In a piece titled ‘Of the obseruation, and vse of things’, the courtier, diplomat, and essayist Sir William Cornwallis (c.1579–1614) wrote:
Pamphlets and lying Stories and News and twoo penny Poets, I would know them but beware of beeing familiar with them. My custome is to read these and presently to make vse of them. For they lie in my priuy; and when I come thither and haue occasion to imploy it, I read them (halfe a side at once is my ordinary), which when I haue read it, I vse in that kind that waste paper is most subiect too but to a cleanlier profit.1
There is little doubt that the early modern broadside ballads fall into Cornwallis’s category, and his observations are a salutary reminder that paper itself was for a long time a physical commodity the value of which could easily outweigh the aesthetic value of whatever was printed on it.2 Still, although the need for paper for hygienic purposes remained, we might imagine that lower down the social scale a greater intrinsic value could have attached to the broadside or chapbook as physical object, a prized possession to be preserved rather than used in the privy.3
Another kind of value to the physical ballad lies in the fact that broadsides were normally illustrated with woodcut images, and could thus be used to provide a cheap form of decoration.4 The practice of pasting ballads on to the walls of alehouses and domestic dwellings is well attested. Izaak Walton, for example, writes in The Compleat Angler (1653), ‘I’ll now lead you to an honest Ale-house, where we shall find a cleanly room, Lavender in the windowes, and twenty Ballads stuck about the wall’.5 Decoration did not, of course, preclude reading. In The Spectator, no. 85 (7 June 1711), the paper in which he goes on to discuss the ballad of ‘The Children in the Wood’ (Roud 288), Addison observes:
This my inquisitive Temper, or rather impertinent Humour of prying into all sorts of Writing, with my natural Aversion to Loquacity, give me a good deal of Employment when I enter any House in the Country; for I can’t, for my Heart, leave a Room before I have thoroughly studied the Walls of it, and examined the several printed Papers which are usually pasted upon them.6
At a later date, the memoirs of the writer Thomas Holcroft (1745–1809) recall how such ballads formed an integral part of his early education:
Even the walls of cottages and little alehouses would do something; for many of them had old English ballads, such as Death and the Lady, and Margaret’s Ghost, with lamentable tragedies, or King Charles’s golden rules, occasionally pasted on them. These were at that time the learning, and often, no doubt, the delight of the vulgar.7
Educated gentlemen like Addison and Cornwallis (but probably not the young Holcroft) might have read ballads silently to themselves, but throughout the period covered by these quotations and beyond, reading would have generally meant reading out loud.
The physical presence of the ballad, it seems safe to assume, must have been both unavoidable and unremarkable for virtually the whole of the time during which printed broadsides and chapbooks were in common circulation, say from the sixteenth to the nineteenth century. Certainly, for the booksellers and printers, and for the pedlars and chapmen who were responsible for their dissemination during much of their history, ballads were first and foremost material objects to be carried and sold around the country.8 The materiality of texts has become the concern of literary scholars engaged in the now well-established discipline that is sometimes termed l’histoire du livre, or materialist criticism, or analytical bibliography. In practice, this embraces a continuum of concerns, from the cultural and economic history of book production, distribution, and preservation, through the sociology of texts, to the signifying functions of physical characteristics such as typography, layout, design, and illustration, paper and binding, iterations of editions, and so forth.9 These last are sometimes termed ‘bibliographic codes’ – that is, ‘the symbolic and signifying dimensions of the physical medium through which (or rather as which) the linguistic text is embodied’.10 The application of this approach to the ballad poses certain special problems, because in principle a single ballad text, both words and music, can simultaneously inhabit a number of material formats. Thus sound can exist in performance and in various recording media and digital files, while writing can exist as manuscript, as print, and again as digital files. This sort of multiplicity requires careful differentiation if signifying characteristics are to be elucidated from such a range of physical forms.
In the first instance, a ‘text’ can be defined as a unique sequence of words and pauses, or of musical notes and pauses.11 Generally, such definitions go on to specify that this sequence must be contained within a single physical form – that is to say, an individual item comprising a discrete physical medium, such as a manuscript, a printed book, a phonograph cylinder, LP record, or CD. This ‘vessel’, as it were, that contains the text is termed a ‘document’. This is very important: because it is often difficult to conceive of a text in isolation from its physical form, the ‘vessel’ or document that it inhabits, it is all too easy to conflate the two separate concepts of text and document. The distinction, however, is not impossible to maintain, if only because exactly the same sequence of words, or musical notes, can inhabit such different physical entities as a collector’s notebook and a phonograph cylinder.12 Ideally, a collector’s accurate transcription of a ballad and a sound recording of the same rendition would contain exactly the same text, but within a quite distinct physical container or document. So it is possible to abstract the text from its physical container, even if it requires some considerable exercise of the imagination in order to do so. But it is certainly very difficult then to do anything with it that does not involve either simply placing it back into the same container, or transferring it into a different one. To consider a text is in practice almost necessarily to allude also to the document within which it is contained – the ‘packaging’ that necessarily surrounds it – so that the unit of discussion is commonly what might be termed the ‘documentary text’.13
In the case of a ballad that is simply sung or recited out loud, the physical document is comprised, not of sequential written or printed marks on paper, the contoured grooves on a phonograph cylinder, or etched pits and lands on a CD, but of sequential vibrations in air – the invisible, gaseous, but nonetheless physical substance that surrounds us. To be sure, the document that is a ballad performance is ephemeral.14 Its transience as a document, however, is nonetheless exactly comparable with that of the grooves of a phonograph cylinder, the wax substrate of which might be shaved in order that the cylinder could be reused.15 Likewise, the pattern of impulses on magnetic recording tape can be destroyed by demagnetizing the tape, and pencil marks on a sheet of paper are always susceptible to being erased. Perhaps it is worth drawing a distinction between these sorts of situations and that of the broadsides that, if Cornwallis is to be believed, ended up as toilet paper. All of these instances witness the destruction of a document, but in the case of those broadsides that would involve the despoilment not just of the writing but of the physical medium, the paper, itself. It is a nice distinction, between what we might call the ‘textual carrier’ – writing, contoured grooves, etched pits and lands – and the ‘substrate’ – paper, wax cylinder, magnetic tape, CD.16 The attenuation of the vibrations in air that carry a sung ballad is not quite to be equated with the degradation of the earth’s atmosphere itself.
Early ethnographic collectors using the phonograph habitually treated their recordings as transient documents, the value of which was simply that they would enable the collector to derive an accurate written transcription, which, presumably, was thought of as having greater permanence (and probably precision).17 Pragmatically, Percy Grainger recommended writing down the words and melodies independently of the phonograph recordings, partly because of the danger of damage to the cylinders, partly because it is not always easy to distinguish unfamiliar words.18 The texts the cylinders contained were carefully (sometimes not so carefully) transferred from one document to another, and from one form of textual carrier to another, and the cylinders, it seems, were often subsequently discarded – a practice that neatly, albeit incidentally, illustrates the point that it is extremely difficult to dissociate text from an awareness of its physical packaging. It is possible to read at the heart of the Latin tag littera scripta manet, ‘the written word remains’, a simple preference for the stability of some physical substrates (papyrus, parchment, paper) over the fragility of others (air, wax, magnetic tape, the flimsy metal coatings of CDs).
Conversely, modern approaches to folk literature tend to embody a preference for the more vulnerable substrates associated with sound rather than writing. There are a number of interrelated reasons for this. They include, from a historical perspective, the Enlightenment ‘discovery’ of oral tradition and the opening up of a perceived dichotomy between orality and literacy, which has enabled oral tradition to continue to enjoy iconic status. From a more theoretical perspective, the preference for sound over writing in the study of the ballad can be considered as a (rather minor) subset of ‘phonocentrism’, the general precedence that, Derrida argues, has been afforded to speech over writing in Western metaphysics.19 The study of folklore, in some manifestations at least, has seen a shift away from the study of texts of any sort and towards context, performance, and ethnography.20 Finally, there is the mere fact that ballads, folktales, and the like are susceptible to vocal performance, which is the source of much aesthetic pleasure – and which in the ballad case gives access to the melody, which in its written form is frequently inaccessible to the non-specialist. Nevertheless, these are simply not very good arguments for the primacy of the substrates associated with sound over those associated with writing.
An acknowledgement of the significance of documentary media is scarcely a new idea. Marshall McLuhan’s famous (if not unproblematic) dictum ‘the medium is the message’ provides the classic expression of this insight.21 Nevertheless, the physical packaging of ballad texts remains worthy of further consideration. Here, the work of Harold Innis, a notable influence on McLuhan, who considered the ‘bias’ towards either time or space that is inherent in different systems of communication, provides a useful starting point.22 Thus stone tablets bearing hieroglyphs exhibit a strong temporal bias, in that they can endure for millennia, but they cannot readily be transported from one place to another. Conversely, texts written on paper, being light and easily transportable, can be disseminated in space, but their very lightness renders them vulnerable to degradation and loss with time.
Now, this observed inverse relationship between time and space is not directly transferable to the ballad field. For example, while printed materials disseminated by chapmen and pedlars from early modern times provide one of the readiest explanations for the presence of recognizably similar sets of ballad words in widely separated places, they also provide some of the oldest records of ballads. On the other hand, broadsides rarely carry music notation, although earlier examples do mostly include the name of a tune, and the dissemination of ballad melodies through both space and time is assumed to be largely dependent on an oral/aural communication system. Paradoxically, such a system is the most ephemeral imaginable, in terms of both space and time. Communicative biases certainly are present in ballads when they occur as different kinds of physical documents, but to trace these it is necessary to consider a range of the signifying qualities that are presented by ballads as documents.
Even supposing an ideal situation where the text of a ballad (the sequence of words and pauses, and of musical notes and pauses) is identical whether carried by a written transcription or by a sound recording, and therefore what that text actually ‘says’ should in principle be unvarying, the documentary signals will be quite different. Most obviously, of course, the textual carriers within the two documents belong to two different semiotic systems. One employs physical marks on paper to be decoded by the human eye and brain; the other, physical arrangements (contoured grooves, patterns of magnetic impulses, etched pits and lands) to be decoded first by a mechanical apparatus and subsequently by the human ear and brain. ‘Biases of the ear and eye’, in Daniel Chandler’s phrase,23 will therefore operate in the apprehending of even a single ballad text through different semiotic systems.
Even prior to that stage, the moment one is presented with a broadside or a CD, one is faced with a distinct set of requirements and expectations. These are instantly signalled by the packaging, even before any direct reference is made to the respective semiotic system – that is, before the broadside is read, or the CD inserted into the player. It is difficult to compile a systematic list of the ways in which different physical formats give out different signals, but one means of suggesting something of the range is to draw a few ready comparisons between some different documentary formats – in particular, phonograph recordings and printed broadsides, with some mention of modern digital recordings and ‘pencil and paper’ collecting, along with the ephemeral materiality of performance. Among the properties of documents that suggest themselves for comparison are those of (i) permanence, (ii) accessibility, (iii) precision, and (iv) information capacity.
(i) Unlike performance, both broadsides and phonograph cylinders constitute documents that signal a degree of permanence, although the broadsides – which, of course, have a much longer history – can more easily be seen to have stood the test of time. Yet it is likely that their cheapness and portability were initially valued more than their potential permanence. Besides their use in the privy, broadsides and all sorts of printed sheets could also be employed as wrapping paper, for book bindings, for lighting fires, and for writing on the verso and in the blank spaces, and it is generally thought that numbers of printed ballads have not survived in so much as a single copy.24 Nevertheless, time has in some degree vindicated the format. The pasting of broadsides on walls testifies to the fact that at least some contemporary consumers were aware of their potential longevity. And treated with moderate care, especially in libraries and private collections, some broadsides have readily survived four centuries or more.
A phonograph cylinder kept in a suitable conservation environment can equally be expected to last for a long time – that is, so long as it is not often played, for frequent listening to a cylinder can cause damage to both the substrate and the textual carrier. Stylus wear results in loss of signal as well as the acquisition of quantities of extraneous noise. Much the same can, in fact, be said about the effects of playing wear on acetate or vinyl records and on recording tape. Although in the domestic context the textual carrier of a CD is pretty well immune to damage from repeated playing, the protective plastic covering and lacquer film which constitute part of the substrate in toto are very vulnerable to scratching, which can on occasion render the whole document unreadable. From an archival point of view, moreover, substrate decay over time, as well as format obsolescence, pose real threats to the longevity of recordable optical media.25 Wax cylinders are fragile, easily broken if accidentally dropped, and subject to ageing effects and especially to mould, which can produce auditory artefacts and ultimately render the cylinder unplayable.26
In contrast, while reading a broadside necessarily entails some handling, which over a long period of time will cause damage to the substrate, short of it becoming so torn or grimy that the words cannot be read (which can happen), or its being deliberately sabotaged by overwriting, just so long as the substrate remains intact then the textual carrier is likely to remain essentially unimpaired. But the real permanence of the printed broadside lies not just in its relative durability, but in its multiplicity. Print runs for broadsides numbered at least in the thousands,27 multiplying the chances of survival of the texts inscribed thereon. The wax cylinder and its text, while quite probably treated with much greater care, might also be unique. Elizabeth Eisenstein describes the preservation of texts as perhaps the most important contribution of the printing revolution, and is at pains to point out that this derives not so much from quality as from quantity.28
(ii) Broadside ballads were readily accessible. They were available at low cost, usually priced at a halfpenny or a penny, and though the relative value of that sum varied over time, the price generally remained low and can be taken as an indicator of access.29 Ballads were widely distributed by pedlars and hawkers, and by the later eighteenth century they were being issued by booksellers/printers across England and Scotland (and Wales). Ballads, as noted above, were also frequently displayed in public. Broadly, accessibility was a consequence of a comparatively cheap paper substrate and the development of a printing process that permitted the multiple imposition of a textual carrier at low cost, although pricing structures within the book trade at large could be complicated by factors beyond the manufacturing costs alone.30
In contrast, phonograph recordings (of ethnographic items such as ballads – we are not concerned here with the wider, commercial phonograph and gramophone market) were relatively expensive manufactured objects, and they required mechanical coding and decoding on a suitable, and again relatively expensive, machine.31 Several of the early folk song collectors (in fact, probably most of them) at least experimented with phonograph recording, and they shared access to the machines.32 The practicalities of transporting the heavy machine and the controlled conditions required for recording imposed spatial limitations on its use, although at various times recordings were made, for example, on the Isle of Skye and in rural Wales, as well as in various places in England. Phonograph cylinders are essentially unique items. The machines themselves were never exactly common and have now become quite rare, imposing a further limitation on the pragmatic longevity of the texts they were used to create. Although digitization has facilitated the (still relatively limited) redistribution of the recordings, it has done so only by transferring the texts to another substrate and carrier – another kind of packaging which, as already suggested, poses problems of its own in relation to the issue of permanence, even if it helps address that of accessibility. Unlike the wide geographical distribution of broadsides and chapbooks, that of phonograph recordings was strictly limited.
Recording technologies at large have accorded to sound the quality of reproducibility, which was long regarded as the particular characteristic of written words and musical notations. This has been a gradual process, however, reaching its fruition well after the close of the period of phonograph recording. On the other hand, once set up to play, in principle anyone could access phonographic sound, whereas access to a printed broadside requires that at least someone possess the basic ability to read it. To some extent, in the past this potential obstacle was bypassed by virtue of the ballads being sung or recited by their vendors, as well as through the existence of the ‘reading communities’ discussed in chapter three. Right up to the twentieth century, the presence of cheap print of the broadside kind – and also such things as newspapers, magazines, hand-bills, and so forth – signalled not so much the activity of private, silent reading, as that of communal, oral, reading aloud, recitation, or even singing. The phonograph, in contrast, remained a more socially privileged technology.
(iii) Phonograph recording required particular environmental conditions in order to be successful. The wax, for instance, needed to be reasonably warm in order to retain its plasticity. Cecil Sharp expressed reservations about the tendency of the phonograph to blur words and to emphasize certain consonants more than others; Anne Gilchrist made similar observations and criticized the machine’s limited dynamic range.33 Modern authorities have endorsed such reservations, additionally pointing out such things as the frequency limitations of the machine.34 Even Percy Grainger, often portrayed as the champion of the phonograph, was prepared to concede some of these points.35 Thus the precision of wax cylinder recordings is strictly limited.36 On the other hand, they can certainly preserve elements of musical style, vocal ornamentation, and so forth, which are extremely difficult to represent graphically in stave notation. Grainger’s phonograph recordings of Joseph Taylor and other Lincolnshire singers provide a prime illustration of this capability.37 His own complex notations of some of the same songs offer a direct point of comparison and invite debate as to whether notation on paper really is capable of attaining the same degree of precision.38
Conversely, printed broadsides generally attain a high degree of precision in the representation of ballad words. Of course, such faults as typographical errors, as well as broken or turned letters, can occur, type can fall out, or a printer can set nonsense.39 By and large, though, the verbal infidelities that do arise are not sufficient to inhibit comprehension, and while they undeniably fall short of the aim of complete precision, they are often much less serious than the shortcomings of phonograph recordings. Broadsides do not generally carry music notation, but many of the earlier examples do carry a tune direction, ‘To the tune of . . .’. This can be taken as a sufficient shorthand representation, at least for anyone familiar with the melody under the title given. A number of late seventeenth- and eighteenth-century broadsides do carry notation, although there is some reason to think that this is sometimes (though certainly not always) decorative in purpose and musically either nonsensical or misleading. Later broadsides mostly do not carry even an indication of the tune, which has to be counted as a loss. A number of popular tunes would have been accessible from other published sources, such as Playford’s dance and music books, and anthologies like Johnson’s Scots Musical Museum. Presumably, though, the general assumption was that the tune would either be known or else learned from the ballad seller.40
At that point, print and performance are closely connected. Performance, however, remains crucially dependent on individual skill, in both the handling of melody and the pronunciation of words, to transmit a ballad text with precision. A garbled word can be much harder to make out than a poorly printed one, a badly rendered tune more difficult to interpret than a scribbled notation. It is not a matter of chance that written or printed words and music have proven to be the favoured tools for scholarly analysis (well beyond the field of ballad studies). The precision of performance is also constrained by environmental factors such as extraneous noise, acoustic properties of the location, physical distance or barriers that might otherwise interfere with transmission of a transient, once-only signal. These are considerations that can impact upon sound recordings, too – even the most sophisticated of modern digital recordings.41
(iv) Both sound recordings and printed documents carry large quantities of information both integral to and supplementary or complementary to the verbal and musical text. Vocal inflections and marks of punctuation, for example, can impart elements of interpretation, such as a lift in the voice or the presence of a question mark at the end of a sentence. These things can reasonably be regarded as integral to the text. Written or printed texts often provide substantial punctuation, indicating such things as a change of speaker through the presence of quotation marks (which just occasionally can prove more ambiguous than otherwise). Conversely, the absence of audible punctuation from sound recordings might invite uncertainty or confusion, but might also preserve a fluidity that can be thought of as integral to ballad texts.
Typographic layout organizes ballad narratives into blocks of lines and stanzas which correspond (most of the time) to syntactic units – a kind of punctuation by spatial layout. At the level of the stanza, though less so at that of the line, this effect is paralleled in sound by the strophic nature of ballad tunes. Just occasionally, too, such patterns may prove disconcerting. Most ballad lines are in some degree end-stopped, with sense and metre coinciding in a pause at the line end, so that the rare case of enjambement, where the sense requires the eye to run on beyond the end of the line without a pause, can come as something of a jolt (and even more so where the syntax actually runs across a stanza break). This sort of occasion prompts a readjustment of expectations on the part of the reader, which may be either inhibited or facilitated by typography – inhibited by the stanza break, or facilitated by the absence of the expected full point. Layout, spelling, and punctuation all enable the closer interrogation of texts, and in very much the same way the details of musical notation enable the analysis of melodies.
While the symbols of musical notation and of writing or print are readily considered as integral to the texts they represent, this is not the case with all the kinds of information that can be carried by a document. The gothic typeface of a black-letter broadside, for example, which can be considered as merely ancillary to the physical nature of the textual carrier, nonetheless also signifies something about the historicity of both the document and the text it contains.42 Indeed, the entire mise en page, or mise en livre, of the broadside situates the document within the economic and ideological history of cheap print, just as the scratchiness of the sound from a phonograph recording situates the document within the early history of recording technology. Similarly, the handwriting in a manuscript or collector’s notebook can signify care taken in transcription, perhaps distinguishing a fair copy from a rough copy, thus signalling the relative authority of the text. Information of this kind largely takes on its significance with a historical perspective, which is akin to saying that its signification is in some sense accidental. The information capacity of documents need not necessarily be reflective of ‘intention’.
Characteristics such as the vocal quality or timbre of a performance or on a recording might also be compared with the choice of typeface on a broadside. The presence of woodcuts and other decorations might be considered roughly equivalent to some of the other kinds of information that can accompany sound, such as movement, gesture, and body language. (Broadside woodcuts, it is perhaps worth observing, are entirely separable from the accompanying words and, it seems, were sometimes used as wall decorations in their own right.43) Fieldworkers often value video recording for its capacity to capture a wide range of extratextual information generated during a performance and to facilitate its analysis. These are all further examples of the capacity for documents to carry supplementary signifying information that is, strictly speaking, outside of the domain of text as it has been defined here.
Clearly, many or all of the aspects touched on in the foregoing comparative remarks could be elaborated in much more detail, but the purpose of this admittedly rather impressionistic account of permanence, accessibility, precision, and information capacity in documents is simply to point to some of the ways in which performance, print, and sound recordings bring with them their own sets of expectations. They do not conform in a straightforward manner with Innis’s model of communication systems and its inverse biases towards time and space, although there is perhaps a trade-off to be discerned between time and space on the one hand, and information capacity on the other. This is especially apparent when the extratextual capacity of live performance is considered against its ephemeral nature. Conversely (from the present historical standpoint), although printed broadsides appear biased towards the dimensions of accessibility and permanence, an awareness of the importance of the broadside bibliographic code(s) expands greatly the range of information that can be read from the document.
A consequence of these sorts of comparisons, then, is to pose a challenge to the idea that primacy can necessarily be ascribed to any particular documentary format, to any one or other substrate. Scholars of an ethnographic persuasion, for instance, are perhaps likely to hold that the primary manifestation of the ballad is in performance. Failing that, a sound recording might represent the primary document for the realization and preservation of the ballad text (as opposed, say, to a collector’s notation). That position no doubt embodies an underlying deference to phonocentrism, but more immediately it reflects the ethnographic interest in aspects of performance (style, expression, vocal quality) that, strictly speaking, lie outside the ballad text per se. Yet, as has already been noted, written or printed words and music are the favoured tools for the critical analysis of words and melody. Other researchers might be equally interested in the physical characteristics of a collector’s notation or, more likely, the typeface and illustrations of a broadside print. These are all things that are preserved in documents, and all make a substantial contribution to the historical, aesthetic, and intellectual import of the various kinds of whole. It is incomplete to consider the ballad in isolation from its ‘packaging’.
The pursuit of a single ballad text through all of its potential documentary formats further demonstrates just how misleading it would be to ascribe ‘intention’ to it. The ballad text in this scenario is simply an intangible object that can migrate from one documentary format to another. History, it is worth observing, has quite undermined the charge that Margaret Laidlaw, James Hogg’s mother, is supposed to have laid against Walter Scott: ‘there war never ane o’ my sangs prentit till ye prentit them yoursel’, an’ ye hae spoilt them awthegither. They war made for singing an’ no for reading; but ye hae broken the charm now, an’ they’ll never be sung mair. An’ the worst thing of a’, they’re nouther right spell’d nor right setten down.’44 The text of this famous anecdote exists in another version which omits the claim that ‘ye hae broken the charm now, an’ they’ll never be sung mair’.45 But perhaps even more curious is the paradox it embodies (in both versions): the ballads were ‘intended’ only to be sung, and yet there was nonetheless a ‘correct’ way to transfer them into print (‘right spell’d’ and ‘right setten down’), of which Scott was accused of having fallen short.46 The question of sound versus writing is considered in chapter five. Meanwhile, a broadside can equally serve to decorate a wall as to preserve for future generations a set of ballad words and at least the name of a tune. A collector’s notebook may be used to analyse narrative and melody or it may simply be put away indefinitely in an archive. And while either of those paper-based documents could enable someone to sing the ballad again, either could just as easily be used to make a shopping list or to wrap a parcel – while the vibrations in the air that constituted a performance have long since attenuated beyond the possibility of detection.
One of the things that this brief consideration of material properties of the ballad is intended to underline is that the text itself – at least in the idealized scenario proposed, where it remains the same from one format to another – is to all intents and purposes intangible. Short of some kind of imaginary telepathic transfer, the text cannot be either expressed or transmitted without the packaging of a physical document, be it a performance, sound recording, printed sheet, or something else. All of those physical documents signify in different ways, so that it becomes extremely difficult to prioritize between forms of ballad ‘packaging’. Rather than comprising a hierarchy of documents, then, each contributes to an intangible idea of what is the ballad.
Scholarship is familiar with the perspective of ‘version’ and ‘type’, and is beginning to awaken to the problems inherent in the notion of ‘version’ vis-à-vis the particular instance or intellectual item – each rendition by the same singer, or each copy of the same broadside printing, for example – which is by definition materially different and therefore distinguishable from any other. There is a parallel here with Leo Treitler’s idea of ‘exemplification’ to describe the status of the individual, and potentially variant, written score in the ontology (that is to say, the nature and conditions of identity or being) of works in Western art music:
Obviously we can’t read any one of these scores as unique identifier of the work, or denotation of the sound pattern that must be executed to produce a flawless performance of the work. Each score can do no more than exemplify the work. In some traditions exemplification is the main task of scores. In such a tradition identification and instruction are accomplished only indirectly, through exemplification. The score in such a tradition is functionally and ontologically parallel, at the same theoretical level, with the performance. The questions ‘what is this piece’ and ‘how does it go’ could both be answered either by writing out a score or by making a performance.47
What Treitler is arguing here is that the individual documentary text, which in this instance comprises a score, merely ‘exemplifies’ the larger, abstract whole that is the musical work – which might equally be exemplified by another score, a performance, or a recording.
The crucial distinction here is between on the one hand the concrete exemplification, and on the other the work as an imagined whole. But the imaginary entity that is a musical or literary work is itself founded upon whatever exemplification(s) of it happen(s) to be available to any particular individual. This may be, for example, the originator, an editor, a reader, or a listener – whose access to particular exemplifications, and whose perspectives on the whole, can be expected to vary accordingly.48 Treitler’s precise equation of score with performance places two quite different kinds of documentary text on exactly the same theoretical level, each to stand as a concrete and discrete exemplification of the work. His score and performance, in the ideal situation, might contain exactly the same sequence of musical notes and pauses (and, on occasion, words and pauses as well) – that is to say, the same text – in just the same way as, say, the idealized collector’s notation and phonograph recording imagined earlier in this chapter. Yet there is little doubt that the respective documents – the score written or printed on paper, and the performance sounded through the air of the concert hall – will continue to signify in very different ways.
In ‘History and the Ontology of the Musical Work’, Treitler is also seeking to demonstrate that multiplicity does in fact pervade Western art music – to redress the view that works have been ‘cast once and for all by a composer in a final form that is represented in a score and presented essentially unchanged in performances’.49 In fact, the ‘work concept’, exemplified by the music of Beethoven, is a product of the history of musical practice and aesthetics which only emerged c.1800.50 Nevertheless, it has become a more or less accepted principle of musical aesthetics (or at least of the aesthetics of Western art music) and it more or less effectively closes off the notion of multiplicity. Treitler maintains, however, that it is not in fact difficult to find exceptions to disprove the assumed rule that the musical work is determinate, fixed once and for all by its composer. He cites examples such as Stockhausen’s Klavierstück XI, where the order of the sections is determined only by the performer at the moment of performance, or a particular performance of a raga by Ali Akbar Khan – and it is not difficult to imagine further examples from genres such as jazz and blues. Multiplicity, moreover, is to be found not just among items that can be realized only in live performance, that are not cast into a final form represented in a score, and where the text can reasonably be expected to vary from one instance to the next, but also among items from the mainstream of Western art music. Treitler points to pieces by Chopin, such as the Nocturne in B major, op. 62, no. 1, which exist (or have existed) in several variant forms, both in writing and in print, as well as in performance. Textual indeterminacy in a work like the Chopin Nocturne can be directly compared with that encountered in medieval chant, as well as in more modern, more ‘experimental’ pieces. Even in rock music, an item as it is performed can vary considerably from a supposedly ‘definitive’ recording.
As we saw in chapter one, these ideas of textual indeterminacy lend themselves well to the ballad situation, with regard to types and versions, and transmission and reinterpretations. What is being added to the argument here is an emphasis on the material aspect of the items that fall under the umbrella of the work. For Treitler and for Western art music, these are primarily score and performance. The ballads, too, are nothing if not multifarious in the documentary sense. When one reads a collector’s manuscript notebook or a printed broadside, places a CD in the player, or attends a performance, the action is accompanied by a set of expectations that are to an important extent dependent on the nature of the substrate and textual carrier that provide the material base for the document in question. Each of these different documents offers a distinct exemplification of the ballad as work, even in the ideal instance where the text remains identical from one document to the next. In that idealized situation, the different documents all provide different access routes to the same thing, but together they render its realization increasingly rich and rounded – permanent and accessible, accurate and detailed, filled out with extratextual information, present to both eye and ear. Indeed, one reason for the resilience of ballad words and music can be posited to lie in their ability to inhabit different material formats – different documents – and to move with apparent ease between them.
Treitler also maintains that the perceived boundaries of the work are historically determined, that they can and do change with time and historical circumstance.51 Modern performers, he observes, tend not to stray from the published scores of Chopin and are usually unaware of the extent to which their predecessors did so. While the ontology of the ballad is informed by the entire range of its actual and potential documentary exemplifications, the conception is also potentially subject to change over time. Thus recorded sound is a fairly recent phenomenon, while the modern publication of ballad books scarcely matches up to the ubiquity of broadsides in, say, the seventeenth, eighteenth, or nineteenth centuries.52 Concerning the extent of actual performances, we are sadly under-informed for the earlier periods, but it is at least worth observing that the elevation of sound to primary status in ballad ontology coincides with the long decline of broadside publication after the middle of the nineteenth century.
Prior to and after a period from, say, the end of the nineteenth century up until the Second World War, as well as being something you (or someone else) might sing, the ballad was something you could pick up and hold in your hand. Earlier, that would have taken the form of a broadside or chapbook; later, it would be a record of some kind, an LP or a CD. During that putative interval in the first part of the twentieth century, there were both publications and recordings, but, for a combination of technological and economic reasons, they were probably rather more thin on the ground. The interval in question happens to coincide with the folk song revival, when large numbers of songs were collected from a supposedly oral tradition, while many (though not all) of the collectors themselves were uneasy about such printed broadside copies as they did encounter. It is a moot point as to whether it would really be possible to draw a causal correlation between the documentary environment and the ideological impulses behind the folk song revival, but it does look as if there was at least something of a temporal correlation.53 It is certainly easier to overlook the material nature of a singer’s performance than of a broadside ballad.54 If it is correct that the perceived boundaries of the work are historically contingent – and the argument certainly appears both sensible and convincing – then it is at least worth giving some consideration to ways in which the timing of the folk song movement might have impacted on its direction and content.
1 Don Cameron Allen, ed., Essayes by Sir William Cornwallis, the Younger (Baltimore: Johns Hopkins Press, 1946), p. 50. As an aside, at a later date, when the Cheap Repository Tracts were introduced at the end of the eighteenth century, the chapmen who distributed them insisted from the beginning that they be printed on soft paper, so that they would find takers even if they were never read. See William St Clair, The Reading Nation in the Romantic Period (Cambridge: Cambridge University Press, 2004), p. 354. Since the Cheap Repository Tracts were intended to price out the long-standing ballads and chapbooks and replace them with their own moral tales, this is a rather gratifying observation.
2 St Clair, Reading Nation, pp. 26–27; Margaret Spufford, Small Books and Pleasant Histories: Popular Fiction and its Readership in Seventeenth-Century England (Cambridge: Cambridge University Press, 1981), pp. 48–50.
3 It is a little difficult to cite evidence on precisely this point (due to the absence of items of such small value from wills and inventories), but ownership of ballads and chapbooks among the common people of early modern England is nonetheless quite well documented. See Spufford, Small Books and Pleasant Histories, pp. 45–82. For a later period, see David Vincent, Literacy and Popular Culture: England 1750–1914 (Cambridge: Cambridge University Press, 1989), pp. 61–62.
4 Hyder E. Rollins, ‘The Black-Letter Broadside Ballad’, PMLA, 34 (1919), 258–339 (pp. 336–38); Tessa Watt, Cheap Print and Popular Piety, 1550–1640 (Cambridge: Cambridge University Press, 1991), pp. 148–49, 194.
5 [Izaak Walton], The Compleat Angler; or, The Contemplative Man’s Recreation (London: printed by T. Maxey, for Rich. Marriot, 1653), p. 49 [ESTC R202374].
6 Angus Ross, ed., Selections from The Tatler and The Spectator of Steele and Addison (Harmondsworth: Penguin, 1982), p, 362.
7 Thomas Holcroft, Memoirs of the Late Thomas Holcroft, 3 vols (London: Longman, Hurst, Rees, Orme, and Brown, 1816), i, 135–36.
8 With respect to the ballad, the ‘middlemen of literature’ (in Robert Darnton’s phrase), and in particular the chapman distribution network, would benefit from much further research. See, for example, Margaret Spufford, ‘The Pedlar, the Historian and the Folklorist: Seventeenth Century Communications’, Folklore, 105 (1994), 13–24; St Clair, Reading Nation, pp. 37–39; Andrew Taylor, The Songs and Travels of a Tudor Minstrel: Richard Sheale of Tamworth (York: York Medieval Press, 2012), pp. 1–2, 19–28.
9 See, for example, Thomas R. Adams and Nicolas Barker, ‘A New Model for the Study of the Book’, in A Potencie of Life: Books in Society, ed. Nicolas Barker (London: British Library; New Castle, DE: Oak Knoll Press, 1993), pp. 5–43; Robert Darnton, ‘What Is the History of Books?’, Daedalus, 111.3 (1982), 65–83; D. F. McKenzie, Bibliography and the Sociology of Texts (Cambridge: Cambridge University Press, 1999); D. F. McKenzie, Making Meaning: ‘Printers of the Mind’ and Other Essays, ed. Peter D. McDonald and Michael F. Suarez (Amherst and Boston: University of Massachusetts Press, 2002). The economic approach developed in St Clair, Reading Nation, also belongs here.
10 Jerome J. McGann, The Textual Condition (Princeton: Princeton University Press, 1991), p. 56. See also Peter L. Shillingsburg, From Gutenberg to Google: Electronic Representations of Literary Texts (Cambridge: Cambridge University Press, 2006), pp. 16–18.
11 This definition is pared down and adapted from Peter L. Shillingsburg, Scholarly Editing in the Computer Age: Theory and Practice, 3rd edn (Ann Arbor: University of Michigan Press, 1996), p. 46: ‘A text is the actual order of words and punctuation as contained in any one physical form, such as manuscript, proof, or book.’
12 Shillingsburg, Scholarly Editing, p. 47, likewise allows that the same text can be stored in more than one place, using more than one set of signs (such as written words, impulses on magnetic tape, or digital 1s and 0s).
13 Peter L. Shillingsburg, Resisting Texts: Authority and Submission in Constructions of Meaning (Ann Arbor: University of Michigan Press, 1997), pp. 73–74, uses the term ‘material text’ to refer to the union of text with document, reaffirming the necessity of the physical medium. For the present purpose, it is desirable to maintain at least a conceptual separation of text from document.
14 Towards the end of his life, Guglielmo Marconi is said to have become convinced that sounds once generated never die, but simply become fainter and fainter until they can no longer be detected.
15 This was the early practice of the American collector James Madison Carpenter, who began collecting and recording in Britain in 1928 (Washington, DC, Library of Congress, American Folklife Center, Archive of Folk Culture, AFC1972/001, James Madison Carpenter Collection, MS p. 09637).
16 Shillingsburg, Scholarly Editing, p. 47, uses the term ‘storage medium’, which effectively combines textual carrier and substrate.
17 Erika Brady, A Spiral Way: How the Phonograph Changed Ethnography (Jackson: University Press of Mississippi, 1999), p. 62; D. K. Wilgus, Anglo-American Folksong Scholarship since 1898 (New Brunswick, NJ: Rutgers University Press, 1959), p. 232.
18 Percy Grainger, ‘Collecting with the Phonograph’, Journal of the Folk-Song Society, 3.3 (no. 12) (1908), 147–242 (p. 148).
19 Jacques Derrida, Of Grammatology, corrected edn, trans. Gayatri Chakravorty Spivak (Baltimore: Johns Hopkins University Press, 1997).
20 Jeff Todd Titon, ‘Text’, in Eight Words for the Study of Expressive Culture, ed. Burt Feintuch (Urbana and Chicago: University of Illinois Press, 2003), pp. 69–98.
21 Marshall McLuhan, Understanding Media: The Extensions of Man (London: Routledge & Kegan Paul, 1964).
22 Harold A. Innis, The Bias of Communication ([Toronto]: University of Toronto Press, 1951).
23 Daniel Chandler, ‘Biases of the Ear and Eye: “Great Divide” Theories, Phonocentrism, Graphocentrism & Logocentrism’ (1994), available at http://www.aber.ac.uk/media/Documents/litoral/litoral.html.
24 St Clair, Reading Nation, pp. 26–28. Even now, there is no clear answer to the question of what proportion of the ballads printed in any particular historical period have survived or, conversely, been lost.
25 Kevin Bradley, Risks Associated with the Use of Recordable CDs and DVDs as Reliable Storage Media in Archival Collections – Strategies and Alternatives, Memory of the World Programme, Sub-Committee on Technology (Paris: UNESCO, 2006), available at http://unesdoc.unesco.org/images/0014/001477/147782E.pdf.
26 Jonathan Sterne, The Audible Past: Cultural Origins of Sound Reproduction (Durham, NC: Duke University Press, 2003), pp. 325–27.
27 Watt, Cheap Print and Popular Piety, pp. 11, 52; St Clair, Reading Nation, p. 340.
28 Elizabeth L. Eisenstein, The Printing Revolution in Early Modern Europe, 2nd edn (Cambridge: Cambridge University Press, 2005), pp. 87–88.
29 Watt, Cheap Print and Popular Piety, pp. 11–12 (and see pp. 260–64); St Clair, Reading Nation, pp. 40, 343–44.
30 St Clair, Reading Nation, pp. 28–29. Some of the variety in format among ballads and chapbooks over time might be related to pricing strategies.
31 Phonographic recording machines were eventually manufactured in substantial numbers for the office dictation market (where ‘dictaphone’ was used as a generic term), and it was such a machine that J. M. Carpenter used for his collecting in Britain in the late 1920s/1930s.
32 C. J. Bearman, ‘Percy Grainger, the Phonograph, and the Folk Song Society’, Music & Letters, 84 (2003), 434–55; C. J. Bearman, ‘The Folk-Song Society and the Phonograph’, Folk Music Journal, 10.3 (2013), 370–74. For an interesting account of phonographic recording by collectors in Slovenia before the First World War, see Drago Kunej, ‘“We have plenty of words written down; we need melodies!”: The Purchase of the First Recording Device for Ethnomusicological Research in Slovenia’, Traditiones [Ljubljana], 34.1 (2005), 125–40.
33 London, EFDSS Archives, Cecil Sharp Collection, Correspondence, Cecil Sharp to Percy Grainger, 23 May 1908 (photocopy); Anne Gilchrist Collection, AGG/8/142, Anne Gilchrist to Lucy Broadwood, 1 and 2 June 1908.
34 Bearman, ‘Percy Grainger, the Phonograph, and the Folk Song Society’, pp. 441–42. See further Drago Kunej, ‘Digitised Early Sound Recordings as Scholarly Resources’, in Trapped in Folklore? Studies in Music and Dance Tradition and their Contemporary Transformations, ed. Drago Kunej and Urša Šivic (Wien: Lit Verlag, 2013), pp. 181–96.
35 Grainger, ‘Collecting with the Phonograph’, pp. 148–49.
36 On the pursuit of fidelity in sound recording at large, see Sterne, Audible Past, pp. 215–86.
37 Some of these can be heard on Unto Brigg Fair: Joseph Taylor and Other Traditional Lincolnshire Singers Recorded in 1908 by Percy Grainger, 12-inch LP (London: Leader LEA 4050, 1972).
38 See Grainger, ‘Collecting with the Phonograph’.
39 Rollins, ‘Black-Letter Broadside Ballad’, p. 263, observes (with perhaps a degree of licence): ‘the [earlier, black-letter] broadsides abound in typographical errors hardly less grotesque and absurd than the amazing woodcuts that embellished the sheets’.
40 See Anthony Bennett, ‘Sources of Popular Song in Early Nineteenth-Century Britain: Problems and Methods of Research’, Popular Music, 2 (1982), 69–89 (pp. 77–78).
41 The fact that the digital signal is only ever an approximation to its analogue original, and the subsequent reconversion to analogue form involves a further approximation, may not impact discernibly upon the issue of textual precision, but should presumably trouble anyone overly concerned with ‘authenticity’ in the abstract.
42 See Charles C. Mish, ‘Black Letter as a Social Discriminant in the Seventeenth Century’, PMLA, 68 (1953), 627–30; St Clair, Reading Nation, p. 341.
43 Barry Reay, ‘The Context and Meaning of Popular Literacy: Some Evidence from Nineteenth-Century Rural England’, Past and Present, no. 131 (1991), 89–129 (p. 116).
44 James Hogg, Memoir of the Author’s Life and Familiar Anecdotes of Sir Walter Scott, ed. Douglas S. Mack (Edinburgh: Scottish Academic Press, 1972), p. 137.
45 Hogg, Memoir and Familiar Anecdotes, p. 62.
46 See further Valentina Bold, ‘“Nouther right spelled nor right setten down”: Scott, Child and the Hogg Family Ballads’, in The Ballad in Scottish History, ed. Edward J. Cowan (East Linton: Tuckwell Press, 2000), pp. 116–41.
47 Leo Treitler, ‘History and the Ontology of the Musical Work’, in With Voice and Pen: Coming to Know Medieval Song and How It Was Made (Oxford: Oxford University Press, 2003), pp. 298–316 (p. 312).
48 Shillingsburg, Scholarly Editing, pp. 42–44.
49 Treitler, ‘History and Ontology’, p. 301. It is only right to record that Treitler adds, in parentheses, ‘Everything depends, of course, on what lies beneath that word, “essentially”’ (p. 301), since every performance is necessarily in some way unique.
50 Lydia Goehr, The Imaginary Museum of Musical Works: An Essay in the Philosophy of Music, rev. edn (New York: Oxford University Press, 2007).
51 Treitler, ‘History and Ontology’, p. 316.
52 A proper account of broadside history has yet to be written. For sound recording, Sterne, Audible Past, is excellent.
53 Yet broadsides were still in commercial circulation around the end of the nineteenth century in England, and somewhat later in Scotland, and it is quite possible that their presence was deliberately played down. This may well be one of the more accurate insights into the revival provided by the revisionist thesis. See Dave Harker, Fakesong: The Manufacture of British ‘Folksong’, 1700 to the Present Day (Milton Keynes: Open University Press, 1985), esp. pp. 141, 182, 193–94.
54 And of a digital download than of an LP or CD – encouraging, perhaps, an illusory sense in some quarters of a late return to a pre-Gutenberg heyday.