¹ Borrowing tools from other disciplines risks involving one in “foreign” debates which, important as they be for the discipline from which one borrows, are of little significance to one's own. Structuralist linguists disagree on the role of semantics: some prefer to exclude it from the analysis of language, others insist that since the purpose of language is to transmit meaning, one should relate signs or sounds to meaning. For discussions on the various approaches to the study of language, see Hertzler, Joyce O., A Sociology of Language (New York, 1965) esp. 90 ff.Google Scholar For arguments in favour of semantics, see Bloomfield, Leonard, Language (New York, 1933).Google ScholarPubMed For applications of the techniques and findings of linguistics to the study of social structures, see among others Levi-Strauss, Claude, “L'analyse structurale en linguistique et en anthropologie,” in Anthropologie Structurale (Paris, 1958).Google Scholar See also Bastide, R., “Sens et usages du terme structure dans les sciences humaines et sociales,” Janua linguarium, 1960–1961, no. 16.Google Scholar

² See Normandin, G., ed., Parliamentary Guide (Ottawa), published yearly since 1914. Prior to that date publication was not as regular. There were, for example, no issues in either 1906 or 1907Google Scholar; hence the selection of 1908 in our sample instead of 1907.

³ Scarrow, H. A., Canada Votes: A Handbook of Federal and Provincial Elections Data (New Orleans, 1962).Google Scholar

⁴ I made two exceptions to this rule of respecting diversity, (a) I always considered the label which appears in the English, to the exclusion of the French, biographies given by the Parliamentary Guide. Ideally, I would have taken the label of French-speaking legislators from the French text of their biographies, but unfortunately bilingualism in the Parliamentary Guide dates from the post-war period, and to maintain comparability with earlier years, I had to draw all the labels from the English text. This concession to standardization is not, however, dramatic and occasionally has some advantages: for example, R. Caouette, who before, as after, 1965 described himself as Ralliement des Créditistes in the French version of the Parliamentary Guide changed from Social Credit to Ralliement des Créditistes in the English text—an important evolution recorded in this case by the English version and not by the French original, (b) The second exception is not having recorded abbreviations (a practice which developed only after the Second World War), as symbols distinct from the names they stood for. I considered P.C. or PC to be equivalent to Progressive-Conservative. Studying abbreviations would have complicated the presentation; not having studied them, however, results in the loss of data (such as the speed of contraction and standardization) which would have enabled us to make a slightly more refined analysis. In 1957, the last of our sample years preceding the standardization of political party labels by the editors of the Parliamentary Guide, 19 per cent of the symbols appear in an abbreviated form (8 per cent in the federal parliament; 25 per cent in the provincial houses). There are two extreme cases: Liberal is never shortened; Co-operative Commonwealth Federation is shortened to CCF 99 per cent of the time. The Progressive Conservative and the Social Credit symbols fall between these two extremes, first in that they have respectively a 23 per cent and an 18 per cent rate of shortening and also in that the Conservative symbol, when shortened, has no single standard form; in the order of popularity they are: Prog. Cons., PC or P.C, Prog. Conservative, Pro. Conservative

⁵ Founded in 1920 by former Boy Scout leaders, the Kibbo-Kift movement rejected industrial civilisation and sought revitalisation of the individual in rites, costumes, and the simple life of woodmanship and handicraft; the movement was anti-democratic and anti-parliamentary. See Macpherson, C. B., Democracy in Alberta (Toronto, 1953), 130–32.Google Scholar

⁶ Justification for sampling systematically rather than randomly is in the greater economy of the former technique. Also important is the fact that since elections do not occur at regular intervals, sampling from all Parliamentary Guides, or even sampling from the Guides of the years following each federal election, would have introduced a bias eliminated in our sample. By jumping decades we insured that at least one election had occurred, federally as well as provincially, between two successive years among those retained for our analysis. Had I analyzed all the parliamentary guides following each new federal election, the proportion of Liberals to Conservatives would have been 53 to 46; in the sample I used it is 60 to 40.

⁷ Under the leadership of H. H. Stevens, a former Conservative, the Reconstruction party obtained nearly 400,000 votes in the election of 1935. See Granatstein, J. L., The Politics of Survival: the Conservative Party of Canada, 1939–1945 (Toronto, 1967), 8–9.Google Scholar See also Wilburn, J. R. H., “H. H. Stevens and the Reconstruction Party,” Canadian Historical Review, 45 (March 1964).Google Scholar

⁸ The Liberal-Conservative party is politically ancestor to the Conservatives; that the symbol Liberal-Conservative dominates Canadian politics at the time of the founding of the confederation and that this form unites the two dominant English political symbols of the time deserves attention. Unity, union, denial of internal division, appear as characteristics both of the political right and of a nation in the making.

⁹ For the origins of the Liberal group see Bradshaw, F., Self-Government in Canada (London, 1903), 49 ff.Google Scholar

¹⁰ For studies of the Progressive movement see Morton, W. L., The Progressive Party in Canada (Toronto, 1950)Google Scholar, and Lipset, S. M., Agrarian Socialism (Berkeley, 1950).Google Scholar

¹¹ See Beck, J. M. and Dooley, D. J., “Party Images in Canada,” Queen's Quarterly, 67, no. 3 (1960) 446Google Scholar; see also Granatstein, The Politics of Survival, 144.

¹² See Scarrow, Canada Votes, 216.

¹³ Except when there is need of contrast or comparison, compound words are usually shortened by omission of the qualifier, the headword surviving by itself. See Gustafstern, , Meaning and Change of Meaning (Bloomington, 1931), 267 ff.Google Scholar

¹⁴ The assumption that words are facts related to but distinct from what they seek to describe is basic to modern semantics. For a discussion of Korzybskian and Aristotelian semantics, see among many others Rapoport, A., “What Is Semantics?” in Hayakawa, S. I., ed., Language, Meaning and Maturity (New York, 1954).Google Scholar A popularized but clear presentation of Korzybski's system is given in Chase, Stuart, The Tyranny of Words (New York, 1938),Google Scholar and Power of Words (New York, 1953).

¹⁵ To the religious individual Jesus Christ becomes either Jesus or Christ; the success of the National Socialist party is measured in its having become Nazi. Time and usage have polished “God be with ye” into “bye.” Economy is, of course, the major factor explaining contractions and shortening of names. A composite name that resists simplification is thus an oddity, like a two-headed or multi-souled individual. For the tendency of the language to shorten verbal and written signs see Salomon, L. B., Semantics and Common Sense (New York, 1966), 31–5Google Scholar; or Bloomfield, Language, 381.

¹⁶ When it changed its name to New Democracy, the newly founded Social Credit party had been steadily declining in its stronghold of Alberta. In the provincial election of August 1935 it had obtained 54 per cent of the vote; in the October 1935 federal election 48.2 per cent; in the March 1940 provincial election 43 per cent; in the March 1940 federal election 34.5 per cent. Similarly, the CCF vote had declined steadily between 1945 and 1961, when the name was changed to NDP, from 15.6 to 13.4 to 11.3 to 10.7 to 9.5; measured in terms of seats the decline was even more significant—from 28 to 8. Similarly, the Conservatives who had had 48.7 per cent of the votes in 1930 had declined to 29.6 in 1935. They changed their name, and running as National Government they obtained only 30.7 in 1940; changing the name again, to Progressive Conservative this time, they declined further to 27.4 in 1945. The figures are from Scarrow, Canada Votes.

¹⁷ See Bloomfield, Language, 400 ff; Salomon, Semantics and Common Sense, 31–5; Hoenigswald, H. M., Language Change and Linguistic Reconstruction (Chicago, 1960), 65.Google Scholar

¹⁸ Ibid. For an example of a word changing its sign from positive to negative take sinistra in Latin, it originally meant the side of the good omens, but being associated with the left, weak, evil side of things it took the negative connotation which has engendered sinister. For this and other examples of the effect of the left taboo, see Wile, Ira S., Handedness, Right and Left (Boston, 1934).Google Scholar

¹⁹ The label Radical appears in the Legislative Assembly of 1821. See Bradshaw, Self-Government in Canada, 113.

²⁰ Only during the period 1815–48 was the term Liberal widely used in France. Later on, while not disappearing, the term passed to the background. One finds it then in such labels as the 1860 Liberal Republican party of L. Gambetta (interestingly, Republican soon dominated Liberal in that compound); in the Alliance Républicaine and in the Alliance Libérate Populaire of the third republic or, more recently, during the 1950s in the splinter parti Libéral of J. P. David. However, neither by itself nor as a qualifier has Liberal played a major symbolic role since the middle of the nineteenth century.

²¹ This was of great importance to Laurier after the condemnation of liberalism by the Holy See. Quebec Conservatives supported by the clergy sought to identify the Liberals as an anti-papacy group, hence the distinction repeatedly made by Laurier between “Continental,” “Social,” or “Catholic” Liberalism on the one hand and “English” or “Political” Liberalism on the other. See Wade, Mason, The French Canadians, 1760–1945 (Toronto, 1956), 360 ff.Google Scholar The association of Canadian liberalism with French liberalism is clear only when, following the French Revolution of 1848, Papineau's influence marks the Liberal leadership with a clearly revolutionary and anti-clerical character. This did not last, however, and following an enquiry by an envoy from Rome, the Catholic Quebec hierarchy issued in 1877 a pastoral letter declaring that the Holy See's censure against Catholic Liberalism was not to be applied to any particular party. See Ibid., 369. See also Pickersgill, J. W., The Liberal Party (Toronto, 1962), 12.Google Scholar

²² As far as I can tell the political usage of bleu, blanc, rouge, and the bleu-rouge dualism dates from the French Revolution—white being the colour of the monarchy, and, to grossly simplify, blue that of the bourgeoisie, red that of the proletariat. Originally, both blue and red were republican symbols—blue because it was the colour of the republican army, red because it was the colour of the Phrygian bonnet, a popular republican symbol. More basically, blue and red, the colours of the flag of Paris had been added to the monarchical white to produce the tricolour. In Quebec, following the hierarchy's interdict on attacking political parties from the pulpit, some priests are reported to have used the reminder that if the sky is blue (the Conservative colour), hell is red (the Liberal colour). See Wade, The French Canadians, 370.

²³ In keeping with my original intention and borrowing the technique from geographical-linguists I do not try to determine whether these terms migrated directly or indirectly, or whether borrowing from the United States was conscious or not. I cannot and need not prove that the word Democrat, which Canadian socialists added in 1961 at a founding convention to what had already for a few years been unofficially known as the New Party, was borrowed consciously from the United States. At the level of my analysis I can ignore the intention of individuals. I infer from geographical distributions as in a different context one infers from statistical aggregates. The advantage of semantics for our analysis is that it forces one to take a global and a historical view of one's data. In Durkbeim's words, the sea can only be seen and understood from a distance. For a closer view at the genesis of the New Democratic Party see Knowles, Stanley, The New Party (Toronto, 1961).Google ScholarPubMed

²⁴ At one time during the period covered by our study, the Liberals were the party seeking to hide its name. At the end of the nineteenth century, for a short while during the electoral decline which followed the devaluation of the term Liberal because of the Holy See's condemnation of Liberalism, Quebec Liberals hid under the symbol National.

²⁵ Sapir uses another metaphor; he sees words as a cloud of points, the language being the axis containing them. “To pass from one language to another is psychologically parallel to passing from one geometrical system of reference to another… the world of points is the same for either frame of reference. But the formal method of approach to the expressed item of experience, as to the given point of space, is so different that the resulting feeling of orientation can be the same neither in the two languages nor in the two frames of reference.” See Mandelbaum, D. G., Selected Writings of Edward Sapir (Berkeley, 1951), 153.Google Scholar

²⁶ Note that provincial originality is not defined by the origins of symbols but by the exclusiveness of their diffusion within the provincial field.

²⁷ Historical linguists note that political boundaries are among the most powerful predictors of linguistic differences when one maps linguistic cleavages. See Bloomfield, Language, 343.

²⁸ The diffusion of languages and forms of speech is related to differences in the prestige and status of individuals. Ibid., 403.