(1) See Schumpeter, J., Capitalism, Socialism and Democracy (New York, Harper and Brothers, 1947), p. 269Google Scholar; Weber, Max, Essays in Sociology (New York, Oxford University Press, 1946), p. 226Google Scholar; see also the article by Plamenatz, J.in McKeon, R., ed., Democracy in a Word of Tensions (Chicago, University of Chicago Press, 1951), pp. 302–327Google Scholar; Lipset, S. M., Political Man (New York, Doubleday, 1960), pp. 45–46.Google Scholar

(2) Parsons, Talcott, Voting and the Equilibrium of the American Political System, in Burdick, E. and Brodbeck, A., eds., American Voting Behavior (Glencoe, The Free Press, 1959), pp. 80–120.Google Scholar

(3) For a comprehensive summary of definitions, see Fairlie, John A., The Nature of Political Representation, The American Political Science Review, XXXIV (1940), pp. 236–48, 456–66.CrossRefGoogle Scholar

(4) For this distinction, as well as the conceptual inspiration of this discussion, see Easton, David, An Approach to the Analysis of Political Systems, World Politics, IX (1957), 383–400.CrossRefGoogle Scholar

(5) They, of course, are far from being the only important such bases. Others such as religious, ethnic or linguistic groups, regions, and rural-urban groupings, have formed the basis for separate parties or differential backings for particular parties. For a detailed discussion of the way in which different groups have varied in support for parties in different democratic countries, see my Political Man, op. cit., pp. 220–282.

(6) Duverger, Maurice, Political Parties (London, Methuen and C°, 1954), p. 204Google Scholar. Duverger's book, however, is the best recent effort to demonstrate the causal effect of electoral systems. A sophisticated critique of Duverger and the general emphasis on electoral systems may be found in Lavau, G. E., Partis politiques et réalités sociales (Paris, Armand Colin, 1953)Google Scholar. There is a comprehensive statement of all the arguments for and against proportional representation in De Grazia, Alfred, Public and Republic (New York, Alfred Knopf, 1951)Google Scholar. A good general discussion of the theory of electoral systems is Hogan, D., Election and Representation (Cork, Cork University Press, 1945).Google Scholar

(7) See Political Man, op. cit., pp. 45–96, and passim.

(8) I am obligated to Carl Friedrich for this definition.

(9) Weber, , op, cit., pp. 263–264.Google Scholar

(10) Bagehot, W., The English Constitution (London, Oxford University Press, 1958), pp. 232–235Google Scholar(emphasisin the original).

(11) See my Political Man, op. cit., pp. 77–79. It should also be noted that Switzerland never had a monarch to over-throw but rather is a historic confederation of small states.

(12) Schumpeter, , Op. cit., pp. 134–139.Google Scholar

(13) Bryce, James, Modern Democracies (New York, Macmillan, 1921), vol. I, pp. 30–32Google Scholar. Briggs, A., Victorian People (London, Odham's Press Ltd., 1954), pp. 153–154Google Scholar. He quotes Matthew Arnold: “It is only in England that this beneficial salutary intermixture of classes takes place. Look at the bottlemerchant's son, and the Plantagenet being brought up side by side […]. Very likely young Bottles will end up by being a lord himself”.

(14) Neumann, Sigmund, Die deutschen Parteien: Wesen und Wandel nach dem Kriege (Berlin, Junker und Dunnhaupt, 1932)Google Scholar; Geiger, Theodor, Die soziale Schichtung des deutschen Volkes (Stuttgart, Ferdinand Enke, 1932), p. 79.Google Scholar

(15) See Michels, Robert, Die deutschen Sozialdemokratie, I: Parteimitgliedschaft und soziale Zusammensetzung, Archiv für Sozialwissenschaft und Sozialpolitik, XXVI (1906), pp. 512–513Google Scholar; Lowie, Robert, Toward Understanding Germany (Chicago, University of Chicago Press, 1954), p. 138Google Scholar. For detailed evidence that the contemporary German Social-Democratic party remains weak among the lower-class workers and relatively strong among the more skilled, a pattern which remains almost unique among left parties, see Political Man, op. cit., pp. 240–241.

(16) Michels, Robert, Sozialismus und Fascismus I (Karlsruhe, G. Braun, 1925), pp. 78–79.Google Scholar

(17) It is interesting to note that in 1918 Max Weber believed that there were four parties “structurally inherent within German Society: a conservative, a democratic [liberal], a socialist, and a Catholic party”. MayerJ, P. J, P., Max Weber and German Politics (London, Faber and Faber Ltd., 1956), p. 101.Google Scholar

(18) See Kirchheimer, Otto, The Political Scene in West Germany, World Politics, IX (1957), pp. 433–445CrossRefGoogle Scholar. and the sources cited there, esp. studies by K. Schutz and G. Schulz in Lakge, M. G. et al. , Parteien in der Bundesrepublik: Studien zur Entwicklung bis zur Bundes-tagswahl 1953 [Schriften des Instituts für politische Wissenschaft, Bd. VI] (Stuttgart-Düsseldorf, Ring Verlag, 1955)Google Scholar; Stern-Berger, D., Lebende Verfassung: Studien über Koalition und Opposition [Parteien-Frahtionen-Regierung, Bd. I] (Meisenheim, Anton Hain, 1956), p. 86Google Scholar; on the change in bureaucratic attitudes, see Bracher, K. D., Die Auflösung der Weimarer Republik [Schriften des Instituts für politische Wissenschaft, Bd. IV] (Stuttgart-Düsseldorf, Ring Verlag, 1955), esp. pp. 174 sq.Google Scholar and Th. Eschenburg, , Stoat und Gesellschaft in Deutschland (Stuttgart, Schwab, 1956), p. 757.Google Scholar

(19) For a detailed discussion of many of these elements see Landes, David, Observations on France: Economy, Society, and Polity, World Politics, IX (1957), 329–350.CrossRefGoogle Scholar

(20) See Landes, David S., “French Business and the Businessman: A Social and Cultural Analysis”Google Scholar, in Earle, Edward, Modern France (Princeton, Princeton University Press, 1951), pp. 334–353Google Scholar; and John E. Sawyer, Strains in the Social Structure, ibid., pp. 293–312.

(21) For a discussion of the strains of industrialization and the ways in which varying patterns of industrialization have affected workers' political behavior in different countries, see Political Man, op. cit., pp. 48–72.

(22) The argument and best evidence for this thesis may be found in Schatt-Schneider, E. E., Party Government (New York, Rinehart and Co., 1942), esp. pp. 65–98.Google Scholar

(23) Weinstein, James, Anti-war Sentiment and the Socialist Party, 1917–1918, Political Science Quarterly, LXXIV (1959) 223–239Google Scholar; Douglas, Paul H., The Socialist Vote in the 1917 Municipal Elections, National Municipal Review, VII (1918), 131–139.CrossRefGoogle Scholar

(24) See Duverger, Maurice, Political Parties, op. cit., pp. 224–228, 246–250Google Scholar; Schattschneider, E. E., Op. cit., pp. 80–84.Google Scholar

(25) “British politics are almost wholly innocent of those issues which cross the social lines in other lands, for example, race, nationality, religion, town and country interests, regional interest, or the conflict between authoritarian and parliamentary methods”. See Bonham, John, The Middle Class Vote (London, Faber and Faber, 1954), pp. 194–195Google Scholar. Lipson, Leslie, The Two-Party System in British Politics, American Political Science Review, XLVII (1953). pp. 337–358.CrossRefGoogle Scholar

(26) For a detailed report on the issues and facts involved, see Butler, D. E., The Electoral System in Britain 1918–1951 (London, Oxford University Press, 1953)Google Scholar. Though many argue from historical evidence (e.g. Lipson, Leslie, op, cit.)Google Scholar that the British two-party system derives from particular national characteristics, since two parties or tendencies preceded the introduction of the present single-member constituency in the mid-nineteenth century, this argument also may be questioned. La-vau, (op. cit.)Google Scholar has pointed out that the House of Commons had unstable majorities, with members shifting their support from government to opposition throughout much of the nineteenth century. Duverger contends that in France itself the contemporary complex political substructure was built “upon the fundamental conflict which dominated the nineteenth century, that between conservatives and liberals […]. The principal actors were a landowning aristocracy, bound to monarchical principles […] and, opposed to this aristocracy, an industrial, commercial and intellectual bourgeoisie, attracted to the principles of political liberty”.

“The first phase in the moulding of the prevailing spirit in modern Europe ended with the appearance and development of the socialist parties […]. Between 1900 and 1914, the bipartisan tendency which had dominated the preceding century was replaced everywhere by a swing towards tripartisanship; the ‘conservative-liberal’ duo now changed to a ‘conservative-liberal-socialist’ trio”. Duverger, Maurice, Public Opinion and Political Parties in France, American Political Science Review, XLVI (1952), p. 1070.Google Scholar

(27) Political Parties, op. cit., pp. 319–320; see also Hermens, F. A., Europe Between Democracy and Anarchy (Notre Dame, University of Notre Dame Press, 1951), pp. 41–44.Google Scholar

(28) For a detailed account of the events leading up to this election as well as an analysis of the vote, see Dupeux, Georges, Le Front populaire et les élections de 1936 (Paris, Armand Colin, 1959).Google Scholar

(29) “It is especially important, in order to understand what the fundamental political problems of France actually are, to investigate why the state for too many years has seemed so completely powerless to hold the antagonisms and the divisions between the parties within reasonable bounds and to make the divergent forces act in concert for the general welfare. This raises the question of political institutions, the most important problem facing France. The present difficulties in this sphere stem from the fact that the political institutions that were adopted in 1946 in no way satisfy the requirements of the economic, social, and political situation […]”. Go-Guel, François, France under the Fourth Republic (Ithaca, Cornell University Press), p. 146 (my emphasis).Google Scholar

(30) The one other country which I know of that has a system akin to the American primaries is Uruguay, the most stable democracy in Latin America. In Uruguay, the various factions within the two major parties may each nominate a presidential candidate. On election day, the voters choose the man they prefer. When votes are counted, the party which has a majority counting the votes of all its presidential candidates wins the election, and the candidate of that party who received more votes than any other one of the party is elected president. In other words, Uruguay combines the primary and the final election on the same ballot. For plans to form a similar system in Brazil, see Szulc, Tad, Brazilians Plan Voting Alliance, New York Times, 12 29, 1959.Google Scholar

(31) Williams, Philip, Politics in Post-War France (London, Longmans, Green and Co., 1954), p. 310Google Scholar. This book contains an excellent discussion of the nature and effects of the French electoral systems. For a detailed description of the way in which the double ballot worked in the first elections of the Fifth Republic in 1958, see Williams, Philip and Harrison, Martin, “France 1958”, in Butler, D. E. ed., Elections Abroad (London, Macmillan, 1959), pp. 13–90.Google Scholar

(32) Truman, David, The Congressional Party: A Case Study (New York, John Wiley, 1959)Google Scholar. See also MacRae, Duncan Jr., Dimensions of Congressional Voting (Berkeley, University of California Press, 1958)Google Scholar and Key, V. O. Jr., Politics, Parties and Pressure Groups (New York, Crowell, 1958), p. 729.Google Scholar

(33) There is in fact one long-term experiment in Finland with a constitutional system that combines a strong president with a relatively weak cabinet which, as in the Fifth Republic, is responsible to Parliament rather than to the president. This Finnish version of a “Gaullist” constitution has not reduced the number of parties in the country; there are currently seven main groupings, and, on average, there has been a cabinet shake-up every year since 1917. It should be noted that although Finland has a strong presidential system, the President is elected indirectly rather than by the people, and both presidential electors and members of parliament are elected under a system of proportional representation. Consequently, there is little scope for the presidential nature of the system to reduce the number of parties. However, the Finnish sociologist Heikki Waris reports that, “By dividing the power between a stable authority, the president, and a temporary authority, the Cabinet, the Finnish Constitution has considerably lessened, if not eliminated, the apparent instability which is the great weakness of the cabinet system in multiparty countries. The Finnish system […] has assured a certain continuity of policy”. Waris, Heikki, “Finland”, in Rose, Arnold ed., The Institutions of Advanced Societies (Minneapolis, University of Minnesota Press, 1958), p. 211.Google Scholar

In European states where local government is highly centralized through the prefectorial system, the argument that strengthening the presidency will stabilise democracy must meet the counterargument that, by reducing parliamentary control over the state bureaucracy, a strong presidency might weaken still further those local centres of power and responsibility which, ever since Tocqueville, have been regarded as the best foundations for democratic order. The formula for responsible republican government, it seems, must include both a strong, directly elected president and elected local officers possessing real power. Where the local elected officials are the puppets of an appointed prefect, they are in a similar position to that of parliamentary politicians in a weak multiparty system: neither can be held accountable for long-term policies. This evil is greatest when, as often in France, local and national politicians form a single class of men. The liberal and democratic argument against centralized internal authority was forcibly restated as recently as 1944 by Luigi Einaudi, subsequently the first president of the Republic of Italy: “Democracy and the prefect are profoundly repugnant to each other […]. Neither in Italy, nor in France, nor in Spain, nor in Prussia has there ever been, nor will there ever be, democracy so long as there exists a centralized type of government of which the prefect is the symbol”. “Via col prefetto”, Il buon governo (Bari, 1954), p. 52Google Scholar; cited by Price, Roy, The Italian Local Elections 1956 (London, Chatto and Windus, 1956), p. 2Google Scholar. It does not seem to have been heeded by any of the countries possessing a prefectorial system, except for some of the German states.

(34) See Seymour Martin LIPSET, Democracy in Alberta, Canadian Forum, XXXIV (1954). 175–177 and 196–198Google Scholar. For British Columbia, see Angus, H. F., The British Columbia Election, 1952, Canadian Journal of Economics and Political Science, XVIII (1952), 518–525CrossRefGoogle Scholar; and Ormsby, Margaret, British Columbia; A History (Vancouver, Macmillan, 1958), pp. 477–489.Google Scholar

(35) See Carter, Gwendolen M., The Politics of Inequality, South Africa since 1948 (New York, Frederick A. Praeger, 1958)Google Scholar; for a further report see Far-Quharson, R. R., “South Africa 1958”, in D. E. Butler, ed., op. cit., pp. 229–275.Google Scholar

(36) Peter Campbell, Politicians, Public Servants, and the People in New Zealand, I, Political Studies, III (1955), pp. 196–197.Google Scholar

(37) Although the Country and Liberal parties usually do not run against each other and act, in effect, electorally as the rural and urban wing of the non-socialist party, the continued existence of two such parties is facilitated by the fact that Australia has adopted the preferential ballot system with the single-member constituency. Under this system voters list the order of preference for all candidates on the ballot. Thus when there are Liberal and Country candidates in the same constituency, a Liberal voter will mark the Country candidate as his number two choice, and Country party supporters will do the same for the Liberals. This system has also encouraged occasional splits from the Labor party, since minority Labor party candidates can pick up first votes without these votes being permanently lost to the major party backed by such protest voters. See Miller, J. D. B., Australian Government and Politics (London, Gerald Duckworth, 1954), PP. 85–86.Google Scholar

(38) See Lakeman, Enid and Lambert, James D., Voting in Democracies (London, Faber and Faber, 1955), pp. 223–230.Google Scholar

(39) Basil Chubb, Cabinet Government in Ireland, Political Studies, III (1955), p. 272.Google Scholar

(40) Hillel, Beth [Society for Social Research in Israel], Electoral Reform in Israel (Tel Aviv, Beth Hillel Publications, 1953), PP. 26, 24.Google Scholar

(41) Duverger, , Op. Cit., pp. 246–247.Google Scholar

(42) Oppenheim, , op. cit., p. 167.Google Scholar

(43) Carter, Gwendolen M., “The Commonwealth Overseas: Variations on a British Theme”, in Neumann, S. ed., Modern Political Parties (Chicago, University of Chicago Press, p. 104.Google Scholar

(44) Leiserson, Avery, Parties and Politics (New York, Alfred A. Knopf, 1958), p. 286.Google Scholar

(45) Detailed discussion of the relation between votes and seats in the British system in McCallum, R. B., The British General Election of 1945 (New York, Oxford University Press, 1947), pp. 277–292Google Scholar; for a table giving votes and seats in British elections since 1900, see Neumann, S. ed., Modern Political Parties (Chicago, University of Chicago Press, 1956), p. 57.Google Scholar

(46) See Friedrich, Carl, Constitutional Government and Democracy (Boston, Ginn and Co., 1950), pp. 416–417Google Scholar; Parsons, Talcott, op. cit.Google Scholar

(47) See Hermens, F. A., The Representative Republic (Notre Dame, University of Notre Dame Press, 1958), p. 201.Google Scholar

(48) “Electoral procedures based on territorial… [as distinct from “representation through interest-groups”] is precisely the technique for the organic integration of the whole. As a matter of principle the individual delegate represents the entire area. The ensuing separation into parties according to political tendencies implies then only differences of belief concerning the means by which the welfare of the nation is to be achieved”. Simmel, Georg, The Web of Group Affiliations (Glencoe, The Free Press, 1955), p. 194.Google Scholar

(49) For a detailed description of the system and an analysis of the events leading to the downfall of the first Austrian republic, see Gulick, Charles, Austria: From Habsburg to Hitler (Berkeley, University of California Press, 1948)Google Scholar. A recent study of elections in the first and second republics demonstrates that the second one is genuinely different from the first in that the two major parties, though based on the same groups as before 1934, have much more support today within “opposition strata” than their predecessors did. Thus the conservative People's party is much stronger today among workers, residents of Vienna, Protestants, and irreligious people, than was the pre-1934 Christian Social party. Conversely, the Socialists though weaker in Vienna are much stronger in the outlying provinces than earlier, and they have increased their vote among peasants considerably. Paralleling the growth of the two parties within segments once overwhelmingly opposed to them has been a sharp decline in ideological cleavage. The conservative party is no longer a Christian or Catholic party, and the Socialists have dropped their adherence to Marxist doctrine. See Simon, Walter B., Ethik, Politische und Struktur, Politische, Kölner Zeitschrift für Soziologie und Sozialpsychologie, XI (1959), 445–459.Google Scholar

(50) Carter, , op. cit.Google Scholar

(51) Cited in Humphreys, John H., Proportional Representation, A Study in Methods of Election (London, Methuen and Co., 1911), p. 58.Google Scholar

(52) Ibid., p. 57; a similar point is made by Duverger, Maurice in L'influence des systèmes électoraux sur la vie politique (Paris, Armand Colin, 1950), pp. 39–40Google Scholar; see also Lakeman, Enid and Lambert, James D., Voting in Democracies (London, Faber and Faber, 1945), pp. 63–64Google Scholar. The latter contend also that “An instance of the unifying effect of proportional representation is Czechoslovakia after the first world war, where representation of their respective minorities prevented what might have been a sharp cleavage between Bohemia and Slovakia”. Loc. cit.

(53) Herbert Tingsten, Stability and Vitality in Swedish Democracy, The Political Quarterly, XXVI (1955), 140–151.Google Scholar

(54) Tingsten, Herbert, Political Behaviour (London, P. S. King, 1937), pp. 219–220, 224–225.Google Scholar

(55) Selznick, Philip, The Organizational Weapon (New York, McGraw-Hill and Co. 1952), pp. 276–291.Google Scholar

(56) See Political Man, op. cit., pp. 77–96Google Scholar; as Walter Bagehot put it almost a century ago: “The functions of […] royalty are for the most part latent […]. It seems to order, but it never seems to struggle. It is commonly hidden like a mystery, and sometimes paraded like a pageant, but in neither case is it contentious. The nation is divided into parties, but the crown is of no party. Its apparent separation from business is that which removes it both from enmities and from desecration, which preserves its mystery, which enables it to combine the affection of conflicting parties—to be a visible symbol of unity […]”. Bagehot, , op. cit., p. 40Google Scholar. Bagehot's use of the distinction between manifest and latent activities and functions is, I think, one of the first uses of these concepts in social science.

(57) Michel Debré, the first Prime Minister of the Fifth Republic, argued in 1947 that the Fourth Republic would inevitably collapse because it used proportional representation. See his La mort de l'État républicain (Paris, Gallimard, 1947).Google Scholar

(58) Dankwart Rustow, Some Observations on Proportional Representation, The Journal of Politics, XII (1950), 107–127.Google Scholar

(59) for a general theoretical discussion of how to deal with these problems comparatively, see Sutton, Francis X., Representation and the Nature of Political Systems, Contemporary Studies in Society and History, II (1959), 1–10Google Scholar. For a detailed description of the party systems and most recent elections in various African states, see Mackenzie, W. J. and Robinson, K. E., Five Elections in Africa (Oxford, Clarendon Press, 1960).Google Scholar

(60) See Duverger, Maurice, Political Parties, op. cit., pp. 275–280Google Scholar; Scott, Robert E., Mexican Government in Transition (Urbana, University of Illinois Press, 1959), esp. pp. 145–243Google Scholar. Karpat, Kemal H., Turkey's Politics; The Transition to a Multi-Party System (Princeton, Princeton University Press, 1959).Google Scholar

(61) Weber, Max, op. cit., p. 226Google Scholar. These concepts are elaborated in an unpublished paper by Richmond, Patricia, “Democracy and Party: The Problem of Political Stability in the Mexican ‘One-Party’ System”Google Scholar, which has greatly influenced my thinking on these issues.

(62) Duverger, M., Political Parties, op. cit., p. 278.Google Scholar

(63) Richmond, Patricia, op. cit., p. 29.Google Scholar

(64) Diamond, Martin, Democracy and The Federalist: A Reconsideration of the Framers' Intent, American Political Science Review, LIII (1959), pp. 56–57.Google Scholar