Elites and Power - 285 tics: Who Gets What, When, How (New York: McGraw-Hill, 1936); and for a more systematic use, H. D. Lasswell and Abraham Kaplan, Power and Society (New Haven: Yale University Press, 1950). 4. The conception of the elite as members of a top social stratum, is, of course, in line with the prevailing common-sense view of stratification. Technically, it is closer to 'status group' than to 'class,' and has been very well stated by Joseph A. Schumpeter, 'Social Classes in an Ethically Homogeneous Environment,' Imperialism and Social Classes (New York: Augustus M. Kelley, Inc., 1951), pp. 133 ff, especially pp. 137—47. Cf. also his Capitalism, Socialism and Democracy, 3rd ed. (New York: Harper, 1950), Part II. For the distinction between class and status groups, see From Max Weber: Essays in Sociology, trans, and ed. by Gerth and Mills (New York: Oxford University Press, 1946). For an analysis of Pareto's conception of the elite compared with Marx's conception of classes, as well as data on France, see Raymond Aron, 'Social Structure and Ruling Class,' British Journal of Sociology, vol. I, nos. 1 and 2 (1950). 5. The most popular essay in recent years which defines the elite and the mass in terms of a morally evaluated character-type is probably Jose Ortega y Gasset's, The Revolt of the Masses, 1932 (New York: New American Library, Mentor Edition, 1950), esp. pp. 91 ff. 6. As in the case, quite notably, of Gaetano Mosca, The Ruling Class (New York: McGraw-Hill, 1939). For a sharp analysis of Mosca, see Fritz Morstein Marx, 'The Bureaucratic State,' Review of Politics, vol. I, 1939, pp. 457 ff. Cf. also Mills, 'On Intellectual Craftsmanship,' April 1952, mimeographed, Columbia College, February 1955- 7. Cf. Karl Löwith, Meaning in History (Chicago: University of Chicago Press, 1949), pp. 125 ff. for concise and penetrating statements of several leading philosophies of history. 30 I Anthony Giddens Elites and Power It is certainly one of the most characteristic emphases of the Marxian perspective that, in capitalism especially (but also, in a general sense, in the prior types of class system), the realm of the 'political' is subordinate to that of the 'economic'. What remains relatively obscure in Marx is the specific form of this dependence, and how it is expressed concretely in the domination of the ruling class.1 The importance of this is not confined to the analysis of the social structure of capitalism, but bears directly upon the question of the classless character of socialism. It relates, in addition, to issues brought to the forefront by the critique of the Marxian standpoint advanced by the 'elite theorists' of the turn of the century. The substance of this critique, in the writings of such as Pareto and Mosca, may be expressed as an attempt to transmute the Marxian concept of class, as founded in the relations of production, into an essentially political differentiation between those 'who rule' and those who 'are ruled'—a transmutation which was, indeed, in part made possible by Marx's failure to specify in a systematic fashion the modes whereby the economic hegemony of the capitalist class becomes 'translated' into the political domination of the ruling class. For if it is simply the case that economic control directly yields political power, the way is open for Anthony Giddens. "Elites and Power," in The Class Structure of the Advanced Societies, pp. 118—124. Copyright © 1973 by Anthony Giddens. Reprinted by permission of HarperCollins, Inc. 286 - Anthony Giddens the assertion that, in socialism, as in capitalism (indeed as in any other conceivable type of complex society), whoever controls the means of production thereby achieves political domination as a ruling class. The movement of history from capitalism to socialism then becomes conceived of as a mere succession of'ruling classes' ('elites'), as in classical 'elite theory', or more specifically as the emergence of the sort of 'managerial' or 'technocratic' ruling class described in Burnham's writings, and more recently in some of the variants of the theory of the 'technocratic society'.2 The points at issue between the Marxian standpoint and 'elite theory' have become further complicated in recent years by the use of concepts drawn from the latter, such as that of 'power elite', as if they were synonymous with that of'ruling class'. It will be useful to clarify the usage of the terms 'ruling class', 'elite', 'power elite', 'governing class', etc., which involves, in part, looking more closely at the structuration of the upper class. In the analysis which follows, I shall be interested primarily in developing a set of formulations which illuminate significant conceptual distinctions, rather than adhering to conventional terminological usage—if it can be said, in any case, that there is any conventional practice in a field in which there has been so much confusion.3 I shall suggest that, given the distinctions set out below, there can exist a 'governing class' without it necessarily being a 'ruling class'; that there can exist a 'power elite' without there necessarily being either a 'ruling' or a 'governing class'; that there can be what I shall call a system of 'leadership groups' which constitutes neither a 'ruling class', 'governing class', nor 'power elite'; and that all of these social formations are, in principle, compatible with the existence of a society which is 'capitalist' in its organisation. To begin with, a few elementary remarks are necessary about the notion of'elite'. As it is sometimes employed, 'elite' may refer to those who 'lead' in any given category of activity: to actors and sportsmen as well as to political or economic 'leaders'. There is evidently a difference, however, between the first and the second, in that the former 'lead' in terms of some sort of scale of 'fame' or 'achievement', whereas the second usage may be taken to refer to persons who are at the head of a specific social organisation which has an internal authority structure (the state, an economic enterprise, etc.). I shall use the term 'elite group' in this latter sense, to designate those individuals who occupy positions of formal authority at the head of a social organisation or institution; and 'elite' very generally, to refer either to an elite group or cluster of elite groups. In these terms, it can be said that a major aspect of the structuration of the upper class concerns, first, the process of mobility into or recruitment to, elite positions and, second, the degree of social 'solidarity' within and between elite groups. Mediate structuration thus concerns how 'closed' the process of recruitment to elite positions is, in favour of those drawn from propertied backgrounds. Proximate structuration depends primarily upon the frequency and nature of the social contacts between the members of elite groups. These may take various forms, including the formation of marriage connections or the existence of other kin ties, the prevalence of personal ties of acquaintance or friendship, etc. If the extent of social 'integration' of elite groups is high, there is also likely to be a high degree of moral solidarity characterising the elite as a whole and, probably, a low incidence of either latent or manifest conflicts between them. There has never been any elite, however solidary, which has been free of conflicts and struggles; but the degree and intensity of overt conflict has varied widely, and thus it is reasonable to speak broadly of differentials in the solidarity of elite groups. Elites and Power - 287 Open P/u from uniform clilt Social Strat. 2e, p. 213 «lablisht'd ctitc Combining these two aspects of structura-tion, we can establish a typology of elite formations [see diagram on this page]. A 'uniform' elite is one which shares the attributes of having a restricted pattern of recruitment and of forming a relatively tightly knit unity. It hardly needs emphasising that the classifications involved are not of an all-or-nothing character. The point has been made that even among traditional aristocracies there was never a completely 'closed' pattern of recruitment, something which has only been approached by the Indian caste system—all elites open their ranks, in some degree, to individuals from the lower orders, and may enhance their stability thereby. A relatively closed type of recruitment, however, is likely to supply the sort of coherent socialisation process producing a high level of solidarity between (and within) elite groups. But it is quite feasible to envisage the existence of instances which approximate more closely to the case of an 'established' elite, where there is a relatively closed pattern of recruitment, but only a low level of integration between elite groups. A 'solidary' elite, as defined in the classification, might also appear to involve an unlikely combination of elements, since it might seem difficult to attain a high degree of integration among elite groups whose members are drawn from diverse class backgrounds. But, while this type of social formation is probably rare in capitalist societies, at least some of the state socialist countries fit quite neatly into this category: the Communist Party is the main channel of access to elite positions, and while it provides an avenue of mobility for individuals drawn in substantial proportions from quite lowly backgrounds, at the same time it ensures a high degree of solidarity among elite groups. An 'abstract' elite, involving both relatively open recruitment and a low level of elite solidarity, whatever its empirical reality, approximates closely to the picture of certain contemporary capitalist societies as these are portrayed in the writings of the theorists of so-called 'pluralist democracy'. The distinguishing of different types of elite formation does not, in itself, enable us to conceptualise the phenomenon of power. As in the case of class structuration itself, we may distinguish two forms of the mediation of power relationships in society. The first I shall call the institutional mediation of power; the other, the mediation of power in terms of control. By the institutional mediation of power, I mean the general form of state and economy within which elite groups are recruited and structured. This concerns, among other things, the role of property in the overall organisation of economic life, the nature of the legal framework defining economic and political rights and obligations, and the institutional structure of the state itself. The mediation of control refers to the actual (effective) power of policy-formation and decision-making held by the members of particular elite groups: how far, for example, economic leaders are able to influence decisions taken by politicians, etc. To express it another way, we can say that power has two 288 - Anthony Giddens rwLn-&(.refla(.h BroarS Cwivjtidalcil P/u from Social Strat. 2e, p. 214 DlflliSCd pa* aspects: a 'collective' aspect, in the sense that the 'parameters' of any concrete set of power relationships are contingent upon the overall system of organisation of a society; and a 'distributive' aspect, in the sense that certain groups are able to exert their will at the expense of others.4 The mediation of control is thus expressed in terms of 'effective' power, manifest in terms of the capacity either to take or to influence the taking of decisions which affect the interests of two or more parties differentially. We may conceptually separate two variable factors in analysing effective power (that is to say, power as differentiated from 'formal authority') in relation to types of elite formation. The first concerns how far such power is 'consolidated' in the hands of elite groups; the second refers to the 'issue-strength' of the power wielded by those in elite positions. While the former designates limitations upon effective power, deriving from constraints imposed from 'below', the latter concerns how far that power is limited because it can only be exercised in relation to a range of restricted issues. Thus it is often held to be characteristic of modern capitalist societies that there are quite narrowly defined limitations upon the issues over which elite groups are able to exercise control.5 By combining these two aspects of effective power as exercised by elite groups, we can establish a classification of forms of power-structure [see diagram on this page]. Like the previous typology, this sets out an abstract combination of possibilities; it goes almost without saying that this is no more than an elementary categorisation of a very complex set of phenomena, and the labels applied here in no way exhaust the variety of characteristics which are frequently subsumed under these terms. According to these definitions, the consolidation of effective power is greatest where it is not restricted to clearly defined limits in terms of its 'lateral range' (broad 'issue-strength'), and where it is concentrated in the hands of the elite, or an elite group. Power-holding is 'oligarchic' rather than 'autocratic' where the degree of centralisation of power in the hands of elite groups is high, but where the issue-strength of that power is limited. In the case of 'hegemonic' control, those in elite positions wield power which, while it is not clearly defined in scope and limited to a restricted range of issues, is 'shallow'. A 'democratic' order, in these terms, is one in which the effective power of elite groups is limited in both respects. Finally, bringing together both classifications formulated above, we can set up an overall typology of elite formations and power within the class structure [see diagram on the next page]. This makes possible a clarification of the four concepts already mentioned—'ruling class', 'governing class', 'power elite' and 'leadership groups'. It must be emphasised that these partially cross-cut some of the existing usages in the literature on class and elite theory. The Paretian term 'governing class' is here not, as in Pareto's own writing, a replacement for the Marxian 'ruling class'; in this scheme, a governing class is 'one step down', both in terms of elite formation and power-holding, from a 'ruling class'. Elites and Power - 289 I" EUre formal iofi ]'C"^cr-hiLild:irii£ Lea.-!*-rsliLp pro^pl ahslrajcl clilc uniJcwmirtrablkOigd riil*_LLLUj P/u from Social Strat. 2e, p. 215 tmoriiC/dirrjfHj.Titjc itratJC/äligarthic ! h^pnüDLCi'ttmacilic I In this scheme, the 'strongest' case of a ruling class is defined as that where a uniform elite wields 'autocratic' power; the weakest is where an established elite holds 'oligarchic' power. Where a relatively closed recruitment pattern is linked with the prevalence of defined restrictions upon the effective power of elite groups, a governing class exists, but not a ruling class. A governing class borders upon being a ruling class where a uniform elite possesses 'hegemonic' power; and comes closest to being a system of leadership groups where an established elite holds 'democratic' power. Where a governing class involves a combination of an established elite and 'hegemonic' power, it stands close to being a power elite. A power elite is distinguished from a ruling class in terms of pattern of recruitment, as is a governing class from a system of leadership groups. The latter exists where elite groups only hold limited power, and where, in addition, elite recruitment is relatively open in character. In terms of the mediation of control, this classification leaves undefined the relative primacy of the power of any one elite group over others. This can be conceptually expressed as referring to the nature of the hierarchy which exists among elite groups. A hierarchy exists among elite groups in so far as one such group holds power of broader issue-strength than others, and is thereby able to exert a degree of control over decisions taken by those within them. Thus it may be that the economic elite, or certain sectors of the economic elite, are able to significantly condition political decisions through the use of 'influence', 'inducement', or the 'direct' control of political positions—i.e., through the fact that members of the economic elite are also incumbents of political positions. We may refer to all of these modes of obtaining, or striving for, control as the media of interchange between elite groups. It is precisely one of the major tasks of the analysis of elite formations to examine the media of interchange which operate between elite groups in any given society in order to determine what kinds of elite hierarchy exist. NOTES 1. Most subsequent Marxist authors have either been content with the most generalised assertions about this issue, or have wanted to have their cake and eat it by insisting that capitalism is dominated by a ruling class who do not actually 'rule'; cf. Nicos Poulantzas, Pouvoirpolitique et classes sociales de I'etat capitaliste (Paris 1970), pp. 36lff. 2. James Burnham, The Managerial Revolution (New York 1941). 3. In this section of this chapter I have drawn upon part of my article 'Elites in the British class structure', Sociological Review 20, 1972. 4. cf. Talcott Parsons, 'On the concept of political power', Proceedings of the American Philosophical Society 107, 1963. The error in Parsons' analysis, however, is to take insufficient account of the fact that the 'collective' aspect of power is asymmetrical in its consequences for the different groupings in society. 5. As in Keller's 'strategic elites'. See Suzanne Keller, Beyond the Ruling Class (New York 1963).