- What I wish to do in this essay is point out an interesting and apparently unnoticed convergence in the doctrines of America's most famous contemporary philosopher of language and those of deconstructionist literary critics, and to discuss the relevance that post-Quinian ways of looking at issues concerning "meaning" might have for critical theory.
- In 1951 W.V. Quine published a devastating attack on philosophers' use of the notion same meaning as well as on the use of the notion true by virtue of the meaning. After the appearance of "Quine (W.V.) - Word & Object" in 1960, Quine's "pessimism" about meaning talk became a large topic of discussion. Quine argued that there are no general rules which determine what does and what does not count as a situation in which a particular sentence is assertible. Assertibility, to the extent that it is rational, is pragmatic and depends on the entire context. When beliefs clash we have trade-offs between different desiderata (for example, simplifying the system and preserving core beliefs), and there cannot be algorithms (much less rules associated with the individual sentences) which tell us how to make the best trade-off in each case. The "meaning" of a sentence cannot be identified with the rule or battery of rules which determine its assertibility conditions, for there are no such rules.
- Quine proposed, in fact, that talk of "the meaning" of a sentence or text only makes sense when relativized — relativized to an interpretation, or, as he put it, to a "translation manual." He advanced the radical thesis that there is "no fact of the matter" as to which translation manual is the right one: all translation manuals satisfying certain extremely weak constraints are formally possible, and the choice of one of the manuals (satisfying the constraints) over the others is subjective.
- In effect, Quine said that we should talk about interpretation and not about meaning. The similarity between Quine's views and those of a number of leading Continental penseurs went, however, totally unremarked. Quine's relentless scientism and his choice of examples (a typical text, in Quinian philosophy of language, is "Lo, a rabbit!") may have had something to do with this.
- There is a difficulty, of course — one pointed out by many of Quine’s critics. If all sentences and texts lack determinate meaning, then how can Quine view his own utterances as anything more than mere noise? (Quine's scientism does lead him to regard all discourse, including his own, as patterns of sound or inscription produced in accordance with Skinnerian schedules of conditioning.) Quine's answer, that his utterances have determinate meanings relative to themselves (if you ask me in my own language, "Under what conditions is 'snow is white' true?" I will answer that this sentence is true if and only if snow is white — using the very sentence you inquired about to state its own truth conditions), hardly seems to acknowledge, let alone respond to, the objection. The difficulty is, in fact, the familiar self-application problem encountered by all relativisms which become total. When one of the European penseurs I mentioned insists that truth and justification, to the extent that there are such things, are determined by criteria internal to a discourse, we naturally ask him whether there is a fact of the matter as to what the "criteria internal to a discourse" really are and as to what counts as fulfilling them. Can Michel Foucault's "archaeology of knowledge" really be objective if all other discourse is subjective?
- In "Putnam (Hilary) - Meaning and the Moral Sciences" I argued that there is a way of preserving Quine's insights without being carried along into total relativism. One can agree with Quine that there are not such things as "meanings," "semantical rules," and so on, and that talk of meanings should give way to study of the activities of interpretation. But one need not agree that there are no better and worse interpretations (among those that satisfy the four constraints listed in Word and Object). What Quine called "the indeterminacy of translation" should rather be viewed as the interest-relativity of interpretation.
- As I use the term, "interest-relativity" contrasts with absoluteness, not with objectivity. It can be objective that an interpretation or an explanation is the correct one, given the context and the interests which are relevant in the context. Something can be interest-relative and "objective humanly speaking," in David Wiggins's excellent phrase. Naom Chomsky's contention (in "Chomsky (Noam) - Rules and Representations") that the doctrine of interest-relativity commits me to total subjectivism assumes, what is not the case, that all interests must be taken to be on a par. There are silly interests, deluded interests, irrational interests, and so on, as well as reasonable and relevant ones (even if there is no general rule for determining which are which). A sane relativism can recognize that there is a fact of the matter in interpretation without making that fact of the matter unique or context-independent.
- With these remarks as background information, I want to turn now to the subject of criticism. Everyone is familiar with the fact that Aristotle's Metaphysics has had many interpretations and with the fact that King Lear has had many interpretations. Not only do critics "read" the Metaphysics or Lear differently; there are recognizable differences in the assumptions of criticism and the styles of criticism in different centuries. If variant interpretations of the sentence "Lo, a rabbit!" exist only as a philosopher's example, the same cannot be said of variant interpretations of Lear. It is surely high time that we brought some of our theory to bear on genuine textual and critical problems.
- I used to say — as a rueful joke — "as I get smarter, Aristotle gets smarter." The notion of an ideal "correct" interpretation seems problematic. Yet the radical view that interpretations are simply the inventions of the interpreter is just the old self-refuting relativism in its latest guise. What should we say?
- If we follow what I have described as Quine's insight, and think of interpretation as correlation — correlation of Aristotle's words and sentences with words and sentences in our present-day language — then some of the mystery evaporates. Aristotle's words depended for their life on particular "contexts" — which is to say, particular institutions, particular assumptions, particular positions that real people once occupied and no longer occupy. Any "translation manual" interprets Aristotle using words which depend for their life on different institutions, assumptions, and positions. It is not surprising, from such a perspective, that each century should require new interpretations, nor that each interpretation should be capable of improvement in an infinity of directions. Once we give up the idea of the Platonic "meaning" that all interpreters are trying to snare and think of interpretation as human interaction — between two or more forms of life — then we will not be dismayed (or driven to an insane relativism) by the open-ended character of the activity.
- In the case of Aristotle, the interest-relativity of interpretation takes a particular form. Even if Aristotle had written present-day English there would still be an interpretation problem. Even if he had written "Happiness is the activity of the psyche according to virtue in the complete life" and not the Greek sentence he actually wrote, we could not answer the exegetical question by saying "The sentence is true if and only if happiness is the activity of the psyche according to virtue in the complete life." For we are interested in the implications of what Aristotle wrote, and in cases where what Aristotle wrote was vague or ambiguous, we are interested in knowing what (more precise) senses it could be given and what it implies about our problems when given these various possible more precise readings.
- There is, to be sure, a difference in this regard between a philosophical text and a literary one; but not a total difference. Whether or not we agree with Henry James in criticizing the explicit moralizing by the omniscient narrator in Middlemarch, we recognize that various meanings can be given to the remark that "Bulstrode was not a hypocrite — he was simply a man whose desires were stronger than his theoretic beliefs, and who always rationalized the latter into satisfactory agreement with the former." (Some of the meanings are indicated by the sentence which appears a little later in the same paragraph: "There is no general doctrine which is not capable of eating out our morality if not accompanied by the daily habit of direct fellow feeling with individual human beings.") George Eliot is trying to "tell us something" — and so is Henry James, even if he uses different devices. Nor is it only the "big" remarks or attitudes in a literary work that prompt the effort at interpretation. Today virtually any aspect may initiate a line of questioning: the kind of audience the writer seems to have in mind, the assumptions the writer seems to take for granted, the relation between narrative devices such as suspense and the deeper issues in the work, and so on. As in the history of philosophy, the questioning clearly can and does go far beyond the literal question: "What is the meaning (truth-conditions) associated with this line?"
- But now it may look as if I have simply overlooked the difference between an exegesis and a commentary. A commentary, it might be said, must depend on what are important questions for us, on our interests, assumptions, even (regrettably) our intellectual fashions. There can never be a final commentary, one that is perfect from the standpoint of every cultural position, every set of interests and assumptions. But why should there not be a perfect exegesis?
- Indeed, Donald Davidson would say that in principle there is. If the work is in our language, then its homophonic or "face value" interpretation is the perfect one (ignoring the problem Jorge Luis Borges more than once raises, that even the text of Don Quixote itself may no longer function as a "literal translation" of Don Quixote into Spanish — even if we are Spanish speakers — that what we mean by it may not be what Cervantes did!).
- But the very fact that even in the ideal case, the case in which the work is written in, say, standard present-day English with no idiosyncrasies whatsoever (if there is such a language), the only "perfect exegesis" would, in general, be the work itself makes my point in another way: in a significant sense, the exegesis / commentary distinction cannot be drawn. Any exegesis that is nontrivial must, to that extent, be commentary-laden.
- As I have already said, I see in this no cause for despair. If new exegeses and new critical interpretations are always necessary, if there is no convergence to One True Interpretation, then, by the same token, the fashion of seeing the interpretations of past centuries as wholly superseded by contemporary "insights" may be recognized as the naive progressivism that it is. Perhaps we can come to see criticism as a conversation with many voices rather than as a contest with winners and losers.
Footnote 1: I couldn’t be bothered to italicise the text in all the places Putnam did.
Text Colour Conventions (see disclaimer)
- Blue: Text by me; © Theo Todman, 2019
- Mauve: Text by correspondent(s) or other author(s); © the author(s)