Social Choice and Cultural Bias

Douglas J, Douglas M, & Thompson M (1983). Social Choice and Cultural Bias. IIASA Collaborative Paper. IIASA, Laxenburg, Austria: CP-83-004


Download (3MB) | Preview


These three essays are the fruits of a little Winter Study, in December 1981, which enabled Mary Douglas and James Douglas (both of Northwestern University, Illinois, USA) to visit the System and Decision Sciences area at IIASA, there to collaborate on an interdisciplinary (or, more properly, non-disciplinary) task. 'Institutional bias' was the provisional title for what we had in mind and our aim was to try to wrap some cultural and political context around the paradoxes of social choice.

The Western liberal tradition holds rationality and individuality in high regard. It has to; otherwise it would not be liberal, nor would it cohere long enough to become a tradition. But too high a regard for reason may exaggerate the part played by conscious design in the conduct of human affairs, and too high a regard for the individual may exaggerate both his ability to identify the things that he values and his scope to arrange them in an order of his own choosing. The unity of these essays lies in their common critical theme; all three, in their different ways, take issue with the liberal tradition.

James Douglas' point of departure is the recognition that actual political systems coped very effectively with the paradoxes of social choice long before Condorcet and Arrow revealed that those paradoxes existed. Since they could not have been consciously designed to do this, these systems must have evolved. The lowly dung beetle, as it decides whether to try to find a new and untenanted cow-pat or to stick with the ever crustier one that it has, follows a personal strategy so subtle as to require integral calculus in its solution. Could it be that we are no better equipped to design our political institutions than is the dung beetle up to doing 'A level' mathematics? Trial and error -- success and failure over countless generations -- we conclude, is what has led the individual dung beetle to the so-sensible strategy that it shares with every other dung beetle. The rational-choice theorist, if he could bring himself to study so distasteful a subject, would have to conclude that, with such a lack of variation in the preference orderings of cow-pats as we go from one individual to another, there is some form of dictatorship operating within the dung beetles' social system. Of course, in the dung beetle case, the lack of individuality -- the dictatorship -- is about as extreme as it could possibly be and it would be foolish to pretend that it provides a more valid model of human social life than does the theory of rational choice. No, our aim is not to jump to the dung beetle's extreme but, rather, to ask: 'extreme from what?' The answer has to be: 'from a situation in which, because the individual preference orderings are so gloriously varied that no parallelisms -- no little clumpings or mutual alignments -- can exist, there can be no dictatorship'. We would argue that such a situation, though intellectually intriguing, has nothing to do with the description of the life of man in society...apart, that is, from saying that it is not like that.

The "invisible dictators"' that the rational-choice theorist conjures up in response to the parallelisms -- the departures from individual perfection -- that he continually bumps up against are, collectively, an old friend of the anthropologist. They are culture. The only trouble is that invisible dictators are plural and culture is singular. To resolve this paradox we begin by defining our extreme at the opposite pole to that defined by the theory of rational choice. Instead of the fine independence of the individual we take as our model the dung beetle. 'To what extent, and in what ways, does our behavior distance us from it?' we ask, rather than 'to what extent does our behavior fall short of the individualist ideal?'. But are not these differences, like a knot in a length of string, simply different ways of measuring the same thing? No, because there is no continuum -- no measuring scale -- between these two extremes. Total dictatorship is attainable; perfect individuality is not. We can measure our divergence from the attainable but not from the unattainable, and the attempt to do the latter we label 'the individualist fallacy ' .

The dung beetle has, over the generations, adapted so as to take advantage of the adoptive possibilities of an environment within which certain laws (such as the progressive drying out of cow-pats) hold inexorable sway. In much the same way, actual political systems have evolved to take advantage of an environment wherein Arrow's impossibility theorem holds away. But what is particularly interesting is that, though all these systems cope with the paradoxes of social choice, they do not all cope with them in the same way. Mary Douglas comes in at this point and, venturing into the untrodden terrain that lies between cultural anthropology and organization theory, sketches out a three-fold typology of socially viable organizations, each one of which stabilizes itself with the aid of its appropriate and distinctive cultural bias. So it is culture -- man's self-reflexive ability -- that distances him from the dung beetle. At the same time, this idea of cultural biases stabilizing their appropriate social organizations ('departures from individual perfection' from the rational-choice viewpoint) allows us to reconcile a plurality of invisible dictators with a singular culture.

The final essay explores the way in which these two levels -- the cultural biases that always intervene to prevent the attainment of perfect individuality and the political systems that cope with the paradoxes of social choice -- fit together. Cultural biases, it argues, are in perpetual contention. One organizational form may, for a time, gain dominance but it can never permanently eliminate the others. Within this flux certain conjunctions of cultural biases (and of their associated organizations) are stabilizable (or, at any rate, change only in slow time) and these persistent regularities we label 'political regimes.

Item Type: Monograph (IIASA Collaborative Paper)
Research Programs: System and Decision Sciences - Core (SDS)
International Forum (IFO)
Depositing User: IIASA Import
Date Deposited: 15 Jan 2016 01:54
Last Modified: 20 Jul 2016 23:37

Actions (login required)

View Item View Item

International Institute for Applied Systems Analysis (IIASA)
Schlossplatz 1, A-2361 Laxenburg, Austria
Phone: (+43 2236) 807 0 Fax:(+43 2236) 71 313