KHAYYAM, OMAR i. LIFE

Uncertainty surrounds the details of the biography of Omar Khayyam (ʿOmar Ḵayyām), including significant dates in his life. Apart from ʿOmar, the other constituents of his name and sobriquets vary from source to source. In some, including his own mathematical works such as his Maqāla fi’l-jabr wa’l-moqābela, his patronymic (konya) appears as Abu’l-Fatḥ (Woepcke, Arabic text, p. 1), while in other sources it is replaced by Abu’l-Ḥafṣ, a more frequently used konya in conjunction with ʿOmar, and the one used by his near contemporary, ʿAbd-al-Raḥmān Ḵāzeni in his Mizān al-ḥekma, written in 515/1121 (p. 270).

Most references in Arabic sources, though by no means all, refer to the poet and philosopher as “al-Ḵayyāmi” while Persian sources opt for “Ḵayyām.” This discrepancy had, in turn, led in the past to further speculations, now discarded, proposing the existence of two entities: a poet Khayyam and a mathematician Khayyami (see KHAYYAM ii).

The dates of Khayyam’s birth and death are also a matter of dispute. In the case of his death date, confusion arises from the different wording in manuscripts of the Čahārmaqāla (q.v.) by Neẓāmi ʿArużi (fl. 504-51/1110-56), where he states (p. 100) that he had visited Khayyam’s grave in Nishapur in 530/1136. Some manuscripts specify the date as four years (čahārsāl) after his death, while others have some years (čandsāl), thereby leaving it unspecified. Regarding his birth, Ẓahir-al-Din Bayhaqi (q.v.; d. 565/1169-70) in his Tatemmat Ṣewān al-ḥekma (ed. Šafiʿ, p. 112) recounts visiting Nishapur in his youth and meeting Khayyam there. He also notes Khayyam’s horoscope (ed. Šafʿi, p. 112). Based on this, Govinda Tīrtha has calculated that his birthday fell on 26 Ḏu’l-qaʿda 439/18 May 1048 (Tīrtha, pp. xxxii-xxxiv).

Bayhaqi’s account also suggests that Khayyam belonged to an old and well-established family in the city. That Khayyam exerted considerable local influence and authority is also implied in a letter addressed to him by the poet Sanāʾi (d. ca. 525/1131), in which he sought Khayyam’s assistance to redress an unfounded accusation against his servant and subsequently against himself during a journey in Khorasan (Sanāʾi, Makātib, pp. 70-77; Minovi, 1950, pp. 209-15). There are also allusions to Khayyam’s prestige and eminence in the divan and letters of Ḵāqāni Šervāni (q.v.). In an elegy on the death of his uncle, Kāfi-al-Din ʿOmar b. ʿOṯmān, Ḵāqāni praises his uncle’s piety and sagacity by putting him on a par with the second caliph, ʿOmar b. Ḵaṭṭāb, and Omar Khayyam (Ḵāqāni, Divān, p. 58). He also recounts an anecdote in a letter to Jalāl-al-Din Šervānšāh (Ḵāqāni, Monšaʾāt, pp. 333-34) referring to the Saljuq sultan Malekšāh’s approval of Khayyam’s scathing comments regarding the superiority of men of science to officials of high bureaucratic status (Dashti, tr., pp. 51-52).

Ebn al-Aṯir (q.v.) reports that in 467/1074-75 the Saljuq sultan Malekšāh (r. 465-85/1072-92) summoned a number of astronomers, including an “ʿOmar b. Ebrāhim al-Ḵayyāmi,” to construct an observatory (X, pp. 67-68; tr. Richards, p. 189). It was probably located in Isfahan (Bayhaqi, p. 163; Sayili, p. 161), and there are indications in some of Khayyam’s own works that he was traveling or residing in that area during that period. His commentary on the problems of certain postulates of Euclid, Resāla fišarḥmā aškala men moṣādarāt ketāb Oqlides, was completed at a city, the name of which is missing in the manuscript, in late Jomādā I 470/end of December 1077 (Leiden University Library, MS UBL Or.199[8]; Homāʾi, p. 222). Khayyam states that he composed his Persian translation of a short sermon (ḵoṭba, q.v.) by Avicenna (q.v.) in 472/1079 at the behest of some “brethren in Isfahan” (Ḥabibi, p. 94; Safina-ye Tabriz, p. 323). In his treatise Jawābanle-ṯalāṯ masāʾel (facsimile in Reżāzāda Malek, p. 419), Khayyam refers to a treatise he had composed in 473/1080 for Abu Ṭāher, the chief judge (qażi al-qożāt) of Fārs. It seems likely, therefore, that Khayyam spent the years 467-72/1074-79 traveling in the western parts of the Saljuq empire while in the service of Malekšāh.

Other references to Khayyam’s travels cannot be dated with any degree of certainty. Ẓahir-al-Din Bayhaqi mentions Khayyam’s status as a companion and attendant (nadim) to Malekšāh, as well as his enjoying an even higher status at the court of the Ḵāqān Šams-al-Moluk (ed. Šafiʿ, p. 115) at Bukhara. The latter is presumably the Ilak-Khanid (q.v.) amir Šams-al-Moluk Naṣr II b. Ebrāhim Tamḡāč-ḵān (460-72/1067-79; on him, see Barthold, pp. 314-16; Zambaur p. 206), and it seems likely that Khayyam’s stay at Bukhara pre-dated his stay in Isfahan in the years 1074-79.

According to Ẓahir al-Din Bayhaqi (ed. Šafiʿ, pp. 114-15), the warm reception that Khayyam had experienced at Malekšāh’s court did not continue into the reign of Sanjar (q.v.). Bayhaqi traces their apparently frosty relationship to the time when the young Sanjar was afflicted by smallpox and Khayyam had been asked by his vizier, Mojir-al-Dawla, to attend to him, and Khayyam had a negative prognosis for the boy’s condition (al-ṣabi maḵuf). Sanjar was displeased when this was reported back to him and was subsequently hostile toward Khayyam. If this report is credible, it is likely that it belongs to the period when Sanjar was the subordinate sultan in Khorasan (490-511/1097-1118) and during Mojir-al-Dawla’s vizierate (490-97/1097-1103; Zambaur, p. 224). Some details are confirmed by other sources: Rāvandi describes Sanjar’s face as marked by smallpox (ābela-nešān, p. 167) and records his age at accession as eleven (p. 185).

According to contemporary or near contemporary accounts, Khayyam had spent most of his time during the reign of Sultan Sanjar in Khorasan and Transoxiana. Neẓāmi ʿArużi, in his celebrated account of Khayyam’s foretelling his place of burial (Neẓāmi ʿArużi, p. 100, tr., p. 71), mentions meeting him and Abu Ḥātem Moẓaffar Asfezāri (q.v.) in Balḵ (q.v.) in 506/1112-13, and in the next anecdote (p. 101, tr., p. 72) locates him in the winter of 508/1114-15 at Marv, lodged in the house of the vizier Ṣadr-al-Din Moḥammad b. Moẓaffar (Ṣadr-al-Din Abu Jaʿfar Moḥammad b. Faḵr-al-Molk, d. 511/1117). Most of the other accounts relating to this period, some possibly apocryphal, appear in Ẓahir-al-Din Bayhaqi’s Tatemmat Ṣewān al-ḥekma and describe Khayyam’s encounters with philosophers, mystics, and astronomers.

Bayhaqi (ed. Šafiʿ, p. 114) mentions an encounter between Khayyam and Abu Ḥāmed Moḥammad Ḡazāli (q.v.; 450-505/1058-1111). Although the passage in Bayhaqi sounds anecdotal and fictitious, there may be some truth in it. Bayhaqi also claims that ʿAyn-al-Qożāt Hamadāni (q.v.; 492-526/1098-1131) was a student of Ḡazāli’s younger brother, Aḥmad Ḡazāli (q.v.; ca. 453-517 or 520/1061-1123 or 1126), which is well attested by other sources, and also of ʿOmar Khayyam. This is possible, given the fact that Bayhaqi goes on to say that, in his Zobdat al-ḥaqāʾeq, ʿAyn-al-Qożāt had combined the discourse of the Sufis with those of philosophers (ed. Šafiʿ, pp. 117-18), and he may well have drawn on Khayyam’s teaching. Other meetings and debates with contemporary philosophers and astronomers are also briefly cited by Bayhaqi, including those with Abu Ḥātem Moẓẓaffar Asfezāri, Šaraf-al-Zamān Ilāqi, and Moḥammad b. Aḥmad Maʿmuri Bayhaqi. The last mentioned is included among those who had taken part in setting up the observatory in Isfahan (ed. Šafiʿ, p. 163).

As well as Bayhaqi’s account of Khayyam’s encounters with friends and other philosophers or students, there are also references to him in the works of some other contemporary philosophers, astronomers, and mathematicians, and these furnish further details of his life and travels. Ebn al-Qefṭi (568-646/1172-1248) refers to his pilgrimage to Mecca as well as his visit to Baghdad (p. 244). Badiʿ-al-Zamān Asṭorlābi (d. 534/1139-40), a famous astronomer and maker of astrolabes (Rosenthal, p. 555; Maʿṣumi Hamadāni, pp. 114-15), describes his encounters, at least once in Baghdad, with Omar Khayyam (“Ḥojjat-al-Ḥaqq ʿOmar al-Ḵayyāmi”) in the preface (f. 168) to a treatise that he had composed on astrolabes (Ketāb al-ʿamalbe’l-korah) and relates his scientific discussion with Khayyam and the latter’s complimentary remarks about him (Bodleian Library, Oxford, MS Marsh 663, no. 8., ff. 164-89; see Rosenthal, pp. 556-57 for the attribution of this text to Asṭorlābi).

In a short treatise entitled al-Resāla al-zājara, Abu’l-Qāsem Maḥmud Zamaḵšari (467-538/1075-1144), the celebrated Muʿtazilite exegete, also describes his meeting and debate with Khayyam in Ḵvārazm (Zamaḵšari, p. 626; Foruzānfar, pp. 1-29). Although Zamaḵšari narrates with relish, in a self-congratulatory manner, how he won the debate with Khayyam, the laudatory epithets that he bestows on Khayyam show how highly Khayyam was regarded by a distinguished contemporary figure.

Taken together, sources dating from the 6th/12th century, contemporary with Khayyam’s lifetime or shortly thereafter, share certain common characteristics. First, though there are some verses in Arabic by him dating from this era, no Persian verses are cited until almost a century after his death. Second, some of Khayyam’s significant and authentic works do not appear in the list of his works from this same period. Third, it was assumed until recently that none of these early sources had expressed reservations or hinted at any criticism regarding Khayyam’s outlook on life and religion and that such a critical stance against the views expressed in his poetry had originated later, from the beginning of the 7th/13th century onward. This assumption had also been used in support of the now discredited theory of the two Khayyams mentioned earlier. As described below (KHAYYAM ii), there were poems by Khayyam in circulation from as early as the second half of the 12th century that support the contention that Khayyam the poet, the mathematician, and philosopher were all one and the same person. Furthermore, as implied in the account of the mixed reception of his quatrains, the very fact that he was regarded as a follower of Avicenna affected the way his verses were interpreted and at times denounced. In other words, much of the criticism leveled against him in the earliest sources was directed at Khayyam as a philosopher following Avicenna (Maʿṣumi Hamadāni, pp. 132-33).

With reference to the accounts of Khayyam’s meetings with Badiʿ-al-Zamān Aṣtorlābi and Abu’l-Qāsem Zamaḵšari, and as already noted above, he was viewed as an esteemed authority by some contemporary theologians and astronomers. He also appears as a figure capable of wielding some local influence, as suggested above in the letter by the poet Sanāʾi (Minovi, 1950, p. 210). But here, as in the case of some other assessments of his legacy, there is an element of ambivalence, suggesting that the acknowledgment of his eminence did not necessarily embrace an approval of his general outlook on life and hereafter as expounded in his verses. In a couple of quatrains, Sanāʾi appears to challenge Khayyam’s attitude toward death by implicitly alluding to his quatrains (Mirafżali, 2015, p. 5).

One of the first direct citations of Persian verses by Khayyam appears in an exegetical treatise on four suras of the Qur’an by Faḵr-al-Din Rāzi (d. 606/1210), entitled Resālat al-tanbihʿalabaʿż al-asrār al-mudaʿa fi baʿż sowar al-Qorʾān al-ʿaẓim (see KHAYYAM ii for further details). Faḵr-al-Din Rāzi quotes Khayyam’s robāʿi:

Dārandačotarkib-e ṭabāyʿārāst/ bāz azčesabab fekandaš andar kam o kāst?

Gar ḵub nayāmad in banāʿayb kerāst? / v’r ḵubāmad, ḵarābi az bahr-ečerāst?

Why did the Maker adorn the forms of creation

And then cast them down to decay and decrease?

Should the forms be ugly, whose fault is it?

And if pleasing they be, why cause their ruin?

(Najm-al-Din Rāzi, tr., p. 54; another tr. in Dashti, tr., p. 36).

The quatrain is quoted in the context of his commentary on sura 95 vv. 4-5: “We have indeed created man in the best of moulds / Then do We abase him (To be) the lowest of the low” (tr. A. Yusuf Ali).

Rāzi cites Khayyam’s lines to point to the problems (eškāl) that the Qur’anic lines pose for those endowed with and relying exclusively on intellect (al-ʿoqalāʾ). Although Rāzi proceeds to rebut these challenges, there is no overtly personal criticism leveled at Khayyam, whose lines are cited in passing as a sample of a philosopher’s questioning stance (Minovi, 1957, pp. 71-72; Dashti, tr., pp. 36-37; Mirafżali, 2003, pp. 23-24; Maʿṣumi Hamadāni, p. 124).

The same verses are cited again a few decades later in an often-quoted manual of Sufism, Merṣād al-ʿebād men al-mabdaʾ elā’l-maʿād of Najm-al-Din Rāzi (q.v.; d. 654/1256). Here, having referred to these verses along with another quatrain by Khayyam, Najm-al-Din launches into a sustained diatribe against Khayyam as “that sightless wanderer” (sargašta-ye nābinā; Najm-al-Din Rāzi, p. 31). Khayyam’s skeptical and confused state of mind is contrasted with the assured certainties of the devout and sincere disciple (morid) who “will perceive who he is, whence, how, and for what purpose he has come; whither and how he shall go; and what his goal and destination are” (Najm-al-Din Rāzi, p. 31, tr., p. 53) once he has faithfully followed the path set out for him in Merṣād al-ʿebād. These disapproving remarks about Khayyam, and the outright denial that he had any Sufi sympathies, are also shared by the translator of Rāzi in a lengthy footnote where he underlines the significance of the passage “as a decisive refutation of claims, ancient and modern, that Ḵayyām was in reality a Sufi” (Najm-al-Din Rāzi, tr., p. 54, n. 10).

As Hamid Algar, the translator, mentions in the same note, Khayyam also appears in an unfavorable light in Farid-al-Din ʿAṭṭār’s (q.v.; ca. 1145-1221) Elāhi-nāma (ʿAṭṭār, p. 326, ll. 4746-53, tr., p. 252). Here he is recalled and described by a seer able to convey Khayyam’s emotional state from beyond the grave as “a man in a state of imperfection” (mardist andar nātamāmi). His soul has remained in perpetual turmoil after his departure from this world, and his lifelong certainties have proved false in retrospect, when viewed from beyond the grave:

Now that his ignorance has been revealed to him / he perspires because of the confusion of his soul.

He is left between shame and confusion / his very studies have made him deficient (ʿAṭṭār, tr., p. 252).

But as the verses that follow this anecdote and appear as direct comments by the poet suggest, “I have traversed the whole world a hundred times / I have found no cure and I am utterly helpless” (ʿAṭṭār, p. 326, line 4758 tr., p. 253), the emphasis is on the absurdity of overconfident metaphysical speculations by philosophers relying solely on their intellect and their refusal to accept with due humility the limitations of human understanding confronted by an unfathomable divine order, a frequently expressed sentiment in both lyrical and narrative poetry in Persian.

There also seems to be a more sympathetic allusion to the quatrain that Najm-al-Din Rāzi had quoted (p. 31) in the two anecdotes a few lines later in the same maṯnawi by ʿAṭṭār. Here, he introduces one of his favorite stock characters, “the delirious fool or madman,” with the fool questioning the Almighty with a direct reference to another similar verse from the Qur’an (13.39; ʿAṭṭār, p. 327, line 4772; ed.’s comm. p. 667; tr., pp. 253-54; Ritter, 1952, pp. 1-15; 2003, pp. 165-87). It is Khayyam’s refusal to acknowledge the limits of human understanding, rather than his questioning tone, which bears the blame.

The description of Khayyam in the discourses (maqālāt) of Šams-e Tabrizi, the spiritual guide of Jalāl-al-Din Rumi (604-72/1207-73), is closer to Najm-al-Din Rāzi’s portrayal of Khayyam as a muddled thinker, clinging to his own set of certitudes and railing against others to compensate for his own deficiencies. He quotes Khayyam as saying that “No one has arrived at the mystery of love, and he who has arrived, remains perplexed.” Šams then proceeds to analyze the reasons behind this quoted paradox. Khayyam’s statement, according to him, reflects his state of mind. Feeling confused and perplexed, he looks outside and beyond himself for targets to blame. In turn, he berates the heavens, or fortune, or the Divine Presence Himself, wavering between negation and proof or prevaricating by inserting a hesitant “if,” taking refuge in dark, disordered, and confused words. Šams, like Najm-al-Din, contrasts this state of wandering and chaos to the clear and coherent world of the believer, rightly secure in his certainties (Šams, p. 301, tr., p. 374).

Other sources present a more ambivalent assessment of Khayyam. One of the most frequently cited sources on the biography of Khayyam is Ebn al-Qefṭi’s Taʾriḵ al-ḥokamāʾ, written most probably after 1231 (and preserved in an epitome composed in 647/1249). On the one hand, the section on Khayyam begins by endowing him with fulsome titles, acknowledging his high status in Khorasan and his matchless erudition. Ebn al-Qefṭi then proceeds to describe him as a proponent of purifying body and soul in seeking the one universal God and as a follower of Greek political philosophy (Ebn al-Qefṭi, pp. 243-44; tr. in Ross, pp. 354-55; tr. in Woepcke, pp. v-vi, Arabic text, p. 52). But the tone of the passage becomes increasingly critical of Khayyam. His avowed sympathy for Greek philosophy is viewed critically as being in direct conflict with the teachings of religion and implicitly as a rival to it. Ebn al-Qefṭi proceeds to express his alarm at the way Khayyam’s verses had ensnared later Sufis who had failed to grasp their true significance and how as a freethinker he had indulged in a life of ingenious dissimulation, referring to his pilgrimage to Mecca as an example of duplicitous subterfuge. He then quotes the same qeṭʿa of four verses that appear in the anthology composed by ʿEmād-al-Din Kāteb Eṣfahāni (q.v.; 519-97/1125-1201; see also below, KHAYYAM ii). As the present contributor has suggested elsewhere (Maʿṣumi Hamadāni, pp. 120-22), for a variety of reasons it seems likely that Ebn al-Qefṭi had based his section on Khayyam on material from ʿEmād-al-Din, or else they had both relied on an earlier source. The indebtedness of Ebn al-Qefṭi to ʿEmād-al-Din had also been briefly referred to in a letter written by Vladimir Minorsky to Moḥammad Qazvini (qq.v.) while Minorsky was engaged in writing his entry on Khayyam for the first edition of the Encyclopaedia of Islam (letter dated 1 August 1934; in Moʿin, pp. 309-10).

In some ways, the various debates about the interpretation of Khayyam’s quatrains resemble those appearing later in the context of Hafez (q.v.; 715-92/1315-90) and his ghazals. ʿAṭṭār’s relatively compassionate but nevertheless critical remarks on Khayyam are to some extent echoed in Bādiʿ-al-Zamān Foruzānfar’s (q.v.; 1903-70) posthumously edited lecture notes. He contrasts Khayyam’s pessimistic negation of afterlife, which endows his carpe diem advice with an underlying note of foreboding and despair, with Hafez’s similar celebratory themes but with an unabashed joyful underpinning, borne out of a mystic’s willing submission to Divine Will (Foruzānfar, 2001, pp. 37-38).

Also, as in the case of Hafez, Khayyam has been subjected to systematic commentaries in order to explicate and recover supposedly esoteric meanings. Khayyam’s sympathetic attitude toward Sufism in his short treatise, Dar ʿelm-e kolliyāt-e wojud (On the existence of universals; see Reżāzāda Malek, p. 389), is used to imply a close connection with his contemporary Sufis (Aminrazavi, p. 136).

In more recent times, ever since the remarkable popularity of Khayyam through translations, particularly after those by Edward FitzGerald (q.v.; 1809-83) in 1859 and J. B. Nicolas (1814-75) in 1867, and as reflected below in several entries on translations of Khayyam, has itself led to a profusion of different readings and interpretations of the poet, often by critics with varying degrees of familiarity with the original Persian, favoring either an esoteric poet in need of decipherment (Aminrazavi, p. 135) or an exceptionally outspoken freethinker writing for an intimate circle of friends (Lazard, 1995, pp. 177-82; 1997, p. 8, p. 11 [quoting Théophile Gautier]). As noted above and in other entries (see KHAYYAM ii), apart from a relatively small number of verses attributed to Khayyam in the oldest sources, the authenticity of many of the quatrains ascribed to him remains in doubt, making it hard if not impossible to offer a comprehensive overview of the corpus, unless of course only those verses are selected that support a pre-conceived notion of the poet, regardless of their origin and authenticity.

• Ḥosayn Maʿṣumi Hamadāni
• EIr.

KHAYYAM, OMAR ii. POETRY

The complete corpus of poetry attributed to Omar Khayyam consists of numerous robāʿis (quatrains, ruba’is) and three qeṭʿas (short topical poems) composed in Persian, as well as approximately forty verses in Arabic.

Khayyam’s poems in Arabic. In contrast to his Persian quatrains, which first appear in sources nearly a century after his death, a scattering of verses (bayts) in Arabic date from the poet’s own lifetime, or shortly afterwards, and appear in anthologies of Arabic verse in the early 6th/12th century.

Ḥosayn b. Moḥammad b. ʿAbd-al-Wahhāb Ḥāreṯi, usually referred to as Bāreʿ Baḡdādi (443-524/1051-1129), was the first anthologist to include some of Khayyam’s verses in his Ṭarāʾef al-ṭoraf. His collection is divided into twelve chapters, and three qeṭʿas of Khayyam’s Arabic poetry (seven verses in all) are recorded in the opening chapter, which is devoted to sayings and adages (nos. 42-44, pp. 40-41). Ṭarāʾef al-ṭoraf was most likely compiled in Khayyam’s lifetime. Perhaps the two had met during Khayyam’s visit to Baghdad (Ebn al-Qefṭi, p. 244), and the anthologist had noted down the verses directly from him.

Another early author who included Khayyam’s verse in an anthology is ʿEmād-al-Din Kāteb Eṣfahāni (q.v.; 519-97/1125-1201) with a qeṭʿa of four verses in his Ḵaridat al-qaṣr wa jaridat al-ʿaṣr (II, p. 85). These verses may date to the years that Khayyam had spent in Isfahan (467-472/1074-1079; Barg-nisi, p. 161). The same qeṭʿa appears in the Taʾriḵ al-ḥokamāʾ (p. 244) of Jamāl-al-Din Abu’l-Ḥasan Ebn al-Qefṭi (568-646/1172-1248). Šams-al-Din Moḥammad b. Maḥmud Šahrazuri (d. 676/1277), in his Nozhat al-arwāḥ wa rawżat al-afrāḥ (Taʾriḵ al-ḥokamāʾ), includes additional Arabic verses under his entry for Khayyam (pp. 325-26; Qazvini, pp. 211-14).

There are differing evaluations of the relative significance of these Arabic verses. Some scholars detect a more intimate and palpable affinity with his stature as a man of science in these extant Arabic verses, rather than in his Persian quatrains. Some go further and maintain that his Arabic verses present a touchstone for ascertaining the authenticity of the Persian verses attributed to him. Others, however, regard them as merely run-of-the-mill poetical exercises, of some interest solely because they happen to have been composed by a celebrated thinker and scientist (Ebn Rasul, p. 187). Moreover, some of the Arabic verses attributed to Khayyam share the fate of many of his Persian ones: their authenticity has been questioned given that they also appear in perhaps more creditable sources under other names, including those of Abu’l-ʿAlāʾ Maʿārri (d. 449/1058), Moʿayyed-al-Din Ḥosayn b. ʿAli Ṭoḡrāʾi (453-514/1061-1121), and Abu Sahl Saʿid b. ʿAbd-al-ʿAziz Nili (353-420/964-1029; Ebn Rasul, pp. 238-40).

Khayyam’s Persian poetry. Almost all of the Persian poems attributed to Khayyam are robāʿis. There are, however, also three qeṭʿas reputedly by him recorded in different sources. A manuscript in Konya of the Rabāb-nāma of Solṭān Walad (d. 712/1312), dated to before 704/1304, includes a poem of eleven verses under his name, couched in the manner of a homiletic qaṣida (Minovi, 1957, pp. 74-75; Homāʾi, p. 167). The second poem, a dialogue in the form of question and answer interrogating the figurative persona of the Intellect, appears in the marginalia of the edition and Turkish translation of Khayyam’s Robāʿiyāt by Ḥosayn Dāneš (q.v.; p. 312; Homāʾi, p. 168). The third poem is a qeṭʿa of three lines that Maḥmud Golestāna (d. 801/1398) attributed to Khayyam in his Anis al-waḥda wa janis al-ḵalwa (p. 113), a compendium of quoted counsel and exempla on sundry topics. The identical verses appear in the divān of Ebn Yamin (q.v.; p. 354). Immediately after the above-mentioned qeṭʿa, there are two others in Anis al-waḥda, which have been erroneously attributed to Khayyam by the editor of Anis al-waḥda and some other authors (Yakāni, p. 404).

The quatrains attributed to Khayyam are among the masterpieces of Persian poetry, and the poet himself is universally regarded as one of the most renowned of all Persian poets. His worldwide fame is based not only on his scientific stature and his learned treatises but also on the memorable verses attributed to him.

These attributed verses have multiplied with the passage of time. Until the early years of the 9th/15th century, they amounted to nearly 140 quatrains (Mirafżali, 2003). Fifty years later, their number had reached 550, as recorded in the Ṭarab-ḵāna of Yār Aḥmad Rašidi Tabrizi (d. 867/1462). The Indian scholar Swāmī Govinda Tīrtha’s comprehensive collection of Khayyam’s quatrains, published in 1941, which drew upon all the manuscripts and printed editions available to him, increased the number to 1,096.

The robāʿi is one of the earliest and most authentic forms of Persian poetry (Elwell-Sutton, pp. 633-39), with its origins traditionally traced back to the poetry of Rudaki (d. 329/941). The earliest extant poems as recorded in different sources such as anthologies, histories, and literary and mystical texts, date from the 4th/10th century. The few predating this period lack sufficient documentary evidence.

Until the middle of the 5th/11th century, the contents of a robāʿi were mainly devoted to a celebration of amatory or panegyric themes. There was as yet no room for philosophic or mystical ruminations. The verses of Sanāʾi (q.v.) and Khayyam ushered in new themes, and endowed the form with a new mystical and philosophically speculative identity. Khayyam’s distinctive ontological ruminations, as expressed in his authentic verses, pioneered a novel and distinct sub-branch in the history of the Persian robāʿi: the robāʿiyāt-e Ḵayyāmāna (Khayyamesque or Khayyamian robāʿis).

The emergence of Khayyam’s poetry and its collection. Khayyam’s quatrains first appeared as dispersed quotations cited in Persian prose works a few decades after his death, usually in support of an observation in the text. Toward the end of the 7th/13th century, a selection of quatrains attributed to him began to appear in the manuscripts of anthologies. The evidence from authentic manuscripts and reliable sources suggests that the actual collection and arrangement of Khayyam’s robāʿiyāt began in earnest in the middle of the 9th/15th century and were gradually followed by the production of numerous manuscripts of his poems. One of the earliest and most famous collections is in the Bodleian Library, Oxford (MS Ouseley 140). It was completed in Ṣafar 865/December 1460, transcribed by Šayḵ Maḥmud Pirbudāqi in Shiraz, and it provided the basis for Edward FitzGerald’s (q.v.) famous rendering of Khayyam. Edward Heron-Allen (q.v.) published a facsimile of the manuscript with a translation and commentary that also included observations on FitzGerald’s idiosyncratic use of the manuscript (London, 1898).

Another significant collection of the robāʿiyāt is Yār Aḥmad Rašidi Tabrizi’s Ṭarab-ḵāna of 867/1462, which is divided into ten chapters based on a thematic classification of the contents. The Istanbul edition of the text, published by Abdülbaki Gölpinarli (q.v.) in 1332/1953, contains 422 quatrains, while the later edition of 1342/1963 by Jalāl-al-Din Homāʾi includes 559 quatrains.

More recent scholars from Iran and elsewhere have offered very different selections of Khayyam, both in number and substance, depending on their chosen criteria for selection, as the following varying numbers of quatrains demonstrate: Ḥosayn Dāneš (249 authentic, 147 doubtful), Ṣādeq Hedāyat (q.v.; 143), Moḥammad ʿAli Foruḡi (q.v.; 179), ʿAli Dašti (q.v.; 74), Moḥsen Farzāna (70), Moḥammad Rowšan (96), Raḥim Reżāzāda Malek (95), ʿAli-Reżā Ḏakāvati Qaragozlu (142), Kāẓem Barg-nisi (36), Esmāʿil Yakāni (492), Louis Jean Baptiste Nicolas (464), Friedrich Rosen (330), Arthur Christensen (q.v.; 121), Gilbert Lazard (101).

This apparent discordance in the quatrains attributed to Khayyam as seen from different perspectives, including aesthetic, stylistic, as well as ethical and metaphysical, has proved a challenge to many scholars. Various stratagems and solutions have been proposed to narrow the distances and present a more homogeneous corpus of poems, which would, moreover, conform to his reputation as an erudite scholar and renowned scientist.

The theory of two Khayyams. One of the theories propounded to address this apparent discordance, and that has attracted some support, proposes the existence of two Khayyams. It distinguishes between ʿOmar Ḵayyām or Ḵayyāmi of Nishabur and a poet using Ḵayyām as his nom de plume (de Blois, p. 356, n. 3). This theory was first put forward by Moḥammad Moḥiṭ Ṭabāṭabāʾi in a series of articles. He based his argument (pp. 17-27) on a reference by Ebn al-Fowaṭi (q.v.; d. 723/1323) to an otherwise unknown poet: ʿAli Ḵayyām from Khorasan. According to Ebn al-Fowaṭi’s Majmaʿ al-ādāb (II, p. 331), ʿAlāʾ-al-Din ʿAli b. Moḥammad b. Aḥmad b. Ḵalaf Ḵorāsāni, known as Ḵayyām, had a substantial divān of Persian poetry and enjoyed fame in both Azerbaijan and Khorasan. It should be pointed out, however, that this single piece of evidence carries little weight when set against the cumulative evidence of Persian sources, from the earliest times on, which have attributed the poems to ʿOmar Ḵayyām (or Ḵayyāmi) Nišāburi and not to a certain ʿAli Ḵayyām. Furthermore, and in spite of Ebn al-Fowaṭi’s reference to his fame, there is not even a passing reference to this poet in biographical accounts (taḏkeras) of Persian poets or elsewhere in Persian literary sources; and the only two verses quoted by Ebn al-Fowaṭi happen to be in Arabic. In his account, Moḥiṭ Ṭabāṭabāʾi (pp. 57-58) referred to ʿAlāʾ-al-Din ʿAli Ḵayyām as a contemporary of ʿOmar Ḵayyām Nišāburi, but Kāżem Barg-nisi (p. 116) suggests that ʿAli Ḵayyām must have been a contemporary of Ebn al-Fowaṭi himself.

Earliest sources. One of the basic requirements for the establishment of Khayyam’s authentic verses is locating the relevant references and citations closest to his time. The earliest attempt in this line of research dates to 1897 and the publication of a pioneering article by Valentin Alekseevich Zhukovskiĭ (q.v.; see Ross, pp. 356, 362). The Russian scholar referred to two early texts, with quotations from Khayyam: Merṣād al-ʿebād by Najm-al-Din Rāzi, known as Dāya (q.v.; d. 654/1256), and Ḵosrow b. ʿĀbed Abarquhi (Ebn Moʿin; fl. 808/1405-6)’s Ferdaws al-tawāriḵ (Barthold, pp. 54-55). The latest contribution and discovery in this particular line of research was made in 2005, noting a reference in the Jong-e Qoṭb-e Širāzi (Bašari, pp. 532-33).

This approach, focused on the retrieval of early sources, proved initially fruitful by itself. However, the discovery in these early sources of some exceedingly pedestrian verses, out of joint with the general tone and temper of Khayyam’s poetry, suggested that the early sources could not be relied upon exclusively and that other criteria such as affinities in embedded thoughts and thematic content within the corpus of the poems should also have a bearing on their authenticity.

The earliest source citing a robāʿi under Khayyam’s name appears in an exegesis on four suras of the Qur’an, Resālat al-tanbih ʿala baʿż al-asrār al-mudaʿa fi baʿż sowar al-Qorʾān al-ʿaẓim, by Faḵr-al-Din Rāzi (d. 606/1210). In the third chapter, in his comments on the concept of resurrection, Rāzi cites and criticizes a quatrain from Khayyam (Minovi, 1957, pp. 71-72; Mirafżali, 2003, pp. 23-24).

Two other early sources should also be noted. The already cited Merṣād al-ʿebād by Najm-al-Din Rāzi (Dāya) quotes Khayyam (p. 31; tr. Algar, p. 54) and pours scorn on his views (Mirafżali, 2003, pp. 27-29). By contrast, ʿAbd-al-Qāder Ahari (d. 1261), in his al-Aqṭāb al-qoṭbiyya, cites Khayyam’s poems approvingly (Ahari, p. 121, pp. 198-99, 203-4; Dašti, pp. 159-64, tr. Elwell-Sutton, pp. 119-23; Mirafżali, 2003, pp. 31-33).

Historical works are another important source for retrieving Khayyam’s verses. These include ʿAlāʾ-al-Din ʿAṭā-Malek Jovayni’s (q.v.; d. 681/1283) Tāriḵ-e jahāngošāy (I, p. 128, II, p. 218; tr. Boyle, I, 164, II, p. 482); Šehāb-al-Din ʿAbd-Allāh Širāzi Waṣṣāf-al-Ḥażrat’s (fl. 702/1303) Tajziyat al-ʿamṣār wa tazjiyat al-ʿaṣār (or Tāriḵ-e Waṣṣāf; p. 407; Mirafżali, 2003, pp. 63-65); Sayf b. Moḥammad Heravi’s (fl. 721/1321) Tāriḵ-nāma-ye Herāt (p. 165); Ḥamd-Allāh Mostawfi’s (q.v.; d. 744/1344) Tāriḵ-e gozida (p. 728; tr. Browne, p. 748); and the already cited Ferdaws al-tawāriḵ (Mirafżali, 2003, pp. 121-22).

The historical sources tend to cite only a few verses of Khayyam as a way of buttressing or rounding off an authorial observation. It is in anthologies and other literary collections that most of Khayyam’s verses appear in some number and are treated from a broader literary perspective. The most important of these sources are as follows: (1) The Nozhat al-majāles (q.v.) of Jamāl-al-Din Ḵalil Šarvāni (mid. 7th/13th century), a fundamental source for the study of Persian quatrains in general, contains 4,000 poems by 300 poets, including 33 quatrains by Khayyam, to whom and to those replicating his themes a separate chapter is specifically devoted (Chapter 15, “dar maʿāni-ye Ḥakim ʿOmar Ḵayyām”; Ḵalil Šarvāni, pp. 671-76). (2) There are seven quatrains by Khayyam, along with some verses by other poets, in the colophon page of a manuscript of a version of the seven sages/ten viziers framed stories, Lamʿat al-serāj le-ḥażrat al-tāj (Baḵtiār-nāma; q.v.), in the Leiden University Library (Codex Or. 593) dated 6 Ḏu’l-qaʿda 695/5 September 1296 (Nöldeke, p. 101; Mirafżali, 2003, pp. 59-61.) (3) The greatest number of verses (forty-four quatrains) recorded in these various literary sources is in a manuscript in the Ayatollah Marashi Library in Qom (no. 1259; 7th/13th century), published as Safina-ye kohan-e robāʿiyāt.

Other early sources include Moḥammad b. Badr Jājarmi’s (q.v.; fl. 741/1340) Moʾnes al-aḥrār fi daqāʾeq al-ašʿār, which has also a section devoted to Khayyam’s poetry (included in Miraf˙żali, 2003, pp. 86-87). An as yet unpublished compendium of Persian poems and correspondence (majmuʿa-ye ašʿār wa morāsalāt) is kept at the Süleymaniye Library in Istanbul (MS Lala İsmail no. 487), apparently copied in Egypt by several scribes in 741-42/1340-41. It contains thirty-three (one repeated) quatrains by Khayyam (included in Miraf˙żali, 2003, 94-95). The already mentioned Anis al-waḥda wa janis al-ḵalwa (written in 750/1349) cites a quatrain by Khayyam not recorded elsewhere (Golestāna, p. 250). Also dating from 750/1349 is a compendium of prose and poetry (majmuʿa-ye naẓm wa naṯr) compiled by Abu’l-Fażl Moḥammad b. Maḥmud b. ʿAli b. Sadid b. Aḥmad (Majles Library, Tehran, MS 633) with eleven quatrains by Khayyam (Miraf˙żali, 2003, pp. 105-8; Mahfuz-ul-Haq, pp. 89-91). Another relevant anthology of Persian and Arabic poems is Rawżat al-nāẓerwa nozhat al-ḵāṭer, composed by a poet and litterateur, ʿEzz-al-Din ʿAbd-al-Aziz Kāšāni (Kāši), in the first half of the 8th/14th century. Several copies of this anthology, some abridged, exist in different libraries, including an early copy belonging to Istanbul University Library (MS 766) and another, an abridgement by the author himself, in the British Library (Or. 9602; Meredith-Owens, p. 82). There are four quatrains directly attributed to Khayyam in the wine poetry (ḵamriya, q.v.) section of the anthology and three without attribution (Miraf˙żali, 2003, p. 97-99). Another anthology of poems from the 8th/14th century with verses by Khayyam is preserved in the Government Oriental Manuscripts Library in Madras (MS 183), transcribed by Moḥammad b. Yaḡmur in Termeḏ. The manuscript has been studied by Sayyed Amir Ḥasan ʿĀbedi (pp. 229-46) who refers to it as Bayāż-e Termeḏ and quotes four quatrains of Khayyam that appear first in this volume (pp. 232-34). Another anthology from the 8th/14th century, acquired by the Ganj Bakhsh Library in Islamabad in 1993 (MS 14456, usually referred to as Jong-e Ganj-Baḵš), contains four quatrains under Khayyam’s name in its chapter on wine poetry and two quatrains elsewhere in the anthology without his name (Miraf˙żali, 2003, pp. 117-18). The voluminous collection of different topics in verse and prose compiled in 782/1380 on the order of an otherwise unidentified patron, Tāj-al-Din Aḥmad Wazir, contains several quatrains by Khayyam. The manuscript is preserved at the central library of Isfahan University; a facsimile edition was published in 1975 (Tāj-al-Din Aḥmad, pp. 310, 295, 306, 796) and a printed edition in two volumes in 2003. Finally, one should mention a quatrain under the name of Ḥakim ʿOmar Ḵayyām (British Library, London, MS Add. 27261, f. 148v; Rieu, II, p. 871) in the wine poetry section in the exquisitely illustrated and wide-ranging miscellany, Jong-e Eskandar Mirzā, compiled in 813-814/1410-1411 for Timur’s grandson, Jalāl-al-Din Eskandar b. ʿOmar Šayḵ (q.v.; executed in 817/1414).

The key quatrains. One of the techniques employed by scholars in the past two centuries for assessing the authenticity of Khayyam’s verses is that of identifying key quatrains from the earliest sources and using them as a yardstick for evaluating other quatrains attributed to him. In the west, the pioneer in this line of research was the German scholar Friedrich Rosen (1856-1935). In the preface to his edition of Khayyam’s Robāʿiyāt (Berlin, 1925), he selected twenty-three robāʿis as authentic and suggested that they could be instrumental in evaluating his other quatrains. In Iran, Ṣādeq Hedāyat was the first to follow the same route (1934) and was later followed by Moḥammad ʿAli Foruḡi (in 1941), ʿAli Dašti (tr. 109-28), Rašid Yāsami, Moḥsen Farzāna, Ḥasan Dānešfar, and Moḥammad Rowšan (further details in the bibliography).

This use of key quatrains as a decisive arbiter has also had its critics. According to Jalāl-al-Din Homāʾi, “there are assuredly genuine quatrains belonging to Khayyam himself among the 66 quatrains selected by Foruḡi as key verses,” but he adds the cautionary proviso, “it is also possible that they contain amongst them pieces definitely ascribed to others or of doubtful origin” (Homāʾi, p. 44).

Authenticating and editing the “Robāʿiyāt”. The frequently used descriptive phrase, “the wandering robāʿi” was first coined by V. A. Zhukovskiĭ (Ross, pp. 360-61) in his contribution to the festschrift of Baron V. Rosen in 1897, which was later translated into English by Edward Denison Ross (1871-1940). Zhukovskiĭ extracted 82 quatrains from the 464 in Jean Baptiste Nicolas’ edition that had been attributed to some 39 other poets in various literary sources and divāns. Further research by Denison Ross and Arthur Christensen increased the number of the wandering quatrains to 108 (Dašti, p. 255, tr. p. 179). This trend reached its peak in the study by Swāmī Govinda Tīrtha in 1941, which designated as wandering quatrains (Tīrtha, pp. 162-67) some 753 quatrains out of the 2,213 attributed to Khayyam but actually belonging to 143 poets over the time span of six centuries (Christensen, tr. Badraʾi, p. 29).

Looking at the quatrains of Khayyam from the perspective of the wandering robāʿis can only be partially helpful for it merely suggests which robāʿi might not be authentic. According to ʿAli Dašti (p. 256, tr. pp. 179-80), this line of enquiry is not applicable to all cases, and given the doubtful nature of some of the information in the sources, there is scope for error. The problem of authorial attribution regarding the robāʿis, whether the verses belong to Khayyam or to another poet, is a perennial one, from the earliest times to the present.

The quatrains from the perspective of their rhyme scheme. Little research has so far been carried out in stylistics regarding Khayyam’s quatrains, and no thorough scrutiny of the authentic robāʿis in the context of poetry in Khayyam’s time, the Saljuq era, has yet been published.

Having studied 2,399 quatrains extant in the collected poetry of ten poets contemporary to Khayyam, the present contributor has found that 2,088 quatrains (equivalent to 87 percent) have a four-rhyme scheme (aaaa) and the remaining 13 percent follow a three-rhyme scheme (aaba). L. P. Elwell-Sutton (q.v.; 1912-1984) carried out a similar study for the poets of the 5th/11th century and concluded that 70 percent of the quatrains in this period follow a four-rhyme scheme and 30 percent follow the three-rhyme scheme (Elwell-Sutton, p. 640; Šamisā, pp. 21-22). But his statistics, derived from the poetical works of Farroḵi, ʿOnṣori, Abu’l-Faraj Runi, Moʿezzi, Azraqi, Masʿud Saʿd-e Salmān (qq.v.), and Qaṭrān, cast doubt on his overall conclusion, for, according to him, these poets have 905 quatrains in the four-rhyme scheme (90 percent), and 91 quatrains in the three-rhyme scheme. Elwell-Sutton had not gauged these statistics in relation to the verses attributed to Khayyam and seems to have been mainly interested in deciphering which of the rhyme schemes was the oldest, the four-rhymed or the three-rhymed.

Mohammad Iqbal (q.v.; 1877-1938) was the first scholar to fully appreciate the significance of this distinction in the rhyme scheme in the context of research on the authenticity of Khayyam’s poetry. In a paper presented at a conference in 1933, he investigated the statistics of the occurrence of four-rhymed quatrains in poetical anthologies and divāns of poets of the 5th/11th and 6th/12th centuries. Given the perennial existence of doubtful attributions in the anthologies, as well the dearth of trustworthy manuscripts, he had to exercise caution and present his conclusions tentatively. Nevertheless, by referring to three robāʿis of Khayyam in two sources with an early date, namely Merṣād al-ʿebād and Tāriḵ-e jahāngošāy, he emphasized the importance of the four-rhymed quatrains, to which he refers as “du-baitīs” in order to distinguish them from the three-rhymed robāʿis, and concludes that “the du-baitīs in a genuine collection of the quatrains of Khayyam must very much outnumber the rubāʿīs” (Iqbal, p. 914).

In a more recent contribution, published posthumously in 2012, Alexander Morton investigated the Khayyamian quatrains in a relatively obscure manual of disparate advice perhaps composed sometime between 503/1109 and 509/1115 during the reign of the Ghaznavid (q.v.) ruler Masʿud (III) b. Ebrāhim (q.v.; r. 492-508/1099-1115). The author is named as Abu’l-Qāsem Naṣr b. Aḥmad b. ʿAmr Šādāni Nišāburi. He was a contemporary of Khayyam as well as originating from the same city. The work is referred to as Ganj al-ganj in several of the manuscripts, most probably a later scribal addition. It is divided into twelve chapters on various topics replete with anecdotes and historical exempla and numerous citations of poetry (Imāni, p. 15), including many robāʿis. Basing his figures on an incomplete manuscript, Morton has calculated that of 86 robāʿis in Ganj al-ganj, “67, that is, nearly 85%” have a four-rhyme scheme (Morton, p. 60). His figures therefore closely resemble the findings of this contributor regarding the divāns of poets contemporaneous with Khayyam.

In this context, a statement by Naṣir-al-Din Ṭusi (q.v.; d. 672/1274) confirms the present contributor’s research: Ṭusi (p. 62) was of the opinion that the more ancient poets composed quatrains in the four-rhyme scheme (aaaa), and that later poets dropped the rhyme from the third half-line (aaba). His view is supported by the observation of Jamāl-al-Din Qarši (circa 702/1302) that poets in Khorasan and western Iran composed quatrains in the four-rhyme scheme (Qarši, p. 3). Given the above evidence, we can conclude that one of the essential guides for ascertaining the authenticity of the quatrains is their rhyme scheme. Confronted by different variants, the verses following the four-rhyme scheme are more likely to be the original ones. This is a more reliable and tested touchstone than the others on offer provided that the authenticity of the available texts is also carefully scrutinized.

A survey of some of the collections of quatrains attributed to Khayyam displays their disregard for the above historical and literary premises. Among the 559 quatrains in Rašidi’s Ṭarab-ḵāna, only 112 are based on a four-rhyme scheme, and in the case of Foruḡi’s edition, 51 out of 179 quatrains follow the four-rhyme scheme. One of the reasons for questioning the merits and value of such collections as the Ṭarab-ḵāna is the stylistic discordance of their offering with what we know of the literary style of Khayyam’s era from other and earlier sources.

Thematic classification of the quatrains. Classifying the quatrains attributed to Khayyam according to their subject matter is another line adopted by scholars to delineate the various thematic contents, key concepts, and images embedded in his robāʿis. As already mentioned, Ḵalil Šarvāni had devoted a specific chapter to the significant themes shared by Khayyam and those influenced by him (pp. 671-76). The fact that in Moʾnes al-aḥrār fi daqāʾeq al-ašʿār the already cited Badr Jājarmi had also allotted a chapter (II, pp. 1144-46) exclusively to Khayyam, indicates that these anthologists were conscious of the particular traits and style peculiar to Khayyam’s quatrains. The anthologist of the Safina-ye kohan-e robāʿiyāt (7th/13th century) compiled most of his selection of Khayyam under the rubric of “On reproaching [the fickle ways] of the firmament and others” (dar maḏemmat-e falak va ḡayr-e ān; pp. 113-21). Railing against the capriciousness of fate appears to be one of the focal topics of Khayyam’s quatrains.

In general, the collections devoted solely to Khayyam’s quatrains are not based on a thematic taxonomy, but Rašidi’s Ṭarab-ḵāna is an exception here. His compilation of the robāʿiyāt is divided into ten sections. The essential topics in his classification are: Divine transcendence (tanzih-e ḵodāvand); rational and philosophical questions (masāʾel-e ḥekami va falsafi); advice and counsel (naṣiḥat va andarz); seizing the moment and the evanescence of pleasing pastimes (eḡtenām-e forṣat va tangnā-ye ʿayš); bacchanalian themes (ḵamriyāt); and the transience of life (gardeš-e ayyām).

Among more recent writers, Ḥosayn Dāneš (1870-1943) was perhaps the pioneer in classifying Khayyam’s quatrains. In 1922, he published in Istanbul a collection of 396 robāʿis with their Turkish translations. His volume comes in two parts: authentic verses and doubtful verses. The first part contains 250 quatrains arranged according to their themes and matter (p. 101): agnosticism (lā-edria); mutability of the world (degarguni-e ʿālam); nihilism (nist-engāri); pessimism (badbini); relishing transient moments (carpe diem; dam ḡanimat šomāri); predestination and fate (jabr and qadar); ironic derision (estehzāʾ). Another set of thematic subdivisions was presented by Ṣādeq Hedāyat in his Tarānahā-ye Ḵayyām (1934): the mystery of creation (rāz-e āfarineš); the pain of existence (dard-e zendegi); the pre-destined decree (az azal nevešta); the transience of times (gardeš-e dowrān); the whirling atoms (ḏarrāt-e gardān); whatever will be will be (har če bādā bād); all for naught (hič ast); savor the moment (dam rā daryābid). The listed captions are not precise enough or sufficiently informative. For example, those quatrains collected under “whatever will be will be” are mostly in praise of wine and drinking and intoxication (Hedāyat, pp. 92-100).

The existence of a so-called Khayyamian School in the context of the robāʿi itself and its thematic contents poses several inherent problems of its own. Many poets, from the earliest times to the present, have written quatrains in this genre and with their passing, many of their verses have found their way into the collection of quatrains attributed to Khayyam. The thematic classification is not, therefore, of much use in ascertaining the authenticity of doubtful robāʿis. The only merit in such a classification is that we can define and demarcate the thematic contents of the so-called Khayyamian verses within the broader horizon of the Persian robāʿi, particularly as we have not, so far, arrived at a consensus regarding their range and scope. For example, there are some quatrains of Khayyam susceptible to a mystical interpretation. Some critics (Dašti, p. 286; Qanbari, pp. 114-15) have questioned the authenticity of such verses as: Dar jostan-e jām-e jam, jahān peymudam / “I scoured the world in search of Jamshid’s bowl” (Ahari, p. 198; Safina-ye kohan, p. 114 with further references; Dašti, tr. Elwell-Sutton, p. 198); or Delhā hama āb gašt-o-jānhā hama ḵun / “Hearts melted all into water and life itself into blood” (Ahari, p. 121; Safina-ye kohan, p. 85 with further references) because of their clear mystical overtones. This in spite of the fact that in general Khayyam seems to have had a favorable attitude toward Sufism and, as expressed in his short treatise, Dar ʿelm-e koliyāt-e wojud (On the existence of universals), had proposed the mystical path as the most suitable approach toward a better understanding of the divine (Reżāzāda Malek, p. 389). Ebn al-Qefṭi’s claim that later Sufis had incorporated the exoteric images and content of Khayyam’s verses into their esoteric discourse (Ebn al-Qefṭi, p. 244; Barg-nisi, p. 65) indicates the inherent potential in his verses for a mystical interpretation and exegesis.

Surveying the various lines of research in recent decades for deciphering the authenticity of Khayyam’s quatrains, one is led to the conclusion that an eclectic approach, drawing upon all the methods, could be the most promising and fruitful. We have no choice but to rely on the more trustworthy sources compiled up to the end of the 8th/14th century. In the later centuries, we face the perennial propensity of the anthologists for expanding the number of the quatrains and their disregard for earlier sources, thereby throwing doubt on their evidence. The older sources too, need to be carefully scrutinized and evaluated. A quatrain that appears in several early sources (the principle of congruence) can be deemed as more reliable (Dašti, p. 23, tr. pp. 37-38). Greater care is needed in the case of the so-called “wandering robāʿis,” attributed simultaneously to several poets. In such cases, the collected works of the poets (divāns) are usually a better guide to authenticity than citations in anthologies and miscellanies.

Close attention to the particular stylistic features of the poetry of the Saljuq period, and in particular the quatrains composed in this era, can prove helpful in establishing the authenticity of the quatrains. Given the nature of the quatrains attributed to Khayyam, and the lack of a trustworthy and closely related corpus of quatrains, we are forced to look at external factors and other criteria. One of these indicators is if the third half-line (meṣraʿ) also rhymes with the rest of the couplet, suggesting an early date for the verse. As for the content, the insertion of a philosophical observation, posing a moral or metaphysical dilemma, or the format of a disputation involving a negation and response are indications of the authenticity of a Khayyamian quatrain (Mirafżali, 1995, p. 11; Barg-nisi, p. 168).

Critics of Khayyam. From the outset, Khayyam’s quatrains have encountered protests in literary circles and amongst poets and writers. For example, Sanāʾi Ḡaznavi (q.v.; ca. 1087/1130), a contemporary of Khayyam—and they may even have sat at the same master’s feet—has two quatrains that appear as a riposte to a quatrain in which Khayyam questions how the notion of death can be accepted as just. In his rebuttal, Sanāʾi likens corporeal existence to the husk or skin of a fruit or a builder’s scaffolding, to be discarded or dismantled once the fruit has ripened or the dome erected, thereby defining and redeeming death as the final stage toward perfection for mankind (Mirafżali, 2015, p. 5).

Two writers from Rayy, both mentioned already, also cite Khayyam so that they can then negate his moral outlook. Faḵr-al-Din Rāzi accused him of shedding spurious doubts on death and resurrection in one of his quatrains (Faḵr-al-Din Rāzi, pp. 59-60). Najm-al-Din Rāzi (Dāya) dismisses him as a wanderer, hopelessly lost, and a believer in timeless eternity (dahri, q.v.) and quotes two of his quatrains to prove his point (Najm-al-Din Rāzi, pp. 30-31, 400; tr. p. 54, 387). During the same early period, Jamāl-al-Din Abu’l-Ḥasan Ebn al-Qefṭi chastises Khayyam’s poetry in his Taʾriḵ al-ḥokamāʾ (p. 244) and points to what he conceives as its inherent dangers, “The later Ṣúfís have found themselves in agreement with some part of the apparent sense of his verse, and have transferred it to their system, and discussed it in their assemblies and private gatherings, though its inward meanings are to the [Ecclesiastical] Law stinging serpents, and combinations rife with malice” (tr. Browne, 1906, p. 250; Barg-nisi, p. 65).

While ʿAbd-al-Qāder Ahari in his al-Aqṭāb al-qoṭbiya bestows effusive epithets on Khayyam (Ahari, pp. 121, 198), in another passage in quoting Khayyam’s quatrains, he follows the same line as Najm-al-Din Rāzi and includes Khayyam among the deniers of the day of resurrection and prophethood (Ahari, p. 175, 204). Among the collected sayings of Šams-al-Din Moḥammad Tabrizi (d. ca. 646/1248) there is also a passage where Khayyam is described, in contrast to men of faith, as lost in bewilderment, capable of naught save confused and dark thoughts (Šams-e Tabrizi, p. 301, tr. de Fouchécour, p. 374). Finally, in the section below, we will discuss the translation into Arabic of one of Khayyam’s quatrains by Qāżi Neẓām-al-Din Eṣfahāni (d. 680/1281). The translation is not offered as a token of approval but, on the contrary, as a proem to criticize his verse and is immediately followed by four quatrains as a rebuttal (Neẓām-al-Din, f. 176; Mirafżali, 2002, pp. 23-24).

Translations of the “Robāʿiyāt”. There are numerous translations of Khayyam into different languages, as can be surveyed in the major bibliographies of Khayyam including those of A. G. Potter (1929), Fāṭema and Zahrā Angurāni (2002), and Jos Coumans (2010) as well as in other entries in this Encyclopaedia (see KHAYYAM iv, vi, vii, viii, ix, x).

The earliest translation of Khayyam is the verse translation by the above-mentioned bilingual poet, Qāżi Neẓām-al-Din Eṣfahāni, in his collection of Arabic quatrains, Noḵbat al-šāreb waʿojālat al-rākeb. It contains 550 quatrains in Arabic and 50 in macaronic (molammaʿ) form, mixing Arabic with Persian in the same verses. As pointed out above, after translating a quatrain by Khayyam into an Arabic quatrain, Neẓām-al-Din proceeds to present four Arabic quatrains of his own as a critical response and rejoinder. In the modern period, Khayyam has frequently been translated into Arabic, one of the most famous translations being that of Ṣafi Najafi (351 robāʿis, 1926); see also below KHAYYAM x.

The earliest references to Khayyam in pre-modern European sources is perhaps in Joseph Scaliger’s (1540-1609) study of calendar systems, Opus de Emendatione Temporum (1583), where “Omar Elhaiamu” is listed along with “Elbiruni” and “Aben Sina” and other men of science (Scaliger, p. 304). Reference to Khayyam as a poet and Latin versions of his quatrain first appeared in England in Thomas Hyde’s (q.v.; 1636-1703) Historia religionis veterum Persarum (pp. 498-500).

Forged or altered manuscripts. The popularity of the Robāʿiyāt worldwide and the perennial quest for old and authentic manuscripts have led to a series of claims of discoveries of significant early manuscripts that have later turned out to be forgeries, although at first defended, at times half-heartedly, by eminent scholars (Csillik, p. 61). These have been discussed in other entries along with forgeries of other Persian manuscripts (see FORGERIES iv. OF ISLAMIC MANUSCRIPTS), as well as in the entry on English translations (see KHAYYAM iv. English Translations of the Rubaiyat) regarding the case of a mysterious and perhaps non-existent 12th-century manuscript of which only a prose translation exists by Omar Ali Shah and a free verse rendering of that by Robert Graves (Bowen, pp. 63-73).

Along with the forged manuscripts briefly referred to above, there are a few genuinely old but doctored manuscripts whose dates have been tampered with to advance their age and enhance their value. Among these is the manuscript dated 721/1321 that Rosen used for his Berlin edition of 1925. It has 300 quatrains and is transcribed in the nastaʿliq style (see CALLIGRAPHY). Arthur Christensen suggests (p. 47) a date around 900/1500 as its true date, a view repeated by A. G. Potter (p. 159, no. 526). Another manuscript (dated 790/1388), containing 333 quatrains, was presented in Esmāʿil Yakāni’s book on Khayyam and his Robāʿiyāt in 1963. The manuscript originally belonged to Ḥosayn Naḵjavāni’s private library in Tabriz but is now in the collection of the late Aṣḡar Mahdavi. After inspecting the manuscript, it became clear that the date had been tampered with (Mirafżali, 2003, p. 250; 2005-6, pp. 377-80; Ḏakāvati Qaragozlu, p. 155).

Sayyed ʿAli Mirafżali

KHAYYAM, OMAR iii. IMPACT ON LITERATURE AND SOCIETY IN THE WEST

The first scholar outside Persia to study Omar Khayyam was the English orientalist, Thomas Hyde (q.v.; 1636-1703). In his Historia religionis veterum Persarum (1700), he not only devoted some space to the life and works of Khayyam, but also translated one quatrain (robāʿi) into Latin. The first quatrain in English was published in 1816 by Henry George Keene (1781-1864) in the famous magazine Fundgruben des Orients/Mines d’Orient. Although the founder of the Fundgruben, Joseph von Hammer-Purgstall (q.v.; 1774-1856), translated a few of Khayyam’s poems into German in 1818, and Sir Gore Ouseley (q.v.; 1770-1844) into English in 1846, Khayyam was to remain relatively unknown for some time (Dole, I, pp. ix-xv).

In 1859, the London bookseller Bernard Quaritch published the first edition of The Rubáiyát of Omar Khayyám. The translator, Edward FitzGerald (q.v.; 1809-83) had 250 copies printed anonymously, of which 40 copies were for his own use. He distributed copies among a few friends, but although some advertisements tried to draw attention to the poem, it remained “spectacularly unsuccessful” (Decker, pp. xxxiii-xxxiv). In 1861, the booklet ended up in Quaritch’s remainder box, where it was offered for a penny a piece. No copies were sold until Whitley Stokes, a Celtic scholar, bought one in 1861. He came back to buy additional copies, one of which he gave to Dante Gabriel Rossetti. From Rossetti, the poem found its way to Algernon Swinburne and George Meredith, both of whom sang its praises and passed on their enthusiasm to other members of the Pre-Raphaelite Brotherhood, including William Morris and Edward Burne-Jones. The latter showed the book to John Ruskin, who, in 1863, wrote a letter to the still unknown translator/author of the Rubáiyát in which he declared that he had never read anything so glorious to his mind than this poem and begged for more. The Pre-Raphaelites were fascinated by the Rubáiyát. So it was that Swinburne wrote his Laus Veneris in the Omarian stanza (1866), and Morris and Burne-Jones wrote out and illuminated a copy on vellum, which was given to Burne-Jones’ wife, Georgiana, in 1872. It was also through Burne-Jones that the young Rudyard Kipling discovered the poem. The poem’s rise in fame is described in detail by Carl J. Weber (pp. 17-31) and John Arthur Arberry (pp. 23-30). In the 1890s, the popularity of FitzGerald’s Rubáiyát had risen to great heights, not only in Britain, but also in America.

In America, the poem had been introduced by Charles Eliot Norton (1827-1908), the renowned scholar and man of letters. When he visited England in 1868, Georgiana Burne-Jones showed him her husband’s copy of the Rubáiyát. Norton got hold of a copy of the second FitzGerald edition (1868) and brought it to the attention of American friends, including James Russell Lowell and Ralph Waldo Emerson (q.v.). In 1869, he published a laudatory article in the North American Review. The article aroused the interest of many fellow Americans and also drew the attention of readers in England. Many people wanted to own a copy of the poem. A pirated printing of the 1868 edition appeared in Columbus, Ohio, in 1870, and, in 1872, FitzGerald had a third edition printed by Quaritch. This found its way to many admirers in America where, as in England, there was an increasing interest and demand for the work. A further impetus to the book’s popularity came from the artist Elihu Vedder, whose Rubáiyát edition, in an attractive cover and embellished with 56 drawings, caused a sensation when it was published in 1884. The exhibition of the original drawings at the Arts Club, Boston, drew up to 2100 visitors a day. The book sold out in six days and was reprinted several times (Soria, pp. 183-86). Other editions were reprinted many times as well: “Competing editions cropped up like dandelions all over the literary lawn” (Weber, p. 30). As John D. Yohannan points out, the adoration of FitzGerald and the epicurean vein that runs through his translation “produced the true agents of the fin de siècle cult of the Rubaiyat; namely, the Omar Khayyam Clubs of England and America” (Yohannan, pp. 202).

In 1892, admirers of the poem founded the Omar Khayyám Club of London. Mr. Justin Huntly McCarthy, who had published his own Robāʿiyāt version in 1889, was elected its first president. Similar clubs were founded in other places but, with the exception of the American club, left almost no traces. The main activity of the members of the London club was (and still is) gathering twice a year to dine and to commemorate Omar, Fitz (as he is called as a term of endearment), and the Rubáiyát, in more or less serious and comic rituals. The London club issued two books, in 1910 and 1931 respectively, with contributions by its members: poems praising Fitz and Omar, pictures of menus, miscellanea, and lists of members and guests. Among these can be found many famous names: scholars, literary men and artists such as Edward Heron-Allen (q.v), Edmund Gosse, Andrew Lang, Thomas Hardy, Max Beerbohm, Lawrence Alma Tadema, Arthur Rackham, G. K. Chesterton, Arthur Conan Doyle, Aldous Huxley, and W. B. Yeats, to name but a few (Book of the Omar Khayyám Club, I and II, pp. 211-20, 168-81). Membership was originally restricted to men, but from 1910 onwards women were allowed to attend special dinners as guests.

Eight years after the foundation of the London club, the Omar Khayyám Club of America was founded. In 1900, on the 31st of March, FitzGerald’s anniversary, Eben Francis Thompson, Nathan Haskell Dole, and others held the first session of the club in Boston. The club was formed “on the basis of good fellowship as well as Oriental learning, with good fellowship as the predominant feature” (Twenty Years, p. 7). Just as in London, members of the club had drawings made for the menu cards of their dinners, often in an elegant Art Nouveau style. The character of the American club differed from the London club. Its members did not only pay attention to Khayyam’s poetry in FitzGerald’s rendition, they also explored a more scientific approach to the subject. They published their own translations, printed beautiful and costly books and contributed to meetings with lectures on the philosophical and mathematical aspects of Omar Khayyam. The club passed into oblivion around 1930, but from its publications, most notably Twenty Years of the Omar Khayyám Club of America (1921), one can appreciate its many contributions.

In the chapter “The Cult of the Rubaiyat,” Yohannan describes how the Omar cult developed in England and America, and produced an anti-cult to combat it (Yohannan, pp. 199-244). He gives a lively account of this strife. But Omar Khayyam, introduced to the public in the form FitzGerald had shaped him into, was not only the hero (or anti-hero) of scholars, men of letters and artists; for “Omaritis” had taken hold of businessmen, schoolgirls, and soldiers as well. Especially in America, we find a real “Omar craze.” In Ambrose Potter’s Bibliography, published in 1929, we find not only a long list of books and articles pertaining to Omar Khayyam, but also an enumeration of him in different arts and in commercial advertisements: music, drama, films, dances, bookplates, and also tobacco, cigarettes, cigars, fountain pens, coffee, chocolate, perfume, toilet soap, pottery, post cards and crossword puzzles, deriving their name and attraction from the old Persian poet (Potter, p. 205). Omar Khayyam had become an everyman’s poet. Eminent men of letters, like Rudyard Kipling, Mark Twain, and Arthur Quiller Couch, had borrowed FitzGerald’s quatrains in order to turn them into parodies. After them countless versifiers made their own Rubaiyats, in which almost every subject that moved people could be parodied (Biegstraaten, p. 33). The Omar Khayyam craze began to fade away in the 1920s.

No doubt Omar Khayyam had the greatest impact on literature and society in countries where English was spoken. Among the writers who were deeply influenced by FitzGerald’s Khayyam were T. S. Eliot and Ezra Pound (D’Ambrosio, passim). But Omar Khayyam was translated and discussed in other countries as well, especially in France and Germany. In his bibliography in 1929, A. G. Potter (pp. 133-58) listed 114 versions of Omar Khayyam’s Robāʿiyāt in 25 languages, of which 18 versions are in French and 37 in German. And an endless stream of editions and translations was to follow in the years after 1929.

In France, Omar was introduced by F. Woepcke, who published L’Algèbre d’Omar Alkhayyàmì in 1851. In 1867, J. B. Nicolas published Les Quatrains de Khèyam, containing 464 quatrains in Persian and French, with extensive comments and notes. Although FitzGerald rejected the Sufi interpretations of Nicolas, the translation of the latter was an important source for his 1868 rendering. Nicolas’ work has been reprinted and translated many times. After Nicolas, many scholars and translators published their own versions of the Robāʿiyāt, among them Charles Grolleau, Franz Toussaint, Claude Anet and Mirza Muhammad [Qazvini, q.v.], Arthur Guy, A.-G. E’tessam-Zadeh, and Mahdi Fouladvand.

In his translation, Rubaijat von Omar Chajjam, Henry Nordmeyer mentions 31 translations of the Robāʿiyāt in German from 1881 to 1963 (Nordmeyer, pp. 102-4). Among the translators we find well-known figures, such as Friedrich Bodenstedt, Graf Adolf von Schack, Friedrich Rosen, Hector Preconi, Hans Bethge, and Christian Herrnhold Rempis. Rempis, a professor at Tübingen University, founded a German version of the English and American Omar Khayyám clubs. The German club had its own publishing company, Verlag der Deutschen Omar Chajjám-Gesellschaft, where Rempis issued his translation, Omar Chajjam und seine Vierzeiler, in 1935. After the Nuremburg laws of 1935, Jewish members of the club were no longer able to take part in its activities, and the club was dissolved in 1937. In Nazi Germany, Khayyam was out of favor. As Rempis stated in 1960, “Omar Khayyam’s attitude towards life was not particularly compatible with Nazi doctrine.” It was only after the end of the war that the reprints and new translations of Omar Khayyam found their way to German readers (Gittleman, pp. 189-93).

Although the impact of Omar Khayyam on the literary and social scene did not hold the fascination it had at the end of the 19th and the beginning of the 20th century, there remains to this day a lively interest in his life, philosophy, and poetry. Freethinkers, pictorial artists, composers, choreographers, and poets all over the world were inspired by his work. Mostly, they were influenced by FitzGerald’s renditions.

There even appeared a new club in The Netherlands in 1990: Het Nederlands Omar Khayyám Genootschap (The Dutch Omar Khayyám Society). Members gather twice a year and organize plenty of activities. So far the club has published four Jaarboeken (Year Books) in Dutch, printed by the “Avalon Pers,” a private press belonging to one of its members. It was also responsible for two exhibitions on Omar Khayyam, one in The Hague Museum of the Book, and the other in the Library of Leiden University (Aminrazavi, pp. 275-77).

Jos Biegstraaten

KHAYYAM, OMAR iv. ENGLISH TRANSLATIONS OF THE RUBAIYAT

Over the past 150 years, the quatrains of Khayyam have been translated into English more often than the verse of any other Persian poet. The bibliographies of Ambrose Potter and Jos Coumans together list nearly one hundred translators and editors for the Rubaiyat in English. Out of this mass of material, however, only a few dozen translations enjoyed considerable circulation or exerted lasting influence on the tradition of the Rubaiyat in English. These can be heuristically divided into two categories: those based directly on the Persian, and those based on previous translations in English or other languages.

Scattered quatrains by Khayyam had been translated into English by Gore Ouseley (q.v., 1770-1844) and the Rev. Henry George Keene (1781-1864), but it was the verse paraphrase of Edward FitzGerald (q.v., 1809-83) that made Khayyam a household name for Anglophone readers (PLATE I). FitzGerald’s rather free translation was based on two manuscripts, the now famous Ouseley manuscript in the Bodleian Library and a much later manuscript held in the library of the Asiatic Society in Calcutta, which has since been lost (Elwell-Sutton, p. 173). Both manuscripts were available to FitzGerald in copies provided to him by his friend Edward Cowell (q.v.), who, in 1858, published an article on Khayyam in the Calcutta Review that included lineated prose translations of a number of quatrains. One year later, FitzGerald published his own translation in iambic pentameter following the common aaba rhyme scheme of the Persian. Although robāʿis are always independent poems, FitzGerald’s Rubáiyát forms a larger structured whole. He writes of the poem as “most ingeniously tesselated into a sort of Epicurean Eclouge in a Persian Garden” (FitzGerald, 1980, II, p. 323). It initially failed to sell, but the work’s popularity increased steadily over the next decade. In 1868, a second, revised edition was published; in addition to other changes, it included an expanded introduction in which FitzGerald criticized J. B. Nicolas’ 1867 translation of the Rubaiyat, done into French prose, in which Khayyam was represented as a devout Sufi (see KHAYYAM vii). Two more editions were to follow during FitzGerald’s lifetime, and it has been reedited and reprinted many times since.

PLATE I Cover of first edition of FitzGerald’s Rubáiyát.

As the popularity of the Rubáiyát grew in the final decades of FitzGerald’s life and after his death, a wave of new translations was produced by amateur orientalists—often ex-civil servants of the British government in India—keen to offer more “literal” representations of Khayyam’s poetry. Among verse translators, one of the earliest and most influential was E. H. Whinfield of the Bengal Civil Service. In 1882, he published a translation of 253 quatrains; in 1883, he published an edition of the Persian text of 500 quatrains, from a variety of sources, along with verse translations; and in 1893, he published a new translation of 267 quatrains. His translations keep the form established by FitzGerald, but hew closer to the original Persian, and they are organized alphabetically according to the Persian rhyme letter. Another early translator in colonial service was E. A. Johnson (a.k.a. Johnson Pasha), who, between 1887 and 1913, published several versions of his Khayyam translation based on a lithograph edition from Lucknow. In 1898, the English litterateur John Payne released a translation of 845 quatrains, which was also largely based on the Lucknow lithograph. Payne’s translations do not follow a fixed English meter, but rather seek to imitate “the different rhythms” of the Persian and preserve the radif (Payne, p. lxxi); the result is not always easy to decipher. A similar approach was taken Michael Kerney, who worked as a cataloger for Bernard Quaritch, the famous bookseller and distributor of FitzGerald’s Rubáiyát; Kerney produced a handful of quatrains “literally rendered into the meter and according to the rhyme of the original” (Garrard, p. 143; Dole, p. lxxxii). A translation by Jessie Cadell was published posthumously in 1899; she had learned Persian in India, where her husband had been stationed, and had published a scholarly article on FitzGerald and Khayyam in Fraser’s Magazine, in which she criticized FitzGerald’s translation as “a poem on Omar, rather than a translation of his work” (Cadell, 1879, p. 650). Like FitzGerald, she renders the quatrains into iambic pentameter, but the rhyme scheme varies. In 1910, Alexander Rogers, a prolific translator of Persian and former member of the Indian Civil Service, published a translation of 160 quatrains. In 1915, a “line for line” verse translation in tetrameter was published by John Pollen, president of the British Esperanto Association; his translation is introduced by Sultan Muhammad Shah (1877-1957), the third Aga Khan (see ĀQĀ KHAN iii), and the book includes an appendix with Esperanto renderings of the opening quatrains of FitzGerald’s version.

Contemporary with the above-mentioned verse translations, Khayyam was also being translated into English prose. The first major prose version was published in 1889 by Justin Huntly McCarthy, an Irish politician and litterateur who taught himself Persian expressly to read Khayyam. He writes that he chose prose because it “can give the meaning more nearly than any verse could” and that it would be “absurd” to attempt a verse translation and thereby invite comparisons with FitzGerald (McCarthy, p. xi). He includes 466 quatrains, presumably taken from the Whinfield and Nicolas editions. The influence of Nicolas’ translations is apparent in some of his own (Dole, p. lxxix). In 1898, the polymath Edward Heron-Allen (q.v.) published a facsimile and transcription of the Ouseley manuscript that included a lineated prose translation for each quatrain. The following year he compiled a concordance of Fitzgerald’s Rubáiyát, attempting to identify the Persian original (or originals) behind each English rendering, and again including his own translations. (A similar concordance, but accompanied by “literal translations” in verse, was compiled by E. H. Rodwell in 1931). In 1928, the German diplomat and orientalist Friedrich Rosen, who had previously translated Khayyam into German, published a lineated English prose translation of 329 quatrains from an allegedly early manuscript, the Persian text of which he had edited in 1925.

FitzGerald’s paraphrase was received even more enthusiastically in the United States than in Britain, and a number of Americans were inspired to translate Khayyam during the late 19th and early 20th centuries. John Leslie Garner, a resident of Milwaukee and a prolific translator from European languages, published a translation of Khayyam in 1888 entitled The Strophes of Omar Khayyam; it was republished in 1898 as The Stanzas of Omar Khayyam. In 1906, E. F. Thompson, a Massachusetts lawyer, bibliophile, and secretary of the Omar Khayyám Club of America, published a translation of 878 quatrains in the standard FitzGeraldian form, by then practically a recognized English verse form in its own right. One of his fellow club members, George Roe of San Antonio, Texas, published his own translation the same year; it included only 122 quatrains, but with copious notes and references. At that time, the president of the Omar Khayyám Club of America was Nathan Dole; he did not produce a translation himself, but he did compile a massive “multi-variorum edition” of FitzGerald’s quatrains, collated with other English, German, French, Italian, and Danish translations. Also noteworthy is the translation by the Rev. Isaac Dooman, who was born in Persia, educated in the United States, and served as an Episcopalian missionary in Japan. It was published in 1911 and contains 180 quatrains in decasyllables in the standard rhyme scheme. After the onset of World War I, new translations were published at a considerably slower rate, in both the United States and in Britain. Nevertheless, they continued to be produced. In 1933, for example, a new translation in (mostly) unrhymed hexameter was published by David Eugene Smith, a professor of mathematics at Columbia University; he did not know Persian, but worked from a “verbatim translation” provided by his collaborator, Hashim Hussein (PLATE II).

Beginning in the 1920s and continuing throughout the century, Khayyam was rendered into English by a number of Indian (and later Pakistani) scholars and translators, including versions by Jamshedji Saklatwalla (1922), A. R. Tariq (1968), and A. C. Bose (1977, published posthumously). Govinda Tīrtha’s study of Khayyam, The Nectar of Grace (1941), also included copious original translations.

Early translators such as Cadell and Whinfield recognized that many of the quatrains in the Khayyamian corpus were likely spurious, and concerns about the authenticity of their source material intensified in the first decades of the 20th century with the work of Valentin Alekseevich Zhukovskiĭ (q.v.) and Arthur Christensen (q.v.).

Thus, it was with great excitement that scholars and translators learned of two early manuscripts, dated 1259-60 and 1207-8 and acquired by Chester Beatty (see CHESTER BEATTY LIBRARY) and the Cambridge University Library in 1948 and 1950, respectively. Neither manuscript contained new quatrains, but they suggested that Khayyam was recognized as a poet earlier than previously thought, and that a core of several hundred poems lay at the heart of the ever-expanding later corpus. In 1949, A. J. Arberry (q.v.) published a transcription of the Chester Beatty manuscript along with a “literal prose translation,” and, in 1952, he published a complete verse translation of the Cambridge manuscript. Instead of following the poetic form set by FitzGerald, he opted to translate each quatrain into two abba stanzas in iambic tetrameter on the model of Tennyson’s In Memoriam. In 1961, J. C. E. Bowen published a translation of sixty quatrains from the Cambridge manuscript using several different verse forms, accompanied by Arberry’s prose translations. The authenticity of these early manuscripts, however, was challenged by scholars such as Vladimir Minorsky and Mojtabā Minovi (qq.v.), and now most scholars agree that they are forgeries, as Bowen himself acknowledged in the introduction to the 1976 edition of his translation.

The specter of forgery was raised yet again in 1967 with the publication of a controversial collaborative translation by Omar Ali-Shah, brother of the popular Sufi teacher Idries Shah, and Robert Graves, the famed English poet. Their translation is allegedly based on a 12th-century manuscript in the possession of the Shah family, from which Ali-Shah made a literal prose translation and Graves then rendered into free verse. In the introduction to the work, they both criticize FitzGerald for taking liberties with the text and obscuring its sufistic orientation. Bowen and L. P. Elwell-Sutton (q.v.), however, have convincingly argued that this manuscript (which has never been made available to scholars) cannot be authentic, if it exists at all, and that the ultimate source of the translation is Heron-Allen’s 1899 concordance.

PLATE II The Smith translation of the Rubaiyat, 1933.

Beginning in the 1930s, attempts were made by Iranian scholars to isolate an authentic core of quatrains by comparing the themes and style of the earliest attested verses with those in the expanding corpus, assuming that genuine quatrains by Khayyam should exhibit some level of consistency. Ṣādeq Hedāyat, ʿAli Dašti, and Moḥammad ʿAli Foruḡi (qq.v.) all produced small collections using this procedure, which is admittedly rather subjective. Nevertheless, these collections, especially the 178 quatrains collected by Foruḡi, have formed the basis of many translations since the 1970s, including the lineated prose translation of Parvine Mahmoud (1969), the prose translation of Parichehr Kasra (1975), the free verse rendering of Peter Avery (q.v.) and John Heath-Stubbs (1979), and the rhyming translation of Sunil Ray (1988).

Since the 1970s, more and more translations of Khayyam into English have been made by native speakers of Persian, often Iranians living abroad or with professional interests in English translation. In 1971, the lexicographers Abbas Aryanpur-Kashani and Manoochehr Aryanpur published a rhyming translation of 154 quatrains. In 1973, Mehdi Nakosteen, a professor of education at the University of Colorado Boulder, rendered 398 quatrains “from various Persian editions of Khayyam” in the standard FitzGeraldian form (Nakosteen, p. xiv). In 1991, Ahmad Saidi published a translation of 165 quatrains, arranged thematically, with meticulous notes and cross-references; he too uses the traditional iambic pentameter and rhyme scheme. A large, bilingual collection of Persian quatrains from different poets was published in 2000 by Reza Saberi under the name A Thousand Years of Persian Rubaiyat, including 173 quatrains by Khayyam. Although they do not rhyme, a radif-like structure has been preserved in some of them. Shahrokh Golestan’s Wine of Nishapur, published in 1988, combines the free verse translations of Karim Emami (q.v.) with the calligraphy of Nassrollah Afje’i and Golestan’s own quietly meditative photographs; it is a welcome alternative to the orientalist imagery found in many editions of FitzGerald’s Rubáiyát.

Although most translators working directly with the Persian have presented Khayyam as a hedonist and skeptical epicurean, or at the very least a multifaceted individual prone to moments of doubt and free-thinking, others such as Nicolas, Tariq, Tīrtha, and Graves have advocated mystical readings of the quatrains, a tradition that has continued in recent decades. In 1984, a set of free verse translations by Iftikar Azmi was published in a luxurious, folio-sized edition by The Whittington Press under the title The Mirror and the Eye. Its introduction characterizes Khayyam as a Sufi and presents the quatrains as products of his various mystical states. A slim set of translations by Nahid Angha, published by the International Association of Sufism (of which Angha is a co-founder), includes Khayyam as a “Sufi poet.” Another example is the collaborative translation of Mary O’Connell and Roshanak Vahdani, published in 2004, which includes a glossary of the allegedly symbolic meaning of Khayyam’s bacchic and amatory imagery.

Many English versions of Khayyam are not based directly on the Persian, but are reworkings of French, German, or previous English translations. Such “indirect translations” were especially popular at the end of the 19th century and in the first decades of the 20th. Several were made by well-known poets, but most were composed by amateurs who simply felt an affinity with Khayyam and FitzGerald and wanted to engage with the material on the model of the latter. When they display some level of fidelity to their sources, they can be distinguished from the many parodies, responses, and original poems inspired by the Rubaiyat (often misleadingly presented as “translations”), which are beyond the scope of the present survey.

The earliest indirect translation was by Louisa Costello, who, in 1840, translated a handful of quatrains into English from Joseph von Hammer-Purgstall’s (q.v.) 1818 German translation (Potter, p. 103). She was followed by Ralph Waldo Emerson (q.v.), who was deeply influenced by Persian poetry, which he also accessed via the German; he translated verses from Saʿdi and Hafez (qq.v.) into English, as well as three quatrains from Khayyam (Yohannan, pp. 299-302). In 1862, Whitley Stokes, who is said to have first “discovered” FitzGerald’s Rubaiyat and circulated it among the Pre-Raphaelites, rendered a handful of quatrains into English verse from the prose translations of Cowell and Garcin de Tassy, and included them in a pirated reprint of FitzGerald’s poem that he produced in Madras (Drew, p. 96).

The most successful indirect translation, in commercial terms, was doubtlessly that of Richard Le Gallienne, first published in 1897. In his introduction, he proclaims his inability to read Persian an advantage, and professes his chief obligation to be to McCarthy’s prose translation (Le Gallienne, p. 13). But his ultimate model was, of course, FitzGerald; he maintains the form FitzGerald popularized and adopts his method of arrangement. Although Le Gallienne’s version was a critical flop, it sold very well, with multiple editions published in the United States and Britain (Potter, pp. 110-11). The Le Gallienne translation itself became source material for further renderings. In 1936, Frank Ankenbrand reworked several of Le Gallienne’s quatrains into a collection of self-styled “vignettes”—short haiku-like poems—that he published under the title A Persian Rose Garden.

McCarthy’s prose was an especially popular source for non-Persophone versifiers. The famed folklorist and poet Andrew Lang reworked a handful of McCarthy’s translations for his Ban and Arrière Ban in 1894. A few years later, Frederick York Powell, a professor of modern European history at Oxford, published twenty-four quatrains “turned into English on the familiar model from M. Nicolas and Mr. Justin McCarthy’s versions” (Powell, p. 18). In 1901, Charles G. Blanden, amateur poet and former mayor of Fort Dodge, Iowa, published Omar Resung, a versification based on McCarthy’s prose; he reworked each quatrain into eight lines of iambic trimeter. In 1899, the New England poet and playwright Elizabeth Alden Curtis published a collection of one hundred quatrains, introduced by Richard Burton, the famous translator of The Arabian Nights. Although she does not specify her sources, she seems to have been working from Heron-Allen’s prose translations, just as she rendered his prose translations of Bābā Ṭāher ʿOryān’s (q.v.) quatrains into verse in 1902. In 1909, Arthur Talbot also published a “literal” versification based on Heron-Allen, as did C. S. Tute in 1926.

Other indirect translators relied primarily or exclusively on Nicolas, including Baron Corvo (a.k.a. Frederick Rolfe) an English writer, artist, and photographer who produced a rather stilted prose translation in 1903; Francis Dyson, who translated 172 of Nicolas’ quatrains into “musical verses of varied metre” in 1916 (Dyson, p. 6); and Horace Thorner, who published a verse translation of 101 quatrains in 1955, based on Nicolas and FitzGerald’s posthumous fifth edition.

The production of indirect translations slowed as the century progressed, excepting a blip in 1959, the centennial of FitzGerald’s first edition. That year, Ankenbrand published another collection of vignettes, this time based on McCarthy’s prose translation, under the title Kings in Omar’s Rose Garden. I. D. du Plessis, a major South African poet who had previously written only in Afrikaans, also published an English version of the Rubaiyat, based on Arberry’s translation of the Cambridge manuscript. In 1968, W. G. Burton released one of the last widely available indirect translations of note, based on his readings of FitzGerald, Winfield, Payne, and Arberry. Neither a poet nor an orientalist, but an agricultural scientist, Burton represented a continuation of the amateur literary engagement that had characterized the Rubaiyat tradition in English for the past hundred years.

Austin O'Malley

KHAYYAM, OMAR v. ILLUSTRATIONS OF ENGLISH TRANSLATIONS OF THE RUBAIYAT

The Rubaiyat (Robāʿiyāt, ‘quatrains’) of Omar Khayyam (ʿOmar Ḵayyām) contain some of the best-known verses in the world. The book is also one of the most frequently and widely illustrated of all literary works, a remarkable feat for a work that is relatively short in length and abstract in content. The stimulus to illustrate Khayyam’s Rubaiyat came initially from outside Persia, in response to translations in the West, particularly the famous version by Edward FitzGerald (q.v.), first published in London in 1859. In subsequent years, modern Iranian artists and publishers have also taken up the illustration of the Rubaiyat .

PLATE I Miniature by Behzād, claimed as one of the first illustrations of the Robāʿiyāt of Omar Khayyam. The miniature is contained in a manuscript dating ca. 1500, published in facsimile by the Indian scholar M. Mahfuz-ul-Haq in 1939.

The history of Rubaiyat illustration. A great deal of uncertainty surrounds the authorship of the verses that have been attributed to Omar Khayyam. This may partially explain why manuscripts of the poem were seldom if ever illustrated, in contrast to the many miniatures contained in manuscripts of Ferdowsi’s Šāh-nāma or Neẓāmi’s Ḵamsa (q.v.). The earliest known illustrated version of the Rubaiyat dates from around 1500, and was published in a facsimile edition by the Indian scholar M. Mafuz-ul-Haq in 1939 (Mafuz-ul-Haq, pp. 1-18). The manuscript contains several miniature paintings, including at least one attributed to the 15th-century painter Behzād (q.v.; PLATE I). Most other early manuscripts containing collections of the Rubaiyat attributed to Khayyam have, if anything, a simple form of decoration. They include the famous Ouseley manuscript in the Bodleian Library in Oxford, dated 1460-61, which was used by Edward FitzGerald as one of the main sources for his first presentation of the Rubáiyát in English in 1859 (Arberry, pp. 41-42). The earliest translations of Khayyam’s Rubaiyat published in the West do not contain illustrations. Nor do the first three editions of FitzGerald’s version of the poem. FitzGerald’s 4th edition, published in 1879, had a frontispiece Persian drawing, but the picture refers to his translation of Jāmi’s Salāmān o Absāl (see JĀMI i), which was presented in the same volume.

The first fully illustrated version of the Rubaiyat in the West is that published in Boston, Massachusetts, in 1884 by Houghton Mifflin, based on FitzGerald’s third edition, with drawings specially commissioned from the American artist Elihu Vedder (PLATE II; Martin and Mason, pp. 12-13). This lavishly illustrated edition by Vedder was reissued several times in the decade following its first appearance. Meanwhile, from the 1870s onward, there was a regular flow of new editions of the Rubaiyat, but without illustrations, including other translations into English, as well as into French, German, and other languages. It was not until 1898 that the publication of illustrated editions began to take off. There were seven different illustrated versions of FitzGerald’s Rubáiyát in that year, by six new artists; the work by two of them, Gilbert James (PLATE III) and Edmund Garrett, was included in more than one new version (Martin and Mason, p. 21). From then on, as the chart (Figure 1) shows, the trickle of illustrated editions became a flood, reaching a peak in the years 1909-10; 1909 marked the 50th anniversary of the publication of FitzGerald’s first edition and the 100th anniversary of his birth.

PLATE II Illustration by Elihu Vedder (1836-1923), attached to quatrain 43 in FitzGerald’s third edition, published by Houghton Mifflin in 1884.

Many of these new illustrated Rubaiyats were published in the United States as well as in the United Kingdom. The remarkable growth in interest in this one collection of verses reflects both the attraction of the verses themselves and the philosophy inherent in them (as interpreted by Edward FitzGerald), as well as the concurrent developments in printing technology that enabled book illustrations to be presented in a cheaper and more attractive form. In addition, with rising affluence, the book market was expanding, and the Rubaiyat, especially with illustrations or decorations and elegant bindings, made excellent material for attractive, popular versions, presented as gift books, special Christmas editions and calendars (Martin and Mason, pp. 8-10).

Figure 1. History of publication of illustrated editions of the Rubaiyat of Omar Khayyam from 1859 (date of FitzGerald’s first edition) to 2004. The figures cover illustrated editions by all translators and in all languages so far identified. They include new editions and reprints of earlier illustrated versions, where known; the number of reprints is believed to be underestimated.

By the end of 1909, there had been more than 100 new illustrated or decorated editions of FitzGerald’s Rubáiyát, containing the work of more than 50 different artists. Many of them were key figures in the Art Nouveau movement, including Edmund Dulac, Rene Bull, Robert Anning Bell, and Jessie King. Dulac’s famous work was among the 15 new illustrated editions in 1909 alone (PLATE IV). The chart (Figure 1) shows that the number of new illustrated Rubaiyats subsided in subsequent years, particularly in the middle years of World War I. But interest picked up again from 1917, and there are few years in the entire period up to the present day in which there has not been either a new illustrated edition of the Rubaiyat or a reissue of an existing version (Martin and Mason, p. 13); the 2001 edition illustrated by Andrew Peno is an example of a fairly recent work.

PLATE III Illustration by Gilbert James (fl. 1865-1941) for quatrain 13 in FitzGerald’s first edition, published by Leonard Smithers in 1898.

It is of particular note that, since the 1920s, the publication of illustrated versions of the Rubaiyat by translators other than FitzGerald has grown in significance. Interest in the Rubaiyat has spread round the world, with the appearance of versions in more than 70 different languages (Martin and Mason, p. 3). These editions have been illustrated less frequently than those of FitzGerald’s text, but quite a number do contain illustrations, some reissuing work that the artists originally created for FitzGerald editions; work by Dulac, James, and Willy Pogany has been used in this way. In terms of Western countries, illustrations by new artists are particularly evident in editions from France, Germany, Hungary, the Netherlands, and Spanish-speaking countries. A significant number of new editions of Khayyam’s Rubaiyat have appeared in Iran, both before and after the 1979 revolution. These usually contain the original Persian Robāʿiyāt, along with translations in various other languages, usually lavishly decorated with illustrations by modern Iranian artists (PLATE V; Martin and Mason, pp. 25-27).

PLATE IV Illustration by Edmund Dulac (1882-1953) for quatrain 12 in FitzGerald’s second edition, published by Hodder and Stoughton in 1909.

The artists and their work. In the period since 1884, at least 220 different artists worldwide have illustrated or decorated editions of Khayyam’s verses. As might be expected, the variety of types and technique of illustration is enormous, reflecting the general trends in artistic styles and forms in the period. Traditional Victorian engravings, rich art nouveau designs, colorful art deco paintings, line drawings, and more modern abstract approaches are all well represented, in addition to the traditions of Persian miniatures and many personal idiosyncrasies.

The approach adopted to illustrating the text has also varied. Some artists attempted, or were commissioned, to illustrate or “illuminate” a number of specific quatrains (robāʿiyāt). Others aimed to show the general subject matter or feel of the poem, without emphasizing the particular images in the verses. In most cases, artists, other than the Iranians, worked from some translation or other of Khayyam’s “original” text. It is not surprising that some of the interpretations presented are uncompromisingly Western in their imagery; these include the initial version by Elihu Vedder (PLATE II). Many artists adopted what can be called an “orientalist” view of their subjects, while there are some who have retained more of a sense of traditional Persian imagery (Martin and Mason, pp. 14-15).

There are few well-known general artists among the Rubaiyat illustrators. The main exception is Sir Frank Brangwyn (1867-1956) whose artistic output ranged from major paintings and murals through book illustrations and posters to the decoration of furniture and ceramics (Horner, p. 7). His two portfolios of illustrations for the Rubaiyat, published in the early years of the 20th century were based on small oil paintings, some of which are still extant. His colorful impressionistic style has an orientalist feel to it. Sir Edward Burne-Jones (1833-98), the Pre-Raphaelite artist, also created some illustrations for a one-off copy of the Rubaiyat, hand produced by William Morris in 1872 (Braesel, pp. 48-49).

PLATE V Illustration by Ḥojjat Šakibā (born 1949), published in a multilingual edition by Gooya House in Tehran in 1999. The illustration is not related to a specific quatrain but based on the actual tomb of Khayyam in Nishapur.

Most of the other artists illustrating the Rubaiyat were specialist book illustrators, but there are some notable absentees from the list such as Aubrey Beardsley and Arthur Rackham, both well known as illustrators during the early period of production of illustrated Rubaiyats. Probably the best known among those who did illustrate FitzGerald’s version of the Rubaiyat were the following four artists: Vedder, Dulac, James, and Pogany. Each of them created very distinctive sets of illustrations, which have been frequently reissued. For example, there was a new edition with Pogany’s paintings in 1999, and Dulac’s work was reissued in the United States in 1996 (Martin and Mason, pp. 19-21).

Elihu Vedder (1836-1923) was an established artist in the United States when he was commissioned to create the first portfolio of illustrations for FitzGerald’s Rubaiyat. He spent nearly a year in Rome working on ideas for his illustrations and the resulting images have something of a classical feel about them, although his drawings have been called “some of the earliest examples of Art Nouveau in America” (Soria, in Grove Art Online; PLATE II).

PLATE VI Illustration by Doris M. Palmer for quatrain 20 in FitzGerald’s first edition, published by Leopold B. Hill in 1921.

His work is in strong contrast to that of the next major illustrator, Gilbert James (fl. 1865-1941), whose illustrations for the Rubaiyat were first published separately, in black and white, mainly in the periodical The Sketch between 1896 and 1898 (PLATE III). Very little is known about this artist, who created three different sets of illustrations for issues of FitzGerald’s Rubaiyat, during the first decade of the 20th century. He was apparently born in Liverpool, allustrated some editions of fairy tales, and worked as well for a number of British magazines (Houfe, p. 190).

PLATE VII Illustration by Ḥosayn Behzād included in Hossein-Ali Nouri Esfandiary’s multilingual edition of the Rubaiyat (n.p. [Japan], 1970). The illustration is related to quatrain 16 in FitzGerald’s first edition, as well as to a quatrain in the original Persian and a French translation by Abolgassem Etessam-Zadeh.

The life and work of Edmund Dulac (1882-1953) is much better documented. Born and educated in France, he moved to London in 1906 and became a British citizen in 1912. His well-known illustrations for the Rubaiyat were first published in the anniversary year of 1909, following his earlier work on The Arabian Nights and Shakespeare’s The Tempest (Houfe, p. 123-24). Dulac’s art nouveau, orientalist paintings, epitomize, for many, the golden age of Rubaiyat illustration (PLATE IV). 1909 also saw the publication of the first set of illustrations created by the Hungarian artist Willy Pogany (1882-1955). He too had something of an orientalist approach to the imagery of the Rubaiyat, though his work is less elaborate in style. Pogany settled in the United States and produced illustrations for many other books, including two further and rather different portfolios for the Rubaiyat, published in 1930 and 1942 respectively (Greer, pp. 7-49).

Two other artists of particular note in terms of Rubaiyat illustration are Edmund Sullivan (1869-1933) and Gordon Ross (1873-1946), both of whom attempted the difficult task of illustrating every one of the 75 quatrains from FitzGerald’s first edition. In spite of their wide chronological separation (1913 and 1941 respectively), the two portfolios of black-and-white drawings are remarkably similar and have a somewhat cartoon-like character. Both sets of drawings were issued in popular, in some cases, paperback editions of the Rubaiyat .

A couple of artists of Indian origin, Mera K. Sett (dates unknown, published 1914) and Abanindro Nath Tagore (1871-1951), were early illustrators of the Rubaiyat; Sett’s black-and-white work is an example of the very symbolic way in which some artists have approached this poem. Key names in the art deco tradition who tackled the Rubaiyat in the 1920s were the British artists Anne Fish (1890-1964), Doris Palmer (d. 1931; PLATE VI), and Ronald Balfour (1896-1941). In the middle of the 20th century, there were also notable contributions from John Buckland-Wright (1897-1954), also British, who contributed delicate line drawings to a famous edition by the Golden Cockerel Press, and from Arthur Szyk (1894-1951), a well-known American artist of Polish origin (Martin and Mason, pp. 23-25). Original illustrations of continental European translations of Khayyam’s Rubaiyat have often been restrained in style, and based on line drawings. The work of P. Zenker (dates unknown, published 1924) in France, and Endre Szasz (1926-2003) in Hungary, has been frequently reissued in those countries. There are also illustrators of the Rubaiyat from as far afield as South Africa (Hope Beck; dates unknown, published 1950) and Uzbekistan (M. Karpuzas; dates unknown, published 1997).

One of the earliest of the modern illustrated editions of the Rubaiyat from Iran is Sadeq Hedayat’s (q.v.; Ṣādeq Hedāyat, 1903-51) selection of the Persian verses published in 1934. The illustrations in it, attributed to Darviš, are very traditional in style, whereas the work of artists such as Moḥammad and Akbar Tajwidi (fl. 1950s), and Ḥosayn Behzād (q.v.; 1894-1968), which appears in Iranian editions from the late 1950s onward, is more modern in feel, while retaining the format and some of the imagery of the Persian miniatures (PLATE VII). The illustrations of later Iranian artists, like A. Jamālipur (dates unknown, published 1996), Ḥojjat Šakibā (b. 1949; PLATE V), and Maḥmud Farščiān (b. 1930) are much more flamboyant and non-traditional in presentation (Martin and Mason, pp. 25-27).

The range of illustrations for the Rubaiyat is wide, and the work varies in quality as well as popularity. Some artists have produced images that seem to bear little relation to the text that contains them. Other images are not only beautiful in themselves, but also serve to interpret, to “illuminate,” the verses to which they refer. Seen from the standpoint of the early 21st century, it is the phenomenon of Rubaiyat illustration as well as the work of individual artists that is of interest. The continued publication of illustrated editions of this short work for more than 120 years is an amazing tribute to the ability of the writing of Khayyam and his translators to retain the interest of publishers and readers in the changing modern world.

• William H. Martin
• Sandra Mason

KHAYYAM, OMAR vi. FRENCH TRANSLATIONS OF THE RUBAIYAT

Omar Khayyam’s Arabic treatise on algebra, Maqāla fi’l-jabr wa’l-moqābala (see KHAYYAM xiv. AS MATHEMATICIAN), translated by Franz Woepcke (1826-64) as L’Algèbre d’Omar Alkhayyâmî (Paris, 1851), was the first of Khayyam’s works to appear in a French translation. He was therefore first known as a scientist in France. His quatrains were translated a short time later. Thereafter, he achieved fame as a great poet and one of the most frequently translated. From 1857 to 2010, there were no fewer than 119 translations of the quatrains (Coumans, p. 45), the majority based on the original Persian text but some relying on the famous English version by Edward FitzGerald (q.v.; 1809-1883).

The first translation of his poems was by Joseph Héliodore Garcin de Tassy (1794-1878), published in the Journal Asiatique (5th ser., 9, January 1857, pp. 548-54) as “Note sur les Rubâ’iyât de ‘Omar Khaïyâm.” Only ten quatrains were translated, each preceded by the Persian original based on the Bodleian Library manuscript (Ouseley 140) at Oxford, dating from 1460. A comprehensive volume of the French translation of the quatrains was published a decade later, in 1867. It was commissioned by Napoleon III and carried out by Jean-Baptiste Nicolas (1814-75). No other translation appeared until 1900, but since then, Khayyam has been regularly translated into French.

Jean-Baptiste Nicolas worked as an interpreter (dragoman) at the French legation in Persia, and was later appointed as the French consul in Rasht (see FRANCE iii. RELATIONS WITH PERSIA 1787-1918). He first published a translation of fifty quatrains in 1863, and a more comprehensive edition in 1867 with 464 quatrains, including the original Persian text (Figure 1). His translation had a great impact on the perception of Khayyam in France, instigating a debate about Khayyam’s personality and philosophical beliefs. In the introduction, Nicolas presents Khayyam’s work as “so essentially abstract in its philosophical thoughts, so strangely mystical in its figurative expressions (too often presented [by its interpreters] in the form of a repellant materialism),” (“si essentiellement abstrait dans ses pensées philosophiques, si étrangement mystique dans ses expressions figurées[trop souvent présentées sous des formes d’un matérialisme repoussant],” preface, p. I). According to Nicolas, the quatrains have to be understood through the prism of Sufism. He frequented Sufi circles in Iran and was acquainted with their spiritual notions. His symbolic interpretation of the poems was markedly different from FitzGerald’s point of view, as the latter contended that Khayyam’s poetry had to be read literally, from a materialistic point of view. Subsequently, the perception of Khayyam’s quatrains in England and in France differed widely. Although the French philosopher Ernest Renan (1823-92) praised Nicolas’ translation, he took a stance in favor of FitzGerald’s literal reading and paid homage to Khayyam’s subtle attitude, preserving “the free genius of Persia” (le libre génie de la Perse) in spite of the Arab Muslim conquest which had brought a new mode of thinking (pp. 56-57). The poet and dramatist Théophile Gautier (1811-72) wrote a long article in Le Moniteur universel of 8 December 1867 (reprinted in idem, L’orient: Tome second, 1877, pp. 57-72) commenting favorably on Nicolas’ translation: “the work is now as perfect as possible” (“l’ouvrage est maintenant aussi parfait que possible,” p. 58). He agreed with Nicolas that Khayyam was a Sufi (pp. 62-71), though he conceded that the references to wine in several poems should be taken at their face value and not systematically interpreted as symbols of divinity (p. 70).

Figure 1. Title page and end page of Nicolas’ edition and translation of the Robāʿiyāt of Omar Khayyam.

Translators of Khayyam came from different backgrounds and can be divided into three groups. Apart from the orientalists and scholars of Iranian studies including Nicolas himself, who formed the most significant group, there were poets and writers, some of whom lacked sufficient knowledge of Persian to translate directly and relied instead on FitzGerald’s English version. Emile Désiron’s version (1959) is the most celebrated. This raises the question of the accuracy of a translation based on another translation, particularly since the FitzGerald version is well known to be a belle infidèle as he is prone to take great liberties with the original. The third group consists of Iranians (scholars, writers, or poets) resident in France or in Iran. Most of these translations appeared from the 1980s onward, but a few were written earlier. For instance, Abolgassem Etessam-Zadeh (Abu’l-Qāsem Eʿteṣām-zāda)’s Les Rubaiyat d’Omar Khayyam was published in Tehran by the publishing house of Beroukhim (Beruḵim) in 1931 and in France by Maurice d’Hartoy in 1934. The translation won an award from the Académie française, which praised its high quality. Other translations were collaborations, such as Claude Anet and Mirza Muhammad’s Les 144 quatrains d’Omar Khayyam, published by Éditions de la Sirène in Paris in 1920. Although his full name does not appear in this edition, in an article in La Revue de Paris (60, 1 December 1920, p. 597, n. 1) on Omar Khayyam, Anet acknowledged the distinguished scholar Moḥammad Qazvini (q.v.; 1877-1949) as his erudite collaborator.

The choice of the manuscripts and the poems to translate poses its own problems. The number of quatrains varies greatly from one translation to another: Nicolas translated 464 quatrains, while Gilbert Lazard’s translation contains 101 poems (Cent un quatrains de libre pensée, Paris: Gallimard, 2002). This contrast reflects the multitude of available manuscripts and the large differences among them. In addition, the question of authenticity of all the poems adds further to the problem. It is not possible to establish a trend in the manuscripts used for all the French translations of Khayyam. For example, Nicolas does not mention which manuscript he had used. He was not a philologist like his contemporary Jules Mohl (1800-1876), who translated Ferdowsi’s Šāh-nāma and clearly cites the manuscripts he had used and the philological difficulties he had encountered. Some translators did not note whether or not they had used FitzGerald’s version. For instance, Charles Grolleau (1867-1940), the Belgian translator of G. K. Chesterton and Oscar Wilde, implied in the title of his translation that it had been translated from Persian, but he quoted FitzGerald’s translation in his introduction and admitted to using the literal English translation that Edward Heron-Allen (q.v.; 1861-1943) had edited in 1899, based on the Bodleian manuscript (Les quatrains d’Omar Kháyyám: traduits du persan sur le manuscrit conservé à la «Bodleian Library» d’Oxford, publiés avec une introduction et des notes, Paris: Carrington, 1902, p. 26).

The translators had to think about the various ways of translating the poetic language of Khayyam, and their choices differed considerably. Some opted to translate the quatrains into a more or less poetic prose. This was especially the case for the first translators. Garcin de Tassy (1857) and Nicolas (1867) chose plain prose, without even seeking assonance or suchlike stylistic effects. Charles Grolleau (1902) brought only a typographic appearance of verses to his work, and the lines were bereft of meter or rhyme. He used some alliteration and presented a text similar to a prose poem. Anet and Muhammad (1920), and Franz Toussaint (1924) translated the quatrains into poetic prose.

For those who chose a verse translation, further questions arose regarding the choice of meter and rhyme. Concerning the rhyme, it is often an “aaba” structure, i.e. an imitation of the Persian structure of rhymes in a quatrain (for example, de Marthold, 1910; Etessam-Zadeh, 1931). Some translators chose another scheme, such as Jean Lahor (Henri Casalis’ pseudonym) with rhymes in “aabb” and Jean Ruillier (2000), who employed, in addition to the Persian “aaba,” the rhymes “abba,” “aabb,” and “abab,” which correspond more closely to rhyme schemes in French. For the meter, almost all of the poetic translations present a twelve-syllable meter, which is the common French alexandrine. However, as Gilbert Lazard has pointed out, this French meter does not fit the task: the alexandrine is a solemn meter and fails to convey the lightness of the Persian quatrain (introduction pp. 15-16). He therefore opted for the French huitain (eight-line stanza) of heptasyllables, using assonances more than real rhymes. He referred to Pierre Seghers’ translation (1982), in which he chose poems in four octosyllables without rhymes. Some other translators proposed a translation in free verse, with or without rhymes, depending on the poem, and with no fixed meter (e.g. Fouladvand, 1960). This choice of free verse avoids the “unforgiving laws of our metric,” which forces translations to stray far from the original text, as Anet and Muhammad noted in the introduction to their 1920 translation (“lois impitoyables de notre métrique,” p. 3). Arthur Guy (1935) tried to imitate the Persian meter, using alternation between short and long syllables, explaining that, even if short and long vowels do not exist in the French language, there are long syllables by position (before two or more consonants).

These translations made it possible for the cultured readership in France, and particularly writers and poets, to become acquainted with Khayyam. Some authors explicitly noted the influence of Khayyam on their work, using his name in the title, or him as a character. Maurice Bouchor wrote a play Le songe de Khéyam in 1892. The anarchist poet Laurent Tailhade published the essay Omar Khayyam ou les poisons de l’intelligence in 1905, in which he praised Khayyam’s thoughts, but also blamed him for his overindulgence in wine. Employing Toussaint’s translation (1924), the musician Jean Cras composed a piece of chamber music Cinq Robayat for piano and baritone in 1924 (see KHAYYAM xiii. MUSICAL WORKS BASED ON THE RUBAIYAT). Khayyam is the main protagonist in the novel Samarcande, published in 1988 by Amin Maalouf, a Lebanese author writing in French.

The influence is obvious in other authors’ work. The already mentioned Jean Lahor published a translation of Khayyam in En Orient in 1907. The first part of the book consists of the re-editing of Les quatrains d’Al-Ghazali, which was first edited in 1896 and consists of Lahor’s poems written as if Ḡazāli (q.v.) had composed quatrains in the style of Khayyam. In another work, L’Illusion (1888), there are sometimes very similar verses to Khayyam, and the poet used the same imagery: beautiful bodies that have turned into clay, dialogues between the dead and the living who will soon be called to join them in death. In Sub tegmine fagi (1913), Jean Bernard freely adapted twenty-one quatrains. In CVII quatrains, published by Alexandre Arnoux in 1943, Khayyam’s influence is omnipresent, both in themes (death, the place of the human being in the universe, the fragility of existence) and in the form of a single quatrain itself, which does not exist in French poetic tradition. Les roses de la nuit, published by Jean Kobs in 1953, deals with the same themes.

Sometimes the influence of Khayyam is less perceptible, and it is difficult to ascertain proof of its existence. However, it is well known that Théophile Gautier, André Gide, and Marguerite Yourcenar were thoroughly acquainted with Khayyam’s quatrains. In the case of Théophile Gautier, the imprint of Khayyam on his writing was limited because he had discovered the poet when almost of all his works had been written. Concerning Gide, he indicated in the journal Parse (1921, pp. 33-34) that he was influenced by Persian literature, especially Saʿdi, Ferdowsi, Hafez (qq.v.) and Khayyam. Indeed, Hassan Honarmandi (pp. 6-26) shows that some passages in Les nourritures terrestres (1897) subtly reflect Khayyam. Marguerite Yourcenar explained in Carnets de notes de “Mémoires d’Hadrien” (1951) that, aside from the Roman emperor Hadrian, the only historical character she was tempted to write on was Khayyam.

As for the popularity of these numerous translations, Jos Coumans has shown in the statistical part of his study (p. 46), that the third most frequently published and re-edited translation (after the two English translations, FitzGerald’s and Whinfield’s) is Nicolas’. This is the case not only regarding other French translations, but compared with all Khayyam’s translations into foreign languages. Additionally, some of the translations into other languages, such as Italian, were based on this French edition. This said, it should be pointed out that some later translations do more justice to Khayyam than Nicolas’ pioneering rendering.

Agnès Lenepveu-Hotz

KHAYYAM, OMAR vii. GERMAN TRANSLATIONS OF THE RUBAIYAT

Omar Khayyam is generally known in the West as a mathematician, astronomer, astrologer, and poet. His literary fame in Europe, going back to the early 18th century, as attested in the work of Thomas Hyde (q.v.; 1636-1703), is due to his collection of poetry, the Rubaiyat (Pers. Robāʿiyāt, ‘quatrains’). The German discourse on Persian literature can be described as part of the zeitgeist of the 17th century, in that the interest of German poets in Persian culture and literature increased continuously. The 18th and 19th centuries crowned the reception of Persian literature in the German language (Maillard and Tafazoli, pp. 5-22). In the 19th century, one can begin to speak of a German reception and translation of Khayyam.

In the classification of Khayyam translations into German, a categorical distinction is helpful: One category consists of direct translations from the Persian original texts; the other of translations into German from mainly English, but also French, translations. The translations from English into German are based primarily on the work of Edward FitzGerald (q.v.; 1809-83). Within these categories, the German translations of the Rubaiyat in the 19th and 20th centuries are listed here. Given the variety of the interpretations of the poems, as well as debates over the origin and authenticity of the different editions on which the translations were based (Arberry, 1952, pp. 151-159; Rypka, 1959, pp -224; Rempis, 1933, pp. 15-20), the history of the reception of the Rubaiyat is a very complex matter. This problem is compounded by the dearth of biographical information for some relatively obscure translators.

The reception of the Rubaiyat in Germany began in the 19th century with the rise of scholarly interest in Persian literature (Tafazoli, pp. 322-539). The first translation, by Joseph von Hammer-Purgstall (q. v.; 1774-1856), included twenty-five quatrains (Hammer-Purgstall, pp. 80-82). The “Voltaire of Persian poetry,” as he calls Khayyam, seems strange and unique, above all because of the “irreligious content of his poems” (Hammer-Purgstall, p. 80).

The question of the religious aspect of Khayyam’s poems became a common feature in the historical reception of Khayyam’s work. Thus, the writer and critic Julius Hart (1859-1930), in his Khayyam translation, refers to Hammer-Purgstall’s comparison with Voltaire, but without naming the Orientalist (Hart, p. 41). Hart translated a selection of sixty quatrains in his Divan der persischen Poesie (1887). The writer and publicist Anton E. Wollheim (1810-84) published some of Khayyam’s poems in the second volume of his National-Literatur sämtlicher Völker des Orients, which is devoted exclusively to the literature of Persia. In the section on mysticism (Wollheim, pp. 204-52), after commenting briefly on Khayyam’s life and the translations known to him, Wollheim presented some quatrains in his own translation and that of Hammer-Purgstall (Wollheim, pp. 206-9). Friedrich Rückert (1788-1866), in his book Grammatik, Poetik und Rhetorik der Perser (1874), dealing with questions of language and poetics, and in the fortieth volume of his journal Jahrbüchern der Literatur (1827), published a few of Khayyam’s poems: two quatrains in the yearbooks (Rückert, 1827, pp. 208 f.) and two in his book on poetics in connection with the explanations of the robāʿiyāt as a genre with “freedom of diversity” (Rückert, 1874, pp. 65-67). These interpretative selections gradually made Khayyam a familiar name.

The art collector, translator, Hispanist, Orientalist, and poet Adolf Friedrich von Schack (1815-94) published an extensive and free translation of a random selection of 336 quatrains (1878). Schack complained in his epilogue that Khayyam’s reputation had so far reached only the Orientalists, although the Persian poet deserved to be represented as a universal figure. In order to acquaint German readers with Khayyam’s importance, he provided an account of the life and career of the poet (Schack, 1878, pp. 113-15), adorned with legends and anecdotes according to the state of knowledge of the time, and he pointed to what he regarded as a one-sided interpretation of Khayyam’s poems found among Sufi circles. He criticized the Sufi way of explicating Khayyam’s poems through a rigidly mystical interpretation of references to wine and the enjoyment of life. Schack pointed out that by no means all of Khayyam’s poems were mystical, theological, or derisory in tone; rather, they contained a deep and serious content (pp. 115-17). He promised a faithful reproduction of Khayyam’s poetic sense and spirit. His translation was based on a Calcutta edition from 1836, without offering any explicit information about it, as well as on Jean-Baptiste Nicolas’ 1867 French edition and translation (which had become an important source-text for the German reception of Khayyam). Among the 336 quatrains were 110 taken from FitzGerald’s English translation published in London in 1868, which Schack identified with an asterisk (pp. 12 f.). It is clear from this that the translator must have been unaware of a number of stanzas in the original Persian form.

The French translation by Nicolas, which included the Persian text, was also the basis for a translation by Maximilian Rudolph Schenck that appeared around 1897. Schenck’s intention was to provide a German translation that was as fluid in verse and measure as the original (Schenck, [1897], p. 6). After an introduction, which also contains biographical data on the poet (pp. 3-6), Schenck translated 468 quatrains and intended to bring the poet closer to the reader with notes and explanations in a footnote form. Which of these stanzas actually came from Khayyam remains unclear. Friedrich Rosen, himself a Khayyam translator, later judged Schenck’s translation to be “as close as possible to the French rendition … but with the loss of much of its peculiarity and poetic drive” (Rosen, 1909, p. 13; “möglichst an der französischen Wiedergabe…aber eben hierdurch ist viel von der Eigenart und dem dichterischen Schwung verloren gegangen”). In 1911, Schenck published a German text and score, based on FitzGerald and the choral work by Granville Bantock (1868-1946; see KHAYYAM xiii. MUSICAL WORKS BASED ON THE RUBAIYAT), intended to acquaint the public with the musical aspects of the Rubaiyat.

The last translation from the 19th century was that of the cultural historian, writer, and translator Friedrich von Bodenstedt (1819-92). Bodenstedt undertook for the first time a thematic categorization of the quatrains. He divided them into ten books with arbitrary designations: “Die Gottheit des Dichters” (Khayyam, 1881, pp. 1-20), “Der Gott des Koran und sein Prophet” (pp. 21-32), “Schein und Wesen” (pp. 33-42), “Die Grenzen der Erkenntnis” (pp. 43-56), “Schicksal und Freiheit” (pp. 57-76), “Lenz und Liebe” (pp. 77-88), “Der Dichter und seine Gegner” (pp. 89-112), “Welt und Leben” (pp. 113-54), “Der Dichter beim Pokale” (pp. 155-200), and “Verschiedene” (pp. 201-17). In the introduction (pp. ix-xxii), Bodenstedt mentions some collected historical notes about Khayyam’s life and fame. For the translation, he set himself the goal of receiving from the original texts accessible to him everything that, according to connoisseurs, is considered to be true to the poetic content (p. xxii). There is no clarification of how this is achieved. Although Bodenstedt mentions the translations by Hammer-Purgstall, Wollheim, and Schack, he is not precise about the underlying issues. Bodenstedt’s translation does not have a separate section of notes; these are presented as footnotes.

In the mid-19th century, a kind of Khayyam cult appeared in Europe, owing its genesis and development to the writer and translator Edward FitzGerald (q.v.), who came from a well-to-do Anglo-Irish family and achieved fame through his translation or adaptation of Khayyam’s quatrains (Arberry, 1959; Weber, 1959, pp. 35-111; Gray, pp. 1-14). At first, the reception of his translation was confined to the English reading public (Nordmeyer, 1969, pp. 13 ff.), but his work soon started to affect German translations, a trend that has continued (Nordmeyer, 1969, pp. 102-4), serving as the source text for many German translations of the Rubaiyat done in the 19th and 20th centuries, including the English-German edition in 1969 by Henry Waldemar Nordmeyer (1891-1981), an American scholar of German and professor at the University of Michigan from 1935 to 1960, and the translation by Martin Rometsch (1999).

George Dunning Gribble (1882-1956), a British playwright and author of The Master Works of Richard Wagner, made a German translation in 1907 from FitzGerald’s 1859 edition (PLATE I). Gribble’s partly rhymed translation included 101 quatrains (Gribble, 1907, pp. 9-109). Each page presented a quatrain with an attractive initial cap. The book contained an epilogue (Gribble, 1907, pp. 113-16) and annotations whose order followed the numbering of the quatrains (pp. 119-22). Another FitzGerald enthusiast was Arthur Altschul (1910, p. 50). Altschul, using FitzGerald’s fourth edition of 1897 and following the Persian rhyming order, compiled a hundred quatrains in a primer annotated with an afterword and endnotes (pp. 47-55). Another translation based on FitzGerald was by Walther Weibel (b. 1882) in 1911 under the pseudonym Hector G. Preconi. Preconi’s translation of 153 quatrains was based in part on FitzGerald’s second edition (1868) and partly on the work of other translators such as Nicolas and Arthur Christensen (q.v.; 1875-1945). His translation also contained brief information about Khayyam’s life and the Sufi interpretation, as well as some legends and anecdotes (Preconi, 1946, 5-16). FitzGerald’s second edition (1868) was also the source for a translation by the Austrian writer, journalist, and historian Paul Tausig (1881-1923) of 110 quatrains, published in 1917. Tausig added a bibliography on Khayyam and FitzGerald to his translation (Tausig, 1917, pp. 129-32). Also based on FitzGerald, Fritz Segers produced a German translation of 300 quatrains, with some additional remarks (Segers, 1923, pp. 5-8). In his comments and remarks, Seger drew on numerous verses from Persian to introduce readers to the spirit of Persian poetry as comprehensibly as possible. Around 1926, there was a translation of FitzGerald’s first edition by W. D. Kulenkampff. Paul Kinsky’s 1927 translation of 75 quatrains was based on various sources, mainly Persian, with no introduction and no afterword. One can only learn from the comments which quatrains came from FitzGerald’s translation. Philologically, this translation is of no great scholarly importance. In 1930, another German translation of FitzGerald by Richard S. Bak was published.

PLATE I Title page of George Dunning Gribble’s German translation of Edward FitzGerald’s Rubáiyát.

One of the most important in the series of translations based on FitzGerald was that by Christian Hernnhold Rempis (1901-72), Die Vierzeiler ʿOmar Chajjāms (1933). This translation of 101 quatrains followed the first edition of FitzGerald (1859) and included some treatment of philological questions (Rempis, 1933, pp. 20-28) along with a brief evaluation of the German translations published until 1933 in a comparison with English translations (pp. 10 f.). In 1935, Rempis published a translation, which was to supercede the 1933 edition in both literary and philological terms. The selection of quatrains was made from the fifty oldest and best-preserved manuscripts, dating from the 13th to the 16th century (pp. 48 f.). From these sources, Rempis carefully compiled a Persian text that formed the basis for his translation. There was a selection of 255 quatrains, which, as in Bodenstedt, were assigned to different subject areas (Rempis, 1935, p. vi). Rempis intended to bring the reader “as close as possible to the original views of the poet” (p. vi). He divided his translation into four books: “Leben und Gedankenwelt” (pp. 1-54), a lyrical translation (pp. 55-104), a literal prose translation of quatrains (pp. 105-54), and explanatory notes and source references (pp. 155-200).

A more basic selection, without introduction and epilogue, was presented by the poet and Germanic scholar Ernst Bertram (1884-1957). His manuscript, badly damaged by the destruction of the Second World War, appeared in 1944 as the 87th volume of the Insel-Bücherei. Bertram’s translation, which was published three times, contains poems by other poets as well as a selection of twenty-three quatrains (Bertram, 1951, pp. 16-22).

In 1963, the freelance writer and political journalist Max Barth (1896-1970) published a translation of 207 quatrains based on all five editions of FitzGerald (Barth, 1963, pp. 23-76). Barth’s translation was a literal one that retained the English rhyme order (p. 9). His work included an overview of Khayyam’s life and influence (pp. 9-17), his European reception (pp. 18 f.), and an interpretation of Khayyam’s freethinking (pp. 20-22), as well as a numbered and an alphabetical list of quatrains (pp. 96-107).

In addition to the FitzGerald impulse for the reception of Khayyam, there had been a strong interest in Khayyam’s original Persian poems in Germany since 1900. By the mid-20th century, there was the gradual emergence of a German school of Khayyam research that sought not to see Khayyam’s life, spirit, and poetry through “FitzGerald’s glasses” (Rempis, 1935, p. v), but rather to view them from a philological and aesthetic perspective (Rempis, 1937). In the first volume of his Geschichte der Weltliteratur, the literary historian Alexander Baumgartner (1841-1910) dealt with the “Literatur der Perser.” In the section on Sufism (Baumgartner, 1901, pp. 561-81), he gave a brief insight into Khayyam’s world view and quoted three quatrains (pp. 569 f.) from Wollheim’s translation (Wollheim, 1873, p. 209).

One of the most famous German Khayyam translations was that of the traveler to Persia and Orientalist Friedrich Rosen (1856-1935) in 1909. According to its preface (Rosen, 1909, pp. 9-18), the quatrains were translated in a thematically ordered sequence. Unlike Bodenstedt, Rosen distinguished only four themes: “Vergänglichkeit” (pp. 21-36), “Welträtsel” (pp. 37-47), “Lehre” (pp. 49-64), and “Wein und Liebe” (pp. 65-75). It concluded with two quatrains in the epilogue “Schlussworte” (pp. 77-79). Explanations of the quatrains were inserted between the preface and the translation (p. 19). The book ended with a lengthy section on Khayyam’s life and influence (pp. 81-147), notes (pp. 149-52), and some remarks on recent writings concerning Khayyam from 1897 to the end of 1906 (pp. 153f.). In keeping with the character of the translation, which is noteworthy for its elegance and readibility, these were not directed at specialists but at educated readers (p. 18), and Rosen refrained from calling his work a philological or critical edition.

In addition to translations of the poems, some paraphrases and interpretative translations distinguished themselves in the German reception of Khayyam’s poetry. This category includes the free adaptation of 184 quatrains by Richard Hamel (1853-1924) under the pseudonym Omar Khayyam in 1912. This small collection concentrated on the Persian rhymes and its aim was “not in the most enjoyable enjoyment of the moment, but in the nascent, the well-meaning, world-consciousness of the 20th century that is heroic in the sense indicated” (Hamel, 1912, pp. xlvi). The publisher of this collection, F. Braun, often referred to the Khayyam translators Schack, Bodenstedt, FitzGerald, Rosen, and Gribble in the introduction and cited those in connection with his own explanations of Khayyam. Braun distinguished between an “old” Omar, meaning the Persian, standing under the sign of pessimism, and a “new” Omar (Hamel), whose joy in life he valued (Hamel, 1912, pp. xxxii-xxxiii, p. xlv).

Another free adaptation, without consideration of the rhyme order, was by Hans Bethge (1876-1946). Bethge was a German poet who made a name for himself through imitations of Oriental poets, including Saʿdi and Hafez (q.v.). His versions of Khayyam’s quatrains, first published in 1921, were based mainly on the translation by Nicolas but also referred to FitzGerald, Bodenstedt, and Schack (Bethge, 2003, 135 f.). Six years later, Walter von der Porten introduced German readers to his Khayyam translation based on the famous Bodleian manuscript (Ouseley 140; see KHAYYAM ii). Porten attempted an “almost literal translation” (Porten, 1927, p. 7) of this manuscript, which had been copied in Ṣafar 865/December 1460 (pp. 8-10). Porten translated 158 quatrains and another 23 in an appendix (pp. 69-78), the authenticity of which had been repeatedly affirmed (pp. 8f.). The book concluded with a short list of annotations (pp. 79-84). Porten’s translation was published by Khosro Naghed in 1992 in a Persian-German version entitled Wie Wasser im Strom, wieWüstenwind. In 1930, there was a German translation of an anthology of Persian poems by the writer and poet Alfred Henschke (1890-1928) under the alias Klabund. In the section “Persische Lyrik” (Klabund, 1930, pp. 283-320), he included a selection entitled “Das Sinngedicht des persischen Zeltmachers 96 Robāʿyāt” (pp. 302-19). There was an illustrated Khayyam translation by Oscar Klausner in 1933, which is extant today in only fifty copies.

Dieter J. Bellmann (1934-97) summarizes Khayyam’s philosophy of life in two categories: “materialism” and “anti-religiosity” (Bellmann, [1958], p. 12). Poems are cited in accordance with these two categories. Bellmann based his translation on three texts: the Bodleian Ouseley 140 manuscript of the Rubaiyat mentioned above, edited by Edward Heron-Allen (q.v.) in 1898; the 1927 edition by Arthur Christensen; and the Persian text edited in 1941 by Moḥammad-ʿAli Foruḡi (q.v.). In 1965, Bozorg Alavi (1904-97) edited a partial translation of the Rubaiyat by Martin Remané (1962) with comments on the poems and an epilogue by Jan Rypka (1886-1968) summarizing the problem of understanding Khayyam (Remané, 1962, pp. 87-113). One does not know, according to Rypka, whether one should interpret Khayyam’s poems as an “expression of a realism, a pessimism, an agnosticism or even atheism” (p. 89). The translator Remané provided no information about the sources for his translation. Manuel Sommer used the translations by Christensen, Rempis, and Rosen for a German edition (1974) of 234 of the quatrains. Sommer’s translation includes an overview of contemporary and cultural currents in Iran of the 11th century and Khayyam’s influence (Sommer, 1974, pp. 13-37), and it concludes with a short bibliography and a word and subject index (pp. 128-48).

Cyrus Atabay (1929-96) translated 121 of the quatrains without taking into account the Persian rhyme scheme and without any additional commentary (1984). Atabay’s sonorous translation was accompanied by Josua Reichert’s Persian calligraphy. This translation was reprinted in another edition in 1998. It contains in the epilogue by the publisher brief interpretations of Khayyam (Atabay, 1998, pp. 185-92). With a renunciation of the final rhymes, the rhythm, and the meters of the Persian, Franz Gschwandtner translated 151 quatrains in an illustrated edition in 1986. Very different from previous translations of Khayyam’s Rubaiyat is the one composed in English and German by Ulrich Helmke. Although Helmke’s translation is in the spirit of FitzGerald (Helmke, 1987, p. 9), it gathers selected verses from other translators such as Bodenstedt, Rosen, Rempis, and Preconi. Helmke’s approach is unusual and interesting at the same time: He tries to contextualize the poems of Khayyam. This is done on two levels, namely through Khayyam’s biography and by his poetic motifs (wine, tavern, tulip, cheeky maiden, potter, pitcher, dust, and love).

A Persian-German edition of Khayyam’s Rubaiyat, with miniatures, was published in 1997 by Purandocht Pirayech. Pirayech’s translation is one of the more modern translations of Khayyam’s poetry into German. It contains 168 quatrains. Pirayech seeks not a literal, but a spiritual transmission of Khayyam’s world of ideas with numerous explanations in the afterword (Pirayech, 1997, pp. 7-26). She distinguishes verses attributed to Khayyam, marking them with an asterisk, from verses that are documented. Another bilinguial translation is by Jalal Rostami Gooran and Ludwig Verbeek (2006). The translators strive to remain faithful to the rhythm of the poems, but they do not attempt to rhyme in every verse (Gooran and Verbeek, 2006, p. xii). The translation of 153 quatrains, accompanied by Masoud Sadedin’s drawings, was done on the basis of Persian models (p. xiii), with some German translations being used comparatively (pp. 172 f.). A readable edition of the Rubaiyat in German translation was published by Hort Rinner (2007). The translator, who does not necessarily approach Khayyam through questions of poetics, strangely sees in his poetry no literary and philosophically planned work, but a work that the poet has made “whimsically” (Rinner, 2007, p. 6).

Khayyam’s Rubaiyat offers a translator the possibility to interpret and to characterize Khayyam’s life, intellectual world, and poetry by exploiting poetic words for the justification of his/her own views. Otto Rauth used the English translation by the founder of the Mazdaznan movement, Otoman Zar-Adusht Ha’nish (1844-1936), for his German translation (1934). Another such German translation is that of 171 quatrains (1950) by Jehuda Louis Weinberg (ca. 1876-1960). In these translations, for instance, translators go beyond the frame and the context of the original text. It seems as if they try to turn the translation into a kind of transcendental world-view that deploys the words of the poems as instruments for their ideological purposes. In the early 1980s, an Arabic poet named Mohamed Abou-Zaid published a German translation of the Rubaiyat. In the preface to his translation, he referred to Khayyam as a poet from Arab culture, whose poetry includes mystical traits and religious doctrine (Khayyam, [1980], p.3). At least with regard to Khayyam’s cultural affiliation, he contradicts himself in naming Nishapur as Khayyam’s birthplace (p. 5). In a lengthy introduction (pp. 5-19), the translator speaks without any evidence of Khayyam’s life and work as a poet and interprets some of his poetry in a free contextual reference. In total, Abou-Zaid translated 109 quatrains, without mentioning his sources or the editions used. In addition to this deficiency, a series of erroneous comments and interpretations make this translation useless for scholarly purposes.

Khayyam—a mystic? The international reception of Khayyam is often characterized by a purely mystical interpretation of the poet. This interpretation is in some cases emphasized by editors of the older editions of Khayyam translations, such as Khosro Naghed, who uses the term mystic in the title of the 1927 Walter von der Porten edition he published (1992). The characteristic of the poetic word is its ambiguity. But it is true that Khayyam, like Hafez, is often portrayed in rigidly Islamic and Sufi interpretations as a pure mystic. Such narrow interpretations are found less frequently in scholarly than in popular and ideological circles. In the prefaces and introductions of the latter, readers tend to be patronized and offered oversimplified notions so as to bend them to a particular understanding of the poet and his work. In addition, there is usually a kind of confrontation with FitzGerald’s understanding of Khayyam. One of these translations is that by the Indian yogi, philosopher, and writer Paramahansa Yogananda, whose real name is Mukunda Lal Ghosh (1893-1952). He translated Khayyam into American English (1994), which was then translated into German in 1995. Yogananda’s translation is based on the first edition of FitzGerald (1859). In the introduction to his translation, Yogananda attaches spiritual power to the poetry of a poet whose philosophy was not fully understood in Persia. Explanations that apparently stem from such forces form the basis for paraphrases, interpretations, and explanations of words in the translated quatrains. Khayyam’s poems in the translation of Sayed Omar Ali-Shah (1922-2005; see KHAYYAM iv) are also treated as if they were Sufi documents. Ali-Shah’s English translation of Khayyam’s Rubaiyat in 1993 was translated into German in 1995. Omar Ali-Shah’s translation is a purely Sufi interpretation of Khayyam’s Rubaiyat, as well as a reckoning with FitzGerald’s free translation, which in the eyes of Omar Ali-Shah was a mistranslation. This German translation of 111 quatrains is still popular in Sufi circles.

Hamid Tafazoli

KHAYYAM, OMAR viii. ITALIAN TRANSLATIONS

No other Persian poet has enjoyed such enduring fame in Italy as has Omar Khayyam (ʿOmar Ḵayyām). Italian libraries hold only four manuscripts (15th-17th centuries) that together contain eighteen quatrains ascribed to Khayyam, one of which is thought to be unattested elsewhere (Piemontese, 1989, pp. 134-35, 292, 303-4, 339-40; Bertucci). The first English and German versions of some of Khayyam’s quatrains had already appeared at the beginning of the 19th century, but the encounter with and reception of, Khayyam’s poetic work in Italy, as in the rest of Europe, was the result of the translation and rewriting of the English poet Edward FitzGerald (q.v.; d. 1883) in the years 1859-79. Thus, in Italy, the more scholarly approach to Khayyam’s work by a few dedicated Iranists at a fitful pace over many decades has had to contend with the overbearing heritage of the so-called FitzOmar, which has been ardently loved, discussed, translated, and recast into Italian many times. This dual process of reception has taken place in two intense phases: the two decades leading up to World War I, and the two decades immediately following World War II.

In the first key period, around the year 1890, a few samples of Khayyam’s quatrains were translated directly from Persian in the context of academic or occasional publications by scholars: Italo Pizzi, professor of Persian at Turin University (5 quatrains in 1887, and 60 quatrains in his Storia della poesia persiana, 1894, from Nicolas’s edition; these were inserted, together with some historical notes by Pizzi from the same book, in Dole’s revised edition of his comparative work, II, pp. 536-56), and Pizzi’s disciple and brother-in-law Vittorio Rugarli, a close friend of the Italian poet Giosuè Carducci, who also drew on Nicolas’s edition to produce two small wedding booklets in 1895, containing 12 and 10 quatrains respectively. However, the call to Khayyam’s writings was only heeded decisively within the intellectual networks of the Italian bourgeoisie upon the discovery and popularization of FitzGerald’s work. This process took place mainly in the literary milieu of Italian Decadentism tied to the figure of the writer and poet, and, later, politician, Gabriele D’Annunzio, and eventually came on the trail of the aestheticist-oriented reading promoted in Victorian England by Charles Swinburne and the Pre-Raphaelites. Indeed, it was D’Annunzio’s close friend and collaborator, Adolfo De Bosis, who published the first essay on FitzGerald’s third edition of Khayyam with illustrations by Elihu Vedder in the journal Il Convito (June 1895); it included the Italian translation of fifteen quatrains from FitzGerald’s English version.

The first complete translations of FitzGerald’s work were produced later. One was by Diego Angeli, a collaborator on Il Convito, in two editions around 1910 (PLATE I), including the 101 quatrains of FitzGerald’s third edition; these were severely criticized for the translator’s alleged poor knowledge of English. Another was by Fulvia Faruffini in 1914, containing the 75 quatrains of FitzGerald’s first edition. Within the context of general fascination and appreciation shown by most intellectuals toward the newly discovered universal “poet philosopher,” only a few critical voices emerged, including the eminent literary critic Emilio Cecchi, who considered the Khayyam-FitzGerald enterprise as an exotic mis-en-scène submerged in aestheticism and mysticism, and thus the fruit of dangerous cultural decay. The third notable translation from FitzGerald was the valuable work of Mario Chini, first published in the journal Nuova rassegna di letterature moderne in 1907, then in a successful volume in 1916; it still drew on FitzGerald’s third edition, but it was the first version to provide an accurate literary and historical profile of both the original Persian poet and the English translator.

In between these two channels of reception (the first one little developed by then), one can discern a few other attempts at translating and presenting Khayyam’s work to an Italian audience, using sources other than the original Persian texts and FitzGerald’s editions. Edward Heron-Allen’s (q.v.) English revision of the Bodleian manuscript that had been the basis of FitzGerald’s work was the source of an indirect translation by Vittorio Gottardi (1903: 155 quatrains through Grolleau’s French version; this happens to be the first collection of Khayyam’s poems in a volume); it also provided the text for Tommaso Cannizzaro’s 1916 translation (158 quatrains). It is not clear which language provided the basis for the widely disseminated version by Massimo Spiritini (two editions, one in 1907 under the pen name Massimo da Zevio, the other in 1924, enlarged from 77 to 84 quatrains, in a larger collection of Persian lyrics; in 1939, he republished a selection of slightly revised quatrains in an anthology of world poetry). A poet himself and a translator from various European languages, Spiritini claimed that his translation was the result of the collaboration with “a friend from Hamadan” (Spiritini, 1924, p. 52); we nevertheless can rule out the possibility that he himself knew some Persian.

PLATE I Frontispiece in Diego Angeli, Edward Fitzgerald: Quartine di Omar Khayyám, Versione di Diego Angeli, Bergamo, n.d., reproducing one of Elihu Vedder’s illustrations for the journal Il Convito. It depicts FitzGerald’s quatrain 76 (in the 2nd edition): “The Moving Finger writes….”

At this early stage of Italian acquaintance with Khayyam, the Persian poet generally was presented and welcomed as a hero of human free thought, a sceptical enemy of hypocrisy and of religious and social ties, a sort of genius of atheism and a martyr of philosophy. In the context of Italy’s deeply classical culture, frequent comparisons were made with Latin poets such as Lucretius, on account of certain Epicurean traits attributed to Khayyam’s philosophy, and Horace, by way of his Anacreontic and apparently hedonistic lyric poetry. Some commentators pushed the matter even further, in creating audacious bonds with modern authors like François Rabelais, Voltaire, Giacomo Leopardi, and many others (e.g., Spiritini, 1907; Chini, p. XVII; De Lorenzo, pp. 116-31)

This phase also witnessed the production of a few notable poetic tributes to the figure of Khayyam—a testament to his emerging, cross-culture literary status. The first one (1890) was the fruit of Turin’s scholarly environment, composed by the poet and professor of Italian literature, Arturo Graf, who was a colleague of Italo Pizzi (although, actually, this is a paraphrase of FitzGerald’s quatrain XXIX, inserted in a wider poem). The other tributes are rooted in the milieus of Decadentism and its literary offshoots: They are the four-section poem “L’immortalità” (Immortality) by Giovanni Pascoli (1896), who learned about the Persian poet at the time of his collaboration with the journal Il Convito, and the poem “A Omar Khayyàm” (To Omar Khayyam), by Vincenzo Cardarelli, composed in 1914, but published in 1942. References to Khayyam henceforth have been quite frequent in texts by Italian writers across many genres. Moreover, following the example of such English composers as Liza Lehmann and others, some compositions for piano and voice based on Khayyam’s verses also appeared in Italy, like the ones by Elsa Olivieri Sangiacomo (1919), wife of the famous composer Ottorino Respighi; Francesco Santoliquido (1920); Giacomo Benvenuti (1929); and Guido Guerrini (1948). A later one by Azio Corghi (1966), based on four quatrains, was written for a male chorus.

With the exception of a couple of little-circulated collections of quatrains taken from FitzGerald in the early 1930s, the interwar period represents a break in the circulation of Khayyam’s thought and poetry on the Italian peninsula, possibly because it was felt that it would contrast with the dominant pragmatic ideologies of the Fascist era. Then, atfter the end of World War II, with the establishment of new schools of Iranian studies at universities in Italy (see ITALY xiv), further progress was made with the first large collection of quatrains translated directly from Persian. In 1944, Francesco Gabrieli (q.v.), who was mainly an Arabist but also worked on Persian sources, published a volume with historical introduction and annotations. This work, subsequently re-published in 1973 with various reprints, was based on Bertalan Csillik’s editions of the Parisian manuscripts, through a comparison with the one at the Bodleian Library, and included 307 quatrains. The careful prose rendering of the poems by Gabrieli was praised by his friend and colleague, Alessandro Bausani (q.v.), who nevertheless did not refrain from adding a second scholarly edition of Khayyam’s quatrains in Italian, published in 1956, with numerous reprints. Bausani’s selection of 282 poems is based on Moḥammad-ʿAli Foruḡi’s and Arthur Arberry’s (qq.v.) editions, and it constitutes an excellent example of faithful translation and poetic expression in elegant and rhythmic Italian. Among the meaningful contributions of Bausani’s critical “Introduction” is the attempt to reallocate Khayyam’s poetry in its proper historical world and ideological context, overtaking the simplistic interpretive dichotomy between the mystical Khayyam and the atheistic hedonist. According to Bausani (p. xxi), “In the Islamic concept of the world dominated by casualism and occasionalism, the three ways of faith, despair, and irony, are not too far one from another, and in Khayyam the emphasis on each one of these three could well depend on the moment’s mood.”

After the war, a second wave of collections of Khayyam’s quatrains appeared, again by way of a number of intermediary languages. These collections often were mainly based on FitzGerald, but sometimes they were taken from other versions as well, such as the very popular French one by Franz Toussaint, issued in 1924 (apparently used for his translation by the classical philologist and poet Alessandro Zazzaretta, in 1948, republished in 1966). A well-known publication was promoted by Pierre Pascal, at that time chancellor of the Iranian embassy to the Holy See, who translated from Persian 453 quatrains, mainly from the controversial Cambridge University Library and Chester Beatty Library (q.v.) manuscripts, into both French (Rome, 1958) and Italian (Turin, 1960, with the collaboration of G. Degli Alberti).

Without counting the selections of quatrains appearing in various anthologies of Persian or world literature, whose sources are generally hardly traceable, a number of other collections of Khayyam’s poems have been published, with evidence of a renewed vogue during the 1990s and 2000s. These include an interesting attempt at a comparative translation and rewriting from French, English, and Italian (by C. Gasparini, 1991), as well as two versions of Paramahansa Yogananda’s spiritual commentary to the Quatrains (1995), a quite successful recasting from various sources (by H. Haidar, 1997), and an Italian version (1999) of the much-debated translation produced in 1967 by Robert Graves and Omar Ali-Shah. An approximate estimate identifies up to twenty-eight different volumes in Italian (to 2013) devoted to Khayyam’s poetry that include at least seventy-five quatrains (that is, the number in FitzGerald’s first edition), of which about half depend directly on one of FitzGerald’s English versions. This estimation remains approximate because of the difficulty in ascertaining the existence of collections of a commercial nature and scope, often with misleading titles. In recent years some multilingual editions have been published in Iran, which include Italian translations by anonymous authors (Coumans, pp. 196-204). However, despite the lack of continuity in dedicated Khayyamian studies within Italian scholarship to date, Gabrieli’s and Bausani’s versions of the 1940s and 1950s remain the most solid and reliable reference works for reading Khayyam in the Italian language.

Mario Casari

KHAYYAM, OMAR ix. RUSSIAN TRANSLATIONS OF THE RUBAIYAT

There are a great number of Russian translations of the Rubaiyat (Robāʾiyāt) of Omar Khayyam (ʿOmar Ḵayyām). By the early 21st century, more than seventy were made. They vary a great deal in both form and content, depending on the choice of the Persian robāʿis selected for translation and the degree of liberty that the translators had allowed themselves in departing from the original. As a result, we are offered a range of strikingly diverse collections of poems under the title of Rubaiyat, conveying conflicting interpretations of Khayyam’s poetry and worldview.

In contrast to such great Persian poets as Hafez, Saʿdi, and Ferdowsi (qq.v.), who had all been already introduced to Russian readers, there is no evidence of any interest in Khayyam in Russia before the last decades of the 19th century, when the Rubaiyat had already achieved widespread popularity in Europe, largely due to Edward FitzGerald’s (q.v.) famous translation.

The very first rendition of Khayyam’s quatrains into Russian was not, however, derived from FitzGerald’s translation, but translated directly from Persian by the poet and traveler Evgeniĭ Belozerskiĭ (1853-1898). He had studied Persian at the Lazarev Institute of Oriental Languages in Moscow and is known to have made prose translations of 153 robāʿis circa 1886 (Abdullaeva, Chalisova, and Melville, p. 163). Since these never appeared in print, the publication of 16 robāʿis in verse translation by the poet Vasiliĭ Velichko (1860-1903) in Vestnik Evropy (Velichko, 1891) marks the starting point of Khayyam’s multi-stage reception in Russia.

The small number of translations that appeared up to 1916 drew their inspiration, at least partially, from the article by Vasiliĭ Zhukovskiĭ (q.v) on the wandering quatrains (Zhukovskiĭ, 1897). They were composed by different authors but shared a common and lively disregard for the specific formal features and conventions of the original verses in Persian. They appeared sporadically as contributions to journals (Porfirov, 1894; Umanets, 1901; Lebedinskiĭ, 1901; Bal’mont, 1910; Umov, 1911), as well as in poetry collections and anthologies (Velichko, 1894; Velichko, 1903; Danilevskiĭ-Aleksandrov, 1910; Mazurkevich, 1913; Umov, 1916). Almost all were free paraphrases, with a general tendency to elaboration and embellishment of the original: in Velichko’s translation, fifty-two poems in all, only five quatrains could be found, the rest of the so-called “quatrains” contain more lines and vary between five and sixteen lines (Vorozheĭkina and Shakhverdov, 1986, pp. 43-47). The only exception here is Konstantin Bal’mont (1867-1942), one of the major poets of the Silver Age of Russian poetry (circa 1890-1920). He kept to the form of the quatrain in all his eleven poems and was the first Russian translator to follow the formal structure of the robāʿi.

The only edition of the Rubaiyat as a single volume at this stage appeared as a literary enigma without any reference to Khayyam. In 1901, a poet and music critic Konstantin Mazurin (1866-1927) published his Stanzas of Niruzam (Strofy Niruzama; where “Niruzam” is an anadrome for the poet’s name) under the pseudonym K. Gerra (Mazurin, 1901). In the introduction, the author claimed that the source of his translation was an old manuscript of an anonymous 10th-century Persian poet from Khorasan. That Oriental stylization, supposedly penned by Mazurin himself, turned out to be a free paraphrase of Khayyam; 110 robāʿis were rendered, emulated, or used in part in 168 stanzas of “Niruzam” (Vorozheĭkina and Shakhverdov, 1982, pp. 45-46).

In subsequent decades (1920s and 30s), new verse translations appeared, some based on FitzGerald’s English version, others directly from Persian. In 1922, Osip Rumer (1883-1954), a polyglot linguist and a versatile poet-translator, published in Moscow a complete translation of FitzGerald’s third edition of Rubaiyyat (Rumer, 1922), which had been published in 1872 (101 poems). The trend to follow FitzGerald’s translation continued with Omar Khaĭyam: Chetverostishiya (Omar Khayyam: Quatrains), published in Paris in 1928. Its author was Ivan Tkhorzhevskiĭ (1878-1951), a statesman, poet, and translator of modern French poetry who had emigrated to France after the Russian Revolution of October 1917 (Tkhorzhevskiĭ, 1928).

Tkhorzhevskiĭ’s Khayyam collection (194 poems) included selected translations from FitzGerald’s oeuvre and improvisations on its main themes, together with alternative renditions from Louis Jean Baptiste Nicolas’ French versions, and poems authored by the translator himself. Tkhorzhevskiĭ did not know Persian but benefitted from the helpful guidance of Professor Vladimir Minorsky (q.v.) in order to enter the spirit of the original. He took great liberties with the imagery but followed closely the formal conventions of the robāʿi genre. He applied the form of the quatrain, already familiar to Russian readers in the 19th century thanks to the rich tradition of epigrams in verse, but changed the conventional rhyme schemes (abab, abba) to the specific robāʿi rhyming pattern (aaba) and observed a compact rhythmic pattern of iambic pentameter throughout his poems. His masterful imitations shaped the first recognizable image of a sententious, ingenious, and impious Khayyam for Russian readers, although Tkhorzhevskiĭ’s popularity came much later, when his translations finally found their way to Soviet Russia (Rumer, N[ekora], and Tkhorzhevskiĭ, 1954; Rumer and Tkhorzhevskiĭ, 1955). The future success of his imitations of Khayyam was predicted by Vladimir Nabokov (1899-1977; Nabokov, 1928). At the time, the young Nabokov criticized Tkhorzhevskiĭ for the “muddle of sources,” but concluded his review of the book with praise for the translator’s “good poetry,” adding: “I suspect the kind Omar Khayyam, though he possibly did not write it at all, nevertheless would have been flattered and would have rejoiced” (Nabokov, 2008, p. 659).

In the 1930s, Khayyam enjoyed scholarly attention in the Soviet Union, prompting new translations based on the original texts. Leonid Nekora (1886-1935) presented 144 poems (from the famous Bodleian Library manuscript of 1460) in an accurate and highly proficient verse translation, in the volume Vostok II: Literatura Irana X-XV vv. (The East II: Literature of Iran, X-XV cc.), dedicated to the Third International Congress of Persian Art and Archaeology (Nekora, 1935). Sergeĭ Kashevarov published his collection of 122 quatrains (Kashevarov, 1935), based on the texts of Khayyam in the editions by Nicolas (1867) and Arthur Christensen (q.v.; 1927); however, his rhymed but pedestrian literal renderings are only of historical interest now. Some of Nekora’s texts, along with several poems offered by Vladimir Tardov (1879-1938) and Konstantin I. Chaĭkin (1889-1938), were included in a separate Khayyam edition (1935) in which the majority of the translations (43) were by Osip Rumer. For this volume, Rumer translated directly from the original, having by then learnt Persian, prompted and inspired by his previous work, his translation of FitzGerald’s Rubaiyat.

In 1938, Rumer published Khayyam’s Chetverostishiya (Quatrains; Rumer, 1938), his own translation of 300 robāʿis, apparently chosen from Nicolas’ edition, since some of the inaccuracies of that translation are found in it as well (Vorozheĭkina and Shakhverdov, 1986, p. 51). Having followed Tkhorzhevskiĭ’s metrical scheme, he rendered all the quatrains in iambic pentameter with aaba or aaaa rhymes and also managed to combine fidelity to the original with a display of aptly chosen Russian poetic idiom, thereby introducing the authentic motifs and images of Khayyam to Russian poetry lovers. Rumer’s collection rightfully counts as the crowning achievement of the first half-century of Khayyam translation into Russian.

Soon after the end of the Second World War, a large-scale project of literary translation aimed at promoting mutual exchange between Soviet peoples started to gain momentum. A strong stimulus to translate Persian poetry came from its official ideological status as the “cultural heritage of the Tajik people.” Thus the existing translations of Khayyam were republished, and some new versions by Il’ĭa Sel’vinskiĭ (1899-1968) and Anatoliĭ Starostin (1919-1980), both active translators of the “USSR peoples’ literature” from several languages, appeared in numerous editions of Tajik poetry (see Sel’vinskiĭ, 1949; Rumer and Sel’vinskiĭ, 1951; and Rumer, Thorzhevskiĭ, and Starostin, 1957).

An influential attempt to understand Khayyam was made in 1959 by the renowned Soviet Iranologists M.-N. Osmanov (1924-2015) and R. M. Aliev (1929-94). They presented an accurate prose translation of 293 poems from the Cambridge University Library manuscript supposedly written in 1207 CE (Aliev, Osmanov, and Bertel’s, 1959). The manuscript itself later turned out to be a modern forgery, although the poems it contained were mostly “authentic,” i.e. culled from already existing collections. Osmanov and Aliev’s mode of understanding of the poems and their choice of words in their translation had a noticeable impact on numerous poets with little or no knowledge of Persian, who were nevertheless eager to compose their own translations of Khayyam (Pen’kovskiĭ, 1959; Spendiarova, 1971; Semenov, 1972; Strizhkov, 1980; Sedykh, 1983; Severtsev, 1984). Their output varies in terms of literary merit and aesthetic value, but they all appear closer to Russian hedonistic poetry than to the original Khayyam.

The most solid contribution was made by Vladimir Derzhavin (1908-75), a prolific poet-translator who was renowned, inter alia, for his translations of Persian classical literature. He published two collections of quatrains. The first collection (Derzhavin, 1965, with the language of the original noted as “Tajik-Farsi”) included 488 poems, of which 292 were from Osmanov and Aliev’s publication, while 196 were from Govinda Tīrtha’s 1941 edition, with 14 robāʿis translated twice as separate poems (Vorozheĭkina and Shakhverdov, 1986, p. 52). For his second collection, Derzhavin selected only 218 quatrains, and he revised and improved some of his translations (Derzhavin 1972, language of the original noted as “Farsi”). The poet applied various meters with a preference for long ones, and his work fails to evoke the laconic brilliance of the original. Unlike other poets-translators, Derzhavin followed the cribs meticulously in an attempt to preserve the “exoticism” of the translated texts. The strategy, given that he lacked a working knowledge of Persian poetic conventions, made his Khayyam sound verbose and at times even stiff and awkward. Derzhavin’s translations have, nevertheless, been well received by readers and critics and frequently reprinted.

It is, however, German Plisetskiĭ (1931-92) who should be mainly credited for promoting the Omar Khayyam cult in Russia. Plisetskiĭ used cribs made by learned Iranologists (M. N. Osmanov, Michael Zand) and worked in cooperation with his editor Natalia Kondyreva, also a well-known translator from Persian. The result was 450 poems (Plisetskiĭ, 1972), around 300 from Aliev and Osmanov’s edition (1959), the rest chosen from the editions by Moḥammad ʿAli Foruḡi (q.v.; 1943) and Govinda Tīrtha (1941). Plisetskiĭ’s translation reproduces the formal features of the original: the poet used anapestic tetrameter, rhythmically close to robāʿi meters, and observed regular rhymes and radīfs. As to the imagery, Plisetskiĭ relied at times not on the literal meaning, but rather on his own imaginative interpretation. Nevertheless, he succeeded in reproducing the seemingly unattainable simplicity of the Khayyamian robāʿi in a light, limpid, easy-to-remember Russian verse. Plisetskiĭ’s Omar Khayyam turned into a cult poet, first for intellectuals, and then for readers in general.

In spite of the fame and popularity of the Plisetskiĭ’s version, the eminent scholar Iosif Braginskiĭ (1905-89), who edited the volume of Iranian-Tajik poetry in the celebrated “Library of World Literature” series, chose to represent Khayyam by different translations (Braginskiĭ, 1974, pp. 101-24). The editor selected 137 of what he considered to be “authentic” texts and chose their “most reliable translations, which at the same time represented the joint experience of the Soviet school of translation” (Ibid, p. 585). Only six quatrains translated by G. Plisetskiĭ were included in the volume, the rest consisted mainly of translations by L.V. [Nekora], O. Rumer, and V. Derzhavin, with a small addition of contributions by I. Tkhorzhevskiĭ, S. Lipkin, I. Sel’vinskiĭ, and L. Pen’kovskiĭ.

In 1983, a learned Iranologist and brilliant translator of the Šāh-nāma, Tsetsiliya Banu (1911-98), married to the famous poet Abu’l-Qāsem Lāhuti, published her selection of 38 robāʿis (Banu, 1983). This fine and elegantly simple and fluent translation manages to adhere closely to the original in form and meaning. Unfortunately, the selection was relatively small and since the book was printed in Tajikistan, it was not widely distributed and did not attract the attention it deserved (for a much fuller collection of 104 robāʿis, published later, see Banu, 1991).

The enormous popularity of the Rubaiyat in Russia and the rest of the former Soviet Union has instigated an interest in its reception. A representative collection of translations made since the beginning of the 20th century was published by Zinaida Vorozheĭkina and A. Sh. Shakhverdov (1986); it is prefaced with an instructive survey of the methods and achievements of the translators as well as a useful index of first lines of the poems in the original source in Tajik (Cyrillic) characters. The book sets a precedent for later editors; in response to the high level of public interest, all the old and hitherto forgotten renderings have been since recollected and anthologized in numerous editions (Reĭsner, 1999; Safi, 2001; Malkovich, 2004; Butromeev et al., 2005; Sinel’nikov, 2008). A major publication in comparative translations (Malkovich, 2012) includes translations of Khayyam published in Russia by sixty-eight poets from 1891 to 2012. Every robāʿi is provided with a literal prose translation, followed by all the existing poetic versions (more than twenty in some cases).

The more recent Khayyam translations in the post-Soviet period tend to be inadequate both in terms of their fidelity to the original and as specimens of good poetry in general. One can detect supposedly “fresh” interpretations of the Rubaiyat that are actually based on earlier versions. Irina Evsa (2003) once again versified the literal translation made by Osmanov and Aliev in 1959. This book, along with Evsa’s awkward translations of Dante’s Divine Comedy and Goethe’s Faust, brought her a “worst translation” prize from the Moscow Knizhnoe obozrenie (Book review) newspaper in 2013 (http://www.biblio-globus.ru/inter_analytics.aspx?id=2046). Pavel Bunin, a famous painter and book illustrator, included his versified renderings of Khayyamic poems along with brilliant graphic illustrations in an erotic style in a book (Bunin, 2006) that also contained the “originals,” i.e. German translations made by Friedrich Rosen (1856-1935) in 1909.

Igor Golubev, who held a doctorate in technical sciences and who had learned Persian on his own, made a solid attempt to reinvent Khayyam’s image and message. He versified 1,300 of what he considered to be authentic poems, working with the Persian text of the Govinda Tīrtha edition (1941) as well as using some manuscripts (Golubev, 2000). In the lengthy introduction to his translations, Golubev sets out his criteria in detail (based mostly on stylistic and subjective grounds) for distinguishing authentic verses by Khayyam from those falsely attributed to him. Golubev also claims to present his own decipherment of Khayyam’s esoteric philosophic message. That collection has since become extremely popular and has gone through many editions with different publishers in both luxury and less expensive editions.

Presently, Khayyam enjoys the status of a uniquely readable and marketable poet in Russia: Every six months or so at least one or two new editions appear, in modest or fancy design. However, they happen to be mostly reprints of already issued popular translations, such as Plisetskiĭ’s Khayyam, the Butromeevs’ collection of old renderings, or the new versions by I. Golubev. Numerous websites (http://haiam.ru/index.html; http://хайям.рф/; http://hayam.spinners.ru/view_all.php, and others) publish old and modern translations; and one can Google “Kak skazal Omar Khaĭyam” (“as Omar Khayyam has put it”) to find out that the Persian poet has become the pre-eminent authority on wisdom in Russian internet blogs and forums.

• Natalia Chalisova
• Firuza Melville

KHAYYAM, OMAR x. ARABIC TRANSLATIONS OF THE RUBAIYAT

Although the Arab and Persian literary traditions have been in close contact for centuries, Khayyam only came to the attention of the Arabs as a noteworthy poet at the turn of the 20th century. Before the modern period, one can point to several references to Khayyam in Arabic anthologies and other literary texts, and these have provided extremely valuable material for the study of the literary history and reception of his poetry. However, most mention of Khayyam in the earlier Arabic sources is primarily as an astrologer, mathematician, and philosopher. Khayyam also wrote poetry in Arabic, but this too, along with other pre-modern references to him in Arabic, are described elsewhere (see above, KHAYYAM ii).

The early decades of the 20th century witnessed the rise in Arabic literature of various literary trends such as Romanticism and Symbolism, motivated by a keen orientation westward. A whetted appetite for translations from English and French played a seminal role in shaping the literary production of this period and in directing the Arab poets’ and writers’ negotiations with their literary tradition. It was at that moment of attentiveness to western literary imports that Khayyam, through Edward FitzGerald’s (q.v.; 1809-1883) English translation, was re-introduced into Arabic as a major world poet. Thus, FitzGerald’s English translation is the site of the first major encounter with the Persian poet in the modern Arab world, leading to a century of fascination with Khayyam, his life, his representations, and his elusive quatrains. To date, more than fifty-five translations of the Rubaiyat have been published in Arabic, including eight translations into different Arabic dialects (Alsulami, p. 79).

The modern engagement with Khayyam as a poet was launched by Jerji Zaydān’s response to a query by Asʿad Afandi Salim about the FitzGerald translation in al-Helāl 11, 15 January 1903. The journal continued to receive requests for more information about Khayyam, his biography, and the robāʿiyāt as a genre, all triggered by FitzGerald’s work. Consequently, al-Helāl published an extended study on Khayyam authored by the Lebanese ʿIsā Eskandar Maʿluf (1869-1956) titled “Khayyam and What the Arabs Knew about Him.” It included six quatrains translated from the English that Maʿluf rendered into metered Arabic verse.

Although a number of short translations appeared in various journals and magazines earlier, the first Arabic translation of the Rubaiyat in book form was by Wadiʿ Bostāni (1886-1954), published in Cairo in 1912. Bostāni was an Anglophone Lebanese intellectual, lawyer, and poet, who also worked as a translator for the British consulate in Beirut. While he consulted English, French, and German translations, his primary source text was FitzGerald’s translation. Bostāni made no effort to go back to the Persian original, and, in fact, he could not have done so because, as he admits in his introduction, he did not know any Persian at all. Bostāni translated 109 quatrains transforming them into seven hemistich qeṭʿas. Like Maʿluf before him, Bostāni adhered to a single meter (ḵafif). The issue of meter later took center stage in the debates around the Rubaiyat and their translation into Arabic, as translations are sometimes categorized into metered and unmetered (free or in lineated prose) translations. In 1921, the Iraqi Aḥmad Ḥāmed Ṣarrāf (1900-1985) published a lineated prose translation of the Rubaiyat, which was later rendered into metered verse by the poet Moḥammad Hāšemi (1898-1973). Ṣarrāf’s translation subsequently appeared in his book entitled ʿOmar al-Ḵayyām: ʿaṣroho, siratoho, adaboho, falsafatoho, robāʿiyātoho (1931). Another prose translation was done by the Iraqi poet Jamil Ṣedqi Zahāwi (1862-1936), who used it as basis for a consequent verse translation he produced himself. Zahāwi published the two versions in a volume titled Robāʿiyāt al-Ḵayyām, which first appeared in 1928. By then, the Rubaiyat had become a space for reflecting on and experimenting with meter in Arabic poetry. Aḥmad Ṣāfi Najafi’s (1895-1978) translation in verse, published in 1931, is a good example of such experimentation, for he employed multiple meters and organized his translated quatrains alphabetically based on final rhyming letter. His translation ultimately employed all the Arabic poetic meters and used all the letters of the Arabic alphabet as ending rhymes.

However, the Arabic translations of the Rubaiyat are more commonly divided into direct and indirect translations, depending on whether the translator consulted Persian manuscripts or merely relied on translations into other languages. Of course, FitzGerald remained the major source text for the majority of indirect translations. After Bostāni’s translation, several indirect translations appeared, such as those made by ʿAbd al-Laṭif Naššār (1895-1972, Egypt) in 1917, Moḥammad Sibāʿi (1881-1931, Egypt) in 1922, Tawfiq Mofarraj (d. 1968, Lebanon) in 1947, and Aḥmad Zaki Abu Šādi (1892-1955, Egypt) in 1932 and in 1957.

Aḥmad Rāmi’s (1892-1981) translation was hailed as the first direct translation from Persian into Arabic when it appeared in a first edition in 1924 and then in a second revised edition in 1932. Rāmi discovered Khayyam’s quatrains via Bostāni’s translation of FitzGerald’s poem. After graduating from the Teachers Institute in Cairo, where he learned English, Rāmi read FitzGerald in the original English. Like many poets and intellectuals of his generation, Rāmi was deeply influenced by English and French poetry, and especially by English Romanticism. He was a member of the Apollo Group, a collective of poets and writers who published a monthly journal under the same name, consisting primarily of translations of English, American, and French poetry (Hafez). It was FitzGerald and the popularity of his translation that set Rāmi on a quest in search of Khayyam in the maze of the available Persian manuscripts.

Rāmi spent the years between 1922 and 1924 studying Persian at the Institute of Oriental Languages in Paris. He examined different Persian manuscripts of the Rubaiyat and became aware of other major poets in the Persian tradition such as Saʿdi and Ferdowsi (qq.v.). Thus, when translating Khayyam into Arabic, Rāmi looked in two places as he patched his source text together: the various Persian manuscripts on one hand, and Fitzgerald’s English poem on the other, which offered some relief from the elusiveness of the originals.

Rāmi translated 167 quatrains without depending on one manuscript exclusively. However, the role of the Persian manuscripts in his project eventually receded, revealing the influence of FitzGerald’s poem as the most detectable in his selection of quatrains and their organization. Rāmi’s translation is not arranged alphabetically as most of the manuscripts are, further attesting to his reliance on Fitzgerald’s poem when arranging his translation. Rami’s translation remains the most celebrated and circulated translation of the Rubaiyat in Arabic. It has been reprinted twenty-five times, most recently in Cairo in 2006, commemorating the twenty-fifth anniversary of Rāmi’s death.

However, the popularity of Rāmi’s translation and the continued demand for it in print are both a result of its adaptation into song by Omm Kolṯum (d. 1975). In 1950, the Egyptian diva collaborated with Rāmi and the composer Riyāż Sonbāṭi (d. 1981) to create the “poem” that she would later sing. Omm Kolṯum personally chose fifteen quatrains from Rāmi’s translation, editing them and rearranging them in a manner that fit her voice and musical personality. The final product, Omm Kolṯum’s “Robāʿiyāt al-Ḵayyām,” became the most widespread and well-known Khayyam in Arabic, instantly becoming part of the repertoire of popular Arabic song and consecrating Rāmi’s translation (or parts of it) as an Arabic poem in its own right. Omm Kolṯum formulated her selections from Rāmi’s translation into a coherent Arabic poem about spiritual redemption with resonant political undertones which spoke to the historical moment of Egypt in the 1940s and 1950s. She incorporated her “Robāʿiyāt al-Ḵayyām” into a repertoire of songs that expressed dissatisfaction with the continued British presence and voiced a budding nationalist sentiment (Danielson, p. 113-114).

Omar Khayyam’s Rubaiyat was received into modern Arabic literature as a colonial text, shaped and validated by FitzGerald’s translation. The quatrains were, however, gradually absorbed into the repertoire of modern Arabic poetry, not only through the many translations but also through the adaptations, rewritings, and original works inspired by Khayyam and his verses. Arab writers and intellectuals were preoccupied with Khayyam’s personality as channeled through the various rewritings of his quatrains into Arabic, sometimes portraying him as an Epicurean free-thinker and at others as a pious man of God. Each Arabic translation fashioned an image of Khayyam in accordance with the translator’s views and convictions. The Rubaiyat offered Arab poets and intellectuals a space for negotiating the relationship with the West, the role of tradition in the projects of modernization, the evolving definition of poetry and poetic form, and the catalytic intervention of translated texts.

Huda Fakhreddine

KHAYYAM, OMAR xi. TURKISH TRANSLATIONS OF THE RUBAIYAT.

See Supplement.

KHAYYAM, OMAR xii. OTHER TRANSLATIONS OF THE RUBAIYAT.

See Supplement.

KHAYYAM, OMAR xiii. MUSICAL WORKS BASED ON THE RUBAIYAT

The enduring popularity of the verses that make up the Rubaiyat (Robāʿiyāt) of Omar Khayyam (ʿOmar Ḵayyām), both in the original Persian and in translation, is reflected in the substantial number of musical works that have been inspired by this work. Many other poets have also stimulated the creation of musical compositions: Shakespeare has been a perennial favorite for composers in the West, and other poets, including many 19th century poets such as Alfred Lord Tennyson (1809-92) have been frequently set to music (Gooch and Thatcher 1979, pp. 509-629; 1982, pp. 43-177). But given the comparative brevity of the Rubaiyat, no more than 110 verses in the longest version by Edward FitzGerald (q.v.; 1809-83), it is remarkable that well over 150 composers have used this single work as their source of inspiration (Martin and Mason, 2007b).

The variety of music that has been created in response to the Rubaiyat is considerable. The works include modern popular music, as well as classical music from the late 19th century onwards. Most compositions involve either the setting of words from the Rubaiyat, frequently in English and from one of FitzGerald’s versions, or a parallel narration of the verses. But there are also film scores, orchestral works and simple piano pieces, as well as jazz suites and pop records (Coumans, pp. 2-4; Garrard, pp. 224-28). Although much of the music originates in the West, particularly in the United Kingdom and the United States, there are works by a number of composers from Russia, Central Asia and the Middle East, including Iran.

In Iran, there is a long established tradition of declaiming or singing verses from major poets with musical accompaniment (During et al, pp. 153-61; Yarshater, pp. 59-78). The ghazals of Hafez, Sa ʿdi, and Rumi have been perennial favorites in this respect. As far as the Rubaiyat are concerned, we do not know of any specific artists or performances whose music for the Rubaiyat was scored or recorded in earlier decades of the 20th century. There is, however, a well-known recording from the 1970s, still available from the Mahoor Institute, of the original Persian text with music by Fereydun Šahbāziān, recitations by Aḥmad Šāmlu (Shamlu) and vocals by Moḥammad-Reżā Šajariān (Shajarian). Verses from the Rubaiyat have also been included, together with works of other major Iranian poets, in “Ascension,” a composition by Kāmbiz Rošan-Ravān, issued on CD by Technoor in 2002.

PLATE I Cover of the score for Liza Lehmann’s 1896 song cycle “In a Persian Garden” (Designed for American edition, G. Shirmer, 1898). Illustration courtesy of Jos Coumans.

The impact of FitzGerald’s translation. Western interest in music for the Rubaiyat dates from the late 19th century when FitzGerald’s translation (first published in 1859) began to gain in popularity (Martin and Mason, 2007a, pp. 7-8). The earliest known setting of a selection of these verses, and one that became very popular in the early part of the 20th century, was by the celebrated English singer and composer Liza Lehmann (1862-1918). In 1896 she created a song cycle, “In a Persian Garden,” for four soloists and piano (PLATE I), using thirty-one quatrains from FitzGerald’s versions of the Rubaiyat (Garrard, pp. 224-25). Some of the songs in this work became very well known, notably the one entitled “Ah, Moon of my Delight!” which was sung by the American tenor Mario Lanza (1921-1959) among others. The whole work has also been recorded, most recently by the Cantabile Vocal Quartet, on a CD issued by Quattro Voci Records in 2000.

Granville Bantock (frontispiece of G. Bantock, ed., One Hundred Folksongs of All Nations, Boston, 1911).

Lehmann’s setting of verses from FitzGerald’s Rubaiyat was followed by a steady flow of works by other composers. Almost every year from 1900 to 1940, some work or other based on the Rubaiyat was produced in the United States or Europe (Martin and Mason, 2007b). Many compositions were straightforward settings of one or more verses for voice(s), with accompaniment from piano or a chamber ensemble. Some of these works fall into the category of popular or drawing-room ballads, which were much in demand in the early years of the 20th century. Others are more in the nature of the “art song” (Northcote, pp. 96-97). They include works by composers such as Roger Quilter (1877-1953), who wrote a setting of verses from the Rubaiyat for unaccompanied voices in 1902. Vivian Ellis (1904-1996), an English composer of musicals, set three songs from the Rubaiyat for voice and piano in 1921. Some musicians from the rest of Europe were also active in the field; for example, a French composer, Jean Cras (1879-1932), composed a setting of five quatrains for voice and piano in 1925, while, in the Netherlands around 1916, the musician Willem Smalt made a setting for an a cappella choir of some quatrains from the Dutch translation by Petrus (Pieter) C. Boutens (1870-1943).

Not all the works based on the Rubaiyat in the first half of the 20th century were small-scale pieces. Probably the best-known large-scale work of this period is Granville Bantock’s composition “Omar Khayyam” for soloists, chorus and orchestra. Bantock (1868-1946; PLATE II) was an English composer and conductor, most of whose works were substantial compositions, and his “Omar Khayyam” is no exception. It is a three-part work, setting all the 101 quatrains from FitzGerald’s fifth edition, and it lasts nearly 3 hours in performance (PLATE III). Bantock composed the “oratorio” in the period 1906-9, and it was first performed in this period in separate parts with the composer conducting (Foreman, pp. 10-11). The complete work has been broadcast and recorded a number of times in Britain; the most recent recording is from 2007 by Chandos with the BBC Symphony Chorus and Orchestra conducted by Vernon Handley.

A limited number of other composers produced orchestral works based on the Rubaiyat in the years before the Second World War. The American Arthur Foote (1853-1937) created an orchestral suite, “Four Character Pieces after the Rubaiyat of Omar Khayyam” in 1912, based on an earlier work for voice and piano. Charles Cadman (1881-1946), another American, was commissioned to create a musical score for a silent film about Omar Khayyam, finally issued as “A Lover’s Oath” in 1925. Cadman’s music for orchestra was published as “Oriental Rhapsody” in 1921. In 1917, Henry Houseley (1851-1925) published a cantata for soloists, chorus and orchestra, entitled “Omar Khayyam.” A little later, in 1924, the Swiss composer, Robert Blum (1900-1994), gave his first symphony, with baritone soloist, the appellation “Omar Khayyam.” Most of these works seem now to have vanished from the performing repertoire. But the Arthur Foote suite was included in a record issued by the US Library of Congress in the 1980s.

PLATE III First page of vocal score for Sir Granville Bantock’s oratorio “Omar Khayyam” (W. Breitkopf & Hartel, 1906).

World-wide interest in the modern period. Rubaiyat-based compositions from countries other than the UK or the USA were relatively few in number before the First World War. They increased gradually in the interwar years, and more especially in the second half of the 20th century (Martin and Mason, 2007b). This period has seen the creation of song settings from Rubaiyat verses in languages ranging from Dutch to Russian and from Finnish to Portuguese. An Uzbek musician, Firus Bachor (b. 1942), composed an opera on the Omar Khayyam theme in the 1990s. There has also been a regular flow of compositions using verses by FitzGerald in English. Quite a number of pop musicians from different countries have found inspiration from the Rubaiyat, while Khayyam and FitzGerald have been an influence both in the emerging sectors of world music, and in music used in Western explorations of mysticism based on Eastern thought.

The more classical formats of song settings have remained the most common forms of composition. Some of the major names among modern composers have put words from the Rubaiyat to music. Alan Hovhannes (1911-2000) created a major work, “The Rubaiyat, A Musical Setting” for narrator and orchestra in 1975, which has been twice recorded. The 1945 setting by Paul Hindemith (1895-1963) of two quatrains in English for “a cappella” of voices was included in a recording in 1998. In 1959, a verse from the Rubaiyat in its Persian original was included in the composition, “Strophes,” by Krzysztof Penderecki (b. 1933). This, too, has been recorded subsequently. A Dutch composer, Lex van Delden (1919-1988), won the music prize of the City of Amsterdam in 1948 for his Rubaiyat cantata, a setting for solo voices, chorus, two pianos and percussion.

Two films about Omar Khayyam have had original musical scores. The music for the first of these films, The Life, Loves and Adventures of Omar Khayyam, which starred Cornel Wilde and appeared in 1957, was the last film score to be created by the American composer Victor Young (1899-1956). The film including its music was recorded on video. More recently, the film The Keeper: The Legend of Omar Khayyam was released in 2005 with music mainly by Elton (Farrokh) Ahi. The complete work is available on DVD.

The influence of the Rubaiyat is also to be seen in a range of modern pop music, including jazz, folk, soul and rock and roll. Key names that have been documented by Coumans (pp. 3-4) are: the jazz musician, Dorothy Ashby; the American folk singer, Woody Guthrie; soul musicians, Allan Toussaint and Willie Harper. The rock-and-roll singer Van Morrison included a mention of Omar Khayyam in one of his recorded lyrics (“Rave on John Donne”). More recently, texts from the Rubaiyat have been identified in works by pop artists such as Coldcut, In the Nursery, and David Olney, while the Rubaiyat of Dorothy Ashby was reissued on CD by Dusty Groove America in 2007. Looking further afield, the Egyptian singer Om Kolthoum (Omm Kolṯum; 1904-75) both performed and recorded songs in Arabic based on the Rubaiyat (Garrard, p. 227). In the field of world music, an Italian group Milagro Acustico, and a French one led by Abed Azrie, have each combined Rubaiyat texts in various languages with Eastern and Western instrumentation. The American-Iranian group, Axiom of Choice produced, in 2002, a vocal-instrumental recording subtitled “A Trans-Global Exploration of Omar Khayyam’s Mystical Vision.” This followed the production in the mid-1990s of two compact discs by Clarity Sound and Light, with instrumental music by J. Donald Walters that was inspired by the Rubaiyat; one recording is described as “a Persian fantasy for sitar and tabla,” the other is “a musical journey into the inner world of Omar Khayyam’s mystical love-poem.”

It is clear that the name of Omar Khayyam and his Rubaiyat lives on into the 21st century through these musical forms, as well as in the continued publication of the poem in book form (Martin and Mason, 2007a, pp. 29-30). Some of the musical interpretations, from earlier periods as well as modern times, may not resonate very closely with the original Persian verses or their original worldview. But the existence of Rubaiyat-based music, and particularly the recording and distribution of such music on a world-wide basis, has brought awareness of the medieval Persian poet and his Victorian English interpreter to a much wider audience than might otherwise have been the case.

• William H. Martin
• Sandra Mason

KHAYYAM, OMAR xiv. AS MATHEMATICIAN

Three mathematical treatises of Omar Khayyam have come down to us: (1) a commentary on Euclid’s Elements; (2) an essay on the division of the quadrant of a circle; and (3) a treatise on algebra. A fourth treatise, on the extraction of the nth root of numbers, is not extant.

(1) THE COMMENTARY ON EUCLID’S ELEMENTS

Khayyam’s commentary on the difficulties of certain postulates of Euclid’s work (Resāla fi šarḥ mā aškala men moṣādarāt ketāb Oqlides) was completed in 470/December 1077. In this treatise, Khayyam intends to amend and rectify what he considers to be the most important difficulties found in the Elements of Geometry, or simply the Elements, a work in thirteen books attributed to Euclid of Alexandria (fl. ca. 300 BCE). The first part of Khayyam’s commentary deals with the theory of parallel lines, the second with the concepts of ratio and proportionality, and the third with the compounding of ratios.

Theory of parallels. Euclid (Oqlides) had expounded the theory of parallels in the first book of the Elements. In it, he defined parallel lines as “straight lines which, being in the same plane and being produced indefinitely in both directions, do not meet one another in either direction” (Heath, I, p. 154). Yet an important part of the theory was based on a statement that Euclid had postulated in the beginning of the same book, that is, the Parallel Postulate: “That, if a straight line falling on two straight lines makes the interior angles on the same side less than two right angles, the two straight lines, if produced indefinitely, meet on that side on which are the angles less than the two right angles” (Heath, I, p. 155). For nearly two thousand years, mathematicians were dissatisfied with this statement, which they always considered as a proposition to be demonstrated rather than a postulate to be admitted.

The first criticisms of the Euclidean theory of parallels have been preserved in the Commentary on the First Book of Euclid’sElements by the Neo-Platonist philosopher Proclus Lycius (410-85). Lycius remarks that Posidonius of Rhodes (135-51 BCE) had defined parallel lines as “lines in a single plane which neither converge nor diverge but have all the perpendiculars equal that are drawn to one of them from points on the other” (Proclus, p. 138), that is, as equidistant straight lines. Proclus also mentions an attempt by Ptolemy (fl. ca. 125-61) to prove Proposition I.29 of the Elements, the first proposition in which Euclid had made use of the Parallel Postulate, but without resorting to it. Proclus himself attempts to prove this postulate. He says that anyone who wants to prove it “must accept in advance such an axiom as Aristotle [De Caelo 1.5.271b 28 ff.] used in establishing the finiteness of the cosmos: If from a single point two straight lines making an angle are produced indefinitely, the interval between them when produced indefinitely will exceed any finite magnitude” (Proclus, p. 291). By means of this axiom, Proclus is then able to prove the Parallel Postulate, but assuming that the distance between two parallel lines is of a finite magnitude.

In a passage preserved in the commentary of Abu’l-ʿAbbās Fażl b. Ḥātem Nayrizi (fl. ca. 287/900) on the Elements, the Greek philosopher and commentator Simplicius (first half of the 6th century) quotes a proof of the Parallel Postulate made by his colleague Aḡānis, possibly the Athenian philosopher Agapius, who flourished about 511 (Lo Bello, pp. 224-29). Aḡānis first defines parallel lines as “those in one plane [such that] if they are extended with an endless extension, without limit, in both directions, together, the distance that is between them is always one distance” (Lo Bello, p. 158). This definition of parallel lines as equidistant straight lines will then enable him to prove Proposition I.29 of the Elements, as well as the Parallel Postulate.

The first Arabic mathematician who dealt with the Euclidean theory of parallels is ʿAbbās b. Saʿid Jawhari (fl. ca. 215/830), in a treatise, now lost, devoted to Euclid’s Elements. But his attempt to prove the Parallel Postulate has been preserved by Naṣir-al-Din Ṭusi (597-672/1201-74) in his “treatise that relieves from the doubt regarding parallel lines” (al-Resāla al-šāfia ʿan al-šakk fi al-ḵoṭuṭ al-motawāzia). Jawhari notably proves that parallel lines are equidistant; but he assumes implicitly that if a straight line falling on two straight lines makes the alternate angles equal to one another, then any other straight line which falls on the two straight lines will also make the alternate angles equal to one another. He then proves the Parallel Postulate (Jaouiche, pp. 24, 37-44, 137-44; Houzel, p. 170).

After Jawhari, Abu’l-Ḥasan Ṯābet b. Qorra (211-88/826-901) made two attempts to prove the Parallel Postulate. In his treatise on the proof of Euclid’s celebrated Postulate (Maqāla fi borhān al-mosādara al-mašhura men Oqlides), he admits as a principle that if two straight lines cut by another straight line diverge in one direction, they will converge in the other direction. This will enable him to prove that in case the alternate angles are equal, the two straight lines will be equidistant. He then proves the Parallel Postulate by means of the so-called “Axiom of Archimedes.” In the treatise on “The fact that two lines produced according to less than two right angles will meet” (Fi anna al-ḵaṭṭayn eḏā oḵrejā elā aqall men zāwiatayn qāʾematayn eltaqayā), Ṯābet b. Qorra introduces the concept of motion. He notably admits as a principle that any point on a solid that moves according to a uniform and rectilinear translation will describe a straight line. This enables him to produce two equidistant straight lines (Jaouiche, pp. 22-23, 45-56, 145-60; Houzel, p. 171).

Abu ʿAli Ḥasan b. Ḥasan b. Hayṯam (d. after 432/Sept. 1040) defines parallel straight lines through the concept of equidistance. In order to achieve this, he attempts to prove that if a finite straight line moves perpendicularly to a fixed straight line, then its extremity will describe a straight line parallel to the fixed line; this will enable him to prove the Parallel Postulate (Jaouiche, pp. 57-74, 161-84; Houzel, pp. 171-72).

Khayyam considers that the attempts of his predecessors to prove the Parallel Postulate were not satisfactory, in that each one of them had postulated something that was by no means easier to admit than the Postulate itself. He elaborates in particular on Ebn al-Hayṯam’s attempt, rejecting categorically the introduction of the concept of motion into geometry. Khayyam’s intention is to prove eight propositions, notably Proposition I.29 of the Elements, and the Parallel Postulate (Rashed and Vahabzadeh, 2000, pp. 185, 219-20, 225-27, 230-33). What makes his attempt particularly interesting is his philosophical position on this matter. Khayyam thinks that the error his predecessors made, while trying to prove the Parallel Postulate, is that they disregarded some of the principles taken from the philosopher (i.e., Aristotle). He believes that the Parallel Postulate should be proven taking as a starting point certain philosophical premises, which, in his opinion, are immediate consequences of the very notions of straight line and of rectilinear angle; for once these premises be taken as necessarily true, then the geometer can admit them without proof. These premises are: (1) Two straight lines that intersect will diverge while going away from the point of intersection (Proclus had already resorted to this premise, attributing it explicitly to Aristotle). (2) Two converging straight lines will intersect. (3) Two converging straight lines cannot diverge while going toward convergence, and vice-versa. Khayyam also assumes (while proving the third proposition) that parallel lines are equidistant; but since he does not give any explanation whatsoever, it is difficult to know whether he takes this as an obvious consequence of the second premise, or whether he considers, like some of his predecessors, that “parallel” and “equidistant” are synonymous (Rashed and Vahabzadeh, 2000, pp. 185, 224, 226).

It should be noted in this respect that, from Khayyam’s point of view, the fact that the second and third premises are mathematically equivalent to the postulate he intends to prove is not an issue at all. Khayyam is not really concerned with matters of mathematical equivalence; he is rather concerned with the fact that the second and third premises are immediate consequences of the concepts of straight line and rectilinear angle, whereas the Postulate is not; and this is why the Postulate should, in his opinion, be proven through them.

The gist of Khayyam’s argumentation is found in the proof of the third proposition. In it, he considers (Figure 1) a quadrilateral ABCD, in which the sides AC and BD are equal to each other and both perpendicular to the base AB. As a consequence of the first proposition he had just proven, the angles ACD and BDC will then be equal to each other.

Figure 1. Diagram for Khayyam’s proof of the third proposition.

He next examines successively the three possible cases, namely that the angles ACD, BDC are both right, both acute, or both obtuse. He first proves that, if one assumes that these angles are acute, then one will get two straight lines that cut another straight line at right angles and diverge on both sides of this straight line; and this contradicts the third premise. Likewise one will arrive at a contradiction assuming that these angles are obtuse. Therefore the angles ACD, BDC will necessarily be right angles. He can now easily prove Proposition I.29 of the Elements, as well as the Parallel Postulate (Rashed and Vahabzadeh, 2000, pp. 185-86, 226-30). Khayyam ends this part of his treatise by explaining that the eight propositions he has just proven should take the place of Proposition I.29 of the Elements, omitting, however, all the philosophical considerations he has elaborated upon, since these properly belong to the science of metaphysics, not to geometry (Rashed and Vahabzadeh, 2000, pp. 186, 233).

About two centuries later, Naṣir-al-Din Ṭusi took up in part Khayyam’s ideas in two treatise: one is the treatise that relieves the doubt regarding parallel lines (al-Resāla al-šāfia), which is devoted to the proof of the Parallel Postulate and contains extensive passages from Khayyam’s proof; the other is the Redaction of Euclid (Taḥrir Oqlides), a treatise devoted to Euclid’s Elements in their entirety.

Traces of Khayyam’s proof of the Parallel Postulate can still be found as late as the 18th century. In the first propositions of his Euclides vindicatus (Euclid freed of all blemish), the Jesuit mathematician Girolamo Saccheri (1667-1733) considers the very same quadrilateral ABCD, now known as the “quadrilateral of Saccheri,” as well as the three possible cases regarding its equal angles ACD, BDC, which Saccheri calls, respectively, the hypothesis of the right angle, the hypothesis of the acute angle, and the hypothesis of the obtuse angle.

Concepts of ratio and proportionality. Euclid had expounded in Book V of the Elements the theory of proportion applicable to every kind of magnitude (lines, surfaces, solids, time). The whole theory was based on the definitions found in the beginning of the Book, of which the following two were to play a prominent part: “3. A ratio is a sort of relation in respect of size between two magnitudes of the same kind. 5. Magnitudes are said to be in the same ratio, the first to the second and the third to the fourth, when, if any equimultiples whatever be taken of the first and third, and any equimultiples whatever of the second and fourth, the former equimultiples alike exceed, are alike equal to, or alike fall short of, the latter equimultiples respectively taken in corresponding order” (Heath, II, p. 114).

What should be noted regarding Definition V. 3 is that the meaning of the words “a sort of relation in respect of size” is nowhere explained in the Elements. This posed a problem as to how they should be interpreted, notably the Greek word pēlikotēs, which is variously translated as “size,” “value,” or “quantity.”

Definition V.5 was the cornerstone of the theory, since it could apply to all magnitudes, whether commensurable or incommensurable, in contradistinction to the alternative definition of proportionality found in Book VII of the Elements, which, strictly speaking, applied only to numbers but could easily be extended to commensurable magnitudes, namely Definition VII.20: “Numbers are proportional when the first is the same multiple, or the same part, or the same parts, of the second that the third is of the fourth” (Heath, II, p. 278).

Definition V.5 posed some problems. First of all, the comparison of multiples of magnitudes did not seem to have any definite and obvious relationship with the concept of proportionality. Secondly, Euclid did not give any indication on how this definition had been conceived or established, so that nothing relevant could be found within the Elements themselves to enable mathematicians to unravel his intention. Finally, although it was meant to define the concept of “same ratio,” Euclid’s exposition did not permit to establish a relationship between Definition V.3 and Definition V.5 (Vahabzadeh, 2002, pp. 10-11).

No Greek commentary on Definition V.5 has come down to us (Euclid of Alexandria, tr. Vitrac, II, pp. 539-43); however, the situation is quite different with regard to Arabic mathematics. This Definition has indeed given rise to numerous commentaries in Arabic, whose aim was either to justify it by means of a proof, or to substitute for it another definition known as the “anthyphairetic definition of same ratio” (see, e.g., Plooij, pp. 48-56, 61-66). The latter definition consisted in applying to two homogeneous magnitudes a process usually known as the Euclidean algorithm, but which historians also call anthyphairesis after a Greek word meaning “alternating subtraction,” “reciprocal subtraction.” More precisely, given two homogeneous magnitudes, the lesser magnitude is subtracted from the greater a certain number of times, until one arrives at a remainder less than the lesser magnitude. Then this remainder is subtracted from the lesser magnitude a certain number of times, until one arrives at a second remainder less than the first remainder. Then one proceeds in the same manner with every pair of consecutive remainders. The sequence of natural numbers thus obtained can then be considered as a “characteristic” of the ratio of the two magnitudes. Now if the ratio of another pair of homogeneous magnitudes is characterized by the same sequence of numbers, then the four magnitudes are said to be in the same ratio, that is, they will be proportional (see, e.g., Plooij, pp. 57-60). For instance, let us assume that AB and CD (Figure 2) are two homogeneous magnitudes. Let us suppose that AB measures CD once, leaving ED less than AB; that ED measures AB three times, leaving FB less than ED; and that FB measures ED twice, leaving GD less than FB. If we proceed in the same manner with every pair of successive remainders, this process will then yield the sequence 1, 3, 2, …

Also consider KL and MN (Figure 3) to be another pair of homogeneous magnitudes. We suppose that KL measures MN once, leaving ON less than KL; that ON measures KL three times, leaving PL less than ON; and that PL measures ON twice, leaving QN less than PL, and proceed in the same manner with every pair of successive remainders. Now if the process applied to KL and MN yields the same sequence 1, 3, 2 …, then the ratio of AB to CD is said to be the same as the ratio of KL to MN.

Figure 2. Illustration of anthyphairesis.

Figure 3. Illustration of the anthyphairetic definition of same ratio.

As noted by some historians (Plooij, p. 63; Gardies, pp. 80-81, 90-91; Vahabzadeh, 1997, pp. 253-57), one of the features of the anthyphairetic definition of same ratio is that, unlike Euclid’s Definition V.5, it allows for each one of the ratios that make up a proportion to be considered independently of the other, which is necessary in order to give a meaning to the concept of ratio in itself, notably for regarding a ratio between any two magnitudes as a number (see below, Compounding of ratios and irrational numbers).

Euclid had already used the anthyphairetic process for finding the greatest common measure of two numbers and of two commensurable magnitudes (Elements, Propositions VII.2 and X.3). He also used it as a criterion for proving that two numbers are prime to one another, and that two magnitudes are incommensurable (Propositions VII.1 and X.2); he, however, did not use this process to define proportional magnitudes. In Topics VIII.3. 158b 29-35, Aristotle states: “It would seem that in mathematics also some things are not easily proved by lack of a definition, such as the proposition that the straight line parallel to the side which cuts the plane [i.e., the parallelogram] divides in the same way both the line and the area. But when the definition is stated, what was stated becomes immediately clear. For the areas and the lines have the same alternating subtraction (antanairesis); and this is the definition of the same proportion” (Thomas, I, p. 507). In his commentary on this passage, Alexander of Aphrodisias (ca. 210) adds: “For the definition of proportions which those of old times used is this: Magnitudes which have the same alternating subtraction (anthyphairesis) are proportional. But he [i.e., Aristotle] has called anthyphairesis antanairesis” (Thomas, I, p. 507). Many historians consider this as evidence of the existence in pre-Euclidean mathematics of a definition of proportional magnitudes based on anthyphairesis, and some have attempted to reconstruct such a pre-Euclidean theory of proportion; however, no trace of this anthyphairetic definition has been found in any extant Greek mathematical text (e.g., see Fowler, 1999a; Euclid of Alexandria, tr. Vitrac, II, pp. 515-23; and esp. Vitrac, 2002, pp. 158-74).

The first mathematical text in which the anthyphairetic definition of same ratio is explicitly mentioned is the Treatise On the Difficulty Concerning Ratio (Resāla fi’l-moškel men amr al-nesba) by Abu ʿAbd-Allāh Moḥammad b. ʿĪsā Māhāni (fl. 246/860). In it, Māhāni considers the ratio of two homogeneous magnitudes as the state occurring to each magnitude when measured by the other. Following a directive of Ṯābet b. Qorra, he characterizes this measure by means of the anthyphairetic process; two ratios will then be the same if this process yields the same sequence of numbers when applied to each pair of magnitudes (see above). Māhāni also states a definition of greater ratio based on anthyphairesis. He then proves that his definitions of same ratio and of greater ratio are equivalent to Definition V.5 and Definition V.7 (Euclid’s definition of greater ratio) respectively (Vahabzadeh, 2002, pp. 12-14, 31-40).

In his commentary on Euclid’s Elements, Abu’l-ʿAbbās Nayrizi (fl. ca. 287/900) has also interpreted Definition V.3 in terms of anthyphairesis; but, unlike his predecessor Māhāni, he considers that it is not necessary to prove Definition V.5 since, in his opinion, this definition belongs to the principles of Book V. Neither does he study the connection between the anthyphairetic definition of same ratio and Definition V.5 (Plooij, pp. 51-53, 61).

In the second part of his commentary, Khayyam intends to deal thoroughly with the concepts of ratio and proportionality between magnitudes, since in his opinion this matter had never been dealt with in a satisfactory and philosophical manner. Commenting upon Euclid’s Definition V.3, Khayyam says that two things enter into the concept of ratio: the relation between the two magnitudes as to equality and inequality, and the size, or the magnitude, of this ratio. He explains that this concept was first found in natural numbers; that is, when one considers the numbers related to one another, one finds that they are either equal or unequal; and if they are unequal, then the smaller number will either be a part or parts of the larger one. For instance, since 3 measures 9 three times, 3 is one-third of 9, and the size of the ratio of 3 to 9 will be one-third; likewise 2 is two-sevenths of 7, and the size of ratio of 2 to 7 will be two-sevenths. When one considers this concept with regard to magnitudes, one will find, besides the preceding three possibilities, a fourth one, namely that the two magnitudes may be incommensurable, so that the less will neither be a part of the greater nor parts (Rashed and Vahabzadeh, 2000, pp. 188, 234-35).

He then recalls Euclid’s Definition V.5, and adds: “But this does not manifest (yonabbeʾ ʿan) true proportionality. Do you not see that if a questioner inquires, saying: Four magnitudes are proportional according to Euclidean proportionality, and the first is the half of the second, will the third then be the half of the fourth, or not?” (Rashed and Vahabzadeh, 2000, p. 236, with correction). In other words, even though it is fairly easy to prove starting from Definition V.5 that the third magnitude will also be one-half the fourth (for according to Definition V.5 twice the third magnitude is equal to the fourth, so that the third is obviously one-half the fourth), yet this definition is not satisfactory in that it does not manifest an immediate property of proportional magnitudes, as in fact any true definition should. He calls the Euclidean concept of proportional magnitudes “common proportionality” and intends to speak of “true proportionality” (Rashed and Vahabzadeh, 2000, pp. 188, 236).

Khayyam’s conception of ratio and proportionality is essentially the same as that of his predecessors Māhāni and Nayrizi. That is, given two magnitudes, they will either be equal, or the less will be a part or parts of the greater; and if the two magnitudes are incommensurable, their relation will then be characterized by means of anthyphairesis; whence it follows that two ratios will necessarily be the same if the anthyphairetical process applied to each pair of magnitudes yields the same sequence of numbers. Khayyam also defines the concept of greater ratio through anthyphairesis. He then proves that the anthyphairetic definitions of same ratio and of greater ratio are equivalent to Euclid’s corresponding definitions. Consequently all properties of proportional magnitudes that had already been established within the framework of the Euclidean theory will remain valid within the framework of a theory based on the anthyphairetic definitions; therefore, these properties do not need to be proven again (Rashed and Vahabzadeh, 2000, pp. 188-89, 236-49).

Compounding of ratios and irrational numbers. We find in the beginning of Book VI of the Elements a definition (now considered as an interpolation) according to which a ratio is compounded of ratios when, the sizes of these ratios being multiplied together, they produce a certain ratio. But Euclid does not explain anywhere what he means by “the size of a ratio,” not to mention the “multiplication” of these sizes. However, he does make use of the compounding of ratios in Propositions VI.23 and VIII.5 of the Elements. In each case, he admits without proof that, given three magnitudes or three natural numbers A, B, C, the ratio of A to C will be compounded of the ratio of A to B and of the ratio of B to C. It is this last statement, the one mathematicians use when compounding ratios, that has often been considered as a proposition that should be proven. This is in fact what Eutocius of Ascalon (b. ca. 480) applies himself to when commenting on Proposition II.4 of On the Sphere and the Cylinder, in which Archimedes (ca. 287-212 BCE) had used the preceding statement on the compounding of ratios. But although Eutocius intends to provide a general proof that this statement is valid for both natural numbers and magnitudes, he only considers ratios between numbers, so that his proof does not apply to ratios between incommensurable magnitudes (Heath, pp. 189-90, 247-48, 354; Youschkevitch, pp. 86-87; Netz, 2004a, pp. 312-15).

Bernard Vitrac has shown that Eutocius attempted later on to overcome the preceding limitation while commenting on Proposition I.11 of the Conics of Apollonius of Perga (b. ca. 262 BCE). Says Eutocius: “A ratio is said to be compounded of ratios when the sizes of the ratios multiplied by themselves produce something; it is being understood that ‘size’ is obviously said of the number to which the ratio is paronymous. On the one hand, it is with the multiples that the size can be a natural number [e.g., the size of the triple ratio is three]; with the other relations, the size will necessarily be a number and a part or parts [e.g., the size of the sesquialter ratio is one-and-a-half], unless, however, someone maintains that there also exist inexpressible relations, as are those between irrational magnitudes. And on the other hand, it is obvious that, in all the relations, this very size multiplied by the consequent term of the ratio will produce the antecedent” (Vitrac, 2000, Annexes, pp. 99-100). Eutocius then produces a proof essentially the same as the one in his commentary on Proposition II.4 of On the Sphere and the Cylinder, but this time he does not specify whether the ratios are numerical or not. He finally adds: “But the readers must not be worried by the fact that this has been demonstrated by arithmetical means, even if the Ancients indeed made use of these demonstrations by proportions that are more mathematical than arithmetical, and this because the object of the research is arithmetical; for ratios and sizes of ratios and multiplications of numbers belong first of all to numbers, and, from there, to magnitudes, in accordance with the saying: ‘for these mathematical sciences appear to be sisters’” (Vitrac, 2000, p. 100; cf. Knorr, 1989, pp. 157-59).

Comments of Eutocius on the compounding of ratios provides the groundwork for illustrating the fundamental problem that arises in this context. Namely, that in Greek arithmetic a number is considered as a multitude of indivisible units (what we now call a natural number); consequently the size (pēlikotēs) of the ratio of two numbers or two commensurable magnitudes cannot, strictly speaking, be considered as a number, unless the antecedent of the ratio is a multiple of the consequent. Therefore, if one wants to consider as a number the ratio of any two numbers or commensurable magnitudes, it will then be necessary to consider a divisible unit, as in Greek logistic, which deals with concrete rather than theoretical units. In the first case, one will end up with what is called a whole or natural number, and in the second case with a fractional number. And in case one wants to consider as number the size of the ratio between two incommensurable magnitudes, one will end up with what is called an irrational number.

Now Khayyam sets out to prove the preceding statement on the compounding of ratios in the general case, that is, for any three magnitudes; and it is precisely in this context that he engages in a detailed study of the quantitative nature of ratio and brings forward the concept of irrational number.

For Khayyam, every ratio expresses a measure; that is, a certain magnitude is assumed as unit, and the other magnitudes of the same kind are related to it. For example, the meaning of “the ratio of three to five” is “three-fifths of a unit.” In case there be given a ratio between two magnitudes A and B, he then considers the magnitude G such that its ratio to the unit is the same as the ratio of A to B. It is this magnitude G that will then express the measure (i.e., the size) of the ratio of A to B. Khayyam explains: “As to studying whether the ratio between magnitudes includes number in its essence, or whether it is inseparable from number, or whether it is joined to number from outside its essence because of something else, or whether it is joined to number because of something inseparable from its essence without requiring an extrinsic judgment: this is a philosophical study to which the geometrician must by no means devote himself,” for this study is not incumbent upon the geometrician once he has “realized that a ratio between magnitudes is conjoined with something numerical or in the potentiality of number” (Youschkevitch, pp. 87-88; Rashed and Vahabzadeh, 2000, p. 251).

While proving the proposition in question, Khayyam goes back over these notions: “The magnitude G should not be regarded as being a line, or a surface, or a solid, or a time. On the contrary, it should be regarded as being abstracted in the intellect from these adjunct characters and as being attached to number: not as a true absolute number, for it may be that the ratio between A and B is not numerical, so that no two numbers can be found in accordance with their ratio” (Rashed and Vahabzadeh, 2000, p. 253). In this manner, he is able to reduce the compounding of ratios to the multiplication of the numbers that express their respective sizes. He also explains that the unit he considers is a divisible unit (in fact the unit considered by him, being a magnitude, is divisible ad infinitum); and it is only by assuming that a number such as 2 is composed of divisible units that one will be able to speak of “the irrational number ⎷2,” unlike the ancient Greeks, for whom a concept such as “the irrational number ⎷2” did not seem to have had any meaning; they would only speak in this case of the ratio of two incommensurable lines, namely the ratio of the diagonal of a square to its side.

This is how Omar Khayyam, by discussing the connection between the concept of ratio and the concept of number, and by raising explicitly the theoretical problems related thereto, made a decisive contribution both to the theoretical study of the concept of irrational number, and to the understanding of its status as a mathematical entity in its own right. For although Khayyam’s point of view (it is not up to the geometer to justify the connection of ratio and number once he has realized that such a connection exists) might seem mathematically defective, it corresponds in fact to the attitude ultimately adopted by mathematicians for many centuries. Such an attitude can be found, for example, in the beginning of Isaac Newton’s Universal Arithmetick, where he asserts without any kind of justification: “By Number we understand, not so much a Multitude of Unities, as the abstracted Ratio of anyQuantity, to another Quantity of the same Kind, which we take for Unity. And this is threefold; integer, fracted, and surd: An Integer, is what is measured by Unity; a Fraction, that which a submultiple Part of Unity measures; and a Surd, to which Unity is incommensurable” (Newton, p. 2; Youschkevitch, pp. 88-89).

(2) THE ESSAY ON THE DIVISION OF THE QUADRANT OF A CIRCLE

This essay has no title and is not dated; we only know that it was written prior to the treatise on algebra, since in the former, which deals only with one specific cubic equation, Khayyam alludes to the subject matter of the latter, namely, a full treatment of all cubic equations (for this essay, see Amir-Moez, 1961; Djebbar and Rashed).

The aim of this essay is to determine (Figure 4) on the quadrant AB of a given circle ABCD a point G, so that the radius AE is to the perpendicular GH as EH to HB. In order to achieve this, Khayyam uses the traditional method of analysis and synthesis: He first assumes that the problem has been solved, and then deduces certain properties that will enable him to construct the point G, which is looked for.

Figure 4. Diagram for Khayyam’s untitled essay on the division of the quadrant of a circle.

The first analysis leads to the determination of a rectangular hyperbola that passes through the center E of the circle; it is left unachieved because of its difficulty. In the second analysis, Khayyam assumes that the point G is known, and draws the tangent GI to the circle. He is thus led to the determination of the triangle EGI having a right angle at G.

After having examined certain properties of this triangle, he assumes that HG is a “thing,” that is, the unknown of an equation, also called “root” or “side,” and that EH is equal to 10. He is thus led to the resolution of the equation “a cube and two hundred things are equal to twenty squares and two thousand.” He then constructs the solution of this equation by means of a circle and a rectangular hyperbola. He is then able to construct the triangle EGI, and consequently the point G, which is looked for (Rashed and Vahabzadeh, 2000, pp. 97-107, 165-70, 174-79).

In the only known manuscript of this treatise (in the 1751 collection of Tehran University Library), Khayyam’s text is followed by a short problem that is attributed neither to Khayyam nor to anyone else. In it, the point G is determined at once as the intersection of the given circle and a rectangular hyperbola passing through point B, instead of point E as in Khayyam’s first analysis, and having as asymptotes CA produced and the perpendicular to CA drawn from point C.

Khayyam’s essay also contains an important digression on the basic concepts of algebra, and a classification of cubic equations. Khayyam first explains that what the algebraists call “squared-square,” “squared-cube,” “cubed-cube,” … (i.e., in modern notation, x⁴, x⁵, x⁶, …) cannot have any meaning in sensible things, so that these expressions should only be understood metaphorically. He then adds: “And as to things which are used by the algebraists, and which exist in sensible things and in continuous magnitudes, they are fourfold: number, thing, square and cube” (Rashed and Vahabzadeh, 2000, p. 171). He explains that number is something abstracted in the intellect from material things: It is a universal intelligible that cannot exist concretely unless associated with particular objects. As for the thing, its position in relation to magnitudes is that of the straight line. The square will of course be a square whose side is equal to the thing; and likewise the cube will be a cube whose side is equal to the thing (Rashed and Vahabzadeh, 2000, pp. 170-71).

Khayyam then gives a classification of cubic equations in which he follows the methodology inaugurated by Moḥammad b. Musā Ḵᵛārazmi. As is well known, Ḵᵛārazmi had written his treatise on algebra (al-Jabr wa’l-moqābala) during the caliphate of al-Maʾmun (r. 198-218/813-33). In it Ḵᵛārazmi had first introduced the basic notions used throughout his treatise, which he defined as the three kinds of numbers needed in algebraic calculations; these three kinds are “squares,” “roots” (which he also calls “things”), and “simple number” (i.e., ax², bx, and c respectively, where a, b, c are natural numbers or positive fractions). He then considered all the combinations between these three kinds, thus obtaining three equations between two terms (i.e., ax² = bx, ax² = c, bx = c), and three equations involving three terms (i.e., ax² bx = c, ax² c = bx, bx c = ax²). Ḵᵛārazmi explained how to solve each of these six equations once the number of the term of the highest degree had been reduced to one. He solved the equations between two terms through specific examples; but he gave the solution of those between three terms in the form of a general rule applicable to any equation of the same species, and justified each rule by means of a geometrical construction. He then applied these rules to the resolution of various sorts of problems, both theoretical and practical ( Ḵᵛārazmi, tr. Rosen, pp. 5-21, and passim).

Khayyam first recalls that the combinations between number, roots, and square yield six equations that the algebraists have already solved. He then considers all the combinations between number, roots, squares, and cube that yield third-degree equations. These equations are either simple or compound. Simple equations are those between two terms. Compound equations are those that involve more than two terms: they are either trinomial or quadrinomial. Khayyam is thus led to three simple equations (i.e., x³ = ax², x³ = bx, x3 = c), nine trinomial equations (i.e., x³ ax² = c, x³ ax² = bx, x³ c = bx, x³ c = ax², x³ bx = c, x³ bx = ax², ax² bx = x³, ax² c = x³, bx c = x³), and seven quadrinomial equations (i.e., x³ = ax² bx c, x³ bx c = ax², x³ ax² c = bx, x³ ax² bx = c, x³ ax² = bx c, x³ bx = ax² c, x³ c = ax² bx). Discarding those that can be reduced to an equation of a lesser degree, he ends up with fourteen cubic equations, all of which can only be worked out by means of conic sections (Rashed and Vahabzadeh, 2000, pp. 172-73).

He then informs us that nothing had come down to him from the ancients in relation to these fourteen cubic equations, and that Māhāni was the first who dealt with one of them. Māhāni was trying to solve the following lemma that Archimedes had used in Proposition II.4 of his treatise On the Sphere and the Cylinder: Given (Figure 5) two lines DB and BZ, where DB is twice BZ, and given a point T on BZ, to cut DB at a point X so that XZ is to TZ as the square on DB to the square on DX (Rashed and Vahabzadeh, 2000, pp. 173).

Figure 5. Diagram for Māhāni’s solution to Archimedes’ lemma.

Although Archimedes had promised to show later on how to determine point X, the solution of this problem was never found in any of his writings. Eutocius, in his commentary on this proposition, reproduces in full a text he found “in a certain old book,” and which could correspond to Archimedes’ solution: In it the problem is solved geometrically by means of a parabola and a rectangular hyperbola (Netz, 2004a, pp. 318-30; idem, 2004b, pp. 16-29).

Māhāni thought of analyzing this lemma by means of algebra, and he was thus led to the equation “a cube and a number are equal to squares.” He tried to solve it by means of conic sections, but was unable to find its solution; thus “he settled the matter by saying that it was impossible” (Rashed and Vahabzadeh, 2000, p. 173). Until Abu Jaʿfar Moḥammad Ḵāzen (d. between 350-60/961-71) finally solved it by means of conic sections. Then Abu Naṣr b. ʿErāq (10th-11th cent.) solved, also by means of conics, the equation “a cube and squares are equal to a number,” to which he was led by analyzing algebraically a lemma that Archimedes had admitted in order to determine the side of the regular heptagon inscribed in a circle. Abu’l-Jud Moḥammad b. al-Layṯ (10th-11th cent.) solved a particular case of the equation “squares are equal to a cube and roots and a number,” to which mathematicians were led by analyzing the following problem: To divide ten into two parts, so that the sum of their squares, added to the quotient of the division of the greater by the less, be seventy-two (Rashed and Vahabzadeh, 2000, pp. 173-74).

Thus, according to Khayyam’s testimony, there are three cubics, to which he also adds the equation “a cube is equal to a number,” that have already been solved “by our eminent predecessors” (Rashed and Vahabzadeh, 2000, p. 174). He ends his digression, adding that no one had discussed the remaining ten, nor given a classification of all cubics, and that he intends to compose a treatise that will include an exhaustive treatment of these (Rashed and Vahabzadeh, 2000, p. 174).

On the whole, it can be said that the main interest of this essay of Khayyam does not lie in Khayyam’s resolution of the specific problem of the division of the quadrant of a circle, since this problem can be solved at once by choosing the appropriate hyperbola, but in that it provides us with an insight into Khayyam’s methodology, and with important data relating to the history of cubic equations.

(3) THE TREATISE ON ALGEBRA

This is the Maqāla fi’l-jabr wa’l-moqābala (A treatise on algebra; lit. A treatise on restoration and comparison); one manuscript has instead the title Resāla fi’l-barāhin ʿalā masāʾel al-jabr wa’l-moqābala (A treatise on the demonstrations of the problems of algebra; Rashed and Vahabzadeh, 1999, p. 117; Woepcke, Ar. text, p. 1). In this undated treatise, Khayyam realizes the project already mentioned in his essay, that is, an exhaustive investigation of cubic equations. Apart from an introductory section, in which Khayyam takes up and expands the discussions already found in his essay, this treatise can be divided into three parts: the equations that can be solved by means of ruler and compass, that is, by means of Euclid’s Elements and Data; the equations that can only be solved by means of conic sections, that is, by means of Apollonius’s Conics; and the equations that involve the inverse of the unknown.

In the introduction to his treatise, Khayyam defines algebra as “a scientific art whose subject is absolute numbers and measurable magnitudes qua unknown but connected with something known which enables one to determine them” (Rashed and Vahabzadeh, 2000, pp. 112-13). In accordance with Aristotelian philosophy, what Khayyam here means by “absolute numbers” are natural numbers, that is, a discrete quantity; magnitudes are “a continuous quantity, of which there are four: the line, the surface, the solid, and time, as it is mentioned in a general way in the Categories [6, 4b20-25] and in detail in First Philosophy [Metaphysics, δ, 13, 1020a7-33]” (Rashed and Vahabzadeh, 2000, p. 113). Khayyam not only understands mathematical concepts in accordance with Aristotelian philosophy, he also insists on the fact that the proofs in his treatise are based essentially on the works of classical Greek geometers: “It must be realized that this treatise will not be understood except by someone who masters Euclid’s work on the Elements and his work on the Data, as well as two Books of Apollonius’s work on Conics; and that if someone is not well versed in any one of these three [works], he will in no way understand it” (Rashed and Vahabzadeh, 2000, p. 113; the same statement is reasserted on pp. 127, 142, 145).

As in his essay, Khayyam classifies the equations obtained by combining number, roots, squares, and cube. But he considers here all equations of the first, second, and third degree. He thus obtains six simple equations (i.e., c = x, c = x², c = x³, bx = x², bx = x³, ax² = x³), twelve trinomial equations (i.e., x² bx = c, x² c = bx, bx c = x², x³ ax² = bx, x³ bx = ax², x³ = bx ax², x³ bx = c, x³ c = bx, c bx = x³, x³ ax² = c, x³ c = ax², c ax² = x³), and seven quadrinomial equations (i.e., x³ ax² bx = c, x³ ax² c = bx, x³ bx c = ax², x³ = bx ax² c, x³ ax² = bx c, x³ bx = ax² c, x³ c = bx ax²); that is, a total of twenty-five equations.

The equations that can be solved by means of ruler and compass are the linear and quadratic equations, as well as the cubics that can be reduced to an equation of a lesser degree. The only linear equation is: “a number is equal to a root”; and its resolution is straightforward.

The resolution of quadratic equations is demonstrated both numerically and geometrically. The geometrical proof is achieved through the introduction of a unit length; this allows Khayyam to represent the terms of any quadratic equation by rectangular figures, so that the original algebraic equation translates into an equation between rectangles and squares, that is, between geometrical magnitudes; in that manner, Khayyam is able to apply the results established in Euclid’s Elements and Data (Rashed, 1997, p. 44).

For instance, the simple equation “a number is equal to a square” is solved in the following manner: the numerical solution is found by extracting the square root of the number. To solve the equation geometrically, Khayyam first assumes (Figure 6) that the straight line AC is equal to the unit, and draws AB equal to the given number and perpendicular to AC; the measure of the rectangle AD will then be the given number. It is thus required to construct a square E equal to the given rectangle AD, and this construction is shown in Proposition II.14 of the Elements. The side of the square E will then be given, as shown in Proposition 55 of the Data, and will be the geometrical solution of the equation.

Figure 6. Diagram for Khayyam’s treatise on algebra: finding a square equal to a rectangle.

Trinomial equations of the second degree are solved numerically as in Ḵᵛārazmi’s treatise, that is, by means of a general rule applicable to all equations of the same species. Khayyam solves these equations geometrically in the same manner as his predecessors, that is, analytically; but he also adds a synthetic proof.

Let us consider, for instance, the equation “a square and ten roots are equal to thirty-nine.” In order to find the numerical solution, he states the following rule: “Multiply half the number of the roots into itself, add the product to the number, and subtract from the root of the sum half the number of the roots. The remainder will then be the root of the square” (Rashed and Vahabzadeh, 2000, p. 120; cf. Ḵ˘ārazmi, tr., Rosen, p. 8). The fact that “number” here means “natural number” is clearly implied by the statement that follows: “Numerically, these two conditions are necessary: the first one of them, that the number of the roots be an even number, so that it may have a moiety; and the second, that the sum of the square of half the number of the roots and the number be a square number. For otherwise the problem would then be impossible numerically” (Rashed and Vahabzadeh, 2000, p. 120). In other words, here Khayyam discards both fractional and irrational numbers.

As for the geometrical solution, Khayyam provides three different proofs. The first proof is based on Proposition II.6 of the Elements and is virtually the same as the one produced by Ṯābet b. Qorra in his authentication of algebraic problems by means of geometrical proofs (Ṯābet’s proof in Rashed, 2009, pp. 160-65); the second proof reproduces that of Ḵᵛārazmi’s. Both proofs are analytical in that it is assumed in both of them that the side of the square being looked for is given, implying that this square has already been constructed. Khayyam produces a third synthetic proof. He supposes (Figure 7) that the line AB is equal to 10, and that the rectangle E is equal to 39. He then applies to AB a rectangle BD equal to E and exceeding AB by a square AD, as shown in Proposition VI.29 of the Elements. The side AC of the square will be given, as shown in Proposition 59 of the Data. Thus the line AC will be the root looked for.

Figure 7. Diagram for Khayyam’s third solution of equation “a square and ten roots are equal to thirty-nine.”

The equations that can only be solved by means of conics are the fourteen cubics that cannot be reduced to an equation of a lesser degree (see above). Khayyam informs us that neither he nor his predecessors were able to solve them numerically, adding that “possibly someone else will come to know it after us” (Rashed and Vahabzadeh, 2000, p. 114; the numerical rules for solving cubic equations were discovered in the 16th century by the Italian algebraists Scipione del Ferro and Niccolo Fontana Tartaglia). Khayyam solves these equations only geometrically, using the properties of conic sections. The construction of the solutions of these fourteen cubics constitutes the bulk of Khayyam’s treatise.

As was the case with quadratic equations, the construction of the solutions is achieved through the introduction of a unit length; this allowed Khayyam to represent each of the terms of a cubic equation by a rectangular parallelepiped, so that the original algebraic equation would translate into an equation between solids, that is, between geometrical magnitudes. This way, Khayyam was able to base his demonstrations on Euclid’s Elements and Data, and on Apollonius’s Conics.

Khayyam first proves three lemmas. The first lemma enables him to construct a cube equal to a given rectangular parallelepiped; the second and third lemmas are used each time Khayyam needs to represent the number in a cubic equation as a solid having either a given base or a given height. He is now able to solve geometrically the equation “a cube is equal to a number” (its numerical solution being the cube root of the number). He constructs (Figure 8) a rectangular parallelepiped ABCD whose base AC is the square of the unit and whose height BD is equal to the given number. It is thus required to construct a cube KHIL equal to ABCD.

Figure 8. Diagram for Khayyam’s construction of a cube equal to a parallelepiped.

He takes two lines (E and G) that are mean proportionals between AB, BD (i.e., between the unit and the given number) as shown in the first lemma, and proves that the cube KHIL whose side HI is equal to E will then be equal to ABCD, that is, to the given number. Therefore, the side HI will be the solution of the equation.

Each one of the remaining cubic equations is solved by means of two single-branch conics. Khayyam investigates in each case the number of points at which these conics intersect or touch (not considering their vertices): the equation will accordingly have either one or two solutions. In some cases, however, the conics neither intersect nor touch, and the equation is “impossible”: it cannot be solved geometrically (Rashed and Vahabzadeh, 2000, pp. 130-56).

In this part of his treatise, Khayyam only produces synthetic proofs, in which he displays a full mastery of classical Greek geometry. However, the solution of each equation was probably found by means of an analysis; Roshdi Rashed has reconstructed such an analysis for the equation “a cube and sides are equal to a number” (i.e., x3 bx = c) by reverting the order of Khayyam’s synthetic proof (Rashed, and Vahabzadeh, 2000, p. 37). It amounts to the following, when expressed in the mathematical language used by Khayyam: let AB be the side of a square MB equal to the number of the sides (i.e., to b). We construct (Figure 9) a rectangular parallelepiped whose base is MB and which is equal to the given number (i.e., to c), as shown in the second lemma; and let its height be BC.

Figure 9. Diagram 1 for Khayyam’s analysis of the equation “a cube and sides are equal to a number.”

Now we assume that the problem has been solved, and that BE is the side of the cube being looked for (i.e., x); and we complete the square EL. The parallelepiped EM will be equal to the sides (i.e., to bx), but the parallelepiped BN is equal to the number (i.e., to c). Therefore, the remaining parallelepiped EN will be equal to the cube of BE (i.e., c – bx = x³), since BE is by hypothesis the root of the equation. In other words, the solid whose base is the square of AB and whose height is EC is equal to the cube whose side is BE. Therefore their bases will be reciprocally proportional to their heights, and the square of AB is to the square of BE as BE is to CE.

Now (Figure 10) let ED be a line perpendicular to BC and such that AB is to BE as BE to ED. Therefore, AB will be to ED in the duplicate ratio of AB to BE, that is, as the square of AB to the square of BE. But the square of AB is to the square of BE as BE to EC. Therefore, AB is to ED as BE to EC; and alternately AB is to BE as ED to EC. But AB is to BE as BE to ED. Therefore, BE is to ED as ED to EC; therefore the square of ED is equal to the product of BE and EC. Consequently the point D is on a circle whose diameter is BC. Also since AB is to BE as BE to ED, the square of BE will be equal to the product of AB and ED. Therefore, the square of DG is equal to the product of AB and BG. Consequently, the point D is also on a parabola whose vertex is B, whose axis is BG, and whose erect side is AB.

Figure 10. Diagram 2 for Khayyam’s analysis of the equation “a cube and sides are equal to a number.”

Thus the analysis has led to the determination of the intersection of a semicircle and a parabola. In the synthesis, Khayyam constructs the parabola HBD and the semicircle BDC, draws from the point of intersection D the line DE perpendicular to BC, and then proves that BE is the side of the cube looked for. According to Rashed, the remaining cubics were solved using the same method (Rashed and Vahabzadeh, 2000, pp. 37, 130-32); therefore, the previous analysis probably gives us an insight into the very process that led Khayyam to the resolution of third-degree equations.

The last part of Khayyam’s treatise is devoted to equations involving the inverse of the unknown (i.e., x^-1). In order to solve an equation like that, Khayyam takes the inverse as a new unknown and is thus led to one of the twenty-five equations previously studied. He then finds the solution of the latter equation and, taking its inverse, obtains the solution of the original equation. In this part of his treatise, Khayyam departs from the Euclidean rigorous methodology exhibited in the resolution of quadratic and cubic equations; for he only solves here particular equations and does not produce any proofs. Besides, he does not restrict the number concept to natural numbers as before, and mentions explicitly fractional numbers (Rashed and Vahabzadeh, 2000, pp. 156-59). Thus the fact that Khayyam had discarded fractions in the preceding parts of his treatise appears to be deliberate, and must have been due to his desire to follow, when possible, the Euclidean conception of number, that is, a multitude composed of indivisible units. But this would have been virtually impossible when dealing with equations involving the inverse of the unknown.

We have already noted that one of the most striking features of Khayyam’s treatise on algebra is the geometrical nature of its argumentation. Of course, the twenty-five equations he intends to solve are in themselves algebraic concepts, and his classification could have hardly been conceived without the previous work of Ḵᵛārazmi; but once translated into a relation between geometrical figures, these equations are dealt with in a purely Euclidean manner. Also Khayyam constantly speaks of the product of a rectangle and a straight line, where Euclid would speak of the parallelepipedal solid with the rectangle as base and the straight line as height (e.g., Rashed and Vahabzadeh, 2000, pp. 119, 125-26; Heath, III, pp. 345-47); this terminology, however, is not altogether foreign to Greek geometry, for it is also found in Eutocius’s commentary on Proposition II.4 of On the Sphere and the Cylinder, and even in Archimedes’ alternate proof to Proposition II.8 of the same (Netz, 2004a, pp. 227-31, 320 ff.; idem, 2004b, pp. 97-120; see also 2004b, pp. 164-65). On the whole, it appears that in this treatise Khayyam made a deliberate return to the rigorous methods of Greek geometers, and that in this way he was able to build up the theory of quadratic and cubic equations on solid foundations.

(4) THE TREATISE ON THE EXTRACTION OF THE NTH ROOT OF NUMBERS

Apart from the preceding works, Khayyam also wrote an arithmetical work to which he alludes in his Treatise on algebra: “The Indians have methods for determining the sides of squares and cubes based on a restricted induction, that is, on the knowledge of the squares of the nine figures—I mean the square of the unit, of two, of three, and so on—and likewise of their product one into the other—I mean the product of two into three, and so on. And we have written a book to demonstrate the correctness of these methods and the fact that they fulfill the requirements; and we have increased the kinds thereof, I mean the determination of the sides of the squared-square, of the squared-cube, of the cubed-cube, whatever degree it may reach. And no one did it before us. But these demonstrations are only numerical demonstrations based on the arithmetical Books of the Elements” (Rashed and Vahabzadeh, 2000, pp. 116-17, with correction). This book has not come down to us and is known only through the preceding quotation. However, MS Or. 199 in the Leiden University Library (which also contains a copy of the Commentary on Euclid’s Elements) lists on its title page, without including it, a work by Khayyam entitled Moškelāt al-ḥesāb (The difficulties of arithmetic). This work may correspond to his treatise on the extraction of nth roots (Rosenfeld and Youschkevitch, 1973, pp. 325-26; Youschkevitch, pp. 76, 80).

Bijan Vahabzadeh

KHAYYAM, OMAR xv. AS ASTRONOMER

Despite the intrinsic importance of the works of Omar Khayyam in philosophy and in particular mathematics, as well as the worldwide renown of his Rubaiyat, Khayyam certainly owes much of his reputation among Iranians to his role in establishing the Jalāli calendar (see CALENDAR ii. IN THE ISLAMIC PERIOD), and it is surprising that there is not much information in early sources about Khayyam’s role in this regard.

The inconvenience of the Arabian novilunar year (12 lunar months and without any intercalation, i.e., the insertion of an additional/13th lunar month into the ordinary/normal year) or the Persian vague year (the year in the modern Zoroastrian calendar, 365 days without any intercalation, i.e., the insertion of an additional/13th “30 day month” each 120-124 years; see Taqizadeh, 1940, p. 108) for practical purposes in the life of the individual and still more in social life (especially the time for collecting land taxes, ḵarāj) led the Saljuq sultan Malekšāh (q.v.; r. 1072-92) and his vizier Neẓām-al-Molk Ṭusi (q.v.; 1018-92) to look for a solution. They assembled a number of the most prominent astronomers of their time to make observations of the sun and determine the time of the vernal (spring) equinox and, based on their observations, to establish a new time-reckoning (Ger. Zeitrechnung; Pers. gāh-šomārī, see Taqizadeh, 1940, p. 107) in which the beginning of the year will always be in sync with a displacement of less than twelve hours from the vernal equinox.

According to a number of astronomers and historians, including ʿAbd-al-Raḥmān Ḵāzeni (q.v., late 11th-mid 12th centuries), a well-known astronomer and physicist and a contemporary of Khayyam, in his al-Zij al-moʿtabar al-sanjari (lit. “the reliable astronomical tables dedicated to [Sultan] Sanjar”; fols. 105r, 122v and also in an extract of this Zij called Wajiz al-zij) and Ebn al-Aṯir (q.v.; 1160-1233; X, p. 97; Abu’l-Fedāʾ, IV, p. 101; see also Sayılı, pp. 161, 164), the decree to amend the calendar and to summon the astronomers was issued in 467/1075.

Ḵāzeni (regarded by Moḥiṭ Ṭabāṭabāʾi, pp. 689-92, as the founder of the Jalāli calendar), whose astronomical table (Zij) is our oldest source about the Jalāli calendar, in his more or less brief references to the subject, failed to mention any of those who were involved in this project or the place the observations were made. Moḥammad b. Ayyub Ḥāseb Ṭabari (d. after 1092) in his Zij-e mofrad considers the place of observations to be Isfahan (pp. 71-72).

Ẓahir-al-Din Bayhaqi (q.v.; 1097-1169), in his Tatemmat Ṣewān al-ḥekma, along with the biography of Moḥammad b. Aḥmad Maʿmuri, has a brief mention of the observations in Isfahan. According to him, Maʿmuri—whose expertise in physics and the construction of mechanical devices (ḥeyal) was known to Khayyam—went to Isfahan to participate in the observations decreed by Malekšāh (Bayhaqi, p. 163; Pers. tr. pp. 95-96; Eng. tr. p. 198). With the exception of this, Bayhaqi makes no mention of the observations that took place in Malekšāh’s time or Khayyam’s participation in these observations or amending of the calendar. But considering what is said about Maʿmuri and comparing it with other sources, it can be assumed that he traveled to Isfahan at Khayyam’s suggestion in order to build the required instruments for observations.

A number of contemporary researchers, regardless of the sources mentioned, have determined that the observations were carried out in Nishapur, Ray, and even Marv (Suter, p. 113; Sarton, I, p. 760; Sayılı, pp. 162–64). But in addition to what has been said, the phrase contained in the Nowruz-nāma, a Persian treatise attributed to Khayyam, about Malekšāh’s decree summoning the contemporaneous learned authorities from Khorasan (p. 12, “befarmud tā … ḥokamāʾ-e ʿaṣr az Ḵorāsān biāvarand”) also indicates that these observations were carried out outside of greater Khorasan.

Ebn al-Aṯir, when describing the notable events of 467/1075 in his al-Kāmel fi’l-taʾriḵ (completed in 628/1231), presented “the decision to the establishment of the Jalāli calendar” and “the observations that took place during Malekšāh reign” in two consecutive but presumably separate reports. In the first report, there is no mention of those involved in the project to amend the calendar: “In this year, Neẓām-al-Molk and sultan Malekšāh summoned a number of the most prominent astronomers, and they placed Nowruz [q.v.] on the first point of Aries [Ḥamal; i.e., the vernal equinox]. Before that, Nowruz coincided with the arrival of the sun to the mid-point of Pisces [Ḥut; i.e., the first day of the Yazdegerdi year, falling in the middle of Esfand, the twelfth month of the solar Hejri (Hejri-ešamsi) calendar, the current official calendar in Iran, equivalent to the first ten days of March in the Gregorian calendar]. The sultan made what they had determined the basis for the calendars” (Ebn al-Aṯir, X, p. 97).

In the second report, Ebn al-Aṯir (X, pp. 97-98) states: “And also in the same year, the astronomical observations for Malekšāh were carried out. A number of the most prominent astronomers congregated to conduct them, including ʿOmar b. Ebrāhim al-Ḵayyāmi [Khayyam], Abu’l-Moẓaffar Asfezāri [Abu Ḥātem Asfezāri; q.v.], Maymun b. al-Najib al-Wāseṭi, and others. The sultan spent a great amount of money on this project and the observations continued until Malekšāh’s death in 485/1092 and were abandoned after his death.” However, it is implausible to suggest that the leading astronomers who assembled to participate in astronomical observations were not the selfsame ones summoned to amend the calendar, both by the decree of Malekšāh, especially given the level of astronomical expertise required to determine the time of the vernal equinox. (Cf. Abu’l-Fedāʾ, IV, 101, who refers to Asfezāri as Abu’l Moẓaffar Esfarāʾeni; Neẓāmi ʿArużi in the Čahār maqāla [q.v.] claims to have met Abu Ḥātem Asfezāri in the company of Omar Khayyam at Balḵ in 506/1112-13 [p. 100; Eng. tr. Browne, p. 71, 137]).

Zakariyāʾ b. Moḥammad Qazvini (q.v.; 1203-83) in the “fourth climate” of his Āṯāral-belād (q.v.), when he mentions Khayyam among the prominent figures of Nishapur, makes the following remarks: “In the reign of the Saljuq sultan Malekšāh, a great amount of wealth was given to him [Khayyam] so that he might procure astronomical instruments for observing the stars, but with the death of the sultan the task was abandoned” (II, p. 318).

Qoṭb-al-Din Širāzi (q.v.; 1236-1311) in his three astronomical works is the first to give a great insight into the role of Khayyam in establishing the Jalāli calendar: “As for the Tāriḵ-e maleki [i.e., Jalāli calendar and era], it is associated with the Saljuq sultan Jalāl-al-Dawla Malekšāh b. Alp Arsalān, which came about when a number of sages [ḥokamāʾ], such as ʿOmar Ḵayyām and Ḥakim Abu’l-ʿAbbās Lokari and six others gathered in his court, and founded a calendar that began [its year] with the sun reaching the [first point of] Aries. And the first day of the [Jalāli] year is the first day in which, at solar noon [or high noon, i.e., the moment when the sun contacts the observer’s meridian], the sun is in Aries. And this day is called Royal Nowruz” (nowruz-e solṭāni; Qoṭb-al-Din Širāzi, Eḵtiārāt-e moẓaffari, 3rd maqāla, 11th bāb, fol. 149 v; al-Toḥfa al-šāhiya, 3rd bāb, 11th faṣl, fol. 131r; Nehāyat al-edrāk, 3rd maqāla, 10th bāb, fols. 154v-155r).

Mollā Moẓaffar Gonābādi (d. 1622) presents a different account that, unlike the three reports by Qoṭb-al-Din, refers by name to Ḵāzeni as one of those who collaborated with Khayyam in establishing this calendar: “It is reported that in the reign of sultan Malekšāh there were a number of sages such as Omar Khayyam and Ḵᵛāja ʿAbd-al-Raḥmān Ḵāzeni who were instructed by the sultan to make observations in his name. They consulted with each other about this extremely difficult task and after more discussions, they preferred the easier [task] to the difficult one and informed the sultan that ‘it would take at least thirty years to complete the observations …. Then it would be proper for us … to make for the sultan a calendar in which the [new] year would always begin at the same time and not change place in the course of time; which would as a result forever keep alive the name of the king’. Upon winning the sultan’s consent, they constructed the aforementioned calendar in sync with the ‘true solar year’ [i.e., tropical year] in the sultan’s name …” (pp. 21-22).

It seems that as with other observation activities of the Islamic era, a zij was compiled as a result of the observations decreed by Malekšāh. This point is likely to be an origin of the scattered references to a zij attributed to Khayyam (Ḥāji Ḵalifa, III, p. 570, called it Zij-e malekšāhi). Qoṭb-al-Din, after the aforementioned report, notes that the difference between the true solar year and “365 days” [i.e., the length of normal year], is less than a quarter of a day (according to present-day calculations, it is equal to 0.24222 days, i.e., 5 hours, 48 minutes, and 46 seconds). So the intercalary should be less than one day every four years, and after every six or seven “four years intercalation,” a “five years intercalation” should be considered. And then he mentions, according to his claim, the mistake that occurred in Khayyam’s Zij in this case: “… and from what we said, ʿOmar Ḵayyām’s mistake in his Zij will appear, where he stated that all intercalations were in the fourth years, but still [the first day of new year] is in sync with the sun reaching the first point of Aries, which is a great error indeed caused by his lack of attention to this detail to which we have brought your attention” (Qoṭb-al-Din Širāzi, Eḵtiārāt-e moẓaffari, fol. 150r; al-Toḥfa al-šāhiya, fols. 131r-131v; Nehāyat-al-edrāk, fol. 155r). Such an error, however, seems highly unlikely from a person who, according to Qoṭb al-Din Širāzi himself, was one of the founders of the Jalāli calendar.

Apart from this, Qoṭb al-Din refers to another work by Khayyam, which is a supplementary “chapter” (faṣl) that Khayyam has appended (elḥāqkard-e ast) to Abu ʿAli Ḥasan b. al-Hayṯam’s (965-1040) Treatise on the Movement of Eltifāf (the movement, or rather change, in the obliquities; Ebn al-Hayṯam’s treatise is not known to have survived). In Eḵtiyārāt-e moẓaffari (2nd maqāla, 10th bāb, fol. 92r), Qoṭb al-Din has only noted this point and, for the text of Khayyam’s chapter, referred the reader to his other work, Nehāyat al-edrāk fi derāyat al-aflāk (2nd maqāla, 10th bāb, fols. 93r-94v). The topic of this chapter was to resolve one of the problems in the Ptolemaic theory of planetary motion. It can thus be considered part of the history of the non-Ptolemaic models in the Islamic world.

Younes Karamati

KHAYYAM, OMAR xvi. AS PHILOSOPHER

See Supplement.