Content area
Full text
ABSTRACT: Building on Philip Converse's understanding of public opinion, John Zaller sees the evidence for the public's "nonattitudes" as reflecting individuals' ambivalence concerning political issues. Because neither individuals nor the public collectively have what Zaller would call real attitudes, he concludes that the effectiveness of democracy rests on competition among intellectual and political elites. In truth, however, the public has many real attitudes that depend heavily on elite leadership, in ways that Converse did not initially emphasize but that are consistent with both his observations and Zaller's model of mass opinion. The quality of the public's attitudes are, however, a point of serious contention.
Not long ago in Critical Review, Eric R.A.N. Smith (I996) described how various lines of recent research can be combined to offer a new perspective on the public ignorance that Philip Converse had underscored in his seminal article, "The Nature of Belief Systems in Mass Publics" (1964). Smith concluded that the more recent work shows "that although the opinions of individuals may not be rational or well informed, the opinions of the mass public taken as a whole are, if not rational, at least comprehensible. Whether this is enough to allay the worries raised by Converse, however, is, Smith concluded, "a topic for further investigation" (Smith 1996, 104).
Converse (1996) himself has reached sanguine conclusions about the implications of this new work for the competence of public opinion, even though his "The Nature of Belief Systems"-along with The People's Choice, by Paul F Lazarsfeld, Bernard Berelson, and Hazel Gaudet (1948); Voting, by Berelson, Lazarsfeld, and William N. McPhee (1954); and The American Voter, by Angus Campbell, Converse, Warren Miller, and Donald E. Stokes (1960)-seemed to show that public input into the political process could hardly be trusted, because the people are grossly uninformed about politics and about potentially useful issue-oriented and ideological ways of looking at the political world. The public appeared, at best, to rely on unsophisticated substitutes such as party identification and social-group and interpersonal influences. People's unstable and incoherent responses to questions in opinion surveys reflected "nonattitudes," not real attitudes toward issues and politics. While Converse (1964, 1970, and 1996) noticed variations in attention and sophistication among different segments of the public, as well as variations in...