Learning Center
Plans & pricing Sign in
Sign Out



									               Intuitions, Transparency, and the Pictures of Thought:
                        Kantianism in McDowell and Sellars
                               Rafeeq Hasan, University of Chicago

       This paper will explore John McDowell‟s critique of Wilfrid Sellars. After sketching out

the broadly Kantian background in which both philosophers operate, I will address McDowell‟s

criticism of Sellars‟s interpretation of what Kant calls an “intuition.” My aim is not only to

develop and explicate McDowell‟s critique, which will necessitate devoting some time to

explaining Sellars‟s understanding of intuitions, but also to delineate these two thinkers‟

conflicting ideas on what shape a philosophical account of objective constraint—that is, the

constraint provided by the layout of reality on thought—should take. A guiding aim here is to

sketch two fundamentally different ways of making sense of the Kantian revolution, i.e. two

different ways of trying to figure out what role Kant and Kantianism ought to have in any

contemporary account of the relation between mind and world. By the time I am done, I also

hope to have gone some way towards elucidating two different „pictures‟ of the philosophical

task. My ultimate claim is that, in the end, the debate between McDowell and Sellars is not one

that can be adjudicated through the standard philosophical procedure, i.e. examining arguments,

weighing the force of cogent examples, etc. Rather, in adjudicating between these two thinkers

what we have to consider is no less than two perhaps irreconcilable images of where philosophy

begins and where it ends, what we might call two different „paradigms‟ of the philosophical task.

If I am right, the perhaps unsettling consequence is that what is at stake is a case of philosophical

decision. The vindication of one of these pictures can only come after the fact: by what an

adherence to them makes possible.

I. The Kantian Radicalization of Epistemology

        Recently a number of philosophers have articulated the following insight about the

Kantian project: Kant is not, in the first instance, seeking to answer traditional skeptical

questions, questions like „How do I know that my thoughts about the world are, in some sense,

true to the way things really are?‟ Rather, Kant poses a question that is by his lights more

fundamental, not just in the sense of more important or more philosophically pressing, but in the

sense that it asks after something that the first kind of question simply takes for granted: that

thought of the objective world is so much as possible (so that one could know what it means for

thought to be true to reality). In place of the skeptical question, Kant asks, „How is it that I can so

much as have representations of a world? What does it mean for the content of thought to be

world-involving?‟ As James Conant helpfully puts it, the contrast between modern skepticism, as

traditionally conceived by Descartes and others, and the Kantian project is that “Cartesian

skepticism calls into question the veridicality of one‟s experience; Kantian skepticism calls into

question the intelligibility of experience.”i

        Now Kant is not just stubbornly shifting the question, as though the skeptical puzzles

were so insuperable that one had better give up trying to solve them and take up a new project

instead. What Kant shows is that by answering the more fundamental question, by showing what

it takes for thought to have intentional content, i.e. content that is constituted by (and depends for

its truth upon) the way the world is, traditional skeptical questions lose much of their force.

When we understand just how it is that thought can be about the world, we see that it does not

make sense to consider a body of thought as consistently failing to be about the world. This does

not mean that the possibility of error vanishes, only that it ceases to pose the kind of agonizing

philosophical question that professional philosophers have perennially obsessed about.

       The radicalization inherent at the heart of the Kantian project—the move from questions

about the conditions for thought to be true to questions about the very possibility of thought

being about something at all (and so a potential candidate for truth or falsity)—has been put in a

number of different ways. It is, I believe, underlying P.F. Strawson‟s minimalist reading of

Kantian philosophy as seeking to elucidate “the bounds of sense,” i.e. “the fundamental structure

of ideas in terms of which alone we can make intelligible to ourselves the idea of experience of

the world.”ii As I read Strawson‟s agenda, figuring out what kinds of concepts we need to make

“intelligible to ourselves the idea of experience of the world” is a way to put what it means to

figure out how thought can be world-involving, how the world can bear on thought. Elsewhere,

Strawson writes, “we ought to ask, not how it can be that on the basis of perceptual experience as

it is, we come to have the beliefs [that we do], but how it is that perceptual experience is already

such as to embody the beliefs in question; or, perhaps better, what it is for perceptual experience

to be such as to embody the beliefs in question.”iii

       Across the so-called Continental/Analytic divide, a version of the same thought finds an

incredibly suggestive expression in Martin Heidegger‟s reading of Kant. Heidegger writes: “the

Critique is concerned with ontology and not with epistemology,” so that Kant‟s “Copernican

revolution elucidates for the first time the possibility of access to objects themselves,” iv rather

than merely taking the object-involving nature of thought for granted. Heidegger‟s somewhat

contentious idiom is actually quite elucidatory here. Kant‟s project is more like an ontology of

thought—an account of what it is for thought to be thought, or, as Strawson writes in the passage

just cited, “what it is for perceptual experience to be such as to embody the beliefs in

question”—than it is an epistemology; it is radicalization as „ontologization.‟

       Wilfrid Sellars and John McDowell are two of the most notable Anglo-American

philosophers whose work has been consistently oriented by the Kantian radicalization of

epistemology. McDowell begins his John Locke Lectures (1991) by candidly stating, “One of

my main aims is to suggest that Kant should still have a central place in our discussion of the

way thought bears on reality.”v And his Woodbridge Lectures (1997) begin by endorsing what he

takes to be the Sellarsian conviction that “no one has come closer than Kant to showing us how

to find intentionality unproblematic.”vi Sellars subtitled Science and Metaphysics, the self-

professed sequel to Empiricism and the Philosophy of Mind, “Variations on Kantian Themes,”

and begins the former work by saying, “I build my discussion of contemporary issues on a

foundation of Kant exegesis and commentary…because, as I see it, there are enough close

parallels between the problems confronting him…and the current situation and its demands.”vii

       Both McDowell and Sellars have pursued the Kantian radicalization of epistemology less

as a project in traditional Kant exegesis and more as a project of exposing the blindness and

question-begging nature of quite a bit of the twentieth-century metaphysics of mind—in this

respect they are perhaps closer to Heidegger than one might suspect. It is instructive to look

briefly at a few examples of their respective ways of engaging in the Kantian project. Right now

I just want to give some determinate content to the idea that the Kantian radicalization can be

conceived as an on-going project for what could be called theoretical/systematic rather than

simply historical philosophy.

       For McDowell the difference between the Kantian question and traditional

epistemological questions shows up in, among other places, Donald Davidson‟s isolation of the

“dualism of scheme and content” in Quine‟s reconstructed empiricism.viii By McDowell‟s lights,

Davidson is right to hold that Quine‟s neat separation of a conceptual-scheme, i.e. the series of

rationally interlinked commitments about what is a justification for what, from content, figured

in Quine as the raw, unconceptualized impingements of the world on our „sensory receptors,‟

poses huge difficulties for the empiricist project of seeing thought as rationally answerable to,

i.e. justified by, how things are in the world. But according to McDowell Davidson

symptomatically misdiagnoses the problem of the dualism as pointing to Quine‟s susceptibility

to traditional skepticism—his susceptibility to the charge that by founding knowledge on the

testimony of the senses, which, as modern philosophers since Descartes have been so fond of

pointing out, are markedly prone to error and deception, one leaves it open to question whether

or not we are systematically deceived about the very nature of reality. On McDowell‟s

understanding, the problem is not that the dualism of scheme and content lapses into skepticism,

but that it is simply “incoherent” (“SCD,” 91). By placing the mind‟s „experience‟ of the world

outside the domain of the conceptual, outside the network of inferential and justificatory

linkages—what McDowell, in a picturesque turn of phrase, calls “the space within which thought

moves”ix—Quine makes it impossible to see how exercises of concepts could be rationally (and

not merely causally) constrained by the world.

       McDowell‟s strategy is not just to expose Quine‟s argument as a species of what Sellars

famously called the Myth of the Given, which for my current purposes can be provisionally

glossed as the idea that nothing that is not structured like a concept can be available as a

justification for thought (which is by its nature discursive)—though one thing I hope to make

clear in this paper is that there is a much deeper characterization of the Myth, one which sounds

a Kantian key and pertains to the very possibility of thought‟s being contentful. x Rather,

McDowell‟s aim is to develop the Kantian insight that before one can ask about the conditions

for thought to be true to the world, one has to ask about how thought can be of the world at all.

On McDowell‟s Kantian reading of the dualism, by removing experience of the world from the

sphere of the conceptual, Quine has made it impossible to see how thought can have intentional

content. Davidson‟s blindness to this problem can be seen as a result of his fixation on the pre-

Kantian space of epistemological problems.

       Similarly, the Kantian radicalization underlies the crucial but often-overlooked section of

Empiricism and the Philosophy of Mind where Sellars discusses the history of early empiricist

views on the status of mental awareness. Sellars compares the empiricist views on generic sorts

or categories, i.e. concepts like red, with their views on awareness of determinate sorts such as

crimson, or, to give an example of the latter that is more to the point for my discussion here,

concepts like „this particular shade of red,‟ exploited in the presence of a colored physical object

(EPM, 57-64). Sellars shows that the acknowledged problem for empiricists was always how to

account for the awareness of generic sorts. The question they sought to address was, „How do I

get the general concept, e.g. Redness, from my specific experiences of determinate shades of

red?‟ According to Sellars, the empiricists simply assumed that there was no philosophical

problem in understanding how one acquires awareness of determinate sorts. The idea was that

we know about these just by virtue of having sensations.

       On one hand, it is clear that the empiricist views on determinate sorts belie a fairly deep

commitment to the Myth. The empiricists assumed that merely by having sensations and images

of determinates we are aware of them as determinates. As Sellars puts it, they assumed that “the

human mind has an innate ability to be aware of certain determinate sorts…simply by virtue of

having sensations and images” (EPM, 62). Clearly this view must be rejected if “to reject the

Myth of the Given is to reject the idea that the categorical structure of the world…imposes itself

on the mind as a seal imposes itself on melted wax” (“Lever,” 12).

       But Sellars‟s point can also be put as an insistence on the priority of the Kantian

radicalization over the traditional epistemological problematic. The question that would be posed

from within the latter is, „Granted that we have simple thoughts (determinates), how do we get

complex ones (determinables)?‟ Asking a question like this leads, in turn, to a form of skepticism

in the shape of the question, „How do we know that the general concepts I get from my

impressions correspond to the general concepts that you get from your impressions (EPM, 58)?

How do I know that what I mean by „red‟ is what you mean by „red?‟‟ But in asking these sorts

of questions the empiricists blithely bypassed the question, „How do I so much as have

awareness of simple concepts?‟ And when we see that even to have simple, determinate concepts

presupposes acquiring a whole arsenal of complex concepts—when we see that the general

concept of red and the specific concepts of shade are, as it were, given as a package xi—then we

see that the skeptical problem disappears. If we only even have the ability to be aware of

determinate sorts through the process of acquiring language, which involves initiation into a

shared, communal network of concepts—the view that Sellars calls “psychological nominalism”

(EPM, 63)—it ceases to be intelligible that my awareness of a determinate sort as something of

that sort could, at the ground level, fail to correspond to your awareness of it as of that sort.

       Now, Sellars‟s view of awareness is undoubtedly much more complicated than this. As

one set of commentators puts it, through the course of EPM “Sellars eventually back[s] off this

hard-core linguisticism.”xii But even this quick sketch of the one stage of EPM shows how a core

Kantianism informs his procedure. It is of course true that much of what both Sellars and

McDowell do is to expose the Myth of the Given in likely and unlikely places, but it is also true

to say that what they do is to flesh out the Kantian radicalization.

       Yet despite the shared agreement between McDowell and Sellars on the importance of

Kant to the project of understanding the relation between thought and the world, they diverge

sharply not only in what the execution of the Kantian project ought to look like, but also on the

extent to which Kant was successful in his endeavor, the extent to which he needs to be

supplemented or corrected.

II. Kantian Intuitions and Objective Constraint

       Figuring out what exactly Kant means by an intuition has become something of a cottage

industry in historically and philologically oriented Kant scholarship. For anyone who has

attempted to make his way through this literature, it becomes increasingly apparent that no

reading of the text, no matter how close or careful, which operates in the absence of a larger

philosophical understanding of Kant‟s transcendental enterprise, and thus in the absence of a

developed philosophy of mind, will be able to provide decisive evidence in favor of one

interpretation rather than another. The following quotation from Henry Allison gives a good

indication of the shape of the problems surrounding the Kantian intuition and the way in which

many (Allison included) are blind to the kinds of substantive philosophical questions that must

be asked if the problems are to find a solution:

       Kant defined an intuition as a „singular representation”…and he contends that it
       refers „immediately to its object‟…it is precisely in virtue of its ‘immediacy,’ that
       is, its direct, nonconceptual mode of representing, that an intuition can present a
       singular object to the mind and, therefore, serve as a [singular representation].


       Nevertheless, a tension, if not outright contradiction, has often been noted
       between the official definition of „intuition‟ as a „singular representation,‟ and the
       account of sensible intuition [that Kant provides]. The problem is that, according
       to Kant‟s theory of sensibility, sensible intuition provides the mind with only the
       raw data for conceptualization, not with the determinate knowledge of objects.
       Such knowledge requires not only that the data be given to intuition, but also that

         it be taken under some general description or „recognized in a concept.‟ Only then
         can we speak of the „representation of an object.‟xiii

The standard reading of Kant (under which Allison‟s falls) holds that an intuition refers to

something nonconceptual. On this reading, it is only with the understanding, the faculty of what

Kant calls “spontaneity,” that concepts come into play. But as Allison suggests, the standard

reading runs into an obvious problem, for if having “determinate knowledge of objects” requires

conceptual determination, how it can be that an intuition, which is supposedly preconceptual,

gives us such knowledge? In straining to preserve the reading of an intuition as nonconceptual,

Allison‟s recommends the following solution:

         The key to the resolution of this tension is…that a Kantian sensible intuition is
         only „proleptically‟ the awareness of a particular…[A]lthough intuitions do not in
         fact present or refer to objects apart from being „brought under concepts‟ in a
         judgment, they can be brought under concepts, and when they are they do
         represent particular objects (KTI, 67-68) (emphasis in original).

So the standard solution, as Allison represents it, is to argue that there are both nonconceptual

intuitions, something like the raw data of sensibility that are not yet conceptually combined into

a representation of an object, and conceptualized intuitions, where the understanding cooperates

with sensibility to form a singular representation of an object. On this reading, Kant‟s failure to

distinguish between these two types of intuitions shows either that he was philosophically

confused or that he was extremely bad at explaining his thought to the reader. The upshot of this

reading is that one needs to do a huge amount of reconstruction for Kant‟s argument to make any


         But the lack of textual charity is the least of the problems involved in readings like

Allison‟s. Both Sellars and McDowell would hold that the very idea of a “direct, nonconceptual

mode of representing” is a manifestation of the deepest level of the Myth, a manifestation of the

idea that the mind could have an innate, nonconceptually acquired ability to classify something

as something. If Kant thought it made sense to speak of nonconceptual representations, he would

be nothing more than a closeted empiricist in rationalist clothing (rather than someone who

ingeniously combined the truths of both positions while casting off their confusions).

          Sellars provides an alternate reading of Kant—the basic outlines of which McDowell

wholeheartedly endorses—according to which intuitions are always-already conceptually

structured. More specifically, Sellars‟s way of making sense of Kant‟s statement that an intuition

provides an immediate relation to an object is to argue that “„immediate relation‟ ought to be

construed on the model of the demonstrative „this.‟” So that, on “the correct interpretation,

intuitions would be representations of thises and would be conceptual in that particular way in

which to represent something as a this is conceptual” (SM, 3) (emphases in original).

          Ultimately, both Sellars and McDowell agree that it simply cannot be the case that Kant

finds it intelligible to speak of “direct, nonconceptual” representations—at least if he is to be

taken as an (perhaps the) exemplary philosopher of intentionality. Understanding intuitions as

already shaped by conceptuality necessitates a reading of the organization and argument of the

Critique that is radically different from the ones put forth by generations of Kant scholars

(though not, interestingly enough, Heidegger). I cannot go into the details of the standard reading


          Yet while both McDowell and Sellars agree that Kant thought that intuitions were

conceptually structured, they differ on just what to make of this. Sellars argues that this

conception of intuitions does not provide an adequate account of the constraint on thought

provided by the layout of reality. While he does not want to speak of proleptic intuitions, he does

want to insist on a contrast between intuitions and sense impressions. (So in this sense his

reading is a more self-aware, philosophically perspicuous version of the standard reading.) Sense

impressions are “the sheer impact of reality on us [that don‟t] declare anything; [they] just

provide the occasion on which we proceed to construct [a representation]” (KPKT, 84).

       Here we see a point of contact between Sellars‟s discussion of Kant and his analysis of

inner experiences in Empiricism and the Philosophy of Mind. Recall that in that text Sellars

distinguished perceptions, which are conceptually structured takings in of facts, from

impressions, which are nothing more than the merely causal impingement of the world on our

senses. Similarly, Sellars holds that in addition to Kant‟s account of intuitions, which are

propositionally structured, he needs an account of impressions, which are not.xv

       Sellars suggests that by failing to distinguishing intuitions from impressions, Kant fails to

adequately secure the constraint of reality on thought, and thus risks falling into idealism. Thus,

“it is only if Kant distinguishes the radically nonconceptual character of sense [impressions]

from the conceptual character of…intuition…and, accordingly, the receptivity of sense from the

guidedness of intuition that he can avoid the dialectic which leads from Hegel‟s Phenomenology

to nineteenth-century idealism (SM, 16). Much of Sellars‟s later and notoriously obscure

philosophical project is aimed at providing an account of how to think of what this passage calls

receptivity (and what elsewhere Sellars refers to as “sheer receptivity”) (SM, 7) so as to

distinguish it from the conceptually mediated receptivity operative in intuitions.

       For McDowell, on the other hand, if one appreciates all that is entailed by Sellars‟s fine

reading of intuitions as conceptually structured demonstratives one finds one‟s way to

countenancing all the objective constraint that one could possibly need to vindicate the bearing

of the world on thought. Wanting any thing like a more fundamental, ultimate kind of constraint

is inevitably to succumb to the Myth.

       By way of conclusion, I want to unpack McDowell‟s alternate understanding of intuitions

and constraint. This will allow me to give a brief account of the different attitudes toward the

task of philosophy, and the different pictures of thought, underpinning the work of Sellars and


       In “Having the World in View” McDowell‟s basic aim is to show how

perceptions/intuitions—i.e. cognitive takings-in of states of affairs—provide all the constraint

necessary to vindicating the intentionality of thought. To show this, McDowell needs to show

that the conceptual determination of an intuition and the claim-containing character of a

perception are “imposed or impressed on their subject” (“HWIV,” 440), so that qua conceptual,

they might be viewed as operations of receptivity rather than spontaneity. Taking up Kant‟s

claim that the paradigmatic form of conceptual spontaneity is judging, xvi i.e. actively deciding

that things are thus-and-so, McDowell suggests that viewing perceptions as conceptually

determined operations of receptivity allows one to see that having things look a certain way to

one “is not the same as judging that they are that way” (“HWIV,” 339), in that only the latter

involves exercises of spontaneity.

       The cornerstone of McDowell‟s argument comes from a passage in a section of the

Critique entitled “The Clue to the Discovery of All Pure Concepts of the Understanding,” where

Kant writes: “The same function which gives unity to the various representations in a judgment

also gives unity to the mere synthesis of various representations in an intuition” (A79/B104-

105). McDowell takes this remark to suggest that

       as actualizations of conceptual capacities with the appropriate togetherness, the
       judgment and the ostensible seeing would be alike. They would differ only in the
       way in which the relevant conceptual capacities are actualized. In the judgment,
       there would be a free responsible exercise of the conceptual capacities; in the
       ostensible seeing, they would be involuntarily drawn into operation under
       ostensible necessitation from an ostensibly seen object “(“HWIV,” 458).

The remark from the Clue suggests that what is given (not Given) to the mind by the world is

always in the form of conceptual content—its being the case, or looking to be the case, that

things are thus-and-so. This same conceptual content can be incorporated into exercises of

spontaneity; one can actively make up one‟s mind to take it that things are as they look to be.

(Indeed, elsewhere McDowell suggests that exercises of receptivity presuppose the faculty of

spontaneity, so that things can only look to us to be a certain way if we also have the capability

to decide that things are that way.)xvii The content given in an intuition is, as McDowell puts it,

“judgment shaped” (“HWIV,” 461).

       McDowell‟s reading of Kant allows us to see intuitions as transparent; in having an

intuition something about the world is directly disclosed to one (in conceptual form). On the

other hand, Sellars‟s reading of Kant locates the ultimate objective content of an intuition in an

impression. It is impressions, not intuitions, which provide the “constraining element

of…experience” (SM, 9). To put it a bit loosely, on the Sellarsian account an intuition

conceptually processes an impression in order to yield the content available to exercises of

spontaneity. This makes intuitions opaque: in the absence of a story of how we get intuitions out

of impressions, we cannot see how it could be that the content of an intuition is directly world-

involving. This is what I take McDowell to have in mind when he counterposes his reading to

Sellars‟s by saying that he (McDowell) wants to speak “of intuitions as “conceptual shapings of

sensory consciousness” (“HWIV,” 462). If McDowell is correct in his reading, Sellars would

want to say something more like, „intuitions are conceptual shapings of impressions, which are

the result of something impinging on sensory consciousness.‟ Sellarsian impressions function as

intermediaries between conceptual content and the world.

       By contrast, McDowell‟s view of receptivity involves a picture of constraint as “exerted,

in intuition, by objects themselves, the subject matter of the conceptual representations involved

in perception” (“HWIV,” 468). At one point, McDowell suggests the metaphor of voice: “A seen

object as it were invites one to take it to be as it visibly is. It speaks to one…„See me as I am,‟ it

(so to speak) says to one; „namely, as characterized by these properties‟—and it displays them”

(“HWIV,” 468).

       In passages like these, McDowell has, I think, the following picture in mind: experience

only contains claims when we have acquired conceptual capacities, which is just to say, when we

have been initiated into the space of reasons. As so initiated, the world impinges on us as

propositionally structured states of affairs. As animals living within the space of reasons, we

open our eyes to the world and simply see that, for example, „this is a red apple.‟ Our thought,

„this is a red apple,‟ is constrained by the way in which the world made its claim on us—namely,

by „saying‟ to us, „this is a red apple.‟ It is always up to us to revise our thought. Perhaps we are

in non-standard lighting conditions, or we know that we have a hard time distinguishing real

apples from holograms of apples. But even in these cases what our thought must be true to is our

initial perception of a red apple. In the cases of perceptual anomaly or error, the worldly fact to

which we think that our thought is beholden—what we are trying to explain by citing weird

lighting or the existence of holograms—is the fact of its looking to us that the apple is red. So if

one understands idealism as the thesis that “the supposed objects of…conceptual shapings of

consciousness can only be projections of our conceptual activity” (“HWIV,” 489), an account

like McDowell‟s is not a species of idealism, primarily because it shows how the conceptual

shapings of consciousness are operations of our conceptual passivity.

       Thus, to vindicate intentionality is to understand how our thought is constrained by the

world. And if “the same function which gives unity to the various representations in a judgment

also gives unity to the mere synthesis of various representations in an intuition,” understanding

the nature of conceptual activity and understanding how conceptual activity is constrained by the

world are but two sides of the same coin.

       Contrast this story with Sellars‟s sense that without an account of nonconceptual

impingements of the world on the mind the threat of idealism looms large. In his account of

Kantian intuitions, Sellars uses images that are almost exactly like McDowell‟s. He writes, “To

know the language of perception is to be in a position to let one‟s thoughts be guided by the

world”; “In receptivity we do the same sort of thing we do in…spontaneity…but we do it as

receptive to guidance by objects we come to represent”; and “it is perceptual judgments

themselves which are evoked by the action of objects in our perceptual capacities” (“KTE,” 273).

But Sellars thinks that these images are not enough, that they do not dissolve the threat of

idealism. Why not?

       There are some quite specific aspects of Sellars‟s philosophy underlying his sense that

the kind of constraint described in these passages is not enough. But many of these local,

somewhat hermetic aspects of Sellarsianism—most notably, his thesis that “elements in the

conceptual order can stand in content-involving or semantical relations only to elements in the

conceptual order, not to elements in the real order” and his related account of picturing—are but

specific manifestations of a more underling frame of mind.xviii

       In brief, this underlying frame of mind holds that philosophical vindication must happen

from outside of the normative order. For Sellars, it is part and parcel of the task of philosophy to

ask how it is that we so much as have the ability to form conceptual representations. He sees that

Kant does not provide an answer to this question—indeed, that he does not even really ask it—

and takes this to be the fundamental flaw of his project.

        McDowell would not deny that figuring out how we have the ability to form the

conceptual representations we do is a legitimate task for some mode of inquiry—he is not a

reactionary about science. But he would deny that this is the task of philosophy. McDowell

famously describes the view that philosophy ought to account for the normative order from a

space outside of it (say, the space of natural science) as the desire for a view from “sideways on”

(MW, 34 and passim). The question of how we form conceptual representations would, for

McDowell, be a question only about the enabling capacities that allow human beings to be the

discursive creatures that they are.xix

        For McDowell, the task of a philosophy of mind is to make intentional content look

unmysterious, given the available concepts that we all use and share. It is to make explicit an

understanding of how the mind relates to the world that he thinks traditional epistemology has

made opaque. If this is the job of philosophy, the job of cognitive science may be to show how

much architectural complexity at the level of the brain underlies this simplicity at the level of

concepts. Sellars on the other hand thinks that philosophy ought to see how both views—the

view from the level of the normative and the view from the level of the architectural—“fall

together in one stereoscopic view” (“PSIM,” 5). While this desire may be legitimate in the

abstract—and I don‟t think McDowell would deny that at some stage of inquiry this is a laudable

aspiration—conflating these levels has led to so many bad views in philosophy that it seems that

one ought, for the moment, to give up the quest for the stereoscopic view. What McDowell‟s

project shows us is that giving it up does not entail having to find it mysterious that thought can

be answerable to the world.


 James Conant, “Varieties of Skepticism” (unpublished paper), p. 5. Conant notes that what he
characterizes as “Cartesian” and “Kantian” strands of skepticism refer less to the actual
discussions of skepticism in each of these philosopher‟s writings and more to the larger “shape
of [their philosophical] problems” (7). I thank Professor Conant for making this paper available
to me. Much of what is hastily discussed in the first section of my paper is developed at length in
Conant‟s “Varieties of Skepticism.”
 P.F. Strawson, The Bounds of Sense: An Essay on Kant’s Critique of Pure Reason (New York:
Routledge, 1966), p.15.
  Strawson, “Imagination and Perception,” in Freedom and Resentment and Other Essays
(London: Methuen, 1974) (emphasis in original), p. 50.
  Martin Heidegger, Phenomenological Interpretation of Kant’s Critique of Pure Reason, trans.
Parvis Emad and Kenneth Maly (Bloomington: Indiana University Press, 1997), pp. 38, 46.
Hereafter cited as PIKC. A similar thought, coming from this side of the divide, is expressed by
Sellars: “The core of Kant‟s „epistemological turn‟ is the claim that the distinction between
epistemic and ontological categories is an illusion. All so-called ontological categories are in fact
epistemic. They are „unified‟ by the concept of empirical knowledge because they are simply
constituent moments of this one complex concept, the articulation of which is the major task of
the constructive part of the Critique.” See Wilfrid Sellars, “Some Remarks on Kant‟s Theory of
Experience,” in Kant’s Transcendental Metaphysics: Sellars’ Cassirer Lecture Notes and Other
Essays, ed. Jeffrey Sicha (Atascadero, CA: Ridgeview Publishing Company, 2002), p. 270.
Hereafter cited in text as “KTE.” At first glance, it would seem that Sellars‟s claim is actually the
opposite of Heidegger‟s, i.e. it would seem that Heidegger is claiming that Kant‟s project is
fundamentally an ontology while Sellars‟s is suggesting that it is fundamentally an epistemology.
But, I take it, the point that both philosophers are trying to make is that Kant‟s focus on
intentional content undoes the putative distinction between the two domains of inquiry.
 John McDowell, Mind and World, paperback edition with “Introduction” (Cambridge: Harvard
University Press, 1996), p. 3. Hereafter cited as MW.
  McDowell, “Having the World in View: Sellars, Kant, and Intentionality,” Journal of
Philosophy, vol. 95 (September 1998): 431. Hereafter cited as “HWIV.”
  Sellars, Science and Metaphysics: Variations on Kantian Themes (Atascadero, CA: Ridgeview
Publishing Company, 1967), p. 1. Hereafter cited as SM.
   McDowell, “Scheme-Content Dualism and Empiricism” in The Philosophy of Donald
Davidson, ed. Lewis Edwin Huhn (Chicago: Open Court, 1999), pp. 87-108. Hereafter cited as
“SCD.” Similar points are made in Mind and World, particularly pages 129-161.

 McDowell, “Knowledge and the Internal,” in Meaning, Knowledge, and Reality (Cambridge:
Harvard University Press, 1998), p. 408. Hereafter cited as “KI.”
  For an expression of this idea see Sellars, Empiricism and the Philosophy of Mind (Cambridge:
Harvard University Press, 1997), p. 19. Hereafter cited as EPM. Later in his career, Sellars gives
some succinct accounts of the Myth of the Given that operate at a much higher level of
generality. He writes, “The temptation to suppose that experience, even in its most primitive
form, comes to us, as it were, as experiences of items as „of‟ a certain sort, this is what I have
called the „Myth of the Given‟ in my paper „Empiricism and the Philosophy of Mind‟; that is
exactly it.” And: “the [M]yth of the [G]iven, as I defined it in „Empiricism and the Philosophy of
Mind,‟ is the notion that objects present themselves to us [without a process of conceptual
mediation] as being of a certain sort or as being of a certain kind.” Sellars, Kant and Pre-Kantian
Themes: Lectures by Wilfrid Sellars, ed. Pedro Amaral (Atascadero, CA: Ridgeview Publishing
Company, 2002), p. 35, 84. Hereafter cited as KPKT. In one of his last publications, he writes:
“To reject the Myth of the Given is to reject the idea that the categorical structure of the world—
if it has a categorical structure—imposes itself on the mind as a seal imposes itself on melted
wax.” Sellars, “The Lever of Archimedes,” The Monist, vol. 64 (1981): 12. Hereafter cited as
“Lever.” Similarly general formulations of the Myth can also be found within Empiricism and
the Philosophy of Mind, but so far as I know none of them are as lucidly or economically
expressed as these. Unfortunately, showing how the various levels of the Myth fit together,
particularly the level concerned with justification and the level concerned with the categorial
nature of experience, is a task that is beyond the scope of a paper such as this one. I am
convinced that ultimately all of the levels are simply different angles of vision on the same
general problematic, and that understanding how this can be so is central to grasping Sellars‟s
philosophical system.
  McDowell puts the basic idea like this: “Nothing could be immediately present to one‟s senses
unless one already had knowledge that goes beyond what is immediately present to the senses”
(“KI,” 411-412). Once again, this is a deeply Kantian claim. At one point in his lecture course on
the Critique of Pure Reason, Heidegger suggests to his students that it is absolutely key to
understand Kant‟s claim that “the data of sensation…precisely do not account for the essence of
sensibility” (PIKC, 70). As Sellars might put it, the Kantian point at play in both of these
passages is that “sense impressions are passive. The world doesn‟t tell us what it is. It doesn‟t
carry its cognitive heart on its sleeve” (KPKT, 72).
   William deVries and Timm Triplett, Knowledge, Mind, and the Given: Reading Wilfrid
Sellars’s ‘Empiricism and the Philosophy of Mind’ (Indianapolis: Hackett Publishing Company,
2000), p. 58.
   Henry Allison, Kant’s Transcendental Idealism: An Interpretation and a Defense (New
Haven: Yale University Press, 1983), p. 67 (emphasis mine). Hereafter cited as KTI.
   For an extremely detailed reading of the Transcendental Deduction that is very close in spirit
to this Sellarsian understanding of Kant see, “The Transcendental Deduction of the Categories:
An Outline and Interpretation,” by John Haugeland (in collaboration with Jim Conant and John

McDowell” (unpublished manuscript, 1997). Also see Robert Pippin chapter “Kantian and
Hegelian Idealism” in Hegel’s Idealism: The Satisfactions of Self-Consciousness (Cambridge:
Cambridge University Press, 1989), pp. 16-41, although there are aspects of Pippin‟s way of
casting the relationship between Kant and Hegel that are not congenial to McDowell, who
discusses his differences from Pippin in “Hegel‟s Idealism as a Radicalization of Kant”
(unpublished manuscript) and in his reply to Pippin‟s “Leaving Nature Behind: Or, Two Cheers
for Subjectivism.” I am leaving the huge question of the proximity of Strawson‟s reading of Kant
to Sellars‟s and McDowell‟s for another time, as well as the even huger question of why it might
be that Heidegger‟s reading converges so markedly with this tradition. I hope to explore both of
these questions in detail elsewhere.
   Much of the last few paragraphs have been an attempt to make sense of the following passage
from Sellars: “It is often taken for granted that Kant was clear about the distinction between
conceptual and non-conceptual mental states or representings. „Empirical intuitions‟ are
interpreted as non-conceptual and construed…as the epistemically more important members of
the sensation family. Actually the pattern of Kant‟s thought stands out far more clearly if we
interpret him as clear about the difference between general conceptual representings (sortal and
attributive), on the one hand, and, on the other, intuition as a special class of non-general
conceptual representings, but add to this interpretation the idea that he was not clear about the
difference between intuitions in this sense and sensations” (“KTE,” 272) (emphases in original).
  Kant writes: “We can reduce all acts of understanding to judgments, and the understanding
may therefore be represented as a faculty of judgment” (B94) (emphases in original).
    For instance, “empirical intuitions…just happen, outside the control of their subjects. But
since they exemplify kinds of unity whose original home, so to speak, is judgment, they could
not happen except in the lives of subjects who are capable of the free intellectual activity that
judging is.” McDowell, “Hegel‟s Idealism as a Radicalization of Kant,” p. 4.
     Sellars of course has other reasons as well for holding that there must be a deeper form of
constraint. To give just a short list: there is his sense that only science can tell us what is truly
real, his interest in understanding something like the nonconceptual aspect of experience (what
McDowell would call the nonconceptual machinery which lies below experience), and his
extraordinarily intricate reading of how to make the various parts of the Critique hang together.
All I am suggesting is that all of these reasons are variations on a more underlying theme, a habit
of thought that provides a justification for all the things just mentioned in my short list. The
inseparability of local argument and underlying philosophical commitment comes out quite
nicely in the following passage: “Kant‟s failure to distinguish clearly between the „forms‟ of
receptivity proper and the „forms‟ of that which is represented by the intuitive conceptual
representations which are „guided‟ be receptivity—a distinction which is demanded both by the
thrust of his argument, and by sound philosophy—had as its consequence that no sooner had he
left the scene than these particular waters were muddied by Hegel” (SM, 29) (emphasis mine).
An equally careful reader of Kant committed to McDowell‟s view would reply to a passage such
as this by agreeing that Kant does not distinguish between these two forms of receptivity, but
would add that in fact not distinguishing these is what “is demanded both by the thrust of his

argument, and by sound philosophy.” And rather than see Hegel as the one who „muddied the
waters,‟ such a reader would be committed to the view that Hegel‟s genius lay in his fleshing out
and radicalizing the idea of worldly constraint as coming from inside the space of the conceptual.
So we see once again that it is impossible to factor out substantive philosophy from even
something as seemingly local as exegesis of a concept from the history of philosophy.
   I owe my understanding of „enabling capacities‟ to Fred Stoutland. The distinction between a
capacity which allows us to possess a concept and our possession of that concept is, I take it,
central to Wittgensteinian accounts of language learning, mathematics, etc.


To top