¹ We wish to thank Lawrence LeDuc, Jr, Walter Soderlund, and Marianne Stewart of the University of Windsor and two anonymous reviewers for their many helpful comments and suggestions.

² The Canadian Political System (Toronto 1971), 463–4

³ introduction to Canadian Politics and Government (Toronto 1972), 127

⁴ Aiken, Gordon, The Backbencher (Toronto 1974), 86–7.Google Scholar Similarly, Paul Fox states: “… the m.p. may be a social welfare officer, assisting citizens to thread their way through the increasingly complicated maze of the welfare state by acting as their personal representative in Ottawa (looking into complaints about pensions, immigration cases, and the like).” Fox, Paul, “Our m.p.'s – Their Roles and Need,” Politics: Canada, ed. Fox, Paul (Toronto, 3rd ed. 1970), 389Google Scholar

⁵ Ontario Commission on the Legislature, First Report (Toronto 1973), 12

⁶ See Kornberg, Allan, Canadian Legislative Behavior (New York 1967)Google Scholar; Hoffman, David and Ward, Norman, Bilingualism and Biculturalism in the Canadian House of Commons, Documents of the Royal Commission on Bilingualism and Biculturalism, v. 3 (Ottawa 1970).Google Scholar For provincial data, see Smith, David, “The Recruitment, Role Perceptions and Political Attitudes of Saskatchewan m.l.a.'s,” unpublished paper presented at the Annual Meeting of the Canadian Political Science Association, Winnipeg, 1970Google Scholar; McCorquodale, Susan, “Newfoundland: The Only Living Father's Realm,” Canadian Provincial Politics, ed. Robin, Martin (Scarborough 1972), 157–67Google Scholar; Leduc, Lawrence Jr, and White, Walter L., “The Role of Opposition in a One-Party Dominant System,” this journal, VII, no. 1 (March 1974), 86–100Google Scholar

⁷ The role-theory approach to studying legislative behaviour was first proposed by John Wahlke and his colleagues. See Wahlke, John et al., The Legislative System (New York 1962)Google Scholar, ch. 1. The literature reporting correlates of legislators' role orientations has grown to sizable proportions in the last decade. For a convenient summary of findings, see Jewell, Malcolm E., “Attitudinal Determinants of Legislative Behaviour: The Utility of Role Analysis,” Legislatures in Developmental Perspective, ed. Kornberg, Allan and Musolf, Lloyd (Durham, N.C. 1970), 481.Google Scholar

⁸ In the United States, however, Jones has analysed the linkages between representational role orientations and roll-call voting in the Texas legislature. See Jones, Bryan D., “Competitiveness, Role Orientations and Legislative Responsiveness,” Journal of Politics, 35 (1973), 924–47.CrossRefGoogle Scholar

⁹ See Eulau, Heinz et al., “The Role of the Representative: Some Empirical Observations on the Theory of Edmund Burke,” American Political Science Review, 53 (1959), 742–56CrossRefGoogle Scholar

¹⁰ Ibid., 752–3

¹¹ Ibid., 749–50

¹² Concerning the consequences of variations in the level of interparty competition at the level of the province as a whole for “styles of opposition” see LeDuc and White, “The Role of Opposition in a One-Party Dominant System,” passim.

¹³ The consequences of “system level” interparty competition have been debated at length with regard to state politics in the United States. See, for example, Key, V.O. Jr, Southern Politics in State and Nation (New York 1949), 302–10Google Scholar; Dawson, Richard E. and Robinson, James A., “Inter-Party Competition, Economic Variables and Welfare Policies in the American States,” Journal of Politics, 25 (1963), 265–89CrossRefGoogle Scholar; Dye, Thomas R., Politics, Economics and the Public (Chicago 1966)Google Scholar; Cnudde, Charles F. and McCrone, Donald J., “Party Competition and Welfare Policies in the American States,” American Political Science Review, 63 (1969), 858–66.CrossRefGoogle Scholar For data on the relationship between “constituency level” interparty competition and legislative behaviour see Bryan Jones, “Competitiveness, Role Orientations and Legislative Responsiveness,” 934–9. See also Shannon, Wayne, “Electoral Margins and Voting Behavior in the House of Representatives: The Case of the Eighty-Sixth and Eighty-Seventh Congresses,” Journal of Politics, 30 (1968), 1028–45.CrossRefGoogle Scholar For a review of many of these studies and a discussion of the consequences of interparty competition in the Canadian political system see Lovink, J.A.A., “Is Canadian Politics Too Competitive,” Canadian Journal of Political Science, 6 (1973), 341–79.CrossRefGoogle Scholar

¹⁴ “Is Canadian Politics Too Competitive?” 368–70

¹⁵ Ibid., 375

¹⁶ In addition to the rather mixed findings of the literature on interparty competition in the American states referred to in fn. 13 above, see also Prewitt, Kenneth and Eulau, Heinz, “Political Matrix and Political Representation: Prolegomenon to a New Departure from an Old Problem,” American Political Science Review, 63 (1969), 433.CrossRefGoogle Scholar Prewitt and Eulau argue that the higher the percentage of “forced turnover” in municipal elections (a measure of the degree of electoral competition), the more likely it is that councilmen will report responding to the wishes of attentive constituents.

¹⁷ The importance of political ambition for democratic political systems has been argued forcefully by Joseph A. Schlesinger. Schlesinger states: “to slight the role of ambition in politics, then, or to treat it as a human failing to be suppressed, is to miss the central function of ambition in political systems. A political system unable to kindle ambition for office is as much in danger of breaking down as one unable to restrain ambitions. Representative government, above all, depends on a supply of men so driven; the desire for election and, more important, for reelection becomes the electorate's restraint upon its public officials. No more irresponsible government is imaginable than one of high-minded men unconcerned for their political futures.” Schlesinger, Joseph A., Ambition and Politics (Chicago 1966), 2Google Scholar, emphasis added. In Ambition and Politics Schlesinger assumes politicians are ambitious. For empirical data on political ambition and its consequences, see Prewitt, Kenneth and Nowlin, William, “Political Ambitions and the Behavior of Incumbent Politicians,” Western Political Quarterly, 22 (1969), 298–308CrossRefGoogle Scholar; Prewitt, Kenneth, “Political Ambitions, Volunteerism, and Electoral Accountability,” American Political Science Review, 64 (1970), 5–17CrossRefGoogle Scholar; Black, Gordon S., “A Theory of Political Ambition: Career Choices and the Role of Structural Incentives,” American Political Science Review, 66 (1972), 144–59.CrossRefGoogle Scholar

¹⁸ Schlesinger distinguishes between static and progressive political ambitions. Politicians with static ambitions wish to make a long-run career out of a particular office, whereas those with progressive ambitions aspire to a more important office than that currently held; Ambition and Politics, 10.

¹⁹ The extent to which provincial and federal political career lines in Canada are integrated is an important, but neglected, empirical question. A recent study of the recruitment of candidates for the 1972 federal election suggests that there may be separate federal and provincial recruitment “streams” in Canada. See Surich, Joachim E.-C. and Williams, Robert J., “Some Characteristics of Candidates in the 1972 Canadian Federal Election,” unpublished paper presented at the Annual Meeting of the Canadian Political Science Association, Toronto, June 1974, 5Google Scholar and 44. Allan Kornberg presents data showing that since 1930 no more than 16 per cent of mps have been mlas; “Parliament in Canadian Society,” Legislatures in Developmental Perspective, 118

²⁰ It is reasonable to assume that party organizers, when recruiting federal candidates, might use previous success (or lack therof) of a candidate at the provincial level as an indicator of the likelihood that an aspiring candidate is a potential winner at the federal level.

²¹ In his study of mps in the 25th Parliament, Kornberg constructs what he calls a “cosmopolitan-rural” index; Canadian Legislative Behavior, 90, and 101, fn. 5.

²² In choosing to utilize data gathered via a mail questionnaire we recognized that data are not being gathered on legislative behaviour per se, but rather on reports of such behaviour. If one is interested in studying the behaviour of more than a few legislators, however, it appears that there are many types of legislative behaviour where actual observation of such behaviour is impossible given realistic estimates of available scholarly resources and the structure of the legislative process in Canada. For those interested in systematic empirical inquiries into various facets of legislative behaviour, then, survey research is often the only feasible option. Of course, for scholars who cast explanations of legislative behaviour in terms of attitudinal predispositions of various kinds, survey research is mandatory. In recent years criticism of the use of survey research to study Canadian legislative behaviour has been raised, perhaps on the mistaken assumption that utilization of survey methods necessarily implies the adoption of a role-theory conceptual framework unsuited to studying legislative behaviour in Canada. There is, however, no reason why adoption of survey techniques need imply adoption of such a framework. Furthermore, as has been suggested above in the discussion of hypothesis 1, the question of the linkage between role orientations and legislative behaviour is an empirical one which should not be decided on a priori grounds. For those facets of legislative behaviour in Canada where the norm of party discipline does not render explanations of individual variations in behaviour trivial, role theory may indeed prove to be a useful approach to explanation. Only a series of empirical studies of different aspects of legislative behaviour will answer the question of the utility of role theory for understanding legislative behaviour. The lack of utility of role-orientation data for understanding voting behaviour in Parliament has been noted by Jackson and Atkinson. They point out that: “regardless of the role m.p.'s claim to adopt, for the purposes of role enactment when voting on the floor of the House all are party delegates.” Jackson, Robert J. and Atkinson, Michael M., The Canadian Legislative System (Toronto 1974), 148.Google Scholar

²³ Cabinet ministers were not included in the study on the assumption that their response rate would be very low.

²⁴ The variable showing the greatest degree of deviation from the population as a whole was level of formal education. The sample contained a somewhat larger number of respondents with university education than one would expect on a purely random basis.

²⁵ Some evidence supporting this assumption has been recently gathered in a study of the recruitment and role socialization of freshman mps. When asked about the specific types of constituency service tasks in which they expected to be engaged, the freshman mps almost invariably responded by mentioning “special” characteristics of their ridings which would dictate the kinds of constituency service tasks they would have to confront. The data for this study were gathered in October 1974 by Robert Krause, Richard G. Price, and Harold D. Clarke. These data are currently being analysed for presentation in a paper to be delivered at the Annual Meeting of the American Political Science Association, San Francisco, September 1975.

²⁶ The labels used for the various representational role orientations vary somewhat from one study to the next. Compare, for example, Kornberg, Canadian Legislative Behavior, ch. 6 with Hoffman and Ward, Bilingualism and Biculturalism in the Canadian House of Commons, 66–11.

²⁷ Most of these efforts to construct measures of interparty competition have been aimed at developing indices applicable for two party situations. Useful reviews of this work include: Pfeiffer, David G., “The Measurement of Inter-Party Competition and Systemic Stability,” American Political Science Review, 61 (1967), 457–67CrossRefGoogle Scholar; and David, Paul T., “How Can An Index of Party Competition Best be Derived?” Journal of Politics, 34 (1972), 632–8.CrossRefGoogle Scholar For a discussion of the problems involved in measuring party competition in multiparty systems such as the Canadian, see Elkins, David J., “The Measurement of Party Competition,” American Political Science Review, 68 (1974), 682–700.CrossRefGoogle Scholar See also Lovink, “Is Canadian Politics too Competitive,” 356–62.

²⁸ In multimember constituencies the procedure involved aggregating vote totals by party before computing percentage differences.

²⁹ Rae, Douglas and Taylor, Michael, The Analysis of Political Cleavages (New Haven 1970), 32Google Scholar

³⁰ See fn. 17 above.

³¹ The average inter-item correlation (Yule's Q) for the dichotomized variables used in the index is .54.

³² Concerning the emphasis on constituency in the Maritimes, see the essays by P.J. Fitzpatrick on New Brunswick, J.M. Beck on Nova Scotia, and Frank Mackinnon on Prince Edward Island in Canadian Provincial Politics. See also Mackinnon, Frank, The Government of Prince Edward Island (Toronto 1951)Google Scholar, ch. 11; Thorburn, Hugh G., Politics in New Brunswick (Toronto 1961).Google Scholar That legislators in Newfoundland are as oriented towards their constituencies as those in other Maritime provinces may be doubtful given that many of the Newfoundland members do not live in the contsituencies they represent. See Noel, S.J.R., Politics in Newfoundland (Toronto 1971), 284.Google Scholar

³³ Recently, the existence of regional and provincial differences in political attitudes and behaviour has been documented in several studies. See, for example, Blake, Donald E., “The Measurement of Regionalism in Canadian Voting Patterns,” this journal, V (1972), 55–81Google Scholar; Schwartz, Mildred A., Politics and Territory (Montreal 1974)CrossRefGoogle Scholar; Simeon, Richard and Elkins, David J., “Regional Political Cultures in Canada,” this journal, VII (1974), 397–437Google Scholar; Wilson, John, “The Canadian Political Cultures: Towards a Redefinition of the Nature of the Canadian Political System,” this journal, VII (1974), 438–83.CrossRefGoogle Scholar

³⁴ Shifting attention to variables subject to intraprovincial variation brings up questions of the logic of comparative inquiry. While basically agreeing with Przeworski and Teune that “proper name” variables such as province or region should have “residual” rather than “priority” status as explanatory factors, we judge that in an initial inquiry, such as the present one, it is of considerable potential heuristic value to document basic patterns of variation in the dependent variable. On various strategies of comparative analysis see Przeworski, Adam and Teune, Henry, The Logic of Comparative Social Inquiry (New York 1970)Google Scholar, part one. See also the brief comment on this point by Simeon and Elkins, “Regional Political Cultures in Canada,” 399, fn. 10.

³⁵ Kendall's Tau C is an ordinal measure of association based on the conceptualization of the data as a universe of ordered (and “tied”) pairs of cases on the two variables being considered. Due to its method of computation Tau C is a “conservative” statistic, that is, it almost always yields values that are numerically smaller than other ordinal measures such as Gamma. See Blalock, Hubert, Social Statistics (New York, 2nd ed. 1972), 421–6.Google Scholar

³⁶ In the analyses which follow, mlas with personal and; party political ambitions are considered separately. Analyses aggregating together those with either type of political ambition were also performed (data not shown) with similar results.

³⁷ Not unexpectedly, the correlations between “objective” and “subjective” measures of interparty competition are not perfect. Supporting the notion that legislators frequently tend to perceive interparty competition as “two party” competition is the finding that the strongest correlation (Tau C = .218, Gamma = .565) between an objective party competition measure and subjective perceptions of competition involved the “percentage difference” in vote totals between the mla and the runnerup in the last election. The Gamma correlation is identical to that reported by Kornberg in his study of the 25th federal Parliament. See Kornberg, Allan, “Perception and Constituency Influence On Legislative Behavior,” Western Political Quarterly, 19 (1966), 285–92.CrossRefGoogle Scholar

³⁸ The focus here has been on interparty competition at the constituency level. Adequate assessment of the effects of system level variations in interparty competition remains to be done. In this regard see the suggestive work by LeDuc and White, “The Role of Opposition in a One-Party Dominant System.”

³⁹ The theoretical attractiveness of interparty competition is basically derived from the fact that some notion of “opposition” at either the level of the constituency or that of the political system as a whole is an integral part of most contemporary theories of democracy. See, for example, Dahl, Robert, Polyarchy (New Haven 1971)Google Scholar, ch. 1. A classic example of deductive theorizing about political behaviour in democracies based on the assumption of interparty competition is Downs, Anthony, An Economic Theory of Democracy (New York 1957).Google Scholar

⁴⁰ See, for example, Kornberg, Canadian Legislative Behavior, ch. 6; Hoffman and Ward, Bilingualism and Biculturalism in the Canadian House of Commons, 66–83.

⁴¹ “The Role of Opposition in a One-Party Dominant System, passim

⁴² Role socialization may be roughly defined as learning about the behaviour expected of those occupying particular roles (in addition to that of “citizen”) in the political system. See Almond, Gabriel A., “Introduction: A Functional Approach to Comparative Politics,” The Politics of the Developing Areas, ed. Almond, Gabriel A. (Princeton 1960), 32.Google Scholar See also Marvick, Dwaine, “Political Recruitment and Careers,” International Encyclopedia of the Social Sciences, v. 12, ed. Sills, David L. (New York 1968), 273–81.Google Scholar

⁴³ On the relationship between interparty competition and the recruitment of candidates for federal legislative office in Canada see Kornberg, Allan and Winsborough, Hal H., “The Recruitment of Candidates for the Canadian House of Commons,” American Political Science Review, 62 (1968), 1242–57CrossRefGoogle Scholar; and Kornberg, Allan, Clarke, Harold D., and Watson, George L., “Toward a Model of Parliamentary Recruitment in Canada,” Legislatures in Comparative Perspective, ed. Kornberg, Allan (New York 1973), 250–81.Google Scholar

⁴⁴ LeDuc and White note that historically one-party dominance has been a frequent occurrence in most provincial party systems. See LeDuc and White, “The Role of Opposition,” 86–8.