At the center of Susan Sontag's discussion of style in "On Style" is her emphasis on the role of "will" in the creation and reception of art:
Perhaps the best way of clarifying the nature of our experience of works of art, and the relation between art and the rest of human feeling and doing, is to invoke the notion of will. It is a useful notion beacuse will is not just a particular posture of consciousness, energized consciousness. It is also an attitude toward the world, of a subject toward the world.
The complex kind of willing that is embodied, and communicated, in a work of art both abolishes the world and encounters it in an extraordinarily intense and specialized way. This double aspect of the will in art is succinctly expressed by [Raymond] Bayer when he says: "Each work of art gives us the schematized and disengaged memory of a volition." Insofar as it schematized, disengaged, a memory, the willing involved in art sets itself at a distance from the world. . . .
Art must distance itself from the world in order to become visible as art in the first place. It comes into being as a version of the world, as an aesthetic reproduction, and for this to be accomplished as thoroughly as is necessary, both for artist and audience, an act of "will" is required. And this act could be described as "dehumanized," since
in order to appear to us as art, the work must restrict sentimental intervention and emotional participation, which are functions of "closeness." It is the degree and manipulating of this distance, the conventions of distance, which constitute the style of the work.
Although I really don't understand how this effort of aesthetic willing could itself be identified as a work's "style" (more on this below), otherwise the concept of "will" as the imposition of a purely formal status on a text, image, or soundscape seems a cogent enough formulation. Most readers of novels, viewers of paintings or sculpture, and listeners to music want to disregard art's distancing effects and recover a notional "closeness" Sontag duly reminds us is antithetical to the very creation of art.
But again Sontag can't seem to accept the full implications of her position. She must add a codicil:
A work of art is first of all an object, not an imitation; and it is true that all great art is founded on distance, on artificiality, on style, on what Ortega [y Gasset] calls dehumanization. But the notion of distance, (and of dehumanization, as well) is misleading, unless one adds that the movement is not just away from but toward the world. The overcoming or transcending of the world in art is also a way of encountering the world, and of training or educating the will to be in the world. . . .
This encountering of the world is what Sontag calls the "function" of art, which she thus substitutes for "content" in opposition to the art "object." She appears to believe that in so doing she is banishing "content" as a subject of critical discussion, but I can't really see how "function" operates as any less of an obstacle to the appreciation of style--which for Sontag remains the only "substance" of art--as the content it effectively displaces. If previously a work of art could be judged by the moral or social ramifications of its "content," what, under Sontag's formulation of "function, " would prevent it from being judged by how acceptably it performs the task of "educating the will to be in the world"? Art would still be valued at least as much--probably, inevitably, more--for its utilitarian intervention in "the world" as it would as a self-sufficient creation, an act of aesthetic will.
The function of a work of art is to be itself. It doesn't engage in "training" for anything other than subsequent, perhaps more "educated" experiences of art. No doubt some people regard some works of art as having provided them the kind of enhanced re-engagement with the world of "real" experience that Sontag invokes--keeping in mind that works of art themselves belong to the world of experience--but to posit that art has a function that makes it useful to the world for reasons other than being availabe to experience, and this function applies at all times for all people, only gives away to the philistines what Sontag otherwise seems to want to preserve--the integrity of art.
Part of the reason for Sontag's readiness to trade "object" for "function" may lie in the ultimate imprecision of her notion of "style" in art, especially as style is embodied in works of literature: "Style is the principle of decision in a work of art, the signature of the artist's will." "If art is the supreme game which the will plays with itself, 'style' consists of the set of rules by which the game is played." "To the extent that a work seems right, just, unimaginable otherwise (without loss or damage), what we are responding to is a quality of its style." "An artist's style is, from a technical point of view, nothing other than the particular idiom in which he deploys the forms of his art." "[E]very style embodies an epistemological decision, an interpretation of how and what we see." Nowhere in "On Style" is there discussion of color or brushstroke, tone or harmonics, phrases, sentences, or paragraphs. "Particular idiom" in poetry or fiction is never associated with specific effects of language, with the use of words.
The closest Sontag comes to a real analysis of style is this brief discussion of Gertrude Stein:
The circular repetitive style of Gertrude Stein's Melanctha expresses her interest in the dilution of immediate awareness by memory and anticipation, what she calls "association," which is obscured in language by the system of the tenses. Stein's insistence on the presentness of experience is identical with her decision to keep to the present tense, to choose commonplace short words and repeat groups of them incessantly, to use an extremely loose syntax and abjure most punctuation. Every style is a means of insisting on something .
It would be hard not to notice Stein's "circular repetitve style"--her particular idiom of "commonplace short words" and "extremely loose syntax"--but this sort of focus on style as the deployment of language is relevant to all writers worth our notice, and otherwise "On Style" defines style much more abstractly as "principle of decision," "set of rules," and "epistemological decision." And even here Stein's prose style is summed up as an aspect of will, as the "means of insisting on something," rather than as the enlistment of words in an aesthetically compelling verbal composition. A writer's "style" can be examined for its successes and failures in meeting the latter goal; as an embodiment of "will" it remains, for me at least, rather too mistily metaphysical.
On the other hand, Sontag seems correct to me when she concludes the essay by reminding us that "In the strictest sense, all the contents of consciousness are ineffable," that "Every work of art, therefore, needs to be understood not only as something rendered, but also as a certain handling of the ineffable."
In the greatest art, one is always aware of things that cannot be said. . ., of the contradiction between expression and the presence of the inexpressible. Stylistic devices are also techniques of avoidance. The most potent elements in a work of art are, often, its silences.
I would only add that the "silences" cultivated by great art are "present" because the work makes room for them in a concrete way. They are incorporated into the work as "ineffable" but real. (The New Critics might have called this ineffable quality "ambiguity," something half-said but not fully said.) The specific way in which, through its style, the work of art invokes a fruitful silence is always still worth attention.