however, that "we cannot give an account of a social relation without giving a somewhat complete account of the reaction evoked by the social contact" (p. 335). Exactly. This is just what the terms competition, conflict, accommodation, and assimilation are intended to make possible-the giving of somewhat complete accounts of interactions set up by social contacts. The actual social reality as we experience it is a great deal more complicated than it would appear to be if an account of it were to be reduced forthwith to expression in the terms which Hayes proposes. What is urgently demanded at the present stage of development of sociological method is a relatively small number of synthetic concepts, with reference to which as headings or guiding lines particular types of social phenomena can be studied and can be described in their fulness. By this method, opportunity is given for the identification of particular mechanisms of causal interaction which appear to be elemental; while at the same time we are protected against losing sight of important, if not obvious, aspects of the social reality. The actual reality is, as Professor Small has pointed out,13 always one of multiple causation; and if we proceed too hastily to set forth a description of it in terms of a selected list of presumably ultimate types of causal relationship, we are almost certain to lose sight of some features of the complex reality. To be sure, we can make out a number of seemingly elemental aspects of social interaction by a common-sense interpretation of the familiar features of our experience, aided, properly enough, by the light which previous workers have been able to shed upon the matters in question. Terms for these obvious features of the social reality can be related to the comprehensive descriptions which may be written by starting from the concepts we have been discussing. A number of Hayes's terms seem to be of this sort. Although he has not indicated very clearly what he means by them, we may venture to classify some of them as special elements, or components, to be brought under the more inclusive terms, conflict, accommodation, and assimilation.

13

"Emulation" can easily be taken as a special form of conflict, one which is under control by a pre-existent social order. "Dom18 General Sociology, p. 52.

inance-subordination" is one element or phase, perhaps the principal one, of the process of accommodation. "Organization" is, as we have seen, the substantive term under which the results of the process of accommodation are subsumed. "Co-operation" is a term that is perhaps best reserved for the designation of an aspect of the social reality from the economic point of view, as we have shown above.

Assimilation, Professor Hayes says, is a term which designates the result of a social process or of a combination of processes. The distinction which he endeavors to make here is the crux of the whole question from one point of view, as we shall point out presently. Here we may simply note that the term assimilation, because of the language habits which it embodies, can be taken either as an activity or as the result of an activity. When taken in the active sense, social suggestion, imitation, and sympathetic radiation appear to indicate psychological mechanisms involved in all forms of interaction. The establishment and maintenance of personal relations, however, which we have briefly described above as the essence of assimilation, remain to be more thoroughly described, analyzed, and explained through further research work with concrete cases. This process cannot be accounted for, except to a very limited extent, by reference to suggestion, imitation, and radiation. These, in fact, are themselves terms which appear to have the character of hypostatization of certain supposed results of social interaction, the concepts so derived being then cited as causes, after the manner which Comte called metaphysical.

Several of the terms which Hayes proposes appear to be alternative names for some of the more general aspects of the social reality, or results of the social process, taken substantively. This we have already noted as the most plausible interpretation of his term "organization." Probably the same can be said of "inducement" and "deterrence." When he says (p. 334) that all social control which is not a matter of education in the broadest sense (assimilation?) is a matter of inducement and deterrence, is not this equivalent to the assertion that all social processes tend to promote control? That is, they make it possible for the larger group,

in which divergent elements are incorporated, to act for some purposes as a unity. "Social control" is coming into general acceptance among sociologists as the term under which we subsume and study the general fact that by various means individuals are induced to undergo mutual subordination and to carry on some sort of corporate existence and activity which we refer to as "society" or as "groups." Hayes's term "corroboration" is not sufficiently defined or described in his article to make it possible to deal with it intelligently. Apparently, it indicates simply one of the most universal aspects of social interaction or communication.14

SOCIAL RELATIONS AND SOCIAL INTERACTION

Professor Hayes does not seem to be clear as to the logical connection and distinctions between relations and interaction. Each of his terms, he says, indicates a "type of social relationship" (p. 341). While "the distinction between activities and the relations in which they stand is readily confused" (p. 342), there appears to be no doubt in his mind that it is a basic distinction. Is this, however, anything more than the distinction between the substantive or static abstraction of the reality of immediate experience, and the abstractions of the same reality which we make in terms of function, process, or activity? There can be "relation between activities" only if the persons or other elements which are thought of as the actors are interacting. The reality given in immediate experience is not, strictly speaking, capable of being reproduced by any written description whatever.15 We are able to deal with this concrete reality, however, in a more or less sophisticated and purposive way, by categorizing it, that is by subjecting it to a procedure of conscious or unconscious abstraction. Now, as Bergson and others have shown, the most natural kind of abstractions which we make from the reality of experience are probably those in which we reduce the reality to a substantive form; we neglect the activity which is experienced in time, and conceive the reality as fixed in

"John Dewey, Democracy and Education, pp. 5-6; quoted by Park and Burgess, op. cit., pp. 185–86.

15 Henri Bergson, Matter and Memory (London and New York, 1912), p. 30. See also the same author's Time and Free Will, passim.

space. The time-activity aspect of things is, for reasons inherent in the nature of our mental endowment and in the nature of language, more difficult to retain in an abstract account. If our abstraction is to preserve the active aspect of the experienced datum, it must apparently tend to assume one of two forms: that of an ongoing change conceived in time, or that of a process of interaction between factors conceived as fixed at points in space. Psychology makes use of techniques which need not take account of space, though some types of psychological analysis lean heavily upon spatial representations of their data; it appears to be of the very essence of scientific sociological method, however, that it must account for the intercausal connections of the activities of individuals who are spatially external to one another, and who can treat one another as external obstacles and utilities.18 In the effort to take account of this aspect of its data, sociology naturally develops a methodology which is expressed in terms of interaction. There are reasons probably more fundamental than this, however, why it is essential to sociological method to conceive its data in terms of a process of interaction between forces thought of as exerted at points in space, rather than as a process of ongoing change in time, though the actual reality we are seeking to deal with is both at the same time. For natural science demands abstractions which are universal and transferable, and this they can be only if they can be taken out of time. So long as our conception of the datum with which we are dealing is left in the time nexus, it is unique and nontransferable; whereas abstractions of the data which are conceived in space can be thought of as subject to being removable to other locations. This means that for the purposes of natural science, we reduce our abstractions to the spatial-interaction form, if we wish to save the activity feature of our conception of the inter-causal connection of the phenomena. It may be pointed out incidentally that this is the essence of the difference between the Comtean sociology and that of Herbert Spencer; Comte never managed to think of the social reality except as an ongoing process of change

16

Compare Georg Simmel, Soziologie (München and Leipzig, 1922), chap. ix; see also Nicholas J. Spykman, The Social Theory of Georg Simmel, chap. iv.

in time, while Spencer treated the "factors of social phenomena❞ as if they were distributed in space, and the social activity, virtually, as a process of interaction.1

Simmel has formulated the clearest expression of the social reality from this point of view that we have so far; one concise version of his account is given in the following passage:

Society exists wherever several individuals are in reciprocal relationship. This reciprocity arises always from specific impulses by virtue of specific purposes. . . . . Impulses. . . . bring it to pass that men enter into group relationships of acting for, with, against one another; that is men exercise an influence upon the conditions of association and are influenced by them. These reactions signify that out of the individual bearers of those occasioning impulses and purposes a unity, that is, a "society," comes into being. An organic body is a unity because its organs are in a relationship of more intimate interchange of their energies than with any external being. A state is one because reciprocal influences exist. We could not, indeed, call the world one if each of its parts did not somehow influence every other, if anywhere the reciprocity of the influences, however mediated, were cut off. ... 18

Hayes holds that the term "social process" should be restricted to activity and change in activity (pp. 341-42). It is the thesis of this paper that, while the social reality as immediately experienced is indeed in continual change, for the purposes of sociological method, the term "social process" should be taken to refer to the interaction of elements, factors, or forces, which are conceived from logical necessity as located at points in space. Even when we are clearly conscious that the spatial representations which we may make of the matter in hand are pure fictions, we still find them useful for methodological reasons, and therefore justifiable. The changes which take place in the course of the actual ongoing of the social reality can be most conveniently dealt with abstractly as types of changes which arise in connection with, and in the operation of types of interaction. Hayes states that "all explanation that can be called scientific is in terms of conditioning relations." The foregoing paragraphs have sought to show that "conditioning relations" is a conception which may be taken to refer to relations of Compare Spykman, op. cit., pp. 3-20.

17

18

Soziologie, pp. 4-5; translated by A. W. Small, American Journal of Sociology, XV, 296.