Beyond the media's explanation: Examining the determinants of attitudes toward torture

ABSTRACT

The question of what constitutes torture has perhaps never held as prominent a place in US political discourse as it has since the attacks on 9/11. This has fueled a national conversation that causes us to consider to what extent do our actions as a nation reflect our values as a nation? In a post 9/11 world in which the country will confront terrorism at home and abroad and our values will be tested we should strive for a better understanding of citizens' attitudes towards the practices that are (and have been) used to counter terrorism. To the extent that democratic theory is correct in necessitating public policy and public opinions to be reasonably congruent, understanding the distribution and formation of citizen attitudes towards torture is an important endeavor.

Notes on contributors

Mary R. Anderson is Associate Professor of Political Science at the University of Tampa. David L. Richards is Associate Professor of Political Science and Human Rights at the University of Connecticut.

Notes

1. Psychology scholars have begun to address some of these deficiencies and have added to the accumulation of knowledge in this area. This would include Houck and Conway (2013) and Crandall and colleagues (2009).

2. Prior to the 2006 Military Commissions Act, the term torture, for the purposes of the 1996 War Crimes Act, was applied to all breaches of common Article 3 of the Geneva Conventions. As of October 17, 2006, applicability was changed to cover only “grave breaches” of common Article 3.

3. A complete copy of our survey can be obtained by emailing the corresponding author.

4. The survey contained ninety items and took roughly twenty minutes to complete. The survey was in the field from October 9, 2006, to December 15, 2006. There were 853 completed surveys for a completion rate of twelve percent. Please see Appendix B for survey protocol and justification.

5. For a full explanation of this technique and the creation of this index, please refer to Richards et al. (2012).

6. The respondent was provided the following preface and questions: When we talk about political issues and governance, often values come up in the discussion. If you absolutely had to choose between the following two values, which is more important: (1) guaranteeing law and order in society OR guaranteeing individual freedom? (2) narrowing the gap between the rich and the poor OR increasing economic growth?

7. Respondents were presented with a list of groups past and present that have been “liked least” and permitted to consider a group they like even less. Once the group was in mind, respondents were asked to indicate their level of agreement from strongly disagree to strongly agree with the following statements: The [selected group] should (1) be banned from being President of the United States, (2) be outlawed, (3) be allowed to make a speech in [city name] to protest against the government, (4) be allowed to hold public rallies in [city name] to protest against the government, (5) be allowed to teach in public schools, and (6) have their phones tapped by the government.

8. The Bush Administration saw techniques such as waterboarding and “fear up” as constituting activities less than torture under Article 16 of the Convention Against Torture (CAT). In Paragraph 3 of General Comment 2, the Committee Against Torture charged with interpreting the CAT stated that the prohibition of Article 16 activities is indivisible and interrelated with the general prohibition of torture, per se, however.

9. This would include whether respondents believed the United States was using, as a matter of policy, either torture or physical abuse falling just short of torture in the war against terrorism; ordered so that 1 = strongly oppose and 5 = strongly support.

10. Angrist and Pischke (2009) made the argument that, in a situation such as our cross-sectional survey, differences between marginal effects and OLS are trivial, and that the latter is easier to make sense of; thus we have elected that strategy. However, we have included ordered logit results and a distribution of the dependent variable in Appendix A, which demonstrate the results are similar to those we have elected to include here, and thus the story remains the same.