Meisam-Structure Meaning 2

STRUCTURE AND
MEANING IN MEDIEVAL
ARABIC AND
PERSIAN POETRY
ORIENT PEARLS
Julie Scott Meisami

First published in 2003
by RoutledgeCurzon
11 New Fetter Lane, London EC4P 4EE
Simultaneously published in the USA and Canada
by RoutledgeCurzon
29 West 35th Street, New York, NY 10001
This edition published in the Taylor & Francis e-Library, 2006.

“To purchase your own copy of this or any of Taylor & Francis or Routledge’s
collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.”
RoutledgeCurzon is an imprint of the Taylor & Francis Group
# 2003 Julie Scott Meisami
All rights reserved. No part of this book may be reprinted or
reproduced or utilized in any form or by any electronic,
mechanical, or other means, now known or hereafter invented,
including photocopying and recording, or in any
information storage or retrieval system, without permission in
writing from the publishers.
British Library Cataloguing in Publication Data
A catalogue record of this book is available from the British Library
Library of Congress Cataloging in Publication Data
A catalog record for this book has been requested
ISBN 0-203-22055-2 Master e-book ISBN
ISBN 0-203-27545-4 (Adobe e-Reader Format)

ISBN 0–7007–1575–4 (Print Edition)
This book is dedicated to the memories of two great teachers
and dear friends – Elroy J. (Roy) Bundy and Tawfiq Sayigh –
who taught me poetry, and who never grudged either
their time or their talk
CONTENTS
Preface ix
1 Introduction 1
Brief encounters 1
Grounds for comparison 5
Medieval literary theory 9
The search for unity 13
Poetry as craft 15
Theories of composition 19
2 Invention 23
Concepts of invention 23
Invention, imitation, and genre 26
Invention and genre in early Abbasid poetry 30
Invention and genre in the Persian ghazal 45
3 Disposition: The parts of the poem 55

Concepts of disposition 55
Beginnings 60
Transitions 75
Endings 90
4 Disposition: Larger structures 111

Segmentation: Proportion and balance 111
Amplification, abbreviation, and digression 130
The shape of the poem 138
vii
CONTENTS
5 Disposition: The Qası̄da and its adaptations 144

˙
Functions of the Nası̄b 144
)
Nası̄b and Hijā 155
Variations on the Qası̄da Form 162
˙
6 Disposition: Varieties of structure 190
Spatial and numerical composition 190
Other strategies: Letters, dialogues, narratives, debates 207
7 Ornamentation 244
Concepts of ornament 244
(
Ornament and structure: The five figures of badı̄ 246
Ornament and structure: The mahāsin al-kalām 269
˙
Ornament and structure: Other figures 282
8 Ornament: Metaphor and imagery 319

Problems of metaphor 319
Imagery and poetic unity 323
Image as argument 341
The poet and the natural world 347
Earthly Paradises? 355
Esoteric imagery 388
9 Conclusion: The coherence of the poem 404

Word and image: A privileging of styles? 404
Concluding Remarks 427
Notes 431
Bibliography 478
Index 502
viii
2
INVENTION
Arms and violent wars, with meter suited to matter,

Arms and violent wars, all in hexameters,
I was preparing to sound, when I heard a snicker from Cupid;
What had the rascal done, but taken one foot away?
Ovid, The Art of Love
If it is to be praise, the love-song always precedes it;
Is every eloquent voicer of poems a distracted lover?
Abū al-Tayyib al-Mutanabbı̄
˙
Concepts of invention
Invention is the art of finding the material appropriate to the work in hand. For
the classical orator or writer, this meant the arguments necessary to judicial
cases or legislative initiatives and the topics of epideictic compositions; it
comprised (according to Aristotle) three branches: proofs, topics, and
commonplaces. “Natural” proofs are derived from the facts of the case being
argued; “artificial” proofs might be based on character (ethos), especially that of
the speaker, on the effect produced in the audience (pathos), or on logic,
including the use of syllogisms, enthymemes, examples, and maxims. Topics
(topoi) are “ways both of conducting an argument and of analysing a theme or
subject prior to discussing it;” commonplaces (koinai topoi), originally “those
topics of argument . . . common to different subject areas” (for example, “that we
cannot judge the merits of an action until we have scrutinized the motives of
the agent”), came to include “any observation or truth which is pithily
expressed, weighty and serviceable: time flies; death is common to all,” as well
as “matters of perennial interest which might be proposed as subjects for debate
or taken as themes for oratorical or poetic variations: the mutability of things;
the contemplative versus the active life” (Dixon 1971: 24–8).1
Medieval artes poeticae extended invention to encompass the matter of
narrative works, including under its rubric historical (or fictional) subject-
matter as a class of topics from which arguments, proofs, commonplaces and so
on are abstracted (see Kelly 1978b). In the most general terms, invention
involves the choice of genre, subject matter, material and topics, and their
adaptation to the occasion and to the poet’s intent, according to the “mental
archetype” he has of his poem (cf. ibid.: 233). Central to this process is the
poet’s discovery of the meaning inherent in his material – whether that material
23
INVENTION
is “historical”, as in narrative poems such as epic and romance, or generic, as in

lyric – and of the means of conveying that meaning to his audience. The
common factor which links both types of material is their derivation from a
received source or sources or from prior treatments by others.
While Arabic and Persian rhetorical manuals lack both explicit discussions of
invention and terms which would correspond to the Greek heuresis or the Latin
inventio, there clearly( existed both a parallel conception and a parallel discipline: (
the science of ma ānı̄, of the “ideas” of poetry (to be distinguished from the ma ānı̄
al-nahw discussed ( in later works, especially following al-Sakkākı̄ (d. 626/1229),
˙(
when ilm al-ma ānı̄ is placed firmly in the domain ( of grammar), represented both
in rhetorical works ( and in collections of
( ma ānı̄ or poetic topics (see Sadan 1991).
The terms ma nā and its plural ma ānı̄ are used by a variety of disciplines and (
in a variety of senses.2 Its later restriction, within the science of rhetoric ( ilm
al-balāgha), to matters of syntax has obscured the existence of a secondary
discipline (which, Sadan notes, is peripheral to the system of balāgha itself; ibid.:
60–1, 67) concerned( with the invention of topics. The widespread practice of
translating ( ma nā ( “idea” or “meaning” has further obscured the parallel
as
between ilm al-ma ānı̄ ( and invention (cf. ibid.: ( 62–3).
Collections
( of ma ānı̄ such as Abū Hilāl al- Askarı̄’s (d. after 395/1005) Dı̄wān
al-ma ānı̄ functioned in part as sources of invention, providing examples of the
expression of a variety of topics (in prose as well as in verse) arranged under
generic headings (praise, satire, description and so on) and specific subjects
(love, wine). Far from being mere catalogues of tropes (as is sometimes assumed),
these collections point to the importance of finding the right expression for
conveying meaning, to the necessity of expanding and developing topics rather
than merely repeating them. I shall have more to say about what they tell us in
this respect below.3
The invention of new topics and the expansion of old ones is central to both
Arabic
( and Persian poetry; and the critics are well aware of this. When
al- Askarı̄ cites al-Buhturı̄’s preference for the poet al-Farazdaq (d. 110/728)
˙
over( Jarı̄r (d. 111/729) because the former “uses his judgement to vary certain
ma ānı̄ as Jarı̄r does not, and employs them in his verse in each qası̄da differently
˙
than in the next,” whereas Jarı̄r merely repeats the words of others (1986: 24),
he signals the (importance of expanding the meaning of a topic or an image;
when he notes Abd al-Hamı̄d al-Kātib’s (d. 132/750) adaptation from Persian of
˙
topics related to the secretarial art he commends the invention of new topics for
new subject matter (ibid.: 69). In his chapter on husn al-akhdh (“excellence in
˙
borrowing”) he states explicitly that no writers
(
can dispense with treating the ma ānı̄ of their precursors and pouring
(their discourse) into the moulds of those who came before; but they
must, if they borrow them, clothe them in words of their own, present
them in forms of their own composition, and introduce them in other
than their original garb. . . . Individuals are distinguished from one another
24
INVENTION
by the words they use, how they put them together, combine them and
order them. (ibid.: 196)
The concept of invention is also seen in discussions of takhyı̄l, which means
both to create and to induce an imaginative (re)presentation (not, it should be
stressed, a representation in a mimetic sense, but a particular form of expression
designed to produce assent).4 Cantarino defines takhyı̄l as “imaginative
creativity,” that is, “the mental process by which the poet can cause his
mimetic representations [sic] to be imaginative, effective, and creative” (1975:
80–1); in fact, what characterizes such “representations” – and specifically, those
expressed metaphorically – is their basis in imaginative premises and their
conveyance through eloquent language.5 The distinction between logical (or
authoritative) discourse, which appeals to the ( reason, and poetic discourse,
which appeals to the imagination, ( underlies
( Abd al-Qāhir al-Jurjānı̄’s (d. 471
or 4/1078 or 81) division of ma ānı̄ into aqliyya (“intellectual”) and takhyı̄liyya
(“imaginative”) (1954: 241–8); both, however, operate with words, the former
to convince, the latter to create an imaginative impression.6 The notion of
imagination as one of the essentials of poetry must therefore be seen as relating
to the (specific way in which words operate on the mind to convey the “idea”,
the ma nā. In this operation, it is not the object represented which is of primary
importance but its significance; the process of takhyı̄l explores, clarifies, and
represents this significance through finding the best means through which to
create an imaginative impression (cf. Cantarino 1975: 78–9).7
In this sense takhyı̄l encompasses all the means of poetry: words, images,
figures of speech, rhyme, metre, and, of course, structure. This wider use of
the term characterizes its discussion by Hāzim al-Qartājannı̄, who classifies the
˙ ˙
poet’s procedure “in considering the imaginative forms [takhyı̄lāt] in which
he wishes to present his praise or blame, and the conceits [takhayyulāt] which he
wishes to assist in this presentation” into eight stages, “each of which has a place
in the practice of composition” (1981: 109). First, “he conceives [or invents
(yatakhayyalu) the general aims of the genre [gharad] which . . . he wishes to
˙
present.” (Hāzim uses gharad variously: as “genre”, with various subdivisions
˙ ˙
[ibid.: 12]; as the “goal” or “motive”
( which produces a genre [11]; as the “human
motives” which generate ma ānı̄ (18). Second, “He conceives for these genres a
mode [tarı̄qa] and a style [uslūb] or (styles, either similar or contrasting, whose
˙
path he follows with his topics [ma ānı̄].” (The tarı̄qa, “mode”, may be serious,
˙
jidd, or jesting, hazl [ibid.: 327]; “style” [uslūb] is of three types, defined as
“delicacy, roughness, and an intermediate style” [riqqa, khushūna, and wasat],
˙
which may be used alone or in combination [ibid.: 109].) Third, he determines
the order of the topics presented, in particular noting places for transition
(takhallus) and digression (istitrād; these terms will be discussed in Chapter 3
˙ ˙
below). Fourth, “he conceives the material form [tashakkul] of these topics and
their accomplishment in the mind in appropriate phrases,” selects a suitable
rhyme, and determines the opening line of his poem; “he may also consider the
25
INVENTION
place of transitions and digressions at this stage.” “These are the four stages of
general inventions [al-takhāyı̄l al-kulliyya].” )
In the fifth stage (the first of “particular( inventions”, al-takhāyı̄l al-juz iyya),
“the poet begins to conceive his topics [ma ānı̄], one by one, in accordance with
the poem’s genre.” Sixth, “He conceives what will ornament [each] topic and
complete it . . . by conceiving matters which relate to that topic with respect to
their proper positioning and the correspondences and relationships which exist
between the various parts of the topic and with things external to it which are
yet connected with it, to aid him in conveying the intended meaning.” The
seventh stage involves precise metrical considerations based on the effect he
wishes to produce in his audience, the eighth the invention of additional topics,
if needed, for the purpose of filling metrical gaps or completing the rhyme. “The
poem is generated on the basis of this type of progression” (ibid.: 109–11).
Geoffrey of Vinsauf defined invention as comprising the “careful planning [of
the work] before taking up the pen to write,” the “subordination of subsequent
disposition and ornamentation to the plan of the invented materia,” and the
overall ordering of the work with respect to its beginning, middle, and
conclusion (Kelly 1969: 119; see Geoffrey of Vinsauf 1967: 16–18). While
Hāzim (as befits his Aristotelianism) is more systematic than Geoffrey, their
˙
conceptions of how a poem comes into being are essentially the same. The first
stages of this creative process are the most crucial: determining the
appropriateness of the material to the genre and purpose of the poem. It is
to the question of generic invention, and its implications for poetic structure, to
which I shall now turn.
Invention, imitation, and genre

Poetic invention is closely linked to the issue of the relation of the poet and his
poem to the tradition, to prior treatments of the same material. Discussions of
this relationship have often expressed it in terms of “influence”, “borrowing”, or
“copying”, or more generally as “imitation”, terms used largely in a derogatory
sense as implying lack of “originality”; alternatively, stress is placed on the poets’
“individual achievement”, and their relationship to other poets, and other
poems, is not considered a valid tool of interpretation. (See, e.g., Arberry 1964:
352–4, and the criticisms by Rehder, 1974: 59; see also Meisami 1987: 308–9.)
But the Arabic or Persian poet was intimately connected with a poetic world
created, and ever in the process of creation, by himself and his predecessors, a
connection seen in all aspects of composition: invention, disposition, and
ornamentation. Other poets, other poems, play a formative role in the choice
both of the poem’s “mental archetype” and of the materials appropriate to it; on
the level of invention, such choices revolve primarily around questions of genre.
Modern criticism has spent a great deal of effort discussing genre, and in
attempting to systematize a concept which resists systematization. The classical
Western division of poetry into dramatic, epic, and lyric is obviously of no value
26
INVENTION
for a system which, like the Arabo-Persian, is lyric-based; yet this system most
certainly possesses a concept of genre. But that, indeed, is what “genre” is: a
concept, a notion in the mind, of classifications, distinctions, decorum, within
which the poet works.
Indeed, genre is perhaps the most important of literary concepts, as it
provides a framework that the poet may not only write within, but write against.
In the pre-modern west genre was precisely such an informative concept, which
expressed itself in a variety of ways.8 That a strong sense of genre is
characteristic of pre-modern literatures has been shown for both classical and
medieval European literature (cf. Cairns 1972; Colie 1973; Fowler 1982: 31–2;
Jauss 1982: 76–109); that this is no less the case for Arabic and Persian is
demonstrated by the medieval rhetoricians.
There is no single term in Arabic or Persian which corresponds to “genre”.
The term most often used is gharad (generally in the plural aghrād; often
˙ ˙
translated as “theme”), ( which refers to the “ends” or “purposes” of poetry; other
terms include anwā , “kinds”, asālı̄b “types”, and so on. Scholars often lament
such terminological “confusion” (or flexibility), as they do the apparent absence
of any consistent system of generic classification (see e.g. Trabulsi 1955: 237–8;
van Gelder 1999 now provides a fresh look at questions of genre); but this
seeming lack of consistency reflects both a diversity of views and a vigorous and
dynamic concept (of genre.
In his Kitāb al- Umda Ibn Rashı̄q provides an example of this diversity. “Some
scholars,” he begins,) “have said . ). . that poetry is constructed upon four pillars
[arkān]: madh, hijā , nası̄b and rithā .” This division into panegyric, invective, love
˙
elegy and lamentation is the traditional one found – often with some
elaboration – in (most works on rhetoric and poetics. “Others say that [its] basic
principles [qawā id] . . . are four: (the provoking of) desire, fear, pleasure,( and
anger. Madh and (shukr [expression of thanks] are associated with desire, i tidhār
˙
[apology] and isti tāf [asking for sympathy] with fear,) shawq(( [passionate longing]
˙
and the( delicacy of the nası̄b with pleasure, and hijā , tawa ud [threatening] and
severe itāb [reproach] with anger.” This division stresses the expressive-affective
aspect, the emotions which generate certain genres and which they, in turn,
generate in the audience. ( Similarly, the poet Artāh ibn Suhayya, when asked by
˙
the Umayyad caliph Abd Allāh ibn Marwān, “Will you compose poetry today?”
replied, “By God, I am neither sad nor angry nor drinking nor desirous of
anything, and poetry
( is composed
( in one of these (states).”9
The exegete Alı̄ ibn Īsā al-Rummānı̄ (d. 384/994) ) identified the five “most
frequently used” poetic aghrād as nası̄b, madh(, hijā , fakhr (boasting), and wasf
˙ ˙ ˙
(description), and included tashbı̄h and ( isti āra (comparison and metaphor)
10
under wasf. Ibn Rashı̄q’s teacher Abd al-Karı̄m ibn Ibrāhı̄m al-Nahshalı̄
˙
provides a more detailed classification.
)
The types [asnāf ] of poetry are comprised under four headings: madı̄h, hijā ,
˙ ˙
hikma [wisdom] and lahw [pleasure]; each type [sinf ] is divided into funūn.
˙ ˙
27
INVENTION
)
Madı̄h includes elegies, boasting and thanks; hijā includes blame,
˙
reproach, and seeking delay; hikma includes proverbs, exhorting to
˙
asceticism, and admonition; and lahw includes love poetry, incitement
to pleasure, and descriptions of wine and the drunken.
)
Finally, others subsume all types of poetry under the bipartite division madh/hijā .
˙
Madh includes elegy, boasting, love-poetry [tashbı̄b], and related laudable
˙
descriptions such as that of the long-suffering and of great deeds, beautiful
comparisons, and admiration of virtues, such as proverbs, sententiae,)
sermons, renunciation
( of this world, and contentment. Hijā is the
opposite; but itāb [reproach]) occupies a middle ground, having one side
towards each, as does ighrā [incitement], which is neither praise nor
blame. (Ibn Rashı̄q 1972, 1: 120–3. For further discussions of generic
classifications see Heinrichs 1973: 38–43; Trabulsi 1955: 215–47; van
Gelder 1999; and compare Shams-i Qays 1909: 421, on the asālı̄b of poetry.
See also the entry, “Genre,” in Meisami and Starkey 1998 [J.S. Meisami])
Ibn Rashı̄q records the approaches to genre by critics in varying disciplines and
professions: rhetoric, grammar, exegesis, kalām (dialectic), and poetry itself.
None of these schemes is either systematic or exhaustive, though some attempt
to be, and all reflect the attitudes and assumptions of the disciplines to which
they are related. Some) accept a general division of discourse into praise and
blame (madh and hijā ) and marshal sub-genres under these headings; others
˙
stress the relation of the poem to specific emotions or states (a classification
recalling the Aristotelian notion of character, ) or ethos, as a source of
argument). The quadripartite
( madh : hijā :: h ikma : lahw classification and its
˙ ˙
subdivisions posited by Abd al-Karı̄m al-Nahshalı̄ is also found in the Kitāb
al-Burhān of Ishāq ibn Ibrāhı̄m ibn Wahb 11 Another
˙ ( (see Ibn Wahb 1933: 70). (
type of classification is grammatical: “ Abd al-Samad ibn al-Mu adhdhal said,
˙
‘All poetry is contained in three words: . . . When you praise, you say anta [you
are], when you write invective you say lasta [you are not], and when you write
(elegy you say kunta [you were]’” (Ibn Rashı̄q 1972, 1: 123; attributed to the poet
Amr ibn Nasr al-Qisafı̄ [d. 247/861] in Ibn al-Jarrāh’s al-Waraqa [van Gelder
˙ ˙
1982a: 90–1]).
What is notable about these classifications is, first, the absence of formal
criteria (cf. van Gelder 1999: 23; classifications by prosodic
( form, with little or
no reference to content, are found in discussions of arūd, prosody); and second,
˙
that whether the emphasis is on the affective (i.e. “praise/blame”) or the
expressive aspect of poetry, the result is a listing of genres according to content.
It is irrelevant to inquire, as does Trabulsi, how many genres “actually existed”
in Arabic poetry or to criticize the omission of “important genres such as
bacchic and gnomic poetry” from any given list (1955: 215, 219); such lists are,
by definition, idealized, partial, and often merely exemplary, rather than
attempting to be all-inclusive.12
28
INVENTION
The generic repertoire of Arabo-Persian poetry can ultimately be traced back

to the pre-Islamic qası̄da, which employed a variety of genres ( in combination.
˙
The practice of the rhetoricians, and of compilers of ma ānı̄ collections, is,
indeed, to trace specific ( generic topoi to their pre-Islamic origins. Thus for
example
( Abū Hilāl al- Askarı̄ introduces the chapter on madı̄h in his Dı̄wān
˙
al-ma ānı̄ (the first chapter of the book, as madı̄h is foremost among the aghrād)
˙ ˙
with a line by the pre-Islamic poet al-Nābigha al-Dhubyānı̄ which, he says, has
been called “the most encomiastic [amdah] line uttered by the Arabs” (i.e., the
˙
ancients):
Do you not see that God has granted you a high station before which
every (other) ruler falters?
For you are a sun, and (other) kings are stars; when you shine forth, not
one of their stars can be seen. (1994, 1: 19; Abū Hilāl gives a
lengthy explanation of the function of this line in the context of the
poem, from which he cites a number of other verses; see ibid.:
19–20)
Abu Hilāl cites earlier poets who preceded al-Nābigha in the use of this topic,
and later poets who adapted and developed it, among them Nusayb ibn Rabāh
˙ ˙
(d. ca. 108/726) –
He is the moon, and other men stars around him; do stars resemble the
light-giving moon? –
and Abū Tammām (d. ca. 232/845) –
As if the Banū Nabhān were, on the day of his death, stars in a sky
from whose midst the moon had fallen. (ibid.: 20)13
But in actual practice the genres, and their related topoi, appear even in( early
poetry
( not only in the qası̄da but in brief, monothematic poems (qita , sing.
˙ ˙
qit a), and in later periods they undergo extensive development in both
˙
contexts.
Genre plays a formative role in composition. A poem’s genre is related both
to its addressee and to its speaker (e.g., the poet may speak as panegyrist to
patron, as lover to beloved, etc.; both speaker and addressee are, moreover,
implied by the genre), as well as to its broader audience (the assembled court,
members of a drinking party, and so on). Certain types of communication are
presupposed by certain genres; and the act of communication itself implies
specific speakers and addressees in certain situations.14 Each content-oriented
genre, therefore, implies a specific type of speaker. Moreover, a poem may consist
of one genre or of several genres in combination; the distinguishing feature of a
genre (as opposed to a topic) is that it can constitute an independent poem.15
Genres may be combined or included; one genre may become a topic of
another, or a topic may be amplified until it takes on the status of a new,
independent genre.
29
INVENTION
Most important in terms ( of the actual process of composition is the generic

repertoire, the topics (ma ānı̄) and other elements which are characteristic of
each genre but are not necessarily genre-specific. It is these elements which play
the greatest part in the generic manipulation so typical of lyric poetry; and it is
for this reason that they figure so( prominently in Arabo-Persian rhetoric and
criticism and in collections of ma ānı̄. For lyric poetry relies extensively on the
manipulation of generic topics; such, indeed, is the point of criticisms of Abū
Tammām’s line quoted above on the grounds that his use of the “moon/stars”
topic implies, not praise, but blame: the Banū Nabhān are so inferior that they
only shine because of the death of their superior chieftain (cf. n. 13 below).
Generic manipulation is a marked feature of the Arabic poetry of the Abbasid
period, to which I will turn next.16
Invention and genre in early Abbasid poetry

The poets of the early Abbasid period (late second/eighth-early third/ninth
centuries) inherited a poetic tradition that already extended back several
centuries. Formally, this tradition was dominated by the (usually polythematic)
qası̄da, consisting of several sections, or “movements” (a term J. Montgomery
˙
[1986] suggests as preferable): an exordium (nası̄b), usually amatory or erotic in
nature, a “journey” passage (rahı̄l), followed by sections of boasting (fakhr),
˙
praise (madh), or gnomic topics (hikma).17 Generically, it possessed a vast range
˙ ˙
of well-established topics and motifs developed over the centuries and hallowed
by repeated use.
The early Abbasid period saw a sudden explosion of hitherto minor genres –
in particular those dealing with bacchic, erotic and gnomic topics – employed
in independent, generally monothematic poems which provided a counterpart
to the more formal qası̄da, now devoted chiefly to panegyric (see Badawi
˙
1980),18 and perhaps reflecting a desire to elevate such formerly minor genres,
formerly inserted into the larger framework of the qası̄da, to independent status
˙
(see ibid.: 7–12; Schoeler 1990: 294). Another was undoubtedly the
increasingly occasional nature of Abbasid poetry (or rather the increasing
diversity of occasions for which poetry was composed) and the multiplication of
informal gatherings in which it was performed;19 a third, related factor was the
association of poetry with music, and with singers and musicians (cf. Vadet
1968: 78–101; Bencheikh 1975a: 120).20 The independent wine poem
(khamriyya), love poem (ghazal), and ascetic poem (zuhdiyya) which were
developed at this time were later adapted to other uses (e.g., for religious or
mystical poetry), and provided models for the various types of the later Persian
ghazal.
The height of this period of experimentation was during the reigns of Hārūn
al-Rashı̄d (170–93/786–809) and his son and successor al-Amı̄n (193–8/809–13).
)
Following the civil war between al-Amı̄n and his half-brother al-Ma mūn,
which ended in the latter’s victory and al-Amı̄n’s murder, the atmosphere
30
INVENTION
changed, becoming less conducive to “light” poetry and favouring the official
qası̄da (cf. Bencheikh 1977: 35–8). Nevertheless, the new genres left a marked
˙
impact both on later Arabic and on Persian poetry; and they remained popular
at the courts of rulers who did not always share caliphal tastes, such as the
Hamdanids of Syria, the Fatimids in Cairo, and the Buyid “protectors” of
the caliphate.
In developing these independent poetic types poets drew upon both the
earlier Arabic poetry and on materials imported from the cultures which had
come under Arab hegemony, most notably the Persian. Their chief technique
was the adaptation and modification of traditional generic topics for use in these
new poetic contexts. ( The fact that the( generic constituents of such poems
(generally called qit a, in Persian muqatta a, i.e., “fragment”) can be traced back
˙ ˙˙
to the ancient qası̄da has prompted some scholars to view them as representing
˙
“a splitting up of the old qası̄da” (Heinrichs 1973: 25); however, as J. Stetkevych
˙
argues (reminding us that al-Jāhiz (d. 255/868–9) termed the short poem
˙˙
al-qası̄da al-qası̄ra), “The term qit‘a is etymologically misleading, and . . . lends
˙ ˙ ˙
itself to being erroneously viewed as part of the standard qası̄da as the latter
˙
appears in the definition of Ibn Qutayba,” itself a “formal abstraction” rather
21
than a description of actual practice (1967: 2–3). In this view, it is not so
much that “parts of the qası̄da” have been split off from it, but that specific
˙
aghrād which the qası̄da typically employs in combination are isolated and
˙ ˙
developed into independent poetic types.22
The dynamics of Abbasid generic manipulation may be observed in the
interplay of the three major types – khamriyya, zuhdiyya, and ghazal – and the
manner in which they draw both on the ancient repertoire and on one another
(compare Williams 1980: 233 on the exploitation of genre-differentiated views
of love). Central to this process is the modification of the persona of the poet/
speaker: for if the predominant mode of pre-Islamic and much early Islamic
poetry was heroic, Abbasid poetry is often deliberately antiheroic (see Hamori
1974: 3–77). In contrast( to the early portrayal of the poet as tribal hero (or, in
the case of the Su lūk or “brigand” poets, as anti-hero; see e.g. S. P. Stetkevych
˙
1984), the Abbasid poet is more often victim: of love, of fate, or of his own
impulses. The central focus of lyric – the intense presence of the speaker,
the lyric I – takes on a variety hitherto unknown. This corresponds to what
Quintilian terms the invention of fictiones personarum: “we create personae that
are suited to advise, blame, complain, praise, or pity. . . . For it is certainly true
that words cannot be invented without also inventing them for a particular
persona” (quoted by Williams 1980: 212). The most startling examples of this
transformation may be seen in the khamriyya, the chief Abbasid exponent of
which was Abū Nuwās (d. c. 198/813), with whom the description of wine and
its associated topics became a poetic end in itself.23 In his hands, the khamriyya
(like, mutatis mutandis, the love poem and the zuhdiyya) becomes a countergenre
which both draws upon and subverts or parodies the heroic mode of pre-Islamic
poetry (see EI2, art. “Khamriyya” [J.-E. Bencheikh]; Hamori 1974: 47–50; on
31
INVENTION
fakhr in pre-Islamic and early Islamic poetry see Blachère 1952, 2: 409–17). The
central focus of the khamriyya is wine, the object of the poet’s devotion and
praise; it is described in terms resonant of the old poetry.
Hamori suggests that the topic of wine as both old and virginal, seen in Abū
Nuwās’s line –
Does it not cheer you that the earth is in bloom, while the wine is
there for the taking, old and virginal? –
originates in descriptions of war, citing this verse by al-Kumayt –
When, after having seemed a delicate young girl, war shows itself a
graying old woman, quarrelsome and shrill. (Hamori 1974: 49; Abū
Nuwās 1958, 3: 5)
)
It occurs as well in this brief poem attributed to Imru al-Qays:
War is at first a beautiful maiden who appears with her adornments to
every headstrong youth.
But when she becomes heated and her fires blaze, she turns into an
aged crone with no spouse,
A grey-hair hag with clipped
( locks, altered in form, loathsome) to smell
and to kiss. (al-Mas ūdı̄ 1971, §1569, and cf. §1776; Imru al-Qays
1958: 161)
The movement of the topic from one context (gnomic) to another which is
totally different (wine) enables its transformation: the young “daughter of the
vine” imprisoned in the vat, at once ancient and virginal wine, becomes an
object of desire.
2 There is no excuse for your abstention from an ancient one whose
father is the night, whose mother the green vine.
3 Hasten; for the gardens of Karkh are decked out; no dusty hand of war
has usurped them. (Abu Nuwās 1958, 3: 5)
)
In these lines, the juxtaposition (in reverse order) of h)arb “war” and shamtā
˙ ˙
“grey-haired” seems to be a deliberate allusion to Imru al-Qays’s poem; this
allusivity (a concomitant of the transfer of topics into new contexts) is a
prominent 24
) feature of Abbasid poetry.
Imru al-Qays warned that war only appears beautiful to the “headstrong”
(jahūl), one with the quality of jahl, excess. Abū Nuwās’s persona of winebibber
prides himself on his jahl: “Youth was (for me) the steed of rude excess,” he
states (ibid., 3: 233), echoing al-Nābigha al-Dhubyānı̄’s
(
Āmir has spoken rudely (jahlan): the riding animal of rude ignorance is
insult (or: ‘youth’). (Quoted by van Gelder 1988: 19)25
)
The Arab critics noted other echoes of pre-Islamic poets, and particularly of Imru
al-Qays, in the khamriyya. Ibn Qutayba asserted that the source of Abū Nuwās’s
32
INVENTION
In a party where joy laughs wholeheartedly and where wine is

permitted
)
was Imru al-Qays’s,
I am now permitted to drink wine, after having been kept from it by an
all-absorbing occupation. (Quoted by Hamori 1974: 48–9; see Abū
Nuwās 1958, 1: 227; Ibn Qutayba 1981: 426)
Ibn Qutayba’s explanation (which, along with the provenance of the topic,
Hamza al-Isfahānı̄, the redactor of Abū Nuwās’s Dı̄wān, accepts) – “Abū Nuwās
˙ ˙
had sworn not to drink wine until joined with the person he loved in a
gathering; when they came together, (wine) became licit to him” – is expanded
upon by Hamori:
)
[Imru al-Qays’s] occupation was avenging his father’s murder; abstinence
from wine (and from other amenities) was to last until blood had been
shed for blood. Ibn Qutayba picks up the technical aspect of hallat liya
˙
l-khamru [the fulfillment of the oath]. It must be left undecided whether
he is correct in reporting that Abū Nuwās too had a vow: to touch no drop
of alcohol until he secured a coveted rendezvous. He was no doubt
justified in sensing a conscious echo – or parody – of the heroic in the
khamrı̄ya. (1974: 48–9)
The poem in which Abū Nuwās’s line occurs is, we may note briefly, a panegyric to
the governor of Egypt al-Khası̄b, and concludes the five-line wine-song with which
˙
the poem opens. There follows a ten-line rahı̄l describing the camel which bears
˙
the poet to his patron; the final five lines comprise the madı̄h. Wine-song and
˙
madı̄h, precisely balanced, are thus linked: the implication is that it is al-Khası̄b’s
˙ ˙
generosity which has made both gathering and wine available, where they had
been disallowed not by self-abnegating abstinence but by financial hardship.
Abu Nuwās frequently inverts, subverts, or explicitly renounces many of the
conventions of the pre-Islamic qası̄da. Most famous, perhaps, is his rejection of
˙
the atlāl topic (the lament over the ruined encampment of the departed tribe),
˙
a rejection which should not be taken, as is often done, at face value: it serves to
establish the khamriyya as countergenre, and Abū Nuwās could employ the topic
seriously when he wished. In his khamriyyāt he sometimes dismisses it outright:
1 Turn aside from the transformed ruins; leave off describing the ancient
abodes and the sticks of the fire-drill;
2 Forsake the bedouin Arabs; leave them, along with their wretchedness,
to one who is deprived, who likes hardship and (is) mean and
niggardly. (1958, 3: 120)
At others he mocks it:
1 Say to the one who weeps over faded traces standing – and what harm
would it do if he were sitting? –
33
INVENTION
2 “You describe the spring quarter and those who camped there, like
Salmā, Lubaynā and Khanas:
3 “Put the spring quarter, and Salmā, aside, and seek a morning cup of
Karkh wine, bright as a firebrand. . . .” (ibid., 3: 196)
Elsewhere, he modifies it to suit the urban ethos of the khamriyya, substituting
for the abandoned camps places in his own home town of Basra, now deserted
by him (see Meisami 1998b: 76–7).
1 The Musallā is empty, the sand hills deserted by me, as are Mirbadān
˙
and Labab,
2 And the Friday mosque which ingathered chivalry and glory is bereft,
and its courtyards and great halls,
3 Gatherings I frequented in my youth, until grey appeared on my
cheeks. (ibid., 3: 29–30)
(
The( opening line ( Afā l-Musallā wa-aqwati l-kuthubū) echoes that of Labı̄d’s
˙
Mu allaqa:
( )
Afati d-diyāru mahalluhā fa-muqāmuhā bi-Minan ta abbada Ghawluhā
˙
fa-Rijāmuhā
The site of the encampments has been effaced, and the hills of Ghawl
and Rijām, their [former] resting place, have returned to wilderness.
(al-Zawzanı̄ 1933: 112)
Similarly, Abū Nuwās’s
1 May abundant rain fall on (a place) other than the height and the cliff,
and other than the ruins of Mayy’s abode in the flat wastes
was said by the grammarian al-Mubarrad
( (as quoted by Hamza) to be a response
˙
to the opening line of the Mu allaqa of al-Nābigha al-Dhubyānı̄:
O abode of Mayya at the heights and the cliff, now forsaken, over
which long ages have passed.
Hamza comments, “al-Nābigha placed the abode on the heights and the cliffs
˙
because it was a higher place for it, more glorious for its people, and safer for
them from the torrents of floods. Then Abū Nuwās came along, ridiculing [or:
finding fault with] al-Nābigha’s description, and said [the line]” (Abu Nuwās
1958, 3: 103).
In what does Abū Nuwās’s “fault-finding” consist? Some critics take very
seriously Abū Nuwās’s outstpoken contempt for the rough, uncultivated
bedouin and his clear preference for a more( comfortable lifestyle, seeing in it
political motives associated with the Shu ūbiyya movement (see the discussion
and references in Arazi 1979). But the khamriyyāt are, it seems to me, designed
primarily for the entertainment of others, in urban Basra and, more particularly,
Baghdad, who share the poet’s preference – hence the force of Hamori’s
34
INVENTION
designation of the poet as “ritual clown” (see 1974, Chapter 2); and they are
not, moreover, lacking in a certain nostalgia for idealized aspects of the Arab
past (see further Chapter 3). This is a poetic game, in which the topics and
motifs of the Ancients (and not only the Ancients) are wrenched out of context
to supply materials for a new type of poetry.
Thus for example Abū Nuwās adapts the rahı̄l – the poet’s description of his
˙
journey through the desert, the excellences of his mount and the many
hardships he suffers – to depict journeys which begin outside the city walls of
Baghdad and whose destination is the tavern, not the beloved, the tribe or, as in
panegyric, the patron:
3 How many a group of companions, blameless, generous, free, of great
honour, radiant and beautiful,
4 Whose mounts I diverted, weary and weakened – for the winds’ ways
were closed,
5 And shadows had risen over my sandal-strap, like a feather in the fold
of a wing –
6 To wineshops located in trellised vineyards in twisting fields. (1958,
3: 90)
He may also describe his favourite mount:
4 And nights when I depart on a dark chestnut steed, and return at dawn
on a golden one,
5 Steeds of wine, not made to sweat or starve for a racing day. (ibid.,
3: 165)
But perhaps the most striking of these transfers, or transformations, is that
surrounding the poet’s object of desire. For the pre-Islamic poet desire for,
attainment of, or separation from a woman figured complex relationships with
kinsmen or enemies, with the poet’s own resources, with life, or Time (dahr)
itself (see Jamil 1999). For Abū Nuwās, wine is the beloved, the primary object
of desire (although other, more physical ones are associated with it – usually
boys, over whom wine provides the means for triumph). Wine, however, is the
true beloved – always faithful, always dependable (except when unaffordable);
and wine is described in terms, and in contexts, reminiscent of both pre-Islamic
and early Islamic ghazal. ( (
The poem beginning Da anka lawmı̄, introduced by an ( apostrophe shared by
both love and wine poetry (address to a “blamer”, ādhil) followed by the
affliction/remedy oxymoron, contains a description of the “beloved” in language
typical of ghazal:
1 Leave off blaming me, for blame is an enticement, and cure me with
she who was) the affliction:
2 A blonde [safrā ] in whose courtyards sorrow does not alight, and who, if
˙
struck by a stone, would touch it, dispelling its sadness (1958, 3: 2).26
35
INVENTION
( (
Compare the opening of a ghazal by Umar ibn Abı̄ Rabı̄ a (d. 93/712 or 103/721):
1 O my two companions, lessen your blame; do a pious deed for a lover
whom desire has afflicted with remembrance
2 For one white as the wild cows of the sands, tame, seductive, plump
and full-formed as the moon:
3 Tall, living in luxury, plump-jointed like the wild cows who graze amid
succulent flowers,
4 Round-legged, fragile of waist, beauteous of neck, breast and hair. (n.d.:
139)
(
But while Umar continues with a narrative description of an amorous
adventure, which includes his companions’ advice – “Leave off loving her! . . .
Be like one struck down [by drink] who has risen from his stupor” – and
concludes with the lady’s tearful remark (“I shall not forget . . . her words, as the
tears from her eyes
( coursed down her throat: ‘The blood-wit for this heart must
be sought from Umar’”), Abū Nuwās, having described the wine, its server, its
pouring and mixing, its brilliance, and the group of youths (fitya) who enjoy it,
returns to the identification of wine as the beloved:
9 It is she for whom I weep,
) and not for an encampment stopped at by
some Hind or Asmā .
10 God forbid that tents be pitched for (the sake of) a pearl, and that
camels and cattle walk over it!
)
The association between the beloved and wine can be traced back to Imru
al-Qays, who says of his beloved:
11 She walks (slowly) like one giddy with drink, felled on a sand-hill,
breathless and weak:
12 Smooth-skinned, elegant, soft and tender as a fresh bough of the bān
tree in leaf,
13 Heavy to rise, slow to speak, smiling with bright, cool teeth,
14 As if wine, and the clouds’ downpour, the scent of lavender, the
wafting sandalwood
15 Watered the coolness of her teeth as the dawn bird trilled. (1958: 110)
(
Here the
) lady is suffering from ( the debilitating effects of wine; in Umar’s
Hawrā a anı̄satun muqabbiluhā adhābatun ka-anna madhāqahu khamrū, “A dark-
˙
eyed maid whose mouth is as sweet as if it were the taste of wine” [n.d.: 182]),
the association between beloved and wine becomes closer, paving the way for
the substitution of one for the other by Abū Nuwās.
This substitution is of course facilitated (if not inspired) by wine’s feminine
gender, which makes ) easy the transfer) not only ( of ) feminine descriptive
adjectives – shamtā , “grey-haired”, safrā , “blonde”, adhrā , “virgin”, and so on –
˙ ˙
but of the patterns of diction typical of love poetry as a whole – though not
without a certain ambiguity, as in the poem by Abū Nuwās which begins
36
INVENTION
1 Shajānı̄ wa-ablānı̄ tadhakkuru man ahwā wa-albasanı̄ thawban mina

d-durri mā yublā
˙ ˙
1 I am distressed and afflicted by the memory of the one I love; it has
clothed me in garments of hurt which do not wear out.
2 Signs of what the young man’s heart conceals are found in the turning
of his eyes towards the one he loves.
3 But not every one who loves passionately is sincere. The lover wastes
away; he neither dies nor lives. (1958, 3: 14; accepting Ghazzālı̄’s
reading [Abū Nuwās 1982: 118] of ablānı̄ in 1 for Wagner’s adnānı̄
˙
“has wasted me”, which preserves the tajnı̄s ablānı̄/mā yublā)
Such patterns as shajānı̄, ablānı̄ etc., especially in the opening line of a poem,
should create the expectation of ghazal;27 but this expectation is short-lived, as
the poet turns to a narrative of his suit for, and marriage to, one of the
wineseller’s “daughters”, and thence to praise of her:
6 A nectar whose father is water and whose mother the vine, whose
nursemaid was the mid-day heat, grown intense. . . .
8 Christian of lineage, dwelling in Muslim towns, Syrian of provenance
[lit. in the morning], produced in Iraq,
9 A Magian who forsook the people of her faith because of her aversion
to the fires which they kindle. . . .
Abū Nuwās’s praise of his “beloved’s” noble ancestry
( scarcely differs from the use
of the same topic in ghazal, as for example by al- Abbās ibn al-Ahnaf (d. after
˙
193/108?):
14 A maiden of high lineage, noble in both paternal and maternal lines.
15 She gave me to drink of the saliva of her mouth; and how wonderful
the perfume from that girl’s lips! (1986: 85)
Wine possesses all the qualities customarily praised in the beloved – noble
lineage, elevated social status, chastity; brightness of visage, life-giving and
reviving powers, and so on (qualities which, mutatis mutandis,
( also characterize –
in the masculine – the mamdūh of panegyric). Ibn al-Mu tazz observed, “As for
˙
the comparison of wine to the brightness of day, we see that the topics presented
in this connection are transferred [muhawwala] from the poetry of the Arabs and
˙
the descriptions of beautiful faces, from which the muwalladūn have produced
various attributes for wine” (1925: 37, and see 37–40). Abū Nuwās’s
personification of wine as beloved will be discussed further in Chapter 8; let
us merely note, for now, the identification, which draws on the topics of both
ghazal and panegyric (e.g., placing the wine, like the name of the mamdūh, at
˙
the center of the poem; see further Chapter 6) to establish wine as the ultimate
object of desire and praise. To each poet, then, his own beloved; and to each his
own persona. And if Abū Nuwās adapts both the heroic topics of the pre-
Islamic poets and the love motifs of contemporary ghazal to his own purposes, he
37
INVENTION
is no less sparing with the topics of the zuhdiyya, and with religious motifs in
general. Hamori comments on the allusion to the topics of “abandonment of the
world” and right guidance in the lines,
Leave the gardens of roses and apples; direct your steps – may you be
guided aright [hudı̄ta] – towards Dhāt al-Ukayrāh!
˙
No doubt, on hearing a line like this, the audience’s first reaction was
astonishment. “Leave the gardens” misleads you: when taken by itself, it
would better suit an ascetic poem. In the next half-line the optative turns
the sentence into something of a sermon. Right guidance is a frequent
notion in the Koran and in pious exhortations, but the place where the
spiritual pilgrim is advised to go happens to be a monastery of Christian
monks who sell excellent wine. It is all a bit of a joke, and the invitation
is not uncommonly outrageous: people often used to go on outings to such
monasteries. The scandal is in the wording rather than in the contents.
(1974: 51)
(
Abū al- Atāhiya (d. 211/826) employs the same motif in more than one
zuhdiyya, with virtually the same wording; for example,
14 Be just – if you are guided aright [hudı̄ta] – if you yourself seek justice;
do not approve in others what you would not approve for yourself.
(1886: 293)
Indeed, the frequent similarities of phrasing between the two poets suggest that
they were, in fact, parodying each other’s poetry (see further below).
Hamori comments extensively on Abū Nuwās’s utilization (or parodization)
of religious imagery and of topics specific to Islam – the devil as shaykh, or
religious guide, who instructs in debauchery rather than piety; wine as the qibla
towards which the winebibber prays (a motif also prevalent in ghazal, in which
the beloved is the sacred shrine sought by the pilgrim-lover); denial of
resurrection and the afterlife; wine as the illuminating light which guides men
to its worship; and the motif of divine forgiveness as encompassing the sinner
(1974: 50–71). We might add many more examples, such as the famous poem
in which the rituals of Islam – prayer, fasting, pilgrimage, zakāt – are inverted to
become the practices of the winebibber.28 The impact of such procedures,
however, arises not merely from their explicit content – the anti-heroic, anti-
religious “message” of the social rebel – but from the manner in which both
traditional conventions and contemporary practices are subverted to the ends of
the khamriyya.
The khamriyya’s generic opposite, the zuhdiyya, also modifies topics borrowed
from other genres to suit its own specific ends. As
( Abū Nuwās was the foremost
practitioner of Abbasid khamriyya, so Abū al- Atāhiya may be considered the
inventor of the zuhdiyya, which, though it has antecedents in earlier poetry, is
developed in its canonical Islamic form by that poet (see Sperl 1989: 72–82).
38
INVENTION
The self-conscious nature of this specialization on the part of both poets is

suggested by an anecdote retold by Ibn Manzūr (d. 711/1311) which purports to
˙
record their marking out of specific poetic territories.
) (
Abū Makhlad al-Tā ı̄ [related]: Abū al- Atāhiya came to me and said,
˙
“Abū Nuwās will not go against you; I would like you) to ask him to refrain
from composing zuhd. I relinquish to him madı̄h, hijā , khamr, and that sort
˙
of effeminate [raqı̄q] verse that poets write; my passion is for zuhd.” I sent
to Abū (Nuwās; he came to me, and we busied ourselves (drinking); but
Abū al- Atāhiya( would not drink wine with us. I said to Abū Nuwās,
“Ishāq [Abū al- Atāhiya] is one known for his eminence and precedence;
˙
he would like you to refrain from composing zuhd.” Abū Nuwās was taken
aback by this, and replied, “Abū Makhlad, you have anticipated what
I wished to say to you on this matter; for I had resolved to compose
[poetry] about things which any libertine would repent of, and have
indeed done so.” (Ibn Manzūr, Akhbār Abı̄ Nuwās; quoted by al-Dāsh
˙
1968: 331)
Whether this anecdote is veracious or fictitious matters little; it reflects an
awareness on the part of poets and their audiences alike both of deliberate
specialization and of a strong competitive element among the poets of the
period which forces us to consider the development of independent genres in
this period not merely from the point of view of the supposed beliefs or
proclivities of the poets, but from the standpoint of literary dynamics.29
Not only does the zuhdiyya, like the khamriyya, manipulate the form of the
qası̄da to its own ends (as will be discussed further in Chapter 5), but it too
˙
appropriates topics associated with other genres, modifying them to the thematics
of the ascetic poem. If Abū Nuwās invokes the atlāl topos only to reject it –
˙
1 Forget the traces of the abodes and the ruined encampments; forsake
the spring quarter, effaced and obliterated.
2 Have you ever seen the abodes return an answer, or respond to one
who asks a question? (1958, 3: 257) –
(
for Abū al- Atāhiya the atlāl are a sign of mortality and transience:
˙
1 Whose is the ruined encampment I question, its dwellings long
abandoned,
2 That morning when I beheld its ground below announce the death of
those above? (1886: 227–8; translated by Sperl 1989: 209)
Li-man talalun (“whose is the ruined encampment?”) is a formula ) which
˙
introduces many a pre- and early Islamic poem. So, for example, Imru al-Qays:
1 Li-man talalun dāthirun āyuhu taqādama fı̄ sālifi l-ahrusı̄
˙ ˙
Whose is the ruined encampment, its signs effaced by the advance of
many ages of time. (1958: 121)
39
INVENTION
And Abū Nuwās:

( ( (
1 Li-man talalun āriyu l-mahalli dafı̄nū afā ahduhu illā khawālidu jūnū
˙ ˙
Whose is the ruined encampment, its place bare, buried (concealed)
from view, whose familiar signs have been effaced, save for
blackened cooking-stones? (1958, 3: 305)
The motif of the ruins not responding is similarly conventional:
( ( ( ( )
1 Alimmā alā r-rab i l-qadı̄mi bi- As asa ka annı̄ unādı̄ aw (ukallimu
( akhrasa
Visit (companions twain) the ancient spring quarter ) at As asa; it is as if
I addressed or called on a mute (stone). (Imru al-Qays 1958: 117;
cf. also Sperl 1989: 220 n. 16)
)
For Imru al-Qays, the absence of response is testimony that the tribe has
departed, abandoning the encampment; for Abū Nuwās, it is because mute ruins
are not going to answer( anyone, and to address them (let alone describe them)
is pointless; for Abū al- Atāhiya, it is because, though silent, they provide mute
witness of those who have not departed, but will inhabit them till Resurrection,
as the atlāl are transformed into the graves.
˙
1 What is the matter with the graves, that they do not respond when he
who sorrows calls out to them? (1886: 25)
Where the khamriyya substitutes, for the pre-Islamic tribal and heroic ethos, the
anti-heroism of libertinism, the zuhdiyya replaces that ethos with the concept of
man’s helplessness: “If the pre-Islamic hero is . . . depicted as active in the face of
death, man in the zuhdiyya is the passive victim of the forces of destiny” (Sperl
1989: 81). If Abū Nuwās exhorts Tazawwud min shabābin laysa yabqā, “Store ( up
provisions of youth, which does not last” (Hamori 1974: 55), Abū al- Atāhiya
responds, Laysa zādun siwā t-tuqā, “There is no sustenance save piety” (Sperl 1989:
81; both verses allude to Koran 2: 197: Tazawwadū fa-inna khayra z-zādi t-taqwā,
“Store up provisions;
( for the best provision is the fear of God” [Hamori 1974: 55]).
Abū al- Atāhiya also employs topics of ghazal not to praise a beloved, but to
warn against setting one’s heart on that false beloved, this world (dunyā, also
conveniently feminine; see further Chapter 8).
1 I have cut the cords of hope in you, and brought down my baggage
from the back of my mount;
2 And despaired that I might survive to enjoy what I have gained from
you, O world, or that ought would remain to me.
3 I found the coldness of despair in my breast, and have been freed from
my bond and my constant journeying. (1886: 194)
The lover renounces this world) in despair, seeing through her seductive
appearance as Kumayt and Imru al-Qays saw through the equally seductive,
equally deceptive face of war. The “abode” (dār), too, is feminine, and thus it
too may be equated with an unfaithful beloved:
40
INVENTION
1 An abode for love of which I suffered, treacherous to her lover:

2 Each is tried and afflicted by what she gives and takes,
3 By her enchantment [khalb] and her pride, by her farness and nearness,
4 By her praise and her blame, by her love and her insults.
5 If you do not seek help in contentment, she will be strait for you,
despite her expanse.
6 No pleasure will be destined for you save with the fear of her calamity
[khatb].
˙
7 If she approaches in fresh opulence the death-knell cries out beside her.
(ibid.: 35–6)
The figure of this world as a treacherous beauty becomes a commonplace of later
poetry, as well as of prose homilies and sermons. This world – man’s transient
abode – is both enchanting and deceitful; he who loves her suffers for nothing,
as his reward is death. The parody of the central convention of ghazal, the
lover’s devotion to a cruel beloved, is signalled in the lines above by such words
as khalb, “enchantment, charm”, which also means “talons” and occurs in the
metaphor khalb al-maniyya “the claws of death”, and khatb, “calamity”,
˙
associated with khataba “to ask in marriage”, used ironically ( by Abū Nuwās in
˙
the khamriyya as he seeks the daughter of the vine. Abū al- Atāhiya also evokes
the cup; but it is a cup to be avoided:
6 I had forgotten, as the cup of death was passing round in the hand of
one neither heedless of it nor forgetful,
7 That one day I shall surely drink from the cup of death, and be struck
down – just as those who have departed drank of it. (ibid.: 130)
Wine, world, abode, soul (as we shall see in Chapter 7) – all feminine, all
treacherous and deceptive. To put one’s trust in them, to follow their urgings or
succumb to their blandishments, is to make a fatal error.
While the khamriyya and zuhdiyya depend for their effects on their
manipulation of generic conventions and topics derived from earlier poetry,
from each other, and from the independent love poem, the ghazal itself appears
more stable than the other two. This is perhaps in part because it achieved (
independent (status( earlier than they, at the hands of Hijazi poets such as Umar
ibn Abı̄ Rabı̄ a, of Udhrı̄ poets like Jamı̄l (d. 82/701), and of Umayyad poets in
the cities of Kufa and Basra, among them Bashshār ibn Burd (d. c. 167/784).
While the ghazals of the poets of these “schools” vary widely in tone, ranging
from 30
( chaste to licentious, by the time of the form’s chief Abbasid exponent,
al- Abbās ibn al-Ahnaf (d. after 193/808?), its dominant mode is already well
˙
established as the antithesis of the pre-Islamic heroic mode: “what had been a
matter of action was now translated into emotion, and a poetry of passionate
but unfulfilled love – passionate inaction – was born” (Hamori 1974: 38, and see
31–47). Its characteristic diction, marked by simplicity and musicality, was
equally well established.
41
INVENTION
As the libertine’s devotion to wine and debauchery puts him outside the
moral norms of society, so the lover’s obsession, similarly incompatible with
religion, makes him an outsider; his single-minded pursuit of that obsession
makes him at once both victim and hero. Hamori compares Jamı̄l’s line –
Whether to love her means to be guided aright or to stray [ghawāya],
I stumbled upon this love without intent–
to a line by the pre-Islamic poet Durayd ibn al-Simma:
˙
What am I but one of the Ghazı̄ya? If they err, I err [in ghawat
ghawaytu]; and if they follow right guidance, I do too. (Hamori
1974: 42, 44)
We may recall ( the motif of “right guidance” (hudā > hudı̄ta), used by Abū Nuwās
and Abū al- Atāhiya with significantly different meanings in different generic
contexts. The equivalent word in Jamı̄l’s line is rushd (Durayd’s phrase is in
tarshud . . . arshudı̄),( opposed to) ghawāya/ghawat, “straying (from the right path),
(error”. (In the Mu allaqa Imru al-Qays’s beloved reproaches him: wa-mā in arā
anka l-ghawāyata tanjalı̄, “I do not see that error has left you”; 1958: 40.)
Hamori reads both lines as reflecting “assent to a given situation” – in
Durayd’s case, to joining his tribe in a disastrous battle after they had rejected
his pleas for restraint, in Jamı̄l’s to pursuing his love despite similar pleas for
restraint by the “blamer” (1974: 44). There is perhaps more involved, however.
Durayd’s poem – in which the line which precedes that quoted is “When they
rejected my counsel I remained one of them, though well aware of their error
and knowing that I would be entering upon a misguided course [wa-innanı̄
ghayru muhtadı̄]” – is an elegy
( for his brother, who was killed in the battle in
question; Abū Hilāl al- Askarı̄ comments, “He agreed with his brother’s
opinion, even though he saw it was in error, and abandoned his opposition,
even though it was right, fearing to lose his love,” and praises the line as “the
most
( eloquent in which a man supported his brother” (Durayd 1981: 47 n. 17;
al- Askarı̄ 1994, 1: 118–19; translated by Hamori 1991: 15; in Abū Tammām’s
Hamāsa, Hamori’s source, there is an intervening verse). The motives for
˙
“assent” are totally different – solidarity versus individual obsession, jahl (and
moreover the Islamic overtones of words like rushd and muhtadı̄ would not have
been lost on the audience): Jamı̄l’s statement involves elevating the ghazal-
poet’s persona to heroic (or counter-heroic) status by deliberately choosing error
over guidance, much the same as the winebibbing persona of the khamriyya is
similarly elevated. Abū Nuwās writes (1982: 198):
1 When the “father of war” orders his horsemen to war,
2 And death’s banner proceeds openly before the leader,
3 And war grows hot, and burns brightly, kindling flames. . . .
5 We take the bows in our hands; but the bows’ arrows are lilies. . . .
7 And our warfare becomes good company, and we ourselves good friends,
42
INVENTION
8 With youths who consider being slain by pleasure a sacrifice.

If the hedonistic winebibber is slain as a sacrifice to pleasure, the obsessed
lover is the sacrificial victim of love – a topos recalling the Prophetic saying, “He
who loves, and remains chaste, and(dies, becomes a martyr,” developed both in
love poetry (especially that of the Udhrı̄ poets) and in the literature on love
and lovers which flourished later in the Abbasid period. (See e.g. Hamori 1974:
39–47; Giffen 1971: 91–115). Yet if Abū Nuwās chooses wine over warfare, and
if the ghazal-poet chooses to make (or at least to pursue) love, not war, these are
active choices, no matter how fatalistically presented (and it is Durayd who,
bound to honour bonds of tribe and kinship, has far less real choice). The
ghazal-poet, like the wine-poet, is thus less passive than Hamori sees him as
being: he chooses his fate, his victim’s role, as a stance which will bring him, in
the end, far greater glory than the heroic one, as his name will join the roll-call
of famous lovers.
(
Jamı̄l never loved as I do; know this, verily, and neither did Urwa,
love’s martyr. )
No, no, nor was al-Muraqqish ( like me when he loved Asmā until the
appointed, fatal end. (al- Abbās ibn al-Ahnaf 1986: 15)
˙
The love poem presents a closed world dominated by the obsessive persona of
the poet-lover, through whose vision all events and persons are filtered. This
obsessiveness is reflected by a marked fixity in the ghazal’s conventions, a fixity
which has implications for its structure as well, as the repetitive nature of its
conventions means that the poem can be organized in a manner which
suppresses explicit linkages between its parts, since any choice of topics can be
deployed which relate to its focus: the depiction of the emotional state of the
lover. Thus the ghazal often reveals no clear organizing principle beyond a
sequence of generically related segments ordered in a manner which appears
arbitrary because it requires the audience to supply connections which are left
implicit. This type of structure – which may also be associated ( with the
performance context of such poetry31 – characterizes many of al- Abbās’s poems,
for example this love-plaint (ibid.: 25).
1 I concealed my passion, and avoided my beloved, keeping secret in my
heart a wondrous yearning.
2 My avoidance of her was not from anger, but (because) I feared lest
shame fall upon her.
3 I shall guard and conceal her secrets, and keep hidden the pleasure
I had from her.
4 How many who stretched out their hands towards union have received
not a share!
5 (With) some, I was satisfied with what love I received from them,
rightly or wrongly;
43
INVENTION
6 and some to whom passion called me, and I obeyed the summons,
acceding;
7 and some I was fond of in youth – but I grew white-haired before my
time!
8 By my life! they lie, who claim that hearts requite each other faithfully;
9 for were that true, as they claim, no lover would ever be cruel to
another.
10 How can this be what I desire? – that my beloved sees my virtues as
faults?
11 I have seen no one like you in all the worlds – half plump (as a
sandhill), half (supple) as a branch.
12 When you trample upon the earth, you make of earth another perfume.
(
Al- Abbās develops four basic topics of ghazal in linear, ABCD sequence.
Segment A (1–3) announces the topic of “concealment of love” (kitmān) with
katamtu l-hawā, “I concealed my passion,” and amplifies it by stating the cause
for concealment (implying a choice between various possibilities) and the
determination
( to pursue this course; the progression of verb tenses (katamtu . . .
sa-ar ā, “I concealed . . . I shall guard”) helps to unify the segment. Segment B
(4–7), introduced by the wāw rubba construction (wa-kam bāsitı̄na, “And how
˙
many who stretched out their hands”) which is a frequent marker of transition,
catalogues the speaker’s past experiences in love, presented as typical and
functioning to define his present situation by placing it in the context of
inequalities in love; it is unified by the anaphoric repetitions or near-repetitions
with which each line begins. (The “religion of love” motif is suggested by the
use of labbaytu, “I responded,” a term associated ( with the pilgrimage, in 6.)
Segment C (8–10), introduced by the oath la- amrı̄, “By my life,” links the topic
of inequality to that of the beloved’s cruelty, moving from a generalization on
the lack of reciprocity in love to the speaker’s own case. In the fourth and last
segment, D (11–12), the speaker addresses the lady directly, to praise her; this
address is not a necessary outcome of the poem’s movement but occupies the
final position for much the same reason that encomium constitutes the final
portion of the polythematic qası̄da, that is, as the last (hence best) item in a
˙
sequence. (See further Chapter ( 3 below.)
The topics chosen by al- Abbās are used to support the poem’s focus not only
on a single genre, that of love, but in particular on the persona specific to that
genre. This focus on the implied speaker (not to be confused with the
narcissistic
( emphasis of a real speaker on his own person) so characteristic of
al- Abbās’s poetry is no innovation with( him, but is typical (of love poetry in
general and is seen also in such poets as Umar ( ibn Abı̄ Rabı̄ a or Bashshār ibn
Burd, both of whom (in contrast to the Udhrı̄ poets, whose poems, by and
large, are brief and occasional) were major contributors to the( development of
the independent love poem. The loose, linear structure of al- Abbās’s poem is
also found often in the zuhdiyya, which, since its generic content is relatively
44
INVENTION
fixed and predictable, often operates through the accumulation, in linear

sequence, of relevant topics, coming to resemble a sermon in verse. This type of
structure is thus characteristic of much Arabic poetry of the period; and we shall
discuss further examples later, in Chapter 6.
That a poem may rely on implicit links between generically related topics
is an important principle of composition in monothematic poems such as
khamriyya, zuhdiyya and ghazal. When the topics are apparently unrelated we are
in the presence of another type of generic manipulation which has proven even
more problematic for scholars, as it involves different strategies of organization,
of generic recognition and of the evocation of the speaker’s persona. Such
strategies characterize a poetic form which, in its most complex manifestation,
lies at the end of our chronological time-scale: the Persian ghazal, as practiced
by the master of the form, Hāfiz of Shiraz.
˙ ˙
Invention and genre in the Persian ghazal

Many centuries lie between the Persian poet Hāfiz and the Arabic poets of the
˙ ˙
early Abbasid period, centuries during which many generic and formal
developments took place both in Arabic poetry itself and in Persian poetry as
it progressed from the tenth century onwards. We will discuss many of these
developments in subsequent chapters. What is important to note here, however,
is that the generic experimentation which marks Hāfiz’s ghazals assumes an
˙ ˙
engagement with the entire tradition of Arabic and Persian poetry of the seven
centuries and more which precede him – as the Dı̄vān’s opening, and
programmatic, ghazal makes clear (see Chapter 9 below) – and must be seen
against the background of that tradition.
(The Persian ghazal originated as a brief lyric form analogous to the Arabic
qit a, and composed for similar occasions. Over the centuries, and particularly
˙
from the late eleventh century onwards, it was expanded and modified until it
became the most popular poetic form in post-Mongol Persia. (On the various
theories surrounding the origins and development of Persian ghazal see
A. Bausani in EI2, s.v. “Ghazal”.) Although the ghazal was originally composed
to be sung, and long retained its close connection with music, the influence of
its performance context on its style has as yet received little examination, and is
largely beyond the scope of this study.32 The dı̄vāns of poets of the fourth/tenth
and early fifth/eleventh centuries (the Sāmānid and early Ghaznavid periods)
contain relatively few identifiable ghazals – partial exceptions are Rūdakı̄
(d. after 339/950–1), who was an accomplished musician as well as poet, and to
a lesser extent Manūchihrı̄ (d. 432/1140–1?) – perhaps because the informal and
oral
( nature of the sung poem (and perhaps also its non-conformity with the
arūd) meant that it would not, or seldom, be recorded in writing or included in
˙
a collection of “official” poetry. Nor is it clear how many of the poems classified
as ghazals in modern editions are in fact the exordia of panegyric qası̄das which
˙
have not survived in their entirety.
45
INVENTION
By the end of the fifth/eleventh century and the beginning of the sixth/
twelfth, however, the independent ghazal had become a flourishing ( form and
had acquired its normative formal features: the rhymed matla and the use of the
˙
poet’s pen-name (takhallus
) ) in the final or penultimate line, said to have been
˙
introduced by Sanā ) ı̄ (d. 512/1131), who employed this device in many of his
ghazals (see Humā ı̄, in Mukhtārı̄ 1962: 571; and see further Chapter 3 below),
and used by other of his contemporaries as well. Whether this reflects the
ghazal’s increasing “literarization”,
( its passage from oral song to written poem (as
the convention of tasrı̄ , and perhaps also that of the takhallus [see Losensky
˙ ˙
1998a] suggest), is a matter for speculation; what is in no doubt, however, is that
from this time onwards it enjoys both increasing popularity and increasing
adaptation to a variety of uses.
The ghazal has received more than its share of criticism for its “incoherence”,
particularly with reference to Hāfiz. (For overviews see van Gelder 1982a:
˙ ˙
14–22, 194–208;
( Hillmann 1976; Andrews 1973: 97–9, 1992.) Hāfiz’s patron
˙ ˙
Shāh Shujā (r. 759–86/1357–84) is reported to have critized the poet for his
“incongruity”, saying: “The bayts . . . in your ghazals . . . do not happen to be of
one kind, instead in each ghazal there are three or four bayts about wine and two
or three bayts about sufism and one or two bayts about the characteristics of
the beloved. The changeableness of each ghazal is contrary to the way of the
eloquent.” To which Hāfiz is said to have replied that, nonetheless, “the poetry
˙ ˙
of Hāfiz has found consummate fame in all regions of the world and the verse of
˙ ˙
his various rivals has not set foot beyond the gate of Shiraz” (quoted by Rehder
1974: 83 [the source is Khvāndamı̄r’s Habı̄b al-siyar]; see also van Gelder 1982a:
˙
207, and see the discussions by Rehder, Hillmann, and Andrews, cited above.)
This perhaps apocryphal anecdote (which ( may reflect poetic jealousy rather
than literary criticism: Shāh Shujā ’s own verses were of consummate
mediocrity), often invoked as evidence for the ghazal’s formal incoherence,
draws attention to an important point: the proliferation of generic elements
within the compass of a brief lyric (with a corresponding proliferation of poetic
personae), to an extent where genres become combined, inserted into one
another, or even treated allusively.
Generically the ghazal is, par excellence, a love poem, and many of its topics
and motifs derive ultimately from the independent love poems of the Abbasid
period as well as from earlier Arabic poetry; but from its inception in the early
proto-ghazals of Rūdakı̄ it incorporated bacchic topics, and was later extended to
include gnomic, homiletic, and religious (often mystical) themes (see further
Meisami 1990d). Its bacchic and erotic diction and imagery were also used for
brief panegyrics (see Meisami 1987: 271–85). Hāfiz was heir to a long tradition
˙ ˙
of generic) manipulation in the Persian ghazal which began, roughly speaking,
with Sanā ı̄ (and his contemporaries), whose ghazals include both panegyric and
religious-mystical poems as well as more “secular” love poetry. Panegyric ghazal
is especially associated with the )court of the Ghaznavid sultan Bahrāmshāh
(r. 515–52/1118–51), where Sanā ı̄ and Hasan-i Ghaznavı̄ (d. 556/1160–1) in
˙
46
INVENTION
particular cultivated its use, although earlier examples are also found (see de
Bruijn 1983: 152; Meisami 1987: 273–9; and see for example Mukhtārı̄ 1962:
221–2).
It is possible that the increasing length and ornateness of the qası̄da led poets
˙
to employ the simpler form of the ghazal to present praise in informal
gatherings such as the banquets and drinking parties which were a prominent
feature of court life, a practice for which there are precedents in Abbasid poetry
as well; moreover, the convention of homoerotic love characteristic of Persian
love poetry (and seen in the nası̄b of panegyric qası̄das) lent itself to the
˙
depiction of relations between poet and patron in a manner similar to the
Augustans’ use of homosexual love as a “framework for treating a personal
relationship with an amicus”) (Williams 1980: 214, and see 212–16; cf. also the
panegyric ghazal by Sanā ı̄ with the radı̄f dūst, “friend”, discussed in Chapter 5
below).
Such brief, lyrical panegyrics employ erotic and bacchic motifs in preference
to the panegyric topics of the qası̄da (e.g., emphasis on the ruler’s military
˙
achievements); the following ghazal by Hasan-i Ghaznavı̄ (1949: 166), which is
˙
virtually indistinguishable from a love poem, provides a typical example.
1 To the beloved I’ve given heart and life;
to join with him once more: ah, that were life!
2 I’ll patiently endure this; for my hand
by separation’s tyranny is bound.
3 I suffer from his absence pain so sore,
the lofty Sphere itself could not endure.
4 The separation of two intimate friends:
how speak of it? for it cannot be known.
5 Rejoice, Hasan, as you for his sake grieve;
˙
he’s both the affliction and the remedy.
6 I fear it will not reach Sultan’s ear
that grief for him is sultan o’er( my heart:
7 Shāh Bahrāmshāh, son of Mas ūd, who is
the very form of sovereignty, image of life.
Although the panegyric context is suggested by such topics as the poet’s humble
patience (opposed to the tyranny of separation in the elaborate word-play of 2:
pāy dar dāman āram az ān-k/dast dast-i sitam-i hijrān-ast, literally “I must draw my
foot beneath my skirt, for my hand is in the hand of separation’s tyranny”), and
by the word-play on sultān in 6, in which gham-i ū, “grief for him,” is as
˙
ambiguous in the Persian as it is in English, the poem’s panegyric ( purpose is not
announced until the final line, where the beloved (ma shūq) of the opening
becomes identified with the person praised (mamdūh), who thus becomes the
˙
object of both love and praise.
This shift in object, as well as in addressee, is also found in much mystical
poetry, in which both become, implicitly, either God, a spiritual master or an
47
INVENTION
ideal figure (e.g., the Prophet). Mystical poetry adapts the erotic and bacchic
motifs of secular poetry in ways which are both thoroughgoing and ambiguous. (
(This is true in Arabic as well, for example in the mystical poems of Ibn al- Arabı̄
or Ibn
) al-Fārid˙; see Chapter 8 below.) The topics of praise in these lines by
Sanā ı̄, for example, might apply equally to the beloved, the prince, or God
(1962: 807):
1 O moon-faced beauty, the whole world sings your praise;
lovers’ affairs are undone because they dance to your tune.
2 Wherever there is sweet verse, there are the stories of your love;
wherever there is elegant prose, there are the books of your
attraction. . . .
7 Wherever there are seeing eyes, there is the court of your love;
wherever there is an exalted ear, there is the lover of your song.
This ambiguity extends to physical descriptions as well (ibid.: 820):
1 If your face, O heart-illumer, is not like the moon,
why are your two black lovelocks two halves of the full moon?
2 Although your moonlike face is a source of light;
although your black locks are the source of sin,
3 You are the king of idols, and lovers are your army;
you are the earth’s moon, and the heavens your crown.
( (
In due course, at the hands of such later poets as Attār, Irāqı̄, and Rūmı̄ (as
˙˙
well as in the prose writings of other mystics), the descriptive, erotic and
bacchic imagery of the ghazal developed into a symbolic vocabulary capable of
being read allegorically (and leading to an often mechanical interpretation of
many poems, even if their original intent was not mystical, according to this
“lexical code”). Thus for example many descriptive topics, which can often be
traced back to early Arabic love poetry, lend themselves to a mystical
interpretation (although not always without ambiguity). ( Consider this blazon by
the sixth/twelfth-century mystical poet Farı̄d al-Dı̄n Attār (1960: 173):
˙˙
1 O you whose locks are snare, your mole the grain:
may every prey you take be licit to you.
2 The sun continually falls, enmeshed,
into the ringlets of your night-dark snare.
3 Like the black markings visible upon
the sun’s face, is your own black beauty-spot.
4 The heart’s bewitched by your black curling locks;
the soul is thirsting for your limpid spring.
5 From the world of beauty the midwife of grace
brought you forth a hundred thousand years ago. . . .
We might read this ghazal in the terms discussed by Ahmad Ghazzālı̄ (d. 520/
˙
1126) in his Savānih, an early Persian mystical treatise.
˙
48
INVENTION
. . . in the world of Imagination, in order to reveal its face, love sometimes

may show a concrete sign, while sometimes it may not.
(1) Sometimes the sign is the tress of the beloved, sometimes the
cheek, sometimes the mole, sometimes the stature, sometimes the eye,
sometimes the eyebrow, sometimes the glance, sometimes the smile, and
sometimes the rebuke.
(2) Each of these symbols relates to a locus in the lover from which a
specific quest arises. He for whom the sign of love lies in the beloved’s eye,
his nutriment is supplied by the beloved’s sight. . . . If the sign is the
eyebrow, then . . . the quest arises from his spirit. . . .
(3) In the same way, each of the other signs . . . in the physiognomy of
love signifies a spiritual or physical quest or an imperfection or a fault, for
love has a different sign on each of the inner screens, and these features
are its signs on the screen of Imagination. Therefore, her features indicate
the rank of (the lover’s) love. (1986: 52–3)
The same sort of transfer applies to bacchic motifs:
If it becomes possible for the lover to take nutriment from the beloved . . .
that will not happen except in (the mind’s) absence from the world of
manifestation . . . which is similar to a state of intoxication in which the
companion is not there, but the nutriment is there. . . .
Love is a kind of intoxication, [for its] perfection . . . prevents the lover
from seeing and perceiving the beloved in her perfection. This is because
love is an intoxication [of] the organ of . . . perception, hence it is a
prevention to perfect perception. . . . (ibid.: 64–5, amending Pourjavady’s
translation)
(
Attār writes (1960: 154):
˙˙
1 Your nearness makes me drunk; I know nought of my being;
I’m drunken with love’s grief; there’s no other drunk like me.
2 Since my liver’s blood was spilt in the feast of the wine of love,
I of the burnt heart have no refreshment but the liver’s blood.
3 Those drunk on love’s wine have departed into this desert;
I have remained (behind), and there is no sign of my leaving.
One result of such transformations is that the language of the ghazal becomes
increasingly polysemous, lending itself to varied, often contradictory readings
simultaneously encoded into the poem. (Compare Andrews’ discussion of later
Ottoman gazel, 1985). This makes it exceedingly dangerous to take the
statements of an Ahmad Ghazzālı̄, or of the later Mahmūd Shabistarı̄ (d. 726?/
˙ ˙
1326?), who wrote a long poem on the symbolic language of the mystics, or of
later commentators on Hāfiz, as descriptive of poetic practice: Ghazzālı̄ employs
˙ ˙
the poetic vocabulary of love poetry (along with quotations from it) for his own
ends; Shabistarı̄’s motives are similar; and the commentators, on the basis of
49
INVENTION
such cross-fertilization of prose by poetic imagery, developed a hermeneutics of

reading, of decoding the poem in a certain predetermined, not to say
overdetermined, manner.( Even
( in the case of poets known to have been
practising mystics – Attār, Irāqı̄, Rūmı̄, each of whom practised his own brand
˙˙
of spiritualism – the language is much more elusive (and allusive), much less
cut-and-dried than is suggested by their interpreters. How much more so when
we come to a poet like Hāfiz, whose “mysticism” is largely a stylistic feature of
˙ ˙
his ghazals.33
Another result of the extensive tradition of generic transfer is that topics and
motifs which in earlier poetry were essentially concrete (though often none the
less symbolic) undergo a process of abstraction, of textualization; the world of the
poem closes in upon itself, it relates primarily to other texts (cf. Williams 1980:
ix for a parallel situation in Augustan poetry). Generic topics become both
increasingly allusive and infinitely combinable, so that the very mention of one
– for example, the beloved’s night-black curls – evokes a host of associations
(dark obscuring light; snares for the heart; fetters for madmen; separation and
withdrawal of favour, and so on) which, as the products of a long process of
development, need not be explicitly stated to be understood. This process,
somewhat paradoxically, also makes possible the application of such conceits to
specific topical issues, as we shall see. It leads as well to another feature which is
particularly marked in Hāfiz’s ghazal: the allusive combination not only of related
˙ ˙
generic topics, but of genres themselves, within the brief lyric. All these features
– polysemy, the combination of the abstract with the topical, and the allusive
mixture of apparently unrelated genres – are seen in this ghazal by Hāfiz (QG9;
˙ ˙
see Bausani 1958: 146–9; see also Meisami 1991a: 99–101, 1990d: 137–40).
1 The brilliance of youth’s season once more adorns the garden;
news of the rose arrives to the sweet-songed nightingale.
2 Sabā, should you pass by the elegant youths of the meadow,
˙
pray, carry my regards to cypress, rose and basil.
3 If the wine-selling Magian child displays himself like this,
I will sweep the wineshop’s threshhold with the tips of my eyelashes.
4 O you who draw over the moon a polo-stick of pure amber,
do not cause me distress, a wanderer, gone astray.
5 I fear that group who laugh at those who drink the dregs
will, in the end, put their faith in pledge to the tavern.
6 Be the companion of the men of God; for in Noah’s ship
is (one of) earth that gives not a drop for the tempest.
7 Go out of this turning dwelling, and do not ask for bread,
for that (host with) blacked pot in the end kills its guest.
8 The last resting-place of everyone is a handful of dust;
say, what need is there then to raise a palace to the skies?
9 O my Moon of Canaan, the throne of Egypt is yours;
the time has come for you to bid farewell to your prison.
50
INVENTION
10 Hāfiz, drink wine, be a libertine, be happy; but

˙ ˙
do not, like others, make the Koran a snare of hypocrisy.
(
This ghazal, whose mixture of topics recalls Shāh Shujā ’s criticism of Hāfiz, was
˙ ˙
used by Bausani to support his notion of the ghazal’s “formal incongruity” (1958:
149). It begins (1–2) with a description of spring (vasf-i bahār), a characteristic
˙
opening for love poems, wine poems, and panegyrics, suggesting that any of these
genres (ghazal, khamr, madh) may become the dominant one. In fact, though all
˙
three aghrād are combined in this ghazal, the generic dominant will turn out to be
˙
something quite different. The apostrophe to the Sabā (the south wind,
˙
conventionally the lovers’ messenger) indicates that the speaker is elsewhere
than in the garden described; lines 3–4 suggest that he is in the tavern, where the
beauty of the “wine-selling Magian child” (a particularly Persian term for the
beautiful cup-bearer, the sāqı̄) threatens to rob him of his wits and place him in
eternal service to the tavern. All these topics – the handsome young Magian sāqı̄,
his irresistible beauty, his cruelty in concealing his bright face beneath his curling
black locks – are ultimately traceable to the Arabic khamriyya and ghazal.
With lines 5–8 the generic tone shifts from that of the wine song, with its
vague, hedonistic expression ( of carpe diem as the lover seeks solace in wine, to
that of admonition (maw iza) characteristic of the zuhdiyya, marked generically
by the imperatives in lines ˙ 6 and 7 and by the sentential commonplaces of
ascetic poems (consort with men of piety; beware this treacherous world; know
that all men are mortal) in 6, 7, and 8. While this segment remains linked to
the preceding one by the motifs of wine, tavern and sāqı̄, and by the transitional
line (5) which contrasts the pious hypocrites with the honest, if reprobate,
winebibbers, the voice of the implied speaker is a different, sterner one: the rind,
or libertine (a conventional persona of wine poetry which has its origins, if not
its specifically Persian manifestation, in the Arabic khamriyya), has been
temporarily transformed into the sage, the preacher of the zuhdiyya, who exhorts
to sincere piety as opposed to hypocrisy, and warns of worldly ( transience. But
there is a difference: for while Hāfiz invokes, as Abū al- Atāhiya might have
˙ ˙
done, the pious example of Noah and the image of the treacherous “abode” of
this world (khāna-yi gardūn, linked explicitly
( with Time as represented by the
turning sky), he does not – as Abū al- Atāhiya most certainly would have done –
condemn outright all those who seek worldly pleasure, but rather a select group:
those who veil their material desires beneath the guise of piety. Line 9 returns
abruptly to a topic of ghazal, the anticipated appearance of the beloved, now
given a courtly setting through the allusion to Joseph, who rose to become ruler
of Egypt after having been imprisoned (on Hāfiz’s uses of this particular allusion
˙ ˙
see Meisami 1990a). That this topic must be read, in the context of the
preceding segment, as an exemplum of the reward of virtue, rather than simply
as an anticipation of the beloved’s epiphany, is made clear by the final line,
which combines a self-addressed invitatio (invitation to drink) with a further
admonition against hypocrisy.
51
INVENTION
That the generic

( dominant in this poem is neither ghazal nor khamr but
admonition (maw iza) is clear from the central position of the admonitory
˙
segment and the amount of space devoted to it (over half the ghazal, if we
(
include its recapitulation in the maqta and the allusive reference in line 9), and
˙
from its connections (through semantic links, shared imagery and selected
generic topics) with the other generic components, which perform an
accompanying rather than a constitutive function and must thus be read as
supporting the argument that contentment with life’s simple joys is superior
to material aspirations thinly veiled by hypocritical piety. The poem’s homiletic
character makes irrelevant the glossing of the rose as “the supreme, inaccessible
symbol of the divine istighnā” (Bausani 1958: 146): while in ghazal the rose
signifies the beloved, and in panegyric the prince, in homiletic poetry it is an
emblem of the transience both of beauty and of power, a sense which
illuminates its use in this ghazal; a mystical interpretation is gratuitous. The
“extraneous” character of the Magian child is explained (were an explanation
required) by his appropriateness to wine poetry (the speaker sends his regards to
the garden from the tavern, where he is engaged in drinking and nazarbāzı̄,
˙
the contemplation of the beautiful sāqı̄); the child’s “ambiguous appearance” is,
in fact, highly conventional. The over-interpretation of such figures as the
mughbachcha, the rose and the nightingale – who in this poem at least is not
singing “invitations to partake of the mystic wine” – stems from the view that
“tradition” imposes meanings on topics and images that the poet “does not
consciously wish” (ibid.: 148); in fact, the poet, having determined those areas
of meaning he wishes to elaborate, selects his images and topics accordingly,
relying on their conventional associations to lessen the dangers of reductionist
readings, and yet manipulating them in such a way as to make his own meaning
clear.
A pre-determined mystical reading, by relying more on the application of the
“lexical code” than on a close analysis of the text in its specific historical
context, misses that important aspect of the ghazal’s meaning which the poet has
taken care to build into it from the beginning and which is crucial to its
interpretation: the courtly one. Williams calls this sort of technique the
creation of an “unspoken field”: “the ostensible field of the poem” – here, love,
and the topics associated with it –
excites a sense of the unspoken field so that the poem acquires a new
dimension that co-exists with the immediately perceptible dimension.
Seen from the poet’s view-point, it is a technique for transforming subject-
matter which, from whatever motive, he is unwilling to treat directly.
Secondly, the process is one by which the poet achieves a certain
objectivity. He insulates himself from direct involvement in the unspoken
field. . . . Thirdly, the proportionality between the two fields always turns
out to be more complex than can be described by simply naming or even
exhaustively plotting the two fields. . . . In the simplest form, the poet, in
52
INVENTION
relation to the ostensible field, has a different persona from that which he
wears in relation to the unspoken field. . . . Fourthly, there is always an
element in the poem which can be regarded as fulfilling the function of an
index of proportionality between the two fields. (1980: 189–90; see also
the quotation from Quintilian on what he terms illusio, ibid.: 191; and
compare Andrews 1985)
As we shall see, the technique is a favourite one with court poets in particular
(for obvious reasons). Now let us examine it in this ghazal of Hāfiz.
˙ ˙
The garden of the opening lines is a courtly garden, its inhabitants – cypress,
rose and basil – emblems of royalty. That the speaker is not present in that
garden, to which he asks the lovers’ messenger (the Sabā) to convey his
˙
greetings, and that he is troubled by his beloved’s cruelty, suggest that he is out
of favour, has lost his place in the courtly garden; the sāqı̄, concealing his bright
face under dark curls, emblematizes this withdrawal of favour, and may be seen
as an analogue of the prince, who has treated the poet in the same fashion. The
homiletic segment, contrasting false piety with honest love and warning that
the world is treacherous and that lofty palaces will be of no avail to those whose
last( resting-place is a handful of dust (topics again recalling the zuhdiyyāt of Abū
al- Atāhiya), shares with ascetic poems its implicit exhortation to royal justice,
but adds a second, analogous contrast between the honest rind and the pious
hypocrites who criticize him. The allusion to Noah may refer to the poet’s
protector (see Meisami 1990a: 152, 157 n. 43); while the apostrophe to the
“Moon of Canaan”, with its allusion to Joseph, also suggests not a mystical but a
topical reference (see ibid.: 151–2, 157 n. 47).34 We should not overlook,
however, the links of both allusions to the contexts of wine and love: kashtı̄
“ship” is also a type of drinking- vessel (hence the numerous poetic references to
“launching the ship” upon the sea or river of wine), Joseph a type of the ideal,
irresistibly beautiful beloved (who, moreover, repulsed – for pious motives
which have not always gone unquestioned – the love-struck Zulaykhā; cf. Koran,
Sūra 12, and see further Hāfiz’s use of this motif in the “Shiraz Turk” ghazal
˙ ˙
discussed in Meisami 1990a: 151 and in Chapter 4 below).
The final line, which reiterates the contrast between the honest rind and his
hypocritical enemies, recapitulates these links – wine-drinking, and the Koranic
“snare”, the excuse for hypocrisy. (Hāfiz’s commentator Sūdı̄ glosses: “Hāfiz, do
˙ ˙ ˙ ˙
what you will; but do not make the holy Koran a snare for hypocrisy, because
hypocrisy is unbelief [kufr]. His purpose was not to incite corruption; rather, he
says that every type of impiety in the world is bad, but hypocrisy is worse than
any” [1979, 1: 79]).
The ghazal as a whole presents itself as advice to the prince; rather than
exemplifying the “compositional principle” of incoherence, it is in fact
remarkably coherent, first grouping topics related to a particular genre in
balanced units, then recapitulating them in an enhanced context. The Arabic
or Persian poet who has such techniques of invention and manipulation at his
53
INVENTION
disposal is free to make the connections between the parts of his poem elliptical,
or to omit direct connections altogether, in the knowledge that his audience –
by expending that amount of effort all good poets expect, nay, demand, of
sophisticated audiences – will recognize at each successive stage where they are
in the poem’s progress, will relate what they hear at present to what has gone
before, and will be aware of the various generic manipulations taking place –
and, by so doing, will derive both pleasure and meaningfulness from the poem.
He is not, as it were, obliged to “start from scratch”, since familiar expressions
and markers will put the audience in the picture, will direct them towards what
they may expect from the poem (but may not always get). This does not, of
course, preclude the existence of other, more complex structural strategies;
indeed, most poems employ a variety of generic and organizational techniques
to produce their total effect, to convey their total meaning. It is to such means
and strategies that I shall turn my attention in the following chapters, beginning
with a discussion of techniques of disposition.
54

Meisam-Structure Meaning 2

Uploaded by

Copyright:

Available Formats

Meisam-Structure Meaning 2

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Meisam-Structure Meaning 2

Uploaded by

Copyright:

Available Formats

STRUCTURE AND

Julie Scott Meisami

This edition published in the Taylor & Francis e-Library, 2006.

ISBN 0-203-22055-2 Master e-book ISBN

ISBN 0-203-27545-4 (Adobe e-Reader Format)

3 Disposition: The parts of the poem 55

4 Disposition: Larger structures 111

5 Disposition: The Qası̄da and its adaptations 144

8 Ornament: Metaphor and imagery 319

9 Conclusion: The coherence of the poem 404

Arms and violent wars, with meter suited to matter,

is “historical”, as in narrative poems such as epic and romance, or generic, as in

Invention, imitation, and genre

The generic repertoire of Arabo-Persian poetry can ultimately be traced back

Most important in terms ( of the actual process of composition is the generic

Invention and genre in early Abbasid poetry

In a party where joy laughs wholeheartedly and where wine is

1 Shajānı̄ wa-ablānı̄ tadhakkuru man ahwā wa-albasanı̄ thawban mina

The self-conscious nature of this specialization on the part of both poets is

And Abū Nuwās:

1 An abode for love of which I suffered, treacherous to her lover:

8 With youths who consider being slain by pleasure a sacrifice.

fixed and predictable, often operates through the accumulation, in linear

Invention and genre in the Persian ghazal

. . . in the world of Imagination, in order to reveal its face, love sometimes

such cross-fertilization of prose by poetic imagery, developed a hermeneutics of

10 Hāfiz, drink wine, be a libertine, be happy; but

That the generic

You might also like