Philosophy Looks at the Arts
ed. Joseph Margolis
(Third Edition, 1987)
[484]Robust Relativism
JOSEPH MARGOLIS
There seems to be a simple way to refute relativism. Construe it as a conservative thesis: that, for some set of judgments, it is not the case that no judgments can in principle be valid (skepticism) or that judgments can be validly defended on one principle only (what Richard Henson has recently termed «universalism").' Assign truth-values, then, to judgments on relativistic grounds and assume that, in relevantly significant disputes, the correct assignment of incompatible truth-values depends on the use of competing (relativistic) "principles." There is no need to attempt to in- dividuate such principles. The point of the exercise is that, on the hypothesis, relativism leads to contradiction, since judgments would then be able to be validly shown to be both true and false.
The argument is impeccable but indecisive--for an elementary reason. Grant only that a putatively relativistic set of judgments lacks truth-values (true and false) but takes values of other sorts or takes "truth-values" other than true and false. For example, if judgments are said to be probable (on the evidence) rather than true, then it is quite possible that judgments otherwise incompatible- -as true or false-_are equiprobable (on the evidence).? This is not to say that considerations of probability entail relativism; but it is also not to deny that they could be construed relativistically. In any case, the refutation of relativism fails so far forth if there is a set of judgments that relativism claims for its own, to which not truth and falsity but values that, interpreted on the model of truth-assignments, would lead to contradiction do not therefore thus do so. It is of course also possible to hold that judgments are relativized in the sense that every validating "principle" is said to subtend its own sector of judgments and that no two
[485]
principles have intersecting sectors.? But, although this is a possible strategy, it is quite uninteresting, since what we want to consider are the prospects of a robust relativism, that is, a relativism that admits some range of competing claims, claims for which there are at least minimal grounds justifying the joint application of competing principles--hence, that admits not only incompatible judgments relative to any particular principle but also what may be called "incongruent" judgments, judgments that construed in terms of truth and falsity would be incompatible and that involve the use of predicates jointly accessible to competing principles. The weaker form of relativism is uninteresting whether truth itself is thought to be relativized to a particular language or whether a restricted range of judgments is thought to be defensible only in terms of some particular convention or "implicit agreement.»S
Still, the distinction between the two sorts of relativism suggests some necessary constraints that a viable and robust relativism would entail: (1) the rejection of skepticism and universalism for a given set of judgments; (2) the provision that such a set of judgments takes values other than truth and falsity and includes incongruent judgments; (3) the rejection of cog- nitivism (entailed by [(2)], in any case--that is, the rejection of the view that, for the properties ascribed in the judgments in question, we possess a matching cognitive faculty [perception for instance] the normal exercise of which enables us to make veridical discriminations of their presence or absence);" (4) the admission of the joint relevance of competing principles in validating the ascriptions or appraisals in question (entailed by [(2)], in any case--that is, the admission of some theory explaining such tolerance).? On reflection, these four conditions appear to be sufficient as well as necessary for the provision of a robust relativism. They are, in any case, jointly compatible and, together, they undercut what may fairly be taken to be the least specialized attack on relativism that could be mounted. I shall take a theory to be relativistic, therefore, if it meets our four conditions- which, on the analysis sketched, is equivalent to the first two. It is important and useful to note that no constraints at all are placed on the kinds of judgment that may be construed relativistically, for instance, as between judgments that are and are not value judgments.
Having said this much, let me proceed, first, polemically, to provide grounds for thinking that, in the context of aesthetics or of the aesthetic appreciation of the arts, there are at least three distinct ranges of judgment that may be strongly defended as tolerating or even requiring a relativistic construction; and secondly, more affirmatively but very briefly, to sketch a theory in virtue of which those findings may be sustained. In the first portion of the argument, then, I shall try to show that, for each of the three domains to be marked out, well-known arguments (at least im-
[486]
plicitly) opposed to a relativistic construction are inherently indecisive. In the second portion of the argument, then, I shall try to say what it is about the nature of art and judgment that sustains a relativistic thesis. It should be said at once, however, that it is no part of my thesis that all judgments (taken collectively) may be defensibly construed as behaving relativisti- cally-which of course would involve construing truth relativistically. That would be tantamount to retreating to a radical version of the weaker sense of relativism; and, in any case, I am persuaded that such a view is incoherent. So it may be insisted that a further condition (5) should be appended, namely, that relativistic sets of judgments presuppose some range of non-relativistic judgments, or that relativistic judgments are dependent on there being some viable range of non-relativistic judgments. But I take (5) to be entailed by (1). In any case, the provision precludes the possible embarrassment of conceding that we may wish to hold it true that relativistic judgments ("incongruent" in the sense supplied) do have the values (other than true and false) that they are said to have. It may also be claimed that genuinely relativistic theories should be distinguished sharply from theories that merely admit that the validity of any range of judgments is relative to the supporting evidence or supporting considera- tions on which that is said to depend. So a further condition (6) may be required, namely, that a set of judgments is relativistic if their validation is determined by considerations bearing on the individual sensibilities of anyone who relevantly judges. (This may in fact be the fair sense of the relativistic interpretation of Protagoras' dictum.) But, the thesis to which (6) is contrasted is itself tautological; and, also, (6) appears to be entailed by (2). Still, (6) is essential, since it is surely with regard to varying personal sensibilities that we anticipate the relevant specimens to arise; and (6) precludes mere expressions of differing preference, since the expression of a preference is not as such a judgment.
Turn, now, to our specimens.
Frank Sibley, in a well-known series of papers, has argued that a particularly important set of aesthetic properties are not in any positive way "condition-governed" and that their discrimination requires the exer- cise of taste or perceptiveness. As he puts it,
We say that a novel has a great number of characters and deals with life in a manufacturing town; that a painting uses pale colors, predominantly blues and greens, and has kneeling figures in the foreground; that the theme in a fugue is inverted at such a point and that there is a stretto in the close; that the action of a play takes place in the span of one day and that there is a reconciliation scene in the[487]
fifth act. Such remarks may be made by, and such features pointed out to, anyone with normal eyes, ears, and intelligence. On the other hand, we also say that a poem is tightly-knit or deeply moving; that a picture lacks balance, or has a certain serenity and repose, or that the grouping of the figures sets up an exciting tension; or that the characters in a novel never really come to life, or that a certain episode strikes a false note. It would be neutral enough to say that the making of such judgments as these requires the exercise of taste, perceptiveness, or sensitivity, of aesthetic discrimination or apprecia- tion; one would not say this of my first group. Accordingly, when a word or expression is such that taste or perceptiveness is required in order to apply it, I shall call it an aesthetic term or expression, and I shall, correspondingly, speak of aesthetic concepts or taste concepts.About these, Sibley claims, "there are no non-aesthetic features which serve in any circumstances as logically sufficient conditions for applying aesthetic terms. Aesthetic or taste concepts are not in this respect con- dition-governed at all.» Of course, Sibley clearly means to hold that the discrimination involved is, in some sense, perceptual or perception- like- informed by taste or perceptiveness- and that the capacity in question is not to be understood in terms of any form of intuitionism.10
It is, admittedly, not clear whether Sibley can escape intuitionism; and it is entirely reasonable to claim that some of the concepts that Sibley regards as aesthetic are condition-governed and that some that he regards as nonaesthetic (and that are also condition-governed) are, on a perfectly reasonable usage, actually aesthetic or aesthetically important-_-for instance, the discrimination of the fugal form and its complications and the dis- crimination of musical unity."' The central question remains, what of those uses of aesthetic terms that are not condition-governed, in Sibley's sense? Can these be reasonably construed as objectively discerned in the work in question? Here, Sibley himself concedes the possibility that ascriptions of the sort he has in mind ("graceful," "dainty," "moving," "plaintive," "balanced, "lacking in unity," and the like-what he sometimes calls "tertiary or Gestalt properties, among others") may merely be "apt rather than true."'2 Sibley himself opts for the objectivity of such qualities on the basis of considerations that quite clearly fail to exclude a decisive alterna- tive. He notices that simple qualities like color admit of "ultimate proof" (that is, proof that they are present) only in the way in which that proof is "tied to an overlap of agreement in sorting, distinguishing and much else which links people present and past; ... where different sets of people agree amongst themselves thus (e.g., groups of similarly color-blind
[488]
people), it is reference to the set with the most detailed discrimination that we treat as conclusive."
His argument continues in the following way:
When I say the only ultimate test or proof, I mean that, since colors are simple properties in the sense that no other visible feature makes something the color it is, one cannot appeal to other features of an object in virtue of possessing which, by some rule of meaning, it can be said to be red or blue, as one can with such properties as triangular, etc. With colors there is no such intermediate appeal; only directly an appeal to agreement. But if there are aesthetic prop- erties-the supposition under investigation-they will, despite dis- similarities, be like colors in this respect. For though, unlike colors, they will be dependent on other properties of things, they cannot, since they are not entailed by the properties responsible for them, be ascribed by virtue of the presence of other properties and some rule of meaning. Hence a proof will again make no intermediate appeal to other properties of the thing, but directly to agreement. I*Sibley adds that "this agreement is not easy to describe. Not any agreement will do; the fact that some of us, here and now, make identical discrimina- tions need not settle the color of things."'S This shows reasonably clearly that Sibley thinks that the "perception" of aesthetic or tertiary qualities is not dissimilar in an essential regard from the perception of colors: the agreement involved is an agreement about perceived (though dependent) qualities. And so, in effect, Sibley subscribes to some form of cognitivism (if not intuitionism); he must reject our condition (3)-hence, relativism. But that's just it. What Sibley needs is a theory of perception and perceptual qualities that would justify construing the qualities in question as perceptual qualities and not merely as qualities such that, in a sense that conforms to the enormous variability of such judgments, it would be apt but not true to say that this poem or sculpture "has" it. In an earlier paper, Sibley says quite explicitly that "aesthetics deals with a kind of perception," and appeals to the case of the color-blind man to clarify the nature of defective aesthetic perception." On the other hand, in spite of his insistence that "some aesthetic judgments may be characterized as right, wrong, true, false, undeniable," he actually favors the alternative theory at times, conceding that, even for his own cases, "for some range of judgments we prefer terms like 'reasonable,' "admissible,' 'understandable' or 'eccentric' to 'right' and 'wrong. »17
Sibley also has considerable difficulty in explaining how to select the aesthetic "elite" whose discrimination is relatively reliable simply because
[489]
he fails to supply a theory of the requisite perception in terms of which to account for and to correct the discrimination of any would-be aesthetic percipient. But the absence of such a theory places his advocacy of cognitivism in doubt, since, for any claim that putatively relies on the exercise of a cognitive faculty (perceptual, for instance), a theoretical basis must be provided for distinguishing between what actually is the case and what only seems to be so.' This condition must be satisfied whether the properties in question are said to be simple or complex. Sibley's admission, however, of what it may be apt but not true to say is incompatible with his particular claims of aesthetic objectivity, since judgments of what may be apt but not true to say cannot preclude incongruence (in the sense supplied). His concession, in short, precludes the application of the req- uisite "is"/"seems" contrast, where the concession has force; and where he would deny its force, he lacks the requisite theory. Hence, Sibley's position is subject to a complex dilemma: either (a) his aesthetic concepts are condition-governed (since dependent) and thus enter inferentially into judgments that are straightforwardly true or false; or else (b) they are not condition-governed (though dependent) and, since they enter into judg- ments that are straightforwardly true or false, Sibley is committed to some sort of intuitionism; or else (c) they are not condition-governed (though dependent) and enter into judgments that can be apt or inapt or the like but not true or false.
It is, I think, fair to say that the concepts Sibley is chiefly concerned with ("graceful," "moving, "balanced," "unified," and the like) have a definite use in judgments that depend on the individual sensibilities of different persons. He himself concedes the point, which is tantamount to conceding a relativistic thesis. Some of his opponents (Peter Kivy, for instance) 1? wish to show that these concepts are used in an ordinary condition-governed way, but they have not shown (and it is difficult to imagine how they could possibly show) that such concepts are never, or are not even characteristically, used in a way that is either not condition- governed (in Sibley's sense) or if condition-governed not governed in such a way as to lead to judgments that are straightforwardly true or false. Let it suffice, then, that judgments that Sibley says involve taste or perceptive- ness or sensibility--ranging over much of what is typically noted in ap- preciative discourse, without directly involving (but not necessarily exclud- ing) evaluative distinctions?-_-may be construed, and may even need to be construed, relativistically.
The other two quarrels I wish to pick are drawn from Monroe Beardsley's relatively recent book, The Possibility of Criticism,? Again, I mean to argue in each case primarily on the basis of internal evidence.
In speaking about a critic's judgments (he confines himself here to
[490]
judgments of literature), Beardsley has the following straightforward view to present:
What is the point of making a literary judgment and arguing for it? My answer to this question--which I shall defend here- is simple and old-fashioned. It is to inform someone how good a literary work is. But philosophers are rightly suspicious of this so-called "informing, if it merely evokes verbal agreement but brings no further satisfaction to the hearer. [Still] there is a proximate end in judging-namely, to provide information about value.Judgments of course call for supporting reasons. Beardsley claims that the reasons a critic supplies in order to justify his judgment conform to the "'ordinary' sense" of reasons; that is, they "have a bearing on the truth of the judgments." He adds that "the relevance of such reasons presupposes that the judgments can be true or false";, but he concedes in the very same context, noting that this runs contrary to his own view about criticism, that "there might be reasons for making a certain judgment that are not reasons for saying it is true, if it should be the case that judgments cannot be true or false. '23 So Beardsley in effect concedes our condition (2) or at least an essential part of it. He even concludes as a result: "So our first question is whether in fact critical judgments have a truth-value--i.e.,are either true or false'24-which, under the circumstances, we must understand to mean, whether all relevant judgments are true or false.
But he never does show that they are. He does mention P. H. Nowell- Smith's account purporting to show that even so-called verdictives (in J. L. Austin's sense), estimates for instance, though not usually said to be true or false, nevertheless "surely involve a claim to truth, which may be allowed or disallowed.»25 But he considers no other possibilities. His principal effort is actually directed against an argument of Michael Scriven's, which purports to show that critical reasoning is impossible, in the sense that justifying or explaining reasons is impossible to supply in the way required to support ascriptions of truth.3 Scriven's argument, as Beardsley summarizes it, holds that "it must be possible for us to know that the reason is true, and also to know that it is a reason for the conclusion, before knowing that the conclusion is true" (the so-called "independence requirement').27 Beardsley seems to take it that the refuta- tion of Scriven's thesis entails his own favored view--that critical judgments (in particular, estimates, as he puts it, of "the greatest amount of artistic goodness that [e.g.] the poem allows of actualizing in any one encounter with it") have truth-value. "This," he says, "I am convinced, is what the critic estimates. '8 Again, his argument against Scriven runs as follows:
[491]
But if, as I claim, these judgments are estimates, then some reasons must be used by the critics in arriving at them, and therefore there must be some basic features of literary works that are always merits or defects.But this begs the question with which Beardsley originally began. For, if justification or explanatory reasons may be provided for utterances judg- ments) that lack truth-value, then it cannot be shown that if judgments include estimates and if estimates call for supporting reasons, then all judgments have truth-value. Some estimates may and some may not, in the sense given; and some critical judgments may not behave in the way Beardsley claims estimates do. Also, it is difficult to see how, unless by some sort of cognitivism or the weaker version of relativism, judgments of the kind mentioned above ("concerning the greatest amount of artistic goodness, " etc.) could possibly be said to be straightforwardly true or false. In fact, what Scriven's argument tends to show, as a by-benefit, is that a significant range of critical value judgments, though they call for support- ing reasons, rest on considerations that actually preclude the ascription of the value true; for, as Scriven says, agreement about how some valuational condition must be satisfied often (Scriven apparently thinks, always) "does not exceed the degree of our initial agreement about the merit of the work of art.»30 I take this to accord with what I have elsewhere termed "appreciative judgments," that is, judgments that call for pertinent justify- ing or explanatory reasons but that, depending as they do on personal taste, cannot be binding on another, cannot be simply true, cannot be said to support the relevant distinction between a work's actually having the value in question or only appearing to. With respect to such judgments, I claim, we may say only that it is reasonable, extreme, eccentric, etc., to say that a work has this or that degree of merit rather than that it demonstrably has it. Hence, even such judgments conform to whatever semantic constraints obtain on the use of the predicates in question. There are, therefore, judgments (including some that Beardsley himself considers) that would support the relativistic view, and that might even require it-the view, that is, that justifying reasons may be admitted where particular judgments cannot take the value "true" (and do not, in Nowell-Smith's sense, involve a claim to truth).
The third quarrel concerns the nature of the interpretation, as opposed to the description, of a work of art; and here, I am simply responding to Beardsley's criticism of my own earlier statements on the matter.32 The issue directly concerns what may be definitely found in a work and what lies outside it. Beardsley opposes what I and others emphasize as "the element of creativity in interpretation." He says, "I find myself rather
[492]
severe with this line of thought," that is, the suggestion "that the literary interpreter, too [like the performing artist] has a certain leeway, and does not merely 'report' on 'discovered meaning,'... but puts something of his own into the work; so that different critics may produce different but workling lie the The sainte-pocanitions., like two sopranos or two ingenies
But there are difficulties in his account. For one thing, though he subscribes to what he terms "the Principle of Independence" (that is, "that literary works exist as individuals and can be distinguished from other things"), he claims that what he terms "the Principle of Autonomy" is a postulate "that is logically complementary to the first" (that is, "that literary works are self-sufficient entities, whose properties are decisive in checking interpretations and judgments.34) But I would maintain-and have tried to demonstrate elsewhere]S-_-that problems about the numerical identity of a work of art can be managed without any commitment respect- ing the demarcation between description and interpretation, the demarca- tion between what is in a work and what is not. Only if one held, in addition to a theory about individuating works of art, a compelling theory about the nature and properties of works of art, could one hope to sustain the so-called Principle of Autonomy- -by actually providing criteria for deter- mining putatively internal properties to be or actually not to be internal. Beardsley offers no such theory, as far as I know. But then, it follows, as a second consideration, that he may well have misdescribed the "latitudinarian" view of interpretation; interpretations (in the sense he rejects) may not be simple 'superimpositions, as he says, that is, "inter- pretations" that are merely "ways of using the work to illustrate a pre-existent system of thought [say, in taking the story of 'Jack and the Beanstalk' as Freudian symbolism or as a Marxist fable]", they may actually be needed precisely because there is no sharp demarcation line between what is internal and what is external to a work of art and because what is uncertain in this respect may be important in terms of aesthetic appreciation. Thirdly, the admission of so-called superimpositions would itself be a telling concession if (as is in fact the case) Beardsley has not yet provided the requisite theory in virtue of which superimpositions and "genuine" interpretations can be logically demarcated. Fourthly, the im- plied admission that there is a certain latitude that holds in music and the other performing arts raises (unresolved) questions both about whether there is a clear sense, for all the arts, of the tenability of the Principles of Independence and Autonomy and about what may be the formulable (and relevant) differences between literature and the performing arts. Fifthly, Beardsley himself concedes, in a context in which he opposes an extreme view ingeniously supported by Frank Cioffi," that
[493]
Some things are definitely said in the poem and cannot be overlooked; others are suggested, as we find on careful reading; others are gently hinted, and whatever methods of literary interpretation we use, we can never establish them decisively as "in" or "Out." Therefore whatever comes from without, but yet can be taken as an interesting extension of what is surely in, may be admissible. It merely makes a larger whole. But this concession will not justify extensive borrowings from biography.I cannot see how this concession, generously advanced though it may be, can fail to undermine Beardsley's Principle of Autonomy. Even the ques- ton of biographical reference and of intentional interpretation surely becomes moot--which is not to say of course that critical interpretation lacks rigor altogether.
A sixth consideration concerns the nature of language itself, since Beardsley here restricts himself to literary interpretation. First of all, he rests his case on the strength of the thesis that the interpretation of "textual meaning" (as opposed to "authorial meaning" ' in the sense proposed by E. D. Hirsch)3° is "the proper task of the literary interpreter" and that such meaning "lies momentarily hidden" in, say, some poem, "really is some- thing in the poem that we are trying to dig out, though it is elusive."* But even apart from his confidence about determining textual meaning, Beardsley seems entirely prepared to concede that meanings may accrue to a literary text because of the historical conditions under which a living language is used. In his effort to contrast textual and authorial meaning, for instance, he says that "the meaning of a text can change after its author has died.. .. The OED furnishes abundant evidence that individual words and idioms acquire new meanings and lose old meanings as time passes; these changes can in turn produce changes of meaning in sentences in which the words appear." He offers a curious instance from the work of Mark Akenside, acknowledges that a certain eighteenth-century phrase "has... acquired a new meaning," and even speaks of "today's textual meaning of the line" (in question).A
But if he allows changing textual meanings, he cannot preclude the possibility of incompatible and non-converging literary interpretations in rendering a coherent account, unless he also maintains that there is an executive rule (unformulated) for determining which textual meaning (changing through diachronic changes in language itself) to prefer; after all, large portions of an entire text may be subject to similar changes and may therefore support plural interpretations. But secondly, in this regard, the very theory of linguistic meaning to which he subscribes-William Alston's theory of "illocutionary act potentials" (regardless of its own
[494]
defensibility)42 depends precisely on speakers' intentions; consequently, once again, Beardsley cannot, on his own principles, preclude the prospect of defending non-converging ("creative") literary interpretations. Finally, with regard to literature (and certainly with regard to the other arts), the critic's interpretation is not restricted, as Beardsley claims, merely to ferreting out textual meanings; it is often concerned (as even the admission of diachronic changes in meaning confirm) with plausible ways in which what may be called the artistic design (the internal coherent order of the work) may be construed.43 In fact, the case that Beardsley puts before us (introduced by Hirsch) of Cleanth Brooks's and F. W. Bateson's incom- patible interpretations of Wordsworth's A Slumber Did My Spirit Seal bears this point out convincingly.4 Unfortunately, neither interpretation, however plausible, is entirely unproblematic. Bateson's pantheistic inter- pretation cannot be supported on the basis solely of the so-called textual meaning of the lines No motion has she now, no force, | She neither hears nor sees;| Rolled round in earth's diurnal course, | With rocks, and stones, and trees. And Brooks's interpretation (which, rightly understood, em- phasizes the lover's shock-almost in a clinical sense--reacting to Lucy's death and consequent inertness) is somewhat careless about textual mean- ings but not in a way that vitiates his interpretation. The upshot is that Wordsworth's Lucy poem, contrary to Beardsley's claim, does appear to support two different interpretations of the poem's larger meaning or design (that is, roughly the picture of the imaginative world disclosed in the poem), without even entailing different interpretations of the poem's textual meaning. There seems to be no way to preclude the possibility.
There is then no reason to deny that interpretation sometimes serves to convey a sense of virtuosity in fathoming what is hidden (but describable) in a work of art. But there is no reason to insist that interpretation functions only thus. Beardsley discounts incompatible interpretations in accord with what he calls "the principle of 'the Intolerability of Incom- patibles, i.e., if two [interpretations] are logically incompatible, they cannot both be true [and they implicitly claim to be true]." "Indeed, Beardsley says, "I hold that all of the literary interpretations that deserve the name obey the principle."'S He does not deny that there are "inter- pretations" that could not be jointly true and yet may be said to be plausible; but these are not true interpretations, are merely what he calls "superimpositions." And he fails to notice that falsity may be opposed to both truth and plausibility.# Also, he has not provided either an explicit theory of the nature of a work of art or the requisite criteria for deter- mining what is internal and what external to the work of art itself; consequently, he cannot in principle preclude plural and incompatible interpretations of a literary work just as he cannot in principle distinguish between superimpositions and interpretations that specify what is "momen-
[495]
tarily hidden" in a piece of literature. Hence, he cannot preclude a relativistic conception of interpretation--which may well be not merely tolerated but required.
I have now, I hope, shown (polemically) that a relativistic conception of aesthetic appreciation, of critical judgments (of value), and of literary interpretations is viable, not unreasonable, and possibly even required by the ways in which we attend to works of art. I shall have to be extremely brief about my reasons for thinking that a relativistic account is actually required for these and related distinctions. The argument centers on two considerations. First of all , works of art are what I should call culturally emergent entities.47 I wish to avoid here theorizing in too detailed a way about the nature of art, since our issue does not require it and since controversial details may easily deflect us from our purpose. But the most familiar properties of art, its artifactuality, its internal purposiveness, its being assignable meanings (in various senses), forms, designs, styles, symbolic and representational functions, and the like all call for a sensitivity to cultural distinctions that cannot in any obvious way be directly accessible (unless by postulating some ad hoc intuitionism) to any cognitive faculty resembling sensory perception. But culturally freighted phenomena are notoriously open to intensional quarrels, that is, to identification under alternative descriptions; and there is no obvious way in which to show that plural, non-converging, and otherwise incompatible characterizations of cultural items can be sorted as correct or incorrect in such a way that a relativistic account would be precluded. The proliferation of intensional divergences is as close to the heart of the cultural as anything we might otherwise suggest. One has only to think of ideologies, ideals, schools of thought, traditions as well as the deep informality of the so-called rules of language and of artistic creation. This suggests why it is that the appreciation, the interpretation, and the evaluation of art should behave in accord with relativistic expectations. In particular, the relativistic theory of interpretation is sometimes resisted because one wishes to avoid the somewhat unfortunate habit of speaking of art's being inherently incomplete or defective and awaiting the interpretive critic's contribution in order actually to finish the work. What is initially defective or incomplete, of course, is our understanding, not the work ; but the nature of the defect is such that, for conceptual reasons, we cannot be certain that what is supplied by way of interpretation is really in principle descriptively avail- able in the work itself on the basis of any familiar perceptual or percep- tion-like model, which after all offers us the best prospect of the requisite control. One can expect, therefore, a certain conceptual congruence be- tween the theory of art and the latitude tolerated in the practice of critical interpretation.
The second consideration concerns the nature of values themselves. I
[496]
should hold (controversially, I admit) that persons like works of art are culturally emergent entities--not natural creatures like the members of Homo sapiens: chiefly, because the mastery of language is essential to being a person. ' If this were granted, then the possibility of defénding any form of cognitivism (moral, aesthetic, or any other) with respect to the values appropriate to persons or to their characteristic work is radically undermined. Consequently, the prospects of avoiding a relativistic account of values (and of value judgments), even were it possible to avoid such an account of the presumably descriptive and interpretive levels of our appreciative concern with art, is nearly nil. But noticeably with respect to values, if cognitivism is defeated,"' then we can either retreat to the robust or weaker form of relativism or else, even further but with inevitable dissatisfaction, to a skepticism about values.
There appear to be no other promising strategies.
Notes
1. In an untitled and as yet unpublished book on ethical relativism.
2. Cf. C. G. Hempel, "Inductive Inconsistencies," in Aspects of Scientific Ex- planation (New York, 1965). Cf. also G. H. von Wright, "Remarks on the Epistemology of Subjective Probability," in Ernest Nagel et al., eds., Logic, Methodology and Philosophy of Science (Stanford University Press, 1962).
3. This is, roughly, the theme of conventionalism in values.
4. Cf. Alfred Tarski, "The Semantic Conception of Truth," Philosophy and Phenomenological Research IV (1944), 341-376; cf. also W. V. Quine, Word and Object (Cambridge, 1960), pp. 23-24; and Donald Davidson, "Truth and Mean- ing," Synthese III (1967), 304-322. The requirements of the coherence of interlin- guistic communication entail the inadequacy of such a conception: even if ascrip- tions of "truth" are relativized, we require a conception of truth that is not language-relative even if what is true can only be formulated in a way that is subiect to the local features of particular languages.
5. The thesis has been defended most recently by Gilbert Harman, "Moral Relativism Defended," Philosophical Review LXXXIV (1975), 3-22. Harman's thesis is relativistic not merely in the trivial sense that supporting reasons are relative to considerations of some sort but because "the source of the reasons" (for doing something-_-Harman's concern here is with moral relativism) is one's "sincere intention to observe a certain agreement," ibid., 10.
6. The argument is indifferent to the kind of property considered, though moral properties have traditionally been the principal object of concern. Cf. Joseph Margolis, "Moral Cognitivism," Ethics LXXXV (1975), 136-141.
7. It is possible that one might argue that (2) signifies only that, were the relevant judgments interpreted so as to take truth and falsity as truth-values, we
[497]
should be commited to a single arena of dispute; that the admission of other values does not entail the relevance of competing validating principles. But the intention here is to formulate constraints for the robust, not the weaker, version of relativism. Hence, provision must be made- -even if separately, via (4)- for the joint relevance of competing principles.
8. Frank Sibley, "Aesthetic Concepts," Philosophical Review LXVIII (1959), 421-450; reprinted (with extensive minor revisions) in Joseph Margolis, Philosophy Looks at the Arts (New York, 1962).
9. Loc. cit.
10. Cf. Frank Sibley, "Objectivity and Aesthetics," Proceedings of the Aris- totelian Society, Supplementary XLI1 (1968), 31-54.
11. Cf. Peter Kivy, Speaking of Art (The Hague, 1973), chs. 1-3; also Joseph Margolis, The Language of Art and Art Criticism (Detroit, 1965), ch. 8.
12. "Objectivity and Aesthetics."
13. Loc. cit.
14. Loc. cit.
15. Loc. cit.
16. Cf. Frank Sibley, "Critical Judgments of Aesthetic Value," Philosophical Review LXXIV (1965), 135-159.
17. "Objectivity and Aesthetics."
18. Cf. Isabel Hungerland, "The Logic of Aesthetic Concepts" (Presidential Address of the Pacific Division of the American Philosophical Association, 1962), Proceedings and Addresses of the American Philosophical Association XXXVI, 43-66. Hungerland's statement is, however, too extreme in that it fails to provide for condition-governed concepts.
19. Loc. cit.
20. Cf. Kivy, op. cit., p. 17.
21. Detroit, 1970.
22. Ibid., pp. 63-64.
23. Ibid., p. 71.
24. Ibid.
25. Ibid.
26. Cf. Michael Scriven, "The Objectivity of Aesthetic Evaluation," The Monist L (1969), 159-187.
27. Beardsley, op. cit., pp. 77f.
28. Ibid., P. 75.
29. Ibid., p. 82.
30. Scriven, op. cit., p. 179.
31. Cf. Joseph Margolis, The Language of Art and Art Criticism (Detroit, 1965), ch. 10; and Values and Conduct (Oxford, 1971), ch. 1.
32. The Language of Art and Art Criticism, chs. 5-6.
33. Beardsley, op. cit., pp. 39-40.
34. Ibid., p. 16.
35. The Language of Art and Art Criticism, ch. 4.
36. Beardsley, op. cit., pp. 43-44.
37. Cf. Frank Cioffi, "Intention and Interpretation in Criticism," Proceedings
[498]
of the Aristotelian Sociery LXIV (1963-1964), 85-103. Cf. also, Stuart Hampshire, "Types of Interpretation," in Sidney Hook, ed., Art and Philosophy.
38. Beardsley, op. cit., p. 36.
39. Cf. E. D. Hirsch, Validity in Interpretation (New Haven, 1967); also Joseph Margolis, review of above, in Shakespeare Studies Il (1970), 407-414.
40. Beardsley, op. cit., pp. 32, 44, 47.
41. Ibid., pp. 19-20.
42. Cf. William Alston, Philosophy of Language (Englewood Cliffs, 1964); and Joseph Margolis, "Meaning, Speakers' Intentions, and Speech Acts," Review of Metaphysics XIX (1973), 1007-1022.
43. Cf. The Language of Art and Art Criticism, ch. 3.
44. Cf. Cleanth Brooks, "Irony as a Principle of Structure," in M. D. Zabel. ed., Literary Opinion in America (New York, 1951); and F. W. Bateson, English Poetry: A Critical Introduction (London, 1950).
45. Beardsley, op. cit., p. 44.
46. Ibid., pp. 42-44.
47. Cf. Joseph Margolis, "Works of Art as Physically Embodied and Culturally Emergent Entities," British Journal of Aesthetics XIV (1974), 187-196.
48. Cf. "Works of Art as Physically Embodied and Culturally Emergent En- tities"; also Joseph Margolis, "Mastering a Natural Language: Rationalists vs. Empiricists," Diogenes no. 84 (1973), pp. 41-57.
49. Cf. "Moral Cognitivism",; also Values and Conduct.
50. I had seen, in manuscript, Professor Annette Barnes's paper, "Half an Hour Before Breakfast," criticizing my theory of the logic of interpretive judgments (JAAC, Spring, 1976). I have taken no account of her charges here, both because the paper had not been published at the time this essay was completed (it was in fact completed before I saw her paper) and because I had attempted to answer her in detail, by letter. Suffice it to say that her charges go wrong in a number of ways: (i) I do not maintain that contradictory accounts of anything can be defended as true, either separately or jointly; only that what would, on a model of truth and falsity, be contradictory, may be jointly defended as plausible or reasonable or the like; (in) Barnes offers a set of alternative versions of a tolerance principle for admitting diverging interpretations, that she wrongly takes to be exhaustive and therefore, to capture my own view in a multiple dilemma: not exhaustive both because several of her alternatives involve self-contradictory features (which I explicitly avoid) and because no provision is made for truth- values other than "true" and "false"; (in) Barnes make no provision for what I term the asymmetry of truth and falsity; that, for instance, the "false" is opposed both to the "true" and the "plausible" and the considerations of plausibility do not entail the relevance of considerations of truth; (iv) my argument regarding the logic of interpretation is only applied to those entities that we call works of art; it does not presuppose any theory of the nature of a work of art, though it is compatible with an independent theory that I also support; (v) on my view, it follows that, if interpretation behaves as I claim it does, the properties ascribed to a work of art by way of interpretive criticism cannot be said to be in the work in the sense in which description would require; some of the paradoxes that Barnes attributes to my position simply fail to take account of this important point. “