Philosophical Studies A NONCONCEPTUALIST READING OF THE B-DEDUCTION --Manuscript Draft-- Manuscript Number: PHIL-D-15-00946R2 Full Title: A NONCONCEPTUALIST READING OF THE B-DEDUCTION Article Type: Original Research Keywords: non-conceptual content; object; objectivity; B-Deduction, cognition (Erkenntnis) Corresponding Author: roberto horácio Sá Pereira, Ph.D University of Rio de Janeiro Rio de Janeiro, Rio de Janeiro BRAZIL Corresponding Author Secondary Information: Corresponding Author's Institution: University of Rio de Janeiro Corresponding Author's Secondary Institution: First Author: roberto horácio Sá Pereira, Ph.D First Author Secondary Information: Order of Authors: roberto horácio Sá Pereira, Ph.D Order of Authors Secondary Information: Funding Information: Abstract: In this paper, I propose a new nonconceptual reading of the B-Deduction. In the reading that I am proposing, categories are not conditions for representing something (I call this the intentionality thesis) nor even conditions for representing something objectively (I call this the objectivity thesis). Instead, they are conditions for cognition (Erkenntnis). In the first step of the B-Deduction, this cognition takes the form of the propositional thinking (transcendental apperception) that the nonconceptually represented object of the sensible intuition exists objectively. In contrast, in the second step of the B-Deduction, this cognition takes the form of the apprehension (figurative synthesis) of what our human senses represent nonconceptually as existing objectively. Response to Reviewers: Powered by Edit orial Manager® and ProduXion Manager® from Aries Syst em s Corporat ion Title Page With All Author Information Roberto Horácio de Sá Pereira Professor, Department of Philosophy at the Federal University from Rio de Janeiro/UFRJ Robertohsp@gmail.com Manuscript Click here to view linked References 1 A NONCONCEPTUALIST READING OF THE B-DEDUCTION ABSTRACT In this paper, I propose a new nonconceptual reading of the B-Deduction. As Hanna correctly remarks (2011: 405), the word “cognition” (Erkenntnis/cognitio) has in both editions of the first Critique a wide sense, meaning nonconceptual cognition, and a narrow meaning, in Kant’s own words “an objective perception” (A320/B377). To be sure, Kant assumes the first meaning to account for why the Deduction is unavoidable. And if we take this meaning as a premise of the B-Deduction, then there is a gap in the argument since the categories are certainly not conditions for non-conceptual cognition (Kantian nonconceptualism). Still, I believe it is not this wide meaning but rather the narrow one that figures in any premise of the B-Deduction. Thus, in the reading that I am proposing, categories are not conditions for representing something (I call this the intentionality thesis), or even conditions for representing something objectively (I call this the objectivity thesis). Instead, they are conditions for the recognition that what we represent through the senses exists mind-independently. In the first step of the B-Deduction, this cognition in the narrow sense takes the form of the propositional thinking (transcendental apperception) that the nonconceptually represented object of the sensible intuition exists objectively. In contrast, in the second step of the B-Deduction, this cognition in the narrow sense takes the form of the apprehension (figurative synthesis) of what our human senses represent nonconceptually as existing objectively. INTRODUCTION The second half of the two-steps-one-proof B-Deduction has represented a challenge for every Kantian scholar since Henrich’s seminal paper (1982), regardless of her provenance. To begin with, the text presents great exegetical problems. For example, it is not clear at all why Kant, having proven that categories are valid or apply to the objects of the sensible intuition in general, still has to prove that categories are valid or apply to the objects of our human sensible intuitions. Considering that our human sensible intuition is just a species of sensible intuitions in general, having proven that the categories of the understanding necessarily apply to intuitions in general, has Kant not proven by the same token that categories necessarily apply to the objects of our intuition? 2 However, the major challenge is to make Kant’s statement in the second half of the B-Deduction (B161) compatible with what he claims at A901/B122-3. At A90-1/B122-3, Kant restates the core of the doctrine of his Transcendental Aesthetic, according to which “appearances would nonetheless offer objects to our intuition, for intuition by no means requires the function of thinking.” In contrast, at the end of the B-Deduction, he states, “categories are necessary a priori conditions of the possibility of all objects of experience” (B161), through which he seems to suggest that nothing can appear independently of categories. The problem becomes even more acute when we consider the troublesome footnote at B160n. Therein, Kant states that the unity of space, and a fortiori of everything that appears in it, is now dependent on a synthesis speciosa according to categories, seemly contradicting the core of Kant’s Transcendental Aesthetic. Recently, the reading of the second half of the B-Deduction has opened a new philosophical front, setting up an opposition between the new nonconceptualist and the new conceptualist readers of Kant1. The reason for this is fairly obvious. At A90-1/B122-3, Kant seems to raise the nonconceptualist claim that we do represent what appears to us without the need for concepts. In contrast, he seems to take this back in the conclusion of the B-Deduction (B161) when he claims that all syntheses of apprehension fall under the categories, and that categories apply a priori to all objects of experience. Hanna claims that the B-Deduction fails. He suggests that there must a gap in the B-Deduction. Assuming that at A90-1/B122-3 Kant envisages a real metaphysical possibility (as all of us nonconceptualists claim), Hanna 1Two prominent names in the recent nonconceptualist trend in the Kantian scholarship are Hanna (2011; 2013; 2015) and Allais (2009). Also worth mentioning are the recent works of McLear (2011) and Tolley (2013). The prominent names that arise in the conceptualist reaction are Wenzel (2005), Ginsborg (2008), Grüne (2011), and Gomes (2014). But we cannot forget that all of the major names in the Kantian scholarship have been conceptualist readers of Kant: Allison (1984; 2015), Longuenesse (1998), Strawson (1966), and so on. 3 claims that the B-Deduction cannot conclude “that all objects of empirical intuition must fall under the categories and be objects of experience” (Hanna, 2013: 14). He believes that the B-Deduction must have left room for the existence of essentially “rogue objects,” in his words, “objects that cannot even in principle be conceptualized2 (2013: 13).” Nevertheless, what is modus ponnes for nonconceptualist readers is modus tollens for conceptualist readers. For example, Grüne (2011), convinced that the argument for the proof of Kant’s B-Deduction (B161) is valid and that its conclusion is sound, concludes that Kant was never in fact a nonconceptualist. Gomes (2014) comes to the same conclusion: since nonconceptualism cannot account for the B-Deduction, it must be rejected as unKantian. Both are committed to providing a conceptualist reading of A90-1/B122-3 as an allusion to a mere “epistemic” possibility to be ruled out in the second half of the B-Deduction, rather than a real “metaphysic possibility,” to use Gomes’s own words. In this paper, I propose a new nonconceptual reading of the B-Deduction. In the reading that I am proposing, categories are not conditions for representing something (I call this the intentionality thesis), nor even conditions for representing something objectively (I call this the objectivity thesis). Instead, they are conditions for cognition (Erkenntnis). In the first step of the B-Deduction, this cognition takes the form of the propositional thinking (judging) that the nonconceptually represented object of the sensible intuition exists objectively. In contrast, in the second step of the BDeduction, this cognition takes the form of the apprehension (figurative synthesis) of what our human senses represent nonconceptually as existing objectively. This paper is divided into five sections. The first is devoted to explaining the crux that divides conceptualist from non-conceptualist readers of Kant, 2Allais (2009) seems to follow Hanna here since, for her, the B-Deduction aims to show that the categories are necessary conditions for the possibility of thinking of something as an object of self-consciousness rather than a condition for perceiving or apprehending something as an object (see, 2009: 405). 4 namely Kant’s statements at A90-1/B122-3. I take the side here of the nonconceptualist reader by arguing that the conceptualist’s attempts to read A90-1/B122-3 as if Kant was evoking a mere epistemic possibility (or a specter, to use Allison’s words) to rule it out later is untenable for several reasons, the most important of them being that the conceptualist reading is incompatible with the whole Transcendental Aesthetic. The second section is devoted to presenting the putative gap in the BDeduction. My focus here is the claims made by Hanna in 2011, 2013, and 2015. To be sure, if the Kantian view is that rational and non-rational animals can represent what appears to their senses without the need for any concepts, Kant has no means to prove that all objects fall under categories. But there is a way out: to adopt a nonconceptualist reading of the B-Deduction. This is exactly what I aim to undertake in this paper. The third section is devoted to rebutting the conceptualist reading of the final step of the B-Deduction. I shall argue here that conceptualists show no argument in favor of their claim that without understanding, we cannot represent space and time as intentional objects. Instead, recalling Strawson (1966: 86), conceptualists tell us just a big story. The fourth is devoted to rebutting Strawson’s reconstruction of the Deduction as an argument in favor of the objectivity thesis. The fifth and last section is devoted to presenting and supporting my own nonconceptualist reading of the BDeduction. THE BONE OF CONTENTION BETWEEN NONCONCEPTUALIST AND CONCEPTUALIST READERS According to its standard definition, conceptualism is the claim that mental states only possess a representational content when the subject of these states possesses the required concepts to specify canonically the putative content that the mental state is representing (Bermúdez, 1998). In 5 contrast, according to its standard definition, nonconceptualism is the opposite claim that a creature’s mental state may have content even when she lacks the required concepts to specify whatever she is representing. To be capable of representing something by the senses, the subject need not possess the concepts required to specify what she is representing. Naturally, there are different ways of understanding nonconceptualism as a general claim about perceptual experience: state view versus content view, strong and weak variations, etc. However, nothing important hinges on those distinctions for our case. What is always quite remarkable is that, regardless of how you understand nonconceptualism, Kant is invariably seen as the founding father (Hanna, 2011) of conceptualism not only by the mainstream of Kantian scholarship, but also even by those who support the opposite nonconceptualist view (Gunther, 2003: 6). When the contemporary debate is transferred to a Kantian context, it usually assumes the following form. The nonconceptualists claim that sensible intuition, and even the synthesis of imagination, represents or refers to its object independently of any concepts and in particular independently of any categories. The rational or non-rational creature is able to refer to objects by means of its sense without the need to possess any concept involved in the specification of what its mental states refer to or represent 3 . In contrast, the conceptualist claims either that sensible intuitions already involve concepts (what Schulting (2015) calls strong Kantian conceptualism) or that without concepts sensible intuitions are nothing more than a manifold of sensations without reference or devoid of representational content (what Schulting calls weak Kantian nonconceptualism). As Schulting reminds us, usually strong Kantian conceptualists are not Kantian scholars because they end up denying the I use “refer to” and “represent” alternately to mark the difference between representationalists and relationalists. The former claims that perception has a content of its own: it projects satisfaction conditions that are or are not fulfilled. Bermúdez (1998) is the best example here. The latter claims that perception is just a relation that puts the subject in direct contact with the world. The prominent name here is Campbell (2011). In the case of Kant’s interpretation, both Hanna and Allais seem to assume the relational view. 3 6 Kantian duality between sensible intuitions and discursive concepts (2015: 565)4. In this paper, I am not concerned with the textual evidences provided by each side of the debate. I am clearly on the side of the nonconceptualist readers. I have been a nonconceptualist reader since the eighties, when the contemporary debate on the philosophy of mind did not yet exist. Moreover, my conviction grew stronger after reading Hanna and Allais. What really concerns me here is the bone of contention at the very heart of the First Critique, namely Kant’s statements at A90-1/B122-3 where he explains what makes the Deduction unavoidable: Objects can indeed appear to us without necessarily having to be related to functions of the understanding. (A89/B122. Emphasis added) Appearances could after all be so constituted that the understanding would not find them in accord with the conditions of its unity.... [and] in the succession of appearances nothing would offer itself that would furnish a rule of synthesis and thus correspond to the concept of cause and effect, so that this concept would be entirely empty, nugatory, and without significance. Appearances would nonetheless offer objects to our intuition, for intuition by no means requires the function of thinking. (A90–1/B122–3. Emphasis added) Nonconceptualists claim that when Kant states that objects can appear to us without categories, what he has in mind is a real metaphysical hypothesis (to use Gomes’s words (2014)). We really can and do represent or refer to objects independently of categories or any other concepts. That is what the Transcendental Aesthetic is all about: the metaphysical possibility of representing by outer and inner sense what is independent of categories and any concepts in general (See Allais, 2009). In contrast, conceptualists emphatically deny that Kant at A90-1/B122-3 is contemplating a real metaphysical possibility. According to strong Kantian conceptualists, it is not even conceivable that objects can appear to us without necessarily having to be related to functions of the understanding. Yet according to weak or moderate Kantian conceptualists, what Kant states at A90-1/B122-3 is conceivable, but metaphysically impossible. According to Gomes, Kant is contemplating “a mere epistemic 4McDowell (1994) and Sellars (1967) are the best examples of strong Kantian conceptualists. 7 possibility to be eliminated later (in the Deduction) as an unreal metaphysical possibility” (Gomes, 2014: 6). Gomes reminds us (2014: 6) that Kant uses the indicative “can” (können) in the formulation at A89/B122, as opposed to the subjunctive “could” (könnten) at A90– 1/B122–3. The first is a stark hint signaling that he takes the possibility of objects appearing without categories as real, while the second is a mere epistemic possibility to be eliminated later. Bowman reads Kant’s statements in a similar way. He suggests that objects can appear to us without categories, but only “in the sense of a formal logical possibility” (2011). The mere logical possibility does not entail a real transcendental possibility. However, the most interesting conceptual reading of A90–1/B122–3 is Allison’s (2004: 160). Following Strawson (1966) and Henrich (1982), Allison suggests that Kant here is evoking a “specter” to be exorcised later, in the second step of the B-Deduction. He reiterates the same reading in his recently published book (Allison 2015): I refer to this possibility as a specter because its realization would result in a cognitive chaos, and I argue that the Transcendental Deduction can be regarded as Kant’s attempt to exorcise it. Although this specter may call to mind the famous Cartesian specter…it is significantly different from it. While the latter…is at the bottom of the worry about the lack of correspondence between our experience and a mind-independent reality, the Kantian specter concerns the fit between two species of representation…in the Kantian specter the problem is that…nothing would be recognizable and our experience would be nothing but what William James famously referred to as “one great blooming, buzzing confusion” (2015: 54). Allison’s assumption here is that without the categories of the understanding, our experience would undergo a radical phenomenological change. It would be reduced to a cognitive chaos or, to use the famous words of William James, to a great blooming, buzzing confusion. That is exactly what Strawson (1966) called the sense-datum theory or hypothesis. Why does Allison think so? Because as a conceptualist, he truly believes that the understanding is not only the power that makes us understand what is given to our senses and the power that makes us understand that what we intuit and perceive exists 8 mind-independently as an object. As the rule-giver for a synthesis of imagination, the understanding is also the power of creating intentional objects out of the chaotic sensory manifold given to our senses. It is as if the unification of the manifold of sensory states in accordance to rules were a real mental act that assembles the pieces of a puzzle to form a picture of reality. However, Allison’s reading, that without concepts our cognitive life would be reduced to a great blooming, buzzing confusion, lacks solid textual support. There are only a few passages in the Deduction that could, when misread, suggest Allison’s skeptical scenario. One of them is Kant’s statement at A107 that the “inner perception is empirical and forever variable.” But this certainly does not mean that without apperception and categories our introspective self-knowledge (inner perception) would be a chaotic manifold of sensory states. Nevertheless, the most important passage is this one: Unity of synthesis in accordance with empirical concepts would be entirely contingent, and, were it not grounded on a transcendental ground of unity, it would be possible for a swarm of appearances (ein Gewühle von Erscheinungen) to fill up our soul without experience (Erfahrung) ever being able to arise from it. But in that case all relation of cognition (Erkenntnis) to objects also disappear, since the appearances would lack connection in accordance with universal and necessary laws, and would thus be intuition without thought, but never cognition (Erkenntnis), and would therefore be as good as nothing for us.” (A111. Emphasis added) On closer look, however, Kant’s swarm of appearances is not James’s blooming, buzzing world of appearances: a chaotic manifold of sensory states devoid of representational content. Kant is clearly assuming that that a swarm of appearances can fill up our souls, that is, that objects can appear to our senses without experience (Erfahrung) and cognition (Erkenntnis). Allison’s mistake is to take experience and cognition as mere representations of objects. Instead, they are technical terms (Burge, 2010: 155). Cognition is neither the representation of objects nor the representation of mind-independent particulars. Instead, cognition is the realization that what we represent nonconceptually and independently by the senses in fact exists mind-independently. mind- 9 Yet, the most compelling argument is the simplest. If Kant meant his statements at A89/B122 and A90–1/B122–3 as mere epistemic possibility, why did he write his Transcendental Aesthetic? How could Kant claim therein that space and time (and whatever is represented in them) are not concepts but pure intuitions if he truly did not believe that objects can appear without necessarily having to be related to functions of the understanding? Longuenesse (1998) is the only conceptualist reader who is coherent in this respect. She clearly sees that if Kant’s statements at A89/B122 and A90–1/B122–3 are taken as mere epistemic possibility to be excluded later, at the end of the B-Deduction, we face the challenge of rereading the Transcendental Aesthetic (1998: 216). Considering that Kant rewrote his Deduction many times and his Refutation dozens of times, why did he never change his Aesthetic? Pace Longuenesse (1998), any reading of the second step of the B-Deduction that entails a rewriting of the Transcendental Aesthetic is self-rebutting. IS THERE ANY GAP IN THE B-DEDUCTION? Thus, let us assume now that Kant in his statements at A89/B122 and A90–1/B122–3 is contemplating a real metaphysical possibility: we can (in the metaphysical sense) represent and nonconceptually cognize what appears to us independently of categories and hence independently of the recognition that what we are representing exists mind-independently. At this point, the nonconceptualist that has a hard time trying to make sense of the conclusion of the Deduction where Kant claims to have proven that categories necessarily apply to all objects of experience. In the ADeduction Kant states: The pure understanding is thus in the categories the law of the synthetic unity of all appearances, and thereby first and originally makes experience possible as far as its form is concerned. But we did not have to accomplish more in the transcendental deduction of the categories than to make comprehensible this relation of the understanding to sensibility and by means of the latter to all objects of experience, hence to make 10 comprehensible the objective validity of its pure a priori concepts and thereby determine their origin and truth. (A128. Emphasis added) Thus, assuming that Kant with his statements at A89/B122 and A90– 1/B122–3 has a real metaphysical possibility in mind (we can cognize what appears to us nonconceptually), and that Kant’s concluding remarks are conceptualist, Hanna (2013; 2015) concludes that there must be a gap in the A-Deduction. If it is metaphysically possible that objects can appear to our senses without necessarily having to be related to functions of the understanding, Kant should also assume the possibility of the existence of “essentially rogue” objects, that is “objects that cannot even in principle be conceptualized” (2013: 13). To close the gap in the A-Deduction, Kant made three steps in the BDeduction (Hanna, 2013). As Hanna correctly remarks (2011: 405), the word “cognition” has in both editions a wide sense, meaning nonconceptual cognition, and a narrow meaning, in Kant’s own words “an objective perception” (A320/B377). Thus, in the first step of his B-Deduction, Kant redefined his concept of cognition in a way that guarantees there cannot be an object of cognition that resists categorization. In the second step, Kant identified “the experience of objects” with “the objects of experience,” thereby also ruling out, by stipulation, the existence of rogue objects. Therefore, the conjunction of the two first steps amounts to the following: “blind” intuitions are not real cognitions (Erkenntnis), and the objects of blind intuitions are not real objects. I do not disagree. To be sure, Kant seems to assume the wide sense to account for why the Deduction is unavoidable: we can nonconceptually cognize what appears to us without categories. And if we take this meaning as a premise of the B-Deduction, then there must a gap in the argument since the categories are certainly not conditions for non-conceptual cognition (Kantian nonconceptualism). Moreover, I believe it is not this wide meaning but rather the narrow one that figures in any premise of the B-Deduction. In opposition to Hanna, however, I do not think that Kant’s assumption of the narrow sense of cognition does result from an arbitrary 11 decision. Let us put ourselves in Kant’s shoes. He must prove that categories of the understanding apply to the objects of our sensible intuitions; otherwise, they would be empty. However, he knows that categories are not conditions of nonconceptual cognitions (nonconceptualism). Thus, the essential premise of the Deduction could certainly not be the fact that we do nonconceptually cognize what appears to us, but rather the further fact that we recognize (in the narrow sense) that what appears to us exists mind-independently. Therefore, every object of sensible intuition that we recognize (in the narrow sense) as existing mindindependently must fall under categories. This leaves plenty of room for Hanna’s “rogue objects” because it is not the case that whatever we nonconceptually cognize (in the wide sense) we also recognize in the narrow sense of existing mind-independently. How can I support my reading? I can do so by showing that, in assuming my reading, there is no gap in the B-Deduction, and that it is compatible with the metaphysical reading of the statements at A89/B122 and A90– 1/B122–3. First, if we assume that cognition (Erkenntnis) is the recognition that the mind-independent objects nonconceptually represented by the senses in fact exist mind independently, there is no gap. Kant states: Since experience is cognition (Erkenntnis) through connected perceptions, the categories are conditions of the possibility of experience, and are thus also valid a priori of all objects of experience. (B161) “All objects” does not refer to whatever we do represent nonconceptually by the senses. Instead, “all objects” refers to everything nonconceptually represented by our senses and that we recognize as existing mindindependently. In this regard, when Kant claims that “the categories are conditions of the possibility of experience, and are thus also valid a priori of all objects of experience“ (B161), what he is stating is that without categories we cannot recognize the nonconceptually represented mindindependent objects of our senses as existing mind-independently. In my reading, there is no gap in either the A or the B-Deduction and there is plenty of room for Hanna’s rogue objects. Furthermore, the reading is 12 entirely compatible with the metaphysical reading of the statements at A89/B122 and A90–1/B122–3. Even though categories are conditions for the recognition that what we represent by the senses are in fact mindindependent objects, they are not conditions for objects appearing to us. THE CONCEPTUALIST READING OF THE B-DEDUCTION Let us assume for the sake of argument that Kant in his statements at A89/B122 and A90–1/B122–3 is actually evoking Allison’s specter (a mere epistemic possibility in which he never believes) to be exorcised in the second step of the B-Deduction. In this conceptualist reading, it is up to the understanding to provide some intentional object out of the chaotic manifold of sensations. Categories are conditions for representing something as an object out of the chaotic manifold of sensory states (James’s blooming, buzzing confusion). Kant's concern is with intentionality or the aboutness of our sensory states. This is what I am calling here the intentionality thesis. Therefore, when Kant says that the “categories are conditions of the possibility of experience, and are thus also valid a priori of all objects of experience” (B161), what he is stating is that without categories, nothing out of the chaotic manifold of sensation would become an intentional object for me. Thus, in the second step of the Deduction Kant must provide the proof that categories are conditions for something that becomes an intentional object of our human sensibility. According to Longuenesse, the second step of the B-Deduction is nothing but trivial. According to her, “Kant’s aim is not simply to winnow down the scope of his demonstration. His aim is rather to radicalize his deductive procedure by reinterpreting …the manner in which the objects are given to us, that is, the forms of intuition expounded in the Transcendental Aesthetic”(1998: 213, original emphasis). Thus, while in the Aesthetic Kant claimed that we do represent objects by our senses independently of 13 concepts, in the second step of the B-Deduction Kant now claims that without the understanding, no object could ever be represented. Longuenesse finds support for her reading in the troublesome footnote of § 265: Space, represented as object (as is really required in geometry), contains more than the mere form of intuition, namely the comprehension of the manifold given in accordance with the form of sensibility in an intuitive representation, so that the form of intuition merely gives the manifold, but the formal intuition gives unity of the representation. In the Aesthetic I ascribed this unity merely to sensibility, only in order to note that it precedes all concepts, though to be sure it presupposes a synthesis, which does not belong to the senses but through which all concepts of space and time first become possible. For since through it (as the understanding determines the sensibility) space or time are first given as intuitions, the unity of this a priori intuition belongs to space and time, and not to the concept of the understanding (§ 24). (B160n. Original emphasis) The careful reader must remember that in the Transcendental Aesthetic, Kant not only claims that space and time are the forms of sensible intuition. He also claims to have proven that space and time are pure intuitions, that is, they are not only the form of what appears to our outer and inner sense, but also immediate and singular representations of space (A25/B39) and of time (A32/B47), that is, immediate and singular representations of the spatiotemporal forms. In the particular case of space, Kant quite clearly claims that without any concepts whatsoever, including the concept of space, we are already able to represent an “infinite magnitude” (B40), the intentional object of our outer sense. Pure intuition of space is a paradigmatic case of nonconceptual content: without the category of quantity or any other spatial concept whatsoever, the subject is able to represent an infinite magnitude (as the intentional object of her outer sense), of course without recognizing or understanding what “an infinite magnitude” means. Kant goes beyond this and wonders how such pure intuitions are possible. It is at this moment that he introduces a further crucial concept: forms of human sensibility. We can only immediately represent a priori the forms of what appears to our outer sense and inner 5 I do not need to reiterate here that in my view any reading of this note that implies a rereading of the entire Transcendental Aesthetic is self-rebutting. 14 sense because those forms of appearances lie a priori in us as formal constitutions of our human sensibility (B41). In the troublesome footnote mentioned previously, Kant reminds us that space and time precede all discursive concepts, including the discursive concepts SPACE and TIME as the form of the sensible intuition. However, he adds there that the unity of space and time presupposes a synthesis that cannot be given by the senses. The product of such synthesis is what he calls a formal intuition, that is, the result of the determination of the sensibility by the understanding. The key phrase for making sense of this footnote is Kant’s “represented as an object.” In the mainstream of the Kantian scholarship, that phrase is understood as the concept of intentionality or aboutness.6 Along those lines, what Kant is saying is that without categories, space could not be represented as an intentional object of the outer sense in the first place. A fortiori, without the representation of space as the object of our outer sense, nothing that appears to us under the spatiotemporal forms could be apprehended and perceived as an object of our intuitions. The pressing question is how the Kantian can now prove that without a synthesis of the understanding, we cannot represent space and anything in it as intentional objects. Here there is a divide in the conceptualist camp. According to Longuenesse’s reading of this obscure footnote, what connects the categories to the spatiotemporal sensible intuition is the Kantian figurative synthesis (synthesis speciosa) described in §24: a conceptual determination of sensibility by the understanding. This determination results from “an act of the Vermögen zu urteilen, an act of the understanding. Still, it is prior to the actual production of any discursive judgment, hence prior to the reflection of any concept and a fortiori to the subsumption of intuitions under the categories” (1998: 216). Here, we are back to the core of the conceptualist 6This tradition is so long that the list is endless. I limit myself here to quoting only a few representative names: Paton (1970), Henrich (1994), Longuenesse (1998), Allison (1984) and (2015), George (1981), Stern (1990), et alia. 15 reading: as rule-giver for sensible intuitions (rather than a discursive power), the understanding is able to perform real acts constituting intentional objects out of a chaotic sensory manifold. Yet, as Longuenesse recognizes, such reading forced her to reread the entire Transcendental Aesthetic: “space and time are given only if understanding determines sensibility” (1998: 216). In this rereading, one thing is for sure: we must reinterpret Kant’s old notions of pure intuitions (the nonconceptual representation of space as an infinite magnitude) according to the new concept of formal intuition, that is, the conceptual or quasi-conceptual representation of space as an infinite magnitude7. We are told that, before concepts and categories, space is already determined by the “non-discursive” (Longuenesse) or “pre-conceptual” (Waxman, 1981) activity of a synthesis speciosa that unifies the manifold of places, shapes, etc. into a single homogeneous infinite magnitude. The intriguing question is where is the proof of this? Perhaps Longuenesse and Waxman are right in their similar readings, but it reminds me of the wise words of Sir Peter Strawson about the Deduction: Let us note, first, that though the Transcendental Deduction is indeed an argument, it is not only an argument. It is also an explanation, a description, a story. To understand its role as a story, we must consider again all those elements of the Kantian model which we eschew in our austere interpretation…our awareness of objects must be spatiotemporal in character because this is how our faculty of sensibility is constituted. We must think about the objects in accordance to the categories because so much is demanded by the constitution of our faculty of understanding. If this is so, it is indeed true that no further proof is required and that only by means of categories an object can be thought. (1966: 86. Original emphasis) The big story is the following. The synthesis figurative or speciosa of the understanding is the tertium that connects the two heterogeneous faculties: understanding and sensibility through its transcendental activity. Because Before Longuenesse, Waxman (1981) suggested a similar reading of the same footnote. According to him, “once it is recognized that Kant explicitly ruled out only conceptual understanding and the spontaneity of thought, the B160 note should cease to occasion any qualms on this score” (1981:82). Like Longuenesse, he also claims that only through a synthesis of imagination not belonging to the senses are space and time first given as intuitions. Moreover, he also equates the formal intuitions of §26 with space and time described in the “Transcendental Aesthetic,” which result from a “pre-conceptual” determination of the sensibility by the understanding (1981:82). 7 16 it is understanding and sensibility/imagination at the same time, the synthesis links the categories to what is given to the human sensibility. With all due respect, the big story is nothing but a rhetorical solution. In contrast to Longuenesse, Allison (2004) refuses to assume that there is a pre-categorical unity of space and time (115-116). For one thing, without categories, for him the forms of space and time in the Transcendental Aesthetic are not synthetized unities, that is, they are a blind discontinuous manifold of places, forms, and figures. In his account, the Kantian concept of pure intuition covers three different phenomena. First, following what Kant explicitly states in the footnote, we must distinguish between forms of intuition and formal intuitions. The former is the indeterminate form of pure intuition, what Allison suggestively calls “pre-intuition” (2004: 116), while the latter is the determinate form of pure intuition. Second, we must in the first case (forms of intuition) distinguish between the innate capacity, or the disposition of intuiting things spatiotemporally, and what is actually intuited (2004: 115). The conceptualists assume that the subject only becomes able to represent space as an object when the understanding unites what appears as a chaotic discontinuous manifold of places as a homogeneous infinite magnitude determined by the category of quantity (B40). If Allison is right, before the determination of categories, what is formally given is just an indeterminate manifold of forms and places (we are unable to see a determined form or shape). However, Allison faces the question raised before: how can the Kantian prove that without the category of quantity we are unable to see or represent the intentional object of our outer sense as a homogeneous infinite magnitude? Moreover, if there is in fact such an argument, the further question is where is this argument? Here we are back to Strawson’s big story: just like a puzzle, the understanding can only make a coherent picture of reality by assembling the chaotic manifold pieces given to our senses. 17 THE OBJECTIVITY THESIS As we have seen, the key phrase for making sense of the second half of the B-deduction is Kant’s “represented as an object of experience.” As I have said, in the mainstream of the Kantian scholarship, this concept is understood as the concept of intentionality or aboutness. Still, we can find an alternative reading: that categories are not conditions for the intentionality of our sensory states (the intentional thesis). Instead, they are conditions for representing what subjectively appears as existing objectively (the objectivity thesis). Therefore, when Kant says that “the categories are conditions of the possibility of experience, and are thus also valid a priori of all objects of experience” (B161), what he is stating is that without categories we could not represent objectively what is given minddependently. Sir Peter Strawson has been the most expressive Kantian scholar on this point. According to him: As the investigation proceeds, however, we become aware that the word “object” is to be taken more weightily than merely a particular instance of a general concept. It carries connotations of “objectivity.” (1966: 73) The major part of the role of the Deduction will be that to establish that experience involves knowledge of objects in the weighty sense… (1966: 88) Strawson seems to recognize that sensible intuitions do represent without categories in the “light” sense of referring to particular instances of general empirical concepts. However, without categories, the particular instances of general concepts would not show that rule-governed connection that characterizes the object in the weighty sense and would be rated “as merely subjective, illusionary, or ‘seeming,’ not a true representation of how the world objectively is” (1966: 89). Thus, categories are not conditions for representing something (object in the light sense), but rather conditions for representing something objectively or mind-independently (object in the weighty sense of objectivity). The welcome consequence of the Strawsonian reading is that we no longer need to reread the entire Transcendental Aesthetic to make sense of the 18 second step of the B-Deduction. We can take Kant’s statement about the unity and the rule-governed connection belonging to the understanding rather than to the sensibility (B160n) as the statement that without the understanding, we cannot represent mind-independently what we first represented mind-dependently. Thus, we can still distinguish our pure intuition of space as an object of the outer sense (in the Strawsonian weak sense of intentionality or aboutness) from the formal intuition of the representation of space as being mind-independent (in the Strawsonian weighty sense of objectivity). Now, because Strawson sees categories as conditions for representing particulars objectively (rather than for representing something), it is noteworthy that Strawson was the first to introduce the skeptic-like hypothesis that Allison calls the specter of a chaotic manifold of sensory states (James’s blooming, buzzing world). According to Allison, the specter is an epistemic possibility rather than a metaphysical possibility (to reuse Gomes’s words), but one that is engendered by the Kantian system itself: the duality between intuitions and concepts. Without concepts, sensible intuition would be a chaotic manifold of sensory states devoid of representational content. In the same vein, Strawson calls the Kantian specter the skeptic-like sense-datum hypothesis according to which our experience could be reduced to a bundle of sense-data (1966: 109). Regardless of whether this hypothesis is unKantian or not, or whether or not it comes from the Kantian duality between intuitions and concepts, Strawson has a simpler argument against it: it does not leave room for the self-ascription of experience in the first place (1966: 98). If the potential for the self-ascription of experience is in fact a necessary condition of experience, that condition could not be fulfilled under the assumption that our experience lacks the rule-governed unity. The Strawsonian reading of the Transcendental Deduction as an argument against the skeptic-like sense-datum theory is idiosyncratic and, like Allison’s specter, faces the same problem: it lacks solid textual support in 19 the Deduction for the same reasons. Nevertheless, Strawson insists, “there are passages in the first version of the Deduction that might almost be read as comments to this suggestion” (1966: 99). He quotes the already commented passage of A111 (see page 9 of this paper) and the passage of A112 where Kant states that, without categories, (t)hese (the manifold of perceptions) would then belong to no experience (Erfahrung), and would consequently be without an object, and would be nothing but a blind play of representations, i.e., less than a dream (A112). Again, cognition (Erkenntnis) and experience (Erfahrung) are technical terms in the Kantian system. Following closely Burge’s original insight (2010: 155), I would state that cognition (Erkenntnis) or experience (Erfahrung) is a self-conscious meta-representation (the recognition) that our sensible representations are objective or are of mind-independent things. Thus, the blind play of representations and the lack of objects do not mean the lack of an objective reference, but rather the lack of re-cognition (Erkenntnis) that our senses represent mind-independent things. Commenting on Strawson’s reading, Burge claims that Strawson reduces “the problem of explaining minimum conditions on experience of objective reality to the problem of explaining necessary conditions on our conception of the relation between perceptions and their objects” (2010: 161). I would formulate Strawson’s confusion a little differently. In my view, he reduces the problem of explaining how it is possible to re-cognize (Erkenntnis) that we do represent mind-independent particulars by our senses (what I call here the recognition-objectivity thesis) to the unKantian problem of how a representation of mind-independent particulars is possible. THE NONCONCEPTUAL READING OF THE BDEDUCTION 20 Let me now provide you with a sketch of the B-Deduction in the light of my nonconceptualist reading of Kant. The starting point must be the exact statements at A89/B122 and A90–1/B122–3 as a metaphysical possibility. That is to say, independently of any concepts, we do represent mindindependent particulars, albeit unknowingly that they exist mindindependently. The first thing to notice is that if this is really Kant’s starting point, there is little sense in reading Kant’s B-Deduction as an anti-skeptical argument. At A89/B122 and A90–1/B122–3, Kant is not contemplating an epistemic possibility (Allison’s specter or Strawson’s skeptic-like sensedatum theory) to be ruled out later. Moreover, since according to nonconceptualism we do possess direct access to objects, it makes little sense to assume that Kant took Allison’s specter or Strawson’s skeptic-like sense-datum hypothesis seriously. Let us remember why the Deduction seemed necessary to Kant. Since we do not possess an intellectus archetypus, there is no direct link between the categories of the understanding and our sensible intuition: the understanding cannot create an object, which means that its concepts can be empty. Likewise, sensibility cannot make sense of what it represents, which means that it can represent blindly. That is the problem with the Deduction: since categories are not conditions of the nonconceptual representation of objects by sensible intuition and vice-versa, how can we prove that categories apply to the object nonconceptually represented by the senses? How can we prove that objects nonconceptually represented by the senses fall under categories? In face of the heterogeneity of sensible intuitions and concepts, the Deduction requires a tertium (ground of proof) that links categories to the appearances of the sensibility. In the Deduction, this tertium first assumes the form of the transcendental apperception. If mind-independent objects could indeed be represented nonconceptually by sensible intuitions without necessarily having to be related to functions of the understanding, then “something would be represented in me that could not be thought at all, which is as much as to say that the 21 representation would either be impossible or else at least would be nothing for me”(B132). Here emerges another distinction between my reading and Strawson’s reading. As Strawson reconstructs the Deduction as an argument against the skeptic-like sense-data theorist, the Strawsonian is forced to construe the transcendental apperception as the subjective selfascription of mental states in the usual sense of self-consciousness (knowing self-reference and self-attributions of mental states), roughly, I think that I am ϕ, where ϕ stands for a mental state 8 . And this subjective self- consciousness is supposed to entail the consciousness of the objective world. That is Strawson’s contrastive argument: I could only self-ascribe experiences as my own ϕ experiences if those are taken to be a subjective picture of the objective world (1966: 109). Unfortunately, the Strawsonian reading of the transcendental apperception does not fit any of Kant’s texts. Kant never claimed that the subjective unity of consciousness entails some consciousness of the objective world. On the contrary, he states that we must distinguish it from the “objective unity of self-consciousness.” He is quite explicit about this: The transcendental unity of apperception is that unity through which all of the manifold given in an intuition is united in a concept of the object. It is called objective on that account, and must be distinguished from the subjective unity of consciousness, which is a determination of inner sense, through which that manifold of intuition is empirically given for such a combination. (B139. Original emphasis) Thus, the subjective unity of consciousness as the determination of the inner sense is the best candidate for what Strawsonians call the selfascription of mental states ϕ. Based on Kant’s direct connection between transcendental apperception and the concept of object, I suggest the following nonconceptual reading. I can nonconceptually represent whatever appears to the senses as a mind-independent intentional object without concepts and categories (nonconceptualism). Nevertheless, when the I think accompanies my sensible intuition, that is, when I think about 8 Actually, Strawson has never stated this explicitly. But I (1986), Cramer (1989), and Almeida (1993) developed an account along those lines. Since then, I have changed my mind completely. 22 what the senses represent nonconceptually, I start to think of what appears to me as something independent from myself. Here the Strawsonian idea of contrast makes sense: whenever I think about what my senses represent, I realize that I am putting something against me as the thinking subject of something that exists independent of me. According to Kant’s own example, “when I carry a body, I feel the pressure of weight” (B142) (representing it as weighty), without realizing that “weighty” is a mind-independent property of some mind- independently existing body (B141). In contrast, when I start to think about what my touch and vision represent, I must realize that both weight and body are mind-independent entities. That is why Kant claims that the objective self-consciousness is the logical form of judgment in general (B140-3). Whenever I think about the nonconceptual content of my representations of body and of heaviness, I thereby judge that bodies are heavy. But the pressing question is why does the B-Deduction need a second step? Let us take a further look at the passage of §24: The pure concepts of the understanding are related through the mere understanding to objects of intuition in general, without it being determined whether this intuition is our own or some other but still sensible one, but they are on this account mere forms of thought, through which no determinate object is yet cognized. (B150. Emphasis added) In the first step of the B-Deduction, the tertium that links the categories of the understanding to objects nonconceptually represented by the senses is the propositional thought that those objects represented by the senses exist objectively or mind-independently. Thus, concluding the first step of the BDeduction, I could only think of those objects represented by the senses as existing objectively or mind-independently, e.g., by judging that bodies are heavy if I think of them according to categories. For, as Kant put in his Prolegomena, categories are just “the condition for determining judgments as objectively valid” (Prol, §39, AA 4:324). According to the example provided by Kant, my categorical judgment that bodies are heavy can only be objectively true or false if I think of bodies as material substances in 23 space and heaviness as one of their properties (B142). Until now, Kant has proven (if anything) that the nonconceptually represented objects of a sensible intuition in general must fall under categories whenever I think about them and make judgments about them. Now, according to Kant, the new tertium is the so-called figurative synthesis or synthesis speciosa “as an effect of the understanding on the sensibility” (B154). According to Schulting (2015), this represents an insuperable obstacle for Hanna and Allais’s nonconceptual reading. For one thing, for both Allais and Hanna, Kant’s figurative synthesis is nonconceptual and as such independent from the intellectual synthesis of the understanding according to categories (2015: 577). Again, the obstacle is easily removed when we remember that what is in question is not the possibility of representing objects (intentionality thesis) or the possibility of representing what is subjectively given to the senses as existing mind-independently (objectivity thesis). Instead, what is in question is the possibility of cognition (Erkenntnis). In the first step of the BDeduction, this cognition takes the intellectual form of a thought or recognition that something exists mind-independently (transcendental apperception). Categories are conditions for the recognition (thinking and judgment) that what is given exists mind-independently. In contrast, in the second step, this cognition takes the sensible form of the apprehension of something given to our senses as something that exists objectively or mind-independently. This is Kant’s figurative synthesis or synthesis speciosa, defined metaphorically as “an effect of the understanding on the sensibility” (B154). Now, the categories are conditions of apprehending space and everything in it as mindindependent objects. Why did Kant need this second step? The answer is in the quoted footnote at B160n. Kant must provide the grounds for natural science and geometry. Without showing that categories are conditions for the apprehension of the objects of our senses as existing mind-independently, natural science and geometry would be groundless. 24 Kant’s major argument of the second step of the B-Deduction can be formulated in a very simple and persuasive way. The first premise is the factual premise according to which we do in fact apprehend space as existing mind-independently (figurative synthesis or synthesis speciosa). The second is conditional: we do apprehend that space is a mind-independent particular if we represent it as a homogeneous magnitude according to the category of quantity9. Now, by applying modus ponnes to both premises, we are entitled to conclude that the category of quantity applies to space and a fortiori to everything in it. This insight also provides an easy reading of the troublesome footnote. What Kant had in mind with “space, represented as an object as is really required in geometry” (B160n. Kant’s own emphasis in bold) is not space as the intentional object of our outer sense, neither is it space as a particular existing mind-independently. Instead, what he meant is the apprehension of space as something existing mind-independently. Likewise, “the formal intuition that gives unity of the representation” (B160n) is not a replacement for the pure intuition, the representation of the form of intuition, but rather the apprehension that the representation of space is as a mind-independent object. WORKS OF KANT References to Kant’s works are given in the German Academy edition: Gesammelte Schriften, herausgegeben von der Königlich Preussischen Akademie der Wissenschaften, 29 vols. (Berlin: 1902–1983; 2nd ed., Berlin: De Gruyter, 1968, for vols. I–IX). They are indicated as follows: abbreviation of the title of the work, followed by AA., volume, and page. 9Think about non-rational animals like dogs. They certainly represent space as a mindindependent entity; otherwise, we could not make sense of their complex behaviors in space. However, dogs do not apprehend space that they represent as existing mindindependently. Thus, their nonconceptual representation of space does not fall under categories. 25 For the Critique of Pure Reason, the references are shortened, in keeping with current practice, to the pagination of the original edition, indicated by A for the 1781 edition and B for the 1787 edition. KrV.: Kritik der reinen Vernunft (1781). Critique of Pure Reason, ed. and trans. Paul Guyer and Allen W. Wood (Cambridge: Cambridge University Press, 1998). REFERENCES Allais, Lucy (2009). Kant, non-conceptual content and the representation of space. Journal of the History of Philosophy 47 (3): 383-413. Allison, Henry E. (2004). Kant’s transcendental idealism. An interpretation and defense. New Haven: Yale University Press ------------------------(2015). Kant’s Transcendental Deduction: An AnalyticalHistorical Commentary. Oxford: OUP Almeida, G. (1993). Consciência de si e Conhecimento objetivo na “Dedução Transcendental” da “Crítica da Razão Pura”. Analytica 1 (1): 187-219). Bowman, B. (2011). “A Conceptualist Reply to Hanna's Kantian NonConceptualism.” International Journal of Philosophical Studies 19 (3): 417 - 446. Burge, T. (2010). Origins of Objectivity. Oxford: OUP Bermudez, Jose L. (1998). The Paradox of Self-Consciousness. MIT Press Campbell, John (2011). “Consciousness and Reference.” In Brian McLaughlin, Ansgar Beckermann & Sven Walter (eds.), The Oxford Handbook of Philosophy of Mind. OUP Oxford Cramer, K., Fulda, Hans F., Horstmann, Rolf-Peter, & Pothast, U. (1989). “Theorie der Subjektivität.” Zeitschrift für Philosophische Forschung 43 (2): 387-390. Ginsborg, H. (2008). “Was Kant a nonconceptualist?” Philosophical Studies 137 (1): 65 - 77. George, R. (1981). “Kant's sensationism.” Synthese 47 (2): 229-255. Gomes, A. (2014). “Kant on Perception: Naive Realism, NonConceptualism, and the B-Deduction.” Philosophical Quarterly 64 (254): 1-19. Grüne, S. (2011). “Is there a Gap in Kant's B Deduction?” International Journal of Philosophical Studies 19 (3): 465 - 490. Gunther, Y. (2003). “General Introduction.” In Essays on nonconceptual 26 content. York H. Gunther (ed.). Cambridge: MIT Press Hanna, R. (2005). “Kant and nonconceptual content.” European Journal of Philosophy 13 (2): 247-290. ------------(2006). Kant, Science, and Human Nature. Oxford University Press -------------(2008). “Kantian non-conceptualism.” Philosophical Studies 137 (1): 41-64. -----------(2011). “Kant's Non-Conceptualism, Rogue Objects, and The Gap in the B Deduction.” International Journal of Philosophical Studies 19 (3): 399-415. Henrich, Dieter (1982). “The proof-structure of Kant's transcendental deduction.” In Ralph Charles Sutherland Walker (ed.), Review of Metaphysics. Oxford University Press, 640-659. ------(1994). The Unity of Reason: Essays on Kant's Philosophy. Harvard University Press Longuenesse, B. (1998). Kant and the capacity to judge. Princeton University Press McDowell, J. (1994). Mind and world. Cambridge, MA: Harvard University Press McLear, C. (2011). Kant on Animal Consciousness. Philosophers' Imprint 11 (15). Paton, H. J. (1970). Kant’s Metaphysic of Experience. London: George Allen & Unwin Schulting, D. (2015). “Probleme des ‚kantianischen‘ Nonkonzeptualismus im Hinblick auf die B-Deduktion.” Kant-Studien 106 (4): 561-580. Sellars, W. (1968). Science and Metaphysics: Variations on Kantian Themes. New York, Humanities Press Stern, R. (1990). Hegel, Kant and the Structure of the Object. Routledge Strawson, P. F. (1966). The bounds of sense. London: Methuen Tolley, C. (2012). “Kant on the Content of Cognition.” European Journal of Philosophy 20 (4): 200-228. Waxman, W. (1991). Kant's Model of the Mind: A New Interpretation of Transcendental Idealism. Oxford: OUP Wenzel, Christian H. (2005). “Spielen nach Kant die Kategorien schon bei der Wahrnehmung eine Rolle?” Peter Rohs und John McDowell. KantStudien 96 (4): 407-426.