functional explanation

Fig. 11 Functional(ist) explanation. In contrast with other kinds of explanation, e.g. purposive explanation (b) functional explanation (a) explains the occurrence of a phenomenon in terms of the functional needs of a system. One argument against functional explanation is thatwhere the ‘mechanisms’ of the system in question remain unclear, reference to ‘system needs’ adds nothing to purposive explanations and will mislead. Unintended consequences may still be acknowledged, but not system needs.

functional(ist) explanation

Often functional(ist) explanation involves a recourse to organic analogies, in which societies or social systems are seen as akin to biological organisms in which specified organs can be identified which fulfil specific system ‘needs’. For example, just as the function of the heart is to circulate the blood, the indispensable function of government may be suggested as establishing and implementing ‘policies for society’ (see also FUNCTIONAL PREREQUISITES, EVOLUTIONARY SOCIOLOGY). Alternatively analogies with servo-mechanical systems may be proposed (see also SYSTEM, SYSTEMS THEORY, CYBERNETICS).

According to Hempel (1959), in so far as it proves an adequate basis for explanation, functional(ist) explanation can be a form of deductive nomological explanation (see COVERING-LAW MODEL AND DEDUCTIVE NOMOLOGICAL EXPLANATION), possessing the following characteristics (where ‘i’ stands for any item or trait, within a system 'S ’ at a time ‘t’, and where ‘c’ stands for the ‘conditions’ or 'S etting’ in which the system operates):

1. at a time t, s functions adequately in a setting of kind c, characterized by specific internal and external conditions (statement of initial conditions);
2. s functions adequately in a setting of kind c only if a certain necessary condition, n, is satisfied (law);
3. if trait i were present in s then, as an effect, condition n would be satisfied (law);
4. (hence) at a time t, trait i is present in s (the EXPLANANDUM).

Criticisms of proposed examples of functional explanation are made by Hempel and others on a number of grounds:

1. that human societies and social systems are different from biological organisms in that they lack precise boundaries, change in form over time, do not exist as clearly demarcated species, are not born and do not die in the clear-cut way in which biological organisms do, and thus do not have clear-cut ‘needs’ or ‘functional requirements’ or clearly identified organs fulfilling these. Thus, a major problem exists in specifying the conditions for the 'S urvival’, ‘efficiency’, ‘adjustment’, ‘adaptation’, etc, of social systems in general, or of types of social system;
2. that no one has yet provided an adequate overall functionalist (and evolutionary) account of human societies which identifies the ‘units selected’ (compare NATURAL SELECTION, GENETICS) or confirms that institutions have in fact evolved on the basis of such units;
3. that even if broad ‘functional needs’ can be identified, any number of different items or traits may fulfil the required functions, e.g. a system of social welfare in modern societies may be provided by the state, by private insurance, or by political parties or criminal organizations, or any mixture of these (see FUNCTIONAL ALTERNATIVES, MERTON. POSTULATE OF FUNCTIONAL INDISPENSABILITY);
4. that even if it might explain the persistence of a system trait or institution (e.g. in terms of its present contribution), functional explanation is no substitute for a historical causal explanation of how an institutional arrangement came into existence, especially so where ‘evolutionary accounts’ are not seen as providing an adequate account;
5. given the problems outlined in (a), (b) and (c) (and also that human social systems are ‘open’ systems), functional explanation adds nothing to causal or purposive explanations (see Fig. 11) (GIDDENS, 1976). In particular, the argument can be made against functional explanation, that would-be functional explanations for the persistence of institutions in sociology fail to identify mechanisms which link the suggested functional need with the appearance of its claimed consequences (see also SOCIAL REPRODUCTION). Thus, taking this all into account, for GIDDENS, a so-called ‘latent function’, such as that of the Hopi rain dance in providing social integration and a ‘unitary value system’ for a small society, can be perfectly well restated without loss, as simply an ‘unintended’ outcome. We only mystify the phenomenon by suggesting that ‘functional needs’ account for this. In such circumstances the concept of function is redundant. We may use the term ‘function’ to refer to the intended purpose of a machine, etc, e.g. the function of the ‘ballcock’ is to regulate the quantity of water supplied to a water closet.

In the same way, the ‘functions’ of a political party, where intended functions are involved, may also be quite properly referred to. But reference to latent functions or to system needs in other terms than with reference to intentions, as in ‘functional explanation’ of the above type, remains obscure; (f) that, in general, one may always ‘define’ a particular institution or trait as ‘essential’ to a particular system, but this risks becoming a ‘tautology’, and merely ‘descriptive’ rather than explanatory, unless some further independent reasons can be given for regarding the relations between the system and the institutions identified as functionally significant.

Notwithstanding such criticisms, and whatever the problems of functional explanation, functional analysis remains a common feature of sociological accounts. The notion that particular societies or social systems have particular requirements seems inescapable in sociological analysis (compare Davis, 1948), even if these cannot always be precisely specified. Whether or not these requirements should be called ‘functional’ requirements, risking misleading biological analogies, would seem to be the issue.