Abbreviations
Following standard practice, I refer to the 1781/1787 editions of the Critique of Pure Reason using the A/B pagination. All other texts are referred to using the abbreviations below, followed by volume and page number from the Academy edition: Kants Gesammelte Schriften, ed. Akademie der Wissenschaften (Berlin: Walter de Gruyter, 1900–). Unless otherwise noted, translations of Kant's texts are from The Cambridge Edition of the Works of Immanuel Kant, ed. Paul Guyer and Allen Wood (Cambridge: Cambridge University Press, 1992–).
Anth.: Pragmatic Anthropology
CJ: Critique of the Power of Judgment
CPrR: Critique of Practical Reason
G: Groundwork for the Metaphysics of Morals
JL: Jäsche Logic
MM: Metaphysics of Morals
M-D: Dohna Metaphysics
M-L1: L1 Metaphysics
M-L2: L2 Metaphysics
M-Mrong.: Mrongovius Metaphysics
M-Vig.: Vigilantius Metaphysics
Progress: What Real Progress has Metaphysics Made in Germany?
Prol.: Prolegomena to Any Further Metaphysics
Introduction
There is a long and ongoing debate about how to understand the central arguments of the Groundwork of the Metaphysics of Morals (1785) and the Critique of Practical Reason (1788), Immanuel Kant's two foundational works in moral philosophy, in relation to each other.Footnote 1 Is a single theory being given two complementary presentations?Footnote 2 Or does a significant shift in doctrine, perhaps even a ‘great reversal, ’ as one philosopher put it (Ameriks Reference Ameriks2000, 226), occur between the two texts? And if there is a substantive change, is it an improvement?Footnote 3 At stake is an issue of special importance: the justification of morality. What kind of justification, if any, can be given of a fundamental moral law? What kind of justification, if any, do we need of a fundamental moral law?
This essay focuses on Kant's thought during the same time period, but it foregrounds a different pair of texts: the Groundwork and the second edition of the Critique of Pure Reason (1787). Kant explicitly claims that this second edition of the Critique of Pure Reason (hereafter ‘B-Critique’) and the Critique of Practical Reason are in ‘precise agreement’ (CPrR 5.106; also 5.6–5.7). If we accept this, even just as a characterization of how Kant would like to be interpreted, then we would expect that at least some of the differences said to obtain between the 1785 Groundwork and the 1788 Critique of Practical Reason would be visible between the Groundwork and the 1787 Critique of Pure Reason, or at least that there would be some degree of ineliminable and disquieting tension between the two texts. But this is not, I will argue, what we find.
One Kant scholar who reaches the opposite conclusion is Karl Ameriks, and a quick summary of his account will provide me with a useful backdrop against which to situate my proposal. In a series of groundbreaking studies, Ameriks lays out the following narrative (Reference Ameriks2000, 191, 211–219, Reference Ameriks2003). As student transcripts of metaphysics lectures delivered by Kant in the mid- to late-1770s reveal, during this period, Kant defends a set of rationalist arguments about the nature of the soul, arguments that purportedly prove that the soul is simple, substantial, single, and absolutely spontaneous – that is, transcendentally free, an uncaused cause. But a few short years later, in the A-Critique (1781), Kant rejects all of these arguments as fallacious – all, that is, but one. There is a conspicuous silence about spontaneity. In Ameriks's view, this is because Kant still endorses his pre-Critical position on the matter: the soul is absolutely spontaneous and we can know that it is. The central piece of evidence for this, according to Ameriks, is the Groundwork. There, Ameriks argue, the pre-Critical argument for our absolute spontaneity is essentially reprised, as Kant infers our absolute spontaneity from some self-conscious capacity of theoretical reason (cf. G 4.452 and M-L1 28.268–269; Ameriks Reference Ameriks2000, 211–219, Reference Ameriks2003, 225–247). But such a view would inevitably ‘suffer…shipwreck, ’ as Ameriks puts it (2000, 191), for it lays claim to just the kind of metaphysical knowledge that the First Critique puts beyond our grasp. On Ameriks's reading, this failure shapes not just the doctrine of the Second Critique but also the revisions Kant makes to the Critique of Pure Reason for its second edition. In particular, Ameriks takes Kant to be at pains to develop his theory of self-consciousness in such a manner as to ‘systematically block…even the suggestion of any kind of argument to absolute freedom’ (Reference Ameriks2003, 258).
Ameriks's framework is, in my view, deeply illuminating. But whereas Ameriks sees a decisive break between the Groundwork and the B-Critique, it is possible to read the texts – especially Groundwork III and the B-Transcendental Deduction – as not just consistent with each other but strongly continuous, developing a single line of thought between them. I will suggest that the concerns about self-consciousness that Ameriks draws attention to in the B-Critique thread through both texts, and that the B-Critique fills in an account of self-consciousness, the shape of which is first outlined in Groundwork III. This essay is thus primarily devoted to laying out a reading of the Groundwork as prefiguring the B-Transcendental Deduction's account of self-consciousness. This task occupies part I; part II turns more programmatically to the B-Transcendental Deduction, delineating some of the parallels that emerge but also pointing to an important difference that obtains between the two. On the reading I end up with, a central aim of both texts is to argue for the necessity of a kind of self-consciousness that is fundamentally a consciousness of oneself as at once sensibly determined and free. The struggle to articulate this conception of self-consciousness and defend its possibility is central not only to the B-Transcendental Deduction, but also to Groundwork III.
Part I: Groundwork III
The first edition of the Critique of Pure Reason is published in 1781, after a long period of gestation – the so-called ‘silent decade’. But just two years after it is published, Kant identifies two sections of it that he is ‘not fully satisfied with’: the Transcendental Deduction and the Paralogisms (Prol. 4.381). These two sections of the text present Kant's theory of the cognitive role of self-consciousness: the Transcendental Deduction identifies the essential role of self-consciousness in making empirical cognition possible, while the Paralogisms reveal that the same self-consciousness generates an illusion of self-knowledge. Kant would indeed re-write these sections of the Critique of Pure Reason wholesale for the 1787 edition of the Critique, but it would not be surprising if we could see Kant grappling with these issues in the intervening years. I will suggest that they are central to Kant's thought in the 1785 Groundwork for the Metaphysics of Morals.
But the main purpose of the Groundwork is, of course, not to give an account of self-consciousness, but to undertake the ‘search for and establishment [Festsetzung] of the supreme principle of morality’ (G 4.392; all emphases in original unless otherwise noted). The search, which unfolds over the first two parts of the text, begins with ‘common cognition’ and, by ‘proceeding analytically, ’ excavates the moral principle underlying such cognition (G 4.392; see also G 4.445). In the course of this ‘search, ’ it emerges that morality confers on humanity a special value as an end-in-itself (G 4.428–4.429); that as moral subjects, human beings possess dignity and command respect (G 4.436); that morality is an exercise in autonomy (G 4.431). But whether we really are such moral subjects as our everyday moral cognition takes us to be – that is left an open question until the third and final part of the Groundwork. There Kant turns to ‘establishing’ the moral principle, and with it, the value of humanity and the reality of human dignity and autonomy. It is this closing argument that I focus on in what follows.
§1. The Problem of Groundwork III
The search for the moral principle ends with a formulation of it as a principle of autonomy. Groundwork III picks up here, identifying freedom as the condition of autonomy. Kant defines freedom as the property of a will whereby it ‘can be efficient independently of alien causes determining it’ (G 4.446). This is a merely ‘negative’ definition, he notes, but ‘there flows from it a positive concept of freedom’ as the property of a will by which it can be a ‘law to itself’ (G 4.447). And since this positive concept of freedom just describes a will under the principle of autonomy, and since the principle of autonomy is a formulation of the moral principle, it turns out that the ‘free will and a will under moral laws are one and the same’ (G 4.447).Footnote 4
It thus appears that to establish the moral principle, we need to establish the reality of our freedom. But what Kant argues next is not that we are free – only that we must think of ourselves as free. The argument begins with the claim that ‘reason must regard itself as the author of its principles, independently of alien influences’ (G 4.448). Since heteronomy obtains just when a ‘foreign impulse…give[s] the law’ (G 4.444) (and since Kant takes heteronomy and autonomy to be exclusive and exhaustive), Kant's premise is that reason must regard itself as autonomous (G 4.448).Footnote 5 If this is true for reason in general, it is true for practical reason in particular; ‘consequently, as practical reason or as the will of a rational being[, reason] must be regarded of itself as free’ (G 4.448). Indeed, it follows that ‘freedom must be presupposed as a property of the will of all rational beings’ (G 4.447; emphasis mine). But Kant goes on to argue that a being that ‘cannot act otherwise than under the idea of freedom is actually free, in a practical respect’ – where to be free in such a ‘practical’ manner is to be subject to ‘all laws that are inseparably bound up with freedom…just as if [we] had been validly pronounced free’ (G 4.448). From this it follows that if we must think ourselves free, we must take ourselves to be bound by the laws of freedom; and since, as established earlier, ‘freedom must be presupposed’ of all rational beings, we must take ourselves and all other rational beings to be bound by the moral law.
But what does not follow from this argument is that we really are free. In these opening paragraphs, then, Kant renders the problem of Groundwork III more acute, rather than resolving it. Kant has shown that reason compels us to regard ourselves as free and thus as subject to the moral law – but he has given us no assurance that our freedom is not illusory. There is thus a justificatory task that remains outstanding, but there is another problem, too. As Kant puts it, were we to be asked why we acknowledge the moral law and the demands it places on us, why we hang our own ‘personal worth’ on our moral self-assessment, ‘we could give…no satisfactory answer’ (G 4.449–4.450). On the account just given, reason compels us in a way that leaves us fundamentally unintelligible to ourselves.
§2. The Solution to the Problem
What is the solution? Kant notes that ‘one resource…still remains’:
whether we do not take a different standpoint when by means of freedom we think ourselves as causes efficient a priori than when we represent ourselves in terms of our actions as effects that we see before our eyes (G 4.450).
I will suggest that Kant takes the account just given in §1 – that we must presuppose our freedom, and that with this presupposition comes a commitment to the moral law – not to be wrong or mistaken, but to be importantly incomplete: it leaves out a shift in standpoint that occurs along the way. But what are these two standpoints? Instead of immediately characterizing them, Kant takes a circuitous route that starts from (what turns out to be) one standpoint and ends up in another standpoint. At least one of his reasons for proceeding this way is, I take it, to show that there is a natural route from the first standpoint to the second, even for ‘the commonest understanding’ – and thus to show that the second standpoint is as ‘natural to our reason’ as the first standpoint (CPrR 5.99). And the first standpoint is one that we all clearly do inhabit, for it is the one we occupy while engaged in prosaic everyday cognitive activities.
From this standpoint, we come to recognize that in order to cognize objects, we must be affected by them; and because of this, we cognize objects only if and as they appear to us, not as they are in themselves (G 4.450–451). As this is true of everything we are aware of through sensibility, it is true also of myself, since I am aware of myself through sensibility. I accordingly cognize the appearance of myself in the world of sense, not my ‘ego as it may be constituted in itself’ (G 4.451). From this first standpoint, then, the common understanding comes to recognize, even if but dimly, a distinction between what Kant calls the ‘world of sense, ’ the familiar world constituted by appearances and cognized by us, and the ‘world of understanding, ’ a world of things in themselves, conceived of as the metaphysical ground of the world of sense. And we view ourselves as we view other objects of the world of sense: all under ‘laws of nature (heteronomy)’ (G 4.452). We thus relate ‘our actions as appearances to the sensible being of our subject’ (CPrR 5.99).
But something complicates this picture. The ‘human being really does find in himself, ’ Kant says, ‘pure self-activity’ (G 4.451). What is this pure self-activity? It is, I take it, the activity of reason in generating ideas, ideas that could not have originated in sensibility, since they outstrip not only what is given in sensibility but also what could be so given, ideas such as the idea of freedom or the idea that there is a way that the world ought to be.Footnote 6 Because we are conscious of this capacity of reason in us, Kant argues, we must think of ourselves, insofar as we are reasoners (or ‘intelligences’), as members of the world of understanding. It is here that the transition to the second standpoint starts to occur – but it is far from clear exactly how this second standpoint is to be understood. A key but puzzling passage is the following:
If we think of ourselves as free [wenn wir uns als frei denken], we transfer ourselves as members into the world of understanding, and cognize autonomy of the will, along with its consequence, morality; but if we think of ourselves as bound by duty we consider ourselves as belonging to the world of sense and yet at the same time to the world of understanding (G 4.453; translation modified).Footnote 7
I will suggest that to understand this passage, it is necessary to distinguish between the world of understanding, the ‘Verstandeswelt’ explicitly mentioned in the first half of the passage, and the intelligible world, or ‘intelligibelen Welt, ’ which is described, I will argue, in the second half.Footnote 8 Kant uses both terms frequently throughout Groundwork III, and most interpreters take them to be used interchangeably. But some passages strongly suggest that there is a difference the terms. For example, Kant writes that
The concept of a world of understanding is thus only a standpoint that reason sees itself constrained to take outside appearances in order to think of itself as practical, as…is…necessary insofar as he is not to be denied consciousness of himself as an intelligence and consequently as a rational cause active by means of reason, i.e., operating freely. This thought admittedly brings with it the idea of another order and another lawgiving than that of the mechanism of nature, which has to do with the sensible world; and it makes necessary the concept of an intelligible world (i.e., the whole of rational beings as things in themselves) (G 4.458).Footnote 9
This passage, naturally read, suggests that there is a distinction between the idea of the world of understanding and a different idea that it brings in its train, viz., the idea of an intelligible world. So how should these two ideas be understood?
The thought of a world of understanding arises, as we have seen, from the standpoint of the common understanding engaged in everyday cognition. It is the thought of the metaphysical ground of the world of sense. Thus, insofar as there must be a metaphysical ground to my existence as a member of the world of sense, I locate it in the world of understanding. Moreover, insofar as I am conscious of myself as a reasoner, I must regard myself as reasoner to be free from ‘alien influence’, independent of the world of sense, and thus a member of the world of understanding. Whenever I reason, then, I take myself to do so as a free member of the world of understanding, and with the thought of my freedom comes, as Kant noted in the beginning of Groundwork III, the ‘consciousness of a law for acting’ (G 4.449). But I cannot get further than the mere thought of myself as free. I cannot cognize myself as free, since for that, as Kant puts it in the Second Critique, ‘an intellectual intuition would be required’ (CPrR 5.31). The world of understanding therefore remains something that I think myself into, but which is of unclear relevance and reality to me.
What about the intelligible world? Here we can find some clues in Kant's use of the term ‘intelligible’ in other contexts. In the Antinomies, for instance, Kant writes, ‘I call intelligible that in an object of sense which is not itself appearance’ (A538/B566; my emphasis). Notice the reference to an ‘object of sense.’ A similar connection with the sensible occurs in Kant's characterization of an ‘intelligible cause’: an intelligible cause is the non-sensible cause of an event in the world of sense, whereby the event is rightfully said to be ‘free in regard to its intelligible cause’ (though also fully determined by the order of efficient causes in the world of sense) (A537/B565). The same point is made in metaphysics lectures delivered by Kant while drafting the Groundwork. ‘A foreigner called it wild fantasy to speak of the intelligible world, ’ he reportedly says, ‘but this is just the opposite, for one understands by it not another world, but rather this world as I think it through the understanding’ (M-Mrong. 29.850; emphasis mine). I will argue that the intelligible world, unlike the world of understanding, pertains specifically, in some sense, to the sensible world; and that it is specifically the idea of an intelligible world, not the idea of the world of understanding, that provides the second standpoint from which we can resolve the difficulties of Groundwork III.Footnote 10
Recall that Kant says that it is ‘when…we think ourselves as causes efficient a priori’ that we ‘take a different standpoint’ (G 4.450). But what is it that we ‘think ourselves…causes’ of? It must be our existences as sensible beings. When the existence of a being in the sensible world is taken to be the appearance of an intelligence that is ‘efficient a priori’, the sensible existence itself is rendered the intelligible existence of a being determined (at least in part) by reason and comprehensible (at least in part) as the appearance of a free intelligence in the world of sense.Footnote 11 When we thus think ourselves members of such an intelligible world, we ‘consider ourselves as belonging to the world of sense and yet at the same time to the world of understanding, ’ as Kant puts it (G 4.453).
To summarize the two standpoints, then: from the first standpoint, we view our existence in the world of sense as the existence of a purely sensible being, and we relate ‘our actions as actions as appearances to the sensible being of our subject.’ But from the second standpoint, we view our existence – again, our existence in the world of sense – as the existence of an intelligible being, for ‘this sensible being is itself referred to the intelligible substratum in us’ (CPrR 5.99). From the first standpoint, ‘we represent ourselves in terms of our actions as effects’ unfolding in the order of efficient causes; from the second, ‘we think ourselves as causes efficient a priori’ (G 4.450, emphases mine). From the first standpoint, we view ourselves ‘under laws of nature (heteronomy)’; from the second, we view ourselves under a law ‘grounded merely in reason’ (G 4.452), which law ‘is to furnish the sensible world, as a sensible nature…with the form of a world of understanding’ and thus render it, on the reading I’m proposing, intelligible (CPrR 5.43).
The question that now arises is, Why is it legitimate to take myself to be a member of the intelligible world? I am essentially identifying the intelligence in the world of understanding that I think myself to be with the metaphysical ground of my existence. But though I must think that the metaphysical ground of my existence lies in the world of understanding, I have no knowledge of it – the metaphysical ground of my existence might well be an intelligence that is not me, for instance. Thus again, what licenses such an identification?
I take it that Kant's answer in the Groundwork, much like his answer in the Critique of Practical Reason, points to the reality of our experience of moral obligation. Recall that Kant says that ‘when we think of ourselves as bound by duty we consider ourselves as belonging to the world of sense and yet at the same time to the world of understanding (G 4.453; translation modified) – that is, we consider ourselves members of an intelligible world. To experience moral obligation is to be conscious of the moral law as actually binding on me. It is an experience that is possible only for someone who is a member of both the world of understanding and the world of sense ‘at the same time’ and who is, as member of the world of understanding, the ‘efficient cause a priori’ of her appearance in the latter. That is, moral obligation is an experience that is possible only insofar as I do inhabit an intelligible world. For only a sensible self can feel constraint; but the constraint can be moral only when it arises from the autonomous self in the world of understanding. Thus, moral constraint arises because I am, as intelligence, ‘efficient cause a priori’ of the appearance of my sensible self. What therefore secures my right to take myself as inhabiting an intelligible world is everyday moral experience, my repeated encounters with the demands of duty as real constraints. Kant thereby returns to the common moral cognition with which he opened the Groundwork and reveals that our everyday moral experience of duty is the experience of a member of an intelligible world.Footnote 12 As Kant puts it a little later, through our consciousness of the moral law as binding on us,
that unconditioned causality and the capacity for it, freedom, and with it a being (I myself) that belongs to the sensible world but at the same time to the intelligible world, is not merely thought indeterminately and problematically…but is even determined with respect to the law of its causality and cognized assertorically; and thus the reality of the intelligible world is given to us (CPrR 5.105; emphasis mine).Footnote 13
Thus in the experience of moral obligation, the reality of the intelligible world and the validity of the moral law is established. So, too, is the value of humanity. It is not as mere intelligence that I am an end-in-itself; rather, it is in my humanity, as a human being existing in the intelligible world, that I am an end-in-itself.Footnote 14
§3. ANew Problem, Left for Speculative Philosophy
Though Groundwork III thus resolves many of the problems it so acutely raised in its opening, a new problem is generated. We have left the subject, Kant says, with a ‘seeming contradiction.’ For ‘if we think of ourselves as put under obligation we regard ourselves as belonging to the world of sense and yet at the same time to the world of understanding’ (G 4.453). But can we really regard ourselves in this way? As both free and not free?
Kant says that in this case, it is actually the appearance of a contradiction that is illusory: ‘something that is unifiable is represented as contradictory’ (Prol. 4.343). And it is a ‘duty incumbent upon speculative philosophy to remove the seeming conflict’ (G 4.456). Kant scholars often take the Third Antinomy to be where theoretical reason discharges this duty (e.g., Timmermann Reference Timmermann2007, 146–147). But what Kant demands is not just the proof that it's possible for freedom and determinism both to be true. As Kant puts it,
it is an indispensable task of speculative philosophy at least to show that its illusion about the contradiction rests on our thinking of the human being in a different sense and relation when we call him free and when we hold him, as a part of nature, to be subject to its laws, and to show that both not only can very well coexist, but also must be thought as necessarily united in the same subject (G 4.456; see also G 4.457).Footnote 15
The Third Antinomy accomplishes only part of this task: it shows that it is possible that we are both free and determined, but not that freedom and determinism ‘must be thought as necessarily united.’ How is this further task to be discharged?
Kant hints at an answer in the Groundwork:
The human being…puts himself in a different order of things and in a relation to determining grounds of an altogether different kind when he think of himself as an intelligence endowed with a will…than when he perceives himself as a phenomenon in the world of sense….and subjects his causality to external determination…. Now he soon becomes aware that both can take place at the same time, and indeed must do so. For, that a thing in appearance…is subject to certain laws from which as a thing or a being in itself it is independent contains not the least contradiction; that he must represent and think of himself in this twofold way, however, rests as regards the first on consciousness of himself as an object affected through the senses and as regards the second on consciousness of himself as an intelligence (G 4.457; my emphases).Footnote 16
This passage begins by reiterating the compatibility of freedom and determinism; it then points to the nature of self-consciousness to account for the necessity of thinking the necessary unity of freedom and determinism. Just such self-consciousness is, as Kant first makes clear in the B-Transcendental Deduction, a necessary condition of the possibility of cognition. It is not just the moral subject but also the cognizing subject who is necessarily conscious of herself as at once an intelligence and an appearance among other appearances in the sensible world, the former through apperception and the latter through inner sense. Thus the cognizing subject must think of herself as a subject in whom freedom and determinism are ‘necessarily united’ – and it is naturally the task of speculative reason to demonstrate that this is so. Kant sets himself to this task in the B-Transcendental Deduction; and it is to that difficult chapter of the B-Critique of Pure Reason that I now turn.
Part II: The B-Transcendental Deduction
The Transcendental Deduction aims to show that the categories apply to ‘all appearances of nature’ (B165) and make experience, ‘cognition through connected perceptions, ’ possible (B161). The first half of the B-Deduction is centered on one kind of self-consciousness: apperception, the consciousness we have of ourselves as thinking. The second half focuses on a second kind of self-consciousness: the consciousness we have of ourselves in inner sense. If Kant is pursuing the program I have suggested he sets for himself in the Groundwork, we would expect him to argue that the consciousness of ourselves in apperception as thinking grounds or otherwise forms the consciousness of ourselves in inner sense as beings in the world of sense, and that we thus take ourselves as beings in the world of sense as appearances of ourselves as thinking subjects. In the following discussion of the B-Deduction, my aim is to explore whether there is room in the text for such a reading.Footnote 17 I begin by discussing Kant's understanding of apperception; I then segue into a discussion of the argument of the B-Transcendental Deduction, pausing occasionally to compare it to the argument in the Groundwork.
Pure apperception is the ‘pure consciousness of the activity that constitutes thinking’ (Anth. 7.141), ‘the consciousness of myself in mere thinking’ (B429; see also B157-158, B413, Anth. 7.135fn and 7.142).Footnote 18 As thinking is, in turn, ‘an act of the spontaneity of the power of representation’ (B130), so apperception is consciousness of the spontaneous ‘activity that constitutes thinking’ (Anth. 7.141).
As Kant noted in the Groundwork, the spontaneity of understanding, the faculty of thinking, is not unfettered. For the categories of understanding ‘serve merely to bring sensible representations under rules…without which use of sensibility it would think nothing at all’ (G 4.452). Thus the consciousness of myself in thinking is also presumably the consciousness of spontaneous but not unfettered activity. But in the B-Deduction, Kant introduces another element of spontaneity. ‘[O]riginal apperception, ’ he says, ‘in an act of spontaneity…produces the representation I think, ’ a representation that ‘cannot be regarded as belonging to sensibility’ but is an expression of the very spontaneous and self-conscious thinking that it represents (B132). This introduction of the ‘I think’ into the Transcendental Deduction is, I note, new to the B-Critique; in the A-Critique, the ‘I think’ is first encountered in the Paralogisms (A341/B399). But what motivates its introduction into the B-Transcendental Deduction? And is this a significant change? To answer this question, I will pick out a path through the B-Transcendental Deduction that mirrors, to some degree, the argument of the Groundwork. The significance I attribute to this, and the reason I take the ‘I think’ to end up being central to the argument of the B-Transcendental Deduction, will hopefully become apparent as I proceed.
I begin with a quick synopsis of the relevant argument in Groundwork III. Kant begins by defending the claim that the ‘free will and a will under moral laws are one and the same’ (G 4.447). His next move is to argue that this premise does indeed apply to me, for I am subject to the moral law. That is clearly a synthetic claim. How can Kant assert it? What is the ‘third thing’ that puts me in relation to the moral law and thus grounds the synthetic claim? I suggested above that it is the intelligible world, which I occupy when I take myself, as intelligence and moral subject, to be the ground of my sensible existence. Perhaps it is not too much of a stretch to say that to take this second standpoint is to engage in a kind of practical ‘synthesis’ of the world of sense and the world of understanding.
Compare now the B-Transcendental Deduction. It also begins with a claim that Kant explicitly calls ‘analytic’: ‘the I think must be able to accompany all my representations’ (B131; for ‘analytic’, see B135 and B138). The ‘I think’ ‘accompanies’ representations that are combined in a judgment. Judgments combine representations by bringing them into a kind of unity that Kant calls the ‘objective unity of apperception’ (B141): a unity that obtains when representations are all united in the concept of an object, i.e., in a way that is determined by the object, not by the subject (B137, B139). When objective unity obtains, representations relate to the object in a way that is ‘objectively valid’ (B142) and amounts to cognition of the object. In every judgment, then, the ‘I’ that thinks ‘accompanies’ all the representations taken up in its judgment, which representations that ‘I’ calls ‘my representations’. And for any ‘I’ that takes itself to make several judgments, it must be able to combine them all in a single objective unity; and all the representations thereby taken up in this unity will be called ‘my representations’ by that judging ‘I’.Footnote 19 What makes Kant's analytic claim analytic, then, is Kant's conception of judgment: for any ‘I’ that judges, all her manifold of representations ‘has a necessary relation to the I think in the same subject in which this manifold can be encountered’ (B132).Footnote 20
Kant then argues that this analytic principle presupposes a synthesis (B132, B133, B134, B135, B138 and B143). As Kant puts it, ‘the thought that these representations given in intuition all together belong to me [i.e., are my representations] means…the same as that I unite them in a self-consciousness’ (B134); or again, ‘only because I can comprehend their manifold in a consciousness do I call [these representations] all together my representations (B134, emphasis mine; see also B132 and B133). The claim is that a synthesis is necessary in order to make it possible to speak of ‘my representations’ in the first place. But why might this be? Recall that apperception is a consciousness of myself that accompanies thinking. I am thus conscious of myself as long as I am engaged in an activity of synthesizing or thinking. But it is only if I go on to effect a synthetic unity that I first become conscious of my identity: that the ‘I’ who thinks the effected synthetic unity is the ‘I’ who synthesized the manifold into the unity and the ‘I’ for whom each of the manifold representations is ‘my representation’. This synthesis therefore first makes it possible to talk of ‘my representations’ because it makes it possible to talk of an ‘I’; until such a synthesis takes place, there is as ‘multicolored, diverse a self as I have representations’ (B134).
The analytic premise thus presupposes a synthesis. Kant calls this synthesis the ‘original-synthetic unity of apperception’ (B131, B136). And he argues that the ‘supreme’ principle of the understanding is, simply, to bring all manifold of intuition to the original-synthetic unity of apperception, and thus to the synthetic unity that is necessary for judgment to be possible (B135; see also B136 and B137). The original-synthetic unity of apperception is thus a synthesis that must obtain in order for there to be a self-conscious cognizer, an ‘I’ that can think, an ‘I’ to whom the ‘supreme’ principle of the understanding is addressed; it is also the synthesis that the supreme principle of understanding enjoins us to realize. It is thus akin to the idea of the intelligible world: we must think ourselves into the intelligible world in order to undertake the standpoint of moral agency, the standpoint of a subject to whom the moral law is addressed; and what the categorial imperative enjoins is that we act as members of the intelligible world, as free intelligences answering only to principles of reason.Footnote 21 It is this idea of the original-synthetic unity of apperception that I focus on in the rest of this discussion.
At the very end of the B-Transcendental Deduction, Kant characterizes the original-synthetic unity of apperception as ‘the form of the understanding in relation to space and time, as original forms of sensibility’ (B169). This is the claim that Kant needs to defend to complete the argument of the Transcendental Deduction. For if space and time and everything given in space and time stand under the original-synthetic unity of apperception, then, given what Kant has already shown, it follows that judgment about such spatiotemporal objects is always possible and that the categories apply to them.
It is time that is of central interest to this project. Time is the form of inner sense; it is thus, Kant says, the ‘fundamental’ form of sensibility – presumably because everything given in space is also given in time (B150). But space and time are also themselves intuitions, with the unity of an intuition (B136n; also B160-161 and B160-161fn). Now Kant has already argued that ‘all manifold of intuition has a necessary relation to the I think in the same subject in which this manifold is to be encountered’ (B132). This is as true for the intuition of time as it is for our other representations, and it is true in virtue of the nature of our understanding and its requirements. As Kant puts it,
time…merely as intuition in general, which contains a given manifold, stands under the original unity of consciousness, solely by means of the necessary relation of the manifold of intuition to the one I think, thus through the pure synthesis of the understanding (B140).Footnote 22
But how is it possible to bring time, ‘as intuition in general’, to the original-synthetic unity of apperception? What does that mean, when time is also the form of intuition?
Here is one way I think the account might be fleshed out. Kant writes that
The understanding, as spontaneity, can determine inner sense through the manifold of given representations in accord with the synthetic unity of apperception, and thus think a priori synthetic unity of the apperception of the manifold of sensible intuition, as the condition under which all objects of our (human) intuition must necessarily stand, through which then the categories…acquire objective reality (B151).
Kant points here to the doctrine of self-affection: that our apperceptive activities of reasoning and of judging affect us and are represented in inner sense. Perhaps one way to understand the passage above is to take it as saying that insofar as we are thinking as we should, bringing representations given in outer sense to the unity of apperception, the manifold of inner sense, which is populated by the workings of the understanding, will also stand under the unity of apperception (see also B153 and B155). Another piece of the account is given by Kant's introduction of the ‘transcendental synthesis of the imagination’. He characterizes this synthesis as ‘an effect of the understanding on sensibility’ that ‘determine[s] the form of sense a priori in accordance with the unity of apperception’ (B152). The understanding's activity of synthesis affects sensibility to produce a succession of representations. Insofar as the understanding synthesizes the spatial manifold in the way it ought to, bringing it to the unity of apperception, the activity produces a single temporal succession in inner sense, again in accordance with the unity of apperception, that relates to and unifies the spatial manifold.
On the account roughly sketched out, the form of inner sense and its unity is determined by the unity of apperception. Here, as in the case of the intelligible world, I am thus a unity in the sensible world only insofar as the ‘I’ as intelligence and its demand for unity is the ground of this appearance. In the moral case, a causally determined heap of desires and impulses can be rendered the intelligible appearance of a finite and flawed but moral agent whose commitment to the moral law confers unity on this sensible existence. In the case at hand, a set of perceptions and other mental representations, loosely related by associative ties, is rendered the intelligible appearance of a finite but rational thinking subject insofar as the thinker brings these representations to the unity of apperception. The supreme principle of the understanding thus applies to us, and judgment is possible, insofar as our existence in the sensible world is determined by our rational activity as intelligences. If this condition obtains, however, my sensible existence is that of a rational thinker: ‘I as a thinking being am one and the same subject with myself as a sensing being’ (Anth. 7.142; see also B156, B157-158, and B429).
I have been arguing that it is possible to see striking similarities between the arguments of the Groundwork and the B-Deduction; and I have suggested that this reflects an ongoing project on Kant's part to sketch out and clarify the nature of self-consciousness. There is, however, one important difference between the two contexts. In its account of theoretical reason and the cognizing subject, the Critique of Pure Reason, Kant says, ‘place[s] reason in its proper territory, namely the order of ends that is yet at the same time an order of nature’ (B425). In our various cognitive activities, we may think of ourselves as free intelligences in the world of understanding, and we may drive our inquiry in various self-determined ways, but we cannot thereby gain cognition of our freedom. Thus, in the Metaphysics of Morals, Kant writes that
although a human being has, in his understanding, something more than [animals] and can set himself ends, even this gives him only an extrinsic value (MM 6.434; also 6.418).
It is only in moral contexts, according to Kant, that the reality of our freedom is secured. But of course, ‘moral contexts’ are not isolable aspects of life; we thus find ourselves encountering our moral obligations – and our freedom – when we are engaged in our aesthetic and scientific projects just as much as when we are involved in the more practical matters of life.
Conclusion
In this paper, I argued that the Groundwork and the B-Deduction both develop an account of our moral and cognitive life as driven by the consciousness of ourselves as at once sensible and spontaneous beings, members of both the world of sense and the world of understanding. This is a point Kant returns repeatedly to in years to come. In a late metaphysics lecture series, for instance, Kant reportedly says,
the striking phenomenon with a human being is that freedom united with natural necessity is found in him. […] [H]e is…affected by lower powers but determines himself by the independence of reason, and so he appears as ordered under reason and nature not successively, but rather at the same time (M-Vig. 29:1019-20; see also Bxxvii-xxviii, CPrR 5.6fn, and M-L2 28.583).
Kant says that to be conscious of oneself in this way, a person just needs to view his ‘existence as determinable only through laws that he gives himself’ (CPrR 5.97). I have argued that it is specifically our sensible existence that we thereby come to see as intelligible. And though much more work would need to be done even to characterize this conception of self-consciousness adequately, one attractive feature of this view, at least from my perspective, is that it makes it possible to encounter people and projects with absolute value, dignity, and moral worth in this world: the world that we’re living in.
Acknowledgments
Thanks to Angela Breitenbach, John Callanan, Alix Cohen, Samantha Matherne, and especially Thomas Land for helpful suggestions and discussions of these issues; to Chen Liang for a thoughtful response to a very early version of this paper at the German Philosophy Workshop in Chicago; and to other participants of the German Philosophy Workshop, the Berlin Summer Colloquium, and conferences at Ryerson University and the University of Southampton.
