¹ The leading work of this sort thus far has been done by the late Key, V. O. Jr., See, e.g., his “A Theory of Critical Elections,” Journal of Politics, Vol. 17, pp. 3–18 (1955)CrossRefGoogle Scholar, and his American State Politics (New York, 1956)Google Scholar. See also such quantitatively oriented monographs as Howard, Perry, Political Tendencies in Louisiana, 1812–1952 (Baton Rouge: Louisiana State University Press, 1957)Google Scholar.

² The most notable survey-research effort to date to develop politically relevant theory regarding American voting behavior is Campbell, Angus, Converse, Philip E., Miller, Warren E. and Stokes, Donald E., The American Voter (New York, 1960)Google Scholar, especially ch. 20.

³ Key, V. O. Jr., “The Politically Relevant in Surveys,” Public Opinion Quarterly, Vol. 24, pp. 54–61 (1960)Google Scholar; Key, V. O. Jr., and Munger, Frank, “Social Determinism and Electoral Decision: The Case of Indiana,” in Burdick, Eugene and Brodbeck, Arthur J., eds., American Voting Behavior (Glencoe, Ill., 1959), pp. 281–99Google Scholar.

⁴ Miller, Warren E. and Stokes, Donald E., “Constituency Influence in Congress,” this Review, Vol. 57, pp. 45–56 (1963)Google Scholar. The authors observe that in the 1958 Hays-Alford congressional race in Arkansas, the normally potential nature of constituency sanctions against representatives was transferred under the overriding pressure of the race issue into an actuality which resulted in Hays' defeat by a write-in vote for his opponent. The normally low issue- and candidate-consciousness among the electorate was abruptly replaced by a most untypically intense awareness of the candidates and their relative postions on this issue.

For an excellent cross-polity study of voting behavior based on comparative survey analysis, see Alford, Robert R., Party and Society (Chicago, Rand McNally, 1963)Google Scholar.

⁵ Key, V. O. Jr., Public Opinion and American Democracy (New York, 1961)Google Scholar, based largely on survey-research data at the University of Michigan.

⁶ This effort, to which the author was enabled to contribute, thanks to a Social Science Research Council grant for 1963–64, has been supported by the Council and by the National Science Foundation. This article is in no sense an integral part of that larger project. But it is proper to acknowledge gratefully here that the S.S.R.C., by making it possible for me to spend a year at the Survey Research Center, has helped to provide conditions favorable to writing it. Thanks are also due to Angus Campbell, Philip E. Converse, Donald E. Stokes and Warren E. Miller for their comments and criticisms. They bear no responsibility for the defects of the final product.

⁷ V. O. Key, Jr., American State Politics, op. cit., pp. 71–73, 197–216.

⁸ Schattschneider, E. E., Party Government (New York, 1942)Google Scholar and The Semi-Sovereign People (New York, 1960), pp. 78–96Google Scholar.

⁹ Scarrow, Howard A., “Patterns of Voter Turnout in Canada,” Midwest Journal of Political Science, Vol. 5, pp. 351–64 (1961)CrossRefGoogle Scholar; Robinson, James A. and Standing, William, “Some Correlates of Voter Participation: The Case of Indiana,” Journal of Politics, Vol. 22, pp. 96–111 (1960)CrossRefGoogle Scholar. Both articles—one involving a political system outside of but adjacent to the United States—indicate patterns of contemporary participation which seem at variance with the conclusions of survey studies regarding the behavior of the American electorate. In Canada rural turnout is higher than urban, and no clear-cut pattern of drop-off between federal and provincial elections exists. Voter participation in Indiana apparently does not increase with the competitiveness of the electoral situation, and does increase with the rurality of the election jurisdiction. With the possible exception of the relationship between competitiveness and turnout, all of these are characteristics associated with 19th-contury voting behavior in the United States; see below.

¹⁰ In computing turnout data, note that until approximately 1920 the criteria for eligibility to vote differed far more widely from state to state than they do now. In a number of states west of the original thirteen—for example, in Michigan until 1894 and in Wisconsin until 1908—aliens who had merely declared their intention to become citizens were permitted to vote. Woman suffrage was also extended piecemeal for several decades prior to the general enfranchisement of 1920. The turnout estimates derived here have been adjusted, so far as the census data permit, to take account of such variations.

¹¹ If one computes the off-year total vote of the years 1950–62 as a percentage of the total vote cast in the preceding presidential election, a virtually identical correspondence is reached with estimated off-year turnout as a percentage of turnout in the immediately preceding presidential year:

¹² Campbell, Angus, “Surge and Decline: A Study of Electoral Change,” Public Opinion Quarterly, Vol. 24, pp. 397–418 (1960)CrossRefGoogle Scholar.

¹³ Ibid., p. 413. The percentage of drop-off from 1956 to 1958, as computed from aggregate voting data, was 25.6%.

¹⁴ See, e.g., Lane, Robert E., Political Life (Glencoe, Ill., 1959), pp. 18–26Google Scholar.

¹⁵ There are, of course, very wide divergences in turnout rates even among non-Southern states. Some of them, like Idaho, New Hampshire and Utah, have presidential-year turnouts which compare very favorably with European levels of participation. A detailed analysis of these differences remains to be made. It should prove of the utmost importance in casting light upon the relevance of current forms of political organization and partisan alignments to differing kinds of electorates and political subsystems in the United States.

¹⁶ Report of the President's Commission on Registration and Voting Participation (Washington, 1963), esp. pp. 5–9Google Scholar. Hereafter cited as Report.

¹⁷ Ibid., pp. 11–14, 31–42.

¹⁸ See, e.g., Kelley, Stanley, “Elections and the Mass Media,” Law and Contemporary Problems, Vol. 27, pp. 307–26 (1962)CrossRefGoogle Scholar.

¹⁹ E. E. Schattschneider, The Semi-Sovereign People, op. cit., pp. 102–3.

²⁰ La Palombara, Joseph, Guide to Michigan Politics (East Lansing, Mich., Michigan State University Press, 1960), pp. 22–35Google Scholar.

²¹ This recalls Robinson and Standing's conclusion that voter participation in Indiana does not necessarily increase with increasing party competition. Of the eight Michigan gubernatorial elections from 1948 to 1962 only one was decided by a margin of 55% or more, while three were decided by margins of less than 51.5% of the two-party vote. Despite this intensely competitive situation, turnout—while of course much higher than in the 1920s—remains significantly below normal pre-1920 levels.

²² Angus Campbell, “Surge and Decline,” op. cit., pp. 401–4.

²³ Tingsten, Herbert, Political Behavior (Stockholm, Stockholm Economic Studies, No. 7, 1937), pp. 10–36Google Scholar. See also Merriam, Charles E. and Gosnell, Harold F., Non-Voting (Chicago, University of Chicago Press, 1924), pp. 26, 109–22Google Scholar, for a useful discussion of the effect of woman suffrage on turnout in a metropolitan area immediately following the general enfranchisement of 1920.

²⁴ Survey-research estimates place current turnout among American women at 10% below male turnout. Angus Campbell et al., The American Voter, op. cit., pp. 484–85. This sex-related difference in participation is apparently universal, but is significantly smaller in European countries which provide election data by sex, despite the far higher European level of participation by both sexes. The postwar differential has been 5.8% in Norway (1945–57 mean), 3.3% in Sweden (1948–60 mean), and 1.9% in Finland (1962 general election). While in 1956 only about 55% of American women went to the polls, the mean turnout among women in postwar elections was 76.1% in Norway and 79.4% in Sweden.

²⁵ Ibid., pp. 402–40.

²⁶ The estimated rates of turnout in presidential elections from 1876 through 1896, mean turnout in the period 1936–60 and estimated turnout in 1964 were as follows in these states:

²⁷ V. O. Key, Jr., and Frank Munger, “Social Determinism and Electoral Decision,” op. cit., pp. 282–88.

²⁸ Mills, C. Wright, The Power Elite (New York, Oxford University Press, 1956), pp. 298–324Google Scholar. See also Vidich, Arthur J. and Bensman, Joseph, Small Town in Mass Society (New York, 1960), pp. 5–15, 202–27, 297–320Google Scholar.

²⁹ Fenton, John H., Politics in the Border States (New Orleans, Hauser Press, 1957), pp. 117–20Google Scholar.

³⁰ However, Ohio's modern pattern of splitticket voting, formed several decades ago, seems to have been little (if at all) affected by the 1950 change from party-column to office-block ballot forms. See Figure 3.

³¹ Flinn, Thomas A., “Continuity and Change in Ohio Politics,” Journal of Politics, Vol. 24, pp. 521–44 (1962)CrossRefGoogle Scholar.

³² Benson, Lee, The Concept of Jacksonian Democracy (Princeton, Princeton University Press, 1961), pp. 123–207, 288–328Google Scholar.

³³ Flinn, op. cit., p. 542.

³⁴ Fenton, John H., “Ohio's Unpredictable Voters,” Harper's Magazine, Vol. 225, pp. 61–65 (1962)Google Scholar.

³⁵ This would seem to suggest a limitation on Key's findings, American State Politics, op. cit., pp. 169–96.

³⁶ This designation is given the state's political system in Benson, Oliver, Holloway, Harry, Mauer, George, Pray, Joseph and Young, Wayne, Oklahoma Votes: 1907–1962 (Norman, Okla., Bureau of Government Research, University of Oklahoma, 1964), pp. 44–52Google Scholar. For an extensive discussion of the sectional basis of Oklahoma politics, see ibid., pp. 32–43, and V. O. Key, Jr., American State Politics, op. cit., pp. 220–22.

³⁷ In 1936, 34 states (71%) elected governors for either two- or four-year terms in presidential years, and the three-year term in New Jersey caused major state elections to coincide with every fourth presidential election. By 1964, only 25 of 50 states (50%) still held some of their gubernatorial elections in presidential years. Two of these, Florida and Michigan, are scheduled to begin off-year gubernatorial elections for four-year terms in 1966.

³⁸ American State Politics, op. cit., pp. 169–96.

³⁹ In the period 1956–62 there have been 840 general-election contests for the Pennsylvania House of Representatives. Of these all but six, or 0.7%, have been contested by both major political parties. No Pennslvania state Senate seat has been uncontested during this period. Despite the 1962 Republican upsurge in Oklahoma, however, there were no contests between the parties in 11 of 22 Senate seats (50.0%) and in 73 of 120 House seats (60.9%). All the uncontested Senate seats and all but two of the uncontested House seats were won by Democrats.

⁴⁰ Mean national partisan swings in presidential elections since 1872 have been as follows: 1872–92, 2.3%; 1896–1916, 5.0%; 1920–32, 10.3%; 1936–64, 5.4%.

⁴¹ If a presidential landslide is arbitrarily defined as a contest in which the winning candidate received 55% or more of the two-party vote, only the election of 1872 would qualify among the 16 presidential elections held from 1836 to 1896. Of 17 presidential elections held from 1900 through 1964, at least eight were landslide elections by this definition, and a ninth—the 1924 election, in which the Republican candidate received 54.3% and the Democratic candidate 29.0% of a three-party total—could plausibly be included.

⁴² The total vote in Georgia declined from 92,203 in 1848 to 62,333 in 1852. Estimated turnout declined from about 88% to about 55% of the eligible electorate, while the Democratic share of the two-party vote increased from 48.5% in 1848 to 64.8% in 1852. The pattern of participation in the Pennsylvania gubernatorial and presidential elections of 1872 is also revealing:

Estimated turnout in October was 82.0%, in November 68.6%. The Democratic percentage of the two-party vote was 47.3% in October and 37.9% in November.

⁴³ The only apparent exception to this generalization in the 19th century was the election of 1840. But this was the first election in which substantially full mobilization of the eligible electorate occurred. The rate of increase in the total vote from 1836 to 1860 was 60.0%, the largest in American history. Estimated turnout increased from about 58% in 1836 to about 80% in 1840. This election, with its relatively one-sided mobilization of hitherto apolitical elements in the potential electorate, not unnaturally bears some resemblance to the elections of the 1950s.

⁴⁴ The Semi-Sovereign People, op. cit., p. 81.

⁴⁵ Ibid., esp. pp. 78–113. See also his “United States: The Functional Approach to Party Government,” in Neumann, Sigmund, ed., Modern Political Parties (Chicago, University of Chicago Press, 1956), pp. 194–215Google Scholar.

⁴⁶ Kerr, Clark, Dunlop, John T., Harbison, Frederick S. and Myers, Charles A., Industrialism and Industrial Man (Cambridge, Harvard University Press, 1960), pp. 47–76, 98–126, 193, 233Google Scholar. Rostow, Walt W., The Stages of Economic Growth (Cambridge, Cambridge University Press, 1960), pp. 17–58Google Scholar.

⁴⁷ Duverger, Maurice, Political Parties (New York, 2d. ed., 1959), pp. 1–60Google Scholar.

⁴⁸ Chicago, Milwaukee & St. Paul Railway Co. v. Minnesota, 134 U. S. 418 (1890).

⁴⁹ Pollock v. Farmers' Loan & Trust Co., 157 U. S. 429 (1895); (rehearing) 158 U. S. 601 (1895).

⁵⁰ The literature on this process of judicial concept-formulation from its roots in the 1870s through its formal penetration into the structure of constitutional law in the 1890s is extremely voluminous. Two especially enlightening accounts are: Twiss, Benjamin, Lawyers and the Constitution (Princeton, Princeton University Press, 1942)Google Scholar, and Paul, Arnold M., Conservative Crises and the Rule of Law (Ithaca, Cornell University Press, 1960)Google Scholar.

⁵¹ Paul, ibid., pp. 131–58.

⁵² See, among many other examples, Congressional Quarterly Weekly Report, Vol. 22 (May 1, 1964), p. 801Google Scholar.

⁵³ Eulau, Heinz, “The Politics of Happiness,” Antioch Review, Vol. 16, pp. 259–64 (1956)CrossRefGoogle Scholar; Lipset, Seymour M., Political Man (New York, 1960), pp. 179–219Google Scholar.

⁵⁴ Ibid., pp. 216–19; Herbert Tingsten, Political Behavior, op. cit., pp. 225–26.

⁵⁵ Key, V. O. Jr., Southern Politics (New York, 1949), pp. 526–28Google Scholar; E. E. Schattschneider, The Semi-Sovereign People, op. cit., pp. 114–28.

⁵⁶ Levin, Murray B., The Alienated Voter (New York, 1960), pp. 58–75Google Scholar, and his The Compleat Politician (Indianapolis, 1962), esp. pp. 133–78Google Scholar. While one may hope that Boston and Massachusetts are extreme case studies in the pathology of democratic politics in the United States, it appears improbable that the pattern of conflict between the individual's expectations and reality is entirely unique to the Bay State.

⁵⁷ Angus Campbell et al., The American Voter, op. cit., pp. 143–45.

⁵⁸ The line of reasoning developed in this article—especially that part of it which deals with the possible development of political alienation in the United States—seems not entirely consistent with the findings of Almond, Gabriel A. and Verba, Sidney, The Civic Culture (Princeton, Princeton University Press, 1963), pp. 402–69, 472–505CrossRefGoogle Scholar. Of course there is no question that relatively high levels of individual satisfaction with political institutions and acceptance of democratic norms may exist in a political system with abnormally low rates of actual voting participation, just as extremely high turnout may—as in Italy—be associated with intense and activist modes of political alienation. At the same time, the gap between American norms and the actual political activity of American individuals does exist, as Almond and Verba point out on pp. 479–87. This may represent the afterglow of a Lockean value consensus in an inappropriate socio-economic setting, but in a polity quite lacking in the disruptive discontinuities of historical development which have occurred during this century in Germany, Italy and Mexico. Or it may represent something much more positive.

⁵⁹ “Whereas less than sixty-five percent of the United States population of voting age cast ballots for Presidential electors in 1960; and

“Whereas popular participation in Government through elections is essential to a democratic form of Government; and

“Whereas the causes of nonvoting are not fully understood and more effective corrective action will be possible on the basis of a better understanding of the causes of the failure of many citizens to register and vote …” (emphasis supplied) The full text of the executive order is in Report, pp. 63–64. Compare with Schattschneider's comment in The Semi-Sovereign People, op. cit., p. 112: “A greatly expanded popular base of political participation is the essential condition for public support of the government. This is the modern problem of democratic government. The price of support is participation. The choice is between participation and propaganda, between democratic and dictatorial ways of changing consent into support, because consent is no longer enough.” (Author's emphasis)