CIA Director John Brennan during a news conference at CIA headquarters in Langley, Va., on Dec. 11, 2014. Brennan defended his agency from accusations in a Senate report that it used inhumane interrogation techniques against terrorist suspects with no security benefits to the nation. (Pablo Martinez Monsivais/AP)

The following is a guest post from political scientists Courtenay R. Conrad (University of California, Merced), Justin Conrad (University of North Carolina, Charlotte), James A. Piazza (Pennsylvania State University), and James Igoe Walsh (University of North Carolina, Charlotte).  The article on which this post is pasted has been ungated by the journal Foreign Policy Analysis and will be available for free for the next six months here.


The Senate Select Committee on Intelligence’s recent report on torture of suspected terrorists by the Central Intelligence Agency (CIA) after 9/11 has renewed the debate about the efficacy and morality of torture. Shortly beforehand, the United Nations Committee Against Torture (CAT) highlighted allegations of torture and other abuses by the CIA as well as members of the United States military. Left unaddressed, though, is a discussion of the conditions under which government agencies view torture as an effective and acceptable response to the threat of terrorist attacks. Many seem to assume that public demands to “do something” in response to terrorism create overwhelming pressure for officials to use torture to obtain intelligence and to deter future attacks.

However, there is surprisingly little evidence that governments regularly respond to terrorism with torture. David Charters and colleagues found that democracies facing threats of terrorism in the 1970s and 1980s were able to craft effective counterterrorism policies without resorting to torture.  A more recent study by Walsh and Piazza finds that governments react to terrorism by engaging in more physical integrity violations like extrajudicial killings and disappearances but do not engage in heightened levels of torture specifically.

In an article forthcoming in Foreign Policy Analysis, we delve further into the effects of terrorist attacks on a country’s human rights picture by arguing that incentives to violate human rights differ across government agencies and that the type of terrorist attack experienced matters. Because an important objective of the military is to defend against external enemies, we argue that military forces respond to transnational terrorist attacks — but not domestic terrorist incidents — with heightened torture. Militaries have historically devoted most of their attention to planning for war, not counterterrorism. Torture is a practice that they can implement quickly and (seemingly) cheaply to gain intelligence about terrorist threats, making it a tempting solution to a novel policy challenge. Police and prison officials, in contrast, are less likely to view responding to transnational threats as central to their organizational missions, and thus do not respond by increasing the degree to which they torture.

We assess this argument with data from the Ill-Treatment and Torture Data Collection Project, which disaggregates the agencies engaging in torture for countries around the world from 1995 through 2005. Figure 1 illustrates this relationship by depicting the predicted probability that a country’s military will engage in greater degrees of torture.

Occurrence of Torture as a Result of Transnational Terrorism (Data: Ill-Treatment and Torture Data Collection Project (ITT); Figure Conrad et. al 2015.)
Occurrence of Torture as a Result of Transnational Terrorism (Data: Ill-Treatment and Torture Data Collection Project (ITT); Figure Conrad et. al 2015.)

The likelihood of widespread, systemic torture by military forces increases sharply with the number of transnational terrorist attacks, while the chance that the military will refrain from torturing declines.

We further find that this response is most likely in established democracies. At first glance, this claim is surprising because democracies are less likely to engage in abuses of human rights, are more likely to cease torturing and long-established and stable democracies are the least likely to torture. Yet most democracies engage in torture, suggesting that they see utility in the practice or at least view the costs of stopping torture as unacceptably high. The value of torture for democratic states increases during periods of foreign threat, including as that posed by transnational terrorists. Democracies have long responded to external threats by increasing repression at home. Citizens are less likely to object to the torture of suspected terrorists who are members of “out-groups,” including foreign nationals, and their preferences carry greater weight in democratic regimes.

These findings suggest at least two reforms that could be put in place to prevent a repeat of the post 9/11 experience of torture by American military and intelligence agencies. The first of these reforms is greater oversight of these agencies, including in long-standing democracies such as the United States. As Michael Colaresi and others have suggested, requiring military and intelligence agencies to explain and justify their treatment of detainees to legislative or judicial bodies might restrain the resort to measures such as torture that promise quick intelligence returns but that threaten to undermine the legitimacy of counterterrorism policies and might lead to more terrorism. A second useful response would be to consider now how the United States can best respond to catastrophic terrorist attacks in the future. The Senate report suggests that the torture and detention policies of the CIA (and likely of the military) were devised hastily and in secret. Coolly debating assessing the range and value of responses to terrorism before the next attack might ensure that more effective and ethical policies can be put in place should the United States homeland be targeted by transnational terrorists in the future.