Plausbility and Aesthetic Interpretation

Canadian Journal of Philosophy 7 (1977): 327-40.

Denis Dutton


www.denisdutton.com

 

If a catalogue were made of terms commonly used to affirm the adequacy of critical interpretations of works of art, one word certain to be included would be “plausible.” Yet this term is one which has received precious little attention in the literature of aesthetics. This is odd, inasmuch as I find the notion of plausibility central to an understanding of the nature of criticism. “Plausible” is a perplexing term because it can have radically different meanings depending on the circumstances of its employment. In the following discussion, I will make some observations about the logic of this concept in connection with its uses in two rather different contexts: the context of scientific inquiry on the one hand, and that of aesthetic interpretation on the other. In distinguishing separate senses of “plausible,” I shall provide reasons to resist the temptation to imagine that because logical aspects of two different types of inquiry, science and criticism, happen to be designated by the same term, they may to that extent be considered to have similar logical structures.

I begin by turning to a writer who takes a view of these things somewhat different from my own. In “The Logic of Interpretation,” an interesting and valuable article which raises a number of important issues, Joseph Margolis comments on aesthetic criticism’s “tolerance of alternative and seemingly contrary hypotheses” - for example, when a single novel is found to admit of several impressive interpretations by critics using Freudian, Adlerian, Marxist, or Thomist concepts and vocabularies.[1] His explanation of this rests finally on an analogy with scientific inquiry.

I should like to suggest that an analogue of this curious tolerance of incompatibles obtains in the physical sciences; and that certain arguments there are based on much the same model that is appropriate to interpretive disputes. (p.116, italics added)

Margolis notes that in science questions arise for which there are at present no answers: “How did life originate?” or “How was the solar system created” In trying to answer these questions it is possible to suggest a number of hypotheses, each perhaps incompatible with the others and yet all quite consistent with available facts. Even though we may have no means for arriving at decisive answers to these questions, and even though we may never at any future date find ourselves in a position to answer them, Margolis holds, we are still able to assess the plausibility of the various hypotheses which we now have. And just as we can judge the plausibility of hypotheses in natural science, so “in aesthetic criticism, correspondingly, we have ... procedures for determining the plausibility of interpretive statements.” (p. 117) Having tried to establish what he takes to be the significant similarities between criticism and science, Margolis ends his paper by briefly recounting some of the differences between the two disciplines:

It must be admitted that, though they are always marginal to the main effort of science, considerations of plausibility are more nearly central to aesthetic criticism. In fact, scientific speculations of the sort illustrated are treated in terms of plausibility only because of a technical inability to gain the desired information; critical interpretations on the other hand, are, in principle, logically weak. It is this probably that gives the appearance of lack of rigor to critical pronouncements. (pp. 117-178)

Even if the nature of the logical weakness of critical interpretation is left unstated, Margolis does seem in the end to want to say that a lack of rigor is endemic to criticism. For he finally admits, with an Aristotelian shrug, “At any rate, we cannot ask for mere precision than the subject will allow.” (p. 118)

Here it is a central view of Margolis that science and criticism operate in certain respects on “much the same logical model,” and it may be best approached by considering the analogies he draws between the place of the concept of plausibility in science and its place in criticism. First, it is worth noticing that he begins by likening a critical interpretation to an hypothesis in natural science, and throughout much of his paper the argument depends on this analogy. But as Margolis himself briefly seems to recognize at the very end of his discussion, a plausible hypothesis provided as a candidate to answer the question, “How was the solar system formed?” is something which we may have forever to regard as merely plausible only because we do not have, and might never have, data sufficient to answer to question with finality. This is not to say that the question is in principle unanswerable. For instance, we might someday develop the means to detect distant planetary systems in the process of formation, and this could conceivably give us very good reason to consider our hypothesis verified or disconfirmed. Yet it is seldom the case that a critical interpretation of a work of art cannot be decided upon because of a similar “technical inability to gain the desired information.” On the contrary, even when all the information is at hand-all the words, all the notes, or all the pigment-a critical interpretation may always remain “merely plausible.” Does this sort of fact really support Margolis’s conclusion that critical interpretations are “in principle, logically weak"? No, it does not. It simply appears to support it if we make the mistake of construing the word “plausible” to mean the same thing whether it is used in connection with art criticism or is employed for purposes of doing science.

Let us imagine a group of critics discussing Paradise Lost. One of them, a Marxist, has just finished a long disquisition in which he has tried to present a Marxist interpretation of the poem. His colleagues find the interpretation remarkably consistent. Not only has he used Marxist concepts to make sense of the dramatic progression of the poem, he has even been able to give a Marxist reading to each of the major symbols Milton has employed. All of the critic’s colleagues, except one, seem satisfied. “Yes,” they agree, “your reading of Paradise Lost is entirely plausible-we would never have imagined it possible to give such a systematically Marxist interpretation to the work.” But that one critic is impatient with the whole business. “Certainly,” he says, “I’ll agree that you have given us a plausible interpretation of the poem. But what I want to know is, is what you say really true?” Now this is a strange question, somehow out of place in a critical discussion. It may be that this critic would call for some investigation into Milton’s intentions in writing the poem in order to satisfy his demand for a true interpretation. The Marxist, however, might well allow that Milton had no specific political or revolutionary program in mind when he composed it; that in any event the interpretation is not at all an hypothesis about Milton’s intentions. The interpretation, he insists, has nothing to do with the poet’s biography. It is just a straightforward interpretation of Paradise Lost. Of course, a demand for the truth would not be odd in the least if this were a group of biologists discussing a plausible hypothesis about the etiology of a disease. But these are not scientists, they are critics. The fact that the question may strike us as being quite out of place in a critical dialogue is indicative of an important difference between the logic of criticism and the logic of science.

In point of fact, truth is a notion seldom used in connection with critical interpretations. Critical interpretations are more commonly said to be “sound,” “proper “ “apt,” “persuasive,” “revealing,” and so forth. If the critic had asked whether the Marxist reading of the poem was “fitting” or “appropriate” to the work, we would find nothing strange in his response. The essential point is this: where a scientist puts forward a plausible hypothesis and then seeks to determine its truth, a critic may only seek to determine the plausibility of his interpretation. In discussing a work of art, a critic may carefully build his case, trying to establish an interpretation. Nevertheless, his ultimate aim is to establish the plausibility of his interpretation. Once a critic has presented us with an interpretation which is seen to be plausible, it is senseless to demand that it be in some further sense a true interpretation. In natural science, the relationship between the plausibility of an hypothesis and its truth is a fairly loose one, since sometimes even the most plausible hypothesis will in the end be disconfirmed. In aesthetic criticism, however, critical interpretations are not considered to be merely plausible-as if being seen to be plausible were a feeble consolation prize compared to the accolade of being known to be true. Indeed, if and when it does make sense at all to speak of the truth of a critical interpretation, there seems to be not a loose, but an essentially close relationship between the plausibility of an interpretation and what we might call its critical truth; one could even say that a plausible interpretation may be called true in such contexts only by virtue of its very plausibility. Popular usage would occasionally have us respond to an apt or convincing interpretation of a work of art by saying “Yes, that’s true.” But our saying this depends on whether or not the interpretation is true to the work; whether, for example, it helps unify and make coherent sense of the disparate elements in our experience of the aesthetic object.

It is important to understand that ordinary usage tends to favor a sense of ”plausible” much more closely related to the (empiricist) scientific sense of the term, as designating an assertion deemed to be only provisionally acceptable, a mere hypothesis. This being so, it should not surprise us that it is possible to say without feeling a sense of contradiction “This interpretation is plausible, but not really satisfactory.” What is meant by such a remark is that we might have thought it critically plausible, perhaps when it was initially suggested, or that it seemed critically plausible, though on closer inspection or longer reflection it turned out not to be so. Critical interpretations can be advanced tentatively, but such hypotheses are rendered acceptable by establishing their critical plausibility. For them, such plausibility is not distinct from truth, let alone inferior to truth.

Objections to my thesis might well be imagined. After all, one may argue, is it not the case that in criticism interpretations are proposed and must, as in science, be verified by going to the work? If this were the case, a critical interpretation would indeed seem to be rather like a scientific hypothesis. We could grant that the confirmation involved here might not have the conclusiveness of a well-validated scientific hypothesis. But we could still allow that many of the basic concepts and procedures connected with natural science-proposing an hypothesis, checking it against known facts, and so on-would be involved in the critical procedure. Of course, we do indeed very often go to the work to check if a critical interpretation fits. This fact, however, should not be allowed to mislead us. For what really happens when a work of art is thus interpreted? I f we do feel that we must go to the work, it is not because we have in hand a plausible interpretation and we wish to determine if it is a true one as well. Rather, if we are in any doubt at all, we go to the work to see if it is plausible. Or, we might say, when an interpretation seems antecedently plausible, or plausible before one’s crucial confrontation with the work of art, then that confrontation will not be a test of truth so much as a final and crucial test of plausibility. Whether a critical interpretation is suitable to a work of art is not always something immediately or self-evidently obvious, and it is certainly not a matter to be decided by mere whim. Intricate argument, careful analysis, and long familiarity with the work may be required before we can be properly persuaded to accept the interpretation as adequate. What may be initially respected as a plausible interpretation may in the light of subsequent discussion and observation be seen to be inappropriate to the work. And there is naturally no denying that similar considerations as well apply in assessing the plausibility of a scientific hypothesis: one can marshal evidence and analyze a subject-matter as much to judge the plausibility of a scientific hypothesis as its truth. But again, the scientist aims to advance hypotheses which might ultimately be shown to be true. For the scientist, plausibility is never an ideal resting place.[2]

To help sharpen these points, let us look at three brief examples from music criticism:

(1) In discussing the last movement of the Schubert D major piano sonata, Robert Schumann remarks that the strongly contrasting passages are rather like the faces one might make to frighten children.

(2) Daniel Gregory Mason has written that a friend once said of the last section of the slow movement of Beethoven’s late A minor quartet that it reminded him of trying to ride a bicycle as slowly as possible without falling off.

(3) The New York Times once carried a review of a recital by a piano virtuoso in which the critic reported that the fast runs and arpeggios in the pianist’s performance of the four Chopin Scherzi zipped by rather like power lines as seen from the window of a speeding train.

These remarks suggest that the typical reader already has considerable familiarity with the music in question, and a reader who does not know this music may well wish to go to a score or a recording to confirm what the critic says. Yet it is not suggested that the critic has in hand a plausible hypothesis which he would like to see others test with an eye toward giving it a higher degree of confirmation. On the contrary, the reader is being invited to test (if a test is needed at all) whether the interpretation is plausible or not. Take, for example, the suggestion that the contrasting sections of the rondo of the Schubert sonata are like ugly faces one might make to scare children. If you doubted this, I suppose we could sit down at the piano or play the record and I could make faces at the appropriate points. But such a procedure would not determine the truth of an assertion which had hitherto been understood as only plausible: it could only be used to decide on the very plausibility of the suggestion, when that was itself in question. The same goes for the comparison of the slow section of the Beethoven quartet with trying to ride a bicycle as slowly as possible, or for likening the virtuoso’s runs to the meaningless passing-by of electrical lines as observed from a train. We do not need to take a bike into the driveway or hop on the next train to test these comparisons; we are familiar enough with the suggested experiences. The question is, do they tell us anything interesting or illuminating about the music with which they are compared? Once the reader has checked whether the critic’s remarks do justice to the music under consideration, there can be no further demand to verify their truth, as though they were hypotheses. The truth of the connection made-if truth is in doubt at all-lies precisely in the appropriateness of the suggested image to the music in question. Thus it is the plausibility of the comparison which validates the interpretation in the sense desired in criticism.

One theorist who would take a dim view of the position endorsed here is Monroe Beardsley, who has criticized Margolis’s article from a standpoint opposed to my own. In disputing Margolis’s stress on the centrality of the concept of plausibility in critical interpretation, Beardsley sees himself engaged in a “vindication of critical rationality.” He rejects the notion that plausibility is the standard for adequacy in critical interpretation and insists that “any statement that is plausible must be in principle capable of being shown to be true.” Thus to Beardsley’s way of thinking, “all of the literary interpreations that deserve the name” obey what he calls “the principle of ’the Intolerability of Incompatibles’ i.e., if two of them are logically incompatible, they cannot both be true.” [3]

Beardsley’s position rests in part on a distinction between what he calls superimpositions and interpretations. Only the former must be relegated to the realm of mere plausibility, while the latter may be shown to be true or false. The difficulty here is that he fails to provide, and indeed I would argue cannot possibly provide, any general criterion for distinguishing superimpositions from legitimate interpretations. I accept Beardsley’s surmise that a critical account of “Jack and the Beanstalk” which takes the story as a Christian allegory will probably be more of an exercise in validating the critic’s own cherished presuppositions than in understanding the story. Still, we could not be absolutely certain of this until we had read the account itself: it might turn out to be a revealing interpretation, one which throws much light on the story.[4] Beardsley’s example does not help, nor does he offer others which decisively support his claim that critical interpretations can be characterized as true or false essentially beyond the senses of “plausible” and “implausible” I have advocated as central to critical discourse. Though it may be partly miscast, the emphasis placed by Margolis on the importance of plausibility is far more attuned to the actual practices of criticism than Beardsley’s insistence on determining the truth or falsity of interpretations of works of art.

It is worth mentioning in passing that there are critics (certainly not friends of Beardsley) who suppose there exists one particular kind of evidence which is decisive in verifying or disconfirming critical interpretations. Return to one of my earlier examples. It might be imagined by some that a discovered diary which reveals that Schubert precisely had in mind making ugly faces when he wrote the rondo of the D major sonata would count as evidence confirming the critical assertion. But it does nothing of the sort. If we are faced with an hypothesis about Schubert’s state of mind when composing the piece, or his intentions in it, the diary could verify something the music had suggested. However, the point of Schumann’s remark is not to tell us something about the man Schubert; his remark is about the qualities of a piece of music as they are present to experience.

It is impossible to overlook the fact that so much criticism employs just such biographical (some call it “intentionalistic") reference in support of purely aesthetic assertions. But it would seem that such references perform more of a persuasive than a verifying function in critical disputes. In this way, such appeals in criticism are analogous to appeals to authority in scientific disputes. If two scientists have argued at length over the truth of an hypothesis, one may finally attempt to settle matters by citing the fact that various eminent figures in science have agreed with his evaluation of the hypothetical alternatives. Though such resorts to authority do not address the issue in dispute, they often have a certain psychological force, at least insofar as they will perhaps compel an opponent to take more seriously evidence for the other side of the question.

Suppose, for instance, we are arguing over the figure of Eve in Michelangelo’s rendering of the expulsion from Eden. You claim that the figure has a distinctly masculine character. I say that I don’t see it: she looks feminine to me. I point out the feminine lips of the figure, and you answer by describing the breasts as grapefruits stuck on to a male torso. Now someone might attempt to settle our argument by informing us that Michelangelo used male models for his figure of Eve; and indeed, such bits of information are commonly brought to bear in critical disputes in favor of one interpretation or another. Although such information does not speak directly to the issue at hand, it may persuade me to reexamine more closely the opposing view. Still, it need not force me to abandon my strictly critical thesis as regards the masculine character of the figure of Eve. Having used a male model for a female figure is neither a necessary nor a sufficient condition for that figure’s taking on a masculine character. And yet the sex of the model is not wholly irrelevant to a dispute over the character of the figure. It is indirectly relevant, though only to the extent that it can persuade one to evaluate sypathetically the plausibility of the alternative interpretation in question.

I noted that in opposition to the view I am expressing it might be said that such concepts as those of an hypothesis, a fact, certainty, error, verification, and the like-concepts which we usually associate with scientific inquiry-can be said to have a place in criticism. But it should be evident that although each of these concepts may have a place in the structure of critical inquiry, they do not occupy positions quite analogous there to the positions they occupy in natural science, and hence to not have the same meaning in criticism as in science. [5] In science, an hypothesis may or may not be plausible, but it stands as a mere hypothesis so long as its truth is not yet clear. In terms of critical interpretation, on the other hand, an hypothesis would be an interpretive assertion whose very plausibility (as applied to a work of art) is in question or is as yet undetermined. Confirmation or verification in criticism thus involves determinations of the plausibility or implausibility of a critical interpretation, which might often require considerable discussion and analysis both of an art work and of a work of criticism. Interpretation, furthermore,would be considered certain in criticism when its plausibility is firmly recognized and established by careful inspection of the work in question. By contrast, critical error is not, as error usually is in science, a matter of believing a plausible but false critical account to be true. It is a matter of thinking an interpretation to be plausible when in fact subsequent reflection and analysis reveals that it is not. One can always ask whether it is possible ever to establish beyond any doubt the plausibility of critical interpretation. But this should not make us doubt the objectivity of good criticism. For precisely the same sort of question can be raised in connection with the verification of scientific hypotheses. And it would often appear in the context of critical practice to be as possible to determine conclusively whether a critical interpretation should be judged plausible as it is to determine the truth of an hypothesis in the context of scientific inquiry.

Bearing these distinctions in mind, let us now return to Margolis’s article and a list which he provides of what he terms the principal differences between the concepts of truth and plausibility. It is important to note that this list is intended to apply to these concepts across the board, whether in connection with science or criticism.

First, we invoke plausibility only when we cannot actually determine truth.

Second, no plausible account may be incompatible with an admittedly true statement.

Third, neither true nor false statements may be viewed as merely plausible or implausible and neither plausible nor implausible statements are logically precluded from being judged true or false.

Fourth, where the statements “P is true” and “Q is true” are contraries, the statements “P is plausible” and “Q is plausible” are not contraries.

Fifth, statements are judged plausible or implausible in virtue of their use of preferred explanatory models in any given domain.... (p. 117)

By suggesting that “mere” plausibility is to be tolerated only when truth cannot be arrived at, the first position reveals clearly that it should not be applied to criticism, but only to science. Critics do not invoke the plausibility of their interpretations only when they are “unable to determine their truth.” Similarly, the second part of Margolis’s third distinction is wrongly applied to criticism because, as has been shown, it is mistaken to suppose that a critical remark found to be plausible can later be judged to be true or false; again, this is appropriate to science but not to criticism. There is no way in principle to uphold the lasting truth or correctness of some critical account of a work of art over and above assuring-or reassuring-ourselves of its plausibility.

The second position of the list and the first clause of the statement of the third involve an interesting ambiguity as far as criticism is concerned. Certainly these points are consistent with the logical structure of science: if a scientist knows that statements X and Y are incompatible, and if he has evidence establishing the truth of X then he cannot-as one who knows X-continue to assert the plausibility of Y. And if he knows the truth or falsity of any statement, then it is odd for him from his standpoint as one who knows its truth value to term it either plausible or implausible. From that standpoint, plausibility for the scientist has been displaced by certainty. Again, however, the situation in criticism is significantly different. This can be seen by examining a piece of work in which critical interpretation and historical scholarship are brought together. Samuel Butler’s The Authoress of the Odyssey was first published in 1897 and is still available in a modern edition. Butler argues on behalf of two main theses: first, that the Iliad and the Odyssey were composed by two different authors approximately two hundred years apart, the Odyssey being the later of the poems; second, that the Odyssey was written by a young woman living on Sicily.

It is this notion that the Odyssey is the work of a woman which concerns the present issue. In order to make his case, Butler first tries to remove initial doubts on the part of the reader by showing that there were women poets at the time the epic was composed. He then goes on to examine closely internal evidence which the poem itself provides that could lead one to suppose that it was the work of a woman. The poet, for example, has made numerous errors which Butler tells us no man would make, such as describing a boat as having a rudder fore and aft. Much of the story is built around feminine interests of home and family life. In contrast to the foolish foibles of mankind, womankind in the Odyssey is portrayed as something to be taken very seriously. Men are consistently placed in the position of needing the support and wise counsel of women.

I would not wish to defend Butler’s argument as faultless. In particular, I mistrust any chain of reasoning from which the authoress of the Odyssey comes off sounding so much more like a Victorian than a Sicilian young lady. Still, I do not think the reader can go through Butler’s chain of reasoning and numerous citations from the text of the poem without feeling that his case has at least some force. But exactly what is the case one feels the force of here? Homeric scholars have generally ignored his book, having found in it historical inaccuracies and altogether too much unsubstantiated speculation. Butler, on the other hand, was aware of the reception his book was bound to get from the scholars and seemed not to care. Here are the closing sentences of the book:

And now as I take leave of the reader, I would say that if when I began this work I was oppressed with a sense of the hopelessness of getting Homeric scholars to take it seriously and consider it, I am even more oppressed and dismayed when I turn over its pages and see how certain they are to displease many whom I would far rather conciliate than offend. What can it matter to me where the Odyssey was written, or whether it was written by a man or a woman? From the bottom of my heart I can say truly that I do not care about the way in which these points are decided, but I do care, and very greatly, knowing about the way they are decided by sensible people who have considered what I have urged in this book. I believe that I have settled both points sufficiently, but come what may I know that my case in respect of them is amply strong to justify me in having stated it. And so I leave it. [6]

One way to construe these remarks, particularly the penultimate sentence, is to suppose that Butler means that even if it were to be determined finally that the Odyssey was the work of a man, the evidence available in the 1890s nevertheless justified the effort to show that it was a woman. Analogously, we might imagine a medical researcher following a fruitless line of investigation and later justifying his efforts: “We had to pursue that avenue of research given what we then knew, though of course we now see that we were wasting our time.” But this is certainly not what Butler means. It would not make Butler’s book worthless if an ancient tomb in Asia minor were to someday yield unimpeachable evidence that the author of the Odyssey was a man. For it would still be clear that Butler’s case for the author’s being a woman would continue to have critical plausibility and would be worthy on that account alone of continued study by students of the Homeric epics.

It is true that from one perspective Butler's book can be treated as an attempt to provide enough evidence to infer the sex of some ancient bard. From that perspective, The Authoress of the Odyssey is to be taken as a piece of purely historical scholarship, and the discovery of the tomb would naturally force us to abandon the book’s purely historical thesis. But what is much more central to an understanding of Butler’s book is the recognition that it is a work of literary criticism, one which offers a respectably persuasive interpretation of the Odyssey. As a work of criticism it cannot be rendered valueless by finding out that the Odyssey was composed by a man. For it would remain an interesting literary fact, permanantly valuable for lovers of the epic, that the poem can plausibly be read as the product of a woman. No purely historical discovery could be used to deny the essential soundness of the critical interpretation.Only another critical argument could be brought to bear against it. I do not dispute that a determination that the Odyssey was the work of a man would be relevant to the persuasiveness of the interpretation. I do insist that such a determination would not alone render the critical interpretation invalid in the same decisive way in which it would invalidate the historical thesis.

This is why it is so misleading to say, as Margolis does, that “no plausible account may be incompatible with any true account.” Confirmation of the historical hypothesis, “The Odyssey was composed by a man,” would not serve by itself to render implausible the critical remark, “The Odyssey could well have been composed by a woman.” Similarly, Margolis’s assertion that “neither true nor false statements may be viewed as merely plausible or implausible,” while it expresses something which scientific inquiry takes quite for granted, does not apply in the same way to critical practice. Butler might greet the awkward discovery of conclusive historical proof that the author was a man by completely accepting it and yet insisting that it was highly implausible from an artistic standpoint. The plain fact is that there need be no contradiction between Butler’s critical thesis and any historical fact, insofar as his critical interpretation is not an historical inference but is an account of what he takes to be the literary character of the Odyssey itself. In such a case, history and criticism may be relevant to each other, but they may also pass one another by.

As mentioned previously, there is a marked tendency in ordinary discourse to favor a sense of “plausible” which can be more clearly associated with the use of the term in scientific contexts. Hence we find such standard dictionary definitions as “having the appearance or show of truth” or “superficially reasonable or pleasing.” Interestingly, however, this suggestion of superficiality cuts in two directions. By implying that plausible assertions are only superficially reasonable and may not reflect some deeper truth, the framer of such a definition ignores the fact that aesthetic interpretations can be unqualifiedly praised as “plausible” without such implication. Yet there is a relevance of the notion of superficiality to aesthetic criticism, and it lies in the fact that the critic is so often concerned to analyze and elucidate the character of an aesthetic surface. Aesthetic interpretation concerns itself largely with the nature of appearance, whereas science has ever set for itself the goal of understanding the essential nature of a reality which underlies outward (and perhaps misleading) appearance. Critical interpretation of a work of art is always directed toward understanding the nature of the work as it stands before us, as it is present to experience. Insofar as criticism is concerned, that aesthetic object is never a mere stepping stone to a grasp of some further primary reality.

It is just this distinction which prompts us to want to accept for scientific inquiry the traditional empiricist differentiation between the context of our groping towards discovery and the context of our reaching some final justification for the acceptance of an hypothesis. The empiricist wishes us to separate the initial plausibility of an hypothesis, its mere appearance of truth, from its ultimate validation, from the question of whether it can be shown adequately to reflect the underlying structure of reality. The empiricist is clear in his insistence that the plausibility of an hypothesis in the context of discovery is neither a necessary nor sufficient condition for its truth as established in the context of justification. An initial, tentative scientific hypothesis in the context of discovery is a first report; it may be little more than a hunch about the actual nature of things. The best critic display a similar care and humility in the practice of aesthetic interpretation. They too have their hunches, they too issue first reports and are called upon to justify their interpretations. But once arrived at, an interpretation which provides an illuminating way of looking at a work of art, though it be described as “plausible,” need no longer be thought of as a hunch or hypothesis about the work. Its plausibility is emphatically not that of an empirical hypothesis. Far from giving us the mere appearance of truth, a plausible critical interpretation provides us with a truth about how the aesthetic object may be understood. To demand something more under the guise of a desire for greater rigor in criticism is a demand which makes no sense.

This brings me to a concluding remark. When invidious comparisons are drawn between science and criticism, it is often pointed out that explanations in the sciences tend to supersede one another; a new theory, once confirmed, may render its predecessors obsolete. Because so many competing explanations thus exclude one another, there is said to be progress in science (natural science, at least) but not in criticism, where radically different interpetations of the same work can stand side-by-side, none superseding any other. Some theorists see this as pointing toward an inherent weakness in aesthetic criticism as a form of inquiry. They talk as though criticism lacked the requisite apparatus to determine the final truth or falsity of the interpretations it proposes. I have, of course, tried to discredit this view and to replace it with one which construes natural science and aesthetic criticism as fundamentally different kinds of inquiry. But even beyond this, it should not be forgotten that the reason for criticism’s peculiar tolerance of incompatible interpretations is less a matter of the structure of criticism than the nature of works of art themselves. However, if the fact is to be explained ultimately, it is inescapable that the finest works of art are “rich” in precisely the sense that they seem never to exhaust their potential as objects of critical discourse. They may be endlessly interpreted. They may never be definitively captured in any final interpretation or critical analysis. Indeed, this is a measure of their greatness. [7]

 

1. Joseph Margolis, “The Logic of Interpretation,” in Philosophy Looks at the Arts, ed. Joseph Margolis (New York, 1962), pp. 708-718. This is an early statement of Margolis’s position. In later writings, particularly “Robust Relativism,” Journal of Aesthetic and Art Criticism 35.1 (1976): 37-46, he espouses a view closer to the one outlined here.

2. Obviously, I am closely adhering here to analyses of scientific method which can be identified as “empiricist,” as part of a tradition derived from Bacon and Mill and having among its outstanding modern exponents such figures as Ernest Nagel and Carl Hempel. At the same time, I recognize a competing perspective, represented by such writers as Thomas Kuhn, Paul Feyerabend, and N.R. Hanson, which would doubtless resist relegating matters of plausibility so exclusively to the “context of discovery.” (If, for example, truth is a concept which can only be properly used in connection with hypotheses put forward within what Kuhn calls a “paradigm,” it might well be argued that paradigms themselves can only be treated in terms of their ultimate plausibility.) Be that as it may, my attention here is limited to the differing logics of scientific and critical confirmation as they are reflected in the day-to-day practices and procedures of art critics and scientists. At this level the analysis given us by the empiricists is a valid and useful one to contrast with criticism.

3. Monroe C. Beardsley, The Possibility of Criticism (Detroit, 1970), pp. 38-61. For another criticism of Margolis, see Annette Barnes, “Half An Hour Before Breakfast,” Journal of Aesthetics and Art Criticism 34 (1976): 261-71.

4. See also Denis Dutton, “Criticism and Method,” British Journal of Aesthetics 13 (1973): 232-42.

5. For a dissenting view based on Popperian fallibilism, see Ralph Rader, “Fact, Theory, and Literary Explanation,” Critical Inquiry 1 (1974): 245-22. The same volume of Critical Inquiry contains discussions by Stanley E. Fish and Jay Schleusener with a response from Rader, pp. 883-911.

6. Samuel Butler, The Authoress of the Odyssey (New York, 1922), p. 170.

7. Versions of this paper were presented at a meeting of the Michigan Academy of Science, Arts, and Letters, April 4, 1975, and at the Tenth Conference on Value Inquiry, held at the State University College at Geneseo, New York, April 16, 1976. I am indebted to Edward Sayles, George Kerner, Joseph Margolis, and the ever diligent editors of this Journal for their thoughtful commentaries on it.