¹ Peter Regenstreif, who had polled the province, was a rare exception. For his pre-election forecast, giving at least 20 seats to Social Credit, see his The Diefenbaker Interlude: Parties and Voting in Canada (Toronto, 1965), xi.Google Scholar On the element of surprise and the reactions of the candidates and the mass media to the results of the election, see Dion, Léon, “The Election in the Province of Quebec,” in Meisel, John, ed. Papers on the 1962 Election (Toronto, 1964), 109–28, esp. pp. 121 ff.Google Scholar

² In the previous federal election (1958), the Social Credit party had contested only fifteen seats in Quebec and had obtained 0.6 per cent of the total vote. See Scarrow, Howard A., Canada Votes: A Handbook of Federal and Provincial Election Data (New Orleans, 1962), 176–8.Google Scholar

³ The survey data are from a multi-stage area-stratified sample of Quebec citizens 19 years of age and over, collected a few months after the 1962 federal election. The sample will be described in greater detail in The Rise of a Third Party (in preparation).

⁴ For an explanation of this figure, see notes to Table I.

⁵ Smelser, Neil J., Theory of Collective Behavior (New York, 1963), 14 ff.CrossRefGoogle Scholar and passim.

⁶ One-party dominance is generally defined as a situation in which the main opposition party has been unable to maintain on the average more than 33 per cent of the votes in a few previous elections.

⁷ The source of these data is Scarrow, Canada Votes. On the weakness of the Conservatives in Quebec as well as in some other provinces during that period, see Williams, John R., The Conservative Party of Canada: 1920–1949 (Durham, NC, 1956), chaps. 4 and 6.Google Scholar

⁸ On this see, for instance, Filley, Walter O., “Social Structure and Canadian Political Parties: The Quebec Case,” Western Political Quarterly, 9 (1965), 900–14.Google Scholar

⁹ Theory of Collective Behavior, 278 ff., esp. 284.

¹⁰ The Progressive Conservatives were elected as a minority government in 1957 and provoked a new general election in 1958, in which they won the greatest majority of seats (208 out of 275) in Canadian history.

¹¹ For the first time since 1891 they also obtained a plurality of the total vote: they got 49.6 per cent of the votes as opposed to 45.7 per cent to the Liberals. See Scarrow, Canada Votes.

¹² For an account of this period in Canadian electoral politics, and in particular in Quebec, see Meisel, John, The Canadian General Election of 1957 (Toronto, 1962)Google Scholar and Regenstreif, The Diefenbaker Interlude.

¹³ Reflecting the general trend, no district changed from Conservative in 1957 to Liberal in 1958.

¹⁴ In a different and at first sight more refined analysis, one could consider the percentage increase in the Conservative vote from 1957 to 1958 as an indication of the strength of that party: the larger the increase, the stronger the party. But this creates problems since the percentage increase is dependent on the level previously obtained. If we take out the seven seats already in the Conservative camp in 1957, we obtain results similar to those of Table II: the larger the increase in the Conservative vote in previous elections, the weaker the Social Credit success in 1962.

¹⁵ What is most likely the case is that the dominant Liberals are strong more or less equally in all districts; that is, there were no districts characterized by Conservative one-party dominance.

¹⁶ These results can be considered from a different angle, which is particularly revealing on the areas of weakness of the Conservatives: as shown in the following Table, two-thirds of the times, the Conservative candidates lost their deposits in 1962 in the districts which were presumed weak, while this never happened in their few presumed strongholds. The relationship runs the other way for the Social Credit candidates. Only rarely did the Liberal candidates lose their deposits in all types of districts.

¹⁷ In a private communication (my translation).

¹⁸ N's equal to 13, 34, and 7 respectively, with an a ₁ (effect of PC organization strength on Social Credit outcome, comparing extremes) of .56; p ( ≤ 0) = .002.

¹⁹ It is interesting in this respect to consider the ecological distribution of the Social Credit victories: the districts won by the Social Credit candidates form a solid block within which only one seat escaped them (in Quebec city) and outside of which they did not win a single district. But it is interesting to note that the Social Credit tide was literally stopped by districts where the Conservatives tended to be stronger. For instance, five of the seven districts where the Conservatives were the strongest are contiguous to the Social Credit block of districts.

²⁰ It might be added that, in some districts, the local organization of the party was so weak that the only activity that would take place during the electoral campaign would be the official nomination of a candidate—often coming from outside, e.g. Montreal, and completely unknown in the district—and the sticking of posters and the like; not a single popular meeting or other appearance of the “mysterious” candidate would take place. See also Meisel, The Canadian General Election of 1957. Meisel writes that in 1957 the party's “organization in many Quebec constituencies was almost non-existent” and that “some of the Conservative standard-bearers were largely token candidates without effective organizations to back them” (133, 174).

²¹ Though claiming that the thesis presented above is “unimpugnable,” Lemieux wrote that our demonstration at the district level is debatable and, choosing the results of the 1953 and 1957 elections rather than those of the 1957 and 1958 elections to characterize the Progressive Conservative party's strength, he claimed to find a positive rather than negative relationship between that party's strength and the Social Credit party's successes. See Lemieux, Vincent, “Les dimensions sociologiques du vote créditiste au Québec,” Recherches Sociographiques, VI (1965), 181–95, esp. 185–90.Google Scholar In our comment on this paper, we have shown that Lemieux' positive relationship between the strength of the two parties does not exist when his data are subjected to a refined analysis. See Maurice Pinard, “La faiblesse des Conservateurs et la montée du Crédit Social en 1962”, ibid., VII (1966), 360–3. See also Lemieux' reply, ibid., pp. 363–5. The remaining “ambiguity” lies in the fact that there can be more than one operational definition of a nominal concept.

²² In the twenty-one districts of Montreal, Social Credit obtained only 6.1 per cent of the votes, compared to 35.8 per cent in the fifty-four other districts of the province.

²³ Though most of these districts have been rated by my informant in the same category— 15 out of 21 were rated as “fairly strong”—the relationship obtained with these ratings again confirms the results of Table III: 33 per cent of the “very strong” Conservative districts give more than 5 per cent of their vote to Social Credit, as opposed to 40 per cent and 67 per cent of the “fairly strong” and of the “fairly weak” districts respectively. (N's equal to 3, 15, and 3 respectively).

²⁴ For instance, in 1957, the Conservatives had less than 30 per cent of the votes in 67 per cent of the Montreal districts, as compared to only 18 per cent in the rest of the province (N = 21 and 54 respectively). In 1958, the Conservatives had less than 45 per cent of the votes in 43 per cent of the Montreal districts and in only 13 per cent of the others.

²⁵ This is documented at length in my The Rise of a Third Party (in preparation).

²⁶ As a matter of fact, organizations are probably much weaker in the Montreal districts than in the rural districts. Contrary to the USA, there is no such thing in Quebec as big city political machines. It is interesting to note in this regard—contrary again to what happens in the USA—that the turnout is much lower in the Montreal districts than elsewhere in Quebec: 93 per cent of the rural districts had turnouts of 80 per cent or more in the 1958 federal elections, while the corresponding figure for Montreal is 5 per cent (N = 54 and 21 respectively).

²⁷ N equals 320, 238, and 332 respectively. The most rural districts of the sample are the provincial electoral districts of Argenteuil, Chicoutimi, Mégantic, Montmorency, Rivière-du-Loup. The least rural are Chambly, Trois-Rivières, and those of Quebec City. On this, see also Regenstreif, , The Diefenbaker Interlude, 95.Google Scholar

²⁸ See Shirer, William L., The Rise and Fall of the Third Reich (New York, 1962)Google Scholar, and Lipset, S. M., Political Man (Garden City, NY, 1960), 146.Google Scholar

²⁹ Irving, John A., The Social Credit Movement in Alberta (Toronto, 1959).Google Scholar

³⁰ Lipset, S. M., Agrarian Socialism (Berkeley and Los Angeles, 1950), chaps. 6 and 8.Google Scholar

³¹ See Key, V. O. Jr., Southern Politics (New York, 1949), 117, 162, 177, 190.Google Scholar On the other hand, the CCF or New Democratic party has more recently fared better in urban rather than in rural areas and generally the same thing holds for Communist and Socialist parties in different countries (see Zakuta, Leo, A Protest Movement Becalmed (Toronto, 1964), 159 Google Scholar, and Lipset, , Political Man, 231, 249 ff.Google Scholar). This is undoubtedly due to greater organizational efforts in urban areas and to ideological differences between cities and country; but we suspect that it does not challenge the above structural argument: we suggest that, everything else constant, these parties would succeed more easily and faster in rural areas than in urban areas because of these structural differences.

³² This means paradoxically that, in districts of Conservative weakness, the chances of Social Credit were better, the larger the proportion of previous Conservatives. This apparent paradox accounts, we suggest, for Lemieux' finding that in a district where the Conservatives were very weak (Lévis), the stronger the Conservatives in 1958 in a community, the stronger Social Credit was in 1962 (an ecological finding within a district contrary to that reported in Tables II and III above when comparing districts). On that basis, he indeed suggested “that the Social Credit party recruited its support more from the Conservatives than from the Liberals.” See Lemieux, Vincent, “Election in the Constituency of Lévis,” in Meisel, , ed., Papers on the 1962 Election, 46–50.Google Scholar

³³ Similar results are reported by Regenstreif, , The Diefenbaker Interlude, 117–18.Google Scholar Regenstreif suggests (119-20) that not only were the Conservatives more likely to vote Social Credit but they were also more likely to be earlier recruits.

³⁴ Since the 1958 Conservative support was more likely to be working class and, since this was also true of the Social Credit support, a control of class was made on the above data to check whether the findings were spurious, but they were not.

³⁵ All the effects reported in the last two paragraphs about weak and intermediate rural districts generally hold for the eight Montreal and Quebec districts (where the Conservatives were weak). Moreover, these results hold with economic strain constant, both having independent effects.

³⁶ Key, , Southern Politics, chap. 13, esp. 292 ff.Google Scholar Key writes: “Southern Republican leaders … are not politicians in the usual sense of the word. They might be called palace or bureaucratic politicians, since their chief preoccupation is not with voters, but with maneuvers to gain and keep control of the state party machinery. They exert themselves only to keep the party weak in the South in order that there will be fewer faithful to reward.”

³⁷ This is discussed at length in The Rise of a Third Party.

³⁸ MacRae, Duncan Jr., “Occupations and the Congressional Vote, 1940–1950.” American Sociological Review, 20 (1955), 339.CrossRefGoogle Scholar

³⁹ The evidence so far accumulated supports the idea that new parties in Canada arose only in situations of one-party dominance. It seems that whenever—and only whenever—the main opposition party fails to retain at least a third of the electorate, it is likely, in periods of strain, to be replaced by new parties. The evidence is presented at length in my The Rise of a Third Party. Preliminary data were presented in my “Political Factors in the Rise of Social Credit in Quebec.” Thus the political structure, I suggest, rather than alleged differences in political culture would account for the presence of third parties on the Canadian political scene. For a cultural view, see Horowitz, G., “Conservatism, Liberalism, and Socialism in Canada: An Interpretation,” this Journal, XXXII, no. 2 (05 1966), 143–71.Google Scholar

⁴⁰ Macpherson, C. B., Democracy in Alberta: Social Credit and the Party System (Toronto 2nd ed., 1962).Google Scholar

⁴¹ See Lipset, S. M., “Democracy in Alberta,” Canadian Forum, 34 (11, Dec. 1954), 175–7 and 196–8.Google Scholar See also C. B. Macpherson, “A Reply,” ibid., 34 (Jan. 1955), 223–5. Notice, however, that Lipset does not include in his thesis that the development of one-party dominance is the intervening factor between the existence of cleavages and the rise of political movements. See also Lipset, , Political Man, 270–9Google Scholar; Wrong, Dennis H., “Parties and Voting in Canada: A Backward and Forward Glance in the Light of the Last Election,” Political Science Quarterly, LXXIII (1958) 3, 397–412 CrossRefGoogle Scholar; Lipset, S. M., The First New Nation: The United States in Historical and Comparative Perspective (New York, 1963) chap. 9.Google Scholar For an earlier view on the crucial importance of cleavages in explaining party systems, see Lipson, Leslie, “The Two-Party System in British Politics,” American Political Science Review, 47 (1953), 337–58.CrossRefGoogle Scholar

⁴² Macpherson, , Democracy in Alberta, 6 ff., esp. 9.Google Scholar

⁴³ See Pinard, Maurice, “Structural Attachments and Political Support in Urban Politics: The Case of Fluoridation Referendums,” American Journal of Sociology, 68 (1963), 513–26.CrossRefGoogle Scholar

⁴⁴ Though the Maritime provinces, and especially Prince Edward Island, would present an exception here.

⁴⁵ Though at the provincial level it has not been used uniformly throughout the periods considered in all provinces; see Scarrow, Canada Votes.

⁴⁶ On this, see Lipset, “Democracy in Alberta,”; on the effects of electoral systems on party systems generally, see Duverger, Maurice, Political Parties (London, 1954)Google Scholar, and Lipset, , The First New Nation, esp. 301–2.Google Scholar But these views have possibly been overstated: it seems that the electoral mechanics act as a buttress rather than as a foundation of the party system and some evidence would indicate that the party systems determine the electoral systems rather than the opposite; see Lipson, “The Two-Party System in British Politics,” and mainly Grumm, John G., “Theories of Electoral Systems,” Midwest Journal of Political Science, 2 (1958), 357–76.CrossRefGoogle Scholar

⁴⁷ Democracy in Alberta, esp. chaps. 1 and 8.

⁴⁸ That the Prairies have developed a fundamental non-partisan tradition and have been consistently opposed to the old party system is challenged by the facts. As pointed out by Smith, the Liberal party in Saskatchewan, ruling almost without interruption from 1905 to 1944, was most of the time in open, even defiant, union with the federal Liberal party; and the Progressives in Manitoba allied themselves with the Liberals after 1928 and from 1931 on described themselves as a Liberal-Progressive coalition. See Smith, Denis, “Prairie Revolt, Federalism and the Party System,” in Thorburn, Hugh G., Party Politics in Canada (Toronto, 1963), 126–37.Google Scholar On the other hand, there have been groups explicitly rejecting the party system even outside the Western provinces; see for instance the discussions on this among Ontario farmers in Morton, W. L., The Progressive Party in Canada (Toronto, 1950), 75.Google Scholar

⁴⁹ “Democracy in Alberta.”

⁵⁰ Macpherson, , Democracy in Alberta, 217 Google Scholar

⁵¹ Ibid., 205.

⁵² The Conservatives had never won an election in that province and though, during that period, they kept slightly more than a third of the votes on the average (38.4 per cent) there is general agreement that they were very weak. The best indication of this is that in the 1921 election they practically withdrew from the scene, contesting only sixteen of the sixty-one seats. On this, see Thomas, L. G., The Liberal Party in Alberta: A History of Politics in the Province of Alberta, 1905–1921 (Toronto, 1959), esp. 171–2, 203–4Google Scholar; and Morton, , The Progressive Party in Canada, esp. 111 ff.Google Scholar The period from 1921 to 1935, on the other hand, was one of one-party dominance by the UFA, the Liberals retaining an average of 28.2 per cent of the votes only. A new party, Social Credit, came to power in 1935.

⁵³ See also Mallory, J. R., “The Structure of Canadian Politics,” in Thorburn, , ed., Party Politics in Canada, 22–9, esp. 24–5.Google Scholar

⁵⁴ There was definitely no consensus on this even among the élites, as the maintenance of the old parties clearly indicates. See Thomas, , The Liberal Party in Alberta, 3–4 Google Scholar, on the disagreements among Albertan élites about the appropriateness of party politics.

⁵⁵ Some of these parties, however, may one day profit from the fact that over time there are more and more people who have at least once supported them and therefore may find it more easy to do it again. If they were to do it simultaneously, these parties might be more fortunate.

⁵⁶ This last paragraph was written before the 1966 provincial election in Quebec. Our prediction was borne out by the results of that election: while the Union Nationale defeated the Liberals, the two new parties on the provincial scene, the Ralliement pour l'Indépendence Nationale (RIN), a separatist party, and the Ralliement National (RN), a créditiste and right-wing separatist alliance, both failed to recruit any serious following: the first got only 5.6 per cent of the vote, and the second only 3.2 per cent.