The text of the Divān. Perhaps the greatest progress in research on Hafez during the past century has been made in the domain of philology. Critical editions have been published which begin to provide a reliable basis for the study of Hafez’s poetry. This is not to say, however, that a textus receptus of his divān, or collected poems, is available; on the contrary, intensive investigation of the manuscript sources has revealed many discrepancies among the oldest documents, dating from the first half of the 9th/15th century, which make it difficult, if not impossible, to posit the existence of a single, definitive original. These differences pertain to the presence or absence of poems or individual couplets, the internal order of the verses, and the readings of couplets, phrases, or words. Barring spectacular discoveries of new manuscripts, it must be accepted that the establishment of a uniform version of the Divān is probably an unattainable goal. In the following brief discussion of Hafez’s poetry the second edition by Parviz N. Ḵānlari (1362 Š./1983) has been used for purposes of reference. The numbers of the poems mentioned in the article conform to this edition.

There are no indications that any major part of Hafez’s poetic output has been lost. And as the poet already enjoyed a great reputation during his lifetime, the transmission of his poems may be assumed to have been continuous. If these assumptions are correct, it seems clear that Hafez was not a particularly prolific poet. His Divān consists mainly of ghazals (q.v., lyrics of normally between 7 and 12 lines); poems in other forms—qaṣidas(odes), moqaṭṭaʿāt (occasional poems), maṯnawis(poems in couplet form), robāʿis (quatrains), etc.—are few in number and generally of less importance. In the case of the quatrains, serious doubts have been raised with regard to their authenticity (see Ḵānlari, II, p. 1094); for this reason they were omitted entirely from Sāya’s edition of 1993 (cf. his Introduction, p. 45). The total number of ḡazals that are generally accepted as genuine is less than five hundred: 495 in the edition by Qazvini and Ḡani, 486 in Ḵānlari’s 2nd edition, and 484 in Sāya’s edition.

Hafez and the azal tradition. Many features of Hafez’s ḡazals have been pointed to as original contributions. By Hafez’s time, the ḡazal already had a long history that can be traced back for at least two centu-ries (see ḠAZAL). It would be difficult to name any individual element of Hafez’s poems, either formal or thematic, which cannot be attested in the works of his predecessors.

According to Šebli Noʿmāni (V, p. 43), Hafez was the first to extend the range of the ḡazal, which so far had been mainly devoted to erotic themes, so as to include the treatment of ethical, philosophical, mystical, homiletic and even political subjects, while keeping intact the lyrical idiom of the genre. Šebli (p. 55), and many other critics after him, also regarded the complex of imagery and motifs centred upon the figure of the “debauchee” (rend) as an important characteristic of his poetry (see, e.g.,Ḵorramšāhi, pp. 27-28, 403-13). In the mid-20th century, the focus of attention in the interpretation of Hafez, in which hitherto a mystical reading had prevailed, shifted to the social context of his poetry. Inspired by the historical studies of Qāsem Ḡani, R. Lescot (1944, p. 59) regarded the use of the ḡazal as a poem of praise, evident from many overt or veiled references to patrons, as an innovation by Hafez through which he assigned to this form a function formerly reserved for the qaṣida.

Perhaps Hafez may have been less an innovator than the poet who brought an already well-established tradition to its highest point of perfection. It cannot be denied that his poetry makes a striking impression of newness, but it is difficult to express this in more precise terms. Tentatively, the following two points may be mentioned by way of a general characterization: first, the greatly increased density in the use of the various elements which the tradition of the previous centuries had delivered into his hands, and second, a stylistic and rhetorical virtuosity unmatched by any other ḡazal writer.

In order to evaluate Hafez’s personal contribution to the development of theḡazal, his debt to his predecessors, and even to his contemporaries who cultivated the same genre, must be taken into proper account. A proper assessment of his originality can only be attempted when all possible influences on his work have been examined (see in particular Ḵorramšāhi, pp. 40-90). This is however beyond the scope of the present article.

Modern scholars have attempted to identify certain conventions according to which what has been called the “technical ḡazal” (ḡazal-e eṣṭelāḥi), a term introduced by A. M. Mirzoev (cf. ḠAZAL, p. 354), ought to be composed. These conventions consist of a prescription for the poem’s length (though not very precise), the use of the poet’s pen name (esm-e taḵalloṣ or simply taḵalloṣ) in the last or the penultimate line as his signature, and a few other prosodic features (such as monorhyme and rhyme in the opening couplet) which have become typical of this poetic form, though they were never exclu-sive to it. By the 8th/14th century, deviations from this pattern had become exceedingly rare; however, these “rules,” so prevalent in poetic practice, were never properly discussed by writers on poetics, who continued to use the term ḡazal in its generic sense of “love poetry.”

Prosodic features. The majority of Hafez’s ḡazals (366, or ca. 75 percent of his output [based on Ḵānlari’s edition]) are between seven and nine couplets in length. There is a small number on both sides of shorter or longer poems. The shortest are four ḡazals of no more than five couplets (Ḵānlari 96, 104, 144, 420, 444). Poems of more than twelve couplets are exceptional: there are five of thirteen couplets (12, 149, 251, 454, 464), three of fourteen (425, 443, 480) and only one (354) of sixteen. Beyond the fact that many of the longer-than-average poems contain a panegyric extension, there are no distinctive features that would allow a classification into types based on the length of the poems only.

According to the statistical survey provided in Elwell-Sutton’s Persian Metres (based on the Qazvini and Ḡani edition), Hafez used twenty-three different metrical patterns in all, but in a very unequal proportion. Only eight patterns occur in 477 poems (98percent), the three most frequent being ramal-e maḵbun-e maḥḏuf (143 ḡazals, or 29percent), mojtaṯṯ-e maḵbun-e maḥḏuf (128 ḡazals, or 26 percent), and możāreʿ-e aḵrab–e makfuf-e maḥḏuf (75 ḡazals, or 15.4 percent). These three meters have all four feet to the half verse (meṣrāʿ), which amounts to fourteen or fifteen syllables.

The radif, the addition to the rhyme of a morpheme, a word (a verb, noun, adjective, pronoun, particle or adverb) or a short phrase repeated at the end of each couplet, is found in most poems. Verbs are particularly frequent and occur in many different forms of conjugation. Some verbs with a particularly wide semantic spectrum have determined to a large extent the choice of motifs and images in the ḡazal (e.g.,17, where the radif andāḵt “threw” is used to display a pattern of aggressive behavior on the part of the beloved; Ḵ 18, in which besuḵt “burned” sets the tune for variations on the theme of fire). Some nouns used as radifs are also of special semantic interest because they belong to the repertoire of terms specific to the theme of love, such as abru “eyebrow” (404), čašm “eye” (331), dust, “friend” (61, 62, 63), ferāq “separation” (291), ḥosn “beauty” (386), and šamʿ “candle” (289). Remarkable from a thematic point of view are further: ḡarib, “stranger, strange, poor and homeless” (15), which throughout the poem stresses the position of the uprooted and destitute lover begging for the attention of his beloved, the “sultan of the beautiful ones”; and the word darvišān, used as the main part of a radif in a poem praising the virtues of the “dervishes” within the context of the panegyric of a vizier (50).

Taḵallosá. The use of the poet’s pen name (taḵallosá) is an almost universal feature; it is absent in no more than eleven ḡazals, a number of which are very brief drinking songs (Ḵānlari 13, 104, 109, 140, 470). The last line (maqṭaʿ or maḵlaṣ) is the most common position for the taḵalloṣ; however, it occurs also in the penultimate couplet (37 poems), more rarely in the antepenultimate (six poems), and in both cases the line often contains a panegyric reference. Five poems where the taḵalloṣ occurs several lines before the end (12, 149, 354, 443 and 453) are panegyric ḡazals in which the device serves as a marker of transition to the section of praise (taḵalloṣ in the original meaning of the term). In one instance the poet’s name is mentioned in the opening couplet (280, where ḥāfez is used in its literal sense of someone who knows [the Koran] by heart, probably is ambiguous as the pen name, which is otherwise not mentioned in the poem); in four others it occurs twice in one poem (251, 285, 309, 431; in the last-mentioned instance this is probably due to textual variants).

Although the pen name may be introduced as a vocative or as a reference to a third person, it is usually the speaking voice of the poet to whom the address can be attributed. Sometimes other voices are involved; for example, the beloved may address the poet/lover, often in the context of a dialogue (e.g., Ḵānlari 5, 68, 193, 227, 266 and 330); personified entities may have, as it were, the last word (e.g.,194: Reason extols the poet’s verse; 382: Reason bids the poet drink wine; 380: the wind exhorts the poet to sing of wine and sweet-mouthed beauties). Remarkable twists on the pen name include the modest banda Ḥāfeẓ “your humble servant Hafez” (341), used probably in a panegyric context, and the affectionate but mildly satirical Ḥāfeẓ-e mā “our Hafez” (46), playing on the literal meaning of ḥāfeẓ. This play on the literal sense of the poet’s pen name—“he who has the Koran by heart—”either directly (e.g., 250: “Hafez . . . grieve not, as long as your litany is prayer and study of the Koran”) or indirectly (e.g., 93: “Love will come to your aid if you, like Hafez, recite the Koran by heart in fourteen versions “rewāyat,” is a marked feature of Hafez’s ḡazals. The religious connotation allows him to exploit the taḵalloṣ in the treatment of one of his favorite themes: the paradox of piety and antinomianism in his own conduct (e.g., 344: “ I am a ḥāfeẓ in [religious] gatherings, a drinker of dregs at parties;” 347: “Whether I am the libertine of taverns or the ḥāfeẓ of the town . . .”). The possibility of internal rhyme is seized upon to contrast Hafez with a typical representative of the religious establishment, the wāʿez “preacher” (e.g., 83: ʿAyb-e Ḥāfeẓ makon ey wāʿeẓ, “Preacher, don’t find fault with Hafez,” 127: Ḥadiṯ-e ʿešq ze Ḥāfeẓ šenow na az wāʿeẓ “Hear the tale of love from Hafez, not from the preacher”). With ironic ambiguity he inserts the pen name in a catalogue of other representatives of false piety (195: “The shaikh, the ḥāfeẓ, the mufti and the moḥtaseb [enforcer of morals] . . . all practice hypocrisy”).

The use that Hafez makes of the taḵalloṣ in his ḡazals is essentially not different from that of other poets. The basic function of the device is to provide the poem with a signature and an elegant conclusion for which a separate motif could be chosen, and which need not be particularly connected to the subject matter of the poem (Arberry’s “clasp theme,” cf. “Orient Pearls,” pp. 706-7). This is made explicit, for instance, in 393 (Ḵatm kon Ḥāfeẓ “Conclude, Hafez”). A very common motif for winding up the poem is the expression of faḵr, “professional pride or conceit” in his own poetical skills or the success of his poetry. A few examples are: 37 (“Why, poetaster, do you envy Hafez much?”), 42 (“You conquered Iraq and Pārs with your poetry”), 202 (“Already in Adam’s time Hafez’s poetry adorned the pages in the album of the roses in the garden of Eternal Paradise”), 445 (“Come to our gathering so that you may learn from Hafez how to make āgazal”). In one instance he compares himself to contemporary poets (251: Ḵvāju, Sal-mān, Ẓahir).

Whenever the taḵalloṣ is a conclusion in which the poet recapitulates the central theme of the poem, he is very often admonishing himself, and occasionally commanding himself to silence (e.g., 300: “Suffer the pain of love and be silent, Hafez”). Being divorced from his beloved, he exhorts himself to sing poems about separation (451: ḡazalhā-ye ferāqi), or begs the beloved to return so that he may come to life again (23). He also addresses the assembled friends (41), the cupbearer (6) or the singer (169: “Sing āgazal from Hafez’s poetry”). In one poem he refers to his own grave (230: “When the wind of your presence passes over Hafez’s tomb / one hundred thousand tulips bloom from the dust of his body”).

The conclusion of āgazal may also be a declaration of the poet’s adherence to “debauchery” (rendi) and to the cult of the “ruins” (ḵarābāt), e.g., 43 (šeʿr-e rendāna), 322 (“Hafez became renowned for his debauchery”), which sometimes introduces a brief panegyric (279, 281, 347).

Language and rhetoric. Hafez is particularly renowned for his refined but quite natural use of language and his effective application of rhetorical devices. Within this brief survey, this important aspect of his poetry can only be mentioned in passing. To begin with, he is a learned poet whose familiarity with both Arabic and Persian becomes clear both through his use of tażmin (the insertion of quotations in both languages) and by his own composition of macaronic poems (molammaʿāt) in which Arabic and Persian verses alternate (see 429, 451-54, 460; also 13, āgazal devoted to the morning drink (ṣabuḥ) with three Arabic hemistichs). The Arabic insertions include verses from the Koran, pious maxims and proverbs, as well as lines of poetry. Most of these are incidental inserts in Persian poems. However, in some instances, a structural purpose can be discerned, as for example when they open and close a poem (1), when they are connected with the taḵalloṣ (443, 475), or are part of a panegyric address (463).

Without the analysis of extensive examples, Hafez’s use of rhetorical devices can only be treated briefly. At least one, and usually more, of the standard devices of sound and meaning can be detected in almost every line of his poetry. These devices include harmony of imagery (tanāsob), e.g.,in 40/1, where the poet describes his love as a garden in which the cypress (sarv) and the fir-tree (ṣanawbar) are rejected in favor of his “home-grown box-tree” (šemšād-e ḵānaparvar); the contrast between “garden” and “house” in the same couplet provides an instance of antithesis (moṭābaqa). Hafez uses various types of word-play such as paranomasia (jenās), e.g., bādapeymā (“drinking wine”) / bādpeymā (“travelling”) in 5/5, or Ṣufi / ṣāfi (“pure”) in 7/1, and amphibology or double entendre (ihām), e.g.,where hawā is used in the sense of “air” in a context where the other meaning “desire, love” (227/5) would not be inappropriate or, involving a harmony of imagery (ihām-e tanāsobi), “Adam abandoned (behešt) the Mansion of Peace,” i.e., Paradise (behešt), in 7/4. Euphonic effects such as internal rhymes and alliteration are extremely frequent and are often seen as one of the most remarkable features of his poetical language. Hafez also became famous for his use of irony, which adds to the allusiveness of his expression. It is especially in evidence whenever the poet takes aim at hypocrisy, e.g.,when he introduces a drunken faqih or doctor of holy law, who gives a ruling (fatwā) to the effect that “wine is forbidden but it is better than [appropriating] the property of orphans” (45/4) or states that “an animal does not become human for not drinking wine” (220/2).

Thematics and imagery. Even before Hafez’s time, the ḡazal was already a complex poem as far as its thematic material was concerned. The genre of ḡazaliyāt combined various poetic themes as they had been current for centuries in Arabic and Persian literature. The most striking subjects of Hafez’s poems are, on the one hand, the trials and tribulations of the poet-lover (taḡazzol), combined with a defiant celebration of drinking (ḵamriyāt), and on the other, spiritual themes such as the reflection of the transcendental in earthly beauty, centered upon the concept of the beautiful person as a “witness” (šāhed) of eternal beauty, criticism of conventional Sufi piety, derived from the tenets of malāmati (self-deprecating) mysticism, and the renunciation of this world (zohdiyāt). Of these themes, perhaps none, with the possible exception of the poet’s description of love and beauty, so much engages the poet’s passion more than his intense desire to unveil the hypocrisy of the figures of the clerical establishment, such as the preacher, the mofti, the judge, the enforcer of public morals, and more particularly, the one who poses as an ascetic (zāhed). He applies the same passion and insistence to uncovering the duplicity of the materialistic and corrupt Sufi establishments. The contrast between these topics is further enhanced by their often being treated together within the framework of a single poem. Moreover, they may also be combined with a wide range of subsidiary themes: the vernal flower of nature, complaints about the cruelty of fate and the transience of this world, praise of himself, his city, and his province and other references to the poet’s environment and the events of his time, and extolling of his own poetry. All this gave the ḡazals an appearance of disconnection, and often makes it difficult to ascertain what the main subject of a given poem really is.

To his stock of images also belong the exempla drawn from the spheres of history, mythology and literature, and the motifs drawn from the sciences and various other spheres of life. The poet refers many times to stories about Solaymān: i.e., Solomon, the Biblical king who is treated in Persian literature as a legendary monarch—his knowledge of the language of the birds, the hoopoe as his messenger to the queen of Sheba, the ant who brought him a modest gift, the demon who stole a magic ring and ruled for a while in his place, his vizier Āṣaf; the disappearance of all this splendor makes him also into a suitable metaphor for the transience of worldly power (20/7). In this respect he equals Eskandar (i.e. Alexander 285/5), but to Hafez the latter is of special interest as the owner of a miraculous mirror, compared to the cup of wine (5/5, see also 145/8, 174/1), and because of his failed attempt to find the fountain of life (268/3, 402/7). The clearest allusion of a literary kind is to famous love stories such as “the burden on Majnun’s heart and the curls in Layli’s locks” and “Maḥmud’s cheeks and the sole of Ayāz’s foot” (41/6). The incidental mention of shaikh Ṣanʿān (79/6) is possibly a reference to the famous story in ʿAṭṭār’s Maṇteq al-ṭayr.

The social setting which is either referred to, or implied, in many poems, is often the convivial circle (majles) of friends, who are sometimes addressed as the majlesiān (41/9); or it may be a courtly feast or drink-ing party, often set in a garden (e.g.,477). Among those present at such gatherings, the singer/musician (moṭreb) and the cupbearer (sāqi) are often singled out for mention, or addressed directly. The sāqi in particular is sometimes identified with the beloved, although his primary role is to comfort the suffering lover with wine. In some cases, the ḡazal is little more than a drinking song. To the call for wine a motivation for becoming drunk is usually added, for instance the need of a cure for the problems and woes of love, the wish to forget the inexorable progress of time, or the return of spring promising the fulfillment of love. Other persons at the gathering (or, if not present, important in the context of the dramatis personae of the ḡazal) are also addressed or referred to: the moddaʿi “pretender,” the poet’s rival in poetry and in love (e.g.,34/9-10); the raqib, “warden (of the beloved), challenger,” who is a constant obstacle in the way of the poet’s access to his beloved, and the ḡammāz, “telltale,” who conspires to expose the poet’s secret (89/7); religious figures (wāʿeẓ, mufti, faqih, judge, moḥ-taseb, and so on); and, in the panegyric context, the prince or other patrons. (On the dramatis personae of the ḡazal see Meisami, 1987, pp. 265-70, 295-96; Meisami 1991).

The ḡazals abound in very short and sketchy narrative passages, used in particular in the opening lines of a poem. For instance, in āgazal the beloved is described as a nightly visitor coming to the poet’s bedside “with disheveled hair, sweating, laughing and drunk, wearing a torn shirt, singing āgazal and holding a bottle in his hand” (22), in another as someone who appears among the drinkers in a “convent of the moḡān,” a metaphor for the taverns, since Zoroastrians, unlike Muslims, had no prohibition against wine drinking (23). In the garden the poet encounters a nightingale, which utters a complaint while holding a rose petal (79). Several times he says that he went out to seek the advice of the pir-e moḡān in the tavern (e.g.,136). Indications of time are added often to such scenes: they are marked as reports on an experience during the preceding night (duš), for instance when we are told that “last night I saw angels knocking on the door of the tavern” (179), or they are said to take place in the early morning (e.g.,13, especially devoted to the morning drink). However, these sketchy stories are never a goal in themselves, but merely introduce a discourse which is usually of a reflective or even paraenetic kind. Many of the unusual attributes of the “beloved” can be understood by recalling that in the classical lyric poetry the image of the poet’s sweetheart refers more often than not to a male figure, normally a youth (see BELOVED, and Yarshater 1953, p. 19).

The mode of presentation in the ḡazals is equally varied. Most often the speaking voice in a poem can be identified with the poet, in one or another of his personae: poet-lover, libertine, sage, seeker of wisdom (see Meisami, 1990). Occasionally, a dialogue between lover and beloved may encompass an entire poem constructed on the “figure” of “question and answer” (soʾāl o jawāb, e.g.,in 193 and 227, in 266 only in the conclusion of the poem), known from the earliest Persian poetry. Other persons or personifications also converse with, advise or exhort the poet, for instance in 37, where an angel from the hidden world (soruš-e ʿālam-e ḡayb), a mystic mentor (pir-e ṭariqat) and a beggar admonish him in succession. A mysterious voice (hātef) is often heard (e.g.,91/6). Another common device is to put the wisdom imparted into the mouth of the pir-e moḡān (the old man of the magi, i.e., the tavern keeper). The poet entreats the wind (bād, ṣabā) or the hoopoe (hodhod; both associated with the legend of Solomon and queen of Sheba) to carry a message to his departed beloved, or bring him the scent of the beloved’s perfume (see, respectively, 1/2 and the opening lines of 91 and 170).

It is abundantly clear that Hafez’s poetry cannot be properly understood without considering its close association with his historical environment. The glorification of his home town and its pleasure grounds, like Rokn-ābād and Moṣallā (3/2, 97/9 and 274/2), are instances of city panegyrics, known also from the ḡazals of earlier, especially Shirazi, poets, such as Saʿdi. The frequent references to persons of political or social importance prove beyond doubt that, as a poet, Hafez was involved in courtly life and was in all likelihood dependent on the patronage of the rich and powerful. The use of the ḡazal for panegyric was not, however, an innovation by Hafez, but is already seen two centuries earlier in the ḡazals of poets of the Ghaznavid courts of Ghazna and Lahore (see Meisami, 1987, pp. 275-76).

Nowhere is the hermeneutical problem of Hafez’s poetry more acute than in the case of the antinomian stance frequently adopted by the poet. In this respect, too, he continues a tradition with which earlier poets had enriched the palette of the ḡazal. The complex of motifs centered around the figure of the “tramp” (qalandar) is already a predominant element in the poems of Sanāʾi, ʿAṭṭār and ʿErāqi, who were all mystical poets of the twelfth and thirteenth centuries. They used it as a forceful metaphor in denunciation of false piety in sufism and in adhortations to a radical renunciation of the world. Its main features were the celebration of intoxication and debauchery and the proclamation of a non-Muslim cult, an imaginary mixture of Zoroastrian and Christian elements. The thought behind all this was that, at a higher level of piety, it becomes necessary to hide one’s spiritual progress behind a screen of sinful behavior so as to avoid the social respect that could so easily be won by a show of piety. The latter was seen as one of the most dangerous pitfalls on the path of the advanced mystic. He should guard himself by seeking the condemnation of the people rather than their veneration. Initially, this was probably no more than a literary ploy of which preachers and mystics availed themselves in their admonitions, but from the thirteenth century onwards qalandar is also known as the appellation of a dervish practicing extreme forms of asceticism and living in a group of the like-minded. (On the qalandars and other practitioners of socially deviant mysticism, see Ahmet T. Karamustafa, God’s Unruly Friends: Dervish Groups in the Islamic Later Middle Period, 1200-1550, Salt Lake City, 1995; on the qalandariyāt in Persian poetry, see Ritter, 1959 and de Bruijn, 1992).

Hafez frequently poses as a rend (“debauchee”), the term which he prefers to qalandar, although it is evident from the expression rendān-e qalandar (479/3) that he refers to the same figure. The rend despises conventional piety as mere hypocrisy (riā), and seeks a refuge from the mosque and the cell (ṣawmaʿa) of the ascetic in a tavern (mey-ḵāna) or “ruined places” (ḵarābāt) of ill repute (e.g.,54, 160, 171). He not only prides himself in his intoxication (masti), but even regards this as his predestined fate to which he is bound since the day of the pre-mondial covenant (ruz-e alast, e.g.,107/5, 144/5). He refers to the alternative rite he adheres to under various names: the doctrine of the ʿayyārs (149/5), of the pir-e moḡān (193/6), of the Zoroastrian religion (178/8), or of love (119/7); but he also calls it just “our doctrine” (maḏhab-e mā; e.g.,47/3: it only forbids the drinking of wine when the beloved is not present). The tavern is the place where the rite is celebrated and where the pir-e moghān, “the elder of the Magi” (e.g.,70, 154, 335) acts as a spiritual guide who reveals esoteric wisdom about the world by looking into the jām-e Jam, the beaker which by legend is ascribed to the mythical Iranian king Jamšid (136/4-5). This vessel, comparable to the Holy Grail of Christian lore, is the object of a life-long quest by the poet and is in the end only to be found in his own heart (136, 137). An erotic trait is added to the imagery when a “youngster of the Magi” (moḡ-bačča, qualified as a “seller of wine” (bāda foruš)) appears, who lures the poet into the tavern when the return of spring invites to drinking and making merry (9/3). On the opposite side are those whom the poet reproaches that “they have given me the reputation of a rend” (409/3) and whom he ridicules and scolds for being hypocrites. They are the representatives of the established religious order, including the doctors of Islamic law (faqih), the preacher of conventional piety (wāʿeẓ) and the inspector of public morality (moḥtaseb), but also the Sufis and the ascetic recluses.

Hellmut Ritter, who examined a large number of Hafez’s poems with qalandari motifs and compared them with similar poems by ʿAṭṭār and Sanāʾi, concluded that Hafez was not really a mystical poet but merely a rend-mašrab: “like the qalandar, he does not withdraw from the pleasures of the world, mocks those who renounce the world and their kind, excusing his scandalous way of life by pointing to predestination, and for the rest puts his hope, in the manner of popular piety, in God’s great mercy” (p. 56). This interpretation supports the view of those modern critics who see in the ḡazals of Hafez an immediate reflection of the life he must have led in fourteenth-century Shiraz. Further support for this view could be found in the frequent mention of a relationship with secular patrons and other elements which unequivocally point to the poet’s social and political environment.

On the other hand, it should be borne in mind that ever since the twelfth century antinomian motifs had been used in a figurative sense by Persian poets. It is difficult to assume that Hafez would have reversed this unmistakable trend in the development of poetic expression, which in his own time even tended towards a complete allegorization of this imagery. One of the problems standing in the way of reaching a conclusion in this matter is the lack of reliable information concerning the life of the poet, which would enable us to determine with what orientation his ghazals were written. The fact that his gha-zals were from a very early date interpreted as mystical poems must be taken seriously, even if the interpreters of this persuasion often disqualify themselves through their zeal to disclose transcendental meaning in every word.

Another issue that was much debated, especially in the second half of the twentieth century is the internal coherence of his poems. Particularly widespread is the view that Hafez brought about a revolutionary change in the genre by giving a much greater independence to the individual couplets than earlier poets, and by treating more than one theme within a single poem. A. J. Arberry argued that Hafez was searching for a new concept of the ghazal after the conventional form had been exhausted artistically—notably by Saʿdi—and devised a “thematic technique,” which means “that he constructs each lyric upon the basis of a limited number of themes selected from a repertory which is itself definitely restricted, and to a great extent conventional” (“Orient Pearls,” pp. 704-5). The discussion which Arberry’s theory started was continued by (among others) G. M. Wickens, R. Rehder (see Bibliography)and in particular Michael Hillmann, who in his monograph described a number of different structural types in Hafez’s Divān. The debate has so far not resulted in a consensus.

Through frequent changes of images and motifs, Hafez creates the illusion that he also changes his subject, which may not in fact be the case. The proper analysis of the typical Hafezian ghazal demands, therefore, a fine distinction between the things expressed and the manifold ways used to express them. The poet’s kaleidoscopic deployment of a rich and complex imagery often creates the impression of a great amount of independence for the individual couplets, and the absence of an overall encompassing structure. This may, indeed, have been one reason for the variation in the order of the couplets in different recensions that has been brought to light by philological analysis.

There can be no doubt, therefore, that the use of multiple images to illustrate a single theme is a genuine aspect of Hafez’s poetic art. A good example is provided by the alternative images used in the first four lines of a poem for the beloved, who is first a gazelle, then a seller of sweets, a rose, and a hunter of birds (4). In the same poem the lover appears under the guise of a tramp, a parrot, a nightingale, and a “wise bird.” The public for whom the ghazals were composed was challenged with the reconstruction of a total meaning from this wealth of suggestive details. The modern interpreter of Hafez should realize that this audience was familiar with the repertoire drawn upon by the poet, and was therefore much better equipped for the reception of this poetry than we are today.



Ḥāfeẓ, Dīvān, ed. Moḥammad Qaz-vini and Qāsem Ḡani, Tehran, 1320 Š./1941; ed. Parviz Nātel Ḵānlari, Tehran, 1362 Š./19832; ed. Sāya (Ḥušang Ebtehāj), Tehran, 1372 Š./1993.

A. J. Arberry, “Orient Pearls at Random Strung,” BSO(A)S 11, 1943, pp. 688-712.

Idem, Fifty Poems of Ḥāfīzá, Cambridge, 1953. Idem, “Three Persian Poems,” Iran 1, 1963, pp. 1-12.

Alessandro Bausani, “The Development of Form in Persian Lyrics: A Way to a Better Understanding of the Structure of Western Poetry,” East and West, N.S. 9, 1958, pp. 145-53.

Idem, EI2, s.v. Ghazal ii. Mary Boyce, “A Novel Interpretation of Hafiz,” BSOAS 15, 1953, pp. 279-88 (a refutation of Wickens’s article).

J. T. P. de Bruijn, “The Qalandariyyāt in Persian Mystical Poetry,” in L. Lewisohn, ed., The Legacy of Medieval Persian Sufism, London and New York, 1992, pp. 75-86.

Daniela Meneghini Correale, The Ghazals of Hafez: Concordance and Vocabulary, Rome, 1988.

Laurence P. Elwell-Sutton, The Persian Metres, Cambridge, 1976.

Qāsem Ḡani, Baḥṯ dar āṯāro afkāro aḥwāl-e Ḥāfezá, 2 vols., Tehran 1321-30 Š./1942-51.

Michael C. Hillmann, Unity in the Ghazals of Hafez, Minneapolis, Minn., 1976.

Bahāʾ-al-Din Ḵorramšāhi, Ḥāfeẓ-nāma, 2 vols., 7th ed., Tehran, 1375 Š./1996.

Roger Lescot, “Essai d’une chronologie de l’oeuvre de Hafiz,” Bulletin d’études orientales 10, 1943-44, pp. 57-99.

Julie Scott Meisami, Medieval Persian Court Poetry, Princeton, 1987.

Idem, “Persona and Generic Conventions in Medieval Persian Lyric,” Comparative Criticism 12, 1990, pp. 125-51.

Idem, “The Ghazal as Fiction: Implied Speakers and Implied Audience in Ḥāfiẓ’s Ghazals,” in Michael Glünz and Johann-Christoph Bürgel, eds., Intoxication, Earthly and Heavenly: Seven Studies on the Poet Ḥāfiẓ of Shiraz, Bern, 1991, pp. 89-103.

Idem, Structure and Meaning in Medieval Arabic and Persian Lyric Poetry: Orient Pearls, Richmond, UK, 2002 (in press).

Robert Rehder, “The Unity of the Ghazals of Ḥāfiẓ,” Der Islam 52, 1974, pp. 55-96.

Helmut Ritter, “Philologika XV: Farīduddīn ʿAṭṭār. III. 7. Der Dīwān,” in Oriens 12, 1959, pp. 1-88.

Hans Robert Roemer, “Probleme der Hafizforschung und der Stand ihrer Lösung,” Abhandlungen der Klasse der Litteratur, Akademie der Wissenschaften und der Litteratur, 1951, pp. 97-115.

Šebli Noʿmāni, Šeʿr al-ʿAjam, Persian translation by M. T. Faḵr-e Dāʾi Gilāni, Tehran, 1318 Š./1939.

G. M. Wickens, “An Analysis of Primary and Secondary Significations in the Third Ghazal of Ḥāfiẓ,” BSOAS 14, 1952, pp. 627-38.

(J. T. P. de Bruijn)

Originally Published: December 15, 2002

Last Updated: March 1, 2012

This article is available in print.
Vol. XI, Fasc. 5, pp. 469-474