All reader-response critics focus on readers during the process of reading. Some examine individual readers through psychological observations and participation; others discuss reading communities through philosophical speculation and literary intuition. Rejecting the Affective Fallacy of American New Criticism, all describe the relation of text to reader. Indeed, all share the phenomenological assumption that it is impossible to separate perceiver from perceived, subject from object. Thus they reject the text’s autonomy, its absolute separateness, in favor of its dependence on the reader’s creation or participation. Perception is viewed as interpretive; reading is not the discovery of meaning but the creation of it. Reader-response criticism replaces examinations of a text in-and-of-itself with discussions of the reading process, the “interaction” of reader and text.

Stanley Fish’s early essay, “Literature in the Reader: Affective Stylistics” (1970) presented one influential version of this reader-response criticism. Fish viewed a sentence in the text not as “an object, a thing-in-itself, but an event, something that happens to, and with the participation of, the reader.” His claims were aggressively descriptive: “In my method of analysis, the temporal flow is monitored and structured by everything the reader brings with him, by his competences; and it is by taking these into account as they interact with the temporal left-to-right reception of the verbal string that I am able to chart and project the developing response.”¹ And the developing response was that of the “informed reader,” a reader with the ability to understand the text and have the experience the author intended.² In this “affective stylistics,” Fish talked as if a text manipulated the reader—the text forced the reader to perform certain cognitive acts—and Fish, as practical critic, described that manipulative process. As a critical theorist Fish attacked formalist approaches, especially American New Criticism, for ignoring “what is objectively true about the activity of reading.” He claimed that his own approach was in contrast “truly objective” because it recognized the “fluidity … of the meaning experience” and directed our attention “to where the action is—the active and activating consciousness of the reader.”³

In a major reversal, Fish rejects these claims in “Interpreting the Variorum” (1976), where he argues that all texts are in fact constituted by readers’ interpretive strategies and that the process he formerly claimed to describe is actually a creation of his critical theory: “What my principles direct me to ‘see’ are readers performing acts; the points at which I find (or to be more precise, declare) those acts to have been performed become (by a sleight of hand) demarcations in the text; those demarcations are then available for the designation ‘formal features,’ and as formal features they can be (illegitimately) assigned the responsibility for producing the interpretation which in fact produced them.”⁴ This radical revision of Fish’s theory has two consequences: a change in the relation of reader to text and a change in the relation of criticism to reading. Fish now claims that in reading the interpreter constitutes the text and that in reader criticism the interpreter’s description constitutes the nature of the reading process according to his interpretive strategies.

Fish has moved from a phenomenological emphasis (which describes the interdependence of reader and text) to a structuralist or even post-structuralist position (which studies the underlying systems that determine the production of textual meaning

This schema locates each critic on a continuum of reader-oriented approaches. A detailed examination of these approaches will reveal not only the interrelations among the critics placed here but also the problems within each of their reader-response theories. All five critics construct a theory consisting (in more or less detail) of an account of interpretation, a model for critical exchange, and a model of reading. These critics’ theories of interpretation try to account for meaning-production in both reading and criticism. Their models of critical discussion specify the nature of critical procedures (observation, description, explication, and explanation) and the ways interpretations are exchanged in critical dialogue. These hermeneutic theories and critical models are based on models of reading, accounts of how readers actually interact with the text during the temporal reading process. Norman Holland’s work provides a useful starting point for the following discussions of reader-response criticism because his writings carefully examine all three of these components making up a critical theory.⁵

Holland’s transactive criticism “takes as its subject-matter, not the text in supposed isolation, as the New Criticism claimed it did, nor the self in rhapsody, as the old impressionistic criticism did, but the transaction between a reader and a text.”⁶ The notion of an “identity theme” is central to Holland’s approach: “we can be precise about individuality by conceiving of the individual as living out variations on an identity theme much as a musician might play out an infinity of variations on a single melody.” A person brings this “unchanging inner core of continuity” to all transactions between Self and Other, including reading.⁷

Holland’s model of reading proceeds from his more general theory of the relation between personality and perception. Perception is a “constructive act,” not merely reflecting but forming reality: “the individual apprehends the resources of reality (including language, his own body, space, time, etc.) as he relates to them in such a way that they replicate his identity.”⁸ That is, perception is also interpretation, and “interpretation is a function of identity, specifically identity conceived as variations upon an identity theme.” Holland particularizes this view of perception in his central thesis about reading: identity re-creates itself. “All of us, as we read, use the literary work to symbolize and finally to replicate ourselves. We work out through the text our own characteristic patterns of desire and adaptation.”⁹

Within this principle of identity re-creation, Holland isolates four specific modalities, which he conveniently organizes under the acronym DEFT—-defenses, expectations, fantasies, and transformations. “One can think of these four separate principles as emphases on one aspect or another of a single transaction: shaping an experience to fit one’s identity and in doing so, (D) avoiding anxiety, (F) gratifying unconscious wishes, (E) absorbing the event as part of a sequence of events, and (T) shaping it with that sequence into a meaningful totality.”¹⁰ The concept of a “meaningful totality” or unity is pivotal for Holland’s reading model (and is equally important in his general theory of interpretation).¹¹ According to Holland, the reader makes sense of the text by creating a meaningful unity out of its elements. Unity is not in the text but in the mind of a reader. “By means of such adaptive structures as he has been able to match in the story, he will transform the fantasy content, which he has created from the materials of the story his defenses admitted, into some literary point or theme or interpretation.”¹² For Holland, meaning is the result of this interpretive synthesis, the transformation of fantasy into a unity which the reader finds coherent and satisfying. As with all interpretation, “the unity we find in literary texts is impregnated with the identity that finds that unity.” Each reader creates a unity for a text out of his own identity theme, and thus “each will have different ways of making the text into an experience with a coherence and significance that satisfies.”¹³ Therefore, Holland’s model of reading accounts exceptionally well for varied responses.

On the other hand, Holland’s present theory has trouble with the phenomenon of similar responses. Similarity was easily explained by his earlier model. In The Dynamics of Literary Response (1968), he spoke as if “fantasies and their transformations were embodied in the literary work, as though the work itself acted like a mind”; different readers could take in (“introject”) the same text and “participate” in whatever psychological process was embodied there. Accounting for recurrent responses has become much more difficult in Holland’s revised model, in which “processes like the transformation of fantasy materials through defenses and adaptations take place in people, not in texts.”¹⁴ No longer embodying psychological processes, autonomous texts no longer serve as a guarantee of recurrence in Holland’s present model of reading. Instead, similar identity themes must somehow account for similar response.¹⁵ This psychological explanation contrasts with Fish’s “sociological” ones. In Fish’s earlier theory, all informed readers had the same basic reading experience because they shared linguistic and literary competence; in his present theory, communal reading strategies account for similar interpretive responses. The comparison between Fish’s and Holland’s reading models becomes more complex when we examine the precise status of the text in their revised theories.¹⁶

Holland and Fish both claim that perception is a constructive act: we interpret a...