Monroe C. Beardsley, Aesthetics: Problems in the Philosophy of Criticism, 1st ed., 1958; 2d ed., 1981.

6. Expression

The highly critical account of the Expression Theory of art given in the first edition (Chapter 7) owed much to a few philosophers (such as O. K. Bouwsma and John Hospers) but knowingly ran counter to the view of art that then very widely prevailed. Not that the Expression Theory had a single form, or even a canonical form, for its main progenitors (Bosanquet, Croce, Collingwood) had offered different formulations. But there was a general assumption (especially among critics, who seldom examined it) that artworks -- or successful artworks -- are created by a process in the course of which an artist expresses his emotions, and that the special character and value of an artwork is the result of its having been brought into being in this way.

Against this assumption it was necessary to argue, in various ways, that such a view could not be very coherently worked out; that it did not by any means give a true description of how artworks are generally made; that the special character and value of an artwork can be accounted for quite independently of assumptions about its origin or manner of production, and so forth. It was noted that an artwork can be expressive, precisely in displaying fairly intense human regional qualities, such as gaiety, charm, forcefulness, or gloom, without being an expression of anything, in the Expression Theory's sense -- and, indeed, could share many such qualities with natural objects, to which the Expression Theory hardly applies. I think it is fair to say that in the past two decades this general view of the arts -- that what was true and important in the Expression Theory can be stated in terms of (let us say) aesthetic qualities -- has come to be the prevailing view. (For a systematic critique see Haig Khatchadourian, "The Expression Theory of Art: A Critical Evaluation.")

Not that this Anti-Expressionist view has been unresistingly acceded to, or that it has been thoroughly understood and its implications heeded by those whose business it is to talk about the arts. What purport to be modified versions of the Expression Theory have occasionally been presented, and excellent points made, although it is questionable that they have been properly named. For example, R. K. Elliott ("Aesthetic Theory and the Experience of Art") has argued persuasively that our experience of artworks is properly quite different from our experience of other objects, whatever qualities they may offer us: we can experience the work "from within," that is, "as if one were the poet or artist" (p. 112), experiencing its "expression after a certain imaginative manner as one's own." This experience is eloquently described: we can hear the musical nassage "as if someone were expressing his emotion in and through the sounds as a person does in and through his voice" (p. 119). This is like finding in the music (or painting, or whatever) something analogous to the dramatic speaker of a poem; we need not at all attribute the expressed emotion to the composer, or indeed anyone very definite. But then it seems clear that this phenomenology of aesthetic experience is not really, in the traditional sense, an Expression Theory, but something better -- and something quite consistent with the Anti-Expressionist view, as I understand and defend it.

The key distinction, it seems to me, has to do with the way a theory of art connects two concepts: (1) what the artwork expresses, and (2) what the artist expressed. Essential to an Expression Theory is that it explains (1) in terms of (2); it is the act or process of expression that is basic, and statements about the expression in the work carry an implicit reference to the expressing agent. On the opposed view, one can still find occasion to speak of an artist as expressing something (although this idiom may be misleading and thus thought avoidworthy), but the act of expression will be regarded, roughly, as the act of creating something expressive. Thus (2) is explained in terms of (1): the basic notion is still that the work has its own forms and qualities, and our relationship to them in no way depends on any assumptions about their creation or creator.

Hence Alan Tormey's illuminating analysis of expression (The Concept of Expression) tells us about the conditions of expressing, but rejects the Expression Theory -- the core of which he formulates as the proposition that if an artwork has an expressive property Q, then the artist, in creating that artwork, expressed a feeling state F of which Q is the "qualitative analogue" (p. 103): tender music has the qualitative analogue of the artist's tenderness in composing it. Expressive properties are properties whose names also designate intentional states of persons: anguish, nostalgia (p. 128). It seems to me that this is a somewhat too narrow definition of expressive properties, for I think we would want to include such (metaphorical) properties as splendor, charm, majesty, dignity, wit which can (literally) be properties of human beings, although they are not intentional states. Tormey also makes a very interesting proposal about the relationship between expressive and nonexpressive properties (compare the discussion above, Section 3, of aesthetic and nonaesthetic properties). Since intentional states of persons are expressed in behavior that is partly constitutive of them (making an angry face or gesture is part of what it is to be angry), it follows that an analogous relationship holds between expressive properties (say) of music and such properties as rhythm, melodic intervals, chord-progression (pp. 130-31). This proposal is not quite definite, since the closeness of the analogy is not specified; but if it entails that the nonexpressive properties of music are partly constitutive of its expressive properties, I think it runs into the difficulty of understanding how local qualities can be ingredients of regional ones.

In his complex and idea-rich book, The Mind in Art, Guy Sircello has expounded and defended an original form of the Expression Theory, in explicit correction and supplementation of what he calls the current "Canonical Position." The key to his theory is the concept of an "artistic act" -- such an act as we might describe by saying that "In such-and-such an etching (of a nude middle-aged woman) Rembrandt has treated (depicted, regarded, considered) his subject coldly and dispassionately." The act is an act of the artist, involving his attitude; the act is said to be "in the work," in that we discover its nature from the work. Certain "anthropomorphic qualities," then -- coldness, irony, compassion -- belong to an artwork "in virtue of what the artist does in that work" (p. 26). Thus Sircello has called attention to an important and neglected feature of artworks, and with numerous telling examples. The doubt that lingers with me, however has to do with the still puzzling notion of an act being in a work. Certainly painting the work and giving it a coldness of attitude is an act, but the coldness is only contingently related, I think, to Rembrandt's actual attitudes. I am drawn rather to Jenefer Robinson's view ("The Eliminability of Artistic Acts") -- which has affinities for that of R. K. Elliott -- that the attitude we find in the work (because the wrinkles are clearly etched and the sagging flesh exposed in a cold light, and so forth) is best attributed to an implicit viewer, a fictional creation (cf. p. 378 below).

Nelson Goodman's analysis of expression (Languages of Art, Chapter 2) is really an analysis of expressive qualities, and, I would say, belongs to a Semiotic Theory of art, rather than an Expression Theory. He rejects the Canonical View because it neglects or ignores an essential element that must be present if an artwork is to express serenity, say: that is, a semantic dimension. When serenity is expressed by a painting, three conditions are met: (1) the predicate "is serene," or some coextensive predicate, applies to it (Goodman reluctantly permits us to say loosely that the painting has the property of serenity); (2) the painting refers to that predicate (or property); and (3) the predicate is applied to the painting metaphorically (only sentient beings can be literally serene). Conditions (1) and (3) would, of course, not be contested by the Canonical Position, but (2) is a departure. It is Goodman's contention that artworks are characters in exemplificational symbol systems (as well as other kinds): they refer to some of their properties.

"Reference" has to be taken very broadly and understood to be constituted by a wide variety of conditions, actions, resolutions, stipulations, and so forth. But if a painting displays serenity to a notable degree and if the painting itself is displayed (in another sense) on the wall of an art-gallery, so as to call attention to its qualities, then it does seem defensible to say that the painting refers (is used to refer) to serenity (see "Understanding Music"). I don't think one can manage by sticking to Goodman's preferred idiom and talking only of predicates, and I think there are some puzzles about the notion of referring to properties, once properties are separated from predicates (see "Semiotic Aesthetics"). Nor do I think the aesthetically interesting and significant properties of artworks are always the ones referred to, even in this weak sense. Still, Goodman's concept of exemplification has already proved to be a valuable contribution to aesthetic inquiry.


Beardsley, M. C.
"Semiotic Aesthetics and Aesthetic Education." Journal of Aesthetic Education 9 (1975).
"Understanding Music." In Kingsley Price, ed., On Criticizing Music: Five Philosophical Perspectives. Baltimore: The Johns Hopkins University Press, 1981.
Brentlinger, Ann F.
"Exemplification of Predicates." Nous 4 (1970).
Elliott, R. K.
"Aesthetic Theory and the Experience of Art." Proceedings of the Aristotelian Society (1967).
Gombrich, E. H.
Meditations on a Hobby Horse. London and N.Y.: Phaedon, 2d ed. 1971.
Goodman, Nelson.
Languages of Art. Indianapolis: Hackett, 2d ed. 1976.
Hospers, John.
"Professor Sircello on Expression." The Personalist 56 (1975).
Howard, Vernon.
"On Musical Expression." British Journal of Aesthetics 11 (1971).
Jensen, Henning.
"Exemplification in Nelson Goodman's Aesthetic Theory." Journal of Aesthetics and Art Criticism 32 (1973).
Khatchadourian, Haig.
"The Expression Theory of Art: A Critical Examination." Journal of Aesthetics and Art Criticism 23 (1965).
Robinson, Jenefer.
"The Eliminability of Artistic Acts." Journal of Aesthetics and Art Criticism 36 (1977).
Sircello, Guy.
Mind and Art: An Essay on the Varieties of Expression. Princeton: Princeton University Press, 1972.
Sparshott, Francis.
"Goodman on Expression." The Monist 58 (1974).
Tormey, Alan.
The Concept of Expression. Princeton: Princeton University Press, 1971.