10. Functionalism in Social Science. In social studies, a first variety of functionalism is the rational-choice approach. If one postulates that all choices maximize the expected utility (gain), then one must assume that those with unforeseen negative consequences correspond to wrong evaluations and mistaken probability estimates. This renders the “rationality postulate” unfalsifiable. In other words, if the choice is successful, the action is deemed to have been objectively rational; if not, the action is said to have been only subjectively rational. In either case, the formalism is saved at the expense of substance or, rather, mechanism, namely social interaction. For this reason, we regard the rational-choice approach as an instance of formalist functionalism in social science.
Not referring to social mechanisms, the rational-choice approach fails miserably to explain the very existence of social systems, from the family to the transnational corporation. This failure is a necessary consequence of its explicit adoption of ontological and methodological individualism. (Bunge 1996, Ch. 14) Still, when used in moderation, this approach may have some heuristic value. Indeed, it suggests looking for the reasons, good or bad, that motivate decisions. (Boudon 1999) It also suggests explaining some failed actions as results of miscalculations. (By contrast, success may be explained either by correct calculation or by favorable circumstances.)
What is known as social functionalism is a version of adaptationist functionalism. It focuses on social systems and their specific functions3 or activities. It also studies both the cohesive or system-preserving (“functional4“) and the divisive or system-interfering (“dysfunctional”) consequences of a system’s activities3. In other words, it investigates whether social functions3 are aptations (functions4) or even adaptations (functions5). Social functionalism is an alternative to both Marxist economicism and the traditional or culture-historical approach centered on statesmen and battles. (Trigger 1989) Though fathered by E. Durkheim at the turn of the century, functionalism flourished particularly in the UK and the US between ca. 1920 and ca. 1960, particularly in the works of B. Malinowski and E. R. Radcliffe-Brown in anthropology, G. Childe in archaeology, and T. Parsons in sociology.
The social functionalists postulated that all the social items (mechanisms, roles, norms, patterns, institutions, etc.) come into being and persist because they are useful to the social system concerned, or even to society at large. Put negatively: the social items that have ceased to discharge any useful function eventually disappear. Hence, the pattern of functional explanation would be this:
For all x: If x is a useful social item, then x is a standard social item.
b is a useful social item.
*(This character cannot be represented in ASCII text) b is a standard social item.
Obviously, this is not an explanation proper, but only a subsumption, or inclusion of a particular into a universal. Hence, it is at most a functional account, not a functional explanation. Moreover, it is not even satisfactory as a functional account, because it has a doubtful presupposition. This is the assumption that the social system is homogeneous, so that what is useful for its cohesion or preservation is good for every member of it. This presupposition is not even true for primitive societies, all of which retain negative items such as crippling kinship conventions and counterproductive superstitions. Thus, it is an instance of the fallacy of division.
Having said this, we grant that, although adaptationist-functionalist accounts may be shallow or even wrong, some functionalist questions may be rewarding. The reason is that many social functions are indeed aptations, or even adaptations. And the existence of such functions4-5 poses the problem of their origin and persistence. Now, every such problem can be analyzed non-teleologically as the sequence of questions: What is the internal activity in question? What is its role? Are the activities and roles valuable to the group (or the society) as a whole? If in fact they are aptations, are they also adaptations?
For instance, the persistence of religion in what is said to be the Age of Science and Technology raises the problem of the latent or collateral function(s) of some churches. Again, the corruption of political democracy by an economic elite invites looking at the way political campaigns are funded and candidates are selected. Both are empirical problems, not questions that can be solved a priori.
On the other hand, a moderate version of social functionalism requires an account of the distinction and inter-relation between microsocial units (individuals) and their subjective dispositions and activities on the one hand, and macrosocial features (which are now constraints, now stimuli to individual action) on the other. (Bunge 1998) Thus, in contrast to radical functionalism, its moderate counterpart does not proscribe the search for mechanism. On the contrary:
Functionalist analysis in sociology, as in other disciplines like physiology and psychology, calls for a ‘concrete and detailed’ account of the mechanisms which operate to perform a designated function. This refers, not to psychological, but to social, mechanisms (e.g., role-segmentation, insulation of institutional demands, hierarchic ordering of values, social division of labor, ritual and ceremonial enactments, etc.) (Merton 1957, 52)
To conclude, social functionalism can be barren or fruitful. It will be barren if it only restates Doctor Pangloss’s thesis that we live in the best of all possible worlds: that in which everyone maximizes his or her expected utilities. But it will be fruitful if it analyzes social systems and attempts to discover what makes them tick–that is, their mechanisms.
11. Functionalism in Technology. Recall from Section 5 that there is an asymmetry between function2 and mechanism (function1): one and the same function2 can often be implemented by different mechanisms (functions1). In other words, the functions2 right arrow mechanisms (functions2) relation is one-many. For example, the manager of a factory poses functional questions to the engineer, such as “Can you design a better mousetrap?” The technologist attempts to solve this problem by figuring out a novel mechanism, one that may involve stuff other than the traditional steel spring-such as electric shock, poison, infertility drug, or what have you. In short, technological problems may be posed in functional terms, but they can only be solved in mechanismic terms. Thus, black box functionalism does not help the technologist.
However, the users of artifacts can be forgiven for adopting a black box functionalist stand. For example, the function2 of a car’s catalytic converter is to transform the toxic carbon monoxide and nitric oxide, resulting from incomplete combustion, into inoffensive gases. The converter’s mechanism, which the driver and the car mechanic need not know, consists of two chemical reactions that are made possible by platinum and rhodium particles acting as catalyzers. By contrast, the designer of the converter must know its mechanism. Indeed, every technological function3, be it symbol-processing, transportation, or social control, must be “realized” in concrete artifacts. Hence, the designers must master such mechanisms; in particular, they must know the behavior of the “stuff” in question–physical, social, or what have you.
Another example is this: Accountants are uninterested in such features of a firm as the precise nature of its operations, organization, personnel morale, or public image; they focus on the company’s net inputs and outputs. The firm’s managers and consultants, on the other hand, will ignore those aspects only at their own risk: They must identify the underlying mechanisms if they wish to control them and, afortiori, to improve their performance. So much for black box functionalism in technology.
All technology is presumed to be adaptationist-functionalist because its practitioners are expected to design artifacts that perform pre-assigned functions (which are at the same time teleofunctions4), such as grinding, symbol-processing, or conflict-solving. However, technological design can be construed in either of two ways: as subordinating everything to a key function, or as combining the key function with collateral functions. For example, whereas some cars are only efficient vehicles, others are also elegant toys or status-symbols. Houses, TV sets, business firms, pieces of legislation, public-health programs, and other technological items are parallel. Therefore, one may speak of two kinds of adaptationist functionalism in technology: radical and moderate. Whereas the former sacrifices everything to short-term efficiency, the latter makes room for other factors and wider temporal and social horizons. It is the answer to such critics of performance-centered and context-free design as Vanderburg (2000).
In architecture and industrial design, radical adaptationist functionalism was born in the 1920s. The Bauhaus is the best-known functionalist school. Its mottos were “Form follows function” (actually first stated by Louis Sullivan) and “Less is more.” That is, it shunned the convolutions typical of the Renaissance and Baroque plastic arts which served no (key) functional purpose. In other words, technological functionalism is utilitarian and cuts the ties with tradition. Post-modern architecture is–or rather was–a reaction against this cold functionalism. Like all reactions, it failed.
Examples of radical adaptationist functionalism in sociotechnology are the fashionable methods for containing inflation and cutting costs, namely monetarism and downsizing, respectively. Monetarist policies are occasionally effective in reducing inflation; but they are inefficient in that they cause enormous social costs, such as unemployment and the accompanying increase in social expenditures. Likewise, downsizing effectively cuts the payroll; but it rips the firm’s social fabric and is thus inefficient in the long run. Like almost any other single goal, that of increasing profits can be attained in alternative ways. What matters in a firm is not just how much profit it makes, but also the technological and social means it employs to make it, as well as aiming at such additional goals as quality enhancing and increased market share.
Successful technology is only moderately functionalist: it admits values other than efficiency, and it balances the latter with social service. Moreover, it is mechanismic throughout, since it involves the design or control of mechanisms. For example, medical therapy–a biotechnology–consists increasingly in tampering with disease mechanisms; and management science–a sociotechnology–relies increasingly on knowledge of the social mechanisms that operate in the firm. In contradistinction to the traditional arts and crafts, all modern technology is characterized by the utilization or design of the mechanisms that mediate between input and output, and that maximize the output/input ratio. Hence, both formalist and blackbox functionalism are out of the question in technology.
12. Concluding Remarks. If we acknowledge that, in science and technology, there is no single all-purpose concept of function, we should try to re-analyze and systematize the various notions of function proposed in the literature. This is what we have attempted to do in this paper by distinguishing amongst five non-teleological concepts and their teleological counterparts, as well as by analyzing the logical relations amongst them. Moreover, this systematization has allowed us to propose a classification of those views that come under the label functionalism, showing that the term ‘functionalism’ inherits the ambiguities of the term “function.’ Whether or not our systematic attempt is deemed to be completely satisfactory, it should at least show that a synthetic perspective on functions, functional explanation, and functionalism in science and technology is a worthwhile goal.
a Received March 1999; revised October 2000.
b Send requests for reprints to Martin Mahner, Center for Inquiry-Europe, Arheilger Weg 11, D-64380 Rossdorf, Germany; email: firstname.lastname@example.org.
c We thank Michael Kary and Dan A. Seni, as well as an anonymous referee for useful suggestions that helped to improve the original manuscript.
DIAGRAM: Figure 1. Relations among the five concepts of function (function1-5) in biology. The figure can be read from left to right (“is needed to define”), or else from right to left (“implies”).
DIAGRAM: Figure 2. In psychology, social science, and technology, any one function(1-5), or part of it, can be attributed a purpose or goal, and thus have an intentional aspect: it can be a teleofunction.
Mahner, Martin, Bunge, Mario, Philosophy of Science