Friday, December 24, 2010

Merry Christmas, or, Ryle's Idiotic Idea

I fondly remember last Christmas Eve, when Jason Stanley said Ryle's view of propositions was "idiotic." We were nearing the end of a brisk yet short-lived correspondence, the bulk of which spanned about 30 emails over the preceding 48 hours. I was home in bed, alone and barely mobile, recovering from a herniated disc in my lower back. My wife had taken the kids to her family's house, leaving me glued to my computer, surprised and inspired by Jason's interest in my ideas. My view was (and is) that Jason Stanley and Timothy Williamson (S&W) profoundly misinterpret Ryle in their oft-referenced 2001 paper, "Knowing How." I am not alone in thinking this. The same point is made in a number of published papers, though in a variety of different ways. Still, I couldn't convince Jason that S&W misinterpret Ryle, and he couldn't convince me that they don't.

At least we agreed on the distinction I had made between propositionalism and intellectualism.

Propositionalism is the view that all mental states (or, in a purely epistemological version, all varieties of knowledge) involve propositions. For example, if I know that snow is white, my knowledge involves the proposition that snow is white. If I think or believe that snow is white, I similarly have some attitude towards that proposition. Jason Stanley is a propositionalist, at least of the epistemological variety.

Intellectualism, as identified by Ryle in Chapter 2 of The Concept of Mind, is the view that all intelligent acts are consequences of intellectual acts; that behaviors which we characterize as intelligent are the result of antecedent acts of intellection; that to do something intelligently you must first think about what you are going to do. Jason Stanley is not an intellectualist.

Jason agreed with me that propositionalism does not entail intellectualism. Or, more accurately, what I believe is that only some varieties of propositionalism entail intellectualism, and that there could be varieties which do not. Jason said that Ryle's mistake was in thinking that propositionalism entailed intellectualism. I don't think that's accurate. Given the only view of propositions which Ryle found acceptable, propositionalism does entail intellectualism. Furthermore, I am not convinced that there is a coherent alternative to Ryle's view on the table.

Look at "Merry Christmas." When people say "Merry Christmas" to each other, are they stating a proposition? Most often, "Merry Christmas" is not a statement of fact. We might say it is short for "I wish you a Merry Christmas," which might look more like a fact. However, when I say "Merry Christmas" (in the right context), I am not reporting a sentiment I had previously made. I am rather just forming (or performing) the sentiment. My utterance does not correspond to some fact, and so could not be either true or false. It is not the case that all speech acts are propositional, in the sense that they all have contents which can be either true or false.

One variety of propositionalist--the intellectualist--might respond that the speech act does report a fact, that some inner thought process formulated the wish which was later expressed by the utterance. This will not do for Ryle, however, because the postulated inner formulation of the wish does not seem markedly different from the one we see and hear coming from a person's mouth. If some inner wish-making is required to make sense of the outward behavior, then why isn't some other inner wish-making required to formulate the inner wish, ad infinitum?

Another variety of propositionalist (Jason's variety) might agree that "Merry Christmas" does not report an inner wish, but simply performs the task of making a wish. This propositionalist will insist, however, that the making of the wish is itself a propositional act; that it entails or manifests a relation between a person and a proposition via some propositional state. Unfortunately, I don't see any good way of making sense of that. It is certainly untenable with Ryle's view of propositions, and it is not clear how an alternative view of propositions could work here.

Before I explain this, I have to come clean about an unfortunate error I made during my exchange with Jason Stanley. I suggested that Ryle's regress argument against intellectualism cannot be framed in terms of knowing-how and knowing-that. Jason wouldn't stand for that, and rightly so. Ryle's regress argument can be formulated in those terms, and Ryle does suggest such a formulation, but not in the way S&W claim.

S&W say Ryle adopts the following two premises:

S&W-P1: If one Fs, one employs knowledge-how to F.
S&W-P2: If one employs knowledge that p, one contemplates the proposition that p.

I don’t think Ryle adopts either of these premises. On the one hand, he defines knowing-how in terms of intelligent behavior; on the other hand, he does not claim that every employment of knowing-that entails an act of contemplation. Rather, examples of knowing-that just entail the acknowledgment or statement of a fact. So a more accurate representation of his regress argument might look like this:

P1: If one Fs intelligently, one employs knowledge how to F.
P2: If one employs knowledge-that p, one states or acknowledges the fact that p.

He then observes that, if knowing-how is reducible to knowing-that, then,

R: If one employs knowledge how to F, one employs knowledge that r is a rule for F-ing.

It follows that, if knowing-how is reducible to knowing-that, then one cannot act intelligently without stating or acknowledging a fact about a rule for that behavior. The problem is that facts can be stated or acknowledged correctly or incorrectly, appropriately or inappropriately. The fact must be intelligently acknowledged or stated, which increases the number of intelligent acts by one. This second intelligent act would also have to be guided by another one, ad infinitum.

The propositionalist might try to avoid the regress by claiming that some intelligent actions just are instances of stating or acknowledging a fact about a rule for that action. The propositionalist may thus reject the intellectualist's claim that the relevant employment of knowing-that is antecedent to the behavior in question. This is Greg Sax's approach in "Having Know-How" (forthcoming). Greg interprets Ryle's knowing-how/knowing-that distinction as a distinction between implicit and explicit propositional knowledge, so that the intelligent behavior itself is an implicit expression of propositional knowledge. Greg's conclusion is that Ryle's argument against intellectualism is consistent with S&W's propositionalism (though Greg does not frame it in these terms).

Greg's interpretation is not quite consistent with Ryle, however. Ryle acknowledges that the rules explicit in our exhibitions of knowing-that are implicit in our demonstrations of knowing-how, that when we act intelligently we apply criteria. However, this does not mean knowing-how is an implicit (or "practical," in S&W's terminology) form of propositional knowledge. At least, it is not clear how it could mean that. While rules might be implicit in our intelligent behavior, we can distinguish those rules from propositions which enjoin them.

Intelligent acts do not seem like implicit acknowledgments or statements of facts about rules for those acts. A musical improvisation, for example, does not seem to be a statement or acknowledgment of a fact about a rule (or rules) for itself. One's know-how, as demonstrated through a musical improvisation, does not seem to consist in knowledge that some particular rule is a rule for that performance--though an intelligent performance can be used as a rule for future performances. So I am not inclined to agree with this response to Ryle's regress argument.

A second strategy is to deny P2 and claim that employments of knowing-that do not always entail the statement or acknowledgment of a fact. I just don't know what could count as an expression of factual knowledge, other than the statement or acknowledgment of a fact. Ryle's characterization of knowing-that is intuitively appealing, and I am not aware of any compelling alternatives.

What is at stake here is just the notion of proposition. S&W are in the Fregean-Russellian tradition, which is marked with deep conceptual difficulties. (Here is a very good, recent paper by Stewart Candlish and Nic Damnjanovic on the topic.) Ryle contested this approach in 1931, arguing that "there are not substantial propositions," but only facts and symbols which are used to make statements of fact; and that the word "proposition" denotes the same as the words "sentence" and "statement," or "might be extended to cover all other symbols which do or might function as symbolic presentatives of facts." (See Ryle, "Are There Propositions?", in Collected Papers Volume 2: Collected Essays 1929-1968, p. 39) For Ryle, any exhibition of propositional knowledge entails some symbolic presentation of a fact. On this view, propositionalism does entail intellectualism. Perhaps some other view of propositions can save propositionalism from the intellectualist's fate, but I do not know how.

I don't think S&W fully appreciate Ryle's view of propositions and knowing-that, and this is part of the reason they misinterpret his argument against intellectualism. Ryle regards knowing-that in terms of abilities, specifically competences related to "the jobs of didactic discourse" (The Concept Of Mind, 1949, Chapter 9). He does not regard it in terms of a relation between a person and a Russellian or Fregean proposition. Ryle does not accept the Fregean-Russellian conception of propositions. Yet, in their paper, S&W put a traditional, Russellian view of propositions in Ryle's mouth, and say that Ryle does not regard knowing-that as an ability or anything similar. Thus, as I wrote to Jason last Christmas Eve, I think S&W are talking past Ryle. That's when Jason said he thinks Ryle "carved out an idiotic notion."

See also: