Watkins’s Objection to McDowell
Watkins argues, pace McDowell, that the Myth of the Given is in fact captured by Non-conceptual Content and there is nothing mythical about Crossing the Line." Watkins endeavors to show this by appeal to what he takes to be Kant’s view that sensations are non-normative states that play a role in the justification of belief:12
The crucial task for the Kantian here lies in showing that the Myth of the Given, as McDowell understands it, is not necessarily a myth at all. ... In short, the goal is to explain further and more clearly how sensations could guide our concepts and judgments from without, but without themselves being characterized conceptually or as propositionally articulated judgments.
(Watkins 2008, 519)
Watkins says here that, since sensations are not characterized conceptually and are not propositional, they do not belong in the logical space of reasons, but rather “below the line,” in terms of McDowell’s image. Nonetheless, Kant’s account of how sensations relate to judgments -which do have standing in the space of reasons - makes it intelligible that sensations can have a justificatory role in relation to belief. This is what is meant by saying that sensations can “guide our concepts and judgments from without” - from without the logical space of reasons, that is. If this is right, it shows that McDowell is wrong to suppose that Crossing the Line captures the Myth of the Given.
Note the shape of Watkins’s proposed argument against McDowell: Kant’s view is presented as a counterexample to the thesis that Crossing the Line is a guise of the Myth of the Given. The characterization of Kant’s position Watkins proceeds to give is intended to strike us, uncontroversially, as assigning a justificatory role to sensations, which are conceived as non-conceptual states. I will argue that this counterexample fails. One way to do this would be to question the accuracy of Watkins’s interpretation of Kant. This is not what I wish to do here.13 Instead, I will argue that once the full implications of the claim that the intellect is spontaneous are spelled out it becomes clear that sensations, conceived along the lines Watkins proposes, cannot play the justificatory role his reading of Kant assigns to them.14
Let me add one further point of clarification before I present Watkins’s counterexample. It is sometimes alleged that unless our beliefs are “guided from without” the logical space of reasons, they are not properly constrained by perceptual experience - and so a view like McDowell’s either threatens to degenerate into an unacceptable idealism or else makes it impossible to see how beliefs could have empirical content.15 Watkins has sympathies for this objection, but it is important to note that this is not the issue his Kantian counterexample is intended to address. Here the issue is to demonstrate, not the possible need for guidance from without, but instead the ability of sensations to guide from without. In other words, what is at issue is the coherence of the very idea of a justificatory relation that crosses McDowell’s line.
Central to Watkins’s reading of the relevant area of Kant’s thought is the claim that a concept is, or involves, a certain type of function, which is defined over sensations and yields conceptually articulated representations - what Kant calls judgments - as outputs. The judgments that are the output of a conceptual function are normative, since the ability to stand in relations of justification is essential to them, and so are the conceptual functions themselves. The sensations that are the input to these functions, however, are not as such normative. Yet, they play a role in the justification of judgments because they act as input to the conceptual functions. Here is Watkins’s own description of the position:
The crucial line of argument here is that if one understands a concept in terms of a function that not only requires that a manifold be given as its input, but also specifies what can and cannot serve as the given input, then one is in a position to assert that the input for empirical concepts must be sensations, that is, certain naturalisti-cally describable causal effects that objects have on us. . . . Such a requirement does not commit the naturalistic fallacy, e.g., by treating sensations as if they both are and are not normative entities or as if they could, on their own, justify knowledge without themselves being justified.. . . Instead, this line of argument exploits the asymmetry in how sensations and concepts operate in cognition, by pointing out that justification can occur when normative functions that require a certain kind of input are then given that input, where the input, by necessity, takes the form of sensations.
(Watkins 2008, 523)
So, on Watkins’s view, sensations have a bearing on the justificatory credentials of items whose proper home is above McDowell’s line without being themselves located above the line. Why do they have such a bearing? Because “the input that is given to us [i.e., sensations] and the function that takes that input [i.e., the conceptual function] jointly determine the output” (Watkins 2008, 521).
According to this picture, the normativity that is required to avoid the Myth of the Given derives from the fact that conceptual functions “require” a particular input. What kind of requirement is this? In particular, can such a requirement ground normativity of the right sort? There are at least two different ways of understanding the claim that a conceptual function requires a particular input.
On the first way of understanding the claim, to say this is to say that for certain kinds of input the function does not yield an output. Just like the mathematical function of division does not yield a value for the argument 0, so Kantian conceptual functions do not yield an output for, for example, numbers. But this simply means that the function is defined only for certain types of input rather than others. While this claim can be expressed in normative language - there is a sense in which those inputs for which the function is not defined are “incorrect” - this does not ground normativity in any interesting sense. In particular, there is no connection with the issue of justification. So, this is presumably not what Watkins has in mind.
An alternative way of understanding the claim is in terms of the notion of an application-condition for a concept. Here the idea is that the concept F is correctly applied when, and only when, certain kinds of sensation are present. Thus, we might say that the concept
Let us assume, then, that this is the position Watkins advocates.16 In the remainder of this chapter, I wish to argue that this position faces a fatal objection. The objection is that, appearances to the contrary, it fails to show that sensations - qua items situated below McDowell’s line - play a role in the justification of belief. As a consequence, it cannot vindicate the claim that the Myth of the Given should be understood along the lines of Non-conceptual content, as opposed to Crossing the Line.
But before I turn to this objection, I wish to note that the position Watkins attributes to Kant bears a striking resemblance to the view Lewis develops in Mind and the World Order. To begin with, Lewis also holds that perception-based beliefs are jointly determined by two elements -one that belongs below McDowell’s line and one that belongs above the line. These are, on the one hand, what Lewis variously calls the immediate presentation, the given element, and the qualia of immediate experience, and, on the other, what he characterizes as the interpretation of the given contributed by our conceptual activity; that is, by taking a case of the given to be the presentation of an instance of some concept F. Here is how Lewis puts it:
The world of experience is not given in experience: it is constructed by thought from the data of sense. This reality which everybody knows reflects the structure of human intelligence as much as it does the nature of the independently given sensory content. It is a whole in which mind and what is given to mind already meet and are interwoven.
(Lewis 1929, 29ff)
That the “data of sense” have their place below McDowell’s line is clear from the fact that they are unstructured and can be enjoyed independently of the possession of conceptual capacities. Thus, Lewis characterizes as “entirely erroneous” (1929, 101) the view
that human experience, at the lowest level at which it can be discerned -the level here indicated by “givenness” - already possesses a structure which reflects the nature of “human reason”.
(Lewis 1929, lOOff)
On the contrary:
There is no reason to think that the absence of human categories [that is, of concepts, TL] affects the content of [a creature’s] given experience in the least.
(Lewis 1929, 101)
Nonetheless, just like Watkins’s sensations, Lewis’s immediate experience constrains conceptual thought, since it sets correctness-conditions on the application of concepts. Although no single encounter with the given -which, on Lewis’s view, is not temporally extended - by itself can do so, classifying such an encounter as a presentation of an F (or an instance of F-ness) commits the thinker to a class of conditionals that predict what future experiences she will enjoy (and would enjoy in relevant counter-factual situations), if the classification is correct. Thus:
The qualia of the given are the clue to the applicability or inapplicability of concepts, and set the limits of conceptual interpretation.
(Lewis 1929, 157)17
Just like Watkins, then, Lewis holds that there are sensory states that do not themselves have standing in the space of reasons - they belong below McDowell’s line - but that nonetheless play an essential role in the justification of empirical judgments. Moreover, though Lewis does not employ the notion of a function in developing his account of concepts, this notion is applicable to it, since the account shares crucial features of Watkins’s view. Thus, as already indicated, Lewis holds that it is an essential feature of an empirical concept that it relates a current experience to which it is applied to future (actual and merely possible) experiences, since it contains, as Lewis puts it, an implicit prediction regarding the future course of experience:
The predication of a property on the basis of momentarily presented experience is in the nature of an hypothesis, which predicts something definitely specifiable in further possible experience.
(Lewis 1929, 131)
But this feature of concepts can equally well be described in terms of Watkins’s notion of a conceptual function: thus, for Lewis, too, an empirical concept is a function that maps a certain class of sensations onto the class of judgments in which it is correctly applied. As a result, the feature that, according to Watkins, accounts for the fact that sensations play a role in the justification of empirical judgments is equally present in Lewis’s account: “the input that is given to us [i.e., sensations] and the function that takes that input [i.e., the concept that is applied] jointly determine the output” (Watkins 2008, 521) - that is, they jointly determine the correctness or otherwise of the resulting judgment. This suggests that the objection I am about to raise for Watkins’s position will apply to Lewis as well.
