Dewey to me one more time
Thu 01 Jun 2006 11:00 PM
Thanks to a kind invitation from Matt, I've been sitting in on a reading group here in San Diego. They had already read two-thirds of the way through Dewey's Logic when I joined them. Fortuitously, they had just reached chapters that speak to the question at issue between Matt and I in our discussion here.
In chapters XIII and XIV, Dewey distinguishes between generic and universal propositions. The former are about kinds and the latter are about categories.
Kinds, for Dewey, are aggregates of things that share certain qualities.
Categories, on the other hand, are kinds that we have picked out in the course of inquiry. Categories are the symbol with which we model the kind.
I do not think that Dewey would selectively privilege some aggregates as genuine kinds, because any arbitrary aggregate will share some properties. As he notes, "Everything in the world is like everything else in some respects, and is unlike anything else in other respects" [p. 268]. In terms of my previous example, LIZ and ARD are genuine kinds because there really are lizard back-ends and lizard fronts. Prior to my formulating them, they were not categories.
There are such an enormous number of kinds that we could not have symbols for all of them. We choose our categories in the course of inquiry to symbolize those kinds that we think are important.
So far this is like Kitcher's realism. Since the cartoon version of Dewey is a crude pragmatist anti-realist, however, one might worry that Dewey's view really can't be realism at all. This concern would be misplaced for two reasons:
First, for Dewey, our categories as free-floating concepts are responsible to the kinds they are meant to capture. As Dewey puts it: "Since existence is existence and facts about it are stubborn, ascertained facts serve to test the hypothesis employed" [p. 266].
Second, I am here interested only in the pluralist strand of Kitcher's realism. He and Dewey would disagree on the value of the analysis of truth as correspondence, but that alone does not make Dewey an anti-realist. (Especially since Philip is wrong about it.)
To move beyond the exegetical point: This string of posts has considered a maximally-promiscuous realism about kinds. While I only knew of one adherent to that position, I could abide by calling it Kitcher's realism. Now that I have a second, I need a general name for the view. I propose maximally pluralist realism, but I would appreciate something more euphonious. Any suggestions?
Comments
from: Ignotus
Sun 04 Jun 2006 12:56 AM
...maxiplurealism?
from: jay
Fri 09 Jun 2006 07:39 PM
PD,
First, my suggestion for a name is 'replete realism'.
Second, it seems to me that on Kitcher's view, the following must be true:
Some K is a natural kind just in case K is a set.
If you are a mathematical realist about sets, then you are about kinds. But, surely kinds are more than just sets. For example: my nose, Michael Keaton, and the Eiffel Tower surely are not members of a kind just because their are elements of the set {Jay's nose, Michael Keaton, Eiffel Tower}. Why is this? Each member of a kind must share kind-constituting properties and elements of sets need not.
So, LIZ and ARD are real kinds only if they as categories have kind-constituting properties which individuate them. I suspect that they do not. Hence, I would not allow them in my ontology.
cheers
jay
p.s. How about 'tropical rainforest realism'?
from: P.D.
Sun 11 Jun 2006 12:51 PM
Hello, Jay!
Jay writes: "So, LIZ and ARD are real kinds only if they as categories have kind-constituting properties which individuate them. I suspect that they do not."
Of course LIZ and ARD have properties which individuate them, as shown by the fact that I could give a concise definition of both terms. You might still say that these properties are not kind-constituting, but I do not see any reason to say that unless we have already decided that LIZ and ARD are not legitimate kinds.
Generally, any group of things is similar in indefinitely many respects. So any set will have properties that distinguish its members from things that are not its members. You might say that these properties are unnatural, disjunctive properties and not wholesome, kind-constituting properties. Again, this seems circular to me-- but that is not to say that it is inconsistent.
from: jay
Tue 13 Jun 2006 06:48 PM
PD,
I appreciate your thoughtful comments. Here are a few in return since I find a "pluralistic realism" attractive.
You write,
"Of course LIZ and ARD have properties which individuate them, as shown by the fact that I could give a concise definition of both terms."
a. Here I would just deny providing a "concise definition" is evidence of a property designated by such a definition. Example: 'grue' can be given a "concise definition" but i am suspicious that is designates a genuine property.
You write,
"You might say that these properties are unnatural, disjunctive properties and not wholesome, kind-constituting properties."
b. That is precisely right. Conjunctions of properties are properties but I resist the notion that negations or disjunctions of properties are properties. Hence, my view about 'grue'.
Moreover, I think that this view of properties is principled. Consider the set {Jay's nose, Michael Keaton, Eiffel Tower}, my suggestion is that this set is not a natural kind because (1) our notion of a property consists in each element sharing some feature in common, (2) that feature not being "a member of the set consisting of {Jay's nose, Michael Keaton, Eiffel Tower}" (i.e., circularity), and (3) given the relevant sense of 'natural', must be mind-independent (i.e., not the set of things Jay is thinking of at...).
c. Having said all this, I don't have any beef with LIZ and ARD. "My" pluralistic realism is not as pluralistic as Kitcher's or maybe yours.
Hopefully this helps clarify some of the cool issues.
cheers
jay