Chapter 7 (pt.1)
RATIONALITY AND PLANNING
Problems of Utilitarian Rationalism
Individuals are Variable Actors
Individuals are Complex Decision-Makers
Individual Actions Have Unintended Consequences
The Insufficiency of Rational Social Organization
Implicit in our discussion thus far has been a conception of social action involving
varying degrees of decision, varying extent of dependency constraining choice, and
varying time spans of accounting and planning. In this chapter we shall provide a
more explicit consideration of such variables, the relation of the conceptualization
to other positions, and its implications. We shall treat rationality as a particular
type of social action, and shall critique perspectives which take it as a postulated
universal human characteristic. Among the variables of social order we shall set
out in chapter 8, this discussion has particular hearing on term of planning and
considered or assumed adequacy of information, along with lesser relations to stability,
authority, similarity of definitions and multiplexity of bonds. Of necessity, this
chapter and the last are very closely linked and their treatments overlap. Thus the
mention made of problems in Arrow's analysis of social choice in the last chapter
is directly related to the critique of utilitarian rationalism in this chapter.
The relation in the case of Arrow, and the most important one in general, is through the notion of individuality. I have chosen not to make it a separate variable, despite (or indeed because of) its importance. Rather, our set of variables reflects individuality in a number of its dimensions. As the term is commonly used, it is not only value-laden, but carries a number of diverse connotations which at some levels are self-contradictory. Thus personal individuality, in my conception, may be greatest when the clearest possible distinction may be made between the given 'individual' and his environment and fellows. This occurs, however, precisely when individuality as a variable of frequency (rather than intensity) is least, when there are the firmest and most sociated units for contraposition. In sum, individuality is a complex issue, and one of the major problems of previous discussions of rationality is that they have tended to assume individuality as an invariant human characteristic, and rationality as an attribute of individuals. This view is untenable both for the reasons given in the last chapter and for others to be discussed below.
Rationality, as we shall consider it, is not an attribute of human nature but a variable, as we have noted, a type of action. As such, it is both the result of increased individuality, and productive of further increase. It is a variable among populations in its sociological significance. As a type of action (decision) rationality is closely linked to planning, for it involves the anticipation of outcomes and choices made on the basis of such anticipations. Clearly, the time spans at which one makes plans vary, and the shorter time spans necessitate more frequent decisions. Social instability, inasmuch as it is immanent, is seen to be primarily a result of this sort of variation in term of planning, and a further condition of it. Of course it is not a prime cause, but is produced by other factors as we shall consider in chapter 8. In this chapter our order of procedure will be to detail some of the problems which inhere in utilitarian rationalism as a perspective in social analysis, and set out some of the dimensions of rationality as we conceptualize it. Next we shall consider the matter of time-spans and planning directly, and lastly, we will discuss the reasons why rationality is insufficient as a principle of social organization, that is, in kernel, why the more people plan, the less predictable is the world.
Problems of Utilitarian Rationalism
That 'man does not live by bread alone' has been a difficult proposition for utilitarian economics and its descendants to accept. There are further problems in the facts that the definition of 'bread', the choice to seek it, the means used and the effort expended in the search are neither obvious nor the product of purely rational decision series. Not least of all, the definitions of seeker and eater are variables, not given. It is important for social (including economic) theory to apprehend irrationality in order to frame adequate considerations of the real world. This is not only a matter of apprehending the subjective or experiential in social life, important though that may be. It is also a matter of dealing effectively with the variable nature of rationality and the variable identities of social actors. In order to accomplish this needed apprehension of irrationality, social theory must come to grips with the determining conditions of action for actors as social individuals in concrete situations rather than assuming that patterns of action can be deduced from the postulated existence of rational actors who maximize their individual interests. The assumption, although it appears to make the individual central to the consideration, in fact eliminates the individual (as both psychological and sociological actor) by assuming him to be invariant.
One effect of this deception is to limit the discussion by defining a particular social and economic ordering implicitly into the concept of rationality. As Godelier puts it:
Although they start from individuals and their preferences and invoke a formal definition of economics as the form taken by the purposive behaviour of these individuals, contemporary theories actually introduce the relations between entrepreneurs and workers, that is, capitalist production-relations. They thus actually start from the system in order to analyze the behaviour of individuals, while appearing to deduce the system from the individuals. (1966, pp. 74-5)
The problem is not merely one of capitalist apologetics obscuring scientific truth, however. The problem is more basic. It inheres in utilitarian assumptions about the basic and knowable nature of the individual. Individuals are, in this perspective, seen as unchanging, identical, equally salient and equally individual in all situations and societies, and as at once simplistic and penetratingly omniscient and accurate in their decision-making. I suggest that the irrational conditions of human action are not merely impediments to be overcome, but necessary condition for stable social life. We shall take this up further in the last part of this chapter. Here we are concerned with the assumptions the utilitarians make and the reasons why they are incapable of producing a theoretical perspective which can deal with a wide range of social phenomena effectively, and particularly to account for differences among social organizations.
In utilitarian theory, individual rationality is both explanatory principle and goal. I attempt to show that it is weak in the former case, and potentially too strong in the latter, since it may be a destructive as well as a liberating force. I attempt no comprehensive discussion, but rather set out as concisely as possible three problem-areas with individualism-rationalism as a theoretical perspective on social action, dealing quite selectively with the literature. Before detailing the central problems themselves, we must at least briefly make our definition explicit. We can productively do this in the course of considering the usages of rationality in the work Godelier on the one hand and Olson on the other. These two are leading writers in varying traditions: choice theory in utilitarian economics (Olson), the perspective we are concerned to critique, and a variant of marxism (Godelier), a perspective from which another critique (partially overlapping our own) has been based. To be sure neither tradition is limited to a single writer, but Olson and Godelier are among the better representatives of each. We may briefly contrast their discussions as being about self-interest, on the one hand, and 'system-rationality' on the other. Neither is, I think, quite adequate. Both make value-judgments implicit in their usages, seeing rationality as an ultimate goal to be sought after. As I have suggested and will elaborate, I see it as an important but insufficient element in social organization, and as insufficient even in individual (as well as collective) decision-making.
In the most comprehensive treatment of the subject in the recent literature, Godelier manages to avoid defining the term, but fails to escape a clear value orientation: for him rationality is an end, the achievement of which is obstructed by the inadequacy of economic theory. Indeed, it is not only an end, but a means to the ultimate end of a society free from contradictions. Just as for the 18th century rationalists, an additional element creeps into the paradigm: rationalism is to be known not merely by the objective qualities of particular decisions, but by their consequences. Thus, for Godelier, rationality can only be evaluated in terms of an external set of standards representing the ends/ means series of a given theoretical position or paradigm:
The question of the 'rationality' of economic and social systems is thus at the same time the critical question of the 'rationality' of the various theories that have been put forward by the humane sciences, together with the forms of action and the objectives of the social groups representing the different economic and social systems that confront each other on the scene of history. (1972m p. xlii)
This is not merely a matter of saying that social theories reflect social conditions and social interests. Godelier is suggesting that a correct theory is one which correctly envisions the ends toward which the underlying determinate system of a given social organization is working.
Instead of a system such as this, I wish to consider rationality as an aspect of the way in which the actor makes a decision. This is a distinction which Godelier implies in his often insightful book when he notes:
In current writing the theme of economic rationality is presented in the form of two questions:
(1) How, in a given economic system, must economic agents behave in order to secure the objectives they set themselves?
(2) What is the rationality of the economic system itself, and can it be compared with that of other systems?
The first question aims at making explicit a rationality of intention that is adhered to by individuals, the second at throwing light on an unintended rationality -- the capacity, for example, possessed by a number of systems to ensure growth of the means of production, improvement in the standard of living, and so on. (1966, p. 10-11)
It is only by positing an evolutionary sequence that the capacity for growth can be considered very clearly a matter of 'rationality', that the change of the socio-economic system can be considered to be achieving an end. This, I think, is a misleading anthropomorphization of the social system. The crucial distinction between any sort of system-rationality (determination by hypostatized ends) and a variety of decision-making by actors remains at best implicit here, as Godelier does not bring out the decisional aspect of a 'rationality of intention'. In this he falls afoul of his own (earlier quoted) interdiction against deducing individual behaviour from system properties. He thus goes on to say that the first of his questions implies the second, that one can only understand the rationality of an actor in terms of the system in which he operates. It would be more accurate to say that one can only understand the decision an actor makes in terms of the system in which he operates, and even then one must remember that the system does not provide full understanding. It is feasible, I think, to understand "unintended rationality" in only one of two ways: either as meeting a goal of which the observer but not the actors are aware, or as working out the implications of a given structural arrangement. In either case it seems misleading to describe the events as the result of rationality. Both of Godelier's levels of questions seem valid, but a more straightforward statement could be made.
Let us distinguish, then between rationality as a characteristic of decisions, and immanent states (similar or different) which may be implied by any given set of social conditions. Godelier is clearly concerned with the latter, but his very identification of them in terms of rationality prevents him from considering certain of their crucial conditions. These are the frequency and pattern of effect of social decision-making. As we have observed, his second variety of rationality is defined in terms of results rather than decision-making. It is this which could keep the two questions separate. We should ask how economic agents do behave in attempting to secure objectives rather than how they must behave in succeeding to secure objectives. Once we do this it is possible to recognize the role of decision-making, and, especially to see that some matters are not and cannot be subject to intentional creation. Further, the accelerating frequency of decisions which rationalism produces becomes apparent. Whatever coherent or beneficial unintended social occurrences there may be, they are precisely not the results of rationality. The only way to accomplish Godelier's goal of a comparative enterprise in the sociology of economies is to regard a system as 'rational' in proportion to the 'rationality' of its constituent actors, not to make any value judgement about the virtues of its results.
In these terms, then, rationality is the process of defining alternatives, viewing them as (with various probabilities) representative of different end-states, and choosing among them as (often implicitly) different measures of some common value. It should be emphasized that this is value as it appears to the chooser, not underlying value (as in Marx's or Ricardo's labour-value). I choose this definition self-consciously, not because I think there are no underlying values, is no 'infrastructure', or is no 'hidden determination' beneath dominant forms. Rather, it seems to me that too much subtlety of linguistic usage is lost by generalizing 'rationality' beyond immediate situations of choice or analysis. Rationality is a property of consciousness, and thus may not directly be a property of systems. The discussion of the fetishism of commodities in the first chapter of Capital includes brilliant examples of the analysis of the significances of certain 'outward appearances' through the faulty rationality of actors (Marx, 1867). The celebrated underlying 'irrationality' of the capitalist system depends on actors planning rationally on the basis of just such inadequate understandings of economic life. What makes the capitalist system irrational for Marx, is, then, in part that it makes the rational actions of actors insufficient, renders rational planning inadequate. This is, of course the meaning behind the 'anarchy of production' and the message of doom read by Engels and others in cycles of business crises.
Rationality, then, is no simple 'good'. Neither is rationality simply self-interest. It is, of course, one way in which individuals may determine their interests. How closely the two concepts are linked for the tradition of utilitarian economics may be seen in Olson's attempt to distinguish the two. He judges rationality still by its effects, though in this case its measure is its efficiency. Rationality is distinguished from self-interest, Olson tells us, by the fact that one can be a more or less effective philanthropist:
Selfless behaviour that has no perceptible effect is sometimes not even considered praiseworthy. (1971, p. 64)
Aside from the far from universally accepted utilitarian morality implied here, we may note that what has happened is that self-interest has been defined in terms of economic benefit, profit-seeking, so that a person who maximizes another end, another value even though no less his own, is not seen as acting in his own interest. Self-interest becomes an imputation of the analyst on the basis of what he considers to be of interest. Rationality comes to be understood
in the sense that their [individuals'] objectives, whether selfish or unselfish, should be pursued by means that are efficient and effective for achieving these objectives. (197§, p. 65)
This means, however, that individuals may only be rational to the extent to which they are omniscient. That is, the quality of individual's information and the predictability of ends influence the evaluation under this scheme of the individuals' rationality. The problem here resolves to a particular case of the problem of assuming (as do Olson and most of his fellow formalists and utilitarians) a 'purely rational' actor. With such an assumption standing centrally behind them, their statements may only have meaning as they define the maximum limit of rational efficiency under certain circumstances, but their utility diminishes as one moves away from limiting cases and into consideration of the real world. Individuals here turn out to have conflicting objectives, varying strengths of motivation, and varying (and differentially accurate) perceptions of the alternatives available to them and their relative efficiency. As I shall attempt to show later, there are many situations where social or cultural limitations on an individual's consideration of alternatives, even where he pursues seemingly clearcut and apparent ends, may be of the utmost social significance in maximizing long-term collective benefits over short-term individual (or collective) gains.
Godelier and Olson are more perceptive than many thinkers in their respective traditions. The bulk of marginalist economics, for example, tends to assume a simplistic and ascertainable ordering of preferences which (supposedly) accounts for the individual's particularities in his representation of the general principle of the maximization of self-interest. This is carried over into Blau's (1964) version of sociological 'exchange theory' and Barth's formulation of 'transactional analysis' (1966, 1967) as individuals are made out to be rational about nearly all their social relations in the same way the marginalists think they are about their market behaviour. Homans (1958, 1961) even finds organismic roots for this in behaviourist psychology: rationality is avoiding punishment and seeking reward. In social situations, however, rewards are both near and distant, the intensity and duration of punishment is unpredictable, and people are able to understand and/or control more or less of the events in their lives. They act rationally indeed in organizing their behaviour, but that is not the only source of pattern to their action, nor could it be so. As Davis has suggested, in a critique of exchange theorists and transactionalists:
a rational person is one who strives to maximize profits. It is one of the contributions which anthropology can make to economics, to point out that profit is not the only possible relation between things, and that there are actually existing cases where people maximize relations other than profit. (1973, p. 164)
As we shall see, when individual rationality does become the dominant pattern in social life, continuity fades and society weakens.
Let us now look at three sets of problems which beset ratio al-man assumptions as principles of social analyses. These stem from the facts that (a) individuals are variable actors, (b) individuals are complex decision-makers, and (c) individual actions have unintended consequences. Bear in mind as well the fact (I hope already established) that individuals are salient social actors in varying degree.
Problems of Utilitarian Rationalism - Individuals are Variable Actors - Individuals are Complex Decision-Makers - Individual Actions Have Unintended Consequences - Time Spans and Plans - The Insufficiency of Rational Social Organization