7

ORIGINAL ARTICLE

None Dare Call It Torture: Indexing and the Limits of Press Independence in the Abu Ghraib Scandal

W. Lance Bennett1, Regina G. Lawrence2, & Steven Livingston3

1 Departments of Political Science and Communication, University of Washington, Seattle, WA 98195

2 Department of Political Science, Hatfield School of Government, Portland State University, Portland, OR 97207

3 Elliott School of International Affairs, George Washington University, Washington, DC 20052

This paper considers the extent to which leading news organizations use independent

documentation to build interpretations of events that challenge official framing. The

data presented in this study show that despite available evidence and sources to support

a counterframing of the Abu Ghraib prison story in terms of a policy of torture, the

leading national news organizations did not produce a frame that strongly challenged

the Bush administration’s claim that Abu Ghraib was an isolated case of appalling

abuse perpetrated by low-level soldiers. The press struggled briefly, and in limited fash-

ion with the question of whether events at Abu Ghraib and elsewhere reflected an

administration policy of torture, but ‘‘abuse’’ was by far the predominant news frame.

The case of Abu Ghraib offers a critical test of agreement and differences among

theories of event-driven news, cascading activation, and indexing. Although all the

3 models were implicated in this case, the data, drawn from a content analysis of the

Washington Post, CBS Evening News, and a sample of national newspapers, fit most

closely with the predictions of the indexing model.

doi:10.1111/j.1460-2466.2006.00296.x

When U.S. Army Specialist Joseph Darby arrived at his post at the Abu Ghraib prison in Iraq in November 2003, he heard about a shooting in Tier 1A. He asked the military police officer in charge of the area, Specialist Charles Graner, if there were

any photos of the site. Graner gave him two CDs of photos, but they were not what Darby expected to see. As a Washington Post story later put it, those images would

soon ‘‘become iconic, among them, the naked human pyramid, the hooded man standing on a box hooked up to wires ..’’ (Higham & Stephens, 2004). It is a mea-

sure of the photographs’ impact that they could be described as ‘‘iconic’’ only 3 weeks after CBS’s 60 Minutes II made the photos public on April 28, 2004.

The photos may have become icons for the world, but inside the United States, their meaning became the object of a political framing contest that raised questions

Corresponding author: W. Lance Bennett; e-mail: [email protected].

Journal of Communication ISSN 0021-9916

Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association 467

about how much independent evidence the mainstream press needs about an event before it can challenge official government definitions. President Bush, administra-

tion officials, and other Republican party leaders immediately and emphatically labeled the events at Abu Ghraib isolated cases of ‘‘mistreatment’’ and ‘‘abuse’’ at

the hands of low-level soldiers, rejecting the claims of some commentators that the photos signified a new departure in U.S. foreign policy—the deliberate torture of terror suspects. The data reported here allow us to assess the relative power of

dramatic events, independent documentation, and elite debate in shaping news framing of U.S. foreign policy. These news dynamics are important to understand

because they determine levels of U.S. journalistic independence that enable the ‘‘watchdog’’ role that is so important to democratic theory (Bennett & Serrin,

2005; Cook, 2005). In particular, this study critically examines the degree to which press indepen-

dence is enhanced by event-driven news dynamics (Bennett & Livingston, 2003; Lawrence, 2000; Livingston, Bennett, & Robinson, 2005; Livingston & Van Belle, 2005). Abu Ghraib initially became big news because digital cameras in the hands of

military personnel enabled the press to build a story that was largely buried behind Pentagon walls before the photos emerged. The question is how that high-profile

story subsequently unfolded in ways that might have affected opinion formation and government responses (Lawrence & Bennett, 2000).

Models of news events and framing

Previous research has shown that dramatic and troubling events can provide legit-

imizing pegs to support relatively independent and critical news narratives, even allowing the news media to set the agenda more proactively than usual (Bennett & Lawrence, 1995; Lawrence, 2000; Molotch & Lester, 1974). For example, two major

U.S. news events that seemingly demanded interpretation—the 1991 beating of Black motorist Rodney King by officers of the Los Angeles Police Department and the 1999

shooting at Columbine High School in Littleton, Colorado—became opportunities for the news to deliberate about problems those events seemed to signify (Lawrence,

2000). The Columbine shooting was framed in terms of a variety of social problems, such as the corrosive effects of widely available guns and a violent popular culture

(Lawrence, 2001; Lawrence & Birkland, 2005). Examples of the event-driven news model in foreign policy include the so-called CNN effect, in which the media pur- portedly use gripping events to draw U.S. policy makers into the foreign policy fray

(Livingston, 2000; Potter, 2002). The ‘‘indexing’’ hypothesis (Bennett, 1991) would predict that most of this event-

driven news soon becomes constrained by the standard journalistic practice of tying, or indexing, story frames to the range of sources and viewpoints within official

decision circles, reflecting levels of official conflict and consensus (Bennett; Bennett & Klockner, 1996; Entman & Page, 1994; Hallin, 1986; Mermin, 1999; Zaller & Chiu,

1996). Most independent story frames introduced by journalists are soon reigned in

None Dare Call It Torture W. L. Bennett et al.

468 Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association

by officials who strive to ‘‘get on top’’ of an event-driven story, particularly in foreign policy news, where journalists may be especially reluctant to probe far beyond the

cues offered by highly placed U.S. sources. Livingston and Bennett (2003) examined 8 years of international stories on CNN and found that even though new news-

gathering technologies have allowed ever greater numbers of event-driven stories to appear, officials ‘‘seem to be as much a part of the news as ever’’ (Livingston & Bennett, 2003, p. 376).

There are, of course, some chinks in this routine relationship between the govern- ment and the press. In foreign policy, some observers point to the weakening of the

Cold War bipartisan consensus as a possible opening for more independent reporting (Entman, 2004a). Althaus (2003) similarly proposes that journalists may exercise

greater ‘‘discretion in locating and airing oppositional voices’’; yet, his empirical examinations conclude that today’s mainstream press does not ‘‘produce many bold

statements of fundamental criticism’’ of U.S. foreign policy (Althaus, 2003, p. 404). Overall, this research indicates that the press act more as ‘‘guard dogs’’ of the

foreign policy establishment than its watchdogs (Donahue, Tichenor, & Olien,

1995). At some point, the Internet may make it possible for more Americans to gain a different perspective through foreign news outlets, but the arc of news framing and

public opinion during the Iraq war suggests that indexing still cues opinion. What becomes important for theory building is to understand more precisely the

limits of indexing and the room provided by events for more independent press framing. The above research suggests that event-driven and indexing dynamics are

not diametrically opposed models of the news. Rather, these dynamics coexist and are often intertwined. Some (though not all) of the predominant frames in the

Columbine story, for example, were suggested to reporters by leaders of the Repub- lican and Democratic parties who were eager to capitalize politically on the event (Lawrence & Birkland, 2005). The curious result of the interplay of events and official

framing is a ‘‘semi-independent press’’ characterized by moments of relative inde- pendence within a more general pattern of compliance with government news man-

agement (Bennett & Livingston, 2003). Entman’s (2004a) model of ‘‘cascading activation’’ (borrowing a metaphor from

literature on ‘‘knowledge networks’’ and ‘‘spreading activation’’) incorporates aspects of the event-driven and indexing models but adds finer gradations of political

power. According to this model, ‘‘the ability to promote the spread of frames is stratified’’ across several levels of U.S. political players, from the White House and the Departments of Defense and State at the top, to members of Congress, ex-

officials, and well-placed experts one step below, to news organizations further down (p. 11). News frames are most likely to be activated at the top and spread down the

ladder, but counterframes may also work their way up. As in the indexing model, Entman (2004a) posits that journalists are more likely to create counterframes,

when officials at the top levels are not united, but his model adds additional counterframing circumstances: when midlevel sources pushing alternative frames

are readily available and when the events being covered are culturally ambiguous.

W. L. Bennett et al. None Dare Call It Torture

Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association 469

Seen through these models, news frames are predictable responses of the media to particular sets of event characteristics, cultural values, and political power rela-

tions in government. The event-driven model suggests that the greatest press dis- cretion in framing choices may be in the immediate aftermath of an event, when

news organizations offer dramatic imagery that challenges the news management skills of government. The indexing model suggests that subsequent news framing will narrow or widen depending on how officials respond to the story: If officials present

a unified front, critical counterframes may be curtailed; on the other hand, if serious official debate breaks out, particularly with policy-making implications, counter-

frames may expand. According to the cascading activation model, the degree of counterframing also depends largely on divisions among elites, but troubling cultural

resonances and the efforts of peripheral sources can also result in more assertive journalistic counterframes. The Abu Ghraib story offers an opportunity to critically

assess these three perspectives on news frames, with implications for a general model of press–government relations.

Applying the models to Abu Ghraib

The case of Abu Ghraib offers a critical test of agreement and differences among these theories, based on several elements. The event-driven model is in play because

photographic and documentary evidence emerged that was not fully controlled by government elites. Leading news organizations independently publicized the photos

and advanced the story by publishing images of hooded and leashed prisoners, pyramids of naked bodies, and other indelible scenes (although CBS, at the Bush

administration’s request, initially held its Abu Ghraib story for 2 weeks). Moreover, the events depicted in the photos demanded interpretation as they were highly challenging to Americans’ social identity as a morally upright nation (Jones &

Rowling, 2006). Indeed, interpretations varied across the political spectrum, from cultural critic Susan Sontag’s assessment that the ‘‘torture’’ pictured in the pho-

tographs showed U.S. policies and attitudes at their starkest (Sontag, 2004) to radio pundit Rush Limbaugh’s assertion that the photos showed nothing more than the

‘‘need to blow some steam off ’’ among beleaguered soldiers (quoted in Meyer, 2004). These characteristics support an event-driven news prediction of critical counter-

frames to compete with official explanations. Our data show an early, modest emer- gence of such frames, followed by their startling disappearance.

A cascading activation model is engaged by the relative difficulty for the admin-

istration of framing Abu Ghraib in a favorable light, and by the presence of credible midlevel sources that framed events at Abu Ghraib in terms of larger problems of

neglect, abuse, and even officially authorized torture at U.S.-run facilities in several countries. The Washington Post reported for the record at least 12 distinct sources of

evidence of systemic problems pointing to a possible pattern of torturing detainees at detention sites in Iraq and elsewhere (Bennett, Lawrence, & Livingston, 2005). For

example, journalists quickly gained access to a summary report of 14 investigations

None Dare Call It Torture W. L. Bennett et al.

470 Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association

of U.S. detention facilities in Iraq conducted by the International Committee of the Red Cross (ICRC) in 2003, which charged that military police repeatedly engaged in

‘‘excessive and disproportionate use of force . resulting in death or injury’’ of detainees, and described the routine stripping, humiliation, and physical mistreat-

ment of prisoners as ‘‘tantamount to torture’’ (ICRC, 2004). Credible accounts by independent journalists also supporting a torture policy frame were also available to mainstream news organizations throughout the story (Danner, 2004a, 2004b; Han-

ley, 2003a, 2003b; Hersh, 2004a, 2004b, 2004c). In addition, a series of government investigative reports made public throughout the spring and summer of 2004 con-

tained evidence that the photographs captured only the tip of the iceberg in terms of conditions in U.S. detention centers (Danner, 2004a; see, e.g., Schlesinger, 2004). A

general who authored one of the reports conceded that some of the events at Abu Ghraib qualified as torture (White, 2004). Moreover, leaked governmental memos

showed a list of legally questionable interrogation techniques—some of which appear in the photos from Abu Ghraib—that were authorized by the Defense Department for use at Guantanamo Bay, Cuba (Priest & Stephens, 2004; UN

Commission on Human Rights, 2006). Additional memos revealed that White House Counsel Alberto Gonzales (who soon would be nominated for U.S. Attorney

General) had loosened the official definition of torture to allow more coercive interrogation techniques against detainees in the war on terrorism (Zernike,

2004). According to the event-driven and the cascading activation models, these sources might lead the mainstream press to counterframe this event-driven story.

Our data indicate that such framing was at best limited. The indexing hypothesis is also in play in this case. As various levels of the

Executive Branch and Congress became publicly engaged with the story, public statements were made, documents released, and hearings held. Yet, a consistent and organized political opposition within government or from presidential candi-

date John Kerry failed to arise. The absence of such official opposition supports an indexing prediction that Bush administration framing would trump the available

evidence and lesser sources that challenged it. In short, this case enables us to assess the relative workings of indexing, event-driven, and cascading activation models in

accounting for the framing of the story.

Why the frame matters

It is important to make clear that we do not claim that torture was the only correct

label for the events at Abu Ghraib. Nor do we argue that competing available frames for Abu Ghraib ‘‘should’’ have had absolutely equal footing in the news. However,

given the array of sources described briefly above, ‘‘torture policy’’ was a counter- frame worthy of public discussion, and it was a frame supported by evidence and

sources available to mainstream journalists.1 The empirical question becomes, to what extent and under what circumstances did news organizations highlight the

torture frame versus the administration’s preferred ‘‘isolated abuse’’ frame? We seek

W. L. Bennett et al. None Dare Call It Torture

Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association 471

the best empirical explanation of press behavior to better understand what the press responds to when it frames news stories.

As an empirical assessment tool, we employ Entman’s useful standard of ‘‘coun- terframing.’’ Significant counterframing occurs when the media ‘‘provide enough

information independent of the executive branch that citizens can construct their own counterframes of issues and events,’’ which requires an alternative frame to attain ‘‘sufficient magnitude to gain wide understanding as a sensible alternative to

the White House’s interpretation’’ (2004a, p. 17). In other words, a meaningful frame contest involves at least two coherent frames presented often and prominently.

Our data allow us to assess whether leading national news organizations (The Wash- ington Post, CBS Evening News, and a sample of 10 national papers) offered a counter-

frame as measured by this standard, and why they did or did not. Beyond our concerns about theory, the quality of mainstream national discus-

sion about Abu Ghraib was important for illuminating in a timely fashion various political questions in a timely fashion: damage to America’s global reputation, pos- sible high-level legal and policy transgressions, and the cultural self-understanding of

Americans. For example, if policies were created that separated the Bush adminis- tration from both domestic and international laws on torture, then the United States

had become a rogue nation with its troops subject to torture by other nations and its leaders subject to war crimes charges—matters that might warrant public scrutiny

(Sikkink, 2005). If, on the other hand, the events at Abu Ghraib were isolated cases of prisoner mistreatment, then existing procedures for prosecuting individual soldiers

were adequate to contain the problem. A sustained press debate including a torture counterframe could also have had

domestic political consequences. A survey by the Program on International Policy Attitudes (PIPA) released in July, 2004 (PIPA/Knowledge Networks Poll, 2004), showed 66% of respondents agreeing that ‘‘governments should never use physical

torture’’ and 60% agreeing that the United States should extend war treaty rights to unconventional combatants (see also Pew Research Center, 2004). Large majorities

regarded the United States as a ‘‘moral leader’’ that ‘‘should not lower itself by engaging in torture or cruel or degrading treatment.’’ Yet, only 35%–55% of respon-

dents knew that the secretary of defense had authorized some practices similar to those depicted in the photos: hooding, using dogs to frighten detainees, and forcing detain-

ees to go naked. Among those who knew that these were official policies, 59% said they were less likely to vote to reelect President Bush. For these reasons, in addition to building news framing theory, we find this study interesting and important.

Methods

A crucial element of frames is their ability to arouse or suppress moral and political

judgments by categorizing events in particular ways (Edelman, 1988; Entman, 2004a, pp. 36–42). We track four labels that were most prevalent in news coverage of Abu

Ghraib: mistreatment, ‘‘scandal,’’ abuse, and torture. Mistreatment and abuse

None Dare Call It Torture W. L. Bennett et al.

472 Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association

include neglectful behavior, consistent with administration framing, whereas torture is strongly intentional and has a stronger connection in both common usage and

legal terminology to interrogation policies and practices.2 Moreover, the label torture might bring attention to policy initiatives up the chain of command to the secretary

of defense and possibly even the president. The labels assigned to Abu Ghraib thus offered broad cues to policy makers and publics about the meaning of events and appropriate reactions.

We began our analysis with the Washington Post, which quickly became the lead U.S. news organization on the story, fueled by some 1,000 photos that it continued to

review and release over time. We searched the Nexis database for all the Post’s news and editorial items mentioning Abu Ghraib between January 1, 2004, and August 31,

2004. The specific search term used was ‘‘Abu Ghraib or (Iraq and prison!).’’ This time frame begins just before sketchy reports of a Pentagon investigation of Abu

Ghraib appeared in the press on January 17 and ends with the final congressional hearing that put the story to rest before the 2004 election campaign entered its final stage. Our analysis concentrates on the Post’s ‘‘big story’’ period between the end of

April and the end of August, 2004. Prior to April 29, only two stories appeared in the Post, on March 21 and 22, noting that criminal charges were filed against soldiers for

abusing prisoners. Scattered reports of an investigation at Abu Ghraib were largely ignored by the media until CBS broke the story of the photos on April 28, after which

the Post steadily published from its cache of photos and assigned its large reporting force to the story.

The Washington Post sample of 609 items included many articles that made only passing reference to Abu Ghraib or Iraqi prisons in the context of stories about other

topics on the war or the Bush administration. We selected only those articles that focused directly on events at Abu Ghraib and/or U.S. policies related to treatment of prisoners, prisons, interrogations, and related matters. In addition, letters to the editor

were removed because our focus is on journalistic framing. This left a sample of 294 items (242 news articles and 52 masthead editorials, columns, and op-ed pieces).

These 294 items were then coded by a trained coder who was unfamiliar with the theoretical propositions of this study. The coded variables reported here are the

‘‘primary label’’ and, if present, the ‘‘secondary label’’ assigned to prisoner treatment in each article. The labels were chosen from a list previously identified by the authors

through a close reading of stories, with labels winnowed by automated word searches to eliminate infrequent terms. The final list from which the coder was instructed to choose included abuse, mistreatment, scandal, torture, or ‘‘none of the above.’’ The

coder was also asked to determine whether a primary label appeared in the headline or lead paragraphs of the story (‘‘primary label placement’’).

Coding reliability was assessed using a second coder and a subsample of 61 articles. Intercoder reliability scores were high, with intercoder percentage agree-

ments of p = .807 for the primary label, p = .912 for the placement of the primary label in the headline or lead paragraph, and p = .754 for identification of the

secondary label. Because percentage agreement is often too liberal a measure of

W. L. Bennett et al. None Dare Call It Torture

Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association 473

reliability, we calculated both Cohen’s k and Krippendorf’s a, which also proved strong, with primary label identification reliabilities of a = .776 and k = .775.

Because error in the first label code magnified error in the second label code, we also merged the ‘‘primary’’ and ‘‘secondary’’ labels into a single, ‘‘prominent label’’

variable that turned out to be the most generous measure of whether torture appeared as a counterframe (by counting it either in the first or second position), while producing strong intercoder reliability coefficients of p = .917, a = .801, and

k = .800. The authors also conducted coder discussions of their disagreements, which revealed nuances such as a number of instances in which the torture label appeared

as a denial rather than an affirmative description of events at Abu Ghraib. A simplified version of this coding strategy was repeated on the full text of Abu

Ghraib stories aired on the CBS Evening News between April and August 2004, chosen because that network’s 60 Minutes II first broke the Abu Ghraib photo story.

The stories were gathered and culled using the same procedure as for the Washington Post, yielding a sample of 54 stories. Intercoder reliability on the primary and sec- ondary label data for the entire sample of CBS Evening News was high due to the

shorter and simpler TV story format (p = .98 and p = .96, respectively). Out of space considerations, the CBS data are not reported in tables but are discussed in the text.

Finally, we assessed our Washington Post and CBS findings against a national newspaper sample with an extended time period to include the Senate confirmation

hearings for Alberto Gonzales, who participated in drafting the White House policy memos justifying relaxed conventions against torture in the war on terror. This study

involved machine coding Nexis search results to ascertain the frequency of the same labels from the above content analyses. The national sample included news and

editorials from 10 newspapers (the Atlanta Journal Constitution, Boston Globe, Chicago Sun-Times, Los Angeles Times, New York Times, Cleveland Plain Dealer, San Francisco Chronicle, Seattle Times, St. Petersburg Times, and USA Today) between

April 2004 and mid-January 2005.3 The results of these analyses are reported in the next section.

Findings

In the pages of the Washington Post, even at the height of the Abu Ghraib story, the

most prominent categorization by far was abuse, with torture barely appearing in the news coverage, and only slightly more often in editorials. Table 1 shows the primary label frequencies in the news stories and editorials. The frame imbalance in the news

was overwhelming, with just 3% (9) of the stories offering torture as the primary frame, compared to 81% (188) offering abuse as the primary frame. Adding mistreatment and

scandal accounted for the balance of the primary frames, meaning that our four terms comprised 99% of the primary framing categories. The editorials were a bit more likely

to introduce torture, but only 17% led with torture, whereas 61% led with abuse. Recall that we also constructed both a tougher and a more relaxed measure of

frame strength. The tougher standard is whether the frame term appeared in the

None Dare Call It Torture W. L. Bennett et al.

474 Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association

headline or lead paragraph. By this measure, only 2 of 242 news articles in the Post offered torture as a strong cue for reader interpretation. Even in these two articles,

qualitative analysis reveals subtle distancing of torture from Abu Ghraib. The open- ing paragraph of the Post’s May 11 article, headlined ‘‘The Psychology of Torture,’’

reads, ‘‘The U.S. troops who abused Iraqis at the Abu Ghraib prison near Baghdad were most likely not pathological sadists, but ordinary people who felt they were doing the dirty work needed to win the war, experts in the history and psychology

of torture say’’ (emphases added). Even the rare news mentions of torture tended to evoke the term in the abstract, whereas abuse was used to describe events at

Abu Ghraib. Our most relaxed measure counted a first or a second frame term anywhere in

the article as equal. By this standard, torture was still a remote challenger to the abuse frame. Table 2 shows that 91% of news articles and 82% of editorials in the

Washington Post used abuse as the first or second label to describe events at Abu Ghraib, compared with 11% and 30%, respectively, that used the term torture.

The Post’s preference for the term abuse was most pronounced on the news pages, where it was used 10 times more often than torture. Editorials used abuse three times as often as torture. A similar pattern prevailed on CBS, where 50 out

of 54 irst or second label, compared with 10 (18%) that used torture as either first or second frame.

To assess whether event-driven news can sustain independent framing, however, weak it may be, we looked at the frequency distribution of the frames over time. The

results show that torture was not only a weak frame but also not a consistent back- ground element in the developing story. Torture appeared most prominently in the

Table 1 Primary Labels Used to Describe Abu Ghraib, by Type, Washington Post, April 1,

2004, to September 29, 2004a

Abuse Torture Mistreatment Scandal

News (n = 242) 81% (188) 3% (9) 3% (7) 12% (29)

Editorials (n = 52) 61% (32) 17% (9) 3% (2) 13% (7)

a These data are based on the first label used in each article. Numbers in parentheses are the

counts for each cell; percentages are not rounded.

Table 2 Percentage of Articles Using Each Label Prominently (either first or second label), by

Type, Washington Post, April 1, 2004, to September 1, 2004 (n = 294)a

Abuse Torture Mistreatment Scandal

News (n = 242) 91% (222) 11% (28) 16% (40) 33% (80)

Editorials (n = 52) 82% (43) 30% (16) 13% (7) 36% (19)

a Data in this table reflect labels used as either the first or the second label in each article; for

this reason, row totals do not equal the total n of stories. Numbers in parentheses are the

counts for each cell; percentages are not rounded.

W. L. Bennett et al. None Dare Call It Torture

Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association 475

2 weeks after the story broke, and then faded quickly as event-driven reportage on the photos was displaced by managed governmental activities such as investiga-

tions, reports, and hearings. In the Washington Post, 15 out of 28 (54%) appearances of torture as either a first or second frame occurred during the first 2 weeks after

the photos broke, then dropped to one the following week and never exceeded three in any week the rest of the summer. By the third and fourth weeks, abuse was prom- inent in 56 stories, scandal in 28, and torture in just three. Similarly, on CBS Evening

News, 8 of the 10 uses of torture as either a first or second label appeared between April 29 and May 12. Thereafter, torture was used prominently only twice to describe

events at Abu Ghraib. Befitting its role as a leading national paper, the Post was followed by other

mainstream national news organizations in framing the story. (‘‘Following’’ is used in a special sense here because indexing permits the national press to behave as

a relatively homogeneous institution without slavishly sacrificing competitive rela- tions.) Our sample of 10 national newspapers from April 1, 2004, through January 18, 2005, produced a total of 895 news articles and editorials about Abu Ghraib. Fully

97% mentioned at least one of the three labels, mistreatment, abuse, and torture; 60%, however, did not mention the term torture at all, and 35% (primarily edito-

rials) used it in conjunction with one of the other labels. Only nine (1%) items in the national press used the term torture alone.

The data reveal some variation across the newspapers, but there was far greater variation in volume of coverage than in framing. For example, the New York Times

published 210 stories about Abu Ghraib during this time period, and the Cleveland Plain Dealer published only 10. The Times also showed the highest proportion of

counterframes, but even in the Times, torture stood alone in only five items (2%), whereas 62% of Times articles and editorials made no reference to torture at all, putting it close to national averages on both counts.

These findings are displayed graphically in Figure 1, which shows the results of four distinct searches of stories about Abu Ghraib in these 10 newspapers. The top

line (‘‘any label’’) shows all stories that mentioned either abuse, mistreatment, or torture; for ease of analysis and presentation, and because we discovered that scandal

so often appeared in conjunction with abuse, scandal was dropped from this stage of the analysis. The second line (‘‘no torture’’) shows all articles that mentioned

either mistreatment or abuse but did not mention torture. The third line (‘‘other 1 torture’’) shows all stories that mentioned torture but also used the terms abuse or mistreatment. The bottom line (‘‘torture only’’) represents all stories in which

only the label torture and none of our other main labels appeared.4

This torture-only line shows how rarely the torture label stood alone. Even at the

height of coverage in early May, the number of items solely using the term torture was small, and almost all appeared on the editorial pages, not in the news itself. If

editorials are excluded from Figure 1, the torture-only line literally disappears. The predominance of the torture 1 other trend line over the torture-only line shows that

when torture did appear, it generally was paired with—and softened by—other

None Dare Call It Torture W. L. Bennett et al.

476 Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association

labels. Overall, the bulk of stories on Abu Ghraib did not include the term torture at all.

As with the Washington Post, most mentions of torture in both CBS and the national news sample appeared immediately after the photos broke. For example,

CBS anchor Dan Rather led a May 3 story by saying that ‘‘evidence of mistreatment, even torture of Iraqi prisoners has inflamed many in the Muslim world.’’ Reporter

Bill Plante also led with torture in a May 5 story: ‘‘Around the world newspapers and magazines have seized on the torture to pour scorn on American promises of

democracy for Iraq.’’ During this early period, torture seemed to enter the news as an event-driven model might predict, as in CBS’s May 6 report that the Post had ‘‘published new images of humiliation and degradation, one of which seemed to

verge on outright torture’’ (Martin, 2004). A glimmer of support for cascading activation was also provided by the initial use of peripheral sources such as the

ICRC. Yet, the failure to achieve a full cascade leading to a robust counterframe is evidenced by journalists’ general reluctance to independently introduce the term

torture after the first few days. By contrast, abuse was often used by journalists on their own. Our Washington Post coding reveals that from April through the end of

September, 95% of the instances of abuse in the news pages (n = 179) were in reporters’ own words, contrasted with only 55% (n = 5) of the instances of torture.

0

50

100

150

200

250

300

4/ 1/

20 04

4/ 15

/2 00

4

4/ 29

/2 00

4

5/ 13

/2 00

4

5/ 27

/2 00

4

6/ 10

/2 00

4

6/ 24

/2 00

4

7/ 8/

20 04

7/ 22

/2 00

4

8/ 5/

20 04

8/ 19

/2 00

4

9/ 2/

20 04

9/ 16

/2 00

4

9/ 30

/2 00

4

10 /1

4/ 20

04

10 /2

8/ 20

04

11 /1

1/ 20

04

11 /2

5/ 20

04

12 /9

/2 00

4

12 /2

3/ 20

04

1/ 6/

20 05

Any Label No Torture Other + Torture Torture Only

Figure 1 Mentions of ‘‘torture’’ and other labels in connection with Abu Ghraib, national

newspaper sample, April 1, 2004, to January 19, 2005.

W. L. Bennett et al. None Dare Call It Torture

Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association 477

Consistent with an integrated event-driven indexing dynamic, journalistic license occurred early in the story before official news management was fully established.

The torture frame briefly reemerged—again diminished by other labels—in August with the release of the official Fay and Schlesinger investigation reports.

These reports occasioned two hard-hitting masthead editorials in the Post, one arguing that they ‘‘have dragged the Bush administration and Pentagon brass a cou- ple of steps closer to facing the truth about how and why U.S. soldiers and inter-

rogators committed scores of acts of torture and abuse in Iraq and Afghanistan’’ (Closer To the Truth, 2004). Notably, the same editors had all but purged the term

from the news articles that cue public opinion, yet introduced it in editorials that speak largely to other elites.

As Figure 1 shows, torture also reentered the national news in December and January in conjunction with the Gonzales confirmation hearings, almost always in

reference to the infamous ‘‘torture memo,’’ as official sources at congressional hear- ings, followed by reporters, labeled it when it first emerged in May. Yet, stories and editorials seldom linked the memo or the practice of torture directly to Abu Ghraib.

Indeed, the Post mentioned ‘‘Gonzales’’ and torture together in 37 items appearing between mid-December 2004 and January 2005—far more than it prominently used

that term in nearly 300 articles between April and September of 2004. Yet, only nine of those items linked torture to Abu Ghraib, and only two of those were news stories

(each mentioned Abu Ghraib only once, deep within each story). Thus, in much of the coverage, the possible connection between Gonzales’s memo and the scenes

depicted in the photos remained unclear. The photos that might have enabled some greater degree of (event driven) press independence—supplemented by ample

sources available to activate a counterframe of torture—ended up being framed, following the administration’s lead, as abuse by the end of the story.

Discussion

Why the mainstream press declined to counterframe Abu Ghraib in a way that challenged Bush administration framing is all the more interesting in light of the

fact that editors at two leading organizations did not view the framing as particularly difficult or even conscious choices. The New York Times’ public editor Daniel Okrent

asked the paper’s editors why they settled on abuse rather than torture—a choice that Okrent described as ‘‘comparatively delicate’’ given the evidence at hand. Okrent reported that the editors ‘‘were surprised when I raised the issue.’’ Both denied that

the Times had a policy one way or another, but acknowledged that ‘‘readers may be right’’ that torture was a more appropriate label. One responded simply, ‘‘Now that

you tell me people are reading things into our not using torture’ in headlines, I’ll pay closer attention’’ (Okrent, 2004). Washington Post editor Leonard Downie held an

online chat with readers, and his response to a similar question suggested that his choice was more deliberate: ‘‘Abuse is obvious from the information and images

we have, and is serious in its own right. Torture is a more loaded term and its

None Dare Call It Torture W. L. Bennett et al.

478 Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association

use requires more information about whether the abuse constitutes torture’’ (Downie, 2004).

Because journalists generally try to present at least two sides of every story, Downie’s rationale is puzzling, as support for another side of the story was available.

By the time Downie made this statement on May 21, considerable documentation had already emerged about high-level policy discussions on coercive interrogation procedures, along with the other evidence of torture noted earlier—evidence the Post

had drawn upon in its own reporting (Bennett et al., 2005). For example, the Post’s own report on the Gonzales memo quoted the White House counsel as saying, ‘‘In

my judgment, this new [war] paradigm renders obsolete Geneva’s strict limitations on questioning of enemy prisoners and renders quaint some of its provisions’’

(Kessler, 2004). A cascading activation analysis might predict that in conjunction with the photos, such evidence would prompt amplification of a torture policy

frame. Instead, the Post’s initial thousand-word story on the Gonzales memo used the word torture only once, in an ironic observation that the memo had caused the administration to postpone a State Department report on the U.S. commitment to

international human rights—a report that included fighting the practice of torture. Entman (2004b) has cited Abu Ghraib as an example of comparatively aggressive

reporting that challenged the government. We agree that the continued publication of photos pushed the government to respond. However, this seems only limited

support for a cascading activation explanation because the requisite counterframe did not fully develop. And although Entman singles out Seymour Hersh’s reporting

as particularly important in the critical framing of Abu Ghraib, our data challenge the notion that such midlevel relatively peripheral sources can either establish or

sustain a counterframe absent higher level support. Although Hersh’s (2004a, 2004b, 2004c) reporting may have shaped the inside-Washington conversation about Abu Ghraib, his strongest claim—that the United States had embarked on a new

policy of torture in the war on terrorism—was not picked up prominently or con- sistently by the rest of the national media.

Perhaps, the reason is that the idea of Americans engaging in torture represents a cultural incongruence that ‘‘short circuits’’ critical frames (Entman, 2004a; Jones &

Rowling, 2005). Following this logic, the torture frame would not fit easily within the ‘‘defensive’’ reporting style documented in studies of war coverage. ‘‘When one’s

own combatants are involved in killing civilians,’’ one study finds, ‘‘journalists usu- ally adopt techniques that lower the emotional impact of such stories’’ (Wolfsfeld, Frosh, & Awabdy, 2005, p. 5; see also Fishman & Marvin, 2003). This pattern extends

to coverage of domestic policing as well, where the term ‘‘police brutality’’ occurs only under particular circumstances (Lawrence, 2000).

The cultural explanation, while plausible, is not fully satisfactory because torture did appear, if only briefly, as a candidate for a counterframe at the outset of the story.

Moreover, the term torture eventually became more prominent in the news as Senator John McCain and other leaders pressured the White House to support an

amendment further limiting the cruel and inhuman treatment of prisoners overseas.

W. L. Bennett et al. None Dare Call It Torture

Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association 479

A search of the Nexis database shows that of 54 articles mentioning the McCain amendment in the Washington Post between October and December 2005, fully 42

articles, or 77%, included the term torture. These patterns suggest that, rather than meeting with a blanket cultural prohibition on discussing torture, the Abu Ghraib

photos enabled an event-driven news pattern to briefly and tentatively challenge the news management capacities of officials, but the fragile event-driven news dynamic faded as the administration aggressively took over the framing virtually unchallenged

by other top-level officials. Within days of the release of the photos, the story began to fall into a familiar pattern of indexing and the counterframe of torture was pushed

out of the news by a deluge of official events that promoted the terms abuse and mistreatment. (Some of these events were no doubt planned during the timely

reprieve from an earlier release of the pictures that the CBS granted the government.) Among the official actions that arguably blocked a potential cascade of other

sources and evidence supporting a challenging counterframe were the following (with the dates they appeared in the Washington Post). On April 30, 2 days after the CBS photo story aired, an initial anonymous government official said the gov-

ernment ‘‘had taken several steps to stop the mistreatment of prisoners,’’ whereas President Bush introduced the frame that he would consistently repeat, ‘‘The actions

of a handful of soldiers . should not taint the tens of thousands who serve honor- ably in Iraq.’’ On May 1, a televised statement by President Bush inserted in a Rose

Garden photo op with the Canadian prime minister referred to the ‘‘treatment’’ of prisoners but promised that ‘‘abuses’’ would be punished. (The Post echoed the term

in its account, but also reported international shock, and introduced the term tor- ture, citing an article in Tehran Times). On May 2, an army investigative report by

Major General Antonio Taguba was released, localizing the abuse problem to, as the Post reported it, ‘‘the willful actions of a small group of soldiers’’ and ‘‘a failure of leadership’’ at the prison level. On May 5, Bush addressed the Arab world in a tele-

vised speech that characterized events at Abu Ghraib as regrettable abuses. On May 7, Secretary of Defense Rumsfeld, testifying before a joint Senate-House Armed Serv-

ices Committee hearing, called the incidents abuse and mistreatment; the same day, the House overwhelmingly passed a resolution that deplored the mistreatment of

Iraqi detainees and repeated the president’s frame: ‘‘The alleged crimes of a handful of individuals should not detract from the commendable sacrifice’’ of U.S. soldiers in

Iraq. On May 14, Rumsfeld made a surprise visit to Abu Ghraib and told reporters the soldiers involved in the scandal would ‘‘be brought to justice.’’ On May 15, the Defense Department issued this punctuating statement: ‘‘No responsible official of

the Department of Defense approved any program that could conceivably have been intended to result in such abuses.’’ This series of official events ensured that the abuse

label would be prominent in news linked to Abu Ghraib. By May 19, while the Senate Armed Services committee held hearings on the abuses, the term torture had all but

disappeared from the news. The weakness of the torture frame is not attributable merely to this administra-

tion news management campaign but to the absence of a strong democratic challenge

None Dare Call It Torture W. L. Bennett et al.

480 Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association

either from Congress or from the presidential campaign. Although some prominent Democrats used the congressional hearings to probe high-level culpability for the

scandal, few framed events at Abu Ghraib as torture. In our Washington Post data, no prominent instances of the torture frame were pegged to members of Congress,

although a broader search found Senator Edward Kennedy cited deep in one story as saying that ‘‘Shamefully, we now learn that Saddam’s torture chambers reopened under new management: U.S. management’’ (VandeHei, 2004).4 Though he would

later call for Secretary Rumsfeld’s resignation, Democratic presidential nominee John Kerry also repeated the president’s framing virtually verbatim, saying ‘‘We

cannot let the actions of a few overshadow the tremendous good work that thou- sands of soldiers are doing every day in Iraq and all over the world’’ (Shanker &

Steinberg, 2004). When asked to react to Kennedy’s statement, Kerry said, ‘‘He’s my friend and I respect him, but I don’t agree with the framing of that’’ (VandeHei,

2004). With President Bush affirming his faith in Rumsfeld early in the frame contest, the avenues for sustaining a challenging story were closed, at least as played by the rules of the U.S. press system.

Conclusions

For all the photos and available evidence suggesting a possible policy of torture laid

bare at Abu Ghraib, the story quickly became framed as regrettable abuse on the part of a few troops. The early limited appearance of the torture frame followed by its

quick demise suggests that event-driven frames, particularly in matters of high consequence, are seriously constrained by mainstream news organizations’ deference

to political power. Lacking any consistent counterframing by high-level officials, the national media declined to challenge the administration. Leonard Downie’s dictum about lacking enough information to raise sustained questions about torture in the

Post’s coverage might be translated as the operating code of the elite press: who (in the political hierarchy of sources) offered what (officially sanctioned) evidence of

torture is the essential question. These kinds of framing rules separate the main- stream press from the alternative press, which did in many cases apply the torture

frame to the story (see, e.g., Danner, 2004a; Fairness and Accuracy in Reporting, 2004; Gilson, 2004).

In conclusion, the framing of events at Abu Ghraib by the mainstream press followed the predictable pattern of indexing. Yet, we see how, with a bit more division in policy circles or on the campaign trail, a counterframe might have been

activated and, perhaps, sustained in a cascading fashion by the event-driven dynam- ics of continued publication of photos. Indeed, the McCain torture bill, appearing

well over a year later, indicates that the Abu Ghraib story lingered on to influence public debate in a more indirect and long-term fashion. However, at the height of

attention to the Abu Ghraib story, when public opinion was in the formative stage, the mainstream media allowed the administration’s ‘‘isolated abuse’’ frame to dom-

inate the news and declined to offer the public a coherent alternative frame.

W. L. Bennett et al. None Dare Call It Torture

Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association 481

In this way, we see important links among theories of event-driven news, cas- cading activation, and indexing. In particular, it appears they operate in hierarchical

fashion in which openings created by indexing (either through divisions among policy makers or failures to mount news management activities) may enable events

to continue to drive challenges from sources beyond the inner circles of power. More broadly, we see rather vividly the limits on press independence: Even when provided with considerable photographic and documentary evidence and the critical state-

ments of governmental and nongovernmental actors, the nation’s leading media proved unable or unwilling to construct a coherent challenge to the administration’s

claims about its policies on torturing detainees. As it turned out in this case, the photos may have driven the story, but the White House communication staff ulti-

mately wrote the captions.

Acknowledgments

The authors thank Tim Jones, Carolyn Lee, and Jonathan Werbin for their research assistance.

Notes

1 Some readers may believe that the term torture was simply not appropriate to describe

the activities at Abu Ghraib. However, some, though not all, of the activities pictured in

the photographs (such as the use of unmuzzled dogs and the stripping of prisoners)

clearly violate elements of the Geneva Conventions and other laws against torture to

which the United States is a signatory (see Danner, 2004a; Sikkink, 2005). As noted

above, one U.S. military investigator even conceded that some of the Abu Ghraib abuses

rose to the level of torture (White, 2004). Moreover, most of the documented injuries

and deaths in U.S. facilities in Iraq and beyond were not pictured in the photographs

from Abu Ghraib (see Danner, 2004a, 2004b), meaning that the photos were not the full

story. Finally, the scope of incidents at multiple U.S. facilities suggests something more

than low-level lapses of judgment. As one legal expert has noted, ‘‘A widespread practice

in multiple locations implies an institutional policy, not human error’’ (Sikkink, 2005,

p. 22).

2 The Encyclopedia Britannica, for example, defines torture as ‘‘the infliction of

excruciating physical or psychological pain for such reasons as punishment,

intimidation, coercion, the extraction of a confession, or the obtainment of

information’’ (http://www.britannica.com/eb/article-9073000?query=tortureampct=,

accessed October 18, 2005.). The International Convention Against Torture includes

the purpose of gathering intelligence or extracting information as one element in its

definition of torture (see United Nations Commission on Human Rights, 2006).

3 The search term used to establish the baseline of articles about Abu Ghraib in each

newspaper was ‘‘hlead (Abu Ghraib or (prison and Iraq*),’’ meaning that we asked Nexis

to find all articles in which the term ‘‘Abu Ghraib’’ or the terms ‘‘prison’’ and ‘‘Iraq’’ were

mentioned in the headline or lead of the article. Only articles focusing on Abu Ghraib

and/or U.S. treatment of detainees were included in the analysis.

None Dare Call It Torture W. L. Bennett et al.

482 Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association

4 A broader search of the Nexis database for the terms ‘‘abu ghraib and senate and torture’’

in proximity yielded only three instances between April and June 2004 in which a

Senator used the term ‘‘torture’’ in conjunction with Abu Ghraib.

References

Althaus, S. L. (2003). When news norms collide, follow the lead: New evidence for press

independence. Political Communication, 20, 381–414.

Bennett, W. L. (1991). Toward a theory of press-state relations. Journal of Communication,

40(2), 103–125.

Bennett, W. L., and Klockner, J. D. (1996). The psychology of mass-mediated publics. In A. N.

Crigler (Ed.), The psychology of political communication (pp. 89–109). Ann Arbor, MI:

University of Michigan Press.

Bennett, W. L., & Lawrence R. G. (1995). News icons and the mainstreaming of social change.

Journal of Communication, 45(3), 20–39.

Bennett, W. L., & Livingston, S. (2003). A semi-independent press: Government control

and journalistic autonomy in the political construction of news. Political Communication,

20, 359–362.

Bennett, W. L., Lawrence, R. G., & Livingston, S. (2005, March). In the Camera’s Frame: the

Contest to Define Abu Ghraib. Paper presented at the annual meeting of the International

Studies Association, Honolulu, HI.

Bennett, W. L., & Serrin, W. (2005). The watchdog role. In G. Overholser & K. H. Jamieson

(Eds.), The press (pp. 169–188). New York: Oxford University Press.

Closer to the Truth. 2004. Washington Post, August 22, A22.

Cook, T. (2005). Freeing the presses: An introductory essay. In T. Cook (Ed.), Freeing the

presses (pp. 1–28). Baton Rouge: Louisiana State University Press.

Danner, M. (2004a). Abu Ghraib: The hidden story. New York Review of Books, October 7,

44–50.

Danner, M. (2004b). Torture and truth: America, Abu Ghraib, and the war on terror.

New York: New York Review Books.

Donahue, G. A., Tichenor, P. J., & Olien, C. N. (1995). A guard dog perspective on the role of

the media. Journal of Communication, 45(2), 115–132.

Downie, L., Jr. (2004). Iraq: New abuse details. The Washington Post. Retrieved February

2, 2005, from http://www.washingtonpost.com/wp-dyn/articles/A44952-2004May21.html

Edelman, M. (1988). Constructing the political spectacle. Chicago: University of Chicago Press.

Entman, R. M. (2004a). Projections of power: Framing news, public opinion and U.S. foreign

policy. Chicago: University of Chicago Press.

Entman, R. M. (2004b, August). Punctuating the homogeneity of international news; abusing

prisoners vs. killing civilians. Paper presented at the 2004 Annual Meeting of the American

Political Science Association, Chicago.

Entman, R. M., & Page, B. (1994). The news before the storm: Limits to media autonomy

in covering the Iraq war debate. In W. L. Bennett & D. Paletz (Eds.), Taken by storm:

The media, public opinion, and U.S. foreign policy in the Gulf War (pp. 82–101).

Chicago: University of Chicago Press.

Fairness and Accuracy in Reporting. (2004, June 10). Activism update: NY Times responds

to FAIR alert on torture. Extra!

W. L. Bennett et al. None Dare Call It Torture

Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association 483

Fishman, M. J., & Marvin, C. (2003). Portrayals of violence and group difference in

newspaper photographs and media. Journal of Communication, 52(1), 32–44.

Gilson, D. (2004). Torture and truth: Tracing the origins—and aftermath—of what happened

at Abu Ghraib. Mother Jones. Retrieved January 15, 2005, from http://www.motherjones.com/

news/qa/2004/12/12_401.html

Hallin, D. C. (1986). The uncensored war: The media and Vietnam. Berkeley: University of

California Press.

Hanley, C. J. (2003a, October 29). Former Iraqi detainees tell of riots, punishment in the sun,

good Americans and pitiless ones. Associated Press.

Hanley, C. J. (2003b, October 28). More than 5,000 detainees sit and wait behind U.S. razor

wire, prison walls in Iraq. Associated Press.

Hersh, S. (2004a). Chain of command: The road from 9/11 to Abu Ghraib. New York:

HarperCollins.

Hersh, S. (2004b, May 24). The gray zone: How a secret Pentagon program came to Abu Ghraib.

The New Yorker. Retrieved from http://www.newyorker.com/fact/content/?040524fa_fact

Hersh, S. (2004c, May 10). Torture at Abu Ghraib. The New Yorker. Retrieved from

http://www.newyorker.com/fact/content/?040510fa_fact

Higham, S., & Stephens, J. (2004, May 22). Punishment and amusement: Documents

indicate 3 photos were not staged for interrogation. The Washington Post, p. A1.

International Committee of the Red Cross. (2004). Report of the International Committee

of the Red Cross (ICRC) on the Treatment by the Coalition Forces of Prisoners of War

and Other Protected Persons by the Geneva Conventions in Iraq During Arrest,

Internment and Interrogation. Retrieved February 20, 2005, from http://

www.globalsecurity.org/military/library/report/2004/icrc_report_iraq_feb2004.htm

Jones, T. D., & Rowling, C. (2006, January). Abuse vs. Torture: How Social Identity, Strategic

Framing, and Indexing Explain U.S. Media Coverage of Abu Ghraib. Paper presented at the

Southern Political Science Association Conference, Atlanta, GA.

Kessler, G. (2004, May 18). U.S. releases human rights report delayed after abuse scandal.

Washington Post, p. A15.

Lawrence, R. G. (2000). The politics of force. Berkeley: University of California Press.

Lawrence, R. G. (2001). Defining events: Problem definition in the media arena. In R. P. Hart

& B. H. Sparrow (Eds.), Politics, discourse and American Society (pp. 91–110). Lanham, MD:

Rowman & Littlefield.

Lawrence, R. G., & Bennett, W. L. (2000). Civic engagement in the era of big stories. Political

Communication, 17, 377–381.

Lawrence, R. G., & Birkland, T. A. (2005). Guns, Hollywood, and criminal justice: Defining the

school shootings problem across public arenas. Social Science Quarterly 85(5), 1193–1207.

Livingston, S. (2000). Transparency and the news media. In B. Finel & K. Lord (Eds.), Power

and conflict in the age of transparency (pp. 257–285). New York: St. Martins Press.

Livingston, S., & Bennett, W. L. (2003). Gatekeeping, indexing, and live event news: Is

technology altering the construction of news? Political Communication, 20, 363–380.

Livingston, S., Bennett, W. L., & Robinson, W. L. (2005). International news and advanced

information technology: Changing the institutional domination paradigm? In Phil Seib

(Ed.), Media and conflict in the 21st century (pp. 33–55). New York: Palgrave.

Livingston, S., & Van Belle, D. (2005). The effects of new satellite newsgathering technology

on newsgathering from remote locations. Political Communication, 22(1), 45–62.

None Dare Call It Torture W. L. Bennett et al.

484 Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association

Martin, D. (2004, 6 May). Ongoing calls for Rumsfeld to resign or be fired over Iraqi prisoner

scandal [Television broadcast]. New York: CBS Evening News.

Mermin, J. (1999). Debating war and peace: Media coverage of U.S. intervention in the

post-Vietnam era. Princeton, NJ: Princeton University Press.

Meyer, D. (2004, May 6). Rush: MPs ‘just blowing off steam.’ CBS News.com. Retrieved

February 2, 2005 from http://www.cbsnews.com/stories/2004/05/06/opinion/meyer/

main616021.shtml

Molotch, H., & Lester, M. (1974). News as purposive behavior: On the strategic use of

routine events, accidents, and scandals. American Sociological Review, 39(1), 101–112.

Okrent, D. (2004, June 1). ‘‘Torture’’ vs. ‘‘abuse’’ in The Times’s coverage of Iraqi prisons.

The New York Times. Retrieved February 24, 2005, from http://forums.nytimes.com/top/

opinion/readersopinions/forums/thepubliceditor/danielokrent/index.html?offset=31

Pew Research Center. (2004, August 18). Foreign policy attitudes now driven by 9/11 and

Iraq. Retrieved February 16, 2005, from http://people-press.org/reports/

print.php3?PageID=871

Program on International Policy Attitudes/Knowledge Networks Poll. (2004, July 22).

Americans on detention, torture, and the war on terrorism. Program on International Policy

Attitudes. College Park, MD: The University of Maryland.

Priest, D., & Stephens, J. (2004, May 9). Pentagon approved tougher interrogations.

Washington Post, p. A1.

Potter, E. (Ed.). (2002). The new information environment and diplomacy. In E. Potter (Ed.),

Cyber-diplomacy: Managing foreign policy in the 21st century (pp. 110–127). Georgetown,

ON: McGill-Queen’s University Press.

Schlesinger, J. R. (2004, August). Final report of the independent panel to review DOD

detention operations. Independent Panel to Review DOD Detention Operations.

Retrieved from http://fl1.findlaw.com/news.findlaw.com/wp/docs/dod/abughraibrpt.pdf

Shanker, T., & Steinberg, J. (2004, May 1). Bush voices ‘disgust’ at abuse of Iraqi prisoners.

New York Times, p. A1.

Sikkink, K. (2005, March). U.S. compliance with international human rights law. Paper

presented at the Annual Meeting of the International Studies Association, Honolulu, HI.

Sontag, S. (2004, May 23). Regarding the torture of others. New York Times Magazine, p. 25.

United Nations Commission on Human Rights. (2006, February 15). Situation of detainees at

Guantanamo Bay. Retrieved February 16, 2006, from http://www.ohchr.org/english/

bodies/chr/docs/62chr/E.CN.4.2006.120_.pdf

VandeHei, J. (2004, May 13). Kerry assails Bush on Iraq. Washington Post, p. A1.

White, J. (2004, August 26). Generals point to contractors, military intelligence soldiers.

Washington Post, p. A1.

Wolfsfeld, G., Frosh, P., & Awabdy, M. T. (2005, May). Journalistic mechanisms for covering

death in violent conflicts: News about the second Intifada on Israeli, Palestinian, and

Al-Jazeera television. Paper presented at the Annual Meeting of the International

Communication Association, New York.

Zaller, J., & Chiu, D. (1996). Government’s little helper: U.S. press coverage of foreign policy

crises, 1945–1991. Political Communication, 13, 385–405.

Zernike, K. (2004, June 27). Defining torture: Russian roulette, yes. Mind-altering drugs

maybe. New York Times, p. WK7.

W. L. Bennett et al. None Dare Call It Torture

Journal of Communication 56 (2006) 467–485 ª 2006 International Communication Association 485