The transactional theory of reading and writing (Louise M. Rosenblatt)

by northtrap

Rosenblatt, Louise M. (2004). The transactional theory of reading and writing. Theoretical Models and Processess of Reading, 5th edition, Robert B. Ruddell, & Norman J. Unrau, editors, International Reading Association, article 48, 1363–1398.
The Transactional Theory of Reading and Writing
Louise M. Rosenblatt
Terms such as the reader are somewhat misleading, though convenient, fictions. There is no such thing as a generic reader or a generic literary work; there are in reality only the potential millions of individual readers of individual literary works…. The reading of any work of literature is, of necessity, an individual and unique occurrence involving the mind and emotions of some particular reader. (Rosenblatt, 1938/1983)
That statement, first published in Literature as Exploration in 1938, seems especially important to reiterate at the beginning of a presentation of a “the- oretical model” of the reading process. A theoretical model by definition is an abstraction, or a generalized pattern devised in order to think about a subject. Hence, it is essential to recognize that, as I concluded, we may generalize about similarities among such events, but we cannot evade the realization that there are actually only innumerable separate transactions between readers and texts.
As I sought to understand how we make the meanings called novels, po- ems, or plays, I discovered that I had developed a theoretical model that covers all modes of reading. Ten years of teaching courses in literature and composi- tion had preceded the writing of that statement. This had made possible obser- vation of readers encountering a wide range of “literary” and “nonliterary” texts, discussing them, keeping journals while reading them, and writing spontaneous reactions and reflective essays. And decades more of such observation preceded the publication of The Reader, the Text, the Poem (Rosenblatt, 1978), the fullest presentation of the theory and its implications for criticism.
Thus, the theory emerges from a process highly appropriate to the prag- matist philosophy it embodies. The problem arose in the context of a practical classroom situation. Observations of relevant episodes led to the hypotheses that constitute the theory of the reading process, and these have in turn been ap- plied, tested, confirmed, or revised in the light of further observation.
From Ruddell, R.B., Ruddell, M.R., & Singer, H. (Eds.), Theoretical Models and Processes of Reading (4th ed., pp. 1057–1092). Copyright © 1994 by the International Reading Association.

Fortunately, while specializing in English and comparative literature, I was in touch with the thinking on the forefront of various disciplines. The inter- pretation of these observations of readers’ reading drew on a number of different perspectives—literary and social history, philosophy, aesthetics, linguistics, psy- chology, and sociology. Training in anthropology provided an especially impor- tant point of view. Ideas were developed that in some instances have only recently become established. It seems necessary, therefore, to begin by setting forth some of the basic assumptions and concepts that undergird the transactional theory of the reading process. This in turn will involve presentation of the transactional view of the writing process and the relationship between author and reader.
The Transactional Paradigm
The terms transaction and transactional are consonant with a philosophic posi- tion increasingly accepted in the 20th century. A new paradigm in science (Kuhn, 1970) has required a change in our habits of thinking about our relationship to the world around us. For 300 years, Descartes’ dualistic view of the self as distinct from nature sufficed, for example, for the Newtonian paradigm in physics. The self, or “subject,” was separate from the “object” perceived. “Objective” facts, completely free of subjectivity, were sought, and a direct, immediate perception of “reality” was deemed possible. Einstein’s theory and the developments in sub- atomic physics revealed the need to acknowledge that, as Neils Bohr (1959) ex- plained, the observer is part of the observation—human beings are part of nature. Even the physicists’ facts depend to some extent on the interests, hypotheses, and technologies of the observer. The human organism, it became apparent, is ulti- mately the mediator in any perception of the world or any sense of “reality.”
John Dewey’s pragmatist epistemology fitted the new paradigm. Hence, Dewey joined with Arthur F. Bentley to work out a new terminology in Knowing and the Known (1949). They believed the term interaction was too much asso- ciated with the old positivistic paradigm, with each element or unit being pre- defined as separate, as “thing balanced against thing,” and their “interaction” studied. Instead, they chose transaction to imply “unfractured observation” of the whole situation. Systems of description and naming “are employed to deal with aspects and phases of action, without final attribution to ‘elements’ or presump- tively detachable or independent ‘entities,’ ‘essences,’ or ‘realities’” (p. 108). The knower, the knowing, and the known are seen as aspects of “one process.” Each element conditions and is conditioned by the other in a mutually constituted sit- uation (cf. Rosenblatt, 1985b).
The new paradigm requires a break with entrenched habits of thinking. The old stimulus–response, subject–object, individual–social dualisms give way to
1364 Rosenblatt
recognition of transactional relationships. The human being is seen as part of nature, continuously in transaction with an environment—each one conditions the other. The transactional mode of thinking has perhaps been most clearly as- similated in ecology. Human activities and relationships are seen as transactions in which the individual and social elements fuse with cultural and natural ele- ments. Many current philosophy writers may differ on metaphysical implications but find it necessary to come to terms with the new paradigm.1
The transactional concept has profound implications for understanding language. Traditionally, language has been viewed as primarily a self-contained system or code, a set of arbitrary rules and conventions that is manipulated as a tool by speakers and writers or imprints itself on the minds of listeners and readers. Even when the transactional approach has been accepted, this deeply ingrained way of thinking continues to function, tacitly or explicitly, in much theory, research, and teaching involving texts.2
The view of language basic to the transactional model of reading owes much to the philosopher John Dewey but even more to his contemporary Charles Sanders Peirce, who is recognized as the U.S. founder of the field of semiotics or semiology, the study of verbal and nonverbal signs. Peirce provided concepts that differentiate the transactional view of language and reading from structural- ist and poststructuralist (especially deconstructionist) theories. These reflect the influence of another great semiotician, the French linguist Ferdinand de Saussure (Culler, 1982).
Saussure (1972) differentiated actual speech (parole) from the abstractions of the linguists (langue), but he stressed the arbitrary nature of signs and minimized the referential aspect. Even more important was his dyadic formulation of the re- lationship between “signifier and signified,” or between words and concept. These emphases fostered a view of language as an autonomous, self-contained system (Rosenblatt, 1993).
In contrast, Peirce (1933, 1935) offered a triadic formulation. “A sign,” Peirce wrote, “is in conjoint relation to the thing denoted and to the mind….” The “sign is related to its object only in consequence of a mental association, and depends on habit” (Vol. 3, para. 360). The triad constitutes a symbol. Peirce repeatedly refers to the human context of meaning. Because he evidently did not want to reinforce the notion of “mind” as an entity, he typically phrased the “conjoint” linkage as among sign, object, and “interpretant,” which should be understood as a mental operation rather than an entity (6.347). Peirce’s triadic model firmly grounds language in the transactions of individual human beings with their world.
Recent descriptions of the working of the brain by neurologists and other sci- entists seem very Peircean. Although they are dealing with a level not essential to
The Transactional Theory of Reading and Writing 1365
our theoretical purposes, they provide an interesting reinforcement. “Many lead- ing scientists, including Dr. Francis Crick, think that the brain creates unified cir- cuits by oscillating distant components at a shared frequency” (Appenzeller, 1990, pp. 6–7). Neurologists speak of “a third-party convergence zone [which seems to be a neurological term for Peirce’s interpretant] that mediates between word and concept convergence zones” (Damasio, 1989, pp. 123–132). Studies of children’s acquisition of language support the Peircean triad, concluding that a vocalization or sign becomes a word, a verbal symbol, when the sign and its object or referent are linked with the same “organismic state” (Werner & Kaplan, 1962, p. 18).
Though language is usually defined as a socially generated system of communication—the very bloodstream of any society—the triadic concept re- minds us that language is always internalized by a human being transacting with a particular environment. Vygotsky’s recognition of the social context did not prevent his affirming the individual’s role: The “sense of a word” is
the sum of all the psychological events aroused in our consciousness by the word. It is a dynamic, fluid, complex whole, which has several zones of unequal sta- bility. Meaning [i.e., reference] is only one of the zones of sense, the most sta- ble and precise zone. A word acquires its sense from the context in which it appears; in different contexts, it changes its sense. (1962, p. 46)
Vygotsky postulated “the existence of a dynamic system of meaning, in which the affective and the intellectual unite.” The earliest utterances of children evidently represent a fusion of “processes which later will branch off into refer- ential, emotive, and associative part processes” (Rommetveit, 1968, pp. 147, 167). The child learns to sort out the various aspects of “sense” associated with a sign, decontextualize it, and recognize the public aspect of language, the collective language system. This does not, however, eliminate the other dimensions of sense. A language act cannot be thought of as totally affective or cognitive, or as totally public or private (Bates, 1979, pp. 65–66).
Bates provides the useful metaphor of an iceberg for the total sense of a word to its user: The visible tip represents what I term the public aspect of mean- ing, resting on the submerged base of private meaning. Public designates us- ages or meanings that dictionaries list. Multiple meanings indicated for the same word reflect the fact that the same sign takes on different meanings at different times and in different linguistic or different personal, cultural, or social con- texts. In short, public refers to usages that some groups of people have developed and that the individual shares.
Note that public and private are not synonymous with cognitive and affec- tive. Words may have publicly shared affective connotations. The individual’s private associations with a word may or may not agree with its connotations for the group, although these connotations must also be individually acquired. Words
1366 Rosenblatt
necessarily involve for each person a mix of both public and private elements, the base as well as the tip of the semantic iceberg.
For the individual, then, the language is that part, or set of features, of the public system that has been internalized through that person’s experiences with words in life situations. “Lexical concepts must be shared by speakers of a common language…yet there is room for considerable individual difference in the details of any concept” (Miller & Johnson-Laird, 1976, p. 700). The residue of the individ- ual’s past transactions—in particular, natural and social contexts—constitutes what can be termed a linguistic–experiential reservoir. William James especially sug- gests the presence of such a cumulative experiential aura of language.
Embodying funded assumptions, attitudes, and expectations about language and about the world, this inner capital is all that each of us has to draw on in speak- ing, listening, writing, or reading. We “make sense” of a new situation or trans- action and make new meanings by applying, reorganizing, revising, or extending public and private elements selected from our personal linguistic–experiential reservoirs.
Linguistic Transactions
Face-to-face communication—such as a conversation in which a speaker is ex- plaining something to another person—can provide a simplified example of the transactional nature of all linguistic activities. A conversation is a temporal activ- ity, a back-and-forth process. Each person has come to the transaction with an in- dividual history, manifested in what has been termed a linguistic–experiential reservoir. The verbal signs are the vibrations in the air caused by a speaker. Both speaker and addressee contribute throughout to the spoken text (even if the lis- tener remains silent) and to the interpretations that it calls forth as it progresses. Each must construct some sense of the other person. Each draws on a particular linguistic–experiential reservoir. The specific situation, which may be social and personal, and the setting and occasion for the conversation in themselves provide clues or limitations as to the general subject or framework and hence to the refer- ences and implications of the verbal signs. The speaker and addressee both pro- duce further delimiting cues through facial expressions, tones of voice, and gestures. In addition to such nonverbal indications of an ongoing mutual inter- pretation of the text, the listener may offer questions and comments. The speaker thus is constantly being helped to gauge and to confirm, revise, or expand the text. Hence, the text is shaped transactionally by both speaker and addressee.
The opening words of a conversation, far from being static, by the end of the interchange may have taken on a different meaning. And the attitudes, the state of mind, even the manifest personality traits, may have undergone change. Moreover, the spoken text may be interpreted differently by each of the conversationalists.
The Transactional Theory of Reading and Writing 1367
But how can we apply the conversation model of transaction to the rela- tionship between writers and readers, when so many of the elements that con- tribute to the spoken transaction are missing—physical presence, timing, actual setting, nonverbal behaviors, tones of voice, and so on? The signs on the page are all that the writer and the reader have to make up for the absence of these other elements. The reader focuses attention on and transacts with an element in the en- vironment, namely the signs on the page, the text.
Despite all the important differences noted above, speech, writing, and read- ing share the same basic process—transacting through a text. In any linguistic event, speakers and listeners and writers and readers have only their linguistic– experiential reservoirs as the basis for interpretation. Any interpretations or new meanings are restructurings or extensions of the stock of experiences of language, spoken and written, brought to the task. In Peircean terms, past linkages of sign, object, and interpretant must provide the basis for new linkages, or new structures of meaning. Instead of an interaction, such as billiard balls colliding, there has been a transaction, thought of rather in terms of reverberations, rapid oscillations, blendings, and mutual conditionings.
Selective Attention
William James’s concept of “selective attention” provides an important insight into this process. During the first half of this century, a combination of behav- iorism and positivism led to neglect of the concept, but since the 1970s psychol- ogists have reasserted its importance (Blumenthal, 1977; Myers, 1986). James (1890) tells us that we are constantly engaged in a “choosing activity,” which he terms “selective attention” (I.284). We are constantly selecting out of the stream, or field, of consciousness “by the reinforcing and inhibiting agency of atten- tion” (I.288). This activity is sometimes termed “the cocktail party phenome- non”: In a crowded room where many conversations are in progress, we focus our attention on only one of them at a time, and the others become a background hum. We can turn our selective attention toward a broader or narrower area of the field. Thus, while language activity implies an intermingled kinesthetic, cogni- tive, affective, associational matrix, what is pushed into the background or sup- pressed and what is brought into awareness and organized into meaning depend on where selective attention is focused.
The transactional concept will prevent our falling into the error of envis- aging selective attention as a mechanical choosing among an array of fixed enti- ties rather than as a dynamic centering on areas or aspects of the contents of consciousness. The linguistic reservoir should not be seen as encompassing ver- bal signs linked to fixed meanings, but as a fluid pool of potential triadic sym- bolizations. Such residual linkages of sign, signifier, and organic state, it will be seen, become actual symbolizations as selective attention functions under the shaping influence of particular times and circumstances.
1368 Rosenblatt
In the linguistic event, any process also will be affected by the physical and emotional state of the individual, for example, by fatigue or stress. Attention may be controlled or wandering, intense or superficial. In the discussion that fol- lows, it will be assumed that such factors enter into the transaction and affect the quality of the process under consideration.
The paradoxical situation is that the reader has only the black marks on the page as the means of arriving at a meaning—and that meaning can be con- structed only by drawing on the reader’s own personal linguistic and life experi- ences. Because a text must be produced by a writer before it can be read, logic might seem to dictate beginning with a discussion of the writing process. It is true that the writer seeks to express something, but the purpose is to communicate with a reader (even if it is only the writer wishing to preserve some thought or ex- perience for future reference). Typically, the text is intended for others. Some sense of a reader or at least of the fact that the text will function in a reading process thus is implicit in the writing process. Hence, I shall discuss the reading process first, then the writing process. Then, I shall broach the problems of com- munication and validity of interpretation before considering implications for teaching and research.
The Reading Process
Transacting With the Text
The concepts of transaction, the transactional nature of language, and selective attention now can be applied to analysis of the reading process. Every reading act is an event, or a transaction involving a particular reader and a particular pattern of signs, a text, and occurring at a particular time in a particular context. Instead of two fixed entities acting on one another, the reader and the text are two aspects of a total dynamic situation. The “meaning” does not reside ready-made “in” the text or “in” the reader but happens or comes into being during the transac- tion between reader and text.
The term text in this analysis denotes, then, a set of signs capable of being interpreted as verbal symbols. Far from already possessing a meaning that can be imposed on all readers, the text actually remains simply marks on paper, an ob- ject in the environment, until some reader transacts with it. The term reader im- plies a transaction with a text; the term text implies a transaction with a reader. “Meaning” is what happens during the transaction; hence, the fallacy of think- ing of them as separate and distinct entities instead of factors in a total situation.
The notion that the marks in themselves possess meaning is hard to dis- pel. For example, pain for a French reader will link up with the concept of bread and for an English reader with the concept of bodily or mental suffering. A sen- tence that Noam Chomsky (1968, p. 27) made famous can help us realize that not
The Transactional Theory of Reading and Writing 1369
even the syntax is inherent in the signs of the text but depends on the results of particular transactions: Flying planes can be dangerous.
Actually, only after we have selected a meaning can we infer a syntax from it. Usually, factors entering into the total transaction, such as the context and read- er’s purpose, will determine the reader’s choice of meaning. Even if the reader rec- ognizes the alternative syntactic possibilities, these factors still prevail. This casts doubt on the belief that the syntactical level, because it is lower or less complex, necessarily always precedes the semantic in the reading process. The transaction- al situation suggests that meaning implies syntax and that a reciprocal process is going on in which the broader aspects guiding choices are actively involved.
Here we see the difference between the physical text, defined as a pattern of signs, and what is usually called “the text,” a syntactically patterned set of ver- bal symbols. This actually comes into being during the transaction with the signs on the page.
When we see a set of such marks on a page, we believe that it should give rise to some more or less coherent meaning. We bring our funded experience to bear. Multiple inner alternatives resonate to the signs. Not only the triadic link- ages with the signs but also certain organismic states, or certain ranges of feeling, are stirred up in the linguistic–experiential reservoir. From these activated areas, selective attention—conditioned, as we have seen, by multiple physical, person- al, social, and cultural factors entering into the situation—picks out elements that will be organized and synthesized into what constitutes “meaning.” Choices have in effect probably been made simultaneously, as the various “levels” transact, conditioning one another, so to speak.
Reading is, to use James’s phrase, a “choosing activity.” From the very be- ginning, and often even before, some expectation, some tentative feeling, idea, or purpose, no matter how vague at first, starts the reading process and develops into the constantly self-revising impulse that guides selection, synthesis, and or- ganization. The linguistic–experiential reservoir reflects the reader’s cultural, so- cial, and personal history. Past experience with language and with texts provides expectations. Other factors are the reader’s present situation and interests. Perusing the unfolding text in the light of past syntactic and semantic experience, the reader seeks cues on which to base expectations about what is forthcoming. The text as a verbal pattern, we have seen, is part of what is being constructed. Possibilities open up concerning the general kind of meaning that may be developing, affect- ing choices in diction, syntax, and linguistic and literary conventions.
As the reader’s eyes move along the page, the newly evoked symbolizations are tested for whether they can be fitted into the tentative meanings already con- structed for the preceding portion of the text. Each additional choice will signal certain options and exclude others, so that even as the meaning evolves, the se- lecting, synthesizing impulse is itself constantly shaped and tested. If the marks on the page evoke elements that cannot be assimilated into the emerging synthesis,
1370 Rosenblatt
the guiding principle or framework is revised; if necessary, it is discarded and a complete rereading occurs. New tentative guidelines, new bases for a hypothetical structure, may then present themselves. Reader and text are involved in a complex, nonlinear, recursive, self-correcting transaction. The arousal and fulfillment—or frustration and revision—of expectations contribute to the construction of a cu- mulative meaning. From a to-and-fro interplay between reader, text, and context emerges a synthesis or organization, more or less coherent and complete. This meaning, this “evocation,” is felt to correspond to the text.
Precisely because for experienced readers so much of the reading process is, or should be, automatic, aspects of the reading process tend to be described in impersonal, mechanistic terms. Psychologists are rightfully concerned with learning as much as possible about what goes on between the reader’s first visu- al contact with the marks on the page and the completion of what is considered an interpretation of them. A number of different levels, systems, and strategies have been analytically designated, and research has been directed at clarifying their nature. These can be useful, but from a transactional point of view, it is important to recognize their potentialities and their limitations. A mechanistic analogy or metaphor lends itself especially to analyses of literal reading of sim- ple texts. Results need to be cautiously interpreted. Recognizing the essential nature of both reader and text, the transactional theory requires an underlying metaphor of organic activity and reciprocity.
The optical studies of Adelbert Ames (1955) and the Ames–Cantril “trans- actional psychology” (Cantril & Livingston, 1963), which also derived its name from Dewey and Bentley’s Knowing and the Known (1949), deserve first mention in this regard. These experiments demonstrated that perception depends much on the viewer’s selection and organization of visual cues according to past experi- ence, expectations, needs, and interests. The perception may be revised through continued transactions between the perceiver and the perceived object.
F.C. Bartlett’s theory of Remembering (1932) (which I regret having dis- covered even later than did his fellow scientists) and his term schema are often called on to explain psychological processes even broader than his special field. It is not clear, however, that those who so readily invoke his schema concept are heeding his fears about a narrow, static usage of the term. Rejecting the image of a warehouse of unchanging items as the metaphor for schemata, he empha- sized rather “active, developing patterns”—“constituents of living, momentary settings belonging to the organism” (Bartlett, 1932, p. 201). His description of the “constructive character of remembering,” his rejection of a simple mechanical linear process, and his concepts of the development and continuing revision of schemata all have parallels in the transactional theory of linguistic events. His recognition of the influence of both the interests of the individual and the social context on all levels of the process also seems decidedly transactional.
The Transactional Theory of Reading and Writing 1371
The Reader’s Stance
The broad outline of the reading process sketched thus far requires further elab- oration. An important distinction must be made between the operations that pro- duce the meaning, say, of a scientific report and the operations that evoke a literary work of art. Neither contemporary reading theory nor literary theory has done justice to such readings, nor to the fact that they are to be understood as rep- resenting a continuum rather than an opposition. The tendency generally has been to assume that such a distinction depends entirely on the texts involved. The char- acter of the “work” has been held to inhere entirely in the text. But we cannot look simply at the text and predict the nature of the work. We cannot assume, for instance, that a poem rather than an argument about fences will be evoked from the text of Frost’s Mending Wall or that a novel rather than sociological facts about Victorian England will be evoked from Dickens’s Great Expectations. Advertisements and newspaper reports have been read as poems. Each alterna- tive represents a different kind of selective activity, a different kind of relationship, between the reader and the text.
Essential to any reading is the reader’s adoption, conscious or uncon- scious, of what I have termed a stance guiding the “choosing activity” in the stream of consciousness. Recall that any linguistic event carries both public and private aspects. As the transaction with the printed text stirs up elements of the linguistic–experiential reservoir, the reader adopts a selective attitude or stance, bringing certain aspects into the center of attention and pushing others into the fringes of consciousness. A stance reflects the reader’s purpose. The sit- uation, the purpose, and the linguistic–experiential equipment of the reader as well as the signs on the page enter into the transaction and affect the extent to which public and private meanings and associations will be attended to.
The Efferent–Aesthetic Continuum
The reading event must fall somewhere in a continuum, determined by whether the reader adopts what I term a predominantly aesthetic stance or a predominantly efferent stance. A particular stance determines the proportion or mix of public and private elements of sense that fall within the scope of the reader’s selective attention. Or, to recall Bates’s metaphor, a stance results from the degree and scope of attention paid respectively to the tip and to the base of the iceberg. Such differences can be represented only by a continuum, which I term the efferent–aesthetic continuum.
The Efferent Stance
The term efferent (from the Latin efferre, to carry away) designates the kind of reading in which attention is centered predominantly on what is to be extracted and retained after the reading event. An extreme example is a man who has ac- cidentally swallowed a poisonous liquid and is rapidly reading the label on the
1372 Rosenblatt
bottle to learn the antidote. Here, surely, we see an illustration of James’s point about selective attention and our capacity to push into the periphery of awareness or ignore those elements that do not serve our present interests. The man’s at- tention is focused on learning what is to be done as soon as the reading ends. He concentrates on what the words point to, ignoring anything other than their barest public referents, constructing as quickly as possible the directions for future ac- tion. These structured ideas are the evocation felt to correspond to the text.
Reading a newspaper, textbook, or legal brief would usually provide a sim- ilar, though less extreme, instance of the predominantly efferent stance. In ef- ferent reading, then, we focus attention mainly on the public “tip of the iceberg” of sense. Meaning results from abstracting out and analytically structuring the ideas, information, directions, or conclusions to be retained, used, or acted on after the reading event.
The Aesthetic Stance
The predominantly aesthetic stance covers the other half of the continuum. In this kind of reading, the reader adopts an attitude of readiness to focus attention on what is being lived through during the reading event. The term aesthetic was chosen because its Greek source suggested perception through the senses, feelings, and intuitions. Welcomed into awareness are not only the public referents of the ver- bal signs but also the private part of the “iceberg” of meaning: the sensations, im- ages, feelings, and ideas that are the residue of past psychological events involving those words and their referents. Attention may include the sounds and rhythms of the words themselves, heard in “the inner ear” as the signs are perceived.
The aesthetic reader pays attention to—savors—the qualities of the feel- ings, ideas, situations, scenes, personalities, and emotions that are called forth and participates in the tensions, conflicts, and resolutions of the images, ideas, and scenes as they unfold. The lived-through meaning is felt to correspond to the text. This meaning, shaped and experienced during the aesthetic transaction, constitutes “the literary work,” the poem, story, or play. This “evocation,” and not the text, is the object of the reader’s “response” and “interpretation,” both during and after the reading event.
Confusion about the matter of stance results from the entrenched habit of thinking of the text as efferent or aesthetic, expository or poetic, literary or non- literary, and so on. Those who apply these terms to texts should realize that they actually are reporting their interpretation of the writer’s intention as to what kind of reading the text should be given. The reader is free, however, to adopt either predominant stance toward any text. Efferent and aesthetic apply, then, to the writer’s and the reader’s selective attitude toward their own streams of con- sciousness during their respective linguistic events.
To recognize the essential nature of stance does not minimize the impor- tance of the text in the transaction. Various verbal elements—metaphor, stylistic
The Transactional Theory of Reading and Writing 1373
conventions or divergence from linguistic or semantic norms, even certain kinds of content—have been said to constitute the “poeticity” or “literariness” of a text. Such verbal elements, actually, do often serve as cues to the experienced reader to adopt an aesthetic stance. Yet it is possible to cite acknowledged literary works that lack one or all these elements. Neither reading theorists nor literary theorists have given due credit to the fact that none of these or any other arrangements of words could make their “literary” or “poetic” contribution without the reader’s pri- or shift of attention toward mainly the qualitative or experiential contents of con- sciousness, namely, the aesthetic stance.
The Continuum
The metaphorical nature of the term the stream of consciousness can be called on further to clarify the efferent–aesthetic continuum. We can image consciousness as a stream flowing through the darkness. Stance, then, can be represented as a mechanism lighting up—directing the attention to—different parts of the stream, selecting out objects that have floated to the surface in those areas and leaving the rest in shadow. Stance, in other words, provides the guiding orientation toward activating particular areas and elements of consciousness, that is, particular pro- portions of public and private aspects of meaning, leaving the rest at the dim pe- riphery of attention. Some such play of attention over the contents of what emerges into consciousness must be involved in the reader’s multifold choices from the linguistic–experiential reservoir.
Efferent and aesthetic reflect the two main ways of looking at the world, of- ten summed up as “scientific” and “artistic.” My redundant usage of “predomi- nantly” aesthetic or efferent underlines rejection of the traditional, binary, either–or tendency to see them as in opposition. The efferent stance pays more attention to the cognitive, the referential, the factual, the analytic, the logical, the quantitative aspects of meaning. And the aesthetic stance pays more attention to the sensuous, the affective, the emotive, the qualitative. But nowhere can we find on the one hand the purely public and on the other hand the purely private. Both of these aspects of meaning are attended to in different proportions in any linguistic event. One of the earliest and most important steps in any reading event, therefore, is the selection of either a predominantly efferent or a predominantly aesthetic stance toward the transaction with a text. Figure 1 indicates different readings by the same reader of the same text at different points on the efferent–aesthetic continuum. Other read- ers would probably produce readings that fall at other points on the continuum.
Although many readings may fall near the extremes, many others, per- haps most, may fall nearer the center of the continuum. Where both parts of the iceberg of meaning are more evenly balanced, confusion as to dominant stance is more likely and more counterproductive. It is possible to read efferently and as- sume one has evoked a poem, or to read aesthetically and assume one is arriv- ing at logical conclusions to an argument.
1374 Rosenblatt
The Efferent–Aesthetic Continuum
Reading or Writing Events


Public Aspects of Sense

Private Aspects of Sense

Proportion of Reader’s or Writer’s Selective Attention
Efferent Aesthetic Stance Stance
Any linguistic activity has both public (lexical, analytic, abstracting) and private (experiential, affective, associational) components. Stance is determined by the proportion of each component admitted into the scope of selective attention. The efferent stance draws mainly on the public aspect of sense; the aesthetic stance includes proportionally more of the experiential, private aspect.
Reading or writing events A and B fall into the efferent part of the continuum, with B admitting more private elements. Reading or writing events C and D both represent the aesthetic stance, with C according a higher proportion of attention to the public aspects of sense.
Also, it is necessary to emphasize that a predominant stance does not rule out fluctuations. Within a particular aesthetic reading, attention may at times turn from the experiential synthesis to efferent analysis, as the reader recognizes some technical strategy or passes a critical judgment. Similarly, in an efferent read- ing, a general idea may be illustrated or reinforced by an aesthetically lived through illustration or example. Despite the mix of private and public aspects of meaning in each stance, the two dominant stances are clearly distinguishable. No two readings, even by the same person, are identical. Still, someone else can read a text efferently and paraphrase it for us in such a way as to satisfy our ef- ferent purpose. But no one else can read aesthetically—that is, experience the evocation of—a literary work of art for us.

The Transactional Theory of Reading and Writing 1375
Because each reading is an event in particular circumstances, the same text may be read either efferently or aesthetically. The experienced reader usually ap- proaches a text alert to cues offered by the text and, unless another purpose inter- venes, automatically adopts the appropriate predominant stance. Sometimes the title suffices as a cue. Probably one of the most obvious cues is the arrangement of broad margins and uneven lines that signals that the reader should adopt the aesthetic stance and undertake to make a poem. The opening lines of any text are especially important from this point of view, for their signaling of tone, atti- tude, and conventional indications of stance to be adopted.
Of course, the reader may overlook or misconstrue the cues, or they may be confusing. And the reader’s own purpose, or schooling that indoctrinates the same undifferentiated approach to all texts, may dictate a different stance from the one the writer intended. For example, the student reading A Tale of Two Cities who knows that there will be a test on facts about characters and plot may be led to adopt a predominantly efferent stance, screening out all but the factual data. Similarly, readings of an article on zoology could range from analytic abstracting of factual content to an aesthetic savoring of the ordered structure of ideas, the rhythm of the sentences, and the images of animal life brought into consciousness.
Evocation, Response, Interpretation
The tendency to reify words is frequently represented by discussions centering on a title, say, Invisible Man or The Bill of Rights. These titles may refer to the text, as we have been using the word, that is, to the pattern of inscribed signs to be found in physical written or printed form. More often, however, the intended reference is to “the work.” But the work—ideas and experiences linked with the text—can be found only in individual readers’ reflections on the reading event, the evocation and responses to it during and after the reading event.
Thus far, we have focused on the aspects of the reading process centered on or- ganizing a structure of elements of consciousness construed as the meaning of the text. I term this the evocation to cover both efferent and aesthetic transactions. The evocation, the work, is not a physical “object,” but, given another sense of that word, the evocation can be an object of thought.
The Second Stream of Response
We must recognize during the reading event a concurrent stream of reactions to, and transactions with, the emerging evocation. Even as we are generating the evocation, we are reacting to it; this may in turn affect our choices as we pro- ceed with the reading. Such responses may be momentary, peripheral, or felt sim- ply as a general state, for example, an ambiance of acceptance or perhaps of
1376 Rosenblatt
confirmation of ideas and attitudes brought to the reading. Sometimes some- thing unexpected or contrary to prior knowledge or assumptions may trigger con- scious reflection. Something not prepared for by the preceding organization of elements may cause a rereading. The attention may shift from the evocation to the formal or technical traits of the text. The range of potential reactions and the gamut of degrees of intensity and articulateness depend on the interplay among the character of the signs on the page (the text), what the individual reader brings to it, and the circumstances of the transaction.
The various strands of response, especially in the middle ranges of the efferent–aesthetic continuum, are sometimes simultaneous, interacting, and inter- woven. They may seem actually woven into the texture of the evocation itself. Hence, one of the problems of critical reading is differentiation of the evocation corresponding to the text from the concurrent responses, which may be projec- tions from the reader’s a priori assumptions. Drawing the line between them is easier in theory than in the practice of any actual reading. The reader needs to learn to handle such elements of the reading experience. The problem takes on different forms in efferent and aesthetic reading.
Expressed Response
“Response” to the evocation often is designed as subsequent to the reading event. Actually, the basis is laid during the reading, in the concurrent second stream of reactions. The reader may recapture the general effect of this after the event and may seek to express it and to recall what in the evocation led to the response. Reflection on “the meaning” of even a simple text involves the recall, the reacti- vation of some aspects of the process carried on during the reading. “Interpretation” tends to be a continuation of this effort to clarify the evocation.
The account of the reading process thus far has indicated an organizing, synthesizing activity, the creation of tentative meanings, and their modification as new elements enter into the focus of attention. In some instances, the reader at some point simply registers a sense of having completed a sequential activity and moves on to other concerns. Sometimes a sense of the whole structure crys- tallizes by the close of the reading.
Expressed Interpretation
Actually, the process of interpretation that includes arriving at a sense of the whole has not been given enough attention in theories of reading, perhaps be- cause reading research has typically dealt with simple reading events. For the term interpret, dictionaries list, among others, several relevant meanings. One is “to set forth the meaning of; to elucidate, to explain.” Another is “to construe, or understand in a particular way.” A third is “to bring out the meaning of by per- formance (as in music).” These tend to reflect the traditional notion of “the mean- ing” as inherent in the text.
The Transactional Theory of Reading and Writing 1377
The transactional theory requires that we draw on all three of these usages to cover the way in which the term should be applied to the reading process. The evocation of meaning in transaction with a text is indeed interpretation in the sense of performance, and transactional theory merges this with the idea of interpreta- tion as individual construal. The evocation then becomes the object of interpreta- tion in the sense of elucidating or explaining. The expressed interpretation draws on all these aspects of the total transaction.
Interpretation can be understood as the effort to report, analyze, and ex- plain the evocation. The reader recalls the sensed, felt, thought evocation while at the same time applying some frame of reference or method of abstracting in or- der to characterize it, to find the assumptions or organizing ideas that relate the parts to the whole. The second stream of reactions will be recalled, and the rea- sons for them sought, in the evoked work or in prior assumptions and knowl- edge. The evocation and the concurrent streams of reaction may be related through stressing, for example, the logic of the structure of ideas in an efferent evocation or the assumptions about people or society underlying the lived through experience of the aesthetic reading.
Usually, interpretation is expressed in the efferent mode, stressing under- lying general ideas that link the signs of the text. Interpretation can take an aes- thetic form, however, such as a poem, a painting, music, dramatization, or dance.
Interpretation brings with it the question of whether the reader has pro- duced a meaning that is consonant with the author’s probable intention. Here we find ourselves moving from the reader–text transaction to the relationship between author and reader. The process that produces the text will be consid- ered before dealing with such matters as communication, validity of interpreta- tion, and the implications of the transactional theory for teaching and research.
The Writing Process
The Writing Transaction
Writers facing a blank page, like readers approaching a text, have only their in- dividual linguistic capital to draw on. For the writer, too, the residue of past ex- periences of language in particular situations provides the material from which the text will be constructed. As with the reader, any new meanings are restruc- turings or extensions of the stock of experiences the writer brings to the task. There is a continuing to-and-fro or transactional process as the writer looks at the page and adds to the text in the light of what has been written thus far.
An important difference between readers and writers should not be mini- mized, however. In the triadic sign–object–interpretant relationship, the reader has the physical pattern of signs to which to relate the symbolizations. The writer facing a blank page may start with only an organismic state, vague feelings and
1378 Rosenblatt
ideas that require further triadic definition before a symbolic configuration—a verbal text—can take shape.
Writing is always an event in time, occurring at a particular moment in the writer’s biography, in particular circumstances, under particular external as well as internal pressures. In short, the writer is always transacting with a personal, social, and cultural environment. Thus, the writing process must be seen as always embodying both personal and social, or individual and environmental, factors.
Given the Peircean triadic view of the verbal symbol, the more accessible the fund of organismically linked words and referents, the more fluent the writ- ing. This helps us place in perspective an activity such as free writing. Instead of treating it as a prescriptive “stage” of the writing process, as some seem to do, it should be seen as a technique for tapping the linguistic reservoir without be- ing hampered by anxieties about acceptability of subject, sequence, or mechan- ics. Especially for those inhibited by unfortunate past writing experiences, this can be liberating, a warm-up exercise for starting the juices flowing, so to speak, and permitting elements of the experiential stream, verbal components of mem- ory, and present concerns to rise to consciousness. The essential point is that the individual linguistic reservoir must be activated.
No matter how free and uninhibited the writing may be, the stream of im- ages, ideas, memories, and words is not entirely random; William James reminds us that the “choosing activity” of selective attention operates to some degree. Like the reader, the writer needs to bring the selective process actively into play, to move toward a sense of some tentative focus for choice and synthesis (Emig, 1983).
This directedness will be fostered by the writer’s awareness of the transac- tional situation: the context that initiates the need to write and the potential reader or readers to whom the text will presumably be addressed. Often in trial-and-error fashion, and through various freely flowing drafts, the writer’s sensitivity to such factors translates itself into an increasingly clear impulse that guides selective at- tention and integration. For the experienced writer, the habit of such awareness, monitoring the multifold decisions or choices that make up the writing event, is more important than any explicit preliminary statement of goals or purpose.
The Writer’s Stance
The concept of stance presented earlier in relation to reading is equally important for writing. A major aspect of the delimitation of purpose in writing is the adop- tion of a stance that falls at some point in the efferent–aesthetic continuum. The attitude toward what is activated in the linguistic–experiential reservoir mani- fests itself in the range and character of the verbal symbols that will “come to mind,” and to which the writer will apply selective attention. The dominant stance determines the proportion of public and private aspects of sense that will be in- cluded in the scope of the writer’s attention (see Figure 1).
The Transactional Theory of Reading and Writing 1379
In actual life, the selection of a predominant stance is not arbitrary but is a function of the circumstances, the writer’s motives, the subject, and the relation be- tween writer and prospective reader or readers. For example, someone who had been involved in an automobile collision would need to adopt very different stances in writing an account of the event for an insurance company and in describing it in a letter to a friend. The first would activate an efferent selective process, bringing into the center of consciousness and onto the page the public aspects, such as statements that could be verified by witnesses or by investigation of the terrain. In the letter to the friend, the purpose would be to share an experience. An aesthetic stance would bring within the scope of the writer’s attention the same basic facts, together with feelings, sensations, tensions, images, and sounds lived through dur- ing this brush with death. The selective process would favor words that matched the writer’s inner sense of the felt event and that also would activate in the prospec- tive reader symbolic linkages evoking a similar experience. Given different pur- poses, other accounts might fall at other points of the efferent–aesthetic continuum.
Purpose or intention should emerge from, or be capable of constructively en- gaging, the writer’s actual experiential and linguistic resources. Past experience need not be the limit of the writer’s scope, but the writer faced with a blank page needs “live” ideas—that is, ideas having a strongly energizing linkage with the linguistic–experiential reservoir. Purposes or ideas that lack the capacity to con- nect with the writer’s funded experience and present concerns cannot fully activate the linguistic reservoir and provide an impetus to thinking and writing.
A personally grounded purpose develops and impels movement forward. Live ideas growing out of situations, activities, discussions, problems, or needs provide the basis for an actively selective and synthesizing process of making meaning. The quickened fund of images, ideas, emotions, attitudes, and tenden- cies to act offers the means of making new connections, for discovering new facets of the world of objects and events, in short, for thinking and writing creatively.
Writing About Texts
When a reader describes, responds to, or interprets a work—that is, speaks or writes about a transaction with a text—a new text is being produced. The impli- cations of this fact in terms of process should be more fully understood. When the reader becomes a writer about a work, the starting point is no longer the physi- cal text, the marks on the page, but the meaning or the state of mind felt to cor- respond to that text. The reader may return to the original text to recapture how it entered into the transaction but must “find words” for explaining the evoca- tion and the interpretation.
The reader-turned-writer must once again face the problem of choice of stance. In general, the choice seems to be the efferent stance. The purpose is mainly to explain, analyze, summarize, and categorize the evocation. This is usu- ally true even when the reading has been predominantly aesthetic and a literary
1380 Rosenblatt
work of art is being discussed. However, the aesthetic stance might be adopted in order to communicate an experience expressing the response or the interpreta- tion. An efferent reading of, for example, the U.S. Declaration of Independence might lead to a poem or a story. An aesthetic reading of the text of a poem might also lead, not to an efferently written critical essay, but to another poem, a paint- ing, or a musical composition.
The translator of a poem is a clear example of the reader-turned-writer, being first a reader who evokes an experience through a transaction in one language and then a writer who seeks to express that experience through a writing transaction in another language. The experiential qualities generated in a transaction with one lan- guage must now be communicated to—evoked by—readers who have a different linguistic–experiential reservoir, acquired in a different culture.
Authorial Reading
Thus far, we have been developing parallels between the ways in which readers and writers select and synthesize elements from the personal linguistic reser- voir, adopt stances that guide selective attention, and build a developing selective purpose. Emphasis has fallen mainly on similarities in composing structures of meaning related to texts. If readers are in that sense also writers, it is equally— and perhaps more obviously—true that writers also must be readers. At this point, however, some differences within the parallelisms begin to appear.
The writer, it is generally recognized, is the first reader of the text. Note an obvious, though neglected, difference: While readers transact with a writer’s finished text, writers first read the text as it is being inscribed. Because both read- ing and writing are recursive processes carried on over a period of time, their very real similarities have masked a basic difference. The writer will often reread the total finished text, but, perhaps more important, the writer first reads and carries on a spiral, transactional relationship with the very text emerging on the page. This is a different kind of reading. It is authorial—a writer’s reading. It should be seen as an integral part of the composing process. In fact, it is necessary to see that writing, or composing, a text involves two kinds of authorial reading, which I term expression oriented and reception oriented.
Expression-Oriented Authorial Reading
As a reader’s eyes move along a printed text, the reader develops an organizing principle or framework. The newly evoked symbolizations are tested for whether they can be fitted into the tentative meanings already constructed for the preced- ing portion of the text. If the new signs create a problem, this may lead to a revi- sion of the framework or even to a complete rereading of the text and restructuring of the attributed meaning.
The Transactional Theory of Reading and Writing 1381
The writer, like readers of another’s text, peruses the succession of verbal signs being inscribed on the page to see whether the new words fit the preced- ing text. But this is a different, expression-oriented reading, which should be seen as an integral part of the composing process. As the new words appear on the page, they must be tested, not simply for how they make sense with the preced- ing text but also against an inner gauge—the intention, or purpose. The emerging meaning, even if it makes sense, must be judged as to whether it serves or hinders the purpose, however nebulous and inarticulate, that is the motive power in the writing. Expression-oriented authorial reading leads to revision even during the earlier phases of the writing process.
The Inner Gauge
Most writers will recall a situation that may illustrate the operation of an “inner gauge.” A word comes to mind or flows from the pen and, even if it makes sense, is felt not to be right. One word after another may be brought into conscious- ness and still not satisfy. Sometimes the writer understands what is wrong with the word on the page—perhaps that it is ambiguous or does not suit the tone. But often the writer cannot articulate the reason for dissatisfaction. The tension simply disappears when “the right word” presents itself. When it does, a match between inner state and verbal sign has happened.
Such an episode manifests the process of testing against an inner touchstone. The French writer Gustave Flaubert with his search for le mot juste, the exact word, offers the analogy of the violinist who tries to make his fingers “reproduce precisely those sounds of which he has the inward sense” (1926, pp. 11, 47). The inner gauge may be an organic state, a mood, an idea, perhaps even a consciously constituted set of guidelines.
For the experienced writer, this kind of completely inner-oriented read- ing, which is integral to the composing process, depends on and nourishes an increasingly clear though often tacit sense of purpose, whether efferent or aes- thetic. The writer tries to satisfy a personal conception while also refining it. Such transactional reading and revision can go on throughout the writing event. There are indeed times when this is the only reading component—when one writes for oneself alone, to express or record an experience in a diary or journal, or per- haps to analyze a situation or the pros and cons of a decision.
Reception-Oriented Authorial Reading
Usually, however, writing is felt to be part of a potential transaction with other readers. At some point, the writer dissociates from the text and reads it through the eyes of potential readers; the writer tries to judge the meaning they would make in transaction with that pattern of signs. But the writer does not simply adopt the “eyes” of the potential reader. Again, a twofold operation is involved. The emerg-
1382 Rosenblatt
ing text is read to sense what others might make of it. But this hypothetical inter- pretation must also be checked against the writer’s own inner sense of purpose.
The tendency has been to focus on writing with an eye on the anticipated reader. My concern is to show the interplay between the two kinds of authorial reading and the need, consciously or automatically, to decide the degree of em- phasis on one or the other. The problem always is to find verbal signs likely to ac- tivate linkages in prospective readers’ linguistic reservoirs matching those of the writer. A poet may be faced with the choice between a personally savored ex- otic metaphor and one more likely to be within the experience of prospective readers. Or a science writer may have to decide whether highly detailed precision may be too complex for the general reader.
Writers must already have some hold on the first, expression-oriented kind of inner awareness if they are to benefit from the second reading through the eyes of others. The first becomes a criterion for the second. The experienced writer will probably engage in a synthesis, or rapid alternation, of the two kinds of authorial reading to guide the selective attention that filters out the verbal ele- ments coming to mind. When communication is the aim, revision should be based on such double criteria in the rereading of the text.
Communication Between Author and Readers
The reader’s to-and-fro process of building an interpretation becomes a form of transaction with an author persona sensed through and behind the text. The im- plied relationship is sometimes even termed “a contract” with the author. The closer their linguistic–experiential equipment, the more likely the reader’s inter- pretation will fulfill the writer’s intention. Sharing at least versions of the same language is so basic that it often is simply assumed. Other positive factors af- fecting communication are contemporary membership in the same social and cul- tural group, the same educational level, and membership in the same discourse community, such as academic, legal, athletic, literary, scientific, or theological. Given such similarities, the reader is more likely to bring to the text the prior knowledge, acquaintance with linguistic and literary conventions, and assump- tions about social situations required for understanding implications or allusions and noting nuances of tone and thought.
Yet, because each individual’s experience is unique, differences due to social, ethnic, educational, and personal factors exist, even with contemporaries. The reading of works written in another period bespeaks an inevitable differ- ence in linguistic, social, or cultural context. Here, especially, readers may agree on interpretations without necessarily assuming that their evocations from the text fit the author’s intention (Rosenblatt, 1978, p. 109ff).
Differences as to the author’s intention often lead to consultation of extra- textual sources. For works of the past especially, scholars call on systematic
The Transactional Theory of Reading and Writing 1383
methods of philological, biographical, and historical research to discover the per- sonal, social, and literary forces that shaped the writer’s intention. The contem- porary reception of the work also provides clues. Such evidence, even if it includes an author’s stated intention, still yields hypothetical results and cannot dictate our interpretation. We must still read the text to decide whether it supports the hypothetical intention. The reader is constantly faced with the responsibility of de- ciding whether an interpretation is acceptable. The question of validity of inter- pretation must be faced before considering implications for teaching and research.
Validity of Interpretation
The problem of validity of interpretation has not received much attention in read- ing theory or educational methodology. Despite the extraordinary extent of the reliance on testing in our schools, there seems to be little interest in clarifying the criteria that enter into evaluation of “comprehension.” Actual practice in the teaching of reading and in the instruments for testing of reading ability has evi- dently been tacitly based on, or at least has indoctrinated, the traditional as- sumption that there is a single determinate “correct” meaning attributable to each text. The stance factor, the efferent–aesthetic continuum, has especially been neg- lected; operationally, the emphasis has been on the efferent, even when “litera- ture” was involved.
The polysemous character of language invalidates any simplistic approach to meaning, creating the problem of the relationship between the reader’s inter- pretation and the author’s intention. The impossibility of finding a single ab- solute meaning for a text or of expecting any interpretation absolutely to reflect the writer’s intention is becoming generally recognized by contemporary theorists. “Intention” itself is not absolutely definable or delimitable even by the writer. The word absolute, the notion of a single “correct” meaning inherent “in” the text, is the stumbling block. The same text takes on different meanings in transactions with different readers or even with the same reader in different contexts or times.
Warranted Assertibility
The problem of the validity of any interpretation is part of the broader philo- sophical problem cited at the beginning of this piece. Perception of the world is always through the medium of individual human beings transacting with their worlds. In recent decades, some literary theorists, deriving their arguments from poststructuralist Continental writers and taking a Saussurean view of language as an autonomous system, have arrived at an extreme relativist position. They have developed a reading method that assumes all texts can be “deconstructed” to re- veal inner contradictions. Moreover, the language system and literary conventions are said to completely dominate author and reader, and agreement concerning inter- pretation simply reflects the particular “interpretive community” in which we find ourselves (Fish, 1980; Rosenblatt, 1991).
1384 Rosenblatt
Such extreme relativism is not, however, a necessary conclusion from the premise that absolutely determinate meaning is impossible. By agreeing on cri- teria of evaluation of interpretations, we can accept the possibility of alternative interpretations yet decide that some are more acceptable than others.
John Dewey, accepting the nonfoundationalist epistemological premises and foregoing the quest for absolutes, solved the scientists’ problem by his idea of “warranted assertibility” as the end of controlled inquiry (1938, pp. 9, 345). Given shared criteria concerning methods of investigation and kinds of evi- dence there can be agreement concerning the decision as to what is a sound in- terpretation of the evidence, or “a warranted assertion.” This is not set forth as permanent, absolute truth, but leaves open the possibility that alternative expla- nations for the same facts may be found, that new evidence may be discovered, or that different criteria or paradigms may be developed.
Although Dewey used primarily scientific interpretation or knowledge of the world based on scientific methods to illustrate warranted assertibility, he saw the concept as encompassing the arts and all human concerns. It can be applied to the problem of all linguistic interpretation (Rosenblatt, 1978, chap. 7; 1983, p. 151ff). Given a shared cultural milieu and shared criteria of validity of inter- pretation, we can, without claiming to have the single “correct” meaning of a text, agree on an interpretation. Especially in aesthetic reading, we may find that al- ternative interpretations meet our minimum criteria, and we can still be free to consider some interpretations superior to others.
In contrast to the notion of readers locked into a narrow “interpretive com- munity,” the emphasis on making underlying or tacit criteria explicit provides the basis not only for agreement but also for understanding tacit sources of dis- agreement. This creates the possibility of change in interpretation, acceptance of alternative sets of criteria, or revision of criteria. Such self-awareness on the part of readers can foster communication across social, cultural, and historical differ- ences between author and readers, as well as among readers (Rosenblatt, 1983).
In short, the concept of warranted assertibility, or shared criteria of validi- ty of interpretation in a particular social context, recognizes that some readings may satisfy the criteria more fully than others. Basic criteria might be (1) that the context and purpose of the reading event, or the total transaction, be consid- ered; (2) that the interpretation not be contradicted by, or not fail to cover, the full text, the signs on the page; and (3) that the interpretation not project meanings which cannot be related to signs on the page. Beyond these items arise criteria for interpretation and evaluation growing out of the whole structure of shared cul- tural, social, linguistic, and rhetorical assumptions.
Thus, we can be open to alternative readings of the text of Hamlet, but we also can consider some readings as superior to others according to certain explicit criteria, for example, complexity of intellectual and affective elements and nature of implicit value system. Such considerations permit comparison and “negotiation”
The Transactional Theory of Reading and Writing 1385
among different readers of the same text as well as clarification of differences in assumptions concerning what constitutes a valid interpretation (Rosenblatt, 1978, 1983). On the efferent side of the continuum, current discussions of alter- native criteria for interpretation of the U.S. Constitution provide another complex example.
Criteria for the Efferent–Aesthetic Continuum
Precisely because, as Figure 1 indicates, both public and private elements are present in all reading, the criteria of validity of interpretation differ for readings at various points on the efferent–aesthetic continuum. Because the predomi- nantly efferent interpretation must be publicly verifiable or justifiable, the crite- ria of validity rest primarily on the public, referential aspects of meaning and require that any affective and associational aspects not dominate. The criteria for the predominantly aesthetic reading call for attention to the referential, cog- nitive aspects but only as they are interwoven and colored by the private, affec- tive, or experiential aspects generated by the author’s patterns of signs. Especially in the middle ranges of the efferent–aesthetic continuum, it becomes important for writers to provide clear indications as to stance and for readers to be sensi- tive to the writer’s purpose and the need to apply relevant criteria.
“Literary” Aspects of Efferent Reading
In recent decades, in one scientific field after another, the opposition between sci- entific and “literary” writing has been found to be illusory. Writers in the natu- ral and social sciences have become aware of the extent to which they engage in semantic and syntactic practices that have usually been considered “literary” and that they, too, have been using narrative, metaphor, and other rhetorical de- vices. Examples are the importance of metaphor in writings about economics or the idea that the historian writes narrative and that he can never be completely ob- jective in selecting his facts. Sensitivity to sexist and racist tropes has increased awareness of the extent to which metaphor permeates all kinds of texts and, in- deed, all language. Sometimes the efferent–aesthetic distinction seems to be com- pletely erased (for example, the historian is sometimes said to write “fiction”).
It becomes necessary to recall that the stance reflecting the aesthetic or ef- ferent purpose, not the syntactic and semantic devices alone, determines the ap- propriate criteria. For example, in a treatise on economics or a history of the frontier, the criteria of validity of interpretation appropriate to their disciplines, which involve primarily verifiability and logic, would still apply. When an econ- omist remarks that “the scientists had better devise good metaphors and tell good stories” (McCloskey, 1985), the concept of a dominant stance becomes all the more essential. The criteria for “good” should be not only how vivid and ap- pealing the stories are but also how they gibe with logic and facts and what val- ue systems are implied.
1386 Rosenblatt
The relevance of the efferent–aesthetic continuum (Figure 1) may be illus- trated by the example of metaphor: The scientist speaks of the “wave” theory of light, and we focus on the technical concept at the extreme efferent end of the continuum. Shakespeare writes, “Like as the wave makes toward the pebbled shore / So do our minutes hasten to their end,” and our aesthetic attention to the feeling of inevitability of the succeeding waves enhances the feeling of the in- evitability of the passage of time in our lives. A political analysis suggested sur- rendering to the inevitability of fascism by calling it “the wave of the future…. There is no fighting it” (Lindbergh, 1940, p. 934). Despite the vividness of the metaphor, efferent attention should have remained dominant, applying the ef- ferent criterion. Did logic and factual evidence support the persuasive appeal?
Implications for Teaching
Reading and Writing: Parallelisms and Differences
Parallelisms between reading and writing processes have raised questions con- cerning their connections, especially in the classroom. The reading and writing processes both overlap and differ. Both reader and writer engage in constituting symbolic structures of meaning in a to-and-fro, spiral transaction with the text. They follow similar patterns of thinking and call on similar linguistic habits. Both processes depend on the individual’s past experiences with language in particular life situations. Both reader and writer therefore are drawing on past linkages of signs, signifiers, and organic states in order to create new symbolizations, new linkages, and new organic states. Both reader and writer develop a framework, prin- ciple, or purpose, however nebulous or explicit, that guides the selective attention and the synthesizing, organizing activities that constitute meaning. Moreover, every reading and writing act can be understood as falling somewhere on the efferent–aesthetic continuum and as being predominantly efferent or aesthetic.
The parallels should not mask the basic differences—the transaction that starts with a text produced by someone else is not the same as a transaction that starts with the individual facing a blank page. To an observer, two people perus- ing a typed page may seem to be doing the same thing (namely, “reading”). But if one of them is in the process of writing that text, different activities will be going on. The writer will be engaged in either expression-oriented or reception- oriented authorial reading. Moreover, because both reading and writing are root- ed in mutually conditioning transactions between individuals and their particular environments, a person may have very different experiences with the two activ- ities, may differ in attitudes toward them, and may be more proficient in one or the other. Writing and reading are sufficiently different to defeat the assumption that they are mirror images: The reader does not simply reenact the author’s
The Transactional Theory of Reading and Writing 1387
process. Hence, it cannot be assumed that the teaching of one activity automati- cally improves the student’s competence in the other.
Still, the parallels in the reading and writing processes described above and the nature of the transaction between author and reader make it reasonable to expect that the teaching of one can affect the student’s operations in the other. Reading, essential to anyone for intellectual and emotional enrichment, pro- vides the writer with a sense of the potentialities of language. Writing deepens the reader’s understanding of the importance of paying attention to diction, syntac- tic positions, emphasis, imagery, and conventions of genre. The fact that the sign–interpretant–object triad is, as Peirce said, dependent on habit indicates an even more important level of influence. Cross-fertilization will result from rein- forcement of linguistic habits and thinking patterns resulting from shared trans- actional processes of purposive selective attention and synthesis. How fruitful the interplay between the individual student’s writing and reading will be depends largely on the nature of the teaching and the educational context.
The Total Context
Here we return to our basic concept that human beings are always in transaction and in a reciprocal relationship with an environment, a context, a total situation. The classroom environment, or the atmosphere created by the teacher and stu- dents transacting with one another and the school setting, broadens out to include the whole institutional, social, and cultural context. These aspects of the trans- action are crucial in thinking about education and especially the “literacy prob- lem.” Because each individual’s linguistic–experiential reservoir is the residue of past transactions with the environment, such factors condition the sense of pos- sibilities, or the potential organizing frameworks or schema and the knowledge and assumptions about the world, society, human nature, that each brings to the transactions. Socioeconomic and ethnic factors, for example, influence patterns of behavior, ways of carrying out tasks, even understanding of such concepts as “story” (Heath, 1983). Such elements also affect the individual’s attitude to- ward self, toward the reading or writing activity, and toward the purpose for which it is being carried on.3
The transactional concept of the text always in relation either to author or reader in specific situations makes it untenable to treat the text as an isolated en- tity or to overemphasize either author or reader. Recognizing that language is not a self-contained system or static code on the one hand avoids the traditional ob- session with the product—with skills, techniques, and conventions, essential though they are—and, on the other, prevents a pendulum swing to overempha- sis on process or on the personal aspects.
Treatment of either reading or writing as a dissociated set of skills (though both require skills) or as primarily the acquisition of codes and conventions (though both involve them) inhibits sensitivity to the organic linkages of verbal signs and
1388 Rosenblatt
their objects. Manipulating syntactic units without a sense of a context that connects them into a meaningful relationship may in the long run be counterproductive.
Nor can the transactional view of the reading and writing processes be turned into a set of stages to be rigidly followed. The writer’s drafts and final texts—or the reader’s tentative interpretations, final evocation, and reflections—should be viewed as stopping points in a journey, as the outward and visible signs of a con- tinuing process in the passage from one point to the other. A “good” product, whether a well-written paper or a sound textual interpretation, should not be an end in itself—a terminus—but should be the result of a process that builds the strengths for further journeys or, to change the metaphor, for further growth. “Product” and “process” become interlocking concerns in nurturing growth.
Hence, the teaching of reading and writing at any developmental level should have as its first concern the creation of environments and activities in which students are motivated and encouraged to draw on their own resources to make “live” meanings. With this as the fundamental criterion, emphasis falls on strengthening the basic processes that we have seen to be shared by reading and writing. The teaching of one can then reinforce linguistic habits and semantic ap- proaches useful in the other. Such teaching, concerned with the ability of the in- dividual to generate meaning, will permit constructive cross-fertilization of the reading and writing (and speech) processes.
Enriching the individual’s linguistic–experiential reservoir becomes an un- derlying educational aim broader than the particular concern with either reading or writing. Especially in the early years, the linkage between verbal sign and ex- periential base is essential. The danger is that many current teaching practices may counteract the very processes presumably being taught. The organization of instruction, the atmosphere in the classroom, the kinds of questions asked, the ways of phrasing assignments, and the types of tests administered should be scrutinized from this point of view.
The importance of a sense of purpose, of a guiding principle of selection and organization in both writing and reading, is being increasingly recognized. The creation of contexts that permit purposive writing and reading can enable the student to build on past experience of life and language, to adopt the appropriate stance for selective attention, and to develop inner gauges or frameworks for choice and synthesis that produce new structures of live meaning.
Collaborative Interchange
In a favorable educational environment, speech is a vital ingredient of transac- tional pedagogy. Its importance in the individual’s acquisition of a linguistic– experiential capital is clear. It can be an extremely important medium in the class- room. Dialogue between teacher and students and interchange among students can foster growth and cross-fertilization in both the reading and writing processes. Such discussion can help students develop insights concerning transactions with
The Transactional Theory of Reading and Writing 1389
texts as well as metalinguistic understanding of skills and conventions in mean- ingful contexts.
Students’ achievement of insight into their own reading and writing process- es can be seen as the long-term justification for various curricular and teaching strategies. For example, writers at all levels can be helped to understand their trans- actional relationship to their readers by peer reading and discussion of texts. Their fellow students’ questions, varied interpretations, and misunderstandings dramatize the necessity of the writer’s providing verbal signs that will help readers gain re- quired facts, share relevant sensations or attitudes, or make logical transitions. Such insights make possible the second, reader-oriented authorial reading.
Similarly, group interchange about readers’ evocations from texts, whether of their peers or adult authors, can in general be a powerful means of stimulat- ing growth in reading ability and critical acumen. Readers become aware of the need to pay attention to the author’s words in order to avoid preconceptions and misinterpretations. When students share responses to transactions with the same text, they can learn how their evocations from the same signs differ, can return to the text to discover their own habits of selection and synthesis, and can become aware of, and critical of, their own processes as readers. Interchange about the problems of interpretation that a particular group of readers encounters and a col- laborative movement toward self-critical interpretation of the text can lead to the development of critical concepts and interpretive criteria. Such metalinguis- tic awareness is valuable to students as both readers and writers.
The teacher in such a classroom is no longer simply a conveyor of ready- made teaching materials and recorder of results of ready-made tests or a dis- penser of ready-made interpretations. Teaching becomes constructive, facilitating interchange, helping students to make their spontaneous responses the basis for raising questions and growing in the ability to handle increasingly complex reading transactions (Rosenblatt, 1983).4
The Student’s Efferent–Aesthetic Repertory
The efferent–aesthetic continuum, or the two basic ways of looking at the world, should be part of the student’s repertory from the earliest years. Because both stances involve cognitive and affective as well as public and private elements, students need to learn to differentiate the circumstances that call for one or the other stance. Unfortunately, much current practice is counterproductive, either failing to encourage a definite stance or implicitly requiring an inappropriate one. Favorite illustrations are the third-grade workbook that prefaced its first poem with the question “What facts does this poem teach you?” and the boy who com- plained that he wanted information about dinosaurs, but his teacher only gave him “storybooks.” Small wonder that graduates of our schools (and colleges) often read poems and novels efferently or respond to political statements and advertisements with an aesthetic stance.
1390 Rosenblatt
Despite the overemphasis on the efferent in our schools, failure to under- stand the matter of the public–private “mix” has prevented successful teaching even of efferent reading and writing. Teaching practices and curriculums, from the very beginning, should include both efferent and aesthetic linguistic activity and should build a sense of the different purposes involved. Instruction should foster the habits of selective attention and synthesis that draw on relevant ele- ments in the semantic reservoir and should nourish the ability to handle the mix of private and public aspects appropriate to a particular transaction.
Especially in the early years, this should be done largely indirectly, through, for example, choice of texts, contexts for generating writing and reading, or im- plications concerning stance in the questions asked. In this way, texts can serve dynamically as sources from which to assimilate a sense of the potentialities of the English sentence and an awareness of strategies for organizing meaning and ex- pressing feeling. Emphasis on analysis of the evocations, or terminology for cat- egorizing and describing them, has no value if they overshadow or substitute for the evoked work. Such activities acquire meaning and value when, for example, they answer a writer’s own problems in expression or explain for a reader the role of the author’s verbal strategies in producing a certain felt response.
The developmental sequence suggested here is especially important in aes- thetic reading. Much teaching of poetry at every level, including high school and college, at present takes on a continuously repeated remedial character be- cause of the continued confusion about stance through emphasis on efferent analysis of the “literary” work. Students need to be helped to have unimpeded aesthetic experiences. Very young children’s delight in the sound and rhythms of words, their interest in stories, and their ability to move easily from verbal to oth- er modes of expression too often fade. They need to be helped to hold on to the experiential aspect. When this can be taken for granted, efferent, analytical dis- cussions of form or background will not be substitutes for the literary work but become a means of enhancing it. Discussion then can become the basis for as- similating criteria of sound interpretation and evaluation appropriate to the vari- ous points on the continuum and to the student’s developmental status.
Implications for Research
Research based on the transactional model has a long history (Applebee, 1974; Farrell & Squire, 1990). Until fairly recently, it has generated research mainly by those concerned with the teaching of literature in high schools and colleges, rather than by those concerned with reading per se in the elementary school (Beach & Hynds, 1990; Flood et al., 1991; Purves & Beach, 1972). It is not possible here to survey this already considerable body of research, much of it exploring as- pects of response to literature; nor does space allow discussion of recent volumes dealing with applications of transactional theory in elementary school, high
The Transactional Theory of Reading and Writing 1391
school, and college (Clifford, 1991; Cox & Many, 1992; Hungerford, Holland, & Ernst, 1993; Karolides, 1992). I shall instead suggest some general considera- tions concerning research topics and theoretical and methodological pitfalls.
The transactional model of reading, writing, and teaching that has been pre- sented constitutes, in a sense, a body of hypotheses to be investigated. The shift it represents from the Cartesian to the post-Einsteinian paradigm calls for removal of the limitations on research imposed by the dominance of positivistic behavior- ism. Instead of mainly treating reading as a compendium of separate skills or as an isolated autonomous activity, research on any aspect should center on the human being speaking, writing, reading, and continuously transacting with a specific en- vironment in its broadening circles of context. And as Bartlett (1932) reminds us, any secondary theoretical frameworks, such as schemata or strategies, are not stable entities but configurations in a dynamic, changing process. Although the focus here will be on reading research, the interrelationship among the linguistic modes, especially reading and writing, broadens the potential scope of problems mentioned.
The view of language as a dynamic system of meaning in which the affective and the cognitive unite raises questions about the emphasis of past research. Researchers’ preoccupation with the efferent is exemplified by their focus on Piaget’s work on the child’s development of mathematical and logical concepts and the continuing neglect of the affective by behaviorist, cognitive, and artificial in- telligence psychologists. This is slowly being counterbalanced by growing inter- est in the affective and the qualitative (e.g., Deese, 1973; Eisner & Peshkin, 1990; Izard, 1977). We need to understand more fully the child’s growth in capacity for selective attention to, and synthesis of, the various components of meaning.
Research in reading should draw on a number of interrelated disciplines, such as physiology, sociology, and anthropology, and should converge with the general study of human development. The transactional theory especially raises questions that involve such broad connections. Also, the diverse subcultures and ethnic backgrounds represented by the student population and the many strands that contribute to a democratic culture present a wide range of questions for research about reading, teaching, and curriculum.
Developmental Processes
The adult capacity to engage in the tremendously complex process of reading de- pends ultimately on the individual’s long developmental process, starting with “learning how to mean” (Halliday, 1975; Rosenblatt, 1985b). How does the child move from the earliest, undifferentiated state of the world to “the referential, emo- tive and associative part processes” (Rommetveit, 1968, p. 167)? Developmental research can throw light on the relation of cognitive and emotional aspects in the growth of the ability to evoke meaning in transactions with texts.
Research is needed to accumulate systematic understanding of the posi- tive environmental and educational factors that do justice to the essential nature
1392 Rosenblatt
of both efferent and aesthetic linguistic behavior, and to the role of the affective or private aspects of meaning in both stances. How can children’s sensorimotor explorations of their worlds be reinforced, their sensitivity to the sounds and qualitative overtones of language be maintained? In short, what can foster their ca- pacity to apprehend in order to comprehend, or construct, the poem, story, or play? Much also remains to be understood about development of the ability to infer, or make logical connections, or, in short, to read efferently and critically.
How early in the child’s development should the context of the transac- tion with the text create a purpose for one or the other dominant stance, or help the reader learn to adopt a stance appropriate to the situation? At different de- velopmental stages, what should be the role or roles of reflection on the reading experience through spoken comments, writing, and the use of other media?
An overarching question is this: How can skills be assimilated in a con- text that fosters understanding of their relevance to the production of meaning? How can the young reader acquire the knowledge, intellectual frameworks, and sense of values that provide the connecting links for turning discrete verbal signs into meaningful constructs? The traditional methods of teaching and testing recognize the important functions of the symbolic system, the alphabetic and phonological elements (the “code”), and linguistic conventions by fragmenting processes into small quantifiable units. These are quantitatively and hence eco- nomically assessable. But do such methods set up habits and attitudes toward the written word that inhibit the process of inferring meaning, or organizing and synthesizing, that enters into even simple reading tasks? How can we prepare the way for increasingly rich and demanding transactions with texts?
Assessment of performance level is usually required as a means of assuring the ac- countability of the school. Whether standardized tests accurately measure the stu- dent’s ability is currently being called into question. Research on correlation of reading ability with factors such as age, gender, ethnic and socioeconomic back- ground, and so on has confirmed the expectation that they are active factors. However, such research reports a state of affairs that is interpreted according to vary- ing assumptions, not all conducive to the development of mature readers and writ- ers. The transactional emphasis on the total context of the reading act reinforces the democratic concern with literacy and supports the call for vigorous political and social reform of negative environmental factors. At the same time teachers must rec- ognize that the application of quantitatively based group labels to individual students may unfairly create erroneous expectations that become self-fulfilling prophecies.
Teaching Methods
In the current transition away from traditional teaching methods, there is the danger that inappropriate research designs may be invoked to evaluate particular teaching
The Transactional Theory of Reading and Writing 1393
methods. What criteria of successful teaching and what assumptions about the nature of linguistic processes underlie the research design and the methods of measurement? Any interpretations of results should take into account the various considerations concerning reader, text, and context set forth in the transactional model.
Results of research assessing different teaching methods raise an important question: Did the actual teaching conform to the formulaic labels attached to the methods being compared? The vagueness of a term such as reader-response method can illustrate the importance of more precise understanding of the actual teaching processes being tested in a particular piece of research. The same term has been applied to teachers who, after eliciting student responses to a story, fall back on habitual methods of demonstrating the “correct” interpretation and to teachers who make the responses the beginning of a process of helping students grow in their ability to arrive at sound, self-critical interpretations.
Much remains to be done to develop operational descriptions of the ap- proaches being compared. Studies are needed of how teachers lead, or facilitate, without dominating or dictating. Ethnographic study of classroom dynamics, records of interchange among teacher and students, videotapes of classrooms, and analyses of text give substance to test results.
Students’ empirical responses to a text (mainly written protocols) form the basis of much of the research on methods generally referred to as reader response or transactional. (The term response should be understood to cover multiple activi- ties.) Protocols provide indirect evidence about the students’ evocation, the work as experienced, and reactions to it. Such research requires a coherent sys- tem of analysis of students’ written or oral reports. What evidence, for example, is there that the reading of a story has been predominantly aesthetic?
The problem of empirical assessment of the student’s aesthetic reading of a text offers particular difficulties, especially because no single “correct” inter- pretation or evaluation is posited. This requires setting up criteria of interpreta- tion that reflect not only the presence of personal feelings and associations, which are only one component, but also their relationship to the other cognitive and at- titudinal components. In short, the assessment must be based on clearly articu- lated criteria as to signs of growing maturity in handling personal response, relating to the evoked text, and use of personal and intertextual experience vis- à-vis the responses of others.
In order to provide a basis for statistical correlation, content analysis of protocols has been used largely to determine the components or aspects of re- sponse. The purpose is to distinguish personal feelings and attitudes from, for ex- ample, efferent, analytic references to the sonnet form. This requires a systematic set of categories, such as The Elements of Writing About a Literary Work (Purves & Rippere, 1968), which has provided a common basis for a large number of
1394 Rosenblatt
studies. As the emphasis on process has increased, refinements or alternatives have been devised. The need is to provide for study of the relationship among the various aspects of response, or the processes of selecting and synthesizing ac- tivities by which readers arrive at evocations and interpretations (Rosenblatt, 1985a). Qualitative methods of research at least should supplement, or perhaps should become the foundation for, any quantitative methods of assessing trans- actions with the written word.
Experimental designs that seek to deal with the development of the ability to handle some aspect of literary art should avoid methodologies and experi- mental tasks that instead serve to test efferent metalinguistic capacities. For ex- ample, levels of ability to elucidate metaphor or to retell stories may not reflect children’s actual sensing or experiencing of metaphors or stories so much as their capacity to efferently abstract or categorize (Verbrugge, 1979).
The dependence on single instances of reading in assessing an individual’s abilities is currently being called into question. The previous reminder that we are dealing with points in a continuing and changing developmental process is especially relevant. Habits are acquired and change slowly; it may be found that the effects of a change, for example, from traditional to response methods of teaching literature, cannot be assessed without allowing for a period of transition from earlier approaches and the continuation of the new approaches over time.
Basal readers have in the past offered especially clear examples of ques- tions and exercises tacitly calling for an efferent stance toward texts labeled sto- ries and poems. There has been little to help students assimilate and make automatic the aesthetic mode of relating to a text. Here, preparations for read- ing, the teacher’s questions both before and after reading, and the mode of as- sessment, which powerfully influences teaching, should be scrutinized.
Studies that seek to generalize about the development of abilities by si- multaneous testing of the different age levels have the problem of taking into account the factor of schooling. To what extent do changes in children’s ability to retell or comment on the grammar of a story reflect schooling in the appropri- ate way to talk about a story? Similarly, to what extent are reported changing literary interests in the middle years not a reflection of personality changes but of too narrow definitions of literary?
Research Methodologies
The preceding discussion has centered on suggesting problems for research im- plied by the transactional model. Research methods or designs have been men- tioned mainly in reference to their potentialities and limitations for providing kinds of information needed and to criteria for interpretation of data. Quanti- tatively based generalizations about groups are usually called for, but currently there is interest in clarifying the potentialities and limitations of both quantitative and qualitative research. Empirical experimental designs are being supplemented
The Transactional Theory of Reading and Writing 1395
or checked by other research approaches, such as the case study (Birnbaum & Emig, 1991), the use of journals, interviews during or after the linguistic event, portfolios, and recordings in various media. Because the single episode test has various limitations, research in which researcher and teacher collaborate—or care- fully planned research carried on by the teacher—provides the opportunity for extended studies. The transactional model especially indicates the value of ethno- graphic or naturalistic research because it deals with problems in the context of the ongoing life of individuals and groups in a particular cultural, social, and educa- tional environment (Kantor, Kirby, & Goetz, 1981; Zaharlick & Green, 1991). The developmental emphasis also supports the call for longitudinal studies (Tierney, 1991). Interdisciplinary collaboration, desirable at any time, seems espe- cially so for longitudinal studies. Research will need to be sufficiently complex, varied, and interlocking to do justice to the fact that reading is at once an intensely individual and an intensely social activity, an activity that from the earliest years involves the whole spectrum of ways of looking at the world.
I want to thank June Carroll Birnbaum and Roselmina Indrisano for reading this manuscript, and Nicholas Karolides and Sandra Murphy for reading earlier versions.
.    1  The 1949 volume marks Dewey’s choice of transaction to designate a concept present in his work since 1896. My own use of the term after 1950 applied to an approach developed from 1938 on.
.    2  By 1981, transactional theory, efferent stance, and aesthetic stance were sufficiently current to be listed and were attributed to me in A Dictionary of Reading and Related Terms (Harris & Hodges, 1981). But the often confused usage of the terms led me to write “Viewpoints: Transaction Versus Interaction—A Terminological Rescue Operation” (1985).
.    3  The transactional model of reading presented here covers the whole range of similarities and differences among readers and between author and reader. Always in the transaction between reader and text, activation of the reader’s linguistic–experiential reservoir must be the basis for the construction of new meanings and new experiences; hence, the applicability to bilingual instruction and the reading of texts produced in other cultures.
4 Literature as Exploration emphasizes the instructional process that can be built on the basis of personal evocation and response. Illustrations of classroom discussions and chapters such as “Broadening the Framework,” “Some Basic Social Concepts,” and “Emotion and Reason” indicate how the teacher can democratically moderate discussion and help students toward growth not only in ability to handle increasingly complex texts but also in personal, social, and cultural understanding.
Ames, A. (1955). The nature of our perceptions, prehensions, and behavior. Princeton, NJ: Princeton University Press.
Appenzeller, T. (1990, November/December). Undivided attention. The Sciences.
Applebee, A.N. (1974). Tradition and reform in the teaching of English. Urbana, IL: National Council of Teachers of English.
Bartlett, F.C. (1932). Remembering: A study in experimental and social psychology. London: Cambridge University Press.
Bates, E. (1979). The emergence of symbols. New York: Academic.
Beach, R., & Hynds, S. (1990). Research on re- sponse to literature. In E. Farrell & J.R. Squire (Eds.), Transactions with literature
1396 Rosenblatt
(pp. 131–205). Urbana, IL: National Council
of Teachers of English.
Birnbaum, J., & Emig, J. (1991). Case study. In
J. Flood, J.M. Jensen, D. Lapp, & J.R. Squire (Eds.), Handbook of research on teaching the English language arts (pp. 195–204). New York: Macmillan.
Blumenthal, A.L. (1977). The process of cogni- tion. Englewood Cliffs, NJ: Prentice Hall. Bohr, N. (1959). Discussion with Einstein. In P.A.
Schilpp (Ed.), Albert Einstein, Philosopher-
Scientist (p. 210). New York: HarperCollins. Cantril, H., & Livingston, W.K. (1963). The concept of transaction in psychology and neu- rology. Journal of Individual Psychology, 19,
Chomsky, N. (1968). Language and mind. New
York: Harcourt Brace.
Clifford, J. (Ed.). (1991). The experience of
reading: Louise Rosenblatt and reader re- sponse theory. Portsmouth, NH: Boynton/ Cook.
Cox, C., & Many, J.E. (Eds.). (1992). Reader’s stance and literary understanding. Norwood, NJ: Ablex.
Culler, J. (1982). On deconstruction. Ithaca, NY: Cornell University Press.
Damasio, A.R. (1989). The brain binds entities by multilingual activities for convergence zones. Neural Computation, 1.
Deese, J. (1973). Cognitive structure and af- fect in language. In P. Pliner & T. Alloway (Eds.), Communication and affect. New York: Academic.
Dewey, J. (1938). Logic: The theory of inquiry. New York: Henry Holt.
Dewey, J., & Bentley, A.F. (1949). Knowing and the known. Boston: Beacon.
Eisner, E.W., & Peshkin, A. (1990). Qualitative inquiry in education: The continuing debate. New York: Teachers College Press.
Emig, J. (1983). The web of meaning. Portsmouth, NH: Boynton/Cook.
Farrell, E., & Squire, J.R. (Eds.). (1990). Trans- actions with literature. Urbana, IL: National Council of Teachers of English.
Fish, S. (1980). Is there a text in this class? Cambridge, MA: Harvard University Press. Flaubert, G. (1926). Correspondance (Vol. 2).
Paris: Louis Conard.
Flood, J., Jensen, J.M., Lapp, D., & Squire, J.R.
(Eds.). (1991). Handbook of research on teaching the English language arts. New York: Macmillan.
Halliday, M.A.K. (1975). Learning how to mean. New York: Elsevier.
Harris, T.L., & Hodges, R.E. (Eds.). (1981). A dic- tionary of reading and related terms. Newark, DE: International Reading Association.
Heath, S.B. (1983). Ways with words: Language, life, and work in communities and classrooms. Cambridge, UK: Cambridge University Press.
Hungerford, R., Holland, K., & Ernst, S. (Eds.). (1993). Journeying: Children responding to literature. Portsmouth, NH: Heinemann.
Izard, C.E. (1977). Human emotions. New York: Plenum.
James, W. (1890). The principles of psychology (2 vols.). New York: Henry Holt.
Kantor, K.J., Kirby, D.R., & Goetz, J.P. (1981). Research in context: Ethnographic studies in English education. Research in the Teaching of English, 15(4), 293–309.
Karolides, N.J. (Ed.). (1992). Reader response in the classroom: Evoking and interpreting meaning in literature. White Plains, NY: Longman.
Kuhn, T. (1970). The structure of scientific rev- olutions (2nd ed.). Chicago: University of Chicago Press.
Lindbergh, A.M. (1940). The wave of the future. New York: Harcourt Brace.
McCloskey, D. (1985). The rhetoric of econom- ics. Madison: University of Wisconsin Press. Miller, G.A., & Johnson-Laird, P.N. (1976). Language and perception. Cambridge, MA:
Harvard University Press.
Myers, G. (1986). William James: His life and
thought. New Haven, CT: Yale University
Peirce, C.S. (1933, 1935). Collected papers
(Vol. 3, Vol. 6) (P. Weiss & C. Hartshorne, Eds.). Cambridge, MA: Harvard University Press.
Purves, A.C., & Beach, R. (1972). Literature and the reader: Research in response to lit- erature. Urbana, IL: National Council of Teachers of English.
Purves, A.C., & Rippere, V. (1968). Elements of writing about a literary work: A study of response to literature. Urbana, IL: National Council of Teachers of English.
Rommetveit, R. (1968). Words, meanings, and messages. New York: Academic.
Rosenblatt, L.M. (1978). The reader, the text, the poem: The transactional theory of the lit- erary work. Carbondale: Southern Illinois University Press.
Rosenblatt, L.M. (1983). Literature as explo- ration (4th ed.). New York: Modern Language Association. (Original work published 1938)
The Transactional Theory of Reading and Writing 1397
Rosenblatt, L.M. (1985a). The transactional theory of the literary work: Implications for research. In C. Cooper (Ed.), Researching re- sponse to literature and the teaching of lit- erature. Norwood, NJ: Ablex.
Rosenblatt, L.M. (1985b). Viewpoints: Trans- action versus interaction—A terminological rescue operation. Research in the Teaching of English, 19, 96–107.
Rosenblatt, L.M. (1991). Literary theory. In J. Flood, J.M. Jensen, D. Lapp, & J.R. Squire (Eds.), Handbook of research on teaching the English language arts (pp. 57–62). New York: Macmillan.
Rosenblatt, L.M. (1993). The transactional theory: Against dualisms. College English, 55(4), 377–386.
Saussure, F. (1972). Cours de linguistique générale. Paris: Payot.
Tierney, R.J. (1991). Studies of reading and writing growth: Longitudinal research on lit- eracy development. In J. Flood, J.M. Jensen, D. Lapp, & J.R. Squire (Eds.), Handbook of research on teaching the English language arts (pp. 176–194). New York: Macmillan.
Verbrugge, R.R. (1979). The primacy of meta- phor in development. In E. Winner & H. Gardner (Eds.), Fact, fiction, and fantasy in childhood. San Francisco: Jossey-Bass.
Vygotsky, L.S. (1962). Thought and language (F. Hanmann & G. Vakar, Eds. & Trans.). Cambridge, MA: MIT Press.
Werner, H., & Kaplan, B. (1962). Symbol for- mation. New York: Wiley.
Zaharlick, A., & Green, J. (1991). Ethnographic research. In J. Flood, J.M. Jensen, D. Lapp, & J.R. Squire (Eds.), Handbook of research on teaching the English language arts (pp. 205–223). New York: Macmillan.
1398 Rosenblatt