Analysing a Poem From Sociolinguistic

Narrative Discourses and Discoursing in Narratives: Analyzing a Poem from a SociolinguisticPerspectiveAuthor(s): Timothy R. AustinSource: Poetics Today, Vol. 10, No. 4 (Winter, 1989), pp. 703-728Published by: Duke University PressStable URL: http://www.jstor.org/stable/1772807Accessed: 22/03/2010 07:09

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available athttp://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unlessyou have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and youmay use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained athttp://www.jstor.org/action/showPublisher?publisherCode=duke.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

Duke University Press is collaborating with JSTOR to digitize, preserve and extend access to Poetics Today.

http://www.jstor.org

http://www.jstor.org/stable/1772807?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp

http://www.jstor.org/action/showPublisher?publisherCode=duke

Narrative Discourses and Discoursing in Narratives: Analyzing a Poem

from a Sociolinguistic Perspective

Timothy R. Austin English and Linguistics, Loyola of Chicago

We will be using observation as a basis for theorizing. Thus we can start with things that are not currently imaginable, by showing that they happened. (Sacks 1984: 25)

1. Language, Sociology, and Literature In the past decade, linguistics has evolved as a discipline in any number of exciting directions. In the core theoretical subfields, scholars have developed multiple models of linguistic behavior, some comple- mentary, some in open competition with one another. In syntactic theory, for instance, a growing number of generative grammars (each sporting the almost obligatory identifying initials: GB, UCG, LFG, HPSG) make markedly different claims and predictions about linguistic data. At the same time, the range of natural languages subjected to relatively thorough syntactic analysis has also been significantly extended. But in an altogether different trend, those areas in which linguists' concerns overlap with those of neighboring disciplines such as neu- rology, artificial intelligence, and psychology have taken their own long strides towards independent respectability. These domains of interdisciplinary investigation offer particularly exciting prospects for future growth because of the opportunities they foster for the mul-

The impetus of this paper derives from courses and lectures that the author at- tended at the 1987 LSA Summer Institute at Stanford University. The author is particularly grateful to the instructors of two courses, Deborah Schiffrin and Herbert Clark, for their clarity of exposition and enthusiasm. Financial assistance toward the preparation of this material was generously provided by Loyola Uni- versity of Chicago.

Poetics Today 10:4 (Winter 1989). Copyright ? 1989 by The Porter Institute for Poetics and Semiotics. ccc 0333-5372/89/$2.50.

704 Poetics Today 10:4

tiple application of particular research findings. In the optimal case, that is, the discoveries of psycholinguists will inform, and may even influence, further work both by research psychologists and by theoretical linguists. By means of this process of cross-fertilization, each of the intersecting disciplines may gain access to concepts, methods, and

perspectives first developed by the other.

Anthropologists and sociolinguists were among the first to join con-

temporary linguists in such essentially collaborative work (see Gum-

perz 1982: ch. 2). Psycholinguistics has enjoyed a somewhat shorter

history (see Fodor, Bever, and Garrett 1974). In recent years, however, specialists in both of these areas have displayed a shared interest in what one might term a second-generation interdisciplinary subspecialization: discourse theory. Two characteristics mark linguistic theories of discourse as departures from the sentence-based com-

petence grammars of the 1960s and early 1970s. In the first place, they take as their domain language units that may be either smaller or, more likely, larger than the canonical sentence (Stubbs 1983: 1). In the second, they study "language in use" in a particular setting (Brown and Yule 1983: 1), seeking to explain its function as well as its form and, indeed, the interconnectedness of those two aspects of

language behavior. Given such a broad definition of the notion of discourse theory, of course, any number of technically quite disparate approaches satisfy it: speech-act theory and natural language semantics, for example, as well as much recent work in artificial intelligence. It has been socio- and psycholinguists, nevertheless, who have until now devoted the most sustained attention to discourse issues. Such scholars as William Labov, Erving Goffman, John Gumperz, Dell Hymes, and Herbert Clark have with some frequency employed discourse as an ideal forum in which to promote alliances among their native disci-

plines and linguistic science, alliances that may then serve the kind of

cross-fertilizing purpose that I have described above. Brief consideration of the grounds on which discourse theory ap-

peals to both psychologists and sociologists, however, surely suggests that literary scholars might also be attracted to this kind of study. Literary texts, after all, represent language in use, and they almost

invariably exceed the length of a single sentence. And it is certainly true that some theories of discourse phenomena have in recent years been invoked in individual stylistic analyses. Speech-act theory, for

example, informs both Stanley Fish's (1980: ch. 3) treatment of Corio- lanus and an excellent recent paper on W. H. Auden's "Song V" by Ronald Carter (1983). By the same token, striking similarities can be found between the theories of narratologists like Wayne Booth, Sey- mour Chatman, and Gerald Prince, on the one hand, and those of

psychologically based discourse theorists like Bertram Bruce, on the

Austin * Narrative Discourses and Discoursing in Narratives 705

other. All the same, one does not sense any dynamic mutual engage- ment of linguistic and literary interests in these studies. Where insights from one field are drawn on in the other, the direction of flow seems almost invariably to be from discourse theory into literary criticism rather than vice versa.

In part, this lack of interplay may be due to a certain snobbishness on the part of the literary community. Some blame, however, also ac- crues to discourse theorists, who repeatedly prioritize what Michael Agar (1985: 147) calls a "favorite type of discourse... 'natural conversation.'" To be fair, some authors who routinely exclude written, let alone literary, discourse from their work do take time out to deny that they do so because of "any theoretical primacy we accord conversation" (Schegloff and Sacks 1973: 289). At the opposite extreme, however, are repeated and, so far as I can tell, unargued assertions by other scholars that "conversation is a more basic, unmarked mode of communication than other communicative genres" (Schiffrin 1988: sec. 0). Such overt privileging of conversation seems to me unhelp- ful for reasons which this paper will demonstrate. For the moment, however, I mention it only as one possible reason for the very evident absence to date of fruitful interaction between discourse theory and literary studies.

In this paper, I explore some ground shared by linguists, sociologists, and literary theorists in a way which illustrates the complex interconnectedness of these three fields.' First, in sections 2 and 3, I use the techniques of one discourse-analysis approach, conversation analysis, to isolate and characterize an apparently anomalous discourse in a literary text, William Wordsworth's "Resolution and Independence." Next, in section 4, I argue that a second type of discourse theory, structural narratology, allows us to explain how it is that the conversational infelicities in this poem are generally overlooked by readers who encounter them in their literary context. Finally, in section 5, I suggest that evidence from this literary case may be reapplied by discourse theorists in nonliterary contexts to account for some otherwise baffling data from everyday conversation.

2. "Resolution and Independence" William Wordsworth wrote "Resolution and Independence" late in the spring of 1802. It constitutes one contribution to what may be seen as a protracted poetic debate between him and Samuel Taylor

1. Gumperz (1982: 15) notes one other rather isolated case of interaction among these three fields at the level of theory: the extension of Kenneth Pike's etic/ emic distinction from phonological theory into both anthropological and critical vocabularies.


Coleridge on the transitoriness of poetic inspiration, a debate which also encompasses "Tintern Abbey," "Dejection: An Ode," and "Ode: Intimations of Immortality." We shall have reason to return to this broader context for the poem later in our discussion. For the moment, I shall concentrate only on Wordsworth's narration at the core of the poem of a meeting between its first-person narrator and a leech

gatherer, a passage that occupies the final twelve stanzas of this twenty- stanza poem. We know something of the real-life circumstances on which Wordsworth based this part of the poem, since Dorothy Words- worth, the poet's sister, was with him when the meeting took place and recorded the facts succinctly in herJournal (de Selincourt 1944:

510-11). But the commonplace origins of this episode in an ostensibly objective report of an everyday occurrence that happened to involve some interpersonal social discourse may lead us to overlook some star-

tlingly atypical features of the dialogue at its center. Let us begin by segmenting Wordsworth's account of the meeting

informally into a series of steps familiar from our own experience of similar meetings:2

Orientation (52-77) The first-person narrator of the poem, out walking in a "lonely place" (52), meets and describes at length the figure of "a Man."

Greeting/Response (78-86) In lines which themselves represent a rather insightful commentary on the purely social function of conversational openings, the narrator

explains, "And now a stranger's privilege I took; / And, drawing to his

side, to him did say, / 'This morning gives us promise of a glorious day'" (82-84). As Clark and Marshall (1981: 56) point out, remarks about the weather are indeed the "stranger's privilege," since the current state of the weather "is mutually identifiable by people in the same locale" and hence an impeccable source of that basis of "mutual

knowledge" on which almost all conversations have to be constructed. The old man's response is, evidently, equally formulaic, since the narrator fails to report it in detail: "A gentle answer did the old Man make" (85).

First Question-Answer Pair (87-105) At this point, the narrator makes the first substantive conversational

move, though again in a highly predictable direction. Since he has as yet no sense of his interlocutor's "community membership," on which basis a suitable topic for talk may be selected (ibid.: 35; Gum-

perz 1982: 142), the narrator selects one possible community type (profession or occupation) and seeks to discover which of its many

2. References are to lines in the edition of de Selincourt (1944: 239-40).


members the old man belongs to: "'What occupation do you there

pursue?'" (88). Before we hear the old man's reply, Wordsworth's narrator once

again supplies us with an apparently extraneous but useful and ex- tremely realistic nonlinguistic detail: "Ere he replied, a flash of mild surprise / Broke from the sable orbs of his yet-vivid eyes" (90-91). A number of discourse theorists have noted that conversations constitute creations to which speaker and listener alike contribute vital material of both verbal and nonverbal kinds (Merritt 1976: 317; Clark and Schaefer 1987: 19). Yet we have relatively few opportunities to

explore the nonverbal dimension of discourse structure because, as Susan Philips (1976: 83) laments, tape recordings, the most common source of analysts' data, "do not capture the listener's contribution to the regulation of interaction." (Students of literary discourse must of course struggle with a still more impoverished corpus; what one

might call "standard" literary reports of spoken dialogue lack even the limited information that tape recordings can supply about pauses, most details of intonation, and speed of delivery.) Wordsworth's narrator, therefore, is unusually helpful in alerting us to the fact that the old man shows heightened interest at this point, where the ban- ter of everyday conversational pleasantries gives way to signs of an impending more sustained and (referentially as opposed to socially) meaningful conversation.

It should not surprise us, then, that the old man replies at some length to the narrator's simple query. Rather than merely name his occupation, he both describes and, to some extent, even evaluates it:

He told, that to these waters he had come To gather leeches, being old and poor: Employment hazardous and wearisome.

And in this way he gained an honest maintenance. (99-105)

Commentary (106-17) Up until this point, as I have suggested by sporadically citing the work of discourse theorists who discuss standard conversational techniques, Wordsworth's narrator has reported a remarkably stock instance of what we might term a meeting-between-strangers speech event (Mer- ritt 1976: 318). It is only at line 106 of the text, in fact, that matters take a somewhat less conventional turn. The narrator reports, "The old Man still stood talking by my side; / But now his voice to me was like a stream / Scarce heard; nor word from word could I divide" (106-8). Harvey Sacks, Emanuel Schegloff, and Gail Jefferson (1974: 727) have pointed out that the "turn-taking system" that underpins all conversational behavior "builds in an intrinsic motivation for listen-


ing to all utterances in a conversation, independent of other possible motivations, such as interest and politeness." Where there are only two parties to a dialogue, the likelihood that the current listener will imminently be called upon (or will himself or herself choose) to speak enhances that motivation.3 The narrator's self-reported inability to separate one word from another entails, of course, that he will fail to glean the information that he originally requested in his posing of the question "What occupation do you there pursue?" but also, more seriously, that he will still lack any common ground on which to pursue the conversation. To that extent, we would predict, the conversation either will have to undergo major "repair" (Clark and Schaefer 1987: 22-23) or will simply go nowhere, since, as Stephen Levinson (1979: 370) argues, if the constraints that govern verbal activity of a particular type, such as questioning and answering, are not met, the activity breaks down.

Second Question-Answer Pair (118-26) What immediately follows in the text confirms our worst fears: "My question eagerly did I renew, /'How is it that you live, and what is it you do?'" (118-19). Anyone reviewing this conversation, regardless of theoretical viewpoint (philosophical, sociological, anthropological, linguistic), must surely recognize that with this development matters have gone seriously awry. Following the lead of John Searle (1969), scholars like Marilyn Merritt (1976: 347) have insisted that, in an unmarked discourse context, "a query is appropriate ... only if the answer is not readily available." What we see here is a particularly egregious violation of that principle, since the leech gatherer himself is the source of the information which, since he has just supplied it in the previous turn at talk, he has every reason to suppose the narrator already knows.

William Labov and David Fanshel (1977: 95), although they do not discuss precisely the conversational anomaly that we encounter in this

poem, do consider the way in which "repeated requests" of other kinds

generate severe social tensions between interlocutors. One common practice, they allege, is "for speakers to mitigate their repetitions." Some examples of mitigatory utterances that forewarn the other inter-

3. It might be objected that, historically, there were not two but three parties to this conversation: William, the leech gatherer, and Dorothy. I would respond that no mention is made of a third party in the text of this poem, which, as a narrative, represents a wholly autonomous fictional event. That argument then dovetails with another to be made later in the paper, to the effect that it would be equally wrong-headed to identify Wordsworth himself with the first-person narrator of "Resolution and Independence."


locutor that he or she may have furnished inadequate information are listed by Clark and Schaefer (1987: 26): "I didn't hear you"; "Pardon me?" Yet the narrator in Wordsworth's poem offers the old man no such verbal flag of truce, and we would surely anticipate as a result that the discourse should at this point dissolve, the leech gatherer re-

tiring in what Goffman (1967: 23) rather charmingly calls "a visible huff" (cf. Gibbs 1986; Gumperz 1982: 132; Tannen 1984).

Returning to Wordsworth's poem, then, we are surprised to find that the narrator's gauche repetition of his question does not earn him

any rebuff at all. Quite the contrary; the narrator blandly reports, "He with a smile did then his words repeat" (120). And indeed the narrator himself obligingly repeats them, indicating, by his use of direct speech on this occasion, his heightened alertness to what the leech gatherer is telling him.

Brief Commentary, Further Dialogue, and Coda (127-40) With the end of the old man's second reply (126), the narrator loses almost all interest in reporting the balance of the conversation, wind-

ing up the entire narrative in the space of fourteen lines. He dismisses the remaining dialogue with the remark "Soon with this he other matter blended" (134) and slides quickly into the aphoristic close that has caused considerable critical comment: "'God,' said I, 'be my help and stay secure; / I'll think of the Leech-gatherer on the lonely moor!'" (139-40).

From the point of view of conversational analysis, then, we are left with one very puzzling feature: the inclusion, in an otherwise remarkably realistic account of a dialogue, of a complete "echo" question- and-answer pair. We can look at this strange phenomenon from the viewpoint of either of the two interlocutors. On the one hand, why does the inattentive narrator not hedge ("mitigate") his socially unac- ceptable reiteration of a simple, previously fulfilled request for information? On the other, why does the leech gatherer not take umbrage, as we would expect him to? Why is he so compliant in the face of clearly improper conversational protocol?

Before proceeding to a fuller discussion of these questions, it is interesting to note that Lewis Carroll, a reader whose ear for the lin- guistically offbeat seldom let him down, detected and exploited precisely the conversational infelicity I have been isolating when parody- ing "Resolution and Independence" in Through the Looking-Glass. The first-person narrator of "The White Knight's Song," Carroll's bur- lesque version of the poem we have been examining, resembles the Wordsworthian figure on whom he was modeled in his marked insensi- tivity to his "aged" interlocutor. In Carroll's poem, however, rudeness is allowed to run amok:


"Who are you, aged man?" I said. "And how is it you live?"

And his answer trickled through my head Like water through a sieve.

So, having no reply to give To what the old man said,

I cried, "Come tell me how you live!" And thumped him on the head.

I shook him well from side to side Until his face was blue;

"Come tell me how you live," I cried, "And what it is you do."

(5-8, 21-24, 37-40) To Carroll's Alice, as to most of his Victorian readers, this conversation, like the Wordsworthian dialogue it parodies, was either absurd or cruel, perhaps both. It is therefore by no means a trivial question whether we can establish that "Resolution and Independence" applies its unconventional conversational forms to some legitimate artistic end, whereas "The White Knight's Song" (designedly, of course) does not.

3. Explanations In seeking an explanation for the puzzle described in the previous section, we might first consider whether discourse theorists who spe- cialize in conversational analysis have noticed any naturally occurring circumstances in which either a request by a speaker for information

presumably available to him or her already or, the more marked case, repetition by one speaker of the same request in successive turns at talk ever constitutes acceptable conversational practice. As it happens, the literature does offer a number of such examples. One subclass of these cases derives from Levinson's (1979: 367-68) observation that lan-

guage use is situated in specific contexts, all of which are also "activity types,... goal-defined, socially constituted, bounded, events with constraints on participants, setting, and so on." (As "paradigm examples" Levinson offers "teaching, ajob interview, ajural interrogation, a foot- ball game.") Participants in each activity type fill certain roles that are

broadly defined by the goals they are pursuing within this context. Most important for our purposes, however, is Levinson's (ibid.) additional remark that each activity type also imposes "constraints ... on the kinds of allowable contributions" that speakers can make to the

ongoing dialogue. Levinson's theory of activity types (which has approximate ana-

logues in Charles Ferguson's [1983] "genres" and Goffman's [1974] "frames") allows him to explain a number of everyday contexts in

Austin * Narrative Discourses and Discoursing in Narratives 71 1

which speakers routinely request information they already possess without endangering the talk in which they are engaged. Legal counsel in a courtroom, he explains, is constrained by accepted defini- tions of what constitutes "a prosecution" (or "a defense") to ask questions which "request details that are already known to the questioner" (Levinson 1979: 380). Posing these questions serves no purpose what- soever if we assume that the only discourse function of a question is to make new information available to the questioner. But reference to counsel's goals in this particular activity reveals that such seemingly anomalous utterances serve two ends. First, though Levinson himself does not point this out, they publicize or institutionalize the information contained in the reply by inscribing it on the official transcript of the proceedings. More intriguingly, they permit counsel to order the flow of information, juxtaposing individual replies in such a way as to invite the judge or jury to infer causal or logical connections. They enable counsel, in fact, to "build a case" (ibid.: 381). Indeed, the witness who treats the questions of a skillful cross-examination merely as requests for information, to be answered truthfully one at a time, may find himself or herself condemned by an implicit, unsuspected argument.

What other activity types permit the posing of questions to which the speaker already knows the answer? Teachers certainly use this technique to induce their pupils to display knowledge that they are supposed to have acquired (ibid.: 384-90; Labov and Fanshel 1977: 89-90). Parents characteristically grant young children an opportu- nity gracefully to accept responsibility for some presumed misdeed: "George, did you cut down that tree in the yard?" And in so-called So- cratic arguments, speakers may begin a complex series of interrelated questions with one or more interlocutors, whose answers they expect to be uncontroversial, so as to establish a firm basis for the increasingly disputable propositions that will follow (ibid.: 102-3).

Granted that particular activity types may lift the most neutral re- quirement that speakers must not request information already known to them, such an approach still will not help us to address the self- reported behavior of Wordsworth's narrator. Deliberately set in the unremarkable context of a casual encounter between strangers, this conversation serves no highly specialized social purpose, nor can we assign any specific goals to its participants. If anything, we might feel tempted to invoke a distinction noted by Deborah Schiffrin (1984b: 315) between the "referential" function of talk and its purely "sociable" use. Schiffrin (ibid.) suggests that even though all talk is naturally used at some level "to convey referential information," in certain cases "the meaning of that information [may become] subordinated by the meaning of the talk itself." Suppose, then, that we locate Wordsworth's


scenario at the extreme opposite end of the spectrum from Levinson's specialized "activity types." Might not requesting known information, under the right circumstances, constitute a purely sociable use of lan-

guage? As we have seen, Wordsworth's narrator and the old man certainly indulge in sociable "small talk" (Goffman 1979: 2) when they first meet. Brief reflection, however, soon leads us to reject this ex-

planation as well. Even at cocktail parties, overt demonstrations of a failure to attend to one's collocutor (such as asking the same question twice) are not tolerated; they appear either inept or condescending.

If conversational analysis cannot provide a suitable contextual frame within which to set Wordsworth's narrator's apparent pursuit of knowl-

edge already available to him, perhaps we should reexamine our (thus far implicit) assumption that the leech gatherer has ever answered the

question posed to him in the first place. After all, one reason for repeating one's utterance is that the first has failed to elicit the desired response (for a question, has failed to elicit the information sought). Clark and Schaefer (1987: 37) consider a rather specialized case in their analysis of telephone calls made to directory assistance operators. These operators are trained specifically to request information from callers in a particular (functionally determined) order. If their standard opening question ("For which town, please?") is answered in-

appropriately ("Could you give me the number of Mr. E. Michaels?"), the operators repeat their first question without comment ("In which town?").

Less contextually bound are cases discussed by Merritt (1976: 330- 33) under the heading of "calls for replay." In a tantalizing footnote, Merritt observes that a questioner "may decide to replay a question" because he feels "that the focus of the question was not properly in-

terpreted" (ibid.: 332). In the case we have been considering, this translates into the suggestion that the old man's reply to the narrator's first query, though lengthy, does not actually address the narrator's

request; hence the second attempt to isolate that crucial information. Such an analysis in this particular instance becomes more plausible after the additional observation that the narrator's question falls into the rather narrow class of "Who is/are X?" interrogatories. Asking who somebody is, it has been pointed out (Boer and Lycan 1975), can be undertaken only with reference to some specific discourse context; the question "Who is that woman in the red dress?" may be answered with any of the following sentences, even if in practical terms they all identify precisely the same individual: "That's Angelica Norburg"; "That's Bill's wife"; "That's the new vice president for marketing"; "That's our hostess." In any given conversational setting, just one of these replies will be particularly apposite; the others will miss the

point, prompting the response "Yes, I know that, but who is she?"

(ibid: 300). Perhaps, then, the narrator's repetition of his question


should be viewed as an attempt to imply that the old man has misiden- tified the discourse context in which the query is framed.

It is one of the advantages of analyzing conversation that the (un-) acceptability of individual parts of a discourse need not be alleged as the expert opinion of "professional analysts." Instead, discourse- internal evidence may be adduced for the success or failure of each contribution. In the case of failure, in particular, "the parties themselves address the talk as revealing a misunderstanding in need of repair" (Schegloff 1987: 204). Indeed, upon close examination, each contribution to a discourse can be shown to reveal (to the interlocutors themselves as well as to the analyst) how prior contributions have been understood by the participants (see Clark and Schaefer 1987). In our

particular case, the availability of such rich evidence puts a quick stop to our emerging analysis of the narrator's reiterated question as mere

refocusing of his initial query. Not only does the narrator himself refer to his second question as an unelaborated "renewal" but, as we have observed, it prompts only a repetition of the old man's story, not any adaptation of it to some previously unnoticed discourse context. Reluctantly, then, we must abandon our explanation of this dialogue as "replay as recontextualization."

Why else might a replay be requested by a participant in a conversation? Alas, we are now driven back to the simplest but least attractive of all explanations: the possibility that the narrator's question reveals simply that he has altogether failed to process the leech gatherer's first response. His repetition thus constitutes the functional equivalent of the conversational "What?" or "What did you say?" (Merritt 1976: 332 n. 29 and sources cited there). This hypothesis fails, of course, to explain the old man's remarkable good humor in the face of the narrator's rudeness, and it brings us uncomfortably close to Lewis Carroll's view of the poem. All the same, it does seem to be the only explanation compatible with the findings of the conversational analysts to date.

In an everyday context, of course, this conclusion would not be so surprising; misunderstandings and even social gaffes commonly occur in discourse, and repairs are not always made verbally. What then disturbs us so much when it occurs in a poetic context? The root of our unease lies, I suspect, in the fact that this conversation is part of (indeed, represents the bulk of) a narrative, a discourse form which imposes its own constraints and its own expectations. It may repay us, therefore, to review this puzzling passage from this second discourse perspective.

4. The Narrative Dimension Most sociologists who have investigated narrative discourses, and they are many, have focused their attention on oral narratives (Labov 1972; Schiffrin 1981, 1984a; Polanyi 1985). A strong initial hypothesis, how-


ever, would surely be that what holds for oral narratives should also be true in written, and even in consciously literary, texts. We might further suppose that our hypothesis should hold most firmly for those analytical findings about discourse structure that seem most theoreti-

cally fundamental and uncontroversial, and for those literary texts that most clearly simulate first-person oral narration. In particular, it seems improbable that a literary narrative such as "Resolution and In-

dependence" should evade that most basic of all narrative imperatives: that it "make a point" (Polanyi 1985: 187).

"A story," Bruce (1981: 278) observes, "is told by someone to someone with some purpose." Indeed, it has been repeatedly pointed out in the literature that each story we tell or hear is carefully designed and delivered so as to forestall that most damning of all criticisms of narrative technique: "So what?" (Labov and Fanshel 1977: 105, 108).4 Furthermore, since all narratives consist at their most elementary level of a series of reported events (Labov 1972: 360), it must be in those events that the integrity of a given narrative as a "tellable" story finally resides. "No event in and of itself," to borrow Livia Polanyi's (1985: 196) summary, "is important-it is significant only in some context.... 'A man got murdered sometime' is not, in itself, a particularly tellable

story." The reader who wishes to understand the dynamics of Words- worth's account of his meeting with the leech gatherer as narrative, then, is confronted with two tasks: first, to determine what constitutes "the events" in this particular story; second, to relate those events to some overall narrative purpose that they may be seen collectively to serve. Let us take those tasks one at a time.

I shall not rehearse here the characteristics of "event clauses" in En-

glish narratives; Polanyi (ibid.) summarizes them effectively and cites relevant discussions from the literature. In any case, it seems to me

4. Livia Polanyi (1985) gives formal status to the issue of the purposefulness of narratives by explicitly distinguishing a story, "a recital of events and circumstances

[that] must ... be told to communicate some message about the world in which the speaker and hearer actually live," from a narrative, which may be used without such a context. The latter case, she argues, results in hearers perceiving that the

speaker has "abused his access to the floor, .... adding nothing substantive to what was being said" (ibid.: 189). A still stronger claim is made by William Brewer and Edward Lichtenstein (1981: 367), who would reserve the term story for "narrative structures organized so as to produce surprise and resolution."

The issue of the degree and quality, so to speak, of the point a narrative makes is

by no means trivial, even in a literary context (Polanyi 1985: 197); contemporaries criticized Wordsworth and Coleridge's Lyrical Ballads both for sensationalism (ex- cessive "surprise and resolution") and for pointlessness. One has also to bear in

mind, of course, derivative or self-referential narratives, such as Carroll's parody, in which the point is precisely to comment on how some other story has made its

point. Despite these complications, however, a more generic approach at this stage in our discussion will not seriously misrepresent the majority of cases.

Austin ? Narrative Discourses and Discoursing in Narratives 715

Table 1. Narrative events in the narrator's story

O I saw a Man (55)

he the pond / Stirred (78-79)

D a stranger's privilege I took / And ... to him did say, ...

(82-83)

D A gentle answer did the old Man make (85)

D And him ... I thus bespake, . . . (87)

a flash of mild surprise / Broke from... his yet-vivid eyes

(90-91)

D he told that... (99)

My former thoughts returned (113)

D My question eagerly did I renew (118) D He ... did then his words repeat / And said that ...

(120-21) D He . . . the same discourse renewed (133) D with this he other matter blended (134)

C "God," said I, "be my help .. ." (139-40)

that one can quite reasonably appeal to our natural sense of narrative intuition to distinguish clauses that move the narrative forward in what Polanyi calls the "story-world" from those that merely relate states of affairs, habitual actions, evaluative commentary, and so on. I have re- lied on just such a naive notion to draw up the array of event clauses from stanzas 8 to 20 of "Resolution and Independence" that appears as Table 1. The first and most obvious remark to make about this list is that it contains an overwhelming preponderance of conversational events (many of them, indeed, realized as conventional conversational "tags," such as "he told" and "I said"). The first event clause listed in Table 1 need not concern us at this point; I have labeled it O to indicate its status as part of the narrator's fairly extensive orientation section, to which we shall return below. The clause in lines 139-40, by the same token, constitutes the coda (C) to the narrative and will also attract our attention in due course. But, as the D (for discourse) prefixed to the great majority of the remaining clauses in Table 1 is meant to indicate, the principal impetus propelling this narrative is in


fact nothing more exciting than the inherently sequential, turn-taking nature of human talk.

In what, then, might the substance, the "point," of this peculiarly "uneventful" narrative consist? Two immediate possibilities present themselves. On the one hand, Wordsworth might expect the reader to infer something from the structure of the conversational exchange itself; the point could be the very way in which that talk evolves. I shall return to this alternative shortly. On the other hand, the poet could be using the conversational frame transparently, intending us to draw meaning from the substance of the talk itself, from the views that are expressed by the participants. Like the conversationalist who reports hearsay ("I met Mr. Jones in the bar last night and he told me that.. ."), Wordsworth could be using dialogue merely as a convenient channel through which to convey some opinion of his own, employing the narrator and the leech gatherer as mouthpieces for ideas he himself wishes to express.

To see how this option contrasts with the first, it may help to por- tray the narrative structure of this poem diagrammatically, adapting a scheme developed by Bruce (1981: 287; cf. Chatman 1978; Chafe 1980: 36), which allows us to view the entire episode as a story within a story. As Figure 1 illustrates, the old man "narrates" his autobiog- raphy to the narrator, "I," who in turn relays it to us, but at a quite different level or layer of narrative structure. If we suppose that the focus of this narrative lies at level 1, in the words spoken by the old man to the narrator, we shall need to examine the substance of those remarks themselves with some care.

The leech gatherer's "story" tells of him pursuing without complaint a thankless, difficult, and depressing occupation in thoroughly un-

pleasant surroundings even as circumstances beyond his control make success ever harder for him to achieve: "'Once I could meet with them on every side; / But they have dwindled long by slow decay'" (124- 25). The point would seem to have to do with perseverance in the face of adversity, and that indeed is the lesson the narrator himself claims to have learned when he tacks on his own summation of the tale: "I could have laughed myself to scorn to find / In that decrepit Man so firm a mind" (137-38). Polanyi (1985: 193) suggests at one point that a "story can be justified as worth telling . .. because of the significant alterations in the world resulting from the events around which the

story was built." If this is so, then the narrator's closing remarks would seem to have been custom-designed to pick out just this "moral" as the level 1 point of the narrative.

A moment's thought, however, reveals that such an account of this narrative helps not one jot to resolve our original problem with the violations of conversational norms described in sections 2 and 3. Since


Narrator - Reader

Leech gatherer -- Narrator \

Figure 1. Narrative layers in the narrator's story

the leech gatherer repeats his words a second time, apparently ver- batim, the level 1 point we have just posited would be just as effectively presented if Wordsworth made it only once. Indeed, under this analysis, the repetitiveness of the episode's dialogue appears to be a problem not only in terms of conversational realism but also in terms of the narrative strength of the poem, since it artificially prolongs, or even distends, the slightly self-righteous moralizing of the old man.

Let us move, then, to reconsider the text from the perspective of level 0, and begin with the question of the storyteller's motive for em- barking on his narrative in the first place. At level 1, after all, the leech gatherer's use of autobiographical narrative occurs in response to the discourse prompt of the narrator's first question: "What occupation do you there pursue?" His reply actually exceeds what is strictly necessary under the circumstances,5 but because he is answering a direct request for information, the old man is under no obligation to establish the contextual relevancy of what he says or to secure a conversational turn in which to place it. The narrator at level 0, by contrast, resembles more closely the typical speaker characterized by Labov (1972: 366), who needs to indicate early in his or her narrative "its raison d'etre: why it [will be] told, and what the narrator is getting at" (see also Chafe 1980: 41-42). Narrators employ various means to achieve this end. In particular, they may use an abstract to preview their story or insert contextualizing material into the orientation section, which characteristically precedes the first event clause in a narrative, perhaps establishing characters or settings that will be inherently interesting to their auditors. In many instances, therefore, the opening section of a narrative will furnish important clues as to its point when that issue has not been settled a priori by the immediate discourse context in which it occurs.

Wordsworth's narrator offers no clear abstract for his encounter with the leech gatherer, but his orientation section occupies most of

5. This in itself is a slightly risky conversational "move," as Goffman (1967: 36) notes.


the first eleven stanzas of the poem. In these lines, the narrator describes the pleasant pastoral setting for the beginning of his country walk ("on the moors / The hare is running races in her mirth" [10-11]); he reconstructs his own mental state immediately before the meeting with the old man ("I thought of Chatterton" [43]; "When I with these untoward thoughts had striven" [53]); and he sketches the bleaker scenery amidst which he will later find the leech gatherer ("Beside a

pool bare to the eye of heaven" [54]). How does this extensive and detailed narrative orientation help us to understand the "complicating action" that follows it?

The mention of Chatterton may offer a starting point; his inclusion, after all, seems at first quite extraneous to the events of the leech

gatherer episode, although our expectations of coherence between the orientation and complication sections of narratives lead us to assume, and to seek, some measure of relevance. Of Chatterton, the narrator stresses three things: his youth ("marvellous Boy"6), his animated nature ("sleepless Soul"), and his untimely death ("that perished in his pride"). This brief portrait contrasts strikingly with the impression created by the leech gatherer when the narrator first observes him several stanzas later. "The oldest man ... that ever wore grey hairs" is described only as "not all alive nor dead, / Nor all asleep" (my italics), barely more animate than the drab surroundings in which he lives, "motionless as a cloud." Indeed, the narrator thereafter takes great care to reiterate and even reinforce these characteristics of his interlocutor. The only two event clauses marked in Table 1 as not denoting conversational exchanges refer to the narrator's startled recognition first of the old man's animation and then of his humanness: "He ... stirred" and "a flash of . . . surprise / Broke from . . . his yet-vivid eyes."

This dominant physical contrast between Chatterton and the old leech gatherer surely leads us to anticipate contrast at a second level: As Chatterton was a gifted poet, the logic goes, so this old fellow, who is in other respects his opposite, should turn out to be virtually mute, as paralyzed verbally as he seems to be in other, physical re-

6. Wordsworth's use of the term boy here is more heavily weighted than might at first appear. I discuss below the relationship between "Resolution and Indepen- dence" and "Ode: Intimations of Immortality," which Wordsworth had composed a few years before. Of interest to the present discussion is that boy is one of a number of words that appear in both texts, so that its appearance here necessarily evokes the fuller context of its use in the earlier work. The central section of the

ode, strophe 5, describes the progress of man from birth (58) through boyhood (68), youth (71), and manhood (75). Applying this scale to Chatterton, we find that

he, as a "Boy," "beholds the light" of poetic inspiration but is never subject to the "shades of the prison-house" that "close" around the maturing adult.


spects. It follows that the reader will be as surprised as the narrator to find in the old man an enthusiastic and even eloquent speaker. From the moment when the leech gatherer first responds to the narrator's formulaic greeting, his manner of speech is given particular emphasis:

A gentle answer did the old Man make, In courteous speech which forth he slowly drew.

His words came feebly, .... But each in solemn order followed each, With something of a lofty utterance drest- Choice word and measured phrase, above the reach Of ordinary men; a stately speech;

Cheerfully uttered, with demeanour kind, But stately in the main.

(85-136) And at the end of their meeting, we share the narrator's "troubled" reaction to the discomforting triple juxtaposition of "the lonely place," "the old Man's shape," and, crucially, his "speech" (127-28). For the leech gatherer's eloquence stands as proof that one need not die young, like the unfortunate Chatterton, to continue to command great power over the language. Both the old man's pride and, still more important, his capacity to articulate that pride remain only slightly dimmed by age and by "the ways of men, so vain and melancholy" that had earlier loomed as such daunting obstacles on the narrator's own horizon.

It is in this context, I believe, that we can finally begin to discern the source of the strangeness that characterizes the conversation at the core of this text. For one component of the point of the narrative told at level 0 in "Resolution and Independence" is that the old man, despite his long and wearisome life, can speak as he does. In the context of forebodings about the loss of inspiration and poetic insight that accompany maturity, forebodings that had been troubling the narrator before the meeting, the old man's garrulousness is note- worthy regardless of the content expressed by the leech gatherer at level 1. Indeed, so remarkable is it that, rather like a child with a talking teddy bear, the narrator excitedly makes the old man "perform" a second time. Since it is that he speaks and how he speaks rather than what he says that most matter to the narrator, the fact that the content of the answer is already known to him becomes irrelevant and the usual rules of conversational good behavior are, from his admittedly unorthodox perspective, moot.

In the preceding discussion, I have used the notion of narrative levels to tease apart two possible points to the story of the encounter


with the leech gatherer. At level 1, the leech gatherer consciously ad- vocates perseverance in the face of adversity; at level 0 he unknowingly provides the skeptical narrator with living, breathing proof that old

age need not render one mute. Furthermore, I have suggested that my level 0 analysis accounts for one of the more baffling aspects of the dialogue that passes between the narrator and the old man: the reiterated question. We are, however, not yet out of the woods, for we have not yet explained the old man's good humor despite his being treated like a clockwork toy by his nosy and inattentive interlocutor. Conversational analysis, as the preceding sections of this essay have shown, would predict that this essentially nonserious use of language should anger a participant like the leech gatherer, who after all is not

privy to the narrator's motives at level 0 and thus should expect his contributions to be treated in the usual, "serious" way. To understand the point of the old man's mysterious smile and even temper, then, we shall have to look elsewhere. We begin by noting that Figure 1 does not, in fact, give a comprehensive picture of the narrative form of this

poem. The author of "Resolution and Independence" did not in the first

instance intend it for commercial publication. Rather, as I observed at the beginning of section 2, the poem occupies a special place in the remarkable decade-long exchange of poems between Wordsworth and Coleridge that took them from their extraordinarily productive collaboration between 1797 and 1800 to the beginnings of the rift that would part them completely in 1810 (see Prickett 1970: ch. 6; Margo- liouth 1953). One may dispute where exactly to fix both the first and last contributions to what Stephen Prickett (1970: 167) repeatedly and

insightfully refers to as this "poetic dialogue"; it may begin as early as

July 1798 with Wordsworth's composition of "Tintern Abbey," and its echoes may still be detected in Coleridge's lines to Wordsworth after his reading the so-called 1805 Prelude in 1806-7. But the conceptual core of this dialogue undoubtedly consists of three great statements about poetic inspiration: Wordsworth's "Ode: Intimations of Immor-

tality," Coleridge's "Dejection: An Ode," and finally "Resolution and

Independence" itself. Dating the three poems is complicated by the fact that all three

underwent considerable revision. The "Immortality Ode" consists of two sections known to have been written at different times, and most of "Dejection" originated, though not in print, as "Letter to Asra," a work with a quite different addressee and purpose. What matters for our concerns is that, thematically, these three poems do represent essentially ordered contributions to the two men's debate over the eva- nescence of poetic insight. Thus the opening lines of the "Immortality Ode," "There was a time when meadow, grove, and stream .. .," are


Wordsworth -o- Coleridge \

Narrator Reader \

Leech gatherer -- Narrator

Figure 2. Narrative layers in Wordsworth's poem to Coleridge

echoed but then turned to a more somber end by Coleridge in "De- jection": "There was a time when, though my path was rough ..." Similarly, while "Dejection" is set before a storm ("This night, so tran- quil now, will not go hence / Unroused by winds"), Wordsworth deliberately describes a storm's-one might almost say the same storm's -aftermath in the opening lines of "Resolution and Independence": "There was a roaring in the wind all night; / The rain came heavily and fell in floods; / But now the sun is rising calm and bright" (1-3). And even if such incidental verbal cross-referencing were not enough to establish these poems' interrelatedness (see also note 6), we would still have the most obvious connection of all. The narrator of "Reso- lution and Independence," in describing his mood at the beginning of the poem, alludes to the title of the ode by Coleridge, which had provided the context for this reply: "As high as we have mounted in delight / In our dejection do we sink as low" (24-25; my italics).

This interlacing of phrases from one poem in the series into the language of the next seems to represent the literary counterpart of a process described for natural discourses by Schiffrin (1987: 17): "By repeating key phrases from prior conversation ... [speakers] use a cohesive device to show that understanding the interactional meaning of the story requires reference to prior conversation." If this is so, however, it also suggests that we need to stipulate a more complex narrative structure for "Resolution and Independence" than we have worked with thus far. I propose the overall form shown in Figure 2, in which I have modified Bruce's numbering system to allow posi- tive integers to denote narrative layers immediately evident within the text of the poem, while layers carrying negative indices can only be inferred by the reader.

We may note in passing that independent considerations confirm the necessity of the additional narrative layer shown in Figure 2 as level -1. For although, as we noted in section 2, the narrator claims


as a part of his level 0 narration not to be able to follow the old man's first reply ("nor word from word could I divide"), the text that we read still contains a meticulous report of its substance. Readers accept here, I suggest, the tacit intrusion of the poet himself at level -1 into the telling of the level 0 story.7

The relationship between level 0 and level -1 in this poem is an intriguingly indirect one. Whereas level + 1 fits into level 0 by means of standard conversational practice, as we noted above, a more conscious artifice on Wordsworth's part relates the two "higher" levels. The crucial clue to the nature of that relation lies in Oswald Doughty's (1981: 205) observation that "a portrait of Coleridge at this time, a slightly critical one, appears in 'The Leech-Gatherer'-now better known as 'Resolution and Independence.'" Doughty points out the many bio- graphical parallels that link the description of the narrator early in this poem, who questions why "others should / Build for him, sow for him, and at his call / Love him, who for himself will take no heed at all" (40-42), with Coleridge's own cloying dependency on William and Dorothy in 1801-2. Coleridge, a hypochondriac, unhappily married and already addicted to opium, leaned heavily on his friends during these years while still suffering frequent pangs of guilt over his own lack of productivity. Thus it is Coleridge's voice, not Wordsworth's, that we hear echoing the strains of "Dejection" at the beginning of "Resolution and Independence." Wordsworth daringly takes up the story of Coleridge's stormy night exactly where it left off, extending it into the following day without shifting from the first-person narration of the original into the third-person framework that we would

naturally expect. The ostensibly first-person I is categorically not the

poet himself-another justification, if we needed one, for the concept of narrative layering.

But if it is Coleridge who is characterized as the narrator in this

poem, then where (if anywhere) does Wordsworth figure in the level 0 or + 1 narrative? Wordsworth, I suggest, realizes his own point of view at level + 1, in the person of the old man. The doctrine of stoic en- durance, after all, of pursuing "dwindling" opportunities for whatever they are worth, is exactly what is espoused more enthusiastically in the

"Immortality Ode"; and the "flash of mild surprise" that is glimpsed in the leech gatherer's eyes picks up tellingly on Wordsworth's meta-

phor, recurrent in other contributions to this poetic debate, for those evanescent "gleams" of poetic insight that represent the last hope of

7. The usefulness of this line in my attempt to justify a "higher" narrative layer for the text was first suggested to me by some similar instances detected by Herbert Clark (personal communication) in quite different, conversational contexts (see section 5).


Wordsworthx -- " Coleridge

Narratory i Reader \

Leech gathererx -- Narrator \

Figure 3. Narrative layers and aligned roles in Wordsworth's poem to Cole- ridge

the maturing poet. It may help, therefore, if we extend Bruce's sche- matization once again by employing subscripts to illustrate the varying alignment of participants at different narrative levels within the text. Such an elaboration yields Figure 3 as a reformulation of Figure 2.

Naturally, we must be cautious as we interpret Figure 3. The narrator at levels 0 and +1 is not the Coleridge of level -1 in fact, and Wordsworth's attribution to the narrator at level 0 of a change of heart after he hears the leech gatherer's story is nothing more than a clever persuasive device of narrative. Rather like role modeling, it seeks to seduce Coleridge into perceiving the reasonableness of Wordsworth's own resolution of the problem: If the fictional narrator is convinced after expressing doubts every bit as black as Coleridge's own, then why should Coleridge himself not also come around? At the same time, we must not lose sight of the fact that Wordsworth cannot be sure of his own success, and the desired conflation of the y-indexed roles in Figure 3 may in fact never occur.8

And so we return, one last time, to our original conundrum and to the only piece of the puzzle as yet unexplained: the leech gatherer's smiling tolerance of his interlocutor's conversational clumsiness. The key here, I believe, lies in the nature of the point of the narrative contribution made at level -1. As Schiffrin (1987: 16) again notes, "What is intended and understood as the point [of any narrative] is strongly dependent on social, cultural, conversational, and personal contexts" (my italics). Such is assuredly the case here. On a purely interpersonal level, Wordsworth displays, through the imperturbability of his chosen persona, both his good-humored patience with Coleridge's recurrent worries (hence, of course, the smile) and, surely, his conviction

8. Indeed, as Coleridge's later poems like "Work without Hope" clearly show, Wordsworth's strategy was unsuccessful.


that the only possible response to them has already been given and that "renewing the question," however earnestly or eagerly, cannot educe a different response, at least from this responder. To that extent, Wordsworth may be seen as seeking politely to terminate this dialogue. Indeed, if one characterizes the dialogue as having consisted of a dynamic exchange of mutually responsive positions on a common theme, "Resolution and Independence" does occupy the final turn, since whatever poems one seeks to add to it as afterthoughts are neither as closely linked nor as personal as those that have gone before.

5. Closing Remarks This paper begins with the observation of an apparent violation in the text of "Resolution and Independence" of the conventions that, conversational analysts have proposed, govern "turns of talk." No amount of contextualization, I argue in section 3, can explain that violation

naturally, that is, explain it as having occurred in some realistic though highly specialized discourse setting. Instead, we find it necessary to

begin by acknowledging the preemptive importance of a series of embedded narrative contexts for the dialogue in which the violation has occurred. Since one or more points are being made with the same material at each of a series of different levels, various aspects of the

perceived conversational deviancy can be satisfactorily accounted for

through appeal to those functions in narrative context. There is much that has not been considered here. Prickett (1970:

166) says of this poem that "it is hard to read it as not addressed to

Coleridge." The fact remains that, with a brief autobiographical in-

troductory note that served only to muddy the waters, "Resolution and Independence" was eventually published and hence directed to a wholly new audience, as shown in Figure 4. A full treatment of the

poem would naturally need to take note both of level -2 and of the

complexities of the relationship between it and level -1, itself only an extratextual construct. My purpose in this essay, however, has not been to conduct an exhaustive analysis of a single work. Rather, I

hope to have established by demonstration the practical usefulness of discourse theories of several types both in raising and in resolv-

ing important questions about literary texts. The anomaly detected in the course of my analysis at level + 1 of Wordsworth's poem is quite adequately handled through appeal to structural aspects of levels 0 and -1; the overarching presence of level -2 in no way complicates or simplifies this particular argument.

A final note of interest to linguists, stylists, and literary scholars alike: This analysis indicates that, at least in the case of this poem, the narrative imperative of making a point may, under the right condi-


Wordsworth - - Reader W o thoede

Wordsworth x -- Coleridge \

Narratory -- "Reader" 0

Leech gathererx - Narrator +, A ~~~y

Figure 4. The public poem

tions, obviate the necessity of observing more detailed rules of conversational discourse. This raises a number of interesting questions: Which rules or rule types can be affected, how many, and in what way(s)? Does this possibility obtain in all narratives or only in written, literary, or perhaps even poetic narration? Is the relationship indeed

asymmetrical, or can the "narrative point constraint" itself be usurped by rules of conversational well-formedness or by some other discourse consideration?

One piece of data is highly suggestive. Recordings of spontaneous speech collected by Deborah Schiffrin and William Labov include one story whose narrator attempts to support his point that it sometimes pays to talk to a friend or a neighbor rather than to a doctor when one is sick.9 The narrative is told in the third person for the most part, the narrator relating events that have happened to his friend Louie. The sad story of Louie's medical misfortunes also involves at least three different doctors, all, apparently, male, and Louie's neighbor, the hero of the story, who finally diagnoses correctly what all of the professionals have missed. Towards the end of his narration, the speaker utters the following sentence: "He said it would've been a little bit more, he could've strangled t' death."

One can adduce plausible arguments for interpreting the first he in this sentence in any number of particular, determinate ways. Strict application of sentence-level syntactic rules of anaphora, for example,

9. I am most grateful to Deborah Schiffrin for her permission to refer to this material, which she employed in her course "Sociolinguistic Approaches to Discourse" at Stanford University in the summer of 1987. Many of the competing analyses of those data that I present briefly in the following paragraphs were contributed by Schiffrin's students.


in the absence of any contrastive stress, leads us simply to search back

through the narrative for the most recently mentioned, eligible ante- cedent noun phrase-one of the three doctors. Pragmatically, though, the malapropism that occurs in the indirect speech-Louie has risked death from a strangulated hernia, not from being strangled-points to an inexpert speaker, perhaps Louie himself, as the referent for he. Then again, the narrator's choice of generic they in the immediately preceding clause to refer to the medical personnel who finally oper- ate on Louie might lead us to rule out at least those individuals as candidates for this honor.

In the end, particularly in light of our discussion of Wordsworth's

poem, 1 suspect that all of this searching is, quite literally, missing the point. The narrator here, after all, has somehow to work his way around to demonstrating to his audience that "it sometimes works to talk to a friend . . ." Furthermore, he has selected as his exemplum a story that deals with "the danger of death or of physical injury," matters which, as Labov (1972: 370) notes, "occupy a high place on an unspoken permanent agenda" of discussible topics. But unless, in his story, the advice of some friend proves effectual where more tra- ditional resources have failed, the narrative will not count as having illustrated that appealing to friends "works" in this kind of extreme circumstance. In this light, the evaluative comment "It would've been a little bit more, he could've strangled t' death," embedded as a source of evaluation internal to the narrative and succinctly characterizing its

crisis, seems structurally all but inescapable.10 To whom the remark is attributed, then, may make sense as the question of a theoretical

syntactician. For the discourse theorist, however, it may well be moot, since the "he said" tag merely provides an ad hoc slot into which the

triumphant narrator can drop the clinching evidence of disaster nar-

rowly averted and thus the evidence that talking to friends really does work.

This case confirms prima facie the evidence gleaned from our lit-

erary study and supports Schiffrin's (1987: 14, 22) insistence on "the

integrated nature of discourse," in which the sometimes competing demands of "different levels of analysis" have to be resolved by reader- hearers who are intent, first and foremost, on making "overall sense out of a particular segment of talk." Useful as the many available approaches to discourse may be, and valuable as we may already have found the results, we have only scratched the surface when it comes to describing the ways in which different components of our discourse

10. Indeed, as Schiffrin pointed out in her Stanford lectures, an exactly parallel sentence occurs in a story discussed by Labov (1972: 387), the announced point of which is also almost precisely the same.


theory interact. Adding literary narratives and discourses in poetic form to the fundamental database may well complicate the task of achieving a comprehensive account. As I hope to have illustrated, however, the long-term gains from such a move, whether in conversational, narrative, or literary theory, far outweigh the short-term frustrations that it will undoubtedly occasion.

References Agar, Michael

1985 "Institutional Discourse," Text 5: 147-68. Boer, S. E., and W. G. Lycan

1975 "Knowing Who," Philosophical Studies 28: 299-344. Brewer, William F., and Edward H. Lichtenstein

1981 "Event Schemas, Story Schemas, and Story Grammars," in Attention and Per- formance 9, edited by John Long, 363-79 (Hillsdale, NJ: Lawrence Erlbaum).

Brown, G., and G. Yule 1983 Discourse Analysis (Cambridge: Cambridge University Press).

Bruce, Bertram 1981 "A Social Interaction Model of Reading," Discourse Processes 4: 273-311.

Carter, Ronald A. 1983 "Poetry and Conversation: An Essay in Discourse Analysis," Language and

Style 16: 374-85. Chafe, Wallace L.

1980 "The Deployment of Consciousness in the Production of a Narrative," in The Pear Stories: Cognitive, Cultural, and Linguistic Aspects of Narrative Production, edited by Wallace L. Chafe, 9-51 (Norwood, NJ: Ablex).

Chatman, Seymour 1978 Story and Discourse (Ithaca: Cornell University Press).

Clark, Herbert H., and Catherine R. Marshall 1981 "Definite Reference and Mutual Knowledge," in Elements of Discourse Under-

standing, edited by A. K.Joshi, B. L. Webber, and I. A. Sag, 10-63 (Cambridge: Cambridge University Press).

Clark, Herbert H., and Edward F. Schaefer 1987 "Collaborating on Contributions to Conversations," Language and Cognitive

Processes 2: 19-41. de Selincourt, Ernest, ed.

1944 The Poetical Works of William Wordsworth, II (Oxford: Clarendon Press). Doughty, Oswald

1981 Perturbed Spirit: The Life and Personality of Samuel Taylor Coleridge (London: Associated University Presses).

Ferguson, Charles A. 1983 "Sports Announcer Talk: Syntactic Aspects of Register Variation," Lan-

guage in Society 12: 153-72. Fish, Stanley

1980 Is There a Text in This Class? (Cambridge: Harvard University Press). Fodor, Jerry A., Thomas G. Bever, and Merrill F. Garrett

1974 The Psychology of Language (New York: McGraw-Hill). Gibbs, Raymond W., Jr.

1986 "What Makes Some Indirect Speech Acts Conventional?"Journal of Memory and Language 25: 181-96.


Goffman, Erving 1967 Interaction Ritual (New York: Anchor Books). 1974 Frame Analysis (New York: Harper and Row). 1979 "Footing," Semiotica 25: 1-29.

Gumperz, J. 1982 Discourse Strategies (Cambridge: Cambridge University Press).

Labov, William 1972 Language in the Inner City (Philadelphia: University of Pennsylvania Press).

Labov, William, and David Fanshel 1977 Therapeutic Discourse (New York: Academic Press).

Levinson, Stephen C. 1979 "Activity Types and Language," Linguistics 17: 365-99.

Margoliouth, H. M. 1953 Wordsworth and Coleridge 1795-1834 (Oxford: Oxford University Press).

Merritt, Marilyn 1976 "On Questions Following Questions in Service Encounters," Language in

Society 5: 315-57.

Philips, Susan Urmston 1976 "Some Sources of Cultural Variability in the Regulation of Talk," Language

in Society 5: 81-95.

Polanyi, Livia 1985 "Conversational Storytelling," in Handbook of Discourse Analysis, edited by

Teun van Dijk, 183-201 (London: Academic Press). Prickett, Stephen

1970 Coleridge and Wordsworth: The Poetry of Growth (Cambridge: Cambridge Uni-

versity Press). Sacks, Harvey

1984 "Notes on Methodology," in Structure of Social Action: Studies in Conver- sation Analysis, edited by M. Atkinson and J. Heritage, 21-27 (Cambridge: Cambridge University Press).

Sacks, Harvey, Emanuel A. Schegloff, and Gail Jefferson 1974 "A Simplest Systematics for the Organization of Turn Taking for Conver-

sation," Language 50: 696-735.

Schegloff, Emanuel A. 1987 "Some Sources of Misunderstanding in Talk-in-Interaction," Linguistics 25:

201-18.

Schegloff, Emanuel A., and Harvey Sacks 1973 "Opening up Closings," Semiotica 7: 289-327.

Schiffrin, Deborah 1981 "Tense Variation in Narrative," Language 57: 45-62. 1984a "How a Story Means What It Says and Does," Text 4: 313-46. 1984b "Jewish Argument as Sociability," Language in Society 13: 311-35. 1987 Discourse Markers (Cambridge: Cambridge University Press). 1988 "Conversation Analysis," in Cambridge Survey of Linguistics, IV, edited by

Frederick Newmeyer (Cambridge: Cambridge University Press). Searle, John R.

1969 Speech Acts (Cambridge: Cambridge University Press). Stubbs, Michael

1983 Discourse Analysis: The Sociolinguistic Analysis of Natural Language (Chicago: University of Chicago Press).

Tannen, Deborah 1984 Conversational Style: Analyzing Talk among Friends (Norwood, NJ: Ablex).

Analysing a Poem From Sociolinguistic

Documents

conditions of use

jstor archive

noncommercial use

jstor transmission

youmay use content

publisher contact information

stanford university

information technology