Wednesday, 5 August 2015

The final section of Avner Baz' 'On when words are called for'

I have been reading Avner Baz’s paper, slightly confusingly called ‘On when words are called for’. It is an interesting Travis-like attack on McDowell's representationalist account of experience. But I am not sure I follow the argument because it doesn't seem as forceful as I expected it would.

The final section requires the defence, in an earlier section, of Stanley Cavell’s Wittgensteinian thought that the meaning of utterances cannot be read off the words alone but turns on their point or interest. Their point is not accidental to meaning but partly constitutes it. The final section applies this to McDowell’s account of experience in Mind and World which, like Sellars’ Myth of Jones, suggests that the content of experiences can be modelled on that of judgements.

Baz reports that:

We are invited by McDowell to think of the content of our experience as the very same content that judgments or claims might have… I think it would be more precise, and more to McDowell’s point, to say that to judge is to make up our mind what we are going to call something, what we are going to count something as (an object in front of us as a red cube; a surface as smooth; a government as stable). [485]

The first objection now runs as follows:

Already at this point we might begin to feel uncomfortable with the idea that ‘a judgment of experience simply endorses the conceptual content, or some of it, that is already possessed by the experience on which it is grounded’ (MW, p. 49, my emphases). For it is not clear what room this formulation leaves for faithfulness or unfaithfulness to our experience – and I mean this not primarily in the sense of telling or hiding the truth about it, or being right or wrong about it, but rather in the sense of being true to it, or failing to. [486]

I don’t follow this. I’m not sure I understand the difference between, on the one hand, being right or wrong about an experience and, on the other, being true to it or not. If that difference is vital then it undermines what I initially thought that this was: a worry that judgement can fail to fit experience so a model in which the former inherits the latter’s content cannot be correct because it cannot account for such failure. Had that been the worry then there are responses in the McDowellian account. To mention one: although a judgement may inherit the very same content as an experience it may also have a different content made as a recognition grounded in the experience. (Particular kinds of birds for example.) Such recognition could misfire.

But this seems not to matter as Baz warns us that this isn’t the way he will proceed.

If we press this worry too soon, however, we run the risk of remaining within the grip of the picture that underlies such formulations. And then we might end up still thinking (that we know what ‘judgment’ means, what judgment is, what faithfulness to the world requires) instead of looking (at what goes into our use of the word ‘judgment’, at what we, in actual cases, call ‘judgment’, at what faithfulness to the world requires). [486]

So what follows will respect the Wittgensteinian injunction to look rather than to think and hence to look at how ascriptions of what is seen are made.

The next objection presses the point set out in the previous section from Stanley Cavell that the meaning of utterances is partly determined by the point of making them, by the particular cares or concerns they express. This is then – I think – directed at the idea that a judgement might share the content of an experience. The question is: what is the content of the judgement?

[I]n judgments… what we decide to call the thing, and what we mean in calling it this or that, is inseparable from the nature of our interest in that thing – from that which leads us to ask ourselves what we want to count it as. Whether or not we decide, for example, to call a surface ‘smooth’ would depend on the nature of our specific interest in that surface. If you came to me and asked me, out of the blue, whether my desk, or face, was smooth, I would not know what to say. I would need for you to tell me what you meant; unless I could see, or was somehow able to guess, why you were asking me that – what the nature of the interest you were expressing with your words was. Similarly, if you asked me, under whatever circumstances, whether there was, whether I would say (judge) that there was, a red cube in front of me, I would not know what you were asking, unless I were able to see the point of your question. [486]

This passage suggests that the questions that might evoke the judgement that the table is smooth or that there is a red cube in front of one are varied depending on the matter of interest and, hence, that the content of the judgement to be made in response varies. In paradigmatic cases such as Sellars’ example of the red cube, it may seem that there isn’t this variability.

[I]t is extremely tempting to think that we can and do know what is being asked by ‘Is there a red cube in front of you?’, even in cases in which we cannot see the point of any such question. [486]

But even the meaning of utterance ‘There is a red cube in front of me’ is underdetermined by the words alone. To show this, Baz constructs a Travis-like example.

For suppose you say ‘There is a red cube in front of me’; I look, and see what for all the world is a green cube. Do I know what you said? Have you, for example, said something false? Will forming for myself an image of a red cube, or staring hard at the cube in front of you, or both, help me find out? The next thing you do is to scratch the green paint to show me that underneath the green paint there is red paint. Do I now understand what you said? Has it turned out that what you originally said was true? But then, shouldn’t you simply have said that underneath the green paint there was red paint? I’m still puzzled: What judgment (or claim), if any, did you express with your words? But suppose you now add: ‘Those idiots thought they could fool me and get away with it. Everybody knows that their green cubes are far superior to their red cubes. But I could tell (judge) immediately, just by holding it in my hand, that that wasn’t a green cube’. Now I finally understand you and know what judgment or claim you made: you weren’t just trying to be clever; and it wasn’t really the color of the cube you were concerned about, but its quality. [487]

This example presses the general claim to follow:

A judgment (or claim), I argue, requires a particular context in which it is called for. Without such a context, it would not be a judgment (or claim), nor have any determinate content. [487]

The paper doesn’t pause here. But I take it that a consequence of this is that there is a tension between two thoughts: that the content of a judgement answers to a context of inquiry but also that the content of some judgements is supposed, by McDowell, to be inherited from experience. The passage continues, directly, though with a possible contrast between judgements and experiences.

Now what about an experience? Isn’t it clearly different in that respect? It would appear that unlike a judgment or claim an experience does not need to be called for; or anyway not in the same way. It just happens to us, as we saw McDowell saying, whether we want it to or not. Whereas a judgment requires a reason if it is to be a judgment, with an intelligible content, it would seem absurd to say that you need a reason in order for you to see this or that that lies in front of your eyes – wouldn’t it? [487]

So one possible line of argument would stress a prima facie difference between the properties of judgements and experiences in order to undermine the connection – via shared contents – that McDowell suggests. But it seems not to work like that. Baz approaches the nature of experiences by looking at how we ascribe seeing something (rather than judging something) to someone. In this case too, he also wants to suggest, against his interlocutor’s intuition, the ascription depends on a context of interests or concerns.

What my interlocutor really wants to say is that the fact that there is a red cube in front of me, and that I’m looking right at it, and perhaps also that I can tell colors and shapes, justifies him in saying that I’m seeing a red cube in front of me and ensures not only the sense, but also the truth, of his words. But is that true? Is this what we call ‘seeing that this or that is such and such’? Is this what we normally mean when we say of someone that he sees this or that (to be the case)? What does determine – how do we tell– what someone sees? [498]

Again, Baz deploys a Travisian example of the possibility of different meanings dependent on context.

Here is a tree and there is Jones walking in its direction. And let us even assume that the color of the tree is of some significance to Jones, or ought to be; let us assume that we told him to wait for us by the tree whose leaves have turned red. Does Jones see that the tree is green? It is easy enough to imagine a case in which it would make perfect sense for me to say that he does. However, if I choose to say of Jones that he sees that the tree is green, my reason for saying the words would determine their – that is, my – exact meaning. [489]

That last phrase is key. Just as in the case of a judgement, so in the case of the ascription of seeing something, the claim, the ascription, about what is seen depends on context. The same Cavellian message is pressed.

I’ve been urging, following Cavell, that crucial to the conditions of someone’s saying that Jones sees this or that particular fact is that there be reason for that person to say that Jones sees this or that particular fact. [490]


It might be said, ‘But aren’t you confusing the issue of your reason for saying that Jones sees this or that with the question of whether or not he sees it? Granted that you need to make some specific point if you are to say something intelligible about what Jones sees; but this has nothing to do with his seeing what he sees…But we already saw that what you’ve just said of Jones (‘He sees that the tree is green’) can have more than one sense, depending on the specific point of saying it, and is not even obviously a ‘saying that he sees that the tree is green’. [490]


And different questions would make for different contexts and therefore for different things for him to be able to say on the basis of what faces him. [490]

For even in the case of the store manager facing the cube, or in the case of Jones facing the tree, different occasions for saying what they see at any given moment would make for different (correct) determinations of the content of what they then see.30 Even the perceiver himself can find himself offering different specifications of what he sees, or saw, given different possible occasions, or reasons, for him to attend to that matter. And so, given that most of what encounters us never actually gets articulated – even if it is something that we clearly do need in one way or another to attend to (say, where we step, or where we place something, or how someone has responded to something that we did or said); and given that different contexts of articulation will make for different specifications of the content of what we see, or saw;31 what might it mean to insist, as McDowell does, that what we take in in experience has the very same content that a judgment or claim might have? [492]

So I infer from this that the problem is not that seeing is a state with an interests-independent content whilst explicit judgement is interests-dependent and hence they cannot share the same content. Rather, the problem is that both are context-dependent in such a way that one cannot say, absolutely, what either content is.

If this is the right interpretation – and thus I think it cannot be – I’m not sure how decisive it is. Couldn’t one hold the McDowellian-Sellarsian-Myth-of-Jones idea of the judgement ‘containing’ the same content as the experience with both fixed by words used in a particular context? It is hard to assess such a thought because McDowell himself has moved on. But if one were sufficiently relaxed, could one not refuse to be embarrassed by the question of which concepts and simply grant that the content of the experience depends on what interests the subject? (My assumption is that the McDowell of ‘Avoiding the Myth of the Given’ could not take this line because that paper hints at a priori limits on the conceptual articulation of experience such that [bird] cannot be an element but [perching animal] might.)

There is a further and distinct objection.

[W]e may be tempted to pursue another understanding of the idea that our experience of the world is to be understood on the model of speech – the idea that our experience contains claims (‘contains’ ‘claims’)… Suppose we want to be able to say something about what must be happening in or to us when we see this or that. Suppose we want to insist, in response to philosophical pressures of the kind expressed in the previous paragraph, that what a competent speaker sees when she opens her eyes has the very same content as what she might say; that our experience presents us with content, in virtue of our being competent speakers whose senses are functioning, and that all that is then left for us to do is to endorse that very same content in judgment, or to express it in an utterance. Then, I’m afraid, we may find ourselves having to rely on words, those things ordinarily used for saying things, to be drawn into operation and capture – by themselves, as it were – the content of our experience, apart from some specific context in which there would be something in particular for them to mean. And then we will be proposing that what we experience is somehow already articulated (by us? for us?), and that all that is left for our judgments to do is simply to ‘select from among a rich supply of already conceptual content’ (MW, p. 49, fn. 6). [493]

But I don’t think that this would counter the contextualist response above.

PS: Avner Baz generously emailed me about this post thus.