Arabic Literature in the Post-Classical Period

The book known in English as The Arabian Nights’ Entertainments or A Thousand and One Nights bears the imprint of many different times, places and individuals. The history of its transmission, translation, expurgation and falsification is nearly as fabulous as the tales told by its most famous character, Shahrazād (Scheherezade). The oldest evidence for the work’s existence, curiously enough, only came to light relatively recently. In 1948, Nabia Abbott, the first female faculty member of the Oriental Institute in Chicago, was examining a rare piece of early medieval paper from Syria when she suddenly realized that the text she was reading was familiar. Writing in no less than six different hands covered every available space on both sides of the sheet of paper: the draft of a personal letter, a legal attestation to a contract, a crude drawing of a human figure, a few scattered phrases scribbled in the margins and the now famous passage from A Thousand and One Nights. Abbott’s painstaking analysis led not only to the deciphering of all of these texts but also to a rather precise dating for the fragment to the early ninth century.¹ The short passage from the Nights that she had discovered proved to be over 1,100 years old – the earliest physical evidence of Shahrazād’s literary existence.

Other than Abbott’s fragment, the oldest pieces of historical evidence are found in two tenth-century Arabic texts. The Baghdadi bookseller Ibn al-Nadīm (d. between 990 and 998) offers an account of the Nights and how it first appeared in Arabic literature in his Fihrist (Catalogue of Books) in the section dealing with ‘Story-tellers and Raconteurs’ (al-musāmirūn wa᾽l-mukharrifūn). He states that the first people to collect and preserve fictional stories (khurāfāt) in books were the ancient Persians and that many of these collections were translated into Arabic and then refined and embellished by later literary figures. He further notes that the first book of this sort ever written was a collection known in Persian as the Hazār afsān (A Thousand Stories), and proceeds to give a précis of the famous frame-story of the Nights, complete with an account of the king who married a new wife each day and had her killed on the following morning and how the fascinating tales of Shārāzād (later Shahrazād) delayed her fate for a thousand nights at which time the king decided to spare her life. After describing the translation of the Nights into Arabic, Ibn al-Nadīm proceeds to give a lengthy list of other story collections and individual tales translated into Arabic from Persian, Indian and Greek sources, including another work in Persian entitled Hazār dastān (A Thousand Tales) and a possible reference to a Byzantine version of the Shahrazād frame-tale.

The second reference, found in the work Murūj al-dhahab (Meadows of Gold) by al-Mas῾ūdī (d. 956) confirms this account, but adds that the work was commonly known in Arabic as Alf layla (A Thousand Nights) and indeed this is the title found on Nabia Abbott’s fragment described above. The earliest known citations of the expanded title, Alf layla wa-layla (A Thousand and One Nights), both date to the twelfth century and are found in a historical work by al-Qurṭī and a document from the Cairo Geniza.²

Although the Nights was clearly translated from Persian into Arabic, some scholars have looked further, towards ancient Indian literature, for the origins of the work. Two literary devices found in the Nights are, in particular, quite common to older Indian literature: (a) the framing device of placing stories within stories by having a character in one tale narrate another tale; and (b) the use of talking animals as the protagonists of didactic fables. D. B. Macdonald pointed out that the frame-tale and the opening three short fables all have clear parallels in Indian texts.³ No physical evidence of the Persian Hazār afsān has survived, however, and earlier Indian sources can only be a matter for conjecture.

Once translated into Arabic, as Ibn al-Nadīm notes, men of Arabic letters did indeed begin to make emendations of their own. Some of the earlier Persian tales may have survived within the Arabic tradition altered such that Arabic Muslim names and new locations were substituted for pre-Islamic Persian ones, but it is also clear that whole cycles of Arabic tales were eventually added to the collection and apparently displaced most of the Persian materials. One such cycle of Arabic tales centres around a small group of historical figures from ninth-century Baghdad, including the caliph Hārūn al-Rashīd (d. 809), his vizier Ja῾far al-Barmakī (d. 803) and the licentious poet Abū Nuwās (d. c. 813). Another cluster is a body of stories from late medieval Cairo in which are mentioned persons and places that date to as late as the thirteenth and fourteenth centuries.

Little is known about how and where the Nights circulated in the Arabic-speaking world between the time of the early medieval references cited above by Ibn al-Nadīm, al-Mas῾ūdī and al-Qurṭī and the arrival of the Nights in Europe in the early eighteenth century. The few extant manuscripts of the Nights from this period, however, point to a work of a rather different nature than that of modern recensions and raise a number of significant questions. First of all, were there ever really a thousand and one nights? At the end of his entry on the Hazār afsān, Ibn al-Nadīm states that he saw copies of the complete work a number of times and notes that, although the narrative was indeed spread over a thousand nights, the collection included only about 200 stories as most of the stories were recounted over more than one night. It is not clear whether he is referring at this point to the Arabic or Persian text, but, in the context of his description, the latter seems more likely. No pre-eighteenth-century Arabic text of anything near this size has survived. In fact, the extant manuscripts from before the eighteenth century for the most part fall into two basic groups: Syrian versions that contain 282 nights and the medieval Egyptian versions containing only 200. No references later than that of Ibn al-Nadīm indicate that the Arabic text ever included a full complement of one thousand nights.

In addition, it appears that the Nights was neither a highly regarded nor even a particularly popular work during these centuries. Medieval Arabic literature includes a number of lists of popular and fictional works located in a wide variety of sources: sermons decrying wasteful pastimes such as reading fictional and fantastic tales; autobiographies in which the authors speak of their childhood fascination with such texts; Koranic commentaries that explain which texts are to be considered laghw (nonsense) and lahw (foolish diversions) and so forth. Together, these sources, along with the Fihrist of Ibn al-Nadīm, offer a rather detailed, albeit incomplete, portrayal of Arabic popular literature. Although the tales of Sindbad, the ‘Epic of ῾Antar’ and many other works of romance, adventure and marvellous voyages are cited over and over again, the Nights appear nowhere in these diverse sources. Confirming this lack of prominence is the notably small number of extant Arabic manuscripts of the Nights; every European scholar or translator who worked on the Nights before the widespread availability of the nineteenth-century printed editions complained bitterly about the difficulty of obtaining a copy of the work.

The pre-eighteenth-century Arabic manuscripts that do survive indicate that the collection was never entirely fixed in either size or content, but rather that medieval redactors continued compiling their versions (much as Ibn al-Nadīm had indicated already in the tenth century) by embellishing, substituting, emending and adding to the material that came down to them. Nevertheless, a core set of tales, including the famous frame-tale of Shahrazād, is found in the majority of the Arabic manuscripts of the Nights; many of these tales, however, are also found in separate analogue forms in unrelated manuscripts, either alone or in compiled collections.⁴ In fact, a number of the tales found in the Nights circulated more widely independent of the Nights than they did as part of that work.

Thus an accurate portrayal of the Nights in pre-modern Arabic literature must situate it within the rather fluid stratum of popular fictional literary production and as a work that was in many ways undifferentiated from a large body of similar texts. Many of these popular narratives and collections were compiled and copied not for personal, silent reading, but rather to function as the basis for public story-telling performances. Though it is nearly impossible to determine at this late date the degree to which the tales in the manuscripts may have been drawn from oral tradition, it is clear that these written texts often acted as scripts for professional oral performances. Antoine Galland, the first translator of the Nights into a Western language, made numerous references to such public performances of popular tales (though notably not of the Nights themselves) in his travel journals and even described bookstalls in Constantinople whose sole commerce was renting copies of such tale collections to public story-tellers for a small fee.⁵ Scholarly descriptions of story-tellers performing from written texts include accounts of public recitations in nineteenth-century Cairo, in twentieth-century Iranian coffeehouses and recent open-air performances in Marrakesh, Morocco;⁶ less detailed reports also abound in European travellers’ accounts of the Middle East. It would therefore be singularly inappropriate to approach a work drawn from this domain of popular medieval literature with preconceptions derived from the modern Western idea of a bounded, cohesive ‘text’ produced to be silently consumed by solitary readers.

The core cycle of tales which occupy the first 200 to 282 nights of the Syrian and Egyptian manuscript groups are organized in a manner that has sparked interest and awe among readers and literary scholars from many different cultures. Frame-tales are a device found in texts from a relatively large number of literary traditions: a story-teller or a story-telling scene is presented to the reader and within that frame a series of independent tales are told. Boccaccio’s Decameron, Chaucer’s Canterbury Tales, The Golden Ass, the Arabic Tales of Kalīla and Dimna and its Sanskrit predecessor, the Panchatantra, are but the most famous of many possible examples. In the Nights, however, this device is conspicuously exploited not only to produce a bewildering array of tales within tales, but also to provide, as many scholars have noted, a fascinating series of echoes and reflections upon certain themes, chief among them the nature of relations between women and men, the vicissitudes of fate, and the power of story-telling itself to control, pacify and educate.⁷

The embedding of tales within tales reaches one of its apogees in a series of tales that culminate in the 170th night of the Egyptian manuscript tradition at which point five independent acts of narration are taking place simultaneously in a sequence of twelve interrelated tales. At the outermost level, the narrator is recounting to us the readers the story of King Shahriyār, within which Shahrazād is narrating her tales. On the 102nd night she begins the ‘Tale of the Hunchback’, set in China, in which the dead body of a hunchback is passed from a tailor and his wife first to a Jewish physician, then to a Muslim steward and finally to Christian broker. At the end of the tale, when the strange incident comes to light and is recounted to the King of China, he asks – rhetorically it would seem at first – whether anyone has ever heard a stranger or more marvellous tale! All four of the main characters, however, immediately reply that indeed they have, which leads to a series of four new tales told by the protagonists of the ‘Tale of the Hunchback’. In the last of these – the ‘Tale of the Lame Young Man from Baghdad’ – narrated by the tailor, the young man of the title recounts the story of his encounter with a barber to an assembly of dinner guests, and the barber, also present, then recounts his own tale, within which he narrates to the Caliph of Baghdad not only one of his own misadventures, but also the stories of each of his six brothers. The narrator is speaking to us, Shahrazād to King Shahriyār, the tailor to the King of China, and the barber to the assembled dinner guests about the tales he told the Caliph in Baghdad. Needless to say, this places the reader in handsome company indeed.

The multiple levels of embedding and the twelve larger stories that form the sequence (several of which include shorter narratives as well) are bound together by a series of motifs that echo back and forth through the narratives. The sequence begins and ends with the Hunchback, and each of the tales within the sequence involves the acquisition of a physical defect: the loss of a hand, a pair of thumbs, an eye, loss of speech, being made lame or paraplegic, or having the ears or lips cropped. Each of the four tales told by the tailor, steward, physician and broker, for example, involves a handsome young man of means who falls in love with a beautiful maiden and then suffers an amputation or maiming for his love. The first young man has a hand cut off; though he eventually wins his love, she dies soon afterwards and his happiness is of short duration. The second has his thumbs and big toes cut off, but he is able to marry his love, acquires great wealth, and they live together happily for many years. The third has a hand cut off and eventually is married to the sister of his love since his original lover turned out to be an evil-hearted murderess. In the final story, the young man is made lame, but does not win his love, nor receive any form of compensation. Four different fates for four similar loves. Perhaps, tellingly, the first three young men achieve their heart’s desire by recounting their tales immediately after their maiming to a king or other powerful figure and only the fourth, who does not tell his tale until much later, receives no compensation. The motif of physical defects is carried forward through the remainder of the tales until the very end of the sequence when it turns out that the Hunchback is not in fact dead and is resuscitated by the Jewish physician, a final echo of the twinned motifs of physical defect and compensation.

The literary style of the Nights has received a variety of assessments from critics and translators over the centuries ranging from complete condemnation to fulsome praise. It is a style heavily influenced by colloquial Arabic and is quite distinct from the norms of high classical Arabic literature. The Arabic manuscripts are filled with ‘misspellings’ that reflect colloquial pronunciations and usages, and they contain a basic vocabulary that eschews erudite vocabulary and opts instead for more commonly used words except in some of the flowery descriptions of clothing, banquets and so forth, and in the – generally mediocre – poetry. The language is at times prodigiously repetitive and at times vulgar. In essence, it is quite simply the language of popular literature which, if not oral in provenance, has been composed in a manner that clearly reflects the milieu of oral story-telling. For non-Arabic speakers a comparable example in English might be works of nineteenth-century popular fiction, such as those by Mark Twain, in which the dialogue is liberally sprinkled with spellings such as ‘gonna’ (going to), ‘hafta’ (have to), ‘betcha’ (bet you) and so forth, a style that might strike some, particularly readers familiar with the colloquial in question, as free-flowing and natural, and strike others, particularly defenders of ‘good taste’, as uneducated and graceless.

The ongoing controversy in Arab society over writing the colloquial language has even led some Western scholars to claim that the Nights is actually improved by translation since in a foreign tongue the tales are freed from this polemic. Whether that be the case or not, each of the translators who attempted to render the Arabic of the Nights into a foreign language was faced with a critical dilemma at the outset: to translate the Nights in a style that reflected its original, at times awkwardly colloquial, language, or instead to couch it in a more respected literary idiom for its new readership. Their various decisions produced astonishingly different results.

The Nights would no doubt have continued to exist only in semi-obscurity had it not been for the intervention of one man: Antoine Galland. Born to a family of modest financial means in Picardy, France, and orphaned at an early age, Galland showed a strong aptitude for languages and studied Latin, Hebrew and Greek before moving to Paris in 1661 to seek his fortune. He studied at the Collège Duplessis, which was incorporated into the Sorbonne in 1646, and then at the Collège Royal where he first studied Arabic. In 1670, he was offered employment as secretary to the Marquis de Nointel who had been appointed royal ambassador for Louis ⅪⅤ to Constantinople. Galland eventually lived in Constantinople for fifteen years during three different sojourns (1670–5, 1676–7, 1680–8) and also travelled to Smyrna, Aleppo, Beirut, Damascus, Jerusalem, Gaza and Cairo. Throughout much of this time he worked as a translator and ‘antiquary to the king’ for whom he purchased manuscripts, coins, medallions and other curiosities. He acquired a great reputation as a bibliophile, numismatist and scholar of eastern languages. He also maintained a diary for most of the fifteen years he lived abroad which continues to provide scholars with striking insights into both French and Ottoman culture of the period. Yet, during the entire fifteen years he spent purchasing books and manuscripts in the Middle East, Galland never once encountered nor even heard mention of A Thousand and One Nights; this work entered his life only later, once he had returned to France, and in a rather mysterious manner.

Having returned to France, Galland aspired to an appointment as professor of Arabic, an honour he would only finally achieve in 1709 at the age of sixty-three in the Collège Royal. In the intervening twenty-one years, he accepted a series of appointments as translator, research assistant and curator in the service of various scholars and noblemen. From 1696 to 1708, he worked as the curator for Nicolas-Joseph Foucault’s private collection of books and orientalia in Caen. Although his relationship with his patron was satisfactory, the position was in Normandy, far from the academic centres of Paris. Here, in the relative isolation of the provinces, Galland began to translate a manuscript of the Voyages of Sindbad which he was dedicating to the Marquise d’O, daughter of the French ambassador in Constantinople during his third voyage to Turkey and a former pupil of his. In the épître published in the first volume of his Mille et une Nuit in 1704, in which he apologizes for the tardy publication of the volume, he gives this account of a remarkable discovery:

The delay, Madame, comes from the fact that having begun the publication [of Sindbad], I discovered that these Tales are drawn from a prodigious collection of similar Tales, in several volumes, entitled the Thousand and One Nights. This discovery compelled me to suspend the publication [of Sindbad] and to dedicate my attentions to acquiring that collection. It was necessary to have it brought from Syria and then to translate into French the first volume, which you have here before you, of only four that have been sent to me.

How Galland managed to discover a reference to the Nights, which had escaped his notice for fifteen years in the Middle East, while working in a small town in Normandy has remained a mystery. Even more curious is the fact that he was mistaken – for the voyages of Sindbad had never in fact formed part of the Nights. An additional mystery surrounds his references to the Arabic manuscript, for in his personal correspondence he refers once to three volumes, another time to five volumes, in the work itself he refers to four volumes, and when he died only three volumes were found in his possession. Nearly three hundred years later scholars are still in disagreement about the relationship between Galland’s translation and his Arabic text of the Nights. Some argue that the many divergences between the two texts are due to Galland’s free-handed translation style which allowed him to add details to the tales as he saw fit, while others argue that Galland must have possessed another text of the Nights that has been lost because some of his ‘additions’ simply make no sense within the context of the stories themselves and must have been suggested by an alternate source.

It was only two centuries later that MacDonald and Zotenberg were separately able to provide reasonably accurate accounts of how most of the stories in the Galland translation came to be included in the twelve-volume French ‘translation’.⁸ The first two volumes of Galland’s French text follow the Arabic manuscript closely. For volume three, however, Galland inserted the Sindbad tales he had translated from a separate source and which had never previously been part of the Nights. Volumes four to six return to the tales in the Arabic manuscript, although Galland for some reason skipped two tales and completed volume six with the tale ‘Qamar al-Zamān’ which was incomplete in the Arabic manuscript, but for which Galland supplied an ending either from his own imagination or from another manuscript. Finally, volume seven comprised the two tales which had been dropped earlier. All of these were published between 1704 and 1706, but with volume seven Galland had exhausted all of the material in his Arabic manuscript. Though he tried for years through a number of different intermediaries to obtain additional Arabic manuscripts of the Nights that contained more tales, his efforts were in vain.

When volume eight appeared three years later in 1709, it was a complete shock to Galland for he had not submitted any new tales to his publisher. The Nights was selling well and since Galland had refused to supply new tales, his publisher took matters into his own hands: volume eight included a tale entitled ‘Ghānim’ that Galland had previously translated or composed on his own (no Arabic manuscript or analogue text has ever been found), as well as two tales translated from Turkish by Galland’s colleague, Pétis de la Croix (‘Zayn al-Aṣnām’ and ‘Khudādād’), who had been working on a collection he entitled Mille et un jours drawn from a Turkish manuscript. Although both Galland and Pétis de la Croix denounced the volume, these tales have been retained in nearly all later versions of the Nights.

The remaining volumes of Galland’s Nights were drawn not from written sources, but, strangely enough, from tales told at Parisian dinner parties. While visiting his friend and fellow traveller Paul Lucas on 17 March 1709, Galland met a visiting Syrian Maronite priest from Aleppo named Ḥannā who proved to be a gifted story-teller. After hearing Ḥannā recount stories to his amused hosts, Galland wrote brief summaries of the tales in his diaries. At some point Galland made a conscious decision to complete his version of the Nights with tales collected from Ḥannā, for Galland noted in his journal that, at his request, Ḥannā provided a handwritten Arabic version for the tale of Aladdin; this transcription, however, was never found after Galland’s death. In order to publish the tales he had only heard Ḥannā recount orally, Galland reworked his own brief summaries of three to eight pages into stories that ranged from 100 to nearly 200 printed pages. Thus the tales recounted by Ḥannā form the final four volumes – a full third – of Galland’s Nights. Among these tales are found the stories and motifs that have most come to represent the Nights in the Western imagination: ‘Aladdin’ and ‘Ali Baba and the Forty Thieves’, as well as several of the most famous motifs, such as the magic lamp and the flying carpet, neither of which are found in earlier Arabic sources. As a culminating contribution, Galland provided the concluding thousand and first night to the collection, finishing off the entire collection, in proper French style, with a happy ending.

In hindsight, it is easy to see how those tales which Galland freely retold from the barest plotline recorded in his diary achieved far greater popularity than most of the tales which had been translated more or less directly from Arabic written sources. Galland recreated these tales as much as two and a half years after having heard Ḥannā tell them, which certainly allowed a great deal of leeway for his own creative impulses. As a result, he succeeded in creating ‘Arabic’ tales that were far more steeped in French structure and style than those translated from Arabic originals, a literary product perfectly crafted to sell to a French readership. His efforts inspired the emergence of an entirely new genre of European literature, the Oriental Tale, at which authors as diverse as Voltaire, Johnson, Scott, Beckford, Disraeli, Irving, Dickens, Dumas and Pushkin were later to try their hand.⁹

In the final count, only nine of the twenty-one stories in Galland’s Nights are drawn from Arabic manuscripts of Alf layla wa-layla. Galland, however, never informed his readers that the later volumes of his Mille et une nuit (Galland himself always used the singular in French – nuit – echoing the Arabic) were not drawn from Arabic manuscripts; in fact, he deliberately misled his audience by indirectly referring to the manuscripts in the introductions to the final volumes.

Why was the publication of the Nights such a resounding and instant success in French? Why did Galland grow so interested in the Nights in the first place and why did he choose to spend his time translating this of all texts? After all, he aspired to a professorship in Arabic and had hitherto published only scholarly works and translations. The answers to these questions lie in part in a rather unusual development that occurred in French fiction during the decade and a half just prior to Galland’s translation. In 1690, a new genre of fiction emerged and rapidly achieved great popularity among upper-class readers, particularly women: the contes des fées or fairy tales. The authors of this new genre were predominantly women of the lesser nobility and the tales were at first concocted at salons with the input of all present and subsequently written down by one of the participants almost as a type of party game. Later, individual authors emerged, the most successful of whom were Madame D’Aulnoy, Louise d’Aulneuil, Marie-Jeanne Lhéritier and Charles Perrault, better known for his Contes de ma mère l’Oye, the famous Mother Goose Tales.¹⁰

The new fairy tales were filled with magic and wit, including veiled references to members of the court. The authors in fact enjoyed referring to themselves publicly as fées (fairies) and wrote to each other with such titles as ‘Queen of the Fairies’. Unlike the earlier style of epics and lengthy romans, these tales often involved characters from different levels of society in rather commonplace settings. More intriguing is that they first took the form of story-telling events within longer narratives and the figure telling the story was almost always a woman. This structure was established in the very first example of the genre found in Histoire d’Hypolite published in 1690 by Madame D’Aulnoy (c. 1650–1705). Thus, in an astonishing historical coincidence, the most popular genre of French fiction just prior to Galland’s translation of the Nights roughly paralleled the structure of the Nights: tales within tales, told by a female narrator, filled with magic and marvellous events, involving a variety of social characters often in quotidien settings, replete with sumptuously detailed descriptions of banquets, palaces and treasures.

There is little doubt that when Galland first read the Nights he read them as the Arabic equivalent to the then extraordinarily popular French contes des fées (the contes were a quite lucrative publishing venture and 114 were published between 1690 and 1715). Indeed, it may well have been these striking similarities that motivated him to translate them at all and to dedicate them to the Marquise d’O, a patron of the same social class as the female authors of the most popular of the contes des fées. In a letter written to Pierre-Daniel Huet dated 25 February 1701, Galland wrote of his Nights: ‘I also have another little translation from Arabic, stories just as good as the fairy tales published these last years in such profusion’.¹¹ The Nights was translated not merely into French, but into a specific genre of French literature, the fairy tale, the existence of which is probably the single most important motivation for the Nights’ translation and its remarkable success in the West.

With its exotic locations and characters, the Nights offered a French reading public already accustomed to the basic features of the contes des fées a new twist on a familiar pattern. Towards the end of the eighteenth century when a massive compilation entitled the Cabinet des fées was published in forty-one volumes, the already close connection between the French fairy tale and the Nights was further strengthened: the series opened with a reprint of Galland’s Nights (volumes one to five) and concluded with a ‘continuation’ of the Nights entitled the Suites des mille et une nuits (volumes thirty-eight to forty-one), while in between appeared the fairy tales of Madame D’Aulnoy, Charles Perrault and their circles along with a variety of tales from other European sources. The European literary establishment of the day made no distinction between Galland’s translated stories, European fairy tales, and ‘Arabic’ and ‘Persian’ tales composed by European authors.

The success of Galland’s creation was unprecedented. His Mille et une nuit was translated into English, German, Italian, Dutch, Danish, Russian, Flemish and Yiddish by the end of the eighteenth century. During the same time period it was reissued in over twenty French editions, with an additional forty French editions appearing in the nineteenth century. Yet all of these texts were derived in one manner or another from Galland’s French text. For nearly a century there were no new translations from the original Arabic.

By the end of the eighteenth century, the great success of the Nights in Europe began to engender a variety of different continuations, retranslations and outright fakes. Of the many scholars who have worked to unravel the labyrinthine histories of the many Arabic manuscripts and translations that mysteriously cropped up in Europe from the late eighteenth to the late nineteenth century, three have provided detailed attempts at classification but have at times reached quite different conclusions.¹²

The Chavis-Cazotte continuation of the Nights involved a collaboration between a monk from Syria, Dom Denis Chavis (Dionysius Shāwīsh), who had been brought to France to work in the Bibliothèque du Roi, and the French author Jacques Cazotte. Chavis claimed to have in his possession a Baghdadi manuscript of the Nights, but, according to Mahdi, who has undertaken the most detailed examination of this text, Chavis began by copying Galland’s three-volume Arabic manuscript which was then present in the bibliothèque, in the hopes of selling this ‘new’ manuscript of the Nights.¹³ In the first volumes he made only a few changes (notably substituting another story where Galland had included the Sindbad cycle), but he quickly faced the puzzle of what to do with the Galland tales for which there were no Arabic texts: he chose to translate them from French into Arabic and thereby created what was for a time mistaken as the Arabic ‘original’ of Galland’s Nights. In addition to translating Galland’s tales from French into Arabic (including the two Persian tales that Galland’s publisher had added to the collection), he inserted other tales from an independent Arabic story collection, the manuscript for which still exists. This collection contains two tales which are also found in Arabic manuscripts of the Nights (‘Jullanar and the Sea’ and a section from ‘The Porter and the Three Ladies of Baghdad’) and twelve others, but this collection makes no claim to be connected in any way with the Nights.

It soon became apparent to Chavis, however, that there was little financial reward to be had from all this effort in creating a new Arabic manuscript, whereas there was a great deal of interest in publishing French translations of ‘new’ and ‘authentic’ tales from the Nights, especially ones not found in Galland. After being introduced to Cazotte by the publisher Paul Barde, Chavis abandoned his work on the Arabic manuscript and turned his efforts to the collaborative production of a French translation. The eventual result was the Continuation des mille et une nuits (published in the Cabinet des fées as the Suites) which contained a mishmash of tales Chavis had in his possession, tales which he apparently outlined to Cazotte and which Cazotte then freely retold, as well as tales that are almost entirely of Cazotte’s creation. The Continuation enjoyed two decades of success and was translated more than once into English, but was eventually decried as a fake in the first decades of the nineteenth century. The small collection of tales that Chavis had brought from Syria was later translated separately by Caussin de Perceval as a ‘supplement’ to an 1806 reprint of Galland.

Another counterfeit manuscript which was apparently created by Mīkhā’īl Ṣabbāgh, a Syrian Christian employed in the Bibliothèque du Roi a number of years after Chavis, experienced a much lengthier acceptance as an authentic copy of the Nights among Western scholars. Both Zotenberg and MacDonald deemed it reliable, while Mahdi has more recently rejected it as a clever forgery cobbled together from parts of Galland, Chavis and at least one other source, though this judgement is still open to debate.¹⁴ The Ṣabbāgh manuscript had a much more enduring impact on the history of the Nights than the Chavis-Cazotte collaboration since it greatly influenced two of the most commonly used Arabic printed editions of the Nights produced in the nineteenth century (see below).

Arabic manuscripts of both ‘Aladdin’ and ‘Ali Baba and the Forty Thieves’, two of the tales Galland had collected from Ḥannā and for which no Arabic source was known, also mysteriously surfaced in Europe in the early nineteenth century and provoked a great deal of scholarly discussion. They are for the most part now discredited as they appear to have been created from Galland’s French text; even if the isolated manuscripts (each contains only a single tale) should eventually be proven authentic, the tales would still have no known historical connection with the Nights.

Around the year 1775 a new redaction of the Nights appeared in Egypt that included the traditional core 200 nights of the earlier Egyptian recensions along with a large number of additional stories that filled out the collection to its ‘complete’ form of 1001 nights. This text, which spawned a large number of manuscript copies and eventually printed editions as well, is commonly referred to by Western scholars as Zotenberg’s Egyptian Recension (ZER) and represents a modern development that was almost certainly motivated in some manner by European requests for manuscripts of the ‘complete’ Nights for there is no evidence of an Arabic manuscript tradition containing more than 282 nights earlier than the ZER which only emerged seventy years after Galland’s translation. Precisely how the ZER came into being is not known, but numerous Western travellers, collectors and diplomats throughout the eighteenth century had purchased texts and commissioned copies of manuscripts of the Nights. The earliest historical documentation regarding the ZER is found in the diary of a German traveller, Ulrich Seetzen, and, although the information is somewhat garbled, a clear distinction is already drawn between the first 200 nights and the remaining tales which are said to have been drawn from other sources.¹⁵ In any case, the ZER later became the base text for both the Bulaq and the Calcutta Ⅱ printed editions.

The earliest Arabic printed edition of the Nights, known as Calcutta Ⅰ, was created at the College of Fort William that had been established by the British East India Company in 1800 to train employees and officials in Indian languages and cultures. The printed edition was designed as a textbook for the teaching of Arabic and was published in two volumes, each containing exactly 100 nights. The main text was drawn from a manuscript of the same Syrian grouping as Galland’s, and was similar to it in most respects. In order that each volume conclude on the proper number of nights, however, the editor added the tale ‘al-Ma’mūn and Būrān’ to the end of volume one and ‘The Guiles of Women’ to volume two, both of which were drawn from a work published by Louis-Mathieu Langlès in 1814 in Paris. In a rather peculiar development, although he had already reached the 200th night, the editor also added the Sindbad cycle, without night divisions, which thus forms the final section of volume two.

The Habicht or Breslau text is further removed from the Arabic manuscript tradition than the three other major printed editions and is often deemed an outright fabrication. Christian Maximilian Habicht, born in Breslau in 1775, studied Arabic in Paris where he met and later collaborated with a Jewish Tunisian, Murād ibn al-Najjār. By compiling tales from texts of the Nights and freely adding stories from other sources, Habicht created a German translation containing 1001 nights. The combining of tales from multiple sources, of course, had been and remained the practice of nearly every European translator of the Nights until the twentieth century; but Habicht went one step further in that he also published this pastiche as an Arabic text which he claimed was based upon a Tunisian manuscript that in fact never existed. The degree to which his Arabic ‘edition’ was based upon a composite of various manuscripts and constituted little more than a miscellany of tales the majority of which had no historical connection to the Nights was demonstrated by Macdonald, but these tales nevertheless found their way into many later editions and translations.

The Būlāq text is the only major Arabic printed edition that was not directly initiated and financed by Europeans. It was one of the first books printed on the Egyptian government presses at Būlāq, a suburb of Cairo, established by Muḥammad ῾Alī. The edition presents the ZER text and was reprinted several times in the latter half of the nineteenth century.

Calcutta Ⅱ, also known as the Macnaghten edition, claims to be based on an Egyptian manuscript brought to India containing the full complement of 1001 nights; the manuscript, however, has never been found. The most significant characteristic of the Calcutta Ⅱ edition is the fulsome detail with which many of the tales are told such that they are noticeably longer than in any earlier source. It was precisely this dilated style that attracted the attention of later scholars and translators and has made it a favourite edition for many. Mahdi has convincingly demonstrated, however, that this manuscript could have been compiled no earlier than 1824–5 due to the otherwise inexplicable presence of details drawn from the Habicht edition of the Nights.¹⁶

A number of partial translations were published by scholars during the late eighteenth and early nineteenth centuries from Arabic manuscripts other than that used by Galland. Nearly all of these tales were from independent sources that claimed no connection to the Nights but were only associated with, or incorporated into, the Nights by their translators or publishers: Jonathan Scott published new tales first in 1800 and added more to his 1811 edition of Galland; Caussin de Perceval, as noted above, published the tales in Chavis’s manuscript as a supplement to the Nights in 1806; and Eduard Gauttier published tales from a variety of sources by scattering them among tales from Galland in his 1822–5 edition of the Nights. Once associated with the Nights, many of the tales continued to be included in later editions.

Two notable translations from actual Arabic manuscripts of the Nights date to this period as well. The first was a ZER text which was translated into French by Von Hammer-Purgstall after which both the Arabic manuscript and the French translation were lost, but not before having been translated into German by Zinserling (1823), which was later translated back into French by Trébutien (1828) and into English by Lamb (1826). The second was an English translation of the first fifty nights from the manuscript of Calcutta Ⅱ published by Torrens in 1838. Once the printed Arabic editions, with their complex genealogies, had been published, however, they soon became the base texts for almost all subsequent Western translations. The problems of the manuscript tradition now receded from the literary scene and even scholarly translators happily worked from the printed editions with little or no question of how these had been compiled.

When Edward Lane sat down to provide a new translation of the Nights in the mid-nineteenth century, much had changed since Galland’s day: European concepts of the East had dramatically shifted; literary tastes had changed and an entire genre of literature had been created in imitation of Galland’s work – the Oriental Tale – against which the Nights was now inevitably juxtaposed; with the spread of colonialism, political relations between Europe and the Ottoman empire had been redefined; academic methodologies of textual analysis, as well as theories of translation, had evolved; and with Champollion’s deciphering of the ancient hieroglyphic writing system, Egypt had to a great degree displaced Constantinople as the primary focal point for Western imaginings of the ‘East’. Lane wrote for a public that was fascinated by the East, but was by now equally as attracted to travel accounts and ethnographic description as they were to tales of marvel. The Nights, with its seductive combination of daily life details couched within fabulous fantasies, proved a perfect vehicle for Lane’s linguistic expertise and ethnographic interests.

About Edward Lane’s early life we know surprisingly little. Born in 1801, it is not known when or how he first studied Arabic, though a short manuscript grammar of colloquial Arabic culled from diverse sources from 1822 written by him indicates that he began his studies before he left for Egypt in 1825. His first sojourn, during which he lived in a decrepit tomb at the outskirts of Cairo dressed as an Egyptian, lasted three years and resulted in his A Description of Egypt. In 1828 he returned to England and four years later travelled again to Egypt where he stayed from 1833 to 1835. Once back in England after this second sojourn, he published his most famous work, Manners and Customs of the Modern Egyptians and then embarked upon his translation of the Nights.¹⁷

Seeing Galland’s version of the Nights very much with the eyes of his own century, Lane deplored Galland’s ‘occidentalization’ of the details of Eastern manners and dress in the tales and therefore annotated his own edition with so many notes that they were later collected and published as an independent volume.¹⁸ Railing against Galland’s prettified language, he couched his edition in an archaic, almost biblical, prose style in which he adhered closely to Arabic word order and syntax even when this produced bizarre effects in English. Although Galland had himself excised some sections of the tales which he felt immoral, Lane, who lived in one of the most morally restrained periods of English history, expurgated his text so drastically that his version was eventually only two-fifths as long as the Būlāq edition from which he was working. Although Lane also included tales from other sources, notably the Habicht edition, he meticulously notated the source of each story for his readers. Since Lane had travelled nowhere in the Middle East other than Egypt, he gave all aspects of the Nights an essentially Egyptian interpretation, often betraying his own unfamiliarity with manners and customs in Turkish- and Persian-speaking cultures. In the end, Lane’s redaction, though filled with voluminous scholarly notes, is so distant from the Arabic text in structure and in style, that almost no aspect of the Nights can be usefully studied from his work.

In 1842, less than two years after the publication of Lane’s Nights as a complete work in 1839–40 (it had appeared in monthly serialized form in 1838–40), John Payne, Lane’s successor, was born. Payne’s talent for languages was apparent early in life during his boyhood studies of French, Greek and Latin; Italian, Spanish, Portuguese, German, Arabic, Persian and Turkish he learned later on his own. A middling poet but a prodigious translator, Payne published several volumes of his own verse, complete translations of the poetic works of François Villon, Heinrich Heine, the Persian poet Ḥāfiz, the Rubā῾iyyāt of ῾Umar Khayyām, Boccaccio’s Decameron, all of the novels of Matteo Brandello, in addition to the most comprehensive English translation of the Nights up to that time. Payne was very scrupulous about his sources and at first translated directly and only from the Calcutta Ⅱ edition, considered the most authentic in his day. In 1884 he added translations of the remaining Calcutta Ⅰ and Breslau tales, but refused to publish ‘Aladdin’ and ‘Zayn al-Aṣnām’ until Zotenberg sent him copies of the above-mentioned manuscripts that had come into his possession. With his 1885 publication of these last two tales, Payne had basically published the entire corpus of the Nights from its various Arabic editions. Payne’s translation was remarkably enough the first English text of the Nights to be published wholly unexpurgated and one of the few to include all of the over 1,400 passages of poetry as well. In order to avoid potential legal difficulties relating to the erotic (‘pornographic’) nature of the text, the translation was released exclusively through private subscription and only 500 copies were printed, even though the demand was several times that number. During his lifetime, Payne refused to allow the work to be reprinted, though it has been reprinted several times since his death.

Enter Richard Francis Burton, by far the most colourful and flamboyant Westerner to add his imprint to the Nights. Though there is abundant material for reconstructing Burton’s life, there are two main obstacles for arriving at any semblance of truth: the first was Burton himself and what one of his biographers has termed ‘his bizarre relationship with fact’,¹⁹ and the second was his wife Isabel, who spent much of her life being ignored by her husband, but who, upon his death, destroyed his private papers and spent the last years of her own life attempting, rather unsuccessfully, to gloss over the lustier aspects of Richard’s career. His life has been recounted numerous times in a series of biographies.

Burton was born in 1821 to a family that had once held a respectable position. His father had served in the army in one of the queen’s regiments, but, through no fault of his own, had become embroiled in a national scandal involving Princess Caroline of Brunswick. After being forced to testify at a sham trial, he was decommissioned at half-pay for the remainder of his life. To save money, the family moved to the Continent where Richard was raised. A brief stint at Oxford ended with Richard being sent down, so, with few other options open, the family purchased him a commission in the Bombay Infantry of the East India Company. There his linguistic abilities led him to acquire at least a basic proficiency in a large number of Indian languages (among which the British counted Persian and Arabic). Burton did not fare well in the army, however, and devoted himself to the pursuits to which his eccentric personality was most suited: a life of adventuring and the writing of prodigiously lengthy and heavily annotated accounts of his feats. The most famous of these latter include his expedition with John Speke in search of the sources of the Nile and his pilgrimage in disguise to Mecca and Medina (the account of which should always be read in tandem with his equally sensationalist book about his journey to two other ‘forbidden cities’ – Salt Lake City and San Francisco! – entitled The City of the Saints and across the Rockies to California).

Burton’s path crossed that of the Nights late in his life when he was, as always, struggling to raise money. He read an announcement in the Athenaeum (5 November 1881) that the first volume of John Payne’s translation of the Nights was going to press. Burton wrote to Payne claiming to be working on a translation of his own and in response Payne graciously asked Burton if he would like to collaborate on the remaining volumes. When Burton actually met Payne, however, it became clear that he had nothing more than a few rough notes, and so Payne went on to publish his text. Once the initial 500 copies of Payne’s text were sold and it became clear that Payne was not going to allow a reprint, Burton stepped in to fill the gap. He did so to a great extent by plagiarizing Payne’s translation. Though the first sections of Burton’s text differ significantly from Payne’s, the subsequent sections are ever more closely drawn from Payne’s work until whole paragraphs and pages are found in which only a few words and the poetry have been changed. The story of this plagiarized translation has been public knowledge since Thomas Wright’s twin biographies of Burton and Payne, but this has never diminished the popularity of the Burton Nights.

To this basic text drawn from Payne’s work, Burton did, however, add a distinctive flavour. Like Lane, he attached massive annotations to the tales; unlike Lane, he eroticized and exoticized the text whenever possible such that ‘boy’ is often translated as ‘catamite’, the mention of a ‘black slave’ is transformed into ‘a big slobbering blackamoor’, and antiquated English vocabulary such as ‘whilome’, ‘kemperly’, ‘wittol’, ‘brewis’ and ‘anent’ appears throughout. Like Payne, Burton chose to make his text available through private subscription to avoid possible obscenity charges; this he did through the Kama Shastra Society, an association dedicated to the translation and publication of Eastern erotica. When his Nights (or rather Payne’s, with Burton’s emendations) were published (1885–6) and proved to be a financial success, Burton immediately compiled a second 1001 nights which were published under the title of the Supplemental Nights (1886–8). Together, the two sets include tales from every possible previous edition and translation, and for the Galland tales for which no Arabic original existed, Burton chose to translate Hindi translations of Galland’s French in order to retain the ‘oriental flavour’ of the tales! The Burton ‘translation’ continues to be popular in English even today, though most editions have been thoroughly expurgated of precisely those elements that were Burton’s contribution to the history of the Nights.

A final nineteenth-century translation was published in 1889–1904 by J. C. Mardrus and has remained throughout the twentieth century the most widely read French edition. The Mardrus edition was based on the Būlāq and Calcutta Ⅱ editions, but great liberties were taken in the ordering of the tales, several of which are inserted from collections unrelated to the Nights, and the translation in general is nearly so free as to merit the label of a ‘retelling’ rather than a translation.

Nearly all of the versions, editions and collections of the Nights available to Western readers even today are derived from the translations of Galland, Lane, Payne-Burton and Mardrus. The notable exceptions are a handful of new scholarly translations: (1) Enno Littmann’s German translation (1953), which gives a complete and meticulous rendering of Calcutta Ⅱ along with the additional tales from Galland, Calcutta Ⅰ and Breslau; (2) Francesco Gabrieli’s Italian translation (1948); and (3) three recent translations of the Muhsin Mahdi edition (1984–94) of the Arabic manuscript once owned by Galland – Husain Haddawy’s English translation (1990), Richard van Leeuwen’s Dutch translation (1993–9) and Claudia Ott’s German translation (2004). The Mahdi edition and its translations represent the only versions of the Nights that are faithful to one and only one Arabic manuscript of the Nights.

If the Nights, as argued above, was originally deemed by its French readers a work in some manner cognate to their own contes des fées, it was also presented to the public as a guidebook to the cultures of the East. Readers encountered this idea already in the Preface to the first volume of Galland’s Nights:

Although it might at first seem outrageous to vaunt a collection of fairy tales as a representation of real-life cultures, the claim was not wholly unjustified in the context of early eighteenth-century France. A number of Arabic works were available in French translation in Galland’s day, but these were primarily works of theology, philosophy, science and wisdom literature. None offered anything that even approximated the portrayal of Middle Easterners in shops and marketplaces, in their homes and occupations, going about their daily lives (albeit constantly interrupted by marvellous and supernatural forces), as they appeared to do in the Nights.

The idea that the Nights was the key to the East was to live on for well over two centuries. It is possible to see in the eighteenth- and nineteenth-century writings of European travellers how quickly the Nights became the template against which Westerners came to gauge their own personal experiences of the Middle East. At times the Middle East lived up to its reputation and offered scenes ‘like something out of the Thousand and One Nights’, but quite often the reality was found wanting; many disappointed tourists wrote of how the dirty and shabby inhabitants of the East failed to live up to the expectations raised by the Nights. Nearly every major literary and political figure of the nineteenth century at some point commented on the Nights (for England, as an example, see quotes by Bagehot, Beckford, Brontë, Carlyle, Coleridge, Dickens, Disraeli, Eliot, Fielding, Gibbon, Keats, Morris, Palgrave, Ruskin, Southey, Stevenson, Tennyson, Thackeray, Wordsworth and Yeats).²¹ In 1836, having been asked to review six new editions of the Nights, literary critic Leigh Hunt exuberantly declared it ‘the most popular book in the world’²² and in 1915, when the guns of World War Ⅰ were echoing over Europe, D. B. Macdonald could still write:

Such books [Arabic popular works] must always, for the home-staying student, take the place of contact with the Moslem world itself, and the best known of them is, of course, the Arabian Nights. They do not mislead nor misinform, as does that contact so often until it is controlled, and as still oftener do books of travel, and I would bear testimony now that when I did meet the Moslem world face to face, the picture of its workings and ideas and usages which I had gained from these romances, poems and religious tales needed modification in no essential point – almost, even, in no detail.²³

A full study of the re-entry of the Nights into the world of Arabic literature has yet to be undertaken, but would be a rich contribution.²⁴ It departed the Middle East in the early eighteenth century an obscure work that drew no special notice and returned a century and a half later as the foundational text through which the Western world had chosen to base its understandings of the Middle East. The Nights have at times been accepted in Arab countries as the ‘masterpiece’ that the West has proclaimed it, and excerpts from it are found in school primers and textbooks throughout the Arab world. On the other hand, however, the work has on occasion been criticized and rejected. In 1985, for example, a series of articles in major Egyptian newspapers attacking the obscenity and eroticism of the Nights led to a government ban of the work and the withdrawal of certain editions from the shelves of bookstores throughout the country.

The Nights was a relatively unknown collection of fabulous tales, one of many such collections that formed a part of late medieval popular Arabic literature, its unique embedding of tales and its compelling heroine notwithstanding. By chance, this particular work was snatched from obscurity and given a new existence by Western scholars, translators, publishers and readers who acclaimed it both as a literary masterpiece and as a trustworthy guide to Middle Eastern cultures. All of the Western alterations, additions and substitutions that shaped and reshaped the Nights over several centuries could perhaps be understood merely as the continuation of time-honoured practices in the production of popular literature – borrowing, compilation, redaction, rewriting – except for the fact that Westerners at the same time conceived of the text in decidedly modern, Western terms. For two centuries Western scholars sought in vain for ‘authentic’, ‘original’ and ‘complete’ manuscripts of the Nights. They harshly criticized each other’s scholarship and editorial policies and at times vehemently denounced new editions and translations. Western readers, though, for the most part simply regaled themselves with the astonishing ingenuity of the tales, the exoticness of their characters and settings, and their powerful ability to entertain. By the late nineteenth century the Nights had also become a vehicle for the inscription of Western erotic fantasies. Whether as a literary work, a cultural guidebook or as a manual of erotic desire, Westerners for generations measured the physical reality of the Middle East against what was for them the ‘real’ East, the East of A Thousand and One Nights. Certainly no other literary text can claim such a central role in reflecting, over several centuries, the changing relations between two great civilizations.

¹ See Abbott, ‘A Ninth-Century Fragment’.

² See Littmann, ‘Alf layla wa-layla’, p. 361; Goitein, ‘The Oldest Documentary Evidence’.

³ Macdonald, ‘Alf laila wa-laila’.

⁴ Pinault, Story-Telling Techniques, pp. 252–3.

⁵ See Galland, Journal, vol. Ⅰ, p. 242.

⁶ Lane, Manners and Customs, chs. 21–3; Page, ‘Naqqālī and Ferdowsi’; Ott, Metamorphosen des Epos.

⁷ Gerhardt, The Art of Storytelling.

⁸ See the works of MacDonald listed in the Bibliography; also Zotenberg, Histoire d’ ῾Alâ al-Dîn.

⁹ See Conant, The Oriental Tale.

¹⁰ See Seifert, Fairy Tales, Sexuality, and Gender.

¹¹ See Mahdi, The Thousand and One Nights.

¹² For Zotenberg and MacDonald, see fn. ⁸; for Mahdi, see fn. ¹¹.

¹³ See Mahdi, The Thousand and One Nights, vol. Ⅲ (1994), pp. 51–61.

¹⁴ Ibid., pp. 61–72.

¹⁵ Ibid., pp. 98–9.

¹⁶ Ibid., pp. 124–5.

¹⁷ See Ahmed, Edward W. Lane.

¹⁸ Lane, Arabian Society in the Middle Ages.

¹⁹ See Hastings, Sir Richard Burton.

²⁰ See Galland, ‘Epître’, in Mille et une nuit.

²¹ See Ali, Scheherazade in England.

²² Hunt, ‘Arabian Nights’, 106.

²³ Macdonald, ‘Concluding Study’, pp. 215–16.

²⁴ Walther, ‘Modern Arabic Literature and the Arabian Nights’.

Edition	Dates	Translations
Calcutta Ⅰ	vol. I 1814/vol. Ⅱ 1818	Rasmussen (Danish)
Breslau	8 vols. 1825–38 + 4 vols. 1842–3	Habicht (German)
Bulaq	2 vols. 1835	Lane (English)
		Henning (German)
		Mardrus (French)
Calcutta	4 vols. 1839–42	Torrens (English)
Ⅱ (Macnaghten)		Payne (English)
		Burton (English)
		Littmann (German)