Escribir el título del artículo

CON-TEXTOS KANTIANOS.

International Journal of Philosophy

N.o 6, Diciembre 2017, pp. 18-41

ISSN: 2386-7655

Doi: 10.5281/zenodo.1092766


Analytic Kantianism:

Sellars and McDowell on Sensory Consciousness


Kantismo analítico: Sellars y McDowell sobre la conciencia sensorial

JOHANNES HAAG

Universität Potsdam, Germany


Abstract


Wilfrid Sellars and John McDowell can both be read as proponents of Analytic Kantianism. However, their accounts differ in important detail. In particular, McDowell has criticized Sellars’s account of sensory consciousness in a number of papers (most notably in LFI and SC), both as a reading of Kant and on its systematic merits. The present paper offers a detailed analysis of this criticism and a defense of Sellars’s position against the background of a methodology of transcendental philosophy.


Keywords


Kant, Sellars, McDowell, Transcendental Philosophy, perception, intuition, judgment


  1. Analytic Kantianism & Transcendental Philosophy


    ‘Analytic Kantianism’ is not a well-defined term. It is mostly used very loosely to group philosophers who engage with Kant’s writings in a systematic, philosophical spirit and who themselves have a background in Analytical Philosophy, broadly conceived. If we look back in the history of discussions of Kant in Analytic Philosophy, Peter Strawson and Wilfrid Sellars probably first come to mind. Any list of important contemporary authors would have to include Robert Brandom, Michael Friedman, Hannah Ginsborg, Beatrice Longuenesse, Barry Stroud – and certainly John McDowell.



    [Recibido: 10 de octubre de 2017

    Aceptado: 20 de octubre de 2017]

    Analytic Kantianism



    But can Analytic Kantianism be described in an at least somewhat less loose and superficial way? I think it is possible both with respect to its subjects and the specific form in which they are addressed – at least for the field of theoretical philosophy I will concentrate on in what follows. John McDowell, whom I take to be one of the leading contemporary representatives of Analytic Kantianism, in one place writes:

    “If we understood Kant’s theoretical philosophy, we would understand how to think about the limits of intelligibility – the bounds of sense, in one interpretation of P.F. Strawson’s intentionally ambiguous title. That would put us within reach of the insight only glimpsed, I think, by Kant himself that those limits are not well-conceived as a boundary, enclosing a territory by leaving other territory outside it. But we can approach that connection with the theme of boundaries and limits only by dealing with the details of the first Critique, and some of that is all I will be doing here.” (SC1 108).

    In this quote, we find a number of features common to what I take to be characteristic for philosophers that could be subsumed under the label ‘Analytic Kantianism’: Most prominently, maybe, there is the vocalization of a deep reverence for Kantian thought, a reverence that Sellars once famously expressed by talking about philosophy being on a “slow climb ‘back to Kant’ which is still underway” (SM 29). Then we find the theme – a philosophical topos one might call it, by no means restricted to Kant – of a great philosopher apparently not being able to understand the depth and dimensions of his or her own thought in the way we today can understand his or her thought: As dwarfs on the shoulders of giants we ca see farther.

    But most importantly, there is the emphasis of the subject of ‘limits of intelligibility’, and with it, we might add, the nature of intelligibility itself that we supposedly are in a better position to understand if we engage in the understanding of Kantian philosophy. It is the latter subject, I would like to suggest, that gives us a grip on a more substantial characterization of Analytic Kantianism.

    What does this subject of intelligibility and its limits and boundaries consist in? Something’s being intelligible always, it seems, implies that it is intelligible as something of one kind or another for some kind of intelligent or rational being. Consequently, something is intelligible if and only if a rational being can intentionally refer to it as something.

    The topic of ‘limits and boundaries’ of intelligibility is thus closely connected to, and indeed: turns out to be only another perspective on, the subject of the possibility of intentional reference and the limits and boundaries of such a reference for rational beings. ‘To be intelligible’ just means ‘to be the potential object of intentional reference by beings of a certain kind’. Thinking about intelligibility and its limits and boundaries, consequently, means thinking about intentional reference and its limits and boundaries. I


    1 For a list of the abbreviations used for the works of Immanuel Kant, John McDowell and Wilfrid Sellars cf. the end of this paper.

    CON-TEXTOS KANTIANOS

    International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41 19

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766


    Johannes Haag


    would like to suggest that this topic of intentional reference and its limits is indeed characteristic for the thought of the authors that can be classified as exponents of Analytic Kantianism.

    However, it should be obvious that not any arbitrary way of asking (and answering) this question should make a philosopher an exponent of this philosophical family. More, certainly, is required to distinguish a way of analytic engagement with questions of intentional reference (even if they include its limits and boundaries) as a specific Kantian way of engagement. Let me, in a first step, try to specify the kind of questions which do make the inquiry a specific Kantian inquiry.

    It seems clear that in this connection the limits and boundaries of an intelligibility for beings that are like us in the relevant way is of special (if not of the only) interest for us. Ultimately, what we want to understand first and foremost is what is intelligible for us – and how that can be the case. But this question can be asked in very different ways, as Kant reminds us when he distinguishes questions of fact from questions of justification (cf. CPR A84/B116).

    What Kant understood was that a pure, even true description of an epistemic process could never amount to a justification of the epistemic ends of the process in question. While questions of fact are concerned with the description of epistemic processes, questions of justification are concerned with the normative nature of the epistemic purport of our representative acts: Justification is, as Sellars often points out, “a higher- order thinking” (Sellars, SK 325/6). In turning from the question “What is intelligible?” to the question “Why are we justified in believing something to be intelligible?” we thus leave the object- level of matter-of-factual truth and ascend to the meta-level where we reflect upon the question what makes a truth the truth it is – what accounts for its epistemic purport. Analytic Kantianism, in first approximation, is concerned with questions of the later kind.

    But, again, not any question about the epistemic purport of intentional reference makes the questioner an exponent of Analytic Kantianism. More, still, is needed to justify this classification. Fortunately this difference is not the only difference in a possible understanding of the question of what the boundaries of intelligibility consist in for beings who are like us in the relevant way. At least two further specifications are possible: One concerning the claim of exclusiveness of any satisfying answer to the question, the other concerning the level of generality or abstraction of the question itself.

    Let me first turn to the claim of exclusiveness: Questions concerning the objective purport of intentional reference can be (and often are) answered by indicating how we factually justify the claim to successful intentional reference. And these factual claims are often amenable to interesting philosophical answers as well – answers that certainly can shed some light on the issues in question. Those answers, however, will be restricted to the (albeit highly abstract) description of the facts in question or rather the factual causes and


    CON-TEXTOS KANTIANOS

    20 International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766

    Analytic Kantianism



    reasons for it. But this kind of pure description is not that what a question in truly Kantian spirit is aiming at. They are asked as questions after the conditions that need to be fulfilled in general so that we could possibly justify the claim to successful reference. The modality that is involved proves to be decisive: This way of putting the question indicates a lack of conceptually possible alternatives that is characteristic of the philosophical level of abstraction at which it is properly located. A satisfying answer to questions of that kind has to explain, why something has to be the case and cannot be otherwise. Philosophically substantial claims that are this demanding are characteristic for what we following Kant we call Transcendental Philosophy. The proper methodological framework for questions of this kind – indeed the proper methodological framework for Analytic Kantianism in general – therefore has to be the framework of Transcendental Philosophy.

    What amounts to Transcendental Philosophy, of course, is a matter of dispute. It has often been connected to the question of the possibility of transcendental arguments. I would instead suggest to sidestep this hotly contested issue and give an alternative characterization of the issues involved.2 In order to do this in a first step it has to be specified what is meant by the answer lacking alternatives. It certainly cannot mean that the answer is somehow logically necessary. That this is not the case is shown by the fact that we ordinarily can think alternative answers without committing ourselves to any inconsistencies. However, those merely logical alternatives are no alternatives for us, i.e. they aren’t alternatives that can be given a determined content by us.

    This determination of content always has to be a determination of thought with respect to its object: as soon as the possibility of this kind of determination is missing, the theory in question only gives us a purely formal analysis of thought. But this is not the kind of answer we expect where we are concerned with the properties of our intentional reference to intelligible objects. For, questions aiming at these properties are questions after the intentional reference to a world of which we conceive ourselves to be a part, i.e. the intentional reference to a certain reality and ourselves as a part of this very reality. The analysis of this reference necessarily has to be neglected in a purely formal analysis of the sort envisaged.

    What the alternatives are that can be thought with a determined content is dependent on the factual limits and boundaries that constrain the space of possibilities of our intentional reference to the world. The starting point of the requested thinking of intentional reference therefore has to be the reflexive reference to our own intentionality: There simply is no alternative starting point for the investigation of the conditions of possibility of intentional reference. In this sense our actual presuppositions with respect to this reference are a necessary element of the analysis of its conditions of possibility. They are the starting point



    2 Parts of the following characterization of Transcendental Philosophy are taken from my “Personhood, Bodily Self-Ascription, and Resurrection: A Kantian Approach” (Haag 2010, p. 130/1). More detailed discussions of the subject can be found in Haag, J. 2007, ch. 1 and Haag 2012.

    CON-TEXTOS KANTIANOS

    International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41 21

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766


    Johannes Haag


    of a reflection on which of these actual presuppositions really are necessary in the sense elucidated.

    However, unlike the analysis of our actual or factual employment of concepts, i.e. classical analysis of concepts, the analysis of the conditions of possibility implies the impossibility of alternatives to those conditions under the so descriptively specified presuppositions.

    The question of intentional reference and its limits and boundaries, understood as a question of Analytic Kantianism, consequently needs to be concerned with necessities that result from the analysis of concepts we have to use, if we can use concepts at all – i.e. if we can intentionally refer to anything at all. Accordingly the answering of those questions is to be conceived of as an analysis of the conditions of possibility of our way of ascription in the transcendental-philosophic sense: It is concerned with the most general conditions of the intentional relation to a world of which we conceive ourselves to be a part. The results of such a transcendental analysis are only justified if they are – under the preconditions thus specified – without an alternative that can be thought with a determined content. I would suggest it to be a defining feature of Analytic Kantianism proper that it shapes the question of intentional reference to intelligible objects as a question that aims at the conditions of the possibility of this kind of reference.


  2. Levels of Abstraction


    I already hinted at yet a further differentiation that is necessary to fully understand the question of intentional reference as a question of Analytic Kantianism – this time not to distinguish Analytic Kantianism from alternative positions, but to differentiate questions within Analytic Kantianism and thereby putting into perspective the scope of the question Analytic Kantians are asking.

    In order to further differentiate our understanding of the question for the intentional reference to intelligible objects, we have to pay attention to the fact that the phrase ‚beings like us in the relevant way‘ is a placeholder that is far from philosophically innocent. It requires (and allows for) distinctions with respect to the ways relevant to certain features of intelligibility and intentional reference: Features that can and should be ordered according to the degree of abstraction from our own human ways of understanding – an abstraction that starts with the limits of intelligibility for human beings and reaches its own boundary only when it tries to abstract even from those features that are part of what is needed for every possible epistemic subject in order to generate knowledge of a world to which itself belongs. In Sellars’s words: it is thus as „to rule out the possibility that there could be empirical knowledge not implicitly of the form ‘such and such a state of affairs belongs to a coherent system of states of affairs of which my perceptual experience is a part’…” (KTE 635).



    CON-TEXTOS KANTIANOS

    22 International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766

    Analytic Kantianism


    An epistemic subject of this kind needs to be finite – otherwise it could not be a (proper) part of the world it wants to generate knowledge of. Furthermore, it needs to be rational – otherwise it could not be interested in intelligibility of something in the first place. Furthermore, for Kant any finite rational being, not being able to spontaneously generating the objects of its knowledge, needs to be receptively given some material in order to subsume it under concepts. In other words, beings like us in the relevant way cannot cognize intuitively, but only discursively.3

    We consequently could describe the most abstract level of philosophical reflection on the limits of intelligibility as the level where we reflect on the limits of intelligibility for every possible finite rational, and hence discursive being that conceives of itself as part of the world (broadly conceived) it tries to understand. The most concrete level would be a level at which we were concerned with the limit of intelligibility for specifically human ways of intentionally referring to objects – a level that, for instance, clearly involves the specifically human forms of intuitive knowledge, i.e. space and time.

    In fact, Kant himself ties what I referred to as different levels of abstraction to the philosophical problem of intentionality as it occurs in transcendental philosophy. The differentiation in question is focussed by Kant as the question of what we do and of what we do not justifiedly presuppose with respect to the objects our intentional representations appear to be referring to. The close connection of this question to the famous question of a letter to Marcus Herz in 1772 concerning the relation between representations and their objects is obvious:

    “What is the ground of the relation of that in us which we call ‘representation’ to the object?” (Kant AA 10:130)

    In order to answer this question, Kant insists that transcendental philosophy is “not concerned with objects at all” (Kant Metaphysik Mongrovius AA 29:756). Its subject- matter, instead, is the possibility of epistemic or, more general, intentional reference to intelligible objects about whose existence nothing may be presupposed in advance. This is the only way, Kant insists, we can work out the conditions of the possibility of epistemic or intentional reference to objects. One could, therefore, describe the ideal aim of transcendental philosophy as making explicit the presuppositions of our thinking or our discourse about objects.

    There are questions on the possibility of reference even at the highest level of abstraction. Transcendental logic, being concerned with intentional reference to intelligible objects begins its reflection on a level of abstraction at which nothing whatsoever is presupposed about those putative objects of reference.


    3 For the important difference between discursive and intuitive understanding cf. Förster 2012, ch. 6 und Haag 2014.

    CON-TEXTOS KANTIANOS

    International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41 23

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766


    Johannes Haag


    Kant marks this unprecedented level of abstraction in thinking about reference to objects by introducing a specific technical term, the concept of a ‘Gegenstand überhaupt’. (The natural English translation ‘object in general’ has the serious drawback that it invites confusion with the (equally technical) Kantian term ‘Objekt überhaupt’.) The term ‘Gegenstand überhaupt’ serves, as Eckart Förster writes, as an “accusative of this relation (of a thought) ... without already presupposing a specific object as actual or even as merely possible.” (Förster 2012, 4/5.)

    But transcendental philosophy must not stop its investigation at this most abstract level: it can only succeed by lifting this abstraction step by step and thus progressing to the insights that can properly labelled to be transcendental philosophical – given that insights of this kind can be achieved at all. It can treat everything which has been shown to be a condition of the possibility of reference to objects at this level as explicitly presupposed at the next (lower) level of abstraction. Thus, after discerning the conditions of possibility of reference to ‘Gegenstände überhaupt’ it has to descend to the next level of abstraction internal to the transcendental hierarchy of these levels: that is, to the level at which the transcendental philosopher discusses the conditions of possibility of ‘Objekte überhaupt’. And from there he finally descends to thinking about the conditions of possibility of objects of experience (Gegenstände der Erfahrung), that is, spatio-temporally located objects.

    Incidentally, these lower levels of abstraction turn out to coincide with the differentiation between the relevant classes of ‘beings like us’ adumbrated above. Thus, if it is possible to substantiate the above distinction with respect to the distinction of Objekte überhaupt and objects of experience just introduced, we will have made considerable progress.

    A very helpful description of the means we have at our hands at each level of abstraction that helps us along these lines can be found in Wilfrid Sellars’s writings on Kant: He discerns the different steps of abstraction by showing that each of these steps involves a different use of the Kantian categories broadly conceived. (I say ‘broadly conceived’ because, at the highest level, categories are, strictly speaking, not yet in the play as categories.) Here is the quote from Sellars:

    “The categories are in first instance simply identical with the forms of judgment ... . These forms of thought would be involved in thinking about any subject-matter from perceptual objects to metaphysics and mathematics.The so-called pure categories are these forms of thought specialized to thought about objects (matter-of-factual systems) in general. Such objects need not be spatio-temporal, as are the objects of human experience. The full- blooded categories ... are the pure categories specialized in their turn to thought about spatio-temporal objects.” (Sellars IKTE §44/5)

    The full-blooded categories are, as Sellars makes clear, the schematized categories which are the result of the Schematism of the pure concepts of the understanding. He therefore closes by adding:


    CON-TEXTOS KANTIANOS

    24 International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766

    Analytic Kantianism



    “The relation of the forms of thought to the pure categories is that of genera to species, as is the relation of the pure categories to schematized categories.” (Sellars IKTE §45.)

    Sellars is very clear about how the categories, at whichever level, are spared the fate of traditional ontological thinking – whose possibility, remember, it is, among other things, what is at stake. Categories are not treated as names for obscure abstract entities.

    “Medieval logicians began the process of reinterpreting the categories that culminated in Kant’s Critique, by recognizing that certain statements (thus ‘Man is a species’) which seem to be about queer entities in the world are actually statements that classify constituents of conceptual acts.” (Sellars KTE 641)

    This development culminated in a treatment of the categories as “grammatical classifications”:

    “And, of course, Kant’s categories are grammatical classifications. They classify the grammatical structures and functions of Mentalese. Thus the category of substance- attribute is the structure ‘S is P’, the form of subject-attribute judgment. The category of causality is the form ‘X implies Y’. The category of actuality is the form ‘that-p is true’. More accurately, the categories are these forms or functions specialized to thought about spatio-temporal object.” (Sellars IKTE §§ 39/40)

    What these distinctions, together with the distinctions concerning the respective accusative of the referential relation in question, give us is an outline of the internal structure of the justificatory or transcendental levels of abstraction: The internal structure is threefold, starting, at the highest level of abstraction, with theorizing about the conditions of possibility of thinking about ‚Gegenstände überhaupt‘ – objects in general writ large – in terms of the most general forms of judgement; turning at the next level to the possibility of our reference to objects in general, i.e. objects thought under the condition of an ‘intuition in general’ which does not have to be an intuition subject to our specific human forms of sensibility; and adding, at its last level, spatio-temporal structure, that is, the categories adapted to our specific human forms of intuition and thus scrutinizing the conditions of possibility of reference to spatio-temporal objects.

    This internal structure is characterizing the higher level in a just two-leveled hierarchy of levels of abstraction whose defining feature is whether it does or whether it does not presuppose the possibility of reference to objects.

    But we have yet to add an essential ingredient to our picture of levels of abstraction. We have to be more perspicuous about what we do when we think about the conditions of possibility of our reference to objects. Kant (according to the lecture-transcript Metaphysik Mongrovius) puts it in the following way:

    “In transcendental philosophy we consider not objects, but reason itself ... One could therefore also call transcendental philosophy transcendental logic ... Transcendental logic


    CON-TEXTOS KANTIANOS

    International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41 25

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766


    Johannes Haag


    abstracts from all this, it is a kind of self-knowledge (Selbst Erkenntniß).” (Kant AA 29:752, 756; quote Förster 2012, 4.)

    We already heard Sellars referring to the categories – at every of the aforementioned internal levels of abstraction – as classifying the structures and constituents of conceptual acts. What we therefore have to add is a distinction between grammatical classifications which do and classifications which do not add to the kind of self-knowledge Kant refers to in this remark. Otherwise we would have to include any old quality and every relation, since every quality and every relation may serve as characterizing a concept or object. And every concept is, on Kant’s and Sellars’s view, a rule for the classification of ideas.

    Which kinds of classification of conceptual acts, then, are the proper subject-matter of transcendental philosophy? It is exactly those kinds that are necessary for an understanding of the ‘ground of the relation of that in us which we call ‘representation’ to the object’. In other words, it is exactly those kinds that are the conditions of possibility of an intentional relation between us as thinkers and the world of intelligible objects, i.e. the conditions of the possibility of intentionality.


  3. Givenness, Guidance, and Sensory Consciousness


    With these methodological observations in place, let me turn to some of the central topics in connection with the question of intentionality as it is addressed in Analytic Kantianism. The common denominator of these topics is their connection with an account of the role of sensory consciousness in a Kantian conception of intentionality. I will unfold these topics by discussing the respective takes of Wilfrid Sellars and John McDowell, two of the foremost representatives of Analytic Kantianism, and relating them to each other. Kant’s original account of these topics will serve as the backdrop for this discussion.

    In Science and Metaphysics (1968) Sellars observes:

    The [manifold of intuitions; J.H.] has the interesting feature that its existence is postulated on general epistemological or, as Kant would say, transcendental grounds, after reflection on the concept of human knowledge as based on, though not constituted by, the impact of independent reality. (Sellars SM 9; my emphasis)

    This impact of an independent reality corresponds to the ‘guidedness’ of our perceptual content Sellars shows himself so impressed by in the opening pages of Science and Metaphysics. 4 This guidedness, for Sellars, is an ultimately phenomenological fact grounded in the passivity of our experience. Kant, throughout his critical writings, emphasizes this passivity with respect to the content of our experience.5 There has to be


    4 Cf. Sellars, SM 16. Robert Pippin discusses these remarks in his Kant’s Theory of Form (1982, 46-51).

    5 Cf. e.g. Kant, CPR A 50 / B 74; Kant, GMS 4:452; Kant, Anthropologie, 7:141.


    CON-TEXTOS KANTIANOS

    26 International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766

    Analytic Kantianism


    something that explains the basic phenomenological fact that we are passive with respect to the actual content of our experience.6

    This guidance is the joint effect of independent reality and the sense-impressions brought about by its impact – Sellars’s ‘sheer receptivity’7. The independent reality is the Kantian thing in itself, guiding us from without via the impressions of sheer receptivity. Only the latter are immediately accessible for the working of our spontaneity.

    Even this immediate contact with sense-impressions is, however, guidance from without the conceptual order in the sense that these impressions are not given as what they are in themselves, but for both Kant and Sellars are always synthesized by the synthesis of the imagination. (It is in this context that Sellars in his later writings introduces the concept of an image-model. Cf. below sec. 5.)

    This fact, in turn, connects the subject of guidance to another important philosophical subject in connection with our intentional reference to a world that exists independently of us: the repudiation of the Myth of the Given. This Myth in its “most basic form” (Sellars, FMPP, I § 44) consists in the following principle:

    „If a person is directly aware of an item which has categorical status C, then the person is aware of it as having categorical status C.“ (ibd.)

    And he adds:

    „To reject the Myth of the Given is to reject the idea that the categorial structure of the world – if it has such a structure – imposes itself on the mind as a seal imposes an image on melted wax.“ (ibd.)

    The rejection of the Myth therefore involves a rejection of every form of direct or immediate awareness of something with a certain categorial structure as having this very categorial structure unless one already developed a conceptual framework, which forms the background of this direct awareness.8

    The connection to the subject of guidance via the synthesis of the receptively given sense- impressions should be obvious. It can be summarized in the following question: In sensory consciousness, are we immediately aware of the receptively given sense-impressions guiding us as what they are in themselves? In answering this question, everything depends on how the imagination in its synthesizing activity transforms what is receptively given.

    McDowell chooses a completely different approach to sensory consciousness and declares himself an intentionalist in the treatment of sensation.9 Intentionalism concerning sensation


    6 Although Sellars thought himself in disagreement with Kant in this respect, I tried to show on another occasion that for both authors this guidance has to be strictly ‘from without’ the conceptual order. Cf. Haag 2014.

    7 Cf. for example Sellars, ibd.

    8 Cf. Rosenberg 2007, 285-289.

    9 Cf. McDowell, SC 119 n. 22.


    CON-TEXTOS KANTIANOS

    International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41 27

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766


    Johannes Haag


    is the position that sensations themselves are or can become intentional states. 10 For a Sellarsian, this is something like a philosophical fall from grace: In countless places Sellars criticizes this position which he considers to be one of the paradigmatic forms of the Myth of the Given.11

    I will not take issue with McDowell’s systematic position directly, but with his understanding of Kant as an intentionalist about sensation. What McDowell suggests as a reading of Kant, is a sophisticated attempt to bring together the two stems of knowledge in a way that does not picture their respective outputs as two “components” (McDowell, SC

    124) of experiences, but as the sensory output of a sensibility already “informed” (ibd.) by the understanding.

    “[T]he intentionality of intuitions is accounted for by the fact that in intuitions sensory consciousness itself is informed by the higher faculty. The thinkings that provide for the intentionality of perceptual cognitions are not guided by sensory consciousness, as it were from without. They are sensory consciousness, suitably informed.” (ibd. 119)

    This technical concept of informed sensory consciousness is important, but at the same time difficult to elucidate. As it is introduced it clearly is brought into play to provide an alternative to Sellars’s concept of guidance from without through sheer receptivity; but it seems that the two conceptions of informed sensory consciousness and guidance from without are not mutually exclusive. To show that will be part of the burden of my argument.

    What is meant by the concept of informed (or conceptually shaped 12 ) sensory consciousness? In a first approximation, sensory consciousness being informed must include not being neatly separable in an intentional and a sensory (or conceptual and sensible, or spontaneous and receptive) component, as McDowell insists.13

    It seems further safe to maintain that informing in this technical sense has to amount to more than simply the (new) arrangement of given sensory input by the understanding. Otherwise there would hardly be any difference in this respect between McDowell’s and Sellars’s reading of Kant: Sellars acknowledges that the sensory material is (re)structured by the synthetic activity of productive imagination. The latter is guided itself not only from without, but also from within by concepts which provide the recipes for this activity.14 In order to count as informed the sensory consciousness itself has to somehow become spontaneous through this interaction with the higher-faculty – otherwise we would be


    10 The cautious amendment “or can become” is meant to incorporate positions, which are prepared to admit that “sensibility alone does not yield cognition” (McDowell, SC 119), but insist that sensory states “informed by the higher faculty” (ibd.) can. McDowell subscribes to such a view.

    11 Cf.. e.g. Sellars, BBK, 46; Sellars, SRLG 335/6; Sellars, EPM 132-4, 154-6.

    12 Cf. McDowell, LFI 34.

    13 Cf. ibd., 124.

    14 Cf. Sellars, IKTE § 28. McDowell explicitly seems to grant as much in a reversal of his earlier criticism in the Woodbridge-Lectures. Cf. ibd. 114 f. I will discuss some of the details of this process and the concepts involved below.


    CON-TEXTOS KANTIANOS

    28 International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766

    Analytic Kantianism



    restricted to merely associatively connected representations, that do not amount to cognition.

    Informed sensory consciousness should not only be merely a formal (re)arrangement of the receptive input. Moreover, it seems necessary that there should not be an unformed sensory element that can be identified as present in the more sophisticated products of mental activity, be they image-model or the even intuition. Otherwise this element would not have been sufficiently transformed by the higher-faculty in the way just adumbrated.

    I would like to point out, though, that the concept of informed sensory consciousness does not seem to imply that a conceptual representation which directly refers to informed sensory consciousness, cannot be distinguished from the representations so informed. There seems to be some philosophical wriggling room here, that will prove important in what follows.

    Nevertheless, some of McDowell’s remarks indicate that he seems to take this implication largely for granted, for instance by writing that in his picture “what provides for an intuition, say, to belong to sensory consciousness is not apportioned to an item other than one whose characteristics provide for the intuition to be of an object, as in Sellars’s picture” (McDowell, SC 118/9).

    But intentionalism in the treatment of sensations does not need to amount to making intuitions themselves sensory items – although it seems McDowell has exactly this in mind. In the dimension of our thinking that concerns intuitive content it could still make sense to separate a concept of an item that is a conceptual tool for direct, demonstrative reference from an item that supplies intuitive thinking with informed sensory consciousness.

    Even if one (like McDowell) rules out an account along these lines, this leaves the possibility that there could still be a purely sensory component that feeds into our mental activity that cannot in any meaningful way be identified in synthesized experience, and yet might prove indispensable for different reasons. In his critical remarks on Sellars, McDowell seems to allow for at least the conceptual possibility of this sort of sensory component, notwithstanding his own rejection of any such sensory items as superfluous.

    According to the picture to be developed, Sellarsian sheer receptivity would operate ‘below the line’, thereby guiding the informing of sensory consciousness above the line from without in Sellars’s use of this term; while in a second step the results of this operation would be sufficiently transformed by spontaneity to justify the classification of the resulting sensory consciousness as informed.

    Notice that in this case, there ultimately will exist not one, but two different kinds of sensation in a complete account of sensory consciousness: The first kind of sensibility would belong completely to receptivity (‘below the line’), while the second would be permeated with spontaneity from the very beginning (‘above the line’).


    CON-TEXTOS KANTIANOS

    International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41 29

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766


    Johannes Haag


    This possibility of a concept of informed sensory consciousness compatible with a conception of a sensory level ‘below the line’15 that separates the receptivity of sense from the spontaneity of understanding, will prove of vital importance for assessing the true extent of disagreement between Sellars’s and McDowell’s Analytic Kantianism with respect to the role of sensory consciousness in intentionality.


  4. The very same function of unity?


    McDowell, in discussing Sellars’s conception of sensory consciousness in connection with his reading of Kant’s Critique of Pure Reason frequently comes back to a key-statement of the first chapter of Kant’s Transcendental Analytics of the Critique of Pure Reason, called “On the Clue to the Discovery of all Pure Concepts of the Understanding”:

    “The very same function that gives unity to the different representations in a judgment also gives unity to the mere synthesis of different representations in an intuition which, expressed generally, is called the pure concept of understanding.” (Kant CPR A 79 / B 104/5)

    This statement is the central sentence of the so-called ‘Metaphysical Deduction’ whose purpose it is to give us the pure concepts of the understanding. It is at the same time the first whole-hearted expression of the intimate connection of the logic of judgment with the logic of intuition – but, unfortunately, it is not one of Kant’s clearest expressions of this relationship. McDowell interprets this statement as follows.

    „We can recast the remark from the ‘Clue’ to say: the function that gives unity to the various representations in an ostensible seeing is the same as the function that gives unity to the mere synthesis of various representations in an intuition.” (LFI 31)

    In order to assess this interpretation it is important to first explain the usage of the technical term ‘ostensible seeing (OS)’. McDowell in this usage follows Sellars who uses it as a term for representations that purport to be instances of visual perceptions which, in the veridical case (unlike in the subjectively indistinguishable case of illusion or hallucination) they indeed are. OS are not the result of a spontaneous commitment of the subject but are involuntary reactions to external stimuli. That these reactions are involuntary does, however, not preclude them from being thoroughly conceptual:

    „Sellars shows us how to understand visual experiences as ostensible seeings, occurrences in a subject's life that ‘contain’ claims about an ostensibly visible region of objective



    15 To borrow an illuminating metaphor from McDowell’s Woodbridge-Lectures. Cf. McDowell, SPE 5. (I consequently do not share Jay Rosenberg’s criticism of this metaphor: Rosenberg thinks that „McDowell’s reification of Sellars’s references to ‚sheer receptivity’ [is] unfortunate“ (Rosenberg 2007, 279). The reason for this criticism seems to that sense-impressions below the line are the material for the conceptually guided synthetic construction of image-models above the line. My analysis of the role of sense-impressions in this construction will implicitly address this point. Cf. section 5.


    CON-TEXTOS KANTIANOS

    30 International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766

    Analytic Kantianism



    reality. That they ‘contain’ claims is the same fact as that they are conceptual occurrences, actualizations of conceptual capacities with a suitable ‘logical’ togetherness. In that respect they are like judgments. But they are unlike judgments in the way in which they ‘contain’ their claims. Judgments are free exercises of conceptual capacities [...]. But in an ostensible seeing whose content includes that of a given judgment, the same conceptual capacities are actualized [...] in a way that is ostensibly necessitated by the objective reality that is ostensibly seen. A visual experience is a case of being under the visual impression that things are thus-and-so in the ostensibly visible environment.” (McDowell, IR 44)

    OS have or contain the same content as the corresponding judgments that are not the result of an involuntary reaction to stimuli. Consequently, they truly are conceptual shapings of sensory consciousness (cf. LFI 33). That these shapings have to be of the same ‘logical togetherness’ is a nod to the fact that we have to exercise a number of different conceptual abilities in order to see something as e.g. a pink ice-cube: ascribing a color, a form, a material etc. (cf. SPE 1 f.). But the connection to the corresponding judgment, according to McDowell, is even closer: OS do have the same content in the same (i.e. propositional) structure. An OS is an (ostensible) seeing-that with direct object-reference (cf. LFI 32). Hence, OS are mental states with propositional content that – unlike judgements – are not the result of a decision. In an OS we see that something is the case without judging, i.e. committing ourselves, that it is. They are distinguished from the corresponding judgment only in that they are not affirmations. McDowell later (in AMG 269) admits that it is indeed hard to deny them the status of judgments. His official doctrine, however, from the Woodbridge-Lectures onwards is that OS are ceteris paribus authorizing the corresponding judgment without being identical to it.

    This makes plausible the replacement of ‚judgment’ with OS in his paraphrase of the statement from the ‘Clue’. After all, the connection Kant is after has to be particularly salient in the relation between intuitions and the corresponding judgments of perception. And since commitment seems to be irrelevant with respect to the logical unity Kant is interested in, it makes perfect sense to switch to OS, thus focusing on the question of conceptual structure.

    Intuiton, unlike OS, does have the task of presenting the object of experience ore perception itself. Intuitions qua immediate representations of an object are themselves not mere products of sensibility but require the involvement of the understanding that is able to guarantee in their synthesis the same kind of logical structure that characterizes judgments. McDowell affirmatively quotes Sellars’s calling intuitions ‘this-suches’. If an OS has the content that there is a pink ice-cube, the corresponding intuition is an intuition of this pink ice-cube over there, thanks to the involvement of the very same conceptual abilities in both cases:

    “If an ostensible seeing is a seeing, then the conceptual shaping of visual consciousness that constitutes it, those very conceptual capacities actualized in visual consciousness with


    CON-TEXTOS KANTIANOS

    International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41 31

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766


    Johannes Haag


    that very »logical« togetherness, constitute – looked at, as it were, from a different angle – an intuition: an immediate presentness of an object to sense. [...] This seeing that..., in describing which we explicitly place an expression for the concept in question in a predicative position, is the very same conceptual occurrence – an actualization of the same conceptual capacities with the same »logical« togetherness – as the intuition.” (LFI 33f.)

    This should not be misunderstood as implying that intuitions somehow ‚give’ us an object that the OS then refers to in a second step. McDowell’s view is the more radical one that intuitions and OS are, indeed, two sides of the same coin: “Visual intuitions of objects simply are seeings that..., looked at as it were from a different angle.” (LFI 34) It is in order to facilitate this that both need to have the same judgeable content (cf. LFI 35). “In fact, visual intuitions just are the actualizations of conceptual capacities, with the requisite togetherness, that constitute those ostensible seeings that are seeings.” (IR 45f.) This, for McDowell, in the Woodbridge-Lectures constitutes the core of the insight that Kant voiced in the key-passage from the ‘Clue’ about the ‘very same function that gives unity’ to an judgment and an intuition.

    It seems clear that McDowell’s interpretation of Kant offers a reading that is compatible with this statement. How, on the other hand, is a Sellarsian reading of this pivotal statement even possible given his conception of sensory consciousness sketched above? In order to make this plausible, we have to take a deeper look at Sellars’s theory of conceptually shaped sensory consciousness as it is developed in his concept of an image- model, in particular. I will introduce this important Sellarsian concept as it were through the lens of McDowell’s discussion of this concept. This will allow me in what follows to develop the opposing views of Sellars and McDowell in the form of a critical investigation of McDowell’s interpretation of Sellars’s Kantian conception of sensory consciousness.


  5. Image-Models and Intuition


    The concept of an image-model (IM) does not appear in Science and Metaphysiscs (based on his 1964/5 Locke-Lectures). It is officially introduced no earlier than in the late paper “The Role of Imagination in Kant’s Theory of Experience” (IKTE, 1978), though there is a number of remarks, starting from “Kant’s Theory if Experience” (KTE, 1967) 16 that foreshadow this decisive twist in Sellars’s conception of sensory consciousness.


    16 “This difficult doctrine requires that the logical powers of the concept cube involve not only the inferential powers characteristic of its role as predicate of full-fledged judgment, but also the powers involved in “constructing” or “drawing” determinate “this-cube”-representings in accordance with a rule, and knowing that this is what one is doing.” (KTE 643) Though some of the important elements of the IM-conception are already in place, it not yet the IM-terminology, of course, that Sellars is using here and, in particular, the relation of these ‚drawings’ to intuitions is not clarified in this context.

    CON-TEXTOS KANTIANOS

    32 International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766

    Analytic Kantianism



    Concluding from some of McDowell’s remarks in the Preface to Having the World in View (2009) IKTE seems to have contributed to a correction of his Sellars-interpretation that then is developed for the first time in his paper on “Sensory Consciousness in Kant and Sellars” (SC, 2008; reprinted in 2009; all references are to the later edition):

    “When I wrote [the Woodbridge-Lectures; JH], I thought Sellars’s picture included this informing of sensory consciousness by capacities that belong to the understanding, and that he added external constraint, by what he calls “sheer receptivity”, as a distinct further role of sensibility. I retract that reading in [SC; JH].“ (HWV vii)

    A reading along these lines would have made Sellars’s account of what happens ‚above the line’ that separates the intentional realm from the non-intentional very similar to McDowell’s own conception of sensory consciousness (and, of course, his interpretation of Kant’s theory), while it would mark the difference between the two accounts as a difference concerning mostly what happens ‘below the line’. Instead he now opts for an interpretation that, in fact, distances Sellars from his own conception in a crucial detail:

    “[M]y question had a false presupposition. Sellars comes close to Kant in saying experiences contain claims. [1] But all he can make of the idea is that experiences are composites, with claim-containing items accounting for their intentionality and sensations accounting for their sensory character. And this reflects his not arriving at what I take to be the authentically Kantian view. [2] Sellars does not envisage claim-containing occurrences that are themselves shapings of sensory consciousness.“ (SC 122; emphasis and numbering JH)

    The discussion in SC makes clear that responsible for this change was McDowell’s integration of the Sellarsian IMs in his picture of Sellars.17

    In the remaining parts of this paper I will be mainly concerned with showing that McDowell is correct in high-lighting the introduction of this concept and the importance of the conception of synthesis of imagination that Sellars develops in this context. However, I will argue that, nevertheless, McDowell misunderstands the function of this concept when he ascribes to Sellars (cf. remark [1] of the quote above) an understanding of experience as a composite of non-conceptual and conceptual elements. This precludes him from seeing the real affinity of Sellars’ position to his own.

    On the other hand, it will be seen that McDowell is, in fact, correct with respect to his claim that “Sellars does not envisage claim-containing occurrences that are themselves shapings of sensory consciousness” (SC 122; remark [2] above). This claim, however, if true, will turn out to point to a less dramatic difference between the two accounts – especially given the modification that McDowell’s own conception seems to undergo in his


    17 There is a brief footnote in the Woodbridge-Lectures where McDowell mentions IMs in passing. Vgl. LFI 26/7 n. 7.

    CON-TEXTOS KANTIANOS

    International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41 33

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766


    Johannes Haag


    later AMG where he himself seems to commit to the view that experiences do not contain claims.18

    The upshot of McDowell’s criticism of Sellars’s concept of IMs is this: “What the productive imagination generates is a unity involving both sensibility and understanding – not an amalgam, however intimately bound together, of components that belong severally to sensibility and understanding.“ (SC 124) What McDowell is describing as an ‚amalgam’ here is the result of Sellars’s division of the activity of the productive power of imagination in (a) the production of the OS’s this-suches and (b) the IMs as the descriptive core of perception.

    While McDowell acknowledges that IMs cannot be produced without the cooperation of the understanding that provides the ‘recipes’ necessary for their construction, he emphasizes that it is still a construction in sensibility: a construction according to conceptual recipes out of sensible material which, consequently, should indeed be seen as an ‘amalgam’ from sensory and conceptual components (cf. SC 117). This clearly is not conceptually informed sensory consciousness in McDowell’s sense.

    But is it an adequate representation of what Sellars’s IMs are? This clearly depends on the exact role of the sensory material in this process. Given that we must distinguish between the construction of IMs and the taking up of the items so construed into perceptual consciousness in an intuition, and given, furthermore, that the productive imagination is somehow responsible for both, there are indeed two distinguishable acts of this capacity in play in experience. However, it does certainly not go without saying that in the construction of IMs the imagination simply arranges the receptively given sense- impressions into structured aggregates of impressions which would belong to the conceptual order simply because they are thus ordered.

    Yet this seems to be what McDowell has in mind: He insists that according to Sellars it is solely due to the participation of the understanding that provides the ‘recipes’ for the construction in question, that the receptivity of sense-impressions gets conceptually enriched (cf. SC 114/5). The IMs, consequently, now are the elements in our perceptual engagement with the world that provide the guidance that proved so important for Sellars’s epistemology – a guidance that cannot be a guidance ‘from without’ anymore as McDowell notices, implicitly retracting his criticism of Sellars’s insistence on a guidance ‘from without’ (cf. SC 115 n. 14).

    This change in Sellars’s position, given the importance he gives to the conception of a guidedness ‘from without’, would indicate a dramatic change. But it strikes me as premature to ascribe it to him on the basis of the introduction of the concept of the IMs. It seems to rest on a wrong understanding on McDowell’s side of what IMs are and what is their task in Sellars’s account of sensory consciousness.


    18 Cf. below sec. 7


    34


    CON-TEXTOS KANTIANOS

    International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766

    Analytic Kantianism


    IMs, for Sellars, are not simply structured aggregates of sensory material. Sense impressions are modifications or states of a perceiving subject. In IMs they are now represented as properties of imagined three-dimensional models of objects, imagined from any given perspective of a perceiving subject (and, hence, essentially perspectival).19

    In being so represented they are already essentially modified. The pinkness of a pink sense-impression (i.e. of a mental state) must be very different from the pinkness of an imagined pink three-dimensional object. In the very act of arranging sense-impressions according to recipes the productive imagination is transforming the sense-impressions themselves and not just ordering them into patterns, as McDowell implies in remark [1].

    What is at stake here is what Sellars in another place calls “in a sense most difficult to analyze, a thinking in color about colored objects” (SK 305). The emphasis here is on ‘thinking’ – and the ‘thinking in ccolor’, of course, has to be supplemented by a ‘thinking in shape’ (and a ‘thinking in’ every other kind of sensibly given properties).20 Another way to make essentially the same point is that, unlike the sense-impressions of ‘sheer receptivity’, IMs are already subjected to a structuring synthesis guided by the (mathematical) categories of quantity and quality.

    The dynamical categories of relation and modality on the other hand come into the picture only when the essentially perspectival IMs are conceived of as objects of experience with causal and dispositional properties that we refer to in intuitions and the corresponding OS and judgments. It is, in other words, not the (essentially perspectival) IM itself that is the sensibly given object we refer to in perception, as McDowell seems to think (cf. SC 224). Instead, we refer to objects of experience whose sensible properties we conceptualize on the basis of IMs. IMs qua arrangements of categorially transposed sense-impressions contribute the aspect of sensible presence that is necessary for the demonstrative reference being truly perceptual or intuitive. But the reference to object of experience in addition to these sensible properties has to ascribe to them at least some causal and dispositional properties that make them objects of experience in the first place, i.e. empirical objects existing independently of their being perceived by us.

    In as much as McDowell’s criticism depends on their being two distinct elements in perceptual reference – a sensibly represented object and an intuitive reference to that very object – McDowell is correct. But this distinct sensible element, although it accounts for the sensible presence of an object in experience, is itself not an element that incorporates sense-impressions from ‘below the line’. In being subjected to the activity of productive imagination it is transformed into an element that clearly belongs ‘above the line’: Sense- impressions are conceptualized through and through as perceptible properties of these objects of experience.


    19 For an extensive discussion of IMs in Sellars’s theory cf. Haag 2007 ch. 7 and Haag 2013, 67-71.

    20 McDowell, though quoting this remark, largely ignores it and goes on to discuss IMs, “Sellars’s most sophisticated treatment of the relation” (SC 114), thereby missing the opportunity to do justice to the relevance of this remark for the proper understanding of IMs.

    CON-TEXTOS KANTIANOS

    International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41 35

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766


    Johannes Haag


    McDowell, for his part, thinks that the resulting conception introduces guidance ‘from without’ through the backdoor. And this cannot be right since the “thinkings that provide for the intentionality of perceptual cognitions are not guided by sensory consciousness, as it were from without”, but are “sensory consciousness, suitably informed” (SC 119).

    Somewhat ironically, Sellars could subscribe to this statement without submitting to McDowell’s position: First, he would insist that it is not sensory consciousness that guides us from without. This task would fall to the sense-impressions of ‘sheer receptivity’ that are never apperceived (cf. Sellars, SM 10). This has been his position for a long time and Sellars does not change it in IKTE. Second, the objects of experience we refer to in perception indeed are conceptually informed sensory representations. In a sense that certainly differs from what McDowell has in mind, these objects of experience supply our perceptual representations with the required intentionality by serving as objects of reference that are thought and experienced at the same time.


  6. Guidance and Phenomenalism


    Consequently, it is at least possible to read Sellars as subscribing to a conception of “sensibility ... that is informed by conceptual capacities in the experience of rational subjects“ (HWV, Preface vii). At the same time, sense-impressions and the concept of guidance ‘from without’ remains firmly in place. If this is right, McDowell’s critical question from the Woodbridge-Lectures – i.e. why does Sellars even need a level of ‘sheer receptivity’ and a guidance from without? – contrary to what McDowell thinks (cf. SC 122), would still be left unanswered. For McDowell, this additional layer is just superfluous.

    It might be helpful to address this deep point of disagreement in connection with the criticism expressed in McDowell’s second remark above: Sellars, he correctly insists,

    „does not envisage claim-containing occurrences that are themselves shapings of sensory consciousness“ (SC 114). For neither do this-suches contain sensible elements nor do IMs or objects of experience contain claims. And this is a very serious point for McDowell, since it not only makes Sellars’s account ultimately incompatible with his own conception that insists that “thinkings of a pink cube can include items that are sensory consciousness informed by the higher faculty” (SC 123); it furthermore is a consequence of a feature of the account of sensory consciousness in question that allows Sellars, but not himself to make sense of a critical point in Kant’s philosophy: a ‘brute fact about the shape of our subjectivity’ (cf. SC 102) that Kant introduces into his account through the specifically human forms of intuition, space and time. This-suches as such cannot do justice to this ‘brute fact’. Hence, as we will see, McDowell with respect to this aspect of a theory of sensory consciousness realizes that he has to give up not only Sellars but Kant as well and


    CON-TEXTOS KANTIANOS

    36 International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766

    Analytic Kantianism



    ultimately side with Hegel. This point is closely connected with McDowell’s criticism of Sellars’s (and Kant’s) phenomenalism.

    McDowell’s own position concerning sensory consciousness (which he believes to be at least compatible with Kant’s own approach until HIRK) as developed in the Woodbridge- Lectures and SC is indeed ontologically sparse and philosophically less demanding than Sellars’s conception. Through a methodology of indirect characterization and abstraction he is able to develop a detailed account of sensory consciousness, 21 which – though closer to Sellars’s account than he believes it to be – yet consistently abstains from all attempts to characterize the sensible elements of perception more directly or substantially.

    For Sellars’s phenomenalist account of perception this direct characterization is not only possible, but unavoidable. 22 (Sellars to this end develops a methodology of analogical concept formation. Cf. SM ch. 1.) There is no explanation on the basis of the conceptual framework provided by the categories alone of the spatio-temporal constitution of the objects of our experience. This is solely dependent on our specific forms of qualitative- sensory access to the world. In his sense-impression inference he concludes from this fact to the phenomenality of the empirical reality or manifest image of the world.

    While McDowell in “Self-Determining Subjectivity and External Constraint” (SDS; originally 2005) still sounds somewhat optimistic about his own common-sense realistic reading of Kant, in “Hegel’s Idealism as a Radicalization of Kant” (HI; originally 2007) he acknowledges the element of radical subjectivity in Kant’s theory of experience – notwithstanding some moves in the right direction that could be found in the B-Deduction in particular:

    “A Kantian conception of empirical intuitions – intelligibility of objects by virtue of exemplifying unities of the kind characteristic of judgment – almost succeeds in showing how the very idea of objective purport can be understood in terms of free intellectual activity. … What spoils things is that when we widen the picture to take in transcendental idealism, it turns out that the “objects” that we have contrived to see empirical intuitions as immediately of … are after all, in respect of their spatiality and temporality, mere reflections of another aspect of our subjectivity, one that is independent of apperceptive spontaneity.” (HI 81)

    McDowell now understands Hegel’s position as the result of a critical examination of Kant’s account. Following Hegel, he endorses all-encompassing objectivity of apperceptive spontaneity as opposed to the phenomenalist subjectivity of sensory consciousness of the Kant/Sellars kind. This is why, for him, the perceived objects themselves can guide us as what they are in themselves – without guiding us ‘from without’ in any but the most innocent sense.


    21 Cf. SC 119-122 and 124.

    22 I have argued in some detail for this claim in Haag 2016.

    CON-TEXTOS KANTIANOS

    International Journal of Philosophy

    N.o 6, Diciembre 2017, pp. 18-41 37

    ISSN: 2386-7655

    Doi: 10.5281/zenodo.1092766


    Johannes Haag


  7. A revision in McDowell’s theory of intuition


This important and deep difference between their respective accounts concerning the role of sensory consciousness in intentionality that underlies McDowell’s remark [2] about Sellars not “envisaging claim-containing occurrences that are themselves shapings of sensory conscioussness“ (SC 122) does remain in place even if we take into account one of McDowell’s most recent developments of his own conception of sensory consciousness.

In AMG McDowell surprisingly seems to revise his position that intuitions contain claims: “What we need is an idea of content that is not propositional but intuitional, in what I take to be the Kantian sense.” (AMG 260) Since conceptual content is always oriented towards judgment as the paradigmatic product of discursive activity, an elaboration of this idea cannot be easy. Since intuitions do not have discursive content we must ask ourselves whether they have conceptual content nevertheless. Charles Travis denies just that, but for McDowell this amounts to a lapse back into the Myth of the Given: Intuitions would be completely alien to the conceptual order and would provide reasons from without this order. For this reason, intuitive content has to be another form of conceptual content. “The conceptual conent that allows us to avoid the myth is intuitional.” (AMG 269) Intuitional content, unlike discursive content, is not “articulated” (AMG 262), but nevertheless can serve as a foundation for the ‘carving out’ (cf. AMG 263) of the articulation of our discursive capacities: “The unity of intuitional content is given, not a result of our putting significances together.” (AMG 263) It is conceptual in its own right precisely because discursive content can thus be ‘carved out’: „The content of an intuition is such that its subject can analyze it into significances of discursive practices.” (AMG 264) Its content is “in the intuition in a form in which one could make it, that very content, figure in discursive activity” (AMG 265). However, this content is constrained to general categorial classification on the one hand and the sensibilia on the other, including their location in space and time, which can be supplemented in a judgment on the basis of our knowledge about causal and dispositional properties of the intuited objects (cf. AMG 265/6).

As indicated, the resulting view is closer to Sellars’s own account of sensory consciousness concerning some of the details of the process, though not concerning the overarching repudiation of his phenomenalism. Now the function of intuitions in McDowell is mainly “to bring particular objects before the mind for its consideration” (Sellars, IKTE §48). But McDowellian intuitions now they have lost their propositional content are much leaner than Sellarsian intuitions: For Sellars the this-suches (linguistically represented as complex demonstrative phrases) that are the intuitions have the same complete, though merely implicit content as the potential judgments that make this content explicit – including the causal and dispositional properties we ascribe to the objects of our experience we refer to in intuition. That these properties should be included, in large part constituted the difference between the (perspectival) IMs and their conception as independently existing in objects of experience.


CON-TEXTOS KANTIANOS

38 International Journal of Philosophy

N.o 6, Diciembre 2017, pp. 18-41

ISSN: 2386-7655

Doi: 10.5281/zenodo.1092766

Analytic Kantianism



Even very complex demonstrative phrases (or, for that matter, their corresponding intuitions) will not be able to encompass the rich content that intuitive representations of objects as a matter of phenomenological fact have (cf. AMG 263). It is this richness that motivated the change in McDowell’s position in the first place in his discussion of Travis. Sellarsian intuitions, McDowell claims, rather than having truly intuitive content, seem to have merely fragmentary discursive content (cf. AMG 270).

But on closer looks, this does not pose a serious problem for Sellars. The demonstrative phrase gets its content from its reference to the intuitively present object of experience. Its content, consequently, is as rich as the involved sensory-cum-conceptual representation of this object. Sellars’s conception of intuition can only be understood in its relation to the object of experience. That is why in intuition the activity of the understanding (in its guise as the productive imagination) the is not the same as in judgment: “If we think of a taking as a special case of believing, it is best to think of it as ‘believing in’ rather than ‘believing that’.” (SM 18/9)

Once again, it is Sellars’s conception of IMs that underlies this account of the richness of intuitive content. Whatever one ultimately might want to say about the question of a guidance from without and the connected subject of phenomenalism with respect to Sellars’s view of intentionality, his conception of sensory consciousness seems to be remarkably resilient to McDowell’s criticism.


Literature


Kant’s works are quoted in the usual manner as after Kants gesammelte Schriften, vols. I- XXII ed. Preußischen Akademie der Wissenschaften, Berlin 1902 ff., vol. XXIII ed. Deutschen Akademie der Wissenschaften, Berlin, 1956, vols. XXIV-XXIX ed. Akademie der Wissenschaften zu Göttingen, Berlin 1966 ff. The translation mostly follows the Cambridge Edition of the Works of Immanuel Kant, eds. P. Guyer and Allen

W. Wood.


Works by Wilfrid Sellars:

BBK“Being and Being Known,” in: Sellars, W. Science, Perception and Reality. London (1963), 41-59.

EPM “Empiricism and the Philosophy of Mind”, in Sellars, W. Science, Perception and Reality. London (1963), 127-196.

FMPP “Foundations for a Metaphysics of Pure Process” (The Carus Lectures) The Monist

64 (1981) 3-90.

KTE “Some Remarks on Kant's Theory of Experience” Journal of Philosophy 64 (1967) 633-647.

KTI “Kant’s Transcendental Idealism” Collections of Philosophy 6 (1976) 165-181.



CON-TEXTOS KANTIANOS

International Journal of Philosophy

N.o 6, Diciembre 2017, pp. 18-41 39

ISSN: 2386-7655

Doi: 10.5281/zenodo.1092766


Johannes Haag


IKTE “The Role of Imagination in Kant’s Theory of Experience” (The Dotterer Lecture) in: Johnstone, H. (Hrsg.) Categories: A Colloquium. Pennsylvania State University Press (1978) 231-245.

SK “The Structure of Knowledge” (The Matchette Foundation Lectures for 1971 at the University of Texas) in: Castañeda, H.-N. (Hrsg.) Action, Knowledge, and Reality: Studies in Honor of Wilfrid Sellars. Indianapolis: Bobbs-Merrill (1975) 295-347.

SM Science and Metaphysics. Variations on Kantian Themes. London: Routledge and Kegan Paul (1968)

SRLG “Some Reflections on Language Games”, in Sellars, W. Science, Perception and Reality. London (1963), 321-358.


Works by John McDowell:

AMG “Avoiding the Myth of the Given” in HWV 256-272.

HI “Hegel’s Idealism as a Radicalization of Kant” in HWV 69-89.

HWV Having the World in View. Essays in Kant, Hegel, and Sellars. Cambridge, Ma.: Harvard University Press (2009)

IR “Intentionality as a Relation” in HWV 44-65.

LFI “The Logical Form of an Intuition” in HWV 23-43.

SC “Sensory Consciousness in Kant and Sellars” in HWV 108-126.

SDS “Self-Determining Subjectivity and External Constraint” in HWV 90-107. SPE “Sellars on Perceptual Experience” in HWV 3-22.

Förster, E. 2012 The Twenty-Five Years of Philosophy, Cambridge, Ma.: Harvard University Press.

Haag, J. 2007 Erfahrung und Gegenstand. Das Verhältnis von Sinnlichkeit und Verstand.

Frankfurt a.M.: Klostermann.

Haag, J. 2010 “Personhood, Bodily Self-Ascription, and Resurrection: A Kantian Approach“ in: G. Gasser (ed.) Personal Identity and Resurrection. How Do We Survive Our Death?, Farnham: Ashgate, 127-143.

Haag, J. 2012 “Philosophische Abstraktionsebenen” in: Barth, C. & Sturm, H. (eds.) Robert Brandoms Expressive Vernunft. Historische und Systematische Untersuchungen, Paderborn: mentis, 261-285.

Haag, J. 2013 „Kant on Imagination and the Natural Sources of the Conceptual“ in: Lenz,

M. & Waldow, A. (Hrsg.) Contemporary Perspectives on Early Modern Philosophy: Nature and Norms in Thought. Studies in History and Philosophy of Science 29. Dordrecht: Springer, 65-85.

Haag, J. 2014 “Faculties in Kant and German Idealism” in: Perler, D. (ed.) Faculties.

Oxford Philosophical Concepts. Oxford: Oxford University Press, 198-246.


CON-TEXTOS KANTIANOS

40 International Journal of Philosophy

N.o 6, Diciembre 2017, pp. 18-41

ISSN: 2386-7655

Doi: 10.5281/zenodo.1092766

Analytic Kantianism



Haag, J, 2016 “A Kantian Critique of Sellars’s Transcendental Realism” in: Reider, P. (ed.) Wilfrid Sellars, Idealism and Realism: Understanding Psychological Nominalism. London: Bloomsbury, 149-171.

Pippin, R. Kant’s Theory of Form. An Essay on the Critique of Pure Reason, New Haven: Yale University Press 1982, 46-51.

Rosenberg, J. “Divergent Intuitions: McDowell’s Kant and Sellars’s Kant” in: Rosenberg,

J. Wilfrid Sellars. Fusing the Images. Oxford: Oxford University Press 2007, 266- 290.


CON-TEXTOS KANTIANOS

International Journal of Philosophy

N.o 6, Diciembre 2017, pp. 18-41 41

ISSN: 2386-7655

Doi: 10.5281/zenodo.1092766

Enlaces refback

  • No hay ningún enlace refback.
Utilizamos cookies propias y de terceros para realizar análisis de uso y de medición de nuestra web para mejorar nuestros servicios. Si continua navegando, consideramos que acepta su uso.


Creative Commons by-nc 3.0 Logo

ISSN: 2386-7655

URL: http://con-textoskantianos.net

DOAJ LogoErih Plus LogoCitefactor logoredib Logo
LatIndex LogoISOC Logo MIAR Logo
SHERPA/RoMEO Logo
MLA LogoZenodo Logo
ESCI LogoEBSCO LOGOWeb of Sciencescopus logo