[as published in SPIEL: Siegener Periodicum zur Internationalen Empirischen Literaturwissenschaft Jg. 8 (1989), Heft 1: 107-130]
 

"Mit Texten Kommunizieren Lernen: Einige Reale Leser, Einige Reale Texte und die Pragmatische Situation" Die Annahme, die Pragmatik gesprochener Sprache sei für das Lesen "literarischer" Texte oder für das Verstandnis der Entwicklung solchen Lesens irrelevant, hat in drei verschiedenen Kontexten -- Literaturtheorie, Psychologie, und (am wichtigsten) Pädagogik -- zu Schwierigkeiten geführt, literarisches Lesen zu verstehen und zu erklären, und führt deshalb heute zu Überlegungen, grundlegende Annahmen zu verändern. Neuere Untersuchungen mit Methoden, die es ermöglichen, pragmatische Aspekte von Leseprozessen zu beobachten, begründen vor allem die Hypothese, solches Lesen eher als soziale Transaktion denn als kognitiven Prozeß anzusehen. Bietet man nämlich Lesern Texte in ihrer tatsächlich veroffentlichten Gestalt und nicht als für Experimentierzwecke präparierte Artefakte an, und führt man "diskurs-orientierte Interviews" durch, so stellt man fest, daß manche Leser anders damit umgehen, als es gewöhnlich in experimentellen Untersuchungen der Fall ist. Das gilt besonders für das Verständnis der Entwicklung hoherer Lese- und Schreibfähigkeiten und damit für die pädagogische Praxis.

Even so, it is still common, even among those whose aim is to emphasize the role of social interactions and the interpersonal functions of language -- pragmatics -- in literary activity and learning, to add hastily that we don't really mean to suggest that written language is very much like oral language. Consider this formulation, for example:

It is more than legitimate to question the pertinence of many pragmatic presuppositions for the analysis of literature. Even a minimal definition of the pragmatic project engages aspects of the communication situation which are not properly part of the text- object: questions of producer's meaning, and of situational context, for example, have long since been expurgated from literary consideration, or simply do not form part of the communicative situation. Moreover, the formal definition of these pragmatic components of communication being as yet far from rigorous, the application to literature remains a somewhat dubious enterprise, open to the criticism of metaphorization. (Randall, 1985:416).

There are three realms in which I think the effects of accepting such a view of written language has had unfortunate effects: roughly, we can call them psychology, literary theory, and education.

First, our attempts to bring to bear the knowledge and strategies of psychology in order to understand how written language works -- and particularly how the most complex and rich forms of written language are produced and comprehended -- have been dominated by the kind of work which concerns the reading of fragmentary or trivial or synthetic texts (isolated words flashed by a tachistoscope, or "reading test texts," or "stories" written to embody certain arbitrary characteristics) in peculiar circumstances (with one's head locked in a frame and an infrared beam bouncing off one's eyeball, or reading from a CRT in a laboratory, with instructions to remember as much as possible).

Fundamental problems with such work have been pointed out by figures like Beaugrande (1980, 1987) and Bleich (1984). It is focused on creating mechanical processes which might emulate the logic of reading (e.g., Gough, 1972; LaBerge & Samuels, 1974, 1977), or concerned with how text structure affects what people remember (e.g., Meyer, 1975), or it concerns how people process, understand and remember artificially constructed or simplified "stories" (e.g., Rumelhart, 1975; Mandler & Johnson, 1977; van Dijk, 1980), or, as Kintsch & Vipond (1979) point out, it aims at evaluating texts for "readability."

It is extremely rare for this empirical work to concern itself with the reading of authentic, whole texts in non-laboratory or potentially "dialogic" situations (Hoppe-Graff & Schell, 1988). As Beaugrande (1987:56) points out, "So far, most psychologists who have used texts rather than word lists or isolated sentences have preferred non-literary samples of narrative and expository discourse"; and, as Larsen and Seilman (1987) observe, the texts which psychologists synthesize are usually "lousy stories." Such work is easily caricatured as studies of the processing of what I have called "textoids" (Hunt, 1989). And while it offers some interesting analogies to things readers in authentic, real situations might be doing, I believe that the reading it is focused on is fundamentally different from the kind of reading with which we should be primarily concerned, and that its methodological strategies preclude the possibility of conversation between reader and text.

In the area of literary theory, the situation is more complex. At least two of the principal kinds of activity in which a practicing literary scholar might be engaged embrace as fundamental assumptions the kind of separation Randall assumes between written text and pragmatically embedded text.

In order to facilitate the continuing production of new interpretations of canonical works and writers (e.g., Hunt, 1975), it has been useful (if not, indeed, necessary) to rule out of bounds "questions of producer's meaning, and of situational context." This has meant, in practice, the treatment of literary works as though they had no authors, contexts, and intentions, and as though their readers (when they are considered at all) were acting in a space quite separated from the world of practical action, reading in a text museum, with "that distinter[est]ed contemplation, which according to Adorno's argument in his essay on Beethoven's Missa Solemnis, neutralizes art and casts it deliberately into a useless limbo" (Sanguinete, 1967:393).

Similarly, the continuing elaboration of literary theory as an end in itself, a complex and often brilliant intellectual exercise, is one which I (and, of course, many others) have questioned as, in the course of keeping literature separate from workaday considerations, abdicating "responsibility to what might be called the social question of the humanities" (Goodheart, 1983:468), if not as self-regarding and self-indulgent (cf. Eagleton, 1983: Mitchell, 1986; Kamuf, 1986: Hunt, 1987b, in press b).

Most immediately and practically, it seems that these activities assume, and foster, an attitude toward literature as a special and different kind of language, a set of texts (in the case of interpretive criticism) or practices (in the case of theory) which have no practical consequences whatever, which exist in the realm of purely aesthetic experience and purely intellectual contemplation. Pratt's convincing argument (1977) that "literary" texts do not differ from other texts in any linguistically definable way is countered (if acknowledged) by saying that the difference is in the way such texts are treated. In Archibald MacLeish's once-celebrated formulation, "A poem should not mean / But be."

It should be clear that these are serious problems only from certain particular standpoints. If you assume that literature and its criticism and theory should be socially engaged (cf. Lentricchia [1983], for instance) you will be concerned about this separation of literature from "ordinary language": if not, it will not seem a problem. It is clear that many literary professionals would not share my concern.

It is more common to find concern for such issues among those who think of themselves primarily as teachers. And indeed, in my view the arena in which these assumptions have had the most important effects is that of language development and education. As a literature teacher, it has seemed to me for some time that to treat literature in the way that literary theory and psychology insist we must is to render it innocuous (Hunt, in press a) and unengaging and -- most important -- to teach the tacit lesson that written language in general is a sterile and empty exercise rather than an immediately socially useful tool like oral language (Hunt, 1989). Further, and worse, if Vygotsky (1986: see also Wertsch, 1986: Forman & Cazden, 1985) was right that it is through our using language in authentic, socially embedded ways that language comes to be for us a tool for thinking -- indeed, becomes thought itself -- disengagement of this powerful and rich kind of language from immediate social motives and concerns is to render it less accessible as such a tool.

Even though work on literacy done in the context of educational interests often seems more relevant to such concerns (cf. especially Smith [1982] and Goodman [1982]), it tends to be most often focused on reading that occurs among young and learning readers and in school situations, and on reading in order to learn facts and information or, at best, to write formal interpretive essays on literature (e.g., Purves & Rippere, 1968). Again, such motives for reading, while related to the situation in which students actually find themselves, might characterize only the kind of reading done in schools. It is at least plausible that the class room context is as inauthentic, and produces a reading as different from "real" reading, as what might occur in reading laboratories, under the watchful eye of an experimenter with a stopwatch.

If in fact there were no reason to think that the separation of pragmatic texts from aesthetic ones was ill founded in fact, of course, any suggestion that it is educationally and theoretically counterproductive to accept the separation would have to be dismissed as flying in the face of experience. However, in recent years I have come to think that the separation does not actually adequately describe what readers do.

Although there are increasingly strong suggestions in other areas that such a separation is difficult to maintain, my own convictions about this derive from my involvement in research into literary reading. Let me quickly summarize that work, stressing the implications for thinking about the similarities and connections between socially embedded uses of language and those we have come to think of as "aesthetic."

About five years ago my colleague Douglas Vipond (in collaboration with whom all my work in reading study has been done) and I began an attempt to find out more than seemed known about what happens when people read literature. Our basic research question could be phrased as having three components: What do (1) actual people do when they read (2) "literary" texts in (3) authentic or "naturally occurring" situations? We were convinced that the basic research strategies had to be empirical. We wanted, not the results of introspection into our own reading processes (though we have done that) nor the results of abstract theory building (though we've done that as well), but rather observable data that might be subject to intersubjective verification. We also wanted our studies to reflect what readers did with authentic texts in actually occurring situations -- that is, it was our attempt to generate real readings, and not professional readings, or "student-style" "learn from text" tasks.

At first we used traditional experimental and quantitative (statistical) strategies. Our earliest studies (e.g., Vipond & Hunt, 1984; Hunt & Vipond, 1986) involved, for instance, more than 250 readings (primarily by university freshmen) of short stories (most were of John Updike's widely anthologized story, "A & P"). We used various strategies in an attempt to discover what was going on behind the silence of our readers. We tape recorded, coded, and quantified oral protocols. We taped readers explaining, during the reading, their choices between various "branches" of the text in which alternative continuations were presented. We taped, coded, and quantified protocols from retrospective interviews (particularly from responses to "probes," or things we told readers had been said about the stories by other readers). We kept records of the time it took readers to read various sections of the stories. We used various devices to ascertain what readers noted while reading and remembered after reading. The basic findings from this reasonably traditional sort of research led us to posit three different modes of reading. It seemed to us that each instance of reading we studied could be categorized as predominantly guided and shaped by one of three sets of purposes or intentions. We have described these categories elsewhere at some length, so here I will merely sketch them.

1. information-driven reading: when this mode of reading is dominant, the reader's overriding purpose is to acquire information from the text. It is similar to what Rosenblatt (1938) has called "efferent" reading, in that it is the reader's main intention to "carry something away" from the text.
2. story-driven reading: in a situation where this mode is dominant, the reader's overriding purpose is to be engaged in a story-world, to imagine the world as fully as practicable and to treat its characters and events in a way parallel to how those characters and events would be treated were they real. The readers of romances described by Radway (1984) seem normally to engage in this kind of reading.
3. point-driven reading: in a situation dominated by this mode, a reader's central purpose is not to identify and remember facts, nor to live through the vicarious experience; rather it is to be engaged with the text as discourse, in a way which seems to be very similar to how, in conversation, we attend to oral stories (cf. Labov, 1972; Polanyi, 1985; Tannen, 1984). We expect the discourse to be, in Labov's terms, "pointed"; we think that the teller probably has a comprehensible reason for uttering this discourse, and that that reason will probably be socially relevant. Accordingly, we treat information and events as subordinate to such purposes, and look for shared "evaluations" (Labov, 1972; Polanyi, 1979) embedded in the text by an intending author.

First consider the text. It would seem, for instance, that a text such as a telephone book or a copy of Books in Print must have characteristics that "push" readers toward reading it in an "information- driven" way. But the only way we can know that it has such characteristics is that people in fact usually do read it in such ways -- in practice, that is, it appears to "afford" such reading. It is, however, clearly conceivable that a certain reader in a certain situation might read the telephone book for other reasons than to take away a number -- read it to share the values of the telephone company, for example, or read part of it which had been "framed" as a "found poem" (and thus as presented or authorized by someone to share values other than those of the telephone company). Or consider, at the other end of the spectrum, a modernist text such as Italo Calvino's If on a Winter Night a Traveler -- or even an almost traditional story like "A & P." Such a text may "afford" a reading whose aim is to "make contact" with a speaker or author in an immediately social way, to share evaluations and understandings with that real or imagined person; but, obviously it can also be read for a vicarious experience of an imagined story world or -- as Louise Rosenblatt (1938) first pointed out -- efferently, as a source of facts and propositions and information to be acquired and remembered. What the text "is," above the level of a particular set of black marks on a white page, is determined by how it is read, which is a function of the reader and the situation.

Similarly, it might be expected that a given reader, as Norman Holland (1976, 1985) or David Bleich (1975, 1978) have argued, unilaterally determines the nature of the transaction in any given instance, bringing to bear on the experience her own preconceptions, expectations, abilities and habits as a unique person -- a student, a habitual reader, a habitual non-reader, a professional, or whatever -- pretty much regardless of the nature of the text. Yet it can be observed, in practice, that some texts actually are usually read in certain predictable ways, and that some situations do affect in principled ways how readers act. Further, investigation of the ways in which readers' concepts of literature and views of themselves as participants in the social systems organized around literature affect their reading (e.g., Meutsch and Schmidt, 1985; Meutsch, 1987, in press) also make clear that a reader's activity is profoundly affected by her more general social position and role.

And finally, the situation -- both the larger social situation and, more specifically, the immediate physical and social context (the classroom, the laboratory, a testing situation, a living room; a text is issued by an institution or offered by a friend) -- plays a role in the way a reader will act in reading. But, again, this role is not a simply defined one: situations which an observer might think of as "the same" will be perceptually constructed by different readers in radically different ways, and will thus affect the reading of even the same texts in different ways.

All of this can be understood as merely an attempt to stipulate as many consequences as possible of saying that any instance of reading is, and thus can only be comprehended as, a transaction. The process of generalizing about reading is, as we came slowly to understand, far more complex than we had expected. But as we struggled to understand our results, the pattern I have just described emerged as a conceptual frame which seemed to us to begin to account for them.

It was clear, however, that the broad categories of reading we identified, and the general patterns we saw among the entities we had identified as fundamental, all might be subject to a great deal more analysis. If there were differences between what a reader was doing when she was driven mainly by the desire to acquire information and when her intention was to participate in construction of evaluations and points, what -- specifically, operationally -- were they? If which of these she does is subject to the influence of the situation, what elements of situations might be salient to which readers and which texts? We decided that the most immediate question had to do with the nature of point-driven reading, which seemed to us to be the kind of reading typical of readers engaged most fully with "literary" texts. Thus our new question became: can we refine the notion of "point-driven reading"? In what way is it, for example, personal or social in the sense which might be suggested by the analogy with conversational listening? This question was not one we had been eager to ask.

For one thing, it was one which it seemed to me implicitly questioned many assumptions of the literary theory with which I was familiar. There was reason to think reading should not be personal in this sense, reason based not only on the assumptions of the New Criticism I had grown up with (Wimsatt & Beardsley, 1954), but also on those of deconstruction and most other current literary theory, which have argued that literature is separate, aesthetic, private, and irrelevant to "authorial intention" and thus not to be seen to be seen as a social transaction in this sense.

It was also one which ran counter to some fundamental assumptions of psychological reading research, which offered reason to hope it wouldn't turnout to be social in this sense, partly because language processing must be separable from immediate social circumstances and goals (or else it becomes impossible to study by traditional psychological means).

And finally, it seemed to contradict some of the traditional assumptions of educational research and practice, which assumed that reading was a set of decontextualized cognitive skills, and learning was a process of acquiring information and "facts" which existed somehow "out there," prior to and separate from the educational context.

In spite of this, it seemed to us we were driven inexorably to investigate the possibility that when readers were reading texts they deemed "literary" -- or, as Schmidt (1989) describes it, participating in the social system organized around literature -- they were often reading in ways that could be closely analogized to the way they might participate in a conversation. And indeed, the more we thought about the question, the more it seemed we could focus it precisely by looking at specific aspects of conversational interaction. To what extent can we say that a point-driven reader is acting as though she were in a conversation? Can we say it at all? What does it mean and entail to say it? Does it mean that she is engaged in a pragmatic transaction with another person? That she makes the same assumptions, employs the same comprehension strategies, and embraces the same social goals -- that, in other words, she's trying to do the kinds of things that sociolinguists tell us (and that we know intuitively) participants in conversation are trying to do -- to "make contact," to share what Labov and Polanyi call evaluations, to say "I think this is funny, I think that is bizarre, I think this is surprising and that is appropriate and that is outrageous"? Is that what these readers seemed to be doing? Were such readers sensitive to evaluations? Did they impute motives to a hypothetical "author"? Did they attempt to create and negotiate situationally relevant points? How did people learn how to do this, and why did so many seem not to learn it at all?

In an attempt to explore these questions we were forced to come up with what was for us a new, more "ecologically sensitive" research strategy. Such "naturalistic" strategies are commonplace, of course, in many other areas of investigation, as Lincoln & Guba (1985) and Guthrie & Hall (1984) have pointed out. We decided to work with a greatly reduced number of readers, and try to put them in situations which were far less controlled and structured, and to examine the readers' activities and statements for regularities and patterns in the areas with which we were concerned. As Rubin (in press) has phrased it, we were prepared to trade control for regularity. Another consideration is one Andringa (in press) has explained elegantly: it seemed that if we were interested in literary reading, it was contrary and inconsistent to continue to insist on statistically "averaging down" to a mean, when what we were really concerned with was the creative, the individual, the unique, drawn from various departments within the university -- not including literature departments, because while the sort of "professional" reading that often characterizes such readers is clearly of interest, it poses problems we preferred to postpone. We asked each of these readers to read four complete texts. Each reader was interviewed over a four hour period (divided into two hour sessions, separated by a few days). In each case, the reader read a text, had a conversation with our interviewer about it, and then read another. At the end of the two-part session, they talked about the whole experience and went back to the individual texts for final comments and wrapup. We attempted to make the physical situation as comfortable for the readers as possible by giving them an upholstered chair and incandescent lighting and so forth, and by handing them the text so that it was in their possession: further, our interviewer busied himself with other work in another corner of the room while they read. The interview itself was only semi-structured: the interviewer primarily tried to get the reader to say as much about the reading as possible (for a fuller description of this study, see Vipond, Hunt, Jewett & Reither, in press). The sessions were tape recorded and transcribed, and then read and reread, by the participants and the interviewer as well as the experimenters, looking for patterns among readers and within readings -- and particularly, on the experimenters' parts, for evidence of the kinds of engagement which it seemed to us might typify "point-driven" reading.

One aspect of the method seems worth commenting on at some length here, because of the way in which it opened up an unexpected aspect of the reading to our observation. The texts were not presented as dot-matrix printed "experimental texts," or on computer screens, or even as photocopies: the readers were handed the actual publications in which the texts had appeared, e.g., a commercially published book that someone might buy on a paperback rack (as opposed. to an artificial text that some university professor has chosen and possibly tampered with). We used four texts. One was an unsigned "Talk of the Town" essay from The New Yorker (McPhee, 1987); in this case the interviewer handed the readers an actual July 6, 1987 issue of the magazine. Two others were recently-published short stories by an author who taught on our campus and lived in the local town (Thompson, 1986): in this case the readers were handed a copy of Leaping Up, Sliding Away. The fourth story was a parodic metafiction by Ursula Le Guin (Le Guin, (1982): readers were handed a paperback copy of The Compass Rose.

Offering our participants the texts as they appeared in their original venue was consistent with our general and long-standing commitment to using "authentic" texts (as opposed to textoids), and was conceived of as a way to "advertise" that authenticity. We discovered that this was even more important than we had had expected, and entailed some methodologically interesting consequences. One of the strongest and most consistent kinds of distinctions we found between more and less pragmatically engaged readings is that the more engaged (or point-driven) readers tended to take aspects of the physical context -- particularly, the nature of the actual publication -- into account as likely to be relevant to the reading, while other readers seemed to assume that such concerns were unlikely to be relevant. They tended, that is, not to use the cues provided by the venue when thinking about matters such as the author, the date, the occasion, or the genre of the piece. Had our study presented these texts as uniform, unidentified, homogenized and disconnected examples of text -- pamphlets printed by computer, for example, of the kind we have used in other research -- it is clear that what turned out to be one of the most important differences between these kinds of readings could not have been observed.

This has, we believe, profound methodological implications for the study of reading -- and not just for the study of "literary" or "aesthetic" reading. If, in constructing the reading occasion, the experimenters inadvertently block out of possibility important kinds of activities, they shouldn't be surprised that all their readers seem to do the same things. As Anne Freadman (1987) has pointed out in another context, a "pretend" tennis match does not afford the employment of strategies and activities that a real one does.

In general, we tried to create a situation which afforded (but did not -- could not, of course -- compel) point-driven reading. The readers were chosen to be a range, some of whom presumably were used to reading in such a way; the situation was, as far as possible, one which did not powerfully afford information-driven reading; and the texts were ones which it seemed to us afforded a range of readings -- we expected, on the basis of our own readings of the texts, that the two short stories would afford both story- and point-driven reading more than information-driven; that the New Yorker essay would afford point- and information-driven reading more than story-driven; and that the Le Guin parody would strongly afford point-driven reading.

The transcribed interviews are still undergoing scrutiny. We have already found some interesting patterns among them, which I will not describe here in detail; we have done so elsewhere (Vipond, Hunt, Jewett & Reither, in press). Here I wish to stress what seem to me some patterns which suggest a good deal about the social nature of point-driven reading that our previous, more experimentally manipulative and statistical, methods had not allowed us to see. As I have said, our primary purpose in this study was to unpack the some of the general notions we've outlined in previous work, and move toward new and increasingly specific hypotheses to drive further exploration. Let me describe three of the ideas which seem to me most important, exploring one in particular at a little more length.

I think the data show that there are indeed some regularities among read- ers who are engaged with texts, who see themselves as having had satisfactory experiences (not that they necessarily liked or admired the text, but that they seem to have been engaged with it); habitual readers who describe themselves as satisfied with their own reading and whom we would describe (in this situation, with this text) as reading for point. Some of these regularities were surprising, others were not.

For instance, it was a surprise to me (not that I had expected something different; rather it hadn't occurred to me to make a prediction in this area, to ask the question at all) that such readers (as I have already suggested) tended also to be those who see the immediate situational and contextual frame as relevant, as something that mattered, that they talked in terms of voluntarily. They tended to have responded to the generic cues of the physical presentation of the text, to have not iced, for instance, that the two stories appeared in the same book and were by the same author; to have noticed that the essay appeared in a a certain magazine; that the Le Guin piece was not framed as scientific exposition, but rather as fiction. Readers who were less engaged did not; many of our readers, for instance, persisted in reading the Le Guin as a (largely incomprehensible) physics article, and when one, at the end of the interview, was told that it was fiction, she expressed amazement and decided that she no longer liked the piece because it wasn't true that, as the article proposed, the reason we're all so short of time is that time is leaking out of a tiny hole in the universe.

It was less of a surprise to us, because we had seen evidence of this in our previous work, that readers who were engaged seemed to act as though the author's intentions were an important issue: for instance, they regularly referred to authorial intention as a way of explaining the difference between two alternative passages. I do not, incidentally, mean to suggest with Hirsch (1967) and others that a historically reconstructed "authorial intention" should be the ultimate arbiter of "the text's meaning." Our readers were constructing authorial intentions, often quite different from what other readers, or we ourselves, might have constructed. As Beaugrande has observed, "like authorship, authorial intention readily becomes our own construct to account for our responses and values. (1987:67).

This is important, I think, because it suggests that in order to treat generic conventions and evaluations (often signaled, as Labov (1972) has phrased it, as "departures from the local norm of the text") as intentional, and to expect explanations and coherence rather than to dismiss them as inexplicable anomalies, readers must treat text as intentionally coherent, and thus must at least tacitly imagine an author or authority "behind" it, warranting that intentional coherence.

All of these considerations came together in a nice contrast between two of our participants' comments about the authorship of the two Thompson stories. Thompson, author of "Metaphors" and "The Sun on Mount Royal," teaches at the University of New Brunswick, located on the same campus as St. Thomas University. Therefore some of our readers not only knew of Thompson, but had actually met him and heard him talk about his new "postcard stories." Only some of these readers, however, considered this knowledge to be relevant when reading and discussing the stories. Don, for one, clearly did:

Jim: The question will be, what did you make of "The Sun on Mount Royal"?

Don: This must be one of Kent Thompson's postcard stories. I talked to him about this.

Jim: Is that right?

Don: Yeah, last year, I was at a party at his house and we got talking about this and so I kind of know what he's trying to accomplish with these about giving everything and forcing yourself to do it, you know, on the size of a postcard. He suggested I try it and I haven't got around to trying it. This is the first one of his I've read actually.

Jim: Would you happen to remember the name of the individual who wrote "Metaphors"?

Kevin: "Metaphors," no . . .

Jim: How about "The Sun on Mount Royal"? Kevin: Again, no -- I can't picture the name.

Jim: All right, do you think that those two pieces may have been written by the same individual or do you think they were written by two different people?

Kevin: ( . . . ) I think they could have been written by the same person.

Jim: All right, they were written by the same person, his name's Kent Thompson --

Kevin: Yeah, I like Kent Thompson.

Jim: Are you familiar with Kent Thompson?

Kevin: Yeah, I took a course from Dr. Jones two years ago on Maritime literature and we studied some of Kent Thompson's work and Dr. Jones has a, as you probably know, a lecture series throughout the year for that course and he was one of the persons who came in and read for us. I know he teaches at UNB himself. Yeah, I like his work . . . .

Jim: ... There are some pieces in there that are only five or six lines long and. . .

Kevin: He started this thing a few years ago about writing on postcards, is that inhere, too? . . . He was talking about that when he read for us.

Finally -- we have of course found other regularities, but these are the ones I think may be most interesting here -- readers who are reading in this way expect that the text will have pragmatic consequences, they expect it to be connected to their life in some way. I don't mean, though this can also happen, that the story means something that will change one's life. Rather, what I'm referring to is signaled by remarks like, "I know someone who'd like that story: can I get a copy of it?" or, "Who wrote that again? I want to write that down because I want to give that book to a friend." That taking of the story and "retailing" it seemed very important to me, because of course that's what we do with stories in conversation. If a story strikes you as pointed and interesting and powerful and memorable, you will often, in another situation, remember it and retell it. ("Oh, I heard about a guy who . . . ") That seems to me analogous to what many of our point-driven readers were doing, or expecting to do. And that was a surprise -- not, again, because we'd expected otherwise, but because we'd had no expectations in the area at all. This, it seemed to us, was the sort of benefit Rubin had predicted when experimenters swap control for regularity.

The way we formulated all this for ourselves was to say that the readers who were, in our terms, successful with these texts in this situation seemed be doing something that looked very much as though they were engaged in dialogue with the text (or, better, with an imputed or imagined intentional author by means of the text). I should make it clear, by the way, that we're not asserting that something "is true" about these readers as people: we are saying only that on this occasion, reading this text, these readers seemed to be acting in this way. I see no reason not to expect that everyone is capable of reading in a point-driven way; after all, we are all point-driven listeners.

That at least some readers read in such a "dialogic" way -- and that many did not -- seems to me worth knowing in part because it has important consequences for literary theory and psychology, but primarily for educational and developmental reasons.

To a literary theorist it insists, for instance, that we need to include the pragmatic dimension in our concept of the reading of imaginative texts, of "literature," which people do for their own purposes. If we are to understand what reading is at all, we must see it as a process in which, at least potentially, this kind of social interchange and activity is entailed. This view is one which, as I have admitted, challenges many of my basic assumptions about appropriate approaches in literary theory; it suggests that although most of the "aesthetic" and "interpretational" things which I as a hermeneutic practitioner have done may be interesting, they don't have a lot to do with how and why people read texts -- including the "classics" of literature that I was interpreting and evaluating.

To a psychologist it insists we need to introduce the pragmatic dimension into our understanding of any reading. This would challenge some basic assumptions in the psychological study of reading -- the assumption, for instance, that reading is pretty much the same thing from one reader to the next, from one text to the next and from one situation to the next; the assumption that it's not a social phenomenon, but a cognitive one. In my view, it now seems far more useful to consider reading as only in relatively trivial ways a cognitive phenomenon: the most important things about it can be understood, I now think, only by viewing it as primarily a social phenomenon. But finally, and of most immediate importance here, this new emphasis on the rhetorical and the pragmatic has powerful implications for education and teaching. Consider, for example, the status of written language in schools: what often happens is that the printed texts which are used in schools -- and, as well, the written texts which are produced in schools by students -- are consciously and deliberately designed not to afford this kind of reading, consciously and deliberately designed to be "fact dumps," to be collections of propositions stored in an orderly way -- not to be pointed, not to be contact-driven, not to be human. They are written in many cases by committees, and often by corporations (Goodman, Shannon, Freeman, & Murphy, 1988; Luke, 1988). If you look at reading texts for young children, if you look at school textbooks in the sciences and humanities, if you look at the introductions and the footnotes and the explanatory material for the texts and anthologies which are usually used in North American universities as the basis of literature courses, you find that everything about them is written as though to make it very difficult to read them in a point-driven way. Not impossible, of course -- but possible only for those with a strong predisposition to read them for human contact, and possible only in situations which afford such reading. Schools, in general, do not provide such situations.

If all that is true, and it seems to me no exaggeration, then we are offered a plausible explanation of why it is that we get so much information-driven reading in school situations, and so very little point- driven reading. Perhaps it offers us a partial explanation, too, of why some people develop into powerful manipulators of written language -- both as writers and readers -- and why others do not. Here is a hypothesis about that question.

As I have already noted, in dealing with oral language in comfortable social situations, everyone has learned to be a point-driven listener. Why should written language develop so differently?

The usual argument works like this. It seems undeniable that oral language is learned with a minimum of direct instruction. Children learn the basic strategies and grammar of their native tongue within the first few years of life. Although those "basic" strategies and grammar are unimaginably complex, virtually no direct explanation of them is ever offered the child. Indeed, it is usually the case that the adults the child has contact with would be unable to explain even the most basic of them; and it's arguable that no one understands all of them with anything like the thoroughness it would require to explain them all to anyone, much less an infant. How language is learned seems to be a matter going well beyond our power of description -- or perhaps even understanding -- but it is clearly connected with the child's practice of using language to make and maintain contact with the people around her, to achieve ends which are important -- indeed, crucial -- to the child's life and growth. If we accept Vygotsky's model of the role of oral language in the development of human intelligence, indeed, we would argue that the learning of oral language is the means by which the individual, defined as a social being, creates herself -- or, as Kenneth Kaye (1982) has shown, opens and operates the dialogic relationships that help the society around her create her. And all this language is learned through use.

Written language, however, is argued to be quite different. For instance, it is not dialogic in the immediate way that oral language is: the employer of written language doesn't have an interlocutor physically present to supply the sorts of cues language learners need, and language without an interlocutor needs, as Vygotsky himself pointed out, to be consciously shaped to provide the information -- about what syntactic structures and signals of coherence and so forth need to be utilized. This is, it is argued, a very complex business and one which renders written language a form which cannot be read in the helter-skelter, hit-or-miss fashion with which young children regularly perform the miracle of language development.

Because written language is so lacking in the support offered learners of oral language, then, it is argued, we must supply this lack with direct instruction. Learners need, for example, to be able to think about language as a structure, quite separate from action (Cook-Gumperz, 1977). A young child will learn, for instance, the oral structure of the relative clause by engaging in a situation where the form is necessary, using it in imperfect ways and having interlocutors be confused or hesitant, hearing it used effectively in context where it's meaningful and related to the child's concerns and goals, etc. In writing, however, there is no interlocutor to be confused or hesitant, and no one to use the construction back, in effective ways, about something with which the language learner is genuinely concerned. So we must discursively explain the relative clause -- or the nominative absolute, or the structure of narrative and expository genres, or the shape of the five-paragraph theme, or the rule about where the apostrophe goes in "it's," "its'," and "its" -- or the "meaning" of a literary text.

It seems to me, however, that this argument is based on an erroneous assumption -- that written language must by its very nature be entirely decontextualized. This assumption depends upon accepting the view that all writing is a lot like published print -- created in one context and consumed in quite another, with no hint to the consumer of the context of the creator. Not only that, it is assumed that the consumer is somehow removed from any social context -- not only that the consumer doesn't share a context with the producer, but also that the consumer is totally alone as a language user: not a member of a language or discourse community, and not a participant in an immediate, social situation. Now this may in fact be true of some situations where written language is used, but it is not necessarily true -- and there is no reason to think that, even if it were widely the case, it is a situation which is either unavoidable or desirable. If it were true that written language were habitually encountered in such a way, then it might be true that we could not rely on the support of social context to facilitate language learning, and we might feel it was necessary to resort to explicit explanation (if, of course, we were sure we understood enough of the structures to construct lucid and useful explanat ions). If not, however, it seems clear that it would be preferable to allow the kinds of processes which are so powerful in children's learning of their first, oral language -- or, indeed, the kinds of processes which writers like Stephen Krashen (1981) argue are indispensable in the learning of second languages -- to operate.

Why, then, do some children develop into readers who can readily shift into "point-driven" mode and others not? It is, I think, at least plausible that those readers who learn about written language are often those who encounter it in situations where the sort of support available for oral language is also available. They are those for whom parental and caregiver story reading has helped bridge the gap between written and oral, where written language has been regularly used by those in the child's environment, with the child, in some of the many ways in which writing actually does function in social contexts -- notes, lists, apologies, invitations, instructions, etc. (Baghban, 1984; Bissex, 1980). If a child's written language develops in the same contextualized ways as her oral language, the child will, it seems likely, be prepared to synthesize or create contexts for written texts, even when the actual situations and texts afford such creation only with difficulty. The child, in other words, may be prepared to read in point-driven as well as information- driven and story-driven ways.

Finally, all this may suggest some ways in which the educational situation can be changed. It suggests, for instance, using authentic texts, texts which were not written to impart information to someone but which were written to make contact, to invite mutual participation in the creation of truth. Such texts may not necessarily or primarily be published, public texts, but texts produced in the classroom context itself, for the immediate social and educational purposes of the learners. It suggests a principled rationale, that is, for the educational strategies often called "Whole Language" (Harste, Woodward & Burke 1984; Newman, 1985).

It suggests, further, that the best way to teach a field that has a defined knowledge base may not be to present students with texts incorporating in an efficiently organized way "the truth" about the field (for instance, in "the" course psychology text or "the" course physics text, or even in "the" truth as presented in the instructor's lectures about eighteenth century literature), but rather to find a way to give the creative initiative to the students, help students participate in the social construction of the knowledge of the field, to invent for themselves a role in the social dialogue -- written and spoken -- which constitutes a discipline (Hunt, 1987a, 1987b, 1989; Hunt, Parkhill, Reither & Vipond, 1988; Parkhill, 1988; Reither, 1985, 1988; Reither & Vipond, submitted).

References

Andringa, Els. "Verbal Data on Literary Understanding: A Proposal for Protocol Analysis on Two Levels." Poetics [in press].

Baghban, Marcia. Our Daughter Learns to Read and Write: A Case Study from Birth to Three. Newark, Delaware: International Reading Association, 1984.

Beaugrande, Robert de. "Schemas for Literary Communication." In Literary Discourse: Aspects of Cognitive and Social Psychological Approaches, ed. Laszlo Halasz. Berlin: deGruyter, 1987. [49] - 99.

Beaugrande, Robert de. Text, Discourse, and Process: Toward a Multidisciplinary Science of Texts. [Advances in Discourse Processes, Vol. 4]. Norwood, N. J.: Ablex, 1980.

Bissex, Glenda L. Gnys at Wrk: A Child Learns to Write and Read. Cambridge, Massachusetts: Harvard University Press, 1980.

Bleich, David. "Reading Research and Reading Experience." Paper presented at the National Reading Conference, St. Petersburg, Florida, 30 November 1984.

Bleich, David. Readings and Feelings: An Introduction to Subjective Criticism. Urbana: NCTE, 1975.

Bleich, David. Subjective Criticism. Baltimore: Johns Hopkins University Press, 1978.

Clark, Katerina, and Michael Holquist. Mikhail Bakhtin. Cambridge, Massachusetts: Belknap Press of Harvard University Press, 1984.

Cook-Gumperz, Jenny. "Situated Instruction: Language Socialization of School Age Children." In Child Discourse, ed. Susan Ervin-Tripp and Claudia Mitchell-Kernan. New York: Academic Press, 1977. 103-121.

Dewey, John, and Arthur F. Bentley. Knowing and the Known. Boston: Beacon Press, 1949.

van Dijk, Teun A. "Story Comprehension: An Introduction," Poetics 9: 1-3 (1980), 1-21.

Eagleton, Terry. Literary Theory: An Introduction. Oxford: Blackwell, 1983. Minneapolis: University of Minnesota Press, 1983.

Eco, Umberto. The Role of the Reader: Explorations in the Semiotics of Texts. Bloomington, Indiana: Indiana University Press, 1979.

Forman, Ellice A., and Courtney B. Cazden. "Exploring Vygotskian Perspectives in Education: the Cognitive Value of Peer Interaction." In Culture, Communication, and Cognition: Vygotskian Perspectives, ed. James V. Wertsch. Cambridge, Cambridge University Press, 1985.

Freadman, Anne. "Anyone for Tennis?" In The Place of Genre in Learning: Current Debates, ed. Ian W. Reid. Geelong: Deakin University Centre for Studies in Literary Education, [1987]. 91-124.

Gibson, James J. The Ecological Approach to Visual Perception. Boston: Houghton Mifflin, 1979. Goodheart, Eugene. "The Function of Matthew Arnold: 1. Arnold at the Pres ent Time." Critical Inquiry 9 (March 1983), 451-468.

Goodman, K. S. Process, Theory, Research. (Language and Literacy: The Selected Writings of Kenneth Goodman, Vol. 1), ed. F. V. Gollasch, Boston: Routledge and Kegan Paul, 1982.

Goodman, K. S., P. Shannon, Y. Freeman, and S. Murphy. Report Card on Basal Readers. Katonah, New York: R. C. Owen, 1988.

Goodman, Yetta M., and Bess Altwerger. "Print Awareness in Pre-School Children: A Working Paper. A Study of the Development of Literacy in Preschool Children." Program in Language and Literacy, Occasional Papers, No. 4. Tucson: Arizona Center for Research and Development, University of Arizona, September 1981.

Gough, Philip B. "One Second of Reading." In Language by Ear and by Eye, ed. J. F. Kavanagh and L. G. Mattingly. Cambridge, Massachusetts: MIT Press, 1972, 331-358. Repr. in Theoretical Models and Processes of Reading, Second Edition, ed. Harry Singer and Robert B. Ruddell. Newark, Delaware: International Reading Association, 1976, 509-535.

Guthrie, Larry F., and William S. Hall. "Ethnographic Approaches to Reading Research." In Handbook of Reading Research, ed. P. David Pearson. New York and London: Longmans, 1984. 91-110.

Harste, Jerome C., Virginia A. Woodward, and Carolyn L. Burke. Language Stories and Literacy Lessons. Portsmouth, New Hampshire: Heinemann, 1984.

Hirsch, E. D., Jr. Validity in Interpretation. New Haven: Yale University Press, 1967.

Hirsch, E. D., Jr. Cultural Literacy: What Every American Needs to Know. Boston: Houghton Mifflin, 1987.

Holland, Norman. "The New Paradigm: Subjective or Transactive?" New Literary History 7 (Winter 1976), 335-346.

Holland, Norman N. "Reading Readers Reading." Researching Response to Literature and the Teaching of Literature: Points of Departure, ed. Charles R. Cooper. Norwood, New Jersey: Ablex, 1985. 3-21.

Holmes, Betty C., and Nancy L. Roser. "Five Ways to Assess Readers' Prior Knowledge." The Reading Teacher 40: 7 (March 1987), 646-649.

Hoppe-Graff, Siegfried, and Martin Schell. "The Comprehension of Literary Texts: Developmental Considerations." In Comprehension of Literary Discourse: Results and Problems of Interdisciplinary Approaches, ed. Dietrich Meutsch and Reinhold Viehoff. Berlin: deGruyter, 1988. [89] -110.

Hunt, Russell A. "A Boy Named Shawn, a Horse Named Hans: Responding to Writing by the Herr von Osten Model." Writing and Response: Theory, Practice and Research, ed. Chris M. Anson. Champaign-Urbana: National Council of Teachers of English, 1989.

Hunt, Russell A. "'Could You Put in Lots of Holes?' Modes of Response to Writing." Language Arts 64:2 (February 1987a), 229-232.

Hunt, Russell A. "A Decade of Change: A Correspondence on Theory and Teaching." Reading-Canada-Lecture 5:3 (Fall 1987b), 148-153. Hunt, Russell A. "Modes of Reading, and Modes of Reading Swift." Festschrift for Louise Rosenblatt, ed. John Clifford [in press b].

Hunt, Russell A. "The Parallel Socialization of Reading Comprehension Re- search and Literary Theory." Beyond Communication: Reading Comprehension and Criticism, ed. Stanley B. Straw and Deanne Bogdan. Upper Montclair, New Jersey: Boynton/Cook [in press a].

Hunt, Russell A. "Whitman's Poetics and the Unity of 'Calamus'." American Literature 46: 4 (January, 1975), 482-494.

Hunt, Russell A., Thomas Parkhill, James A. Reither, and Douglas Vipond. "Writing Under the Curriculum: Learning to Write by Using Writing to Teach." Panel Presentation, Conference on College Composition and Communication, St. Louis, Missouri, March 1988.

Hunt, Russell A., and Douglas Vipond. "Evaluations in Literary Reading." TEXT 6:1 (1986), 53-71.

Johnston, Peter H., and Peter N. Winograd. "Passive Failure in Reading." Journal of Reading Behavior 17: 4 (1985), 279-301.

Kamuf, Peggy. "Floating Authorship." Diacritics (Winter 1986),3-13.

Kaye, Kenneth. The Mental and Social Life of Babies: How Parents Create Persons. Chicago: University of Chicago Press, 1982.

Kintsch, Walter A., and Douglas Vipond. "Reading Comprehension and Readability in Educational Practice and Psychological Theory." In Perspectives on Memory Research: Essays in Honor of Uppsala University's 500th Anniversary, ed. Lars Goran Nilsson. Hillsdale, New Jersey: Lawrence Erlbaum Associates, 1979, 329-365.

Krashen, Stephen D. Second Language Acquisition and Second Language Learning. Oxford: Pergamon, 1981.

Labov, William. Language in the Inner City: Studies in the Black English Vernacular. Philadelphia: University of Pennsylvania Press, 1972.

LaBerge, David, and S. Jay Samuels. Basic Processes in Reading: Perception and Comprehension. Hillsdale, New Jersey: Lawrence Erlbaum Associates, 1977.

LaBerge, David, and S. Jay Samuels. "Toward a Theory of Automatic Information Processing in Reading." Cognitive Psychology 6 (1974), 293-323. Repr. in Theoretical Models and Processes of Reading, Second Edition, ed. Harry Singer and Robert B. Ruddell. Newark, Delaware: International Reading Association, 1976, 548-579.

Larsen, Steen F, and Uffe Seilman. "Remindings During Reading Literature." Paper, International Congress for the Empirical Study of Literature, Siegen, West Germany, December 1987.

Le Guin, Ursula K. The Compass Rose. Toronto: Bantam Books, 1982.

Lentricchia, Frank. Criticism and Social Change. Chicago: University of Chicago Press, 1983.

Lincoln, Yvonna S., and Egon G. Guba. Naturalistic Inquiry. Beverly Hills: Sage Publications, 1985.

Luke, Allan. Literacy, Textbooks and Ideology: Postwar Literacy Instruction and the Mythology of Dick and Jane. London/Philadelphia: Falmer Press, 1988.

McPhee, John. "In Virgin Forest." The New Yorker (6 July, 1987), 21-23.

Mandler, Jean M., and Nancy S. Johnson. "Remembrance of Things Parsed: Story Structure and Recall." Cognitive Psychology 9 (1977), 111-151.

Meutsch, Dietrich. "Cognitive Processes in Reading Literary Texts: the Influence of Context, Goals, and Situations." Empirical Studies of the Arts 5: 2 (1987), 117-138.

Meutsch, Dietrich, "How to Do Thoughts with Words II: Degrees of Consciousness During the Comprehension of Literature and Exposition with Different Types of Readers." Poetics (in press).

Meutsch, Dietrich, and Siegfried J. Schmidt. "On the Role of Conventions in Understanding Literary Texts." Poetics 14 (1985), 551-574.

Meyer, Bonnie J. F. The Organization of Prose and Its Effects on Memory. Amsterdam: North-Holland, 1975.

Mitchell, W. J. T., ed. Against Theory: Literary Studies and the New Pragmatism. Chicago: University of Chicago Press, 1986.

Newman, Judith M. Whole Language: Theory in Use. Portsmouth, New Hampshire: Heinemann, 1985.

Parkhill, Thom. "Inkshedding in Religion Studies: Underwriting Collaboration." Inkshed 7:4 (September 1988), 1-4.

Polanyi, Livia. Telling the American Story: A Structural and Cultural Analysis of Conversational Storytelling. Norwood, New Jersey: Ablex, 1985.

Polanyi, Livia. "So What's the Point?" Semiotica 25: 3/4 (1979), 207-241.

Pratt, Mary Louise. Toward a Speech-Act Theory of Literary Discourse. Bloomington: Indiana University Press, 1977.

Purves, Alan C., and Victoria Rippere. Elements of Writing About a Literary Work: A Study of Response to Literature [NCTE Committee on Research: Research Report No. 9]. Urbana: NCTE, 1968.

Radway, Janice. Reading the Romance: Women, Patriarchy, and Popular Literature. Chapel Hill: University of North Carolina Press, 1984.

Randall, Marilyn. "Context and Convention: The Pragmatics of Literariness." Poetics 14 (1985), 415-431.

Reither, James A. "Writing and Knowing: Toward Redefining the Writing Process." College English 47: 6 (Oct. 1985): 620-628. Repr. in The Writing Teacher's Sourcebook. 2nd ed. Ed. Gary Tate and Edward P. J. Corbett. New York: Oxford UP, 1988. 140-148.

Reither, James A. "The Writing Student as Researcher: Learning From Our Students." Paper delivered at The Miami University Conference on the Teacher as Researcher, Oxford, Ohio: Summer 1988.

Reither, James A., and Douglas Vipond. Writing as Collaboration. MS submitted for publication, 1988.

Rosenblatt, Louise M. Literature as Exploration. Third Edition. New York: Barnes and Noble, 1938.

Rosenblatt, Louise M. "Towards a Transactional Theory of Reading." Journal of Reading Behavior 1:1 (Winter 1969), 31-49.

Rosenblatt, Louise M. "Transaction Versus Interaction: A Terminological Rescue Operation." Research in the Teaching of English 19 (1985), 96-107.

Rubin, D. C. "Issues of Regularity and Control: Confessions of a Regularity Freak." In Everyday Cognition in Adult and Later Life, ed. L. W. Poon, D. C. Rubin, and B. A. Wilson. New York: Cambridge University Press, in press.

Rumelhart, David E. "Notes on a Schema for Stories." In Representation and Understanding: Studies in Cognitive Science, ed. Daniel G. Bobrow and Allan M. Collins. New York: Academic Press, 1975.

Sanguinete, Eduardo. "The Sociology of the Avant-Garde." Litterature et Societe, Institut de Sociologie, Universite de Bruxelles, 1967, 11-18; Repr., tr. Petra Morrison, in Sociology of Literature and Drama, ed. Elizabeth and Tom Burns. Harmondsworth: Penguin, 1973, 389-397.

Schmidt, Siegfried J. Die Emergenz des Literatursystems im 18. Jahrhundert MS, Siegen: Institut fur Empirische Literatur- und Medienforschung, 1989.

Smith, Frank. Understanding Reading: A Psycholinguistic Analysis of Reading and Learning to Read. Third Edition. New York: Holt, Rinehart and Winston, 1982.

Tannen, Deborah. Conversational Style: Analyzing Talk Among Friends. Norwood, New Jersey: Ablex, 1984.

Thompson, Kent. Leaping Up, Sliding Away. Fredericton, New Brunswick: Fiddlehead Poetry Books and Goose Lane Editions, 1986.

Vipond, Douglas, and Russell A. Hunt. "Literary Processing and Response as Transaction: Evidence for the Contribution of Readers, Text, and Situations." Comprehension of Literary Discourse: Results and Problems of Interdisciplinary Approaches, ed. Dietrich Meutsch and Reinhold Viehoff. Berlin: DeGruyter, 1988. [155]-174.

Vipond, Douglas, and Russell A. Hunt. "Point-Driven Understanding: Pragmatic and Cognitive Dimensions of Literary Reading." Poetics 13 (June 1984). 261-277.

Vipond, Douglas, Russell A. Hunt, James Jewett, and James A. Reither. "Making Sense of Reading." Becoming Readers and Writers During Adolescence and Adulthood, ed. Richard Beach and Susan Hynds. Norwood, New Jersey: Ablex, [in press].

Vygotsky, Lev S. Thought and Language. Translation newly edited and revised by Alex Kozulin. Cambridge, Massachusetts: MIT Press, 1986.

Wertsch, James V. Vygotsky and the Social Formation of Mind. Cambridge, Harvard University Press, 1986.

Wimsatt, William K., and Monroe C. Beardsley. "The Intentional Fallacy." The Verbal Icon: Studies in the Meaning of Poetry, by W. K. Wimsatt. University of Kentucky Press, 1954. [3]-18.