By Henry Pickford
Martin Shuster’s Autonomy after Auschwitz is an ambitious and impressive work, from which I have learned a great deal. It is ambitious because it aims to situate Adorno’s thought within both specific contexts of the German Idealist tradition (Kant’s theoretical and practical philosophy, Hegel’s philosophy of history) and within a certain region of contemporary Anglophone philosophy oriented around Wittgenstein and neo-Aristotelianism. Shuster’s work is impressive not least because of the extent to which those ambitions are realised. The book undertakes not only a novel and expansive reading of Adorno’s practical and moral philosophy in relation to Kant, on the one hand, and Cavell, on the other, but also a careful exposition of Kant’s changing conception of the highest good within his rational theology, and a re-interpretation of Hegel’s philosophy of history to complement Adorno’s moral theory. My focus here will be on a central line of argument that connects Chapter 1 to Chapter 3 and centers on autonomy, agency and action. Shuster reads Horkheimer and Adorno as claiming that Kantian autonomy itself undermines agency, and then reconstructs Adorno’s moral theory as a response to that deficiency. Shuster and I first discussed these issues on a panel at the meeting of the Association for Adorno Studies in New York City in October 2015, and again on a panel at the Pacific conference of the American Philosophical Association in April 2016; I want to thank him for his clarifications at those events, and for continuing the conversation now.
In Chapter 1, Shuster offers a compelling interpretation of the first essay of Dialectic of Enlightenment according to which Adorno and Horkheimer submit Kantian autonomy itself, rather than, say, instrumental rationality, to critique. Kantian autonomy is understood here not primarily in a metaphysical sense (e.g. as the ability to initiate novel causal sequences in the world), but rather as a normative status or standing in the Sellarsian “space of reasons”, that is, the space “of justifying and being able to justify what one says” and does. On this picture Kantian autonomy necessarily involves one’s responsiveness to, and endorsement of, reasons as reasons. Such a self-reflexive attitude to reasons as norms can in turn be constitutive and expressive of one’s identity as a rational agent (pp. 3, 29). Adorno and Horkheimer claim, on Shuster’s reading, that
Kant’s notion of autonomy, if adopted as a norm, dissolves our capacity for reason, especially practical reason, and thereby our very standing as agents. (p. 2)
For Shuster, Horkheimer and Adorno’s critique is directed at a general and fundamental element of Kantian autonomy, namely its epistemological framework that is logically prior to the material conditions of reality encountered, and the specific rationality deployed, by the Kantian agent. Thus, regarding material conditions, Shuster holds that “Kantian autonomy destroys our status as agents, and this would remain a conceptual issue even if we lived in utopia” (p. 2). And regarding rationality he holds that instrumental rationality is “merely symptomatic of a deeper problem centering on identity, both the identity of the subject (with herself) and the identity of the object (with itself and with the subject’s appraisal of it)” (p. 17). Rather, for Shuster the problem is to be found in Kantian self-consciousness itself, which requires a discernible distinction between subject and object, which in turn rests on a distinction between concept and intuition (p. 20). For Shuster, this epistemological picture alone suffices to entail the dissolution of the normative intelligibility of Kantian autonomy.
He lays out the overall argument in three broad claims, each of which requires careful unpacking.
If I understand the totality claim correctly, Shuster argues as follows. Kantian self-consciousness rests on the epistemic duality of subject and object, concept and intuition. However, according to the Transcendental Aesthetic and the Transcendental Deduction of the First Critique, the objectivity of experience is constituted by the a priori forms of intuition of sensibility and the categories of the understanding, hence by subjectivity. In fact, transcendental subjectivity (the “I think”) and transcendental objectivity (the “object in general”) are both constituted through the acts of synthesis: they are the two sides as it were of the transcendental unity of apperception. Thus self-consciousness requires the strict dualism of subject and object, concept and intuition on the one hand, but on the other hand undercuts the independence of the object by constituting the conditions of its possibility. This instability, in extremis, just is the dialectic of enlightenment:
There is no being in the world that knowledge cannot penetrate, but what can be penetrated by knowledge is not being. Philosophical judgement, after Kant, aims at the new yet recognizes nothing new, since it always merely repeats what reason has placed into objects beforehand. […] Both subject and object are nullified. The abstract self, which alone confers the legal right to record and systematize, is confronted by nothing but abstract material, which has no other property than to be the substrate of that right. The equation of mind [Geist] and world is finally resolved, but only in the sense that the two sides cancel out. (DE 19-20/5:48–49) (pp. 23–4)
As Shuster puts it:
[T]he more of the world that subjectivity seeks to identify and bring under its domain, the more it destroys its own conditions of possibility and thereby itself. Since all objectivity merely becomes the mirror of subjectivity, the distinction between the subjective and the objective disappears, and the pole of pure subjectivity looks entirely like the pole of pure objectivity; hence the return of fate, of myth, and of nature. […] To the extent that seemingly anything can be an object of knowledge […] the dialectic of enlightenment can apply to everything. (pp. 22–3)
Having shown that the dialectic of enlightenment, and more specifically the dissolution of subjectivity, can apply to any object of knowledge, Shuster’s second large-scale move—the necessity claim—is to argue that it must apply to every object of knowledge. The argument runs as follows:
Because [by the totality claim] a correspondence between how reason conceives the world to be and how it is cannot be presupposed, the only path left open to reason is to analyze its own functioning. (p. 27)
The Critical turn, the pursuit of justification by means of rational self-reflexive inquiry, is an exercise of autonomous reason. But in that pursuit of justification, as Kant explains in the Dialectic of Reason, reason seeks the unconditioned, which it ultimately postulates in the form of rational ideas. These rational ideas—God, the immortal soul, the highest good—cannot be objects of experience, and thus are elements of “myth” and “also fall under the sway of the dialectic of enlightenment” (p. 32). As Shuster rightly notes, Horkheimer’s and Adorno’s conception of the dialectic of enlightenment is a formal analogy to Kant’s dialectic of reason, now applied to all reasoning.
As we strive for autonomous agency, and considering that only self-reflexively can we ground such autonomy, we repeat this dialectical procedure at every level of human activity. (p. 30)
I take this to mean: if we take ourselves to be agents, and if being an agent requires autonomy, “then no external justification, no justification apart from one’s own reason, can serve to ground that autonomy” (p. 30), but this undercuts the objectivity of the subject/object, concept/intuition dualism that is required for self-consciousness, thus dissolving subjectivity and agency.
The final large-scale claim by Shuster—the practical Reason claim—is that “[p]ractical reason, in that it serves as the capacity for generating practical reasons, under the dialectic of enlightenment and with subjectivity, is dissolved” (p. 38), that “a certain sort of subjectivity (enlightenment thinking—Kantian autonomy) does undermine practical reason (and everything else subjective besides)” (p. 39). Practical reason presupposes an agent reasoning what to do. Kantian agency presupposes autonomy, which (by the totality and necessity claims) undermines the conditions of self-consciousness, hence of subjectivity and hence of agency. These then are the three broad claims of the argument as I understand it.
“The dialectic of enlightenment is a conceptual problem that must be resolved, or subjectivity ceded” (p. 39). The conceptual problem, I take it, is that “Kantian autonomous agency cannot allow for any standpoint that grounds its stance in something external to itself” (p. 8), or “without some objective realm to oppose the subject, the subject will vanish—that is, everything is ‘reduced to a single common denominator,’ namely, ‘the subject’ (DE 4/5:29)” (p. 22). Call this the external constraint requirement. Shuster speaks of this constraint as ‘objectivity’, ‘the objective realm’, or (in speaking of Kant’s rational ideas) “objects that constantly anchor our subjectivity through their presence ‘outside’ us” (p. 41). It strikes me that how we understand this constraint is crucial to understanding the nature of the dialectic of enlightenment and its baleful consequences for subjectivity and autonomy. Since Shuster’s is an argument about normativity (not metaphysics), we should understand the constraint in normative terms: an external constraint is ‘objective’ in that it is in some sense independent of subjectivity, by which the subject can justify a belief or action; put more forcefully, for thought to be intelligible it must be about something, to which the subject is normatively accountable or answerable. In this sense objectivity potentially constrains the free self-determination of the subject that characterises one picture of Kantian autonomy.
One way of understanding the constraint is as an independent objective realm for empirical cognitions. And one (orthodox) way of understanding the dialectic of enlightenment is that as the epistemic subject abstracts from particular intuitions, first in particular cognitions of the understanding and then in systematising nature via reason’s search for the unconditioned, the result is that subjectivity is reduced to the epistemic act of classifying and syllogistically reasoning, and ‘the objective realm’ is reduced to greater abstraction. To be sure this is a gross impoverishment of experience, but this picture does not amount to the collapse of subjectivity or objectivity, nor their numerical identity or indiscernibility, for they are still qualitatively different. Consider the case of perceptual knowledge: the same conceptual capacities that are freely and responsibly (that is, autonomously) exercised in judgement are here involuntarily drawn into operation by the receptive faculty of sensibility. The passive quality of sensory consciousness distinguishes the presence of objectivity from that of subjectivity in consciousness. Moreover, the passive quality of sensory consciousness can provide a constraint in a distinctly normative sense. The second analogy can be read as arguing for the discernibility between the subjective realm (a subjectively produced sequence of appearances; Kant’s example is the sequence of perceptions of a house) and the objective realm (a causally necessary objective sequence of appearances, as in the sequence of perceptions of a moving boat) that is grounded just in the difference between what we can and cannot control. Hence autonomy and agency—practical freedom, autonomy as self-determination—is there taken as a presupposition of categorial causal judgement. If the objective constraint is taken in the sense of a normatively independent objective realm, then the passive undergoing of perceptual appearances in receptive sensory consciousness still fulfils that constraint, as when the ordering of such appearances is beyond our free determination. Therefore, if we conceive subjects as receptive with respect to objects then—no matter how else we characterise subjectivity, objectivity and kinds of reasoning—we are entitled to the thought that objects are independently there for subjects, and hence that our standing as epistemic subjects is not entirely undermined. So in this Kantian picture of empirical objectivity there seems to remain an external constraint on cognition and subjectivity that, I suggest, qualifies Shuster’s totality claim that Kantian self-consciousness entails the dissolution of subjectivity and objectivity.
Another way of understanding the externality constraint upon autonomy is as a mind-dependent but nonetheless normatively objective, rational constraint. I think this kind of constraint can be brought to bear in considering the necessity and practical reason claims in Shuster’s argument. Because reason cannot point to a reality outside itself that ultimately justifies it or serves as its foundation (the asserted conclusion of the totality claim), it can only pursue justification by means of some self-reflexive procedure (p. 28). reason undermines itself, on Kant’s view, when it generates, in the form of rational ideas, entities that cannot be encountered in sensible experience. (Shuster admirably discusses the postulate of the highest good in Chapter 4.) This argument decidedly turns on the specific employment and specific objects of rationality in view: in its search for the unconditioned, reason is led beyond the bounds of sense to posit rational ideas (God, immortal soul, highest good) that by rights should only serve a regulative use. Shuster’s necessity claim holds that Horkheimer and Adorno extend Kant’s dialectic of reason (by using it as a general, formal analogy) to “all human reasoning and cognition” so that, as we seek to ground our autonomy self-reflexively “we repeat this dialectical procedure [viz. reason originating ideas that usurp genuine empirical experience and hence become ‘myth’] at every level of human activity” (p. 30).
Two quick observations: first, this unrestricted extension of the dialectic of reason appears, by the lights of Kant’s own system, to be an illicit expansion of scope, and so, secondly, one would like to see some argument for why this extension into and usurpation of empirical knowledge is necessary. The argument by Horkheimer and Adorno that comes most readily to mind here claims that the fundamental motive of cognition and reasoning is fear before and mastery of nature (cf. p. 22), but this apparently anthropological or historical premise seems external to the purely conceptual problem Shuster is discussing. But back to the second kind of externality constraint.
Because rational ideas transcend the bounds of sense we cannot know their objects, but we can think these ideas consistently, and in that sense they meet an externality constraint of objectivity. Shuster sees this as a “servitude” by which “we anchor our subjectivity to an object we take to be distinct from that subjectivity (which we thereby see as objective), and to this extent such object checks the dissolution of subjectivity that the dialectic of enlightenment engenders” (p. 40), so that “ultimately, the dialectic of enlightenment is avoided to the extent that I must carve out a space outside my subjectivity, outside myself” (p. 41). The dilemma appears to be as follows. On the one horn, Kantian autonomy understood as grounded in subjective self-reflexivity, lacks an objective external constraint, thereby undermining the subject-object pair, and hence dissolving subjectivity. On the other horn, ‘mythic’ rational ideas (e.g. postulates of practical reason) fulfil the externality constraint by being ‘outside’ the subject and taken as normatively authoritative but thereby limit one’s autonomous reason by its reverting to such ‘mythic’ ideas. Might there be a conception of the relation between autonomy (self-determination) and rationality that avoids this dilemma?
Autonomy according to Kant’s master metaphor requires that one endorse only those principles and norms that one could treat has having legislated oneself. So reason in its practical employment must become self-reflexive: it cannot rely on other, external authorities (on pain of heteronomy), but must through its employment ground its own rational authority to act. Shuster writes: “Reason then must both subject itself to itself and give itself the law by which it subjects itself” (28). ‘Self-legislation’ can be read in a misleading way, to mean that the subject establishes, constitutively institutes the authority and validity of reasons, principles, and norms. However, if the legislative act is not already subject to the norm of reason, it must appear arbitrary. And nothing instituted by an act that is arbitrary could be intelligible as the authority of reason. So a better reading of Kantian autonomy is that the metaphor of self-legislation means that one recognises a norm that is authoritative anyway. This “anyway” is all the objectivity that we need: the normativity of reasons in this sense is ‘outside’ the subject and provides an external objective constraint on what one might take as one’s reason for believing or acting. This conception of the relation between autonomy and rationality avoids the dilemma, for now autonomy is grounded in the intrinsic (and perhaps mysterious) capacity we have to respond to reasons, to come to inhabit the space of reasons, as the distinctly rational animals that we are.
Chapter 3 unfolds Adorno’s positive concept of autonomy, which centres on an original and fascinating interpretation of the moral addendum (das Hinzutretende) and “addendum actions” (p. 85) more generally. In contrast to a voluntaristic view of freedom of the will as ‘could have done otherwise’ or as mental causation, Shuster understands an ‘addendum action’ as an abrupt impulse that is discontinuous with the agent’s current intention-in-action and hence judged to be ‘irrational’, but which, upon reflection, may be rationalised and thereby acknowledged by the agent as her action after all. Such self-ascription and justification via rational teleological explanation amounts to a kind of self-determination after all, and this more modest form of self-determination is one feature of Adorno’s concept of autonomy. Addendum actions are both somatic and mental, bodily and available to consciousness, and are “drawn out” (p. 87) of the agent by the “affordances” (p. 84) of her normatively saturated environment.
Thus, when he turns from action in general to morality, Shuster situates this theory of action within a sensibility theory (e.g. McDowell, Wiggins, Murdoch, Cavell), according to which the ethically attuned agent immediately (non-inferentially) ‘sees’ what is to be done morally:
Ethical actions too will be ‘pulled’ out of agents; the world may come to be imbued with ethical affordances. (p. 99)
Since impulses are responsive to, drawn out by, moral features of one’s environment that become salient to the agent, freedom—another feature of Adorno’s concept of autonomy—is here understood as the possibility of one’s ‘framework for affordances’ to become ‘expanded and enlarged’ by the mind’s ability to transcend the immediately given, which Adorno calls “the speculative element” (p. 102) and which for Shuster is itself a variety of non-reflective impulse that relies on actualised imaginative capacities that can “jump start possibilities for practical reason and thereby open new regions for action” (p. 105).
The self-ascription requirement for addendum actions entails that such actions “fundamentally express who an agent is” (p. 106). So Adorno’s concept of autonomy is not a procedural (nor even a substantive) autonomy as much as, surprisingly, an autonomy of authenticity, understood as one’s being able to identify oneself fully with one’s actions. The concept of freedom bound up with this sense of autonomy, therefore, is freedom of self-expression, that is, freedom from (self-)alienation, which in turn requires that one’s environment—society—“draws out of one actions that can then be understood as expressing who one is” (p. 106), paradigmatically the expression of suffering and the acknowledgement of another’s suffering.
Shuster rightly acknowledges that one’s framework of affordances is subjective (“there might be cases where suffering registers solely through recourse to my sensibilities, to who I am“; p. 112) and so a threat of “narcissism” (p. 114) looms. Shuster enlists Cavell’s thoughts on expression (language) as a “shared form of life” (p. 119) and his conception of morality as the site of open, ongoing, contestations and conversations between individuals and societally differentiated spheres such as politics and religion. Objectivity amounts to the two observations that morality demands universal agreement and that suffering objectified in expression becomes at least potentially “objectively accessible” (p. 121).
I am very sympathetic to Shuster’s project to bring Adorno’s thought into dialogue with contemporary Anglophone philosophy, and indeed, this chapter could easily have been twice its length to make more explicit the dimensions of that rich and productive conversation. On the other hand, there are antinomian elements of Adorno’s philosophy that lose their sharpness in Shuster’s domestication of his thought.
One brief example is that Adorno holds that the complexities of modern administered society thwart a virtue ethics of immediate moral insight and environmentally solicited moral action. Immediate moral insight is thwarted because “such things as moral philosophy or virtue are only possible in a circumscribed universe, in contrast to the immeasurably expanding universe of today which is incommensurable with our experience” (PMP 98) for which Adorno’s shorthand term ‘context of delusion’ (Verblendungs-zusammenhang) might stand. And moral action is thwarted because were you to undertake such an action you would soon find yourself “caught up in a dialectic without end, one in which the good you are trying to achieve has to be paid for with infinite quantities of the bad and the dubious, with injustice, unkindness and forgetting” (HF 263), for which Adorno’s shorthand term ‘context of guilt’ (Schuldzusammenhang) might stand. As a result, for Adorno, “[t]he more society develops into an overpowering, objectively conflict-ridden totality, the less will any individual moral decision have claims to be judged authentic and right” (HF 262). Others have developed this interpretation of Adorno’s critique of virtue ethics (twinned with his critique of moral rationalism), and Shuster fully acknowledges that the theory of action he reconstructs from Adorno’s writings must be qualified in light of the modern world’s distorting and disruptive effects on the conditions of action, but this casts his positive theory of action, which is presented separately for heuristic reasons (p. 76), in a distinctly counterfactual mode.
I turn to my second example of how Adorno’s antinomian thought is somewhat smoothed over in Shuster’s presentation.
Adorno describes the historical development of the will, as an epicycle of the dialectic of enlightenment, as a “unity of mutually contradictory elements” (HF 215) that are also “mutually interdependent” (HF 238): as a “unity in tension” (HF 249) between archaic, involuntary impulse/addendum (itself developed from simpler mimetic reflexes, HF 235–8) and reason/consciousness. Thus,
what is needed for a willed act or for practice in general is the coincidence of two antagonistic elements that do not become completely fused. On the one hand, there is intellect, reason, about which […] if you take the notion of practice very seriously, it contains or presupposes the idea of the unrestricted, highly progressive theoretical consciousness. On the other hand, there is what I have labeled the additional factor, the bodily impulse that cannot be reduced to reason. (HF 239)
Adorno’s account is read most straightforwardly as a response to Hegel’s critique of Kant’s moral theory, because the addendum, as the “voluntaristic element in a narrower sense” (Adorno) explains how practical reasoning can relate to action: the relation is one of discontinuity, between a reason endorsed by reflection, and “an involuntary adjustment to something extramental” (HF 213). Moreover, on this account agents know what they want to do and what they are doing (they have practical knowledge in Anscombe’s sense) without it being the case that a purely cognitive mental state caused their action; rather their action, while involving consciousness to a lesser degree, to a higher degree involves bodily comportment to the environment and in that sense is involuntary.
This account also illuminates an aspect of the two main examples adduced by Adorno: Hamlet’s abrupt acts of violence and the lawyer von Schlabrendorff’s joining the July 1944 plot against Hitler. Perhaps in Hamlet’s case, but certainly in von Schlabrendorff’s case of joining a conspiracy, the agent’s ‘addendum action’ is itself an intention-in-action that demonstrates the means-ends, processual, and nesting characteristics of regular intentions-in-action. Adorno emphasises the German conspirator’s case, in all its historical and individual contingency (it is, after all, “the coincidence of two antagonistic elements”; HF 239), as “the true primal phenomenon of moral behavior [that] occurs when the element of impulse joins forces with the element of consciousness to bring about a spontaneous act” (HF 240), which suggests that this is an answer to Hegel’s objection to the apparently unbridgeable gulf in Kantian moral theory between pure practical reason and action.
Shuster to some extent resolves Adorno’s potentially antinomian characterisation of will (e.g. it is rare when impulse and reason “join forces”) by as it were disarticulating the two elements—bodily adjustment and reasoning consciousness—into before and after, which yields the counterintuitive position of the agent of an ‘addendum action’ coming to have practical knowledge of her action only post festum. Shuster writes:
Such a jolt, or break, in the chain of reflection appears irrational from the perspective of that chain [the current intention-in-action by which the agent first-personally unifies her actions as exhibiting a proper means-ends rational order that expresses her intention, H.P.]. As a jolt, it seems to come from nowhere. Such a jolt, however, can serve as a reason, and this means that, after the act/jolt in question, it is acknowledged as my act because I can give reasons for it, reasons that are genuinely mine. (p. 82; original emphasis)
On Anscombe’s and Shuster’s view, the kind of intention in question here is not that of a prior mental state, separable from and causally responsible for a subsequent action, as voluntarism holds; rather intention here is a kind of continuant whose instances alter their shape as time passes, i.e. as the intended action unfolds. The intention (typically an answer to the question “Why are you 𝜙-ing?” put to an agent) runs through and unites the various actions that comprise the intended action as it is carried out (e.g. whisking eggs, turning on the stove, making an omelet, giving my wife breakfast in bed, etc.). Practical knowledge (knowing what one is doing) is knowing the intention-in-action that unifies one’s doings. Shuster seems to be saying that the addendum is an action that, when it occurs, does not figure in the contents of one’s practical knowledge, and hence appears to the agent to be ‘irrational’, but that, upon reflection, can be incorporated by the agent into a different rational, teleological ordering of actions that thereby establishes its rationality and authenticity just in virtue of the agent’s attributing that new ordering, that new intention-in-action, to herself.
Such an account, however, perhaps invites a virulent form of consequential moral luck: how the addendum action is ultimately rationally embedded by the agent into larger teleological orderings of means-ends or instance-generality relations, etc. will determine the description under which the action is practically known by the agent herself. Addendum actions thus prove to be constitutively susceptible to context-reversals that are only exacerbated by modernity’s ‘context of guilt’ (Schuldzusammenhang) that can potentially invert the moral valence of virtually any action.
How far does the antinomian current run in Adorno’s moral philosophy in general, and in his conception of autonomy in particular? In his description of the “coincidence” and “unity in tension” between rational reflection and archaic impulse that generates one’s spontaneous acts, he first gestures towards an autonomy of authenticity:
With this impulsiveness, freedom extends into the realm of experience. If we behave spontaneously we are no more simply blind nature than we are suppressed nature. We feel that we are ourselves.
But he then continues:
[I]n yielding to impulse we find that what I have called the Hamlet syndrome has for a moment been overcome. The sense of being divided, of being between inner and outer, is overcome as in a flash. Thus we believe that so long as we obey our impulses we shall find ourselves once again in the realm of objects from which we had withdrawn by an absolute necessity, albeit perhaps only in appearance. (HF 237; emphasis added)
“Yielding” and “obeying”, of course, suggest that an element (Moment) of heteronomy is intrinsic to Adorno’s concept of autonomy.
Invited: 17 August 2016. Received: 8 September 2016.
 Since at least some of the Anglo-American philosophers to whom Shuster refers (Thompson, McDowell, Wiggins) understand their own projects as distinctly neo-Aristotelian, we can also understand Shuster’s project too, like Fabian Freyenhagen’s recent Living Less Wrongly, as pursuing Aristotelian themes in Adorno. Shuster’s nuanced and meticulous interpretation of Adorno’s relationship to Kant and Hegel is itself an extraordinary contribution to scholarship, adding to earlier work including O’Connor (2004) and Bernstein (2001).↩
 Sellars (1997), § 36. In a talk held in New York City, Shuster characterised the stakes of the argument in Chapter 1 as “either the dialectic of enlightenment or the myth of the given”. My comments here aspire to present an alternative to this dilemma.↩
 “Instead of being historical in nature, these claims are conceptual” (p. 19); “Nonetheless, the ‘problem’ of the dialectic of enlightenment is not reducible to worries about these [material] conditions” (p. 34; cf. pp. 36, 39, 105).↩
 “Ultimately, while claims about instrumental rationality are crucial to Dialectic of Enlightenment, they are hardly the focus […] I suggest that for Horkheimer and Adorno instrumental rationality is merely symptomatic of a deeper problem centering on identity” (pp. 16–17); “The omnipresence of identity thinking, the dissolution of subjectivity, the rise of instrumental rationality, the disenchantment of the world, and so forth, are grounded in our very standing as autonomous agents” (p. 32).↩
 Cf. Kant: “The conditions of the possibility of experience in general [i.e. the subjective conditions required for an epistemic subject to have a coherent, unified experience accompanied by a continuous ‘I think’, H.P.] are at the same time conditions of the possibility of objects of experience” (A158/B197). At the NYC conference, Shuster very helpfully provided a detailed reading of the A-edition of the Transcendental Deduction along these lines. Because the categories of understanding must apply to all experience (that is, sensibility and understanding), all experience is conditioned by apperception. Apperception requires the dualism of subject/object. But it follows from the transcendental deduction that “in any important sense, the ‘object’ part of that pair is in fact entirely constituted (albeit not created) by the subject in the form of ‘object in general’” (Shuster, NYC talk). Hence the subject/object dualism is undercut, which dissolves the subject’s normative standing.↩
 Adorno & Horkheimer (2007).↩
 “[…] without some objective realm to oppose the subject, the subject will vanish—that is, everything is ‘reduced to a single common denominator,’ namely, ‘the subject’ (DE 4/5:29)” (p. 22). “Since I am the one taking things a particular way, then the pole of objectivity is already always compromised, contaminated by the pole of subjectivity, and from here, for Horkheimer and Adorno, it is a small dialectical step to the complete dissolution of subjectivity” (p. 32).↩
 Cf.: “If one understands oneself as an autonomous agent […], then no external justification, no justification apart from one’s own reason, can serve to ground that autonomy” (p. 30).↩
 I am making a distinction in the use of ‘active’ between understanding’s spontaneity on the one hand (active in the sense of the synthetic unity of apperception, e.g. A50/B74, etc.) and freely determining what to think (active in the sense of autonomous self-determination) on the other. Kant hints at this when he writes “The same function that gives unity to the different representations in a judgment also gives unity to the mere synthesis of different representations in an intuition” (A79/B104–5), but not all instances of this kind of unity must be viewed as a result of free cognitive acts.↩
 See O’Neill (1989).↩
 At this point in the dialectic one might invoke the distinction between ‘mere appearances’ and ‘things in themselves’, arguing that while the former might provide a highly qualified, anthropocentric form of objectivity constraint, the latter, understood as what ‘mere appearances’ are appearances of, provide no epistemic objective constraint. At Bxxvii Kant insists on an identity of things as they appear in our knowledge and “the very same things as things in themselves”, that is, when abstracted from the constitutive conditions determining how they appear to us. It is a further, distinct move to claim that things in themselves possess other, humanly unknowable properties than they appear to possess in our knowledge of them. A two-aspect interpretation of the distinction thus does not impugn the objectivity (external) constraint provided by the receptive nature of sensory consciousness of appearances. Of course, much more would need to be said here.↩
 Shuster cites Adorno’s similar judgment “Kant’s rescue of the intelligible sphere […] is […] an attempted intervention in the dialectics of enlightenment, at the point where this dialectics terminates in the abolition of reason (ND [Adorno 1973] 385/6:378)” (p. 41).↩
 “In this tradition, when I self-legislate […] I come to regard the law as my own, as if it had been authored by me. I must thereby come to understand my legislation as expressive of myself as a rational agent, that is, as expressive of my innermost commitments as a rational human” (p. 3).↩
 Adorno at times seems to suggest this unfortunate reading, as when in the lecture course on history and freedom he writes: “Thus freedom does not mean that I do not act in accordance with laws, that I am not subject to laws; it means that these laws are to be identical with the laws governing my own rationality. However, since in Kant all the laws that actually exist are the laws of my own reason, it follows that, in the light of the definition of freedom [in Kant’s Groundwork] that I have just read out to you, the theory of freedom is profoundly restricted and even revoked by Kant himself” (HF 248, cf. 254).↩
 For thoughts along these lines, see McDowell (2009).↩
 How one comes to inhabit the space of reasons must be a historical, contingent story on pain of Platonism, which is certainly a mythic conception of reason.↩
 Here too the potential for reflective endorsement is essential, and indicates a disanalogy between value perception and, say, colour perception: unlike colours, values or ethical affordances do not merely draw or pull out, solicit or elicit responses, they merit such responses.↩
 This inflection of an autonomy of authenticity should not be conflated with something like a perfectionistic ethics of monadic bourgeois interiority. Cf. Adorno: “Not only is the self entwined with society; it owes society its existence in the most literal sense. All its content comes from society, or at any rate from its relation to the object. It grows richer the more freely it develops and reflects this relation, while it is limited, impoverished and reduced by the separation and hardening it lays claim to as an origin” (MM §99 [Adorno 1974:154]).↩
 Adorno (2000).↩
 Adorno (2006).↩
 On this argument, see part III of Menke (2005), which Shuster terms “a different reading” (p. 128, fn. 130). But it seems to me to be an antithetical reading to Shuster’s, and so how Shuster-Cavell-Adorno might respond here to Menke-Adorno would be intriguing.↩
 “If we think of the will as a unity in tension between Reason and this other factor, then we might speak of it as the voluntaristic element in a narrower sense” (HP 249, cf. 257, 260).↩
 A toy example. I go to see a baseball game with my glove because I have a standing desire to catch any fly balls that come my way. A fly ball does come my way, and I form the more specific, occurent desire or intention to catch that particular ball, and voluntarily get myself into the optimal posture to do so. The actual catching of the ball, however, is largely my body’s reacting to the incoming ball, and is in that sense involuntary, although I clearly know what I want to do (‘catch the ball’) and what I’m doing (‘trying to catch the ball’). We can (try to) understand Adorno’s thought here as attempting to maximally extend responsive bodily comportment vis-à-vis the environment, including of course the environment qua second nature.↩
 Cf. “Addendum actions, therefore, might appear irrational because they seem to violate a certain currently conscious means-end relation (A-B-C-D), but they need not remain so, because upon reflection they come to be seen as occupying an alternative (whether unconscious, implicit, or unexamined) means-end relation (A-B-C-E). In this way it is not inappropriate to speak of an ‘irrational aspect’ to such actions. That aspect, however, appears or emerges only through a process—through time, and that aspect might also dissipate in time” (p. 91); “Such actions might appear irrational in a process of reflection immediately following the action, because they seem initially unanticipated from the view of that process, but ultimately they are not irrational, since they still reference a space of reasons (which the agent can later come to acknowledge)” (p. 93).↩
 Sometimes Shuster, and Adorno himself, complicates this story by suggesting that the “addendum action” is the expression of an unconscious motive (p. 91). This complicates the fulfilment of the “self-ascription requirement” (p. 79). In explicating Adorno’s reading of Hamlet, Shuster glosses that the impulse action is not without reason: “Instead, certain reasons seem foreign to Hamlet’s current state of mind. Because of his particular context, some reasons might be so foreign as to be consciously inaccessible (but nonetheless they are still his, in that if they were presented, by, say, a psychiatrist during a session, Hamlet would accede to them)” (p. 82). I can leave my session with my psychoanalyst newly persuaded that my inexplicable behaviour was in fact ‘my’ unconscious wish to 𝜙 or unconscious fear of x, but it does not follow that I have thereby rationalised my behaviour in the sense of attributing it to myself first-personally, rather than to my unconscious, to which I maintain a third-personal perspective. These considerations complicate the distinction Shuster makes between reflex actions attributable to organisms and impulse actions attributable to agents (p. 98).↩
 Adorno specifically cautions against hypostatising the impulse/addendum for similar reasons: “If the will were nothing but what I have called ‘the additional factor,’ it if were not more than an impulse as the so-called decisionistic theories teach, then the will would be at the disposition of every conceivable purpose, just as much as instrumental reason is according to the analysis with which are you familiar. As long as the two aspects remain entirely separate from one another, they tend to converge by virtue of the fact that they become available in a quite arbitrary way—and precisely this random availability is incompatible with the idea of the will” (HF 260). We can understand one motive for Kant’s postulates of practical reason to remove this worry of consequential moral luck.↩
 Shuster rightly disavows this interpretation of the self-attribution requirement of intention-in-action, which has recently been defended by Pippin in his interpretation of Hegel’s practical philosophy (our NYC discussion).↩
 Another example of the fundamental antinomian nature of morality for Adorno: “To the question what is to be done with defeated Germany, I could say only two things in reply. Firstly: at no price, on no conditions, would I wish to be an executioner or to supply legitimations for executioners. Secondly: I should not wish, least of all with legal machinery, to stay the hand of anyone who was avenging past misdeeds. This is a thoroughly unsatisfactory, contradictory answer, one that makes a mockery of both principle and practice. But perhaps the fault lies in the question and not only in me” (MM §33 [Adorno 1974:56]).↩
Adorno, Th. W. (1973), Negative Dialectics, trans. E. Ashton (New York: Continuum) (=ND).
——— (1974), Minima Moralia, trans E. Jephcott (London: Verso) (=MM).
——— (2000) Problems of Moral Philosophy, trans. R. Livingstone (Stanford: Stanford University Press) (=PMP).
——— (2006) History and Freedom, trans. R. Livingstone (Malden, MA: Polity) (=HF).
Adorno, Th. W. & M. Horkheimer (2007), Dialectic of Enlightenment, trans E. Jephcott (Stanford: Stanford University Press) (=DE).
Bernstein, J. (2001), Adorno: Disenchantment and Ethics (Cambridge: Cambridge University Press).
McDowell, J. (2009), ‘Self-Determining Subjectivity and External Constraint’, in J. McDowell, Having the World in View: Essays on Kant, Hegel, and Sellars (Cambridge, MA: Harvard University Press), pp. 90–107.
Menke, C. (2005), ‘Virtue and Reflection: The “Antinomies of Moral Philosophy”’, Constellations 12(1): 36–49.
O’Connor, B. (2004), Adorno’s Negative Dialectic: Philosophy and the Possibility of Critical Rationality (Cambridge, MA: MIT).
O’Neill, O. (1989), ‘Reason and Autonomy in Grundlegung III’, in O. O’Neill, Constructions of Reason: Explorations of Kant’s Practical Philosophy (Cambridge: Cambridge University Press), pp. 59–63.
Sellars, W. (1997), Empiricism and the Philosophy of Mind (Cambridge, MA: Harvard University Press).
© Henry W. Pickford, 2017.
Henry Pickford is Associate Professor of German and Philosophy at Duke University, USA. He works in philosophy and literature, primarily within the German tradition. Areas of interest include German Idealism, Marx, Wittgenstein, and Frankfurt School Critical Theory. He has edited and translated two collections of essays by Adorno published together as Critical Models (Columbia UP, 1998, 2005). Pickford’s latest book is Thinking with Tolstoy and Wittgenstein: Expression, Emotion, and Art (Northwestern UP 2016). Together with Gordon Finlayson he is currently working on a monograph entitled Adorno: A Critical Life.