Friday, September 21, 2007

Malory and Scheherazade:

[Morte d'Arthur]

A Study in Narrative Method

Needless to say, [Wilkie] Collins’ frame stories come nowhere near equalling the unforgettable plight of Scheherazade. The trouble with the Nights, however, is that the frame is more of a situation than a plot. In this respect at least, After Dark and The Queen of Hearts may be judged superior; they do have a sustained storyline. Moreover, in Collins’ handling of the time-gaining scheme, there is variety in the story-tellers. (Caracciolo, 1988, p.149)
This comment, taken from Peter L. Caracciolo's collection of studies on the influence of the Thousand and One Nights on British culture, The Arabian Nights in English Literature, comes as quite a surprise, even in context. The emphasis of the book up to then has been on the stimulating effect the images and situations of the Nights have had on a group of European writers and artists ranging from Coleridge, Thackeray, Doré and Dickens to Conrad, Joyce, Dulac and Yeats. Any exegesis of the Nights themselves has been left to footnotes and asides: speculations on which version of a particular story Dickens, say, might have seen.

Here, however, the convention is set aside. Caracciolo reveals an irritation with the lack of ‘plot’ (i.e. development) in the overall frame-story of the collection, and concludes with what he considers the unquestionable superiority of the ‘sustained storyline’ and ‘variety in the storytellers’ to be found in Collins’ linked short-story collections After Dark (1856) and The Queen of Hearts (1859).

I don't want to imply that this is an absurd view to hold. On the contrary, it would accord with the taste of most modern readers. Nevertheless, it is difficult to see how Collins’ frame-stories can ‘come nowhere near echoing the unforgettable plight of Scheherazade’ and yet simultaneously ‘be judged superior’ - even if only in terms of ‘sustained storyline’ and ‘plot’.

Another revealing remark can be found in Katharine Slater Gittes’ 1983 PMLA article, ‘The Canterbury Tales and the Arabic Frame Tradition’:

In all Arabic literature, the experiences of a single man, a group of men, or events in one part of the world act as connectives or, in a loose sense, frames. All the works emphasise each fragment of the whole, as well as the whole itself. The Arabs did not demand that knowledge have a focus or that literature have a unified structure in a Greek sense; yet their works exhibit a means of organisation, an external one. It is no wonder, then, that the Arabs ... produced great frame narratives like the Book of Sindbad [sic: for Sindibad] and the Thousand and One Nights. These works, because their original versions are lost, are not useful in this study; however, the idiom “thousand and one,” meaning a large, indefinite number of nights, is in keeping with the Arabic preference for loose, open frames that do not place limitations on the framed material. (Gittes, 1983, p. 242).
The authority cited for this set of generalisations is an article on ‘Literature’ by Franz Rosenthal from the Oxford anthology The Legacy of Islam (1974). This may help us to understand how Katharine Gittes could manage to fall into so many of the traps of ideological Orientalism (as defined by Edward Said (1985)). It is, however, a trifle disturbing to find her article still cited as a principal source of information on the relation between medieval Western and Islamic literatures.

Beginning modestly with the admission that ‘we cannot define the Arabic outlook ... any more than we can the outlook of any other cultural group’, Gittes goes on to say that ‘certain patterns of thought do seem to be peculiar to Arabic literature, art, music, and mathematics’ (1983, pp. 238-39). Among these ‘patterns of thought’ are (of course) the Arabs’ failure to ‘demand that knowledge have a focus or that literature have a unified structure in a Greek sense’, already noted above, but this turns out to be a consequence of the direction in which they read texts:

Because Arabs read numbers as they do texts, from right to left, in reading a large number like 1,034 they read the smallest number (4) first and the largest number (1,000) last. Westerners read the largest number (1,000) first and the smallest (4) last. The Arab must first grasp the units, the parts, before moving on to the whole. A Westerner does the reverse, comprehending first the thousands before moving on to the smallest part, the unit. Arabs, then, tend to view a number as an expanding entity; with the smallest part comprehended first, the progression moves toward the limitless. (Gittes, 1983, p. 240)

Hence, too, the Arabs’ ‘greater sense of boundlessness’.

It's tempting simply to ridicule this line of argument. The ancient Egyptians read texts in whichever direction the characters happened to be facing - does this explain their extreme need for rootedness, manifested in so many pyramids and tombs? Hebrews, too, read from right to left: hence their early discovery of the concept of an omnipotent and omnipresent God ... There are, however, more serious, because more insidious, implications in Gittes’ article as it applies to our particular subject.

Because their original versions are lost' (my emphasis), works such as the Thousand and One Nights ‘are not useful’ to her study of the Arabic Frame Tradition. What precisely does this mean? It means that the comparative study of Arabic and Western frame-stories promised by Gittes’ title is in fact another origin study. What Gittes is telling us is that a line of direct influences must be discovered or postulated between particular works (the Panchatantra, the Canterbury Tales), or their ‘traditions’ (a word useful because so vague) before they can reasonably be compared.

The dearth of information on what Arabic or ‘Oriental’ works Chaucer may have been acquainted with in one form or another (we do not even know if he had read the Decameron), does explain - though it hardly excuses - Gittes’ tendency to dwell on far-fetched anthropological conjectures about numbers and reading from right to left in order to substantiate her lofty claim that ‘the structure of the Canterbury Tales can be most appropriately compared not with the cathedral but with the mosque’ (1983, p. 237) - or, even more picturesquely, with ‘a tradition that originated not in European villages but at distant Bedouin campsites’(p. 250).


Having criticised Caracciolo’s characterisation of the Thousand and One Nights as a (failed) proto-novel, and - in rather stronger terms- Gittes’ ethnological fantasies about the the philosophical tendencies of Arabic culture, the onus now falls on me to propose something better.

I should begin by saying that my own conception of Comparative Literature sees it more as an attempt to weigh up the divergences between different traditions, than an assertion of their common origins.

Seen thus, the point of making detailed comparisons between works from the Islamic and Western spheres of influence is in order to compare their solutions to common problems of technique and style.

Malory, then, is a better subject for my purposes than Chaucer because both the Morte d’Arthur and the Thousand and One Nights are works which have had to contend with the problem of presenting large bodies of prose narrative in a coherent form, with only oblique assistance from the conventions of verse patterning and oral story-telling well-established in each culture.

Chaucer may have a subject in common with the Nights (’The Squire’s Tale’), but there the analogy ends (in terms of technique, at least). Firdausi’s Shah-Nameh (1985) or Attar's Conference of the Birds (1984) might be more helpful referents for him.

One problem with this type of study, however, is that it is forced to pay far greater attention to the detail of the works under examination, and therefore requires more knowledge in both critic and audience. I'd better begin, then, by setting the few indisputable facts about the history and nature of the Nights, against the same basic information about the writings of Sir Thomas Malory.


1/ Editio Princeps:

Caxton (1485) - Galland (1704-1717)[1]

2/ Standard Text:

Eugène Vinaver (1947) - Bulaq Edition (1835)
(Winchester Ms.)[2] - (Zotenberg's Egyptian Recension)[3]

3/ Date of Composition:

c. 1464-1470 - c. 16th Century[4]

4/ Source Material:

The ‘Matter of Britain’ - Middle-Eastern Folktales
(Welsh, Breton, French) - (Persian, Iraqi, Egyptian)[5]

5/ Immediate Source:

French Prose Vulgate - Hazár Afsáneh[6]
(late 12th-13th Century) - (8th-9th Century)

The frame-story of the Nights, then, comes from a Persian collection which had certainly been translated into Arabic by the ninth century , since it's referred to by at least two tenth-century Arabic writers, al-Mas’údí and al-Nadím (Littmann, 1960, p. 361).

Some of the stories - especially fairy-tales ‘in which the ghosts and the fairies act independently’ (Littmann, 1960, p. 363) - are Persian, and may be as old as the collection itself - though Littmann claims an Indian origin for the structural notion of the framestory. Burton (1885) sums up as follows:

The oldest tales, such as Sindibad (the Seven Wazirs) and King Jili’ád, may date from the reign of Al-Mansur, eighth century A.D.

... The thirteen tales mentioned ... as the nucleus of the Repertory ... may be placed in our tenth century.[7]

... The latest tales, notably Kamar al-Zaman the Second and Ma’aruf the Cobbler, are as late as the sixteenth century.

... The work assumed its present form in the thirteenth century. (1885, 10: 88).
The oldest surviving manuscript known in Europe is the Ms. Galland, purchased by him in Syria, and dated to the fourteenth century by Hermann Zotenberg. The main printed editions are as follows:

Calcutta I: an edition of the first two hundred nights only, printed in two volumes at Calcutta in 1814-1818.

Bulaq: the first complete edition, printed at Bulaq near Cairo by the State Printing Office in two volumes (1835).

Calcutta II: also known as the ‘Macnaghten edition’, after its editor. The most complete text of ZER (‘Zotenberg’s Egyptian Recension’), printed at Calcutta in four volumes (1839-1842)

Breslau: edited by Maximilian Habicht, allegedly from a ‘Tunisian recension’ of the Nights, and printed in Breslau between 1825 and 1843. Modern scholarship tends to regard this as an eclectic text constructed from a number of different sources. (Littmann, 1960, p. 360).
European readers, accordingly, had access to the Nights (in the form of Galland’s translation) for at least a century before the text was first printed in Arabic.

In this essay, I have chosen to use Richard F. Burton’s translation as my base text. Despite its notoriety as a manual of sexual lore, and its alleged plagiarisms from John Payne’s complete translation (1882-84), it must still be regarded as the standard edition in English. It owes this status to its completeness (the translation appeared in ten volumes, heavily annotated, in 1885), its inclusiveness (a further six volumes of material from the Breslau and Wortley-Montague texts, as well as the stories by Galland which do not appear in ZER, were published between 1886 and 1888), and its refusal to expurgate or tone down any aspects of the original (including prose rhyme, interminable interpolated verses, and sexual peccadilloes).

Its main drawback as a translation is the archaic and convoluted English which Burton (like Payne before him), saw as appropriate for such a ‘classic’. Mia I. Gerhardt, in her pioneering work The Art of Story-Telling: A Literary Study of the Thousand and One Nights,, therefore prefers to employ Enno Littmann’s six-volume German translation (1976), translating the relevant portions into English as required. This procedure, too, has much to recommend it, as both Burton and Littmann use Calcutta II as their standard text, but only Littmann is sufficiently scrupulous to record all his divergences from the original.


If some people find it distressing to have a work which cannot be assigned to any single author, let me remind them that in another art we are familiar with this sort of thing. I am thinking of a great cathedral, where Saxon, Norman, Gothic, Renaissance, and Georgian elements all co-exist, and all grow together into something strange and admirable which none of its successive builders intended or foresaw. Under Malory’s work lies that of the French prose romancers; under theirs, that of Chrétien, Wace, and other poets; under that, Geoffrey, and perhaps the Breton lais; deepest of all, who knows what fragments of Celtic myth or actual British history? Malory is only the last of many restorers, improvers, demolitionists; if you will, of misunderstanders. (Lewis, 1963, p.25)

The author [of the Nights] is unknown for the best reason; there never was one: for information touching the editors and copyists we must await the fortunate discovery of some MSS. (Burton, 1885, 10: 88)

Another good reason for comparing the story-telling technique of Malory and the unknown redactor/s of the Nights, rather than attempting to ascertain the origins of their materials (Gittes’ approach), or disparaging their artistic prowess by comparison with modern novelists (Caracciolo’s), is this doubt about the true nature of their work.

Caxton, Malory’s first editor, considers him in two roles only: as the chronicler of a real historical king (‘And yet of record remayne in wytnesse of hym in Wales, in the toune of Camelot, the grete stones and mervayllous werkys of yren lyeng under the grounde, and ryal vautes, which dyvers now lyvyng hath seen ... al these thynges forsayd aledged, I coude not wel denye but that there was suche a noble kyng named Arthur’), and as a translator (‘I have ... enprysed to enprynte a book of the noble hystoryes of the sayd kynge Arthur and of certeyn of his knyghtes, after a copye unto me delyverd, whyche copye Syr Thomas Malorye dyd take oute of certeyn bookes of Frensshe and reduced it into Englysshe’ (Vinaver, 1990, 1: cxlv).

Malory undoubtedly reshaped the texts he translated, but he is always careful to give the ‘French book’ as his ultimate authority - supplemented, on occasion, by pieces of geographical and historical information more accessible to him than to the authors of these sources.[8] Like any historian, he is vitally concerned with narrative flow, and like any translator, with making his original read smoothly in the adopted language - even if this means ‘improving’ it in various respects. Asking whether he considered himself a creative writer or not is therefore no longer necessary within these frames of reference (as well, of course, as being a question with little real meaning for any medieval writer).

The reason that my essay’s title juxtaposes Scheherazade, the fictional protagonist of a frame-story, with the historical Sir Thomas Malory is because of the almost infinitely flexible nature of story-collections such as the Nights. There need not be a single discoverable author even for the specific tales contained in our chosen text.[9] As with other folk-tales, one identifiable story or subject can recur in different languages and cultural traditions with appropriate changes made for each era and setting.

Thus, in going from India to Persia, and thence into Arabic and the far-flung world of Islamic popular culture, the Thousand and One Nights retain as their distinguishing feature the situation of the 'nights' themselves - Scheherazade, like Esther in the Bible, talking for her life before a despotic king. That, and the thirteen or so ‘tales common to all texts’ mentioned by Burton.

In order to make up the tally of 1001 nights, not only every manuscript but every translator resorts to addition and subtraction from the vast corpus of stories which it would be possible to include. Having stated this, though, the paradox remains that there is one accepted text for most of the Nights - Zotenberg's Egyptian Recension - presumably compiled in Egypt by a particular editor / scribe in the early sixteenth century (though his role may have been confined to adding one or two stories to an already well-established version). The narrative processes of this text can - even for those, like myself (and Mia L. Gerhardt), with no Arabic - be investigated through the medium of a reliable translation.

I have therefore selected one story from Burton’s translation of the Nights, and one romance from Vinaver’s edition of Malory’s Works in order to perform just such a comparison. They are:

1. ‘A Noble Tale of Sir Launcelot du Lake’
(Vinaver, 1990, 1: 249-87)
[See here for a plot summary and structural breakdown]

2. ‘The Porter and the Three Ladies of Baghdad’
(Burton, 1885, 1: 82-186)
[See here for a plot summary and structural breakdown]
It would not be feasible to discuss these stories in detail, given the space at my disposal. Nevertheless, points which need to be made about the relationship between them, within the context of their ‘master-narrative’, have persuaded me to choose a diagrammatic representation as the best way of conveying the maximum amount of information in the minimum space. I shall thus, I hope, also be enabled to use one story from each work as the template for discussion of the others.

Of ‘The Noble Tale of Sir Launcelot du Lake’, Eugène Vinaver says:

The Tale falls into three distinct sections of approximately equal length, each corresponding roughly to two folio leaves of Malory’s ‘French Book’. The three sections belong to three different parts of the text. It is tempting ... to speculate that Malory’s source was simply a gathering of three leaves which had dropped out of a volume of the Prose Lancelot. The alternative, and the more likely, explanation is that he deliberately chose those three sections in order to give a moderately continuous account of Lancelot’s adventures and so avoid the typical ‘cyclic’ method of interweaving a variety of different themes. (1990, 3: 1408)
He sees the ‘three distinct sections’ as 1/ from the departure with Lionel until he leaves the court of King Bagdemagus, a series of episodes which ‘centre upon the redoubtable figure of Tarquyn’ (1990, 3: 1408-9); 2/ the killing of Tarquyn and Perys de Foreste Savage and the liberation of the people of Tintagel from the two giants; and 3/ ‘As for the remaining portion of the Tale ... only the first six pages of it have a parallel in the Lancelot’(3: 1410). As will be clear from my analysis, I do not see the actual working of the story in quite these terms, though I accept that these are the divisions in Malory’s source-materials.

Elsewhere, in a discussion of the digressive nature of French prose romance, Vinaver characterises it in terms of ‘the technique of tapestry. Just as in a tapestry each thread alternates with an endless variety of others, so in the early prose romances of the Arthurian group numerous seemingly independent episodes or “motifs” are interwoven in a manner which makes it possible for each episode to be set aside at any moment and resumed later’ (1990, 1: lxvi). Vinaver goes on to explain that the most convenient way of representing this interweaving is to designate each strand or motif by a letter, giving various examples in the discussion of the ‘Tale of King Arthur’ in his introduction (1990, 1: lxviii-lxi).

By comparison with the French romancers, then, in Malory’s version of the story ‘The order of events is not a1 b c1 a2 c2 , but a1 a2 b c1 c2’; the three threads of the narrative are unravelled and straightened out so as to form in each case a consistent and self-contained set of adventures’ (Vinaver, 1990, 1:lxx). Vinaver sees this as ‘closely approximating to the conventional modern technique of exposition’ (1990, 1: lxxi), which seems to me an over-simplification, but I shall be returning to this point later. Applying Vinaver’s notation to the ‘Noble Tale of Sir Launcelot du Lake’, I find the first section at least a perfect example of interlocking or ‘nested’ motifs:

A - ‘Sir Tarquyn’
B - ‘Bagdemagus’
C - ‘Sir Belleus’
B - ‘Bagdemagus’
A - ‘Sir Tarquyn’
In other words, Launcelot and Lyonell set out together, but Lyonell has the first adventure (A) - being defeated by Sir Tarquyn while Launcelot is sleeping under the apple-tree.[10] The latter wakes to find himself a prisoner of Morgan le Fay, but is freed by the first ‘damsel’, the daughter of King Bagdemagus (B) on condition he helps her father in a tournament. On the way there he has his misadventure with Sir Belleus and his ‘lemman’ (C). After the tournament, Launcelot meets a second damsel, who leads him to Sir Tarquyn (A), and then (after the prisoners have been freed by Gaheris) to the equally wicked Sir Perys. This damsel is therefore the bridging device between the closing of the first set of motifs and the opening of the second:

A - ‘Sir Gaheris’
B1- ‘Sir Perys’
B2 - ‘Tintagel’
C1 - ‘Sir Kay’
C2 - ‘Chapel’
B3 - ‘Sir Phelot’
B4 - ‘Pedyvere’
A - ‘Queen’
The killing of Sir Perys (B1) is therefore linked to that of Sir Tarquyn (A) by this common intermediary, but Launcelot’s next, seemingly unrelated adventure (B2) is also related to the Tarquyn episode by the fact that there are ‘three score’ knights kept prisoner by him in his castle just as there are ‘three score of ladyes and damesels’ imprisoned by the two giants in Tintagel. I therefore read these two episodes as parallel ‘B’s’, as it were, rather than as separate motifs in their own right.

We then have the triple conflict with three groups of ‘three knights’ - first by Launcelot in defence of Kay, then disguised as Kay, and then against Ector, his own cousin, and three of his best friends (C1). This, in its turn, is parallelled by the triple encounter with, first, the dead Sir Gylberde the Bastarde, then the wounded Sir Melyot de Logyrs, and then the Chapel Perilous itself, with its presiding sorceress Hallewes (C2); again, her declaration of love to Launcelot parallels that of the damsel-guide after the encounter with Sir Perys.

Two more incidents, each involving a lady and her knight - the first lady helping her husband try to kill Sir Launcelot (B3), the second lady seeking Launcelot’s protection against her husband (B4) - invert each other perfectly. In the first case the knight is killed, in the second case the lady. I have classified them as ‘B’s’ because of the linking-up with the theme of the oppression of damsels and ladies (as in the case of Sir Perys and the giants of Tintagel) in the episode of Sir Pedyvere, but there is obviously a less powerful sense of interlocking stories here than in the first section of the story.

Finally, in Launcelot’s arrival back at Camelot, we have a resumé of the various adventures and stories to date.

While the ‘Noble Tale’ does have some features of the ‘conventional modern technique of exposition’ claimed for it by Vinaver, perhaps what is most interesting about the story is the over-determined nature of its narrative method. Repetition of numbers (especially groups of three - or three plus one) and motifs (helping one member of a couple against the other - or being used by one of them to benefit the other) is far more frequent and pervasive than would be common in modern fiction.

Indeed, if one goes on to compare this story with ‘The Tale of Sir Gareth’, one finds, again, number symbolism of a nature which makes more sense as a structuring device than as a contribution to the themes of the narrative (each of the knights of successive colours - Black, Green, Red and Blue - defeated by Gareth has a larger number of retainers whom he pledges to the latter’s service), as well as a set of repetitive motifs which can make it a more rewarding subject for the annotator or editor than the casual reader (first, the succession of duels against multi-coloured knights; then, the various love-trysts with Dame Lyonesse disturbed in various ways; and finally, the attempts to avoid Sir Gawain in order to conceal his true identity).

So much for Vinaver’s analysis of Malory; now let's apply these considerations to the Nights:

Andras Hamori, in his On the Art of Medieval Arabic Literature, has the following comments to make about the ‘Porter and the three Ladies of Baghdad’, to which he devotes an entire chapter:

The reader will immediately notice that the tale is a series of echoes and reflections ... My purpose is to examine the relation between the formal coherence of events in the story and the storyteller’s evaluation of the moral coherence among the events he narrates. It could be argued that the structural properties of the tale, the neat and interesting relations among motifs and variations on motifs, are not unlike musical relations, and that the pleasure the audience derives from them is a musical pleasure. It could also be argued that periodicity is a storyteller’s device for holding the audience’s attention. You know that something is destined to happen again, but you also know that it will come in by a different door or even a window. Of course you are interested to see how an analogous result comes about through a different process, but you are not necessarily concerned with the good and evil of it, or with the question whether or not the actors in the different episodes deserved the same fate. (Hamori, 1975, p. 172)
Hamori’s argument about the moral elements concealed in the seemingly random universe of the ‘Three Ladies’ is ingenious in the extreme, and he deserves credit for being virtually the first to take the tale’s original teller seriously as an artist.

Mia Gerhardt, by contrast, does little more than speculate that ‘it was the original creator of the story who made an ambitious effort to set up a sort of double frame [‘the well-known plot that turns on the presence of Harûn incognito; on the other hand, the ransom device that is intended to motivate the telling of the mendicants’ stories’], but, having attempted too much, did not quite succeed’ - basing this conclusion mainly on the asymmetrical absence of a story for the lady-cateress (1963, pp. 408-10).

Another of Hamori's major points is:

The invariants in the dervishes’ narratives ... all three are blind in one eye ... Each is a prince; each has been involved with a splendidly equipped underground hiding place; each has contributed in one way or another to the destruction of the occupants of these catacombs. (1975, p. 173)
One reason why I find myself finally unconvinced by the subtly nihilistic motives detected by Hamori in the mass marriage which concludes the whole intertwined set of stories - the just and the unjust alike paired off by the Caliph according to symmetry rather than mutual inclination - is because the ‘structural properties’ of ‘motifs and variations on motifs’ which form so large a part of his argument can be parallelled not only among the other stories of the Nights, but also in Malory.

I have already discussed the number symbolism and repetitive motifs to be found in the ‘Noble Tale of Sir Launcelot’ and the ‘Tale of Sir Gareth’. As befits a story on a larger scale, ‘The Porter and the Three Ladies of Baghdad’ plays the changes on seven as well as three. The repetitive nature of certain motifs in the story is not altered by the fact that here events tend to be doubled or mirrored rather than simply repeated. The idea of a frame-story, or a series of stories told by the protagonists of other stories is something for which the Nights is justly celebrated - but in narrative effect, it could surely be parallelled with the ‘nesting’ technique of strands of different stories employed by Malory?[11]

I can claim little originality in applying the analytical methods devised by Vinaver and Hamori to the very texts which they discuss. Already. however, I think it can be seen that an exchange of information between these two systems can be mutually illuminating. Compared with the French Prose Vulgate version of the Arthurian romances, Malory does indeed look ‘modern’ - but when the doubled motifs and the (surely thematically arbitrary) repetitions of numbers such as ‘three’ and ‘three plus one’ and ‘three-score’ are seen in other contexts as well, it is clear that another term must be found to characterise this narrative method.

Similarly, seen in isolation, it is tempting to postulate a cunning subversion of the dominant values of ‘romance’ (love, marriage, loyalty, kingship) in the unnaturally symmetrical patterns of the ‘Three Ladies of Baghdad’ - but when an almost equally obsessive attention to such details of externalised structure can be found in ‘The Tale of Sir Gareth’, substantiating such an argument becomes far more difficult.


In conclusion, then, it remains to offer at least some possible reasons for the common patterns in these two examples of medieval story-telling.

Let me begin by saying that I want to disavow any intention of claiming these features of numerical repetition, doubling of motifs, and nesting or framing stories as the discovery of some ‘link’ between these two literary traditions. On the contrary, I see them as two not dissimilar responses to a not dissimilar problem of literary technique - one which arises (I imagine) inevitably at the point of intersection between oral and written story-telling.

Both Malory’s romances and the stories of the Thousand and One Nights are intensely written texts - copied (with modifications) from other writings which were themselves responses to earlier written texts. The French Prose Vulgate was itself an elaboration of earlier layers of Arthurian romance in both verse and prose. The Nights had the Persian Hazár Afsáneh as its immediate original, but finally grew to contain entire cycles of stories, both Arabic and Persian, within its compass.

Nevertheless, regardless of the literacy of Sir Thomas Malory or the ZER scribe’s immediate circle of acquaintanceship, their works were intended (whether primarily or secondarily) to be read aloud to an audience. Thus, the patterns within their narratives had to be comprehensible when communicated orally.

Hence, I imagine, what I have referred to above the ‘over-determined’ nature of both sets of stories. Hamori’s remark about the ‘musical pleasure’ to be derived from contemplation of the ‘relations among motifs and variations on motifs’ is, to my mind, more accurate than he realises. Like music, story-telling depends on the possibility of grasping the structure as well as the discrete nature of the sounds which one listens to.

At this point it might be as well to make some acknowledgement to the one precedent I am aware of for this particular comparison, the Rev. Cameron Mann’s ‘The “Thousand and One Nights” and the “Morte d’Arthure”’, published in the North American Review on January 18th, 1907. The Bishop of North Dakota concludes his article by avowing:

Yes, it would be hard to find two other books so alike in their origin - each a composite of myths and legends, each with a strict theological creed, each with its Bible in the background and its Paradise ahead, yet so utterly unlike and repugnant in their contents. In the one, we enter a palace where fountains plash in the court, where the walls reek with glowing and erotic decoration, where wild music clangs and wilder dancers spin .... where silk-clad men and women idle out the time with libidinous toying and coarse jest. In the other, we wander through the grave forest, where we meet, now a gladsome company who have been a-maying and return laden with blossoms, now a priest with solemn eyes bent upon his breviary, now a stalwart knight in full armor riding after some perilous but noble deed. (Mann, 1907, pp. 155-56)
Strangely enough, while I remain unmoved by his argument that ‘these two books stand for the ideals of their respective communities; for what the disciples of Jesus and the disciples of Mahomet felt and wished, long after the founders of their religion had passed away’ (1907, p. 151), I feel that there is a certain amount to be said for the manner in which Mann conducts his discussion. The difference between us is, of course, that he wants to contrast the systems of morality implicit within the two works, whereas I'm setting out to compare their respective narrative methodologies.

Even in these terms, I feel the Reverend Mann overestimates the extent to which the Nights are ‘thoroughly, unblushingly, callously sensual’ (p.152). He certainly overestimates the moral virtues of Malory (perhaps a clue lies in the fact that the article ends with an invocation of Tennyson’s Arthur as the ‘faultless man’ (p. 156) - The Idylls of the King would actually suit his argument far better than the deadpan violence of the ‘Noble Tale of Sir Launcelot’).

Nevertheless, he does set two highly-developed products of different cultural traditions against each other with the explicit intention of increasing our understanding of both (however tendentiously), and this still seems to me preferable to the mystical quest for their common origins or descendants which has dominated so much subsequent criticism.


1. Antoine Galland's translation appeared in 12 volumes between 1704 and 1717. The name ‘Arabian Nights’ Entertainments’ was first coined by an anonymous English translator of this version (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 ... first published in London between 1706 and 1717; my edition dates from Edinburgh, 1781.

2. The Winchester manuscript of Malory's works was first discovered in 1934, and subsequently (1990) edited by Eugene Vinaver. This is the text to which I shall be referring throughout.

3. The so-called ‘Vulgate’ or standard text of the Nights, derived from an Egyptian version of the whole work, was first identified by Hermann Zotenberg, keeper of Eastern manuscripts at the Bibliothèque Nationale in Paris. It was first employed by Edward William Lane for his selected and expurgated version (1839-1841), and later by Francesco Gabrieli for his complete Italian translation (1972).

4. I should perhaps say ‘compilation’ rather than ‘composition’ because all of the elements of this edition are far older (as are Malory’s sources). Enno Littmann (1960, pp. 358-64), gives an account of the various Arabic editions and their source-material.

5. See also, in this connection, E. J. Ranelagh’s The Past We Share (1979, pp. 195-242) for an amusing account of various parallels between the Nights and folk-tales from different traditions.

6. This Persian collection (’A Thousand Tales’) is no longer extant, but its frame-story at least had been translated into Arabic by the ninth century - see Manzalaoui (1973, I: 39).

7. Burton (1885, 10: 78): ‘the tales common to all [texts and MSS.] are the following thirteen: - 1. The Introduction ... 2. The Trader and the Jinni ... 3. The Fisherman and the Jinni ... 4. The Porter and the Three Ladies of Baghdad ... 5. The Tale of the Three Apples. 6. The Tale of Núr al-Dín Ali and his son Badr al-Dín Hasan. 7. The Hunchback’s Tale ... 8. Nur al-Dín and Anís al-Jalís. 9. Tale of Ghánim bin ‘Ayyúb ... 10. Alí bin Bakkár and Shams al-Nahár ... 11. Tale of Kamar al-Zamán. 12. The Ebony Horse; and 13. Julnár the Seaborn.’

8. The information about Camelot's now being called ‘Wynchester’ in "The Tale of King Arthur", for example (see Vinaver, 1990, 1: 92). Caxton disagrees, thinking it a town ‘in Wales’.

9. This is illustrated by Burton (1886-88, 15: 45) when he notes of a story which is a ‘replica’ of one in Volume IV: ‘I have retained it on account of the peculiar freshness and naïveté of treatment which distinguishes it, also as a specimen of how extensively editors and scriveners can vary the same subject’. The search for the author of a ‘subject’, rather than the actual text of a story, must be the object of any quest for the origins of the various components of the Thousand and One Nights.

10. Refer to my summary of the ‘Tale’ for a more precise demarcation of these episodes.

11. The fact that Vinaver describes this as a corruption of the twelfth century ‘tapestry’ technique, and as a move in the direction of the ‘modern’ is neither here nor there. Frame-stories, too, existed both before and after the Thousand and One Nights. My intention in these analyses of different stories is to point to the workings of these techniques in actual texts, not their possible provenance or authority.

Works Cited:

  • Abbott, Nabia. (1949). A Ninth-Century Fragment of the 'Thousand Nights.' Journal of Near-Eastern Studies, 8: 129-64.

  • Attar, Farid ud-Din.(1984). The Conference of the Birds. Trans. Afkham Darbandi and Dick Davis. Harmondsworth: Penguin.

  • Burton, Richard F, trans. (1885). 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. N.p. [= Boston]: The Burton Club, n.d.

  • Burton, Richard F., trans. (1886-88). Supplemental Nights to the Book of the Thousand Nights and a Night with Notes Anthropological and Explanatory. 6 vols. Benares [= Stoke-Newington]: Kamashastra Society. 7 vols . N.p. [= Boston]: The Burton Club, n.d.

  • Caracciolo, Peter L., ed. (1988). The Arabian Nights in English Literature: Studies in the Reception of The Thousand and One Nights into British Culture. London: Macmillan, 1988.

  • Chaucer, Geoffrey. (1989). The Riverside Chaucer. Third Edition. Ed. Larry D. Benson. 1987. Oxford: Oxford University Press.

  • Ferdowsi. (1985). The Epic of the Kings: Shah-Nama, the National Epic of Persia. Trans. Reuben Levy. 1967. Rev. Amin Banani. London: Routledge & Kegan Paul.

  • Gabrieli, Francesco, ed. (1972). Le mille e una notte: Prima versione integrale dall’arabo. Trans. Francesco Gabrieli, Antonio Cesaro, Constantino Pansera, Umberto Rizzitano and Virginia Vacca. 1948. Gli struzzi 35. 4 vols. Torino: Einaudi.

  • Gerhardt, Mia I. (1963). The Art of Story-Telling: A Literary Study of the Thousand and One Nights. Leiden: E. J. Brill.

  • Gittes, Katharine Slater. (1983). The Canterbury Tales and the Arabic Frame Tradition PMLA 98: 237-51.

  • Hamori, Andras. (1975). On the Art of Medieval Arabic Literature. Princeton, NJ: Princeton University Press.

  • Lane, Edward William, trans. (1839-41). 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.

  • Lewis, C. S. (1963). The English Prose Morte. In J. A. W. Bennett, ed. Essays on Malory. Oxford: Oxford University Press. Pp.7-28.

  • Littmann, Enno. (1960). Alf Layla wa-Layla. In The Encyclopaedia of Islam. Leiden & London: E. J. Brill. Pp. 358-64.

  • Littmann, Enno, trans. (1976). Die Erzählungen aus den Tausendundein Nächten: Vollständige deutsche Ausgabe in zwölf Teilbänden zum ersten mal nach dem arabischen Urtext der Calcuttaer Ausgabe aus dem Jahre 1839 übertragen von Enno Littmann. 1921-28. 2nd ed. 1953. 6 vols in 12. Frankfurt am Main: Insel Verlag.

  • MacDonald, Duncan B. (1924). The Earlier History of the Arabian Nights. Journal of the Royal Asiatic Society, 1924: 353-97.

  • Malory, Sir Thomas. (1953). Le Morte d’Arthur. 1485. Everyman’s Library. 2 vols. London & New York: Dent & Dutton.

  • Mann, Rev. Cameron. (1907). The Thousand and One Nights and the Morte d’Arthure. North American Review, 184: 150-56.

  • Manzalaoui, Mahmoud A. (1973). The Arabian Nights. Cassell’s Encyclopaedia of World Literature. 3 vols. London: Cassell. 1: 38-41.

  • May, Georges. (1986). Les Mille et une Nuits d’Antoine Galland, ou le chef-d’oeuvre invisible. Paris: PUF.

  • Mosès, François, ed. (1991). Lancelot du Lac: Roman français du XIIIe siècle. Lettres gothiques. Paris: Livre de Poche.

  • Payne, John, trans. (1882-84). 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.

  • Ranelagh, E. J. (1979). The Past We Share: The Near Eastern Ancestry of Western Folk Literature. London: Quartet.

  • Said, Edward W. (1985). Orientalism. 1978. Harmondsworth: Penguin.

  • Schacht, Joseph, and C. E. Bosworth, ed. (1974). The Legacy of Islam. 2nd ed. Oxford: Clarendon Press.

  • Sommer, H. Oskar, ed. (1909-16). The Vulgate Version of The Arthurian Romances, Edited from Manuscripts in the British Museum. 8 vols. Washington, D. C.

  • Vinaver, Eugène, ed. (1990). The Works of Sir Thomas Malory. 3 vols. 1947. 3rd ed. rev. P. J. C. Field. Oxford: Clarendon Press.

[This paper, under the title “Malory and Scheherazade: A Study in Narrative Method” was read as a Department of English Staff Seminar at Auckland University on 16 April, 1992.]

No comments: