By Jim Virtel
The latest issue of BBS includes a précis of Edouard Machery’s Doing Without Concepts
—the book that boldly argues that the term “concept” should be eliminated from psychology. The fourth tenet of Machery’s Heterogeneity Hypothesis (HH) proposes that prototypes, exemplars, and theories
—three types of concept
—are used in distinct cognitive processes. Gualtiero Piccinini and I wrote a short response arguing that Machery has not provided enough evidence that prototypes and exemplars are used in distinct cognitive processes. If we are right, then Machery’s argument for concept eliminativism as he presents it doesn’t go through.
Interestingly, Machery chooses to respond to us by weakening his argument. Machery maintains that even if exemplars and prototypes are used in the same cognitive processes, theories are used in different cognitive processes from exemplars and prototypes. Thus, Machery concludes, “there are no generalizations about how concepts are used in cognitive processes” (2010, 237).
But Machery’s response is less satisfying than it may appear. Our commentary focused on prototypes and exemplars because that’s where Machery made his strongest case for differences in the cognitive processes. Machery offers little, if any, support for the claim that theories are used in distinct cognitive processes from prototypes and exemplars. In fact, everything Machery says is consistent with the hypothesis that theories are prototypes plus some causal information of a category (incidentally, many psychologists also believe that theories are enriched versions of prototypes). If theories are just augmented prototypes (i.e., a kind of prototype), then a fortiori they are used in the same cognitive processes. The burden is on Machery to provide evidence that theories are truly distinct from prototypes and are used in distinct cognitive processes.
Unless, of course, theories are seen as what Piccinini (forthcoming) calls “linguistic concepts”.
I really enjoyed the commentary you wrote together with Gualtiero. Overall, in fact, I enjoyed the experience of writing the precis, reading the 27 commentaries, and developing a reply. I think the outcome is a worthwhile read!
Concerning your post, I do not so much weaken my position than make the point that even if it is true that exemplars and prototypes are not used in two distinct kinds of processes, theories still would. And that would be sufficient to support one of the key premises of the eliminativist argument.
But is there good evidence that prototypes and exemplars are used in distinct types of processes? If there is one thing I have learned from your commentary and from a few others (including Glymour’s review in Mind) is that I have underestimated how difficult it is to provide evidence for the nature of cognitive processes. In a sense, it is curious that I should have forgotten this given the importance in my own thoughts of the controversy between Anderson and Pylyshyn in the 1970s and 1980s.
In any case, I still think that overall the evidence supports the view that exemplars and prototypes are used in two distinct kinds of processes, involving different types.
But rather than discussing this question, I would to like to answer briefly the challenge made in the post. What’s the evidence for the claim that theories are used in a kind of process distinct from the types of processes defined over exemplars and prototypes?
I think that the answer is already in DwC. Research on categorization and on induction inspired by the theory theory view of concepts suggests that the processes involving them do not involve matching and similarity-computation. The processes defined over exemplars and prototypes do. I take that this is good evidence that the relevant processes form two distinct types.
As you know I’m a fan of your work.
But I think it’s fair to point out (as Jim does) that in your reply you have weakened your fourth tenet (from three kinds of concepts being used in different processes to two).
As to the evidence, as far as I remember the evidence you mention shows that when causal information about a category is used, categorization and induction judgments may be different than when it is not used. This is consistent with maintaining that theories are prototypes plus causal information, used in the same processes that prototypes are used in.
As you know, I don’t even think that there is a radical difference between (implicit) theories and prototypes, so a fortiori I don’t believe that theories and prototypes are used in different cognitive processes. But I do believe that explicit concepts are a different matter, and in many cases are used in different cognitive processes. So some of what you might take as evidence that theories are used in different cognitive processes I would reinterpret as evidence that explicit concepts are used in different processes.
If theories are just augmented prototypes (i.e., a kind of prototype), then a fortiori they are used in the same cognitive processes.
I don’t see how that follows. If a unicorn is just an augmented horse, it doesn’t follow that unicorns _are_ used, or even can be used (might not be enough vertical clearance for the horn!) everywhere horses are used.
I would probably go (a lot) farther than Machery, but I’ll certainly agree that “concept” is a vague and overbroad … concept.
I do not understand how asserting the conditional “If prototypes and exemplars turn out to be used in the same processes, it is still the case that theories don’t” amounts to weakening my position.
Anyhow, there is a bit more to the evidence than what you suggest! To repeat, the evidence is that similarity does not matter when theories are used, it does when prototypes are used. This is why it is just not the case that prototypes and theories are used in the processes.
I will try to put the response to the Dialogue symposium, in which your article was published today.
It follows because the intended point is that under the hypothesis being presented, prototypes and theories are not two different kinds of concepts but two different accounts of (the same kind of) concepts. BTW, that’s how many psychologists think of them.
Whether similarity matters depends on how you measure similarity. If you include causal information in your measure of similarity, perhaps you can do everything with a similarity computation. But in any case, my main point is that if theories are not a separate kind of concept but just a more accurate description of the same kind of concept, then a fortiori they are used in the same processes.
As to weakening, I just meant it in the technical logical sense. If you change your fourth tenet from A&B to A, you “weaken” it (meaning, roughly, that it rules out fewer possibilities). When I said you weakened your argument, I just meant that you weakened one of its premises. It’s not a criticism; it’s a good and honest way to avoid an objection.
Just for the sake of clarity let me repeat that I assert a *conditional* “f prototypes and exemplars turn out to be used in the same processes, it is still the case that theories don’t*, which of course is not the same as asserting the antecedent of this conditional. In fact, I *deny* the antecedent of this conditional.
Re-similarity: I accept your conditional, but since the consequent seems to be false (for the reason already mentioned), the antecedent has to be false too.
An intended point is better if it has an argument supporting it.
Wow, there’s a novel observation! The burden of proof is on Edouard to show that theories are actually not an alternative (to prototypes) account of concepts, as most people take them to be (because that’s how they were introduced in the first place), but a separate kind of concept that coexists with prototypes.