My response addresses the implications of Thomas Pfau's discussion of the concept of interiority vis-à-vis Kantian and post-Kantian discussions of pleasure and criticism by mapping, then commenting on, key turns in his argument.

As Pfau traces the modern history of interiority, its authenticity depends on a paradox: that "voice" is at once inalienable and socially iterable. For this reason, the notion of interiority names a complex dialectical relation between 1) formal-aesthetic productivity in the work of art and 2) the array of modern disciplines that seek to describe or, as Pfau puts it, "secure," the epistemological and historical consequences of formal productivity. Further, this dialectical relation takes two forms according to Pfau. In the first of these, the form of the work of art is homologous with its beholding intelligence—a position I take to be roughly assimilable to Romantic claims about organic form. In the second, there is a divide between the work's affective quality—which is identified with aesthetic production—and a post-lapsarian consciousness of the discursive world and "its often incompatible interests." As I imagine it, this second form is philosophical kin to the consciousness Schiller ascribed to the sentimental, as opposed to the naïve, poet.

Pfau productively (and polemically, as it turns out) links his inquiry to modern, critical questions about the telos of aesthetic pleasure. Is the formal trajectory of such pleasure its critical articulation or is pleasure instead the basis for the perceiving, experiencing subject's claim to have cognitive acuity and therefore social authority? What, in brief, is the operative relation between aesthetic pleasure and criticism? If pleasure exceeds the range (and grasp) of critical discourse, then what we do is pointless to the exercise of art. If pleasure deludes the subject (be that subject artist or reader) by concealing from subjective gaze implicit critical and social values, criticism might be thought of as the "other " to pleasure—a mental spy, as it were, who discloses what the work of art conceals. The relevance of these alternatives to recent critical debates about Romanticism is clear. Pfau suggests that even before Wimsatt's influential attack on assorted critical "fallacies" pleasure had become a "problem" to be solved (or not) by philosophical aesthetics and its "subsidiary" disciplines, such as "poetics, compositional theory and musical aesthetics." I would query this application of the term "subsidiary," which makes those disciplines that are more practically concerned with art forms less, rather than more, authoritative about such forms. In one sense, making philosophy prior rightly imagines its centrality to all questions—modern and ancient—about the work of representation. But in another, putting philosophy first may put the materiality of art objects, including poems, below the horizon of criticism.

If, as Pfau suggests, the aesthetic is somehow "proto-articulate," it is so either because it is redeemed by criticism or because the "voice" of the aesthetic inevitably constrains or limits criticism. Pfau finds an instructive resolution of this question in Kant's Critique of Judgment. Kant supposes that pleasure arises from the subject's reflexive understanding of its own subjective condition in the moment of aesthetic experience. Imagined in this way, aesthetics is "an encryption of the very intelligence that will constitute itself" in and through the act of interpretation . The confounding wiggle in Kant's claim has of course to do with the corollary he offers for this argument—that such aesthetic experience is "universally communicable." Thus Kant is a key source for the paradox of interiority with which Pfau began. His interrogation of this familiar Kantian difficulty turns to consider his conviction that sensation conveys to the subject the contingent unity of relation thus established for the imagination and the understanding. Pfau astutely characterizes Kant's abrupt turn in his analytic of the beautiful from "a purely abstract, formal dynamic said to occasion the 'feeling' of the beautiful to the ostensibly empirical and material vocabulary of 'sensation'" as a "conceptual tremor" in the third Critique —one that de Man generalized as having to do with Kant's "purely tropological system."

Pfau chooses instead to pursue options suggested by Kant's reliance on "voice." Recalling that Rousseau links pleasure to continual sound —the ebb and flow of water "continued but magnified at intervals" (quoted by Pfau)—Pfau sketches the philosophical implications of duration, particularly that of music, wherein duration makes form—as harmony, as tonal proportion, as design—possible. This assessment is in many ways the most crucial moment in this argument, for it asserts that aesthetic experience is, as Kant insisted, "contemplative"—and, as such, invested in its own prolongation." As the contemplation of the beautiful strengthens and reproduces itself" (says Kant), we gain and thereby recognize our interiority. We need sensation, in other words, to inaugurate the prolongation for which musical duration is a trope as well as key instance.

In modernist aesthetics, Kant's legacy takes the two forms described above. Either the aesthetic provides "a formal rehearsal of the subject's cognitive mobility" or (in the view of its critics) it evokes feeling that is empty of cognitive ground—as though, to return to the musical examples that make Pfau's analysis particularly instructive, to bathe the listener in a emotional soup of sound that refuses or blocks contemplation, as does Andrew Lloyd Webber's score for Les Misérables—my choice of a representative instance of what Pfau calls (by way of other examples) "evacuated liturgy." His more detailed exposition of these alternatives invites analysis and some queries. The first requires a subject who is on the lookout for ways to exercise cognitive mobility, whether that exercise involves tracking a motif in a piano sonata, a figure in Hegel, or fossil and geological evidence of evolution. In each case, Pfau argues, the "empirical practice" of an "observing intelligence" reflects and, in doing so, discerns (Pfau uses the more agent-directed verb "to extract" to characterize this work) "a developmental pattern." What troubles me about this account is the form of pleasure it is said to produce—a species of pleasure that understands itself as correcting sensation, as "the formal device that aims to redeem the materiality of being from its vagrant and unreflected drift through time." Applied to music, this model presupposes a listener who is potentially more aggressive than reflective. Imagine for a moment singing polyphonic music: the subjective pleasure of this activity surely has to do with sustaining, however briefly, tonal patterns and as surely this pleasure must increase with a singer's greater awareness of the inner structure of the polyphony. But it is delusory to imagine that one can redeem the materiality of sounds thus produced. Perhaps more to the point, I understand the pleasure of such song as experienced along side the recognition that the sounds so essential to this pleasure will fade, even when the mind remains conscious of the contrapuntal pattern thus achieved. This double recognition is in part why fading musical notes grip singer and listener. When Pfau suggests that this aesthetic approach involves a "consciousness eager to reaffirm its cognitive authority by isolating recursive, imitative, antithetical or otherwise differential patterns," I balk at the implied generalization of this description. The evolutionary example Pfau offers for this model, Darwin by way of Richard Dawkins's brief for the "selfish gene," is similarly pitched toward contest and victory over time and evolutionary change. What, I wonder, would an evolutionary biologist like Stephen Jay Gould make of presenting evolutionary patterns as this tightly focused and monitored? Pfau's point here is that such a model conveys the first of the two modern approaches to aesthetic pleasure that emerges from Kant and which Eduard Hanslick extends in the mid-nineteenth century. But does he? When Hanslick says that to specify the "content" of a motif for someone, "we will have to play for him the theme itself" (quoted by Pfau), this statement does not as I read it mean that pleasure "has been absorbed into the cognitive play of an attentively listening, analytic intelligence" (Pfau). Why would pleasure and sensation be put aside by such musical attentiveness? The difference between Kant's communicable aesthetic judgment and Hanslick's closed circuit model of sound and a listening intelligence is, as Pfau notes, not good news, for it ushers in a model in which tight little mutually confirming and essentially narcissist relations between speakers and addressees obtain.

The second model of a listener/subject absorbed in sensation and pleasure and for this reason unable to scrutinize or evaluate aesthetic experience is worse. For Pfau as for Charles Rosen, Keats (that is, "To Autumn") and Mendelssohn (of the large symphonic pieces) offer formal brilliance without content—that world of sensation that the young Keats once declared he wished to have. This characterization of Keats looks very peculiar and certainly partial to a certain way of reading this poem and this letter. It would not, I think, stand up well to other examples, say Lamia or the Hyperion poems. Yet it would perhaps be unfair to distract attention from Pfau's larger historical theme by disputing instances. That theme traces a diminuition of Kantian (and Enlightenment) knowability and communicability into Schopenhauer's presentation of feeling and sensation as untainted by thought, the other, or thoughts about others. So construed, the aesthetic becomes co-extensive with the motives of will and thus the body at the expense of cognitive work. For Pfau, this model offers an experience of music in which its "sheer sonority is said to absolve us from the contingent"—from the world of representation as we know it in modernity. I assent wholly to Pfau's sense of the self-deception of this rendering of aesthetic experience and subjective response, though I do not necessarily assent to the examples he offers. The anti-theoretical and anti-selfconscious impulse of this model of aesthetic response produces meaning that depends on "transference or self-projection" as much as "consensus." Against this model, we can perhaps array a different genealogical project in modernity, akin to that indicated at the end of Toni Morrison's Beloved. There passing on or not passing on—the verb Pfau uses to talk about aesthetic genealogy—the story of Beloved is provocatively and ironically enmeshed in the act of story-telling and listening. This is not a story to pass on, Sethe warns, and it is precisely the story that the novel does pass on, but without a comfortable fit between teller and addressee.

In its larger gestures, Pfau's argument seeks to discover a modern ground for the ethical imperative of Kant's notion of "voice" as at once interior and communicable. That ground, he suggests, might be found where acts of naming (such as poetic figures) "hover[ ] between the creative and the recursive, the tropological and the referential." It is not sufficient to read Romanticism as wholly expressive, immersed in the young Keats's "world of sensation rather than thoughts." Neither is it sufficient to oppose the inventive and non-recursive features of Romantic writing to its culture as a field of references from which Romantic tropes are then said to deviate. What is needed instead is a criticism flexible enough and ironic enough to move between these poles of aesthetic response—within and without Romanticism.