As far as the historical dimension of inner sense is concerned, Udo Thiel (1997) maintains that there was little examination of the notion of inner sense, at least until the late 1760s. It was probably the result of the orthodox belief, throughout the eighteenth century, that humans had an ability to perceive our inner states and acts. However, inner sense became more studied by German empirical psychology in the 1770s, through writers such as Christoph Meiners (1747-1810) and Michael Hissmann (1752-1784). Inner sense is still today a problematic subject matter in philosophy[2] which, even before Kant, had been regarded as an accepted condition under which we can observe or “introspect” our own mental states. In this vein, inner sense constitutes a source of knowledge of ourselves as individuals, namely, of the human mind in general (Thiel, 1997). However, our capacity of saying these inner states (e.g. states of anger, pleasure, fear, etc.) are my states does not rely on the inner sense but rather on apperception. Moreover, Sahabeddin Yalcin rightly claims that “Kant’s concept of self-knowledge depends upon his notion of inner sense” (Yalcin, 2002, p. 182).

1. Inner sense and apperception

Kant points out in CPR[3] B152 that inner sense was identified with the faculty of apperception in the systems of psychology during the XVIII century. He, nonetheless, argued for its differentiation, claiming that inner sense should constitute the subject matter of psychology, whereas apperception should be that of logic (Kant, 1900, AA 7:141; 1998, CPR A106-7, B132; Schulting, 2015). Of course, such a distinction can be correctly identified if we precise the term apperception, mentioning that apperception can be either pure or empirical. Empirical apperception is related to the empirical consciousness of ourselves, and is tantamount to inner sense. Empirical apperception is identified with a certain form of self-consciousness in the CPR and in Metaphysika Dohna (1792-3), where Kant claims: “we name only one inner sense - the faculty of the consciousness of one’s own existence - in time empirical apperception” (Kant, 1900, AA 28:673). Pure apperception, by contrast, is related to the synthesis of the empirical manifold given in sensibility, the objective formulation of judgements and the transcendental self-consciousness (1998, CPR B68, A105, B138-9, B142, B157).

Inner sense provides us with an empirical manifold that only can be connected by the pure apperception through the categories (1998, CPR B154). It means that inner sense alone does not give the consciousness of such states as belonging to ourselves, for “all manifold of intuition has a necessary relation to the I think [emphasis added] in the same subject in which this manifold is to be encountered. But this representation is an act of spontaneity [emphasis added], i.e., it cannot be regarded as belonging to sensibility” (1998, CPR B132). In other words, we demand the representation ‘I think’ by which all our representations are related to our consciousness (Mohr, 1991; Howell, 2001; Aquila, 1983). As a result, pure apperception is an active faculty grounded on the understanding, while inner sense is a passive faculty, grounded on sensibility, which does not contain yet any determinate intuition (1998, CPR B158; 1900, AA 7:140-1).

Moreover, Kant attributes different terms to two kinds of self-consciousness, which take place in an empirical or transcendental level respectively. These terms can be grouped as follows: i) discursive and intuitive self-consciousness, ii) consciousness of oneself divided into reflection and apprehension, iii) intellectual and empirical consciousness of my existence and iv) transcendental and empirical apperception. I shall provide an explanation of every group.

First, Kant holds that “consciousness of oneself is either discursive in concept or intuitive in the inner intuition of time. - The “I” of apperception is simple and binding” (1900, AA 7:143 footnote). In the discursive self-consciousness the “I” is regarded as a simple representation, while in the intuitive consciousness the “I” is regarded as a manifold of associated representations in time. Kant indirectly equates these forms of self-consciousness with discursive and intuitive apperception. These two forms of apperceptions represent ‘a doubled consciousness of the I’ (ein doppeltes Bewußtseyn dieses Ich); the former is the “I” of mere thinking, which belongs to logic and has no content, while the latter is the “I” of inner perception which belongs to anthropology and has a content provided by inner sense (1900, AA 7:397-8; Dessoir, 1924). This discursive consciousness is nothing but a reflection, which is a consciousness of the activity in the combination of the manifold of representation according to the rule of unity of that manifold (1998, CPR B158-9; 1900, AA 7:141). Moreover, discursive consciousness is regarded as a pure apperception of one’s mental activity, so that “the “I” of reflection contains no manifold in itself and is always one and the same in every judgment, because it is merely the formal element of consciousness” (1900, AA 7:141). Therefore, Kant ascribes an epistemological function to this form of consciousness, without which the synthetic power of the understanding could not connect the manifold of intuition through concepts and judgements.

Moreover, the intuitive consciousness refers to the manifold of our inner states which is represented through our inner sense as related in time: “the I [emphasis added] of apprehension is a matter of a manifold with representations joined to one another in the I as object of intuition” (1900, AA 7:143 footnote). Inner sense should not be identified with inner experience, but as a necessary condition of inner experience (Monzel, 1920), for inner sense contains the material of our consciousness, namely a manifold of empirical inner intuition which can be apprehended and represented as something that belongs to the “I”. In this form of consciousness, the “I” is represented through inner empirical intuition, that is to say the ‘I’ is affected inwardly by experiences according to successive or simultaneous relations of time (1900, AA 7:141-2).

Second, Kant states that the consciousness of oneself can be divided into reflection and apprehension; the first one is consciousness of the understanding, while the second one is consciousness of inner sense. In Kant’s own words, “inner activity (spontaneity), by means of which a concept (a thought) becomes possible, or reflection; and receptiveness (receptivity), by means of which a perception (perceptio), i.e., empirical intuition, becomes possible, or apprehension” (1900, AA 7:134 footnote). The first kind of consciousness provides us with the simple representation ‘I’ that lacks any intuitive content, whereas apprehension provides the ‘I’ that contains an empirical manifold which makes an inner experience of our states possible (1998, CPR B408; Sturm, 2017). Kant’s view of apprehension, regarded as a condition for the possibility of empirical intuition, can be identified with the synthesis of apprehension performed by imagination, which is nothing but “the composition of the manifold in an empirical intuition, through which perception, i.e., empirical consciousness of it (as appearance), becomes possible” (1998, CPR B160; B164).

Third, Kant maintains that the intellectual consciousness of our existence is concerned with the representation I am, which is not an intuition, but rather an intellectual representation that accompanies all judgements and actions of our understanding (1998, CPR BXXXIX-XL, B277-8; 1900, AA 7:134 footnote). This intellectual consciousness seems to be no other than the synthetic original unity of apperception according to which “I am conscious of myself not as I appear to myself, nor as I am in myself, but only that [emphasis added] I am. This representation is a thinking, not an intuiting” (1998, CPR B157). In the early 1790s, Kant declares that we represent us to ourselves in every thought through intellectual consciousness. However, he further holds that “I” cannot “cognize myself” by means of it. I only cognize, by means of it, that I am that which makes the act of the understanding, namely, that “I am” (1900, AA 28:712). He, by contrast, describes empirical consciousness as that by means of which we are conscious of our existence in time. This empirical consciousness informs us about changes in our mind, which can only be determined, if we represent the existence of something persistent through outer sense. Kant emphasizes that “inner experience itself is consequently only mediate and possible only through outer experience” (1998, CPR B277).

I am in agreement with Arthur Melnick’s contrast between the empirical self and the transcendental self. He claims that the first one is nothing but “the perceiving states that appear to me in inner attending”, whereas the second one is the “intellectual thinking subject I am”. However, he attacks those who, like Peter Strawson and David Carr, inflate the transcendental self to a noumenal a-temporal entity. Melnick’s approach to the problem of inner attending starts by not considering as conclusive the claims: i) “the self appeared to” is noumenal and ii) we are mere appearances, since we can only reach a non-noumenal grasp of ourselves and our identity (Melnick, 2009). He further suggests that there is a third mode of existence which is neither noumenal nor phenomenal:

It is this existence which is appeared to, which grasps itself in transcendental self-consciousness, and which, through being thus appeared to, is fixed or identical or abiding. This existence is the accompanying intellectual action I am in determining (unifying) my inner attending and is itself determined (fixed through variation) by being the identical intellectual action that keeps up with the attending. (Melnick, 2009, p. 125)

I contend that Melnick’s proposal of this third mode of existence does not help us to understand Kant’s account of the self and that such mode of existence is a combination of the first two modes of existence, which finds no place in Kant’s texts. On the one hand, Kant is reluctant to admit a noumenal cognition of the self since human beings are incapable of reaching noumenal cognition in general. It means that we can only attain a phenomenal cognition of the self which involves intuitions, feelings and other representations related in time, present in appearances. On the other, his reference to B157–158 provides no evidence of this third mode of existence, but rather explains the existence of a transcendental “I” that cannot be perceived as any other appearance. Instead, it is a primordial unity of consciousness whose existence must be presupposed at a logical level.

Furthermore, Kant holds in the CPR and in the Anthropology that we cognize ourselves as we appear to ourselves, not as we are in ourselves (1998, CPR B152-3; 1900, AA 7:398). Time, as form of inner sense, rules all our representations, including those intuitions of our inner states:

However, that we only cognize ourselves through inner sense as we appear to ourselves is clear from this: apprehension (apprehensio) of the impressions of inner sense presupposes a formal condition of inner intuition of the subject, namely time, which is not a concept of understanding and is therefore valid merely as a subjective condition according to which inner sensations are given to us by virtue of the constitution of the human soul. (1900, AA 7:142)

In my view, the transcendental function ascribed to time in the CPR is fundamental for comprehending the pragmatic-observational doctrine of human nature in the Anthropology, since time grounds human thought and action. However, Kant does not focus on physiological anthropology, which deals with the investigation of the human being’s limitations determined by nature, but on pragmatic anthropology that is concerned with the human being’s potentialities as a free acting being (1900, AA 7:119; Pappe, 1961; Foucault, 2008).

Fourth, Kant maintains in the first edition of CPR that transcendental apperception is an original and transcendental ground of “the unity of the consciousness in the synthesis of the manifold of all our intuitions, hence also of the concepts of objects in general, consequently also of all objects of experience” (1998, CPR A106). In other words, this form of apperception is an objective condition of experience, because provides a unity of consciousness without which we cannot intuit nor think of objects (Schulting, 2015). On the contrary, Kant claims about the empirical apperception that: “the consciousness of oneself in accordance with the determinations of our state in internal perception is merely empirical, forever variable; it can provide no standing or abiding self in this stream of inner appearances” (1998, CPR A107; translation modified slightly). Kant further notes in his handwritten draft of the Anthropology that by means of empirical apperception, the subject attends to itself, i.e. it affects itself, and, as a result, certain representations related in time (simultaneously or in succession) are brought to consciousness (1900, AA 7:399). In the same way, Kant claims in Metaphysik K2 that empirical apperception is “when I am conscious of myself by means of inner sense”[4] (1900, AA 28:712; my own translation). H. J. Paton suggests that empirical apperception is not only concerned with the consciousness of our mental states but also with the very possibility of cognition of external objects:

Empirical apperception is said to be concerned with the states of mind, but it must at the same time be concerned, not only with the order in which these arise in our mental history, but also with the particular way in which (as appearances) they are combined in the object. (Paton, 1939a, p. 402)

I deem problematic his interpretation in as much as it destroys Kantian boundary between empirical and pure apperception, that is, between the empirical consciousness of a representation given to us in experience and the a priori condition of that consciousness. It means that the particular way in which appearances are “combined” in the object cannot be an object of the empirical apperception, because it is the result of the synthesis a priori of the manifold given in intuition through the categories of the understanding. In the second edition of the CPR, Kant suggests that the empirical unity of apperception is derived from the transcendental unity of apperception and cannot be necessarily and universally valid but it has merely a subjective validity, because it is concerned with an empirical association of representations that may differ in each person (1998, CPR B140).

To sum up, the first group of terms related to the transcendental selfconsciousness deal with the consciousness of both the connection of the manifold and the fact that all representations belong to the same subject. Needless to say, this group operates at the pure, discursive, objective or logical level of human cognition. The second group related to the empirical self-consciousness, elicited by our inner sense, points out an empirical consciousness of oneself that operates at the empirical, intuitive, subjective or psychological level of human cognition. However, the latter form of cognition should not be thought of as a source of chaotic unrelated representations but as representations temporally related of our own existence, which are subject to the synthesis of our understanding. On this point, I am in agreement with Aquila’s claim that “what Kant calls “empirical apperception,” or “inner sense,” is the awareness of oneself insofar as one is aware of particular stretches of intuited time “synthesizable” together with others into the right sort of whole” (Aquila, 1983, p. 175).

2. The content of inner sense

Any attempt to show the object of inner sense must deal with the difficulty that there is not a specific object but an array of different things that can be regarded as object of inner sense (Mohr, 1991). It is to an extent comprehensible Andrew Brook’s claim: “Kant’s doctrine of inner sense is a mess, and to sort it out would take more space than the task is worth” (Brook, 1994, p. 78). This difficulty is partly the result of an unsystematic treatment of inner sense in the CPR and Kant’s customary statement that all appearances in general belong to inner sense (1998, CPR A98-9, B67, A34/B50).

I contend that inner sense contains the material of consciousness and a manifold of empirical inner intuition, namely the “I” of empirical apprehension (1998, CPR A22-3/B37; 1900, AA 7:134 footnote, 142-3). Kant identifies inner sense with empirical apperception, by which the subject is conscious of empirical representations of its own mental states (1998, CPR A107; 1900, AA 28:673; 7:399). However, inner sense does not provide us with the intuition of one object given in experience, to which the terms substance, one, abiding, and identical over time can be attributed. That is to say, we cannot obtain from experience any intuition of the “unity” of the self, nor a representation of something numerically the same, or abiding in us (Brook, 1994; Allison, 2004). In contrast, we can only reach a cluster of intuitions, perceptions, imaginings, memories, and so on:

The consciousness of oneself in accordance with the determinations of our state in internal perception is merely empirical, forever variable; it can provide no standing or abiding self in this stream of inner appearances, and is customarily called inner sense or empirical apperception [emphasis added]. (1998, CPR A107)

Accordingly, no “fixed” or “abiding” self can be object of empirical consciousness because the empirical apperception can only provide the intuition of appearances that are nothing but subjective occurrences governed by time (1900, AA 8:154; 7:397). On the contrary, we represent us to ourselves, via inner sense, as a ‘stream’ (Fluss) of inner appearances, in which no intuition of numerical identity can be found (1998, CPR A363-4; Brook, 1994). However, the flow of mental representations in empirical consciousness does not imply that time itself is a flow of consciousness, nor that it is a sort of psychological event (1998, CPR A36-7/B53). On the contrary, time, as form of inner sense, is a necessary condition of the flow because it links our representations according to relations of present, past, and future, succession and simultaneity (Pacheco Acosta, 2018). Thus, I agree with Mathew S. Rukgaber who holds that “Kant must mean that the “form” of inner sense describes a structural feature of our perspective that enables determinate temporal extents like the flow of consciousness” (Rukgaber, 2009, p. 180). Strawson has taken the same direction and has suggested that the self is not given to us as a single, simple thing, but rather as “a flux of inner representations, a succession of constantly changing, albeit connected, perceptions” (Strawson, 1997, p. 266). In other words, our inner sense can only inform us about relations between representations of our inner states.

I suggest that both David Hume and Kant are reluctant to admit that human being is capable of an ‘intuition’ or ‘perception’ of the unity or singleness of the self. Such idea is also present in the Handschrift of the Anthropology, where Kant states: “the I which has been observed by itself is a sum total of so many objects [emphasis added] of inner perception” (1900, AA 7:399). It follows that we would become conscious of the “plurality” of the self rather than of its “unity”. My interpretation is in line with Aquila’s claim that

Kant himself provides what might appear the most natural suggestion, if we assume that inner sense provides no direct awareness of the “self” or the “subject” of experience. In that case, it would seem, inner sense must provide an awareness of at least some of those items that we might regard as “determinations” of the self. (Aquila, 1983, p. 154)

Aquila suggests that the list of these determinations includes not only Vorstellungen but also “feelings and inclinations and perhaps even in some sense “acts” of thinking and willing” (1983, p. 154). On the contrary, what subject perceives, according to Hume, is “nothing but a bundle or collection of different perceptions, which succeed each other with an inconceivable rapidity, and are in a perpetual flux and movement” (Treatise, p. 252; see also Howell, 2001)[5].

Collin Marshall argues that the Humean bundle view does not fit with Kant’s view of the self and offers two arguments. First, Marshall thinks that the relation between the self and its representations should be understood as follows: “the self is the thing that combines representations in synthesis” (Marshall, 2010, p. 11); this relation, according to Marshall, simply cannot be admitted by Hume. Against this argument, I hold that he misses the crucial distinction between the empirical and the logical self (1900, AA 20:270; 7:134 footnote; 1998, CPR B407-8). Instead, he reduces Kant’s account of the self to its logical expression as unity of consciousness, overlooking the empirical representations such as feelings, thoughts, desires, etc., which are associated to the empirical self.

As to the second argument, Marshall correctly holds that for Kant we do not know ourselves as we are in ourselves but as we appear to ourselves. He goes on, however, to claim that “appearances are representations, so if we are bundles of representations, then there needs to be another level of representations that is the appearance of that bundle. But this is clearly not Kant’s view” (Marshall, 2010, p. 11). It is far from clear why he thinks that Hume’s view of bundle would be suitable for Kant’s account of the self when an “appearance of a bundle” is presupposed. Marshall turns the bundle into an appearance. On my account, inner sense provides with a stream of representations of our states which, in themselves, are nothing but empirical appearances (1900, AA 7:398; Prauss, 1971). Thus, it seems there is no reason to reject that the empirical self is, for Kant, a “bundle of appearances”, which are not unarranged unrelated representations, but rather representations related in time and determined by the understanding through categories (1998, CPR B157-8). The important point is the use of appearance as a “limiting” term that distinguishes the empirical cognition of the self as it is given in experience from the (unattainable) cognition of the self as it is in itself (1998, CPR B164; 1900, AA 7:398). I, nonetheless, agree with Henry E. Allison on admitting two essential differences between Hume and Kant: first, Kant distinguishes inner sense from apperception, and such distinction implies that the impossibility of intuiting the self as unity does not lead to reject the thought “I”. Second, unlike Hume, Kant justifies the possibility of a genuine inner experience (Allison, 2004).

3. Change and permanence in inner experience

This is where I want to express my disagreement with Rudolf Makkreel, who claims: “the representations of inner sense cannot be made clear and determinately fixed. They constitute an indeterminate temporal stream” [emphasis added] (Makkreel, 2014, p. 19). In contrast, I argue that the temporal relation of those representations already constitutes a kind of determination made by imagination, whose synthesis (figurative) submits the manifold of those representations under the abovementioned temporal rules (1998, CPR B151; Carl, 1992). In other words, Kant is underlining that those representations do not constitute “one object” but rather a stream of representations, so that when we are conscious of ourselves, we become aware of a stream or flow of states (e.g. thirsty, angry, etc.), in which none abiding state is continuously present throughout our life.

On top of that, Kant states in the Anthropology that outer objects in space appear next to each other and abidingly fixed, while inner sense “sees the relations of its determination only in time, hence in flux, where the stability of observation necessary for experience does not occur” (1900, AA 7:134). As Aquila notices, there is no reason to assume a parallel characterization of the material provided by inner (concerning ourselves) and outer sense (Aquila, 1983). Perhaps the most notable asymmetry between outer appearances, given in space and time, and inner appearances, given only in time, consists in the abiding character of the material given in the former and the “fleeting” character of the latter. As he puts it:

Although both are appearances, the appearance before outer sense has something standing and abiding in it, which supplies a substratum grounding the transitory determinations, and thus also a synthetic concept, namely that of space and of an appearance in it; whereas time, which is the only form of our inner intuition, has in it nothing abiding, and hence gives cognition only of a change of determinations, but not of the determinable object. (1998, CPR A381)

This suggests that we can only cognize, via empirical self-consciousness, the determinations of our mind which change in us, so that the cognition of a persistent enduring object is beyond our inner experience. The fact that “in that which we call the soul, everything is in continual flux, and it has nothing abiding” (1998, CPR A381) need not preclude us from arguing for a “dynamic” theory of the self, which regards the “self”, or rather the group of its states, as a living being that changes relentlessly (Aquila, 1983). Kant, nevertheless, holds that we can only determine the change in ourselves, if we have the perception of something persistent in relation to which the change is determined (1998, CPR BXXXIX footnote). Given that we cannot find the perception of something persistent in ourselves, then it should come from the actual existence of things outside us:

The perception of this persistent thing is possible only through a thing outside me and not through the mere representation of a thing outside me. Consequently, the determination of my existence in time is possible only by means of the existence of actual things that I perceive outside myself. (1998, CPR B275)

Kant suggests that inner sense is a necessary, but not sufficient condition of inner experience, since the empirical consciousness of our inner changes is made possible by outer experience (1998, CPR BXL footnote, A205/B250; Mohr, 1991). Here a contrast between Descartes and Kant concerning self-consciousness may be convenient, since Kant claims that self-consciousness relies on consciousness of outer objects while, for Descartes, the certainty of the subject’s existence is prior and apart from cognition of external things. Thus, it makes sense to hold that Descartes “understands self-consciousness as immediate, not as mediated by consciousness of anything different from the self” (Rockmore, 2012, p. 307). Kant, by contrast, conceives of empirical self-consciousness in terms of awareness of flowing representations that can be conceptualized by reference to an outer perception of substances (Aquila, 1983). Perhaps the human living body, which is changing over time, is not as persistent as a mountain, but even our spatiotemporal representations of it have a certain level of persistency, by which we are conscious of our inner changes:

The persistence of the soul, merely as an object of inner sense, remains unproved and even unprovable, although its persistence in life, where the thinking being (as a human being) is at the same time an object of outer sense, is clear of itself. (1998, CPR B415)

Certainly, the human being does not consist merely of a mind but also of a body that can be object of outer sense, but the connection between body and soul is not relevant in Kant’s anthropological agenda (1998, A342/B400; 1900, AA 20:308; 23:31-32). Unlike Béatrice Longuenesse, I strongly believe that empirical self-consciousness should not be reduced merely to a consciousness of both “outside objects” and “the distinction between the temporal determinations of those objects and the temporal determinations of one’s perceptions and experience of them” (Longuenesse, 2006, p. 302). Of course, the empirical self-consciousness relies on outer sense inasmuch as the temporal organization of inner representations is only possible by means of external objects. However, Longuenesse overlooks the fact that the human being, via self-consciousness, is aware of a plentiful set of representations such as thoughts, memories, and feelings whose existence does not demand the current presence of external objects.

Of course, Kant contradicts the problematic idealism approach to selfconsciousness, suggesting that the temporal relation of our inner representations in the consciousness of our existence depends upon our experience of external things of the world in which we live. For instance, we can be conscious of the fact that some thoughts or feelings take place regularly in our mind after, before or during other representations. We can also be conscious of the fact that those representations and the occurrence of certain external events take place at the same time. However, Mohr (1991) is right to say that the data of inner sense do not have a propositional structure, but our empirical ability to conceptualize all empirical data is conditioned to the language we learn in the community with others (1900, AA 7:127, 324; 25:1195-6, 1417). The human being has undoubtedly the capacity to formulate judgements concerning to its inner experience in which the I is the subject and various empirical are the predicate; for instance, “I am cold”, “I am dubious”, “I am ashamed”, etc., (1900, AA 18:186).

4. Feelings, sensations and perceptions as components of inner sense

Kant holds that inner sense and time constitute a totality in which all our representations are contained and takes inner sense for “the sum of all representations” (1998, CPR A155/B194; A177/B220). It is precisely this wide range of notions, contained by inner sense, what has motivated a disagreement on what is the matter of inner sense. For instance, Yalcin declares that “thought”, “desires”, “willing” and “decision-making” are not an inner manifold and, therefore, do not belong to inner sense. These, by contrast, belong to “the active aspect of the self” (Yalcin, 2002, p. 185). Kant, nevertheless, admits that the human being is empirically conscious of its mental operations. And if it is so, these operations would not be only representations of inner sense (empirical self-consciousness) but also would be determined by time, as long as the latter is the form of inner sense (1998, CPR A33/B49, A357). This seems to be borne out also by J. Vogel’s claim: “there is a way to characterize or provide content to the notion of oneself: as the subject of one’s various experiences, states and mental activities” (Vogel, 1993, p. 881).

As far as the matter of inner sense is concerned, H. J. Paton (1936) suggests that the immediate content of inner sense corresponds to feelings, desires and the stream of ideas. I deem correct G. Mohr’s interpretation that “concerning to different versions of the «object» of inner sense it follows that not the I but my thoughts and representations (my inner state) are the object of inner sense”[6] (Mohr, 1991, p. 71). In a similar manner, Patricia Kitcher holds that this matter is constituted by thoughts, perceptions and the temporal succession of thoughts and perceptions (Kitcher, 2016). According to Corey Dyck (2016), there are two objects of inner sense, namely thoughts (the mental states) and the subject itself. Aquila (1983) collects a large number of things that can be regarded as the matter of inner sense, by saying that the “domain” or the “manifold” of inner sense can be constituted by: the mind, its inner state, inner determinations, modifications, alterations, thoughts, feeling, inclination, decision, representations, will, thoughts and thinking.

Roughly speaking, if we admit that we intuit our inner states through inner sense (1998, CPR A38/B55) and ‘feelings’ (Gefühle) are nothing but a state of the mind, then we may say that feelings are an object of inner sense (1900, AA 7:231; 20:208, 230). However, one might think that feelings cannot be an object of inner sense because Kant claims that:

Everything in our cognition that belongs to intuition (with the exception, therefore, of the feeling of pleasure and displeasure and the will, which are not cognitions at all) contains nothing but mere relations, of places in one intuition (extension), alteration of places (motion), and laws in accordance with which this alteration is determined (moving forces) […] the representations of outer sense make up the proper material with which we occupy our mind. (1998, CPR A49/B66-7)

At first blush, this passage suggests that feelings and the will cannot reach the status of intuition and, therefore, they cannot be object of inner sense. This has led some commentators to downplay inner sense, suggesting that external objects alone are the object of inner sense (Wolff, 1963; Collins, 1999; Longuenesse, 2006). Against this reading, I hold that feelings are excluded from inner sense in that passage, because Kant is concerned in the CPR with the cognition of outer objects rather than with the cognition of an immediate I-intuition in inner sense. In other words, Kant restricts the material of inner sense to external objects, because his “interests are primarily epistemological, and his account of our sensible faculties largely reflects that focus” (Valaris, 2008, p. 2)[7]. By contrast, S. Yalcin (2002) seems to hold a more flexible position, for he admits both an “official view”, according to which the manifold of inner sense is derived from outer sense and an “alternative view”, according to which inner sense has its own manifold constituted by the mind itself and its inner states. Accordingly, I think that inner sense per se should not be reduced to either an outer or inner manifold, but the sort of manifold depends on the object of cognition (ourselves or external objects).

Kant explicitly declares in CPR A357 that thoughts, consciousness, desires, etc., cannot be externally intuited, inasmuch as they belong to inner sense. In the same way, feelings, inclinations or decisions are not contained by outer sense and, then, their existence in our mind should be contained by inner sense (1998, CPR A358). Feelings belong to sensibility (1900, AA 7:200; 6:211 footnote) and they provide our consciousness with a source of empirical representations of ourselves, but they cannot reach the status of intuition (1900, AA 7:239-40; 20:206, 222; 5:189), for feelings of pleasure and displeasure are not related to objects but to the subject alone, and they are not nor can provide cognition about outer objects:

Any relation of representations, however, even that of sensations, can be objective (in which case it signifies what is real in an empirical representation); but not the relation to the feeling of pleasure and displeasure, by means of which nothing at all in the object is designated, but in which the subject feels itself as it is affected by the representation. (1900, AA 5:203-4)

Feelings of pleasure and displeasure are not, according to Kant, sensations but the effect of a sensation, for they are subjective representations that express only states of the subject by reference to the representation of an object that might or not exist in experience (1900, AA 6:211-2; 1998, CPR A29/B44). Even though feelings do not contribute to the cognition of external objects, they still contribute to a pragmatic self-knowledge, by informing us the sort of effect (pleasant or displeasing) produced by our sensation of objects. This view is endorsed by J. H. V. Kirchmann who believes that self-perception, by its own nature, can only convert existing states of the soul into a knowledge that would be composed by feelings, desires and other existing representations which are mixed in the human soul (Kirchmann, 1869). I further support H. J. Paton’s reading that feelings are the material of inner sense. As he puts it: “by inner sense we are immediately aware, not only of our feelings and desires, but also of the stream of ideas which, whatever else they are, are for Kant modifications or states of our minds” (Paton, 1936, pp. 99-100). In a similar manner, N. Kemp Smith holds that the content of inner sense is double:

On the one hand we have feelings, desires volitions, that is, states of the mind in the strict sense, subjective non-spatial existences. On the other we have sensations, perceptions, images, concepts, in a word, representations (Vorstellungen) of every possible type. These latter all refer to the external world in space. (Kemp Smith, 2003, p. 293)

Kant holds that feelings are the effect of sensation on states of the mind, so that they force the subject to leave a particular state is disagreeable to it or to remain in a specific state, if it is agreeable (1900, AA 7:230-1; 1998, CPR A29/B44). The acts of leaving or remaining in one state depends on the change of sensation that occurs in time, for it entails a temporal sequence in the subject’s thoughts and in the consciousness of such change:

We are led along irresistibly in the stream of time and in the change of sensations connected with it. Now even if leaving one point of time and entering another is one and the same act (of change), there is still a temporal sequence in our thought and in the consciousness of this change; in conformity with the relation of cause and effect. (1900, AA 7:231)

Accordingly, the occurrence of change in our feelings and the act of being conscious of such change through inner sense are equally conditioned to time. For all our representations are related as one after other in conformity with time’s dimension (1998, CPR A31/B47). Life turns out to be a set of continuous opposite states in time, for “enjoyment is the feeling of promotion of life; pain is that of a hindrance of life. But (animal) life, as physicians also have already noted, is a continuous play of the antagonism of both” (1900, AA 7:231). Of course, all changes of the mind’s states are determined by a succession of tensed points (instants) of time, in which they are present, past or future events (Pacheco Acosta, 2018). In fact, since the human being exists in the stream of time, wherein the present becomes past while future becomes present now, each state of its mind, no matter whether it is pleasant or not, will be replaced by an indeterminate different state. Therefore, the correct combination between pain and enjoyment produces health in human beings. That is, health does not consist in a continuously felt wellbeing but in a set of intermittent agreeable feelings (1900, AA 7:231). I am against those commentators, who exclude prematurely feelings from inner sense without noticing that Kant is concerned in that passage with the knowledge of external objects rather than with the self and its inner states (Allison, 2004; Yalcin, 2002; Melnick, 2009; Schmitz, 2013).

Moreover, Kant holds in Anthropology: “the latter [inner sense], as a mere faculty of perception (of empirical intuition), is to be thought of differently than the feeling of pleasure and pain” (1900, AA 7:153). This passage should be carefully interpreted, since Kant is not denying that feelings are object of inner sense, he is rather pointing out that inner sense cannot be equated with feelings. This is natural because inner sense is fundamentally a receptive faculty of cognition that cannot produce spontaneously representations, but it is subject to the understanding. Feelings of pleasure and displeasure are nothing but particular modifications of the mind.

Furthermore, Kant holds that if a determination of the feeling of pleasure or displeasure is regarded as sensation, the latter “is related solely to the subject, and does not serve for any cognition at all, not even that by which the subject cognizes [emphasis added] itself” (1900, AA 5:206). However, this statement is nuanced by Kant himself, who does not exclude feelings from inner sense but, rather, wants to differentiate sensation from feelings by appealing to a subjectiveobjective criterion. According to this criterion, feelings (unlike sensation) are merely subjective and cannot be objective representations of an object. As a result, the idea that “the subject cannot cognize itself” would mean that the subject represents itself through feelings but these representations could not reach the status of “cognition” of its own self, because these cannot constitute objective representations of itself. That does not mean, however, that we cannot reach an anthropological cognition of ourselves grounded on subjective representations.

Inner experience contains linked-in-time representations of our states, which we are consciousness of, through empirical apperception (1900, AA 7:142). Inner and outer experiences rely on inner and outer sense respectively, which provide us with empirical representations of ourselves and of external objects. I am in agreement with A. Cohen, who suggests that “through one’s inner experience, one can observe the play of motives, inclinations, desires, and intentions, and derive from it empirical knowledge of oneself” (Cohen, 2009, p. 52). It is noteworthy to mention that Kant does not use the expression ‘self-feeling’ (Selbstgefühl) in his theory of inner sense and the use of this term is very fragmentary and out of discussion (1900, AA 15:58, 689, 725). This despite the fact that, as Thiel (1997) notices, the term of “Selbstgefühl” was present in 1770s discussions of inner sense and it was probably introduced into philosophical terminology by J. B. Basedow in 1764, taken up from Basedow afterwards by J. G. Feder.

According to Kant, an empirical intuition is a representation that “is related to the object through sensation” (1998, CPR A19-20/B34), while sensation is “a perception [emphasis added] that refers to the subject as a modification of its state” (1998, CPR A320/B376). It implies that a sensation (e.g. color, sound, sharpness, etc.) is not yet intuition, nor an a priori condition of our knowledge of objects, nor an intrinsic property of objects in themselves. Sensation is rather a subjective representation that belongs to the particular constitution of sense in the subject, that is, a modification of our state (1998, CPR A28-9/B44). Kant further divides ‘senses of physical sensation’ (die Sinne Körperempfindung) into vital sensation and organic sensation:

Sensations of warm and cold, even those that are aroused by the mind (e.g., by quickly rising hope or fear), belong to vital sensation. The shudder that seizes the human being himself at the representation of the sublime, and the horror, with which nurses’ tales drive children to bed late at night, belong to vital sensation. (1900, AA 7:154)

Since sensation is a particular content of inner sense, these two kinds of sensations would belong to the content of inner sense. Inner sense, nonetheless, contains not only sensations but also intuitions and perceptions, i.e. empirical representations accompanied by our consciousness (1998, CPR B155-6, 1900, AA 7:144). It is my contention that these representations do not arise through a purely “internalist” activity of the subject regardless of the influence of outer sense but these arise in social intercourse. That is to say, the play of ideas that exist in inner sense, without any reference to outer sense, should not be considered as ‘experiential knowledge’ (Erfahrungserkenntnis) but as fiction:

The tendency to retire into oneself, together with the resulting illusions of inner sense, can only be set right when the human being is led back into the external world and by means of this to the order of things present to the outer senses. (1900, AA 7: 162)

Accordingly, the content of inner experience which is concerned with ourselves does not consist in a set of fictitious, nor a priori representations. This rather relies on the material derived from sensibility. However, human beings can be aware not only of their sensations and feelings, but also of sensationless representations that compel them to do any activity: “even if no positive pain stimulates us to activity, if necessary a negative one, boredom, will often affect us in such a manner that we feel driven to do something harmful to ourselves rather than nothing at all” (1900, AA 7:232-3). Indeed, the human being tends to leave the state of boredom, insofar as it produces a fearful oppressive difficulty in its interior. As far as the human being attends to the relation between its life and time, it undergoes the oppressive and frightening arduousness of boredom. In my view, Kant is here referring to a sort of “phenomenal” time, namely, time in the first-person experience.

In the first-person experience, the human being at times wishes to jump from one moment to another in order to avoid unpleasant experiences: “this pressure or impulse to leave every point of time we are in and pass over into the following one is accelerating and can grow until a man makes the resolution to end his life” (1900, AA 7:233). Indeed, the human being avoids to perceive ‘the empty of sensations’ (Leere an Empfindungen) in its existence, which produces a presentiment of a slow death, which is taken for more painful even than death. This is why things that shorten time are regarded as enjoyments, and the quicker the human being makes the time pass, the more this will feel refreshed (1900, AA 7:233-4). Thus, despite Kant rejects, from a theoretical perspective, any empirical intuition of time (1998, CPR A176/B219, A166/B207, A172/B214, A182-3/B225- 6), he reflects, from an anthropological perspective, on the way in which the human being experiences time.

5. Conclusion

Kant’s theory of inner experience contains an empirical doctrine that describes the way in which we become aware of representations derived from our own inner experience. In this doctrine we deal with concepts like empirical apperception, intuitive self-consciousness, the I of apprehension, the empirical consciousness of one’s existence. These concepts emerge from experience and are employed to provide an anthropological or psychological description of the mental phenomena in humans. However, this theory of inner experience integrates also an a priori doctrine that explains how our experience of objects is possible. In this doctrine we deal with concepts like pure or transcendental apperception, discursive self consciousness, the I of reflection, intellectual consciousness of one’s existence. These concepts are necessary conditions for the possibility of experience, because they play a logical role in the possibility of human cognition. That is to say, these describe the way in which the manifold of intuition is connected by the categories of the understanding as well as the synthetic unity of the understanding and its domain over all our representations.

Referencias

Allison, H. E. (2004). Kant’s transcendental idealism. Yale University Press.

Aquila, R. E. (1983). Representational Mind: A Study of Kant’s Theory of Knowledge. Indiana University Press.

Brook, A. (1994). Kant and the mind. Cambridge University Press.

Carl, W. (1992). Die Transzendentale Deduktion der Kategorien in der ersten Auflage der Kritik der reinen Vernunft. Vittorio Klosterman.

Carruthers, P. (2011). The Opacity of Mind: An Integrative Theory of Self-Knowledge. Oxford University Press.

Cohen, A. A. (2009). Kant and the Human Sciences: Biology, Anthropology and History. Palgrave Macmillan.

Collins, A. W. (1999). Possible Experience: Understanding Kant’s Critique of Pure Reason. University of California Press.

Dessoir, M. (1924). Kant und die Psychologie. Kant-Studien, 29, 98-120.

Dyck, C. W. (2016). The Scope of Inner Sense: The Development of Kant’s Psychology in The Silent Decade. Con-Textos Kantianos: International Journal of Philosophy, (3), 326-344.

Foucault, M. (2008). Introduction to Kant’s Anthropology. (R. Nigro & K. Briggs, trans.). Semiotext.

Howell, R. (2001). Kant, the ‘I Think’, and Self-Awareness. In P. Cicovacki (Ed.), Kant’s Legacy: Essays in Honor of Lewis White Beck (pp.117-152). University Rochester Press.

Hume, D. (1739/21978). A Treatise of Human Nature. Selby-Bigge & P. H. Nidditch (Eds.). Oxford University Press.

Kant, I. (1900ff.). Kant‘s gesammelte Schriften, Akademie Ausgabe. In Königlich Preussischen Akademie der Wissenschaften (Ed.). G. Reimer [later W. de Gruyter].

Kant, I. ([1781/1787] 1998). Critique of pure reason. In P. Guyer., & A. W. Wood (Eds.). Cambridge University Press.

Kemp Smith. N. (2003). A commentary to Kant’s “Critique of Pure Reason” (3rd ed.). Palgrave Macmillan.

Kirchmann, J. H. (1869). Erläuterungen zu Kant’s Anthropologie in pragmatischer Hinsicht. Verlag von L. Heimann.

Kitcher, P. (2016). What Can Humans Cognize about the Self from Experience? Comments on Corey Dyck’s “The Development of Kant’s Psychology during the 1770’s”. Con-Textos Kantianos: International Journal of Philosophy, 1(3), 345-352.

Longuenesse, B. (2006). Self-consciousness and consciousness of one’s own body: Variations on a Kantian theme. Philosophical Topics, 34(8), 1–2, 283-309.

Lycan, W. (1996). Consciousnes and experience. MIT Press.

Makkreel, R. (2014). Self-Cognition and Self-Assessment in Kant’s Lectures on Anthropology. In A. Cohen (Ed.) Kant’s Lectures on Anthropology (pp. 18-37). Cambridge University Press.

Marshall, C. (2010). Kant’s Metaphysics of the Self. Philosophers’ Imprint, 10(8), 1-21.

Melnick, A. (2009). Kant’s theory of the self. Routledge.

Mohr, G. (1991). Das sinnliche Ich: Innerer Sinn und Bewußtsein bei Kant (Vol. 81). Königshausen & Neumann.

Monzel, A. (1920). Kants Lehre vom inneren Sinn und der Zeitbegriff im Duisburg’schen Nachlaß. Kant-Studien, 25 (1), 427-435.

Pacheco Acosta, H. (2018). Theoretical assumptions in Kant’s theory of time. Sincronía, (74), 29-58.

Pappe, H. O. (1961). On Philosophical Anthropology. Australasian Journal of Philosophy, 39 (1), 47-64.

Paton, H. J. (1936). Kant’s Metaphysic of Experience: A Commentary on the first half of the Kritik der reinen Vernunft (Vol 1). Allen & Unwin.

Picciuto, V., & Carruthers, P. (2014). Inner sense. In D. Stokes, M. Matthen, S. Biggs (Eds.), Perception and Its Modalities (pp. 277-296). Oxford University Press.

Prauss, G. (1971). Erscheinung bey Kant: ein Problem der “Kritik der reinen Venunft”. De Gruyter.

Roche, M. (2013). A Difficulty for Testing the Inner Sense Theory of Introspection. Philosophy of Science, 80(5), 1019-1030.

Rockmore, T. (2012). Kant on Unconscious Mental Activity. In P. Giordanetti, R. Pozzo & M. Sgarbi (Eds.), Kant’s Philosophy of the Unconscious (pp. 305-326). De Gruyter.

Rukgaber, M. S. (2009). The Key to Transcendental Philosophy: Space, Time and the Body. Kant-Studien, 100(2), 166-186.

Schmitz, F. (2013). On Kant’s Conception of Inner Sense: Self-Affection by the Understanding. European Journal of Philosophy, 23(4), 1044-1063.

Shoemaker, S. (1994). Self-Knowledge and “Inner Sense” Lecture I: The Object Perception Model. Philosophy and Phenomenological Research, 54(2).

Schulting, D. (2015). Transcendental Apperception and Consciousness in Kant’s Lectures on Metaphysics. In R. R. Clewis (Ed.), Reading Kant’s Lectures (pp. 89-114). De Gruyter.

Strawson, P. (1966). The Bounds of Sense: An Essay on Kant’s Critique of Pure Reason. Methuen & Co. Ltd.

Strawson, P. (1997). Entity and Identity: And Other Essays. Oxford University Press.

Sturm, T. (2017). Reines und empirisches Selbstbewusstsein in Kants Anthropologie: Das “Ich” und die rationale Charakterentwicklung. In G. Motta & U. Thiel (Ed.), Immanuel Kant: Die Einheit des Bewusstseins (pp. 195-220). De Gruyter.

Thiel, U. (1997). Varieties of inner sense. Two pre-Kantian theories. Archiv für Geschichte der Philosophie, 79(1), 58-79.

Valaris, M. (2008). Inner sense, self-affection, and temporal consciousness in Kant’s critique of pure reason. Philosophers’ Imprint, 8(4), 1-18.

Vogel, J. (1993). The Problem of Self-Knowledge in Kant’s “Refutation of Idealism”: Two Recent Views. Philosophy and Phenomenological Research, 53 (4), 875-887.

Wolff, R. P. (1963). Kant’s Theory of Mental Activity. Harvard University Press.

Yalcin, S. (2002). The Problem of Inner Sense in Kant’s First Critique. Journal of Academic Studies, 4(15), 181-190.

Notas

Notas de autor