A key question in the philosophy of perception concerns the nature of empirical content and its relation to the content of thought. Conceptualists answer that the content of perceptual experience is exclusively conceptual in nature. On their view, conceptual capacities that belong to spontaneity already are operative in receptivity, albeit in a passive manner, which guarantees that the contents of experience and judgment are formally of the same kind. This enables us to justify empirical judgments non-inferentially simply by pointing to the actuality of their content in experience. Non-conceptualists deny this and maintain that experience and judgment operate with categorially distinct kinds of content. On their view, the contents of judgment and thought are conceptual, while perceptual experience is an essentially non-conceptual informational state induced in the subject by the world’s efficient-causal impact on its senses. The subject, however, is capable of conceptualizing the non-conceptual contents of experience, i.e., of transforming them into conceptual contents that can figure in judgment and thought. Although an expression of our spontaneity in the formation of concepts, this act of conceptualization is limited in its freedom by the non-conceptual contents receptively given to us in experience. Such contents thereby act as a foundation for empirical judgment and belief, which non-conceptualists maintain is necessary to ensure the objectivity and world-directedness of our thinking. Conceptualist, however, doubt that such a categorial transfiguration is possible and argue that the very notion of non-conceptual content, as independent of the subject’s rational capacities yet still capable of rational uptake, is an instance of the ‘Myth of the Given’. Non-conceptualists, on the other hand, worry that conceptualism entails a subjective form of idealism that renders the world a mere product or reflection of thought. The debate between both views is an extensive and complicated one.Footnote 1

In this paper, I focus on one central point of contention, namely whether conceptualists can give an adequate account of the fine-grained phenomenology of experience. Gareth Evans first formulated an objection to conceptualism along these lines:

[N]o account of what it is to be in a non-conceptual informational state can be given in terms of dispositions to exercise concepts unless those concepts are assumed to be endlessly fine-grained; and does this make sense? Do we really understand the proposal that we have as many colour concepts as there are shades of colour that we can sensibly discriminate?Footnote 2

Put differently, the concepts that supposedly constitute the content of experience will have a certain generality. Yet, experience is full of particular detail and fine-grained sensory distinctions, and it is questionable whether our conceptual scheme can keep up with this degree of particularity. If not, then experience exhibits a far greater density of content than our concepts could account for, and this surplus content therefore cannot be conceptual. John McDowell has resisted this objection by arguing that our conceptual scheme in fact matches the fine-grained character of experience because it includes demonstrative concepts, such as ‘colored thus’ or ‘this shade of color’, which can achieve the desired degree of particularity. Yet, his reply has met with a plethora of further objections and the prevalent opinion now seems to hold that non-conceptualism has won the debate on this point.

I believe that McDowell’s critics underestimate the true force of his reply. To show this, I introduce a new hylomorphic reading of McDowell’s account of experience and argue that the objections to his reply all depend on a specific empiricist assumption that is untenable because it conflates the form of experience with its content. Consequently, I begin by reconstructing McDowell’s account of experience. Since the notion of conceptual or—as I prefer to call them—rational capacities plays a central role in this account, I first outline the nature of such capacities (§1). I then explain how they structure our conceptually shaped receptivity on McDowell’s account and introduce my hylomorphic reading of it (§2). Subsequently, I present McDowell’s reply to the objection from fineness of grain and provide a systematic survey of the additional objections raised against it (§3). I then identify the shared assumption underlying all of these objections by further explicating the hylomorphic structure of experience (§4) and subsequently answer them in detail (§5). I conclude with some reflections on why conceptualism so understood is dialectically superior to non-conceptualism (§6). The ultimate end of this paper is to introduce experiential hylomorphism as a promising yet neglected option in the philosophy of perception.

1 The nature of rational capacities

Given his quietism, McDowell is somewhat reticent about the nature of rational capacities, despite their central role in his account of empirical content. Nevertheless, he regularly attributes a variety of features to them. Thus, he connects them to concepts and maintains that we exercise them with a certain logical togetherness in acts of judgment. He disambiguates the meaning of these concepts into their Fregean sense and reference, or Sinn and Bedeutung. He understands meaning generally in terms of rule-following and therefore situates concepts and their associated capacities within language-games and practices. Finally, he connects rational capacities to spontaneity and our responsiveness to reasons and norms.Footnote 3 Yet, how does this wild mixture of Geach, Frege, Wittgenstein, and Kant fit together?

In giving a unified account of these features, I draw substantially on work by Andrea Kern and Sebastian Rödl, which appears sufficiently compatible with McDowell’s approach.Footnote 4 I also foreground the recognizably Kantian provenance of his project. On the resultant picture, rational capacities essentially are capacities that actualize a concept, in that their exercise consists in the self-conscious execution of a rule of synthesis. Such rules are norms for combining elements into the determinate kind of unity designated by the concept; norms that determine how we ought to perform an act of synthesis if it is to realize this unity. Rational capacities and their exercises therefore have a teleological structure, and the specific telos that individuates each capacity is the actuality of a particular kind of unity. Since all rational capacities share this basic structure, McDowell sometimes elucidates features of the theoretical capacities that interest him by analogy to practical capacities, most notably in discussions of disjunctivism.Footnote 5 I will follow the same strategy and first elaborate on this basic structure for practical capacities, since they provide the simplest illustration, and then draw out some implications for theoretical capacities.

We exercise rational capacities for action, such as for dancing or baking, in particular actions. Such actions often run through several steps or phases, which come together in a particular kind of temporal unity when executed successfully. Dancing consists in a succession of movements that aim at actualizing or instantiating a certain choreography, and baking involves a course of action that comprises several steps, as describable in a recipe, the completion or actuality of which is productive of baked goods. The different phases of such actions are constituent parts of these temporal unities and, as such, related internally, in that their nature is determined by their position or function within the whole and cannot be understood apart from it. Thus, a certain rotation of the body only counts as a pirouette when viewed within the functional context of ballet and a certain movement of the hand only counts as folding in whipped egg whites because of its role in the practice of baking. Practices thereby impart a meaning or Sinn to their constituent acts, which renders them intelligible by functionally relating them to other such acts.Footnote 6

A practice is a system of rules that constitutes the order and cohesion among the phases of a distinctive activity and thereby determines how to perform it successfully. In their totality, these rules describe the paradigmatic or ideal structure of the temporal unity enacted by executing the activity. Concepts of practices capture this structure and therefore signify a determinate form of action or practice-form, i.e., the eidos of the respective activity, which is exemplified more or less well in particular performances. Hence, rational capacities actualize such concepts by enabling our self-conscious execution of the practices captured by them, and we acquire such capacities by rehearsing and habituating the rules that govern these practices. Their self-conscious character entails that an agent, in order to possess such a capacity, must understand the relevant concept, i.e., be able to act from a practical representation, usually implicit, of the respective rule(s) of synthesis. While itself general, this representation plays both an explanatory and a normative role in relation to particular acts.

Its explanatory role ensures that it is no accident when a rational capacity generates a course of action that conforms to a practice and exemplifies its eidos more or less well. For the phases of that course of action do not represent an accidental succession of acts but form a progression that is derived from the agent’s consciousness of the rule and understood as rationally necessary in light of it. This derivation is explicable as a practical syllogism. Since the agent in exercising their capacity aims at the actuality of a practice-form, the rule of synthesis describing that form, i.e., the ideal structure of the intended action, enters into its major premise. Thus, the agent may intend the realization of a particular dance choreography or cake recipe, where the concepts of these actions capture the distinctive succession of steps one must take to perform them, i.e., their ideal structure. The syllogism’s minor premise then contains the particular circumstances under which the agent exercises their capacity by performing these actions. This combination of premises allows the agent to view their current circumstances in light of the practice-form, i.e., to locate the here and now at the relevant point within the ideal structure of the action and thereby to map out what is necessary to progress with realizing it. The baker, after a brief absence, may recognize that the egg whites are folded in already and the next step is to pour everything into the baking pan. The premises together thus yield a conclusion in the guise of an essentially practical determination of the particular act(s) the agent must subsume under the rule if it is to be exemplified successfully here and now; a determination that is an essential part of what the capacity is a capacity to do.Footnote 7

Since possession of a rational capacity, as acquired by practice and habituation, thus involves proficiency in the application of a rule of synthesis, such capacities act as causes of the actuality of their respective practice-forms. It allows the agent to place potential acts in the wider context of this form, as representing a necessary element in its ideal progression, and thereby enables the agent to continue spontaneously a pattern of action by generating instances that conform to and exemplify it. Crucially, this does not require the agent to act on an explicit or theoretical representation of the rule. Rather, their capacity will manifest in a practical representation of the rule that consists in the perceptually situated recognition of a practical pattern and informs a suitably experienced agent of what their next steps should be. Put differently, once an agent has habituated a rule successfully, anything relevant to determining its proper execution will become perceptually salient to them due to its associative connection with the rule, i.e., with the other elements unified in the respective practice-form. That is, the agent is able to see immediately how whatever is perceptually present to them fits into the wider context or structure of their intended action because their productive imagination adds the non-present elements of that structure by means of association. It thereby imaginatively enriches the agent’s experience of the situation and renders it pregnant with practical implications for executing the rule, in the guise of practical reasons that suggest ways of proceeding and lead beyond what is immediately present to the agent’s senses. The habitual character of the agent’s rule-following, their knowledge-how, then consists in spontaneously following the lead of such a chain of practical-perceptual associations without reflective resistance. Habituating or internalizing a rule thus canalizes the agent’s spontaneity in specific ways. Rational capacities therefore represent particular shapings or formations of our spontaneity, which have been stabilized by habituation and find expression in such ‘automatized’ rule-following when we perform particular acts of synthesis.

Given the syllogistic structure of their actualizations, rational capacities do not merely explain why we perform particular acts of synthesis in one way rather than another, they also justify the shape these acts take on any occasion. For the elements combined in a practical syllogism stand in rational relations to each other, in that its major and minor premises rationally ground its conclusion. On the one hand, the practice-form in the major premise stands to particular acts falling under it as a ground or reason in the sense of a norm. Just as pirouettes presuppose the practice-form of ballet, so any movement is identifiable as an act of a particular kind only when viewed within the wider context of the relevant practice-form, and therefore taking its assigned place within that form is at least partially constitutive of any such act. Hence, form and act are related internally in such a way that the form normatively binds the acts that instantiate it by requiring them to be embedded functionally into the whole in a specific way. This entails that a practice-form can justify the shape such acts take in a given situation as rationally necessary for actualizing it.

On the other hand, such norms also contain the principle of their own actualization by determining the content of the minor premise, i.e., by governing the selection of which facts are relevant for enacting the practice-form under the given circumstances and therefore count as reasons for executing it one way rather than another. A fact counts as a reason here when it has implications for how we must proceed to complete the temporal unity of the action we aim at. Hence, that the crust has browned nicely counts as a reason for taking the cake out of the oven rather than for turning up the heat because I aim at baking and not burning it. Its status as reason ultimately derives from the function it plays in shaping the concrete realization of a particular act of synthesis and therefore is rooted in the teleological structure of our acts. Together, major and minor premises ground the syllogism’s conclusion by determining how we need to order particular acts to accomplish their synthesis in the temporal unity described by the practice-form. Rational capacities, as products of habituation, therefore involve a stable and determinate manner of transitioning from grounds to consequences and thus a responsiveness to reasons and norms that is equally spontaneous and habitual. Again, once an agent has habituated a rule, such reasons are perceptually salient to them due to their associative connection with the other elements interrelated by it.

Of course, not every act of synthesis aims at a form of temporal unity, i.e., unifies diachronically disparate elements. It also can bring synchronically disparate elements to the unity of a concept, as when we exercise a rational capacity for theoretical cognition in recognizing elements of a sensible manifold as a magpie, cup, or tree. The actualization of such capacities also has a syllogistic structure. In this case, the syllogism’s major premise is filled in with the form of the relevant object, its eidos or ideal structure, as it is captured by the object’s concept, while the syllogism’s minor premise is filled in with the sensory circumstances under which the capacity is exercised. This enables the subject to align the elements of the sensible manifold with the ideal structure of this object, e.g., with the structure characteristic of magpies. It thereby unifies and grasps this manifold as a particular empirical object of this kind. In this, the subject again is aided by its productive imagination, which supplies by association the elements of that structure not sensibly given, e.g., because they form part of the object’s backside. The conclusion of the syllogism thus consists in the recognition of a particular empirical object and is articulable in a judgment, such as ‘This is a magpie!’, which is self-conscious of its own ground precisely because it represents the conclusion of a syllogism.

Exercising a rational capacity for theoretical cognition in this manner thus involves a perceptually situated recognition of a pattern or Gestalt in a sensible manifold. Since all elements of this manifold are present synchronically, grasping them in a unitary manner as a determinate empirical object requires no temporal progression but is done instantaneously. Exercising the respective theoretical capacity therefore is no temporally extended act here but a momentary or instantaneous act; and the same holds for the conceptual unity of sensory impressions it brings about. The act constitutes this unity by structurally interrelating the elements of the manifold according to a rule of synthesis, which describes the formal structure of the object and is captured by the object’s concept. It thereby renders these elements intelligible as constitutive parts of a whole. Executing the rule thus imparts a specific meaning or Sinn to the matter synthesized by subsuming it under a concept.

This allows us to adapt Frege’s distinction between a concept’s Sinn and Bedeutung to the ontology of the conceptual underlying the present account, which considers concepts neither mental representations nor abstract objects or classes but abilities of the subject which it manifests in particular acts of synthesis, of grasping a manifold in a unitary manner. A concept’s sense and reference then characterize the structure of such acts.Footnote 8

For Frege, a concept’s reference is its relation to some bit of extra-linguistic reality, while its sense is the mode in which it presents that reality, i.e., the specific form the reference relation takes.Footnote 9 On the present ontology of the conceptual, we can understand a concept’s reference as an act of the subject by which it singles out the relevant bit of extra-linguistic reality. This singling out is done by executing the rule of synthesis corresponding to the concept, i.e., by an act of synthesis that constitutes the very object falling under the concept and referred to by it. Hence, we can identify the concept’s sense with the specific form this act of synthesis takes, since this is the mode in which the act presents the concept’s referent. Now, the form of an act of synthesis is its teleological structure, which consists in the way it interrelates the elements of a manifold to attain its specific telos, viz. the actuality of the kind of unity designated by the concept. Thus, an act that aims at visually recognizing a magpie does so by interrelating and apprehending the relevant elements of a sensible manifold in the structural unity designated by ‘magpie’. This unity both explains and justifies, in the guise of a norm, why the elements unified are selected and belong together in this way and hence why the concept refers in the way it does. Since this structural unity is described by the corresponding rule of synthesis, a concept’s Fregean sense just consists in this rule and, ultimately, in the eidos traced out by it and captured by the concept.

Since these rules form part of intersubjectively shared practices, Fregean sense is not a private matter but subject to public criteria in Wittgenstein’s sense. We learn to understand it, and thus how a given concept refers, by initiation into practices such as baking, dancing, or recognizing magpies. This initiation consists in internalizing the rules constitutive of a practice by repeated performance and habituation, which leads us to develop the rational capacities required for following these rules. As stated above, following such rules, and thus understanding a concept’s sense, does not require a discursively explicit or theoretical representation of the rule but only the ability to apply it successfully in particular acts of synthesis; an ability consolidated and stabilized by acquiring the respective rational capacity. Such capacities thus safeguard our understanding of Sinn and thereby enable us to grasp the world in action and cognition. In other words, they essentially structure the specific teleology or intentionality of consciousness, i.e., our relation to its objects. Crucially, when I speak of objects, empirical objects, or objects of experience throughout this paper, I primarily intend a logical notion of object, in the sense of ‘something’s being there for a subject’, irrespective of whether this is a substance, property, quality, relation, or other kind of entity. Thus, I do not restrict such objects to ‘medium-sized dry goods’.

Finally, it is important that the above distinction between synchronic and diachronic unification does not mark a generic difference between practical and theoretical rational capacities, since the latter are equally open to exercise in discursive thought or argumentation, i.e., in temporal syntheses. Whether we exercise them synchronically or diachronically, our theoretical capacities fulfill the same explanatory and normative functions as our rational capacities for action. They likewise represent particular shapings or formations of our spontaneity, which are acquired by habituation and involve both an ‘automatized’ rule-following and its concomitant responsiveness to reasons and norms. Thus, possession of rational capacities generally places subjects in what McDowell calls the logical space of reasons, which is constituted by rational relations among conceptual contents, i.e., Fregean senses, and structured in its specific topography by our practices and their ends. With this rough account of rational capacities in hand, I now will consider how such capacities are operative, on McDowell’s view, in shaping our experience of the world.

2 McDowell’s account of experience

McDowell’s account of experience is motivated by the question how experience, as a product of receptivity, can have standing in the space of reasons, i.e., act as a reason for belief that normatively binds the freedom of spontaneity in making judgments about the world. His basic contention is that, since this space is structured by our rational capacities, experience can play this justificatory role only if our receptivity is itself a rational power of cognition, which essentially draws on such capacities in its actualizations. Consequently, McDowell maintains that ours is a conceptually shaped receptivity, which entails that the experiences we receive by the world’s impact on our senses are conceptually structured. Put differently, experience relates us to its objects essentially under a conceptual mode of presentation, by means of Fregean senses that form the content of experience. This enables us to transition such content immediately into judgments and beliefs about these objects, which are justified non-inferentially because our experience of them has the very same content.

How do our rational capacities accomplish this? In §1, rational capacities were defined as capacities that actualize a concept, in that their exercise is the execution of a rule of synthesis. Actions, including acts of judging and inferring, are syntheses the subject performs actively, i.e., spontaneously and voluntarily, by choice. Yet, McDowell holds that rational capacities often also actualize themselves passively, i.e., involuntarily and habitually, without an act of choice, and that this is what typically happens in experience. That is, empirical objects usually are constituted through a passive synthesis, in which the subject involuntarily grasps the elements of a received sensible manifold in the structure described by the appropriate rule of synthesis, in a habituated and immediate response to the world’s impact on the subject’s senses and its locus of attention. Since every rule of synthesis simultaneously is a rule of differentiation, it thereby separates these elements out from the totality of the manifold and experiences them as belonging together. Hence, the subject is aware of these elements, from the outset, in the unity of the ideal structure described by the rule, and the empirical object so constituted thereby exemplifies its eidos or concept.Footnote 10

Consequently, the passive and involuntary operation of rational capacities in receptivity aims at actualizing their associated concepts just as much as their active and spontaneous exercise does, and the actuality of these concepts is disclosed here in the guise of the empirical objects we apprehend by means of them. In this sense, the teleology or finality of rational capacities for theoretical cognition generally consists in apprehending the actuality of their corresponding concepts, in either experience or judgment.

Objects of experience exhibit the actuality of their concepts because the passive actualization of our rational capacities, in structuring and unifying the sensible manifold, imparts a specific Fregean sense to it. On McDowell’s view, this Fregean sense is the content of experience; its mode of presenting objects to us. Crucially, such content is not constituted by a fragmented actualization of individual capacities but through the simultaneous operation of a multitude of them, which are not actualized in isolation from each other but jointly and in a certain logical togetherness. Hence, experience does not disclose the actuality of isolated contents such as ‘pink’, ‘cube’, and ‘ice’ but their unity in the guise of a pink ice cube, i.e., of an object of experience that exemplifies several concepts at once and in a togetherness merely separable in thought. This togetherness of several concepts in experience exhibits a logical unity that the subject can reconstitute, in principle, by means of an active synthesis of the same concepts in a judgment.Footnote 11 In his original account, this led McDowell to attribute an essentially propositional content to experience:

In a particular experience [...], what one takes in is that things are thus and so. That things are thus and so is the content of the experience, and it can also be the content of a judgment.Footnote 12

On this view, experience and judgment both operate with the same forms of conceptual unity, and therefore the synthetic unity of conceptual contents in experience also exhibits a propositional structure, i.e., corresponds strictly to the possible ways in which we can join subject and predicate in judgment. Hence, experience enables us to grasp particular propositions in a special manner. It consists in the sensible intuition of facts, of states of affairs, and presents them as obtaining, as actually being the case. Thus, it can be an object of my experience that there is a storm clamoring outside now or that I have a pink ice cube in my drink. Since experience, for McDowell, is our immediate access to the world, this seems to entail that the world itself is structured propositionally. We can picture it in a Tractarian manner as everything that is the case, as the totality of all facts obtaining. Hence, a true empirical judgment, since it adequately grasps a fact, also discloses an aspect of the world as it actually is.Footnote 13

However, certain objections to this account have moved McDowell to modify his view in important respects. Thus, some critics doubt whether his original account allows for a clear distinction between experiences and beliefs. Since empirical content is propositional, experience itself apparently amounts to an odd kind of propositional attitude, namely the involuntary holding to be true of a proposition, as caused by an actualization of our receptivity due to the world’s impact on our senses. If so, then experience effectively saddles the subject with a passively held empirical belief. The content of that belief could then enter into inferences and ground other beliefs, while its involuntariness presumably would serve to avert an infinite inferential regress. McDowell’s view would differ then only terminologically from the sort of coherentism defended by Donald Davidson since such ‘experiences’ merely amount to “beliefs in disguise”Footnote 14 with an efficient causal origin; a view McDowell explicitly rejects.Footnote 15

Others object that McDowell’s original account cannot capture the true richness of experience because it pictures its content merely as the sum of all propositions exemplified in it. In principle, then, it should be possible to fully articulate any experience’s content in a finite number of empirical judgments. Yet, when we consider the phenomenal differentiation of experience in its intricate detail, it seems doubtful whether even the most complex conjunction of propositions could ever exhaustively describe its content. For it is always possible to give a more detailed and precise description of what we experience, to catch more of its gradual transitions and inconspicuous differences. Since experience is always richer and denser in content than any description we could give of it, any experience potentially warrants an infinite number of empirical judgments. Yet, if empirical content is not fully articulable in propositional terms because there always remains an as of yet unarticulated residue, it cannot be essentially propositional.Footnote 16

In response, McDowell has undertaken two modifications to his account, only the second of which I accept in revised form.Footnote 17 First, he now upholds a quasi-Kantian distinction between categorial and empirical concepts and restricts the content of experience to the former. Categorial concepts, on his view, comprise our concepts of the proper and common sensibles. The latter include formal concepts of objects, such as ‘substance’ or ‘animal’, by which we unify the former into modes of presentation that relate us to empirical objects. Empirical concepts, such as ‘cardinal’, then specify these formal concepts into less formal, material concepts by subdividing and categorizing the empirical objects constituted by them. In doing so, empirical concepts remain external to experience proper and merely supervene on its categorial content, which is why McDowell associates them with ‘recognitional capacities’ operating on, rather than in, experience. McDowell’s stated motivation here is to guarantee that two subjects will have essentially ‘the same’ present in experience even if one of them lacks such contingent concepts as ‘cardinal’ or ‘okapi’. My discussion of the objection from phenomenal difference in §5 will remove this motivation by showing that it equivocates on ‘the same’. In addition, this modification also threatens to undermine McDowell’s non-inferential account of empirical justification since judgments containing ‘merely empirical’ concepts are then no longer justifiable in recourse to the passive actualization of these concepts in experience. Consequently, I will reject McDowell’s first modification and continue to treat all concepts as potentially constitutive of empirical content.Footnote 18

McDowell’s second modification rescinds his original claim that empirical content is propositional yet retains his basic contention that it is conceptual. For he now distinguishes two kinds of conceptual content, viz. the propositional content of judgment or belief and the intuitional content of experience. Both qualify as conceptual and therefore belong under the same genus, first, because they both consist in the Fregean sense of the concepts unified in either judgment or experience, and second, because the respective forms of propositional and intuitional conceptual unity, while categorially distinct, still share an analogous structure. For they are products of the same ‘unifying functions’, which merely actualize themselves differently in each case. Thus, the same logical ‘function’ or form of togetherness that joins subject and predicate in a judgment also unites a substance and its properties, under a conceptual mode of presentation, in an intuition. This is meant to preserve the internal relation between judgment and experience required by McDowell’s account of justification, i.e., the essential availability of empirical content to judgment, and simultaneously to establish a formal difference between experience and belief in terms of their content. For McDowell, their contents are formally distinct chiefly because propositional content is discursive, i.e., actively synthesized, and therefore articulated or explicit, while intuitional content is passively given and therefore unarticulated or implicit, although open to discursive explication.Footnote 19

It is doubtful, however, whether McDowell’s distinction between explicit and implicit is sufficient to mark a difference in the very nature of content rather than its mode of givenness. Either way, as some critics maintain, we face a dilemma. For if the contents of experience and judgment are formally identical, then it seems that intuitional contents are merely dormant judgments, i.e., still beliefs in disguise. Yet, if they are formally distinct, then intuitional content is not available to thought without some prior act of explication (or conceptualization?) that effects a categorial transfiguration in its very nature; and if such a transfiguration is necessary for it to figure in judgment, then in what sense can intuitional content still count as genuinely conceptual? It is unsurprising then that some have charged McDowell with accepting non-conceptualism now in all but name.Footnote 20

I believe this dilemma is avoidable on a hylomorphic reading of McDowell’s account, which marks the formal difference between experience and belief not in terms of their content but its mode of givenness. On this revised account, the adjectives ‘propositional’ and ‘intuitional’ do not signify a categorial difference in the nature of content but in our form of awareness of it. That is, although experience and judgment operate with essentially the same cognitive matter, viz. with formally identical conceptual contents, they differ in how they present these contents to mind. This preserves the internal relation between them in terms of sameness of cognitive matter and accounts for their essential difference in terms of how that matter is structured and given. Specifically, experience and judgment are distinct forms of awareness of conceptual content because they formally differ in the carrier of Fregean sense.

Thus, in judgment, the carrier of Fregean sense is the linguistic signs that stand in for our concepts, and since the active synthesis of concepts into propositions is undertaken by stringing together such signs in appropriate ways, judgment presents conceptual content to mind in the form of linguistically expressed propositions. In experience, however, the carrier of Fregean sense is the sensible manifold structured and unified into empirical objects by the passive actualization of our concepts, and therefore experience presents conceptual content to mind in the form of perceptible substances, properties, relations, and acts. Consequently, a difference in the carrier of Fregean sense marks a categorial difference in the form of awareness because it accounts for a difference in how formally identical conceptual contents are given to mind: either linguistically or sensibly.

This difference in the carrier of Fregean sense makes good on McDowell’s claim that the same ‘unifying functions’ are operative in both experience and judgment but actualize themselves differently in each case. For the operation of these functions and thereby also its results are bound by the nature of the medium in which Fregean sense is expressed. Thus, although the same forms of conceptual unity are operative in both, the synthetic unity of conceptual contents in experience will exhibit a significantly greater complexity and density than is achievable by stringing together linguistic signs in judgment. Given the multi-dimensional whirl of the sensible manifold being synthesized, the eidetic structure of experience manifests a multitude of parallel syntheses that crisscross and interlock, and therefore its conceptual content does not confront us in the tidy shape of discrete and clearly demarcated propositions but clumps together in a highly intricate network, which is so polymorphous that no proposition could ever hope to capture it. For our power of judgment, when it actively synthesizes concepts into propositions by means of the same forms of conceptual unity, remains bound by the comparatively one-dimensional medium of linguistic signs and therefore can isolate and carve out only particular strands of that network. In doing so, we selectively and discursively reconstitute some of the combinations of concepts passively given in experience, thereby asserting truthfully that they obtain as fact, without ever being able to map the actual structure all these interlocking combinations exhibit in experience.

Although the conceptual content of experience is not essentially propositional for this reason, it is still accidentally so, in that it is articulable in propositional form. Since this attributes propositionally structured content only to judgment, it entails that experiences do not represent facts. Contrary to McDowell’s original view, undergoing an experience does not amount to perceiving that such and such is the case but merely to perceiving as something, in that some sensory experience is apprehended in terms of its Fregean sense, e.g., as a pink ice cube or magpie. Hence, as indicated above, experience consists in the sensible intuition of substances, their properties, relations, and acts, and it does not present any of them as an isolated, irreducibly particular this, which stands in relations of absolute difference to everything else, but always as a this-such, as an instance of something general, viz. a concept. Thus, empirical objects represent our concepts by exemplifying them, and experience therefore involves a sensible awareness of the actuality of our concepts.

This hylomorphic reinterpretation of his account allows McDowell to maintain his basic contention that empirical content is essentially conceptual and answers the objections to his original account. For the above distinction between propositional and intuitional forms of awareness clearly demarcates experiences from beliefs and renders plausible why we cannot give empirical content a full and exhaustive propositional articulation.

To further demonstrate the potency of this reinterpretation, I now will apply it toward resolving the objection from fineness of grain. Although related to the objection from richness, it differs in that it attacks McDowell’s view on a more fundamental level. For the objection from richness only questions the propositional structure of empirical content and thus allows McDowell an option for retreat. The objection from fineness of grain, however, also attacks his basic contention that empirical content is conceptual.Footnote 21 In what follows, I first set out the core objection and McDowell’s reply and give a survey of the additional objections raised against it (§3). Since I maintain that all these objections rest on a shared assumption that is untenable on my hylomorphic reading of McDowell’s account, I subsequently undertake a deeper analysis of the hylomorphic structure of experience in order to identify this assumption (§4) and to answer the objections based on it (§5).

3 Fineness of grain: Surveying the debate

The objection from fineness of grain rests on the observation that experience is markedly more fine-grained than our conceptual scheme since we are capable of sensibly discriminating intuitional contents, such as subtle shades of color, for which we have no concepts. Yet, lacking these concepts, we do not possess corresponding rational capacities either, which entails that such fine-grained contents cannot result from their passive actualization. If experience can give us contents even though we lack corresponding rational capacities, however, then these contents cannot be conceptual. Thus, the objection’s intent is to repudiate McDowell’s conceptualism in favor of a non-conceptualist account of empirical content.Footnote 22

In reply, McDowell has argued that our conceptual scheme is exactly as fine-grained as the contents of experience because it includes demonstrative concepts, which we form by combining some general concept(s) with a deictic term, as in ‘this cup of tea’, ‘this sound’, or in something’s being ‘colored thus’. Such concepts are object-dependent, not merely in that they refer to some perceived object but in that the object, via the deictic term, enters into their Fregean sense. The concept’s mode of presentation of its referent is then, at least partially, a sensory one. Hence, demonstrative concepts are available only when, and as long as, their objects are sensibly present to the subject. On McDowell’s view, there also are demonstrative concepts of phenomenal properties, which enable us to apprehend some sensory experience in terms of its Fregean sense, e.g., as ‘this shade of blue’ or ‘this color’. Given their deictic component, such expressions function as indexicals in that their sense and reference differ depending on context. This allows for a flexible situational extension of our conceptual scheme to include new concepts that are just as fine-grained as we need them. Thus, on McDowell’s account, our very power of discriminative perception involves a power of spontaneous conceptual differentiation, which guarantees that our concepts always are exactly as fine-grained as the intuitional contents we can discriminate.Footnote 23

McDowell’s reply, however, has met with considerable resistance. We can divide the objections roughly into three groups. First, some critics object that McDowell’s reply suffers from circularity since the object-dependence of demonstrative concepts entails that we cannot form such concepts unless their objects of reference are already given. For otherwise they could not enter into their sense. Hence, the fine-grained intuitional contents that furnish the objects of demonstrative reference cannot yet be conceptual themselves but must consist in non-conceptual contents, which we then conceptualize in demonstrative concept-formation.Footnote 24 This is an instance of a general problem with empirical concept-acquisition that the critics attribute to conceptualism. For our very ability to form general concepts of empirical objects, such as ‘ball’, ‘chair’, or ‘blue’, and to do so based on experience, also presupposes that experience already confronts us with such objects before we acquire their concepts. Yet, conceptualism assumes that experience only has content, and thus makes objects available, once we possess the requisite concepts. It then seems impossible to acquire concepts, at least basic ones, from experience at all. Unless one is prepared to endorse some form of concept nativism, as McDowell is not, the only way to render empirical concept-formation intelligible therefore is to admit that experience furnishes us with non-conceptual contents and that we then conceptualize these.Footnote 25

Second, some critics doubt that demonstrative concepts are genuine concepts since their object-dependence prevents them from fulfilling a necessary requirement for concept-possession. On this Re-Identification Requirement, subjects possess a concept only if they are capable of reliably re-identifying its instances across differing contexts. That is, they must be able to classify occurrences that differ synchronically or diachronically as falling under the same concept and hence able to recognize something identical in them. In diachronic respect, this seems impossible because demonstrative concepts are available only as long as their referents are present to the subject. As soon as the relevant sensory experience fades from view, the expression ‘colored thus’ loses its situation-specific sense. Yet, if we lose the very concept together with its original instance, we simply cannot recognize subsequent sensory experiences as presenting further instances of it. Since McDowell accepts the Re-Identification Requirement, he maintains that, e.g., demonstrative color concepts remain available briefly even after the original sample has faded because we can retain it in memory, at least for a short time. As his critics rightly point out, however, there is ample psychological evidence attesting to how limited our recollective powers actually are in this respect.Footnote 26

In synchronic respect, there are two problems with the extension of demonstrative concepts. Thus, the objection from graduality subjects McDowell’s account to a sorites paradox based on the observation that differences between color shades are often exceedingly gradual. Consider a spectrum of three color samples, a, b, and c. The samples a and b are phenomenally indistinguishable, and so are b and c, while a and c are clearly distinct. Since a and b are indistinguishable, a demonstrative color concept ‘this shade’ whose situation-specific sense is determined by a would also include b in its extension. Moreover, since b and c are indistinguishable, the same should hold for c. That is paradoxical, however, since a and c, ex hypothesi, are clearly distinct shades. McDowell maintains that we can avoid this paradox by abandoning the underlying principle of transitivity, i.e., by restricting the concept’s extension only to samples that are immediately indistinguishable from the original sample a. This is insufficient, however, since b, in virtue of being indistinguishable from both a and c, still would fall under both color concepts ‘a’ and ‘c’ simultaneously, while our color concept ‘b’ still would include two distinct shades, a and c, in its extension. Hence, McDowell’s account classifies shades that do not belong together as instances of the same concept and identifies the very same sensory experience as two distinct shades. Arguably, this spoils his reply to the fineness of grain objection, which depends on an exact fit between our concepts and the intuitional contents we can discriminate.Footnote 27

The objection from contextuality, on the other hand, rests on the observation that the very same shade of color can differ phenomenally depending on perceptual context. Thus, various parts of a uniformly white wall may look quite different depending on how the light falls on them, and so may a particular shade of blue when instantiated in a dusty woolen carpet rather than a shiny steel ball. Such context-dependence confronts McDowell’s account with a dilemma: If we subsume what in fact differs phenomenally under the same demonstrative concept and refer to it unitarily as ‘this white’ or ‘this blue’, then such concepts cannot match the fine-grained character of intuitional content after all. Yet, if we form separate demonstrative concepts to capture such differing looks, we end up with a surplus of conceptual differentiation that prevents us from recognizing that they instantiate the very same color.Footnote 28 Hence, both of these objections identify, in synchronic respect, significant problems with determining the proper extension of demonstrative concepts. Consequently, demonstrative concepts cannot fulfill the Re-Identification Requirement for concept-possession and therefore cannot count as genuine concepts.

A third group centers on what I call the objection from phenomenal difference, which rests on the observation that we can describe the very same experience with differing demonstrative concepts, such as ‘this olive’, ‘this shade of green’, or ‘this color’. Yet, given McDowell’s account, passively actualizing these concepts in experience yields altogether different intuitional contents, and hence two subjects differing in the concepts they possess also will differ in what experience presents to them. In other words, they will not see the same shade of color. Thus, some critics claim that, on McDowell’s account, differences in conceptual repertoire entail differences in phenomenal appearance, which is highly implausible and undermines the empirical world’s role as an intersubjectively shared frame of reference for mediating conceptual differences.Footnote 29 Likewise, some argue that, on McDowell’s account, the empirical world will differ phenomenally for rational and non-rational animals since the latter lack concepts and therefore cannot experience the same contents as the former. Yet, since many non-rational animals have perceptual systems structurally similar to our own, what they perceive should be relevantly similar too. Hence, if we want to guarantee that all subjects, rational and non-rational, perceive essentially the same world, we must posit the same kind of content for them, i.e., non-conceptual content.Footnote 30

4 Identifying the underlying assumption

Having set out the objections, I now will proceed to diagnose their underlying assumption. My central claim is that this assumption conflates the form of experience with its content and therefore is untenable on a hylomorphic reading of McDowell’s account. Hence, to identify it, I need to analyze the hylomorphic structure of experience more deeply by investigating how precisely the matter/form-distinction applies to it. On my reinterpretation of McDowell’s account, experience consists in the sensible intuition of particular objects, or more precisely in a sensible awareness of the actuality of our concepts insofar as these objects instantiate them. This shift in emphasis indicates that, in applying the matter/form-distinction, we should distinguish between (a) objects of experience and (b) our experience, i.e., consciousness, of such objects. For the results will differ depending on whether we analyze the object or the act by which we grasp it. A difficulty in marking this distinction clearly is that the object of experience and our consciousness of it are easily conflated, especially on idealist accounts that picture the object as somehow constituted by our consciousness of it. Therefore, differentiating the hylomorphic structure of the act from that of its object requires some care.

I begin with empirical objects, such as magpies, cups, or trees. As stated in §2, the subject apprehends such objects through a passive synthesis, in which it brings the elements of a sensible manifold to the unity of a concept by involuntarily interrelating these elements according to the rule of synthesis corresponding to the concept. Thus, the rule that corresponds to the concept ‘magpie’ traces out a certain structural arrangement that the subject is able to recognize in a manifold of sense impressions; a certain Gestalt pattern, if you will. When it involuntarily actualizes this ability in an appropriate sensory context, the subject grasps that these elements belong together in the determinate formal structure characteristic of magpies and then experiences them as a unified empirical object of that very form. Thus, the sensory character of this empirical object is its matter, while the arrangement, structure, or Gestalt pattern instantiated by its sensory character is the object’s form.

Moreover, in §1 the Fregean sense of a concept was identified with the rule of synthesis belonging to this concept, and thus with the eidos or ideal structure traced out by that rule. This entails that the Fregean sense of a concept passively actualized in experience is identical with the formal structure of the empirical object it refers to. The structural arrangement that I designate in abstracto with the concept ‘magpie’ is the same I recognize in concreto when confronted with a particular magpie in experience. Hence, the form of an empirical object simply is the concept of that object; the eidos or ideal structure that individuates it as an object of a particular kind and thereby constitutes its nature.

Crucially, this distinction between an empirical object’s form and matter does not reintroduce the dualism of conceptual scheme and empirical ‘content’ so adamantly opposed by McDowell. This dualism pictures our concepts as externally related to the sensible manifold, as essentially independent of and applied to it from the outside. In a well-known characterization of our conceptually shaped receptivity, McDowell rejects this, writing that “[t]he relevant conceptual capacities are drawn on in receptivity. […] It is not that they are exercised on an extra-conceptual given.”Footnote 31 Thus, in contrast to this dualism, McDowell does not picture the sensible manifold apprehended, i.e., structured and unified, by a concept as externally related to it. For him, an empirical object’s matter and form are not distinct realities but two aspects of the very same reality, of the same unified object, and therefore related internally. They are separable merely in thought but not in actuality, i.e., as given in experience. That is, we never experience a pure and unstructured sensible manifold. Rather, as mature human beings, we experience it always as already structured eidetically in determinate ways and apprehend it as an ordered array of distinct empirical objects.

On McDowell’s view, the subject receives these objects by an act of receptivity that is mediated by its concepts. That is, experience, as the reality of the receptive act, discloses objects to us always under a conceptual mode of presentation, which forms its content. Put differently, the content of experience consists in the Fregean senses imparted to the sensible manifold by the passive actualization of our concepts, and by extension in the formal or eidetic structure of the objects we thereby receive. The fact that the content of the receptive act is numerically identical to the formal structure of the object secures their internal connection, i.e., that the act actually grasps its object. Hence, it is in virtue of its content that experience refers to objects at all.Footnote 32 Yet, this content itself is given to us in a specific form, viz. not discursively or propositionally but intuitionally—by means of the sensory character of experience that acts as the carrier of Fregean sense. This makes it empirical content and differentiates it from other forms in which content can be present to mind.

Experience, then, is a species of intentional object-consciousness whose distinctive form is the sensory mode in which its content is given. That is, when I experience a particular magpie, the content of this experience is the ideal structure designated by the concept ‘magpie’, and this content is present to me because this structure is realized in a manifold of sense impressions. These impressions, which constitute the sensory character of the object, thereby act as the carrier of this conceptual structure; they constitute the sensory form in which I become conscious of its actuality. Hence, the content or matter of this consciousness—what I cognize in an act of receptivity—consists in the concepts present to me in the sensory guise of the objects I experience. Empirical content is therefore “thinkable content”,Footnote 33 i.e., the Fregean sense of thoughts that we can explicate in empirical judgment and thereby transition from intuitional into propositional form.

In this context, McDowell’s famous metaphor of the unboundedness of the conceptual ultimately implies that the sensory character of experience as such contributes nothing to its content but, qua carrier of Fregean sense, merely is the form in which we apprehend its actuality. In abstraction from our self-conscious act of grasping it as something, we could not even be aware of the sensible manifold as such because, as the empty form of experience, it would be nothing to us. Hence, we can sum up the nature of the hylomorphic unity formed by spontaneity and receptivity in experience as follows: While spontaneity provides us with concepts, i.e., the matter or possible content of both experience and thought, receptivity’s proper contribution consists in actualizing such concepts in sensible form, which discloses their objective reality in the guise of particular empirical objects and thereby provides an empirical grounding for thought.Footnote 34

This application of the matter/form-distinction now enables us to articulate the empiricist assumption that underlies the objections to McDowell’s reply: All of them implicitly equate the sensory character of experience with its content.Footnote 35 Consequently, they maintain that intuitional content is exactly as fine-grained as what we experience sensorily and therefore cannot be constituted even by demonstrative concepts. Thus, the circularity objection claims that demonstrative concepts are derivative of intuitional contents because the sensory experiences such concepts refer to enter into their sense. The objections from contextuality and graduality implicitly take sensory differences to entail differences in content and therefore state a lack of fit between our concepts and the intuitional contents they supposedly constitute. Finally, the objection from phenomenal difference presupposes that two subjects experience the same intuitional content when confronted with the same sensory experience, which is why such content cannot depend on both coincidentally possessing the same concepts. Yet, as I will argue, the conceptual structure of experience, its intuitional content, differs from and need not be as fine-grained as the sensible matter it unifies. If this diagnosis is correct, the objections should lose their force once we abandon their underlying assumption and distinguish more clearly between the form and content of experience. To show this, I now will apply the account outlined here toward answering them.

5 Answering the objections

Generally, that empirical content is constituted by concepts, which always have a certain generality, is rendered unproblematic once we recognize that the fine-grained phenomenology of experience is mostly a function of its sensory aspect, i.e., its sensible form, and therefore operates, as it were, on a different level than its content. Given §4, the sensible form of experience is more fine-grained than its conceptual content because the relation between concepts and the sensible manifold they unify is a one-to-many relationship.

Thus, the sensible manifold given in an experience only carries the content ‘magpie’ because we perceive its elements in the unity of the respective concept or eidos, i.e., in a Gestalt pattern that constitutes a unitary experience of a fluctuating sensible manifold. As the magpie I perceive moves around, my experience of it involves an intricate flow of simultaneously changing color and sound impressions. Yet, the formal structure ‘magpie’ that unifies these impressions, i.e., the pattern or Gestalt I recognize in and track through this flow, remains constant through all these changes and allows me to experience them as belonging to a particular and determinate empirical object. Put differently, in apprehending some sensory experience as something, we institute a conceptual identity among its sensible diversity and are self-conscious of both these aspects simultaneously.

That is, although we are conscious of a sensible manifold as such only because we grasp it in the unifying structure of some pattern, this identity of grasp does not cancel out the diversity being grasped. Nor does a concept, as an abstract Gestalt pattern, fully determine the phenomenal character of the empirical object instantiating it, since the same pattern can be exemplified in many ways. We can think of an empirically realized concept, eidos, or Gestalt as an abstract frame filled in by a sensible diversity not fully determined by the frame itself; a frame that acts as our mode of representing this diversity as a unitary object of reference. In experience, such concept-frames can be nested, e.g., in that the wider frame ‘magpie’ contains narrower frames such as ‘wing’, ‘beak’ or ‘hopping’, each designating a distinctive Gestalt pattern. Yet, we can never narrow our concepts down enough to actually arrive at a one-to-one relationship with elements of the sensible manifold. Even when we perceive ‘simple’ qualities such as ‘red’, what counts as red can be phenomenally quite diverse, even within a single experience. In this, a concept functions like a mathematical variable that can take on a variety of particular values and merely determines or constrains the range of possible values. That is what its generality vis-à-vis its sensible instances consists in. Consequently, since any sensible manifold stands to the concept unifying it as a plurality, the sensible form of experience always is more fine-grained than its conceptual content.

This entails that sensory differences do not as such constitute differences in conceptual content and therefore do not automatically make a difference to truth-conditions. For it is possible to apprehend the phenomenally diverse elements of a sensible manifold either as identical under the same concept or as differing in conceptual content in some respect. Whether a sensible diversity is apprehended conceptually in a unitary or a differentiated manner ultimately is a function of the rules of synthesis that habitually impart a determinate eidetic structure to the sensible manifold. Put differently, the sensible manifold itself, considered apart from the cognitive and practical needs that inform our practices of perception, does not strictly prescribe how to synthesize and structure it conceptually, even though it represents an enabling condition for our pattern-apprehension. Its fine-grained sensory character ultimately constitutes only a potential and a limit for apprehending the manifold in a conceptually differentiated manner. We can actualize this potential according to our cognitive and practical needs by spontaneously generating sufficiently fine-grained concepts, but nothing in this potential itself requires us to actualize it fully.

Thus, according to popular myth, the Inuit command an unusually differentiated vocabulary for snow, which corresponds to rational capacities for visually recognizing subtly different kinds of snow that simply look the same to other people. Yet, this does not entail that what the Inuit see differs in sensory terms from what other people see but merely that they have learnt to see it differently. They have developed some particularly fine-grained concepts that allow them to differentiate and synthesize the sensible manifold in a more complex manner than is possible to subjects lacking these concepts. Hence, while their snow-experience does not differ in its sensible form from that of other people, it differs in having a comparatively richer and more differentiated conceptual structure or content. Although admittedly a myth in case of the Inuit, this illustrates a general principle. For wine connoisseurs, musicologists, geologists, and economists arguably also experience wines, musical performances, rock formations, or economic data in a considerably more structured, complex, and differentiated manner than laypersons because such experts have acquired the conceptual apparatus necessary for such advanced powers of pattern-perception.Footnote 36

This principle now allows us to concede, in response to the objection from phenomenal difference, that two subjects differing in color vocabulary indeed also will differ in the content of their color experience. Yet, this difference in content does not entail that their experiences must differ in sensible form. Just as some words are homonymous, so identical sensory experiences may differ in Fregean sense; and just as different words can be synonymous, so differing sensory experiences still might have the same general content. Consequently, two subjects can apprehend the same color experience differently, despite its sameness in sensory character, and then experience it respectively as ‘this shade of green’ or ‘this olive’, i.e., under different descriptions or modes of presentation. Another illustration of this principle is the famous ‘duck-rabbit’; a figure we can experience either in the Gestalt of a duck or a rabbit, i.e., with differing conceptual content, even though its sensory character is exactly the same in both cases. Although two subjects indeed may experience the world differently then, intersubjective understanding remains possible since each subject can acquire, in principle, the other’s concepts and thereby enter into their way of seeing the world. They can undergo a fusion of horizons that enables them to share essentially the same point of view.Footnote 37

The same principle also illuminates the essential difference between the perceptual powers of rational and non-rational animals. Both kinds of animal, a sufficient structural similarity in their perceptual systems granted, differ mostly in the content of their experience but not in how the world appears to them in purely sensory terms. Rational animals can apprehend some sensory experience as a cup or as a magpie, i.e., as a determinate empirical object that is distinct from them and thus a possible object of thought towards which one could assume a variety of attitudes. On McDowell’s view, non-rational animals are incapable of such a distanced and flexible relationship to their surroundings. Rather, they react instinctually to their sensory experiences by apprehending them in light of their needs as dangers, opportunities, or obstacles. Consequently, rational and non-rational animals possess categorially distinct powers of perception that, while not essentially of the same kind, still represent two species of the same genus. Their forms of receptivity differ specifically in that they structure the received sensible manifold according to different principles. Rational animals possess a conceptually shaped receptivity and therefore the power to structure the manifold spontaneously by generating a conceptual scheme. In contrast, the receptivity of non-rational animals is structured fundamentally by fixed biological imperatives, wherefore it is not a rational power of cognition. For unlike rational animals, mere animals cannot perceive their environment differently based on reasons since they do not have a subjectivity capable of thinkable contents, i.e., of a self-conscious world-view one can flexibly adjust to the world because it is rationally responsive to how things stand within it, in a way that transcends mere biological need. On McDowell’s view, human animals begin exhibiting such a distinctively rational form of receptivity once they learn language.Footnote 38

A hylomorphic approach also helps to resolve the two objections pertaining to the synchronic extension of demonstrative concepts. In both cases, demonstrative concepts did not fit sensory differences exactly, and from this McDowell’s critics inferred a mismatch between such concepts and intuitional contents. Yet, this mismatch no longer follows if sensory differences merely characterize the sensible form of experience, in which conceptual content is given. As stated, we can apprehend phenomenally differing elements of a sensible manifold as identical under the same concept and thereby refrain from fully actualizing its potential for conceptually differentiated apprehension. This explains why empirical objects, although fully unified, need not be phenomenally uniform, as when we apprehend the phenomenally diverse parts of a wall as identically white. Likewise, one and the same concept can be instantiated in phenomenally diverse ways because divergent sensory experiences can carry the same general conceptual content and thereby represent phenomenally differing objects of the same kind, as when the same shade of blue differs in appearance depending on context. While the form of experience is more differentiated in these cases than its content, this content nevertheless fits the concepts that constitute it exactly. Hence, there is no mismatch here between concepts and intuitional contents. This answers the objection form contextuality.Footnote 39

Similar considerations apply to the objection from graduality. Hence, there is no such mismatch either when we apprehend a gradually differing color spectrum a, b, and c as a unified object that instantiates the demonstrative color concept ‘b’. Since a and b as well as b and c are phenomenally indistinguishable, the concept ‘b’ is exemplified as a continuous intentional object b with vague boundaries at its margins a and c. Such vagueness precludes neither demonstrative reference to the object b nor an exact fit between our concept ‘b’ and the intuitional content it constitutes. Likewise, there is no mismatch between intuitional contents and demonstrative concepts merely because the same color sample b is subsumable under both color concepts ‘a’ and ‘c’. For one can invest the same sensible matter with a variety of possible contents by subsuming it under different concepts, as the ‘duck-rabbit’ figure illustrates. This holds of two subjects differing in conceptual repertoire, as discussed above, but also of the very same subject, since it can pursue different cognitive ends whose fulfillment may require different ways of structuring the sensible manifold, i.e., the actualization of different concepts in the same sensible form depending on context. Consequently, there is no lack of fit between demonstrative concepts and their perceptible instances, and therefore no problem with the synchronic version of the Re-Identification Requirement.Footnote 40

To answer the objections from circularity and concept acquisition, we need to complement McDowell’s reinterpreted account of experience with an account of empirical concept-formation. Demonstrative concepts provide a useful case study due to their special object-dependence. For McDowell’s critics, this dependence entails that demonstrative concepts are derivative of their referents because they enter the concept’s sense via the deictic term, and therefore we can form such concepts only if the intuitional contents that enable demonstrative reference are already given. Consequently, forming a demonstrative concept must involve conceptualizing some prior non-conceptual content, such as a particular shade of blue. As before, however, the matter/form-distinction allows us to avoid this conclusion because what actually enters into the sense of ‘this shade of blue’ via the deictic term is not a content but something sensory belonging to the form of experience. It thereby is invested with a Fregean sense, i.e., a conceptual content, and apprehended as this shade of blue, which constitutes it as a determinate object of both experience and reference in the first place.

Crucially, this act of object-constitution is inseparable from the corresponding act of concept-formation, which consists in combing the deictic term with the relevant general concept(s). For this combination effects a conceptual differentiation within the general concept, which corresponds to a parallel differentiation of the empirical content constituted by that concept. Thus, where a subject previously apprehended phenomenally diverse elements of a sensible manifold identically as blue, application of the deictic term now allows it to differentiate some of the elements subsumed under the general concept ‘blue’ from the others and to grasp them separately as this blue, i.e., to synthesize them into a new and distinct object of experience. Hence, our power of spontaneous conceptual differentiation grounds our very power of discriminative object-perception, which therefore is a rational power. In exercising it, the subject actively modifies the eidetic structure of experience by spontaneously generating new ways of differentiating and synthesizing, and therefore of understanding, the sensible manifold. Since they effect this modification of experience, newly formed demonstrative concepts are from the outset immediately actualized in sensible form. Consequently, we do not form such concepts by abstracting prior non-conceptual contents from experience. Rather, their formation is itself a content-constitutive act of the subject by which it comes to perceive the elements of a sensible manifold in new Gestalt patterns. On the account of empirical concept-formation outlined here as a complement to McDowell’s conceptualism, concept-formation essentially is content-formation.

This places the ultimate source of empirical content, qua thinkable content, in the subject’s spontaneity but does not entail that the spontaneous generation of new content occurs in a void, i.e., fully independent of and unconstrained by experience. For spontaneity and receptivity remain in hylomorphic interaction during concept-formation, in that the spontaneous generation of new rules of synthesis depends on a continuous interplay between the extant eidetic structure of experience and its sensible potential for further structuring. As stated, this potential resides in the fine-grained sensory character of experience and is actualized in particular acts of spontaneous conceptual differentiation and synthesis, which the subject then can habituate into stable rational capacities. We can characterize this empirically mediated process of concept-formation, since it depends on an ongoing hermeneutic circle between the extant conceptual content and the sensible form of experience, as a hermeneutics of sensibility by which the subject continuously develops, refines, and corrects its conceptual scheme in a self-conscious effort to understand the input of its senses. Sensibility thus remains the indispensable medium in which we unfold our conceptual scheme, understood as the totality of the rules of synthesis we have developed. Concept-formation, as the spontaneous generation of an eidos or Sinn, is essentially dependent on receptivity and impossible without it, which underwrites McDowell’s contention that there is no content for thought at all unless spontaneous capacities are operative in receptivity.

As indicated, the object-dependence of demonstrative concepts consists in the fact that we necessarily actualize them in experience, as particular objects of reference, whenever we possess them. Although we may possess and exercise other concepts even when they are not actual in experience, this does not apply to demonstrative concepts because their sense partly consists in the sensory experience they refer to, and this intimate connection with the sensory both guarantees their necessary actuality and restricts their availability to the duration of that experience. Consequently, such concepts do flout the diachronic version of the Re-Identification Requirement and thus apparently confirm the related objection that they are not genuine concepts because we cannot reliably recognize their instances across diachronically disparate contexts. Due to its inherent context-dependence, we are unable to habituate the act of synthesis that constitutes the object of demonstrative reference in the present sensory context and therefore unable to repeat this act and recognize it as the same in future sensory contexts. Thus, the object-dependence of such concepts entails that no rule of synthesis corresponds to them because a rule is something general that transcends any one context of application. It enables us to recognize differing objects as instances of the same concept, i.e., as exemplifying the very same rule, because we remain conscious of its numerical identity when actualizing it in different contexts. Without a context-transcending rule of synthesis, it seems there is no concept, and therefore demonstrative ‘concepts’ do not qualify.

This objection, however, rests on equating concepts with the rational capacities we acquire by habituating them. Hence, it trades on an ambiguity in the notion of a capacity, which means either an ability in the sense of a mere potentiality, i.e., that it is not entirely impossible for us to perform acts of a particular kind successfully (mere ability); or an ability in the sense of a first actuality, i.e., a stable disposition to repeatedly perform such acts with reliable success (full-fledged capacity). In the first sense, I do not have the capacity to fly by flapping my arms but generally the capacity to dunk basketballs, even if I fail invariably. In the second sense, I only have the capacity of dunking basketballs once I have mastered it well enough to hit the basket with reliable success. Similarly, we can differentiate between the mere ability to use a concept and the full-fledged capacity to do so, which results from mastering its use through practice and habituation.

While equating both is an easy step to take, we can understand the possibility of such habituation only if we distinguish them. Admittedly, an act of synthesis performed for the very first time cannot yet follow a rule since rules are something general. They only exist once we are able to repeat the very same act in different contexts, thereby transcending the particular context of the initial act. Yet, this first-time act still must be capable of serving as a precedent that guides us in repeating it. Otherwise, we could not recognize that any further act is qualitatively the same as the first, i.e., a repetition, and thus an instance of something general. In order to serve as precedent, the first-time act already must exhibit a rule-like structure, for it is by repeating acts of this structure that we acquire, through memory, a habit of rule-following when performing such acts, i.e., a rational capacity.

This rule-like structure consists in how the elements unified in the act, irrespective of whether they are its diachronic temporal phases or a synchronically given sensible manifold, are arranged in a determinate order, pattern, or Gestalt in performing it. Since such a structure is open, through memory, to self-conscious repetition in further acts, we can take the very first act exhibiting it as a rule-forming precedent. This enables us to characterize this first-time act of synthesis as one in which a new concept is formed. Arguably, this characterization also holds when its self-conscious repetition fails or is impossible, as with demonstrative concepts, due to the act’s special context-dependence. For even in such cases, the act still must exhibit a determinate structure individuating it as this particular act. Consequently, we are in possession of a concept, however short-lived, even in acts of synthesis not open to self-conscious repetition. Since a concept’s momentary formation and exercise therefore do not presuppose its diachronic availability, we can reject the diachronic version of the Re-Identification Requirement and affirm the genuineness of demonstrative concepts.Footnote 41

Moreover, while available, such concepts perform exactly the same functions in our thinking as any other concepts. Thus, we can employ demonstrative concepts in truth-apt judgments, such as ‘This garish shade is a terrible bedroom color’ or ‘Curtains which are colored thus have a calming effect’, and in related arguments and inferences.Footnote 42 Although such thoughts are thinkable only while the relevant concept is available and its object thus perceptually present, nothing in principle prevents us from re-forming this very concept over and over again in suitable perceptual circumstances and thus from grasping the very same thought anew. Since we cannot retain or habituate such concepts, we admittedly may not recognize them immediately as the very same concepts and thoughts we previously had. Nevertheless, we can recognize their identity inferentially, through an extra step, as when looking at a color sample board we remember that the shade we referred to as ‘this blue’ yesterday was the first one on the right in the second row.Footnote 43 Consequently, although demonstrative concepts differ in that we cannot habituate them into stable rational capacities, they still are genuine concepts, and this final objection too is therefore unsuccessful.

6 Conclusion

In summary, a hylomorphic reinterpretation of McDowell’s conceptualism can resist the objection from fineness of grain. Its distinction between the sensible form and the conceptual matter or content of intuition guarantees an exact fit between concepts and intuitional contents, i.e., their numerical identity, and enables us to differentiate such contents, where necessary, up to the very limit of phenomenal discriminability. In the remainder of this paper, I will try to indicate why conceptualism so understood is dialectically superior to non-conceptualism and why we should classify it as a special, objective form of idealism.

As interpreted in §4, non-conceptualism identifies the sensory character of experience with its content and therefore posits categorially distinct kinds of content for experience and thought. I characterized this as an unacknowledged empiricist presupposition, which underlies non-conceptualism, and now will try to substantiate this claim. In its classic form, empiricism is a species of indirect realism since it maintains that subject and object are numerically distinct and externally related. There is an ontological gap between them across which they interact causally, in that the object’s impact on the subject’s senses produces sensory states in the subject that supposedly are isomorphic to the object, representing it within subjectivity. Thus, although experience relates externally to the object because it consists in sensory states internal to the subject, it still has representational content and can act as an epistemic intermediary between them. For experience so understood furnishes the subject with evidence regarding its efficient cause, i.e., its object. Once suitably conceptualized, such evidence can justify judgments about the object by entering into inferences about it as a premise; it thereby acts as a foundation for belief. Empiricism thus grounds empirical content in the supposed fact that the subject’s sensory states are, as Gareth Evans puts it, informational: In being isomorphic to their cause, they make available their own etiology.Footnote 44

Yet, since empiricism places experience in an external relation to its object and cause, it cannot contain, qua effect, any meaningful awareness of its own etiology. Even if our sensory states were isomorphic to their cause, we could not know that since we have access to nothing but them and therefore cannot compare cause and effect. Consequently, experience so understood cannot serve as evidence and does not in fact represent its object and cause. Instead, it confronts us as a purely sensory state devoid of any representational content; a subjective state that bears a causal but not an intentional relation to external objects since intentionality is a relation internal to subjectivity. The often-implicit assumption that experience so understood nevertheless bears representational content, can act as evidence, or simply presents us with objects for reference ‘directly’, i.e., absent any mode of presentation, forms the core of the ‘Myth of the Given’. An underlying picture of this kind thus likely explains why non-conceptualists implicitly identify the sensory character of experience with its content, why conceptualists doubt that we can conceptualize such ‘contents’, and why the attempt draws McDowell’s charge of falling into the Myth.

A hylomorphic reinterpretation of McDowell’s conceptualism is dialectically superior because it succeeds at avoiding the ontological gap between subject and object. It conceives of experience as a kind of intentional object-consciousness that relates us, under a conceptual mode of presentation and thus in virtue of its conceptual content, immediately to worldly objects and thereby renders them accessible within subjectivity itself. At the same time, it therefore has a content that can justify empirical judgements about these objects because it is available to thought. For the subject relates primarily to its own concepts or eide in experience but comes to recognize their subject-independent actuality in the sensible form of the object, and this supplies its thinking with external constraint.

This entails that empirical content has its ultimate source in the subject’s interpretive activity, and the same applies to the determinate structure of the world that experience discloses. Put differently, since our form of receptivity relates us to empirical objects only under a conceptual mode of presentation, such objects ultimately are themselves products of interpretation. On the present account, McDowell’s conceptualism therefore comes out as a form of idealism. Of course, McDowell himself has spent much effort in Mind and World on rebutting an anticipated charge of idealism.Footnote 45 I want to submit, however, that his arguments there are directed only against subjective idealism, i.e., a view that renders the world a product of the unconstrained and hence arbitrary acts of some particular subject and thereby abolishes the distinction between true and false interpretation. This still leaves open the possibility that his conceptualism might instead be a form of objective idealism.

And McDowell indeed attributes objectivity to our concepts or eide. Importantly, for McDowell, ‘objectivity’ does not equal ‘absolute subject-independence’, as this would imply a Platonic conceptual realism rejected by him under the label of ‘rampant platonism’. Rather, our concepts are objective because, generally, they are rules that constitute the social practices and language-games we are initiated into in upbringing. While practices depend ontologically on the subjects engaging in them, their rules precede any particular subject and confront it as a social and historical matter of fact. As such, these rules are not at our free disposal; we cannot change them arbitrarily or at will. Consequently, our practices and concepts enjoy a qualified subject-independence that is central to McDowell’s ‘naturalized platonism’ and provides substantial constraints on truthful interpretation.

On this view, interpretation can be true or objective because it is subject to constraints that are material, in that concept-formation depends on the sensible medium interpreted; pragmatic, in that this interpretive activity serves various ends in our lives and is subject to their realization conditions; and intersubjective, in that these ends and our historically accumulated knowledge of how to achieve them are enshrined in the social practices that normatively govern our interpretive endeavors and embed us in a shared horizon of understanding. When interpretation is in fact true, the eidetic structure of our subjectivity simply is the formal structure of the world as it objectively is. Experience then constitutively contains and relates us to objects that are objectively real, in a way fully compatible with the lessons of McDowell’s disjunctivism. This objective idealist reading of his conceptualism thus considers the world not an alien realm brutely external to subjectivity but understands it as our intersubjectively shared life-world, as we collectively make sense of it in our practices. It is the world of the manifest image, which McDowell’s project aims to vindicate. I have argued here that experiential hylomorphism is crucial to the prospects of this project.Footnote 46