3. Textual Authority and the Sources of Variance
The previous chapter touched on the iconic status accorded to oral tradition in ballad studies, the corollary of which is the denigration of print. Philip Bohlman writes: ‘So strong is the correlation of oral tradition with folk music that most definitions treat oral tradition as fundamental to folk music, if not its most salient feature.’1 Loosely applied, ‘oral tradition’ might indicate nothing more precise than that ballads in general are amenable to singing or recitation, and/or that a particular item was collected from a singer who had memorized it. However, the implication of non-literacy is frequently there in the qualifier ‘oral’, and that of transmission from person to person over time and/or place in the substantive ‘tradition’. Accordingly – seductively – oral tradition comes to be equated with the very engine of ballad variation, that complex of processes involving remembering and forgetting, addition, subtraction, and rearrangement, substitution, borrowing, and consolidation (to name just a few) that scholars have codified to account for textual and melodic stability and change.2 After all, the argument implicitly runs, it is much harder to reproduce a text learned by word of mouth than one learned from a printed source. Tunes are even more prone to variation than words.3 It is generally understood that the melodies have nearly always been learned by ear; although musical literacy was not that rare in the past, ballad tunes were not so frequently printed and, given their relatively simple, strophic nature, seem unlikely to have required recourse to notation.
The oral argument finds its apotheosis, in the wake of the publication of Albert Lord’s The Singer of Tales in 1960,4 in the attempt to apply the theory of oral-formulaic improvisation directly to the Anglo-Scottish ballads, most notably in David Buchan’s The Ballad and the Folk.5 It is fair to say that this particular application of the oral-formulaic theory has been effectively refuted;6 but it is equally fair to say that the presence of ballads in printed form – in particular, the cheap printed formats of broadsides and chapbooks – is still widely regarded as, at best, a conundrum. Bohlman calls it a ‘blatant paradox’.7 Print is acknowledged as a means of ballad transmission and a source of new material, but there is a general urgency to hurry the song along into oral tradition.8 The distrust of print goes all the way back to early folk song collectors such as Sabine Baring-Gould and Cecil Sharp.9 Likewise, Francis James Child’s work on ballads was marked by an uneasy relationship with broadside print.10
All of this is problematic, now that the true depth and breadth of the print culture that lies behind the English and Scottish ballads is becoming apparent, thanks on the one hand to the painstaking efforts of researchers in libraries and archives, and on the other to the increasing availability of digitized images of broadside and chapbook collections and related materials. It is thought that as many as 90 per cent of the English and Scottish ballads and folk songs collected during the Romantic period and in the late nineteenth/early twentieth century could also be found in cheap print.11 Moreover, while it is widely (and quite reasonably) assumed that printers copied from one another, it is not the case that printed ballads are devoid of variation. In fact, as Dianne Dugaw was perhaps the first to demonstrate in depth, it is possible to identify just the same sorts of textual stability and change among printed ballads as are found among those collected from singers.12 Variants in printed ballads frequently take the form of lexical omissions and additions, rearrangements, substitutions, and the like, and (usually over a broader chronological and/or geographical span) it is also possible to identify instances of wholesale revision and rewriting, such as the recasting of The Berkshire Tragedy as The Cruel Miller considered in the previous chapter.
Rarely is it possible to be certain whence these printed variants originated. While some may be the result of lax typesetting, faulty copy, constraints of the printed format, and the like, others could just as easily represent more deliberate ‘authorial’ interventions, and it is quite likely that some of them may have been influenced by the ballad in its orally circulating form. Where something like a date or a proper name is substantially different as between one broadside ballad and another, it is difficult to imagine that the substitution was not in some way deliberate.13 Different printings of ‘The Mountains High’ (Roud 397), for example, name the main character variously as Reynardine, Rinordine, or Ryner Dyne. Similarly, in ‘The Bold Princess Royal’ (Roud 528), the date of the encounter with the pirate ship is usually given on broadsides as the 14th of February, but other dates are found among copies collected from singers, and at least one broadside has the 6th of January. In ‘The Fatal Ramillies’ (Roud 1266), the complement of the battleship appears on broadsides as seven hundred and seventy or seven hundred and twenty, which is close to the historical figure, but also as the much exaggerated seventeen hundred and seventy. If these various broadside texts were indeed copied from one another, then it would seem that the printers must have had some other model in mind when introducing such variant details – that is to say, either some consciously creative input, or some extraneous knowledge of how they thought the ballad ‘ought to go’. This whole area demands further investigation and systematization – and it also demands a theoretical model of variation capable of integrating both the printed and the oral ballad.
Just to set out the ground quite clearly: there is no question here of denying that English-language ballads can be, and certainly have been, learned and passed on without the aid of written or printed copies, over the five centuries or so during which the genre has been in existence, and that this may have been a source of textual and melodic variation. Oral transmission has perhaps had a particular importance in remote regions such as Appalachia, or among social groups such as the Scottish Travellers.14 Ballad melodies, as noted above, were rarely written down as music notation, meaning that a degree of transmission by word of mouth and learning by ear must have been more or less essential. A number of ballads stand out as not having any known record in cheap print, and therefore presumably must have been passed on by oral means. Among them are widely collected items such as ‘Lord Randal’ (Child 12) and ‘The Maid Freed from the Gallows’ (Child 95), which are strongly repetitive and incremental in nature, suggesting that they would have been particularly easy to learn and remember. Indeed, one of the most cogent reasons for rejecting the theory of oral-formulaic improvisation in relation to the English-language ballads is that it is simply not necessary: they are just not that difficult to learn by heart. A further reason is that the historical context and literary culture out of which the English-language ballads emerged was entirely different from that surrounding the South Slavic epic songs that provided the material evidence for The Singer of Tales.
In post-Conquest England, oral and literate strands of culture were intertwined from the beginning, with a gradual shifting of emphasis away from the oral and towards the literate – given a tremendous boost, of course, by the invention of printing – being more or less complete by the end of the nineteenth century.15 The evidence for this is necessarily very diverse in form and chronology. A small selection of the key loci that witnessed the interface of oral and literate culture would include the gradual replacement of oral testimony by written documents in legal proceedings; the ubiquity of preaching as a means of conveying the written word of the Bible to the non-lettered in vernacular languages; the repetition in print from the sixteenth century onwards of classic popular fictional narratives; the pasting up in public and reading aloud of written and printed matter, ranging from official proclamations to libellous verses; the reading aloud of newspapers in coffee houses, facilitating the dissemination of news and the growth of a political culture; the symbiosis of private study and rote learning in the advancement of education and self-improvement, with or without the assistance of the church and/or state, in the industrial era.
It is extremely difficult to cite meaningful figures for the growth of literacy.16 Using the measurable criterion of people’s ability to sign their names to a document, by the 1640s perhaps 30 per cent of men and 10 per cent of women could do so, rising to 45 per cent and 25 per cent, respectively, by the mid-eighteenth century. On the same measure, England had achieved universal literacy by the outbreak of the First World War. However, writing and reading were entirely separate skills, and it is widely believed that such figures spectacularly underestimate the numbers who could read. By the mid-eighteenth century – and perhaps much earlier – probably a majority of the adult population had attained a degree of functional, pragmatic literacy, enabling them to read print to some extent. Nevertheless, literacy rates were extremely variable, being generally lower among the poorer sections of society, among women than among men, and in rural as opposed to urban areas – with rates in London, the epicentre of ballad printing, being consistently well above the national average. Throughout much of the period in question, those who encountered ballads in cheap printed form might have been just as likely to be able to read them as not – but that is a very general statement indeed.
Moreover, it is no more than a small part of the story. Access to written material was never confined to the literate alone but was made much more widely available through what Roger Chartier terms ‘communities of readers’.17 Through the agency of a literate interpreter, reading could be turned into a collective experience. That might be in the context of the church, the family, the marketplace, the school, the workplace, the coffee house, the trade organization. Right up until the early years of the twentieth century, the experience of reading most characteristically meant reading out loud, and reading was most frequently not a solitary activity at all but one rooted in human interaction and sociability.18 Some historians of literacy are now inclined to view the dichotomy of literacy vs. illiteracy, or literacy vs. orality, as little more than a nineteenth-century (or maybe eighteenth-century) ideological construct, tied to a modern construction of interiority, personality, and subjectivity.19 Perpetuated by influential theorists such as Walter Ong and Marshall McLuhan, the so-called ‘great divide’ theory of literacy vs. orality actually reveals rather little about the communal reception of texts in England and Scotland, or even in western Europe at large, at an earlier time.
Instead, for virtually the whole time for which there is anything in the English language that might reasonably be termed a ballad, the picture is one of mutual interdependence of oral and written forms, and of a shared experience of texts regardless of the ability to read. Keith Thomas provides a succinct statement of the matter as it stands in early modern England: ‘Literates and illiterates need not necessarily have different mental habits; and the illiterates of early modern England were in quite a different position from the non-literate inhabitants of purely oral societies. They lived in a world which was to a great extent governed by texts, even though they could not read themselves.’20 Even in post-Roman Britain, the development of a written Old English vernacular, manifest through utilitarian documents such as charters and laws, as well as the more celebrated works of poetry and heroic literature, along with the residual influence of Latin writing, testifies to the continuity of the idea of literacy.21 Early medieval England did not approach anything like what Ong would consider a state of ‘primary orality’.22 (The world of Beowulf is probably something of a literary fiction – mentioned here largely because it seems to correspond in some degree with the putative condition of society, ‘in which the people are not divided by political organization and book-culture into markedly distinct classes, in which consequently there is such community of ideas and feelings that the whole people form an individual’, that Child envisaged as necessary for the genesis of ballads.23) By the thirteenth century, for example, the replacement of oral testimony by writing in legal contexts was well established and its impact was experienced even among the peasantry.24 Across Europe, too, the growth of markets, towns, and commerce from the eleventh to the thirteenth centuries was associated with the spread of new forms of documentary record-keeping.25 The assimilation of the ideas of spoken and written language as means of communicating, preserving, and transmitting texts that these developments represent was in train well before the era of print, and must have taken place even while overall levels of actual, measurable literacy were quite low.26
Here, Brian Stock’s notion of ‘textual communities’ is pertinent.27 Textual communities can be defined as ‘microsocieties organized around the common understanding of a script’.28 Like Chartier’s communities of readers, such microsocieties need not necessarily be composed of literate people, the minimal requirement being just one interpreter with some sort of access to a set of texts and the ability to pass them on to others. The physical presence of a written text was not essential: oral record, memory, and re-performance would suffice.29 Hence ‘textual’ rather than ‘reading’ communities, and the charting of a rise of ‘textuality’ rather than merely an increase in literacy.30 Stock envisages the process thus: as the written word gained ground in western Europe, existing oral practices realigned themselves to function in a reference system based on texts; a culture emerged that was both oral and written; texts were not always written down, but they were understood as if they were (my italics).31
This, then, is the environment in which even the earliest of the English-language ballad-like songs, such as ‘Judas’ (Child 23) and ‘Twelfth Day’, need to be situated. The presence of these two pieces within a manuscript miscellany containing items in English, French, and Latin, of probable Franciscan provenance, is indicative of the ballads’ potential to function within reading/textual communities.32 Subsequently, there is nothing in The English and Scottish Popular Ballads for another couple of centuries, until c.1450, right at the beginning of the era of print and well into an age that was, in Keith Thomas’s phrase, ‘to a great extent governed by texts’.
Even today, relatively little is known for certain about ballad singing, reading, and/or recitation during the early modern period, although there are enough contemporary references to merit the assumption that ballads were ubiquitous, or at least were perceived as being so.33 There are thousands of broadsides which could be – and perhaps were intended to be – sung to named tunes, and at least some of them are the ‘same’ as ballads collected from singers during the folk song revivals of much later date. There are also plentiful literary references to semi-professional ballad singers (Autolycus and his kind) and ballad writers.34 There are also Addison’s Spectator papers of the early eighteenth century, and the beginnings of the appearance of ballads in printed miscellanies. In the eighteenth century, ballads are found among anthologies that can be associated with the growth of musical entertainments in the theatres and metropolitan pleasure gardens. A string of references indicate that it was not uncommon to find broadsides pasted on walls in alehouses and domestic dwellings. Vic Gammon makes the very reasonable point that singing was the most readily available form of artistic expression for much of the population.35
Stock’s specific examples of textual communities are mostly based around ideological movements, such as the Waldensians, where a group of people would coalesce around vernacular texts of the scriptures and their exposition through (unlicensed) public preaching, establishing an agreed meaning for those scriptural texts which came to comprise, for the group in question, the ‘real text’.36 An example from much closer to home would be the religious revival of 1742 at Cambuslang, Lanarkshire, which culminated in outdoor meetings where tens of thousands gathered from miles around to hear the evangelist George Whitefield.37 Like the earlier movements studied by Stock, the revival at Cambuslang was focused on the experience of hearing texts expounded through preaching. The most notable difference was that while the earlier episodes were not generally founded on an immediate continuity with a literate tradition (Peter Waldo’s initial contact with the scriptures was essentially an oral experience), virtually all of the converts at Cambuslang were well able to read the Bible for themselves. At least one girl is recorded as saying, ‘I used to read the Bible at my parents desire tho’ I had more delight in reading story books and ballads.’38 But in so far as the shared experience can be understood in terms of the (admittedly temporary) emergence of a textual community focused around the revivalist preaching of the scriptures, the ability to read or otherwise is largely irrelevant.
While textual communities established around religious ideological movements provide a particularly potent instance for the shared engagement of a group of people with a text, it is clear that the general principle can apply to a much broader range of groups united by a general agreement over the meaning(s) of a text, or set of texts. These might embody aspects of everyday civic and cultural life – the requirements for commercial activity, for example, or the ethical values arising from a body of literature – in which similar social origins could be expected to comprise a sufficient (though not a necessary) condition for participation.39 A number of essays in The Cambridge History of Medieval English Literature make reference to the idea of textual communities when discussing the readership and reception of Middle English literature (testimony to the far-reaching influence of Stock’s concept on medieval studies), including Lollard textual networks in London and Essex; an East Anglian readers’ circle centred on regional collections of saints’ lives; the vernacular ‘self-help’ afforded by early Middle English guides to spiritual well-being; monastic communities responsible for sponsoring literary genres such as hagiography, exegesis, or biblical redactions; women’s textual communities, both as they may have existed in real life and as imagined through literature; implied readerships for Piers Plowman, ranging from the vernacular to the university-trained; and the importance of lowland Scots in the development of a national consciousness, exemplified by Gavin Douglas’s translation of the Aeneid shortly before the Scottish defeat at Flodden (1513).40
Then there are the several audiences for the Middle English romance, which have been characterized at different times as courtly and aristocratic, gentry and middle-class, popular, male or female.41 This variety arises not least because the genre was an exceptionally heterogeneous one, extending over a very long period of time, well into the age of print. Indeed, the continuing popularity of romance stories in chapbooks into the eighteenth and nineteenth centuries should not be overlooked.42 Clearly, different audience groups would have their own ideological interests, which would inform their shared understanding of romance stories. Regardless of social composition, the audience for romance is commonly characterized as being a ‘listening’ audience, and while the romances were literate compositions, they lent themselves to the whole spectrum of receptive experience from listening to singing, recitation, or reading aloud, right through to private reading.
It is accordingly not difficult to envisage textual communities united around a shared understanding and appreciation of certain ballad texts. Angela McShane, for example, argues that political ballads of the period from the beginning of the civil wars to the Glorious Revolution ‘not only supplied a commentary on the upheavals of the period but provided too a political frame of reference for their often unsophisticated, but not uninformed readers’.43 There is a risk, however, of imparting too much weight to the (presumably) shared experience of ballad singing/recitation/reading. The notion of textual community involves a significant degree of the fashioning of a group identity out of the shared experience of interpreting a text, and Richard Green has given voice to a certain scepticism about the blanket application of the idea to groups of medieval English texts and their readers/audiences.44 Instead, he suggests, the later medieval period witnessed the emergence of something more like the ‘interpretive communities’ of reader-response theory, the term originated by Stanley Fish to describe groups of readers who bring a shared ‘horizon of expectations’ to the interpretation of a text.45 That would certainly fit the potentially differing purposes in authoring and presenting medieval romances to courtly and/or gentry audiences, and the corresponding differences in the way they might be received and understood. It would fit, too, with the production and reception of political ballads. Later on, it is sometimes claimed that certain ballads could be expected to have special meaning for particular social groups. The bothy ballads sung by the farm servants of north-east Scotland, the texts of some of which relate directly to their personal and professional circumstances, provide a good example.46 The idea, sometimes expressed, that ‘The Gypsy Laddie’ (Child 200) should be especially dear to the hearts of Gypsy/Traveller singers might fall into the same category. Equally, ballads are as open to variety of interpretation as any other kind of verse or music.47
What this exploration of the textual environment into which the genre emerged does indicate is that, quite regardless of questions of literacy vs. non-literacy and written vs. oral transmission, ballads would be conceived and understood as texts. They were not always written down, but they were understood as if they were. It is reasonable to speak of ballad ‘textuality’ without reference to a specific means of transmission. Although it is even conceivable that a text learned by ear might be memorized in a manner qualitatively different from one learned by eye, these still amount only to different routes to access the same thing – like viewing an object without touching it, or handling it in the dark.
The ballad text, on this model, has an autonomy of its own – not an essentialist, New Critical, signification of its own, but an imaginary, abstract existence, separable from any intended meaning that might accompany its production (singing/recitation/printing) or its reception (listening/reading). A simple illustration of this can be found in the deictic or ‘orientational’ features of ballad language. Deictics are linguistic markers that ostensibly relate to the immediate situation of utterance and hearing (to use terms that deliberately invoke an oral situation, while considering properties that belong to written and oral texts alike), but which in actual fact relate only to imagined constructs in the mind of speaker and listener. A moment’s thought will reveal that, far from introducing immediacy and presence, their effect is to distance and depersonalize the content from both speaker and listener alike.48 Examples of deictics in poetic usage include personal pronouns, anaphoric articles and demonstratives, adverbs of place and time, and the use of the historic present tense.
Thus the poetic ‘I’ and the activity mentioned in the common folk song incipit ‘As I walked out on a midsummer morning’ references neither singer nor listener but a construct – or, rather, separate constructs – in the mind of each, which may or may not largely overlap with one another and/or with a common understanding among a putative ‘balladic’ (ballad-singing, ballad-listening, ballad-reading, ballad-inscribing) textual community. This may be difficult to appreciate because at first blush, ‘midsummer’ might seem temporally objective; but it can still refer either to a specific day or to a less precisely defined season, personally experienced in many different ways. Midsummer has been associated historically, for instance, with bonfires and processions, and practices for love divination – but these customs are strictly localized (temporally as well as geographically).49 The extent to which any of these, or other, associations might actually be relevant is highly dependent on the (real and imagined) identity/ies of the ‘I’ of the incipit.
Consider further the opening stanza of ‘The Outlandish Knight’ (Child 4), one of the most widely collected of English-language ballads, as it appears on a number of printed broadsides of the nineteenth century:
An outlandish knight came from the north lands
And he came a wooing to me
He told me he’d take me to the north lands
And there he would marry me.
Connotations for outlandish listed in the Oxford English Dictionary range from ‘Of or belonging to a foreign country; foreign, alien; not native or indigenous’, through ‘Looking or sounding foreign; unfamiliar, strange. Hence, in extended use: odd, bizarre; going beyond what is considered normal or acceptable; outrageous, extravagant’, to ‘Out-of-the-way, remote; far removed from civilization’.50 The English singer Shirley Collins recalled of her neighbours in Etchingham, Sussex, in the 1970s: ‘Talking one day to Fanny, who was then in her eighties, about her early life, she told me she’d been courted by another man before she married Will, but couldn’t marry him “because he was too outlandish”. “What was so outlandish about him?” I asked, intrigued. “Oh, he came from a village about three miles away,” she replied!’51 Conversely, the analogous character who gave rise to Child’s standard title for the ballad type, ‘Lady Isabel and the Elf-Knight’, is seemingly a supernatural being (Child 4 A) – and there are sufficiently diverse characters in romances alone for a knight to be associated with chivalry and courtly love on the one hand and violence, seduction, and betrayal on the other. Outlandish strangers like the one in Sir Gawain and the Green Knight bring the idea of supernatural shape-shifting, as well as an element of moral examination, into such tales.52
The imagined geographical location of the ‘north lands’, and the sort of inhabitants who might be expected to reside there, are likewise very dependent on pre-existing spatial and social horizons. For someone living in England, the north lands might suggest Scotland; but for a native of southern England the suggestion might just be of a generic ‘north of Watford’, to borrow the modern idiom. Conversely, for a ballad collected in the north-east of Scotland, the north lands might have to be equated with the Highlands; even while an English reader might envisage precisely the Scottish north-east. Not to mention the possible perspectives on ‘wooing’ and ‘marry[ing] me’ which requires an act of imaginative identification on the listener’s part, dependent perhaps on their own gender. But no one believes any of it literally happened to the singer.
Interpretive ideas like these have little to do with the physical presence or absence of a speaker/singer and everything to do with the horizon of expectations of the reader/listener, or of the textual or interpretive community. The physical presence of a speaker would be most unlikely to resolve any of the inherent uncertainties. It is mistaken to conceive of the ballad, whether conveyed orally or in writing, in terms of a transmission or ‘informational’ model of communication, whereby a sender conveys a message of their own devising through a particular channel to a receiver (with, in later modifications of the model, the possibility of a feedback loop from receiver to sender, enabling modifications of the message).53 Now essentially discarded by communication theorists in favour of a diversity of ‘constructivist’ approaches which recognize that in virtually all communicative situations meaning is actively constructed by both sender and receiver, the transmission model is particularly inappropriate in relation to works of imaginative literature, which exploit the potentials of linguistic excess and redundancy, which conversely constitute ‘noise’ in the transmission model.54 Yet the transmission model has a seductive appearance of conformity with ‘common sense’, and (at the risk of setting up a man of straw) it seems that it might also embody something of the immediacy and subjectivity that are sometimes thought to characterize sound or speech – and hence the iconic oral tradition – as opposed to writing.
Yet, so far as we know, oral performance and transmission of ballad texts has always taken place alongside written and printed ‘performance’ and transmission. Indeed, it still does, as a couple of examples – exceptional only in terms of their quality – will illustrate. The Copper family of Rottingdean, Sussex, sing songs some of which have probably been in the family for a couple of centuries and which were written down in songbooks in the 1920s and 1930s, and they still make use of a family songbook in performance even though they must surely know all the items by heart.55 The Traveller culture of north-east Scotland is to a great extent an oral one, even while many of their ballads can be identified with items printed in the eighteenth and nineteenth centuries, and members of one of the celebrated families of Traveller singers, the Stewarts of Fetterangus, have written down their songs in order to enrich their culture and to ensure its continuity.56 In neither of these cases is writing a ‘substitute’ for oral performance and transmission: the two paths exist in parallel, providing complementary routes of access to the respective song traditions.
The idea that, in the sort of textual environment sketched in over the course of this chapter, those who learned texts by ear should find it more difficult to commit them accurately to memory than those who learn them by eye, would seem to be the product of a modern, almost exclusively literate, sensibility. Conversely, there is good historical evidence that, without the visual support of writing, people’s mnemonic powers were often highly developed.57 The clergyman and writer Nicholas Bownd (d.1613) could complain that there were too many people who showed no interest in singing psalms, ‘though they can sing some other vain songs very perfectly; and though they cannot reade themselues, nor any of theirs, yet will haue many Ballades set vp in their houses, that so they might learne them, as they shall haue occasion’.58 In Izaak Walton’s Compleat Angler (1653) a proud mother promises the anglers her (admittedly fictional) daughter, ‘shal sit by and sing you the good old Song of the Hunting in Chevy Chase, or some other good Ballad, for she hath good store of them: Maudlin hath a notable memory’.59 The eponymous hero of an eighteenth-century broadside ballad, The Proud Pedlar, claims: ‘I never sung a Song in all my whole Life, / But I could sing it again.’60 Ballads were composed using familiar linguistic patterns, refrains, and so forth, all of which could serve as mnemonic devices, as well as endeavouring to engage a prospective audience through the choice of narrative material.61 And there is reason to think that melody and words generally have a mutually reinforcing mnemonic effect.
Knowledge of the Bible, in particular, was often excellent among people who could not actually read it for themselves.62 This is a particularly good instance, because it can be presumed that at least some of the imperative towards accurate memory and recall derives from the authority vested in that particular text, considered as the Word of God. With the best will in the world, ballads do not aspire to be considered the word of God, and in most instances not even the word of any particular author. Thus far, the idea of ballad textuality has been used in Stock’s sense of a reference system based on access to a corpus of texts, not always written down but understood as if they were. Ballad textuality, however, must embrace not just the reference system and its routes of access, but the authority that resides in the system of texts. So, while verbatim recall of the Bible might be assumed to be a desideratum, clearly that is not necessarily the case for the ballad.
Variation is frequently stated to be the key, defining, characteristic of ballads and folk songs, and of folk literature at large. Yet the actual level of variation from one instance to another is not necessarily any greater than, say, the different published texts of modernist poets such as W. B. Yeats and Marianne Moore, to name just a couple of examples from mainstream, canonical literature.63 The bulk of collected texts of a single ballad type such as ‘The Outlandish Knight’ (Child 4), or ‘The Cruel Mother’ (Child 20), or ‘Barbara Allan’ (Child 84), or ‘The Maid Freed from the Gallows’ (Child 95), or ‘The Battle of Harlaw’ (Child 163), or ‘The Golden Vanity’ (Child 286), actually vary rather little, although certainly some are more complete than others. Substantial change is usually apparent only with equally substantial chronological and geographical spread. The difference, of course, is that the authority for variation in modernist poetry is generally assumed to lie with the individual, named author,64 whereas in the case of the ballad it is often ascribed to a process, namely oral tradition. Yet if oral tradition is seen to be actually rather beside the point in relation to the transmission and reception of the English-language ballads, and if oral transmission is also perfectly capable of ensuring verbatim reproduction, then the motor for ballad variation must be located elsewhere. What is being posited here is that this motor lies in just the opposite direction from the apparent situation of modernist poetry: it lies in the absence of the sort of authority over the text that is deemed to belong to the individual author.
What can be envisaged for ballads – oral and written alike – is a condition very much akin to what Bernard Cerquiglini has posited to account for variation in medieval vernacular literature that cannot be readily ascribed to an individual author, under the term variance.65 Cerquiglini argues that in the Middle Ages the literary work was ‘a variable’, usually anonymous, to which the fact of an originating authorship was less important than the continual rewriting of the work, which then ‘belonged’ to whoever prepared it and gave it form again.66 A state of purposeful variation – variance – was intrinsic not only to the transmission but to the very aesthetic of medieval vernacular literature. The varying written realizations of a medieval vernacular work represent not ‘errors’, scribal departures from a single text authenticated by the persona of the author, but a state of ‘generalized authenticity’.67 What Cerquiglini seems to envisage is a dispersed textual authority shared across all of the renderings, and perhaps all of the possible renderings, of what can be loosely identified as the ‘same’ literary work.
Middle English literature is indeed generally lacking in a sense of stable textual authority, with many works being either actually or effectively anonymous, becoming varied and expanded in the course of textual transmission, and being realized in manuscripts of varying content and layout.68 Variation characterizes some of the romance texts, where it might be the consequence of scribal activity or more directly of vocal performance, in each case rerouted back into the extant written manuscripts.69 But these are all models founded in the study of written texts, notwithstanding the undoubted importance for medieval literature of oral expression and transmission, and of an oral dimension to scribal copying – scribes are thought sometimes to have worked by reading aloud, or at least sub-vocalizing, and therefore ‘hearing’ at least a part of their texts.70 Cerquiglini’s variance is thus specifically distinguished from Paul Zumthor’s probably rather better-known concept of mouvance, which is tied specifically to the precedence of the voice over writing.71
Cerquiglini, moreover, is prepared to extrapolate his conception of dispersed textual authority: ‘When our literary presuppositions have become sufficiently unhinged that Shakespeare is affected, it is not hard to conceive of the disturbance gradually spreading to most of premodern writings.’72 Here he is alluding to the now widely acknowledged instability that pervades Shakespearian texts – the variant Hamlets, King Lears, Macbeths, and so on, which mean that it is impossible to pin the play down to a single authoritative version, and instead the work we call Hamlet, or King Lear, or Macbeth exists only as an abstract compounded of all its instances. ‘Shakespearian writing is no longer presented as a closed, original, and seminal utterance; it is constant and multiple production.’73
Certainly, this ‘medieval’ model for textual authority – or, rather, the absence thereof – is particularly well suited to the ballads in English. In this manner the circle can be squared: ballads can be understood as texts in Stock’s sense of being not always written down but being understood as if they were, but also as texts that can accommodate variation which may be variously accidental or deliberate. Variation, in turn, though not evidently bounded by an identifiable authority such as the name of an author, is nonetheless in practice constrained by, it would seem, a mutual sense of textual community. That is to say, a shared set of understandings and expectations ensures that in practice a ballad does remain recognizable from one instance to another, one rendition to the next, and that ballad transmission is actually fairly conservative. Empirically, this means that ballad types demonstrate relative stability across time and space, even while their versions can be distinguished one from another – and that variation is present between texts transmitted orally and/or in print, without distinction.
Some ballads have remained quite recognizable, in some cases even down to verbal details, over a period of as much as half a millennium. Either end of that time span might be marked by a record in the form of writing (and at the latter end recorded sound), but frequently there is rather little evidence as to precisely what went on in between. Such ballads may have been written down and/or printed, or they may have been recalled from memory, or more probably there is a largely unknown history of interaction of these processes through time. That is to say, they were not always written down, but they could have been – the converse of which is that they were not always memorized, but they could have been. Both processes permit access to the text, and both can be expected to leave deliberate or accidental textual traces on it – formulaic language, recurrent structural patterns, deictic markers, moralizing conclusions, purposeful variants, as well as the arguably more casual effects of ‘faulty’ memory and typesetting, and so on and so forth. Cerquiglini asserts, ‘in the Middle Ages the literary work was a variable’, and his chapter title celebrates the condition as one of ‘joyful excess’.74 We now know that in one sense or another Hamlet and King Lear and Ulysses are variables, too. The textuality of the English-language ballads was characterized by variance well before the idea began to adhere to such literary icons.
1 Philip V. Bohlman, The Study of Folk Music in the Modern World (Bloomington and Indianapolis: Indiana University Press, 1988), p. 14.
2 Bohlman, Study of Folk Music, pp. 17–24. See also Roger D. Abrahams and George Foss, Anglo-American Folklsong Style (Englewood Cliffs, NJ: Prentice-Hall, 1968), pp. 12–36; Tom Burns, ‘A Model for Textual Variation in Folksong’, Folklore Forum, 3 (1970), 49–56; Eleanor R. Long, ‘Ballad Singers, Ballad Makers, and Ballad Etiology’, Western Folklore, 32 (1973), 225–36.
3 Ian Russell, ‘Stability and Change in a Sheffield Singing Tradition’, Folk Music Journal, 5.3 (1987), 317–58.
4 Albert B. Lord, The Singer of Tales, 2nd edn, ed. Stephen Mitchell and Gregory Nagy (Cambridge, MA: Harvard University Press, 2000).
5 David Buchan, The Ballad and the Folk (London: Routledge & Kegan Paul, 1972). See also James H. Jones, ‘Commonplace and Memorization in the Oral Tradition of the English and Scottish Popular Ballads’, Journal of American Folklore, 74 (1961), 97–112. Lord himself wrote briefly and inconclusively about ballads in an incomplete article, ‘The Ballad: Textual Stability, Variation, and Memorization’, published in Albert Bates Lord, The Singer Resumes the Tale, ed. Mary Louise Lord (Ithaca, NY: Cornell University Press, 1995), pp. 167–86.
6 Albert B. Friedman, ‘The Oral-Formulaic Theory of Balladry – A Re-rebuttal’, in The Ballad Image: Essays Presented to Bertrand Harris Bronson, ed. James Porter (Los Angeles: Center for the Study of Comparative Folklore & Mythology, University of California, Los Angeles, 1983), pp. 215–40. See also Flemming G. Andersen and Thomas Pettitt, ‘Mrs. Brown of Falkland: A Singer of Tales?’, Journal of American Folklore, 92 (1979), 1–24; Holger Olof Nygard, ‘Mrs. Brown’s Recollected Ballads’, in Ballads and Ballad Research, ed. Patricia Conroy (Seattle: University of Washington, 1978), pp. 68–87; Kenneth A. Thigpen, Jr., ‘A Reconsideration of the Commonplace Phrase and Commonplace Theme in the Child Ballads’, Southern Folklore Quarterly, 37 (1973), 385–408.
7 Bohlman, Study of Folk Music, p. 28.
8 Bohlman, Study of Folk Music, pp. 28–30.
9 S. Baring Gould and H. Fleetwood Sheppard, Songs and Ballads of the West (London: Methuen, [1891–95]), p. viii; Cecil J. Sharp, English Folk-Song: Some Conclusions (London: Simpkin; Novello, 1907), p. 101. See also Cambridge, MA, Harvard University, Houghton Library, MS Eng 863, Appendix no. 5, Sabine Baring-Gould to Francis James Child, 23 August 1890, available at http://pds.lib.harvard.edu/pds/view/22250626.
10 Mary Ellen Brown, ‘Child’s Ballads and the Broadside Conundrum’, in Ballads and Broadsides in Britain, 1500–1800, ed. Patricia Fumerton and Anita Guerrini (Farnham and Burlington, VT: Ashgate, 2010), pp. 57–72.
11 Figures in this region were first advanced by Robert S. Thomson, ‘The Development of the Broadside Ballad Trade and its Influence upon the Transmission of English Folksongs’ (unpublished doctoral thesis, University of Cambridge, 1974), p. 274; Rainer Wehse, ‘Broadside Ballad and Folksong: Oral Tradition versus Literary Tradition’, Folklore Forum, 8 (1975), 324–34 [2–12] (p. 333 [11]). They are confirmed by researchers currently working on the material (personal communication).
12 Dianne M. Dugaw, ‘Anglo-American Folksong Reconsidered: The Interface of Oral and Written Forms’, Western Folklore, 43 (1984), 83–103.
13 See Steve Roud, ‘Introduction’, in Street Ballads in Nineteenth-Century Britain, Ireland, and North America: The Interface between Print and Oral Traditions, ed. David Atkinson and Steve Roud (Farnham and Burlington, VT: Ashgate, 2014), forthcoming.
14 But see, respectively, David E. Whisnant, All That Is Native & Fine: The Politics of Culture in an American Region (Chapel Hill: University of North Carolina Press, 1983); Chris Wright, ‘Forgotten Broadsides and the Song Tradition of the Scots Travellers’, in Street Ballads in Nineteenth-Century Britain, Ireland, and North America: The Interface between Print and Oral Traditions, ed. David Atkinson and Steve Roud (Farnham and Burlington, VT: Ashgate, 2014), forthcoming.
15 See, for example, Jonathan Barry, ‘Literacy and Literature in Popular Culture: Reading and Writing in Historical Perspective’, in Popular Culture in England, c.1500–1850, ed. Tim Harris (Basingstoke: Macmillan, 1995), pp. 69–94; M. T. Clanchy, From Memory to Written Record: England 1066–1307, 3rd edn (Oxford and Malden, MA: Wiley-Blackwell, 2013); Adam Fox, Oral and Literate Culture in England, 1500–1700 (Oxford: Clarendon Press, 2000); Barry Reay, Popular Cultures in England, 1550–1750 (London: Longman, 1998), pp. 36–70; David Vincent, Literacy and Popular Culture: England 1750–1914 (Cambridge: Cambridge University Press, 1989). For Scotland, see R. A. Houston, Scottish Literacy and the Scottish Identity: Illiteracy and Society in Scotland and Northern England, 1600–1800 (Cambridge: Cambridge University Press, 1985).
16 For the earlier period, see, for example, David Cressy, Literacy and the Social Order: Reading and Writing in Tudor and Stuart England (Cambridge: Cambridge University Press, 1980); Thomas Laqueur, ‘The Cultural Origins of Popular Literacy in England, 1500–1850’, Oxford Review of Education, 2 (1976), 255–75; R. S. Schofield, ‘The Measurement of Literacy in Pre-Industrial England’, in Literacy in Traditional Societies, ed. Jack Goody (Cambridge: Cambridge University Press, 1968), pp. 311–25; Lawrence Stone, ‘Literacy and Education in England, 1640–1900’, Past and Present, no. 42 (1969), 69–139; Keith Thomas, ‘The Meaning of Literacy in Early Modern England’, in The Written Word: Literacy in Transition, ed. Gerd Baumann (Oxford: Clarendon Press, 1986), pp. 97–131. For the later period, see Barry Reay, ‘The Context and Meaning of Popular Literacy: Some Evidence from Nineteenth-Century Rural England’, Past and Present, no. 131 (1991), 89–129; Vincent, Literacy and Popular Culture, pp. 22–32. For Scotland and northern England, see Houston, Scottish Literacy, pp. 20–109. There is an excellent overview of the subject in Adam Fox and Daniel Woolf, eds, The Spoken Word: Oral Culture in Britain, 1500–1850 (Manchester: Manchester University Press, 2002), pp. 21–33.
17 Roger Chartier, The Order of Books: Readers, Authors, and Libraries in Europe between the Fourteenth and Eighteenth Centuries, trans. Lydia G. Cochrane (Stanford: Stanford University Press, 1994), pp. 1–23.
18 Vincent, Literacy and Popular Culture, p. 275.
19 Patricia Crain, ‘New Histories of Literacy’, in A Companion to the History of the Book, ed. Simon Eliot and Jonathan Rose (Oxford and Malden, MA: Wiley-Blackwell, 2007), pp. 467–79.
20 Thomas, ‘Meaning of Literacy’, p. 107.
21 Clanchy, From Memory to Written Record, pp. 30–35.
22 Walter J. Ong, Orality and Literacy: The Technologizing of the Word (London: Routledge, 1988 [1982]), pp. 6, 11.
23 F. J. Child, ‘Ballad Poetry’, in Johnson’s New Universal Cyclopædia, eds-in-chief Frederick A. P. Barnard and Arnold Guyot, 4 vols (New York: A. J. Johnson, 1881 [1874]), i, 365–68 (p. 365).
24 Clanchy, From Memory to Written Record, esp. pp. 48–53.
25 Brian Stock, Listening for the Text: On the Uses of the Past (Philadelphia: University of Pennsylvania Press, 1996 [1990]), pp. 126–27. A remarkable example of everyday written documentation is provided by the Ely Farming Memoranda, early eleventh-century farming records noted down in Old English. See Clanchy, From Memory to Written Record, pp. 32, 67.
26 On the further complication of the coexistence in post-Conquest England of three languages – English, Latin, and French – with distinct spoken and written functions, see Clanchy, From Memory to Written Record, pp. 199–225.
27 Stock, Listening for the Text, pp. 22–24, 150; Brian Stock, The Implications of Literacy: Written Language and Models of Interpretation in the Eleventh and Twelfth Centuries (Princeton: Princeton University Press, 1983), esp. pp. 90–92.
28 Stock, Listening for the Text, p. 23.
29 Stock, Listening for the Text, p. 37.
30 Stock, Listening for the Text, pp. 20, 140, 146; Stock, Implications of Literacy, p. 7.
31 Stock, Listening for the Text, pp. 19–20.
32 See John C. Hirsh, ‘The Earliest Known English Ballad: A New Reading of “Judas”’, Modern Language Review, 103 (2008), 931–39.
33 Christopher Marsh, Music and Society in Early Modern England (Cambridge: Cambridge University Press, 2010), p. 225.
34 Marsh, Music and Society in Early Modern England, pp. 238–50. See also extracts reprinted in Natascha Würzbach, The Rise of the English Street Ballad, 1550–1650, trans. Gayna Walls (Cambridge: Cambridge University Press, 1990), pp. 253–84.
35 Vic Gammon, Desire, Drink and Death in English Folk and Vernacular Song, 1600–1900 (Aldershot and Burlington, VT: Ashgate, 2008), pp. 10–11.
36 Stock, Listening for the Text, pp. 24–29, 146.
37 T. C. Smout, ‘Born Again at Cambuslang: New Evidence on Popular Religion and Literacy in Eighteenth-Century Scotland’, Past and Present, no. 97 (1982), 114–27.
38 Smout, ‘Born Again at Cambuslang’, p. 124.
39 Stock, Listening for the Text, p. 37.
40 David Wallace, ed., The Cambridge History of Medieval English Literature (Cambridge: Cambridge University Press, 1999), pp. 677–78, 686–87 (Lollards), 626–27 (saints’ lives), 90–91 (spiritual well-being), 349–50 (religious communities), 109–21, 307, 350 (women’s reading), 527–30 (Piers Plowman), 247 (vernacular Scots).
41 Derek Pearsall, ‘Middle English Romance and its Audiences’, in Historical & Editorial Studies in Medieval & Early Modern English for Johan Gerritsen, ed. Mary-Jo Arn and Hanneke Wirtjes, with Hans Jansen (Groningen: Wolters-Noordhoff, 1985), pp. 37–47; Ad Putter and Jane Gilbert, eds, The Spirit of Medieval English Popular Romance (Harlow: Longman, 2000), pp. 20–26; Andrew Taylor, ‘Fragmentation, Corruption, and Minstrel Narration: The Question of the Middle English Romances’, Yearbook of English Studies, 22 (1992), 38–62; Wallace, ed., Cambridge History of Medieval English Literature, pp. 152–76, 690–719.
42 See John Simons, ‘Romance in the Eighteenth-Century Chapbook’, in From Medieval to Medievalism, ed. John Simons (Basingstoke: Macmillan, 1992), pp. 122–43.
43 Angela McShane Jones, ‘“Rime and Reason”: The Political World of the English Broadside Ballad, 1640–1689’ (unpublished doctoral thesis, University of Warwick, 2004), pp. 1–2. See also Angela McShane, ‘Typography Matters: Branding Ballads and Gelding Curates in Stuart England’, in Book Trade Connections from the Seventeenth to the Twentieth Centuries, ed. John Hinks and Catherine Armstrong (New Castle, DE: Oak Knoll Press; London: British Library, 2008), pp. 19–44.
44 Richard Firth Green, ‘Textual Production and Textual Communities’, in The Cambridge Companion to Medieval English Literature, 1100–1500, ed. Larry Scanlon (Cambridge: Cambridge University Press, 2009), pp. 25–36.
45 Stanley Fish, Is There a Text in This Class? The Authority of Interpretive Communities (Cambridge, MA: Harvard University Press, 1980).
46 Peter A. Hall, ‘Farm Life and the Farm Songs’, in The Greig–Duncan Folk Song Collection, ed. Patrick Shuldham-Shaw, Emily B. Lyle, et al., 8 vols (Aberdeen: Aberdeen University Press; Edinburgh: Mercat Press, for the University of Aberdeen in association with the School of Scottish Studies, University of Edinburgh, 1981–2002), iii, xxi–xxxiv.
47 New research into the history of reading, directing attention to the sheer variety of experience, potentially has much to contribute to ballad studies. See, for example, James Raven, ‘New Reading Histories, Print Culture and the Identification of Change: The Case of Eighteenth-Century England’, Social History, 23 (1998), 268–87.
48 Jonathan Culler, Structuralist Poetics: Structuralism, Linguistics and the Study of Literature (London: Routledge, 2002 [1975]), pp. 192–99. For a different perspective, see Würzbach, Rise of the English Street Ballad (but see the review in Folk Music Journal, 6.2 (1991), 238–39).
49 Jacqueline Simpson and Steve Roud, A Dictionary of English Folklore (Oxford: Oxford University Press, 2000), pp. 238–39.
50 OED outlandish, adj. and n.
51 Shirley and Dolly Collins, Love, Death & the Lady, CD (EMI Harvest 7243 8 29860 2 7, 1994), insert notes.
52 Raymond H. Thompson, “‘Muse on þi mirrour . . .’: The Challenge of the Outlandish Stranger in the English Arthurian Verse Romances,” Folklore, 87 (1976), 201–08.
53 Claude E. Shannon and Warren Weaver, The Mathematical Theory of Communication (Urbana: University of Illinois Press, 1949). For a useful introduction, see Daniel Chandler, ‘The Transmission Model of Communication’ (1994), available at http://www.aber.ac.uk/media/Documents/short/trans.html.
54 Jerome J. McGann, The Textual Condition (Princeton: Princeton University Press, 1991), p. 14; Peter L. Shillingsburg, Resisting Texts: Authority and Submission in Constructions of Meaning (Ann Arbor: University of Michigan Press, 1997), pp. 108–09.
55 [The Copper Family], The Copper Family Song Book: A Living Tradition (Peacehaven: Coppersongs, 1995).
56 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.
57 Fox, Oral and Literate Culture, pp. 22–23, 135; Marsh, Music and Society in Early Modern England, pp. 255–56; Thomas, ‘Meaning of Literacy’, pp. 107–10; Vincent, Literacy and Popular Culture, pp. 60–62, 182–84, 194–95.
58 Nicholas Bownde [sic], The Doctrine of the Sabbath (London: printed by the Widdow Orwin, for Iohn Porter, and Thomas Man, 1595), p. 241 [ESTC S113231].
59 [Izaak Walton], The Compleat Angler; or, The Contemplative Man’s Recreation (London: printed by T. Maxey, for Rich. Marriot, 1653), p. 203 [ESTC R202374].
60 The Proud Pedlar ([London?, 1740?]) [ESTC T46054; Roxburghe Ballads 3.656].
61 Marsh, Music and Society in Early Modern England, pp. 284–86.
62 Fox, Oral and Literate Culture, p. 22.
63 George Bornstein, Material Modernism: The Politics of the Page (Cambridge: Cambridge University Press, 2001).
64 Although the gist of Bornstein’s argument, and of much recent materialist criticism, is that the shape of the text is often driven by factors beyond the author’s immediate control – place and format of publication, presumed readership/audience, external political developments, and so on.
65 Bernard Cerquiglini, In Praise of the Variant: A Critical History of Philology, trans. Betsy Wing (Baltimore: Johns Hopkins University Press, 1999).
66 Cerquiglini, In Praise of the Variant, p. 33.
67 Cerquiglini, In Praise of the Variant, pp. 34, 51.
68 Tim William Machan, Textual Criticism and Middle English Texts (Charlottesville: University Press of Virginia, 1994).
69 Albert C. Baugh, ‘Improvisation in the Middle English Romance’, Proceedings of the American Philosophical Society, 103 (1959), 418–54 (esp. pp. 434–40); Nancy Mason Bradbury, Writing Aloud: Storytelling in Late Medieval England (Urbana and Chicago: University of Illinois Press, 1998), pp. 1–21; Taylor, ‘Fragmentation, Corruption, and Minstrel Narration’, esp. pp. 43–46.
70 H. J. Chaytor, From Script to Print: An Introduction to Medieval Literature (Cambridge: Cambridge University Press, 1945), pp. 13–21.
71 Cerquiglini, In Praise of the Variant, pp. 84–85 n. 10. See Paul Zumthor, Essai de poétique médiévale (Paris: Éditions du Seuil, 1972), p. 507; Paul Zumthor, La lettre et la voix: De la ‘littérature’ médiévale (Paris: Éditions du Seuil, 1987), p. 160.
72 Cerquiglini, In Praise of the Variant, p. 39.
73 Cerquiglini, In Praise of the Variant, p. 39. See further John Jowett, Shakespeare and Text (Oxford: Oxford University Press, 2007).
74 Cerquiglini, In Praise of the Variant, p. 33.