What is the myth of the given?

Abstract

The idea of ‘the given’ and its alleged problematic status as most famously articulated by Sellars (1956, 1981) continues to be at the center of heated controversies about foundationalism in epistemology, about ‘conceptualism’ and nonconceptual content in the philosophy of perception, and about the nature of the experiential given in phenomenology and in the cognitive sciences. I argue that the question of just what the myth of the given is supposed to be in the first place is more complex than has typically been supposed in these debates, and that clarification of this prior question has surprising consequences. Foundationalism was only one of Sellars’s targets, and this not only in the familiar sense that the more fundamental issues at stake concern the very ‘objective purport’ or intentionality of our empirical thinking in general. When pushed further still, Sellars’s critique in fact hinged on his diagnoses of implicit framework-relative or ‘categorial’ metaphysical presuppositions he exposes in givenist views. Furthermore, the key to his critique accordingly turns out to rest on implicit assumptions concerning the in principle revisability or replaceability of any such presuppositions, whether ‘innate’ or acquired, and including Sellars’s own. Another key result is that widespread assumptions that Sellars’s famous critique is simply inapplicable or irrelevant to either ‘thin’ nonconceptualist views of the given (such as C. I. Lewis’s), since they are ‘non-epistemic’; or alternatively, irrelevant to ‘thick’ conceptualist and phenomenological analyses (since they, too, reject ‘sense-data’ or the ‘bare given’)–both turn out to be mistaken.

“It will be noted that the account I am giving of physical objects as individuated volumes of color stuff is essentially what I [call] the child’s proto-theory of the objects of visual perception. This proto-theory is part and parcel of what I have called the Manifest Image of Man in the World. That this essay moves largely within the categories of the Manifest Image must be borne in mind throughout what follows. It must also be borne in mind, however, that it also moves within the framework of a theory of categories which denies the authoritative status of the categories of the Manifest Image, i.e., it works within the framework of a theory of categories which rejects the Myth of the Given.”

Wilfrid Sellars, ‘The Lever of Archimedes’ (1981b, FMPP I §75, note 6).

As a phrase, and as a set of interconnected arguments, the idea of the “myth of the given” originates in Wilfrid Sellars’s “general critique of the entire framework of givenness” in his wide-ranging 1956 article, “Empiricism and the Philosophy of Mind” (EPM I §1).¹ Sellars’s critique of philosophical conceptions of the given as a “myth” that conceals a variety of philosophical confusions² was then subsequently defended in highly influential books by Rorty (1979), Brandom (1994), and McDowell (1996). The idea of the given and its alleged problematic status continues to be at the center of heated controversies about foundationalism in epistemology, about “conceptualism” and nonconceptual content in the philosophy of perception, and about the nature of the experiential given in phenomenology and in the cognitive sciences.³ The question of just what the myth of the given is supposed to be in the first place, however, is more complex and multi-layered than has often been supposed in these disputes, and clarification of this prior question turns out to have surprising consequences. My contention will be that the core of Sellars’s original conception of the myth has been obscured in a way that has made it seem to be simply inapplicable and so irrelevant to certain widespread views to which, I argue, it was in fact intended to apply and represents a significant challenge.

In section §I, I lay out some familiar Sellarsian dilemmas that raise questions about how our sense experiences provide warrant for our empirical beliefs about the world. I then suggest that Sellars’s critique has subsequently been assumed to be inapplicable to widespread views of two opposed kinds. It is assumed or objected that either (1) the Sellarsian critique targets an irrelevantly “thin” nonconceptual form of the given (e.g., ‘sense data’) that is simply irrelevant to more phenomenologically adequate or conceptually “thick” conceptions of the experiential given; or conversely, it is assumed (2) that Sellars’s myth had an irrelevantly “thick,” over-intellectualized target concerning conditions on inferential reason-giving justification that makes the myth inapplicable to sufficiently “thin” or non-epistemic versions of the experiential given.

In section §2, I then examine a subtle example of the second sort of assumption in the form of a view among recent interpreters of C. I. Lewis (1929) that Sellars’s myth is not applicable to Lewis’s resolutely non-epistemic version of the sensuous given. In response I argue that a crucial but less well-known line of thought in Sellars’s work concerning the myth of the categorial given⁴ (which I will argue traces back to EPM) was both intended to and does apply to Lewis’s nonconceptual given, and that it is also potentially applicable on similar grounds to any non-epistemic conception of the given depending on wider questions pertaining to categorial revisability that I seek to clarify in some detail. In section §3, I then argue that the myth of the categorial given was on similar grounds also held by Sellars to be applicable to a wide variety of conceptually or phenomenologically “thick” positions that have been assumed to be shielded by the second sort of irrelevancy. Together these diagnoses both explain and provide crucial support for Sellars’s late-career remark that the categorial given is “perhaps the most basic form” of the myth (1981b, FMPP I §4), and thereby also demonstrate the applicability of his critique to positions widely assumed to be immune to it.

1 Sellarsian dilemmas: how do sense experiences justify empirical beliefs?

Both in recent debates and in Sellars’s EPM in 1956, the primary focus of the critique of the given as a myth has appropriately enough been on the alleged insufficiency of any nonconceptual sensory given to fulfil the various foundationalist and other roles that the given has been supposed to play in human knowledge, empirical belief, and intentionality in general. It will be best to begin, then, with this familiar conception of the bare, unconceptualized sensory given and with its critique as a myth as far as epistemic or justificatory foundations are concerned.⁵

Consider first a classical if somewhat outdated empiricist picture of human knowledge as resting ultimately on (to use the traditional terminology) our immediate awareness or sense impression of the proper and common sensible qualities of objects. Suppose that Sue sees a red building brick on the table. To pursue the classical line, strictly speaking at present she sees only the red and rectangular facing surfaces of the brick, the rest of the brick not being visible. More strictly still, in light of the possibility of illusion or hallucination, she is visually experiencing a red and rectangular expanse that she takes to be, or sees as, and perhaps believes or judges to be, the surface of a smooth red physical building brick. On this classical empiricist view, the visually apprehended rectangular expanse of red or red ‘sense datum’ is what is sensuously given in the experience, as far as the relevant properly visual qualities are concerned – though of course this is experienced against a complex background of other color expanses, and overall exhibiting visual depth and other properties that are omitted from this oversimplified description. That the red surface is that of a physical building brick, by contrast, is what the strictly given element is perceptually taken to be or experienced as, and this will depend on the conceptual (and/or perhaps the naturally evolved “proto-conceptual”) recognitional capacities, prior learning, and anticipatory expectations of the experiencer.⁶

Sellarsian criticisms of the myth of the given in the above traditional empiricist shape have generally taken the form of arguing that such sensory states of consciousness neither qualify as instances of empirical knowledge themselves (or even as states with intentional purport), nor are such states by themselves sufficient to provide any warrant for an empirical belief. My concern here is not to rehearse and evaluate these arguments and the usual objections to them. For present purposes, at least to start with, various widely discussed remarks of Sellars’s will for many readers call to mind some of the more familiar arguments that arise in this connection: for example, Sellars’s claim that to characterize something as a knowing is to place it “in the logical space of reasons, of justifying and being able to justify what one says” (1963 EPM §36); or his normativist contention that “the idea that epistemic facts can be analyzed … into non-epistemic facts … [is] a radical mistake – a mistake of a piece with the so-called ‘naturalistic fallacy’ in ethics” (EPM §5); or his view that “Sensations are no more epistemic in character than are trees or tables, and are no more ineffable” (1963 SRLG §40).

The expansion of these remarks into arguments by Sellars and others has taken many forms. Rorty, for example, highlighted Sellars’s emphasis on the social aspects of normative reason-giving in relation to one’s peers, vividly distinguishing questions of causation from questions of justification. Rosenberg (2007), Brandom (1994, 2015), and McDowell (1996, 2009) developed the Kantian strands in Sellars’s thought and argued that not only epistemic justification but more fundamentally the very possibility of having any conceptually contentful empirical thought or intentionality about a world of at all requires a background of conceptual capacities, normally expressible in a natural language, that serve as implicit criteria of correctness and incorrectness of application.⁷ On these neo-Kantian and neo-Hegelian views there is no conceptual content or intentionality properly speaking in the first place without a background “space” of rational normativity of this kind.⁸

In contemporary epistemology these debates have coalesced around what has come to be known in the literature as “the Sellarsian Dilemma” that is argued to confront all foundationalist responses to the sceptical regress of justifications. Here the problem concerns how any allegedly foundational basic belief is supposed to be justified in terms of the supposed immediate experience of or direct acquaintance with the given. A frequently cited articulation and defense of the Sellarsian dilemma is from BonJour (1985) (before his later (2003) conversion to foundationalism):

[T]he proponent of the given is caught in a fundamental and inescapable dilemma: if his intuitions or direct awarenesses or immediate apprehensions are construed as cognitive, at least quasi-judgmental (as seems clearly the more natural interpretation), then they will be both capable of providing justification for other cognitive states and in need of it themselves; but if they are construed as noncognitive, nonjudgmental, then while they will not themselves need justification, they will also be incapable of giving it. In either case, such states will be incapable of serving as an adequate foundation for knowledge. This, at bottom, is why empirical givenness is a myth. (BonJour, 1985, p. 69)

Willem deVries and Timm Triplett put forward a structurally similar account of the Sellarsian dilemma as a reconstruction of what they called Sellars’s “Master Argument against the Given” (2000, Ch. 4). Briefly, the “doctrine of the given requires that for any empirical knowledge that p, some epistemically independent knowledge g is epistemically efficacious with respect to p” (2000, p. 104). The Sellarsian dilemma then takes this form: If the given, g, is non-propositional (for example, a nonconceptual sensory or qualitative state), then arguments are put forward that g cannot itself function as a premise in any justificatory reasoning for any empirical belief, and is thus epistemically inefficacious. (Compare Donald Davidson’s well-known claim (1986, p. 310) that “nothing can count as a reason for holding a belief except another belief.”) However, if g is propositional, then the Sellarsian anti-givenists will argue, on the sorts of holistic and normativist grounds mentioned earlier, that the conceptual response g is no longer epistemically independent or presuppositionless in the way that the givenist assumes.

There have been vigorous objections to the required supporting arguments offered for each of these horns of the Sellarsian dilemma. In my view these issues retain their importance despite not being addressed directly to the considerations that I will be raising in the remainder. As indicated, I am concerned here with the assumption that the Sellarsian critique has a target that renders it irrelevant to certain widespread views even if the epistemic Sellarsian Dilemma’s supporting arguments are sound. As mentioned, on the one sort of reaction, Sellars’s target is alleged to be irrelevantly thin in this sense: while Sellars and others before him rightly rejected the impoverished and atomistic empiricist conceptions of the “bare” nonconceptual “sense-datum” given that Sellars focused on in EPM, those rightly rejected impoverished conceptions are simply irrelevant to the givenist defender’s own (assumed to be) innocently “thick” given, carefully conceived, for example, as a richly phenomenological and concept-laden analysis of our ordinary embodied and social “lived experience,” taken initially on its own terms.⁹ Conversely, however, the other sort of reaction contends or assumes that Sellars’s target in the Myth is irrelevantly thick, in particular due to Sellars’s (alleged) over-intellectualizing focus on the question of propositional justification and intellectual reason-giving, thereby not touching the nature of the defender’s own conception of a safely non-epistemic, non-justification focused phenomenological conception of the experiential given.

Both sorts of cases are difficult to recognize as even possible instances of the Myth, but both were in fact central to Sellars’s concerns, and it will be the burden of the following sections to clarify how this is so. For example, how could a properly “thick” (for example, Kantian) understanding of direct perceptual experience as itself already conceptually informed possibly foul afoul of Sellars’s famous critique? For as mentioned, McDowell, Brandom, and others have rightly stressed that one of Sellars’s deepest insights, shared with Kant and Hegel, is that nonconceptual sensory processes alone must fail to account even for the content or objective purport of our world-directed intentionality; so that the key to avoiding the Myth is precisely to recognize the role of our conceptual capacities as being already operative in sensibility in this regard. Nonetheless, while that is importantly true, we shall see that Sellars held that a wide variety of conceptually “thick” conceptions of direct experience (phenomenological, rationalist, “ordinary language,” directly realist “theories of appearing,”¹⁰ and others) have either explicitly or implicitly assumed that their most basic conceptions of the nature of our experience (i.e., of what sort the experience is) are in principle invulnerable to the sorts of alternative reconceptions and challenges to which Sellars argues they are indeed always potentially subject. Furthermore, it is their seeming invulnerability to the accusation that renders such views all the more important though difficult to diagnose, as Sellars was keenly aware.

Similarly in the opposite direction, where Sellars’s concerns have seemed to be irrelevantly conceptually “thick” to the defender of the given: How could any view that presents the sensory “given” as a resolutely non-epistemic, non-justificatory, or nonconceptual factor in our perceptual experience possibly be subject to Sellars’s “conceptualist” and normativist critique of the given, which was precisely concerned only with conceptually “thick” questions about conditions on epistemic justification regarded solely within the normative “space of reasons”? Furthermore, Sellars himself defended his own non-epistemic postulation of nonconceptual sensory representations, put forward as a best explanation of various aspects of our perceptual cognition, and which he clearly regarded as innocent of the Myth precisely in virtue of it nonconceptual nature and its non-justificatory role. Nevertheless, here again it will turn out that the deepest or “most basic” diagnoses of the Myth concern precisely those aspects of views that have seemed most invulnerable to it.

Or so I will contend. It is in attempting to expose and clarify the grounds in Sellars’s account for replying to each of these two opposed irrelevancy reactions that the question of the intended scope and nature of the myth of the given will itself begin to occupy center stage. In the next section I begin with the nonconceptualist’s “thin given” in relation to which Sellars’s targeted justificatory given has been regarded as “irrelevantly thick.” Following this in section §3 it will also become clear, on structurally similar grounds, why it is that no matter how conceptually and phenomenologically “thick” and embodied a conception of the given might be, the challenge represented by the myth of the given remains a further question to be considered.

2 Lewis’s “thin” sensuous given and Sellars on the myth of the categorial given

C. I. Lewis’s classic 1929 book, Mind and the World-Order: Outline of a Theory of Knowledge (MWO) provides an excellent example of a sophisticated theory of a nonconceptual given, the predominant view of which among recent interpreters of Lewis is that his conception of “The Given Element in Experience” (the title of chapter two) is a safely non-epistemic conception to which Sellars’s Myth is thus held not to be applicable. Furthermore, such interpreters argue that Lewis’s overall conception of knowledge was not foundationalist, despite this having been traditionally assumed. Zarębski has recently articulated this widespread interpretation of Lewis as follows¹¹:

since Lewis’s conception of givenness is not directly epistemological in the Sellarsian sense and his account of empirical knowledge has roughly a Sellarsian [i.e., non-foundationalist] structure, then Lewis’s conception of the given is immune to Sellarsian criticism of the ‘myth of the given’. (Zarębski, 2017, p. 200)

Without conceding the historical claim with respect to Lewis and foundationalism, we can simply set aside that question since the structure of knowledge, whether foundationalist or not, will ultimately be seen not to be Sellars’s primary target in relation to Lewis’s given.¹² In fact, for the sake of argument, let us assume along with some of his recent defenders that Lewis’s nonconceptual given plays no directly or indirectly epistemic or justificatory role, though this certainly appears to conflict with his later writings on the given (e.g., Lewis, 1952). We can also fully recognize the fact that Lewis rejected “sense-data” and “phenomenalism” as Lewis interpreted those positions. In particular, Lewis argues that such views illicitly import conceptualization into what is in fact a nonconceptual given element in our knowledge (O’Shea, 2021). The question to be considered now, then, is whether Lewis’s nonconceptual given element in experience, even in its officially non-epistemic causal and phenomenological roles, is “immune” to, or to the contrary is a proper target of, Sellars’s myth of the given. For this we need at least a brief look at Lewis’s subtle account of the given.

Lewis argued that adult human perceptual knowledge involves two fundamental “elements”: a concept possessed by the knower, and a sensory “presentation” or given element that is received from the world (1929, pp. 37, 52). He sought by philosophical reflection to “isolate” the “thin given of immediacy” from “the thick [conceptualized] experience of the world of things” (Lewis, 1929, p. 54), while emphasizing that knowledge properly speaking pertains only to the latter. He defends two criteria for the given element: the given is what is unalterable by thought and has an immediate sensuous quality. Lewis recognizes the difficulty in seeking thus to abstract, in philosophical reflection, from the inevitable intrusion of concepts into our attempted descriptions of the given:

While we can thus isolate the element of the given by these criteria of its unalterability and its character as sensuous feel or quality, we cannot describe any particular given as such, because in describing it, in whatever fashion, we qualify it by bringing it under some category or other, select from it, emphasize aspects of it, and relate it in particular and avoidable ways . . . So that in a sense the given is ineffable, always. Yet no one but a philosopher could for a moment deny this immediate presence in consciousness of that which no activity of thought can create or alter. (Lewis, 1929, pp. 52–53)

Lewis’s sophisticated attempts to “isolate” what is strictly or “immediately” given in experience, while thus seeking to abstract from any implicit “categorial”¹³ conceptual classifications, is exemplified by the following:

At the moment, a certain ‘that’ which I can only describe (in terms of concepts) as a round, ruddy, tangy-smelling somewhat, means to me ‘edible apple’ . . . An object such as an apple is never given; between the real apple in all its complexity and this fragmentary presentation, lies that interval which only [conceptual] interpretation can bridge. The “objectivity” of this experience means the verifiability of a further possible experience which is attributed by this interpretation. (Lewis, 1929, pp. 119–20)

Lewis’s “thin” sensuous given element in our knowledge thus takes the form of ineffable and fragmentary “qualia” presentations, and the given’s “content is either a specific quale (such as the immediacy of redness or loudness) or something analyzable into a complex of such” (1929: pp. 59–60). Furthermore, while “the presentation as an event” is unique, “the qualia which make it up are not. They are recognizable from one to another experience” as “repeatable content” (1929: p. 60, italics added). At the same time, however, Lewis holds that “such qualia, though repeatable in experience and intrinsically recognizable, have no names. They are fundamentally different from the universals of logic” and from the “objective properties” of conceptualized objects (1929: pp. 61, 131, italics added).

Our question, then, is this: since we are for the sake of argument assuming with some of his defenders that Lewis’s “thin” nonconceptual given is not in the business of providing any epistemic justification for our “thick” empirical beliefs about ordinary objects, is Sellars’s Myth objection thereby rendered inapplicable to Lewis’s careful isolation of strictly nonconceptual sensory qualia repeatables within our experience, as these defenders assume?

This brings us to the heart of the matter. For in fact Sellars’s conception of the myth of the given was not restricted to debates concerning epistemic justification, though it was of course intended to have the anti-foundationalist epistemic implications canvassed in section §1. In an important later attempt to clarify his views on the given in his 1981 Carus Lectures – but we’ll see that the idea was central to EPM as well – Sellars articulates what he characterizes as “perhaps the most basic form of what I have castigated as ‘The Myth of the Given’,” and which he states in terms of the following “principle”: “If a person is directly aware of an item which has categorial status C, then the person is aware of it as having categorial status C” (1981b, FMPP I §44).¹⁴ As mentioned earlier, this has come to be known as Sellars’s myth of the categorial given,¹⁵ and I will argue that Sellars is right to have seen that this is in a sense “the most basic form” of the myth of the given. So, first a few remarks about categories and the “categorial given,” before returning to Sellars in relation to Lewis’s nonconceptual qualia-repeatables in particular.

Categories, on Sellars’s view, are roughly speaking (meta-)conceptual classifications of types of concepts, classifying the latter as functioning within some cognitive-linguistic framework that serves to represent the world as being a certain way and thus as containing various sorts of things and events. The categorial ontologist or descriptive metaphysician can then analyze such everyday, scientific, or philosophical concepts as implicitly categorizing the world in terms of various general sorts of particulars, properties, relations, states of affairs, processes, activities, dependencies, and so on.¹⁶ Sellars holds that our sophisticated common sense conception or “manifest image” of the perceptible world has been analyzed in different ways by different philosophers, and that there is truth and falsity with respect to how accurately or coherently these analyses have succeeded in capturing the implicit categorizations of our common sense framework of persons, bodies, and their qualities and relations (1963 PSIM I–III, 1977 SK I, 1981b FMPP). Furthermore, Sellars stresses, both scientific theorists and metaphysicians have over time offered various alternative categorizations of the fundamental nature of the world and of our knowledge and values within it, in seeking how best to explain how it all hangs together.

Sellars’s views on the myth of the given are articulated within this general conception of categorial conceptual frameworks and the metaphysics of epistemology (see this article’s opening epigraph). The key point in relation to the myth of the given is Sellars’s contention that philosophers have almost always assumed, whether explicitly, implicitly, or by unwitting implication, that some fundamental categorization of the world or of persons is irrevisable, and so is assumed to be just “given” in that sense. These categorizations often cut so deeply and only implicitly that in some cases (though not all – there are many types of cases, including more local ones) the only way to bring out that they are not compulsory conceptions (in cases where they are implicitly assumed to be such) might be to sketch or construct an alternative categorial explanatory framework, for example one that might seek to resolve an ostensible clash between frameworks. This is one important reason why Sellars’s writings¹⁷ are littered with the question, “But what is the alternative?” – followed by Sellars’s various attempts to construct or reveal a different conception and in some cases an alternative categorial reconception of some phenomenon that the given philosopher had, by Sellars’s lights, mistakenly assumed to be innocent or non-controversial. This also helps to account for the dialectical, historical, and indirect nature of Sellars’s modes of argumentation, and for the resemblance many have thought these bear to aspects of both Hegel’s and Peirce’s philosophical methods. As many have noted, it has seemed notoriously difficult (to take the case at hand) to isolate a single, nicely formulable argument in Sellars that is supposed to apply against all cases of the mythic given. It has struck many that Sellars always seems to work with historical philosophical examples, to raise internal tensions within them but not to attempt any knockdown arguments (Sellars states in various places that philosophy does not provide knockdown arguments), and then to sketch an alternative explanation, and in some cases a fundamental categorial reconception, of the view at hand. I am suggesting that this is not an accident, and that in many cases this signature method has to do with “the most basic form” of the myth, the myth of the categorial given.

The question of the ultimately correct categorial ontology of the sensible qualities provided perhaps the most important example of this fundamental kind for Sellars, though as we shall see it was far from being the only type of case. A key motivating interest in the myth of the given for Sellars in this particular case was ontological. The extant categorial ontologies of the sense-datum givenists, for example, but also, alternatively, of the ordinary language philosophers and of directly realist “theories of appearing,” all in his view served to block the road of inquiry and to render invisible what Sellars takes to be at least a “live” alternative, though admittedly a speculative one. The case at hand concerns the ultimately correct categorial analysis of the “location” of color and other sensible qualities in the ontological scheme of things. Sellars’s continual career-long efforts to provide a radically different, systematic alternative to those conceptions of the place of the sensible qualities in the total scheme of things is supposed to jar us from what has seemed to be at least a categorially irrevisable aspect of the given conception of things under examination (however revisable it may be in other ways). The articulation of an at least equally coherent or plausible alternative categorial scheme is sufficient to dispel the myth of the categorial given, without the criticized givenist necessarily having to accept Sellars’s particular preferred alternative conception.

The myth of the given, in perhaps its most basic form, is thus the idea that there is some implicit categorization of whatever is under consideration that is assumed to be in principle not revisable or replaceable by a fundamentally different categorization in this way. This is due to the fact that the given type of presence or entity is thought or assumed to be revealed as it really is (following Sellars, call this “sort” or way of being, ‘C’), simply in virtue of (for the givenist) one’s being directly aware of it under this implicit, and perhaps unwitting categorization, C. Hence, again, the myth of the categorial given: “If a person is directly aware of an item which [ultimately, really] has categorial status C, then the person is aware of it as having categorial status C,” that is, just in virtue of being directly aware of it (FMPP I, §44, interpolation added). In other words, the direct awareness by itself is implicitly assumed to reveal the item as, in this case, the sort of item it must really, ultimately, be correctly conceived to be. To the contrary, Sellars contends that there is no species of “direct awareness,” whether a non-conceptual or a conceptualized awareness (the myth will apply to both types of cases), that is somehow invulnerable to potentially being revealed as an implicit or explicit miscategorization in light of a possible explanatorily superior recategorization of that item’s fundamental nature (using “item” here merely as a placeholder).

To return now to C. I. Lewis’s particular conception of the sensuous-qualitative given element in our knowledge, Lewis as we saw explicitly seeks to characterize the given as a strictly nonconceptual sensuous element in our awareness that receives active conceptual categorization and re-categorization in terms of our ongoing and revisable conceptual interpretations, but which the reflective philosopher nonetheless can “isolate” as an immediate and ineffable qualitative element in our conscious experience. As we saw, “such qualia, though repeatable in experience and intrinsically recognizable, have no names” (1929, p. 61). Sellars, however, contends that our first-order awareness of such sensible qualities as Lewis’s “redness or loudness,” however immediate, always already embodies an implicit categorial ontology, whether this be (the particular list here is not important): (a) a “manifest” naïve realism that represents such qualities as constituent-characters of physical objects (as Sellars argues is in effect the case with our evolutionarily inherited, “innate” way of (partially mis-)representing the world); or (b) such qualities or qualia or tropes as might be claimed to be reflectively isolated in a philosophical sense-datum theory; or (c) in the form of intrinsically recognizable Lewisian nonconceptual qualia-repeatables; or (d) in a nominalist “pure process” metaphysical recategorization of such repeatables (cf. Sellars 1981b, FMPP II); or (e) in a scientific-metaphysical recategorization and “relocation” of such sensible qualities as adverbial states of sensing in the perceiver’s central nervous system; and so on. Sellars’s contention is that Lewis’s ostensibly categorially neutral account of the nonconceptual sensuous qualia-given, which officially abstracts from all conceptual interpretation, has in fact implicitly categorized such allegedly intrinsically recognized repeatables in a way that itself represents a categorial choice among such alternatives as those just mentioned. This point will take some hammering home.

Sellars is in agreement with what Lewis called his pragmatic conception of the a priori (Sellars, 1963 ITSA §7), at least as far as their shared emphasis on the importance of categorial conceptual change is concerned (though there are significant disagreements between them here, too (O’Shea 2018)). But what Sellars thinks is true of Lewis is in essence the same as what he argued in EPM VI is true of Locke, Berkeley, and Hume: namely, that the latter three philosophers all took “for granted that the human mind has an innate ability to be aware of certain determinate sorts – indeed, that we are aware of them simply by virtue of having sensations and images” (EPM VI, §28). In this case, as with Lewis’s description of immediately and intrinsically recognizable qualia repeatables, the form of the myth of the given that Sellars is targeting is “the givenness of determinate kinds or repeatables, say crimson” (EPM §29; cf. Hicks, 2020 similarly on Sellars on H. H. Price).

What Sellars makes more clear in other writings, however, is that it is not the innateness in this case that is the real issue, but rather the unquestioned yet optional (even if it is innate) categorization that is involved. For he argues, for instance, that our innate, “proto-conceptual” perceptual categorization of the world naturally embodies a naïve-realist miscategorization as far as the real ontological nature and location of the properly sensible qualities are concerned. He usually mentions in this regard the views of one of his former Oxford teachers, H. A. Prichard,¹⁸ and more generally he endorses the contention that our natural sensory endowments were designed to get us around the environment rather than to represent entirely accurately the world’s true categorial ontology.¹⁹ What matters is not innateness vs. acquisition per se, but rather the ever-present possibility in either case of an alternative re-categorization in relation to any phenomenon, however “immediate,” which Sellars takes the givenist either explicitly, implicitly, or by unacknowledged commitment to have denied or failed to recognize in the case of our direct awareness or acquaintance with (in this case) determinate sense-repeatables. What matters is that the road of speculative categorial inquiry has thereby unwittingly been blocked, by failing to recognize such a possibility in relation to the seemingly innocent assumption of an immediate nonconceptual qualia-presence to consciousness.

I have several times mentioned Sellars’s notion of “proto-concepts.” Sellars reserved the unqualified notions of “concepts” and “propositions” for reason-giving beings who have the logical resources to explicitly represent reasons as reasons, and who thus have the ability to infer and act in light of principles within a logical space of reasons (1981a, MEV). But he argued that sophisticated non-logic representing animals and infants nonetheless have the cognitive resources, thanks to inherited adaptive capacities and learning mechanisms within a biological “space” of proper-functioning (as it were), sufficient to “track” and represent objects as having various properties and as standing in various relations. Sellars thus held that the “concept of innate abilities to be aware of something as something, and hence of pre-linguistic awareness is perfectly intelligible,” and that conceptual or “propositional form is more primitive than logical form” (1981a, MEV §§56–60). He often calls such “primitive” animal and human infant concepts “ur-concepts” or “proto-concepts” in order to register the relevant functional contrasts and similarities. In a similar way he writes of the human toddler’s “proto-theory” of the world prior to the child’s development of more complex distinctions between how things look or seem as opposed to how they are (1981b, FMPP I). These late-career notions figured in Sellars’s critique of the categorial given in Lewis (and in Roderick Firth’s developments of Lewis) roughly as follows.

Sellars explains that “We, as phenomenologists, can bracket the concept of an expanse of red in that radical way which involves an abstraction from” any particular categorial determination of what sort of item that expanse of red is in fact determinately conceived to be (FMPP I §84), whether as a part of a visible physical object, or alternatively as a state of sensing in the perceiver, for example, as noted earlier. Sellars himself argues that in fact our animal or infant “proto-conceptual” starting point, and so our initial implicit categorial ontology, is most likely of the former kind, while the latter “relocational” recategorization is a much further step along Sellars’s way. The temptation he thinks givenists such as Lewis succumb to, however, is to characterize their supposedly neutral, indeterminate, isolated, or bracketed conception in ways that are in fact not categorially neutral or innocent in the way supposed, or so Sellars or the Sellarsian critic contends.

It is in this way that a merely nonconceptual given can be unofficially and implicitly conceptualized, and thus potentially miscategorized (depending on further arguments), while being advertised as innocently nonconceptually given. In the case of Roderick Firth (discussed in FMPP I), Sellars’s diagnosis of the myth of the categorial given consists in attempting to expose Firth’s (alleged) assumption that the bracketed phenomenal “expanse” must in fact be a subjective state of consciousness in the infant or toddler’s consciousness (while the toddler lacks our explicit adult “is/seems” distinctions). To the contrary, Sellars argues that there is an alternative, and as it happens more likely sketch of an account according to which the toddler’s ur-conceptual representation of the world is interpreted as an unsophisticated direct (or “manifest”) realism about the “location” of the sensible qualities of physical objects. Similarly, in the cases of Locke, Berkeley, and Hume, as well as of Lewis, the diagnosis of the categorial myth takes the same structural form, but in this case in relation to “determinate sense-repeatables.”

So not withstanding the fact that Lewis himself would of course deny that our immediate awareness or “recognition” of the given in the form of such repeatable qualia-contents involves any implicit categorization of what sort of item they are at all, this is in fact the real core of the dispute between Lewis and Sellars on the given, and Lewis is guilty as charged.²⁰ Here is how Sellars had put this in his 1953 article, “Is There a Synthetic A Priori?”:

All classification of objects, however confident and pre-emptory, is a venture, a venture which at no point finds its justification in a pre-symbolic vision of generic and specific hearts on the sleeves of the objects of experience. . . . I am afraid, however, that our agreement with Lewis is more shadow than substance. For while he writes in this manner of the interpretation of the given by means of concepts whose implications transcend the given, he also holds that the sensible appearances of things do wear their hearts on their sleeves, and that we do have a cognitive vision of these hearts which is direct, unlearned, and incapable of error – though we may make a slip in the expressive language by which these insights are properly formulated. In other words, the assumption to which we are committed requires to extend to all classificatory consciousness whatever, the striking language in which Lewis describes our consciousness of objects. (Sellars, 1963 ITSA VII §44; italics added)

Lewis himself, of course, would not call such a vision “cognitive” if this term is taken to entail “conceptualized” or “known,” since the repeatable qualia-givens he seeks to “isolate” by reflection are officially not conceptually categorized in any way whatsoever. The latter, however, is what Sellars’s Myth charge denies has been achieved: Lewis’s bracketing “isolation” is in fact, in the details, conceptually (and so categorially) contaminated rather than innocent. (In this Lewis is in the distinguished company of many other great philosophers.)

Three decades after EPM in the 1981 Carus Lectures (FMPP), therefore, Sellars has now developed the same species of underlying charge more explicitly in terms of the alternative categorial classifications that make clear that there is no avoiding some implicit categorization or other of the sorts of item that are taken to be involved in our direct (non-inferential) awareness of, in this case, the sensible qualities. In FMPP Sellars explicitly mentions Lewis as subject to the Myth only in passing as follows (his focus is on Firth’s defense of Lewis’s view), and of course knowing full well here that Lewis would not himself describe his quale as a “category”: “In the grip of the Myth of the Given, a C.I. Lewis might be tempted to say that to the careful mind the expanse of red presents itself as a quale, the latter being the one and only basic category which is above the pragmatic competition of the market place” (1981b FMPP I §85). Sellars makes clear throughout FMPP that there is an explanatory competition even with respect to Lewis’s immediately given element itself, and indeed in his view with the most momentous consequences. By Sellars’s lights the ontological nature and location of such items is and ought to be wide open to possible alternative fundamental categorizations:

The one thing we can say, with phenomenological assurance, is that whatever its ‘true’ categorial status, the expanse of red involved in an ostensible seeing of the very redness of an apple has actual existence as contrasted with the intentional in-existence of that which is believed in as believed in. But notice that the family of concepts to which this contrast belongs consists of transcendental concepts, i.e., concepts which apply across categories. An expanse of red could be something actual and be either [1] a sense datum in visual space, [2] a manner of sensing, or [3] a spatial constituent of a physical object.” (1981b, FMPP I §88; numbering added)

Sellars believes that there are more competitors in the marketplace of categorial competition than between [1], [2], and [3] – his inquiry here is restricted to competing categorial analyses of the “manifest” framework (as in the epigraph that opens this chapter) – but that fact serves only to highlight further his basic point about the fundamental nature of the myth of the categorial given.

My aim in this section has been to explain what is “perhaps the most basic form” of Sellars’s myth of the given: the categorial given, and in particular how this is meant to apply to the case of Lewis’s “thin” nonconceptual given. The next section will show how this perspective also helps to explain what is common to many of the other very different forms and targets of the myth accusation that notoriously show up throughout Sellars’s philosophy, which will shed a different light on Sellars’s own views as a whole. By the end it should also be clear, if only by implication, why this improved understanding of the myth of the given has widespread consequences for the various contemporary disputes about of the myth of the given mentioned at the outset, whichever side of those disputes one might fall on.

3 “The Framework of Givenness”: varieties of the myth of the categorial given

Although most contemporary debates about the myth concern the sensory given in relation to the problem of perceptual warrant as discussed in section §1, it is well known that Sellars opens EPM by listing a wide variety of other versions of the given:

Many things have been said to be “given”: sense contents, material objects, universals, propositions, real connections, first principles, even givenness itself. And there is, indeed, a certain way of construing the situations which philosophers analyse in these terms which can be said to be the framework of givenness. This framework has been a common feature of most of the major systems of philosophy, including, to use a Kantian turn of phrase, both ‘dogmatic rationalism’ and ‘sceptical empiricism’ … And many who today attack ‘the whole idea of givenness’ – and they are an increasing number – are really only attacking sense data. For they transfer to other items, say physical objects or relations of appearing, the characteristic features of the ‘given’. (EPM I §1)

How could everything on this list possibly be taken to be different versions of a common “framework of givenness”? It will help to look at some of Sellars’s remarks in the context of non-perceptual and non-justificatory versions of the given.

The following is a version of the myth of the given that Sellars remarks on in EPM VII (here bearing in mind, as I have argued in section §2, that the acquisition vs. innateness question is not the important one, but rather that of the neglected availability in principle of a categorial alternative):

There is a source of the Myth of the Given to which even philosophers who are suspicious of the whole idea of inner episodes can fall prey. This is the fact that when we picture a child – or a carrier of slabs – learning his first language, we, of course, locate the language learner in a structured logical space in which we are at home. Thus, we conceive of him as a person (or, at least, a potential person) in a world of physical objects, coloured, producing sounds, existing in Space and Time … In other words, unless we are careful, we can easily take for granted that the process of teaching a child to use a language is that of teaching it to discriminate elements within a logical space of particulars, universals, facts, etc., of which it is already undiscriminatingly aware, and to associate these discriminated elements with verbal symbols. And this mistake is in principle the same whether the logical space of which the child is supposed to have this undiscriminating awareness is conceived by us to be that of physical objects or of private sense contents. (EPM VII §30).

In the course of the second half of EPM it then becomes clear that our adult human categorial “manifest” framework of ordinary persisting and colored physical objects is in principle in turn replaceable, on Sellars’s view, by an explanatorily more adequate competitor “logical space” or categorial framework that he argues is in fact already on the scientific and metaphysical horizon. The first key step is to appeal to the hard-earned rejection of the myth of the given in the first half of EPM:

. . . once we give up the idea that we begin our sojourn in this world with any – even a vague, fragmentary, and undiscriminating – awareness of the logical space of particulars, kinds, facts, and resemblances . . . we now recognize that instead of coming to have a concept of something because we have noticed that sort of thing, to have the ability to notice a sort of thing is already to have the concept of that sort of thing, and cannot account for it. (EPM X §45)

As we saw Sellars argue both in FMPP and in EPM VI (in relation to Locke, Berkeley, and Hume), it is the myth of the categorial given to assume that the direct awareness of any sort of thing by itself provides one with a cognition of that item as the sort of item it ultimately is – that is, as the sort of item it would be correctly revealed to be by whatever is or turns out to be the best overall explanatory account of whatever sorts or categories of things there really are in the world. Such an assumption or presupposition would in effect either dogmatically or sceptically close the door on the marketplace of categorial-ontological and explanatory reinterpretation.

When Sellars himself then notoriously goes on to argue in EPM “that there are no such things as the colored physical objects of the common sense world,” he emphasizes that this is to be taken “not as an empirical proposition... within the common sense frame, but as the expression of a rejection (in some sense) of this very framework itself, in favour of another built around different, if not unrelated, categories. This rejection need not, of course, be a practical rejection” (EPM IX §43). One important target of Sellars’s lectures on the myth of the given in EPM was thus to combat the following implicit dogma:

There is a widespread impression that reflection on how we learn the language in which, in everyday life, we describe the world, leads to the conclusion that the categories of the common sense picture of the world have, so to speak, an unchallengeable authenticity. There are, of course, different conceptions of just what this fundamental categorial framework is. For some it is sense contents and phenomenal relations between them; for others physical objects, persons, and processes in Space and Time. [But each conception] . . . rules out of court as utterly absurd any notion that there are no such things as this framework talks about. (EPM IX §43)

This widespread impression is of course not restricted to mid-century analytic sense-datum theorists, philosophers of ordinary language, or relational theories of appearing. As Sellars sees it this applies also to the rich tradition of continental phenomenology, which he regularly praised and roughly equated with the analysis of the “manifest” or sophisticated common sense framework of the perceptible world (cf. Christias, 2018, O’Shea, 2012, and Sachs, 2014). For he argues that phenomenology ultimately reaches “the end of its descriptive tether” and fails to explain adequately even the perceptual phenomena that fall directly within its own carefully bracketed domain of the manifest life-world (cf. FMPP I §89, and passim).

Finally, at the end of EPM, Sellars puts forward at least a glimpse of his own proffered explanation of the ontology of perception and its ultimate place in the overall scheme of things, by means of a fundamental categorial reinterpretation envisioned in a “micro-theory of sentient organisms” (EPM XVI §61). Without going into the details, Sellars elsewhere (e.g., PSIM VI, §107, “Is there any alternative?...”; and FMPP I–III passim) further clarified and developed this same speculative alternative categorial ontology of ultimate micro-sentience. To use his terminology, this reconception involves “non-physical₂”²¹ qualitative sensings or “sensa” jostling cheek by jowl, in (what in “macro” terms is) the central nervous systems of sensing animals, with other equally fundamental physical₂ micro-physical processes (involving quarks, electrons, etc.), within the all-comprehensive “physical₁” and mostly non-living physical₂ spatiotemporal-causal universe as a whole. (For the details and interpretation of this ontology, see Coleman, 2015, Ch. 2, deVries 2005, Ch. 8, O’Shea, 2007, Ch. 6, Rosenberg, 2007, Ch. 9, and Seibt 2015, Ch. 9.)

Particularly revealing for our present topic of the myth of the categorial given, however, are some passages in the comparatively neglected second Carus Lecture, on “Naturalism and Process” (Sellars 1981b, FMPP II). Here Sellars compares his proposed alternative “pure process” categorial ontology, briefly described above, with the views of neutral monists such as Bertrand Russell, according to whose views, as Sellars puts it: “Of course there are minds and material things. But there really are no minds or material things; for neither minds nor material things are among the ultimate constituents – sensibilia – of which all things are made” (1981b, FMPP II §85).²² For the neutral monists, “statements in which we describe the manifest world of changing things [are] capable of correlation with logically complex statements in a language of which the basic statements ascribe qualities and relations to sensibilia” (§86). Sellars then suggests that something similar is the case for his own ultimate categorial ontology of “sensings” and “quarkings,” et al., as a Heraclitean flux of “pure processes” in nature. Sellars explains that “the correlations” established between “this [Russellian neutral monist] ‘alternative’ framework” and our manifest categorial framework “need not be viewed as offering analysis of manifest statements, i.e., as preserving sense” (§92): “It is rather to say that the one framework is, with appropriate adjustments in the larger context, replaceable by the other – eliminable in favor of the other. The replacement would be justified by the greater explanatory power of the new framework” (§95). Sellars then proceeds to take “the radical step of construing all the ‘atoms’ of our neutral monist model as absolute processes” (§99), that is, in terms of a pure “flux” ontology of ultimate non-physical₂ sensings (e.g., “reddings”) and physical₂ “quarkings,” and so on.

As Rosenberg (2007, Ch. 9) makes clear, this sort of radical categorial replacement ontology was already generically in view, as far as the relevant radical categorial change is concerned, in EPM in 1956. One of Sellars’s primary concerns, accordingly, was that the myth of the given in any of the versions he considers in EPM (whether logical empiricist, phenomenological, ordinary language, relational appearing, or directly realist) would unwarrantedly block the very possibility of such an outlook from the start. Sellars of course defends his own categorial ontology as being the best way, he argues, to provide a fully adequate naturalistic (or at least “physical₁”) solution to the intractable consciousness/body problem (cf. FMPP III). But in these sorts of cases his rejection of the road-blocking myth of the given consists only in exposing the relevant categorial presuppositions and in sketching a coherent categorial alternative, without one’s having to accept, for the purpose of diagnosing the myth, his further arguments in favour of his alternative categorial ontology.

It should be clear at this point that the myth of the categorial given – the myth in “perhaps its most fundamental form” – is indeed Sellars’s primary way of answering both the “nonconceptualist” and “conceptualist” irrelevancy objections discussed in sections §§1–2. Exposing the myth’s applicability where it seems most invisible requires a different set of underlying considerations than those commonly discussed in relation to the problem of perceptual warrant discussed in section §1. In these most fundamental types of cases the diagnosis requires, in particular, categorial-ontological considerations that are additional even to the pivotal and much debated Kantian and Sellarsian claims concerning the holistic and normative “space of reason” conditions on the very possibility of conceptual content or objective purport in general.

Once one sees the myth from this categorial perspective, other very different remarks on the myth of the given made by Sellars fall into place. In a late article²³ on “Hochberg on Mapping, Meaning, and Metaphysics” (1977, MMM), for example, Sellars at one point examines Hochberg’s criticism of Sellars’s nominalism. As discussed in that article, Sellars’s nominalism involves his interpretation of the role of predicates such as “is white” in terms of the functional classification of linguistic tokens (e.g., “‘W₁’ is a •white•”). As Sellars recounts it, Hochberg’s criticism takes place against the backdrop of “the background language in which, presumably, we formulate and dialectically resolve the issues at stake between nominalistic and realistic claims”; but in Sellars’s opinion, Hochberg does not pursue this background aspect further, and he argues that this “accounts for some of the flaws in his [Hochberg’s] argument” (MMM §4, italics added). To come to the key point, Sellars then clarifies, in relation to his own long-standing nominalist categorial analysis, that it:

. . . is intended as a dialectical move in which the background language itself, in so far as it uses such terms as ‘refers’, ‘stands for’, ‘property’, ‘fact’, etc. is being scrutinized. From my point of view Hochberg takes the background language to wear its dialectical heart on its sleeve. If I were name calling I would call him an ordinary language philosopher. My Tractarian search for a perspicuous language is no mere incidental feature of my method. It is reflection within the background language on the background language which leads me to the conclusion that expressions which look like names (for example ‘triangularity’) are not names, but rather distributive singular terms. [Footnote:] The idea that the background language is to be taken at its face value is surely the last bastion of the ‘Myth of the Given’. (Sellars, 1977, MMM §24; italics added)

This makes clear again that the myth of the given, when push comes to shove, is ultimately grounded in some uncritical groundlevel assumption or other, taking different forms in different philosophical contexts, about the background categorial framework that structures the very availability of whatever sorts of entities and corresponding conceptions are taken to be unproblematically available for use or for direct apprehension.

The focus on the myth of the categorial given in the same manner accounts for the various “rationalist” and other forms of the myth of the given that we saw Sellars canvass at the outset of EPM (I §1), quoted at the start of this section. In these cases, too, Sellars’s method in his various examinations of the topic of causality, for example, consists in constructing a categorial (in this case, an epistemically and semantically inferentialist, and ontologically nominalist) alternative to, for instance, rationalist conceptions of the direct apprehension of necessary connections among universals or essential kinds, realistically construed. To put in general terms the central alternative animating principle of his various critiques of the Myth in its more “rationalist” versions, he contends that:

There is nothing to a conceptual apparatus that isn’t determined by its rules, and there is no such thing as choosing these rules to conform with antecedently apprehended universals and connexions, for the ‘apprehension of universals and connexions’ is already the use of a conceptual frame, and as such presupposes the rules in question. (Sellars, 1953 IM VI §47)

To this Sellars adds that “it is characteristic of modern science to produce deliberately mutant conceptual structures with which to challenge the world” (ibid.), thus generating alternative categorial principles and offering potentially better frameworks of explanation as a whole. This is something that he also holds is characteristic of philosophy at its best, when it is not stifled by one or another version of the myth of the categorial given.

4 Conclusion

What are the consequences of this perspective on the myth of the given in perhaps its most basic form for the Myth’s disputed role in the subsequent and ongoing debates that were highlighted in section §1? Some of the most well-known controversies, which are generally focused on the nature and requirements of epistemic justification, in many cases can and should remain engaged as they were before, in the terms originally posed in those debates – for example, as to whether or not propositional, inferential, or social reason-giving are or are not necessary conditions for epistemic justification and perceptual warrant. But as we have seen, it is when one considers either of two of the most widespread types of response (or in most cases, non-response) to the Myth charge, that the most basic categorial form of the myth of the given was brought to bear by Sellars, and this is where I think it becomes most relevant again today, though usually in an unnoticed presuppositional way.

These rejections of the myth as irrelevant, as we saw in sections §§1–2, typically regard the Sellarsian myth accusation as construing the given in such a way that it is simply inapplicable to the defender’s putatively innocent conception of the given. Put crudely, defenders of any phenomenologically and conceptually “thick” conception of the given often assume that they are ipso facto safe from Sellars’s attack on long-forgotten sense-datum theories and other badly impoverished or “thin” misconceptions of the given. Conversely, defenders of any resolutely non-epistemic or nonconceptual “thin” given often assume that they are ipso facto safe from the Sellarsians’ exclusive focus on higher-level epistemic warrant and the holistic conditions on conceptual intentionality within the normative space of reasons. These are not unreasonable reactions, for we have seen that it can seem impossible for Sellars’s Myth to be applicable to either such case, as long as the given is interpreted as either safely non-epistemic and nonconceptual, or safely conceptually informed and phenomenologically rich. What has been argued here, however, is that it is primarily these seemingly innocent cases of the given, assumed to be immune to the myth for either reason, that at the deepest level were targeted by Sellars’s myth of the categorial given, depending on further considerations concerning revisability and the open availability of possible coherent categorial alternatives.

In a way this puts a burden on the accuser of the Myth, as perhaps it should, since no doubt every mature philosophical view regards itself as having already considered relevant alternatives and found them to be wanting by comparison. It is in this way, however, when no further alternatives seem relevant or even potentially applicable, that further inquiry into a given presupposition has often ground to a halt. What Sellars’s dialectical conception of the myth of the categorial given attempted to show was that just about every classic philosophical position has rested on background starting points regarded as innocent through lack of awareness of an alternative reconception that reveals that innocence to have in fact been lost from the start. Whether Sellars’s own view is able entirely to escape this same sort of accusation is of course another question, and one that he would want us to ask.²⁴