Bartels and May’s paper presents the outlines of a theory of practical knowledge. The paper consists of a discussion of intellectualist and anti-intellectualist approaches to knowledge-how, a characterization of a range of behavioral particularities of practical knowledge, and the outlines of a theory that attempts to explain these behavioral particularities in terms of involved underlying mental representations. The discussion is remarkably clear, and the explicit exposition of what is to be explained by a theory of practical knowledge is a great virtue of the paper. For our purposes here, a discussion of the initial characterization of practical knowledge and its attempted explanation in terms of conceptual and non-conceptual capacities would help us assess the import of this paper. To my valuation, however, the discussion also reveals some very important features of the relation between knowledge ascriptions (and, to that effect, ascriptions of propositional attitudes in general) and descriptions of underlying cognitive structures and representations. Most importantly, Bartels and May employ Stanley’s semantic reading of propositionality, according to which the propositionality of some mental state depends on whether a proposition is mentioned in the ascription of that state. As a result, questions concerning cognitive structure and underlying representations are largely detached from considerations concerning ascriptions of propositional attitudes. I think this is a great advantage, because we are not led to read back the relational grammatical structure of ascriptions of propositional attitudes onto psychological states themselves. Here I want to focus on this semantic reading of propositionality and ask about its effects on the relation between Bartels and May’s proposed explanation of practical knowledge and Stanley’s theory of knowledge-how. The result will be that Stanley and Bartels and May attempt to explain quite different things. While Stanley proposes a theory of how we ascribe knowledge-how to each other, Bartels and May are interested in underlying cognitive processes. The semantic reading of propositionality, however, only goes halfway towards disentangling these different endeavors. A further step can be made with the help of the idea that ascriptions of propositional attitudes are (like) measurements. I will call this the measurement view of ascriptions of propositional attitudes. Considerations from measurement theory can then be used to shed further light on the relation between ascriptions of propositional attitudes and the underlying cognitive representations. The result will be that nothing can be inferred about cognitive structure from the structure of ascriptions of propositional attitudes alone. Propositions need not play any role in a theory of cognition. Nonetheless, there is a clear sense in which propositional attitudes are real. They are the measurement-theoretic representatives of behaviorally relevant states. In closing I will note that, given the close connection between concepts and propositions, a semantic reading of conceptuality might be desirable. For Bartels and May, this would mean that the difference between practical and theoretical knowledge should not depend on the conceptuality of the underlying representations. But given their definition of conceptuality, this would merely require a change in nomenclature.
Before going into the discussion of a semantic reading of propositionality, of measurement and its bearing on the relation between Bartels and May’s proposed explanation of practical knowledge and Stanley’s theory of knowledge-how, I will briefly summarize Bartels and May’s line of argument.