Kant, Husserl, McDowell: The Non-Conceptual in Experience

1. Introduction

Recently, discussions on non-conceptual content in Edmund Husserl′s phenomenology have seen a revival. This seems due partially to the work of John McDowell on the role of sense perception in experience. In his already classic work Mind and World (1994), McDowell advocates a conceptualist reading of Kant^[1] according to which there can be no non-conceptual mental states. Those against conceptualism mostly agree that there are at least “some mental states [that] can represent the world even though the bearer of those states does not possess the concepts required to specify their content”.^[2] Conceptualists like McDowell, however, think that this line of thought inevitably results in what Wilfrid Sellars [1963] famously dubbed the “myth of the given”. For that which cannot be conceptually apprehended must in all ways escape the bounds of thought and is therefore, as Kant so diversely put it, “blind”,^[3] “as good as nothing for us”,^[4] or “less than a dream”.^[5] Therefore, McDowell asserts that the very idea of non-conceptual sense data which somehow affect our beliefs while at the same time offering a foothold in external reality must be avoided. But it seems the only alternative is a coherentism according to which thought does not touch upon the world and beliefs are justified by other beliefs only. Such a coherentism, at least on McDowell’s understanding, renders the whole idea of experience impossible. The solution he then proposes takes form as a conceptualist reading of Kant which is supposed to save thought′s bearing on external reality without invoking any kind of non-conceptual content that would be prone to the given.

In this paper I compare McDowell′s conceptualism to Husserl′s early and late phenomenology and recent interpretations of both. I start out by discussing McDowell′s conceptualist reading of Kant and some of the criticisms it has received from Kant scholars. I will argue that their cases for Kantian non-conceptualism fail to rebut McDowell’s conceptualist reading. I will then look at Hopp’s [2010, 2011] phenomenological critique of McDowell, which he bases on the phenomenology of the early Husserl, and show why his arguments similarly have little bearing on McDowell’s conceptualism. Thirdly, I discuss Barber’s [2008] and Mooney’s [2010] interpretations of Husserl’s later work, who assert that the later Husserl and McDowell are in fundamental agreement. I will argue that their position is correct but for the wrong reasons. Contrary to their interpretations, I will show that Husserl’s focus on passive synthesis allows him to move beyond experience in the Kantian and McDowellean sense, which demands the role of concepts in any synthesis, toward wholly passive syntheses which operate freely from the rule of the understanding, and which are therefore non-conceptual. This position is, I argue, inconceivable from the standpoint of Kant′s transcendental philosophy, since in the latter both transcendental apperception and the categories are taken to be involved in the lowest syntheses, even that of perception.^[6] Although this makes sufficient room for non-conceptual content in Husserl’s later philosophy, I will argue that it need not be incompatible with McDowell’s conceptualism after all.

2. McDowell′s Kantianism

The topics on which McDowell [1994, 2009, 2013] and many others with him recently embark broadly concern the epistemological role of the senses in the structure of our experience and the relation between it and higher level acts of consciousness. To talk about senses and experience, however, is already to use highly ambivalent terms. In Kant′s day and age, science and philosophy had less to do with psychological or phenomenological descriptions of passive perception or sensory episodes of consciousness then it does today. For Kant′s purposes, a theory of experience had primarily if not only to incorporate those elements required for mathematical calculations, the description of natural laws, and the limits and transgressions of both.^[7] As is well known, these interests led him to define experience [Erfahrung] in a way that is much more narrow than is ours today: it refers only to that part of our worldly sense-making relevant to the possibility of scientific progression, which in turn is conditioned by a priori intuitions and the categories of the understanding. Experience, for Kant, is a synthesis of an intuition with a concept of which knowledge is the end product.

It is not immediately obvious where we should locate the role of the senses in this Kantian picture of experience. McDowell favors an interpretation of Kant in which the term intuition points to the representations provided by sensible experience, whereas sensations considered on their own are abstractions. If we were to call any of the constituents of experience blind, then, it would have to be the senses and the sensations [Empfindungen] they supply, but only if we render them in an abstract isolation unfaithful to experience. This way, sensations (but not intuitions) are deprived of any epistemological function and by consequence McDowell has excluded one important candidate for the myth of the given. It is in this respect that McDowell departs most from Sellars [1963], who is slightly more favorable of sensations.

Having deprived the senses of the function Kant provided them with in affording intuitions, McDowell still needs to reconsider the meaning of intuition in order not to let it fall prey to the myth of the given. McDowell therefore asserts that what happens in perception is an “opportunity for judging”.^[8] This means firstly that perceptual experience provides representations which form opportunities for discursive acts. But in a second and more profound sense, it means that these perceptual representations are already informed by conceptual activity, regardless of whether a stance of judgment is taken. For McDowell, in any perception, the relevant conceptual capacities that could be put into play upon a discursive apprehension of the represented object are already drawn upon. McDowell has been ambiguous as to whether the content of a perception is different at all from that of a concept. In Mind and World, he writes “that things are thus and so is the content of the experience, and it can also be the content of a judgment […] So it [perceptual experience] is conceptual content”.^[9] But it is not McDowell’s intention to argue that, as some critics have assumed, that the involvement of concepts in perception implies a distance between the subject of the experience and the object he or she intentionally relates to.^[10] For McDowell, the merely perceiving or skillfully coping subject does not actively engage in conceptual activity.^[11] Rather, the idea is that such acts are passively informed by concepts, that is, without having to objectify the content as one does in making a propositional judgment. To see a cup of coffee on one’s desk as well as to skillfully drink from it are acts that are “saddled” with conceptual capacities, which is to say that they are essentially reliant on conceptual capacities in a sense that does not distort their specific phenomenology. That this is so is then explained by adherence to the process of Bildung or cultural development in which human beings engage from their birth onwards.

McDowell’s conceptualist reading of Kant has stimulated plenty new debates over Kantian conceptualism.^[12] Those who, against McDowell, favor a non-conceptualist reading of Kant, often draw on those fragments of the Critique where Kant seems to suppose that intuition can present objects without concepts being in play. These are three of them:

Objects can indeed appear to us without necessarily having to be related to the functions of the understanding. (A89/B122)

Appearances would nonetheless offer objects to our intuition, for intuition by no means requires the functions of thinking. (A90/B123)

The manifold of representations can be given in an intuition that is merely sensible, i.e., nothing but receptivity. (B129–B130)

An additional often heard counter argument is that Kant′s claims on the constitutive role of the understanding for experience are closely tied to his objective notion of experience which does not match ours today. As for instance Hanna [2008] and de Sá Pereira (2013) point out, the relevant passages from the Critique which indicate the indispensable role of the understanding for experience could be interpreted relative to objective experience or knowledge rather than as pertaining to experience in our contemporary sense. The third and, to my view, most important argument against McDowell’s conceptual reading of Kant comes from Hanna’s interpretation of Kant’s pre-critical works. In brief, Hanna [2008] believes that it is essential to Kant’s early position with regard to debates on space that there is an element to experience that is not conceptual. The reason for this is that Kant thought it impossible to conceptually differentiate two objects that are each other’s mirror image.^[13] Thus, in an imagined space with a left and a right hand in it that are otherwise identical, no conceptualist could ever mathematically assert the difference that humans obviously would experience on an encounter with both hands. This problem, addressed by Hanna as the “Two-Hands Argument” led Kant to see that there has to be a non-conceptual element to our experience of space and time. In the Critique, this would be called pure intuition. Later, in What Does It Mean To Orient Oneself In Thinking?, Kant would also refer to it as a sense of orientation.

These three criticisms pose problems for McDowell’s conceptualism both for itself and as an interpretation of Kant. I believe, however, that McDowell’s reading of Kant can be defended against the three main criticisms just outlined. Firstly, the fragments drawn on by non-conceptualists do not as such make a case for non-conceptualism, given that the Critique contains a more or less equal number of passages that state the opposite. They do not by themselves offer a reason to favor these fragments over the others and therefore do not offer unambiguous support for non-conceptualism. The second argument for non-conceptualism is justly based on the difference between Kant’s concept of experience and ours. If we look at the B-Deduction, we find Kant arguing for the transcendental necessity of the categories for all experience. So far, the ambiguity might obtain. However, a bit later on,^[14] in what is sometimes referred to as the second step of the B-Deduction^[15], Kant makes it clear that he wants the categories to play their part in all possible perception as well:

Consequently, all synthesis, through which even perception itself becomes possible, stands under the categories. (CPR B161 – my italics)

The strong claim contained in this sentence is not brought about by an accidental turn of phrase. It is the core of the argumentative structure of the Transcendental Deduction. Kant believes that only this way the transcendental use of the categories for all experience can be maintained. Consequently, the attempt to reduce Kant’s conceptualism to a theory of knowledge in order to make room for a non-conceptualist reading of simple perception, as so many scholars have attempted, is bound to fail.

Hanna’s so-called “Two Hands Argument” (Hanna [2008]) drawn from Kant’s writings on space shows that Kant throughout his pre-critical as well as his critical phase believed experience to contain an element that is intrinsically non-conceptual. The argument, the way I see it, is entirely correct, which means that non-conceptualism has to be appropriated in any complete Kantian theory of perception. Still I do not believe, however, that it should pose a threat for McDowell’s reading. In order to see why, it is necessary to distinguish two kinds of (non-)conceptual content. On the one hand, for a perceptual content to be conceptual could mean it lives up to these two standards: (a) concepts are passively drawn upon in perception (conceptually ‘saddled’) and (b) the content is essentially open to conceptualization (‘opportunity for judging’). This is the kind of content McDowell is interested in when he argues for conceptualism. It is important to see that this reading only includes the intentional or representational content of the experience; it is restricted to that which an act of consciousness is about, its objective content. I will for now call this position general content conceptualism.^[16] But at the same time, any perceptual intentional content could be said to contain “particularity conditions” or what Husserl calls real (reelles) content. For example, right now I might be perceptually related to the cup of coffee on my desk. Although the intentional content of my perception does not change during the few seconds I am looking at the cup, the exact sense in which it is present is in continuous flux. For one, my perspective on the cup changes constantly through the smallest movements of my eyes and body. Given that (i) these particularity conditions are incessantly changing and (ii) I am not intentionally related to them but rather to the object given through them, it is impossible for me to properly conceptualize them. I will call this kind of non-conceptual content real content non-conceptualism. I believe the non-conceptualism Hanna finds in Kant’s writings on space should be taken to match real content non-conceptualism. However, Hanna does not make the distinction just made between real and general content conceptualism. Therefore, he does not see that Kant’s arguments from his earlier works on space can be exhausted in terms of real content non-conceptualism while at the same time maintaining a conceptualist stance at the level of general content. The same distinction could also apply to McDowell. For a general content conceptualist like McDowell, the particularity conditions alluded to by the real content non-conceptualist are irrelevant; they do not constitute reasons for beliefs and therefore do not belong to the space of reasons to start with. Consequently, McDowell could simply set aside the point made by real content non-conceptualists, and the Two Hands Argument thereby fails to make a case against McDowell’s general content conceptualist reading of Kant.

3. The Early Husserl and Non-Conceptual Content

Phenomenologists are, unlike McDowell, primarily interested in providing accurate descriptions of experience. The careful analyses of everyday experiences carried out by some of the most influential phenomenologists of the past have led a number of contemporary phenomenologists to take a critical stance regarding conceptualism. One important phenomenological critique of McDowell comes from Hubert Dreyfus, who in the 2013-book The McDowell/Dreyfus-Debate^[17] turns to the phenomenological descriptions of Heidegger and Merleau-Ponty to argue against McDowell’s conceptualism. Dreyfus focuses specifically on descriptions of pre-reflective, skillful action. For one, he follows Heidegger that we do not have to think about the doorknob on the door in order to use it to enter or leave a room. In fact, the doorknob does not have to be apprehended at all. Such ‘absorbed copings’ are mindless activities; they involve no intentionality, that is, no subject/object-relation, and therefore, it is claimed, no rationality.^[18] The conclusion he then draws is that the operations of the understanding are not involved here, and hence conceptualism is false. I have already shown in the previous section why I believe this argument cannot be held against McDowell. The latter nowhere claims that it would be a necessary condition for conceptual content that an intentional distance between subject and object should obtain. Dreyfus, and Schear [2013] likewise, miss out on McDowell’s notions of Bildung and “second nature”, which point to the fact that human beings can passively employ capacities that belong to the understanding, that is, without actively contemplating the relevant content involved, and thus without distorting the specific phenomenology of skillful coping of which most of our everyday lives are comprised. Schear and Dreyfus leave the entire issue unaddressed whether the pre-reflective action of opening a door would have looked the same – or would have been possible at all – without passive knowledge of how door knobs work, knowledge of what lies behind the door, and knowledge of why I would want to go there. McDowell wants to argue that rationality of this kind pervades skillful coping and that these actions are therefore a part of the “space of reasons” – which is the key term in McDowell’s conceptualism – but Dreyfus´s phenomenological argument does not connect to that.^[19]

Hopp [2010, 2011] provides a different argument for non-conceptual content based on Husserl’s Logical Investigations, which he similarly directs against McDowell. In Perception and Knowledge (2011), Hopp focuses specifically on perceptual content and Husserl’s phenomenological account of “fulfillment”.^[20] Hopp asserts that a perceptual content does not have to be thought about in order to be there for me. What’s more, the perception is precisely not thinking and not conceptual: it is not thought – which is empty or signitive, as Husserl puts it in Logical Investigations – but rather something else, namely something that can epistemically fulfill an empty thought. For instance, when I think about a white floor, I intend something merely emptily, without the white floor actually being given to me sensibly. If this empty thought of a white floor is now combined with a corresponding perceptual givenness of a white floor, then a “synthesis of recognition” takes place in which the empty thought is “fulfilled”. Hopp believes that perception must have non-conceptual content because it can play this role of epistemic fulfillment in a way mere thought cannot. Conceptual thought by itself is empty; perception, however, can offer a distinctive surplus, which must therefore be non-conceptual.

According to Hopp, the conceptualism of for instance John McDowell [1994, 2009] is “utterly incapable of explaining why perceptual experiences play this distinctive and privileged role in the production of knowledge”,^[21] because it allegedly overlooks the important role of fulfillment perception plays:

I can compare my belief that an apple is red with the apple that is red, and when I do, I know something I would not have known merely by believing that that apple is red.^[22]

The above fragment is an argument for a particular surplus of perception relative to thought. Its claim, it seems to me, is convincing. Seeing that an apple is red offers knowledge of a kind I would not have had merely by connecting the ideas “apple” and “red” in my head and trying to belief their combination to be actually so. However, it is entirely unclear why McDowell would have to disagree with this. Conceptualism, as I have shown already, is a thesis about the kind of things that can constitute reasons for beliefs. It does not imply denying that intuition is something very different from thought in a way that would oppose Husserl's early theory of fulfillment. In fact, McDowell knows well that in perception something is given while in thought there is not.^[23] A denial of the specific function of intuition in knowledge would rather be a part of the kind of frictionless coherentism McDowell is eager to avoid. Hopp is certainly right that when “I come home to find my basement flooded, the proposition ‘My basement is flooded’ occurs to me because I am presented with my flooded basement”.^[24] But this is precisely the kind of phrase McDowell would hold in support of conceptualism: does the fact that I already see a flooded basement and the fact that this can function as a reason for believing it to be so not by itself indicate that some form of rationality is involved already at the level of perception? After all, what I see does not await for me to endow it with sense: I see a flooded basement, and because I do, that perception can fulfill my thought of it. Even if Hopp does not accept this as sufficient ground to speak of conceptual content at the perceptual level, he does appear to agree that the perception is a part of the space of reasons – and that is all McDowell is truly after.

Both Dreyfus’s Heidegger-inspired case for non-conceptual content and Hopp’s one based on the early Husserlian account of fulfillment present interesting ways of thinking about non-conceptual content. However, the kind of non-conceptual content they argue for have little bearing on McDowell’s conceptualism. In the next section, I will consider the phenomenology of the later Husserl and some contemporary readings of him that commit to his agreement with McDowell.

4. The Later Husserl and Non-Conceptual Content

About the last twenty years of his life, Husserl reinvented phenomenology by working on what he called passive synthesis: the phenomenological analyses of experiences that the subject does not actively participate in, as for instance in skillful coping and visual perception of some kinds. This allowed Husserl to conceive of experience as layered; building up from passive sense-makings of which the experiencer is unaware to conceptual activities in which he or she actively engages. Roughly speaking, on the basis of his late masterpiece Experience and Judgment, three such layers of experience can be distinguished, which are more or less hierarchically structured: 1. Primary passivity^[25]: the lowest realm of sense-making that happens in complete independence of the active subject; 2. Explicative Contemplation: acts of perception which are realized without the subject’s explicit command but which are nevertheless consciously experienced in some way by him or her. This is, as I take it, the usual case of visual perception; 3. Conceptual activities such as judgments of existence which are brought about by the subject.^[26]

One important transcendental mechanism Husserl thinks governs the lowest two layers of experience is what he refers to as ‘types’. I shall not here try to explain the complex analyses Husserl goes through in order to make that notion plausible, but instead restrict myself to what types are said to do.^[27] Husserl believes that every time I experience a new kind of object, a new type is automatically installed through which objects that are similar to it will immediately be apprehended as familiar and known.^[28] For the adult human being, most visual perception is said to be governed by types. We do not perceive strange, empty forms first which we then actively endow with meaning. Rather, perception already presents us the object as something that we are familiar with. This does not mean we perceive a cup of coffee immediately as being a cup of coffee, which would involve the apprehension of concepts and categorial structures.

For the question of the conceptual saddledness of perception, phenomenologists may thus turn to Husserl’s notion of type to argue that we never have blind intuitions, as McDowell interprets the famous Kantian phrase that ‘intuitions without concepts are blind’.^[29] Therefore, some have argued, intuitions for Husserl too have to be conceptually endowed. A second argument for Husserlian conceptualism, and the most important one also presented by Husserl scholars Barber [2008] and Mooney [2010], is based on a specific interpretation of the relation between these three different strata of experience. Both claim that Husserl does not consider the various layers of experience as truly distinct, from which it would follow that primary passivity – the lowest kind of sense-making which involves no conscious partaking of the subject – can also never be wholly separated from conceptual activities. Mooney adds to this that primary passivity is something we once had to go through but which we have left behind in become adult.^[30] This would mean that the kind of passive synthetic processes Husserl describes as taking place without us being aware of it are in fact abstractions, for in reality they are saddled with concepts much in the way McDowell would have it. If this argument is sound, as Barber and Mooney believe, then the later Husserl would be a conceptualist just like McDowell.

In what follows, I want to show that these arguments are incorrect, but also that Husserl could still be a conceptualist in McDowell’s sense. Let me start with taking on the first argument, that the guidance of our everyday perceptions by types would come down to tacit use of concepts. Husserl regards the realm of passive experience as dominated by an interplay of among others bodily movements, bodily affections and anticipations which together structure the objects of experience before the experiencer is aware of it.^[31] Very often indeed, these passive organizations are guided by earlier conceptual activities through the permanent installment of a so-called type. For instance, I would not have perceived the room I just entered as full of computers had I not once learned what computers are through conceptual effort. Because I once learned the appropriate concept, I can now simply see that the room is full of computers without having to actively think that this is so. The kind of types involved in this case Husserl also calls a secondary passivity.^[32] In my opinion, this corresponds perfectly to McDowell′s notion of perceptions saddled with spontaneity. But the terminology already indicates that there is something more. For Husserl’s phenomenology shows that the mechanism of type is not in fact reliant on conceptual activity.^[33] It is also possible to experience new objects passively, whether through visual, auditory or tactile perception, and still attain a type which will henceforth steer my perceptions of similar formations.^[34] For instance, a pre-linguistic infant (or a cat for that matter) does not have to have learned the concept “rabbit” in order to have some passive predictions about how a rabbit will proceed its way around the garden. This is not to say that the child can conceptually calculate or predict where the rabbit will end up. The child’s body, however, will be able to produce a response in anticipation to the movement of the rabbit when necessary, for instance when the rabbit approaches the child. The point here is that the body can passively structure objects in the visual (or otherwise perceived) environment and produce responses to these objects without any concepts having been in play at any level. This is an example of types at the level of primary passivity: forms of sense-making that are not conceptual or conceptually endowed but nevertheless successfully provide relevant information about the environment by way of association with earlier experiences.^[35]

Unlike McDowell, Husserl has very specific ideas about what counts as a concept and what not. He thinks of conceptual activity as a complex act actively performed by the subject through which something is intuited that is not itself sensibly given. By ´´not itself sensibly given´´ Husserl means that to conceptualize the cup of coffee in front of me is to make thematic categorial structures that obtain between the different aspects of the perceptual object and me. For instance, in saying ‘this is a cup of coffee’, I judge about the being of the cup, which is not as such given to me perceptually, that is, the being does not and cannot figure in receptive experience. Conceptual judgment therefore involves something that cannot be receptively constituted. Husserl writes: ‘objectivities of the understanding [concepts] can never be originally apprehended in a mere act of reception; they are not preconstituted in pure passivity’.^[36] This is not to deny, as I have just shown, that concepts can work their way back into passivity and become a secondary passivity. McDowell’s position seems therefore well supported by Husserl’s later phenomenology. However, McDowell’s theory demands that intuition is thoroughly determined by conceptual capacities. But this is a claim which, taken at face value, Husserl´s phenomenology cannot support. For as I have shown, Husserl also has a phenomenology of primary passivity, which is unaffected by concepts, secondary passivity or the subject´s Bildung. Phenomenologically speaking, then, primary passivity must be a kind of non-conceptual content, even, it seems, on McDowell´s own terms.

It should be clear that the idea of primary passivity and the notion of non-conceptual content I have ascribed to it do not imply an utterly unstructured reception of blind intuitions or a commitment to the myth of the given. Husserl does not ascribe a basic epistemic role to primary passivity, as in fulfilling the foundationalist′s desire for an unmediated apprehension of sensuous content. The idea of primary passivity and the phenomenological mechanisms that operate in it are only an attempt at phenomenological description of how entirely passive experience works.

What I have said thus far should already be sufficient to provide an answer to the second argument posited by Barber and Mooney that primary passivity would be an abstraction from real experience in which the conceptual is always involved. A further argument against their case could, I think, be drawn from the general methodical principles of Husserl’s phenomenology. I find it hard to conceive of Husserl’s analyses of passive synthesis as abstractions in the sense Barber and Mooney endorse, given that phenomenology cannot analyze that which is not immediately and with certainty given upon reflection. If passive synthesis would be an abstraction, then it would not be given upon reflection and therefore could not be analyzed phenomenologically in the way Husserl wants to analyze it. I find Mooney’s additional characterization of these writings as pertaining to something that we have left behind in becoming an adult particularly implausible. The reason for this is, once more, that the adult phenomenologist cannot analyze the experiences of children nor of himself as a child, for these experiences are not given to him immediately upon reflection, as is essential to the phenomenological method.

In contrast with the kind of non-conceptual content Dreyfus, Schear and Hopp argue for, the one I developed on the basis of Husserl's analyses of primary passivity does not ignore the specifically passive sense in which McDowell sees concepts integrated in experience. Therefore, if Husserl's theory is correct, and it plausibly results, as I have suggested, in a kind of content unaffected by concepts and the cultural development of the subject, it would appear to present a viable alternative to McDowell's conceptualism. However, I do not think that this need be the case. It is useful to recall that McDowell's conceptualism is a theory about experience in the light of justification, that is: of what constitutes a reason for a belief, i.e. what belongs to the space of reasons and what not. The fact that the kind of sense-makings involved in primary passivity are not affected by the conceptual abilities of the subject also establishes that those very same sense-makings cannot be a part of the space of reasons. Therefore, Mooney’s and Barber’s conclusions turn out to be right after all that the later Husserl does not contradict McDowell's conceptualism.

5. Conclusion

As I have shown in the first parts of this paper, McDowell’s conceptualist reading of Kant cannot easily be rebutted by reference either to Kant’s pre-critical or critical writings. The arguments for Kantian non-conceptualism I discussed are the best I know of, but I do not think they are good enough to trouble the conceptualist. Hanna's argument from incongruent counterparts can, I have argued, be appropriated by distinguishing between Kantian real content non-conceptualism and general content conceptualism. McDowell's conceptualist reading of Kant is one that focuses on general content conceptualism, that is: on what can figure as a reason for believing something to be the case. Because Hanna's argument can be made to fit real content non-conceptualism, it fails to touch upon these issues, and therefore it cannot trouble McDowell.

The phenomenological attempts at countering McDowell's conceptualism that I have discussed were focused specifically on skillful coping and on the structure of fulfillment as Husserl presented it in Logical Investigations. Although these cases successfully establish interesting phenomenological notions of non-conceptual content, I have argued that it is wrong to take them as having much bearing on McDowell's conceptualism, for the reason that they do not take the specifically passive, habitually sedimentated sense of conceptual activity properly into account.

The later Husserl offers better resources to discuss the conceptuality of experience in the sense McDowell is after. Contrary to the interpretations offered by Barber and Mooney, I have argued that, phenomenologically speaking, primary passivity can be interpreted as a stratum of experience filled with non-conceptual content, even though this need not contradict the kind of conceptualism McDowell proposes.