Introduction

1. The Nature of the Problem

A significant part of Western culture consists of literary works. They have become assimilated into that culture and have come to influence people's beliefs and values by being understood in a certain way. If they were understood very differently, our cultural tradition itself would be likely to be or become very different from what it is. The interpretation of literary works is thus of considerable importance in shaping that cultural tradition. It seems desirable therefore to have a general characterization of what is involved in interpreting a literary work, in saying of a particular literary work that it means so-and-so. The purpose of this book is to provide such a characterization. I shall attempt to analyze our concept of the meaning of a literary work. I want to emphasize that my purpose is not to recommend a certain notion of the meaning of a literary work, nor to recommend a certain wav of interpreting literary works. To put it differently, I am not concerned to prescribe standards or criteria on the basis of which we ought to construe literary works. Nor am I concerned to establish empirical generalizations as to how certain readers or groups of readers tend to interpret literary texts.

I shall try to show rather what our common concept of the meaning of a literary work is. What I shall present then is neither an empirical theory of literary interpretation nor a "persuasive definition," but an analysis of what it is for a literary text to have a certain meaning.

In order to clarify the nature of the question which I propose to examine, as well as to provide a frame of reference for my discussion, I want to briefly mention several representative theories which offer various answers to the question.

(1) One might argue that what a literary work means depends on the reader's purpose. A specialist in Elizabethan literature, for example, might be primarily interested in interpreting a literary work of that period in terms of the beliefs of its author. Another reader may prefer to construe a work in such a way that it accords as much as possible with his own views. Yet a third reader might choose that interpretation of a work which maximizes its aesthetic value. This is not, of course, to say that any interpretation of a literary work is correct provided only that it answers a reader's purpose or interest. A further condition is that the interpretation is linguistically possible — that it does not violate the relevant linguistic rules. Thus, on this view, to say that a literary work has the meaning m is to say (a) that m is one of the readings in accord with the relevant rules of the language (at the time the work was written) and (b) that m answers the purpose or interest of a reader. It is obvious that, on this theory, there will be a fairly large number of equally correct interpretations of any given literary work.

(2) According to another view, similar to the first, interpretive statements are essentially normative. They can be justified in terms of a particular standard or standards of interpretation, but the question what the proper standard or standards are is a genuinely normative one; that is, it calls for an implicit or explicit decision by the critic as to how a work should be read. Suppose a critic accepts the following criterion: 'A work means m if and only if that is how the author understood it.' Then he can justify a particular interpretive claim about a work by appealing to this criterion. But since, on the present theory, the criterion which a critic accepts is a normative claim as to how literary works ought to be read, it can in turn be justified only by considerations bearing on the latter question. One might argue, for example, that a certain standard or a certain method of interpreting literary works corresponds to "habits of mind" that ought "to be cultivated" or, alternatively, that a certain standard (or a particular interpretation) ought to be rejected because it is "indicative of habits of mind that no one ought to have."

Consequently, the explicit or implicit standards which underlie claims about the meaning of a literary work function as imperatives or quasi-imperatives; what a particular work means is in some measure determined by "a critic's own individuality." Again, as with the first proposal, a reader cannot construe a work in any way he likes. The rules of the language in which a work is written set limits to its possible interpretations. Thus, on this theory, a claim about the meaning of a literary' work is, within the limits indicated, a statement about how the work in question ought to be construed; that is, it is a statement (at least in part) about the critic's personal preference based on his beliefs, attitudes, habits of mind, and so on — in short, on his "own individuality."

(3) A variant of this theory defines the meaning of a literary work as "a class of similar experiences, one or other of which those words in that order and arranged in that form, ought to evoke in a reader familiar with the language (or languages) in which [the work] is written." As with the preceding proposal, the question what a particular literary work means calls for a decision. The main difference is that, on the present view, the "ought" in the definition calls for an "aesthetic decision" while, presumably, a critic's decision as to what criteria to accept or how to construe a work need not, on the preceding theory, reflect an aesthetic judgment on his part.

(4) According to another theory, there are at least three different criteria which jointly determine what a literary work means. The first of these, "correspondence," requires that an interpretation be based on historical knowledge of the "subject matter" dealt with in a particular work. If, for example, a poem contains a reference to a certain myth, then a critic needs to be familiar with the myth in question or the version (or versions) of that myth current at the time the poem was written in order to construe the latter correctly. (I assume that the "correspondence" criterion also excludes an interpretation which violates the rules of the language in which the text is written.) The second standard requires an interpretation to accord with what the author intended; the third that, on a given interpretation, the various parts of the work form a coherent whole; In contrast to theories (1), (2), and (3) above, an interpretation is, on this view, subject to these criteria regardless of (i) a critic's beliefs as to which standard or standards ought to be accepted, (ii) his purpose in interpreting a work, and (iii) his belief as to what interpretation maximizes the aesthetic value of a work. Although the advocates of this theory assume that in general their three standards "tend to converge." it is obvious that in any given instance they may lead to different results. When they do, a work will presumably have several equally defensible interpretations. Hence a statement about the meaning of a literary work is, according to this theory, a factual claim about the satisfaction of one or more of the above conditions.

(5) On a prominent phenomenological theory of interpretation, the meaning of a literary work is said to be determined in part by the "historical situation" of the critic. The process of understanding a text involves a "fusion of horizons" (Horizontverschmelzung) of the critic and of the text. This theory is based on the so-called hermeneutic circle, the thesis that the understanding of the whole (text) depends on the understanding of the parts and vice versa. It is obvious that if this thesis is correct, any sort of objective understanding of a text is in principle impossible. The only criterion for determining whether or not an interpretation is correct is the coherence of the various parts of the text under the interpretation in question. But whereas the preceding theory (4) allows at least in principle an objective assessment of the coherence of a text under a given interpretation, the present theory precludes the very possibility of such an assessment. For, on this theory, the only constraint on a critic's interpretation of the parts is his interpretation of the whole. It follows that a statement about the meaning of a work is a statement about a particular critic's subjective understanding, about his personal perspective, about his own Vorurteile. Hence it is somewhat misleading to say that, according to this theory, it is the "historical situation" or "the whole of the objective course of history" (das Ganze des objektiven Geschichtsganges) which determines what a work means.

(6) Perhaps the most widely held view is t h a t the meaning of a work logically depends (i) on the rules (or "public conventions") of the language in which a text is written and (ii) on the coherence and complexity of a literary work under a given interpretation. To say that a particular literary work means m is to say (a) that m is one of the linguistically possible readings of the text and (b) that, of these readings, m maximizes the coherence and complexity of the work. The question whether these two conditions are satisfied in a particular case is held to be, at least in principle, objectively decidable. Thus, for example, whether a word sequence has certain associations or connotations in a particular text, or whether certain possible associations or connotations of a word sequence in a text increase its coherence or the complexity, can be determined objectively; that is, it does not depend on the purpose, the subjective experience, aesthetic decision, or other normative considerations of a particular critic.

(7) Finally, there is the view that the meaning of a literary work is determined by the author's intention. Usually, the qualification is added that the meaning of a work depends on the author's intention only within the limits of what the text could — given the rules of the language — be construed to mean. A statement about the meaning of a work is, on this view, as on (4) and (6), an objective claim, namely, with the qualification noted, a statement about what the author intended to convey by his work.

It is clear that all of these theories have certain implications as to what kinds of facts constitute evidence for the meaning of a literary work. Much of what I shall therefore be concerned with in the following is what considerations are relevant to questions about the meaning of a work. My purpose is not, however, to settle various interpretive controversies or even to add an argument in favor of one or the other of the interpretations of a particular text. Rather, my aim is to help clarify what such controversies are about, to shed light on the logic of interpretation by showing what a claim about the meaning of a literary work entails.

Although the theory I shall present is not an empirical but an analytical theory, it is clearly subject to empirical constraints. It must account in some plausible manner for the sorts of things critics and ordinary readers alike actually do in interpreting a work, what they take into account in arriving at an interpretation. It must be able to account for cases in which there is substantial agreement about the meaning of a work as well as for cases in which such consensus is lacking. Any theory which cannot adequately account for such facts and hence must dismiss as in principle misguided a significant part of what critics and readers do in interpreting literary works — to what criteria they appeal in practice, for example — will be suspect of offering, not an analysis of our common concept of the meaning of a literary work, but rather an implicit recommendation as to how we ought to conceive of its meaning.

2. Why Is a Conceptual Analysis of Literary Interpretation Useful?

An analysis of the concept of the meaning of a literary work does not add to our (empirical) knowledge of what a particular work means. One may wonder therefore what the value of such an analysis is, why such a theory might be worth having. The answer to this question is twofold. By providing an account of the logical structure of statements and arguments about the meaning of a literary work, such a theory makes us aware of what we as critics or readers are doing in interpreting literature. It makes us aware, in other words, of the logical commitments of our claims about the meaning of literary works.

Furthermore, a theory of this sort provides the basis for a principled acceptance or rejection of an interpretation of a literary work. Without such a theory we can accept or reject the practical criteria to which a critic might appeal in support of his interpretation of a work only on intuitive grounds. The theory enables us to justify the acceptance or rejection of such criteria and hence (indirectly) of the interpretations based on them.

Consider, for example, a controversy about the meaning of a text from the sixteenth century. Critic A construes the work in one way, critic B in another. Let us assume that, in support of his reading of the text, A has produced considerable evidence based on the language of the time and on the interpretation of the work by the author's contemporaries. Now B might conceivably insist that facts about the language at the time the work was written as well as the interpretation of the work by the author's contemporaries are simply irrelevant to its meaning. The meaning of a work, B might contend, is determined by our (twentieth-century) usage of the words which constitute the text. And B might be able to show that, by this criterion, his own reading is correct, whereas A's is incorrect.

It is obvious that without a theory of what it is to say that a literary work means so-and-so, we cannot decide the crucial issue between A and B as to what kinds of facts constitute evidence for the meaning of a work. We might, of course, find A's criteria more plausible; but, without a theory, we can defend our acceptance of his interpretation only on intuitive grounds. On the other hand, a well-founded theory would enable us to justify, on other than merely intuitive grounds, a claim about what kinds of facts constitute evidence for the meaning of a literary work. Consequently, we could defend our acceptance of, say, A's interpretation without begging the question whether the kinds of facts A has adduced are indeed evidence of what the work means. Furthermore, we would then be in a position to assess the relative weight of the various kinds of facts relevant to the meaning of a work. To put the point differently, we would also be able to determine what, if anything, would constitute better evidence of the meaning of the text than, say, facts about the language at the time the work was written.

3. The Claims to be Defended

I shall defend three main claims.

(1) I shall attempt to uphold the view that there is a logical connection between statements about the meaning of a literary work and statements about the author's intention such that a statement about the meaning of a work is a statement about the author's intention.

E. D. Hirsch's books Validity in Interpretation and The Aims of Interpretation present undoubtedly the most important and justly influential recent defense of a similar position. The difference is basically this. Hirsch believes that there is no logical connection between the meaning of a work and the author's intention. He holds rather that we ought to accept the author's intention as the decisive criterion of what a text means, since otherwise literary interpretation will be hopelessly subjective and therefore unable to provide genuine knowledge about the meaning of literary works. Whereas Hirsch is more or less explicitly offering a recommendation as to what critics ought to do in interpreting a text — namely, try to ascertain the authors intention — my view is that they are necessarily doing so already, in virtue of what it is for a literary work to have a certain meaning. In the next chapter I shall discuss Hirsch's theory in some detail, partly in order to differentiate more fully my own view from his and to indicate what I take to be difficulties with his position, and partly to bring out the force of one of his most valuable and important contributions to the theory of interpretation, namely, his distinction between meaning and significance.