Stanley's Great Error

I'm not done reading and responding to Stanley's chapter on Ryle, but I've found another big mistake in his interpretation which deserves a post of its own.  Stanley says that Ryle claims the following:  If intelligent action is guided by rules or if it involves the application of criteria, then those rules or criteria must be intellectually acknowledged prior to the intelligent performance.  Stanley couldn't be more wrong.  [Stanley's error leads him to mischaracterize Ryle's knowing-how/knowing-that distinction. See update below.]

Stanley rejects the position which he attributes to Ryle.  He says Ryle's view is unjustified, because Ryle has not demonstrated that the mere application of criteria requires an antecedent acknowledgment of that criteria.  Stanley says (Know How, p. 13):  "Ryle therefore needs an argument that the view that 'intelligent performance involves the observance of rules, or the application of criteria' entails that 'the operation which is characterised as intelligent must be preceded by an intellectual acknowledgment of these rules or criteria; that is, the agent must first go through the internal process of avowing to himself certain propositions about what is to be done ('maxims', 'imperatives' or 'regulative propositions' as they are sometimes called); only then can he execute his performance in accordance with those dictates.'"

In that passage, Stanley is quoting Ryle (The Concept of Mind, page 29 in my edition.) Here is the passage from Ryle in full (it starts on the preceding page):

What is involved in our descriptions of people as knowing how to make and appreciate jokes, to talk grammatically, to play chess, to fish, or to argue? Part of what is meant is that, when they perform these operations, they tend to perform them well, i.e. Correctly or efficiently or successfully. Their performances come up to certain standards, or satisfy certain criteria. But this is not enough. The well-regulated clock keeps good time and the well-drilled circus seal performs its tricks flawlessly, yet we do not call them 'intelligent'. We reserve this title for the persons responsible for their performances. To be intelligent is not merely to satisfy criteria, but to applv them; to regulate one's actions and not merely to be well-regulated. A person's performance is described as careful or skillful, if in his operations he is ready to detect and correct lapses, to repeat and improve upon successes, to profit from the examples of others and so forth. He applies criteria in performing critically, that is, in trying to get things right.

This point is commonly expressed in the vernacular by saying that an action exhibits intelligence, if, and only if, .the agent is thinking what he is doing while he is doing it,and thinking what he is doing in such a manner that he would not do the action so well if he were not thinking what he is doing. This popular idiom is sometimes appealed to as evidence in favour of the intellectualist legend. Champions of this legend are apt to try to reassimilate knowing how to knowing that by arguing that intelligent performance involves the observance of rules, or the application of criteria. It follows that the operation which is characterised as intelligent must be preceded by an intellectual acknowledgment of these rules or criteria; that is, the agent must first go through the internal process of avowing to himself certain propositions about what is to be done ('maxims', 'imperatives' or 'regulative propositions' as they are sometimes called); only then can he execute his performance in accordance with those dictates. He must preach to himself before he can practise.

I discussed this passage with Stanley a couple years ago, and I challenged his interpretation of it. I guess I wasn't able to convince him of his error, because he repeats it again in his book.

To see the error, look at the first paragraph of the passage I just quoted from Ryle.  He says, and I quote again, "To be intelligent is not merely to satisfy criteria, but to applv them; to regulate one's actions and not merely to be well-regulated."  Ryle states, clearly and straightforwardly, that intelligent action entails the application of criteria, and the regulation of one's actions in accordance with criteria.  So Ryle accepts the first of the claims that Stanley quotes:  "intelligent performance involves the observance of rules, or the application of criteria."  On the other hand, Ryle denies that intelligent action must be preceded by an avowal or acknowledgment of a proposition or truth.  So Ryle denies the second claim that Stanley quotes, that "the operation which is characterised as intelligent must be preceded by an intellectual acknowledgment of those rules or criteria."  How could Ryle suppose that the first claim entails the second, when he clearly affirms the first but denies the second?

Why does Stanley accuse Ryle of supposing that the first claim entails the second?  Because in the second paragraph I quoted, Ryle presents an argument to that effect.  However, Ryle is showing us the sort of argument he thinks an intellectualist is "apt" to make. Stanley confuses this hypothetical intellectualist's argument for Ryle's own, and so misunderstands Ryle's point of view entirely.

Ryle states his view again on the next page of The Concept of Mind:  "Some intelligent performances are not controlled by any anterior acknowledgments of the principles applied in them."  Ryle is clear:  intelligent performance is guided by rules, but it is not always controlled by "anterior acknowledgments" of those rules.  It still involves the application of criteria. Ryle thus distinguishes between following a rule and being guided by a proposition.  Stanley's interpretation is simply wrong.

Based on his (mis)reading of that one passage, Stanley believes that Ryle thinks knowing-that entails the application of criteria or the following of rules, and that this requires some kind of antecedent triggering or accessing of propositions.  Thus, he takes Ryle to suppose that manifestations of knowing-that (unlike manifestations of knowing-how) require prior acts of acknowledging or avowing propositions.  He writes (p. 17):  "[Ryle] assumes that manifesting propositional knowledge requires a prior mental act, such as the prior triggering of a maxim or rule . . . Second, he assumes that knowing how in contrast can be manifested without there being any prior mental act whatever. . . . Ryle is operating with a metaphysical picture of knowing how according to which one's know how just is constituted by the fact that when one is so situated, one acts thus."  Thus, Stanley says, Ryle believes that "my state of knowing how to open a door is manifested simply by my being in front of a door and opening it."  Stanley concludes that Ryle's distinction between knowing-how and knowing-that is unmotivated, because intelligently opening a door is properly taken as an employment of knowing-that, or propositional knowledge:  It shows that one knows that such-and-such is a proper way to open the door.

The first major problem with Stanley's interpretation is that Ryle does not suppose that knowing-that entails antecedent acts of intellection.  In fact, he says knowing-that does not always require intellection at all, as it can be demonstrated by rote recitation of rules.  (See Ryle's discussion of the difference between knowing how to play chess and knowing the rules of chess in Chapter 2 of The Concept of Mind, pp. 40-41.) Furthermore, in Ryle's view, knowing-how cannot be evidenced simply by performing an action in a particular situation.  According to Ryle, knowing-how cannot be identified or associated with any particular action type.  It is not a matter of habit, and so cannot be demonstrated by merely doing a particular thing in a particular circumstance.

In Ryle's view, intelligently opening a door can involve propositional knowledge, but it cannot be reduced entirely to propositional competence.  However, merely opening a door out of habit, without the care and innovation that marks intelligent behavior, is not an example of know-how.  It could, however, involve (or perhaps share some interesting features with) propositional knowledge.