"The mercy of Allah be upon you, for this is no island but a gigantic whale floating on the bosom of the sea, on whose back the sands have settled and trees have grown since the world was young! When you lit the fire it felt the heat and stirred. Make haste, I say; or soon the whale will plunge into the sea and you will all be lost!"
[--'The First Voyage of Sindbad the Sailor']
The perennially popular collection of (often very bawdy) morality tales known as The Thousand and One Nights has a fiendishly complicated textual history, which is largely invisible to the non-specialist reader. It certainly was to me, both when I first read a compilation of the Nights as a child, and later when I came to buy NJ Dawood's translation for Penguin (1954; revised edition 1973). It was only when I had to dig deeper into the text, in the course of academic research on Arabic conquest narratives (published last year, eee), that I began to get an idea of its strange history: the nested stories-within-stories structure of the narrative, it turns out, has nothing on the layers of invention, deception, creativity and appropriation that have gone into making the Nights what it is today.
Only a handful of the tales now associated with the Nights originally belonged to the collection, insofar as anyone can tell what was originally in it - or that 'originally' has any useful meaning when discussing an ever-evolving literary tradition. Much of the most visible and significant reshaping happened at the hands of Orientalist European translators. Some of the tales can be traced to medieval Arabic manuscripts of the Nights; some were filched from elsewhere in Arabic literature ('Sindbad the Sailor', for example); some came from Ottoman Turkish tradition; some (such as 'Aladdin and the Enchanted Lamp') seem to have been largely invented, and many others heavily revised, to fit the tastes and assumptions of an 18th-century French reading public with a thirst for the supernatural and the 'exotic' East.
In his introduction to this particular version, Dawood labels the Nights a collection of "folk-tales". He praises them as "the spontaneous products of untutored minds", which can, he says, "be regarded as the expression of the lay and secular imagination of the East in revolt against the austere erudition and religious zeal of Oriental literature generally". Even allowing for the inflated, essentialising rhetoric characteristic of mid-20th-century and earlier Orientalists, there is a level of dramatic irony here: what 'untutored mind', exactly, was writing down such carefully structured stories? And how could anyone who has read even a little medieval Arabic literature really think it was all so humourless and 'austere'?
Where the introduction becomes (unintentionally) funny, however, is with Dawood's assertion that the Nights "have little in common with the refined didacticism of Classical Arabic literature and have therefore never been regarded by the Arabs as a legitimate part of it". Because since Dawood wrote this it has become inescapably clear, thanks to the scholarship of Muhsin Mahdī, amongst others, that in many ways the Nights are indeed not a part of Classical Arabic literature, but not for the reasons he suggests: rather, they're a Persian-Arabic-French co-production, and many of the most famous stories only entered the corpus long after the Classical period.
While the Nights' framing story - of the wily Shahrazad telling tales night after night to the vengefully misogynistic Sasanian shah, pausing on a cliffhanger each dawn in order to win herself another day of life - is set in pre-Islamic Persia, many of the individual tales are set in Baghdad during the reign of the 'Abbāsid caliph Hārūn al-Rashīd (r. 786-809). (Others are set in Egypt, China, and the high seas.) There is evidence that the collection originated not long after this period, a time that soon came to be viewed as a golden age of learning, stability, and prosperity, and one in which Persian culture was beginning to have a real impact on Islam, as more and more prominent Persian writers and politicians came to embrace the religion and translate Persia's rich literary traditions into Arabic. Back in the 1940s, Nabia Abbott found a papyrus fragment of the Nights that she dated to the 9th century; writers in the 10th century also mention it: al-Mas'ūdī (who died c. 955) summarises a story that he says comes from it, while the bibliographer Ibn al-Nadīm (d. 998) includes the work in his great index of Arabic literature, noting that it consisted of about 200 'nights' (this is important, for reasons I'll come back to in a moment).
The earliest manuscript of the Nights that survives, however - at least to the best knowledge of modern scholars - comes from the 14th century, and it is virtually impossible to tell what, if any, 'authentically' 9th-century material it contains. As is fairly typical in medieval literature in any language - before printing, copyright, etc - the Nights does not seem to have been a fixed book, or the product of a single pen.
But the Nights as we know it - that is, the Nights as represented in Dawood's edition, and more recently (with much more self-awareness) in Penguin's three-volume 2008 edition - is a long way from even this 14th-century core of material. These versions of the Nights, like virtually every version available in English translation, are in fact a mish-mash of three things: a) Classical and medieval Arabic retellings of Persian fables, b) bits and pieces from all over Classical and medieval Arabic fictional, historical and geographical literature, and c) a bunch of 18th-century French fairytales dressed up in 'Oriental' clothing.
The literary tradition of the Nights underwent a radical re-invention in the 18th century at the hands of a French booklover and Arabicist named Antoine Galland (1646-1715), and the various other 'translators' who came after him. Dawood acknowledges this to a limited degree (he calls Galland "by no means a faithful translator"), but his introduction gives no proper sense of the scale of what happened. It's exaggerating only slightly to say that this re-invention stems from Galland misunderstanding the work's title when he happened across it during a book-finding trip to the Middle East. The Arabic title of Alf layla wa-layla (literally 'the thousand nights and a night') is simply a pretty, figurative way of saying 'an awful lot of nights'. Galland, however, took the title literally; as Muhsin Mahdī puts it, charmingly, in his brilliant 1984 study of the Nights' textual history, Galland "fell under the spell of the title". (Incidentally, the first 60 pages or so of Mahdī's book, also called The Thousand and One Nights, is available to read for free on Google books; I highly recommend seeking it out.)
Galland became convinced that the text he had found, the narrative of which was divided into only 280 or so nights, was incomplete. Driven both by curiosity and, increasingly, by his publishers back home - for whom the first few volumes of Galland's translations had proven very popular and profitable - Galland tried in vain to hunt down additional manuscripts. When that failed, he published stories from other sources - including the tales of Sindbad the Sailor and Aladdin, both of which are included in Dawood's translation - under the Nights' title, and offered contradictory reports on where he had found them. Altogether, 12 volumes of his French translation/adaptation/creation were published between 1704 and 1717.
The quest to find the 'missing' nights continued after Galland, and the deceptions only grew more elaborate and brazen. In the 1780s, Jacques Cazotte, noted writer of contes de fées (fairytales), was put in touch by his publisher with one Dom Denis Chavis, a Christian Arab priest who claimed to have access to an additional Nights manuscript. In actual fact, said manuscript was a forgery. Chavis copied the Arabic manuscripts Galland had worked from (both the 14th-century Nights text and another containing the Sindbad stories), introducing minor changes (including some contemporary Syrian colloquialisms) to make it look like an independent work; he also translated some of Galland's French material - including 'Aladdin' - 'back' into Arabic, based solely on the French and not on any Arabic original. Finally, he added a bunch of new stories to the Nights that were actually from completely separate texts. Cazotte and Chavis worked together to translate all this into French.
Meanwhile, Michel Sabbagh, an Iraqi and a much more skilled and learned Arabicist than Chavis, set about creating a decidedly more convincing forgery. He created what he claimed was a transcription of a complete (i.e. 1001 rather than 280) Nights manuscript he'd found in Baghdad, whose colophon dated it to 1703 - thus making it appear to be completely independent of the manuscripts Galland had used, and also putting the putative original well out of the likely geographical range of scholars who might go looking for it (the Orientalist stomping grounds in those days being primarily Aleppo, Istanbul, and Cairo).
Scholars have debated back and forth over the veracity of Sabbagh early-19th-century version, but recently the conclusion has become inescapable: Sabbagh's Nights, like that of Chavis, was largely based on Galland's, spurious additions and all. There is no 'complete' pre-18th-century Thousand and One Nights in Arabic, with all 1001 episodes, because such a thing never existed outside the imagination of the early modern French publishing industry. Nonetheless, these forged manuscripts formed the bases for the two main manuscript traditions, Calcutta and Bulaq (Egypt), that have been used for translation and study ever since, on the presumption of their authenticity. In a considerable step forward for the field, Muhsin Mahdī published an Arabic edition of the 'core' Nights - the 14th-century manuscript - in 1984, as a companion volume to the ground-breaking study I mentioned above; an English translation of this core text, by Husain Haddawy, was published in the US a few years later, although to the best of my knowledge it remains unpublished over here in the UK.
What does this mean for the modern reader? On one level, it enriches the reading experience to see how this body of narrative has evolved over time and in multiple cultural contexts. The example I would offer to demonstrate this is 'The City of Brass', a story that features in Richard Burton's famous late-19th-century translation and in the new (2008) three-volume Penguin collection, but not in Dawood or Haddawy. This is not a core Nights tale, in the sense of being from the 14th-century manuscript, but it wasn't wholly a French invention either; in fact, it comes out of the Arabic historical tradition surrounding the Islamic conquest of Iberia.
The conquest itself happened in the year 711, but its history did not begin to be recorded until the middle of the 9th century. The earliest surviving Arabic chronicle of the conquest was written by the Spanish Muslim jurist Ibn Habīb (d. 852). Ibn Habīb's account is essentially a selection of anecdotes about the myriad strangenesses of Iberia before Islam arrived to civilise the place. In one of these, an expeditionary force wandering in the desert encounters a city, which is entirely surrounded by a wall made of copper. The troops can find no gate, and so their commander, Mūsā, orders the building of a scaffold so that they can scale the wall and see inside. However, when the builders reach a suitable vantage point, and look within, they suddenly begin to laugh - and then throw themselves over the wall, apparently to their deaths. Mūsā offers 100 dinars to the first man to climb up and find out what just happened; the unlucky volunteer, however, does exactly the same thing as his predecessors. Mūsā, chastened, declares the city one of the marvels ('ajā'ib) of Satan, and the army beats a hasty retreat.
Most later medieval accounts of the conquest did not include the story of the City of Copper; instead, it was picked up by writers working in the geographical tradition - that is, by people writing descriptions of the various regions of the Islamic world and its neighbours, a practice that was often supplemented with eye-catching little bits of trivia and anecdotes about their subjects. The tale was, of course, embroidered in the retellings. My favourite of the versions I've found so far is that of al-Gharnātī (d. 1169), who wrote a book all about 'ajā'ib (which might as well have been subtitled Weird Stuff From Around The World). He adds a splendidly gory twist. Here, the chaps scaling the walls to scout out the inside emit horrible shrieks after they plunge over the top. The third and final man is sent up with a rope tied around his waist, but to no avail; although the army down below tries to hold him back, such is the force pulling him inside that the supporting rope succeeds only in cutting him in half before the horrified eyes of his fellows down at the foot of the wall. Despite such cosmetic changes, though, the essential structure and message of the story remains the same: there are some deeply strange, deeply supernatural things in the world that man shouldn't dream of tangling with or trying to understand.
Enter the French Nights adapters. 'Hmm, that's odd; let's get out of here' was considered an unsatisfactory ending for a French audience reading these tales in a very different way, and so the story was extended. In this version, the Muslim army finally made it inside the City - one of their number having been able to climb down inside, and find and open a hidden gate - to discover a dead tableau of prosperous urban life, the city's entire population frozen to the spot in the middle of all manner of everyday actions. Exploring further, they find a palace, and inside it the body of the city's queen and several helpful explanatory stone tablets, which explain that the city got mummified on account of being greedy instead of humble and pious. Mūsā immediately renounces plundering and goes back to the east to live a more ascetic life. As a moral, it's hardly any more sophisticated than the original, but the imagery is just as striking, and it's fascinating to be able to watch the layers of a story develop and change across the centuries like this. With all necessary caveats in mind, then, I'd recommend the three-volume Penguin edition on this basis alone.
On another level, knowing that much of the Nights is early-modern French fabulism means rethinking what we take from the Nights about the medieval Islamic world; Dawood's easy conviction that the Nights show us a folksy, "secular" Arabic culture is rather undercut by the realisation that much of the image this text has created for the western world over the past three centuries is an Orientalist fantasy. Many of the stories Dawood includes are an insight into 18th-century France, not 9th-century Baghdad, and there is a strong appeal to reading a version of the Nights, like Haddawy's translation, that contains only the material that we know is both medieval and Islamic. There is still no manuscript evidence for the existence of either 'Ali Baba and the Forty Thieves' or 'Aladdin and the Enchanted Lamp', two of the most famous Nights stories in the western world, prior to Galland's French 'translation'. Galland claimed to have got them from an otherwise unknown Arab informant - in oral form - called Hanna Diab. Both Chavis' and Sabbagh's purported Arabic 'originals' of 'Aladdin' are in fact translations from Galland's French into Arabic, with the odd colloquial and medieval construction thrown in - attempts to make them sound authentic, but not enough to fool the numerous scholars who have pointed to the underlying French sentence structures and cultural assumptions of these tales. (An additional 'Aladdin' manuscript in Oxford's Bodleian library has likewise turned out to be a dead end, albeit one in better Arabic.)
Even without this background awareness, 'Aladdin and the Enchanted Lamp' reads oddly in Dawood's volume. It is set in China and full of terminology (like 'nabob') that belongs more more obviously in an early modern European context, rather than a medieval Arabic one. Its villain is also repeatedly referred to as 'the Moor', although that, at least, could simply be an awkwardly outdated translation - it is certainly not impossible that a medieval Arabic story would cast a Berber as an Othered bad guy, in contrast to more worthy Arabs and Persians. More significantly, though, the main appeal of reading 'Aladdin' lies in recognition - the magic (alas not flying) carpet! the jinnī in a lamp! the beautiful princess and the eeevil vizier! - because as a story it lacks the fun and adventure you might expect, wittering on for 70 or so pages and expecting us to cheer for a, frankly, feckless waste of space of a hero.
Other stories are much more enjoyable. Probably my favourite is 'The Porter and the Three Girls of Baghdad', one of the three core Nights tales from Muhsin Mahdī's edition that is also in Dawood's translation (and, of course, in Haddawy's). I have a hazy memory of reading this story as a child - specifically, of the central female character repeatedly hitting her pet black dogs, and then weeping over them - but I'd forgotten most of it, making it an endlessly inventive delight on re-reading: this is the sort of imaginary ninth-century 'Abbāsid caliphate in which there are random palaces underground, a magnetic mountain in the middle of the sea that destroys ships, and a chap made of brass who spends several days rowing the storyteller in a boat. Marvellous.
Since it appears in both translations, 'The Porter...' also gives me an opportunity for a brief comparison of translation technique. Here is our first description of the girl in Dawood:
Here the young woman unveiled her face and knocked, and the door was opened by a girl of surpassing beauty. Her forehead was white as a lily and her eyes were more lustrous than a gazelle's. Her brows were crescent moons, her cheeks anemones, and her mouth like the crimson ruby on King Solomon's ring. Her teeth were whiter than a string of pearls, and like twin pomegranates were her breasts.
Haddawy's is even more purple, but of particular note is his rendering of the final line: the girl, we're told, has a "bosom like a fountain, breasts like a pair of big pomegranates resembling a rabbit with uplifted ears". I confess myself bemused over the rabbit ears part; my partner suggests it might have something to do with cleavage - any further ideas...?
'The Tale of the Hunchback' - another of the core tales - is a wonderful example of both the style and the tone of the Nights. Set in the cities of Basra and Baghdad in (roughly) the mid-9th century, it offers multiple nested narratives, neat little pen portraits of multicultural urban life, beautiful tricksy women, and drunken lust-addled men getting publicly humiliated. In the first level of the story, a tailor and his wife accidentally(ish) kill a hunchback beggar, after which a series of tragicomic misunderstandings sees the highly symbolic trio of a Christian, a Muslim and a Jew each believing they're to blame for the death, and each in turn being absolved by the other; in the next level, the tailor tells the story of how he met a highly dodgy raconteur who is also a barber (as he so modestly puts it, he is "not only a barber of great repute, but also a master of the arts and sciences: one who is not only versed in alchemy, astrology, mathematics, and architecture, but also (to mention only a few of my accomplishments) well schooled in the arts of logic, rhetoric, and elocution, the theory of grammar, and the commentaries on the Koran"); then we get the barber's tale; this, in turn, involves the barber telling the stories of each of his six brothers.
Taken together, the story is full of people being hoist by their own petard, people getting their just (or unjust) desserts, and men and women finding ways to sneak around social restrictions to meet for assignations. Often all three at once, as in the story of the second of the barber's brothers, who can't believe his luck when he ends up being invited into the apartments of a beautiful young woman (described, as so often in Persian literature especially, as having a face like the moon). Once there, he carries out every request she makes of him - through an older woman acting as a go-between (another typical aspect of Persian fiction) - in the delighted belief he's about to get laid. The fact that these requests include dyeing his eyebrows, shaving off his beard (a major symbol of emasculation in medieval Islamic culture), getting drunk, and stripping naked while she remains fully clothed, doesn't faze him for a moment; the silliness climaxes, so to speak, in the young woman encouraging the man to chase her through her apartments. He does so in "a frenzy of desire", whereupon she leads him through a room with a weakened floor, which gives way beneath him and dumps him unceremoniously out in public:
When the merchants saw al-Haddar drop in their midst – naked, shaven, with eyebrows dyed, his face painted, and his penis erect and rampant, they booed him, and clapped their hands at him, and thrashed his bare body with their hides, until he fell down senseless.
Women are, in general, decorative and/or schemers in the Nights. My favourite is the title character of 'The Young Woman and Her Five Lovers', who manages to trap each of the title's five men in turn in the same, multi-compartment cupboard, before making off with their wealth and her real lover, while the five men bang impotently on the doors. While her goal is dishonest, the real moral failings that the story lampoons are the men's: it is they who have been rendered foolish by their arrogance and lust, each in turn believing both that this woman finds them irresistible, and that at the very moment they are finally alone together in her home, she hears her husband coming and maybe for their safety they'd better hide in this convenient man-sized cupboard compartment...
Men and woman alike display this sort of skill at deception, however; it is the context in which it happens that is gendered, not the technique itself. So women trick the men that trespass (or are enticed) into their homes; men play their mind games in public. In 'The Fisherman and the Jinni' - another of the three core tales in Dawood's translation - an almost identical trick is played on a supernatual creature:
[T]he fisherman asked: "How could this bottle, which is scarcely large enough to hold your hand or foot, ever contain your entire body?"
"Do you dare doubt that?" roared the jinnee indignantly.
"I will never believe it," replied the fisherman, "until I see you enter this bottle with my own eyes!"
Upon this the jinnee trembled from head to foot and dissolved into a column of smoke, which gradually wound itself into the bottle and disappeared inside.
There is one female character whose intelligence is presented as an unambiguous force for good. This is Shahrazad herself, described in Dawood's translation as someone who "possessed many accomplishments and was versed in the wisdom of the poets and the legends of ancient kings". Haddawy's rendering is more detailed, and shows the author(s) of the Nights transposing a typical elite Islamic education onto Sasanian times:
The older daughter, Shahrazad, had read the books of literature, philosophy, and medicine. She knew poetry by heart, had studied historical reports, and was acquainted with the sayings of men and the maxims of sages and kings. She was intelligent, knowledgeable, wise and refined. She had read and learned.
Convinced of women's evil, the shah has been marrying a new girl each night and having her put to death by the morning. Despite her father's attempts to protect her, Shahrazad insists that she be allowed to answer the shah's summons when his eye falls upon her, "so I may either succeed in saving the people or perish and die like the rest". Soon, she is putting her quick wits to good use, keeping the shah occupied each night so as to prevent him from killing any further women.
So there are good reasons to pick up any one of these translations. I'd recommend the Haddawy for the 'real' Nights, for proof that medieval Islamic literature can be enormously entertaining, and for a much better introduction than Dawood's, although the translation lacks flair in places; I'd also suggest the new Penguin version, if you want to see how the Nights was changed by cross-cultural translation and adaptation, and also because the stories you remember will be in there. Get both; call yourselves bibliophiles, otherwise?