Blommaert (2006)- Ethnopoetics as functional reconstruction

Review article Ethnopoetics as functional reconstruction Dell Hymes’ narrative view of the world Jan Blommaert University of London and Ghent University Dell Hymes. Reading Takelma texts. Bloomington: Trickster Press 1998. ix + 76pp. (ISBN 0915305070) Dell Hymes. Now I know only so far. Essays in ethnopoetics. Lincoln: University of Nebraska Press 2003. Xi + 512pp. (ISBN 0803273355) Dell Hymes. In vain I tried to tell you: Essays in Native American ethnopoetics. 2nd edition. Lincoln: University of Nebraska Press 2004. viii + 403pp. (ISBN 0803273436) In vain I tried to tell you was published by the University of Pennsylvania Press in 1981; it quickly gained recognition as an important book.1 It also quickly gained recognition as a difficult book to read, complex in structure and argument and replete with long and extremely detailed analyses and re-analyses.2 Consequently it is doubtful whether it was widely read, and it went out of print some years ago. I considered this a tragic defeat for scholarship, for there are books that deserve to remain in print simply because they are good and important, not because they sell well. In light of this, the new edition of In vain I tried to tell you by the University of Nebraska Press should be warmly welcomed, and one hopes that this new edition will be treated with more courtesy by the readership than its predecessor. here are very good reasons to be hopeful, for whereas the first edition of In vain I tried to tell you (henceforth IV) was a rather lonely book on any shelf, the second edition can be read alongside two other major publications by Hymes on ethnopoetics: the small study Reading Takelma texts (1998, henceforth RT) and the rather more monumental Now I know only so far (2003, henceforth NK). Taken together, they now constitute a voluminous, complex and rich Functions of Language 13:2 (2006), 229–249. issn 0929–998X / e-issn 1569–9765 © John Benjamins Publishing Company 230 Jan Blommaert oeuvre demonstrating the linguistic and anthropological skill, the capacity for meticulous, scrupulous analysis of detail, and the unstinting theoretical and historical insight of Hymes. One could add Hymes’ Ethnography, linguistics, narrative inequality (Hymes 1996; henceforth EL) to the pile of must-reads, for there as well the theoretical argument was underpinned by copious, detailed and rich ethnopoetic analyses, and ethnopoetic analysis is predicated explicitly on concerns for justice and equality. I will refer to EL in what follows, because it can now be read as an introductory volume to the more ‘technical’ ethnopoetic publications IV, RT and NK. In addition to the expansion of Hymes’ own work, there is now a much more widespread appreciation of implicit form in language, of poetic patterning in narrative and of the indexical (i.e. implicit, oten iconic) organization of speech, and prominent scholars have published magnificent surveys and analyses (see e.g. Bauman & Briggs 1990; Moore 1993; Ochs and Capps 2001; Haviland 1996, 1997; Silverstein 1985, 1997, 2005). here is thus now an infinitely richer environment for reading Hymes’ ethnopoetic studies than there was at the time of the publication of IV. his does not mean, to be sure, that the reading is any easier than it was twenty years ago. Having ventured into ethnopoetic analysis on some occasions, I can testify to the fact that it is a demanding, tough kind of analysis requiring skill, patience and analytic insight in a variety of technical domains, from phonetics over grammar to discourse and narrative analysis, sociolinguistics and cultural anthropology. his complexity in analytical process converts in complexity in presentation, and this in turn demands concentrated and careful reading. Ethnopoetic studies are not exactly novels, and ethnopoetics itself is oten misunderstood and misrepresented. hus, unless the fundamental assumptions are well understood, works such as these may be perceived as overly detailed, technical, and dull. Ethnopoetics suffers from the same curse as phonetics: unless one understands its function, value and potential applicability, it is a very unattractive thing. In the case of ethnopoetics, to make things worse, the very term ‘poetics’ is a liability because it projects intertextualities in all directions, from Aristotle to Bob Dylan, and could easily wrong-foot the reader in search of ‘poetry’ or ‘rhyme’ and ‘meter’. Much of what I have to say in this paper will be aimed at a more precise understanding of what ‘poetic’ means in ‘ethnopoetic’.3 In what follows, I will introduce ethnopoetics in general terms, avoiding a technical exposé (for which, anyway, there is no substitute to reading Hymes’ work) but focusing on the main theoretical assumptions underlying it. Next, I will engage in a discussion of the way in which Hymes sees ethnopoetic Ethnopoetics as functional reconstruction 23 analysis as a tactic for restoring, reconstructing and repatriating the functions of narratives. Finally, I will turn to the critical and humanist dimensions of Hymes’ ethnopoetics, arguing for a political reading of his ethnopoetic work. An appendix will offer an illustrative sample of ethnopoetic analysis. But first, let us have a quick look at the books. Books and oeuvres hroughout this essay, the books will be treated as an oeuvre: not a complete one and even less a closed one, but a consistent scholarly effort resulting in different books. It is recommendable — because it is immensely rewarding — to read the three volumes in one effort, as an oeuvre and a serious introductionand-immersion into ethnopoetic theory and analysis. When such is impossible, it is still worth keeping in mind that the books are connected by common lines of argument, visions of what narrative is, and ideologies of research — elements which I will try to spell out in the sections below. Chronologically, IV is the precursor of the two other ones, and Hymes sets out, step by step, to define the challenges, purposes and possibilities of ethnopoetics. he pivot of the book is the essay ‘Breakthrough into Performance’ (Chapter 3) — a text of fundamental importance even decades ater its first circulation. In ‘Breakthrough’, Hymes defines the central theoretical preoccupations of ethnopoetics; he sketches the field in which ethnopoetics plays. It involves issues of competence, real versus potential ability, the development and ‘bringing about’ of genres, different kinds of performance, and the way in which linguistic form (e.g. code- and style-shiting) is mobilized in performance. Around that central essay, Hymes collects studies that describe the state of affairs in scholarship of Native American folklore and studies that re-analyze and retranslate previously published texts from the North Pacific coast of North America, in Clackamas Chinook, Wasco Chinook, Takelma, Kwakiutl and Haida. In IV, Hymes repeatedly emphasizes that there is very little work on Native American oral tradition going on, and that what there is oten suffers from serious methodological defects; Hymes repeatedly insists that there is some urgency here, as the materials, speakers and occasions for performance are disappearing fast. RT and NK both express this sense of urgency: in contrast to the more theoretical ambitions of IV, they both seem to have mainly documentary goals, to present a maximum of ethnopoetically analyzed texts. RT is a careful edition and analysis of a Takelma myth, ‘Coyote and Frog’, narrated by Frances 232 Jan Blommaert Johnson in 1906 and recorded, later published by Edward Sapir. As a standalone study of a single text, it is exemplary, and it can serve as a pocket-format summary of ethnopoetics. Hymes takes us all the way up from ‘discovery’ of the text, the identification of the problematic nature of its first edition, and the careful reconstruction of the story as a poetically organized narrative, oriented towards local and universal motifs and organizing principles. NK is far wider in scope, and it represents Hymes’ second attempt at summarizing his views on ethnopoetics and accomplishments in analysis. Like IV, it is again organized around central essays, two in this case: “Use all here is to Use” (Chapter 3) and “When is Oral Narrative Poetry?” (Chapter 5). Whereas in IV, Hymes focused strongly on issues of competence and performance, the focus in NK has shited towards the potentially universal patterns that Hymes starts identifying in several of the stories he analyzes. he ‘poetic’ — identified as a central function of language use in IV — now becomes a potential universal of human conduct. he range of languages he addresses in NK is, consequently, also wider; Hymes still works from within the Pacific Northwest, but he now also discusses at length the studies done by others on Native American languages and elsewhere, on European languages. He even concludes the book with a chapter on the work of the American poet Robinson Jeffers — a chapter which includes important reflections on what one understands by a poetic ‘line’. As said before, one should not expect easy reading when picking up these books from the library. Even for someone relatively at ease with Hymes’ style, lexicon and arguments, NK is a book that takes time to read. It is packed with data, and transcripts and story profiles fill many, many pages; Hymes also attempts to incorporate and address almost any issue that has arisen in the study of oral narrative — methodological and historical issues, issues of method emerging from discussions with other scholars, and the emergent work done on other communities and traditions. hus, it needs to be read in conjunction with IV, for the fundamental issues discussed in IV are presupposed in NK. Ethnopoetics Ethnopoetics, to Hymes, is part of a larger theoretical vision revolving around narrative and performance and ultimately embedded in a view of language in society. Before discussing ethnopoetics per se, we need to consider some of these larger aspects. Hymes’ efforts in ethnopoetics can be seen from one angle as deviating from his other work, which focused on the ethnography of situated, contextualized Ethnopoetics as functional reconstruction 233 speech events (Hymes himself flags this ‘deviation’ and amply motivates it, NK: 11). Yet, there is more that ties ethnopoetics into his other work than separates it. Hymes’ ethnopoetic work is one way of addressing the main issue in ethnography: to describe (and reconstruct) languages not in the sense of stable, closed and internally homogeneous units characterizing parts of mankind (a view Hymes strongly associates with Chomskyan linguistics), but as ordered complexes of genres, styles, registers and forms of use: languages as repertoires or sociolinguistic systems (not only linguistic systems), in short. And ethnopoetics is urgently needed, because many languages are not only endangered as linguistic systems, but also, and perhaps even more critically, as sociolinguistic systems — genres, styles, ways of speaking becoming obsolete or unpractised.4 Ethnopoetic analyses, as we shall see, attempt to unearth culturally embedded ways of speaking — materials and forms of using them, that belong to the sociolinguistic system of a group (or groups), and that have a particular place in a repertoire due to their specific, characteristic form-function relationships. Such form-function relationships, Hymes argues, are complex and display ‘second linguistic relativity’ — a relativity of functions rather than form (as in Whorf ’s ‘first’ relativity) (Hymes 1966), causing a need to investigate functions empirically, that is ethnographically.5 In that sense, ethnopoetics fits into the general theoretical ambitions of the ethnography of speaking. It also fits into Hymes’ more general concerns with language functions, notably with narrative and performance. Hymes starts from what he calls “a narrative view of the world” (EL: 112), in which narrative is “a universal function” of language. his function, however, is rarely recognized. Narrative seems to be subject to to all kinds of constraints and socioculturally framed restrictions on use: narrative is a way of using language which possesses limited legitimacy and acceptability (EL: 115). Furthermore, it is rarely seen as a vehicle for rational, ‘cognitive’ communication, and oten stereotyped as affective, emotional and interpersonal (remember Bernstein’s ‘restricted codes’, e.g. Bernstein 1971). In contrast to this widespread view (both lay and specialized), Hymes sees narrative as a central mode of language use, in which cognitive, emotional, affective, cultural, social and aesthetic aspects combine.6 hey combine in implicit form, however — and here Hymes’ approach to narrative starts to differ from that of many others (e.g. Labov), who focused on explicit form and explicit contents, and who saw narrative largely as a repository of explicitly voiced facts, images and concerns.7 Consequently (and this defines much of the tradition of folklore) stories could be asked for, elicited, and performance could be invited, while its results were seen as the tradition, folklore, even ‘culture’ of the performers. Hymes’ approach, as said, differs 234 Jan Blommaert fundamentally. To Hymes, the essence of narrative — what makes it poetic — is an implicit level of structure: the fact that stories are organized in lines, verses and stanzas, connected by a ‘grammar’ of narration (a set of formal features identifying and connecting parts of the story) and by implicit organizational patterns, pairs, triplets, quartets etc. his structure is only partly a matter of awareness: it is the ‘cultural’ dimension of narration; most speakers produce it without being aware of its functions and effects, and good narrators are those who can stage a performance organized through “the synchronization of incident and measure” (EL: 166). Consequently, narration involves the blending of at least two kinds of ‘competence’: the competence to organize experience, events, images in a ‘telling’ way, and the competence to do so in a sequentially organized complex of measured form (EL: 198). his is not a random thing: narratives are “organized in ways that make them formally poetry, and also a rhetoric of action; they embody an implicit schema for the organization of experience” (EL: 121). More precisely, “the relationships between verses (…) are grouped in an implicit cultural patterning of the form of action, a logic or rhetoric of experience, if you will, such that the form of language and the form of culture are one and the same at this point” (EL: 139). So implicitness — its recognition and interpretation — is central to Hymes’ concerns. It is by recognizing that a lot of what people produce in the way of meaning is implicit, that we can reflect more sensibly on the general problem of assessing behavioural repertoire, and [alert] students to the small portion of cultural behavior that people can be expected to report or describe, when asked, and the much smaller portion that an average person can be expected to manifest by doing on demand. (Some social research seems incredibly to assume that what there is to find out can be found out by asking). (IV: 84) In other words, it is through investigating implicit form that we get to a vastly wider, richer and complex domain of cultural-linguistic organization, one that has been overlooked by much of twentieth-century linguistics (the main topic of EL). his more complex domain is also a domain of more complex functions, the aesthetic (or presentational, in Hymes’ terms) functions being central to it. And for Hymes, narrative is the mode of language use in which such presentational functions coincide with denotational, cognitive, affective and interpersonal ones. his brings us to ethnopoetics as an analytical technique. Hymes sees ethnopoetics as a form of structural linguistics, more precisely of “practical Ethnopoetics as functional reconstruction 235 structuralism” — “the elementary task of discovering the relevant features and relationships of a language and its texts” (NK: 123).8 It is about describing what exists in language and texts, and when applied to texts, it is a form of philology. But even if “[t]his kind of linguistics is old, known as philology (…), [t]he kind of discoveries it makes are new” (RT: ix), because To the recording of texts as massive documentation, with linguistics as a means to the ends of ethnography and aesthetic appreciation, we can now add (…) the influence of structural linguistics on our ability to perceive poetic structure. (IV: 59) It is an eclectic and composite philology, though, one that has been composed out of classical philological principles (the collection and meticulous analysis of texts), anthropological heuristics (the Boasian and Whorfian emphasis on cultural categories, on culture as an organizing principle for linguistic form), ethnographic epistemology (the principle that things can only be found out by structured attention to situated contextualized behavior), and the influence of two important predecessors to whom we shall turn in a moment. his philology is oriented towards discovering verbal art, organized in a (structurally described) ‘grammar’ of discourse which yields implicit patterns and principles of organization, allowing us to see “artistry and subtlety of meaning otherwise invisible” (NK: 96).9 It comes down to considering spoken narrative as a level of linguistic structure, as having consistent patterns — patterns far less complex than those of syntax, but patterns nonetheless. (NK: 97) his level of linguistic structure revolves around three ‘universal principles’ (NK: 340, also 95). he first principle is that narratives do not consist of sentences, but of lines and relations between lines (verses, stanzas…). Identifying such lines and relations is the bread and butter of ethnopoetics, and considerable skill and technique are required to do so.10 Lines and verses are oten marked by particular formal linguistic features, from discourse markers and particles to syntactic parallelisms and intonation contours, where all of this is subject to what Roman Jakobson (Hymes’ first important predecessor) called ‘equivalence’ (Jakobson 1960). Equivalence is the second ‘universal principle’ that governs this form of art: “a variety of means is employed to establish formal equivalence between particular lines and groups of lines” (NK: 340). hus repetitions of (parts of) lines, similarities in length, number of syllables, intonation contours, grammatical concord and so on can all mark lines and groups of lines, and sudden changes in pattern indicate new 236 Jan Blommaert episodes in the story — new verses, stanzas, refrains etc. Finally — the third universal principle — there is always a general aesthetic organization to the story, a more global form of organization that connects the story to culturally embedded understandings of the logic of activities and experiences. his is the level where a story can become a captivating one, a joke a good one, a poem a beautiful one, and here, Hymes draws on insights from his second important predecessor, Kenneth Burke (e.g. 1969 [1950]). Attention to this level of structure leads to a higher level of abstraction in ethnopoetic analysis. Ater the identification of lines and groups of lines, a ‘profile’ of the story needs to be drawn which brings out the intricate and delicate correlations between linguistic form, thematic development (scenes, episodes) and the general (‘cultural’) formal architecture of the story. In the appendix to this paper, I will provide an illustration of such an architecture. Comparatively investigating such architectures, Hymes argues, could yield universal insights. Especially in NK, Hymes insists that stories are usually organized around numbers of lines — he talks of measured instead of metrical to denote forms of non-metrical formal internal organization of stories: “here are regularities in the relations among measured lines, just as there are regularities in metrical lines” (NK: 96). And these regularities, Hymes suggests, are a limited set: hese regularities have to do with cultural patterns, but also with the explorations and skill of narrators. In terms of cultural patterns, communities appear to build upon one of two alternatives: relations in terms of two and four or relations in terms of three and five. (NK: 96) hus, stories can be organized along series of two and four lines, verses or stanzas, or alternatively along series of three and five — with all sorts of permutations occurring within both alternatives. Hymes here argues for a different kind of universal: an aesthetic-formal universal which simultaneously may be a universal of the discursive sedimentation of human experience. Summarizing, Hymes sees ethnopoetics as a descriptive (structural-philological) tactics capable of addressing (and analytically foregrounding) implicit formal patterns in narratives, that can help identify them as ways of speaking within a culturally embedded speech repertoire. Such patterns are responsible for the poetic, artistic, aesthetic qualities of such narratives, and these qualities are a central part of their meaning and function. At the same time, these aesthetic qualities are deeply cultural, and they may reveal the cultural ‘grammar’ of human experience, both at the level of specific communities (repertoires) and at the level of universals of language and culture. In that sense, Ethnopoetics as functional reconstruction 237 ethnopoetics fulfils (or attempts to fulfil) the promises of linguistic anthropology in the Boas-Sapir-Whorf tradition: to detect and make understood the cultural in language, the relation between culture and linguistic form, and the way in which language use feeds into culture. En passant we take on board conceptions of language form, function and usage that are fundamentally different from those of mainstream linguistics, and we venture into an exciting new world of theory and analysis. Ethnopoetics as functional reconstruction But there is more: ethnopoetic analysis, to Hymes, is a form of restoration: he work that discloses such form can be a kind of repatriation. It can restore to native communities and descendants a literary art that was implicit, like so much of language, but that now, when continuity of verbal tradition has been broken, requires analysis to be recognized (RT: vii) In order to understand this argument, the décor of our discussion needs to be slightly changed, from the texts themselves to the tradition of recording and analyzing them. Hymes is critical of the linguistic and folkloristic traditions of scholarship on ‘oral tradition’, claiming that they produced a record which has dismembered the very traditions as traditions, i.e. as something deeply connected to culture and cultural activity — as performable, poetically organized narrative, operating as a cognitive, cultural, affective way of handling experience. Losing that dimension of language means losing the capacity to produce voice — to express things on one’s own terms, to communicate in ways that satisfy personal, social and cultural needs — to be communicatively competent, so to speak. Consequently: he fact is that one cannot depend upon most published versions of Native American myth. Even if the native language is preserved, its printed form is two steps away from what was said. he first step, from what was said to what was written down, cannot be transcended. We are dependent on what did get written down. But we can transcend the step between what was written down and what was published. Choices were made, mistakes sometimes made, in the course of that step. And words may be given a form they did not have. For generations they have been assumed to be prose and put in paragraphs ad hoc. Experience in recent years has shown that such narratives had an organization of their own, an organization not of paragraphs, but of lines and groups of lines”. (RT: vii) 238 Jan Blommaert he stories, in other words, were not represented as poetry — a form which bespeaks artistry and aesthetic intentions (Burke’s “arousal and satisfaction of expectation” — NK: 340) — but as denotational, linearly organized, ‘sensemaking’ text. Features of narration such as repetition (one of the most common forms of Jakobson’s equivalence, hence usually revealing emphasis or insistence) were oten dropped from printed editions; code-switching or borrowing were similarly oten edited out; likewise with ‘nonsensical’ sounds or utterances, audience responses and so forth: the model for native text was that of literature in European languages. And as a consequence, little was learned about how such stories fitted into local speech repertoires, how they functioned in contrast to other forms of language use, how they operated in a group as a culturally legitimate, relevant, useful way of speaking.11 A lot of what Hymes does in IV, RT, and NK, consequently, is re-transcribing and critically retranslating texts previously published by the likes of Edward Sapir and Melville Jacobs, organizing them in a different presentational format. his is methodologically essential: questions of mode of presentation arise because ethnopoetics involves not only translation but also transformation, transformation of modality, the presentation of something heard as something seen. he eye is an instrument of understanding (NK: 40). In other words, the stories need to be presented not as denotational text but as aesthetically organized poetic text, text containing the implicit forms of organization that make it meaningful culturally as myth, popular story, anecdote or experiential narrative — where such genre differences are a matter of implicit poetic organization triggering generic recognizability. Using old-fashioned anthropological terminology, the ethnopoetic transformation of texts is aimed at visualizing the emic organization of the text, the text as organized in terms of culturally embedded genre features. And such features, it should be underscored, are primarily aesthetic features, features of narrative-poetic shape, not only linguistic form. We could reformulate Hymes’ point of view as the primacy of the aesthetic functions of narrative, and the primacy of narrative as a cultural genre (or genre complex). Analytic interventions of the past, Hymes insists, have erased these aesthetic features, focusing on form instead of on shape, and reducing narrative to surface-segmentable (explicit) denotational expression organized in graphic units belonging to the language-ideological repertoires of the describers, not of the narrators. he essence of the object of inquiry — its implicit, cultural organization — was thus erased from the record, effectively precluding an accurate Ethnopoetics as functional reconstruction 239 understanding of such texts as cultural artefacts, as forms of language use having complex, multiple functions, rather baffling degrees of (non-random) variability, and a unique situatedness in the act of telling.12 Since “[n]arratives answer to two elementary functions of language, presentational as well as propositional” (EL: 205), deleting presentational aspects from the record means the loss of the narrative (behavioural, cultural) aspects of the texts. his is not only a problem for analysts; it is an even greater problem for members of the communities from whom these narratives were taken. For them, the written, published versions of stories are oten the only remains of an endogenous oral tradition, and given the functional dismembering of such stories in scholarship, stories are no longer oral and can no longer be performed as poetry, i.e. as texts organized according to community-specific poetic conventions. hus: One merit of verse analysis (as this work can be called) is that it helps recognize the worth of oral traditions for which we have only written evidence. (…) When lines, verses, and relations are recognized, one can venture to perform the narratives again, given appropriate circumstances. (NK: 98) And in that way, by showing the implicit structure of such narratives, the rules of such implicit art forms could be learned anew, so that narrators can acquire again the tacit, implicit knowledge of form and the conventions of telling culturally appropriate, useful, functional stories. We will come back to the political dimensions of these restorative aims of ethnopoetics in a moment. At this point, a theoretical argument deserves to be underscored, one that leads us back to Hymes’ ‘narrative view of the world’. Ethnopoetics, to Hymes, is about reconstructing the aesthetic functions of narratives, thus reconstituting them as a culturally recognized and valid complex of genres combining cognitive, affective, emotive, aesthetic and other aspects of language. his, then, goes back to his view of functional relativity — the fact that the function of language forms is a matter of their place within culturally configured repertoires, which cannot be posited a priori but need to be determined ethnographically (EL: 44ff). he scholarly tradition of investigating narrative has assigned particular functions to such narratives: those commonly ascribed to denotational, linearly organized, written/printed explicit prose text. And by doing that, such narratives have lost their ‘meaning’ — their usefulness, their functionality as narrative in particular communities. Ethnopoetics is the technique by means of which some of these functions could be restored. Rather than just as repositories of ‘wisdom’ or ‘customs’, such texts could now again become objects of aesthetic pleasure, of entertainment, opportunities for the 240 Jan Blommaert display of narrative skill and virtuosity, for endless variation and renewal, for negotiating and enacting norms, conventions, standards — for culture in the sense of dynamic social-semiotic transmission. he politics of ethnopoetics his could easily be read as a classic instance of salvage linguistics, and nothing would be wrong with that. But once again, there is more. he effort of reconstruction is inspired by an acute awareness of inequality and a desire for equity. Reconstructing the functions of narratives is not just a matter of reconstructing latent cultural heritage, it is a politics of recognition which starts from a restoration of disempowered people as bearers and producers of valuable culture, over which they themselves have control: recognizing one’s language, to Hymes, means recognizing one’s specific ways of speaking. his is how Hymes concludes In vain I tried to tell you: We must work to make visible and audible again that something more — the literary form in which the native words had their being — so that they can move again at a pace that is surer, more open to the voice, more nearly their own (IV: 384) Voice — this is what functional reconstruction is about. Ultimately, what ethnopoetics does is to show voice, to visualize the particular ways — oten deviant from hegemonic norms — in which subjects produce meanings. As mentioned earlier, in Hymes’ view (most eloquently articulated in EL), voice is the capacity to make oneself understood in one’s own terms, to produce meanings under conditions of empowerment. And in the present world, such conditions are wanting for more and more people. he Native Americans of IV, RT and NK are obvious victims of minorization, but Hymes extends the scope of ethnopoetic reconstructions in EL to include other marginal groups in society — African Americans, working-class college students, other minorities. Interestingly, such groups frequently appear to be the victim of a very Bernsteinian phenomenon: the negative stereotyping of part of their repertoire, the dismissal of their ways of speaking as illegitimate, irrational, not-to-the-point, narrative rather than factual (Bernstein would say: restricted rather than elaborate), and one form of inequality of opportunity in our society has to do with rights to use narrative, with whose narrative are admitted to have a cognitive function (EL: 109). Ethnopoetics as functional reconstruction 24 More in general, Hymes observes (alongside many others, e.g. Gumperz, Labov, Bourdieu) that ‘making sense’ oten, concretely, is narrowed to ‘making sense in particular ways’, using very specific linguistic, stylistic and generic resources, thus disqualifying different resources even when they are perfectly valid in view of the particular functions to be realized. It is in this world in which difference is quickly converted into inequality that attention to ‘emic’ forms of discursive organization takes on more than just an academic import and becomes a political move, aimed at the recognition of variation and variability as ‘natural’ features of societies, and at recognizing that variation in cultural behavior can result in many potentially equivalent solutions to similar problems. his, consequently, radicalizes the issue of diversity, because it shits the question from one of latent potential equivalence to one of effective disqualification and inequality. If all languages are equal, how come some (many!) are not recognized even as languages? How come that the latent and potential equivalence of languages, in actual practice, converts into rigid language hierarchies? hat potential equality is matched by actual inequality? that “unfamiliar pattern may be taken to be absence of pattern” (EL: 174)? Part of Hymes’ answers is that diversity still requires deeper understanding as to its actual forms, structures and functions. Misunderstanding of such aspects of diversity, oten resulting from errors in past work or sloppiness in current work, precludes appreciation of diversity as a solution. In this respect, he is particularly hopeful that a different universal dimension of human sense-making may be found in the numbered patterns he discovers in Native American texts. Such patterns, he submits, could recast visions of diversity: In sum, there lies ahead a vast work, work in which members of narrative communities can share, the work of discovering forms of implicit patterning in oral narratives, patterning largely out of awareness, relations grounded in a universal potential, whose actual realization varies. To demonstrate its presence can enhance respect for an appreciation of the voices of others. (EL: 219) his is no longer just about developing a better, more accurate philology of native texts; ethnopoetics here becomes a program for understanding voice and the reasons why voice is an instrument of power with potential to include as well as to exclude. It becomes a critical sociolinguistic method that offers us a way into the concrete linguistic shape of sociocultural inequality in societies. 242 Jan Blommaert Conclusions I have not done justice to the full richness of Hymes’ methodology, having focused instead on the theoretical and methodological, programmatic, character of his ethnopoetic work. It is too oten dismissed (and too easy to dismiss) as an aridly technical toolkit of bewildering complexity, aimed at developing more ‘authentic’ or ‘accurate’ (philological) readings of badly edited Native American texts. It is, to be sure, far more than that, and it has been my attempt to bring out and foreground some of the fundamental assumptions underlying ethnopoetics. hese fundamental assumptions are in line with other lines of work in Hymes’ large and complex oeuvre. Even if ethnopoetics looks like a very different type of language study than, say, Hymes’ papers on communicative competence or the ethnography of speaking, it is inspired by precisely the same deep preoccupations. hese include an ethnographic epistemology and a concern with language-as-praxis, as a socially and culturally conditioned form of human behavior subject to constraints and developments that cannot be predicted a priori but need to be established empirically. he aim of ethnopoetics, furthermore, is to arrive at a reconstruction of languages-as-sociolinguisticsystems: of language as composed of culturally embedded ways of speaking. he fact that language is oten misunderstood because its role in societies is oten only superficially addressed is another thread that shoots through his ethnopoetic work as well as his other work. And here perhaps more than elsewhere, he illustrates the unpredictability of form-function relationships in the structure of language-in-society, as well as the — real, effective — dangers of taking form-function relationships for granted. It not only leads to misunderstanding, it also leads to disqualification, dismissal and erasure for those who produce ‘strange’ patterns. A book such as EL clearly, and convincingly, demonstrates the ways in which ethnopoetics fits into a larger sociolinguistic-programmatic edifice, both of theory and of commitment. here is thus room for exploring ‘applied’ topics for ethnopoetic analysis — for taking it beyond the study of folkloric oral tradition and into other spaces where narrative matters: service encounters, police interviews, asylum applications, trauma narratives, social welfare interviews, political speech, advertisements and promotional discourses, and so forth.13 It would be a great pity if a powerful analytic tool such as ethnopoetics would remain under-used because of it stereotypically being pinned on a small set of particular analytic objects. Ethnopoetics as functional reconstruction 243 Notes . Many of the ideas in this essay became clearer during a series of long talks I had with Dell and Virginia Hymes in late September 2004, and I am very grateful to them for their hospitality and generosity in time and attention. Ethnopoetics has over the years been a consistent topic of discussion with Stef Slembrouck, whose influence is also gratefully acknowledged. Intensive work with Speranza Ndege (University of Nairobi) while she was completing her PhD with me in late 2002 forced me to focus on many of the technical and theoretical aspects of ethnopoetics, and compelled me to adopt more nuanced views on many issues. Speranza’s work resulted in a magnificent dissertation (Ndege 2002), which provided (rare) evidence for Hymes’ claims about the occurrence of numbered patterns in stories. Finally, Ben Rampton and Bill Bright commented perceptively on an earlier version of this paper, and their input is gratefully acknowledged. 2. I once heard students refer to it as In vain I tried to kill you. 3. I am grateful to Bill Bright for pointing this out to me: it was a lexical trap so big that I failed to spot it. 4. Moore (2000: 67) has more recently noted the emphasis “in the ‘endangered languages’ discussion (…) on languages qua grammatical systems (and/or systems of nomenclature), as artefacts (…) of cognition: something akin to the Elgin Marbles, perhaps, in the realm of conceptualization”. See also Blommaert (2005) for an ethnographic critique of such views of language endangerment. 5. According to Hymes, modern linguistics has consistently overlooked the problem of functional relativity, oten wrongly taking functional stability and formal variability as the central assumption of analysis. his point is forcefully developed in EL; see also Hymes (1980: Chapter 1). 6. Observe that this claim resembles that of conversation analysts, who would argue that conversation (dyadic, sequential and rule-governed interaction) would be the most ‘natural’ (hence, sociologically and culturally most interesting) form of language usage. here is no reason why narrative — storytelling, big or small — could not be seen as such, if for nothing else because many conversations are, in fact, narrative, while not every narrative needs to be conversationally organized (though it is usually conversationally embedded, yielding interesting dynamics of triggering, partly addressed in e.g. Sacks’ work on telling jokes). Note that Michael Silverstein’s work on poetics draws on conversational examples: apart from the formal ‘mechanics’ of sequential organization, therefore, conversations also clearly display poetic (measured, even metrical) forms of structural organization (Silverstein 1985, 1997). 7. he scope of this paper does not allow for an elaborate comparison between Hymes’ ethnopoetics and the many other approaches to narrative; I will have to restrict myself to identifying what I believe are the distinctive characteristics of ethnopoetics, and encourage the reader to engage in such comparisons on the basis of this outline. Interested readers might want to start from Ochs & Capps’ superb review paper (1996). 244 Jan Blommaert 8. Hymes emphatically dismisses connections between this ‘practical structuralism’ and ‘structuralism’ as “what has been made of linguistic analysis in anthropology, semiotics, and the like” (NK: 123). It is easy to be misled by terminology here, and Hymes is not always the most helpful writer in this respect (witness famously cryptic lines such as “In aim, the method is structural, but in execution, it must also be philological” — Hymes 1966: 131). Hymes has maintained throughout his career a complex relationship with structuralism (see e.g. Hymes 1983). 9. he ‘practical structuralism’ shines through in statements such as this one: “One must work out a ‘grammar’ of the local world of discourse and work out the internal relations of a text in relation to that grammar before proceeding to analytic comparison and interpretation in terms of relationships found elsewhere”. (NK: 126) 0. here has been some debate on the criteria for identifying lines, and Hymes addresses comments and proposals by other scholars — Labov, Gee, Tedlock, and others — in IV, NK and EL. Along with Hymes, Dennis Tedlock is oten seen as the ‘founder’ of ethnopoetic analysis; see e.g. Tedlock (1983). . his problem of textual conversion — entextualization — is a language-ideological matter in which particular metalinguistic grids are being imposed on the text, recreating it as a particular form of text, culturally recognizable within the repertoire of those who edit it. See Silverstein & Urban (1996) and Bauman & Briggs (1990) for extensive discussions. 2. With respect to this situatedness, Hymes, especially in IV, devotes a lot of attention to the issue of dictation in the field: “Perhaps the most obvious influence on what we know of the traditions of nonliterate groups has been the constraint of dictation, and dictation slow enough to be written down; the effect on sentence length and the internal organization of texts has been increasingly revealed by research with tape recorder” (IV: 86). He also observes that the structure of narratives in fieldwork oten develops according to the informants’ appraisals of the developing competence of the researcher, stories becoming more complex ater long periods of fieldwork and repeated narrations. 3. To my knowledge, very little published research of this sort exists. Partly in collaboration with Katrijn Maryns, I have investigated African asylum seekers’ stories using ethnopoetics (Blommaert 2001; Maryns & Blommaert 2001; Maryns 2004). 4. For more detailed comments and suggestions on ethnopoetic technique, I refer the reader to Blommaert (2000), a working paper originally written for the benefit of students involved in the fieldwork project. Blommaert & Slembrouck (2000) provide an extensive discussion of a range of methodological issues related to ethnopoetic analysis and data representation. References Bauman, R. & C. Briggs (1990) Poetics and performance as critical perspectives on language and social life. Annual Review of Anthropology 19: 59–88. Bernstein, B. (1971) Class, codes and control, volume 1. London: Routledge & Kegan Paul Ethnopoetics as functional reconstruction 245 Blommaert, J. (2000) Analysing African asylum seekers’ narratives: scratching the surface. Working Papers in Urban Language and Literacies 14, King’s College London. Blommaert, J. (2001) Investigating narrative inequality: African Asylum seekers’ stories in Belgium. Discourse & Society 12/4: 413–449 Blommaert, J. (2005) Situating language rights: English and Swahili in Tanzania revisited. Journal of Sociolinguistics 9: 390–417. Blommaert, J. & S. Slembrouck (2000) Data formulation as text and context: On the (aesth)etics of analysing asylum seekers’ narratives. Working Papers on Language, Power and Identity 2, Ghent University. Burke, K. (1969 [1950]) A rhetoric of motives. Berkeley: University of California Press. Haviland, J. (1996) ‘We want to borrow your mouth’: Tzotzil marital squabbles. In C. Briggs (ed.) Disorderly discourse. New York: Oxford University Press. 158–203. Haviland, J. (1997) Shouts, shrieks, and shots: Unruly political conversations in indigenous Chiapas. Pragmatics 7: 547–573. Hymes, D. (1966) Two types of linguistic relativity (with examples from Amerindian ethnography). In W. Bright (ed.) Sociolinguistics. he Hague: Mouton. 131–156. Hymes, D. (1980) Language in education: Ethnolinguistic essays. Washington DC: Center for Applied Linguistics. Hymes, D. (1983) Essays in the history of linguistic anthropology. Amsterdam: Benjamins. Hymes, D. (1996) Ethnography, linguistics, narrative inequality: Toward an understanding of voice. London: Taylor & Francis. Jakobson, R. (1960) Closing statement: Linguistics and poetics. In T. Sebeok (ed.) Style in language. Cambridge MA: MIT Press. 350–377. Maryns, K. (2004) he asylum speaker: An ethnography of language and communication in the Belgian asylum procedure. PhD Dissertation, Ghent University Maryns, K. & J. Blommaert (2001) Stylistic and thematic shiting as a narrative resource: Assessing asylum seekers’ repertoires. Multilingua 20: 61–84. Moore, R. (1993) Performance form and the voices of characters in five versions of the Wasco coyote cycle. In J. Lucy (ed.) Reflexive language. Cambridge: Cambridge University Press. 213–240. Moore, R. (2000) Endangered. In Alessandro Duranti (ed.) Language matters in anthropology: A lexicon for the millennium. Special issue of Journal of Linguistic Anthropology 9: 65–68. Ndege, S. (2002) he Meru folktale: An ethno-poetic study. PhD Dissertation, Ghent University. Ochs, E. & L. Capps (1996) Narrating the Self. Annual Review of Anthropology 25: 19–43. Ochs, E. & L. Capps (2001) Living narrative: Creating lives in everyday storytelling. Cambridge MA: Harvard University Press. Silverstein, M. (1985) he pragmatic poetry of prose: Parallelism, repetition, and cohesive structure in the time course of dyadic conversation. In D. Schiffrin (ed.) Meaning, form, and use in context. Washington DC: Georgetown University Press. 181–199/ Silverstein, M. (1997) he improvisational performance of culture in realtime discursive practice. In R. K. Sawyer (ed.) Creativity in performance. Greenwich, CT: Ablex. 265– 312. 246 Jan Blommaert Silverstein, M. (2005) he poetics of politics: ‘theirs’ and ‘ours’. Journal of Anthropological Research 61: 1–24. Silverstein, M. & G. Urban (eds.) (1996) Natural histories of discourse. Chicago: University of Chicago Press. Tedlock, D. (1983) he spoken word and the work of interpretation. Philadelphia: University of Pennsylvania Press. Author’s address Jan Blommaert Department of African Languages and Cultures Ghent University Rozier 44 9000 Gent [email protected] Appendix: ‘I walked for seven hours’ By way of illustrating ethnopoetic analytic technique, I will try to show how implicit structure can be made visible in a small story, an anecdote. he anecdote is part of a long interview recorded in late 1997 with a seventy-five year old man, a former District Commissioner in the Belgian Congo. he man speaks Flemish Dutch with clear regional (dialect) accent. he topic of the interview was life in the colony and the practice of professional conduct in colonial service — the theme of a fieldwork project for students of African Studies, Ghent University in 1997–1998.14 he interview was transcribed by the students in ‘field transcript style’, i.e. using minimal codes and focusing on general patterning of talk. Present during the interview were the interviewee, his wife, and three female students who do the interview. During the interview, lots of anecdotes are told. hese are generically marked and usually start with an explicit generic framing device (‘once..’, ‘there’…). Let us start from the field transcripts. he Dutch field transcript is the original transcript provided by students; I have provided an approximate English equivalent. Dutch ik heb daar eens zeven uur gemarcheerd om in een dorp te komen /..en euh /.. als ik dan hoorde ik ze lachen /..en ik verstond nie wa da ze zeiden maar ik had altijd ne jachtwachter Kalupeshi heette die die had ik bij en die was van de streek en ik zei wat wat is ‘t groot plezier hij zei dat dat oud vrouwke wat daar zit hij zei die zei /.. ik moest al ik had al jaren gene blanke meer gezien en ik wou absoluut nog eens ne blanke zien ik moest dus naar de weg waar dat ik van kwam daar è/. en nu heb ik hem gezien /. nu hoef ik nie te gaan zei ze Ethnopoetics as functional reconstruction 247 English I once walked for seven hours there to get to a village/ and ehr/ when I then I heard them laugh/ and I didn’t understand what they were saying but a always had a gamekeeper with me Kalupeshi was his name that that one I had with me and he was from that region and I said what what is the big fun he said that that old lady who sits over there he said she said/… I had to I hadn’t seen a white man for years and I desperately wanted to see another white man so I had to go to the road where I came from right/ and now I’ve seen him? now I don’t need to go she said his is a short, at first sight unremarkable micro-narrative, certainly when represented as prose organized in sentences However, when we deploy an ethnopoetic apparatus focusing on line, verse and stanza organization, relations of equivalence and general aesthetic/poetic patterning in the story, we get an amazingly complex and delicate narrative, which shows how the narrator deploys content and form in synergetic, aesthetic moves. In the ethnopoetic transcript of this anecdote (which is the outcome of analysis — see Blommaert & Slembrouck 2000), I am using several procedures and codes. (1) Indentation and clustering of lines indicating the relations between lines. Some lines are subordinate to others, groups of lines can be identified. (2) boldface elements in the transcript indicate particularly salient markers, oten identifying lines and signalling relations among lines. hus, the difference between ‘en’ and ‘maar’ signals a change from one group of lines to another. (3) underlined fragments mark parallelisms: repetitive constructions that suggest themes and emphases on parts of the story, and contribute to the overall aesthetic organization of the narrative. Arrows further mark such repetitive poetic constructions. (4) single or grouped lines can be verses, marked by a, b, c in the transcript. A verse is typically a line identified as a main proposition (and marked by a line-initial narrative marker such as ‘and’), potentially complemented by dependent, subordinate lines. (5) Several verses can form a narrative unit — a scene — in which part of the narrated event is developed. In the transcript, scenes are marked by (I)–(IV) Taken together, we get the ‘architecture’ of this story, and it looks like this: 248 Jan Blommaert In this brief anecdote, three actions are put in a sequence. Together, they form the ‘stuff ’ of the story: 1. I arrive in a village, hear them laugh and don’t know what it means 2. I ask the gamekeeper what it is about 3. He translates the words of an old lady Actions 2 and 3 are both narrated communicative events: dialogues with two turns each. Between actions 1 and 2, the narrator inserts an out-of-sequence scene: ‘I had a local gamekeeper’. his part provides contextual information, it complements the sketch of the situation and introduces a character for actions 2 and 3. hese actions are narratively organized in four scenes, marked by numbers (I)–(IV): (I) (II) (III) (IV) generic framing: deictic anchoring and sketch of the situation. First action: I heard them laughing and did not understand them out-of-sequence contextual element: I had a local gamekeeper Second action and first part of dialogue Turn: I asked him what it was about hird action and second part of dialogue Turn: gamekeeper translates the words of the old lady (reported speech framed by ‘she said’, ‘he said’). Whereas the actions are, so to speak, ‘content’ elements of the story, the scenes are narrative elements in which form and content are blended into a poetic organization of lines and relations between lines. Let us have a closer look at the different scenes: Scene I his scene comprises three verses (a, b, c) marked by (a) a generic framing device for the very first verse of the narrative: explicit deictic anchoring of the story and sketch of the setting (“I once walked there for seven hours” — in italics in the transcript); (b) the use of the connective ‘en’ (“and”) for verses b and c, which both contain the first action of the story. Scene II his scene is an out-of-sequence scene with two verses (a, b) in which contextual information is given: ‘I had a gamekeeper there — his name was Kalupeshi — I had him there — he was from that region’. Note the parallelism: proposition-elaboration // proposition-elaboration. his scene is introduced by ‘maar’ (“but”), an adversative discourse marker that marks a break with scene I as well as with scene III — both are identified by the use of ‘en’. Scene III he action sequence of the story is resumed by means of the connective ‘en’, which establishes cohesive links with Scene I. In this one-line scene, we get the first turn of the dialogue (T1): the narrator asks Kalupeshi what the big fun was all about. he dialogue action is framed by an explicit metapragmatic signal: the phrase ‘ik zei’ (“I said”). Scene IV his complex three-verse scene is the second turn of the dialogue (T2). Like the first turn, it is introduced by a metapragmatic phrase ‘hij zei’ (“he said”). he reported speech of the old lady is framed initially as well as finally (‘sandwiched’) by ‘die zei/zei ze’ (“she said/said she”). he lady’s reported speech itself is a three-verse rhyme with considerable internal parallelism: (a) I haven’t seen a white man in years, (b) (if I wanted to see one) I had to go to the Ethnopoetics as functional reconstruction 249 road, (c) now I don’t have to go anymore (he came to me). he parallelisms mark differences between main-subordinate lines and marking of the punchline: a. IK (main) — BLANKE ZIEN (rhyme) EN (subordinate — BLANKE ZIEN (rhyme) b. IK (main) EN (subordinate) -ZIEN (rhyme) c. NU (punchline) No formal rhyme but ‘semantic rhyme’: ‘now I have seen him’ Especially in Scene IV, the complex poetic patterning (three verses with a clear refrain of three rhyming repetitions) produces a stylistic intensification of the narrative — Hymes would use the term ‘full performance’ for this — and supports the stylistic and frame shit into a doubly layered reported speech: ‘I tell what Kalupeshi said the old lady said’.