A New Translation of the Arabian Nights

[Malcolm & Ursula Lyons (2008)]

A couple of months ago my friend Derek Gordon (aka Bringwonder the Storyteller) asked me which was the best translation of the Arabian Nights. The situation, I had to tell him, was a complex one.

The short answer is that there isn't any completely satisfactory translation of the Nights into English, despite the bewildering number of selections, retellings, adaptations and other attempts at a solution to the problematic nature of the book itself which have appeared over the past couple of centuries. Ridiculous, isn't it?

With the appearance of this new, complete Penguin translation by Malcolm C. Lyons - introduced and partially annotated by Robert Irwin, author of The Arabian Nights: A Companion, and with new translations of "Aladdin" and "Ali Baba" from the French by Ursula Lyons - the situation suddenly looks much brighter. That is if, like me - and, I guess, professional storytellers such as Derek - you're a fanatic for everything to do with the (so-called) Arabian Nights (or Kitab Alf Layla wa Layla [Book of a Thousand Nights and a Night]).

I suppose, though, it was only a matter of time. No-one could seriously question the need for it. There's a brand-new translation in French, just out from the Bibliotheque de la Pleiade; thorough, reliable versions in German and Italian - even in Russian and Czech. So why has English been lagging behind?

Well, the first reason stems from the fact that there are actually two very different manuscript traditions of the Nights.

1. On the one hand, there's the 14th-century Syrian Ms. Galland, which the Frenchman Antoine Galland used for his pioneering multi-volumed translation of 1704-1717. This was the form in which the collection first reached Europe, and the choices Galland made in making his rather free adaptation of the materials available to him have had a huge influence ever since. It was he, for example, who chose to incorporate Sinbad the Sailor (originally from a quite different text) into the Thousand and One Nights. It was he who, forced to supply the insatiable demand of his public for more volumes of stories (rather like an eighteenth-century J. K. Rowling), collected "Ali Baba," "Aladdin" and various other stories from the oral recitations of a visiting Lebanese Christian, Hanna Diyab (the Arabic texts of the stories discovered later turned out to have been adapted from Galland's French, rather than the other way round).


2. On the other hand, there's Z.E.R. [Zotenberg's Egyptian Recension - named for the German scholar who first identified this separate manuscript tradition in the late 1880s]. This forms the basis of most of the "complete" versions of the Nights - i.e. containing 1001 actual nights of storytelling. It was once thought to be the original from which Galland's incomplete version was extracted, but unfortunately it turns out that the traffic was actually the other way. Z.E.R. arose in Egypt largely as a result of the demands of Westerners, brought up on Galland's elegant fairy-tale version of the Arabic tales, for a fuller and more comprehensively "Oriental" version of the whole collection. The 1835 Bulaq edition (printed in Cairo), and the 1839-42 Macnaghten edition (printed in Calcutta by the British East India Company) are the two essential versions of this text. Opinions differ greatly on which of the two is preferable - Macnaghten is fuller, but also contains a lot more editorial interference (though probably not by William Hay Macnaghten himself, who was killed in the British retreat from Kabul in 1841).

So which of all these competing Arabic versions should one translate? Which represents the genuine 1001 Nights? The answer is (unfortunately) that there is no "genuine" text of the Nights. The medieval Arabic encyclopedists who mention the collection say that it was originally translated from Persian into Arabic (and the names Sharazad and Shahryar only make sense in Persian, which tends to substantiate the claim). The Persians may (or may not) have derived the frame-story of the Nights from a lost Sanskrit version - opinions differ on that point. In any case, none of these earlier texts survive, but enough information about them remains to convince us that they can have contained only a very few (if any) even of the stories which have come down to us as the solid core of the collection: "The Merchant and the Jinni," "The Fisherman and the Jinni," "The three Ladies of Baghdad" and (of course) the immortal "Tale of the Hunchback."

The collection as we have it seems imbued with the spirit of Islam, and myriad everyday details of life in Baghdad and Cairo. The Persian Thousand Nights, however, certainly predated the Arabic conquest, and some critics have suggested that this pre-Islamic strand is still discernible in the independent wilfulness displayed by the various Jinnis and supernatural creatures in the first few stories of the Nights. Possibly. Who knows?

What is clear is that there has been a story-collection with a title something like The Thousand and One Nights, with a frame-story something like the Scheherazade-Shahryar one we're all familiar with, extant in the Arabic language since at least the ninth century. The earliest substantial manuscript of this collection is, however, the 14th-century Ms. Galland (it seems a pleasing coincidence that the first translation into a European language turns out to have been made from the oldest extant manuscript). But by far the fullest collection of stories available under the title of The 1001 Nights is represented by the various texts of Z.E.R.

So do you want the oldest and most elaborate version: the sadly truncated and incomplete Ms. Galland? Or would you prefer the hybrid, textually dubious, but incredibly capacious Z.E.R.? As Winnie the Pooh put it, when asked to choose between having jam and honey on his bread, "Both, please, but forget about the bread." Husain Haddawy gave us a beautiful and elegant translation of Muhsin Mahdi's critical edition of the Ms. Galland in 1990. Malcolm Lyons now supplies us with the missing part of the puzzle: a complete and accurate version of the Macnaghten edition into clear, current English.

By my count, then, Malcolm Lyons' new 3-volumed, 2,700-page chef-d'oeuvre is the seventh substantial translation into English - leaving aside for a moment all the innumerable abridgements and retellings. I thought I might list them below, along with any currently-available editions of them in print, in order to underline the significance and timeliness of Lyons' achievement.

[Antoine Galland, ed. Robert L. Mack (1998)]

1. Antoine Galland, 12 vols (1704-17)



Arabian Nights Entertainments: Consisting of One Thousand and One Stories, Told by the Sultaness of the Indies, to divert the Sultan from the Execution of a bloody Vow he had made to marry a Lady every day, and have her cut off next Morning, to avenge himself for the Disloyalty of his first Sultaness, &c. Containing a better Account of the Customs, Manners, and Religion of the Eastern Nations, viz. Tartars, Persians, and Indians, than is to be met with in any Author hitherto published. Translated into French from the Arabian Mss. by M. Galland of the Royal Academy, and now done into English from the last Paris Edition. London: Andrew Bell, 1706-17. 16th ed. 4 vols. London & Edinburgh: C. Elliot, 1781.


We begin at the beginning, with the first introduction of the Nights into Europe, Galland's maddeningly elegant, terrifyingly eclectic French adaptation / translation / transformation, which came out in 12 volumes between 1704 and 1717.

Galland was translated, anonymously, into English almost as soon as his volumes started to appear, and (variously supplemented) his translation provided the basis for all subsequent English versions until Lane's translation began to appear in 1838.

It's no exaggeration to call Galland a storytelling genius. The influence his work has had on European culture is incalculable (whether or not it's true that French guttersnipes used to gather under his bedroom window at night and yell out "Please, my sister, if the Sultan pleases, tell me another one of those tales of which you possess such a great store" until he gave in and cranked out another volume for them, I'll leave you to determine).

The liberties he took with the text were considerable - some would say unforgiveable - but the fact that the various textually-unwarranted additions he made to the "pure" Ms. Galland, have come to be seen as the core of what most characterizes the Nights as a whole (Aladdin, Ali Baba, even Sindbad himself), might lead us to call him more their creator than their translator.

The earliest English translation of Galland (by the so-called "Anonymous Grub-Street Hack": rather a strange name under which to achieve immortal fame) is now fortunately available in its entirety in the Oxford World's Classics series. I've provided a link to it below:


Arabian Nights’ Entertainments. Ed. Robert L. Mack. The World’s Classics. Oxford: Oxford University Press, 1995.




[Lane (1995)]


2. Edward William Lane, 3 vols (1838-41)



The Thousand and One Nights, Commonly Called, in England, The Arabian Nights’ Entertainments. A New Translation from the Arabic, with Copious Notes. 3 vols. London: Charles Knight, 1839-41.


Edward William Lane was a great Arabist, long resident in Cairo, and famous for his book The Manners and Customs of the Modern Egyptians (1836). As a supplement to this work, he undertook a translation of the Bulaq edition of the Nights (or rather the Z.E.R., Cairene text of the Nights rather than Galland's earlier, Syrian text) soon after its publication in 1835.

The translation was wildly popular, partially because of the stunningly beautiful engravings and textual ornaments provided for it by William Knight, but also because it supplied a great many new stories to supplement those familiar from reprints of Galland and his followers.

Lane's extensive annotations about Muslim life were subsequently extracted and published as a separate volume, Arabian Society in the Middle Ages, still in print today (as you can see from the illustration above). Burton, no great fan of the prudish reserve of Lane's translation, put their value more simply: "The traveller who adds Lane's notes to mine will know more about the Moslem East than many a man who has spent half his life there ..."

On the minus side of the exchequer, Lane's prose is rather Latinate and heavy, and he does leave out a vast amount of material - sometimes because he considered it too risque, but other tales were dropped simply because he regarded them as repetitive or tedious: subjective judgements at the best of times.

For this reason, his text is little read today, though it has had almost as great an influence as Galland's on subsequent bowdlerised retellings for children (Andrew Lang's included). Lane's is the last version of the Nights which could be called at all suitable for children.

The last substantial revision and reediting of it was by Lane's nephew Stanley Lane-Poole in the early twentieth century, once available in a beautiful little pocket edition from Bohn's Library:


Lane, Edward William, trans. The Arabian Nights’ Entertainments. Ed. Stanley Lane-Poole. 4 vols. 1906. Bohn’s Popular Library. London: G. Bell, 1925.




[John Payne (2008)]


3. John Payne, 13 vols (1882-89)



The Book of the Thousand Nights and One Night; Now First Completely Done into English Prose and Verse, from the Original Arabic. 9 vols. London: Villon Society, 1882-84.

Tales from the Arabic of the Breslau and Calcutta (1814-’18) Editions of the Book of the Thousand Nights and One Night, Not Occurring in the Other Printed Texts of the Work; Now First Done into English. 3 vols. London: Villon Society, 1884.

Alaeddin and the Enchanted Lamp; Zein ul Asnam and the King of the Jinn: Two Stories Done into English from the Recently Discovered Arabic Text. London: Villon Society, 1889.


Nearly fifty years after Lane had completed his work, the scholar and linguist John Payne decided to translate a complete and faithful version of the Nights (conflated from the Bulaq and Macnaghten texts) for private circulation only.

It's alleged that he did most of his work whilst travelling around London on the roof of a horse-drawn bus, glancing up from time to time to make sure that the weather wasn't going to damage his precious dictionaries and manuscripts. It seems a little hard to believe, but so the story goes, at any rate.

Payne was a devotee of high-flown Wardour-Street English, and his prose is ornate and florid in the extreme, but the sheer magnitude of his achievement is difficult to exaggerate. He and he alone put into English the whole of the 1001 Nights, in their most complete form, and he did it with self-taught Arabic, having never travelled in the East.

He went on, in subsequent years, to translate four supplementary volumes of tales from the other manuscript traditions of the Nights: a magnificent example of scholarly acumen and dedication.

His fatal error was to issue his translation in 500 copies only, with a pledge (presumably for leagal reasons) never to reproduce it in any way. As a result, his work exerted very little influence over the reading public, and the field was left open for his great friend (= rival) Sir Richard Francis Burton, who thus was able to scoop the pool.

His translation has recently been reissued by Borders Classics (details below). Copies of the original edition are understandably somewhat rare, though it has been reissued complete in various clandestine forms over the past hundred years or so, as well as providing the text for Joseph Campbell's Portable Arabian nights.


Payne, John, trans. The Book of the Thousand Nights and One Night. Publisher's Note by Steven Moore. 3 vols. Ann Arbor, MI: Borders Classics, 2007.




[Burton / Zipes (1991)]


4. Richard F. Burton, 16 vols (1885-88)



A Plain and Literal Translation of The Arabian Nights’ Entertainments, Now Entituled The Book of the Thousand Nights and a Night: With Introduction, Explanatory Notes on the Manners and Customs of Moslem Men and a Terminal Essay upon the History of the Nights. 10 vols. Benares [= Stoke-Newington]: Kamashastra Society, 1885.

Supplemental Nights to the Book of the Thousand Nights and a Night with Notes Anthropological and Explanatory. 6 vols. Benares [= Stoke-Newington]: Kamashastra Society, 1886-88.


"Galland for the nursery; Lane for the library; Payne for the study; and Burton for the gutter," said one of the first reviewers of Burton's immense, frustrating, multifaceted masterpiece (according to Burton himself , anyway - he included a vituperative section entitled "The Reviewers Review'd" in one of the later volumes of this library masquerading as a single book).

What can one say about his translation? Well, Burton was indisputably a great linguist and traveller, and his knowledge of the East was unrivalled. All this erudition is on display in the copious notes to his many volumes.

He was also a cranky eccentric, obsessed with "unmentionable" subjects such as female circumcision, castration, bestiality and (above all) male homosexuality, and lavished information on these topics through his notes and the immense "Terminal Essay" which occupies most of the tenth volume of his immense work. These matters seemed far more shocking to the Victorians than they do to most of us, though, and it's increasingly difficult to see much distinction between the professional curiosity of the modern anthropologist and Burton's omnivorous zeal for collecting of information on all aspects of the societies he travelled through and studied.

The immense, ill-assorted detail displayed in his own books of travel and adventure tends to make them very difficult to read. Not so his Nights, where the genius of the original storyteller enables us to travel with far more ease and pleasure through the strange thickets of Burtonian erudition.

Racist, yes; Imperialist, undoubtedly (though in an unorthodox and complex way) - but, like Burton himself, his Nights are a mine of quaint and curious lore. This is probably the fullest version of the Nights which will ever appear, and (till the Lyons translation was published) it still had to be regarded as the standard text in English.

Burton's honesty may be doubted, true. He undoubtedly plagiarised from Payne's preexisting translation (allegedly with the latter's permission, but actually - as Payne subsequently revealed - to his considerable irritation). The real problem with Burton's 16 volumes, though, is the bizarre pseudo-English in which his translation was composed.

If a word had ever appeared in English - in Elizabethan thieves' jargon, in translations of Cervantes or Rabelais, in the obscurer pages of his namesake's seventeenth-century Anatomy of Melancholy - that was sufficient reason for Burton to revive it. He attempted to reproduce all aspects of the Arabic text: the passages in rhyming prose, the long poems in monorhyme (a single rhyme sufficing for up to forty verses). Nor did he hesitate to coin new words whenever he thought he needed them.

The result is fascinating, but harder syntactically than Milton, and every bit as linguistically florid as Spenser. One almost has to learn a new language in order to read it. To be sure, if you persevere, the variety and flexibility of Burton's vocabulary impresses you more and more, and the sheer energy of his diction comes to beguile and enchant as much as it initially frustrated you, but it's a hard row to hoe, and not everyone can be bothered.

A recent attempt to keep these virtues of Burton's translation whilst modernising his English can be found Jack Zipes' two paperback volumes of selected Arabian Nights. I'm not a great fan of abridgements and retellings, myself - better, I think, to spend your time grappling with the original, but I guess these toned-down versions will continue to appear:


Zipes, Jack, ed. Arabian Nights: The Marvels and Wonders of the Thousand and One Nights, Adapted from Richard F. Burton’s Unexpurgated Translation. Signet Classic. New York: Penguin, 1991.

Zipes, Jack, ed. Arabian Nights, Volume II: More Marvels and Wonders of the Thousand and One Nights, Adapted from Sir Richard F. Burton’s Unexpurgated Translation. Signet Classic. New York: New American Library, 1999.




[Mardrus / Mathers (1986)]


5. Edward Powys Mathers, 8 vols (1923)



The Book of the Thousand Nights and One Night: Rendered from the Literal and Complete Version of Dr. J. C. Mardrus; and Collated with Other Sources. 1923. 8 vols. London: The Casanova Society, 1929.


Now this one really is a lot of fun. A purist would probably ignore it altogether, given that it's an English translation made directly from the French rather than from any Arabic text. What's more, it was made from possibly the least accurate "complete" translation ever to appear - that of Joseph Charles Mardrus, an Egyptian French Doctor whose adherence to the aesthetic of the fin-de-siecle is rather embarrassingly apparent in the liberties he took with the original text.

He had no real original text, in fact. He simply embroidered onto the 1835 Bulaq edition a series of ornamentations and expansions of tales he found too bald, not scrupling to invent entirely new stories at times. But he pretended that all of this was justified by a (mythical) "complete" manuscript whch no-one else was ever permitted to see.

Mardrus was writing in the ambience of Pierre Louys and Pierre Loti, and was encouraged by their example to add a great deal of naughty spice to the somewhat bald eroticism he encountered in Bulaq. His translation, then, as a translation, must be regarded as little better than a hoax.

And yet, he had a natural gift for narrative, and to fans of Loti and Louys (I must confess myself rather a fan of the latter, at any rate), his volumes maintain their charm.

Edward Powys Mathers, working at secondhand, and accordingly unburdened by such questions of authenticity, was thus paradoxically enabled to produce something of a minor masterpiece in English - an achievement which might be seen to parallel Scott Moncrieff's long-superseded (but still immensely influential) version of Proust's Remembrance of Things Past - or Arthur Waley's Proustian expansion of The Tale of Genji, for that matter: another icon of early twentieth-century translation.

Mathers turned Mardrus' prose poetry into beautifully-turned English rhyming verses, wrote clean, crisp sentences where Mardrus had a tendency to drag,and generally created a Nights appropriate to the Ballets Russes generation.

It's no accident that his four thick volumes are still in print after eighty years, despite the number of exposes of Mardrus's deficiencies. From the point of view of pure entertainment, Mather's remains perhaps the best "complete" version to read.


Mathers, E. Powys, trans. The Book of the Thousand Nights and One Night: Rendered into English from the Literal and Complete French Translation of Dr. J. C. Mardrus. 4 vols. 1949. 2nd ed. 1964. London: Routledge and Kegan Paul, 1972.




[Husain Haddawy (1990)]


6. Husain Haddawy, 2 vols (1990-95)



The Arabian Nights: Based on the Text of the Fourteenth-Century Syrian Manuscript edited by Muhsin Mahdi. New York: W. W. Norton & Co., 1990.

The Arabian Nights II: Sindbad and Other Popular Stories. New York: W. W. Norton & Co., 1995.


Fun thought it is, though, the Mardrus / Mathers edition could not be left to reign unchallenged (beside unassailable, multi-volumed Burton) forever.

The next major development in Nights scholarship came, serendipitously, from the Arabic scholar Muhsin Mahdi. His re-editing of the Ms. Galland revolutionized ideas of what the Nights were or could be said to be. This, to him, was the true original of the Nights, and all subsequent texts had simply been produced by greater or lesser degrees of prompting by European savants.

Husain Haddawy's timely translation of the text established by Mahdi went a long way towards substantiating this claim. The fullness and elaboration of the versions of the core nights stories made it clear how much had been lost by simply following Galland's lead for almost three centuries.

This is a version which should remain in print as a permanently valuable witness to the character of the earliest (and, in many ways, richest) text of the Nights we are ever likely to possess.

Haddawy slightly weakened the effect of his elegant translation by supplementing it a few years later with some versions of Galland's other, spurious additions to the canon, but it's hard to see this as a serious diminution of the value of his work as a whole. The two together make up a very attractive package.


[Malcolm & Ursula Lyons (2008)]


7. Malcolm & Ursula Lyons, 3 vols (2008)

The Arabian Nights: Tales of 1001 Nights. Introduction by Robert Irwin. 3 vols. Harmondsworth: Penguin, 2008.

Which brings us full circle back to Malcolm Lyons. After Hadawy had completed his task of presenting Muhsin Mahdi's "purified" text of the Nights, the only job left was to provide a full and faithful picture of the true nature and contents of the Egyptian ms. tradition, Z.E.R.

I suppose my initial impression of his translation is just how bald Lyons' prose sounds beside the florid zaniness of Burton. What struck me most, though, was how beautifully simple the verses now appeared. There's a great deal of poetry in the Nights, most of it of a rather banal nature, but Lyons has done a splendid job of decoding it for the reader. Payne was certainly a better versifier than Burton, but both felt compelled to try and reproduce the strict forms of Arabic versification. Lyons has dispensed with all that, and the result is charming as well as being far easier to read.

The translation is so far only available in a deluxe hardback edition, limited to 3,000 copies, but it will be coming out in full next year as a Penguin Classic. For the moment, though, the only readily available text is the following sampler, available (at a very reasonable price) from Amazon.uk:

Lyons, Malcolm & Ursula, trans. Three Tales from The Arabian Nights. Harmondsworth: Penguin Classics, 2008.

[Malcolm & Ursula Lyons (2008)]

To sum up, then, if you want to understand the true nature of the Nights themselves, Lyons and Haddawy together give as complete a picture as most of us would ever need of the two major textual traditions.

If your interest is more in the influence of the Nights on European literature, Galland and Burton are the crucial versions, each more of an addition to the literature of their own country than a faithful mirror of the text they were purporting to translate.

The Lane and Mardrus-Mathers translations are each, in their way, very beautiful books - Lane more for his illustrations than his text, Mathers for the crisp conciseness of his prose, and the elegance of his verses.

We're left with Payne, still unjustly overshadowed by Burton. In a way, though, since he did a good deal of the translation work with which the latter is unjustly credited, I guess he could be said to be as much the architect of those 16 eccentric volumes as Burton himself. His translation remains a curiosity of literature, but his status as the first European scholar to provide a full and unvarnished picture of the whole of Z.E.R. cannot be taken away.

There are (of course) other translations and versions, each with their charms. The earlier Penguin translation of selections from Bulaq by N. J. Dawood is still very readable (it's perhaps best sampled as a talking book, beautifully read by Souad Faress and Raad Rawi). Henry Torrens (1839) and A. J. Arberry (1955) each began complete translations which were abandoned after a single volume. Both remain magnificent fragments.

The story will no doubt continue. For the moment, though, my Christmas plans include a good deal of lying around exploring the intricacies and arabesques of Lyons' Nights.

[Malcolm & Ursula Lyons (2008)]

[NB: If you're curious to know more about the history of the 1001 Nights and their influence in general, I try to maintain an up-to-date series of links on the sidebar of my website Scheherazade's Web, which also includes a number of my own essays on the subject.]