(1) The two terms used in the title of this paper were chosen in preference to Marthe more conventional terms “society” and “state”, although the latter are used in the text as well. My reason is that “social stratification” emphasizes (as “society” does not) the division of individuals into social ranks which provide the basis of groupformation that is of interest here. The term is used in this very general sense with the understanding that individuals who differ from one another are united into groups by a force that overrides the differences existing between them, as T. H. Marshall put it in his definition of “class”, “Political community” in turn emphasizes the consensus between governors and governed within the framework of a polity while the term “state” puts the emphasis upon the administrative aspect of government, at any rate in English usage. Both aspects must be considered together, but I did not wish to emphasize the latter in the title.

(2) For easy identification it would be desirable to label these three approaches, but it is awkward to do so since every label has misleading connotations. “Society as an object of state-craft” may be considered a Machiavellian approach, but this perspective is also characteristic of overthe social-welfare state which is not “Machiavellian” in the conventional meaning of that term. Government considered as a “product of society” might be called the sociological perspective, but this is also characteristic of Marxism which should not be identified with sociology, and then there are sociologists like Max Weber and Robert MacIver who do not adhere to this view, The theory of a partial dualism between society and government is a characteristic feature of European liberalism, but to call it the “liberal orientation” carries overtones of a specific political theory which need not be associated with this approach, In view of such difficulties I have decided to avoid convenient labels and repeat the three phrases mentioned in the text.

(3) The development suggested here is traced in Meinecke, Friedrich, Die Entstehung des Historismus (München, R. Oldenbourg, 1946)Google Scholar, ch. III. The partly scientific orientation of Machiavelli and Montesquieu is discussed in Olschki, Leonard, Machiavelli the Scientist (Berkeley, The Gillick Press, 1945)Google Scholar and Durkheim, Emile, Montesquieu and Rousseau (Ann Arbor, University of Michigan Press, 1960).CrossRefGoogle Scholar

(4) See Meinecke, Friedrich, Die Idee der Staatsräson (München, R. Oldenbourg, 1925)Google Scholar, passim. The work is available in English translation under the title Machiavellism. The relation between this concern with “reasons of state” and the development of factual knowledge about society is discussed in Heckscher, Eli, Mercantilism (New York, MacMillan Co, 1955), II, pp. 13–30, 269 ffGoogle Scholar. and Small, Albion, Origins of Sociology (Chicago, University of Chicago Press, 1925)Google Scholar. See also Friedrich, C. J., Constitutional Reasons of State (Providence, Brown University Press, 1957).Google Scholar

(5) Quoted in Manuel, Frank E., The New World of Henri de Saint-Simon (Cambridge, Harvard University Press, 1956), p. 135Google Scholar. Professor Manuel shows how this theme recurs in ever new formulations throughout Saint-Simon's writings.

(6) Karl Löwith has shown that this contrast goes back to ancient Greek and Christian ideas and he has traced the development of this theme during the nineteenth century. See his book Von Hegel zu Nietzsche (Zürich, Europa Verlag, 1941), pp. 255–65Google Scholar and ff. See also the judicious restatement of Rousseau's position in Palmer, R. R., The Age of the Democratic Revolution (Princeton, Princeton University Press, 1959), pp. 119–127.Google Scholar

(7) Durkheim, Emile, Montesquieu and Rousseau, pp. 65, 137Google Scholar, and passim.

(8) See Durkheim, Emile, The Division of Labor in Society (Glencoe, The Free Press, 1947), pp. 200–206Google Scholar and passim.

(9) Durkheim, Emile, Professional Ethics and Civic Morals (Glencoe, The Free Press, 1958), p. 61Google Scholar. Published for the first time in 1950 in a Turkish edition of the French manuscript, these lectures were delivered by Durkheim in 1898, 1899 and 1900 at Bordeaux and in 1904 and 1912 at the Sorbonne. As will be shown below, these lectures contain Durkheim's political theory and their repeated delivery together with the well-known preface to the second edition of The Division of Labor (published in 1902) indicate that for Durkheim this aspect of his work was of great importance. It is a symptom of the “sociologizing” tendency of our own time that this political aspect has been neglected or ignored by most scholars who have been influenced by Durkheim's sociological theories.

(10) I take this telling phrase from Marcel Mauss's introduction to Durkheim, Emile, Socialism and Saint-Simon (Yellow Springs, The Antioch Press, 1958), p. 2Google Scholar. Durkheim's elaboration of his views on the corporate society may be found in the second preface to his The Division of Labor, pp. 1–31Google Scholar. Cf. especially the following summary statement of his position: “A society composed of an infinite number of unorganized individuals, that a hypertrophied state is forced to oppress and contain, constitutes a veritable sociological monstrosity […] Where the State is the only environment in which men can live communal lives, they inevitably lose contact, become detached, and thus society disintegrates. A nation can be maintained only if, between the State and the individual, there is intercalated a whole series of secondary groups near enough to the individual to attract them strongly in their sphere of action […]” ibid., p. 28. In the preceding account I have only restated in the briefest compass the familiar themes of Durkheim's work, The best analytical exposition of these themes is contained in Parsons, Talcott, The Structure of Social Action (Glencoe, The Free Press, 1949), ch. VIII–XI, though this statement was written before Durkheim's unpublished lectures on the state became available.Google Scholar

(11) Durkheim, Emile, The Division of Labor, pp. 283 ff.Google Scholar

(12) Ibid., pp. 27–28, 218–219.

(13) Durkheim, Emile, Professional Ethics and Civic Morals, pp. 61–63. My italics.Google Scholar

(14) Ibid., p. 45. This formulation is indebted to Montesquieu and Tocqueville.

(15) Cf. the reference to this paradox in Benoit-Smullyan, E., The Sociologism of Emile Durkheim and His School, in Barnes, H. E., ed., An Introduction to the History of Sociology (Chicago, University of Chicago Press, 1948), pp. 518–520.Google Scholar

(16) See Division of Labor, pp. 386–388.Google Scholar

(17) Halévy, Elie, The Growth of Philosophical Radicalism (London, Faber and Faber, 1928), pp. 90–91, 118–120, 489–491Google Scholar and passim.

(18) The following statement relies on the work of Otto Hintze, Weltgeschichtliche Bedingungen der Repräsentativverfassung, Historische Zeitschrift, CXLIII (1930), pp. 1–47Google Scholar and by the same author, Typologie der ständischen Verfassungen, Historische Zeitschrift, CXLI (1929), pp. 229–248Google Scholar. Hintze's contributions are corroborated and extended in Dietrich Gerhard's Regionalismus und Ständisches Wesen als ein Grundthema Europäischer Geschichte, Historische Zeitschrift, LCXXIV (1952), pp. 307–337.Google Scholar

(19) This characteristic feature of medieval political life will be contrasted below with the problematic relation between social stratification and the political community in modern Western societies.

(20) The quotation marks refer to the ineradicable ambiguity of this term in medieval society. The “people” were objects of government who took no part in political life. Yet kings and estates frequently couched their rivalries in terms of some reference to the “people” they claimed to represent. In fact, “consent of the people” referred to the secular and clerical notables whose voice was heard in the councils of government. See the discussion of this issue in Gierke, Otto, Political Theories of the Middle Ages (Boston, Beacon Press, 1958), pp. 37–61Google Scholar. It may be added that this ambiguity is not confined to the Middle Ages, since all government is based in some degree on popular consent and since even in the most democratic form of government the “people” are excluded from political life in greater or lesser degree. These differences of degree, as well as the qualities of consent and participation are all-important, of course, even though it may be impossible to do more than formulate proximate typologies.

(21) Cf. Weber, Max, Law in Economy and Society (Cambridge, Harvard University Press, 1954), ch. v and passim.Google Scholar

(22) In his analysis of traditional domination Max Weber distinguished patrimonial from feudal administration, i.e., the effort of rulers to extend their authority and retain control by the use of “household officials” or by their “fealty-relationship” with aristocratic notables of independent means. These two devices are by no means mutually exclusive, since “household officials” were usually of noble birth and in territories of any size demanded autonomy, while “feudal” notables despite their independence frequently depended upon the ruler for services of various kinds. Contractual obligations as well as elaborate ideologies buttressed the various methods of rule under these complementary systems, For an exposition of Weber's approach cf. Bendix, R., Max Weber, An Intellectual Portrait (Garden City, Doubleday and Co., 1960), pp. 334–79Google Scholar, which is based on Weber, , Wirtschaft und Gesellschaft (Tübingen, J.C.B. Mohr, 1925), II, pp. 679–752Google Scholar. These sections are not available in translation.

(23) A systematic analysis of this role of the church is contained in Weber, Max, op. cit., II, pp. 779–817Google Scholar. A brief résumé of this chapter is contained in Bendix, Reinhard, Max Weber, pp. 320–326Google Scholar. For a detailed historical treatment of the consecration of secular rule cf. Kantorowicz, Ernst, The King's Two Bodies (Princeton, Princeton University Press, 1957).Google Scholar

(24) Weber, , Law in Economy and Society, p. 143Google Scholar. In this connection it should be remembered that the privileges or poliliberties of medieval society were associated with duties that would appear very onerous to a modern citizen. Also, these individual or collective “privileges” frequently resulted from compulsion rather than a spontaneous drive for freedom, as is vividly described in White, Albert B., Self-government at the King's Command (Minneapolis, University of Minnesota Press, 1933)Google Scholar. The title itself illuminates the combination of royal power and compulsory local autonomy, which was typical of England, but not found to the same extent elsewhere in Europe. Still, the privileges of an estate also had the more ordinary meaning of rights (rather than duties), and this was true to some extent even of the lower social orders. Cf. the discussion of this problem by Herbert Grundmann, Freiheit als religiöses, politisches und persönliches Postulat im Mittelalter, Historische Zeitschrift, CLXXXIII (1957), PP. 23–53Google Scholar. A detailed case-study of medieval political life is contained in Brunner, Otto, Land und Herrschaft (Brünn, Rudolf M. Rohrer Verlag, 1943).Google Scholar

(25) Cf. note 22 above for a reference to Weber's distinction between feudalism and patrimonialism as the two aspects of “traditional domination” which were present throughout the European Middle Ages, The development towards absolutist regimes is best seen, therefore, as a relative shift of emphasis in Western European institutions, which varied from country to country.

(26) For a comparative account of this political structure in eighteenth century Europe cf. Palmer, , The Age of the Democratic RevolutionGoogle Scholar, ch. III and passim.

(27) De Tocqueville, Alexis, The Old Regime and the Revolution (Garden City, Doubleday and Co., 1955), pp. 22–77Google Scholar. For a modern appraisal of the survival of corporate and libertarian elements under the absolutist regimes of the eighteenth century cf. Kurt von Raumer, Absoluter Staat, Korporative Libertät, Freiheit, Persönliche, Historische Zeitschrift, CLXXXIII (1957), pp. 55–96.Google Scholar

(28) de Tocqueville, Alexis, op. cit., p. 137.Google Scholar

(29) Ibid., pp. 15–16.

(30) Tocqueville, , Democracy in America (New York, Vintage Books, 1945), II, p. 311Google Scholar. In advancing this thesis Toequeville referred, for example, to the innovative activities of manufacturers that were characteristic of democratic eras. Such men engaged in “novel undertakings without shackling themselves to their fellows”, they opposed in principle all governmental interference with such private concerns, and yet “by an exception of that rule” each of them sought public assistance in his private endeavor when it suited his purpose. Tocqueville concluded that the power of government would of necessity grow, wherever large numbers of mutually independent men proceeded in this manner, See ibid., p. 311, n. 1.

(31) Ibid., p. 333.

(32) Cf. ibid., pp. 114–132.

(33) Durkheim, Emile, Professional Ethics and Civic Morals, pp. 64–65.Google Scholar

(34) A fuller critical appraisal of Tocqueville's facts and interpretations is contained in the essay by Pierson, George W., Tocqueville in America (Garden City, Anchor Books, Doubleday and Co., 1959), pp. 430–477Google Scholar, though Pierson slights Tocqueville's theoretical contribution which is emphasized in the text.

(35) A further theoretical note is in order here. No one doubts the relevance of the distinction between a feudal order and an equalitarian social structure, which Tocqueville analyzed. In any study of social change we require some such long-run distinction so that we can know whence we came and where we may be going, though distinctions of this kind may be tools of very unequal intellectual worth, But while it is the merit of long-run distinctions that they enable us to conceptualize theoretically significant dimensions of social life (within the same civilization over time or between different civilizations), it also follows that these distinctions will become blurred the more closely we examine social change in a particular setting and in the short-run. The following discussion will suggest some concepts that are designed to “narrow the gap” between the long and the short run and hence reduce to some extent the reliance on deductions which characterized Tocqueville's work. But I doubt that the gap can be closed entirely, because in the short-run we are bound to fall back upon Tocqueville's method of logically deduced possibilities of social change, even if we can go farther than he did in comparing actual changes with these artificial benchmarks. Two rules of thumb should be kept in mind, however. One is that this partly inductive and partly deductive study of social change in the short-run should not lose sight of the long-run distinctions, for without them we are like sailors without compass or stars. The others is that this retention of the long-run distinctions imparts a dialectical quality to the analysis of short-run changes. Since we do not know where these changes may lead in the long-run we must keep the possibility of alternative developments conceptually open and we can do this by utilizing the dichotomous concepts so characteristic of sociological theory. For suggestions along these lines cf. Reinhard Bendix and Bennett Berger, Images of Society and Problems of Concept-Formation in Sociology, in Gross, Llewellyn, ed., Symposium on Sociological Theory, pp. 92–118Google Scholar. This perspective is greatly indebted, of course, to the work of Max Weber.

(36) Max Weber's well known concept of “bureaucracy” is based on the assumption that this process of separation of modern from patrimonial administration has been completed. See his Essays in Sociology (tr. and ed. by Gerth, H. H. and Mills, C. W.; New York, Oxford University Press, 1946), ch. VIIIGoogle Scholar. For an exposition of the contrast between patrimonial and bureaucratic administration see Bendix, R., Max Weber, pp. 419–420Google Scholar. An admirably clear, comparative study of administrative history, in which this process of separation is traced since the middle of the seventeenth century, is contained in Barker, Ernest, The Development of Public Services in Western Europe, 1660–1930 (New York, Oxford University Press, 1944).Google Scholar

(37) Admittedly, these matters are in flux and in this respect significant differences exist within Western civilization. Still, no one can be in doubt in the instances in which this fundamental assumption has come into question, as in the American Civil War or more recently in the critical conflict between the national government in France and the French settlers in Algeria. The Southern opposition to school-integration is not a comparable development, I believe, since even in the more extreme cases it is combined with an acceptance of national jurisdiction on which there is no sharp disagreement. But the political reintegration of the South has been and may continue to be painfully slow, because the American political structure appears to militate against the type of statesman-ship that combines principled firmess with tactical flexibility.

(38) Neither medieval political life nor the absolutist regimes of the eighteenth century nor yet many of the “developing areas” of the modern world knew or know a government of this type, because adjudication and administration were and are decentralized, personal, intermittent, and subject to a fee for each governmental service.

(39) Durkheim, E., Professional Ethics and Civic Morals, pp. 64, 82.Google Scholar

(40) For a survey of this line of thought, disguised as it is in theoretical disquisitions, cf. Nisbet, Robert A., The Quest for Community (New York, Oxford University Press, 1953).Google Scholar

(41) Tocqueville tended to obscure this distinction by identifying this reciprocity posiin the earlier estate societies of medieval Europe with the later symbiosis of absolutist rule and aristocratic privilege, though he was quick to point out how absolutism tended to undermine the aristocratic position. On the increase of aristocratic privileges just prior to the French Revolution cf. Palmer, , op. cit., ch. II–IV.Google Scholar

(42) Max Weber has characterized the contrast as follows:

“In the legal systems of the older type all law appeared as the privilege of particular individuals or objects or of particular constellations of individuals or objects-Such a point of view had, of course, to be opposed by that in which the state appears as the all embracing coercive institution […] The revolutionary period of the 18th century produced a type of legislation which sought to extirpate every form of associational autonomy and legal particularism […] This […] was effected by two arrangements: the first is the formal, universally accessible, closely limited, and legally regulated autonomy of association which may be created by anyone wishing to do so; the other consists in the grant to everyone of the power to create law of his own by means of engaging in private legal transactions of certain kinds.” Weber, Max, Law pp. 145–146.Google Scholar

(43) Demands for representation are difficult to distinguish from demands for privileged jurisdictions or outright benefits, because representation in decision-making bodies may be used to obtain these privileges privileges or benefits. It is clear at any rate that voluntary associations are not the unequivocal counter-weight to centralized power for which Tocqueville was searching in his study of American society. Instead, voluntary associations frequently demand governmental assistance even where they reject it in principle, and in this respect they act in much the same way as individual manufacturers tended to do. a century ago according to Tocqueville's observations. Voluntary associations are a protean phenomenon. They are evidence of consensus within the society, especially where they pursue private ends as an alternative to governmental assistance and regulation. But they may also be evidence of dissensus within the national political community, in so far as they enlist the national government in the service of parochial interests, and hence seek to secure from the government privileges that are denied to other groups.

(44) Incidentally this hiatus is reflected in the very widespread and sanguine juxtaposition of patriotism with the most extreme selfishness of individuals and groups.

(45) The term “state” is needed to designate the continuing political identity of the nation irrespective of the governments embodying this identity from time to time. Where monarchical institutions have survived they represent this identity separately from the ruling government. Such institutional separation is not possible under democracies. In this discussion the terms “state” and “political community” or “polity” are used interchangeably, since all three refer with different emphasis to the apparatus and the consensus sustaining the continuous political identity of the modern nation.

(46) Like all such distinctions there is a good bit of overlap between the two types. Affinities of interest which arise from the social structure forever engender, relations of super- and sub-ordination, while the exercise of instituted authority forever produces, and is affected by, affinities of interest.

(47) Hence, the ideas that society is an object of state-craft or that all governmental institutions are the product of social forces represent perspectives which are useful only as long as their partiality is recognized.

In a recent article Raymond Aron stated the case against the Comtean as well as Marxian tendency to reduce all politics and government to forces arising from the socio-economic sub-structure. «Contre l'un et l'autre, nous avons appris que la politique est une catégorie éternelle de l'existence humaine, un secteur permanent de toute société. Il est illégitime de se donner, par hypothèse, l'élimination de la politique en tant que telle ou de caractériser une société par sa seule infra-structure.» See Aron, Raymond, Les sociologues et les institutions représentatives, Archives européennes de sociologie, I (1960), p. 155Google Scholar. The present analysis is in agreement with this position and I assume that Professor Aron would agree that the sociological level of analysis likewise possesses a certain autonomy. Perhaps it is symptomatic for the modern climate of opinion in the social sciences that the most elaborate systematization of social theory to date acknowledges society, culture and personality, but not politics, as relatively autonomous levels of analysis. Cf. Parsons, Talcott, Shils, E. A. et al. , Toward a General Theory of Action (Cambridge, Harvard University Press, 1951), pp. 28–29CrossRefGoogle Scholar. Cf. also the learned critique of this reductionist tendency in Wolin, Sheldon S., Politics and Vision (Boston, Little, Brown & Co., 1960), chs. ix–x.Google Scholar

(48) As I see it, this is the viewpoint from which Max Weber developed the analytical framework of his posthumously published work Wirtschajt und Gesellschaft. The fundamental distinction of that work is not the one between “economy” and “society”, but between society and domination and hence between groups arising from the pursuit of “ideal and material interests”, on the one hand, and relations of superand sub-ordination arising from beliefs in legitimacy, administrative organization, and the threat of force. For details of this interpretation cf. my book Max Weber, passim. A lucid exposition of the fundamental assumptions of this approach is contained in MacIver, Robert, The Web of Government (New York, Macmillan and Co., 1947), ch. XIII.Google Scholar

(49) Machiavelli, Niccolò, The Prince and the Discourses (New York, The Modern Library, 1940), p. 15.Google Scholar

(50) See Lipset, S. M., Party Systems and the Representation of Social Groups, Archives européennes de sociologie, I (1960), p. 51Google Scholar. In this article Professor Lipset presents comparative materials on the interrelation of different representative systems with different social structures.

(51) Cf. Selznick, Philip, TVA and the Grass Roots (Berkeley, University of California Press, 1949)Google Scholar, which may be considered a case-study of this “reverse inter-comchange”.

(52) Cf. the analysis by Hobsbawm, E. J., Social Bandits and Primitive Rebels (Glencoe, The Free Press, 1959), ch. IIIGoogle Scholar. See also Roger Vailland's novel The Law which illustrates the anarchical propensities through which either formal compliance with, or the symbolic re-enactment of, the law is used to subvert all “rule-abiding behavior”.

(53) There is, thus, a close relationship between this gradual establishment of “social rights” and the decline of ideology, although it must be kept in mind that this decline in the West may be the consequence of the Cold War and the rise of ideology in the rest of the world as much as it is the result of the welfare-state. Different aspects of this complex phenomenon are discussed in Marshall, T. H., Citizenship and Social Class (Cambridge, At the University Press, 1950), ch. IGoogle Scholar; Brunner, Otto, Das Zeitalter der Ideologien, Neue Wege der Sozialgeschichte (Göttingen, Vandenhoeck and Ruprecht, 1956), ch. IXGoogle Scholar; Shils, E. A., Ideology and Civility, Sewanee Review, LXVI (1958), pp. 450–480Google Scholar; Bell, Daniel, The End of Ideology (Glencoe, The Free Press, 1960), Part IIIGoogle Scholar; and Lipset, S. M., Political Man (New York, Doubleday and Co., 1960), pp. 403–417.Google Scholar

(54) For a comparative analysis of the cleavages facing the “new nations” and the related liabilities of government cf. Shils, E. A., Political Development in the New States, Comparative Studies in Society and History, II (1960), pp. 268–282, 379–411.Google Scholar

(55) Almond, Gabriel, A Functional Approach to Comparative Politics, in Almond, Gabriel and Coleman, James S., eds., The Politics of the Developing Areas (Princeton, Princeton University Press, 1960), p. 4.Google Scholar

(56) The concept “state” is discarded on the curious ground that it is based on a dichotomy which is incompatible with the existing continuity of the phenomena. Professor Almond suggests that with reference to the “developing areas” only the political “input” functions will be analysed because the formal governmental structure (“output” function) is usually not well developed. This decision would seem to reintroduce the distinction which was discarded. Cf. ibid., pp. 12, 17.