltS-w j C-u ^,5 L^u-^u
\
(n )jjjS LlJ ^ (i-)* i \ i J I \ 1 0 11

(MJ 1—1J 1 ^ 4 ^ ^ ^ CLa-w(^OJJ ^ ^ j LtJ 1 ^* '•Ij ^ 11uJmJ ^ ^^ ^ 2j

* ♦y * o ! Qo o J ^tw • j

'i * ^ * 1 ^(j 1—*.qj3 * O *Y

* ^ #£) * j *jj ^Lrd J l-^U»0 J
• l^w! * u> Lh* 1 *y * ^ t 1 # *|

• ^1 ♦ 1 * jj * 1 *
•

f
tO Jtol<jIj ^ O \^* C-A^ (JJL J * 1Jkriuj1 f
t * j*
• fl— l*i f
t\ f
t^Y * |jj51f)jJ1 * O *11
 pjj\ oljiUjl (jo A^uto UJlj,

Vj l_^.a j uP U 1 y ^ 3 c \ S-lii? U J t jJ _j Vj jj 3 J j)'*! ^ lAJ I

O )
* djj 1J tiUvAj ViI.ii.i..,a »AS^ * ( / ^/ O J ^ ^ l>3'-o j

Wr<

(V ) t
Ujli»QjJ ^Vw^ <X1■■! j^vA ^Jwii^ (j\—i
Oj ^*
j1? VJi I 111101

(r )
^ ^ | y j \j ^ jj J r tlL & A |I||P i ^ . ^ 1 ^J3 ^ o J ^ (J 1 ^ ^ ^ ( | ^ ^ ^ ^

<JVx-,4
U ) (O
£ <j ^-vyJP 1>J 1 j (1 ; li) \ d-J^Vwi UJ\
IjJ \ j„) tj.^w U J Il j„j)
LtnJ jLrw ^1
<JJW t~n1 j*) , j jn)'
‘
^n

Cj ~ * j j j ——*-™-> j ■■*,,■«,r ( j —^ - o— 1— ’
r r r r r r

♦ V^ J, ^ •* * ^ V^tj il IiOuti ^ ^
j^< * Vr^Jil * '—*' ^ •Y
 ())
^ihil1^ (ll«mu.Vl^i1 nAaJ 1I1*1 ^ -..HjAwfL
i
L ii!1^ ^^ i flrr^HpJjJU in*1) ^9

L- ^ V-*o j t^LJ >«t ^ ^ ^ ^ ^ I***) j ^ ^ j taJ I ^ ^ L jj 1*^** i)

La (jL jU J I5 (WuO l i J I o -" 1 I ■& ^ (]kd4rViQL^J i t$ 3~*w 1 jJ \ 3 _j-s-Vj 1

^ ^ ^ { Jij ^ ^ ^ taij 1 4) 1^1 ^ ^jyj ^ I^ li^wj 1 ^ 1 i^ f) 1Jj J it^J I

_ (r)
d '■»>mmft J 'I (J.^W j 1 0 ^S 1vtJ 1 d ^jj9 Li i»^» JJ 4 ^O j L ) j L )J b yL>^

(t). ,,.
<) wl)^i ^ J jJ fcj^-i \ I O \ ^iU.! \$ * j C-il J ^>v«) J 1 L»*)^ ^ ^ iJ1^ y

(*)
^^ ^jI L^) J 4 Lj"W^Auj1 ^fyib ^^^J^^ijI (Jh^^wJ^J ^

(jfl ^«A-~a ^ * £> j j 1jJ I d J jtl jjjS j J ^ ^Jj V—%4 j nJ^lifcAi/1 J ss

(A ) (Y ) (1 }
Jj cJ 1 jj$ 111\ 3*-«Sj j *J Lf ^ 3 J 3I I *-d jjJ 1 jj cix) \ j

(‘I ) + +
• L j« . i J J ^ J i JJ ♦ \-^»>XsjJ> 1 L j jr f£ ) jjjA J ^ {3 J ^ J ^ ^ jJ ^

* L|>jjS\j <1 ^ •»?■• ^ <sjLj *V* *^ V—wftj • ^ *V • (!jJb »^ |

• ^ i 1 • i ♦ * L^uui 1 * ^ 0 » *>. j^yi • £

1 jj.a C ji^rfv * - ( - ♦* * -J- ♦ ♦ d~—

^ j^ w j) ♦ L-1 * ^ • JjM\ » »5“ *y
^J 3 ^ ^ j3 O 3 ^ ^ J 3^ G ^—"° ^ d»A%- l_ o 1 y

WVVfcj ^^ d"fc^-l VlJ 1 ^ ^ J^ ^-^*1 I o | Uj ^w^ j ^ ^ i4j ^ I 1 ^1 I

{Y ) (1 )
* Vlp
w&J i*^-i^ Jq-1^ ^ f i^«lJIj jt^tll^ t ,.mAJ1 I^ |) l^l^J1^

G*—"* Cfm
am
t-f Cj ~~~j 0™~' G — ' Cj—'J G—’■■' 0 —*,J

r r r r r r r r r r r r
(O _ (r)
I (J i ( j j ^ ' i ^ j) (J L ^ u J ! ^jo 1 (3 ^ f lJ 1 i ^ .1 ..^ J.i<5^j I I—* ^

^ (o )
j^ j J—>i*>«U (
1—^l™4 J1 J— JW j1 <1 Ijj1 <J ^jP J»!S^ JiilP • LuJ!

<Y ) (1)
* J ^ *rSJ*j ^ \™**Q^ ^ ^ / V f)wiimi|l.iB^{^ 1 f||—lijG1
-CH^

(A)
L* L^ j dj Lw!) j ^^ 1-w^J ^ J 3*H

^ ^ ^^ |^ ^ j^ jV4-^^ (n> V-«we^^ ^j^ ^j^^ 1 1

.(")... ('• )
.„tfi
jj * A, j„;>. iJj 1^JlsJ 1 LjLlJLsxsJ ut-JW 1 j) I V*J1 ^JjiillI

i G***^ G—*"*•"■' G -^ —" {jr^ 3 ^ 3—*aj 1—.w ^ lH 1 (JajVVJ 1 j $ J cJv ^ J j )£j ^

vA«UT3u9 « <3—wkGL 0 ^ j 1J ^ -■G« ^J<w»G * / . « , J« j) (> (J—JX> J 1 JjU&i! ^ t_SjjJL5*. 2

d —*-* 3 * d G° j3^ ^ \fi—^ ^U J\ G4 CH J 3 *** d —^ ^'•'■•[■'-j J j^ ^

d ^ -J Ij d ^ {j>-^ dlJ «Hs ( 1111 ijja J 3^ VJ 1 J ^ ^ J

* 1 ♦ 1♦ y • J-V.*l& 4
* l~> 1
1 » j

.j : v . % 4 4 1 ♦Y'
# *
* dj^M^ *^ O *'’
j 4 ^fitl1 1 ♦ '—' 4£j
* (J^'H. ♦-> 1 ♦A ♦ / ,J / 3 / - ? '/ G ^ j *Y

♦ * 1 * 1* *^ dd ijJ * ^
* UL^J^JL ♦ O ♦ 5 *j j
 — a 2 * ^ vjwLsw H i 1 t) I (jJ H I I ^J«iiLti U ( Y**1 cj5*) <jX ^ ^ j) ^ d i>^ ^

tiJJ jj * (jXHJ^ JjiuJi jjavI; liJUMj t CJJaSU^ |v*vL* HJ1 (_j£ivS*>

,(Y )
I ^ ^uj? ^ I ^ j 1^ 1 ^ L ij ! 1 |y J n§uf} ^Lwfc^j 1 |^ J i ^«i^j 1 ^ ^■I'1^ ^ ^

jf^L-S \
<• O
.
i ^AjlJ 1 ^jpJ H j
. ^ <5LjIs
. (* )C.^aA^S’*
. . . ^.1 5(j 1 ' ”)y>-\
.-^ L ^p U L i

(y ) . w (n ) _ (* )
1 & . j lr- ut^5, ^ 5 ^ 0^5 ^ L^v&HLi1 ^9 i 3 J LaJ1

^fcS ^ I [*j\s ^ v»^\.tf.i

^<II 1 ^tlI|»^1 ^ IaJ1 Cyi 1 ^ \jiiJ1 ^5 j|<^ii I ^i

(T ) + nA .< _ B m+ (A )
y ^ * ^ 1 ^ fj.oJ cf L_^ j.9 d .,*>Hvi Is^Ls» HJ \ JjAJtJ1

(1 * ) X X
* (Lm4^JfrW <J <3^

• 1 drt— iyAUJ (^yl ^1-^.1 » nQ ^

. 1.
* 1 ,f . J\ * ^ Uj ) * Y • t-£jJUJ\j * ->* «._.- . }
^sj? \ju & £~9 jj.o J \ ^ J \ C .‘I a A A J • »?». {^ y t 3 tJ 1* j • $ • C J L iiL ^ .J \ ♦ t— * * £

•1 ijJL j *y *4^ * I *^ ^ H-ii i

(lA^AF ^ -
t
-
—b
"“ ^^9 IkXw*^ *y , ,.y .<\ * dfi-U^J ^ ^ \ ♦ t-7* \ •A
fi^ J l Lil9 ^ \ ,^)p»^J*5 * i L > . £ ^ , \ j£ *& \S tr ^ ^ ^3 j>oJ \ <wii t) L^-J j * \

+ n,JSa I d ^ J * I > 4 ^ * | * + 1 ^ L jS jp il^ lriJ

 V ' 1w /

(* ) (O (r) (v )
• J**! J^J' O cJijsx* ^ (_J- l^*9 (^ J jw * u < y j ) ik J V j J 1 d— I 1

L»<jt) j <j—^ 1-J1^ ^ j>'i1 d—j^.o o~^ I—o

0 )
I—4j c]-'^'®; ^ o^ ^

>J—*3^ ^j)*t^^j> ^ / ***/ O ^ ^ .s (V r1 ) d.."t.ll£J1^ ^ 15J! (j—

^‘
^■*Cl (J?-
^ ^ / “*/ ^ L>iJ5*jJ1 £—-w \_uj £ 1—4 j 1—OJJj) . j t—»j

:jn-Ji
(V)
O- ^1 O'**** u—■
■' 0 ” -”' ( j—
*■' O—J 0 ■ '* • W
* ■ ^ JmhhijI ill ij J

f r r r r r

♦ I—* ^f£tl 1 1»»t»£} ^rriC~>^dv,5i—e jS 1jj

^..i— 4 1 U_j'-u^ 1tJ„Ci (y)^.u-i jJi=n_rjJ1
j5 Vav jj

(A)
^V-tuJ1 t—w-<,uj1 3—-L(£>UJ1 U-^J? 2_jAjJ1 dLi>1 'J,i£> jj

( 1 )+. +
1-;^ 1 d id.A jj

*> 0

* Lju1 <^L*£) • jv o •y >» jjij Lj J^JLui i ^ i t i -(- ♦ J

* I—
■“* » |3-!U ^ 1 • -£>- • ^JJ JAufJ(| L i\j ^1 > 1 * \u

♦ 1—3 j ♦ »J>* • Y * (JaS.JJ 1—aj * v-t>- * {j

«+ .. ^ H H if M *♦— ttta * !
* lP Cy lP Cy clP <>■' O j u~L*"‘ cP^ *Y
ut^ I * «|a * * « >|h * l\ * [ ^ 3 ^ ipAul I \i3<J 1 ( jP t\ 1^ • i t ^
 . .0 ).........
^ 1 ^ ^ (3 Vju«**.WI dj ^ J jJJ ^JmmQ^J 1 \—^ 1j

(Y ) (V )
t__4.lAuJ 1 L—O d J''"* ^ J ^JQ *<1 3 * d j£Lt I <»CAV ^jS C1.1 ..( J JJ Ll (^) J&J

^ j J (_j1—»-oJ (j)Jj 3 1 (Jjjj 1~4 1— 3 j ^L..J*l>- L_^>1j.4v 1 ^ja J LuJ 1

(0 ) _ (t )
♦ dj Li— 3 * j jj i*lj 1 d —..t— .33 ^ C? j) Laiuu

0—*j O-'-’ cj~*^ O "J u—■' u—J

f r r r r r
0) + t +
O—i
"1 •*■■•■' O— ' O— * O'”-"* O—■
“' * ^rj”^i O0 J ^—45
r r r r r
(a ) . (Y )
^ d—*“«w L.tJ 1 3 jjJ jj 'ill (^j-'i3j-J 1 C^»1 ^dlJ 1 d t-Lft 1 t t j^<,h L—41 ^

. . 0* ) . .(l )
JilKJ? ( Y Y (J ° ) J - ' ^ LJLoL*a—.<> j jj td ^9 ^ ll? ^L*J 1 Lj IA C u ii ^ 1 1 *J 13

(iv ) ^ m t (n )
6J O ^—1(1J ^ (J—4 ^ O ^ (J L) a ^-'f-i—05 ♦ 1 d—J

( u ) ( 1 Y* )
(JP n^ai^J \ j>A Jw J — 0 Jj (j-yvJ Jj^» 3 4 Ji^' J3 J o'— 5J J-*— ’

. J^I <
—# J.iiij

* UP*"* ^ * d ^*4^ 1 Lu 1 * 1 »j
* ^ LJ^ OrilfUy * Y * (^ »1 * y

* * ^>- * 0 • J) LtAUJ » o * £

+ <lL* 1 * ^ *Y * ^ o i^ w * ^ * » * ^. « 'i

qI:jj&l\ L Ift CIajev?jj!*JL?*1*Y • *js O 1 *A

• f^jLj j£l\ 1 ^ U
>La ^J> jj 1 11
-
> I? *w
5». t ^aj 1 1AA fj 1tJ 1 *'—'

* ^<M.UJ • > <3 ♦ 1 * 1 ) * 1 Ljfi Cj-SniC-.u ♦ ^ *

<J— .3 JJ CLtAttAJ J) 1 JJ jjJ ^ ^ 1 * 1 . ^Y" *^ J $^ * t‘T' * 1 V

* D ° 3^dts»aJ 1 j,-,o.J 1 * v^ *11 *
 V-r-rc- j

( V) . 0 )
• ^ ^ (i ^J— ^j.!^ 1_j$ 1^

r r t r
(r).
* L^jt)Vj 1j U V.cs^

<i

As

* 2uSU'jJ'a m*®' t- j j)

*1 cul^ui ♦ y /J * V>> o
*US"* 1
* IJ j.
k ^i ^ » u
i>
™* \J V V/&^ 1 *y*

* 't' {3^S U/°'v‘^ 9 * V* * i.

 (M. (')
* 9~^ 9 * <3™^ ^ j 9 o '-— 4 3 9 ** J 9 “* (jL -o j i L a 1^

(r)
0~*“' 0~ — '
r r

J jY^i Oa t) * l**J\ 1 (1>(_jj)-* (jj^jj"V1 (i^iLJ1 d-Lji?1 ^^iij»j)

(* )
* ^' '»JJ6 ^ (J—>3 1 ^5 J 9 ^ ^ <0—*0 J ^J — J) p~ 1

( r f r r)
,,. (y ) (i )
{^|jj __j J-- '•ioJ 1 5^ ^ ^ ^ “* Cj ^ J} * 3~s*J ^ jJ J _9 rj '’i 1 1 I— •j^i> n

(vrvu»)

J>'

* (j)--^9 * -»*' 1 •V
'— ‘ * (J— 4 ^1 ^_{L j S l ^ I—.* y ^ •<
—* 1 * \

♦ 1 ♦ ^ • i * cA* o*u" O - ^ •V *Y*

♦ iM-ttQV.'jJ1 * 1 *1 •6
• \ b 1wJ <jbbi>^ * "V*
j 5 \~^- fjS jj ♦ li^-j^5!J d y » w^5**1 *Y
 (444)

JL>'

^LjwIsy (> \ ^0 A—J>yn-^j <—• ^tJvU i>

(V )
* J IbIw-vObJ^ Ijj* 3yC> Aj $J 0^~4J!> *

( )
<j— O ■"' •' 1,1 o—■
3--*— * (^— i 0"*^'*l-~-'

(0 ) (1)
<)~L£s *1nUb L^j 1 y jj "'J r

A * t 3J L j * fjp'S''IsJ 1 • <~^* • ^!t4=»\sJ ^ t— » ysgS 1 *

(jrw * ' o* * V * 1

* V* LJa c^J^'Lu * 1 • YJ
• o-*35 u-^ cy <y^ O " O " * ^

* d * 1 ^ i 1 » i>
(445)

O ^ okj'iM 6
 * 1 t ■!Fr* ^ <J>^

V^j1 t ( wLL—o ^5 dJ *"*3iAAJ\ 3.A *)

^“J C
3 ^ 3 ^ 1-^Ia'Wjj ^Y'A (_
j=-^^ <
J»T
-l.^
iwv
j j) djjj? jjj l—o ^ I^Jl—o3 * d
— sJlLAi

^ ^3 <
i-
w <
»*-
5 «
»31J
J J ^ |*j t^-L)i»
l) j lr dlkXiv jyj 1J
) OJ^ ^qL.'i&

. J loJtJ 1 J
j jl!X<vJ
J

^ Is Lj.Jt(.< 1
? jj (jj Ij & jj 3 3 3-*3
3 C
U~-.v1j 1^

. I,.AJsJ ^JLfJU

U/^jIs [£jWt>*3 {Jfl£~lu^‘3 ^ j3 3JL_^s\jj j^ ^ jj_>

c ^3 3 ill3 ^ I—o1j
j

.(M . e _
*3 3-^3 U j-** ^3J 3^ V

,. (Y ) . ., . ,„ . ,(i). ,
t—\^wLu I 3 jj 3^tt.i 1 *.Xr>w X» j Uj 1 3,3—jju^'*

(n ) . (A ) .
^ 1 J i^V LLuI \jj j wCw {J7^ wLv%J1 ^ I? tdlliJ

*:jsuji

(n )
,_j—A-AJ1 J*o \ jj \,;i 3 1 £ ? 3
* St jjo J 1 dJJ 1 jj • 1 4 3 Ij j
 OU J 1 1

(v rA u®) /> (3*<*—- S a < < 5 ^ ji, ^^S3 ^ji <*lt .w cji^

^ 3 * jP3J1 (jU yl^ a«Jl> 3^ 2jSli J^i^iJI ^ u*Lu*^!i

0) .
4 *^ > > ^ y ( ^ ^ ^JS * b ^ < *> *1 1 jf J ," rt .%

„ (r) (t )
J-^ ji{ ^ ^ ^.»VlU ♦ (J^ili (I-P" ^ i ^f**^*3"■♦ L^i3^3 * JiuJ \

^ «^*jJ 1 JU&

J*y' <y jijy (t) . - .

i-tojb

(• )
C (5

H n 1 H

o 3^aJ1

(1 )
4*i& 1 —* ^ |SJUp3 (j—■*3 I jJ . J ^ (^.^5 1>» L

(A) <Y )d
tf .s ».<A*!9 ,.,.*.)\ jJ^Ai 4 » « jL ^ tJ 2 3 ,|4>) L 2 * \m
i~ ^1 3

3-f->* l^ J 3 * *>■ \ * \ *O ^ ^ i ^
j * ) L jw w 1> y 3 ^ c> • y* •3‘
,> 4 • 4U ^*» * \ *t> * ^ ^
*4jjiLlm • i—_» •y *(Jl1^ 1*^ * ^
* 1j^Iu » %>■ U» ♦^
 (449)

i v .s 9 c_i j; b * L
O i— ^

c5 u L? V— ■W I f
C ♦ •*

n T 1 1 1 U

t ^ ' j j < u l* \— «

U -u Aj >
<S * •*
c

t T X 1 1 1 1 A

*
tj ^ (jo J U 4 ^ If j
0 ^ — 5

L? l» w C u dj>
iS
C

n ' 1 1 1 U

i * <; . ...
£ J \ S t u q \ j j U j
j -

C? iS «— W
z

n X 1 1 n u

ItA A («p* j *& j o ' ( p i u ® )

( v )
f l { J l > ^ V J l ^ ^ /U -u j 2— e u U
u > 6 / - ^ / 5- * i J •u -b

( O , , . ( v )
C w » l j ^ 9 A u m ^ 1^ ) C > 3 > tA J 1 i JJ3) U i ( j l e L ...J - L -*S o i j l ^ i i j

, ( V ) . , 0 ) . < ( o )
J « tS U (^ )1 ' ttL l— J — 4 J 2 • / ^ i- o
/ O * -*•£ /

J j d QP lt9 ^ * *>• • j
♦ 1 ♦ (^9 2 j L j * y* • j j w j

♦ lo tft W ^ jl • O * dmmmj J*t/(Xi ( j ^ * »> 1 t tJ y|J i *jS • ^

^■U—S*-» • \ •CjS.• 1*1 */ ££/ / V-i/ O* W*^ * 1 •o

* ->» (*j* <i j 1/ j *v
 (4-50)

(V ) r (T ) (1 )
43?^^3 ^ ^)L >Ai ^,9tjj j ^ \Jufc

. L j J U J *

(o )
Uaj I J* jJ I j^a Cu* I jjS 3-Jb ^

+
c ^ c ^

it t t t n i n i t n t

0 )+
o

(Y )
jt^ls 1 jj$— *? lS j"*?'$ ^»Uc> \^l L «4 aJJ) ^ ^ 1

\j ^ *lSiJl »t— Is |^J 3 J 3 (>y— Jjl^e

*. "««J l|fl i* l lII■< dm—in.—i.<pln.

^...,.....nj1^ "V 1 l^lC

£ Vi £ 3 ^ 5 c ^ Vi ->i Z

It Y t t l t t t t 1 It

J J iS JI....,l'i-^ UO» \ *4Il.i

n*?It9 ^jl..,....i,ft.... ^.. ..^ jj

£ Vr*i £ 5 3 £ t5 ^ 3 ->■

it t i t t t t t t t n \ i t

•l*9^i * O ^ j .Y * ♦ ->• ♦ j
*dJl> • v>> •<l)L*m *1 * ^ nB***l>> (3*3
IW ♦ 1 d.i O® ^ *t*
* I 0* 4 ^. « <t 4*| <~c«*\^5 tj-"* ^ L?* d-i-tj * \ •^
• 4 I •Y
 (4-51)

^ *4 1 &.. "—*.
i
*** I3'*'■1*'1 *-1
,11 ^ J

(iv y v v i v v v v i n

J -'----- \-J-----^ ......................1 '

.
H J---— m
i--P
- ^— 5 ....-J 3 <_P 3

C ^ — C 3 3 £ S-*i £ 3 ?r

H Y t Y Y i Y Y Y Y 1 1 * Y

(1 )
J-o.Pl ^ j-^i ^-Uii <y s— l ^ ( v n u°)

1 - ^ 3 £ l£ 3 £ ^ C 3 C ^ ^ ^ f: ***£

Y 1 1 l Y Y l t Y Y 1 * i t i Y Y

j'3^1 J...i,W4w* (jr*3 i3lk-« J 3i JaSJ ^i^31 <y 3-Jb^

£ Lr« i »~S £ l 5 ^ £ * d 3^ '

Y Y Y t 1 } 1 1 1 1

d C ^ C C ^ C ^ t3^^1

Y Y Y t 1 1 1 1 1 1

Z t£ * r* i C *■** <^ 4** ^ C ^

Y Y Y t 1 1 1 1 1 1

d z ^ c ^ c ^ c ^ ^ c.’^ ’
T T T t 1 1 1 1 1 1
 Z Vi "I'i z ^ 1 ■> 1 UJ1

1 1 1 T Y \ 1 ) )

A (J 1 L? '■?* *£ ' (J^JLmJI

y y y y ) ) y r
£ Vi
iS Vi a> a 1 l? ^uUJl
y y y y i 1 r r
Vi VI Z ^ V 1 v 1 O'"** luJ 1
r ) 1 r t ) ) ) )
V*i (j> Vi A> 1 ^m U)1
i y y i i r r
o) • 1wUSl yS i t o* 3

M'laJI dJJl «A ^ jw ^1>J3 • t ^ 3^L j

x ^JIaj dJJl 3 4J~*w (_j*J "ll ^ 3XAH3 <J«A_>-3

• 3—J }L*C»Jl
dJJl {jLto 3 v* 1— 13 J 1 <uJ \ 3 v* 1 ^ a U 13 ♦V' O* *^ J

i ^l*w 3 (J j n .i> i fjj jJi>liaJ 1 ^^juweipaJl A,.-J 13 tJ 1».•» a lj«Lu«

. \J£xJ \jj±J
 ii

A printed edition of the durrat al-taj exists (ed. Sayyid

Muhammad Mashkut and Nagr Allah Taqwa, Teheran, 1939-4-6),

hut the section on music is unfortunately quite unreliable.

The following text of the passages translated in chapter 7

(pp. 24-4— 80) is based on three MSS.:

( \ ) BM. Add. 7694-. Bated 1020/1611. Small but on the whole

clear nastal llg .

The folio numbers given relate to this MS.

( ) India Office Eth§ 2219- Bated 1075/1664*. Small,

somewhat careless naskhx.

(f ) India Office Ethe 2220. Bated 1177/1764-. Clear naskhi.

All variants are given.

 ^1m )»aJ1 d^.iJ ^Ix)1 JjJ

c*wl ijf^^ij ^ j■«* <5 ^ a I—+.>• jl j o*

oL^Ji y - s ^ (J*

^ i4i j rt»»4J|1m4 y Cmrl j i4 *■£• 0^5

j. cJL i-< «

<1..*.!>i>»j jly>. jj C«»l j_j- ^J

;U.

o' 3' -H

>^CV-~4 j jL JJ Cu»l J ..W-*V .» ^

jt-bp > •2 ^XmJb A

 (455;

(V m u°)

3 '
■■■A jv^ J)139 dJ 3-* d-AtA.
f>» ^tw jJ

^ d*^L& |lLj *1yS.wi l,

..
M>(Xl U*L ^1 JLnWilwl J<J dJ ^

0)
(3 > a * “ * l * i 1& ( 1 y * ^ ***—^ d j U *. t j j, i^>4 c y * ~ * p j** y u j* j ^ j i I. «».<u

(T )
iii»* ^ i ^ <>i> ji 0** * ^«k> ujI^» jj1 imyumt^ >;^ ^1

*X»L 4 £ 4*2 y \j ^«suw j ^ aq ^3 i^t 3 idsL|<4^I^ Ji***3 J3 J $*3 'P\j

3' d*«wl tw*Sy Jf dJ >U<j5 4> ^ l^"^ jul J Jj jl £,-P

*S j J* 3 3 ' c * w i u 5 ^4 ^ j)i 4 pjkw (X u L -4 ^ ^ jrf ^

( r )
* c«*i dJ^ o ^ ' d j w y y ^1^jI ^jaJUA “* i*»-j l t d t-<.>6 ^

o' J 3 ' J'i—• jJ & i jP *4;'■; *-*-$ Oij* o' jij&3

(i)
(-5 3 4
1 ( _ jn i> .*»*J l ^ 3 ' 1^*5 y t iA M iw i j j 3 *^

J d * w ' 3 3 J l 3 i C t< w l ^4 dJ ^ 1 3 ^ vLu ^ L i i jjS ^ J * » l ^JJs*

(J^

. 0) cu*wl33
(^.■■i-dl 3*3 ^1 3*3- . _jl<*> (J^' o't^'^ ^
'
'

dS *U t 3S J ' 3 ^ 3 j 13> 4$ ^ 1 dJ ^ 3 ' ^ *y r. dJ

,
* {j &a j
*
* u-» •
» * 1
* ->< 1 ♦y c?U 13J
„ „ 3*t ♦
* ->• (—* • i^Ma
»» 3^ *
*1
1 * }

* 13 ^ J * U> • 1 3 3 3 ^ # t HtXt 3 ^ * Lj • L .tA.i j* t 3J * ^ i ♦ Y*

• 31 J * y » ^ jL j« it j> l^ • * > J • t>
• C m # 13 * •> • • I3 3 * y 1 » "j
 \1 /

J ' 2 *^' O 13 it dvw '33 J '^o-51* 3' 4J ^ X j j 3 it j byr- 3 j 3 j 3J j \ Ctw 1

j $ J jH J' J* t5 3 ^ 3 * d^«» 1 jji5L ^ 3 “*3 tJj)' <*-S~ * dtA*>i3 3 (3Lpws \

(T )
i 'iM t— Jb 3 ^ Cs,w 1 3 3 jL^ic* j1 dJ tpXj j Aw 3 ^ Cw/jJ j L>t>, 3

* C3^~^9**-^ ' 3 J ij C u w \ 3W 3 *4Jft 1 3^. 3*^ i 3J ^ 1 LjJlsi?L

* Ia j dU 1 h (3 1 ^aw — ^3 ^ ^ Lqd— a ^3* 3 */ C1*a *>1 (3“"i' A^*j 1 1— 4 3

(r) _ ^
t—>
1 3 —
^u«J1 gi 3 J 'i J 1 La«iw CmuSIt— 4 O If* 1 3 J 4 ,n I 3 S3

f * byL_aw Cb-y L—
a ^—
,^taw (
J3 i J-5 >.\ ( j—
Aj Jw 1 CwIft. ^ 1 ^Ji A-^hua "1 3

•W» ( t ) 11
tixii>CW it {^y I ^ .d
LwlwlVlnt.i i\sj .- ^
<ibl L— Jl^/ ^«y-"Md■ \■mfcjj^

^ (l ) (^ )
(J h ij 1 ^ ^ O^i ^ ^ JI ^ ^ Is 0

J) ^ teA*inf.J 1 J L?^ ^ \ ai^J

^ ^J I ^ ^ J ^Lufci ^1 Irn& 1 ^ Lb/ Lu 1 1/ jJb

^ ^J ^ \>iij ^Tk“ | J1 ^I (1^ ^3 1 ^ «bA.**■ C_3 1 1iib^ih^ ^h#.1'

(Y )
* 3*1 .-a t j^5

. (A )
^ Lb*!^ ^ miu*iJ 1^ ^ Xul la ° 3' 3 ' '3 dD ^ 3 ^*?* Oil' O 3 “?"3

1—a *4?w Aim—a ^•■1^i j / (1>li^/ It ^ ^ 1 jjI A / 3™^ 13

* ^3 3 l*'>f* * if* ' * V * l£ 3 Iftf- • (-* 1 » |

t airtjvi 3 j i—**wLu • y ->■ *^ * 3! * -~'- *»/i * i-y i •y
• oImai—* •o * UwVtniS—-♦ •I **| • ■■ *1* ci/L—vt«...4 • o» y •o
• ^yt)£Lfw * -2»> • y atA f*»4 • Lw I *Y » d* It ifii1) a • »s*

• 1ftJ y * O *: 1ft * 1 •^ ♦ i3 ^yiO-tW 3 * *"1* q* * 3' 3 * ' *A

 (D
(3y
»v JjI
d eft w^ wUI
1 tliJtLiuo O
/*)1
1 (J) 3-fu^viM ^w-tiP 3—jvw 3 3 3 0w>u4 ^ jjL> i—<w3a

Aw a 3!,^ —,a Lyj ' >i AJ »1 ,§Xtl ji 3 aw /-J L-a aS

dSw 3-^ (jj3-svj 3 3 3 p 3 OJl{0i j3 <3l^ 3»X c2 63^ 3^'

(V ) _(Y )
(jA 3JU--WS ^1L - V (3 3 ^ 3 J*1™W 33 b Lw^—W 31 ^ Cuww O I —u&jFbJ 0 ^.'4>VU4

* O ' 3 yj 3 Ip \3 J^ Ji—^>w ^Osi 3'3 ^ *^ 3-'

tilw 1 O L—A w -h j 1 Cl<w.v \ <5 -itE ^Lt—a 1-kv 13 3S <£> X b 3^ 3^'3

c i /1 H 2 :> j \ L.1 i -W 1 A ^ I f d * L u 3=W pAJ 3 S i C1/W.V 1 A —x S,.Q 3 13 3 1 L_aj&»Vj^

3i ^3<4jjw j^fwv (jt1 l—ACwSf*3 * 3 3WW A3 3W 3 ».»w ! (j;jj d 0 ' —.,G

d—Swif> A. 1
. IaJIw (jj 33 3 j \ A—■
J3w t-lf A3 L^*j1 j 1 <1} 3wi 3 3*
1—..a (Y V V LJ °)

(j£d Aw 3 ^ X L/w ^j J ^ 3^^^ 33 333 ^3 3 t .w ^ 3 -ui b 313 ^3 3 3 3

• 3 3*1 O .AlJ gA

3 1 ( j —w f J L 1 3ww 1 dS" 3w! A3 3 I 1 pJ 1 1 ft 3 ^3 ' Cy 3 ^ d3 3ww 3 S" 3 b! 3

llw Ias*. ^ 3A.f1/i 33 g™»A-i,^*«-»a3i4 3 3 3-w ^ d—a.^iw

^jw—ww 3 3 b Cw-/^fs- 3 i 3 ^ Clt9 b 0^3^ A—wJ) 3 .^ U.,Afrr.\.<,i> C..4tw1/jPw 3 I 3

(* ) .
. 3 iJl 33 3W-3

^3 3 I d 3 + ->- » (_5 3 ^ i 3 •q i * Y
6 * ^ "'
». * -.
♦ ^jj—w * Lw * ^J11■.0 * I* ^ • ,jo 3iLjL—.a • yy- o .(3.03/aHw—a *• t1 . y*
*33 W *333
» . f . I— • \ *0
 \1 /

('1') ^ ('f) (1 )
o (3 i ^ j 1 3„,*u t 3 S (j^ p*~?- a3 -f5
*3 3 1 3 J
t*
1^ 3 ^'

(O
^ wCj1<\fs« A3 3J I3 dL—.aJft 3„wL ^ 1 L^f.13 Jj 31 ^3’
J' . JJ J} 3 3

3-w b d£) (****^ 3^ 13' i J LJi, 3 ' (3UAW A_^» i d e w i Jftsbw dJLuJLfe» 3133! (—( > l/ ) 3 ^ "'3

c (0 ) c
1— t i J j b _ r> » 4 ljlL X o » i (3 0 — J lw 1 3 jjj L ^ w L L iu b cJ 3 X W 10 |s a v i C Lt 3 3 w L u f» dS

i C1aav^3 a-fo 13 Clew 13 3 3 -J 3 V.U 13»^- daw 13 13 rw 3 11 3 3 dew \ 3 dJJ 1 A....*, ---3 d—a53 - w

* dj _/> (3 / 3-1 Cuv 1 A3 3S 0*^ 2 5 3ws- 33 'j! 31 a5^ A.3 A3 ^buJ 1 3J?b>

0 '— ^ s' g 3 J O „b Jo J 3H

(J 3^ c
A 3 A—.0-S4W Jo d Jo 3 UJ1

cbJbv >1 dL—aw^w

Cleiui dL—
4J!W .ft. 30
£;'3'

c
A—aJlJ Jo "s" 3^5UJ1

(i' 3* C ■» dL— Is if. c3v3 b.u.JI

bJl
Cj d t

ii)31 yv» b„ c£ £ 3 "f’ A—i*_rw 1—/ wf- ,fs>, .,h if . ^34bJ\

kA il™9 13W3
.t— d—aww
<rUll ■

1
0 ]l5D^3 3 A if- 1 A—a-ft—
r i-bjt

* b-3 * d jJ 1.
* /il 1 i*
 a3jS pL-*w 3«J 3 \ i3 ^Uejl q L-aJb 3^3

JY )
d e w I <1/ ^ 3 \ 3J \ ^ ^ j • d / y |i3-** 34 ift 3 3W J 3 3*>. 33

' 3^3 j B'*->" aJ 3^# 3 33 3> i 1 33 31 {jijjw £ 3p iji |» 3JL4-4

3wL£J I j->- * dewi A^.3 (3^ j j (jr— *b*i Jb. 03-^3 * 3 <jlyiu>i 33 Ji — *\

^ 3 3-u 3 ^ aJ 3 |»ft \ 3 dj 3 ^ 3 333 \j f xUl-5' 4> 3 33 3^*w A* 1 ■* /mw dS"

(r)
d ^ * ' ' 3 ( J tf) 3 3 3 3 / ^w 3 >j | \ 3 3 C t< > i 3 3 J 1 3 ! 1- 5* y

• ( 3 < t ~ >Af s J i 3 3 C le w 13 3 3 ^ ! 1 3 3 £ L p U o i 3 1 d »5'3 a 3 3ww 3 ^

d— *uaJ
- <* )d L *bJJ 1? * d ew 1 A 1-A ...5 jl ^ JI ^J Aa93*w L—eJ>-bs d%tl L,^1*

(A ) .(Y ) _ . (1 ) (a )
A—wL A... tC t .4.J Vi!> * 1" lO 3 13 .0 (34 u 5 34 It ySU y ^J

(n)
• Aw Lels Lw4 J pj ' i Awl 'l! I ^ d > 3 } 3 1 .aw.9 i - d y J It J jit

{1♦ )
i d15j J * 3 3pb>4 *
>y <iu>w dew i Aw J — afr —.1^3 !
( J y dLt.^-i* L—a13

& <Xj L>4 3 I £5 3 W— >4 3 ^ (jl j 5> iS 3 J 3 4 t2 3 ' J 3 ^ ^ ^*u 3 A A^ j / A ^ A«*# 3

d< L.4.iw 3# 3 3W J li W iti i"■>iS/.ufti * I y X w 1 >..4.t O W |3ft Aw.jyw diSAfLf*- 3 *3 lep 3

yrl—•JS' dtdbo dwb 3 1 aS « -! c*

, * * »
* jj<J3b * Iw • * I 3 • d • j

.jsuju : .jjuju : v 1 ^ * d 5 34 * >> d * Cm «13 * 1 • y*

* ^3 Lw I i y eJi d .^ jj aij\ : e..'i ^ ti-ji: 1 .0

* * * *
4 k«4>OwJ ♦ 4 4 I ^ • 13 1 3..J ^4 • e>* « i »Li^ 3 ' 3_JS (3* * d 1 * Y
* *
* Aw Aw Jsi . d ^ 4 4i*5^
• 1 • Aw Aw JXtl •At At3LI • d dJM#b
f • ->• * de*vl Aw • d I • j•
 L^-bu;

^ tJ^ii ^ V>CJ>4 ^ jj ^3Im^i^W J I^m^U1 JUcw^_*j

^ ^ >1 O ^ * o '^ * c'4* 3 o ' 1’*-' tp jb * ' * * lX ^ —4 JJ d b lj> t J

dSw 3-w y«W dji L A) pb*A J L J jj djlf> 21 -MUhW 1 jtJ A AftJ (ji

' 3^ b / 3*1^ ^ d jl AjOj 3I—4 a -4JW / l (j/W • 3-4-mLj dXwl J

d w 13 A3 3 W 33 ^wL>4 • Cm) A3 Ji u T 3 3 (
jy
sJ4$ 3i£ t A
td
3 3S

3oaJ / ) j ^ laJJa—4 3jj dew 13 A3 3W 33 3 31 j ^ lltl 3^09 / )

» 3Jtw)340
>>» Cm \j t€sj l_y _ j \j ,3 d _ j fj / ^j ^s, It I
t-

(K-)
C 93 b 3I b J iw 3 j ^ y A1.t . 4 u 0 * l> (3 3J0 3I 2 Ld—X?1 j<J jJ ) 13W

( V Y Y 'Y 'u i>) ^ j 3 *° (^» c»J < j) 4 0 3 3 *9 I j Jj A—4-a iw 2* - ^

(& }3 .,<
»> . ?
1* . 1
J L-a..ft-w 1
. *U )
pw / Ji / a i-4.m
*. <Ti
3 ^ '^ *
e
*’**■* 3^ 333P g 3 * <—■.a

(1 )
• 3 yJjj 3-1*—5 33b*w / / ‘B / £3 / J' * J >«*

(v)
^33 dS "* / 'd / j' *1J it) dS 3JLS) 30- 3Lu&1 (^)<4w j 3j

33 2 ^ 3 -* !}) ^6 UL-4-& j»*9j 2^ ' J*^ 3 * ^ *1 ^ 1 j

. I| • / ,-r
^ > 3^1 3 3w j jl <V jjj-Aw ^ 3 wu d / 31 j L j Ait y C m ju y du j 3

3 3 3JL1 . 3 3 3 j J. ^ 3 »*4—4 ( ^ )d> 3 3*gu (')C m it 3 ^ 3 4 u 1.4 d * 3 j-iA Li ^ 3-w 3W

• L • -fi C i 1 \j * 1 » y • -
5*. **j' il)
ij * Cl
* Ci I1 ♦« ^

-3- <—I «J‘W 3* |k_5

'W *1 ♦ ^ .2bw\ * y..2' ■"*' * v ' *y
» htu-5 • Cl . 3-4—5 * -> i » Y * i
W ,* lw3 **AjLiI 1J
if •
* 4^ I *o

uj jS • —1
I , dp 3.9 « ) *^ * ■1■■■—>■ * \ * * v> L-# * y

■3 3*^ 5 » l-7"' * di 3 3 «a * t, »>• 1 * 1 * ».C m I3^ 3**u4 di 3 jjftt • + 1 • <J

 I J

{))
3' J^ J *■'* t“ ' ' 3 2 "^ h«J Od 3^-p J — C-wL j I \ j\ 3,_su L—a \

(Y )
^Jo-fia A.1—wj.y» C m 13 Al i 3 3 j J A .*?r.-o—a (3 bJ 1 d5 X t* Lb» 3 ^ 33 3 3 jS 1

“"'*X;5
1' j ' i 3-e> 1 j j / 3a Ia 3 3 3 t,iL»1 "J 1 ( Clevu 1 3 3 ^ 3a Lfb 3 j_ t (.11/ 3 3 ^ 0

/ 'i / A—a_ruw ^3 -y.ei—a 3 1 3w—a A\5 3 3 3 3 3 Cl/ 3 b (—■evu-'t*i 3 3 ^ /iLe; ^ LvJLi13

* 3

A l5 3.J 3^ 3 1 Aw;!? 3 P 3 ^. 3 4 3 3j L (_ jv U 3 ( j j j 3W 13 A\
5A-w 3

d lw L jd iia 1 A1
53 i 3 3*-» J — a —'vb .u .a j jj j A— av=vw g ) 3 - f. 3 3W ^3 13JS O 3 3.1.0 ^ jl

O 3~Z' t~ W J 3^ A 1
53 i j 3 L j 0 1
—a fb L 3 3-1 — a..^v*-w a j jj / A—-a .
-
r3 ^ yj>.

( r ) , .,
* J >: H-*;1 a-*.

C u« 13 C li 3 3v[C»w 3 1 g 3a_vu.a d > 3 3.10 2w 1 5^ 1 A\

53 L i- jr - ^yj U s (3 J 3 J p -fl 3

(1) ^ , 1
j ^ j U <0 3 3W j Aw j A —a ..^vw J V— a -fc L w 1 3 3 CwJ b tlU 1 C—t-euiw 3 3W d i 3 31

333 j j A— a-<iw ^3 1 ^ ] l —a-jevo 1 31 CJJw 3^ CleetuJ C u lu u 3 ^ C w v 1 ^w 3 *1 1 3 3

^ %
Jv2
•» ^1 L b / 1*u/j Li^tO
(Y > L ^Luu ^ 1
. j ^ j jj ^ J
J1} .

(A )
b 3 - !iC ^ ( 3 3W d-w 13 O 3 y,\d Aw U a g CI3 3.50 • j Aw j 13 0 p J

> -- 1 ^
3^1 (>5

♦ t
3w 3 3 3 ^ 1 • 1—* • 3-13 3 3 .91 » —
Sh 1 * Y * 2 '^ huw • (— > * be • 1 ’ 1

• 3-1 • Ci «3 3-. . -:r. \ * £ « p V/^dJ1 • -jt- • b-Xi 1 * c 1 . YJ

* 3 a • vf* ♦ |>IUI 3 I 1•
a * C— * Cw \ * 0
* g 3/VMia » —*■ C i » ^ aaft.- » 1 *A I__* 1 * Y

y C i
* 1
 { \) r
J J / ^ i / b /

bw j 1 V-vv j ^ j \j j cl—*<jr-xw 1 ^ 1 ti>u»w I j I ^ jj CUjuuu

? \t
d b jj J td* J Q ^ d^,«.t> Li-_4
S
(0 .AJ \ ^ jj*_«<l d> J <^i£) t—OU4V 1~Lf j)
^ )i «A_l b

* vAj j bui j \t j -i^L><-l dS b-3

b »J ^ cL»bw.o ^1 ^ j)A-«t<j d-> j ^£> tlL> 1 d \S ^i_j i^yj 1*5 ^.-fa ^

d-> b_*i^w j i Lu/ <— j \ j ^ i —^ Lt j m i X ^ i b j ^ j db 1 <■

—t^ t-v x j 6 1j

J™»0—
1%-^MHJ (j 1 ^ J J j '.tL bjs=. \J (y j j j d—>!MU b jh rj l!^ b.9JLW (f Ca-W i

• C^ui I |J»«oJ5 l—i b j J

d—it lb 1 J J b-»4-fVlMV 1 Iw UO W j l l 'O - 0 Cmi l J £ j j / f> U _ , O-OW {J J j 1 j)

(r )
JyJ d-<\Jb^r'l ^ * d**o\j <5yJjj j*A J \—it*}1 d>bji j1 d**1
****!^ ^.Jb

U )
d j b^.:^* jj db Am jj db^ wJ 1j d j~ ) jS i tlb L i^- vJ ^ 1 ^ 1 ^jjQ Lft> *A j -j

* d *tv *J j^)b»0 J (j}J 1 d9 f L-FV.ll. ... a b—Q1 ir y j t^J d 1!^ t)

! b •»* ^ *A j d^ \— jsw j >J j L1 O 1 j b *& l 1 &3 iAj \— 3_i d tX j- u i j)

O b_rd 1 t i-Li 1 j b bw 1 j^ y i 1 b d j b_f& j 1j b b ib «*A»t ■

j \ >J*nSiwi ci.b b_^. 4 *A o^t j ^ j d— ^ d*-<u i j d >A j j > j -)

(0) <.A a

• 1 — a j \ dJ—i - . <1\ 1 £ J J J> X y -* 5 j

i 0-i,<u b.A > O i C u iu .j L it • 1 i 'ju * *4j b ♦ jn ♦ '»Lj b * d> » «-0 b • \ * Y » d^ ♦ -|- *^" * ]

+ »J » V^** b«> * J ^ I II.|Q * i ♦ i, , d b jj i i)j . o i , ^

 1.9 ^ J .0 £ 1 6 b ^ *A A **v *t 3bL—o J 3 J jH d J^ J— ''"'3

db j d*»r^«i jj 4 ^^J.i^J™-a Clv^u 1 j CL- j i5^:<? ^f-bcrw—*s 1 d b dLtv <!Ll^*u j ^ *J C/j

•A ^iJi»«»^»^i.*-t.!i4 «A ^ \ db ^ iLJ d_i*rj^.u ^1 CL- j ^t o a 1

CLt*v 1 j C^_Tidd- (jaw ♦ -A 3 ^— -<i j 1 (Jj! j i-tju d„5 - 4 dj j L w j \ j ~o

• (J) \j£> C*„rvui j 1 (_j* 1a5 jj ^ib jj ^ »AJ i-L-i jj 1 t—

a-EM-u j 1

i f c / ^ A * L* ^ U V * . <■ 5 f \* f *1a."*- o O^ *-A*.L 1 ^ d \5*d~tw J}

(r) (y )

(l)
6J J i aS a^. 3 ^ L - < .^ * p « * J 1 3-= - j f J^r l a 4^

* |r ^ ^ ^

cr* 3 ^ <-5v j b > $ “b*j \ $ s . d ^ ° IS 3 j ^ <A}jUv / .9 / £ \2

(• )
. J d J Ji 3 ' Ap tS 3 J

1A*j 1 tirliibi>» ^jjuau^5- ^j C-1 « » . A 1**•*?}.-*•■«..a*1 ^^

(1 )
—i |V* j ^ J 1 iQ,t^LAV 1 ^ k-tj ^ i#W j 1J ^ -^VtAil-1' j J j d 1

t <J f f j J J j A— b»*_fl*AV 1 J< J d*J V&j)

1 ^j j ^ j ^.-ii.OjJW A 1*^** 1 I^ • A Cfti c.,,.1.1li^iv^d 1 p ^ ■■~*a ^

* (3^-9^A * '"?' S"’ * 13^ *1* * 1 tj? * Vr' ^ * 1

• 1j 1a j 1^ I * ^ o ♦ j 1a ^1^1*1*^ ♦^ * c- • *Ytii * “7* ^ * Y^
(S 3 Jr * ’ "T' d J 3"i d ^ A r' t ? . ? A d ^ J ^ l ^ jl^j 2 O U 1 ^ c3 1 J& iS 3 J* • * V 1 * «»

% lAaJuoj^ * b*** • >«AmU i>m^ ♦ w y - I . ’y . j f t ^ d j ^ 3 ^ A“°

 V^—r w ~ r j

0 )
*A.i—^ j j ^*«^ L ^js5- j)1 ^^9 j1jj J J *j 3'i"'
*o

L
m p
/1 ^^ , 1 ^jlcu i.^..O L-^ \g^^ i LhM^1 ^ ^LiJ^*lj j J l-l j ^ j kr.'^^^Tl u
O^

. (r) _ (V)
» >»0.& ^ Ij ^ I d.!{P ^J-P dd tJwtu Lj

^ CU»wI dj j^fCi J ^I d
ub C-'_i-(.uI d J j j) I ^I^ I i— lil_^J»L<,v L_*QI ^

(O
1^
Ud'3 ♦ dJ dJ 1 ^ Lft> j I j) I ^ 1 d tJ J4»« lib j I ^ 1 jI Ij j I lilj I Li

j^LlauJ dJ j)5-» J I d <J LfO J j) *Xs>* j -J d^ (J_jI Lu-*o jj ^ 0 -i.vI CJ^ L*j~ *0 0 'y-*

^ J J jS (& )^niL^vi ^^ j 5 ^ J ^ 5^5 i) ^ *—&»-! ItCU _} I tili Li-^» ^ CUAjjil^J jj * *) j \ -O

(Y - -. }
t«U ^j ^lw w
O I j J j (O d J ^,j I j yd
5
^
— -lhvI l«^A^U I ( 1

j ^ j jl I J I 'ivJW ^ ^ W^fl— ^ l-h Ai-fc^^ I %-Laj L«*4 ^ V■■-'!-'■ * JI ^

*Aii_5 “ ■ ‘‘^ j^'“' ?v0 CIUjwI

Cmv I^ >J; J-[3 j^ S I a!? ,U
¥

• *AJL1_J^ vlpj>i3 liluuj I j j ' £- j j* i ^ <ArfiLi j / ^ / <xJ ft51» j /c Y £

|^ ^ *> tirLij l!LnrtJ.I l L^LlJ^ J LhUWI ^ ^ J ^ ^ J j) di l— ^^

.J dj£» ^ tAJHLtUJ.J ^ j£> dJ ^ J iy£> ^ jjXno ( j U,>- j) \ QSr**»

lJ) ^pj»AjL«bjl d1"-Alib P d_i ^ .<i l_ » 4 J q ^ * d^ Cl*-w/ 1 <LtS_—

5' ^ \ .j *d 1

^J t) I JI 0“*^I j) * j L-i^ •v-W dLJa_5

• dwji) ^-P • w*- » A^3 j & ♦ '>— ' I ♦Y • ^Ij* *\* **

—JU U-/ i
» ♦ ^

u L *J*- * d <Jjn ^ iJ j d «J j-> * tJSz* 4 * t^ ^ ^

• ^fl_Sr>Jj > '—■» »(Jv2^V(l ♦ *£f~ I •C3
*__ IJS •Jlwy1 I O I *V t diLt..^i ♦ (—^ • c^5j_pi ♦>-'*■I •^
.CJVH__JI5\ : U-* .Jl5— !Ul :wj>.\ *^

*
 J

I dJ j \ J I j\ Lia j \ y \ d>- ( YV L C
sJ0 ) ^ d*-vuI &■£LiJL-^q \jb j I ^ I dJ»—^.?»

Aj (^J•'-*-' « j ^ I d_i j \ j) uJ N I ^ji3u?VJ JJ I <!Ll2 - -pS" J 1^ j I JJ d lj ic>'

j \j j * fcA i-tu V J L -5
—i J^ li ^ ^ ^j ^ kdjL^.v J ^ ^ La*<^vW ^

^*i*^l<0 |5
> i^ J 1 ^j ^ L^-ch11 ^
*
y ^ ^J d^ j^j l»
4-#s
*4 L»/^

O) .
jj dL-*o~>* i ^Sj i j) ^ 1 0 ^3 ^ dj L_»a J <.JL*JLdbJ l—
*o1 ♦ J jj-vv

v
J —
-a,p- C
_<l—
^cy^w jj j^j_P*Jb A.I .q.>- dj I—.o CJ2j-j LdJj ^-?'- A dun 1 I <
—93 b
s*-*

J 1-
---
-4 d) d-S~^} I JJ

( V )
IJ yi 'f£t Lw) jj Uj jJ * pji j^
}- JL di d^W I“U C*-WJ*-JLl I Jj

( Y >
• 1 j).A - <3 J -J ^ j ) p J wJ ^ J(

U ) A
L ~ usei ,L«. «J5U ^js, h J 1j J-*0^ V

0 ) r (O
V-^Jl—r^M1 dcw
JJU^»f di d-P jS- I t-t/iPU di lJ.UA! L* L'w'^J jj * L--*^.^1 1a
»''^5 j piiutJ

(Y )
(Jj \ di lAjj j)-*^*' “* “* _
/>^J ^JwSW-a CldLuftj \ jfS 1 tib LlP*. i-LtluLi dtJAi

d—ui-.nA Cf Cu I .o„,rw Jji I db I Lj >JJ c^S j ^ja,~«!i—.o 1 CltSAS^- J ^ $ * <

—“ uI A

(A) ^ .
L.^ jl 1 J O I J O La I— <D kJ j y (Jj \ L— wl-uju ^jj I j) t iJ ^-» du/P ^ —4

* 1wLl * w*y 1 •(^JIJyl * I *^ * t.3 • l_ua • I

1 •1
A ^
J ^.cy^-y A I .r
I ii i a I 4 Cm^JLU^ 4 ^ ♦ ^

J 1— a— IiI .
’(—- .^JI— 2— Ul ! ^ \ .w( ■*0i * W*' ♦ ^ J ii I—tip ^ i I1M*^->i 1 .0

u I L-cuvu * Ui i 1 \ jI * I i^ * J J j5i * L^ * «-JwJ A : ' •Y

 l_5V—a a—a
^5 1 j j ...»o_i d-> dO 1 j 1k £> c} . >AJu_^Xa j ■ ■■)" \j

i Lj j 1 j 1 LJ,U'J ^ VH**wjX* dL£j iJ^s j Hi ^ * vA^td1 j Ia-va; Jcii& ^ e^j 1

&1 i-Li Li 6 *J^tS1 (j£ tJ tJ j j -^-o».i 'o

1 Cj*L^ (j?* ^ *A ■ 1 ^ J —4X'

^ AaAULu ^ tAvi 1 ■ ^1*A*51 j *Aa/ ^ L,™

<*i ^ ^ ^| Lw^f^AAV1 j\

♦ dj (Jj-Li- 1 —aJ j ^.hj—.d J**-^ j)

(J j) I ^u-4.» v-> ^Ju—.oJi— <1 ClA^tu^JtAxib 1 <3«J ^ jjb £ CU-uui 1«A wli Lu ^

1t-U f^j 1 J 1—«c»S d..^- ^ 1 CUaua^ d >J j- i ^ *lz> C) j^Cv.q ^.•jOJ~i*J j*X 1

C) “U ^»w j 1j i I U _J’'i" ^ dj^Sjj <)»Ajj dtAiJi—^ ^ >AC^J Jb^ ^-hd4

dJ, e^Lt
<?.
j CUSiJij (V 5
*A•• <•
-a1—> ^„ 1 A.-.nl
- ..,<’ > , .,..
j Lu jj i=Jj^.V:(3-uj j J cj j -. •■
CIai—»„nJ

U>0 >l|d () * J (pu L_9ijgi .>0 ir^vi 1t>b jlAi j d) ^ J ijj—>4A-^

( r )
J1 1 ^jib t L*fCtUAi f |"|«Q|I^*^| •fJ1 1 dj ^ \i ^i«1(SbLw 1^ ^AV

» CLmv1 4j» 1 jj J £

* • toAJLi—S * 1 • y d j) * Sr* * J J ^_1’J * ^ * 1

* ■I * wJ*-
* 1—^ ^ ^
 ( J

ft'* d i.0

’k-' I
"*nft Jf
tJ uUj**L* n j^ L j JUJU.

^ »-Ci51 Cl-* Llav \ r .0 ~i..-| ■ b \ J *—

C-ufv-Cw 3 ^ (J^ 1 dS" d*^w \ 3 wL; Lu

(V)
3 ^ 3'^>3 (3^ 3 3 o JL j c-cx*v ^ \ v_.tAv ll_ ^ j j ] J 1—<L*lj1 3 “, 3

&y^s 3 3~^*“ 'q — 3 3d

C^-^i — 30 db J3 Ai i-Uu b dti CUddd Id_.^03 • JJ jJ$ ^jj>«J

^ (j b —
»o jjjA»!U — 33 ^J <V
i uCavL Ab 3 »■ iAaavL> A&jJi^ di/ ^) J 33 jjb

3 3 d (jJ i * ^ 3 3 t-~J3 bw1 1 3_j (J^j j 1 >3_su db 1 *L- ^ 3 »

J 3 -^ wb&s—dj

(0 ) (i) (r)
VAVVilun»«Q
U— *^S j.h> j1 L (33-^d 3 b (3‘^** 3-^ 3 3J *JJJJ 3d 3

. . (V)
' ' <0 )
b ^ .3 3 J —3 <_£ 3 Vib j J3I i-b—o . O J^i'. cJ j 1 L 3d 1

(A)
L3 3' 6 < ^ .3 ^ 3^ 33 3 3-1 b {^33 c3 3bfc 3 ^ 33 J U—33 I

• *-—*’3^s. LJ> 3 .to 3 1 (J» 3 bfe 3 3 ^ 3 3d £_» 3 V 1 33 CUjt-u*d 33 3 to 3 b—.o

(1* ) 0)
U b—o 3wfw cO jjjJmS 3 Ltodv 1 CUdd— o J 3 iAj** 33 Ai Ato 34 ^ 3* rS>* 3dd I—0 3

« ^ 3-1.03^ 3 tj^"'3 ^-33^>. 3>b j1 £d£>33 3 3d ^ 3w^>3^ ~Lnd (3Lb 1 d.*i3 3JI— 0

Or) _ ............... ( n ) (n )
^31 4 O J Lw 1 Lib*Luv Vi—wQ j I ^1,!; L~HkQ

* 1 * V.-- * jjJ^sL * b,3 S * V + > *1

♦
»1*1* ^.^—» * v>* * Cb^uu^u * L—* 1 .< * ^Jj * bj *
t>: 3-13 * -»-' * Y
* ^J * j:** • j\«fiij * 0 1 '\ •— * \ . L : .r-- <— > * £
>

•^ tV
iJ* i *A *k : ^ :V
.l
1 *Y
.6 ^px9 j) cjS" • 1—^ * 3l!
c ^ *9 ^ . 1
• )•
„ A *
4 ^^-*l|-tlf 1 11^ ♦ 4»^* 4 1—4JU*—d> •
»V 1 *n
* * * L1** * 4*1
* 1 * )V L_0 3 3 , v.^- Lad 3 » t-< ♦ L—o13 : 1
* • . 11
 T V_3L ^ J

, , . (v) .

1 3JI—jj jjjX—«4,C* ^35 ^3b >Jv(iXA»..»d toUj\d35 CIbifi3 Jl.Hd

t—a^Vau 0 U j 3^ I IAVLs J3

3 S 34 d jj 3 3 U -v ty l j

dLsv C5’ ^ 3^3

333 3J. O

U^s^l dbj
^aAV>J
.

( r )
0^3 J ' > t~LMViS^
L?

A1
53 *J (_*3L^Jti'31

A15 3 J ■j

^3>. j,L^ 3 '-'to

<J 3^ 3 i£ > 3 j j dj L.4 ti)3 Jd

(1 )
615"3 J 0?3bb 3 <i'3-0 & 3 3J.

(3 1
fto
u ? !“ d 15 3 3 ^ 5” 3

(*)
0 1 33 3 3"* 333r O '? 3

dl$ <u« o 313A d 15 dLvu d l,^ 3^

<=
dj V—0 a-UAAUj^ dU <JUv

* 1 • I
^jj'SdwAV3a jjyij.fi ' '-d » j^L—
aafi * * Y f j ♦ v^* * J * O 1 t j

♦ d 15 3 ^ * tod ♦ (J 1 3J& * 1 # ^ *J'> V *6 ]J* ** •r

* '-t' *t5 1 333 3-j I 1 •&
 dl<5J J 3 J $*

0)
Ub
lP3] dj j 6 \SAm

(r) + _ (f)
• 1->
4<U
AiJIJ wlj \*J ^ Ij^tfO I*1 ^ I I^ ijb fcJ j*

j5 J * $ J J J C u v Lu ( v ^ n u®)

ci j.h j$J* 3J O

(J 3 ^ J

jjl-^JU-9J

tii.E>.jp5 \ yij

C.9^^ J-5 ^^LA-W-TS

dlffl J \ jjJ <Lt L--1

(3 ^ dj1—»q

dJLu. L5^*Id CjLr?^£> (3 ^-

l^JL_L<vC^ j

*t* • u-* * y * *^ *^l^JL*g>l * ^ \ • ^

* Cuv i dJ*Ay j J *-]-*** ^
J- • ^
 (J \ J* ) £> <> jj- 3 J -5 j 3 j 3->

<3€ jL^>- ^ ^ *>r 3 J~^ (J$I-*^

{^ ) + i |- .. +

cU->> £>J3 JJ> ,j

c \3s'
3'

ij—*^ <3\

, (V )
^—'L-»0 l3w«5 d) Cvwj 1ft Lu J jj j O iuv Lu 3 ^*^

d t) j) J C L'^Ij \ j) C 1 tdJL \ 1 j3 1 3 I j W-t.-'- 3 d \5 3 sJ <J^~*1

9 ^ U— »j5"\ ^jj j ^ j j),v^wwa

Ja^vj 3 ^ 3^ dlSjjJ * ^UX> 1 i c5 ^ ^3 d\!)j)wJj) i Ja^v 3 j W*-2“ 3 J 3^

(r)
3 J j^i A.Jjp j j c ^>XLi\ 1j3 \ 3 £ (J I—oJb 3 L'JUl ^ 3 \„<s»j 3 iA>^ j j

(i)
j ”* 3 J C$i ^ (,J ^ 1^ tJ ^ \ i— j ci> \— 0 \5— <s 3 '{ S 3 ^ j

CU.x.v ^iO .S "^r— 0 3^,ui J .iA il— ^<1 di c -*j3 \ j 3 3 i™*„.svt* 3 1 .& kJ *n

O \ «Y * CUau \ & vA^CV O JjJ* JN J ^ # *|* * * * -J- * j

♦ ^ ^^ 1 [^ fjlib^ ^ 1 ^
 index

technical terms
abusalik: see busalxk

adraj 433, 4 3 3, 437, 433, 441

ala ^67, 368, 403, 405

anf J68, 369

agabi* 345, 346

awaz 38n, 95, 131, 141-6, 148, 190, 217, 244-6, 248, 251,

253, 260, 262, 263, 265, 283, 319, 410, 411, 455-7,

463-5

B 65, 151-4, 156, 157, 158n, 183, 186, 191, 200, 215, 2 5 0 ,

572-4, 376-82, 385, 413, 458

bafcr 400-2

bamm 371, 405, 406, 429, 430

baqiyya 109, 372, 377

basta 103

basta (nigar) 103

 ( 7)

bayfla* 114n

bingir 49, 51, 405, 406, 409, 410, 412, 429, 430

~bu‘a 71, 75n, 365, 366, 371, 374-80, 382-5, 392, 455, 457,

463, 467, 469

bu4d alladhi bi 9l-arba4, b u 4d dhi al-arba4 71, 75n, 3 7 2 - 9 ,

381, 382, 392, 402, 455, 456, 458-61, 467, 469

bu4d alladhi bi ?l~khams, b u 4d dhi al-khams 71, 371, 373-6,

379, 381, 455-7, 459, 465, 467, 469

b u 4d alladhi bi *l-kull, b u 4d dhi al-kull 371, 373, 574, 376,

381, 382, 384, 402

b u 4d alladhi bi vl-kull wa 91-arba4 , b u 4d dhi al-kull wa ?1-

arba4 , al-kull wa *l-arba4 573-6

bu4d alladhi bi ?l-kull wa *l-khams, bu 4d dhi al-kull wa 71-

khams, al-kull wa 9l-khams 373-6

bu4d alladhi bi ?l-kull marratayn, b u 4d dhi al-kull marratayn

373-6, 406

b u 4d B: see B

bu4d baqiyya: see baqiyya

b u4d J : see J

bu4d J: see J

ab4ad kubra 200, 374

ab4ad gughra 200, 374, 375

ab4ad wusfra 200, 374

 (474)

bus all k 81, 86, 1 0 7 , 111, 122, 128, 132, 135, 1 5 3 , 166, 173,

175n, 177, 2 0 7 , 210, 214, 219, 225, 227, 238, 239,

241, 248-50, 2 7 0 , 278, 280, 290, 291, 293, 310, 311,

407, 409, 413, 414, 419, 446, 457, 458, 46?, 469

buzurg 59, 89-92, 113-5, 127, 128, 133-6, 139, 143, 146,

148, 159, 163, 166, 167, 181, 188-92, 194, 195, 199,

201-3, 210, 211, 213, 214, 219, 225, 236, 240, 241,

244, 245, 250, 259, 272, 274, 280, 305, 308, 310, 407,

410, 413, 414, 424, 446, 455, 458, 468, 470

buzurg-i agl 89

buzurg-i kamil 127, 219, 227, 236, 239, 241

chahargah 100, 106, 2 5 2 , 256-9, 261, 276, 280, 283-5, 299-

301, 306, 307, 310, 459, 461-4, 470

chahargah wa foijazi 106, 218, 269, 300

c hah argali wa igfahan 106, 218, 269? 300

 Wb;

d a ?i r a : see dawr

flarb 434, 439, 440, 444, 448, 450

darb al-agl 317, 318, 321, 356, 357, 435, 437, 438, 440-3

dasban 47n, 59n, 366, 367, 370, 404, 405

(see also r a ?s)

dawr, daira 71, 262, 316, 365, 366, 379, 383-92, 400-11, 413,

415, 417-29, 431, 433-45, 451, 4 5 5 , 457, 458, 465

dhu al-arba4 : see bu4d

dhu al-khanis : see bu4d

dhu al-kull; see bu4d

dhu al-kull wa ?1-arba4 : see bu4d

dhu al-kull wa *l-khams: see bu4d

dhu al-kull marratayn: see bu4d

dugah 64-6, 100, 2 5 2 , 254-6, 258, 259, 261, 263, 272, 274,

276, 278, 280, 283-6, 291, 297-9, 301, 3 0 2 , 3 0 5 -8 ,

310-2, 344, 459-63, 468-70

dugah-i rash 254, 460, 461

dugah wa fcijazi 280, 3 0 5 , 470

dugah wa rahawi 280, 470

 f

fakhiti 318, 444

farafa. 114n

fagila 316, 452-6, 440, 448

furs: see wusfca

gushe 182n

fradd 405, 406, 429, 430

hazaj 318, 434, 443

fridda 366, 36?, 370, 371, 406, 460, 461, 467, 469

frijazi 84, 85, 90, 92, 98, 100, 102, 104, 106, 108-10, 115,

116, 120n, 121, 127, 128, 130, 133-8, 140n, 145-7, 166,

167, 173, 174, 177, 178n, 194-9, 202-5, 207, 208, 211,

214, 215n, 219, 220, 2 2 7 , 233, 240, 244, 245, 251-3,

2 7 0 , 2 7 2 , 276, 280, 298, 299, 305, 309n, 310, 311, 314,

 WV)

34-5, 407-11, 428, 446, 449, 455, 456, 459, 468, 470

td-sar 96, 97, 117, 118, 187, 201, 239, 240, 248, 251, 272,

283, 310, 311n, 349, 353, 354-, 457, 4-68

frigar wa igfahanak 118, 240

humayun 280, 281, 304, 307, 309n, 470

frusaynl 57, 88, 89, 101, 102, 1 0 5 , 110, 111, 119, 1 2 5 , 133-8,

142, 147, 167, 177, 211, 214, 218, 225, 226, 237,

240 , 24-8 , 251, 254 , 255 , 268*-70 , 272 , 276 , 278 , 280,

283, 290-3, 298, 302, 310, 311, 352, 354, 407, 4-08,

410-4 , 427 , 446 , 4-57 , 460 , 467-9 , 470

ikhtilas 463, 464

Intagal, intiqal 97n, 379 , 4-03 , 408 , 429 , 430 , 459 , 460,

462-7, 4-70

lqa1 75n, 365, 366, 430, 431, 439, 445, 4-60, 462

1iraq 79, 83, 92, 98, 102-4, 108-10, 116, 118, 121, 126, 132,

134, 135, 166, 174, 178n, 196-9, 202-5, 208, 209, 211,

213n, 214, 219, 2 2 7 , 234, 240, 250, 251, 256-61, 274,

278, 280, 2 9 2 , 293, 299, 303, 306, 308, 310, 314, 346,

407, 409, 4-10, 413, 414, 421, 4-46, 458, 461, 4-63, 468-70

1iraq m a ‘ al-baqiyya 109

 K'+foJ

Isfahan 80, 83, 84-, 87-9, 106, 114, 120 123, 124, 132-5, 145,

155, 167, 174, 1 7 5 n, 1 7 6 , 180n, 181-6, 1 9 6 , 2 1 1 , 214,

215, 219, 2 2 1 , 224, 225, 229n, 234, 240, 244-6, 248-

250, 252-5, 263, 2 7 0 , 2 7 2 , 274, 276-8, 2 8 3 , 290, 2 9 1 ,

298-300, 302, 3 0 6 , 3 0 8 n, 309n, 310, 313n, 349, 551,

555, 554, 407, 408, 410, 411, 422, 446, 455-60, 4-68,

469

igfahan-i agl 5 9 , 87, 24-4, 245, 252, 253, 455, 456, 459

igfahanak 103, 104, 118, 142, 218-20, 239, 244-9, 256, 2 5 7 ,

260, 261, 2 7 6 , 2 ? 7 ,283, 299, 3 0 6 n, 309n, 344, 345,

4-55, 457, 4-61, 4-63, 469

igfrilchab 365, 366, 403, 4-29, 4-30

ittafaq, ittifaq 371-3

J 58, 65, 6 6 , 151-4, 156, 157, 158n, 183, 186, 191, 200, 215,

250, 372-4, 376-82, 385, 392, 413, 458

.jahr 33 4

,jam‘ 75n, 244-6 , 260 , 264-, 268 , 455 , 456 , 458 , 463 , 465,

467, 469

,jumu‘-i murakkaba 268, 467

janfaza 111n
lass 3 7 1 - 3 , 3 7 5 , 3 7 7 , 3 7 9 , 403, 415, 429, 430
j
l
jins 59n -

ajnas mufrada 1 3 6 , 190

kamil fi al-tala*um, 1 5 0 , 3 8 6

aniya 97-9, 111-3, 115, 141, 143, 166n, 167, 1 7 6 , 180n,

181, 183-5, 211, 214, 218, 219, 221-4, 237, 240,

244, 245, 248, 249, 263, 411, 412, 455, 457

nirlzi 113, 1 2 2 , 219, 2 2 1 , 240

kardaniya za?id 1 1 3 , 114

kawasht 103, 118, 141-3, 148, 1 5 9 , 167, 171, 181, 187, 2 1 0 ,

211, 213, 214, 219, 240, 245, 324, 344, 411, 412, 450

khafi al-tanafur 155, 1 5 8 , 384, 391

khaffif 333

khafif al-ramal 318, 434, 442

khafif al-thaqil 3 1 8 , 434, 438, 439

khingir 51, 405

khufut 334

(abtad) kubr a : see bu 4 d

kuchek 77, 104, 2 5 6 , 2 7 0 , 2 7 2 , 274, 2 7 6 , 2 7 8 , 280, 291, 293,

299, 310-3, 461, 468-70
 V*tou j

kuchek: see also zirafkand

kuchek-i tamam 104, 218, 220, 239, 251, 291, 347n

kull al-durub 319

kull al-oaagham 319

kull wa 3l-arbat : see bu* d

kull wa kl-khanis : see b u 4&

la* am 377

latin
T HI nfrt—
368, 411, 461, 467 *

labnl 371, 374, 378, 379, 382

m add 334

mafruda 75n, 254, 268, 460, 465, 467 (see also thaqil)

mahur 45n, 114, 223n

mahuri 45, 114n

maqlis afruz 182

magam 125, 127, 182n, 209n, 217n, 247, 268, 269, 280, 467,
470
markaz 59n, 268, 402, 413, 465, 467“8

mathiath 371, 405, 406, 409, 410, 412, 429, 430

mathna 405, 406, 409, 410, 412, 429, 430

mawanlb 345n

ma.ja 91n, 92-4, 130, 141, 142, 143n, 201n, 213, 217, 220, 239,
252, 253, 262, 264, 265, 272, 276, 278, 282, 285, 288,
289, 291, 299, 303, 306, 308, 310, 411, 412, 459, 464,
465, 468, 469

mazmum 8On

midrab 448

mubarqa* 252, 260-2, 285, 286, 344, 345, 459, 465, 464

mudrag 434

mufakhkham 334

mufrad: see jins

mujiayyir jiusayni 44n, 82, 119, 120, 139, 146, 147, 167, 173,
175-7, 179, 211, 214, 217, 226, 240, 244, 245, 248, 249,
283, 301, 332, 333, 350, 351, 455, 457

mufrayyir zlrkesh 120, 184, 185, 219, 225, 240, 301

muj annab 51, 52, 53n, 64, 328, 345, 346, 405, 406, 410, 412,
429, 430, 451

rnuq annab al- sabb aba, 51, 52

mu jann ab al -wu sjla 51

mujannab wus5a zalzal 51

mukhalif 77^

mukhalif-i rast 87 , 89

mukhalifak 77

mukhammas 4-39

rnukhtalas 334*

mula’im 154, 155, 366, 377, 379, 384*, 385, 391, 4-63
murakkab 131, 14-4-, 145, 196, 281, 382, 4*10, 4*11, 4*55, 4*59,
4*67 (see also jam* )

mursal 4-41

mushaddad 334

mushfr 368, 369

mutabaddil 382, 392

mutala* im 156, 386

mutanafir 71, 154, 371, 377, 382-4, 392, 408

mutlag 51, 329, 346, 347, 366, 36?, 370, 403, 405, 406, 409,
410, 412, 429, 4-30, 451

muttafiq 371

naghma, nagham 75n, 97n, 133n, 365-8, 370— -5, 377“85, 391,

403, 405, 413, 415, 429, 430, 446, 448, 449, 455-67
nagra 333, 431“41, 443, 444, 448, 449

nawa 81, 86, 107, 132, 135, 166, 173, 177, 178, 207, 210,
214, 238, 239, 249, 250, 310, 312, 407, 409, 413, 414,
418, 446, 458

nawba 319

nawruz 64-6, 82, 84, 88, 104, 105, 110, 111, 117, 119-21, 124,
125, 130, 133-5, 137, 141-4, 196, 205, 207, 211, 213,
220, 239, 244-7, 249, 250, 254, 255, 258, 259, 261,
268-70, 274, 276, 278, 280, 283, 290, 291, 293, 298,
302, 306, 309n, 310, 311, 321, 343, 346, 34?n, 411,
412, 446, 448, 455-8, 460, 463, 46?-70

n awru z-i t am am 10 5

nihawand 114n, 116

ni haw and $ag hi r 11611

nihuft (al-fri,jazl) 109? 120, 121, 126, 127, 11-6, 11-7, 167,
175, 176, 177, 179, 181, 182, 195, 198, 211, 214, 227,
240, 241, 301

nihuft-i kamil 126, 127, 219, 227, 241

nirizi 59, 98, 122, 195

niriz kabir 122

nlsba 365-7 , 371-5 , 377 , 379 , 385-92 , 403, 1-06, 430 , 460-2
467
 V-TUT/

pan.isah 60, 100, 252, 256-9, 261, 262, 283-8, 459, 462-4

parda 75n, 97n, 217, 218, 244, 246-52, 254, 258, 260, 262-5,

270, 280, 283, 284, 296n, 319, 455-70

proslanibanomenos 255

qawl 332

qlsm 71, 251, 368, 369, 379-82, 384, 3 8 6 - 9 2 , 400-2, 4 5 8 , 459

rag 62

rahawi 80, 83, 84, 101, 102, 135-6, 138, 146, 157-9, 162, 166,

177, 178, 188, 196, 197, 203, 206, 208n, 210, 214, 220,

227, 244, 245, 247, 250, 251, 268-70, 2?4, 276, 278,

280, 291, 298-300, 3C2-5, 3 O811, 309n, 310, 511, 313,

407, 408, 410, 411, 413, 414, 426, 446, 455, ^56, 458,

467-70

rahawl-yi tamam 101, 218, 220, 239, 248, 249, 251, 303, 457
rakbi 111n, 260, 261, 263, 2 7 2 , 274, 285, 286, 298, 3 0 1 , 3 0 2 ,

306, 307, 310, 344, 463, 468

ramal 318, 321, 323-5, 32?, 328, 346, 434, 441-3, 448, 450,

451

r a ’s al-dasatln 51

rast 57, 59-61, 63, 64, 82, 84, 87,89, 97, 99, 100, 108, 112,

113, 116, 119-22, 132, 135, 142n, 144, 166, 174, 178,

186, 196, 205, 207, 209, 211, 214, 222, 225, 240, 244-6,

250-63, 276, 285-5, 297, 306, 308, 310-2, 314, 343, 407,

409, 411, 413-6, 420, 429, 430, 446, 449, 455, 456,

458-64, 469

ru-yi ‘iraq 83, 2 5 1 , 2 5 2 , 260, 261, 265, 282n, 283-5, 31-9,

350, 354, 459, 463

sabab 316, 432-5, 438, 440, 441, 448

sabbaba 51, 52, 405, 406, 409, 410, 412, 429, 430

salmak 128, 130, 141, 142, 213, 262, 263, 266, 267, 282,

285-7, 304, 34-5, 411, 412, 464, 466

sama ' 1 24n

gawt 319, 321, 343, 367, 408, 439, 447-50

 (,48b j

segah 60, 100, 101, 252, 256-61, 274, 276, 283-5, 299-301,

3 0 6 , 3 0 7 , 309n, 3 1 0 , 459, 461-3, 468, 469

segah wa b.i,jazi 1 0 0 , 1 0 1 , 218, 2 6 9 , 500

segah wa frusayni 1 0 1 , 218, 2 6 9 , 300

shadd 3 8 n, 131-3, 138-42, 144-9, 172, 189, 190, 192, 193,

203, 217, 247, 251, 263, 269, 316, 407, 446, 44?

shahnaz 94, 95, 130, 141, 142, 213, 2 1 7 ,221, 239, 2 5 2 , 253,

262, 264, 265, 272, 282, 285, 288, 306, 51On, 311n,

411, 412, 459, 464, 465, 468

shu‘ba 63, 1 0 0 , 175n, 186, 217, 218, 222, 228, 242, 244, 252,

2 5 3 , 2 5 8 , 259, 261, 2 6 3 - 5 , 268, 2 7 0 , 280, 282-7, 289,

296n, 306, 3 1 2 , 315, 319, 344, 345, 455, 459, 463,

465, 1-67 , 468 , 470

(ab‘ad) gughra: see bu 1 a

sunbula 187

systema teleion 25, 255

T 58, 6 5 , 6 6 , 1 5 1 , 1 5 3 , 2 0 0 , 2 5 0 , 372-4, 376-82, 413, 4 5 8

frabaqa 42, 1 2 1 , 147, 249, 263, 268, 3 1 6 ,3 6 6 , 379, 381-4,

392, 401, 402, 408, 409, 411, 415, 417-28, 430, 457,
467
taladum: see kamil

talfrin 457, 465, 467

tanafar, tanafur 45, 1 5 0 , 366, 371, 377, 378, 3 8 0 , 3 9 2 , 408

(see also khafi, gahir)

franinl 3 7 2 , 375, 3 8 2 , 469

tar 4-7, 4-8

tarkib 145, 244, 246, 2 5 1 , 264, 265, 280, 281, 290, 296n,

455, 456, 465, 470

frarlqa„ 17 n, 319 ,3 2 1 , 324 , 328 , 329 , 343 , 346 , 439 ,447 , 448,

450, 451

t a ’thir 22, 133n, 203, 366, 446

tazyin 261

thabit 1 5 4 , 382, 385, 392

thaqll al-awwal 318, 329, 34-6, 4-34-, 4-36, 438-40, 451

thaqil al~mafru<g.at 255

thaqil al-ramal 318, 440 (see also ramal)

thaqil al~thani 317, 318, 434, 436-9

thi gal 366, 367,369, 371, 4-60, 461, 467, 4-69

tunbur baghdadl 17n

tunbur khurasani 42, 46-8

turki 234n
 u

*ud 26,41n, 48-50, 3 6 6 , 405, 417

1ushshaq 61, 63, 81, 86, 99, 1 0 7 ,112-4, 132, 135, 156, 159,

166, 173, 178, 2 0 7 , 210, 214, 224, 239, 248-50, 276,

278, 293, 299, 310, 407, 409, 413, 414, 417, 4-46,

457, 458, 469

‘uzzal 92, 100, 106, 110, 121, 125, 195, 244, 245, 252, 253,

2 7 2 , 310, 311, 455, 459, 468

waqf 334

watad 316, 432-5, 437, 443, 448

watar 365-8, 370-3, 405-5, 417, 429, 430, 448

wigal 123

wusfra 49

(ab* ad) wusfra: see bu* d

wusta al-furs 49n, 5 0 , 51, 405, 409, 410, 412, 429, 430

wusfra zalzal 43, 49-51, 52n, 53n, 59, 60, 64, 133n, 405, 4-09,

410, 412, 429, 430

 y

yakah 127n, 209n

sahir al-tanafur 71, 155, 158, 385, 391

za’id 50-2, 114n, 405, 409, 410, 412, 429, 430

zalzal: see wusfra

zankula 115, 116,. 133-5, 138, 146, 148, 159, 167, 175, 178,
179, 188, 189, 192-7, 2C8n,211, 213, 239, 244, 245,
250, 252, 253, 262, 263, 266, 267, 276, 280, 287,
289, 302-4 , 345, 349, 353, 354, 407, 410, 411, 413,
414, 425, 446, 455, 456, 459, 464, 466, 469, 470

zawll 79, 252, 260, 261, 265, 282n, 283, 284, 459, 463

zir 89, 190, 405, 406, 409, 410, 412, 429, 430

zlrafkand 77, 78,89, 94-6, 103, 117, 133-5, 139, 142-4, 146,
148, 159, 163, 167, 181, 187, 210, 211, 214, 240,
250, 251, 343, 346, 347, 407, 410, 411, 413, 414,
423, 446, 450, 458

zlrafkand-i buzurg; 190

zirafkand-i kuchek 77, 104, 190n, 220, 256, 257, 303n

zirkesh 89

zlrkesh frusayni 88, 120, 260, 261, 263, 306, 463

bibliography
 sources

al-Jafri?, fi frabaqat al-mughannln, in majmu*a rasa’il al-

Jafris, Cairo, 1906

kit ab al-qiyan, in rasa’il al-Jafri^, Cairo, 1964--5

al-kindi, risala fi khubr t a ’lif al-alfran, ed. R. Lachmann

and K* el-Hefni, Leipzig, 1951

risala fi al-lufrun wa *l-nagham, ed- Z. Yusuf,

Baghdad, 1965

risala fi ajza* khabariyya fi al-musiqa

mukhtagar al-musiqa fi t a ’lif al-nagham wa 5 an* at

al-* ud

al-risala al-kubra f 1 al-ta’lif

in m u ’allafat al-Kindi al-musiqiyya, ed- Z. Yusuf,

Baghdad, 1962

Ibn Qutayba, kitab al-ma*arif, Cairo, 1934-

Abu Bakr *Abdallah b. Abl al-Dunya, dhamm al-malahi, ed. J.

Robson in Tracts on listening to music, London, 193<

Abu Jalib al-MufagLgLal b. Salama, kitab al-malahi, in J. Robsoi

Ancient ^rabian instruments (Collection of Oriental

writers on music, iv), Glasgow, 1938

Abu Atimad Yaftya b. ‘All al-Kunajjim al-Nadim> risala f 1

al-rouslqa, BM. MS. Or. 2361, ff- 236v-238

Ibn ‘Abb Rabbihi, kitab al-*iqd al-farid, Cairo, 194-0-53

al-Farabi, kitab al-muslqg. al-kabir, Cairo, n.d.

al-Farabi, kitab ibg a* al-‘ulum, ed. A. Gonzalez Valencia in
Alfarabi catalogo de las cienoias, Madrid, 1932
al-Fias* udi, murut
j al-dhahab, ed. G. Barbier de Meynard, Paris,
1861-77
al-Igbahani, kitab al-aghani al-kabir, Cairo, 1927-
al-Khwarizmi, mafatifr al-culum, ed. van Vloten, Leyden, 1895
Ikhwan al-gafa’, rasa’11, Beyrout, 1957
Ibn Sina, kitab al-shifa9, BM. MS. Or. 11190
risala fx al-musiqa, in m ac
jmuca r as a* iI al - sh aykh
al-ra*is Ibn Sina , Hyderab ad, 1935
Abu Kanpur al-gusayn b. Zayla, kitab al-kafj fi al-musiqa,
ed. Z. Yusuf, Cairo, 1964-
Abu al-Qasim al-Raghib al-Igfahani, mulyadarat al-udaba* ,
Beyrout, 1961
Majd al-DIn al-Ghazali, bawariq al-ilma* , ed. J. Robson in
Tracts on listening to music, London, 1938
Falchr al-D'in al-Razi, t
jami * al - *u lum, BM. MS. Or. 2972
Kayketus, qabus narna, ed. R. Levy, London, 1954
Hagir al-Din al-Jusi, risala fi *ilm al-musiqa, ed. Z. Yusuf,
Cairo, 1964-
gafi al-Din al-Urmawi al-Baghdadi, kitab al-adwar
al-risala al-sharafiyya fi al-nisab al-ta*lifiyya,
Bodleian MS. Marsh 521
Qu£b al-Din al-Shirazi, durrat al-taj
Shams al-Din Muhammad b, Mahmud Amuli, nafa3is al-funun,
Teheran, 1377-9h
Mu Uamm ad b . ‘All al-Kha^ib al -1rb i11 , qawahir al-nigam xi
 m a4rifat al-angham, ed. Cheikho, al-mashriq, 1913
al-Nuwayrl, nihayat al-arab, Cairo, 1923-
Abu 4Abdallah Muhammad al-gaydawi al-Dhahabi, kitab fi
ma4rif at al-angham, Bodleian MS. Marsh 82
anon., kanz al-tubaf, BM. MS. Or. 2361
Muhammad b. 4Isab. Karra al-Baghdadi, ghayat al-matlub fi
4ilm al-angham wa 9l-flurub, BM. MS. Or. 924*7
Lisan al-Din b. al-Khapib, risala f1 al-fraba*i4 wa *1-tabu4
wa ?l-ugul, ed. H. G. Farmer in Collection of
Oriental writers on music, i, Glasgow, 1933
anon. , sharb mawlana mubarak shall bar adwar, BM. MS. Or.
2361, ff. 68v-153
anon., sharb al-adwar, BM. MS. Or* 2361, ff. 33v-68
Ibn al-gabbab al-Dhahabi, kitab fi 4ilm al-musiqa wa ma4rifat
al-angharn, Bodleian MS. Ouseley 102
anon., kitab al-rnizan fi 4ilm al-adwar wa ?1-awzan,Bodleian
MS. Ouseley 102
4Abd al-Rabman al-babbak al-4Udi, urjuza fi 4ilm al-mmsiqa,
in Abmad Taymur, al-musiqa wa *l-ghina? 4ind al-
4ara.b, Cairo, 1963
anon. , al-musiqa ashraf al-4ulum wa alfraf al-funun,in Abmad
Taymur, al-musiqa wa *l-ghineB 4ind al-4arab,
Cairo, 1963
4.abd al-Qadir b. Ghaybl al-Maraghi, jarni4 al~alban, Bodleian
MS. Marsh 282
maqagid al-alban, ed. faqi Binish, Teheran, 1966
anon., untitled treatise dedicated to Mubammad b. Murad,
 BM. MS. Or. 2361, ff. 16Sv-219v
J ami, risala-yi inusiqa, Tashkent, 1960
Muhammad b. *Abd al~];lamid al-Ladhiqi, al-yisala al-fatfoiyya
fl al-musiqa, BM* MS. Or. 6629
Mu^affar b. al-Miusayn al-$agkafi al-Kusiqi, risalat al-
kashshaf fi *ilm al-angharn, Planches ter MS. 676
anon. , al-shat
jara dhat al-akmam, BM. MS. Or* 1535
Abu Zayd ‘Abd al-Ragman b. 4All al-Easi, al-qumu4 fi 4ilm
al-musiqa wa vl-fubu4 , ed. H. G. farmer in
Collection of Oriental writers on music, i , Glasgow,
1933
 other works

Abu 4Awf, A^imad Shafiq, aflwa* 4ala al-musiqa al-* arabiyya,

Cairo, n.d-
A1cel, L ., Studies in the social history of the Umayyad period
as revealed in the Kitab al-Aghani, Ph- D, thesis
(London), 1960
Allah Wirdi, Hikha’il, falsafat al-musiqa al-sharqiyya,
Damascus, 1948
al-4Aluchi, 4Abd al-£tamid, rad id al-nmsiqa al-4arabiyya,
Baghdad, 1964
Avenary, H., Abu al-£altvs treatise on music, musica disciplina.
vi, 1952
al-4Azzawi, 4Abbas, al-musxqa al-4iraqiyya fi 4ahd al-mughul
wa 91-1urkuman, Baghdad, 1951-
Barkeshll, Mehdi, La rnusique iranienne, Encyclopedic de la
Plfeiade: Histoire de la musique, i, Paris, 1960
Barkeshll, K. and Ma‘rufi, Kvisa, radix-i musiqa-yi Tran,
Teheran, 1963
Bart&k, B-, Die Volksmusik der Araber von Biskra und Urngebung,
Zeitschrift fiir husikwis.sensehaft, ii, 1920
Bel5aiev, V., The formation of folk mode systems, Journal of
the international folk music council, xv, 1963
Bern ex1, A. , Btudien zur arabischen Kusik aux Grund der
gegenwaxdigen Theorie und Praxis in ^gypten,
Leipzig, 1937
Carmi-Cohen, I)., An investigation into the tonal structure
of the maqamat, Journal of the international folk
music council, xvi, 1964
Caron, N., and Safvate, P., les traditions musicales: Irani,
faris, 1966
Ohilesotti, 0- Le scale arabico-persana e indu, Samraelbande
der internationalen Husikgesellschaft, iii, 1901-2
Chottin, A., Tableau de la musique marocaine, Paris, 1999?
Christensen, A. , Persian melody names of the Sassanid period.,
The Lastur Hoshang memorial volume, Bomb ay, 1918
Collaer, P., La migration du style mblismatique oriental vers
1 ’Occident, Journal of the international folk music
council, xvi, 1963
D ’Erlanger, R. , Lot musique arabe, Paris, 1930-59
Eisner, J., Zu Prinzipien arabischer Musizierpraxis, Jahrbuch
fur musikalische Volks- und Volkerkunde, iii, 1967
Parhat, H., The udastgahn concept in Persian classical music,
Ph. D. thesis (California), 1967
Parmer, H. G ., A history of Arabian music to the Xlllth century
London, 1929
The lute scale of Avicenna, Journal of the Royal
Asiatic Society, 1937
The music of Islam, hew Oxford history of music, i,
London, 1957
musi&r, Bncyc1opaedia of Islam, Leyden, 1913-38
The old Arabian melodic modes, Journal of the Royal
Asiatic Society, 1965
Farmer, H. G-., The old Persian musical modes, Journal of the
Royal Asiatic Society, 1926
Satadya Gaon on the influence of music, London,

194-3
The sources of Arabian music, Bearsden, 194-0
G-erson-Kiwi, TA , The Persian doctrine of dastga-composition
Tel xxviv, 1963
Husmann, H., Grundlagen der antiken und orientailschen
Kusikku11u r , Berlin, 1961
Idelsohn, A, Z., Die Magamen der arabischen Frusik, Sammel-
bande der internationalen Musikgesellschaft,
XV, 1913
Ka^im, A,, al-igtilatiat al-musiqiyya, tr. al-JJaquti, Baghdad,
1964-
Khatschi, K., Der Dastgah: Studien zur neuen persischen
Husik, Regensburg, 1962
Das Intervallbildungsprinzip des persischen
Dastgah Shur, Jahrbuch fur musikalische Volks-
und VoIkerkunde, iii, 1967
al-Khula‘i, Kamil, al-musiqa al-sharqiyya, Gairo, 1904-
Lachmann, R., Die Kusik in den tunesischen Studten, Archiv
TUr Kusikwissenschaft, v, 1923
Land, JA P. LA, Recherches sur l’histoire de la. gamme arabe,
Proceedings oT the 6th international congress of
orientalists, Leyden, 1883, vol. 11(1), 1834
TonschriTtversuche und kelodieproben aus dem
muhammedanischen Hittelalter, Viertel(
jahrsschr lit
Malrufi, M . : see Barkeshll
Oransay, G., Das Tonsystem der turkischen Kunstmusilc, Die
Rus ikf orsc hung , x, 1957
Die traditionelle turkische Kunstmusik, Ankara,
1962
al-Rajab, Biashim Muhammad, al-maqam a P iraqi, Baghdad, 1961
Rouanet, Jo, La musique arabe, Bncyclop^die de la musique
(Lavignac), 1re partie, v, Laris, 1922
Safvate, D.: see Garon
galab al-Din, Muhammad, miftaft. al-alpan al-4arabiyya, n.p.,
194-7?
Saygun, A . , La musique turque, Bncyclop§die de la Pl&iade:
Histoire de la musique, i, Laris, 1960
Schneider, M., Le verset 94- de la sourate VI du Goran btudie
en une version populaire et en trois nagamat de
tradition hispano-musulmane, Anuario musical,
ix, 1954
al-Sh awwa, Sami, al-qawaAid al-fanniyya fi al-musiqa al~
sharqiyya wa *1-gharbiyya, al-juz’ al-awwal,
Cairo, 1946?
Shiloah, A., Caract&ristiques de l*art vocal arabe au Moyen
Age, Lei Aviv, 1963
laymur, Ajpmad, al-musiqa wa *l-ghina9 4ind al-* arab, Cairo,
1963
Louze, Mo, Sur les modes musulmans, Revue musicale, annee 7?

i, 1925
Yekta, R a 5uf, Les modes orientaux, La revue musicale, vii,

1907; and viii, 1908

La musique turque, Bncy c 1 opedie de la musique

\ re
CLavignac), 1 partie, v, Laris, 1922

Lonis, E . , Contemporary art music in Persia, Musical