Rewriting Hafez: Re-theorizing Untranslatability in Persian Poetry
July 8, 2014
A guest post by Aria Fani. A native of Shiraz, Aria Fani studies towards a Ph.D. in Near Eastern Studies at the University of California in Berkeley. All translations are by the author unless otherwise noted. ariafani@berkeley.edu
و الشعر لا يسطاع ان يترجم و لا يجوز عليه النقل. و متی حول تقطع نظمه و بطل وزنه و ذهب حسنه و سقط موضع التعجب منه و صار كالكلام المنثور. و الكلام المنثور المبتدأ علی ذلك احسن و اوقع من المنثور الذي حول من موزون الشعر– الجاحظ | كتاب الحيوان
Poetry [shi‘r] is untranslatable; it cannot be transferred from one language into another. Translation breaks its metrical arrangements [naẓim] and spoils the rhythm [wazin], ruins its aesthetics [ḥusn], and flattens the element of wonder [muḍi‘ al-ta‘ajjub]. Translation turns poetry into prose, and prose originally written as such is preferred over what has been turned into prose as a result of translating verse– al-Jaḥiẓ | Kitab al-Hayawan.
Near Eastern thinkers, perhaps as early as al-Jaḥiẓ, the eighth century Arab writer and polemicist, have commented on the problem of literary translation, particularly concerning poetry. While al-Jaḥiẓ points to broad challenges encountered in translating lyrical poetry, he gives more weight to translation as an end result, characterized by various layers of loss consequent to this linguistic transfer. One implication of reading translation merely as a product is to indict translation as an intellectual enterprise—as often has been done—and place it within the reductionist model of “betrayal” versus “faithfulness.” However, al-Jaḥiẓ’s view also lends itself to the notion of developing a critical awareness towards translation as a process of linguistic and cultural negotiation between the source and target language and their literary traditions. In other words, literary translation does not end by acknowledging the concept of untranslatability, it begins with it and furthermore testifies to the translator’s difficult task of doing justice to what he or she perceives as the poem’s internal vision.
Untranslatability is a trope standard to writing and rewriting, and is not limited to translation. In this vein, one can view failures, disappointments, and success not as a fixed form of assessment, but as an open discussion of the translator’s vision and accommodations within the fluid context of his or her era, language, target audience, and mode of reading. Untranslatability can be a humbling acknowledgement made by a translator well-acquainted with the poetic system of the source language, one who fully understands components such as naẓim, wazin, ḥusn, and muḍi‘ al-ta‘ajjub, prior to translating the work. “Untranslatable” could indicate that no translator has yet engaged this process for a “general” audience, or existing translations have not found wide readership. It is translation as a process of negotiating between two (or more) literary traditions, models of reading, and target audiences that I am concerned with in this article, and less with its end result.
Translation can be characterized as an interplay between literary traditions (source and target languages are rooted in multiple traditions of their own), a process that illuminates the difference in approaches to and articulation of poeticity. Consequently, linguistic and cultural challenges arise that need to be addressed by the translator regardless of his or her approach to translation. The Persian literary tradition presents unique challenges that are particularly well-revealed in the ghazaliyyāt of Shams al-Din Muhammad Hafez of Shiraz (circa 1315-1389), a poet who is widely read in Persianate societies. Poets and scholars alike have expressed the difficulty of translating Hafez.
In his famous essay, “On Not Translating Hafez” (New England Review, 1999), Dick Davis engages the many facets of translating the Persian poet, maintaining at the end that his verse is untranslatable. Davis, who has now attempted the feat himself in his Faces of Love: Hafez and the Poets of Shiraz (Penguin Books, 2013), characterizes these challenges broadly as linguistic and cultural. Linguistic challenges are often in the form of translating idioms, puns, and rhymed phrases. Cultural resonance, embedded in the source text, constitutes even greater weight for the translator; the ghazals of Hafez are composed against the backdrop of an intertextual dialogue with various dimensions of Islam, a background readily accessible to certain Persian readers and largely missing in cultural and linguistic communities in the West.
As an example of cultural resonance, Davis uses wine metaphors in the poetry of Hafez to further unpack this challenge. He writes: “It would never occur to a Western poet to express the forbidden intoxications of mysticism by alluding to the forbidden intoxications of wine, for the simple fact that the intoxications of wine have never (if we exclude the brief and local moment of prohibition in the United States) been forbidden in the West.” The Hafezian metaphor resides in the realm of culture, not words or images; it is a capacity he exploits throughout his poetry.
Similarly, in an essay entitled dar tarjumah napaziri-i shi‘r [On the Untranslatability of Poetry, Iran shinasi, 2002], Mohammad Reza Shafi‘i-Kadkani highlights the difficulty of communicating the semantic complexity of Hafezian metaphors while arriving at the same conclusion: Hafez is untranslatable. Shafi‘i offers a beyt (line) as case in point, and zeroes in on one phrase of a single misra‘ (hemistich). Here I will unpack the entire beyt, partially relying on the commentaries of Shafi‘i and Baha al-din Khurramshahi [ḤāfiẓʹNāmah, Shirkat-i Intishārāt-i ʻIlmī va Farhangī, 2005]:
Transliteration:
bih may sajādih rangin kun garat pir-i mughān guyad
kih sālik bikhabar nabvad z-i rāh-u-rasm-i manzilhā
The literal translation of this ghazal, closely approximating the Persian syntax, is as follows: “with wine prayer rug make colorful, should pir-i mughan ask /// for salik oblivious should not be to the path and way of Homes.” Reproducing the rhyme and meter of the Hafezian ghazal—a quality intimately associated with classical Persian poetry as with all poetry—is very challenging, one that most translators of Hafez decide not to undertake. The first misra‘ directly addresses readers in second person singular, and is grammatically conditional (do “x” if one were to ask you). Its central concepts are wine, prayer rug, and pir-i mughan: Wine, evoked as an impure element, is positioned against the cleanliness and purity of the prayer rug, a small carpet placed between the bare ground and the Muslim worshipper. Furthermore, both prayer rug and wine can be read as metonymy for orthodox religiosity and the forbidden respectively.
Pir-i mughan is principally a Zoroastrian elder who sells wine, and has later gained different semantic layers through Sufi teachings. However, in this context, the earlier referent is being evoked. The second misra‘ is grammatically ambiguous for nabvad is both imperative and indicative; some have translated nabvad as an opinion or piece of advice (Pilgrims should show each stage’s rule), while others have rendered it as a statement (That seasoned voyager knows the ways of the road). In both translations, the ambiguity of the Persian is necessarily removed.
Salik is approximated by wayfarer or seeker—one who embarks on a journey. Salik is deeply rooted in Sufi traditions, a protean and transnational phenomenon that broadly captures the mystical dimensions of Islam.Rah-u-rasm refers to the path and the way of mystical principles; although grounded in a different tradition, Christian poetry makes use of these concepts as well;one such example is “Enter not into the path of the wicked, and go not in the way of evil men” [Proverbs 4:14, KJV]. Rah-u-rasm has been translated as each stage and the ways respectively by Dick Davis and Peter Avery and John Heath-Stubbs. Last, but not least, the concept of manzil is approximated by the “realm of the divine,” sought by the seeker. Thus far, our rough draft appears as follows; the Persian terms are retained in the English to further invite non-initiated readers to research these concepts:
Make your prayer rug colorful with wine, if pir-i mughan asks you
for salik shouldn’t be oblivious to the way of the higher path.
Having annotated the linguistic aspect of the source text, the translator arrives at a different problem: how does one uproot the poem from its cultural soil and replant it in an unaccustomed earth? This is an area in which a literal translation may not prove to be a viable option. For instance, “make your prayer rug colorful with wine” accurately reflects the Persian, but fails to convey the subversive message of the ghazal’s speaker for English readers unfamiliar with Perso-Islamic culture. Hafez’s metaphor functions through a source domain—readily accessible for Persian readers who are familiar with Qur’anic and pre-Islamic lore and intertextuality—in which wine is conceptualized as najis, or impure, and prayer rug as tahir, or free of impurity.
The concept of najis/tahir is grounded in a cultural understanding of cleanliness, partially informed by sharia, one that is not closely approximated by the English terms clean/unclean.Clean for most English-language readers may simply mean free from dirt, marks, or stains, while pure–free from a contaminating substance–is closer in meaning, all the same fails to convey the cultural significance of tahir in Perso-Islamic culture.
Wine has its own semantic associations in the West, for example it can symbolize the Blood of Christ, and as such can be deemed Holy. The sacrilegious request of the pir-i mughan is entirely lost in a literal translation, and would need to rely on footnotes–or other forms of discussion– to convey the poem’s central metaphor. It must also be added that Hafez’s ghazal speaks from the margins of mainstream Sufism at the time; it is a Zoroastrian wine-seller that shows the seeker “the Path and the Way,” not a spiritual Sufi leader to whose (institutional) order he adheres. From every angle, this ghazal is culturally and socially subversive.
Our discussion of the linguistic and cultural aspects of translating Hafez merely scratches the surface of the dialogue between literary traditions; in this case, subversive, Sufi, Islamic, Persian, Christian, and English traditions are implicated. It is a dialogue shaped by the poetry of Hafez and mediated by his translator(s) as well as publisher(s). It goes without saying that it is absolutely paramount that such mediators be fully bilingual and competently bicultural.
More often than not Hafez’s verse has been rendered into English by translators whose lack of proficiency in Persian does not allow them to engage the source text directly; they are more accurately characterized as “versioners, adapters and impersonators” by Franklin Lewis [Rumi: Swallowing the Sun, Oneworld, 2008, xiv]. The list of individuals who have attempted to use Hafez as a ladder to climb into literary fame is rather lengthy; it will suffice to mention Daniel Ladinsky, contemporary American poet, who has sold several collections of poetry as “translations” of Hafez. Ladinsky does not know Persian while his poems bear little or no resemblance to what Hafez has composed. Ladinsky’s work–and others who adhere to the literary model of cultural appropriation–exists as a symptom of highly unequal power relations between the source and target language and culture, in this case Persian and English are implicated. If examined as a symptom of an asymmetrical cultural context–and not an indictment of translation–such reductive models can critically inform the development of more egalitarian metaphors for translation, namely translation as inter-cultural negotiation.
Many contemporary thinkers have offered us thought provoking metaphors for the difficulty of translation. Walter Benjamin’s influential essay, “The Task of the Translator” (1923), itself a retelling of multiple traditions—Jewish, Kabbalistic, Christian—underlines the importance of allowing the source text to shine through the target language. He treats translation as a “mode” that allows the kinship of languages to be “conveyed,” a relationship best highlighted by “conveying the form and meaning of the original as accurately as possible” (Illuminations, Harcourt, Brace & World, pp. 69-82). The translator’s efforts to meditate on all the possibilities offered by English and bend the target language beyond its comfort zone to accommodate the Persian mode of signification will highlight the kinship between English and Persian and demonstrate what this negotiation can achieve.
Rather than placing translation within the reductive and primitive metaphors of betrayal and faithfulness, one might suggest a creative and scholarly process of inter-cultural and linguistic negotiation—a more productive and egalitarian model—that enriches the target language by expanding its linguistic components, offering new cultural metaphors, and cultivating its theoretical landscape, and equally adds deeper nuances to the reception of the literary work in the source language. Vigorous engagement with the work—consequent to the process of negotiation—often results in rewarding moments, however brief the moment of contact between the source and target language may be, as powerfully underlined by Benjamin. Such translations bring the reader closer to the source text, much like a detour that never loses sight of the main road and attempts to guide the reader via its own road signs, albeit placed in different locations. Only a brief examination of two translations of Hafez’s beyt demonstrates the different creative and scholarly measures taken by its translators:
به می سجاده رنگین کن گرت پیر مغان گوید
که سالک بی خبر نبود ز راه و رسم منزل ها
And if the wine-seller says wine
Should dye your prayer-mat…dye it!
Pilgrims should show each stage’s rule
And seek to satisfy it.
–Dick Davis
(Faces of Love, pg 10)
Stain your prayer-mat with wine if the Master tells you:
That seasoned voyager knows the ways of the road.
–Peter Avery & John Heath-Stubbs
(Hafiz of Shiraz, Handsel Books, 2003, pg. 25)
Dick Davis has opted for a simple and contemporary diction and has remained close to the source text. Although different from the mono-rhyme of the Persian ghazal, his translation has developed its own rhyme at the end of each even misra‘, and every beyt has been expanded to four lines:
Davis’ translation is aligned with Khurramshahi’s interpretation, namely that pir-i mughan in this ghazal alludes to a Zoroastrian wine-seller, and its mystical reading (pir as a Sufi sage) is only suggested in the background. To keep mystical interpretations of Hafezian figures seamlessly in the background, he has rendered salik as pilgrim, as opposed to seeker. The concept of rah-u-rasm that guides the salik to manzilha is expressed through “seek[ing] to satisfy each stage’s rule,” which reflects the notion of traversing levels of mystical Islam, critiqued by Hafez in this ghazal.
Peter Avery and John Heath-Stubbs’ translation does not rhyme; all the same, it has struck its own meter in English. They have opted for stain, rather than rangin kun [make colorful]; their rendition may not reflect the source lexicon, but does convey the Hafezian metaphor, a reasonable compensation. However, they have decided to over-impose a mystical interpretation on the ghazal by translating pir-mughan as “the Master,” which necessarily defangs Hafez’s critique of institutional Sufism and its practices. “Ways of the road” reflects the linguistic metaphor of the source text that “Goals are Destinations,” embedded in rah and manzil. Both versions are close to the source text and, through recalling different possibilities of English, they have put Persian and English poetics in conversation with one another.
Hafez was first introduced to English-speaking readers through Sir Williams Jones’s version of a ghazal, titled “A Persian Song of Hafiz” in 1771. Translators of Hafez have come a long way ever since, taking his poetry in numerous directions based on their local variant, target audience, reading models, and approaches to translation. Julie Meisami’s essay in the Encyclopedia of Literary Translation Into English examines the historical and stylistic trajectory of these works [Routledge, 2000, pp. 600-602]. Hafez’s English translations, all surrounding his work akin to exegetical exercises of the Persian, have histories and traditions of their own. For instance, there are as many equivalents suggested for pir-i mughan and salik as there are English translators of Hafez.
Translation | pir-i mughan | salik | Metaphor |
Arberry, A. J., Hafiz: Fifty poems, 1947. | taverner | [one] whose wont is on this road to go | Let wine upon the prayer-mat flow |
Clarke, H. W. The Divan of Hafiz, 1974. | Pir of the magians (the perfect murshid) | holy traveller (the perfect murshid) | With wine, becolour the prayer-mat |
Kashani, A. A., Odes of Hafiz: Poetical Horoscope, 1984. | The experienced sage | us/my friend | “Wash your prayer-rug in red wine, my friend” |
Bell, G. The Teachings of Hafiz, 1985. | Tavern-keeper | traveller | “With wine, with red wine your prayer carpet dye!” |
Smith, P. Divan of Hafiz, 1986. | Master | This experienced traveller | With wine dye your prayer-mat |
Gray, E. The Green Sea of Heaven, 1995. | Magus | traveller | Stain your prayer mat with wine |
Alston, A. J., In Search of Hafiz, 1996. | Magian Teacher | the true traveller | Dye your prayer-carpet in wine |
Newell, J. R. The Songs of Hafiz, 2001. | master | that holy one | Dye your prayer mat red with wine |
If examined carefully in the light of translation as a process—not merely an end result—of cultural and linguistic negotiation, these translations can point us to new and unprecedented horizons and challenge the idea of languages as fixed and unevolving literary idioms strictly associated with a particular religion or culture. In the past several decades, English has become a common vehicle for the expression of Islamic and Sufi thought, both through creative and scholarly writing. The central metaphor of Hafez’s ghazal may be unique to Persian, but the source concepts it evokes–that of religiosity and subversion–is common among all languages. English will continue to search for the voice of Hafez—and poets in other languages—and with each vigorous process of negotiation, we move a step closer in the Benjaminian direction of highlighting the “kinship” of all languages, their interconnectedness in what they wish to express.
_ _ _
This essay was written in close dialogue with Chana Kronfeld and my colleagues in her graduate seminar, “The Politics and Poetics of Translation” at the University of California, Berkeley. I would also like to thank my colleagues Leyla Rouhi, Tracy Cummings, and Franklin Lewis for their critical comments.