Some Remarks in Response to Bill's Handout
 
     First, re the question whether Katz is offering necessary and sufficient conditions.  We may find some indirect evidence that he takes his dictionary definitions to provide such conditions in the fact that he takes them to support analyticities (which he presents as truths in virtue of meaning.  See p. 144: "Analytic sentences are thus trivially true by virtue of the fact that the determination of the things to which their subject refers already guarantees that these things will have the attributes their predicate asserts of them."  And then: "Since semantic markers represent the conceptual components that form the meaning of a subject and a predicate, the primary condition for 'analyticity' can be stated … in terms of the formal condition that the semantic markers in the reading of the predicate also appear in the reading of the subject."  The key here is tying the semantic analysis to reference (and thus truth) determination.  We'll talk about that next time.

     The above quoted passage, by the way, also suggests the connection between Katz' proposed format for semantic analysis and a theory of (complex) concepts: if we think of concepts – as we've been invited to – as the meanings of lexical items, Katz' semantic markers give you the elements into which complex concepts decompose.  This, of course, leave us wanting a theory of non-complex concepts (I take that to be the 'bite' of Ted's question); and as far as I can see Katz has nothing to say about that.  Indeed, as Dan pointed out after class, anyone who believes there are complex concepts (i.e., everyone) can in principle sign on a Katz-style analysis.  Though there will be wide disagreement about which and how many concepts are complex.

     I want to pause to reflect on Bill's "quickest way to make making Katz' theory into a theory of concepts in a psychological sense".  Bill's way is, of course very natural for a proponent of a Representational theory of mind (what this theory amounts to exactly will emerge soon enough in the seminar, I'm sure).  It trades on the idea that concepts are mental particulars, by which is meant, I take it, constituents of certain mental (=brain) events – episodes of thinking, hoping, judging, etc.  This comes out in Bill's requirement that a concept in the psychological sense play "an active and constitutive role in ordinary thought about ordinary things".  Notice that this requirement doesn't sit very well with the idea that concepts can be thought of as word meanings.  Word meanings, as Bill points out, are not "active" in Bill's sense.  Even when taken as psychologically real, they are either items in a look-up table, or they represent speakers' behavioral and/or mental dispositions involving the relevant words, or some variant of these.  Suppose, as a representationalist is prone to do, we think of episodes of thinking etc. as a matter of manipulating mental representations – say, symbols in our very own machine language, the language of thought.  If concepts are these mental representations, then they are not meanings of (public) words; rather, they are things that have meanings (though these meanings, unlike the meanings of public words, are not acquired).

     Maybe, then, one could take the theory offered by Katz as a theory of (complex) concepts insofar as concepts are thought to be meanings of terms.  What would make it a psychological theory rather than an "abstract quasi-mathematical theory" would be the idea that the semantic analysis he supplies reflects the structure of elements in our cognitive repertoire, as evidenced by the semantic judgments that competent speakers either volunteer or that could be elicited from them in direct or indirect ways.  Bill's complaint about the "nearly constant unclarity" in Katz seems to rely on the idea that to offer a psychological theory of concepts one must take them to be playing an active role in cognitive processing.  But arguably that is a representationalist prejudice.  I'm harping on this not to pick on the representationalist theory of mind (so I don't think we're fighting yet, Bill…).  Rather, it's because I want to get to the bottom of the slippage (ambiguity?) that we've detected between, if you will, two different ways of taking the concept CONCEPT.  I'm starting to worry that the slippage may lead to one way of cross-purpose talk.

      By the way, the above is, I think, related to the problem I tried to raise in class about taking concepts to be both mental particulars and types.  But I won't go on now, not to add to what may be a growing confusion.