Reassessing the Impact of Supreme Court Decisions on Public Opinion: Gay Civil Rights Cases

Article excerpt

The theoretical and empirical debate over the ability of the U.S. Supreme Court to influence public opinion through its decisions is far from settled. Scholars have examined the question using a variety of theoretical perspectives and empirical evidence, but there is no theoretical consensus, nor are the empirical studies without methodological weaknesses. We enter this debate in an attempt to bring some clarity to the theoretical approaches, overcome some of the methodological shortcomings, and bring a yet unstudied issue area, Court decisions on gay civil rights, under scrutiny. We argue that the ability of Court decisions to influence public opinion is a function of the salience of the issue, the political context, and case specific factors at the aggregate level. At the individual level these factors are also relevant, but citizen characteristics must also be taken into consideration. Our analysis of aggregate level and individual level opinion does indeed suggest that Court decisions can influence public opinion. However, the ability of Court decisions to influence public opinion is conditional. Our findings lend support to the legitimation hypothesis and the structural effects model. We conclude with a discussion of the implications of our findings and suggestions for future research.

In a representative democracy we are often concerned with the congruence of public preferences and the decisions of political institutions. Most often we expect that the decisions of institutions will be consistent with public preferences. However, we also know that the decisions of political institutions can influence the policy preferences of the public as well as public confidence in our political system and its institutions (Stimson 2004). When the focus has been on the Supreme Court, scholars have indeed suggested that the Court will defer to public opinion in deciding cases (e.g., Adamany 1973; Dahl 1957; Marshall 1989). However, the Court does not always defer to public opinion, and Court decisions may actually play a role in shaping public opinion (Franklin and Kosaki 1989; Hoekstra 1995; Johnson and Martin 1998; Marshall 1989; Petty and Cacioppo 1986). Indeed, many recent studies suggest that Court decisions might temporarily shape public opinion under certain conditions (see Hoekstra 2003, and Marshall 1989, for reviews). For example, citizens who have knowledge of a decision but also have low prior interest in the issue are most likely to be influenced by a decision (Hoekstra and Segal 1996; Petty and Cacioppo 1986). But scholars continue to debate the potential influence of the Court, the conditions under which influence is possible, and whether or not observed shifts in opinion following Court decisions can be directly attributed to a Court decision (e.g., Caldeira and Gibson 1992; Franklin and Kosaki 1989; Gibson, Caldeira, and Spence 2003; Hoekstra 1995, 2000; Hoekstra and Segal 1996; Johnson and Martin 1998). In part, the conflicting findings may be a result of researchers using different methods and examining opinion at either the individual level or the aggregate level only.

We join this debate by examining the potential influence of Court decisions in an issue area not yet explored, gay and lesbian civil rights. Although there have only been four significant decisions by the Court on gay civil rights, each decision had significant policy implications, and the most recent decision in Lawrence v. Texas (2003) appears to be playing a central role in shaping the national debate over gay rights and same-sex marriage. Our study expands the theoretical arguments of earlier research by more fully addressing the conditional influence of the Court. We also attempt to overcome some of the limitations of previous studies by examining the potential impact of multiple Court decisions on public opinion using both aggregate and individual level analysis-previous research has only examined opinion change at the individual or aggregate level, not both. …