1 For a useful introduction to the recent literature on Quebec see Resnick, Philip, “La Gauche et la question nationale,” this JOURNAL 13 (1980), 377–88Google Scholar, and the extensive new references and commentary in the revised edition of Kenneth McRoberts and Dale Posgate, Quebec: Social Change and Political Crisis (Toronto: McClelland and Stewart, 1980).Google Scholar

2 The obsession with political leadership has been a staple of English-Canadian journaLism and public opinion, and there is evidence of a tendency to make a theoretical virtue of this practice. See Paul Fox's presidential address, “Psychology, Politics, and Hegetology,” this Journal 13(1980), 675–90.Google Scholar The dominant emphasis in English-Canadian political science on governments as independent actors shaping society rather than being shaped by it, and on political conflict as pre-eminently the clash of governmental self-interest, is clearly expressed in Cairns, Alan C., “The Governments and Societies of Canadian Federalism,” this JOURNAL 10 (1977), 695–725Google Scholar; and Hockin, Thomas A., Government in Canada (Toronto: McGraw-Hill Ryerson, 1976).Google Scholar

3 Limited attention is explicitly devoted in the recent English-Canadian political economy literature to the problems of federalism, regionalism and intergovernmental relations, but for this argument see Stevenson, Garth, “Federalism and the Political Economy of the Canadian State,” in Panitch, Leo(ed.), The Canadian Slate (Toronto: University of Toronto Press, 1977), 79–100.Google Scholar The more general argument in Canadian political economy does not specify this attachment of different fractions of the bourgeoisie to different levels of government. It is argued rather that regional economic distortions are a result of the form of concentration of capital in central Canada, and political conflict is expressed in the opposition of the interests of the Canadian bourgeoisie served by the national state to the interests of exploited classes or alliances of classes in the peripheral provinces. See Wallace Clement, “A Political Economy of Regionalism,” and Cuneo, Carl J., “A Class Perspective on Regionalism,” both in Glenday, Daniel, et al. (eds.), Modernization and the Canadian State (Toronto: Macmillan, 1978), 89–110Google Scholar and 132–56; and Henry Milner, “The Decline and Fall of the Quebec Liberal Regime: Contradictions in the Modern Quebec State,” in Panitch, The Canadian State, 101–32.

4 There was of course little debate in English Canada about the desirability of independence for Quebec, or of restrictions on English language rights in that province. Given the near unanimous opposition to Quebec nationalism on these points, the debate in English Canada focussed rather on these other questions conceived in terms of deterrents to support for the PQ or inducements to Canadian federalism.

5 See Fletcher, Frederick J., “Public Attitudes and Alternative Futures,” in Simeon, Richard (ed.), Must Canada Fail? (Montreal: McGill-Queen's University Press, 1977), 28–41Google Scholar; Edouard Cloutier et le Centre de recherches sur l'opinion publique, Sondage sur la perception des problemes constitutionnels Quebec-Canada par la population du Quebec (Montreal, September 1979)Google Scholar; Ornstein, Michael D.. Stevenson, H. Michael and Williams, Paul M., “Public Opinion and the Canadian Constitutional Crisis,” Canadian journal of Sociology and Anthropology 15 (1978), 158–205.CrossRefGoogle Scholar

6 The classic statement of the pluralist position in reference to Canada is Porter, John, The Vertical Mosaic (Toronto: University of Toronto Press, 1965).CrossRefGoogle Scholar An attempt to investigate the process of policy making in a pluralist framework is Presthus, Robert, Elite Accommodation in Canadian Politics (Cambridge: Cambridge University Press, 1973).Google Scholar See also Thorburn, H. G., “Canadian Pluralist Democracy in Crisis,” this Journal 11 (1978), 723–38.Google Scholar

7 See Miliband, Ralph, The State in Capitalist Society (London: Weidenfeld and Nicholson, 1969).Google Scholar

8 Poulantzas, Nicos, Political Power and Social Classes (London: New Left Books, 1972). There is an internal tension in Poulantzas’ argument between an instrumentalist conclusion that the state in advanced capitalist societies enshrines the hegemony of monopoly capital and structuralist argumentation that the state is a “balance of forces” in which the unity of society and capital is expressed as a resultant of the relatively autonomous political and ideological practices that forge policy and political allinaces, these being irreducible to the interests of any single class or class fraction.Google Scholar

9 See Offe, Klaus, “The Theory of the Capitalist State and the Problem of Policy Formation,” in Lindberg, Leon K., Crouch, Colin, and Offe, Klaus (eds.), Stress and Contradiction in Modern Capitalism (Lexington: Lexington Books, 1975)Google Scholar; Holloway, John and Picciotto, Sol (eds.), State and Capital: A Marxist Debate (Austin: University of Texas Press, 1979)Google Scholar; and Block, Fred, “The Ruling Class Does Not Rule: Notes on the Marxist Theory of the State,” Socialist Revolution 33 (1977), 6–28.Google Scholar

10 It might be hypothesized, for example, that foreign controlled business would be more amenable to negotiating with Quebec than was Canadian controlled business, as the logic of multinational operations implies, that business with interests in areas more closely tied to provincial jurisdiction and protection (notably natural resources but also manufacturing) would be more amenable to negotiation with Quebec than commercial and financial business with interests more closely tied to federal jurisdiction and protection; or that “big business,” more able to pass on the costs of political and other changes in the relations of production, would be more amenable to negotiation with Quebec than “small” business.

11 The generalization about the “liberalism” of elites was developed first in research on the United States, notably in Stouffer, Samuel, Communism, Conformity and Civil Liberties (Garden City: Doubleday, 1955)Google Scholar; and McClosky, Herbert J. et al., “Issue Conflict and Consensus Among Party Leaders and Followers,” American Political Science Review 54 (1960), 406–27.CrossRefGoogle Scholar Presthus reaffirms the generalization for Canada, arguing that “the Canadian political elite… shares a generally consistent disposition towards political liberalism as measured here” (Elite Accommodation, 298). Putnam cites studies in several other countries that in his words “have shown that leaders are considerably more tolerant of political dissent and more committed to civil liberties and due process than are ordinary citizens” (The Comparative Study of Political Elites [Englewood Cliffs: Prentice-Hall, 1976], 116).Google Scholar He cautions earlier, however, that “the pages of world history record how remarkably rare is elite commitment to [criticism of, and organized opposition to, incumbent rulers]. Even today most leaders in most countries doubt the capacity of their subjects to participate fully in the governing process and reject the right of discontented groups to contest the status quo” (84). This caution might well have qualified his generalization about Western countries, and is obviously relevant to the Canadian case we are investigating.

12 The key texts on this subject are: Converse, Philip E., “The Nature of Belief Systems in Mass Publics,” in Apter, David E.(ed.), Ideology and Discontent (New York: The Free Press, 1964), 206–61Google Scholar; “Attitudes and Non-Attitudes: Continuation of a Dialogue,” in Tufte, Edward P. (ed.), The Quantitative Analysis of Social Problems (Don Mills: Addison-Wesley, 1970)Google Scholar, and his response to Pierce, John C. and Rose, Douglas D., “Non-Attitudes and American Political Opinion: Examination of a Thesis,” American Political Science Review 68 (1974), 626–49 and 661–66.CrossRefGoogle Scholar

13 The matching of elite and public opinion has been normal procedure in the investigation of the elite versus public liberalism argument, and is desirable for probing the constraint argument. The latter argument has been addressed with Canadian data only, to our knowledge, in MacDonald, Lynn, “Attitude Organization and Voting Behaviour in Canada,” Canadian Review of Sociology and Anthropology 8 (1971), 164–84.CrossRefGoogle Scholar

14 We are aware that the inclusion of the concepts “elite” and “class” in the same sentence involves a kind of “category mistake.” The concepts originate in theories based on radically different assumptions and drawing fundamentally different conclusions about the dynamics of social and political life. Elite theory is explicitly developed, in fact, to invalidate or displace theories of class. This “category mistake” is, however, central to the methodology of recent instrumental theories of the state, as Nicos Poulantzas argues in response to Miliband, The State in Capitalist Society. See Poulantzas, “The Problem of the Capitalist State,” reprinted from New Left Review in Blackburn, Robin (ed.), Ideology in Social Science (Glasgow: Fontana, 1972), 238–53. In our view. Poulantzas is correct in arguing against the supposition that a class force can be empirically determined by the social background and kinship ties of elites. He is wrong, however, to argue that a correct social theory can only be worked out in theoretical abstractions uncontaminated by the terms and data of bourgeois empiricisms, and that the use of the concept elite and the methodology of elite background analysis necessarily entails residues of bourgeois ideology. Marx's method seems clearly to have been to think through the terms and data of bourgeois ideology to a theoretical account that exposed the hidden logic mystified in ideology. In any event, our own use of the term elite entails no more than a reference to the individuals occupying the highest official positions in various institutions in society. Our analysis of the social background and attitudes of these individuals attempts to show that the assumption that they are members of a ruling class is empirically more compatible with the structuralist logic of class structure and state power, with which Poulantzas is pre-eminently associated, rather than with the instrumentalist logic he criticizes.Google Scholar

15 See Clement, Wallace, The Canadian Corporate Elite: An Analysis of Economic Power (Toronto: McClelland and Stewart, 1975), 354ff.Google Scholar, and Continental Corporate Power: Economic Linkages between Canada and the United States (Toronto: McClelland and Stewart, 1977), 29ff.Google Scholar

16 See Olsen, Denis, The State Elite (Toronto: McClelland and Stewart, 1980), chaps. 5 and 6.Google Scholar

17 This is an admittedly crude summary of a very complex argument, most carefully explored in the work of Poulantzas. The difficulties in his argument indicated in our footnote 8 above are alleviated somewhat in dialogue with his critics. See Poulantzas, Nicos, “The Capitalist State: A Reply to Miliband and Laclau,” New Left Review 95 (1976), 63–83.Google Scholar

18 To regulate the size of the sample, new respondents were added when an interview could not be obtained. The total of 602 elite interviews is approximately 90 per cent of the original sample size; in all, approximately 70 percent of the requests resulted in completed interviews. In a small number of cases, substitutions of close colleagues of the originally selected respondents were made. The data were gathered between August 1977 and May 1978. The individual sectors were sampled in the following fashion: a. Business from large firms: The respondents were chief executives of a 50 percent sample of the largest Canadian firms, as identified in the 1977 Financial Post ranking. b. Businessmen from medium-size firms: Using lists obtained with the cooperation of boards of trade in the 12 largest Canadian cities, interviews were obtained with firms' chief executives—the number selected in each city was proportional to its population. c. Lawyers: Interviews were requested from the senior partners of the 35 largest Canadian law firms. d. Federal civil servants: Our respondents were deputy ministers and assistant deputy ministers, sampled randomly from a complete list. e. Provincial civil servants: In each province interviews were attempted with the deputy ministers in the same four ministries—as a result, approximately equal numbers of respondents were obtained in each province. Our reports on the opinions of provincial civil servants include Quebec—but those responses account for only one-tenth of the overall figures. f. Municipal civil servants: This sample includes municipal department heads from the same four departments in the 12 largest Canadian cities. g. Federal politicians: This sample includes only members of parliament; the selection process included cabinet ministers, opposition critics, and backbenchers, h. Provincial politicians: Cabinet ministers from each province were interviewed—from the same four ministries as for the civil servants. i. Municipal politicians: The selections include the mayors and three randomly chosen members of the executive committee (or board of control, in some cases) for the 12 largest Canadian cities. j . Mass media executives: Those selected included the editors of the largest Canadian daily newspapers and the chiefs of the major radio and television networks. k. Academics: The sample was drawn randomly from the membership of the Royal Society of Canada. It therefore strongly represents the physical sciences and humanities, but includes few social scientists or other representatives of recently developed academic areas. 1. Trade union leaders: The respondents are presidents of the largest trade unions and the top officials of the major trade union centrals. m. Agriculturalists: The sample includes the heads of the major farm organizations in each province.

19 Reponses to the second question are complicated by the relatively large proportion of respondents who indicated that the question was impossible to answer—this was true for about 15 per cent of the elites and 20 per cent of the general public.

20 This evidence for the relative salience of issues relating to the future of Canada is amplified somewhat by responses to questions about the costs of the separation of Quebec. Whereas 88 per cent of the general public agreed that the economy of Quebec would suffer if Quebec became independent, only 55 per cent felt that it was likely that other provinces would move towards independence rather than remaining a coherent political unit if Quebec separated. Quebeckers were significantly more likely than other Canadians to say that independence would have an economic cost, so the corresponding proportions of the population in the rest of Canada who appear to have been apprehensive about the costs of Quebec's separation are even lower. This somewhat sanguine response might reflect the relatively low probability assigned to the likelihood of Quebec's separation—60 per cent of the general public in our survey thought it was not very likely or not at all likely that Quebec would become independent. Elite respondents, in contrast, were more inclined to see serious economic costs and to acknowledge the likelihood of Quebec's independence.

21 These figures are not given in the tables, but seeOrnstein et al.. “Public Opinion and the Canadian Constitutional Crisis,” for the public survey data

22 The characteristics of the variables employed in the regression are as follows: a. educational attainment is measured in years: b. age is in years; c. the logarithm of family income in thousands of dollars is used to measure income in the general public: for the elites, the logarithm of the individual's income is taken: d. the six measures of political ideology are all Likert scales using between two and eight survey items: precise wordings are available from the authors.

23 See the mention of the results of this survey in The White Paper on Quebec's Referendum, chap. 5, reported in The Globe and Mail, November 2, 1979, 12.Google Scholar