By Sacha Golob
Comparative work in the history of philosophy is a difficult thing to do well. It requires bringing into dialogue systems and arguments which are, even when close chronological and intellectual connections exist, often driven by very different ambitions and pressures, and which are frequently couched in terminological and conceptual frameworks untranslatable without remainder. Yet such comparative work is also extremely important. This is in part because of the complex and distinctive relation between philosophy and its past. It was for Kant, and for many of his successors within European thought, both natural and necessary to vindicate their work in part by relating it to pre-existing dialectics and texts: above all, by providing a type of error theory, an explanation of how one might plausibly arrive at, say Humean empiricism, and yet why it was nevertheless fundamentally mistaken—a tactic that at times achieves something close to methodological dominance once one reaches Hegel and Heidegger.
In his book, Sorin Baiasu examines what is, I think, a particularly important comparative case study, that of Kant and Sartre. Sartre himself is often scathing of the Critical tradition—for example, his Notebooks for an Ethics dismissively suggests that, in the face of “concrete” moral problems such as “collaboration or resistance”, “Kantianism teaches us nothing” (Sartre 1992:7). Yet there are undoubtedly links between the two thinkers, links which suggest that each might provide a distinctive and informative perspective from which to view the other. Most obviously, perhaps, Sartre places a colossal emphasis on freedom. Indeed, those who oppose the Kantian focus on autonomy often see Sartre as the culmination of that pernicious trend: consider Iris Murdoch (1970:80) with her denunciation—well written, poorly thought through—of ‘Kantian man’ as Milton’s Lucifer, and her closely related and vociferous opposition to Sartre. Yet it is far from obvious how exactly the connections should be spelt out: commentators on Kant, for example, often invoke existentialism as an example of how things might go awry in the sphere of practical reasoning. Consider this from Donald Regan:
At the end of ‘The Normativity of Instrumental Reason’, Christine Korsgaard suggests that perhaps the only way we can account for the bindingness of the Hypothetical Imperative is to think reason requires us to be ‘heroic existentialists’, who choose projects for no reason at all and then stick to them in the face of competing inclination for no reason except that we chose them […]. I suppose someone with existentialist intuitions might find this admirable, but to me it seems pointless and empty. (2002:278; I have inverted the order of these sentences)
Looking beyond freedom, similarly tantalising points of contact arise across a host of other areas. Sartre’s emphasis on consciousness and self-awareness was rendered disreputable within the Continental tradition, first by Heidegger’s disavowal of mental state talk and then by the various funerals for the subject that energised structuralism and post-structuralism. Yet that very same emphasis brings him into close proximity with Kant, and offers the chance of another angle on core Kantian apparatus such as apperception. In short, there is a lot to be gained by a better understanding of the Kant-Sartre relationship.
Baiasu’s text is therefore very welcome: it offers an impressively perceptive and rigorous treatment of many of the key aspects of that relationship. It moves fluidly between Kant’s theoretical and practical works, and across Sartre’s writings from Transcendence of the Ego through to the 1965 Cornell lectures. I agree with much of what Baiasu says, and the subtlety with which he maps Sartre’s evolution in particular, for example in Chapter 6, is very welcome. But I want in this paper to press on three central issues, where I would like to hear a little more from him. The first is the link between consciousness and self-consciousness, a topic obviously of foundational importance for both Kant and phenomenology. The second and third are normativity and freedom: here Sartre’s attacks on Kant, many of which echo points made by German Idealism, are comparatively straightforward, but the rest of the picture is less so.
1. Consciousness and Self-Consciousness
The importance of self-consciousness to Kant’s story is, I take it, obvious. As the Anthropology puts it in the section titled “On consciousness of oneself”:
The fact that the human being can have the ‘I’ in his representations raises him infinitely above all other living beings on earth […]. For this faculty (namely to think) is understanding. (Anthr, AA 7:127)
In the First Critique, the ‘I think’ is likewise identified as “the vehicle of all concepts whatever, and hence also of transcendental concepts” (A341/B399). One of the core aims of the Metaphysical and Transcendental Deductions is then to construct an argument linking this general form of the understanding with the specific capacities characteristic of category deployment, and the rich intentionality of an objective world which such capacities sustain (A201/B246). In line with such an explanatory strategy, Kant insists that the awareness we have of the ‘I think’ is not itself the sort of awareness we might have of external world objects: instead “I exist as an intelligence that is merely conscious of its faculty for combination” (B158).
The question of self-consciousness is also, of course, absolutely central to Sartre. As Baiasu observes, matters are complicated here by developments between early texts such as Transcendence of the Ego and the more ‘mature’ work of Being and Nothingness: given the present context, I do not want to pursue this type of Sartrean exegesis here, but I think that Baiasu’s picture of those changes is essentially correct (p. 50). My interest, rather, is in the relation between Sartre’s magnum opus and Kant’s own position. In Being and Nothingness, Sartre argues that there is a biconditional relationship between consciousness and self-consciousness. As he puts it:
This self-consciousness we ought to consider not as a new consciousness, but as the only mode of existence in which it is possible to be conscious of something. Just as an extended object is compelled to exist according to three dimensions, so an intention, a pleasure, a grief can exist only as immediate self-consciousness. (Sartre 1969:liv)
This basic principle is visible throughout the phenomenological tradition. As Husserl puts it, “[t]o be a subject is to be in the mode of being aware of oneself” (1973:151). One can find the same view in Heidegger, perhaps most straightforwardly on those occasions when he is willing to use mental state talk: his 1928–29 lectures, for example, assert simply that “every consciousness is also self-consciousness” (2001:135). Furthermore, whilst Husserl’s changing terminologies create certain textual complexities here, all of these authors are at pains to insist that such self-awareness is distinct both from the ordinary awareness we have of objects and from the type of explicit, thematic “consciousness of oneself” that one might find in deliberate introspection or reflection. As Sartre puts it, the explanatorily primary form of self-awareness is “non-positional” (1969:liii).
Bringing these points together, we can see that Kant and Sartre share a common commitment to two claims: that self-consciousness plays some essential role in our experience of the world, and that our awareness of ourselves is in some important sense not primarily an ‘object awareness’. These claims are both extraordinarily important and highly controversial; they provide a foundational link between Kant and Sartre. Baiasu’s first chapter provides a sketch of them, but then moves very rapidly to the question of character and personal identity over time—this is understandable given his ultimate focus on ethics, but it is also something of a shame in view of their central status. Indeed, as he observes himself, “there seems to be little point in comparing further aspects of Kant’s and Sartre’s works” unless we can get clear on these basic issues (p. 43). So I would like to prompt him to say a little more.
First, how does he understand the link between consciousness and self-consciousness in the two thinkers? One way to put the issue is this. It often seems as if Sartre endorses non-positional self-awareness because he takes it to be a necessary condition on making sense of the idea of a conscious state (see e.g. Sartre 1969:xxviii). His line of argument would then be similar to that found in self representing theories of consciousness such as Kriegel’s. As Kriegel summarises:
The resulting two-stepped argument unfolds as follows: (1) conscious states are states one is aware of being in; (2) states one is aware of being in are represented states; therefore, (3) conscious states are represented states; (4) when a state is represented, it is represented either by itself or by a different state; but, (5) there are good reasons to think a conscious state is not represented by a different state; therefore, (6) there are good reasons to think that a conscious state is represented by itself. (Kriegel and Williford 2006:10)
In comparison, however, it looks like this cannot be Kant’s understanding of self-consciousness. The reason is that he takes non-rational animals to have conscious states in at least the thin sense of Kriegel’s premise (1) (see e.g. Log, AA 9:64–5) and yet he denies them, as seen in the passage from the Anthropology, the ‘I think’. In other words, Kant and Sartre share a distinctive commitment to self-consciousness and yet appear to differ radically as to its warrant; if that is right, it is a fact of striking importance. Where does Baiasu stand on this?
Second, how does Baiasu understand the joint claim that this self-consciousness is necessarily distinct from object-awareness? In his book, he offers this explanation of the claim:
Recall that, on Sartre’s account, every consciousness is at the same time consciousness (of) itself, which means that this pre-reflective level of consciousness is always at work within human experience. But then, as a necessary condition of cognition, pre-reflective consciousness cannot become the object of cognizing consciousness. (p. 52)
This is not convincing: compare the fallacious argument that, since the eye is a necessary condition of visual experience, it cannot become the object of it. Baiasu further suggests that such self-consciousness lacks “any particular determination […] it is empty” (p. 52). There is surely something right here, but it would be interesting to hear more about the similarities and differences between this and apperception and Kant’s notion of formality: would the table of judgements which cashes the ‘I think’ in the Critique of Pure Reason, for example, count as a “determination”? Baiasu makes a few brief remarks in this area when discussing Transcendence of the Ego (p. 49), but a clear statement on Being and Nothingness would be helpful.
Third, and more broadly, Kant’s account of self-consciousness is routinely criticised within the phenomenological tradition as overly intellectualist, mainly because of the links between self-consciousness and judgement set out in the Deduction. Baiasu provides useful discussion of Sartre’s refusal to privilege deliberation as a decision-making mechanism (p. 60), but it would be good to have a direct statement from him on the canonical charge of intellectualism and apperception.
2. Morality and Normativity
As I mentioned in the introduction, many of Sartre’s direct criticisms of Kantian morality are reminiscent of those found in German Idealism. So, for example, the Notebooks for an Ethics rails against Kant’s focus on the “abstract universal” or the “abstract and formal” which supposedly renders his principles unable to provide substantive guidance in the richly textured situation of concrete ethical action. This line of thought is familiar, of course, from texts such as Hegel’s Philosophy of Right. Baiasu provides a clear analysis as to how Kant might respond to Sartre’s point, for example by emphasising that maxim formulation must be done with a close eye on the changing “circumstances and characteristics” of the individual case. As he notes:
Kant is not guided by the idea of a code of norms that ought to be applied in every situation […] when he formulates the maxim for certain actions, a person has to invent the maxim and the exact formulation of the maxim will depend again on the singular characteristics of the situation the person is in, as well as on that person’s characteristics. (p. 138)
This seems exactly right to me. One might further supplement the Kantian defence with the sort of consideration, for example vis-à-vis the role of the various formulations of the moral law, which Kantians have typically employed to fend off Hegel’s accusations.
When it comes to the relationship between Sartre’s positive proposals and the Critical system, however, there are two issues where I would like to press Baiasu further—what he says at present is suggestive, but I want to try to get absolutely clear on his reading and its commitments. The first concerns the role of interpersonal relations. Sartre’s work in the years after Being and Nothingness offers a much more nuanced picture of interpersonal relations and alterity than that text where “respect for the other’s freedom is an empty word [le respect de la liberté d’autrui est un vain mot]” (1969:409). As Baiasu notes, these relations increasingly take on a foundational normative role for Sartre, a role which is also central to Sartre’s error theory as to what went wrong with Kantianism. Sartre writes:
The freedom that for Kant upholds the categorical imperative is noumenal, therefore the freedom of another. It is separated by that slight stream of nothingness which suffices so that I am not it. It is the projection of the other in the noumenal world. (1969:139)
Part of the issue concerns noumenal freedom which I discuss in Section 3. But what I want to highlight here is the idea that the two-world or two-aspect distinction which drives the Kantian system is really an attempt to model the self–other interaction. In the years after Being and Nothingness, Sartre himself often suggests that it is only by engaging directly with this relation that ethics can be secured. Baiasu summarises the trend as follows:
Sartre offers the following picture of normativity (more exactly, of the ontological source of normativity) and of Kant’s account of normativity. For him, interpersonal relations are a necessary ontological source of normative standards […]. On Kant’s account, however, Sartre notes, a person may follow a moral standard in isolation from the others, since her reason provides the standard which can guide the person in the attempt to act morally. This radical separation between reason and sensible incentives, Sartre suggests, is meant to precisely replace the distinction between two persons and the interpersonal relation, which is required for the existence of moral standards. (p. 140)
From a Kantian perspective here, I would like to hear more. On the one hand, Baiasu doesn’t say much about those Kantian texts which offers vividly concrete accounts of ethics and the interpersonal: I am thinking, for example, of the discussion of the “church” and the “ethical commonwealth” in Religion within the Limits of Reason Alone (RGV, AA 6:101). On the other hand, does Baiasu think there is anything Kant might learn from this Sartrean approach? After all, one can imagine Kantian commentators who would simply dismiss it: appealing to interpersonal relations or a direct encounter with the other, unless it is coded talk for the formula of humanity, simply reintroduces all the issues of partiality that Kant found so problematic in empiricism. After all, how we respond to such an encounter is clearly biased by a host of factors—race, gender etc.—which seem to render it inappropriate as an ethical basis.
The second point is a blunter one. There is a way of reading Being and Nothingness on which it is, frankly, hard to take that book seriously. On this interpretation, Sartre holds that the fact that you choose something is sufficient to make it choice worthy. This reading is a longstanding one—it is at the root of attacks on Sartre by figures as diverse as Merleau-Ponty and Iris Murdoch. It also fits with a certain, cautionary narrative as to the Kant–Sartre relation, one on which Sartre represents a reductio of the consequences of combining a Kantian focus on the will and on autonomy with a loss of faith in the resources of pure reason. Now, it might be that this is indeed how Sartre thought, at least at that point in time. But if so, it seems disappointing and ironic that the existentialist most associated with ethics should be wedded to such a bankrupt and unrevealing model; after all, even Heidegger, no particular friend of first-order normative theory, can be read as deploying the concept of authenticity within a mix of Aristotelian and Kantian strategies to yield something more sophisticated. So the question I have for Baiasu is this: does he hold that this was Sartre’s view in his major work, and if so, why should a Kantian take that text seriously? I ask because there are certainly passages where Baiasu appears to endorse such an interpretation. For example, he writes:
According to Sartre’s account, which I think is incomplete here, in order to determine whether the end of a value is ‘to be realised’ by a person, we should simply determine whether or not the person has chosen that end; actual choice is in this case both a necessary and a sufficient condition for the normativity of an end. (pp. 142–3)
If this is Sartre’s view, “incomplete” seems generous to the point of euphemism. I would be particularly interested in seeing Baiasu flesh out the role of authenticity as he views it. If I understand him correctly, Baiasu confines authenticity to only a comparatively limited role; yet there are certainly passages, for example in the War Diaries, where Sartre’s formulations seem so close to Heidegger’s perfectionist view of it as a ‘duty’, that such confinement is exegetically questionable (p. 226). If, alternately, Baiasu is willing to allow a stronger role for authenticity in Sartre’s ethics, then something needs to be said explicitly about how that role might either meet or avoid Kant’s sustained attacks on perfectionism. It might also be useful here to have a precise formulation of how Baiasu understands L’esprit de serieux which Sartre uses as a counterpoint to his own views.
Before moving on to freedom and its metaphysics, I want also to mention one purely Kantian issue which I would be interested to see clarified. In discussing Kant’s remarks on enlightenment, Baiasu offers a suggestive discussion of the categorical imperative and the various interpretations of contradiction in conception and in willing advanced by commentators such as Onora O’Neill (p. 168). Baiasu’s aim is to offer an alternative gloss on the idea of contradiction, a gloss on which the categorical imperative occupies a “constitutive role” in practical reasoning, parallel to the categories’ constitutive role in experience. I am not sure how immediately helpful the comparison with the categories is here: there are several quite distinct senses in which the various categories—mathematical or dynamical or both?—might be said to be constitutive for several quite different levels of experience. But what interests me most is the suggestion that being bound by the categorical imperative is, quite directly, a condition on engaging in practical reasoning. Baiasu writes:
To raise the problem of how we ought to act means to make appeal to a standard of action. Assuming that this is a standard of moral goodness understood as objective, the standard would have to be universalisable. Hence the problem of the ought already invokes the moral criterion given by the Categorical Imperative. (p. 170)
Yet this is surely too quick. There is perhaps a good sense of universalisation entailed by the very idea of a standard or reason—although even here the issue of agent relative versus agent neutral reasons needs some discussion – but one surely cannot move form that to the much thicker sense of universalisation demanded by Kantian ethics; the coherence of rational egoism testifies to that.
3. Freedom and Determinism
Sartre makes a number of criticisms of Kant’s views on freedom. At times, for example, he charges the Critical system with what Baiasu aptly calls “authoritarianism” (p. 135). The complaint is effectively that the categorical imperative is an external imposition on the agent: to put the point another way, Sartre is suspicious as to how an act may be both autonomous and yet mandated by pure reason. But perhaps the most important aspect of this strand of the Kant–Sartre dialogue concerns the metaphysical underpinnings of freedom. As I touched on in Section 2, Sartre is highly suspicious of the noumenal apparatus posited by Kant to get his theory of freedom off the ground. Indeed, Being and Nothingness argues plainly that the idea of noumenal spontaneity is conceptually incoherent. Baiasu provides a neat formulation of Sartre’s complaint:
[T]here is a contradiction in the idea of an unchanging spontaneity, since the very idea of spontaneity goes against that of a fixed, unchanging element. Hence on Sartre’s account, Kant would congeal the person in an “unchangeable denomination”, that of “the eternal subject who is never a predicate”. As such we do not really have a genuine spontaneity, a person who is really free. (p. 107)
At this point, the comparison with German Idealism is again useful, partly since I assume most Kantians will be more familiar with those texts than with Sartre’s. Recall for example, Hegel’s similar insistence that:
[W]hat the understanding understands by freedom and necessity indeed concerns only the ideal moments of true freedom and true necessity, and the two in their separation amount to nothing true. (Hegel 2010:95 [Encyclopedia, Pt. I, §48, Addition])
Now, there are undoubtedly problems created by Kant’s noumenal/phenomenal split, and individual Kantians will have their preferred methods of dealing with these, depending on familiar issues such as the choice between a two-world, two-aspect or other model of transcendental idealism. The challenge, however, for authors like Hegel and Sartre is that, whilst both stress those problems, neither wants to retreat to the type of straightforward Humean compatibilism which would entirely avoid them. The result is that it can be hard for Kantians to see how exactly either author advances the conversation. Consider, for example, Karl Ameriks, as always an extremely perceptive and charitable reader, who nevertheless summarises Hegel’s position as follows:
What is disturbing about Hegel’s position is that, while he does appear to allow that absolute or “negative freedom’” is at least part of what man has, he does not put any effort into arguing for the possibility of such freedom (he appears to think it obvious on introspection). Kant’s doctrine of transcendental idealism is at least addressed to this issue, and it at least takes seriously the deterministic implications of science and social life. […] Hegel, however, simply tries to have the best of both worlds, that is, he builds on the metaphysical doctrine of our absolute negative freedom, while impugning traditional metaphysics and dismissing all particular arguments to the doctrine. (Ameriks 2000:304)
The pressing worry is that a very similar accusation could be made against Sartre. In the book, Baiasu addresses this in part by discussing the interrelation between motive, cause and action in Sartre’s system. If I understand him correctly, Baiasu’s reconstruction is as follows. The cause of an action lies in the fact that “the current situation is apprehended as having to be changed, and this constitutes the cause of [the] action” (p. 116). But the fact that the situation is so apprehended is a function of our projects; for example, rampant hunger might give me a reason to eat within one project, and a reason to remove myself from all food in another, such as fasting. The result is that:
The cause of an action cannot determine the action; on the contrary, the cause is possible only insofar as there is a project of an action from the perspective of which the current state of the world is perceived as intolerable and having-to-be-changed. (p. 116)
Clearly, it is true that my goals determine what aspects of the world I see as requiring change. But none of this seems to remove the threat which worried Kant, namely that my goals or projects themselves, and so by extension my actions, are already simply the clockwork outcomes of a long running causal process over which I have neither power nor purview. I am unclear if Baiasu takes Sartre to resist this ‘freedom of the turnspit’ threat, and if so how exactly he thinks he does so. For example, there are commentators—McCulloch is one of the clearest Anglophone examples—who think that Sartre can quite legitimately duck the whole question of the metaphysics of freedom, and concentrate solely on its phenomenology, on what Sartre calls its “evidence” (1969:40). Thus McCulloch:
So even if, as universal determinists hold, my original decision causes (along with subsequent events, of course) my ensuing behaviour, it does not function in my consciousness in that way. Yesterday’s decision—that was yesterday. Today I must either renew it or reject it, in full view of the gaming tables. In itself as it now strikes me it has no power over me, whatever its underlying causal role: “it stands behind me like a boneless phantom”. (1994:42)
Where, then, does Baiasu stand on the charge that Sartre, like Ameriks’s Hegel, simply assumes a grade of freedom that can only be delivered by the same Kantian structures which he mocks?
In conclusion, this is an insightful and far-ranging text which raises some fascinating issues. I very much look forward to hearing a little more from Baiasu on some of the topics sketched above.
Invited: 29 October 2014; received: 1 July 2015.
 For a persuasive and influential example, see Herman (1993).↩
 For discussion see, for example, Golob (2014), Chap. 6.↩
 As to the War Diaries, consider this, for example: “Authenticity is a duty that comes to us both from outside and from inside since our ‘inside’ is an outside. To be authentic, is to fully realize one’s situated being, whatever that situation may be” (Sartre 1984:53–4). For the same strand in Heidegger, see Heidegger (1983:246–8).↩
Ameriks, K. (2000), Kant and the Fate of Autonomy. Problems in the Appropriation of the Critical Philosophy (Cambridge: Cambridge University Press).
Golob, S. (2014), Heidegger on Concepts, Freedom and Normativity (Cambridge: Cambridge University Press).
Hegel, G.W.F. (2010), Encyclopedia of the Philosophical Sciences in Basic Outline. Part I: Science of Logic, trans. and ed. K. Brinkmann & D. Dahlstrom (Cambridge: Cambridge University Press).
Heidegger, M. (1983), Die Grundbegriffe der Metaphysik (Gesamtausgabe, vol. 29/30) (Frankfurt a/M: Klostermann).
——— (2001), Einleitung in die Philosophie (Gesamtausgabe, vol. 27) (Frankfurt a/M: Klostermann).
Herman, B. (1993), The Practice of Moral Judgement (Cambridge, MA: Harvard University Press).
Husserl, E. (1973), Zur Phänomenologie der Intersubjektivität (Husserliana XIV) (Den Haag: Martinus Nijhoff).
Kriegel, U. and Williford, K. (2006), Self-Representational Approaches to Consciousness (Cambridge, MA: MIT Press).
McCulloch, G. (1994), Using Sartre (London: Routledge).
Murdoch, I. (1970), The Sovereignty of Good (London: Routledge).
Regan, D. (2002), ‘The Value of Rational Nature’, Ethics 112: 267–91.
Sartre, J.-P. (1969), Being and Nothingness (London: Methuen).
——— (1984), War Diaries: Notebooks from a Phoney War November 1939–March 1940 (London: Verso).
——— (1992), Notebooks for an Ethics (Chicago: University of Chicago Press).
© Sacha Golob, 2016.
Sacha Golob is Lecturer in Philosophy at King’s College, London. His research focuses on the intersection between Kantian and post-Kantian philosophy and contemporary work on the philosophy of mind, aesthetics and philosophical methodology. Among publications in British Journal of the History of Philosophy, European Journal of Philosophy, Kantian Review, Proceedings of the Aristotelian Society and in various edited volumes, he is the author of Heidegger on Concepts, Freedom and Normativity (Cambridge UP, 2014) and co-editor of the Cambridge History of Moral Philosophy (Cambridge UP, forthcoming).