Functional Analysis in Sociology Author(s): Harold Fallding Source: American Sociological Review, Vol. 28, No. 1 (Feb.

, 1963), pp. 5-13 Published by: American Sociological Association Stable URL: http://www.jstor.org/stable/2090451 . Accessed: 05/06/2013 02:51
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

American Sociological Association is collaborating with JSTOR to digitize, preserve and extend access to American Sociological Review.

http://www.jstor.org

This content downloaded from 143.225.109.4 on Wed, 5 Jun 2013 02:51:01 AM All use subject to JSTOR Terms and Conditions

AMERICAN
February, 1963

SOCIOLOGICAL

REVIEW
Volume 28, No. 1

FUNCTIONAL ANALYSIS IN SOCIOLOGY

LIKE other statements which have ap- making, we have not yet the common code peared in recent years,' the papers on which Merton sought. Yet more than ever functional analysis by Kingsley Davis 2 there is need for a code that would serve as and Ronald Philip Dore 3 illustrate the scaffolding for diverse workers to build in amount of spadeworkthat remainsto be done unison. The scaffolding could always be dishere. While the contributionsof Talcott Par- membered mentally by anyone looking into sons 4 and Robert Merton 5 have been epoch- basic questions,6but for anyone interpreting data it needs to stay provisionally fixed. The following is not the code called for but a 1 See Harry C. Bredemeier, "The Methodology rally to persevere in winning through to it. of Functionalism," American Sociological Review,
20 (April, 1955), pp. 173-180; Bernard Barber, "Structural-Functional Analysis: Some Problems and Misunderstandings," American Sociological Review, 21 (April, 1956), pp. 129-135; Walter Buckley, "Structural-Functional Analysis in Modern Sociology," in Howard Becker and Alvin Boskoff (eds.), Modern Sociological Theory in Continuity and Change, New York: The Dryden Press, 1957, pp. 236-259; Francesca Cancian, "Functional Analysis of Change," American Sociological Review, 25 (December, 1960), pp. 818-827. 2 Kingsley Davis, "The Myth of Functional Analysis as a Special Method in Sociology and Anthropology," American Sociological Review, 24 (December, 1959), pp. 752-772. 3 Ronald Philip Dore, "Function and Cause," American Sociological Review, 26 (December, 1961), pp. 843-853. -'See especially, Talcott Parsons, The Social System, Glencoe, Ill.: The Free Press, 1951; Talcott Parsons, R. F. Bales, and E. A. Shils, Working Papers in the Theory of Action, Glencoe, Ill.: The Free Press, 1953; Talcott Parsons and Neil J. Smelser, Economy and Society, Glencoe, Ill.: The Free Press, 1956; Talcott Parsons, "An Outline of the Social System," in Talcott Parsons, Edward Shils, Kaspar D. Naegele and Jesse R. Pitts (eds.), Theories of Society: Foundations of Modern Socio5

logical Theory, Glencoe, Ill.: The Free Press, 1961, pp. 30-79. 5 See Robert K. Merton, Social Theory and Social Structure, Glencoe, Ill.: The Free Press, 1957; Robert K. Merton, "Social Problems and Sociological Theory," in Robert K. Merton and Robert A. Nisbet (eds.), Contemporary Social Problems, An Introduction to the Sociology of Deviant Behavior and Social Disorganization, New York: Harcourt, Brace and World, 1961. 6 An analysis at this basic level has been undertaken by a number of authors. See, e.g., S. F. Nadel, The Foundations of Social Anthropology, London: Cohen and West, 1953, pp. 368-408; Ernest Nagel, "A Formalization of Functionalism," Logic Without Metaphysics, Glencoe, Ill.: The Free Press, 1956, pp. 247-283; Dorothy Emmet, Function, Purpose and Powers: Some Concepts in the Study of Individuals and Societies, London: Macmillan, 1958; Carl G. Hempel, "The Logic of Functional Analysis," in Llewellyn Gross (ed.), Symposium on Sociological Theory, Evanston, Ill.: Row, Peterson and Co., 1959, pp. 271-307; Richard Bevan Braithwaite, Scientific Explanation, a Study of the Function of Theory, Probability and Law in Science, Cambridge: The University Press, 1959, pp. 319-341.

This content downloaded from 143.225.109.4 on Wed, 5 Jun 2013 02:51:01 AM All use subject to JSTOR Terms and Conditions

AMERICAN SOCIOLOGICALREVIEW of any activity contained within a social system we refer to the effect it has in strengthening (or weakening) these productive bonds. For instance, conflict itself has been analyzed by Coser 8 as having positive functions in some social structures, and by this he simply aims to show that it can strengthen existing productive bonds by sealing their corrosions e.g., when a hampering grievance is aired and removed and everyone is "able to get on with the job." Contained within a casket of existing bonds conflict may work like fire to purge them of imperfections; but, without these, it would presumably not even constitute a social phenomenon, much less a functional one. The notion of function in connection with societies and their component groups has been variously employed, as Merton 9 showed. But the usage that has come to prevail takes as the function of an activity within a system the contribution it makes to the whole. We have therefore come to see the importance of specifying precisely both the part and the whole to which a functional statement refers. A practice which is functional within one social region need not be functional in one which is more (or less) inclusive. Other things also have to be specified if functional statements are to mean anything. As Nagel 10 stressed, we should say to what state of the whole the practice in question is contributory. But, more important than any of this, should we not bear in mind all the time what the product of the whole system is, since this itself may or may not be functional for those who bear the cost of the system and so expect to benefit? Very frequently in sociology it is whole, bounded action-systems that are being judged to be functional or dysfunctional for man and, only by transference, any parts within them which may make them so. And this is because social action is prompted by human need. The examination of the properties of the "functional system" that has been under8 Lewis A. Coser, The Functions of Social Conflict, Glencoe, Ill.: The Free Press, 1956. 9 Robert K. Merton, "Manifest and Latent Functions," Social Theory and Social Structure, Glencoe, Ill.: The Free Press, 1957, pp. 19-84. 10 Ernest Nagel, op. cit.

The paper states in synoptic form only some of the set of assumptions which, if assembled, might orient sociologists more uniformly to their task. As it is the tradesman's last that is being fashioned, the writer does not follow into the assumptions behind the assumptions as the philosopher would-yet an exercise like this makes it easier to see what some of these postulates are. The writer is aware, moreover,that what he proposes as definite will still prove controversial. His main purpose is to represent functional analysis in sociology not as part of explanation but as a form of measurement which belongs to the natural history phase of the science. And an attempt is made to elicit some of the specific dimensions of measurement, of the kind Hempel 7 has asked for. As this proceeds, the way in which disorganization is implicated in dysfunction begins to appear plainer. The kinship connection of these two chief mourners has scarcely yet been traced. 1. Functional analysis involves evaluation. Sociology treats behavior in situations which pose a problem of regulation, rather than dealing solely with situations where regulation is achieved. But before we ever speak of social events, a system of regulated interpersonal contact either exists in some degree, or the participants are aware that one is called for and have adumbrated itor, having lost what was won, have abrogated it. To designate the realized desideratum we can speak of "a group" or, alternatively, of "social arrangements," "social organization," "social structure," "social system." While if we speak of the function of a social system as a whole it is simply to specify what product is secured through the bridges and bonds thus established between man and man. It may be bread or bullion, music or medicine, sympathy or salvation. Then if we speak in addition of the function
7 Hempel writes: "For the sake of objective testability of functionalist hypotheses, it is essential, therefore, that definitions of needs or functional prerequisites be supplemented by reasonably clear and objectively applicable criteria of what is to be considered a healthy state or a normal working order of the systems under consideration; and that the vague and sweeping notion of survival then be construed in the relativized sense of survival in a healthy state as specified." Carl G. Hempel, op. cit., p. 294.

This content downloaded from 143.225.109.4 on Wed, 5 Jun 2013 02:51:01 AM All use subject to JSTOR Terms and Conditions

FUNCTIONAL ANALYSIS IN SOCIOLOGY taken by Nagel 11 seems to concentrate on the penultimate question of the sustained functioning of the system of action and does not ask whether, when functioning, it is functional for those who operate it. Nagel's formalization greatly facilitates our analysis of the internal processes of change and compensatory counter-changeby which a system preserves equilibrium. But this has to be linked to the more ultimate question of whether the system itself is functional or dysfunctional in yielding products matched to human need. In the writer's view, asking this ultimate question is what makes it worth while to ask the penultimate one.'2 Malinowski's 13 insistence on this is one of his enduring contributions to the discussion. Furthermore, there is a teleological residue in functional thinking that is scarcely disposed of by Nagel's 14 demonstrationthat the explanatory element in functionalism is simply causal explanation put in a roundabout way. Where a need-satisfaction stands at the end of a process of human endeavour it exercises some directive power over the efforts taken to achieve it. Here we have a case, then, of the kind of process for which Braithwaite 15 has striven to preserve recognition, wherein the anticipated future goal controls the present movement towards it, so that the end achieved is not the passive effect of a causal chain but, to some extent
Ibid. Bredemeier suggests that functional analysis loses its point if it fails to hold in view at least the needs which are induced in the actors by the normative definitions of the dominant culture. Harry C. Bredemeier, op. cit., p. 179. Cancian, on the other hand, passes over the idea that "functional" might mean "fulfilling a basic need" for no reason save that such a view is "inappropriate according to Nagel's concept of a functional system." Francesca Cancian, op. cit., p. 820. Even so, one wonders whether it would be as inappropriate as Cancian says. For we would scarcely ask, as Cancian suggests, what keeps need constant but would make need-satisfaction our G, and ask whether or not this is maintained. 13 B. Malinowski, A Scientific Theory of Culture, and Other Essays, New York: Oxford University Press, 1960. 14 Ernest Nagel, "Teleological Explanation and Teleological Systems," in Herbert Feigl and May Brodbeck (eds.), Readings in the Philosophy of Science, New York: Appleton-Century-Crofts, Inc., 1953, pp. 537-558. 15 Richard Bevan Braithwaite, op. cit., pp. 328336.
12 11

at least, the cause of its own causes. Braithwaite 16 points out that the field of study explored by cybernetics is largely concerned with teleological mechanisms like this. Such processes, once launched, may achieve the end in view-or may fail and so be "in vain." Does not the anticipation of an end to be achieved underlie all our judgments of function or dysfunction, when those judgments are made in such a way as to imply a comparison with the alternative possible outcome? In the writer's view, what we are interested in when we make this comparison is not explanation but evaluation. 2. The evaluation involved in functional analysis is objective and needs no apology. To ask the function of any social arrangement is to call for its justification-or alternatively for its condemnation. The positive and negative polarity inherent in the terms (eu) functional and dysfunctional should betray at once that evaluation is afoot. A great deal of unnecessary hedging in sociological work would be obviated if this could be frankly admitted. At the same time, sociological work could be more easily purged of covert, private evaluations, if it were allowed that evaluation of this objective kind is intrinsic in sociological analysis, and altogether honorable. Yet, in saying this, we have to distinguish the two meanings given to "subjective" when that state of mind is unfavorably contrasted with the "objective." It can mean biased or intuitive. When it has the former meaning subjectivity is to be deplored, because by it the person's perception is distorted. When it has the latter meaning it is simply to be regretted, since the person is not yet able to share his vision of what may well be the truth. The first kind of subjectivity has to be expunged from science altogether, but the second kind is the anlage of science and has to be protected and fostered till its testimony can be objectified. In saying that the evaluation in functional analysis is objective, freedom from subjectivity of the first kind is mainly what is being claimed of course. Yet this gives us grounds for believing that freedom from subjectivity of the second kind can also be achieved with time. We imply objective evaluation of two
16

Ibid., p. 328.

This content downloaded from 143.225.109.4 on Wed, 5 Jun 2013 02:51:01 AM All use subject to JSTOR Terms and Conditions

AMERICAN SOCIOLOGICALREVIEW then be implying (i) that buying political support from diversified groups by dispensing help to them is an efficient means of concentrating power and (ii) that all such groups need some access to power and that the power secured by them needs to be concentrated to some degree. Or finally we may say, with Parsons,22 that two functions of institutionalizing a "collectivity-orientation" in the professional role of the scientist are (a) to protect the public from arbitrary interference by men whose special knowledge gives them an advantage and (b) to expose the ideas of any scientist to the critical scrutiny of his fellows. We would then be implying (i) that role-institutionalization is an efficient means of restraining individuals who, in an intellectual sense, handle dynamite, and (ii) that society needs their contribution to knowledge. But it would save misunderstandingif it could be appreciated that evaluation made in ways like the above is sociological evaluation only. So far as present society is concerned, X is functional, Y dysfunctionalthat is always the implicit stipulation. There may well be supernal heights or historical perspectives from which a socially functional arrangementcan be judged bad and a dysfunctional one good-just as ill-health is sometimes recalled with gratitude because it brought spiritual blessing, or poverty because it put one in the way of great fortune at a later time-but that would not be incompatible with recognizing the arrangements as socially functional or dysfunctional.23 Futhermore, this helps us to see
22 Talcott Parsons, The Social System, London: Tavistock, 1952, pp. 335-345. 23 Merton makes the same point as this when he stresses that judgments about social disorganization are not moralizing judgments but technical judgments about the working of social systems. Robert K. Merton, "Social Problems and Sociological Theory," in Robert K. Merton and Robert A. Nisbet (eds.), Contemporary Social Problems, An Introduction to the Sociology of Deviant Behavior and Social Disorganization, New York: Harcourt, Brace and World, 1961, pp. 719-723. Although he is referring to the notion of social disorganization, Merton attempts in this same essay to relate social disorganization to social dysfunction. He suggests that disorganization may be viewed as the resultant of multiple dysfunctions. Ibid., pp. 731-737. Martindale states that all the critics of the notion of social disorganization base their objection at least

kinds, in fact, whenever we give a function. Basically, we are making a judgment as to whether the expenditure that goes into the creation and maintenance of the arrangement is worthwhile; but we determine this worthwhileness by both a backward and a forward look, as it were. The backward look tries to sum up the efficiency of the arrangement in producing its effects. To the extent that it is inefficient, wasteful, it is dysfunctional in a way. The forward look examines whether the effects themselves are valuable in terms of some schedule of needs which we postulate for the life of man in society. Some instances will make this plainer. We may say, as GluckmanI7 has done, that the function of rituals of rebellion in Africa is to channel off the resentment that the natives feel for their chiefs and so preserve stability in the existing authority arrangements-thereby ensuring the continuing supply of everything those arrangements guarantee. We would then be implying (i) that these ritual expressions of aggression are efficient means of dissipating resentment and (ii) that it is desirable to maintain an uninterrupted need-satisfaction, and hence social stability. We may say, with Davis,18 that the function of social stratification is to "insure that the most important positions are conscientiously filled by the most qualified persons."19We would then be implying (i) that a grading of rewards is an efficient means of motivating suitable persons to accept greater responsibilities,while a division of labor into tasks of unequal importance is an efficient means of securing common needs, and (ii) that it is desirable to satisfy common needs continuously. We may say, with Merton,20that the function of the political boss in the U.S.A. is "to organize, centralize and maintain in good working condition 'the scattered fragments of power' which are at present dispersed."21 We would
17 Max Gluckman, Rituals of Rebellion in SouthEast Africa, Manchester: University of Manchester Press, 1954. 18 Kingsley Davis and Wilbert E. Moore, "Some Principles of Stratification," American Sociological Review, 10 (April, 1945), pp. 242-249; Kingsley Davis, Human Society, New York: Macmillan, 1959, pp. 364-389. 19 Ibid., p. 367. 20 Robert K. Merton, op. cit., pp. 70-82. 21 Ibid., p. 72.

This content downloaded from 143.225.109.4 on Wed, 5 Jun 2013 02:51:01 AM All use subject to JSTOR Terms and Conditions

FUNCTIONAL ANALYSIS IN SOCIOLOGY that intellectual judgments about function and dysfunction contain no ethical imperative. We cannot pass from "The instruction of children is functional" to "Thou shalt instruct children." As sociologists we hold no whip. (Fortunately, though, our sociology is redeemed from the final futility by the fact that men exist who are not sociologists merely.) 3. Evaluating social arrangementsas functional or dysfunctional is equivalent to classifying them as normal or pathological. This is a necessary preliminary to the search for causal explanation. A physiologist cannot arrive at the function of the liver by generalizing directly from a random collection of livers which contains some diseased specimens. He distinguishes between the diseased and healthy organs at the outset and, setting the diseased ones aside, generalizes from the healthy ones. Certainly, by contrast, he gains some understanding of healthy functioning from an examination of the diseased cases, but can only do so if he first sets them in opposition by classifying them apart. His account of the liver would be altogether confounded if it simply averaged the properties of the whole collection. Distinguishing normal from pathological cases is one of his first assignments, and precedes causal knowledge of the conditions of normal or pathological functioning. Social systems are more complex than livers, of course, but the two things are alike in this respect. The instance will therefore serve to show what confusion can invade an intellectual discipline if, being concerned with things that have systemproperties,it fails to recognize that character in them and evaluate them accordingly. It
partly on the fact that valuations are inherent in it. See Don Martindale, "Social Disorganization: The Conflict of Normative and Empirical Approaches," in Howard Becker and Alvin Boskoff (eds.), Modern Sociological Theory in Continuity and Change, New York: The Dryden Press, 1957, pp. 340-367; E. M. Lemert, Social Pathology, New York: McGraw-Hill Book Co., 1951, pp. 3-26; C. Wright Mills, "The Professional Ideology of Social Pathologists," American Journal of Sociology, 49 (Sept., 1943), pp. 165-180; Louis Wirth, "Ideological Aspects of Social Disorganization," American Sociological Review, 5 (Aug., 1940), pp. 472482. But could the critics see that the need for evaluations originates from the data and not the evaluators the practice might seem less objectionable to them.

should show, further, that evaluation, forced upon us in this way, is simply scientific measurement. It amounts to quantification of those dynamic properties, possession of which defines the class of things in question. To evaluate systems is to have appropriated the dimensions for measuring them. In what follows, the three major dimensions for the significant measurement of social systems are proposed, as well as some of the minor dimensions implied in them. It may requirea titanic effort to overcome the clinging prejudice that any view about social desiderata will be ideologically colored and therefore suspect, but the effort has to be made. Through the whole of modern sociology two principal objective social imperatives have been fulminating, forcing recognition for themselves. Adaptive change (which implies rationality) and stability have both had to be assumed necessary in the social arrangements men make. These are the two main components of efficiency which, it was said, we gauge by a kind of backward look. They are the states of the social system (Nagel's G's) without which it cannot be properly productive of anything. They are, as such, nobody's political ideology; and there is no sense in calling a sociologist conservative or radical because his work illustrates the necessity of the one or the other-and as likely as not it will illustrate the necessity of both. Man, an anxious creature who looks before and after, works for his satisfactions over time and has therefore to bind time, and social organization and culture take their origin from this and must develop a certain conservatism to be of service to him. Yet, even in its most colloquial usage, stability has never meant fixity. We must not suppose that stable social arrangements mean arrangements that are fixed forever. They are simply arrangements that materialize as expected for as long as they are wanted. In no sense, either, is stability an opposite of change, so it would be wrong to think that combining them necessarily meant striking some mean of moderation or gradualism. Commitment to stability still leaves men free to adapt to a world which changes, or of which their knowledge changes, by various means. But to be of service stable social arrangements must yield-have a certain plasticity.

This content downloaded from 143.225.109.4 on Wed, 5 Jun 2013 02:51:01 AM All use subject to JSTOR Terms and Conditions

10

AMERICAN SOCIOLOGICALREVIEW -at the very least "preparedto put up with them." Everyone must have his reward or the guarantee of it. Thus the maintenance of stability is essentially a matter of the enlistment of motives. While every person's reward will not be the same, everyone will have the same expectation of a certain reward and hence share a positive regard for, or valuation of, the arrangements.We probably claim too much if we mean more than this when we say that value-consensusis the basis of social order. But where this is lacking in even a single individual there will be disorganization in some degree, however localized. His guarantee of reward is the individual's security, and conformity to the arrangements of the group is what he will give in return for the security gained. In this lies the group's power of control over him, and the insurance of its own organization at the same time. A consensus (of the kind specified) and security and control for individuals, are therefore further indices of social normality. We can also say that communication of a certain effectiveness will be necessary if everyone involved is to know what the arrangementsare and if consensus is to be reached in upholding them. Stability implies all this. But we can say more. Whether individuals are satisfied with their rewards depends largely on whether they think them fair in comparisonwith the rewards of others. Any social system at all is in danger of being convulsed if a supposed injustice is unearthed-although the convulsion does not necessarily occur. The victim may grin and bear it if he thinks rebellion will prove more costly of justice still. Or there may be sanctioned channels through which he will work for a change. But, on the other hand, he may harbor resentment and be uncooperative, or take the law into his own hands and fight for his rights. Whatever the eventuality, the basement beneath every social system is piled with inflammatory tinder. Justice is the thing that damps it, although it is essentially subjective justice. It is justice which is seen and felt to be sufficient ("fair enough") by those concerned. And what is accepted as sufficient today may appear in a different light tomorrow-even without the situation changing. This is the craving in the elementary social nature of man which has to be appeased

We have no choice except to take as normal, healthy or functional those social arrangements which exhibit both stability and adaptive change in the combination demanded by the time and place, and as abnormal, pathological or dysfunctional in some degree those which do not (and it is of course a matter of degree). It is admittedly a delicate calculation to make and often still largely beyond us, so we make it in a rough, rule-of-thumb way (perhaps in a way which is subjective in the second of the above senses.) Yet we have no choice but to make it-and very early in our sorting of the data. To do this does not imply that the abnormal or pathological phenomenon is being pronouced "evil" in some ultimate sense. It simply means it is being classified sociologically, given a negative sociological valency so that it will not be confused with the positive counterpartwhich in other respects will deceptively resemble it. The well-organized family is thus not to be taken for the same sociological phenomenon as the disorganizedone, the high morale department store for the same thing as the low morale store, the nation riddled with suicide and homicide for the same thing as the one where these are rare, the church or school which moves with the times for the same thing as the one made redundant through its archaisms. It is only after we have set such contrasting phenomena in classes apart that we can sample the cases in each class and arrive at causal laws about functional and dysfunctional systems. It would be a pity if the attention that has been given to functionalism were taken for an invitation to make functional analysis the end of sociological inquiry-for then it would be still-born. It is rather the end of sociology to explain-"explained" functions: to show what things have a constant association with various functional or dysfunctional operations. 4. The stabilization of rational arrangements is what we mean by social organization (social cohesion or solidarity being roughly synonymous.) The stability requirement presents an internal challenge, the rationality requirement an external challenge. The internal challenge to social organization is to motivate lasting support. Stability is only secured in social arrangementsif everyone involved in them is "satisfied with them"

This content downloaded from 143.225.109.4 on Wed, 5 Jun 2013 02:51:01 AM All use subject to JSTOR Terms and Conditions

FUNCTIONAL ANALYSIS IN SOCIOLOGY by the institutions man makes if it is not to devour them. Homans,24 having explored this "subinstitutional" level of behavior, points out how men face a problem perennially of reconciling their institutions with it. The social processes of Park and Burgess 25 could be regarded as stages on the way to the reconciliation. As a condition of stability, then, "sufficing justice" must be added to the indices of social normality. Some people's misgivings about functionalism stem from a suspicion that the positive regard it pays to stability implies a condemnation of just revolt. But this is a misreading of functionalism'scase; and it is unfortunate if, because of it, lovers of justice have failed to take functionalists for their friends. Actually, functionalism implies that a system is deficient in respect of stability, if anyone under it withdraws support through being denied justice. 5. The external challenge to social organization is to cope with "present and foreseeable circumstances."Change in their external situation may force a group to "make rearrangements," but a change in their knowledge of the same situation can have the same effect. (A changed view of the justice of existing arrangements is but an instance of the latter.) The ultimate source of the external impulse to change, therefore, is new knowledge-or probably more correct still, novel experience. Washburne26 has proposed the overwhelmingevent as the source of all social change, and thereby simplifies our thinking about it considerably. But there seems no reason why an event should be overwhelming in its proportions in order to force social change, so long as it is overwhelmingin its novelty. Man's intelligence and sensibility impose on him a growing burden of new knowledge; to accommodate, he must change his ways continually. He has no option but to be put at a loss through having one day's practice made unequal to the next day's knowledge. He does have some option, though, in his
24 George Caspar Homans, Social Behavior: Its Elementary Forms, New York: Harcourt, Brace and World, 1961, pp. 378-398. 25 Robert E. Park and Ernest W. Burgess, Introduction to the Science of Sociology, Chicago: Univ. of Chicago Press, 1924. 26 Norman Foster Washburne, Interpreting Social Change in America, New York: Random House, 1954.

11

reaction to this. This state is essentially the one of anomie where procedures and norms for dealing with perceived realities are lacking. Individuals can be torpedoed by this experience or seek recovery: a negative and positive reaction are both possible. One instance of the negative reaction is the anomic suicide which Durkheim 27 identified. Other instances, which do not require going out of life altogether, are the various forms of pseudo-conformitycommonly styled "hypocrisy." Such are the "bohemian" and "philistine" stances described by Thomas and Znaniecki28-and are the "other-directedness" and "tradition-directedness"of Riesman 29 identical with them? But the positive reaction to anomie, whereby recovery is achieved, is one of reorganization. And this is not infrequently (perhaps always) approached by a denouement of collective behaviorand ideology. By collective behavior in some form men make a primitive communication of sharing in a predicament, and then, in resurgent ideology they re-affirm their responsibility to one another as men. This tides them over the break-down in their actual role obligations, and prepares them for a different definition of duties. The resiliency or adaptability we look for in social systems when making functional analysis requires them to meet novelties by absorbing them in this way, rather than be themselves shattered by shock. Once again, it shows misunderstanding when critics of functionalism represent it as opposed to change. It is not opposed to change in social systems but to their death when they have still a work to do. If some novelty should leave men unable to cope completely, that would not be social change but social extinction. It is from knowing that human life is impossible without society and society impossible without adaptive change that the functionalist deprecates this bleak alternative. The writer wonders, though, whether attempts to defend the usefulness of functional analysis for the study of change have sufficientlyexploited the fact that an adapta27 Emile Durkheim, Suicide, edited by George Simpson, Glencoe, Ill.: The Free Press, 1951. 28 William I. Thomas and Florian Znaniecki, The Polish Peasant in Europe and America, New York: Dover Publications, 1958. 29 David Riesman, in collaboration with Reuel Denney and Nathan Glazer, The Lonely Crowd, New Haven: Yale University Press, 1950.

This content downloaded from 143.225.109.4 on Wed, 5 Jun 2013 02:51:01 AM All use subject to JSTOR Terms and Conditions

12

AMERICAN SOCIOLOGICALREVIEW themselves. A third objective social imperative appears in modern sociology, whenever the system to be analyzed has more than one side to it. This is integration. Just as a discourse needs to be coherent and consistent to be convincing, a social system needs to be integrated to be productive. For just as a discourse draws separate facts together, a social system may be compoundedof a number of sets of arrangements. Food production, say, as well as the maintenance of health and the taking of recreation, may all be undertaken by one set of people. These various activities must be so deployed that they do not interfere with one another. If they are all scheduled for one time, for instance, they will. If the product of one is an obstacle to the achievement of another, they will-as when some food which is harmful to health is produced and used. The compatibility of their ends has always ultimately to be faced by a group if they are not to leave themselves open to self-defeat. This forces them to settle on some heirarchy for their preferences, so that they will know which to further when the ends come into competition with one another. It will force them further. They will have to settle whether any of their aims are to be pursued for their own sake and without limit (and perhaps these alone should be called their "values"). All questions of the compatibility of ends are really economic questions, in the most generic sense of that term. For these considerationsare forced on us by scarcity-by how far the material will stretch. Included in "the material" in this case is that of which man himself is made. Being in a state of contradiction, for example, can be intellectually, emotionally or morally insupportable by man because of the way he is made-just as it is impossible for a cake to be both actual and eaten because of the nature of cakes. While all co-ordinating efforts contribute to social integration, legislation and adjudication are two of its most deliberate expressions. But the sanctifying or sacralizing operations of religion probably provide the prototypical instance of it. This is the group's radical approach to integration, in that it is the object here not to deny any normal activity, but to deny autonomy of aim to every activity. All activities, and the

tion requirement implies that readiness to change is itself one of the properties of the system which has to be maintained.30 As well as large-scale changes, changes have constantly to be made within the minutiae of life. Even in those safe places of society where norms are firmly established, the norms rarely fit the case exactly. So it is required that the actors themselves supply an informal, on-the-spot structure like an inner lining to the formal. Both for these minor adaptations and for major ones the recruits to any social system need to come equipped in advance to do both more and different things than the rules prescribe. Here is a critical instance of the past laying its hand on the present. For what re-asserts itself here is nurture, not biological nature. It is the human nature which is fashioned in the primary group which constitutes this regenerative capacity, and if any man has been denied this group he will be denied this nature and capacity also. If ideology is the theory of recovery, the imprint of primary group involvements is the empirical basis of the theory. In the primary group men will have experienced the unrestricted mutuality, the universal love, which acts like embryonic tissue for the secondary structures that can grow anew between them wherever their ways drift into chaos.31 6. If a diversity of arrangementsis made by the one set of individuals, it is required that the arrangements be integrated among
30 Cancian, in replying to critics of functionalism, and to Dahrendorf and Hield especially, comes close to doing so when she states that examining a stable rate of change is one of the four possible ways in which change can be incorporated in functional analysis. See Francesca Cancian, op. cit., pp. 823826; Ralf Dahrendorf, "Out of Utopia," American Journal of Sociology, 64 (September, 1958), pp. 115-127; Wayne Hield, "The Study of Change in Social Science," British Journal of Sociology, 5 (March, 1954), pp. 1-10. But this hardly suffices. It is not a stable rate of change that the normal society must exhibit, but a stable capacity to change at variable rates according to the challenge of the times. 31 Students of disaster draw attention to the way in which primary group solidarity reasserts itself when secondary structures collapse, thus supplying evidence for Cooley's idea that the primary group is the nucleus from which all social organization grows. See Charles E. Fritz, "Disaster," in Robert K. Merton and Robert A. Nisbet, op. cit., p. 689; Charles H. Cooley, Social Organization, New York: Scribner, 1909.

This content downloaded from 143.225.109.4 on Wed, 5 Jun 2013 02:51:01 AM All use subject to JSTOR Terms and Conditions

BUT SOME ARE MORE EQUAL THAN OTHERS society itself, are recalled to the service of an overmastering end, and thereby integrated among themselves. 7. It is because the demand for need-satisfaction through them is unrelenting, that social arrangements must achieve stability, adaptive change and integration. For this reason, making judgments of function or dysfunction, normality or pathology, presupposes a whole catalogue of assumptions about human needs. It was said that part of our judgment that an operation is functional rests on our assumption that the product of the operation is needed by man. This must be one of the main reasons why functional thinking seems unsatisfactory to many people-for who shall say what the things are that man must have under all circumstances?And who shall so disentangle the inherited need from its cultural modification as to evaluate the latter as a means of supplying the former? Yet, unsatisfactory though this admittedly seems, such assumptions must be made if sociological work is to have depth and significance. These assumptions make a back-drop of our own providing to all the empirical observations we make. But the back-drop is not itself non-

13

empirical for all that. It has been woven from the cumulative experience of mankind, as we have each been able to absorb this. We accept as human needs all those satisfactions which men have striven to repeat at many differenttimes and places and by many different means. We assume that men need food, for example, because, under so many varied circumstances, we know they have acted as though they needed it. It is not for any different reason that we assume they need shelter, mutual protection, status, skillfulness, explanations of natural phenomena and the consolations of religion. What any sociologist assumes here should hardly be that minimum range of needs to which all his colleagues will give ready assent, but that whole spectrum to which his own vision has admitted him. It is precisely here that the social scientist is served by his explorations of the arts and humanities, as well as by the diversity of his experience and his range of sympathy and imagination. Perhaps it is by exceptional endowment here that the great sociologist is marked out. If he lends us his eyes for a time we may devise instruments which will compensate other men for their partial blindness.

BUT SOME ARE MORE EQUAL THAN OTHERS *

E.

MOORE t

Princeton University A review of the "functional theory of stratification" indicates some neglect of dysfunctions and other dynamic elements in systems of social inequality, but does not find the critics persuasive on the possibility of eliminating inequality. In behavioral terms inequality is seen as deriving from the probability of differential valuation of performance, qualities (a prospective view of performance) and achievements (a retrospective view of performance). In structural terms these variables are seen as modes of access to positions, but positional differentiation is independently linked to inequality of rewards. Although equity strains inhere in any system of differential rewards, there is no reason to suppose that an equalitarian system would be free from such strains, as fairness and complete equality are scarcely synonymous and equalitarianism as a value is rather restrictive in its context of applicability.
T

HE

theoretical controversy ever the in-

terpretation of social inequality is one of the most enduring disputes in contemporary sociology. Now nearing the end
*The title of course is borrowed from George Orwell'sAnimalFarm, New York: Harcourt,Brace, 1954. t The present essay is partially derived from a joint work with Arnold S. Feldman, Order and

of a second decade since the publication of the essay by Davis and Moore 1 and more
Change in Industrial Societies, now in preparation. I also benefited from critical comments by my collaborator, but accept sole responsibility for the present formulation. 1 Kingsley Davis and Wilbert E. Moore, "Some Principles of Stratification," American Sociological Review, 10 (April, 1945), pp. 242-249.

This content downloaded from 143.225.109.4 on Wed, 5 Jun 2013 02:51:01 AM All use subject to JSTOR Terms and Conditions