Abstract
This study examines the relationship between punitive attitudes toward criminals, two measures of economic insecurity and a measure of blame for stagnating incomes that targets welfare, affirmative action, and immigration. In effect, we are testing whether punitiveness toward criminals is part of a general constellation of resentment toward what Gans (Citation1995) has termed the “undeserving poor” and that Garland (Citation2001) has described as the “politics of reaction.” Survey data involving 1,476 adults are assessed using OLS regression. Results indicate that blame of welfare, affirmative action, and immigration is the strongest predictor of punitiveness. Economic insecurity has variable input to punitive attitudes that depends on the measure used and the sex and race of respondents. Some evidence of an “angry White male” phenomenon is also provided by the results.
Acknowledgments
An earlier version of this paper was presented at the annual meetings of the American Society of Criminology, San Diego, CA, November, 1997. The authors would like to thank Gary Kleck and anonymous reviewers for their comments and suggestions for improvement of this paper.
Notes
1. When first articulated by Blalock, the term used was “power threat,” but subsequent development has seen the concept evolve into “social threat.”
2. Blalock (Citation1967) himself makes reference at various points to such individual‐level factors as the “fear of competition,” “the perception of competition,” and “personality factors” as potential “intervening variables in the analysis” that would test his structural‐level model of racial competition and control.
3. Affirmative action is explicitly concerned with race, and as Gilens (Citation1996, p. 593) has observed, welfare, like crime, is one of those “coded” issues “that play upon race … without explicitly raising the ‘race card’”.
4. For a detailed review of this research, see Blumstein and Cohen (Citation1980).
5. While this measure differs from others typically used, and includes some items that depart somewhat from more traditional measures, all of the items in this index represent specific, punitive criminal justice policies that were being implemented or considered at the time we conducted our survey, and were receiving considerable publicity. Although these policies may seem less significant in the current political climate, at the time we collected the data we were particularly interested in public support for these policies, as opposed to more general “punitiveness.”
6. Income and education were measured on 5‐point scales. To increase their range of variation, income was recoded to the midpoint of each of the five intervals presented to respondents. Education was recoded to approximate the highest grade completed. The values for income are 7.5, 22.5, 40.0, 62.5, and 87.5; for education, they are 8, 12, 14, 16, and 18.
7. The lack of a more direct measure of political orientation was not the result of a conscious decision by the authors but rather a simple omission. We acknowledge that the measure we employ is quite similar to the blame scale, and is in fact moderately correlated with it (r = .42). However, given the importance in the present context of having some control for political conservatism, as well as the absence of multicollinearity discussed below, we feel its inclusion is justified, and even necessary.
8. In analysis disaggregated by race not shown here, the same effect of economic insecurity was found in the White sub‐sample but not among non‐Whites.