The place of accompaniment in folk studies

The status of accompaniment within folk music’s cultural practice can be problematic for performers and scholars. It is widely believed that Britain’s folk tradition was fundamentally a melodic one, and for many within the post-war folk revival the solo, unaccompanied voice of a ‘source’ singer like Harry Cox was the epitome of authentic traditional performance. In The Singing Island (1960) Peggy Seeger and MacColl suggested practical guidelines for accompanists which reflected the notion that folk music was essentially unaccompanied. Despite this, (as the publishers of MacColl and Seegers book recognised with their inclusion of piano accompaniments) most people are used to hearing folk songs and tunes with some sort of harmonic and rhythmic setting.

The creative re-harmonization of folk melody isn’t a recent development: tunes from what the American folklorist Samuel Bayard called the ‘common melodic repertory’ have been reworked and reharmonized by musicians for centuries (Bayard 1950: 145). Before Cecil Sharp and his contemporaries in the English musical renaissance began to grapple with the curious, ‘modal’ strains of English folk melody, Scottish traditional airs appeared in settings by Geminiani, Oswald and even Beethoven. As Matthew Gelbart (2007) notes, mid-18th century collections might include figured bass arrangements that reflected a contemporary tradition of semi-improvised accompaniment.

The advent of commercial recording and the folk revivals of the 20th century generated new approaches to song and tune accompaniment that brought British folk melodies into contact with other styles such as jazz, blues, and rock, and onto the fringes of the commercial mainstream. The work of musicians such as Archie Fisher, Shirley and Dolly Collins and Planxty in the 1960s, and ‘electric folk’ bands Steeleye Span and Fairport Convention kickstarted a wave of experimentation with harmony, timbre, texture and rhythm that continues in the work of contemporary performers such as Lau. All this has meant that in much of the Anglophone world, the guitar as much as the fiddle is the quintessential folk instrument. So why have academics written so little about accompaniment?

Reasons can perhaps be found in the beginnings of folk music research in the 18th century. Rather than studying traditions as part of contemporary musical practice, early scholarship was often concerned with sifting out the pure elements of traditional melody from contemporary performance (textual or musical), while contemporary social-historical theories of musical development equated melodic simplicity with oldness and authenticity. Gelbart shows how the conceptual separation of folk and art music in the 19th century established folk as the realm of timeless communal melody; vast collections of ‘national song’ presenting swathes of disembodied tunes reflected a definition of tradition that privileged text over performance, a position which 20th century developments in ethnomusicology have yet to completely supersede. This intellectual legacy has meant that accompaniment is still often understood as a dimension of re-interpretation rather than as part of ‘the tradition’ proper.

As Gelbart shows, European nationalism and the politics of class have also played a substantial role. Herderian notions of a national, peasant-derived song culture left little room for the refined forms of pastoral found in Edinburgh drawing rooms or the tune-settings of foreign-born virtuosi, making them seem at best inauthentic and at worst forms of corruption and appropriation. 20th century Marxian class-based analyses which identify folk music with ‘working-class culture’ have done little to correct these assumptions. Despite the work of researchers such as David McGuinness and contemporary ensembles such as Concerto Caledonia in recovering pre-20th century harmonic traditions, the conceptual separation between a pure, melodic (vernacular) tradition and secondary (commercial/bourgeois) practices of re-presentation remains in place, colouring our understanding of folk’s performance history.

There is much more to be learned about how musical approaches to accompaniment have developed over the last century as well as how scholarly and popular attitudes to accompaniment have shaped (and been shaped by) perceptions of folk and traditional music’s historical and cultural significance. There is a vast amount of evidence to be sifted through in the form of commercial and field recordings, in addition to which many key 20th century practitioners are still living. There is thus scope for a fusion of ethnographic, textual and historical approaches which would not only recover a great deal of sophisticated musical practice but potentially produce an alternative history of Anglophone folk music which moves beyond the tacit intellectual assumptions of the 19th century to recover the creative work of musicians whose work has so far received little in the way of detailed analysis.

Works Cited 

Bayard, Samuel (1995). ‘Principal Melodic Families of British-American Folksong’ [1950], in Dugaw, Daphne (ed.) The Anglo-American Ballad: A Folklore Case-Book. New York: Garland.

Gelbart, Matthew. (2007) The invention of ‘folk music’ and ‘art music’: emerging categories from Ossian to Wagner. Cambridge University Press.

MacColl, Ewan and Seeger, Peggy. (1960) The Singing Island. London: Mills Music.

Broadsides and traditional ballads

From a fairly early period, folk scholars have tended to distinguish between broadside ballads and traditional ballads, although the precise nature, degree and implications of this distinction have varied across time. The categorical distinction between the ballad of tradition and the printed broadside is complicated by the fact that a high percentage of traditional ballads have circulated in commercial printed form at some point; in fact, broadsides are often the earliest existing version of ballads placed in the traditional category. The distinction is also closely intertwined with a series of paired concepts including print and orality, art and commerce, popular and traditional, old and new, rural and urban, and (often tacit) assumptions about the status and origins of the songs and their relationship with the different cultures that produced them.

The term ‘broadside ballad’ refers to a form of popular literature emerging in the early modern period – songs printed on a single sheet of paper to be sold cheaply in the street. Broadsides dealt with a range of themes including, love and courtship, religion, politics, history, war and crime. They reached the peak of their popularity between 1550 and 1700, though they continued to be produced and consumed in relatively large numbers up until the end of the 19th century and were still being printed until the mid-20th. Broadside ballads were works of known individual or collective authorship and often created expressly for the print market, frequently in response to current events, making it sometimes possible to determine their composition date with some degree of certainty from internal evidence.

The term ‘traditional ballads’ refers to material of uncertain authorship and date that has been passed on through oral transmission. These are songs thought to predate their first printed version and even print culture itself. For early ballad scholars, like Percy, they were thought to be the products of a much earlier culture of minstrelsy, and thus to have a far greater aesthetic and antiquarian value than the typical products of the commercial ballad trade of his own time. Nevertheless, collectors including Percy used the collections of broadsides made by gentlefolk in the seventeenth century (such as Samuel Pepys), as well as the stocks of ballad printers such as Cluer Dicey, as sources for their published collections of ‘ancient’ songs. They were often scathing about these materials of commercial provenance, however; Motherwell referred to broadsides as ‘a great quantity of sad trash’ (Fumerton & Guerrini 2010, 36), while the great American collector Francis James Child famously described the Pepys and Roxburghe broadside collections as ‘veritable dunghills in which, only after a great deal of sickening grubbing, one finds a very moderate jewel’ (Palmer 1996, 158).

So, what were Child and his predecessors looking for, and how did they distinguish the jewels from the trash? Percy hoped to recapture the work of a historical creative elite – the minstrels of the feudal past. Child was looking for the last remnants of a tradition of ‘popular’ song, using the term in a slightly unfamiliar sense: for Child, these songs were the product of a radically different kind of society from the one that had given birth to the ballad culture of the broadside era. Child saw the popular ballads as the natural form of self-expression of a culture at a specific stage in its development:

The condition of society in which a truly national or popular poetry appears … is a condition 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. (Child 1994 [1900], 214)

For Child, the ballads were popular not in the sense of belonging to the lower echelons of society but because they were the product of a period before such distinctions existed. The transition between this older, less stratified society and the world of the broadside ballad was characterised by the decay of traditional ways and their replacement by a highly stratified, commercial society. As part of this process, an older, oral, culture unified by tradition was supplanted by a literate culture characterised by social and political division. The broadsides were part of a commercial culture that drew on but inevitably distorted traditional materials. The goal was to separate the authentic poetic tradition from later commercial derivatives, and scholarly judgement was necessary to reveal the authentically ‘popular’ strand hidden within the mass of highly adulterated material.

Orality and literacy have often been understood as a zero-sum game: as literacy and print culture spread, orality and the domain of tradition have been thought to inevitably recede. According to this way of thinking, the two concepts are not just opposites but distinct developmental stages. The assumption that print culture-followed-manuscript culture-followed oral culture shaped the epistemological stance and the editorial practices of collectors; because scribal culture was felt to represent a transitional stage between orality and print, written sources were closer than print sources to a tradition located firmly in the historical (pre-Gutenberg) past. Manuscript sources were also favoured over oral versions by many collectors. As Dave Harker and others have noted, there was an implicit class dimension underlying this preference for writing over print; MS sources were often acquired from aristocratic informants, and therefore deemed inherently more trustworthy than either print (commercial) or oral (peasant) sources.

Scholars such as Motherwell, Percy and Child thought they could separate out the traditional wheat from the commercial chaff though certain internal markers that indicated the origin of songs in a pre-literate tradition. Motherwell identified the ‘commonplaces’ which helped the reciter unaided by print to find their path through the story as evidence of oral origin. These theories have developed in the work of subsequent scholars into a stylistics of the ballad. Traditional ballads, according to Gerould tell a story by ‘letting the action unfold itself in event and speech, and [tell] it objectively with little comment or intrusion of personal bias’ (quoted in Atkinson 2002, 13). Atkinson (2002, 12), identifies some other key markers including ‘repetitive textual, metrical, and melodic structures’; ‘conventional vocabulary and epithets’ (‘lily-white hand’; ‘milk-white steed’ (ibid.). He also suggests that traditional ballads, rather than being self-contained works, refer to other ballads within the tradition, something he calls traditional referentiality. This intertextual quality presupposes a certain competency on the part of the listener; meaning is not contained in the text but emerges in the relationship between the singer and their audience as joint possessors of a shared cultural heritage. Additionally, traditional ballad culture incorporates a body of common song tunes which have an unstable relationship with texts as well as traditional stylistic traits, a ‘recurrent vocabulary of melodic, rhythmic, and dynamic technique’ (ibid.).

According to a narrative in which industrial print culture gradually replaces an older popular tradition, it makes sense to attempt to strip away the corrupted material to get back to the purer, oral versions that preceded it. However, contemporary scholarship has begun to question the notion of orality and literacy as opposites and recognise that song culture has from a very early period rested on a combination of oral and literate practices; rather than one replacing the other, ‘oral, scribal and printed media fed in and out of each other as part of a dynamic process of reciprocal interaction and mutual infusion’ (Fox quoted in Pettitt 2009, 429). Rather than to attempt to reconstruct an authentic oral tradition, recent scholarship has sought to trace the relationship between these different modes of transmission. While broadsides have often been used as sources for this type of reconstructive research, songs originally written for the broadside market have also passed into oral tradition: Pettitt (2009; 2013) has investigated the ways in which the oral transmission of this material transforms the content and stylistic features of ballad narratives, looking at the example of the early 19th century song ‘The murder of Maria Marten’, which first appeared in print in 1828 and was subsequently recorded from oral sources in the late 20th century.

With their combination of entertainment, moralising comment and reportage, broadsides anticipated the tabloid newspaper. The ballad which told the story of Maria Marten’s death at the hands of her lover William Corder in 1827 existed alongside prose accounts of the murder and included a certain amount of quasi-journalistic detail. Comparing an early broadside version with contemporary prose accounts and later oral versions, Pettitt found that the ‘traditional’ (or ‘memoral’) versions had a reduced text which removing the incipit (the conventional opening which hailed the listener and introduced the main characters), and much of the detail and sentimentality to focus on the ‘emotional core’ (Coffin 1961, in Pettitt 2009), the relationship of the Maria with her mother. This focus on the tragedy of the victim shifts the centre of gravity away from the moral lesson supplied by the murderer’s fate which was a common feature of the Victorian ‘good night’ ballad.

Pettitt also noted the removal of interlinking explanation creating a succession of separate but interlinked images, and a preference for telling the story through direct speech rather than third person reportage. Lines from the ballad were also reformulated to create incremental repetition (e.g., ‘the very first stroke as they struck’) and a parallelism or ‘balancing’ of events (e.g., the words used in the mother’s ‘prophecy’ are echoed in the description of the father’s action when discovering the body); departures (William takes Maria Marten to the barn at the beginning while the father takes her away (dead) at the end – an inversion of the normal marriage rite; the opening conversation between mother and daughter is echoed in that between mother and father. The result is a depersonalisation of protagonists: they become less real, historical individuals than social archetypes; and an emphasis on relationships in which the mother-daughter dynamic takes centre stage.

Given that ‘Maria Marten’ circulated first as a broadside and the evidence which suggests that this version is the origin of the later recorded versions, Pettitt’s work indicates that the ‘oral’ features identified by earlier collectors are indeed a function of oral song culture. He attributes the changes to the processes of memory and oral performance, which, he argues, (using a structural metaphor which echoes 18th and early 19th century collectors) ‘can be compared to the weathering of rock or the wearing of cloth to reveal the underlying structures and their patternings that hold the whole together’, or ‘the decay of a corpse in which the softer tissues disappear and reveal the skeletal structure beneath’.

Such research is of value in beginning to understand the changes that songs undergo as they circulate between print and oral transmission. Recent scholarship has begun to uncover the complex interplay of oral and literate practices at work within popular song cultures. Increasingly, the relationship between the broadside and the traditional ballad is seen less as one of corruption of oral ‘texts’ by commercial forces than as a ‘dynamic, mutual reinforcement’ in which the ‘oral and the written or printed word – function continuously in support of one another’ (Atkinson 2002, 25). Instead of a search for hard definitions, perhaps it is necessary to acknowledge the multiplicity of the ballad, simultaneously a part of literary and popular cultures, an early media phenomenon and a set of principles of oral creation which link the present with the distant cultural past.

Works cited

Atkinson, David. (2002) The English Traditional Ballad: Theory Method and Practice. Aldershot: Ashgate.

Child, Francis J. (1994 [1900]) ‘“Ballad Poetry”, Johnson’s Universal Cyclopaedia, 1900’, Journal of Folklore Research, 31(1/3), 214-222.

Fumerton, Patricia and Guerrini, Anita. (2010). Ballads and Broadsides in Britain, 1500–1800. Farnham: Ashgate.

Roy Palmer. (1996) ‘“’Veritable Dunghills”: Professor Child and the Broadside’, Folk Music Journal, 7/2, 155-66.

Pettitt, Tom. (2009) ‘Written Composition and (Mem)oral Decomposition: The Case of “The Suffolk Tragedy”’, Oral Tradition, 24/2, 429-454.

Tradition and metaphor

What kinds of metaphors have scholars used when writing about traditional music, and what can we learn from them? This post explores some ways that metaphors have been used to structure historical understandings of the concept of tradition in the light of some recent ideas from cognitive linguistics, suggesting that the language we use when writing or talking about tradition is more significant and revealing than it might first appear.

A glance at the historical literature and popular discourse around folk and traditional music shows many kinds of metaphors for tradition. Some examples include:

  • Tradition is an old building (William Motherwell; Walter Scott)
  • Songs are pebbles in a stream (A.L. Lloyd)
  • Groups of tunes are families (Samuel Bayard)
  • Traditional performance is a journey (Motherwell)
  • Songs are ‘beautiful peasants’ (Joseph Ritson)
  • Songs are ‘antique’ statues (Thomas Percy)
  • Transmission is reverse alchemy (Ritson; Scott)
  • A song collection is a museum (Robert Burns and James Johnson)
  • Tradition is a treasure chest; well-spring (Kist o’Riches; Tobar an Dualchais website)

Traditionally, metaphor is understood as a function of poetic language – a piece of decorative illustration, that shows the poet’s skill in describing the world using beautiful or revealing phrases. Cognitive linguists, however, suggest that metaphors don’t just help us describe reality as it exists; they are necessary for us to conceptualize it in the first place, and as such determine our experiences of reality and what we can say and think about it. When we use a metaphor such as tradition is a building we are using a concrete (no pun intended) metaphor to conceptualise something abstract. This is an example of conceptual metaphor, rather than linguistic metaphor. In the latter, we use the word for one thing to describe another (‘my dentist is a butcher’). Conceptual metaphor is where we use one concept to understand another concept – usually a more abstract one. In this case, conceptual metaphor allows us to produce inferential structure about ‘tradition’ (an abstract domain) by borrowing it from the concrete domain of buildings and, specifically, the ways they are affected by the passage of time.

However, the metaphors we use don’t just combine two separate domains – they combine two levels of experience – the basic (sensorimotor) level and the abstract level at which most theoretical discourse takes place. This is a thoroughly everyday process; we habitually use our understanding of objects, actions and processes we can see and hear to understand things we can’t. We can’t, for example literally grasp an idea; but we can grasp an object (at least one and light enough to hold in our hands and carry around with us). We therefore understand ideas as objects. The word ‘comprehend’ in English has as its root ‘prehendere’, the Latin word meaning ‘to grasp’. This verb suggests an understanding of facts or concepts as a special kind of object, namely, a tool. To grasp an object entails taking possession; to grasp a tool means to prepare ourselves confidently to explore its possible uses. Thus, our concept of understanding uses a metaphor from the physical to move up into the abstract domain; in doing so, it brings with it certain metaphorical implicatures. To understand the ‘object’ of discourse is to be able to ‘wield’ it, to put it to work in constructive ways, which can exert a shaping force on reality.

When we talk about abstract theoretical concepts, such as ‘class’; ‘nationhood’, ‘ethnicity’, or tradition, we derive our inferential structures from the domain of objects. We often talk about the nation as if it was a literal container. It isn’t – this is a metaphor; however, the notion of a container allows us to make use of inferential structures based on the imagined qualities of various containers. Some of these are more salient than others, depending on the discourse. Is it a large, relatively empty container, or a small, relatively full one? Is it transparent or opaque? Are its boundaries, rigid, or porous? These kinds of speculations and inferences are at work all the time in abstract sociological and political discourse, and have obvious implications for everyday politics. Getting people to imagine the country as a small, crammed container with leaky sides serves the interests of those who favour tighter laws on immigration; their opponents can argue the opposite, but may fare better by changing the metaphor; for example, by evoking the image of the nation as a well-to-do and nurturing family, capable of supporting all its children.

The most important implication of this approach is that language is not a transparent medium. It doesn’t just describe the world as it exists and thus allow us to speak truthfully about it. Instead, it shapes the inferential logic of discourse. The metaphors used by students of traditional music thus don’t just illustrate a prior reality, they allow us to hold structures of understanding in our minds and, as such, shape the internal logic of how that reality is discussed. They determine what can be said truthfully and logically about things like traditional music and its transmission. Once we have accepted one basic conceptual structure, we are obliged to honour its entailments; otherwise our arguments will appear inconsistent.

According to this way of thinking, we don’t just use metaphors occasionally, as illustrations of factual reality. We construct factual reality and rules and inferential structures using metaphor. The metaphors we use effectively determine whether our statements are true (because logically sound) or not. If reality doesn’t match up with our conceptual models, we must change the metaphor. This is the basis process whereby we refine our knowledge of the world, squaring our theories of reality with the experiences we have of it. In science, the metaphorical system shared by researchers is referred to as a paradigm. A change in this system is called a paradigm shift. A paradigm shift entails changes in the ways we write and act in research: ideas also affect discourse and practice. In the humanities, it is less common for researchers in a field to share a single research paradigm, and we therefore talk about various theoretical approaches which are often in conflict with one another. These are competing paradigms, conflicting metaphorical systems.

Another very important implication of recent research is that these metaphors are not only found in text and speech but also in images, sound and music. This supports the idea that they are not confined to the verbal mode but also structure visual and sonic modes (more on this later).

An important consideration is that writers don’t just use one metaphor to talk about traditional culture; they use all sorts of different metaphors. In the ballad scholarship of the eighteenth century, it is common to find traditional material described using words such as ‘fragments’, ‘monuments’, and ‘relics’. These are not just isolated metaphors, but seem to structure the whole conceptual model used to understand both the nature of the materials, their origins and the role of collectors and antiquaries. William Motherwell (1827) uses the metaphor of a cathedral to describe the operation of oral tradition:

It is granted at once that the ‘expressions and illusions’ of these compositions fluctuate, and that frequently, but these changes never alter entirely the venerable aspect of the whole ballad. It is like repairing gradually the weather worn face of an ancient cathedral, by the insertion here and there of a freshly hewn stone, as need may require. (Motherwell 1827, xi)

This metaphor has certain implications and entailments. Firstly, it implies age, strength and size. Secondly, it has entailments based on the processual metaphor Motherwell adopts for describing transmission which is that of gradual and conservative repair to the structure of a large (sacred) building. It also has entailments for how we understand the other actors involved in the system; those involved in the tradition are portrayed as ‘caretakers’, rather than architects; they are concerned with maintenance, not creativity. It also suggests that tradition is a stable structure; although its physical contents change over time, its shape persists. This notion of tradition as immanent structure, rather than content, is a basic and highly significant entailment of Motherwell’s metaphor with implications for collection and editing practices. The metaphor is not just an illustration of a prior reality – it determines the epistemological terms of enquiry and the attendant practices of Motherwell’s basic approach.

Although scholars use different metaphors to describe tradition, we should expect to find some level of cohesion between them. That is, they might have different specific content, but share certain basic structures and metaphorical entailments. Thomas Percy, for example, writing several decades before Motherwell, used a similar metaphor to describe the tradition. Like Motherwell, he sees the tradition in terms of stone objects persisting over time:

[M]ost of them are fragments too mutilated and imperfect to afford much pleasure to a reader in their present state; and . . . most of them contain charming hints, which might give occasion to very beautiful songs, if supplied and filled up, in the manner that old broken fragments of antique statues have been repaired and completed by modern masters. I think I could fill up the breaches of some of them myself. (Thomas Percy to David Herd, quoted in Harker 1985, 33)

Like Motherwell, Percy sees the songs as ancient stone objects that have been worn down (‘mutilated’) by time and tradition. The transmission process is natural, but basically degenerative. The important entailment shared by Percy and Motherwell is the assumption of ancient authorship; the cathedral had an architect, the statues a sculptor. Change takes the form of decay and weathering. Percy takes the analogy further to suggest an equivalency between editorial reconstruction and the restoration of ancient monuments by ‘modern masters’. The logic of this approach derives from inferential structure provided by his basic metaphor. Again, it has implications for the role of actors involved; are they passive or active; creative or destructive? What remains to be done? What actions should we take relative to the materials to hand?

Where a wholly new metaphor is introduced to discourse, we see a paradigm shift. That is, the change in discourse indicates an underlying change in common ways of thinking about tradition and attendant research practices. This is because the metaphors clash – their basic structures and logical entailments cannot be reconciled with one another. Writing almost two centuries after Percy, A.L. Lloyd (1952) used the notion of a stream to describe the transmission process. In his metaphor, the objects of tradition taken separately, were imagined as stones in a river, continually worn smooth by the water. The basic structure of this metaphor is this:

  • Tradition is a stream
  • Songs = pebbles
  • Water = time
  • Traditional change = the smoothing away of rough edges over time

Notice how this metaphor shares Percy’s and Motherwell’s notion of songs as objects and that of continuity over time. Like Motherwell, he constructs the traditional process as a change in the form and consistency of stone objects, both evoke natural process metaphors (‘weather-worn’; the ‘river of time’). The transmission process is one of shaping by taking away material gradually from an object. In both, the traditional object retains its integrity. But these metaphors have further entailments which diverge from one another:

  • In the role of individuals in the transmission process
  • In the nature of the change undergone
  • In the basic function of tradition in determining the identity of tradition and its objects
  • In the origins and role of composers

Percy and Motherwell both suggest ancient authors whose work has been broken down by time. Unlike Percy, who sees tradition as essentially a ‘wearing down’, Motherwell suggests that tradition (the preservation of traditional materials in oral practice) plays a role in preservation, albeit a passive one; for Lloyd, however, this position is totally reversed. For him, tradition is a process of completion. Tradition itself (though still seen as a natural process) takes the place of the ancient architect or sculptor; it is a creative process which determines the identity and character of traditional songs. Lloyd, however, stops short of attributing authorial status to individual participants in tradition; what makes a song traditional is its involvement in a quasi-natural process – the smoothing action of tradition on musical materials. As such, although these authors are talking about the same subject, Lloyd’s metaphor could be seen as indicating a paradigm shift in scholarly practice – a change in the basic metaphorical mapping used to conceptualise tradition.

As mentioned above, conceptual metaphors are realised multimodally; they can be conveyed using various modes of communication, not just language. The introduction to Scott’s Minstrelsy of the Scottish Border (1802) for example, is prefaced with this image:

Scott Frontispiece

The image introduces his study and can be read as standing for the textual ‘monuments’ (Scott uses this term quite a bit) contained in the book. Instead of expressing the ‘ruined building’ metaphor verbally like Motherwell, he invokes it visually. As will be apparent by now, one possible explanation for the existence of conceptual metaphors cross modally is that the cognitive linguists are correct in arguing that the metaphors used in language and imagery both reflect the same basic conceptual models that are cohesive and allow us to think about abstract and complex subjects such as traditional culture. Importantly, they are prior to language and other modes. This insight reverses the role of metaphor from being a local aspect of discourse – an illustration of a prior reality – to a conceptual basis for constructing a more complex cognitive reality.

This discussion has suggested that paying attention to the metaphors used by writers can reveal underlying conceptual models. When we see new metaphors emerging, new ways of thinking are also emerging. Most importantly, however, these metaphors don’t just illustrate a prior reality independent of thought and language; they provide a framework for thought, discourse and practice, determining the inferential structure (and thus the logical truth conditions) of discourses on abstract subjects. This has important implications for how we understand the role of metaphor in folk music historiography, and how we ourselves use language in our attempts to conceptualise processes such as oral transmission.