Monday, 20 July 2015
Michael Devitt on Analyticity & Conceptual Containment
Michael Devitt tackles the frequently-used
“All bachelors and unmarried men.”
example as an ostensible case of a priori knowledge. This time, however, instead of talking about ‘synonyms’, ‘meanings’ or 'mutual definitions’, he talks in terms of concepts. In fact his statement of the case has a Kantian ring to it. He says that “the content of the concept [bachelor] ‘includes’ that of the concept [unmarried]”. Kant, though, talked of conceptual ‘containment’ rather than inclusion. And, as Quine told us about the Kantian metaphor ‘contained’, the word ‘includes’ is also difficult.
How does one concept ([bachelor]) contain or include another ([unmarried])?Anyway, because these concepts are seen this way, the above is taken to be an analytic statement or taken as being known to be true a priori.
Devitt puts his own slant on this well-known example of an analytic statement. He claims that others believed that
“simply in virtue of having a concept, a person was in possession of a 'tacit theory' about the concept; in virtue of having [bachelor], a person tacitly knew that its content included that of [unmarried]”.
We can say that if a person were asked for a definition of the word ‘bachelor’, he would say that “a bachelor is an unmarried man”. However, because his theory is tacit, he doesn't vocalise (or even sub-vocalise) that definition every time he uses the word ‘bachelor’ – perhaps he's never done so. Though when asked, he could indeed do so. We can assume that simply having heard - and then known - the definition, then that's enough for such tacit knowledge. Of course because it's tacit, he almost certainly wouldn’t talk about “conceptual containment” or one concept being contained within another. Perhaps his only technical knowledge would be definitional.
The former a priori explication of these concepts and the a priori statement above are seen as Cartesian by Devitt. We have a “privileged 'Cartesian' access to the facts about [such] concepts”. Presumably this must mean that we need no experience of concept-use and we certainly don’t need to check a dictionary. The only thing that's required is
“a reflective process of inspecting the contents of concepts to yield knowledge of the relations between them which in turn yield such knowledge as that all bachelors are unmarried”.
This is odd to those of us who see concepts as themselves (often) being the contents of statements or mental states. Concepts are often seen as being pretty primitive and therefore suitable candidates for the contents of various things. On this picture, concepts themselves have content. Thus:
The content of the concept [bachelor] is the concept [unmarried man].
Does that mean that a concept, qua content, contains another concept? Thus we would have:
[ bachelor [unmarried man]] or [unmarried man [bachelor]]
All this also went under the name of “conceptual analysis” and was similarly seen as an a priori process. Indeed at one point, because of the importance of conceptual analysis in philosophy (at least at one time), it was often stressed that philosophy itself was an essentially a priori business. Either that, or it was a way of stressing the already-taken-for-granted view that philosophy is an a priori business. (Some philosophers still believe that conceptual analysis is the primary role of a philosopher. Thus that philosophers are basically apriorists.)
Devitt also brings in the epistemic notion of justification. This seems a little out of place, prima facie, when it comes to the statement “All bachelors are unmarried men”. However, we're talking about a priori justification here. He asks us if we can justify the “proposition that all unmarrieds are unmarried were justified” (6). Well, of course it is – isn’t it? What we have here is the Quinian reduction of the analytic truth
“All bachelors are unmarried men.”
into the logical truth
“All unmarried men are unmarried.”
by the “substitution of synonym for synonym”. Again it seems, prima facie, like an odd question to ask “where does the justification for this proposition come from?” (6). Of course it's justified… Well, if it isn’t exactly justified, that’s because it doesn’t need to be because it's a logical truth. However, the apriorist has previously said that the statement “All bachelors are unmarried men” is known to be true a priori and therefore it's also justified a priori. Now it seems it doesn’t require epistemic justification because it's a disguised logical truth.
The obvious question now follows (asked earlier in Devitt’s paper): What justifies logical truths? This question wasn’t asked much until fairly recently in either philosophy or logic. Though if the analytic truth is a disguised logical truth (even though it was classed as a priori justified), then “we have not described a nonempirical way of knowing” (6) precisely because the apriorist has shifted (if he has shifted) from an analytic statement justified to be true a priori to a logical truth which hasn't received such a justification.
A question Devitt doesn't ask here is: Can logical truths be justified?
Devitt, Michael, 'There is no a Priori' (2005)