Anthony Waterman wrote: > What is 'methodological individualism'? As I understand > it, it is the working assumption that human social > phenomena may be explained without remainder as the > outcomes of action by individuals: and that any additional > explanans (e.g. 'collective' plans, intentions etc., 'laws > of history', 'general will' and so forth) are redundant. > There can be no 'proof' of the 'correctness' of this > working assumption. It is part of the 'hard core' of > economics' which we stick with so long as the models we > construct on that basis seem to work. Why are we > quarrelling about this? And how could it possibly need > a defence? It is hard to know where to begin with such a broad question. I'm somewhat inclined to a response as effortless as the question: http://scholar.google.com/scholar?q=%22methodological+individualism%22&ie=UTF-8&oe=UTF-8&h l=en&btnG=Search As you suggest in your post, methodological individualism is often more of an ideological commitment than an explanatory strategy. And it tends to remain correspondingly vague in its actual explanatory commitments, to the point that the profession has founds itself smiling upon the ad hoc device of a "representative agent" as being somehow methodologically individualist. If I proposed as a working assumption that human social phenomena may be explained without remainder as the outcomes of action the natural laws of the physical universe, and that any additional explanans (e.g. plans, intentions, will, etc.) are redundant, I suppose at some level it would be hard to object. But surely there are pragmatic objections. As in your definition above, the word "may" bears astounding weight. Again, thinking a bit about the neoclassical theory of the firm can be very helpful: although pervasive, the absence of an serious commitment to "methodological individualism" in our profession is seldom as stark. Cheers, Alan Isaac