Torture is a difficult subject to research. Perpetrators almost always attempt to hide their acts, making any kind of systematic data collection challenging. If one is able to gather data, research on such a subject is often received with ambivalence. Raising the question of whether torture “works” can run the risk of seeming to justify the violence, as can inquiries into the reasons states and their representatives engage in torture. Still, it is of great importance to glean what is possible about why torture occurs and how it might be prevented. Makes a significant contribution to that aim.
Many of the studies refute common assumptions. Rejecting the claim that an American majority supports torture, Peter Miller and his colleagues show that this has rarely been the case. Support for torture did increase, however, among Republicans during the Obama Administration, which the authors argue indicates the effect of elite messages on public opinion. Jeremy Mayer and his colleagues find that terrorism does not increase support for torture, contrary to what many researchers assume, but that the degree of political freedom in a country is inversely related to such support. A study of anti-torture advocacy by Courtenay Conrad and Jacqueline DeMeritt fails to show any positive effect of naming and shaming for government practices, and in fact shows that such efforts are associated with increased government repression in other areas. The authors seem to regret this finding, and express the hope that naming and shaming is effective in ways not captured by their study. These studies call readers’ attention to the ways in which common assumptions, such as about the factors that influence support for torture and the measures that prevent it, are not always supported by evidence.
Other chapters provide further empirical and theoretical support to conclusions that are more prevalent or intuitive. Henry Carey points to a country's reliance on international alliances and corresponding willingness to uphold international agreements as key to whether a government engages in torture. Tracy Lightcap shows that governments are more likely to torture when there is an asymmetrical war that threatens the stronger regime's position or projects. James Pfiffner makes a strong argument in two chapters that torture has undermined both the moral authority and strategic interests of the United States, and that it is ineffective at producing good intelligence.
The authors address distinct but interrelated questions, and synthesizing their findings or at least considering them in light of each other would be illuminating. For example, drawing on a previous study he conducted on torture reform in the United States, France, Israel, and Argentina, Carey concludes that a government's willingness to engage in reform depends on perceived national security threats. In another chapter, however, Miller and colleagues find that the actual level of threat does not predict the use of torture by a government. What might explain this discrepancy? Should the reader conclude that the level of real or perceived threat from acts of terrorism is not sufficient to induce a government to begin using torture, but that a government that already uses torture will be unwilling to change its practices as long as the threat prevails? Or does perceived threat matter more than actual threat? Such a discussion between chapters would increase the significance of each.
Moreover, many of the findings point to the need for in-depth domestic research. For example, Carey concludes that domestic institutional reform is crucial for preventing torture. Ethnographic research within domestic institutions can reveal the barriers and potential pathways to such change. Conrad and DeMeritt suggest that one reason they did not find a positive effect of “naming and shaming” on torture reform is that state leaders may have little control over what local law enforcers do. Collaborating with qualitative researchers who conduct research on local law enforcement could test this possibility. Similarly, while they reject the hypothesis that media affects public opinion writ large, Miller and his colleagues guess that the media may have an effect on people who actually engage in torture. Interviews with police and military officers could identify how messages about torture are received, and how such messages might be counteracted.
Ultimately, neither qualitative nor quantitative research is sufficient on its own to understand the complex and multifaceted causes of torture. Quantitative research is not well equipped to assess the role of agentic individuals. Analyses that show whether the level of security threats or economic development predict whether a government uses torture or is willing to reform can give the impression that societal forces produce torture, rather than individuals who make decisions. At the same time, qualitative research is not best suited to understanding how such wider forces can inform the decisions of individuals in ways of which those individuals may not be aware. What is needed is theorizing on the structure-agent problem in regard to the causes of torture. How do individuals make decisions about whether to use or support torture in interaction with wider structures and social forces?
Such theorizing could be especially useful to efforts to prevent torture, as interventions often suffer from the same challenge as research. Training programs for law enforcers may equip them with nonviolent interrogation techniques and may even persuade them that they should not use torture. But they would still be subject to the wider social and political forces that encourage the use of torture. Even state leaders may have only limited capacity to individually create change. Likewise, institutions could be reformed, but without changing the skills and opinions of people within those institutions, torture may still continue. Hence, understanding the relationship between the individuals involved in torture and the wider context in which they work is key to preventing the practice.
Authors in this volume agree that qualitative work can complement their research, and some offer suggestions for how it might do so. Miller et al. explain how their survey research supports or undermines claims made by qualitative researchers, and how such researchers might respond with further studies. Building on such suggestions, as well as synthesizing the different conclusions reached by authors in this volume, offer promising avenues that could further enhance the significance of the volume
