1 In December 1987, a judicial inquiry had concluded that Stevens had breached conflict of interest principles on 14 occasions during his period as a cabinet minister. MrParker, Justice W. D., Report of the Commission of Inquiry into the Facts of Allegations of Conflict of Interest Concerning the Honourable Sinclair M. Stevens (Ottawa: Supply and Services Canada, 1987).Google Scholar

2 See, for example, Cameron, Stevie, “Liberals' Ugly Chaotic Nomination Fights Make Farce of Democracy,” Globe and Mail, July 14, 1988Google Scholar, and Simpson, Jeffrey, “Democracy Diminished,” Globe and Mail, September 14, 1988.Google Scholar

3 For comparative studies, see Gallagher, Michael and Marsh, Michael, eds., Candidate Selection in Comparative Perspective: The Secret Garden of Politics (London: Sage Publications, 1988)Google Scholar, and Norris, Pippa, Carty, R. K., Erickson, Lynda, Lovenduski, Joni and Simms, Marian, “Party Selectorates in Australia, Britain and Canada,” Journal of Commonwealth Studies 28 (1990), 219–245.Google Scholar

4 “Candidate Selection,” in Penniman, Howard, ed., Canada at the Polls, 1979 and 1980: A Study of General Elections (Washington: American Enterprise Institute, 1981), 86–120Google Scholar. This essay provides a summary of the literature on Canadian practices prior to the 1979 and 1980 elections.

5 Few parliamentary democracies place much in the way of legal limits beyond age and citizenship on eligibility for legislative office. See Gallagher, and Marsh, , eds., Candidate Selection, 257.Google Scholar

6 For the 1988 election 1,547 Canadians (out of 17.5 million voters) were nominated. In fact only those nominated by a national party (80 per cent of them all) stood a realistic chance of being elected.

7 In a few states candidates endorsed by the party organization through, for example, party conventions, have legal advantages in primaries, but this is unusual. See Epstein, Leon D., Political Parties in the American Mold (Madison: University of Wisconsin Press, 1986), 148.Google Scholar

8 Epstein, Political Parties, 153–54, and Eldersveld, Samuel D., Political Parties in American Society (New York: Basic Books, 1982), 228–230, 335.Google Scholar

9 These results are found in the surveys of the 1983 Conservative and 1984 Liberal national leadership conventions conducted by George Perlin. For descriptions of these surveys see Perlin, George, ed., Party Democracy in Canada (Scarborough: Prentice-Hall, 1988)Google Scholar. We discovered the same pattern among provincial activists in British Columbia, where both NDP and Liberal activists strongly reject the notion that the party ought to be able to require that a certain proportion of the candidates be women. See Blake, Donald, Carty, R. K. and Erickson, Lynda, Grass Roots Politicians: Party Activists in British Columbia (Vancouver: University of British Columbia Press, 1991).Google Scholar

10 On the lack of political experience among Canada's members of parliament, see Franks, C. E. S., The Parliament of Canada (Toronto: University of Toronto Press, 1987), 72–79.CrossRefGoogle Scholar

11 See Guppy, Neil, Freeman, Sabrina and Buchan, Shari, “Representing Canadians: Changes in the Economic Backgrounds of Federal Politicians, 1965–1984,” Canadian Review of Sociology and Anthropology 24 (1987), 417–430.CrossRefGoogle Scholar

12 On organizational differences in parties across time, see Carty, R. K., “Three Canadian Party Systems,” in Perlin, , ed., Party Democracy, 15–30Google Scholar. Although a systematic analysis of nomination processes remains to be done, contemporary newspaper accounts during the first period (1867–1917) testify to the widespread use of this form.

13 Quebec and Newfoundland are the most often cited examples. In the latter case the provincial premier following Confederation in 1949, Joseph Smallwood, generally determined who would be the federal Liberal candidates.

14 In some parts of the CCF the provincial office exercised a veto over locally selected candidates.

15 Courtney, John C., “Recognition of Canadian Political Parties in Parliament and Law,” this Journal 11 (1978), 33–60.Google Scholar

16 Canada Elections Act, 23(2) (h).

17 Formal public vetoes of locally selected candidates have been rare. There were two widely reported cases in 1988: the Conservative leader and prime minister, Brian Mulroney, vetoed his former cabinet colleague Sinclair Stevens, and the Reform party leader, Preston Manning, vetoed a candidate in British Columbia for his views on immigration and race questions.

18 See Fraser, Graham, Playing for Keeps: The Making of the Prime Minister, 1988 (Toronto: McClelland and Stewart, 1989), 178.Google Scholar

19 For example, see Regenstrief's, Peter description of Liberal party conventions in “The Liberal Party of Canada: A Political Analysis” (doctoral dissertation, Cornell University, 1963), 160.Google Scholar

20 A variant of the general meeting is the “travelling convention” in which series of ballots are conducted in communities across a territorially large constituency. See Williams, “Candidate Selection,” 112, for a description of these.

21 Ibid., 104–06.

22 Ibid., 99–103.

23 Ibid., 100.

24 “Representing Canadians,” 421.

25 Much of the increase in women candidates in the 1970s was attributable to women running as independents or for fringe parties. It was not until the 1984 election that the majority of women candidates were associated with one of the three parties. Brodie, Janine, Women and Politics in Canada (Toronto: McGraw-Hill Ryerson, 1985), 19–20.Google Scholar

26 Ibid., chap. 8.

27 We would like to acknowledge the assistance of the staff of the Office of the Chief Electoral Officer in supplying the list of official agents. Their help in this, and in providing other information, has been invaluable. The questionnaire (with marginals) and data set are available on request.

28 Under the Canada Elections Act, “official agents” must be named by the candidate as part of the formal nomination process. Agents' responsibilities effectively make them the principal instrument through which the state enforces the election financing regime with respect to local constituency activity.

29 Nine per cent of our returns came from Atlantic Canada, 24 per cent from Quebec, 37 per cent from Ontario and 29 per cent from the West. There were also three (of nine possible) returns from the territories: they are not always included in analysis that have regional breakdowns. Of the replies, 22 per cent were in French, 78 per cent in English. In political terms 32 per cent were from winners, 68 per cent from losers which is what one would expect in a three-party system.

30 For one such case see Fraser, Playing for Keeps, 166.

31 See, for instance, “The War for Toronto Ridings,” Globe and Mail, June 27, 1988; “Wading Through Liberal Nomination Mire,” Globe and Mail, August 25, 1988, and “Sinclair Stevens Squares Off Against Stronach,” Globe and Mail, July 16,1988. The latter, written before the prime minister's veto of Stevens' Conservative candidacy, reported how Stronach “admitted to spending between $30,000 and $40,000 during his nomination campaign” against an opponent who himself admitted to spending nearly $10,000 (Globe and Mail, June 30, 1988).

32 One of the Liberal candidates subsequently withdrew so that there were only 884 major party candidates actually running in the election.

33 The Liberal campaign machine attempted to reserve three ridings in Quebec for “star” candidates but this once-accepted practice engendered considerable local conflict during the election and the central party had to retreat. For one of several accounts, see “Supporters of Dery Storm Turner Rally,” Globe and Mail, October 11, 1988.

34 Our survey data give no information on the details of these appeals. One account is provided by Fraser, Playing for Keeps, 163–64. The individual involved, Albina Guarnieri, MP, made this experience the centrepiece of her testimony before the Royal Commission on Electoral Reform and Party Finance, Ottawa, June 12, 1990.

35 Williams, “Candidate Selection,” 118.

36 For an account of one such local case, see “Nomination Defeat Marks Liberal Change,” Globe and Mail, April 29, 1988.

37 These two represented 2 per cent of our sample of incumbents.

38 We suspect that in the case of the national NDP it testifies to still powerful internal norms. In the case of the two older parties it may well reflect competitive differences. In 1984 (when their positions were reversed) there appears to have been rather more of this sort of internal competition in the Conservative party.

39 This is a more competitive rate than for the sample generally. As one might expect, former members tended to seek candidacies in more competitive constituencies and, as we note below, this is where competition is more likely to occur.

40 We asked our respondents “When the party was nominating the candidate, how did the local association assess the chances of victory in the constituency? Was the riding considered by your party to be: safe, good chance, unlikely, hopeless.”

41 Denver, D., “Britain: Centralized Parties with Decentralized Selection,” in Gallagher, and Marsh, , eds., Candidate Selection, 51Google Scholar. British selection processes involve considerable central party activity.

42 This includes associations that rated their electoral prospects as either having a “safe seat” or a “good chance.”

43 Hoffman, David and Ward, Norman, Bilingualism and Biculturalism in the Canadian House of Commons (Ottawa: Queen's Printer, 1970).Google Scholar

44 There is a modest, but significant negative (—.24) relationship between the proportion of francophones and a contested nomination in a constituency.

45 For example, “Fights, Charges of Rigging Mar Liberal Nomination,” “Ethnic Candidates Storm Liberal Party,” Globe and Mail, July 25 and August 16, 1988, or Elaine Dewar's “March of the Third Force,” Toronto Life, December 1988. Our analysis follows these accounts by asking about the impact of these groups on the nomination process. It is beyond the scope of this article to deal with what this activity represents for the groups involved.

46 The questionnaire asked respondents if candidates (for the nomination) actively recruited supporters to come to the nomination meeting. Those who answered affirmatively were asked about the basis of the recruitment and “ethnic groups” were one of the categories provided.

47 The data in this paragraph are calculated from the Report of the Chief Electoral Officer.

48 This was significant to .03.

49 Williams, “Candidate Selection,” 118; emphasis in the original.