Version classiqueVersion mobile

The Anglo-Scottish Ballad and its Imaginary Contexts

 | 
David Atkinson

7. Palimpsest or texte génétique

Note de l’éditeur

An earlier version of this chapter was published as ‘Genetic Foundations for a Palimpsest Model of the Anglo-Scottish Ballad Text: Evidence from the J. M. Carpenter Collection’, Folklore, 121 (2010), 245–67, and the material is reused with permission.

Texte intégral

1The ballad, whatever else it may be, is an aesthetic artefact. Like other works, literary and musical alike, the same ballad can be presented in a number (perhaps any number) of different ways, in manuscript, print, recording, or performance. Of course, what constitutes the ‘same’ is potentially problematic, but common practice is to link ballad type with source and to fall into the habit of speaking of ‘so-and-so’s version of such-and-such a ballad’. Notwithstanding the fact that the ballad, a genre associated with performance, is amenable to unlimited successive and evanescent renditions, and can quite readily vary between one and the next, a silent equation is thus being made between the collected item and the agency or intention of its contributor.

  • 1 That is not wilfully to neglect altogether the role of local ballad composers. See, for example, Ed (...)
  • 2 OED palimpsest, n. 2 a, b.

2Implicit in this standard approach is the tolerance of a degree – but only a quite limited degree – of variance in what one is talking about. Commonly, all that is extant is a single sound recording or set of music and words taken down by a collector, and the name of the contributor stands as a, seemingly unproblematic, organizing principle. The contributor is deemed well-nigh equivalent to the author or composer of a conventional literary or musical work. And yet that is not the case. However ethnographically and artistically important the individual might be, the whole point of most of the ballads under discussion here is that the individual is not its originator.1 And however creative the contributor, the ballad remains a palimpsest of precedent and current words and melody – ‘palimpsest’ in its technical sense being ‘A parchment or other writing surface on which the original text has been effaced or partially erased, and then overwritten by another; a manuscript in which later writing has been superimposed on earlier (effaced) writing’, and in less formal, extended use ‘a thing likened to such a writing surface, esp. in having been reused or altered while still retaining traces of its earlier form; a multilayered record’.2

3Mostly, this is something of an abstract idea, revealed in part by historical research into ballad words and melodies, but still requiring exercise of the imagination. However, the J. M. Carpenter ballad collection, described in the previous chapter, offers something of a special case, where manuscript materials show evidence of possibly multiple layers of revision, and sound recordings differ sometimes substantially from their manuscript equivalents, so that it is suddenly no longer so easy to make the instinctive equation between item and contributor intention. Some editing of the confusing layers of materials seems essential if the collection is to be of value to those who might be expected to want to make use of it – as potential performers, family and social historians, scholars wanting to carry out comparative and other kinds of studies.

  • 3 See, for example, Almuth Grésillon, Éléments de critique génétique: Lire les manuscrits modernes (P (...)
  • 4 Almuth Grésillon, ‘Slow: Work in Progress’, Word & Image 13 (1997), 106–23 (p. 106).

4The Carpenter ballads have shown themselves amenable to the social theory of text, but they also offer a possibly unique opportunity to consider the ballads in relation to the insights of another area of editing theory, the French school of genetic criticism, or critique génétique.3 The defining characteristic of genetic criticism, or textual genetics as it is sometimes called, is that the work, as manifest in documents such as literary manuscripts (most of the manifesto statements about critique génétique refer to literary texts and works, but the theory has wide-ranging application across the creative arts), is envisaged not just as product but as a process. Thus the preference of the genetic critic is for ‘production over the product, writing over what is written, textualization over the text, multiplicity over uniqueness, possibility over the finite, virtuality over the ne varietur, the dynamic over the static, the operation over the opus, genesis over structure, the strength of the act of writing over the form of the printed word’.4 Genetic criticism ties the material basis of literature (the text in the document) to what can be inferred about the process that brought it into being:

  • 5 Deppman, Ferrer, and Groden, eds, Genetic Criticism, p. 2.

Like old-fashioned philology or textual criticism, it examines tangible documents such as writers’ notes, drafts, and proof corrections, but its real object is something much more abstract – not the existing documents but the movement of writing that must be inferred from them. Then, too, it remains concrete, for it never posits an ideal text beyond those documents but rather strives to reconstruct, from all available evidence, the chain of events in a writing process.5

  • 6 Grésillon, ‘Slow: Work in Progress’, p. 106.

5The critique génétique project endeavours to recover, or elucidate, the dynamics seemingly obscured within texts, written, typed, printed, or mechanically recorded. ‘Its vision is of literature as a doing (“unfaire’”), as an activity, as a movement.’6

  • 7 Samuel Johnson, The Lives of the Most Eminent English Poets, with critical observations on their wo (...)
  • 8 Deppman, Ferrer, and Groden, eds, Genetic Criticism, p. 3.
  • 9 Edgar Allan Poe, ‘The Philosophy of Composition’, in Selected Writings of Edgar Allan Poe: Poems, T (...)
  • 10 René Wellek and Austin Warren, Theory of Literature, 3rd edn (Harmondsworth: Penguin, 1963), pp. 90 (...)

6The general approach is not, in fact, an entirely unique or recent one, and it has some eminent forerunners. Samuel Johnson, for example, observes in his ‘Life of Milton’, ‘it is pleasant to see great works in their seminal state, pregnant with latent possibilities of excellence; nor could there be any more delightful entertainment than to trace their gradual growth and expansion, and to observe how they are sometimes suddenly advanced by accidental hints, and sometimes slowly improved by steady meditation’.7 Edgar Allan Poe’s 1846 essay on ‘The Philosophy of Composition’, considered a foundational text of French genetic criticism,8 considers how valuable it would be were an author to ‘detail, step by step, the processes by which any one of his compositions attained its ultimate point of completion’– and then proceeds to do just that for his own poem ‘The Raven’.9 Genetic criticism, however, really gained traction, in France especially, in the last quarter of the twentieth century, partly as a response to the potential apparent in the vast array of surviving manuscript materials relating to important modern authors such as Zola, Balzac, Hugo, Valéry, Flaubert, and Proust, and, in English, Joyce, Yeats, Beckett, and Eliot. In this respect, genetic criticism stands in acute contrast to the mid-century New Critical insistence on the integrity of the finished work, to which earlier drafts were considered as, at best, marginal, discarded, alternative versions.10

  • 11 Jean Bellemin-Noël, Le Texte et l’avant-texte: Les Brouillons d’un poème de Milosz (Paris: Larousse (...)
  • 12 Pierre-Marc de Biasi, ‘What Is a Literary Draft? Toward a Functional Typology of Genetic Documentat (...)
  • 13 Grésillon, ‘Slow: Work in Progress’, p. 115; Laurent Jenny, ‘Genetic Criticism and its Myths’, Yale (...)

7Genetic criticism considers together all of the different, interdependent, dynamic states of a work that are available, up to and including the point of publication. These may be comprised, variously, in such things as manuscript drafts, notebooks, outlines, sketches, letters, corrected proofs, which all belong temporally prior to the ‘achieved’, published text. These are conceived of not as superseded matter that is external to the literary work, but as an integral part of its ontology, designated accordingly by the new term avant-texte, originally coined in 1972 by Jean Bellemin-Noël.11 Generally, genetic critics do acknowledge a dividing line between this system of avant-texte and the ‘frozen shape of a published text’, marked by that moment when the author signs off the text as bon à tirer, or ‘passed for press’.12 At that moment the endlessly pliable avant-texte gives way to a particular manifestation that passes into the public domain, after which the reality of its existence cannot be altered, although that fact need not in itself preclude subsequent modifications in the form of further published editions, author’s handwritten annotations, and so forth. In practice, though, genetic criticism offers a somewhat ambiguous perspective on the relationship between avant-texte and the supposed authority of the published text, between process and product.13

  • 14 Jerome J. McGann, A Critique of Modern Textual Criticism (Charlottesville: University Press of Virg (...)
  • 15 Cited in D. C. Greetham, Theories of the Text (Oxford: Oxford University Press, 1999), p. 60.
  • 16 Hans Walter Gabler, ‘The Text as Process and the Problem of Intentionality’, Text: Transactions of (...)
  • 17 de Biasi, ‘What Is a Literary Draft?’, p. 58.

8Now, it is true that a tradition of Anglo-American textual editing deriving from nineteenth-century philology has long been concerned with recording variant readings. However, variant readings have characteristically been subordinated to an idea of the author’s ‘final intentions’ as embodied in one particular state of the text.14 The variants themselves are recorded in a suitably subordinate position, usually in a critical apparatus at the back of the book or the foot of the page. Genetic criticism, in contrast, is concerned not so much with the variants per se, though it may take account of the aesthetics of different textual states, as with what Daniel Ferrer calls the ‘injunctions’ that appear in the acts of accretion, deletion, and superimposition that the variants represent.15 Together, the variants establish a matrix of binary on/off switches as each variant modifies and is modified by its context. Thus in genetic criticism the object of study becomes the dynamic process of artistic ‘invention’. In that sense, it entirely reverses the perspective of Anglo-American textual criticism geared towards the ultimate shape of the text that best embodies the author’s final intentions.16 Instead, textual genetics seeks to grasp the inherent fluidity, and, as a corollary, the diachronicity, of the literary work. Pierre-Marc de Biasi refers to the way in which literary genetics restores the temporal dimension to criticism, so that ‘the text of the work reclaims possession of its history’.17

9Though scarcely on the scale of, for example, the Flaubert or Joyce manuscripts, it is nonetheless the fact of the presence of a mass of avant-texte among the Carpenter ballads, as described in the previous chapter, that has prompted the idea of applying some of the principles of genetic criticism to the investigation of a theory of text for the Anglo-Scottish ballad. And while the Child ballads in the collection provide the most compelling case, many of the other collected items exist in more than one iteration of the ‘same’ thing, and further examples could be drawn, for example, from among the folk plays. So the textual model described below as a ‘palimpsest’ potentially has a broader application.

  • 18 Washington, DC, Library of Congress, American Folklife Center, Archive of Folk Culture, AFC1972/001 (...)
  • 19 Carpenter Collection, Cylinder 131 04:21; Disc side 310 03:05; MS pp. 04924–04926, 06997–06998, 084 (...)

10Now, as already observed (and notwithstanding the objections raised in chapter six), a particular song as taken down from a particular contributor is most usually said to constitute that person’s version of that particular song type. It would be normal practice to speak of, say, Sam Bennett’s ‘version’ of ‘Our Goodman’ (Child 274),18 or of Sarah Phelps’s ‘version’ of ‘The Outlandish Knight’ (Child 4),19 when they are in fact represented by several variant iterations. But none of those written or recorded iterations are texts ‘intended’ for publication or for issue as recordings, and there is nothing definitive about any one particular rendering of either words or tune. The words on the sound recordings might differ more or less noticeably from those the contributors dictated, and the user of the collection is therefore obliged to consider the two as being at least as distinct – in principle, if not in degree – as the acknowledged versions of certain literary texts – say, the second quarto and folio Hamlet. To minimize potential confusion over terminology, the sound recording and the dictated text could be referred to as different ‘renditions’ of the ‘same’ thing, which are to be considered of equivalent status – this despite the fact that their genesis may be separated in time by no more than a matter of a minute or two. In contrast, the fair copy typescripts derive directly from the rough copy typescripts, even though their genesis may have been separated in time by a space of several years, so those texts need to be considered as further iterations of the same renditions, albeit characteristically in a variant textual state. Much the same can be said of the relationship between the sound recordings and Carpenter’s own music transcriptions, and between the sound recordings and modern transcriptions that can be made from them.

  • 20 For a general introduction to Carpenter’s folk play collection, see Eddie Cass, ‘The James Madison (...)

11The folk plays, likewise, are in typescript but with handwritten alterations, which appear to reflect corrections of a nature more or less comparable to those in the ballad rough copies.20 At least one play (the Lower Heyford play) was recorded in its entirety, as well as taken down from dictation, once again providing different renditions of equivalent status. In other instances, a play from a particular location was taken down from more than one contributor (the Hunton sword dance play, for example). This latter situation is not, however, directly comparable to that of the same ballad type taken down from different contributors, because the implicit demands of collaborative dramatic performance can be presumed to imbue the text-as-learned with much greater authority than in the ballad case. Rather, these instances might be considered again as multiple renditions of the same play.

  • 21 The full set of references is Carpenter Collection, Cylinder 130 11:12; Disc sides 308 04:25, 309 0 (...)
  • 22 Carpenter Collection, MS p. 10333.

12Some of the difficulties of taking an intentionalist approach to this sort of material have been outlined in chapter six, so here, instead, it will be useful to turn to a specific example to explore the nature and potential of the ballad avant-texte. The range of choice among the ballads is huge and it should be emphasized that this one may or may not be the best, or the most characteristic, example. It is, however, one that displays a good range of kinds and concentrations of textual variants. The ballad in question is ‘The Bonnie Banks o Fordie’ (Child 14) and the contributor Mrs Alexander McEwan (spelling unconfirmed).21 Although he does not name names, in the course of several drafts of essays present in the collection, Carpenter writes about a couple who, on the basis of the repertoire items mentioned, must be Mr and Mrs McEwan. Apparently, they were Scottish Travellers, and his description of them becomes more than a little romantic: ‘She was a lithe attractive creature, with her quick animation and flashing black eyes. When I first saw her, in the morning, wandering along a green country lane, glancing swiftly from side to side as she laughed and bant [er] ed with her elderly husband, I had thought, somehow, of a wild animal – perhaps a deer.’22

13The ballad in question exists in a characteristic number of renditions and iterations:

  • a rough copy typescript (designated T1) (Fig. 7.1)

  • a fair copy typescript (designated T2)

  • a cylinder recording, comprising three stanzas

  • a lacquer disc copy of the recording

  • a transcription of one stanza from the sound recording

  • stanzas quoted in draft essays.

Fig. 7.1 Carpenter Collection, MS pp. 04384 – 04387. Courtesy of the American Folklife Center at the Library of Congress, Washington, DC, USA.

14The manuscript pages carry several different titles – ‘The Bonnie Banks o Fordie’, ‘The Banks o Eldrie’, ‘Bubblin Johnnie’– the last of which, taking the name of the outlaw character, is possibly unique as an identifier for this particular ballad. For ease of reference, a critical edition text, founded on T2 as base text and with a simplified critical apparatus, is given in Appendix 1 at the end of this chapter. Stanza and line numbers cited below are keyed to this text, indicated by crit. ed.

15The assumption is that T1 represents what Carpenter took down on his portable typewriter and amended in consultation with Mrs McEwan. The sound recording is evidently of a somewhat different nature, not just because it was created at a different, albeit probably quite close, moment in time, but more importantly because it derives from the process of singing as opposed to that of dictation. Pragmatically, these can be treated as separate renditions and therefore presented independently, which conveniently circumvents the difficulty posed by differences between the words as sung and as dictated. A transcription from the sound recording, alongside the corresponding lines from the critical text, is given in Appendix 2.

  • 23 David Atkinson, ‘The Secret Life of Ballad Manuscripts’, Variants, 8 (2012), 183–206 (esp. pp. 186– (...)

16The rough copy T1 carries alterations made both in ink and in red pencil, the cautious assumption (founded upon general impressions drawn from increasing familiarity with the entire collection) being that alterations in ink were made more or less at the time of dictation, while those in red pencil may have been made at a later date.23 As far as stanza 10 – that is to say, over the part of the ballad that is essentially incremental and hence presumably easier to recall – T1 is quite clean and has been copied into T2 largely without difficulty (the spelling bonnie regularly replacing bonny ). The missing o at the end of 1.2 in T1 is easily explained by the stanza running into the contributor information typed at the top of the page. At 2.2 the typist of T2 has added the pronoun he, not present in T1, following And at the beginning of the line And he wheeled her round and he gart her stand. This might be either an accident of the copying process (8.2 also begins And he, while 5.2 begins He), or a recollection of the way the ballad was actually dictated – there is no way of knowing. At 6.3, there is one substantial handwritten amendment, in ink, correcting Or wid ye be a rank rob [b] er’s wife, apparently a mistaken repetition of the wording of 6.1, to the expected repetition of 6.2 Or wid ye die by my penknife. This could represent the correction of an error on the part of either contributor or collector – perhaps, given the extent of the correction and evident irregularity of what was initially typed (compare 3.3 and 9.3), the former seems more likely. At 10.2 Carpenter began typing Or wi [ll], before making the correction to Or I will, again an evident slip which could be readily charged to either contributor or collector.

  • 24 M. J. C. Hodgart, The Ballads, 2nd edn (London: Hutchinson, 1962), pp. 56–57; Hugh Shields, ‘Textua (...)

17Thus far, nine out of the first ten stanzas have held to a regular, four-stress line, rhyming aaae (where e represents the Eldrie, o refrain); the exception being stanza 1 which rhymes aebe (or, if alone/ them is allowed as a half-rhyme, aeae). Stanza 10 of T1 is followed by a two-line stanza comprising an extrametrical line, For if you kill me, I’ve a brother will kill you, and the bonny banks o Eldrie, o refrain line. In T2, these two lines have been closed up to stanza 10 to give a six-line stanza, rhyming aaaebe. The following stanza (crit. ed. 11) reverts to the aeae rhyme scheme suggested in stanza 1, with an apparently extrametrical first line. Here it is worth noting that, though extrametrical lines might indicate something that is going on with regard to the dictating of the text, in practice they are likely to present no special difficulty to a ballad singer.24 The next stanza (crit. ed. 12) is typed as an unrhymed two-line stanza in T1, with an intervening line, And the tears ran down in sorrow, handwritten in ink (crit. ed. 12.2).

  • 25 Carpenter Collection, MS p. 10333.

18Continuing the romantic vein of the draft essays already noted, Carpenter describes Mrs McEwan’s demeanour when singing this ballad as ‘a striking example of the eerie enchantment once exercised over the minds of primitive folk by ancient ballad lore’.25 Then he types out crit. ed. 1 – 3, after which he describes Mrs McEwan ‘half-chanting, half-dramatizing the sister’s reply’(i. e. crit. ed. 4), and goes on to give crit. ed. 12 (filled out to a quatrain by repetition of the second line). Then he writes:

  • 26 Carpenter Collection, MS p. 10334.

At this point in the ballad her voice seemed somehow to catch in her throat. Then suddenly her English melted into her native Doric, and for several stanzas my Dictaphone recorded, in the half-chanted rhythm of the ancient Gaels, the words of this strange story, crooned to a somber Gaelic melody that sounded like a song of lament. I am not sure that her husband, trying to calm her, did not break off her story. At all events, after gaining time to collect herself, she explained that ‘Bubblin Johnnie’ was under the power of a robber band, who had set him on to do the deed, but that in the end he avenged himself, as follows: [quotes crit. ed. 16 – 17].26

19There are certain difficulties with this account (which may have been written well after the event). This part of the ballad is not in fact on the dictaphone recording; the rough copy T1 has the appearance of having been taken down from dictation, not singing, in accordance with Carpenter’s usual practice; and the description appears to skip over crit. ed. 13 – 15 and to conflate the prose explanation given separately at the end of T1 (see below) with the story as it can be gleaned from the ballad verses. Nevertheless, the indications of a change of language register, and of a break in concentration at this point in the ballad (perhaps prompted by the presence of an ‘audience’ in the form of her husband and/or the collector), are highly suggestive in relation to what follows.

  • 27 DSL-DOST sit, v.; EDD sit, v.; OED sit, v.

20There follows in T1 another unrhymed three-line stanza, the second line of which has a repetition of the phrase and sore sat down, making the line extrametrical, which repetition is simply omitted from T2 (crit. ed. 13.2). Quite possibly influenced by the previous occurrence of the same phrase at the end of the immediately preceding line (crit. ed. 13.1), the assumption was presumably made in the preparation of T2 that this was simply an accidental repetition. On the other hand, the handwritten alteration in T1, in ink, of the last two words of the same line, and lamenting, to with sorrow, is most probably attributable to self-correction on the part of the contributor. Conversely, the handwritten alteration, this time in red pencil, of set to sat in three places in this and the preceding line would seem to be the work of the collector, favouring the standard English form over what is a plausible older Scots and/or dialect form.27 It is not possible to know whether either form has any bearing on what Carpenter thought he had heard – in general, he was more likely to ‘Scotticize’ than to standardize words during the preparation of the fair copies, but there are instances in both directions.

  • 28 CSD cove &c, n., vt.; DSL-SND1 cove, n. 1; DSL-SNDS cove, v.3.

21The next stanza picks up the aeae rhyme scheme again, and has a handwritten alteration, in ink, of This to My (crit. ed. 14.1). This is followed by a stanza rhyming abbe, with a missing word, they, written in, in ink, in the second line (crit. ed. 15.2). The final line initially began On the bonny ban[k]s, presumably in anticipation of the usual refrain line, but this was subsequently typed through with hyphens (and later crossed through again by hand) and replaced by an extrametrical line, To the coves up abeen the banks o Eldrie, o; subsequently banks was altered by hand, in ink, to birks (crit. ed. 15.4). The coves, it would seem, are caves or caverns, specifically recesses worn out of a river bank (the verb cove refers to the action of a river hollowing out its banks).28 So birks (birch trees) seems to shy away from the more obvious connection and further to increase the complexity of idea in this line.

22This is sustained, to the point of confusion, in the next stanza, which initially began What’s the calder cave on the, altered in typescript to What’s the calder coves into the birk (crit. ed. 16.1). Although calder is possible as a comparative form of Scots cald/cauld (cold), the word is also common as a personal or place name, and in T1 the typed lower-case initial c has been raised by hand, in ink, to upper-case. When quoting this stanza in the essay drafts, Carpenter typed Calder Coves, with upper-case C for both words. But the meaning remains unclear. Also, the birk of T1 has become birks in the preparation of T2, possibly under the influence of the plural form in the last line of the preceding stanza (crit. ed. 15.4). The correction in T1 of bides to bide (crit. ed. 16.3), and the replacement of the older spelling wie with wi in the preparation of T2 (crit. ed. 16.5), in contrast, require little notice. This is a six-line stanza of difficult syntax, where the rhyme scheme is largely lost.

23That apparent loss is continued into the penultimate stanza. The first two lines are typed as a single line in T1 and split into two in the preparation of T2 (crit. ed. 17.1–2). The epithet Bubblin is handwritten in, in ink, to correct a presumably accidental omission. The essay drafts quote a single, variant, line, There they’re settin waitin till Bubblin Johnnie comes home. Given the absence of easy syntax and predictable lines in this part of the ballad, the next line, initially beginning Up in the cove of, then altered, in typescript, to read Up in the cove abeen Eldrie, o, then typed through with hyphens and replaced, still in typescript, with Up abeen the banks o Eldrie, o, seems on the face of it likely to be attributable to the contributor’s self-correction (crit. ed. 17.3). In the draft essays, it is given as Up in the Coves abeen Eldrie, o. There then follows a long extrametrical line, When in goes Bubblin Johnnie and there’s none got a chance (crit. ed. 17.4). A fifth line contains some further presumed self-correction, all in typescript, of On the to Up on t [he?] to, finally, Up abeen the banks o Eldrie, o (crit. ed. 17.5).

24The final stanza then comprises just two lines: another extrametrical line, I have killed my sisters and I have got revenge (crit. ed. 18.1); and the Up abeen the banks o Eldrie, o refrain (crit. ed. 18.2), which by this stage has emerged as the regular form (crit. ed. 16.4, 17.3, 17.5, 18.2), replacing On the bonnie banks o Eldrie, o. Notwithstanding the caveat entered above, the increasing prevalence of extrametrical lines in this latter part of the ballad can be read as modulating towards what, judging by the fact that the typed line is indented and within parentheses in T1 and T2, must have been a spoken explanation following the ballad proper: ‘He killed them as they ran out, one by one.’

25If the impression derived from T1 of some self-correction on the part of the contributor, followed by further necessary correction during the preparation of T2, is at all correct, then both metre and rhyme do appear to be giving way over the more complicated, non-incremental, and seemingly more emotional, part of the ballad story. Following the ballad there is, on a new page in both T1 and T2, a prose summary which contains some details that cannot be deduced from the text as it stands, along with some comments on Mrs McEwan’s demeanour and her attitude to the ballad:

Story: The seven robbers induced Bubblin Johnnie, a herd boy whom they thought to be silly, to rob for them. They put him up to attack sisters. When he found that he had murdered his sisters, he went back and murdered them.
The singer was trembling and eyes were shining as she sang (strong face, like lioness) as if under some strange spell. She was loath to sing concluding verses. ‘Too heavy, awful.’ She had heard the ballad preached against in pulpit.

26The implication is that she did have a coherent outline in mind, even if it was not confidently realized in the dictated text.

  • 29 Pointed brackets mark deletions, and arrows ↑↓ mark insertions other than direct substitutions. Fo (...)

27The raw materials of this ballad thus contain the potential for its representation either as process or as product. The fair copy (and, by and large, the critical edition text) presents the ballad as product. But it would be equally possible to present it in a genetic edition, representing the matrix of various forms that embody the ballad as it occurs across the manuscripts T1 and T2. In somewhat simplified form, a genetic edition of stanzas 12–18 might look rather like the following.29

He catched his sister, he kissed her again
↑And the tears ran down in sorrow↓
On the <bonny> bonnie banks o Eldrie, o.

They kissed one another and sore <set> sat down
And sore <set> sat down <and sore <set> sat down> <and lamenting> with sorrow
On the <bonny> bonnie banks o Eldrie, o.

<This> My two sisters I have slain
On the bonnie banks o Eldrie, o
I am sure I will have revenge
On the bonnie banks o Eldrie, o.

He catched his youngest sister by the hand
The tears they wiped, and ↑they↓ walked on
The tears they wiped and they walked on
<On the bonny ban [k] s> To the coves up abeen the <banks> birks o Eldrie, o.

What’s the <calder> Calder <cave on the> coves into the birk↑s↓
Where the robbers had their nest in Eldrie, o?
This where the coves and the rank robbers bide<s>
Up abeen the banks o Eldrie, o
Says Bubblin Johnnie, Come wi<e> me
And I’ll have revenge on Eldrie, o.

There they’re settin waitin
Till ↑Bubblin↓ Johnnie will come home
<Up in the cove <of> abeen Eldrie, o> Up abeen the banks o Eldrie, o
When in goes Bubblin Johnnie and there’s none got a chance
<On the> Up <on t [he]> abeen the banks o Eldrie, o.

I have killed my sisters and I have got revenge
Up abeen the banks o Eldrie, o.

28A genetic text along these lines presents simultaneously the matrix of variant possibilities preserved within the extant documentary evidence. What it depicts on the page is Mrs McEwan endeavouring to reconstruct the ballad for Carpenter, and Carpenter endeavouring to represent her recitation. He may also have prompted her at various points – the evidence for or against this is not recorded – and he might himself have begun to type lines, anticipating what he thought she was going to say, so that traces of his own contribution may be present, albeit unmarked as such. The genetic text, therefore, also represents a social text. Even more than as an editorial procedure, this way of representing the ‘coming into being’ of a ballad text captures something of the process by which it was achieved in the historically situated interaction between contributor and collector. And although it is not quite so easy to envisage what it would look like, there is no reason in principle why the same approach should not be applied to the music transcription (both words and melody), especially as the (poor quality) raw material of the sound recordings relate to a modern interpretation of what is going on therein.

29What the genetic approach offers, therefore, is two things, simultaneously: (i) it depicts the ballad as a process, embodying intentions that are relative and in flux throughout; and (ii) it depicts something of the social interaction involved in (verbal and musical) text creation. The exercise is designed to demonstrate that genetic editing, which is possible in this particular instance of Mrs McEwan’s ‘Bubblin Johnnie’ because renditions in various different states have been preserved, can bring out into the open principles that are most often suppressed on the printed page or in sound recordings. Ultimately there is unlikely to be sufficient practical justification for a genetic edition of ballads, but the potential for such presentation alongside more conventional editing methods remains important to the opening up of ballad ontology and the establishment of a theory of text.

  • 30 de Biasi, ‘What Is a Literary Draft?’, p. 33.
  • 31 G. Thomas Tanselle, ‘Textual Criticism at the Millennium’, Studies in Bibliography, 54 (2001), 1–80 (...)
  • 32 Grésillon, ‘Slow: Work in Progress’, p. 115.

30There might, however, still be a problem. Genetic criticism, as noted above, generally does recognize that moment when the author signs off a text as bon à tirer, passed for press, when a particular state of the work passes into the public domain. The observance of this moment permits the crucial distinction between avant-texte and the ‘frozen shape of a published text’. In doing so it also facilitates the recognition of a temporal and logical ordering within the matrix of avant-texte – a teleology that represents the genesis of a state signalled by the declaration ‘passed for press’.30 Emphatically, this is not a hierarchical, evolutionary, or ‘finalist’ ordering, since chronology is by no means necessarily synonymous with progress or improvement.31 Indeed, the genetic approach can seem to undermine the finality, or at least the ‘sacrosanct auctoritas’, of the published text.32 Nevertheless, it is difficult in practice not to perceive the published text as a point from which to orient the discussion of avant-texte. In the case of the ballad, however, although Carpenter’s fair copy typescript may be the nearest thing to the ‘passed for press’ state that can be identified, clearly it does not bear Mrs McEwan’s metaphorical signature any more or less than do, say, the variant stanzas that she sang for the sound recording. It may be that with materials that are not in a conventional sense ‘authored’, there never actually is a ‘passed for press’ moment.

  • 33 This term ‘conceptual text’ alludes in part to the notion of ‘mental text’ in Lauri Honko, Textuali (...)

31In consequence, while the process of artistic ‘invention’ rendered visible in this way can illuminate the collecting of ballads and suchlike materials, it might not provide the most exact analogy for thinking both about potential multiplicities of ballad renditions, and about a putative ‘conceptual text’ that might lie behind them.33 It is quite reasonable to suppose that a ballad never is ‘passed for press’, in the sense that at least its vocal rendition may be perceived as – indeed, may be intended to be – evanescent. So each rendition attributable to the same source will be of equivalent, independent status; each is, so to speak, an equivalent part of a matrix of something akin to avant-texte, but lacking either logical or even temporal ordering. Of course, there necessarily will be a chronology among such renditions, but that temporal arrangement is without teleological significance. Each rendition represents, more or less accurately, an intention that can be attributed to a particular passing moment in time and the circumstances then prevailing. Variation among such renditions represents merely the relativity of those momentary intentions.

32An alternative way of looking at the same body of renditions, of course, is simply to consider everything as ‘passed for press’, since once rendered aloud, however transiently, the ballad has entered the public domain (regardless of the actual physical presence or absence of some kind of audience). Each rendition momentarily achieves the ‘frozen shape of a “published” [my added quotes] text’, which could potentially be recorded or written down. This is to draw the analogy not so much with manuscript genetics as with the genetics of published editions – although a temporal and logical ordering is still lacking, since each rendition has an exactly equivalent status and their chronology is again without teleological significance. Literary and musical editors are most used to thinking of versions of works as something insubstantial, a form of the work as the author intended it at one particular moment. In the ballad scenario, however, intention is made momentarily explicit by the act of recitation, singing, or writing.

  • 34 There is a possible analogy here with the discussion in Jerome J. McGann, The Textual Condition (Pr (...)

33Nevertheless, it still seems counter-intuitive to consider these renditions just as discrete units, because they do all testify to a single, though not necessarily fixed, conceptual text. It is the felt presence of this emergent conceptual text – largely analogous to the idea of the literary work – that imposes a pattern on to the distinct renditions. In other words, it would be possible, in theory at least, to order ballad renditions rather in the manner of a genetic edition – not so much by establishing the usually unknown avant-texte that underlies a known ‘passed for press’ state, as by assembling the known physical texts and so enabling a more fluid conceptual text to emerge. One way of describing such a combined text would be as the representation of a palimpsest of material texts.34

34The notion of palimpsest is being invoked here in a manner close to its technical meaning for manuscript studies. Ballad renditions can be imagined as layers superimposed upon one another, while each is still able to be read and/or heard. Where a thick corpus of material is physically available, the editorial representation of such a palimpsest might begin to permit the depiction of a putative conceptual text with a certain amount of depth. Whether such an exercise would actually be intellectually fruitful is perhaps more questionable. Certainly, if one considers simply mapping Mrs McEwan’s third stanza from the sound recording on to the dictated text, the fact that, to put it crudely, when singing the ballad she conflated two stanzas of the text she subsequently dictated, is scarcely surprising and probably not particularly culturally or psychologically significant.

  • 35 Quite independently, Peter Harrop, ‘The Antrobus Soulcakers: A Consideration of Site, Mobility and (...)
  • 36 Nevertheless, the possibilities presented by, for example, the Sabine Baring-Gould manuscripts shou (...)

35Where the model is perhaps more illuminating is in disentangling the conventional language of ‘versions’. The recognition that, generally, when one speaks of a contributor’s version of a ballad what one is really referring to is a palimpsest of material texts that equate, with greater or lesser accuracy, to a putative conceptual text, is salutary.35 At the very least, a degree of caution is required when discussing ballad and folk song aesthetics and repertoires, especially at a historical distance, based on limited evidence. Most of the manuscript, printed, and recorded ballad collections – the extant physical texts – offer a mere snapshot, a single layer of the palimpsest.36 If nothing else, the iconic status of great ballad collections is being challenged here.

  • 37 de Biasi, ‘What Is a Literary Draft?’, p. 58.
  • 38 Thomas A. McKean, ‘The Stewarts of Fetterangus and Literate Oral Tradition’, in The Singer and the (...)

36This palimpsest, presumed or actual, is complemented by the perspective derived analogically from genetic criticism, whereby ‘the text remains inseparable from its pre-textual realization’,37 and which therefore posits, so to speak, ‘versions within versions’. The relativity of the notion of intention is highly pertinent to ballads that are always amenable to being variously learned, reconstructed, transmitted, modified. For instance, although when a ballad is first learned, regardless of the nature of its source, the intention may well be to achieve a form that reproduces that source with some accuracy, subsequently that intention might become modified towards a rendering that resembles it much less closely, as the performer seeks to make the song ‘their own’.38 The genetic perspective would nonetheless recognize the earlier, learned form as still integral to that later form, in the manner of avant-texte. The Carpenter collection, with its mass of amended typescripts, seems to indicate that carriers of traditions are not always confident, accomplished performers, but that they may well seek consciously to recover or reconstruct texts. In that case, the contribution of the collector/editor may be both a necessary and a welcome intervention. The reconstructed text, too, would itself then be integral to any subsequent rendition that might emerge.

37The idea of palimpsest and the genetic perspective together offer a textual model for the ballad more or less equivalent to that of the literary work, and more fluid than any of the ballad’s individual manifestations. The model posits a matrix of variance and connectedness, implying at any one moment both its preceding and succeeding variations. The ballad as received is conceived both as a process of accretion, deletion, and superimposition, and as a product that exists, even if only ephemerally, relative to its own contexts of production, collection, publication, and reception.

Annexes

Appendix 1

1 There were three sisters walkin alone
On the bonnie banks o Eldrie, o
As this young man come up on them
On the bonnie banks o Eldrie, o.

2 He catched the older sister by the hand
And he wheeled her round and he gart her stand
He wheeled her round and he gart her stand
On the bonnie banks o Eldrie, o.

3 It’s will ye be a rank robber’s wife?
Or will ye die by my penknife?
Or will ye die by my penknife
On the bonnie banks o Eldrie, o?

4 I will not be a rank robber’s wife
But I will die by your penknife
But I will die by your penknife
On the bonnie banks o Eldrie, o.

5 He took the second sister by the hand
He wheeled her round and he gart her stand
He wheeled her round and he gart her stand
On the bonnie banks o Eldrie, o.

6 It’s wid ye be a rank robber’s wife?
Or wid ye die by my penknife?
Or wid ye die by my penknife
On the bonnie banks o Eldrie, o?

7 I will not be a rank robber’s wife
But I shall die by your penknife
But I shall die by your penknife
On the bonnie banks o Eldrie, o.

8 He’s catched the third sister by the hand
And he wheeled her round and he gart her stand
He wheeled her round and he gart her stand
On the bonnie banks o Eldrie, o.

9 Oh will ye be a rank robber’s wife?
Or will ye die by my penknife?
Or will ye die by my penknife
On the bonnie banks o Eldrie, o?

10 I will not be a rank robber’s wife
Or I will not die by your penknife
Or I will not die by your penknife
On the bonnie banks o Eldrie, o
For if you kill me, I’ve a brother will kill you
On the bonnie banks o Eldrie, o.

11 Come tell to me what is your brother’s name
On the bonnie banks o Eldrie, o
Bubblin Johnnie is his name
On the bonnie banks o Eldrie, o.

12 He catched his sister, he kissed her again
And the tears ran down in sorrow
On the bonnie banks o Eldrie, o.
13 They kissed one another and sore sat down

And sore sat down with sorrow
On the bonnie banks o Eldrie, o.
14 My two sisters I have slain
On the bonnie banks o Eldrie, o
I am sure I will have revenge
On the bonnie banks o Eldrie, o.

15 He catched his youngest sister by the hand
The tears they wiped and they walked on
The tears they wiped and they walked on
To the coves up abeen the birks o Eldrie, o.

16 What’s the Calder coves into the birks
Where the robbers had their nest in Eldrie, o?
This where the coves and the rank robbers bide
Up abeen the banks o Eldrie, o.
Says Bubblin Johnnie, Come wi me
And I’ll have revenge on Eldrie, o.

17 There they’re settin waitin
Till Bubblin Johnnie will come home
Up abeen the banks o Eldrie, o
When in goes Bubblin Johnnie and there’s none got a chance
Up abeen the banks o Eldrie, o.

18 I have killed my sisters and I have got revenge
Up abeen the banks o Eldrie, o.

‘He killed them as they ran out, one by one.’

Key: del – deleted; ins – inserted; om – omitted

recurrent variant: bonnie] T2; bonny T1 (1.4, 2.4, 3.4, 4.4, 5.4, 6.4, 7.4, 8.4, 9.4, 10.4, 10.6, 12.3, 13.3)

1.2 (after Eldrie) o] T2, ins T1

2.2 (after And) he] T2; om T1

6.3 die by my penknife] T2, T1; be a rank rob[b]er’s wife del T1

10.2 after Or] wi[ll] del T1

10.4 – 5] stanza break T1

12.2] T2, ins T1

13.1 sat] T2, T1; set del T1

13.2 sat] T2, T1; set del T1
after
down] and sore sat down (sat] T1; set del T1) T1
with sorrow] T2, T1; and lamenting del T1

14.1 My] T2, T1; This del T1

15.2 (after and) they] T2, ins T1

15.4 before To] On the bonny ban[k]s del T1
birks] T2, T1; banks del T1

16.1 Calder] T2, T1; calder del T1
coves] T2, T1; cave on the del T1
birks] T2; birk T1

16.3 bide] T2, T1; bides del T1

16.5 wi] T2; wie T1

17.1 – 2 line break ] T2; om T1

17.2 Bubblin] T2, ins T1

17.3 before Up] Up in the cove abeen Eldrie, o (after cove] of del T1 ) del T1

17.5 before Up] On the del T1
after
Up] on t [he?] del T1

Appendix 2

cylinder recording

... was walking alone
On the bonnie banks of Eldrie, o [
cylinder skips]
... was walking alone
On the bonnie banks of Eldrie, o
Fan [there?] was a gentleman comin his lone
On the bonnie, bonnie banks o Eldrie, o.
He catched the older sister by the han
He feeled her round and he gart her stand
He feeled her round and he gart her stand
On the bonnie banks o Eldrie, o.

Will you be a rank robber’s wife?
Or will you die by my penknife?
I will not be a rank robber’s wife
On the bo...
But I will die by your penknife
On the bonnie banks o Eldrie, o.

[transcription by Thomas A. McKean and author]

critical edition

There were three sisters walkin alone
On the bonnie banks o Eldrie, o
As this young man come up on them
On the bonnie banks o Eldrie, o.

He catched the older sister by the hand
And he wheeled her round and he gart her stand
He wheeled her round and he gart her stand
On the bonnie banks o Eldrie, o.

It’s will ye be a rank robber’s wife?
Or will ye die by my penknife?
Or will ye die by my penknife
On the bonnie banks o Eldrie, o?

I will not be a rank robber’s wife
But I will die by your penknife
But I will die by your penknife
On the bonnie banks o Eldrie, o.

Notes

1 That is not wilfully to neglect altogether the role of local ballad composers. See, for example, Edward D. Ives, Larry Gorman: The Man Who Made the Songs (Bloomington: Indiana University Press, 1964); Edward D. Ives, Lawrence Doyle, The Farmer-Poet of Prince Edward Island: A Study in Local Songmaking, University of Maine Studies, no. 92 (Orono: University of Maine Press, 1971); Edward D. Ives, Joe Scott: The Woodsman-Songmaker (Urbana: University of Illinois Press, 1978); Thomas A. McKean, Hebridean Song-Maker: Iain MacNeacail of the Isle of Skye (Edinburgh: Polygon, [1997?]).

2 OED palimpsest, n. 2 a, b.

3 See, for example, Almuth Grésillon, Éléments de critique génétique: Lire les manuscrits modernes (Paris: Presses Universitaires de France, 1994); Pierre-Marc de Biasi, La Génétique des textes (Paris: Nathan, 2000). In English, see Jed Deppman, Daniel Ferrer, and Michael Groden, eds, Genetic Criticism: Texts and Avant-textes (Philadelphia: University of Pennsylvania Press, 2004); essays in Yale French Studies, no. 89 ‘Drafts’(1996).

4 Almuth Grésillon, ‘Slow: Work in Progress’, Word & Image 13 (1997), 106–23 (p. 106).

5 Deppman, Ferrer, and Groden, eds, Genetic Criticism, p. 2.

6 Grésillon, ‘Slow: Work in Progress’, p. 106.

7 Samuel Johnson, The Lives of the Most Eminent English Poets, with critical observations on their works, introduction and notes by Roger Lonsdale, 4 vols (Oxford: Clarendon Press, 2006), i, 261.

8 Deppman, Ferrer, and Groden, eds, Genetic Criticism, p. 3.

9 Edgar Allan Poe, ‘The Philosophy of Composition’, in Selected Writings of Edgar Allan Poe: Poems, Tales, Essays and Reviews, ed. David Galloway (Harmondsworth: Penguin, 1967), pp. 480–92 (p. 481).

10 René Wellek and Austin Warren, Theory of Literature, 3rd edn (Harmondsworth: Penguin, 1963), pp. 90–91.

11 Jean Bellemin-Noël, Le Texte et l’avant-texte: Les Brouillons d’un poème de Milosz (Paris: Larousse, 1972). See also Jean Bellemin-Noël, ‘Reproduire le manuscrit, présenter les brouillons, établir un avant-texte’, Littérature, no. 28 (1977), 3–18.

12 Pierre-Marc de Biasi, ‘What Is a Literary Draft? Toward a Functional Typology of Genetic Documentation’, Yale French Studies, no. 89 (1996), 26–58 (p. 37).

13 Grésillon, ‘Slow: Work in Progress’, p. 115; Laurent Jenny, ‘Genetic Criticism and its Myths’, Yale French Studies, no. 89 (1996), 9–25 (p. 14).

14 Jerome J. McGann, A Critique of Modern Textual Criticism (Charlottesville: University Press of Virginia, 1992 [1983]), pp. 37–49; Peter L. Shillingsburg, Scholarly Editing in the Computer Age: Theory and Practice, 3rd edn (Ann Arbor: University of Michigan Press, 1996), pp. 29–39; G. Thomas Tanselle, ‘The Editorial Problem of Final Authorial Intention’, Studies in Bibliography, 29 (1976), 167–211; G. Thomas Tanselle, A Rationale of Textual Criticism (Philadelphia: University of Pennsylvania Press, 1989), esp. pp. 75–78.

15 Cited in D. C. Greetham, Theories of the Text (Oxford: Oxford University Press, 1999), p. 60.

16 Hans Walter Gabler, ‘The Text as Process and the Problem of Intentionality’, Text: Transactions of the Society for Textual Scholarship, 3 (1987), 107–16.

17 de Biasi, ‘What Is a Literary Draft?’, p. 58.

18 Washington, DC, Library of Congress, American Folklife Center, Archive of Folk Culture, AFC1972/001, James Madison Carpenter Collection, Cylinder 031 00:00; Disc sides 064 02:17, 065 00:00; MS pp. 05701–05703, 07592–07593, 08622–08623.

19 Carpenter Collection, Cylinder 131 04:21; Disc side 310 03:05; MS pp. 04924–04926, 06997–06998, 08480–08481, 11721–11722.

20 For a general introduction to Carpenter’s folk play collection, see Eddie Cass, ‘The James Madison Carpenter Collection of British Folk Plays’, Folklore, 123 (2012), 1–22.

21 The full set of references is Carpenter Collection, Cylinder 130 11:12; Disc sides 308 04:25, 309 00:00; MS pp. 04384–04387, 06539–06541, 07930, 10333–10334, 11180–11181, 12187–12189, 12272–12273, 12395–12396.

22 Carpenter Collection, MS p. 10333.

23 David Atkinson, ‘The Secret Life of Ballad Manuscripts’, Variants, 8 (2012), 183–206 (esp. pp. 186–87) [and note the corrigendum to this article in Variants, 10 (2013), 13].

24 M. J. C. Hodgart, The Ballads, 2nd edn (London: Hutchinson, 1962), pp. 56–57; Hugh Shields, ‘Textual Criticism and Ballad Studies’, in Dear Far-Voiced Veteran: Essays in Honour of Tom Munnelly, ed. Anne Clune (Miltown Malbay: Old Kilfarboy Society, 2007), pp. 287–94 (p. 293). More generally, see Hugh Shields, ‘Supplementary Syllables in Anglo-Irish Folk Singing’, Yearbook of the International Folk Music Council, 5 (1973), 62–71.

25 Carpenter Collection, MS p. 10333.

26 Carpenter Collection, MS p. 10334.

27 DSL-DOST sit, v.; EDD sit, v.; OED sit, v.

28 CSD cove &c, n., vt.; DSL-SND1 cove, n. 1; DSL-SNDS cove, v.3.

29 Pointed brackets <> mark deletions, and arrows ↑↓ mark insertions other than direct substitutions. For the sake of clarity, mere corrections of typing errors and the marking of line divisions are ignored, and an editorial system of minimal punctuation has been applied (on punctuation, see further chapter 8 below). The annotation is loosely borrowed from genetic examples in D. C. Greetham, Textual Scholarship: An Introduction (New York: Garland, 1994), pp. 408–10. This is just the kind of editing, however, that most obviously lends itself to the electronic environment.

30 de Biasi, ‘What Is a Literary Draft?’, p. 33.

31 G. Thomas Tanselle, ‘Textual Criticism at the Millennium’, Studies in Bibliography, 54 (2001), 1–80 (p. 27).

32 Grésillon, ‘Slow: Work in Progress’, p. 115.

33 This term ‘conceptual text’ alludes in part to the notion of ‘mental text’ in Lauri Honko, Textualising the Siri Epic, FF Communications, no. 264 (Helsinki: Suomalainen Tiedeakatemia/Academia Scientiarum Fennica, 1998), pp. 92–99; Lauri Honko, ‘Thick Corpus and Organic Variation: An Introduction’, in Thick Corpus, Organic Variation and Textuality in Oral Tradition, ed. Lauri Honko (Helsinki: Finnish Literature Society, 2000), pp. 3–28 (pp. 18–19). I would not, however, want to draw anything like a direct analogy between the English-language ballads and the oral epics studied by Honko. In substituting ‘conceptual’ for ‘mental’ I am hoping, too, to avoid locating this abstract textual idea solely with the contributor. The conceptual text refers simultaneously to the contributor’s dynamic mental image of the ballad, the social interaction that has enabled it to be collected in a particular form(s), and its ongoing (and changing) reception over time–but organized by its identification with the named contributor.

34 There is a possible analogy here with the discussion in Jerome J. McGann, The Textual Condition (Princeton: Princeton University Press, 1991), p. 30, of ‘evolutionary texts’ combining the entire extant corpus of materials relating to a work into an edition representing a single, continuous historical project. In the ballad case, while the notion of continuity is useful, that of ‘evolution’ should be resisted, although change is certainly possible.

35 Quite independently, Peter Harrop, ‘The Antrobus Soulcakers: A Consideration of Site, Mobility and Time as Components of Traditional Performance’, Contemporary Theatre Review, 22 (2012), 267–73, has employed ‘palimpsest’ as a way of describing the cumulative effect of repetition in folk play performances.

36 Nevertheless, the possibilities presented by, for example, the Sabine Baring-Gould manuscripts should not be overlooked.

37 de Biasi, ‘What Is a Literary Draft?’, p. 58.

38 Thomas A. McKean, ‘The Stewarts of Fetterangus and Literate Oral Tradition’, in The Singer and the Scribe: European Ballad Traditions and European Ballad Cultures, ed. Philip E. Bennett and Richard Firth Green (Amsterdam: Rodopi, 2004), pp. 181–207 (p. 198).

Table des illustrations

URL http://books.openedition.org/obp/docannexe/image/1953/img-1.jpg
Fichier image/jpeg, 20k
Légende Fig. 7.1 Carpenter Collection, MS pp. 04384 – 04387. Courtesy of the American Folklife Center at the Library of Congress, Washington, DC, USA.
URL http://books.openedition.org/obp/docannexe/image/1953/img-2.jpg
Fichier image/jpeg, 20k
URL http://books.openedition.org/obp/docannexe/image/1953/img-3.jpg
Fichier image/jpeg, 20k
URL http://books.openedition.org/obp/docannexe/image/1953/img-4.jpg
Fichier image/jpeg, 14k

Acheter

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search