1 Deutsch, Karl W., “Social Mobilization and Political Development,” American Political Science Review 55 (1961), 494CrossRefGoogle Scholar.

2 Rokkan, Stein, with Campbell, Angus, Torsvik, Per and Valen, Henry, Citizens, Elections, Parties: Approaches to the Comparative Study of the Processes of Development (New York: David McKay, 1970), 238–39Google Scholar.

3 See, for example, Lerner, Daniel, “Toward a Communications Theory of Modernization: A Set of Considerations,” in Pye, Lucien (ed.), Communications and Political Development (Princeton: Princeton University Press, 1963), 327–50Google Scholar; Huntington, Samuel P., Political Order in Changing Societies (New Haven: Yale University Press, 1968)Google Scholar. See also essays by Stein Rokkan entitled: “Methods and Models in the Comparative Study of Nation-Building,” “Nation-Building, Cleavage Formation and the Structuring of Mass Politics,” “The Mobilization of the Periphery,” and “Electoral Mobilization, Party Competition and National Integration,” in Rokkan et al., Citizens, Elections, Parties. Additional writings by Deutsch on the relationship between communications and political development include his Nationalism and Social Communication: An Inquiry into the Foundations of Nationality (Cambridge: MIT Press, 1953)Google Scholar; and his contribution in Jacob, Philip E. and Toscano, James V. (eds.), The Integration of Political Communities (Philadelphia: J. B. Lippincott, 1964)Google Scholar.

4 Stokes, Donald E., “Parties and the Nationalization of Electoral Forces,” in Burnham, Walter Dean and Chambers, William Nisbet (eds.), The American Party System: Stages of Political Development (New York: Oxford University Press, 1967), 182–202Google Scholar; and Stokes, Donald E., “A Variance Components Model of Political Effects,” in Claunch, John M. (ed.), Mathematical Applications in Political Science (Dallas: Arnold Foundation, 1965), 61–85Google Scholar.

5 For further details on British Columbia geography see: Robinson, J. Lewis (ed.), Studies in Canadian Geography: British Columbia (Toronto: University of Toronto Press, 1972)Google Scholar; and Robinson, J. Lewis and Hardwick, Walter, “The Canadian Cordillera,” in Warkentin, J. (ed.), Canada: A Geographical Interpretation (Toronto: Methuen, 1968)Google Scholar, chap. 13.

6 See infra, 522–23, for further discussion of these developments.

7 On settlement of the province see Alfred H. Siemens, “Settlement,” in Studies in Canadian Geography: British Columbia, 9–31; and Norris, John, Strangers Entertained: A History of the Ethnic Groups of British Columbia (Vancouver: Evergreen Press, 1971)Google Scholar. A few communities based on traditional ethnic solidarities did take root. Examples include Doukhobor communities in the Kootenays, and the Finnish community of Sointula on the coast.

8 Stokes, “Parties and the Nationalization of Electoral Forces,” and “A Variance Components Model.” In operationalizing his model Stokes introduces a procedure based on two-way analysis of variance. Applying this procedure to variation in party support across districts and elections, he reaches estimates of the importance of national, regional and local electoral forces during periods spanning several elections. We are forced to alter this procedure because a condition necessary for its application—the occurrence of multiple-election “runs” between constituency boundary adjustments—is not satisfied for the pre-World War Two period. A criticism of the Stokes model is put forward in Katz, Richard S., “The Attribution of Variance in Electoral Returns: An Alternative Measurement Technique,” American Political Science Review 67 (1973), 817–28CrossRefGoogle Scholar. See also Stokes's response, “Comment: On the Measurement of Electoral Dynamics,” ibid., 829–31; and Katz, “Rejoinder to ‘Comment by Donald E. Stokes,'” ibid., 832–34. Fora discussion of the bearing of the Katz arguments on our findings see this author's thesis, “The Impact of Modernization on British Columbia Electoral Patterns: Communications Development and the Uniformity of Swing, 1903–1975” (Ph.D. dissertation, University of British Columbia, 1978), 33–46Google Scholar.

9 For a discussion which compares percentage point and proportionate measures of swing see Wilson, “The Impact of Modernization,” 46–57.

10 Twenty-two provincial general elections occurred between 1903, when parties first competed in provincial elections, and 1975. Fifteen of the 21 possible inter-election swings are analyzed. The swings of 1916, 1924, 1933, 1941 and 1966 are excluded because boundary adjustments intervened between these elections and the previous ones. The swing of 1912 is excluded because of the high number of acclamations. Two of the pairs which are included (1933–37 and 1953–56) also flank boundary changes but adjustments involved in these two redistributions affected limited number of constituencies. Ridings affected were dropped from analysis of constituency swings in these elections. Data for the 1903, 1907, 1909 and 1916 elections are from The Canadian Parliamentary Guide. Certain errors and omissions in the returns given for the 1903, 1907, and 1909 elections were corrected using results given in Gosnell, R. E., The Yearbook of British Columbia and Manual of Provincial Information (Victoria: n.p., 1911), 76–80Google Scholar. Data for 1920 and 1924 were from British Columbia, Department of Provincial Secretary, British Columbia General Election, 1920 (and 1924) Statement of the Polls as Certified by Returning Officers on Completion of Final Count. Data for all other electons were taken from British Columbia, Chief Electoral Officer, Statement of Votes: General Election (Victoria, various dates). First-ballot results of the 1952 and 1953 elections were used. In all elections, party percentages in multimember ridings were estimated by first dividing the total votes obtained by the party's candidates by the number of members to be returned from the constituency, and then dividing the result by the figure reached when the total number of votes cast was divided by the number of members to be returned.

11 The sampling strategy can be described as stratified sampling within constituencies. We attempted to include all major nonmetropolitan communities along with a sample of smaller places. The sample was chosen from communities which were listed as polling stations in the 1960 election. In its final form the community sample included about 375 cases, with this number representing close to one-half of all nonmetropolitan communities with polling stations. Those chosen included most places that had more than 500 voters registered forthe 1960 election, about one-half of communities with between 100 and 250 voters (in 1960), and about one-third of communities with fewer than 100 voters. Poll data were obtained from British Columbia, Chief Electoral Officer, Statement of Votes. Poll level results for the 1975 election were not available when this part of the analysis was carried out. Poll boundaries have never been rigidly defined in nonmetropolitan constituencies. For example, the description of polls in these constituencies after the 1966 redistribution reads: the constituency is “divided into the following polling stations, being the places named and the territory tributary to them from the standpoint of accessibility.” Although exact comparability of areas across successive elections cannot be assumed, the correspondence should be very close.

12 The period demarcation points were chosen with an eye to having each period include equal numbers of elections. The 1933 and 1952 elections provide natural cut-off points since each marked the beginning of a distinct period in the history of the party system. The CCF first ran candidates in 1933, while 1952 marked the arrival of Social Credit as a major force and the start of the period in which Social Credit and the CCF/NDP were the principal protagonists. The swings examined in each period were as follows: for the 1903–1928 period, Liberal and Conservative swings; for the 1933–1952 period. Liberal and CCF swings in 1937, Coalition and CCF swings in 1945, 1949, and 1952 (with Coalition swing in the 1952 election taken as the difference between Coalition support in 1949 and the combined support of the Liberals and Conservatives in 1952); for the 1953–1963 and 1966–1975 periods, Social Credit and CCF/NDP swings.

13 If we make the (unrealistic) assumption that swing is produced by vote switching then we can see that in the very small community, switches in preference by a handful of voters could lead to a large percentage point swing. In a larger community a swing of the same magnitude would require switches by large numbers of voters. Ourargument is that the change in the small community is much more likely. The small community is more homogeneous so all of its voters could easily be influenced by one opinion leader. Indeed, all of its voters could belong to one family or logging crew. Thus we would expect to see more extreme variation in swing across a sample containing small places. After communities with fewerthan 100 voters are excluded, the sample ranges in size from 87 communities in 1937 to 199 in 1972.

14 With the sample limited in this way the average populations of the communities included range between 140 and 143.

15 See, for example, Black, E. R., “British Columbia: The Politics of Exploitation,” in Shearer, R. A. (ed.), Exploiting Our Economic Potential: Public Policy and the B.C. Economy (Toronto: Holt, Rinehart and Winston, 1968), 25Google Scholar; Robin, Martin, “The Politics of Class Conflict,” in Robin, Martin (ed.), Canadian Provincial Politics: The Party Systems of the Ten Provinces (Scarborough: Prentice-Hall, 1972), 36–37Google Scholar; Resnick, Philip, “The Political Economy of B.C.: A Marxist Perspective,” in Knox, Paul and Resnick, Philip (eds.), Essays in B.C. Political Economy (Vancouver: New Star, 1974), 3–12Google Scholar; Marchak, Patricia, “Class, Regional and Institutional Sources of Social Conflict in B.C.,” B.C. Studies, No. 27 (1975), 39–49Google Scholar; and Koenig, Daniel J. and Proverbs, Trevor B., “Class, Regional, and Institutional Sources of Party Support Within British Columbia,” B.C. Studies, No. 29 (1976), 19–28Google Scholar.

16 The fact that Victoria is the political metropolis and Vancouver the economic metropolis also advises that the two should be distinguished. On the historical tensions between Vancouver and Victoria, see Ormsby, Margaret, British Columbia: A History (student edition; Toronto: Macmillan of Canada, 1958), 262–97Google Scholar. For a discussion of integration between Vancouver and Victoria, see Walter G. Hardwick, “The Georgia Strait Urban Region,” in Robinson (ed.), Studies in Canadian Geography: British Columbia, 119–33. On regions in the nonmetropolitan area see J. Lewis Robinson, “Areal Patterns and Regional Character,” ibid., 1–8; and K. G. Denike and Roger Leigh, “Economic Geography, 1960–1970,” ibid., 69–86.

17 For a discussion of the factors considered in arriving at this division, and a list of ridings included in each region, see Wilson, “The Impact of Modernization,” 189–92.

18 Before examining the communications thesis we tested four hypotheses which propose straightforward alternatives to the view that communications change, or any other facet of societal modernization, caused the trend to uniformity. Each of these “safety check” hypotheses was set aside after consideration of the evidence. See Wilson, “The Impact of Modernization,” 87–106, 205–14.

19 Surfaced highway is that which is paved, or covered with gravel or crushed stone. See Canada, Dominion Bureau of Statistics (Statistics Canada), The Canadian Yearbook, various years. Rapid growth of paved highway mileage, so often associated with the 1950's, actually characterized most of the period between 1930 and 1960. Paved mileage grew from about 350 miles in 1931 to over 1,500 miles in 1941, and then remained fairly static for the duration of the war. By 1946 the “blacktop” boom was on again. Paved mileage doubled between 1946 and 1956, and more than tripled between 1946 and 1961.

20 For 1921 to 1951, see Urquhart, M. C. (ed.), Historical Statistics of Canada (Toronto: Macmillan, 1965)Google Scholar, Table S235; and for 1961 and 1971, Canada, Dominion Bureau of Statistics (Statistics Canada), The Canada Yearbook, 1963, 1973.

21 Data on gasoline consumption are from Canada, Dominion Bureau of Statistics (Statistics Canada), The Canada Yearbook, various years. Changes in motor vehicle efficiency, or changes in the proportion of fuel consumed by construction or tourist vehicles, are two factors which may make problematic inferences based on fuel consumption.

22 Calculated from data in Canada, Air Transport Board, Origin and Destination Statistics: Mainland Scheduled Traffic Survey of Revenue Passengers, 1955–1959 (Ottawa: 1961)Google Scholar; Canada, Air Transport Board (Committee), Airline Passenger Origin and Destination Statistics: Domestic Report, 1966 (1967) (Ottawa, n.d.); and Canada, Statistics Canada, Airline Passenger Origin and Destination Statistics: Domestic Report, 1975 (Catalogue No. 51–204) (Ottawa: 1976). For details on the estimation procedures used, see the introductory or explanatory notes provided in these reports. The British Columbia airports included in the 1955–1959 statistics (and thus those used as a basis for our comparisons) were: Castlegar, Cranbrook, Fort Nelson, Fort St. John, Kamloops, Penticton, Prince George, Prince Rupert, Quesnel, Sandspit, Smithers, Terrace and Williams Lake.

23 For 1941, see Canada, Dominion Bureau of Statistics, The Census of Canada, 1941, Vol. 9, Table 18; for 1951, see Canada, Dominion Bureau of Statistics, The Census of Canada, 1951, Vol. 3, Table 40. For 1961 and 1971, see Canada, Dominion Bureau of Statistics (Statistics Canada), Household Facilities and Equipment (Catalogue No. 64–202), May 1961 and May 1971. The 1961 and 1971 figures are estimates based on surveys of household facilities carried out in conjunction with quarterly labour force surveys.

24 The fact that local calling areas (toll-free areas) have expanded in geographical size over the last 25 years means that this figure underestimates the growth in extra-local calling. Data on telephone traffic volume were provided by Mr. Tony Farr, company historian, B.C. Telephone Company. Unfortunately, data for years before 1950 have been destroyed by the company. The figure for 1951 is an estimate calculated by multiplying the total number of long distance calls in 1951 by the proportion of total long distance calls which, according to data available after 1959, are made between points within the province.

25 See for 1931, Canada, Dominion Bureau of Statistics, The Census of Canada, 1931, Vol. 5, Table 57; for 1941, Canada, Dominion Bureau of Statistics, The Census of Canada, 1941, Vol. 9, Table 18; for 1951, Canada, Dominion Bureau of Statistics, The Census of Canada, 1951, Vol. 3, Table 40; for 1961 and 1971, Canada, Bureau of Statistics (Statistics Canada), Household Facilities and Equipment, May 1961 and May 1971. The figures for 1931–1951 are percentages of occupied dwellings while those for 1961–1971 are for households.

26 See Canada, Bureau of Statistics (Statistics Canada), Household Facilities and Equipment (Catalogue No. 64–202), September 1953, May 1959, May 1965.

27 As new means of conveyance and communication were adopted, dependence on devices such as trains, coastal and lake steamers, the telegraph, and the mails, declined. Also it cannot be assumed that behaviour (for example mobility, interaction with extra-local people, or movement of politicians) always changed in response to technological change. Where data have allowed a check on the assumption, the evidence indicates a strong correlation, but in other cases it may be that technological change simply made it easier (or faster) to maintain traditional patterns of behaviour.

28 The form of the ballot was changed in two stages. The first, legislated in 1921, provided for identification of candidates’ affiliation in only the multimember Vancouver City and Victoria City ridings. The second and more significant change came in 1939 and provided for inclusion of affiliation on the ballot in all ridings. See British Columbia, Statutes, 1921 (First Session), Chap. 17, “An Act to Amend the ‘Provincial Elections Act,’” Section 9; and British Columbia, Statutes, 1939, Chap. 16, “An Act Respecting Elections of Members of the Legislative Assembly,” Sections 88 (4) and 88 (5).

29 Data for 1939–1964 are from the first editions in each year of Canadian Advertising (Toronto: Maclean Hunter, 1939–1964)Google Scholar, while those for subsequent years are from Canadian Advertising: Rates and Data (Toronto: Maclean-Hunter and Standard Rate and Data Service, 1969, 1975)Google Scholar.

30 See note 29 for sources.

31 For example, the percentage of dwellings with telephones in 1951 ranged from over 45 per cent in the Okanagan to less than 20 per cent in the North Central and Peace River regions. See Canada, Dominion Bureau of Statistics, The Census of Canada, 1951, Vol. 3, Table 41. Or, to take another example, by 1961 television was found in about 70 per cent of dwellings in the Okanagan and 50 per cent of those in the Kootenays, but in less than 40 per cent of dwellings in the Peace and in only about 10 per cent of those in the North Central region. By 1971, these discrepancies had been almost totally erased. See Canada, Dominion Bureau of Statistics, The Census of Canada, 1961, “Living Conveniences” (Catalogue No. 93–527), Table 56; and Canada, Statistics Canada, The Census of Canada, 1971, “Household Facilities” (Catalogue No. 93–737), Table 17.

32 The within-region uniformity of swing in a given election is indicated by dispersal of community swings around the mean swing in that region. In calculating the dispersal rates for an election we average across standard deviations obtained in separate analyses of government party and main opposition party swings.

33 See note 32.

34 On the party system prior to 1933 see Dobie, Edith, “Party History of British Columbia, 1903–1933,” Pacific Northwest Quarterly 27 (1936), 153–66Google Scholar. On the development of the Conservatives prior to 1933 Parker, Ian Donald, “Simon Fraser Tolmie and the British Columbia Conservative Party, 1916–1933” (M.A. thesis, University of Victoria, 1970)Google Scholar; and Alper, Donald Keith, “From Rule to Ruin: The Conservative Party of British Columbia, 1928–1954” (Ph.D. dissertation, University of British Columbia, 1976)Google Scholar, chaps. 1 and 2. On the Liberals prior to 1933 see Sutherland, John Neil, “T. D. Pattullo as a Party Leader” (M.A. thesis, University of British Columbia, 1960)Google Scholar, chaps. 1 and 2.

35 On the British Columbia CCF's first two decades see Roberts, Dorothy June, “Doctrine and Disunity in the British Columbia Section of the CCF, 1932–1956” (M.A. thesis, University of Victoria, 1972)Google Scholar; and Sanford, Thomas Michael, “The Politics of Protest: The Cooperative Commonwealth Federation and the Social Credit League in British Columbia” (Ph.D. dissertation, University of California, Berkeley, 1961)Google Scholar, chap. 4.

36 See Black, Edwin Robert, “The Progressive Conservative Party in British Columbia: Some Aspects of Organization” (M.A. thesis, University of British Columbia, 1960), 29–34Google Scholar; and Alper, “From Rule to Ruin,” chap. 3.

37 See Alper, “From Rule to Ruin,” chaps. 4 and 5; Alper, “The Effects of Coalition Government on Party Structure: The Case of the Conservative Party in B.C.,” B.C. Studies, No. 33 (1977), 40–49Google Scholar; and Black, Edwin R., “Federal Strains Within a Canadian Party,” Dalhousie Review 45 (1965), 307–23Google Scholar.

38 See Horsfield, Barbara, “The Social Credit Movement in British Columbia” (B.A. essay, University of British Columbia, 1953)Google Scholar, chaps. 5 and 6; Sanford, “The Politics of Protest,” chaps. 6 and 7; Sherman, Paddy, Bennett (Toronto: McClelland and Stewart, 1966)Google Scholar, chaps. 4–6; and Robin, Martin, Pillars of Profit: The Company Province, 1934–1972 (Toronto: McClelland and Stewart, 1973)Google Scholar, chaps. 5–7.

39 On the nature of politics in British Columbia after 1952 see Edwin R. Black, “British Columbia: The Politics of Exploitation,” 23–41; Smiley, D. V., “Canada's Poujadists: A New Look at Social Credit,” Canadian Forum, September 1962, 121–23Google Scholar; Sherman, Bennett, chaps. 5–12; Robin, Martin, “The Social Basis of Party Politics in British Columbia,” Queen's Quarterly 72 (1966), 675–90Google Scholar; and Martin Robin, “British Columbia: The Politics of Class Conflict,” 27–68. An argument with certain points of similarity to the one advanced here is in Sproule-Jones, Mark, “Social Credit and the British Columbia Electorate,” B.C. Studies, No. 12 (1972–1973), 34–45Google Scholar.

40 Philbrick, Allen K., “Principles of Areal Functional Organization in Human Geography,” Economic Geography 33 (1957), 299–336CrossRefGoogle Scholar.

41 Rokkan, Stein, Foreword to Cox, Kevin R., Reynolds, David R. and Rokkan, Stein (eds.), Locational Approaches to Power and Conflict (New York: Sage Publications, 1974), 9Google Scholar.