Kuehn on Methodological Individualism
When Daniel Kuehn is not cravenly defending Obama, he's a pretty smart guy. (I'm just kidding, of course, Daniel: you are a pretty smart guy even when you are cravenly defending Obama.)
The main way to convince yourself to be (or remain) a methodological individualist is to arbitrarily and severely narrow the number of options as to how to do social science, and then show that MI is the only valid one. Here is a great example from the comment section of the post linked to above (from "increasingmu"), saying our choices are limited to:
a) methodological individualism
b) attributing of agency to non-individuals
c) arguing that it doesn't matter what is really true because our theories are just tool
Wow, why in the world are those are only options? Let's say I wish to study kinship systems. Yes, of course this study will involve individuals -- kinship systems do not simply float around in space. But they certainly aren't a matter of individual choice: I am born into a kinship system I in no sense chose, and, in general, can exit it only by exiting the society in which it exists. Furthermore, the very idea of a kinship system necessarily involves a super-individual structure. (As does, say, the notion of a language!) I think it is completely sensible to hold:
a) kinship systems are super-individual causal factors in social analysis (contrary to choice a) above);
b) kinship systems are not agents; they don't make choices or have preferences (contrary to choice b) above); and
c) kinship systems, like, say, the English language, are real causal factors -- we don't make them up as social scientists just because we find them handy (contrary to choice c) above).
In fact, my a) through c) are simple common sense; so commonsensical, in fact, that it takes years of advanced education to learn to ignore common sense and become a methodological individualist.
The main way to convince yourself to be (or remain) a methodological individualist is to arbitrarily and severely narrow the number of options as to how to do social science, and then show that MI is the only valid one. Here is a great example from the comment section of the post linked to above (from "increasingmu"), saying our choices are limited to:
a) methodological individualism
b) attributing of agency to non-individuals
c) arguing that it doesn't matter what is really true because our theories are just tool
Wow, why in the world are those are only options? Let's say I wish to study kinship systems. Yes, of course this study will involve individuals -- kinship systems do not simply float around in space. But they certainly aren't a matter of individual choice: I am born into a kinship system I in no sense chose, and, in general, can exit it only by exiting the society in which it exists. Furthermore, the very idea of a kinship system necessarily involves a super-individual structure. (As does, say, the notion of a language!) I think it is completely sensible to hold:
a) kinship systems are super-individual causal factors in social analysis (contrary to choice a) above);
b) kinship systems are not agents; they don't make choices or have preferences (contrary to choice b) above); and
c) kinship systems, like, say, the English language, are real causal factors -- we don't make them up as social scientists just because we find them handy (contrary to choice c) above).
In fact, my a) through c) are simple common sense; so commonsensical, in fact, that it takes years of advanced education to learn to ignore common sense and become a methodological individualist.
You can reduce the concept of a kinship system to the actions of individual human beings. The issue at hand is when aggregate demand is used to justify things that don't map pack to individual behavior (e.g. doesn't make use of stick wages). Or, the "standard social science model" used in outdated anthropology, psychology, and sociology. You can make reference to language and most other things in terms of individuals. But there are some concepts that don't work in the same way, which is why I have no problem with a broadly-defined version of methodological individualism.
ReplyDelete"You can reduce the concept of a kinship system to the actions of individual human beings."
ReplyDeleteYes, you can. So what? You can reduce the behavior of a storm to that of individual atoms. That does not justify "methodological atomism." Storms are good things to study in and of themselves, and we learn something looking at that level we do not learn at the atomic level. And so it is with kinship systems, and language.
The point is that you *can't* think of the two examples in terms of human agency. This is why anthropology, which is supposed to be the study of man, became the study of culture. Groups of human beings are vastly different, but no one individual could affect those differences. Therefore, it was believed, that culture is everything and exists separate from any individual and causes individuals to behave the way they do. "Culture" was the explanation of everything and it existed out in the ether, never to be disputed. You can get to the same empirical facts using a methodologically individualist outlook ("the result of human action but not of human design"), but the Standard Social Science model would lead one astray.
ReplyDeleteWell, increasing, in that anyone ever believed anything so blitheringly stupid as a culture that existed without individuals, MI certainly is an attractive alternative! But I've read plenty of prominent social scientists who reject MI, and not one of the ones I have read believes anything vaguely like what you describe.
ReplyDeleteIn fact, increasing, you are doing exactly what I described in the post: making MI seem attractive by limiting the alternatives to the completely stupid!
ReplyDeleteIn the classic debate (or debates) of the 1960s involving J. W. N. Watkins, Maurice Mandelbaum, and Ernest Gellner ("Sir Ernest," as he later was), one has to conclude that Watkins, the paladin of methodological individualism was not the winner. I suspected this when I read the essays in grad school in the 1970s, and on re-reading them lately I am certain of it.
ReplyDeleteIn addition, it may not be an accident that much of Watkins' work was on Hobbes. Now there's an individualist (and atomist)for you!
On Hobbes: "Now there's an individualist (and atomist)for you!"
ReplyDeleteGranted, I am not the most well-read or intelligent fellow around these parts, but from what I know, I would assume that what I quote above is sarcastically stated. Is this correct, or am I shooting in the dark?
No, Joseph, Hobbes could be considered the creator of atomic individualism.
ReplyDeleteGene is right, Ryan. Sure people are wrong if they think "culture" exists independently. But that's not the claim. The claim is never that storms exist as entities independent of the action of atoms - it's that we can study storms independently without assigning agency to them. The claim is never that aggregate demand exists independently of human action - it's that we can study AD independently without assigning agency to it.
ReplyDeleteMethodological individualism expects us to build up aggregate phenomena associated with non-agents from the action of agents. We know enough about complex systems now to know how dangerous that is.
Thanks, Gene. I get it now. The word "atomistic" was used by the other side of the argument to describe Hobbes' particular individualist philosophy. I've read Locke, but I am definitely lacking in Hobbes' territory (I read Leviathan years ago). I certainly am not read up on the opposing views of the time.
ReplyDeleteI guess that I have a few more books to buy.. Damn you Gene Callahan for making me think about stuff!
Yes, Joseph, you have consider Hobbes's "state of nature" -- each individual is radically on his own, in the "warre of all against all." The isolation of individuals is so extreme that the leviathan is the only solution!
ReplyDelete