MARTIN SHUSTER | Autonomy after Auschwitz: Adorno, German Idealism, and Modernity | Chicago: University of Chicago Press, 2014


By Martin Shuster

Let me say thank you to the three respondents in this wonderful forum. I have learned a lot from all of their responses and they have each given me much to think about, suggesting potential future avenues of inquiry. More specifically, thank you to Henry Pickford for charitable, interesting, and frankly flattering comments about my book. Pickford and I have been in conversation about these issues for quite some time now, and I appreciate the opportunity to engage with such a sharp interlocutor. Robert Hanna’s response is also appreciated, especially for its intellectual generosity. Embedded in it is a sophisticated reading of Kant, which I unfortunately do not here have the space to address in the detail it would require. Finally, I am equally grateful to Charlotte Baumann’s rich, expansive, and highly probing comments. Especially, I am struck by the deep, and to my mind important, methodological issues that emerge from her commentary. In fact, when responding to her, I shall start with them in order to begin to address some of her more specific points. Let me address each critique in turn.

Response to Henry Pickford

Pickford has involved himself quite deeply with elements of my reading of Kant, which I find fruitful, as Kant is certainly a central figure both in the book and, I think, for Adorno more generally. Because of that, I want to start with a few points about Pickford’s presentation of Kant in order to try to respond to the worries he raises about my reading of the dialectic of enlightenment. I think a lot of the issues involved here revolve around the role of time in Kant’s argument, and I shall try to bring that out in my response.

Pickford’s presentation of the ‘totality claim’ strikes me as exactly right, but I find that his presentation of the ‘necessity claim’ seems to miss something that I think is important. Note that Pickford locates the claim of or for necessity roughly in the conditional that if I take myself to be an autonomous agent, then I cannot look to anything outside of my reason to justify myself. This is generally right, but it also seems to me to minimise or understate the way in which this issue falls out of and depends on the function (indeed, functioning) of apperception. To see how this is the case will take some time.

Pickford notes that everything hinges on how we conceive of how objectivity constrains the operations of autonomous human reason. He considers two options of how objectivity is in play: one is what he terms an “independent objective realm” and another is what he terms as “mind-dependent but nonetheless normatively objective”. In fact, though, I think for Kant these distinctions get blurry. Let me say how. For the first—the independent objective realm—Pickford leans heavily on the idea of a passivity to sensibility, and relies on the Second Analogy of the Critique of Pure Reason to make his point. As Pickford notes:

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.

I do not think, however, that the Second Analogy gets us what Pickford claims. First, let me note that the problem of the Second Analogy is whether we can justify causal claims. Kant stresses that causality

must either be grounded completely a priori in the understanding, or must be entirely given up as a mere phantom of the brain. For this concept makes strict demand that something, A, should be such that something else, B, follows from it necessarily. (B123–4; trans. Kemp Smith)

The problem is that “experience teaches us that a thing is so and so, but not that it cannot be otherwise” (B3). In other words, as Hume (and earlier sceptics) pointed out, we never ‘perceive’ or ‘experience’ causality. It does not appear in experience in any sort of way. To situate this point in the First Critique, remember that up to this point, by means of the Transcendental Deduction, Kant has shown that to experience an object in time, I must make a judgement about it. Take two Kant-inspired examples: (1) a ship is moving downstream (B237), or (2) a house has a rooftop pool and a first floor bedroom (B235). Note that in order to make such judgements, I must at least potentially be able to distinguish between (a) cases where successive perceptions are all of one object that is not changing position (the house), and (b) cases where successive perceptions are all of one object that is changing position (the ship). Perceptions, however, are not time-stamped, and thus the sequence of perceptions cannot itself provide grounds for any such distinction.

Importantly, though, our uptake of a sensible manifold is always successive. We apprehend the house by perceiving its parts (first left, then right, or up, then down, or centre, out, or whatever). We apprehend the ship by doing the same and by apprehending its first position, then its second, then its third, as it floats down the river. Nonetheless, the manifold of the house does not require us to insist, perceptually, for change over time in a way that the manifold of the ship does. Why? Because we are aware in the case of the house that we can mix up our succession of perceptions, uptake them in any order. We cannot do the same for the ship, or else we would not see a ship moving downstream. It would be a different event. The question is: what in our representation makes a successive perception the perception of a succession? Kant’s suggestion is that we must somehow be able to think of a certain event as irreversible. This is the only way to distinguish subjective time order from objective time order.

What are the conditions under which it is the case that I must think B as occurring after A? Before answering this question, note that importantly Hume cannot even account for such a question: the question of what it means to be an event. In fact, as Kant notes, such irreversibility can be conceptualised only by means of us having applied an a priori rule. What else could such a rule be but causality? Here’s Kant:

If one were to suppose that nothing preceded an occurrence that it must follow in accordance with a rule, then all sequence of perception would be determined solely in apprehension, i.e., merely subjectively, but it would not thereby be objectively determined which of the perceptions must really be the preceding one and which the succeeding one. In this way we would have only a play of representations that would not be related to any object at all. (A194/B239)

There must ultimately be some way to distinguish between the successive apprehension of a manifold (which is subjective, accidental, fraught with uncertainty, and so forth), and the irreversibility of particular perceptions themselves (which is objective). In order for the latter to happen, I must regard these particular perceptions as irreversible. To do so just is to subsume these perceptions under a rule (the schema of causality). Pickford suggests that such rule-application is “beyond our free determination”, but this seems to me to be insufficiently precise, if not entirely mistaken. In what sense is this “beyond our free determination”? It is I who still, spontaneously and autonomously, apply the schema of causality;[1] and thus it seems that Adorno’s worries are justified (at least as I have conceived them in the first chapter of Autonomy after Auschwitz). But even putting aside this issue—say, that we grant Pickford’s claim—that claim still does not get us an objective realm. It only gets us the necessary distinction between the apprehension of a manifold and the successive order of how that manifold is arranged, that is, it merely gets us the schema of causality (read: the grasping of a distinction between objective and subjective time order).

I want, however, to pursue Pickford’s suggestion about establishing an objective realm with respect to the First Analogy, since I actually think his proposal might have more success there. There, an element of Kant’s problem is the question of how do we come to know time? As he notes:

All appearances are in time, in which, as substratum (as persistent form of inner intuition), both simultaneity as well as succession can alone be represented. (B224)

To have simultaneity and succession, we need time, that is, time is the sort of ‘backdrop’ necessary for simultaneity and succession (and this is important because in the Transcendental Deduction Kant had shown that time is unitary—the transcendental unity of apperception mandates that I project that I am the one having my experience, even into the future). Importantly, though, there is no such thing as ‘time’ perceived in experience (“Now time cannot be perceived by itself”; B225). Unlike modern digital cameras, the parts of a manifold, indeed any part of our experience, does not have a time stamp. So, how does one even come to have experience of a temporal order? There must be some perceptible substratum that allows for the possibility of thinking an event as contained in a unified temporal framework (“that which persists is the object itself, i.e., the substance (phænomenon)”; B227). In the First Analogy Kant shows that the schematised concept of substance is a necessity for the possibility of the experience of things in time. And this seems to get Pickford exactly what he desired, but note that, at this point in the First Critique, Kant also undertakes a quite problematic sleight of hand. He writes:

Therefore in all appearances that which persists in the object itself, i.e., the substance (phænomenon), but everything that changes or that can change belongs only to the way in which this substance or substances exists, thus to their determinations. (B227, my underlining)

Aside from this incredible switch, we also get a more extreme claim, from relative to absolute permanence:

For if that in the appearance which one would call substance is to be the proper substratum of all time-determination, then all existence in the past as well as in future time must be able to be determined in it and it alone. Hence we can grant an appearance the name of substance only if we presuppose its existence at all time, which is not even perfectly expressed through the word “persistence” since this pertains more to future time. (B228)

Kant’s reason for this is apparent as when he notes that

Substances (in appearance) are the substrata of all time-determinations. The arising of some of them and the perishing of others would itself remove the sole condition of the empirical unity of time, and the appearances would then be related to two different times, in which existence flowed side by side, which is absurd. (B232)

Nonetheless, I take this to be an incredibly radical switch. In Aristotelian terms, it might be said that Kant goes from talking about form to talking about matter. Kant is simply not justified in making the claim that he does above. In fact, this claim, as far as I can see, and as Adorno himself notes, can stand only if it is made weaker: at any moment that I have experience, the a priori objective validity of the schematised categories of substance and causality must be affirmed; in this way, we have grounds for asserting such a priori objective validity along the possibilities of temporality and irreversibility, and we thereby grasp the distinction between an objective and subjective temporal order, all of which makes possible the experience of events.

But this is really only a further elaboration of the main conclusion of the Transcendental Deduction: that if I am to be self-conscious, apperceptive, I must then posit an object in general that allows me to unify a manifold from moment to moment thereby maintaining my apperceptive consciousness over time. Thus Adorno stresses, apperception is

not just something in me, but is always and at the same time present in the experiences concerned, because the experiences, the appearances, are in truth always only mine, they are mediated through me. (Adorno 2011:140)

This, however, is exactly the issue that Horkheimer and Adorno aim to diagnose with the dialectic of enlightenment. So when they claim “there is no being in the world that knowledge cannot penetrate, but what can be penetrated by knowledge is not being” (Adorno & Horkheimer 2002:19), they are commenting on the alleged insufficiency of this standpoint: that seeing all of experience as conditioned by apperception is actually to dissolve the subject’s standing. Such a standing, apperception, requires that there be a subject/object pair. In an important sense, though, the ‘object’ part of that pair is in fact entirely constituted (albeit not created) by the subject in the form of ‘object in general’. And, in this sense, in Horkheimer and Adorno’s words,

[t]he 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 and world is finally resolved, but only in the sense that the two sides cancel out. (Adorno & Horkheimer 2002:20; trans. modified)

This is why it is appropriate to extend the form of Kant’s dialectic to every experience, because the temporal quality of apperceptive experience is such that I must project it also into every future experience, if it is to be my experience.[2]

Finally, let me very briefly respond to Pickford’s reading of the third chapter of Autonomy after Auschwitz. There is a lot to say here—more than I can hope to cover in a response—and so my comments will be brief, but not because Pickford has not raised substantive issues. Most importantly, let me note that I’m wary of Pickford’s talk of authenticity because in this context it suggests that there is something from moment to moment to which one can be authentic; I do not think there is, and I do not think Adorno thinks that there is. I do, however, think that we can, if Pickford would like, talk about authenticity in the context of any particular moment, if that is understood as the sort of reflective equilibrium that is suggested by any particular situation and any particular person in that situation. I agree, therefore, with Pickford’s first example: my point is that such a ‘context of guilt’ has everything to do with a context of guilt, not with any formal qualities about how to understand agency. The broader problem is a sort of teleological one, where capitalism warps the content of human action. In this sense, then, absolutely, consequentialist moral luck is a problem, but it is not one that is traced to the form of agency I am outlining, but rather to the content that gets plugged into it. It just is a fact of life in late capitalism that we often do not know how our actions will turn out. This does not mean that things are always determined retroactively, and subject to such, but rather that when plugged into a broader late capitalist order they turn out to be one way rather than another. This does not have to do with the form of the action, but rather with the content of how the various intentions-in-action hang together.[3] Furthermore, it is important to note that Adorno does not subscribe solely to consequentialist moral luck exclusively, but also attaches great importance to constitutive moral luck (who gains insight is often left up to luck). I take this to suggest that a theory of agency can only take us so far: the sort of modalities and procedures that we need to formulate for critical consciousness will have to be worked through and achieved time and time again, all while living under the rapidly changing conditions of late capitalism. Following Adorno, though, I think essential to any such task is rejecting any Kantian voluntarist inspired accounts of agency.

Response to Robert Hanna

I think that Hanna and I agree on a lot when it comes to the relationship between Adorno and the Frankfurt School and Kant. Notably, he is surely right to note that Adorno is a sort of ‘philosophical child’ of Kant—this is true autobiographically as much as intellectually. With all of that said, though, I do have a few hesitations. Hanna pushes my reading towards what he terms the OHP (Oxford-Harvard-Pittsburgh) school of Kant scholarship (a moniker I had not encountered before, and one I have to note I find peculiar, since the three places and the figures he puts here—say, Sellars, Strawson, and Cavell—strike me as distinct enough in their approaches to Kant to make me wary of the label). He notes that,

according to the OHP, we must read Kant in a conceptualist, intellectualist, anti-metaphysical, mainstream-analytic-philosophy-oriented, Hegel-oriented, indeed neo-Hegelian, communitarian, and liberal democratic way.

Apart from judging the veracity of this statement (for example, elements of this strike me as true of McDowell, but not of Sellars or Taylor or Putnam), I can note that the entire thrust of my second chapter—a point Hanna recognises in his response—was to propose and to highlight the metaphysical impulses animating Kant’s deepest intentions. Now, perhaps Hanna means something quite weak by “heavily oriented towards”, but then I think the suggestion raises a deeper question about the utility of painting in such broad strokes and also of framing our approach in a binary that pits OHP vs. Left Kant.[4] Indeed, as should be obvious, even though Hanna and I, broadly speaking, present metaphysical readings of Kant, we do so in quite distinct ways.

More explicitly, though, I have to admit to being quite suspicious of Hanna’s suggestion that Kant is either “highly progressive” or a “radically enlightened anarchist”. Hanna’s suggestion in a very interesting piece that he cites is that Kant’s stance is in fact

a highly original, politically radical, and if not revolutionary, then at least robustly State-resistant, State-subversive, and even outright civilly-disobedient cosmopolitan, existentialist version of anarchism that I call existential Kantian cosmopolitan anarchism, [which] very naturally flows from Kant’s moral philosophy, his philosophy of religion, and his political anthropology. (Hanna 2016:61)

Although I cannot develop this in full, let me note that Kant also explicitly takes his political philosophy to derive from and be linked to his moral philosophy. I do not mean to suggest that Hanna’s procedure of minimising this fact in his interpretation is problematic, only that all of this requires significantly more discussion.

Indeed, at a very high level of generality, all of Kant’s seemingly disparate inquiries are in fact unified by one broad topic, namely autonomy, and the issues that fall out from it.[5] I have wanted to mention this both to flag where the discussion will need to go, but also to voice a hesitation—or at least an uncertainty—about Hanna’s invocation, seemingly on Adorno’s behalf, of a notion of “relational autonomy”, which Hanna unpacks as the idea of

the coordinated practical agency of each of the members of a group of persons, according to shared principles or reasons of their own choosing, aka multiply self-legislated principles or reasons, all of them, ultimately, falling under the Categorical Imperative.

First, it seems to me that such a notion is still fundamentally voluntarist, which I think Adorno finds objectionable, at the very least on metaphysical, not to mention ethical, social, and political, grounds—my sense is that Hanna does not see a problem with the metaphysics here, and believes that Kant can respond to the ethical, social, and political issues that Adorno might raise; these are all issues I take to explicitly address in the third chapter of the book.

Second, it seems to me that this presupposes a particular, and I believe untenable, view of history, since I take it that one of Adorno’s critiques of the very notion of a categorical imperative is its formalism (something I also discuss in the book); now, my sense is that Hanna does not find the formalism critique convincing, but one of my suggestions in Autonomy after Auschwitz was that it becomes significantly more convincing when one takes seriously the broader story that Kant tells about autonomy, about the highest good, and about how they connect with his religious and metaphysical commitments in the Third Critique, especially around the notion of humanity as the final end of creation. These are all large points that require a lot of contextualisation (pursued in the book and so not rehearsed again here).

Noting these points is a way to say that I feel like I do not have much more to add right now except to note these broad issues, issues which would have to be worked out in significant (and lengthy detail), but which I feel like I would be unable to do so in the constraints imposed by the present format.

Response to Charlotte Baumann

Baumann sets up her discussion as one that attempts to delineate between two allegedly and relatively clear and stable ‘contexts’ for philosophy, one that she calls a “pragmatic or analytic, Wittgenstein-inspired” and another that she calls “continental” which she takes largely to mean “German and historically trained”. In doing so, she seemingly lumps into the former camp figures as diverse as Robert Pippin, Wilfrid Sellars, and Peter Strawson, and into the latter figures as diverse as Axel Honneth, Paul Natorp, and Jürgen Habermas. Importantly and interestingly, Baumann notes that she will “withhold judgement about” my alleged “analytic line of interpretation of Kant, Hegel, and Adorno”, that is to say, she will “try to refrain from claiming that [I] ought to have interpreted Adorno as [she] interpret[s] him”.

I appreciate this overture and embrace it as an opportunity to think seriously about what it is we are doing when we do philosophy, what philosophical traditions amount to, and indeed what philosophical ideas ultimately amount to and how. This is important because I have to admit to finding Baumann’s dualism here difficult to make plausible. On the one hand, of course, I see what she is trying to get at: there is a way in which Autonomy after Auschwitz engages contemporary Anglophone debates as much as it engages European or ‘continental’ ones and thus pulls Adorno into contexts that may not immediately seem relevant, and indeed, might have appeared foreign even to himself. On the other hand, this just is what philosophy is and why it is not, say, intellectual history (and all of this without any prioritisation of the value of either of these disciplines; they just do different things). One can see exactly this tension appear when Baumann writes, discussing Hegel and towards the end of her review, that

it is clear, however, that just because there is a problem with Hegel’s argument does not by itself prove that he must have made a different argument; in fact, Adorno and many others were well aware of this and other problems with Hegel.[6]

But it seems to me that this latter point (that others are aware of particular problems with an argument) does not conversely by itself prove that Hegel (or anyone) must have made this argument, i.e. the one that folks take to be problematic. In other words, there is no getting outside of ourselves to ‘check’ to see who is right. We have to make and assess our claims accordingly; furthermore, when we are doing philosophy, as opposed to intellectual history, it seems to me that we also have to employ—at least in some way—the principle of charity, which requires us not just to take the philosopher in question to be arguing rationally, but also to force ourselves to formulate the best, most convincing, and strongest interpretative version of the argument in question. Kant is an inspiration for me here when he stresses that

I need only remark that it is by no means unusual, upon comparing the thoughts which an author has expressed in regard to his subject, whether in ordinary conversation or in writing, to find that we understand him better than he has understood himself. As he has not sufficiently determined his concept, he has sometimes spoken, or even thought, in opposition to his own intention. (B370; trans. Kemp Smith) 

Of course, there are limits here: conceptual, philosophical, indeed, also ethical, but these limits—at least it seems to me—do not come premade, nor, importantly, do they map onto any version of the alleged distinction between ‘analytic’ or ‘Anglophone’ philosophy and ‘continental’ philosophy. (To see what I’m getting at here, we might consider Deleuze’s readings of figures in the tradition, like Spinoza or Hume, or Heidegger’s, of, say, Nietzsche or Leibniz or Aristotle—Deleuze and Heidegger are surely ‘continental’ figures par excellence, although one of them is admittedly French and not German.)

In this way, the distinction that undergirds Baumann’s review is peculiar—it trades on two issues that I want to separate. The first is the largely sociological point about how methodologies are taken up and understood by philosophers, how resources are distributed amidst departments of philosophy and philosophers, how decisions are made about who to read and why, and how we justify these decisions to ourselves (i.e. the so-called ‘analytic’ and ‘continental’ divide). The second is the largely methodological point about how to understand our relationship to philosophers of the past: here, at an eagle’s eye view of things, the two methodological poles are between what I would term philosophy and intellectual history.

Obviously, I do not have the space to pursue a forceful delineation of these two, but my sense is that we can draw a rough distinction between making ideas work, particularly in light of additional interventions, arguments, and innovations (philosophy), and describing that process (intellectual history). That is to say, crudely, it might be said that the former will have ‘built into it’ a higher level of evaluation of the ideas in question. Now, the best of each of these, at the very least, deeply informs the other (the former, say, in making salient what the latter picks out; and the latter in informing the work of the former). Emerging with any such conceptualisation is a wide array of issues, not the least of which is that there certainly are deeply problematic versions of philosophy that take it as given that they can understand the work of philosophy as not necessitating the history of philosophy, which I think is absurd (one thinks here of Gilbert Harman). Of course, this is impossible, and doing history of philosophy just is essential to doing good philosophy, but it is not by itself sufficient: understanding that a particular conception of science informs Descartes’s notion of philosophical method or that a particular conception of sensibility informs Plato’s epistemology, does not, by itself, make the arguments work. There are still better and worse ways of presenting them, of arguing them, and sometimes those ways emerge only through the passage of time; furthermore, there is also an important distinction to be drawn between what a particular philosophical figure asserted and what they were entitled or justified in asserting based on other claims essential to their project (i.e. an immanent critique; see the Kant quote above).

Now, I want to be clear that the above is meant merely as a prelude for addressing the serious and core question that Baumann raises in response to my account, the alleged problematic substitution of talking about ‘reasons’ instead of ‘reason’ (or ‘Reason’). Broadly speaking, Baumann’s claim is that “something forceful and problematic about the dictatorship of reason” is absent from my account, with its talk of a Sellarsian “space of reasons” and with its specific reading of Kant and the Kantian tradition (taken here broadly to include both Hegel, and importantly, Adorno).

The substantive issue on the Kant-Adorno side is my reading of the transcendental unity of apperception. We are in agreement that this concept is “a good starting point for approaching Adorno’s critique of subjectivity” (and I should note that I am thrilled to have Baumann agree with me on this, as it has not been clear or obvious to a range of Adorno scholars). Baumann disagrees with my interpretation of this notion, seemingly both in Kant and also in Adorno. Her reasons seem to be the following. First, that I incorrectly locate the transcendental unity of apperception as a distinct item within the Kantian framework, as opposed to seeing it as the “transcendental subject”. Secondly, that I interpret the notion in a way that allows me to easily slip from talking about ‘reason’ to talking about ‘reasons’, that is, that I perform “the easy transition from the singular to the plural” that she has found problematic.

Baumann summarises my presentation of Adorno’s reading of Kant with the idea that transcendental unity of apperception “entails a problem insofar as objects are always interpreted by me”. This is not exactly what I say, though. Throughout, I argue that Adorno’s problem with Kant’s transcendental philosophy is that I constitute the world—every single possible piece of it—exactly because I am constituting such (transcendental) unity (of apperception). It is not a case of ‘merely’ interpreting things. This is an important point because it undercuts an unwarranted rhetorical step that Baumann appears to invoke throughout. There is a serious and substantive issue that she has diagnosed about the slippage from ‘reason’ to ‘reasons’, but it is not the case that every slippage from ‘reason’ to ‘reasons’ is a problem. To see why this is the case, let me first note that the problem animating the ‘slippage’ issue is the fact that there is a way to take statements about ‘points of view’ or ‘reasons’ as negating the actuality or activity (or, better: the fact) of reason, i.e. something unitary and single that accomplishes such unity. In other words, part of the seeming bite of Baumann’s critique is that when we stop talking about ‘Reason’ or ‘reason’ in the singular and start talking about ‘reasons’ in the plural, we are in fact losing sight of the fact that there is just one activity going on here: reasoning, and that is the more philosophically significant point (and it is this fact also that will allow Adorno to see reason as a sort of authoritarian trap). But this is not what I nor Sellars nor Pippin nor many others mean when they talk about ‘reasons’. Nor Kant. Here’s a passage from the Groundwork. Kant notes:

Now, a human being really finds in himself a capacity by which he distinguishes himself from all other things, even from himself insofar as
he is affected by objects, and that is reason. This, as pure self-activity, is raised even above the understanding by this: that though the latter is also self-activity and does not, like sense, contain merely representations that arise when we are affected by things (and are thus passive), yet it can produce from its activity no other concepts than those which serve merely
to bring sensible representations under rules and thereby to unite them in one consciousness. (GMS, AA 4:452)

In this way, it is this self-activity or spontaneity that defines reason above all, and that spontaneity is paradoxically unitary and one of a kind, exactly the same in all cases and distinctly my own. And that is because it is an activity; it is literally not something that someone else could ever do for me. None of the figures I cite—and neither me nor Adorno—denies that such spontaneity is singular; but it is exactly such spontaneity that produces reasons, plurality. This is not a problematic sort of slippage; it just is a fact (of reason). In other words, it is the human capacity for reasoning (singular) that produces all the wide array of reasons (plural), the former a unified formal property that admits of an essentially infinite amount of content.

I mention this exactly to undercut the (neo-Kantian) interpretation of Kant that Baumann attributes to Adorno. I agree with Baumann that a proper accounting of Adorno’s relationship to neo-Kantianism is a desideratum for future research, but I disagree that understanding Adorno as having been influenced by neo-Kantianism leads us necessarily to adopting this interpretation (or, perhaps, more ecumenically solely this interpretation), of the transcendental unity of apperception merely as “nothing but the (unified system of all) Kantian categories” (as Cohen and other neo-Kantians do). And I mention this exactly because of the spontaneity issue raised above: the unified system of Kantian categories is the understanding, not reason in Kant, and it is essential to Kant’s deepest commitments as a philosopher that the Critique of Pure Reason accounts for the dialectical nature of reason, and that is only explicable if one reads the transcendental unity of apperception as importantly tied to spontaneity. What I mean by all of this is that apperception is not simply the categories, as if they could just be applied mechanically, but rather apperception is the whole totality of the faculties, and thereby something that can never act mechanically, but only spontaneously; I have to be doing something, performing an activity. Now, I shall be the first to admit that this raises a host of issues, in how to understand Kant’s project, the transcendental unity of apperception, and even how to tell properly the history of neo-Kantianism, but those issues do not change the fact that in Adorno there just are multiple points of entry to this problem.

Indeed, Adorno himself suggests my reading as much as the one that Baumann proposes when he notes that

the unity of consciousness is not just something in me, but is always and at the same time present in the experiences concerned, because the experiences, the appearances, are in truth always only mine, that is, they are mediated through me. (Adorno 2001:140)

And there just is—as Adorno also notes—an ambiguity here. But that ambiguity comes to the fore exactly because of the spontaneity issue above: Adorno later moves simply to calling this an aporia, stressing—contra Baumann and the neo-Kantians—that if the transcendental unity of apperception “is really no more than a merely logical unity, we could not imagine how spontaneity or activity could be ascribed to it” (Adorno 2001:213). I do agree with Baumann that if one prioritises my alternative line of interpretation in Adorno—as I do here and in my book—then the strand of Adorno’s thought that she reproduces about ‘market laws’ and the ‘transcendental subject’ loses its force, indeed becomes not particularly convincing if also no longer cogent. I do not have a problem with that as I simply have never found that argument convincing or useful: it is, to my mind, at best a rhetorical move, not a substantive one. I mention this to hearken back to the methodological issues I raised in the beginning: Baumann may have compelling reasons for why this argument ought to be taken seriously (although they are not reproduced in her review), but I have trouble seeing them. Furthermore, I am convinced that there is enough interesting stuff—also there in Adorno—that can be used to present a very different strand of argumentation, allowing us to jettison or ignore this weaker one.[7]

This point takes us to another substantive point that falls out from the above, what, inspired by Baumann, we might term the ‘repression’ thesis: that by “distancing oneself from one’s immediate needs and impulses, the self (in the sense of a coherent, rational, unchanging person) emerges”. As I mentioned in my response to Henry Pickford above, after reading Owen Hulatt’s excellent Adorno’s Theory of Philosophical and Aesthetic Truth,[8] I do agree that this is an important strand to introduce into the account and one that I unfortunately minimised in the book. Nonetheless, I do not think this issue is as simple as it appears. After introducing the issue, Baumann notes:

When Adorno speaks of the emergence of subjectivity, he is basically referring to the emergence of thinking beings, which largely coincides with self-conscious, rational beings, beings capable of reasoning. For Adorno, Kant is making a point about how thinking necessarily functions; how thinking beings operate. (And Kant’s point is not that thinking beings give reasons, but that they necessarily think with the help of the categories thereby positing or presupposing a particular law-like and necessary form of objects that leads human beings not only to overlook the natural particularities and differences of physical things, but also necessarily to misconstrue their own self.) (boldface added)

Yet, there is no reason to so cleanly or easily separate ‘giving reasons’, on the one hand, and ‘thinking’, on the other. (In fact, there are a host of discussions that emerge here, with figures as diverse as Arendt and Brandom having an important role to play—I mention the former especially to highlight that thinking, and thereby, giving reasons might be a broader act than the employment of the Kantian categories.) Furthermore, there is again here too, an important ambiguity or circularity about how any such distancing might be accomplished without the prior existence of self-consciousness. In other words, Baumann’s reconstruction of Adorno’s account seems to suffer from a regress issue: self-consciousness is not a fact, as if one just suddenly became self-conscious, but rather more like an act where one accomplishes something in doing so (as, of course, Fichte stresses above all, but as is also found easily in more and less explicit ways in Kant and the entire German Idealist tradition). To think that the normative standpoint of self-consciousness emerges out of the natural fact of being human strikes me as incomplete. Of course, the repression issue is important here, but that only pushes the questions back a step: how can I go about repressing certain content in order to create self-consciousness, when, in a deep sense, I only exist when I become self-conscious. All of this, of course, just takes us back to the spontaneity issue above.[9]

This issue, in turn, helps to see the fault-lines that emerge in our different proposals for how to read Adorno on Hamlet and on the addendum. Baumann suggests that my reading is off the mark because I do not see the archaic element in Hamlet actions as referring to a phase more akin to the magical pre-conscious stage that is described in the Dialectic of Enlightenment. She goes on to recount that such a reading of Adorno on Hamlet would take up a common Hegelian thesis about the ancient Greeks: that they lacked a certain sense of interiority and thereby are unable to perform certain procedures that moderns can. So Baumann notes:

There is a stage in which individuals are not yet subjects, self-conscious, rational, independent thinkers, who can distinguish their own opinions, thoughts and judgements from what society demands and prescribes. As such a non-subject, they do not know or formulate abstract and universal concepts and principles (which would enable them to measure society against universal standards).

I think that Baumann does an excellent job of presenting this alternative reading of Adorno on this point; it is a common one, but I also felt compelled to avoid it exactly because I have always found this sort of Hegelian line to be quite unconvincing. In fact, it was part of my desire to elaborate Adorno’s philosophy of language in Autonomy after Auschwitz—especially in such close proximity to the philosophy of Stanley Cavell—exactly in order to highlight the implausibility of this view of (the evolution of) human agency, and thereby of human language and art. Instead, with Cavell, I think that human language (and thereby art) has always possessed the same broadly formal and existential properties (and thereby held the same relationship to scepticism, to universality, to particularity, and so forth), but that particular historical configurations exacerbate these existential issues, bring them to a head, and push them to extremes (as Cavell puts it, “modernism only makes explicit and bare what has always been true”). I think it is possible to read the historical theses that German philosophy often presents in this tenor, including notably in Adorno. I mention this not because I can elaborate this claim, especially with Hegel and Marx on this point, in the space here, but only to flesh out the contours of how such a discussion will go (a discussion, again, that I have implicitly pursued in the third chapter of Autonomy after Auschwitz). 

There just are other ways to read Adorno here, ways that still manage to make good sense of his writings (and again, incidentally, the same is true of Hegel, if we acknowledge his innermost assumptions, opting to strike certain things he actually seems to have said in favour of a reconstruction that stresses what he ought to have said or what was entitled to say, again taking us back to the methodological points that opened this response). And for this reason, I do not find compelling Baumann’s suggestion that I do “not see” these claims and that I allegedly exclude “from the outset the possibility that Adorno is arguing in favour of irrationality, or an irrational moment”. Quite the contrary: I wholeheartedly agree that Adorno is arguing for an irrational moment (as I stress throughout), only noting that any such moment—to be mine—must be capable of rationalisation (a standard McDowellian point that I develop chiefly through reference to his work).

For a quick suggestion of how this works, one need only glance at how easily Baumann is able to explain this allegedly “irrational” moment (albeit with reference to a different locus of traditions). The chain certainly appears irrational to the agent, but it cannot stay that way or else it would literally be nothing for the agent (thus my extended discussion of the distinction between addendum actions and reflex ones, a distinction that Adorno himself acknowledges, as when he notably stresses that “we feel that we are ourselves” [Adorno 2006:237]).

When Baumann concludes her critique of my reading by suggesting that the aforementioned McDowellian point (but equally also a point in Anscombe, and in Adorno), namely, the point that all the necessity of an action for being my action requires that it be capable of rationalization, now seems to offer instrumental rationality as “part of the solution” I offer, I have to also say: quite the contrary. The point here is not about instrumental rationality, but rationality altogether. So, the idea is not, with such actions, “there is a different, coherent chain of reasoning, a coherent rational self to which the acts can be attributed”, but rather that with such actions there is the possibility of rationalisation, so that they can become part of a rational self, one that must rationalise something that initially—and potentially perpetually if the agent is unable or unwilling to do so—appears wholly irrational.[10]

This discussion allows me to turn to the most significant point that Baumann raises against my account, that it somehow avoids addressing “which particular chain of reasons it would be best to choose” and “[w]hat is truly and absolutely right and rational”. Apart from leaning on a problematic voluntaristic notion of agency that I take pains to attack in Autonomy after Auschwitz, I agree Baumann raises the core issue. But it is an issue that I explicitly had in mind when I wrote the third chapter and aimed, following Adorno, to locate the answer to this question in the somatic response to suffering. There may be additional issues with that account, but to imply that my account is unable to answer that strikes me as odd: it is at the heart of my development of Adorno’s categorical imperative (an imperative, incidentally, that argues exactly against the idea that there are things that are absolutely right and rational, if ‘absolute’ is meant to imply any sort of ‘timelessness’ and context-independence—as Adorno notes, Hitler forces such an imperative on us: there is a distinct historical origin to it). Furthermore, the imperative forces us to rethink—as I argue and attempt to do—our very notion of reason (and of giving reasons), introducing to the entire procedure a somatic dimension that was overlooked by Kant and that, frankly, has far greater prospects for universality, though it is a universality that must be achieved and does not come premade.

Finally, let me conclude with a few words about my reading of Hegel. Baumann presents a deeply interesting and deeply compelling picture of various elements of Hegel’s thought. It has given me a lot to think about, especially as I continue to work out my own thoughts about Hegel. I do want to raise some small points in response. At one point she notes that “the transition from reason to spirit is explained by the fact that, as mentioned, Hegel treats ‘self-conscious reason’ and spirit synonymously”. I find this to be imprecise: spirit is the far broader category here, at least in the Phenomenology, which is the text to which I tried to restrict myself. Baumann is certainly right to note how Hegel’s thought changes and becomes more complex after the Phenomenology, and especially that the Encyclopædia complicates this story. Nonetheless, I believe that my points about Hegel—about teleology and history—in Autonomy after Auschwitz stand, or at least do not fall because of the story Baumann produces. Any story about Hegel’s later Encyclopædia will still need to account for the historical elements present in the Phenomenology, and especially of how to understand that text in relation to the rest of Hegel’s corpus (i.e. Hegel’s historicism). There are a slew of options here, some minimising one element over another, and I myself am not yet prepared to flesh out this relationship in the sort of detail that I would find convincing. But I importantly think the account one gives here, in response to this issue, will complicate the picture of a ‘non-metaphysical’ Hegel that Baumann highlights. For example, when Terry Pinkard, in his latest two books (Hegel’s Naturalism and Does History Make Sense?), elaborates a naturalism that includes certain teleological elements, it seems that the terrain of ‘metaphysical’ and ‘non-metaphysical’ is starting to shift, and the topography is looking different. I feel like the grounds are shifting yet and so I take all of Baumann’s reflections on Hegel seriously.

Received: 31 August 2017.


[1] For a classic statement of the issues involved with such a claim, see Pippin (1987).

[2] This rejoinder to Pickford should not be taken to imply that the account offered in the first chapter of Autonomy after Auschwitz cannot be expanded. I think it can. In fact, reading Owen Hulatt’s recent book (Hulatt 2016) made me realise that one can give a more thorough account of the dialectic of enlightenment by synthesising my approach with Hulatt’s own (see notably his first two chapters on the “cry of terror” and the experience of objects); nonetheless, I am convinced that the account I present is primary, exactly because apperception (or self-consciousness) remains primary and is required even by the account that Hulatt presents. I hope to address these issues in a future piece.

[3] There is a lot more to say here and more detail that needs to get worked out, something I am also working on for a future piece. But note that it is a common contemporary experience under capitalism that what abstractly (devoid of any particular context) appears as good, causes great suffering when inserted into a new context. (Some prominent cases might include everything from Victor Gruen’s invention of the mall—which he intended as a social space for people to come together, but which required shops for funding—to Arthur Galston’s invention of Agent Orange as an herbicide.)

[4] And this is a point that can be repeated for each of the items that Hanna delineates as essential to OHP, a point that I explicitly pursue in a piece where I show that Kant can block McDowell’s ‘neo-Hegelian’ reading of Kant even as Kant can be understood to be a conceptualist in the way that McDowell suggests (Shuster 2014).

[5] This point is most obviously and forcefully presented in Pippin (1999).

[6] Admittedly, Baumann had noted that in her discussion of my interpretation of Hegel she would be more combative because I had explicitly proposed that Adorno should have interpreted Hegel a certain way. I get this and appreciate it, but I also think that what I say following this note applies to her remarks about the rest of my book, as I shall show.

[7] A reading which falls into the class of readings of Adorno that Robert Pippin (2005) has, to my mind, convincingly shown to be problematic.

[8] See footnote 2 above.

[9] My own sense of this point—which I intend to develop in a future piece—is that Freud is essential to working out this issue properly, especially the Freud of Beyond the Pleasure Principle, and that the relationship between nature and self-consciousness is going to mimic the relationship that Freud sets up between life and death.

[10] Since the writing of my book, I have come to realise that there is a lot more to this story, especially as emerges when one highlights the Freudian resonances here, but none of this cuts against this point, and unfortunately this is something I can only flag here.


Adorno, T. (2001), Kant’s Critique of Pure Reason, trans. R. Livingstone & ed. R. Tiederman (Cambridge: Polity Press).

——— (2006), History and Freedom, trans. R. Livingstone & ed. R. Tiedemann (Cambridge: Polity Press).

Hanna, R. (2016), ‘Radical Enlightenment: Existential Kantian Cosmopolitan Anarchism, With a Concluding Quasi-Federalist Postscipt’, in D. Heidemann & K. Stoppenbrink (eds), Join, Or Die: Philosophical Foundations of Federalism (Berlin/Boston: de Gruyter).

Horkheimer, M. & T. Adorno (2002), Dialectic of Enlightenment. Philosophical Fragments, trans. and ed. E. Jephcott (Stanford: Stanford University Press).

Hulatt, O. (2016), Adorno’s Theory of Philosophical and Aesthetic Truth (New York: Columbia University Press).

Pippin, R. (1987), ‘Kant on the Spontaneity of Mind’, Canadian Journal of Philosophy 17: 449–76.

——— (1999), ‘Dividing and Deriving in Kant’s Rechtslehre’, in O. Höffe (ed.), Immanuel Kant, Metaphysische Anfangsgründe der Rechtslehre (Berlin: Akademie Verlag), pp. 53–85.

——— (2005), ‘Negative Ethics: Adorno on the Falseness of Bourgeois Life’, in R. Pippin, The Persistence of Subjectivity (Cambridge: Cambridge University Press), pp. 98–121.

Shuster, M. (2014), ‘Kant’s Opus Postumum and McDowell’s Critique of Kant’, The Southern Journal of Philosophy 52(4): 427–44.

© Martin Shuster, 2017.

Martin Shuster teaches at Goucher College in Baltimore, USA, where he is the director of Judaic studies and an Assistant Professor in the Center for Geographies of Justice. In addition to many articles in several areas, his most recent publication is New Television: The Aesthetics and Politics of a Genre (Chicago UP, 2017).