Tolerance of Free Speech in the United States and West Germany, 1970-79: An Analysis of Public Opinion Survey Data

Abstract

In previous research on political tolerance in West Germany I found that respondents' level of education had no statistically significant effect on a number of variables measuring tolerance. This result seemed to contradict standard findings for populations of democratic countries. Therefore, article explicitly compares tolerance of free speech for nonconformists in America using replications of Samuel Stouffer's benchmark questions (National Opinion Research Center surveys in 1972 and 1977) with very similar-but not identical-questions asked in West German national surveys in 1970 and 1979. These questions concern tolerancefor a communist, an atheist, and a neo-Nazil militarist speaker. Using log-linear models to analyze scales of these questions and of the disaggregated tolerance items, the initial findings were confirmed: education has little effect on tolerance in West Germany, but considerable effect in the U.S. The influence of generational cohort, occupation, left-right ideological self-placement, party preference, country, and time are also tested. Semantic and historical explanations for the findings are briefly discussed. A thirty-year accumulation of empirical research in liberal democracies has impelled observers to the conclusion that higher levels of education are positively related to higher levels of liberalism-especially political and social tolerance. This relationship has often been interpreted as showing the influence of educational institutions in instilling the dominant (liberal) norms of the society. Although there are several other possible interpretations (which will be briefly discussed), one provides a reasonable starting point for treating the phenomenon in a comparative and historical perspective, as will be done here. ? 1982 The University of North Carolina Press. 0037-7732/82/040973-92$02.00 973 *I am grateful for the kind assistance of the Krupp Foundation, Essen, and of the Zentrum far Umfragen, Methoden und Analysen, Mannheim, and of the Center's directors, Rudolf Wildenmann and Max Kaase. I also thank James A. Davis, Sidney Verba, and Thomas Pettigrew for their comments on earlier versions of paper, one of which was presented at the 1980 meetings of the American Sociological Association. This content downloaded from 157.55.39.243 on Wed, 05 Oct 2016 04:07:05 UTC All use subject to http://about.jstor.org/terms 974 I Social Forces Volume 60:4, June 1982 The seeming consistency of these findings in several Western countries, above all the United States, has led some to generalize relationship. One of the clearest statements of view is given by Lipset: Data gathered by public opinion research agencies which have questioned people in different countries about their beliefs on tolerance for the opposition, their attitudes toward ethnic or racial minorities, and their feelings for multi-party as against one-party systems have showed that the most important single factor differentiating those giving democratic responses from the others has been education. The higher one's education, the more likely one is to believe in democratic values and support democratic practices (39-40). And Lipset's view from twenty years ago is still widely held today. A recent study of tolerance in America in the 1970s cites passage and concludes that this assessment continues to describe research evidence accurately in 1977 (Nunn et al., 58). To substantiate his claim, Lipset makes an explicit comparison between the U.S. and Germany. In his well-known (and still controversial) chapter on Working-Class Authoritarianism in Political Man, he cites evidence on tolerance for the rights of dissidents to speak publicly in the United States and evidence on support for a more-than-one party system in Germany to illustrate the theory that the better educated are more democratic, more liberal, or less authoritarian than the rest of the population (100-3, cf. Tables 3 and 4). It is well known, however, that the use of different indicators to measure the same phenomenon across time or place is a tricky business. In case, one could argue that these two items-dissidents' right of free speech and approval of multi-party systems-measure different things, acceptance of liberal values and support for democratic institutions. The distinction is not trivial, although it may not appear obvious from the perspective of a long-established liberal democracy. Especially in newer democracies, adherence to the regime form and its institutions may simply be part of an otherwise undemocratic citizen's loyalty to the state: liberal values like tolerance may not be part of ideological package. Nor need distinction necessarily be limited to new regimes which aspire to liberal democratic status. For example, it is theoretically possible to have an illiberal democracy of a Rousseauist or communist type a danger clearly recognized by Tocqueville and J. S. Mill in their discussions of the tyranny of the majority-in which a majority allows no deviations or dissent from its decisions; and although we may feel that the definition of democracy is thereby stretched, balanced defenses of distinction have also been made with reference to contemporary empirical cases (Lindblom). 1 If distinction is correct, then there need be no necessary correspondence in the pattern of association between democratic and liberal measures; and there is no necessary basis for concluding that if the better This content downloaded from 157.55.39.243 on Wed, 05 Oct 2016 04:07:05 UTC All use subject to http://about.jstor.org/terms