The INTERFACE Series ‘Language, Literature and Critical Feminist Stylistics Practice Sara Mills Ways of analysing text : 7 lakers ‘Twentieth-century Fiction From text to context Literature, Language and Change Peter Verdonk and Jean Jacques Weber Ruth Waterhouse and John Stephens Literary Studies in Action ‘Alan Durant and Nigel Fabb ‘Language in Popular Fiction Walter Nash Language, Text and Context Essays in stylisies Edited by Michael Toolan ‘The Language of Jokes Analysing verbal play Delia Chiaro ‘Language, Ideology and Point of View Paul Simpson ‘A Linguistic History of English Poetry Richard Bradford Variety in Written English, Exploring the Language of Drama Literature about Language From text to context Valerie Shepherd Edited by Jonathan Culpeper, ‘Twentieth-century Poetry Mick Short and Peter Verdonk From text to context ‘The Discourse of Advertising Edited by Peter Verdonke Second edition ‘Textual Intervention Guy Cook Critical and creative strategies for literary studies Rob Pope ‘The Series Ealitor ‘Ronald Carter is Professor of Modera English Language at the University of ‘Nottingham and was National Coordinator of the ‘Language in the National Curriculum’ Project (LINC) from 1989 to 1992, Narrativ A critical linguistic introduction Second Edition. Michael ToolanIn loving memory of Margaret ‘Mac’ McAloren who told such stories, and never told on us, AGA, First published 1588 ‘Reprinted 1991, 1992, 1994, 1995, 1997 12d 2001 Abingdon, Oxon, OX16 4RN 19 published othe USA and Can by Routledge Seert 270 Manon Ave, New York NY 10016 ‘Typeset in Times by Wearset, Boldon, Tyne and Wear Allsihts seserved. No pat ofthis book cole my be reps or feproduced or ue in any form o by an leone neces 20 keno clined eg sapien erg nay naan org British Library Catal of Congress Cataloging n Pubcon Date ‘Michael J. fad i = 3 linguistic introduction / Michael Toolan.— 302.7 Té6 2001 4014121 ISBN 0-415-231744 (hbk) eee ISBN 0-415-23175-2 (pbk) Printed and bound by CPI Antony Rowe, Eastbourne Contents Preface Acknowledgements Preliminary orientations LL] Teller, tale, addressee 1 1.2 Typical characteristics of narratives 4 1.3 Narratives and non-narratives 8 14 Story-text-narration 10 Farther reading 12 Notes and exercises 13 Basie story structure 21 Storyftabulahhistoire 15 22 Propp’s morphology of the Russian fairytale 17 2.3 Barthes on narrative 22 24 Plot-summarizing: modelling intuitions 29 25. Inssearch of the grammaticization of plot structure 31 ther reading 37 Notes and exercises 37 ‘The articulation of narrative text I: time, foealization, narration 3.1 Narrative text: single level of analysis 41 32 3.3 Temporal refractions in text: Nabokov’s Prin 54 34 3.5. Perceptual focatization as primary 63 3.6 Narrators and narration 64 32.7 Simpson's typology of narratorial modes 68 Further reading 77 Notes and exercises 77 atToT 4 The articulation of narrative text I: character, setting, suspense, film 41 Character 80 42° Greimas’ actant model 82 43° Character traits and attributes 86 44. Distinctive feature characterology 88 43. Setting 91 46 Character and sewing in ‘The Dead’ 94 47 Creating surprise and suspense in narratives 99 48 From prose to film: radical translation 103 49° The grammaticization of character and situation 107 Further reading 112 Notes and exercises 112 3. The articulation of narrative text lk: representing character discourse 5.1 Achieving immediacy in the narration of thoughts 116 5.2 Modes of speech and thought presentation 119 53° Differences between Direct and Indirect Discourse 125 54. Different again: free Indirect Discourse 130 53 Who speaks, who thinks? 133 5.6 FID: functions and effects 134 Further reading 140 Notes and exercises 140 6 Nan approach 8s socially situated: the sociotinguistic 6.1 Labov and narrative structure 143 62 Fixed narrative clauses, free evaluative clauses 145 63 Abstracts and orientations 149 64° Evaluation 151 65° Doing and saying 153 66 Internal evatuation 135 67 Coda 157 68 Stories in societies 159 69 Narrative performance 160 6.10 Dispersed, embedded, and group oral narratives 162 6.11 From Labov to literature 167 Further reading 172 Notes and exercises 173 7 Children’s narratives Bd 7.1 Stories for, by, and with children 178 7.2. Storytelling and emergent literacy 180 73. Differing styles, differing orientations 182 74 Children's narrative development 185 7.5. Children’s narratives and the development of registers and genres: the systemicclinguistic ‘approach 189 7.6 The sysiemic-linguistic account of story genres 193 7.7 Stories for and with children 197 Further reading 203 Notes and exercises 204 8 Narrative as political action 206 81 The contexts of narratives 206 8.2. Hard news stories in the newspaper 206 83. Political narratives in the news 208 84 The unfolding news story: a contemporary example 212 85. The linguistic apparatus of political construal: notes on key resources 221 86 News stories online 230 87 Stories of class and gender 233, 8.8 Prejudice in ethnic narratives 234 89. Stories in court 235 Further reading 239 Notes and exercises 239 Bibliography a Indexmah Acknowleagements: ‘when only Roisin was old enough to enjoy Burglar Bill, and Miciam's story had not even begun. No ven begun. Now all three are old nou explain the new narratives to me. eee The author and the publisher also wish t : is (0 thank the copyr their permission to reproduce the fol eine gue links Labour with murder rise’, by sr tise’, by Webster and Baldwin, © Times Newspapers Limited, 19 December 2000. fa “Race and Policing: Hague’s defiance infla a ames the anger’, by Paul and Andrew Grice, 19 December 2000, © The Independent Syndie fate “Hague race jibe angers ministers’, by Nich 19 December 2000, The Guardian 8 White and Nick Hopkins, “Tory Leader “won't be gagged on crime’ ber 2000, © The Telegraph. + by George Jones, 19 Decem- ‘Crazy Hague defies Dami dad’s plea’, 19 December 2000, © The Mirror. 1. Preliminary orientations does a linguistic approach come in, and how helpful can it re The following are introductory notes on these and other basic issue: which should at Jeast indicate the terrain to be covered, and why tors sometimes begin by stating the truism that any tale involves a teller, and that, therefore, narrative study must analyse two basic components: the tale and the teller. But as much could be said of ere is always inherently a speaker, separable from extended spoken ones, is ‘Tellers of long narratives can be surprisingly prese as they unfold a tale that ostensibly draws all our attention, as readers or listeners, to other individuals who are within the tale. As 2 result we may g our attention between two objects of interest: y itself, and the individual telling us n we road Coleridge’s ‘Rime of the Ancient wering Heighis ot ‘0 the rambling anec- the activity of ‘reading’ or scru- about these. Thus Mariner’ or Bronte’s Wi ote of a friend, part of the experience friend, Already the tw plication. In both texts mentioned, there is more than one the mariner, for instance, is a ‘higher’ teller who writes, ‘It is an ancient Mariner/And he stoppeth one of three’, But we can address such compli cations later, and should concentrate here on narrative’s dual essential spatiotemporally distant: here’s the present teller, seemingly close to the addressee (reader or listener), and there at a distance is the tale end its topic. This selection of effects of closeness and distance can be represented graphically:2 Preliminary orientations << [i ADDRESSEE But since the preser sense, too, in whic uncommoaly pr the sole access to the topic, there isa tive entails making what iat and absent three-way merging father than a division. Dia- grammatically this merging-and-immediacy can be represented as: TALE +» aopressee However, since tellers can become intensely absorbed in their self Benerated sense of the distant topic they are relating, addressees some- times have the impression ¢ taken leave, so as to be more fall third type of felation between and merging) might be cast thus: involved in the removed scene. This and addressee (a withdrawing tt soonesee TALE —_— In short, narratives always involve a Tale, a Tell involv a Teller, es these can Be ‘placed’, notion: cl depiees Of Uti aI s back to the viewer, appears ning boys Who are evidently Yascinsted Absorbed by what he tes them, The old man is using his whole ams to Point out to the sea, visible in the distance. Bi 4 . But the boys’ eyes are on th man and his gesturing arm, not any distant scene he may be designating. i Narrative focusses our attention on to a story, a sequence of events, through the direet mediation of a ‘telling’ which we both stare at and Preliminary orientations 3 through, which is at once central and peripheral to the experience of both absent and present isness of those being story. Like the two young ing aria but our minds are fixed upon what is pointed at. (Hawthorn, 1985: es of narrative concerns its neces source, the narrator, We stare at the narrator rather than interact him as we would if we were in conversation; at the same tim« ‘Thus there is a teller in every tale to a far greater degree than there is a speaker in any ordinary turn at talk, Because narratives are, relative to ‘ordinary turns of talk, long texts and personalized or evaluated texts, there is a way in which, while your conversational remarks reflect who you are (your identity and values), in the course of any narrative the narrator's text describes that narrator. In brief snatches of conversation, a person ‘may be able, through accent-mimicry for example, to ‘pass’ for someone of different class or gender or ethnic identity; but to take on another's iden- ‘a sustained fashion, across a number of personal narratives, is ordi- narily very difficult, and may even imply disabling confusion or a personality disorder. The reflection/description contrast may be chiefly a ‘matter of degree, but itis arguably an important contrast with far-reaching consequences — e.g, even for assessments of mental health of This brings us to another important asset of narrators: narrators are trusted by theit addressees, In at least i to take up the role of knower, the addressees’ adopted role of learner ot those which originate from those who have power, author over us. Any narrator then is ordinarily granted, as a rebuttable presump- is also a granting or asserting of misrepresentation is a complex process, diffic cation of it arises far from literature: in criminal cases of serious fraud. ‘Where, after having pleaded not guilty, a defendant is found guilty, the sentencing judge often refers to the obfuscating detailed deception that has been uncovered as ‘a complex tissue of systematic distortion and fabri- cation’, or uses a similar revealing description. Even before we attempt a working definition of narratives, it is clear4 Preliminary orientations ‘cut off’ in some respects from surrounding co- text and context (their verbal and non-verbal environment, respecti the former comprises any language that precedes or follows the nar they’re just ‘there’, it seem ts or paintings, and you can take them or Ieave them. They differ, at least in degree, from more transac- tional uses of language, as when someone asks you a question, or makes 4 request of a promise or warning: in such cases there is strong expecta tion that the addressee will zespond or act in predictable ways. So some of the normal constraints on how we make sense of discourse seem to be suspended. And it seems we do not always have to relate narratives directly and immediately to their authors, or socio-historical back- grounds, 1.2 Typical characteristics of narratives We can begin to define narrative by noting and inspecting some of its typical characteristics: 1A degree of artificial fabrication or constructedness not usually appar- ent in spontaneous conversation, Narrative is ‘worked upon’ ‘Sequence, emphasis and pace are usually planned (even in oral narra. ve, when there has been some rehearsal — previous performance — of for example, elaborate deserip- ly articles. 2A degree of prefabrication. In other words, narratives often seem to have bits we have seen or heard, or think we have seen or heard, before (recurrent chunks far larger than the recurrent chunks we cal words). One Mills and Boon heroine or hero seems much like anoth Hardy, ete., seem to be thwarted (for a time at least) in roughly comparable ways, And the kinds of things people d res seem to repeat them- selves over and over again ~ with important variations, of course. Again, prefabrication seems common in various types of writing and al spectacle besides narrative, although the kinds of things men- joned above seem particularly to be prefabricated units of narrative. 3 Narratives typically seem to have a ‘trajectory’. They usually go some- where, and are expected to go somewhere, with some sort of develop- ind even a resolution or conclusion provided. We expect them ve beginnings, middles, and ends (as Arist Preliminary orientations 5 And they all lived happily ever after; E since then, the dragon has never been seen again and notice the finality and permanence conveyed by the/ever/never / pair. Or consider the common story-reader’s exit-line: Sco And that is the end of the story. which has near-identical counterpars in the closing sequences of radio and television news bulletins. All these examples mark this attention to the expectation of closure and finality one aspect of the jectory. Relatedly, the diaries.and live commentaries, in that in the latter new intervening acts, beyond the control of the witness/reporter, can dictate the shape and content of the report. In true narratives, arguably, the teller is, ¢ Dickens and the bedtime story- ve a teller, and that teller, no matter how », is always important, In this respect, de special characteristics, narrative is language communication like any other, requiring a speaker and some sort of addressee. es are richly exploitative of that design feature of language isplacement (the ability of human language to be used to refer ings or events that are removed, in space or time, from ei removal or rative. Thus a radio running commentary on si is approaches the status of narrative by virtue of spa is not a ner- rative at all if [am at the football match directly witnessing, and listen- ing to the radio commentary). But live commentaries, like real diaries, breach characteristic 3 above, and are arguably not narratives at all lights of sports and other events ighlights of matches and events temporally remote from the teller and his audience. Compare our practices with those of the honeybee,OS reuminary orventanons I-wagging is no proper narrative in our sense, but merely a kind of reflex observation. As Roy Harris has remarked: Bees do not regale one another they found last week, nor discuss find tomorrow. iscences of the nectar gether the nectar they might (Harris, 1981: 158) This is lovely image (or narrative), partly because in fact itis some- thing that (as far 2s we know) we humans alone do, ant animals ~ even those with simple language systems. A first attempt at a minimalist definition of narrative might be: nn recognizes that a natrative is a sequence of events. But is really a complex term, presupposi some recognized state or set of conditions, and that something happens, causing 4 change to that state. The emphasis on ‘non-random connectedness’ means that a pure collage of described events, even given in sequence, does not count as a narrative. For examy turn supplies a one-paragraph descr these paragraphs are then pasted together, they will not‘count as a narra- tive unless someone cowes to perceive a non-random conn: taken to be motivated and significant. This curious transitional area between sequen- al description and consequential description is one of the bases for the fun of a familiar party game in which people around a table take turns to write a line of a ‘story’, the other lines of which are supplied, in secret, by the other participants. ‘The important role of ‘change of state’ has been celebrated in the more term transformation by the structuralist Tzvetan Todorov (1977 mn of successive facts does not constitute a narrative: these facts must be organized, which is to say, ultimately, that must have elements in common. But if all the elements are in common, there is no longer a narrative, for there is no longer anything recount. Now, transformation represents precisely a synthesis of Preliminary orientations 7 ks two facts without their being able Gifferences and resemblance, to be identified, n of narrative also suggests that consequence is not so iven’ as ‘perceived’: narrative depends on the addressee seeing it as narrative - the circularity here seems inescapable, While most would agree there can be legitimate disagreement as to the status, as narrative, of Jess familiar and complex structures. For ‘example, imagine you enter a cartoonist’s studio and find three frames, on settings, furniture, etc., and seem to be about quite unrelated topics. seem to be rough drafts, because on the corner of one is a coffee-ring, where ‘has carelessly left a cup, and on the second one there’s a food top corner, while the third has some cigarette ash and a coffee ‘That example, however whimsical, tries to touch on a fundamental but problematic feature of narrative study. Perceiving non-random connected ness in a sequence of events is the prerogative of the addressee: itis idle for anyone else (e.g, @ addressee just does not s ultimate authority for ratifying a text as a narrative rests not with the teller buat with the perceiver or addressee. ness, non-randoriness, and sequet terms first highlighted by Aristotle, we expect ends as well as beginnings and middles (something not commented upon iti the quotation from Todorov above). In more twentieth-century terminology, we expect complex motivations and resolutions — even in quite ‘simple’ tales such as, more concern: eliminated or using events and of the way in which the concern is inished. (Colby, 1970: 177) As we shall see in the next chapter, this definition is similar to that of the pioneer Russian narratologist Vladimir Propp. Propp studied the over arching structure of the Russian state of equilibrium is turbulence brings disequi (perhaps an jum and upheaval before some sort of leads to the restoration of a modified y izes the role of the perceiver, but not that of an independent teller. This is because the two roles are not entirely, separate, I have already suggested that the activity of perceiving& Preliminary ortentations consequential relatedness of states is the enabli we might in formed by al addressees. ly to be per- at least if they intend to be tellers or 1.3 Narratives and non-narratives red sequence of non-randomly connected ng, as the experiencing agonist, humans or or other sentient beings, from whose experience we learn’, lefinition introduces (1) one or more foregrounded individuals as mncers, and (2) the idea of addresse ti jients might 1s. And if one wished to elaborate this one step further, it might be to specify that our preference is often for the sequence of connected events to take shape around a state or period of turbulence of crisis, subsequently resolved. That is, while a sequence of events entails some sort of change of state, a sequence containing a resolved crisis or problem entails a pronounced change of state, : 7 All these more contentious elaborations of the. il minimalist defini- tion are useful when one attempts to distinguish narratives from non-narrative texts. For if a narrative is a sequence of logically at logically related events, bound together by a recurrent focus (which may not be constant or exclusive) on one or more individuals (‘charact the reader becomes interested (positively or negatively); is in addition that kind of chronological sequence in which a period of lence, crisis or uncert: superseded by a later stage of calm, sol closure; and if these are indeed defining criteria of narratives; shen’ texts which display few or none of the above features will not be narratives. And texts which have some can summarize those three chief or defining features as: * sequenced and related events; ferary text such as Michelle Roberts’ novella ~ if that is entitled Une Glossaire/A Glossary. This comprises a series of ed glossary entries, with French headword and English- Janguage commentary, which offer brief (typically one-page) glimpses of Preliminary orientations 9 collections of the rural Norman childhood that the narrator, with French mother and English father, had access to. Cumulatively the text of recovery, of beachcombing for fragments, and of proceeds, Is Une Glossaire/A Glossary (as a wh ive, bythe foregoing criteria? three core desiderata hing of a foregrounding of one or more ini ‘sketching of conditions that may lead to a crisis. But is this ‘glossar ‘more a journal than a glossary? If it a journal then, ce a genuine diary, it would lack the poss 3 mapping-out a before e already known and past to the teller, who is now in a con- trolling position from which to recount them. But the very first section of Une Glos hero (seeker or vi Note that an actual chat example, some individu: hharacter role (for Jin and false hero) be filled by several individuals (there could Beople functioning as helper or villain). Demonstrating the is descriptive apparatus to his corpus of stoties in meticu- lous detail, Propp concludes: ee to the particular tales in his compus, sit ‘certain fict themselves to Proppian analysis without too much strain (see the ‘Notes Basic story structure 19 ‘Morphologically, a tale ... may be termed any development proceeding from villainy ... or @ lack ..., through intermediary functions to mar- riage ... or to other functions employed as a denouement, Terminal funetions are.at times a reward ... gain or in general misfortune ... an escape from pursuit each new lack creates a new move. On ay have several moves and when analyzing a text, one must first of all determine the number moves of which it consists. One move may directly follow another, but they may also interweave; a development which has begun pauses, and anew move is inserted (Propp, 1968: 92) ll not spend time summarizing just how Propp applies this morphology my main purpose here is to ie what he means by ‘function’, ‘role’, and ‘move’, so that we can -ntify similar elements in other stories. And the is 1d exercises’ section at the end of this chapter). To take an example from ular culture, consider the Star Wars film Proppian functions and moves, the characters filling six of the seven ‘core roles are easy to ist: Villain: Darth Vader Dispatcher: Luke's uncle Donor/provider: Obi Wan Kenobi (magical power provided is the Force) Helper: Yoda Hero (seeker or victim): Luke Skywalker Princess (+ father): Lei ‘You can apply the Proppian categories to any composed narrative, across e Buffy the Vampire Slayer, and ‘The following story written by a seven-year-ol functions appended:on the left, may serve to demonstrate how easily and appropriately Propp’s grammar can ft simple tales. 1 Once upon a time there was @ bunny named Benjie + magical agent 2 and she had magic powers. 3 One day she was walking in the woods Departure anda ‘bunny boy appeared and they went together for a walk and aman appeared with a big net20 Basic story structure [ 7 and he got the two bunnies and went in a big ship. ilsiny | 8 Poor bunnies. [9 They were caught now, Stragte 10 Butright thea the git! bunny tripped the man. and they got free once agai So the boy bunny thanked the girl bunny for saving him, The boy bunny asked the him and she seid yes. So they had six bunny babies and they lived happily ever after. bunny to marry (text and anal See Christie er al., from King and Rentel, 1982. 1984) If Propp’s schema fits the above story with eloquent ease, we might now ut it to work on a far more complex tale, that of ‘Eveline’, from Joyce's Dubliners collection, (I will be discussing this story in future chapters in relation to a number of issues.) Propp’s very first function seems almost uncannily relevant: ‘One of the members of a fami absents [himself] from home. Relevant, but not applicable mechanically. We might say, for instance, that the ‘action’ of ‘Eveline’ is a dramatization (chiefly a menial cramatization) of a stage within that first function: One of the members of a family reconsiders a decision to absent herself from home, While Propp's fairytales proceed through developmental actions, ‘Eveline’ is very largely a mental projection, both forward to-possible future events and backward fo actual past ones: remembered before-events and imag- ined after-events, Notice how her opening revery, up to sentence 24, is a conspectual review of past circumstances as i ation. As Propp notes, a story hi ation, one into am is introduced by the function of absentation or another of the seven preparatory functions. But in “Evelin tuation itself that is extensively dwelt upon, and none of the first 23 sentences appears to constitu function. Nor does this tendency lapse with sentence 24, sentence 24 comes confirmation that liscourse ~ the of functional departure the chief narrative mode adopted in this story (a mode I will discuss further in Chapter 4). But the Basic story structure 21 mphasis on elaboration of the multiple habitual circumstances that com- rise the initial situation, mere prologue to a story, remains. However, we can do some reconstruction of a simple developmental by him, at first finds this merely ‘an excitement’ but later begins t hhim. All of that, one imagines, would count as simply one fun jerms ~ “The heroine meets with a ¢2) stranger.” iking for Frank seems related, not nically perhaps, to the latter’s implied story-tellings, which themselves form a skeletal story: He had started as a deck boy, had sailed through the Straits of Magellan, 1ad fallen on his feet in Buenos Aires, and had returned to the old country to be his wife At this point in the text it seems that the gep between the story outlined in the two paragraphs above and the characier’s current reflections closes, for the next series of thoughts that are reported, incidents of fami (centred on her mother) both happy and grim, are the direct trigger of the ‘sudden impulse of terror’ she feels - a psychological impulse to act (Escape! She must escape!) quite as real and compelling as an encounter ‘with any forest-dwelling villain. Of course in this psychological story impulses are not pure and simple, but complex and clashing ~ a counter- impulse is to stay, keep her promise to her mother, and submit to a ‘lite of ommonplace sacrifices’. The final paragraphs are all about that clash of impulses, the ‘maze of distress’ that renders her helpless and inert, unable to respond in any way to Frank’s summons. ‘succumbing to the light, with Frank as the hher mother, We shall find the pattern of these contrary readings neatly highlighted by Greimas’ typology of character roles (which he calls |, to be discussed in Chapter 4. But now, and not for the last time, we may want to raise the question of, reductivism, Ts not Propp’s bold anatomy of fairytales a procedure which ly distorts, since it sets aside the important and necessary cultural ily sets out to shear off the ial) with regarcPee eee eee would come the reply! For just as the phonologist, in positing all the core and distinct sound units of a language, the phoneme: Counts all the phonetic variation which does not con: bearing, word-changing variation, so the story str he is identifying the basic narrative u guage of Russian fairytales, and in so ditions within which story me How tenable is the anzlogy? The strength of hard to stand out ive judgments about what are and are not English sounds, and so on. It’s less easy to see, harp boundary to any set of stories supposedly covered by a Proppian analysis — although the boundary \was sharp enough as far as Propp himself was concerned, since his corpus edly, we might ask why Because, Propp is siven corpus. Bi ask, are the grounds of this ‘need? ‘The thirty-one funetions identified (with neither duplication nor unjus- ied merging of types) are claimed, largely intuitively, to be the only fune- rure of the stories in the i question is whether such intuitionism is defensible, or whether descriptive apparatus is invalidated. But we need also to keep in st what the goals and expectations of a Propp or Barthes-based are. The getting at our intuitive judgments (rather than our public and conditioned ways of talking about plots and plot structures) will always present difficulties. But we do readily find ‘groups of readers (even whole communities) disclosing substantial agree- ment over what is essential and non-esse plot, characte so on ~ disclosing, in short, a comshon grasp of structure. This gen agreement and commonality of grasp are the essential justification inductive speculations of Propp, Barthes, and others, 2.3 Barthes on narrative Entirely appropriately, Barthes’ famous ‘Introduct lished in French in 1966) begins with an argument ductive ver deductive methods (in linguistics and narrative study), and defends latter, despite th Narrative analysis is condemned to a deductive procedure, obliged first to devise thetical model of description (what American lin- i ) and then gradually to work down from this model towards the different narrative species which at once conform to and depart from the model. (1977: 81) Basic story structure 23 gui ems reasonable’ as a founding model for bbut notes that discourse study will require a ‘secon guistics’ going beyond the sentence. But he does posit a homological rel tion between sentence and discourse, at least as far as semi ‘message-bearingness’ —is concerned: A discourse is @ long ‘sentence’ ... just as a sentence ... is a short ‘dis- course’. (Barthes, 1977: 83) More particularly, Barthes emphasizes the need to separate different levels of analysis, and the need for a hierarchical typology of units, in this early essay. He proposes three major levels of narrative structure: 1 functions (as in Propp, Bremond) 2 actions (by which he refers to ‘character cussed below — refers to them as actants "rather as Greimas ~ dis- to look for (and find) designed to support, su -and-efifect logic. Barthes proceeds to distinguish two types of functions: (a) functions proper (which we might call ‘Propp-type functions’); and (b) indices, Which are a unit referring not to a complementary and consequential act but to a more of less iffuse concept which is nevertheless necessary to the meaning of the stor} : (Barthes, 1977: 92) They include indices to charactets’ psychological states, notations of imosphere’, and so on. While functions proper are distributional, sequential, ‘completed’ further on in the story — and so have a kind of syn- tagmatic ratification, indices are said to be integrational, hierarchically- oriented, realized by relating them to some higher, integrated level, pacadigmatic ratification, On a broad continuum, Barthes suggests,24° Basic story structure functional narratives such as folktales, to be contrasted with heavily indicial ones such as psychological novels A further cut is now introduced, Functions proper are of two types: (@)L_ Cardinal functions or nuclei or, to use Chatman’s useful term, ‘kernels’ (in Chatman, 1969): these are real his of narra- "); they occur important consequences. Catalysers (not the best of terms): these fl in the naira between nuclei, and are described as parasitic and Barthes, areas of safety and rest, For example, a ringing or a delivered letter may herald a real nucleus in a story — and a pre- liminary ‘hinge’ would be whether the summons is answered or not, the letter opened — bu j (a2 I (o)2 relevance); Informants (depthless, transparent, identificatory data), Indices involve an get of deciphering, the reader is to learn to know a character or an atmosphere; informants bring ready-made knowledge... their functionality is weak. (Barthes, 1977: 96) Finally Barthes notes that a unit can be a member of more than one class. ime: one could be both a catalyser and an index, for example. And he tes that ina sense nucei (Kernels) ae the special group, with the other mit types he unl types being expesons of nice, Nuclei provide the necessary Barthes goes on to appeal for descriptive study not merely of the ‘major ive’ but of the organization of the smallest segm which he sees as combining into coherent sequences: A sequence is a logical succession of nuclei bound together by a rela- tion of solidarity: the sequence opens when one of its terms has no solidary antecedent and closes when another of its terms has no conse- quent. (197: 101) For example, ‘having a drink’ is suggested as a closed sequence with the following nuclei: order a drink, obtain it, drink it, pay for it. (But is paying as obligatory and integrated as the other three nuclei?) Now the business of seeing a sequence in such a string of reported events, and labelling it as having a drink’ (rather than, say, ‘quenching one’s thirst’ or ‘making ‘oneself socially available’) is, for Barthes, the kind of projective interpre- tive activity readers are always doing in their narrative processing, It's Basic story structure 25 he later subsumes under his proaieretie code (Barthes, 1970) jing is a key act of mental processing, under the assumption that the jer does aot remember everything they read, but remembers selec- portance. The reader registers, and threads, in a nari ‘not appear that Barthes drew on psycholinguistic evidence in his smptions and theorizing of ‘sequence naming’, but there are some resting parallels between his proposals here and more recent psy- linguistic research on narrative (see the brief mention in the “Further ” section of Chapter 7). critical demonstration of the Barthesian machinery Chatman (1969), which sets it to work on the same this chapter. Chatman identi- kernels in the story; I list these labellings appended: Ww, with Chatman’s interpret She sat at the window watching the evening invade the avenue. (SITTING AND LOOKING) ‘One time there used to be.a field there in which they used to play . (REMINISCING) Now she was going to go away ING THE DECISION TO G ‘Was that wise? (QUESTIONING) She stood among the swaying crowd in the station at the North W: (PREPARING TO EMBARK) ‘Out of a maze of distress, she prayed to God to direct her, to show her ‘what was her duty. (INDECISION CHANGING TO ANXIETY) She felt [Frank] seize her hand. (FRANK'S URGING HER TO GO) Nol No! No! (REFUSAL) . to leave her home. (REHEARS- Ath the required connected ized glosses of the kernels functions proper reflect the story's atte than physical change: the story is structured around reflection, reminis- ‘cence, thinking about doing something, and getting ready to do something, jer than on actions themselves. In addition, there is ample evidence and catalysers sccompanying the narrative development that is principally ‘driven’ by the functions proper, all contribute to an integrated presentation. Even the ‘smallest textual details, we might argue, play a role. Notice, for example, 1e words swaying and maze that are used in the course of functions S and these sentences, they are also f indeterminacy and vacilla~ , respect Many readers, casting proceed to extract dominating indices,Basic story structure 27 26 Basic story structure is ‘doer’, made the focus ct that John is the ‘underlying subject’, the ‘doer’, ma ough a clefting device. But if we read on and find John isin fact the aceriying object, the ‘done to’ ~ informants, and so on, as they build up a sense of the basic structure of the Dubliners narrative as a whole. Among the indices that many generalize from ‘Eveline’ are qualities of dependence, submission to duty, and inet fectual Despite the attractions of Barthes’ basic four-way categorization of narrative material, problems remain concerning the replicability of articular over how (by what criteria) we can is and is not a nucleus, a catalyser, an index, Proper and an informant. Some of these problems can be outlined by examining Chatman’s explanation of the difference between kernels (ie, nuclei) and catalysers. Kernels are said to be hinges, alternative path openings, and so on, while catalysers (better, ‘sat essential actions (‘business’) accompanying the kernels, b Prospective consequence. In the extract below, Chatman it alleged kernels: Tewas John who the boys attacked. all troubled by the need for revision. Amending the truism, You can begin to parse a sentence or text before you've finished ~ ‘teading it, but you know you may need to revise your analysis. rimedness test mentioned above is related to the gramm: mm between obligatory and deletable material, In_ narrative : while catalyser and indexical : ‘ i stitute a coherent ‘bare’ narra~ One of the telephones rang in the dark room. Bond turned and moved ‘ably related to the full version) ~ the discarded bits quickly to the central desk and the pool of light cast by the green eas ee tees aaie aa Stone, aang Hm. He picked up the back telephone trom the rank Poe upshot of these qualifeatons and reservations shouldbe that we of four. more clearly that Kemnels and catalyses are not so much textual givens’ as analytical constructions; and as Culler explains using a Saus- srean phrase, they are ‘relational terms only’ ‘What we might question here is the assumption that the phone-ringing and answering are inherently nuclei, that Bond’s moving across the room is inherently secondary. Such decisions can only be made retrospoctive! the light of a fuller scanning and assessment of adjacent text, But hen it did not guide our reading If structuralist analysis (in terms of Kernels or functions) isto be of value, we clearly need a robust explanation of the kind of interest or question that a kernel provokes, an explana that sets out the bases of our stronger and weaker impressions of kern work ‘longitudinally’, text's gra a broad sense of that of more local and grammatically-cued marks of the core ely, however, when we offer a determination of Utterances are, we have to operate holistically | ly. And this assumption is necessary in order to apply any is retrospective process of sense-making and plot-determination, 's truism is that you can’t parse a sentence until you've read it, Notice, however, that you can start to parse a sentence before you have ished res ~ I think we typically do. But we know that the is provisional, may not ‘go through’ if we find a configuration of ‘that is out of the ordinary, the unmarked form. We know, then, not to put too much trust in our parsing until the reading is complete, and we've'scen all the structure there is to see, Thus as we read: pu which the heroine récalls could be orgai satellites, but within the story they become sat a kernel such as ‘weighing the evidence’, ... We must a ‘recognize kernels only when we the role of an aston in the plot ot, to put it another way, promote an action to a constitue 2, One cannot determine the role or function of an action ‘without considering its consequences and its place in the story as a whole. (Culler, 1975b: 135-6) ally involved. Thus as we read through 7 8 model of scheme into which the disparate propes fons (it jndow, leaning her head, listening to the man's footsteps, Shout childhood days) can somehow fit. Much ofthis mode ing wl our cultural background, by what we think - guided by our cultural backgr thik = and Ttwas John who28 Basic story structure Both texts and readers are inescapably shaped or framed by prior (but not fixed or eternal) cultural assumptions as to what salient, It is because there is always this context of cultural and s says we have ‘a language of p particular story. .n outward sign of having made sense of the textual data is the pro- duction of reasonable paraphrase, ie. a parephrase that neither we nor ther readers find incongruous or absurd. The paraphrase could be very. | i, Many times the length of the original (as most literary critical art- les on ‘Eveline’ are), oF it could be as: A young woman reflects on her past or as: Reflection But (and here the analyst’s dilemma begins to look remarkably like that of | those who attempt to identify the structure of spoken discourse) does it, make sense to work ‘from the bottom up’, as Barthes and Chatman ¢ to be doing? Do we not have to work ‘from the top down’ up some broad hypothesis as to what happens i op-level constituents of story, ther bound together? Or bottom-up quite simple but fundamental this dramatic opposition of top-down and ’s objection to Barthes’ model is {It] remains strangely atomistic, through the lack of any specifica of what one is moving towards as one collects kernels and sat and groups them into sequences. thatman is able to out [the sentence One time there used to be a field there] as a ke only because he has some sense of an abstract structure towards which he is moving, .... What the reader is looking for in a plot is a passage from one state to another ~ a passage to which he can assign thematic valve. (Culler, 1975b: 138-9) Again, we are back to a linked before and after. The before and the after can be labelled, if we like, oppositionally as problem and solution, or logi ight simply label them as .ces ‘around’ all our involvements with narratives that Barthes § thin us even before we approach any -proposition length, ot just a | single word, The opening paragraphs of ‘Eveline’ could be paraphrased _ Basic story structure 29 4 Plot-summarizing: modelling intuitions Paper on ‘Defining narrative units, which focusses on problems of ability and adequacy in certain proposed models of plot analysis, ler concludes: | Competing theories of plot structure can only be evaluated by their cess in serving as models of a particular aspect of tence: readers’ together similar plots, etc, This int facts to be explained. (Caller, 1975b: 127) this characterization we should add a couple of caveats. The first is t-even more immediately apparent than in relation to the notion of lin- istic competence on is loosely based, literary competence seems be very largely learned rather than innate, and markedly culture- Thus when we talk of ‘readers’ abilities’ we have to keep clearly in nd that we are talking only of the acquired and developed ability of a soup of readers, rather than some universal mental ability, comparable to ‘near-universal ability to walk or subtract. The second related point is that what it means {o call these culture- itive knowledge’ remains unclear. People can be ing, and can get better at it: does this dgural approach and argue that we get good at plot-summarizing largel because, in our kind of world, constructing and communicating summar- plots is a valued skill By ‘our kind of world? I mean a world in which dren and adults frequently face examinations; in which we often have ‘our recall of events put to various kinds of test; in which it is common to sare ‘at second hand? a narrative whose performance one’s addressee has _aiissed; and in which there is particularly sharp awareness of the universal that infiuences the shape of so much of our behaviour, namely, js clear that people, with their various non. jons, can and do become adept at producing and understanding ‘ng the more important characters and events of all, they get good at identifying what, relative of world knowledge and cultural assumptions, is of a story. One of the questions we need to try to answer is "how on carth they do this. That they should do itis surely no great surprise: