“If Torture Is Wrong,What About 24 ?” Torture and the Hollywood Effect

Abstract

Since 9/11, entertainment media has focused on depictions of terrorism and counterterrorism. How do dramatic depictions of counterterrorism practices—specifically torture—affect public opinion and policy? Using a mixed within-subjects and between-subjects experimental design, we examine how framing affects support for torture. Participants (n = 150) were randomly assigned to a condition for dramatic depictions showing torture as (a) effective, (b) ineffective, or (c) not present (control). Participants who saw torture as effective increased their stated support for it. Participants who saw torture—regardless of whether or not it was effective—were more likely to sign a petition on torture. We discuss the policy implications of our findings on how framing affects opinion and action regarding torture.

Keywords

media violence public opinion counterterrorism torture

Introduction

Jack Bauer, the protagonist from television’s 24, seems to mostly get his man. Through dramatic depictions of heroism, while bending the rules, Bauer regularly subverts terrorism, mass destruction, and other horrid outcomes for the U.S. government and its people. Some of the situations Bauer solves seem outlandish. Fans would acknowledge this and suggest 24 and other similar shows are solely for entertainment.¹ It is not clear, however, the extent to which people believe that 24 provides any insight into counterterrorism.

Some military leaders, at least, took Bauer fairly serious. In February 2007, as counterinsurgency in Iraq began to shift in the U.S. and Iraqi government’s favor, Brigadier General Patrick Finnegan traveled to Hollywood to meet with producers of 24 to persuade them to remove Bauer’s illegal actions, such as torture, as they were having detrimental effects on the troops in theater.² The Parents Television Council found that in the five seasons of 24, there were over 60 scenes depicting torture and that the number of torture scenes in the media jumped in the 4 years following 9/11 (also see Flynn & Salek, 2012; Prince, 2009 for a detailed discussion).³ Building on General Finnegan’s suspicions, how are members of the American public affected by dramatic depictions of torture?

Our article is organized as follows: First, we engage with the literature on how media affects attitudes generally, how this pertains to views of law enforcement, and factors that affect attitudes about torture. We then discuss our methodological approach, sample, and analyses. We conclude with a discussion of our results, how this pertains to policy, and avenues for future research.

How Media Affects Attitudes

How an issue is framed can produce a predictable shift in a person’s opinion on a given topic. Even when the change in framing is subtle, this can yield notable differences in how a person views the issue (Tversky & Kahneman, 1981). In many cases, people accept the argument put forth and make a decision without questioning the logic of the argument itself. This is particularly likely when the person is unfamiliar with the underlying subject matter (Kahneman & Tversky, 1982). Even when people are knowledgeable on an issue, framing the outcome as a gain (such as “lives saved”) rather than a loss (such as “lives lost”) can push people to supporting the gain perspective (Kahneman & Tversky, 1984).

Of the myriad mechanisms through which issues are framed, television is one of the most ubiquitous. The average American adult watches over 24 hours of television each week.⁴ In terms of news media, the adage “if it bleeds, it leads” suggests that news coverage tends to focus more on the negative stories. This may explain why people drastically overestimate the risk of negative outcomes like crime and terrorism (Nellis & Savage, 2012).⁵ Likewise, entertainment media tends to focus on sensational, attention-grabbing storylines. As stories are increasingly told in this format versus face-to-face, storylines can be further removed from reality. Built on this, media can cultivate perspectives on issues with which the viewer has no direct experience. Thus, viewers construct a reality that may not truly exist in the real world (Gerbner, 1998).

Media and Law Enforcement

The proliferation of pop-culture media—namely television shows—on crime and reactions to it led to a scholarly debate on the impact that these media have on perceptions of law enforcement and other criminal justice–related outcomes (Callanan & Rosenberger, 2011; Donahue & Miller, 2006; Dowler, 2002; Dowler & Zawilski, 2007). How do media shape perceptions of law enforcement and the policies they use? Some, like Dowler (2002), argue that crime dramas do not affect public opinion about law enforcement and their practices. Numerous studies, however, have found that people’s perceptions of law enforcement are affected by reports about police in the media (Graziano, Schuck, & Martin, 2010; Miller & Davis, 2008; Weitzer & Tuch, 2005; Wu, 2010) and dramatic depictions of law enforcement (Callanan & Rosenberger, 2011; Donahue & Miller, 2006; Donovan & Klahm, 2015; Eschholz, Blackwell, Gertz, & Chiricos, 2002).

As the general public has limited interaction with the criminal justice system, people tend in part to base their opinions on media depictions of law enforcement (Weitzer, 2002). Media depictions of law enforcement may be particularly impactful on people who have not had contact with the police (Adoni & Mane, 1984). Experience with and knowledge of interrogations in counterterrorism are even more rare. Building from this discussion, it stands to reason that members of the public rely heavily on media to form opinions of the practices that are effective and appropriate in counterterrorism.

Public Perception of and Support for Torture

There has been a vigorous debate over appropriate counterterrorism practices, making studies on public perceptions of torture and behaviors in support of these beliefs salient. Some politicians advocate the idea that torture is a necessary evil in the war on terror (Gearty, 2007). Yet the Senate Torture Report, senior members of the intelligence community, and a body of scholarship all conclude that torture is ineffective at gathering actionable intelligence and is strategically counterproductive (Carlsmith & Sood, 2009; Janoff-Bulman, 2007; Santucci, 2008). Despite clear evidence that torture does not work, approximately half of the public still supports it.⁶ Roughly, seven in 10 conservatives support torture while about the same number of liberals oppose it.

How Contextual Factors Affect Public Opinion on Torture

In the past decade, scholarly attention on torture has peaked, especially in the context of counterterrorism. Research has focused on perceptions of the torture (Carlsmith, 2008) and how identity affects such perceptions (Piazza, 2015), support for torture (Gronke et al., 2010), perceptions of what constitutes torture (Nordgren, McDonnell, & Loewenstein, 2011; Norris, Larsen, & Stastny, 2010), why the use of torture persists despite arguments against it (Arrigo & Bennett, 2007), and the efficacy of torture (e.g., Carlsmith & Sood, 2009; Gray & Wegner, 2010; Janoff-Bulman, 2007; Santucci, 2008). In addition to individual-level motivations for support for torture, research has also examined how institutions can change how people perceive the act. Although conventional wisdom suggested democracy and democratic institutions might have a palliative effect, Rejali (2009) shows that these institutions often just shift the kind of torture used. In response to Rejali, Conrad, Hill, and Moore (2017) argue that the type of institutions matters. Beginning with the assumption that government torture is generally targeted at individuals who voters find threatening, they show that institutions that reflect public opinion—like electoral contestation—are associated with higher levels of torture. By contrast, institutions shielded from public opinion like strong, independent courts, will be associated with lower levels of torture.

Public opinion polls on torture frequently ask about the level of support in the abstract without examining other factors that can affect perceptions. Experimental research, however, shows that support for torture can be swayed by situational factors. Members of the public are more supportive of torture when the suspect is an out-group member (Miron, Branscombe, & Biernat, 2010; Norris et al., 2010; O’Brien & Ellsworth, 2012; Tarrant, Branscombe, Warner, & Weston, 2012). These findings are consistent with social identity theory and appear to hold across social groups (Tajfel & Turner, 1979). When the suspect is perceived to be guilty, people are also more likely to support torture (Carlsmith & Sood, 2009), which mirrors views on guilt and punishment more broadly (Golash, 2005). Greater geographic distance from the suspect also increases justification of torture (Gray & Wegner, 2010), which suggests that people may engage in an “out of sight, out of mind” logic when torture is more abstract. Finally, people tend to be more supportive of psychological torture over physical torture (Nincic & Ramos, 2011; Riva & Andrighetto, 2012). This finding likely stems from the (incorrect) belief that physical torture is more painful and damaging than psychological torture (Piwowarczyk, Moreno, & Grodin, 2000; Sanders, Schuman, & Marbella, 2009; Vallacher, 2007). In short, an individual’s support for torture is not fixed. Rather, this body of research demonstrates that individual-level views on torture are largely contingent on contextual factors.

People tend to be unaware of their motivations for supporting torture. When people support torture, they often justify this belief on utilitarian grounds (Carlsmith & Sood, 2009). Carlsmith (2008), however, found that although people abstractly support utilitarian policies toward torture, their behaviors often contradict their stated attitudes in favor of retributive motivations. Malleable perceptions of torture are a double-edged sword. On one hand, malleable perceptions of torture suggest that people can be convinced to oppose torture. On the other hand, research shows that these variations in support of torture often hinge on prejudices and the desire to punish others.

Expert Opinion and Public Support for Torture

Expert consensus is that torture does not work. Military interrogators say that torture is not an effective way to gather accurate and reliable information (Janoff-Bulman, 2007). This was corroborated by the 2014 Senate Torture Report, which states that torture did not elicit actionable intelligence. Recently, the Secretary of Defense, James Mattis, spoke out in opposition to torture and in favor of rapport-building approaches.⁷ Other vocal opponents of torture include Senator John McCain who was, himself, a victim of torture during the Vietnam War, and human rights groups like Amnesty International.

Experimental research suggests that accusatorial questioning increases the likelihood of confessions, both true and false, as compared with information-gathering questioning (Evans et al., 2013; Meissner et al., 2014). U.S. courts have long recognized that confessions obtained during duress are neither reliable nor Constitutional (Redlich, 2007). Beyond these concerns, using torture may actually increase vulnerability to terrorism, the very thing that it is allegedly trying to prevent (Walsh & Piazza, 2010). Despite clear expert opinion that torture does not work, many people still believe that it does or can. In fact, a recent Pew Research Center study shows that roughly half of the American public thinks that torture in counterterrorism is acceptable in certain situations.

How Media Frames Torture

Most people are only exposed to torture through media. Although many of these studies assess attitudes about torture, scholars have rarely addressed how media influence perceptions on torture specifically. However, research shows that entertainment media can increase perceptions of law enforcement effectiveness and decrease perceptions of false confessions under duress (Donovan & Klahm, 2015), so media reasonably affects perceptions of torture as well. Particularly since 9/11, pop-culture media about terrorism and reactions to it has flourished (e.g., 24, Homeland, Quantico, Zero Dark Thirty). When torture is shown in television and movies, there is generally a ticking time bomb scenario that yields the information desired by the torturer. Janoff-Bulman (2007) states that this depiction “seems to fundamentally define how we think about and react to torture interrogations” (p. 431). Horne (2009), among others, argues that such depictions of torture affect public perceptions of the efficacy of torture and may alter support for these tactics. People may assume that torture works when we show them a TV clip where it does (Kahneman & Tversky, 1982). From this, we expect that support for torture will be tied to how effective it appears in media:

Hypothesis 1 (H1): When torture is depicted as effective, support for the practice will increase.

Hypothesis 2 (H2): When torture is depicted as ineffective, support for the practice will decrease.

Hypothesis 3 (H3): When torture is not depicted (control), support for the practice will not change.

Linking Public Opinion and Policy

In the study of war, the so-called CNN effect is one proposed impact that media can have on public policy. In short, when the media frame an issue, it can sway opinion and lead to foreign policy changes (Gilboa, 2005). Although this impact had been debated and contextualized, evidence suggests that the impact can be greatest when policy is uncertain (Robinson, 2000). As public opinion polls and policy discussion demonstrate, views on torture fall into this uncertain category.

Hurting another person is generally considered to be wrong. Yet there are contexts in which people tend to think that it is permissible (Crelinsten, 2003). In part, torture may persist under the argument that it was a “necessary evil” in counterterrorism (Gearty, 2007; Opotow, 2007). Shows, such as 24, that perpetuate this narrative and promote the ticking time bomb paradigm may influence public opinion as well as torture practice (Horne, 2009; Janoff-Bulman, 2007). Although exposure to violent media may not inspire criminally aggressive action on the part of the viewers (Savage & Yancey, 2008), it may sway support for aggression and policy on the use of aggressive tactics by law enforcement. Violent media’s effect reaches beyond the general public; even legal scholars have used 24 to justify torture. In a discussion of constitutional jurisprudence, Justice Antonin Scalia asked if a jury would convict Jack Bauer.⁸ Similarly, John Yoo—author of the so-called Torture Memos—refers to 24 in an argument for the use of torture.⁹ When legal experts are swayed by media depictions, it is likely that media will affect the public at large, which can indirectly influence policy. In addition, elite opinions affect public opinion on terrorism and counterterrorism (Hill, Oliver, & Marion, 2010). This would be of particular concern if media depictions of torture are not just swaying attitude but also inspiring action.

As we know from behavioral economics and psychology, people do not always do what they say they will (Rynes, Gerhart, & Minette, 2004; Yezer, Goldfarb, & Poppen, 1996). Researchers have addressed this problem by adding a behavioral component to the traditional attitudinal measure. For example, in a recent field experiment in Nigeria, Collier and Vicente (2014) gave participants a postcard to mail in, if there were concerns about electoral violence. On divisive issues, there is also greater concern about social desirability bias. To address this concern, participants can be asked about their attitudes and be asked to take action. We would expect people to say the socially acceptable thing but they may be reticent to take action if it is not what they truly believe. Specifically, we expect the following between-subject differences:

Hypothesis 4 (H4): People who see torture as effective will be less likely to take action in support of their posttreatment belief than people in the control condition.

Hypothesis 5 (H5): People who see torture as ineffective will be less likely to take action in support of their posttreatment belief than people in the control condition.

Hypothesis 6 (H6): People who see torture as effective will be less likely to take action in support of their posttreatment belief than people in the ineffective condition.

Experimental Design

Most studies on media and perception of crime and law enforcement are survey-based rather than experimental (Graziano et al., 2010, is one notable exception). One key limitation of survey work is that it relies on self-reported data for exposure to crime-related media. These methods cannot identify causal mechanisms. They also cannot account for selection into watching certain types of media or other factors that may drive the results. Exposing people to these media via a randomized control trial is more similar to how they actually consume it.¹⁰ This can somewhat mitigate concerns about realism in the lab. Given the power of experiments at identifying causal effects, there is a need to employ this methodology more to understand security issues and policies generally (Arce et al., 2011). Thus, to examine the influence of media on support for torture, we designed a randomized control laboratory experiment.

Students from a midsize university in the Mid-Atlantic region were recruited to participate in a 45-min study on “Current Events.” Participants received a US$10 gift card to Amazon.com in exchange for their time. Participants were first asked about their level of support for five current event topics using a 4-point Likert-type scale from Completely Disagree to Completely Agree.¹¹ The five current event topics were the Keystone Pipeline, legalization of marijuana, the use of torture in interrogations, legalization of same-sex marriage, and teaching intelligent design in public schools.¹² Of course, our primary interest is perceptions of torture. The other four issues were included to obscure the true purpose of our study. Participants were then shown a series of five pop-culture video clips on these topics. Every participant saw the same four filler videos: pro-Keystone Pipeline, anti-legalization of marijuana, pro-same-sex marriage, and anti-intelligent design in public schools.¹³ For the clips on torture, participants were randomly assigned to one of three experimental conditions that showed torture as effective, ineffective, and not depicted (control).

The torture clips came from 24. To control for context within the two treatments, the clips in the effective condition and the ineffective condition were the same except the ending. In both, the suspected terrorist was in Jack Bauer’s custody being interrogated about the location of a bomb. As a form of psychological torture, the suspect’s children were shown on a live-feed television, and the suspect was convinced that they would be executed if he did not give up with bomb’s location. The suspect also suffered physical torture by having his fingers broken when he did not divulge information. The ineffective clip ended with the suspect screaming that he would never disclose the bomb’s location. Participants in the effective condition saw an additional scene where the suspect did tell Jack Bauer the location of the bomb and the attack was foiled. The control condition clip depicted Jack Bauer interrogating a suspect about the location of a bomb. This clip did not show torture. It also did not say whether or not the interrogation was successful at eliciting the desired information.

After watching the video clips, participants were again asked about their opinions on these five current event topics. Finally, modeling the notion of behavioral commitments in a laboratory setting, we gave each participant the option to sign petitions at the end of the study. Each participant was presented with a total of 10 petitions—one in support of and one in opposition to each of the five topics discussed. These petitions would then be sent to the Chairman of the United States Senate Committee under which each issue falls.¹⁴ Participants were told that petitions were optional and there would not be a penalty, monetary, or otherwise, for nonparticipation. Participants were also told that by signing any petition, they would waive confidentiality for this portion of the study. Across all experimental materials, the five topics were presented in a randomized order to control for order effect bias (Perreault, 1975). In addition, to control for how the questions were framed (Tversky & Kahneman, 1981), half of the participants were asked if they were supportive of an issue in the first questionnaire and if they were unsupportive of an issue in the second questionnaire, and vice versa.

Participants

One hundred fifty students participated in this study.¹⁵ The study was open to undergraduate and graduate students.¹⁶ Figure 1 depicts the demographic breakdown of participants by age, gender, race, religion, and year in school. Participants were balanced across conditions on these demographic variables.

Figure 1.

Demographic variables.

Results

Stated Beliefs

Participants were asked about their level of support using a 4-point Likert-type scale for five issues both before and after watching the stimulus videos. Lower scores indicate less support for the topic. Figure 2 shows the pretest and posttest mean levels of stated support for torture by condition.¹⁷ Although not identical, pretreatment views on torture are not significantly different across conditions, F(2, 143) = 0.47, p = .625.

Figure 2.

Mean stated level of support for torture in interrogations.

To test the first three hypotheses, we conducted t tests to compare pretest support for torture interrogations with posttest stated support.¹⁸ As expected in Hypothesis 1, participants in the effective condition were significantly more supportive of torture after treatment, t(49) = 2.67, p = .005.¹⁹ The effect size for this analysis (d = 0.24) is small. This indicates that pop-culture depictions of torture as being effective can affect support for the practice, but that it is not a large effect. Contrary to expectation in Hypothesis 2, participants in the ineffective condition had a slightly—but not statistically significant—higher level of support for torture after treatment, t(47) = 0.89, p = .19.²⁰ This suggests that showing torture not working does not make people less likely to support it. As expected in Hypothesis 3, participants in the control condition exhibit no change in stated support for the use of torture after treatment, t(47) = 0.00, p = .50.²¹

Taking Action

Participants were given the opportunity to sign petitions either in support of or in opposition to the five current event issues in the study. Out of the 750 potential actions to take (signed petitions) in this study, participants signed 460 petitions (61.33%).²² There is no relationship between condition, level of change in stated views on torture, and total number of petitions signed. For the purposes of this study, we are only interested in the petitions about torture. Figure 3 shows the percentage of participants in each condition who signed petitions in line with their stated posttest view on torture. There were significant differences in signing a petition on torture across conditions, F(2, 2) = 3.81, p = .024. Participants in the effective condition were most likely to sign a petition, followed by the ineffective condition, and finally the control condition. Interestingly, these results hold for both petitions. In sum, seeing torture as effective seems to inspire greater action both in support of and in opposition to the practice.

Figure 3.

Percentage of participants who signed each petition on torture by condition.

In addition to within-subject differences in stated views, we were also interested in differences in petition signing across conditions. We looked at petition signing in two ways. First, we coded taking action as dichotomous (1 = signed petition in line with stated posttest view; 0 = did not sign petition in line with stated posttest view). Second, we looked at which petition was signed to create three possible outcomes (1 = signed petition in opposition to torture, 2 = no petition signed, 3 = signed petition in support of torture). As shown in Table 1, we estimated two models to examine differences in petition signing between the two treatment conditions and the control condition (Hypotheses 4 and 5). The first is a logit model to examine differences in petition signing overall by condition. The second is a multinomial logit model to examine differences in which petition was signed by condition. We also estimated these models to compare differences between the two treatment conditions (Hypothesis 6), but the results were not significant and thus are not reported.

Table 1.

Differences in Petition Signing by Condition.

Petition signed	Condition	Coefficient	SE
Any petition	Effective	1.09**	0.42
	Ineffective	0.80^†	0.41
	Constant	−0.51^†	0.30
Petition to oppose	Effective	0.98*	0.44
	Ineffective	0.83^†	0.44
	Constant	−0.69*	0.32
Petition to support	Effective	1.49*	0.74
	Ineffective	0.64	0.82
	Constant	−2.30^†	0.61

Note. Results for “any petition” estimated with a logit model. Results for “petition to oppose” and “petition to support” estimated with a multinomial logit model. For both models, “no petition signed” is the base category.

†

p < .10. *p < .05. **p < .01.

We expected the participants in the control condition would be most likely to sign a petition, followed by participants in the ineffective condition and then participants in the effective condition. Results, however, indicated the opposite. Relative to the control condition, participants in the effective condition were significantly more likely to sign a petition (p = .010) while participants in the ineffective condition were not (p = .054). The probability of signing any petition was 69.27% for the effective condition, 64.84% for the ineffective condition, and only 37.5% for the control condition.

We then look at which petition participants are more likely to sign by condition. Relative to the control condition, participants in the effective condition were significantly more likely to sign a petition regardless of whether it was in support of (p = .027) or in opposition to (p = .044) torture. There was no difference in which petition was signed between participants in the effective and control conditions.

As a robustness check, we examined how political ideology affects taking action. We measured political ideology on a 5-point scale ranging from 1 = very liberal to 5 = very conservative. We estimated similar logit and multinomial logit models as above and included the political ideology variable. As shown in Table 2, political ideology has no impact on whether or not a person signed a petition in general (p = .087), but it does affect which petition a person is more likely to sign. More conservative participants were less likely to sign a petition in opposition to torture (p = .001) and more likely to sign a petition in support of it (p = .002).

Table 2.

Differences in Petition Signing by Condition and Political Ideology.

Petition signed	Condition	Coefficient	SE
Any petition	Effective	1.16**	0.43
	Ineffective	0.75^†	0.42
	Political ideology	−0.35^†	0.20
	Constant	0.29	0.55
Petition to oppose	Effective	1.06*	0.47
	Ineffective	0.82^†	0.46
	Political ideology	−0.81**	0.25
	Constant	1.10^†	0.62
Petition to support	Effective	1.65*	0.79
	Ineffective	0.22	0.93
	Political ideology	1.33**	0.44
	Constant	−6.15**	1.54

†

p < .10. *p < .05. **p < .01.

Discussion

We found that participants who saw torture depicted as effective were more likely to support torture. Building from Kahneman and Tversky’s body of work on framing and decision-making, this finding is unsurprising. Interestingly, though, framing torture as ineffective did not have the intended impact on support. Contrary to General Finnegan’s expectation, showing torture as ineffective did not reduce support for the practice. Although seeing torture as ineffective did not have a significant impact on support, with a larger sample size it is possible that it would have actually increased support as well.

Across conditions, just over half of the participants were willing to take action in line with their stated views on torture. This comports with the notion that people do not always do what they say they will. So, when are people more likely to act? We expected that participants in the treatment conditions would be less likely to sign a petition than participants in the control condition. We found the opposite. Beyond changing attitudes, seeing torture as effective also made people more likely to take action by signing a petition for Congress on the issue. Interestingly, though, seeing torture as effective increased the likelihood that a participant signed either petition. Seeing torture work makes people more likely to say they support it. This also seems to inspire people to take action both in support of and in opposition to the practice. In sum, these results indicate that dramatic depictions of torture where it is shown to be effective can change both stated attitudes about the practice and willingness to act on these views via signing a petition in line with stated beliefs. There was no difference in willingness to act between the effective and ineffective groups.

Our findings suggest that being primed on torture may lead people to believe that it works. It may also indicate that showing aggression of any kind inspires people to take action against these aggressive acts. Research on witnessing violence suggests the observer can mimic those behaviors or increase in likelihood of criminality (Bandura, Ross, & Ross, 1961; Eitle & Turner, 2002). As General Finnegan feared, media such as 24 may have an impact on troops in the position to engage in such violence. In the case of our study, being primed with violence might suggest to the observer that it is effective and thus influence an individual’s support for the practice. Exposure to dramatic depictions of torture may also desensitize people to this violence in the long run (Huesmann, 2007).

As we expected, exposure to media that depicts torture as effective moves opinion in favor of torture. As U.S. generals in Iraq in the mid-2000s conjectured, these media may unintentionally influence interrogators and others whose support can create a permissive environment for such behaviors. Research on violent video games, for example, suggests that these media increase aggression and that exposure to shows like 24 might have a similar effect (Bushman & Anderson, 2002).²³ Although their research is similar to ours, the link is more indirect. We are interested in how exposure to violence, and specifically torture, influences support for public policies. Although we expected that the efficacy of the violence would influence support, depicting violence might also prime the respondent into taking actions on the issue regardless of its efficacy.

Future Directions

Like all experiments using a subpopulation, there are limits to the generalizability of the results. In the future, we intend to replicate the study using a nationally representative sample. In addition, we intend to use a targeted population of members of the military and police to examine if they behave similarly to college students when exposed to these media. Given randomized assignment to treatment, we can be more confident that the results are due to the treatment and not to some other factor. We expect the results to be present in other samples. In addition, in the laboratory setting, we measured views of torture immediately posttreatment. In future iterations of this study, we plan to follow up with participants several times to measure the duration of the effects discovered in this study.

As we used clips from actual media, we were constrained by how media depict torture. Torture is not depicted as definitively ineffective on 24. Thus, we had to alter the effective clip to show torture being ineffective. It is possible that participants may not have fully interpreted this as torture not working as there is no negative result shown for this failed interrogation, such as the bomb detonating. Rather than using a clip from a different show or movie that would have introduced a host of potential confounding factors, we opted to use the same clip for both treatment conditions and make alterations based on where we cut the video. This allowed us to control for the suspect’s race and gender and other contextual factors. If torture is depicted more accurately in media, we will use such clips in future iterations of this research. Two additional considerations would be to use multiple depictions of torture as experimental materials and to control for previous exposure to the stimulus material. Showing multiple torture clips could help to minimize any idiosyncratic elements of the clip that may affect the outcome. However, by showing multiple clips on the same issue, we would reveal the true purpose of the experiment. Similarly, by asking participants about their previous exposure to 24, we would be revealing the purpose of the experiment and neglecting to account for the myriad media depictions of torture.

Policy Implications

The results have potentially troubling public policy implications. If exposure to media depictions of torture as effective leads people to support its use, then one consideration could be curbing these dramatic depictions. Of course, in a free and open society, constraining media undermines one of the foundations of a democratic system. But is it appropriate for the public or leaders to ask the producers of 24 and similar content to stop showing torture? Will it be effective, as General Finnegan allegedly asked, for Hollywood producers to “do a show where torture backfires . . . [because] The kids see it and say, ‘If torture is wrong, what about 24’?” Our research does suggest that depicting violence in this context may actually backfire. What can be done then?

The appropriateness of torture has been publicly debated for over a decade. Scholars and high-level military officials understand that torture does not actually yield actionable intelligence. Yet roughly half of the public thinks that torture can be justified in interrogations with suspected terrorists. Media depictions of torture play a role in driving public support. Although academic research and policy papers are inaccessible to much of the public, a television is not. To bridge the gap between expert knowledge and public opinion, the narrative surrounding torture must change. Simply showing that torture is ineffective is not enough. Rather, more nuanced portrayals of torture are necessary. On 24, Jack Bauer does not suffer any psychological or physical ramifications for his actions. Actual interrogators who have conducted or witnessed others conducting torture do suffer lasting effects from the experience (Lagouranis & Mikaelian, 2007). Similarly, media do not depict the myriad forms of damage that torture does to its victims (Rejali, 2009; Sanders et al., 2009). Depicting the long-term impact of torture for both the victims and perpetrators is one step toward humanizing an otherwise abstract practice for the public. Humanizing torture in media to make it less abstract may reduce support for the practice.

Footnotes

Acknowledgements

We would like to thank Evan Padulla for his invaluable research assistance.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received financial support for the research, authorship, and/or publication of this article: This work was supported by the School of Public Affairs Research Scholars Grant at American University.

Notes

Author Biographies

Erin M. Kearns is an assistant professor in the Department of Criminology & Criminal Justice at the University of Alabama. Her primary research seeks to understand the relationship among terrorism, law enforcement, and the public. Her publications include articles on why groups lie about terrorism, conditions that impact perceptions of counterterrorism practices, and relationships between communities and law enforcement. Her work has been funded through a number of sources, including the National Consortium for the Study of and Responses to Terrorism (START), some of which has been featured on CNN, NPR, the Washington Post, and Vox.

Joseph K. Young is associate professor with appointments in the School of Public Affairs and School of International Service at American University. He is an expert on terrorism and political violence, and is the co-founder of the Peace and Violence Research Lab at American University. His work has been published in journals across scholarly disciplines, including political science, criminology, economics, and international studies. His research has been funded by the National Science Foundation, USAID, and the National Consortium for the Study of and Responses to Terrorism (START).

References

Adoni

Mane

(1984). Media and the social construction of reality toward an integration of theory and research. Communication Research, 11, 323-340.

Arce

D. G.

Bakshi

Croson

R. T.

Eckel

C. C.

Fatas

Kass

(2011). Counterterrorism strategies in the lab. Public Choice, 149, 465-478.

Arrigo

J. M.

Bennett

(2007). Organizational supports for abusive interrogations in the “war on terror.” Peace and Conflict: Journal of Peace Psychology, 13, 411-421.

Bandura

Ross

S. A.

(1961). Transmission of aggression through imitation of aggressive models. The Journal of Abnormal and Social Psychology, 63, 575-582.

Bushman

B. J.

Anderson

C. A.

(2002). Violent video games and hostile expectations: A test of the general aggression model. Personality and Social Psychology Bulletin, 28, 1679-1686.

Callanan

V. J.

Rosenberger

J. S.

(2011). Media and public perceptions of the police: Examining the impact of race and personal experience. Policing & Society, 21, 167-189.

Carlsmith

K. M.

(2008). On justifying punishment: The discrepancy between words and actions. Social Justice Research, 21, 119-137.

Carlsmith

K. M.

Sood

A. M.

(2009). The fine line between interrogation and retribution. Journal of Experimental Social Psychology, 45, 191-196.

Collier

Vicente

P. C.

(2014). Votes and violence: Evidence from a field experiment in Nigeria. The Economic Journal, 124(574), F327-F355.

10.

Conrad

Hill

Moore

(2017). Torture and the limits of democratic institutions. Forthcoming in Journal of Peace Research.

11.

Crelinsten

R. D.

(2003). The world of torture: A constructed reality. Theoretical Criminology, 7, 293-319.

12.

Donahue

A. K.

Miller

J. M.

(2006). Experience, attitudes, and willingness to pay for public safety. The American Review of Public Administration, 36, 395-418.

13.

Donovan

K. M.

Klahm

C. F.

(2015). The role of entertainment media in perceptions of police use of force. Criminal Justice and Behavior, 42, 1261-1281.

14.

Dowler

(2002). Media influence on citizen attitudes toward police effectiveness. Policing & Society, 12, 227-238.

15.

Dowler

Zawilski

(2007). Public perceptions of police misconduct and discrimination: Examining the impact of media consumption. Journal of Criminal Justice, 35, 193-203.

16.

Eitle

Turner

R. J.

(2002). Exposure to community violence and young adult crime: The effects of witnessing violence, traumatic victimization, and other stressful life events. Journal of Research in Crime & Delinquency, 39, 214-237.

17.

Eschholz

Blackwell

B. S.

Gertz

Chiricos

(2002). Race and attitudes toward the police: Assessing the effects of watching “reality” police programs. Journal of Criminal Justice, 30, 327-341.

18.

Evans

J. R.

Meissner

C. A.

Ross

A. B.

Houston

K. A.

Russano

M. B.

Horgan

A. J.

(2013). Obtaining guilty knowledge in human intelligence interrogations: Comparing accusatorial and information-gathering approaches with a novel experimental paradigm. Journal of Applied Research in Memory and Cognition, 2, 83-88.

19.

Flynn

Salek

(Eds.). (2012). Screening torture: Media representations of state terror and political domination. New York, NY: Columbia University Press.

20.

Garland

(1991). The mid-point on a rating scale: Is it desirable? Marketing Bulletin, 2, 66-70.

21.

Gearty

(2007). Terrorism and human rights. Government and Opposition, 42, 340-362.

22.

Gerbner

(1998). Cultivation analysis: An overview. Mass Communication and Society, 1, 175-194.

23.

Gilboa

(2005). The CNN effect: The search for a communication theory of international relations. Political Communication, 22, 27-44.

24.

Golash

(2005). The case against punishment: Retribution, crime prevention, and the law. New York: New York University Press.

25.

Gray

Wegner

D. M.

(2010). Torture and judgments of guilt. Journal of Experimental Social Psychology, 46, 233-235.

26.

Graziano

Schuck

Martin

(2010). Police misconduct, media coverage, and public perceptions of racial profiling: An experiment. Justice Quarterly, 27, 52-76.

27.

Griffiths

(1999). Violent video games and aggression: A review of the literature. Aggression and Violent Behavior, 4, 203-212.

28.

Gronke

Rejali

Drenguis

Hicks

Miller

Nakayama

(2010). US public opinion on torture, 2001–2009. PS: Political Science & Politics, 43, 437-444.

29.

Hill

Oliver

W. M.

Marion

N. E.

(2010). “Shaping history” or “Riding the wave”? President Bush’s influence on the public opinion of terrorism, homeland security, & crime. Journal of Criminal Justice, 38, 896-902.

30.

Horne

(2009). Torture: A short history of its prohibition and re-emergence. Judicial Review, 14, 155-169.

31.

Huesmann

L. R.

(2007). The impact of electronic media violence: Scientific theory and research. Journal of Adolescent Health, 41(6), S6-S13.

32.

Janoff-Bulman

(2007). Erroneous assumptions: Popular belief in the effectiveness of torture interrogation. Peace and Conflict: Journal of Peace Psychology, 13, 429-435.

33.

Kahneman

Tversky

(1982). On the study of statistical intuitions. Cognition, 11, 123-141.

34.

Kahneman

Tversky

(1984). Choices, values, and frames. American Psychologist, 39, 341-350.

35.

Lagouranis

Mikaelian

(2007). Fear up harsh: An army interrogator’s dark journey through Iraq. New York: Penguin.

36.

Meissner

C. A.

Redlich

A. D.

Michael

S. W.

Evans

J. R.

Camilletti

C. R.

Bhatt

Brandon

(2014). Accusatorial and information-gathering interrogation methods and their effects on true and false confessions: A meta-analytic review. Journal of Experimental Criminology, 10, 459-486.

37.

Miller

Davis

R. C.

(2008). Unpacking public attitudes to the police: Contrasting perceptions of misconduct with traditional measures of satisfaction. International Journal of Police Science & Management, 10, 9-22.

38.

Miron

A. M.

Branscombe

N. R.

Biernat

(2010). Motivated shifting of justice standards. Personality and Social Psychology Bulletin, 36, 768-779.

39.

Nellis

A. M.

Savage

(2012). Does watching the news affect fear of terrorism? The importance of media exposure on terrorism fear. Crime & Delinquency, 58, 748-768.

40.

Nincic

Ramos

(2011). Torture in the public mind. International Studies Perspectives, 12, 231-249.

41.

Nordgren

L. F.

McDonnell

M. H. M.

Loewenstein

(2011). What constitutes torture? Psychological impediments to an objective evaluation of enhanced interrogation tactics. Psychological Science, 22, 689-694.

42.

Norris

J. I.

Larsen

J. T.

Stastny

B. J.

(2010). Social perceptions of torture: Genuine disagreement, subtle malleability, and in-group bias. Peace and Conflict: Journal of Peace Psychology, 16, 275-294.

43.

O’Brien

Ellsworth

P. C.

(2012). More than skin deep: Visceral states are not projected onto dissimilar others. Psychological Science, 23, 391-396.

44.

Opotow

(2007). Moral exclusion and torture: The ticking bomb scenario and the slippery ethical slope. Peace and Conflict: Journal of Peace Psychology, 13, 457-461.

45.

Perreault

W. D.

(1975). Controlling order-effect bias. Public Opinion Quarterly, 39, 544-551.

46.

Piazza

J. A.

(2015). Terrorist suspect religious identity and public support for harsh interrogation and detention practices. Political Psychology, 36, 667-690.

47.

Piwowarczyk

Moreno

Grodin

(2000). Health care of torture survivors. Journal of the American Medical Association, 284, 539-541.

48.

Prince

(2009). Firestorm: American film in the age of terrorism. New York, NY: Columbia University Press.

49.

Redlich

A. D.

(2007). Military versus police interrogations: Similarities and differences. Peace and Conflict: Journal of Peace Psychology, 13, 423-428.

50.

Rejali

(2009). Torture and democracy. Princeton, NJ: Princeton University Press.

51.

Riva

Andrighetto

(2012). “Everybody feels a broken bone, but only we can feel a broken heart”: Group membership influences the perception of targets’ suffering. European Journal of Social Psychology, 42, 801-806.

52.

Robinson

(2000). The policy-media interaction model: Measuring media power during humanitarian crisis. Journal of Peace Research, 37, 613-633.

53.

Rynes

S. L.

Gerhart

Minette

K. A.

(2004). The importance of pay in employee motivation: Discrepancies between what people say and what they do. Human Resource Management, 43, 381-394.

54.

Sanders

Schuman

M. W.

Marbella

A. M.

(2009). The epidemiology of torture: A case series of 58 survivors of torture. Forensic Science International, 189(1), e1-e7.

55.

Santucci

(2008). A question of identity: The use of torture in asymmetric war. Journal of Military Ethics, 7, 23-41.

56.

Savage

Yancey

(2008). The effects of media violence exposure on criminal aggression a meta-analysis. Criminal Justice and Behavior, 35, 772-791.

57.

Tajfel

Turner

J. C.

(1979). An integrative theory of intergroup conflict. In Austin

W.G.

Worchel

(Eds.) The social psychology of intergroup relations (pp. 33-47), Pacific Grove, CA: Brooks/Cole Publishing.

58.

Tarrant

Branscombe

N. R.

Warner

R. H.

Weston

(2012). Social identity and perceptions of torture: It’s moral when we do it. Journal of Experimental Social Psychology, 48, 513-518.

59.

Tversky

Kahneman

(1981). The framing of decisions and the psychology of choice. Science, 211, 453-458.

60.

Vallacher

R. R.

(2007). Local acts, global consequences: A dynamic systems perspective on torture. Peace and Conflict: Journal of Peace Psychology, 13, 445-450.

61.

Walsh

J. I.

Piazza

J. A.

(2010). Why respecting physical integrity rights reduces terrorism. Comparative Political Studies, 43, 551-577.

62.

Weitzer

(2002). Incidents of police misconduct and public opinion. Journal of Criminal Justice, 30, 397-408.

63.

Weitzer

Tuch

S. A.

(2005). Determinants of public satisfaction with the police. Police Quarterly, 8, 279-297.

64.

(2010). College students’ evaluation of police performance: A comparison of Chinese and Americans. Journal of Criminal Justice, 38, 773-780.

65.

Yezer

A. M.

Goldfarb

R. S.

Poppen

P. J.

(1996). Does studying economics discourage cooperation? Watch what we do, not what we say or how we play. The Journal of Economic Perspectives, 10, 177-186.