Joan Shorenstein Center on the Press, Politics and Public Policy
Harvard Student Paper April 2010

Torture at Times: Waterboarding in the Media
By Harvard Students: Neal Desai, Harvard Law School Andre Pineda, Majken Runquist, Mark Fusunyan, Harvard College Research Team: Katy Glenn, Gabrielle Gould, Michelle Katz, Henry Lichtblau, Maggie Morgan, Sophia Wen, Sandy Wong Advisor: Thomas E. Patterson, Harvard Kennedy School

© 2010 President and Fellows of Harvard College. All rights reserved.

Abstract The current debate over waterboarding has spawned hundreds of newspaper articles in the last two years alone. However, waterboarding has been the subject of press attention for over a century. Examining the four newspapers with the highest daily circulation in the country, we found a significant and sudden shift in how newspapers characterized waterboarding. From the early 1930s until the modern story broke in 2004, the newspapers that covered waterboarding almost uniformly called the practice torture or implied it was torture: The New York Times characterized it thus in 81.5% (44 of 54) of articles on the subject and The Los Angeles Times did so in 96.3% of articles (26 of 27). By contrast, from 2002 2008, the studied newspapers almost never referred to waterboarding as torture. The New York Times called waterboarding torture or implied it was torture in just 2 of 143 articles (1.4%). The Los Angeles Times did so in 4.8% of articles (3 of 63). The Wall Street Journal characterized the practice as torture in just 1 of 63 articles (1.6%). USA Today never called waterboarding torture or implied it was torture. In addition, the newspapers are much more likely to call waterboarding torture if a country other than the United States is the perpetrator. In The New York Times, 85.8% of articles (28 of 33) that dealt with a country other than the United States using waterboarding called it torture or implied it was torture while only 7.69% (16 of 208) did so when the United States was responsible. The Los Angeles Times characterized the practice as torture in 91.3% of articles (21 of 23) when another country was the violator, but in only 11.4% of articles (9 of 79) when the United States was the perpetrator.

2

INTRODUCTION Thousands of news articles have been written over the past several years about the practice that has come to be known as ““waterboarding.”” 1 The New York Times, for example, mentions waterboarding in over 150 articles in 2007 and 2008 alone. Even before the current debate, however, waterboarding appeared with some regularity in the news throughout the 20th century, from the Philippine insurgency to World War II to the Vietnam War. In addressing waterboarding, for more than 70 years prior to 9/11, American law 2 and major newspapers consistently classified waterboarding as torture. However, since the story began receiving significant media attention in 2004, following the Abu Ghraib prisoner abuse scandal and revelations of waterboarding by the United States, media sources appear to have changed their characterization of the practice. Documenting the extent of the discrepancy between the pre––9/11 consensus that waterboarding was torture and the post––9/11 media treatment of the practice is an important first step to explaining how and why this occurred. This study seeks to quantify the treatment provided to waterboarding before and after 9/11 by reviewing coverage of the practice in the nation’’s four widest circulating newspapers. Based on our initial review of media reporting and some secondary literature, we hypothesized that the tone taken toward waterboarding by major newspapers might be somewhat more lenient in the post––9/11 era, particularly after the Bush administration authorized the practice and fear of terrorism was widespread among the public. What we found, however, through our review of thousands of articles in major newspapers, was a dramatic shift in coverage away from nearly a century of practice recognizing waterboarding as torture. This study provides details on the nature of this transformation through an exhaustive examination of over a century of reporting by the nation’’s leading newspapers.

Before 2004, ““waterboarding”” had been referred to variously as ““water torture,”” the ““water cure,”” the ““water treatment,”” el submarino (or the wet submarine), dunking, and forced ingestion, among other terms. 2 For example, Court Martial of Major Edwin F. Glenn, Samar, P.I., April 1902 (reprinted in Leon Friedman, THE LAW OF WAR: A DOCUMENTARY HISTORY, 814 (1972)); Case against Masatoshi Sawamura (U.S. Military Commission, Yokohama, 14 29 April, 1947) (Sawamura was convicted of violations of the laws and customs of war for, inter alia, water torture of American prisoners of war, and was sentenced to 30 years hard labor); United States of America v. Hideji Nakamura, Yukio Asano, Seitara Hata, and Takeo Kita (U.S. Military Commission, Yokohama, 1 28 May, 1947. NARA Records, NND 735027 RG 153, Entry 143 Box 1025); Evan Wallach, Drop by Drop: Forgetting the History of Water Torture in U.S. Courts, 45 COLUM. J. TRANSNAT’’L L. 468 (2007).
1

3

METHODOLOGY This study examines the narrative voice of the nation’’s leading newspapers concerning the practice of waterboarding. Using electronic databases, our research team word searched for the term ““waterboarding”” and its historical synonyms, then read the retrieved articles and classified their reporting of the practice into several categories. These included classifying the practice as ““torture,”” giving it some lesser, negative classification (such as calling waterboarding ““inhuman””), giving it a softer, less negative classification (such as calling waterboarding ““objectionable””), or not characterizing the practice at all. We define waterboarding to be the practice of intentionally inducing the sensation of drowning in the victim. This sensation has been achieved in a number of ways, including but not limited to: (1) placing a cloth or plastic wrap over the face of the victim and pouring water over the cloth or plastic wrap, (2) pouring water directly into the mouth and nose of the victim, (3) placing a stick between the victim’’s teeth and pouring water into his or her mouth, often until the victim’’s stomach becomes distended, then forcing the water back out of the victim’’s mouth, and (4) dunking and holding the victim’’s head under water. We examined coverage of waterboarding in the four U.S. newspapers with the highest daily circulation: USA Today, The Wall Street Journal (WSJ), The New York Times (NY Times), and The Los Angeles Times (LA Times). 3 Using the online databases Proquest, LexisNexis, and the NY Times website archives, we searched the papers for specific terms referring to the practice. 4 The coders continuously added to our list of search terms as new synonyms of waterboarding were discovered. As search terms were added, they were applied to the years that had already been searched. All articles discovered using the new search terms were coded and added to the data set. For the NY Times, we used Proquest Historical Newspapers Database for the years 1851 1986. We used the archives at http://www.nytimes.com to search the years 1987 2008. For the LA Times, we used Proquest from 1881 1985 and

Numbers as reported by the Audit Bureau of Circulation as of March 31, 2007. Available at http://www.burrellesluce.com/top100/2008_Top_100List.pdf 4 These terms were: waterboarding, ““water boarding,”” ””water board,”” waterboard, ““water board,”” water boarding,”” ““water torture,”” submarino, ““simulated drowning,”” ““mock drowning,”” ““near drowning,”” ““feigned drowning,”” submersion head water, submersion water torture, ““water cure,”” ““water treatment,”” ““parrot’’s perch,”” ””torture lite,”” ““tortura del agua,”” ““tormento de toca,”” ““punishment of the pump,”” ““water detail,”” ““form of mock execution,”” ““Asian torture,”” ““Swedish drink,”” ““cold water dash,”” ““cold water process.”” ““Parrot’’s perch”” does not refer to waterboarding, but the authors found other mentions of waterboarding near the term, so searched for it to identify other references to waterboarding.
3

4

LexisNexis from 1986 2008. For the WSJ, we used Proquest from 1881 1991 and LexisNexis from 1992 2008. For USA Today, we used LexisNexis from 1989 2008. Using our search terms, we returned a total of 14,589 results. 5 Articles containing terms associated with waterboarding but not addressing the actual interrogation practice (e.g. the actions of municipal Water Boards) were eliminated. In addition, articles that mention the practice only tangentially or metaphorically (e.g., quoting a stock broker saying that the economy was like Chinese water torture 6 ) were not included in the data set. Finally, to isolate the narrative voice of the paper itself, book reviews, theater reviews, film reviews, and letters to the editor were excluded. Of the 14,589 total returns, 668 articles met our specifications and were coded. This includes 175 coded from the LA Times, 354 from the NY Times, 36 from USA Today, and 103 from the WSJ. Articles were coded along seven possible categories: ““torture,”” ““implying it’’s torture,”” ““others calling it torture,”” ““negative treatment,”” ““softer treatment,”” ““no treatment,”” and ““miscellaneous.”” The categories were defined as follows: 7 ““torture”” –– the narrative voice of the article itself explicitly and directly equates waterboarding with torture. For example, ““six former inmates testified that they were tortured in the rural East Texas jail from 1976 to 1980 by having towels draped over their faces and water poured over them.”” 8 ““implying it’’s torture”” –– the article does not explicitly call the practice torture but strongly and directly implies that it is. This category applied to situations in which the practice is grouped with other practices that are called torture, but waterboarding itself is not explicitly called torture. For example, ““The interrogation techniques themselves have been repeatedly discussed, and administration officials have been forced to explain why waterboarding, a simulated drowning technique of torturers dating back to the Spanish Inquisition, was not torture when used by the C.I.A.”” 9 ““others calling it torture”” –– the article refers to or quotes someone else calling the practice torture. For example, ““critics suspect the tapes
The exact breakdown of search term results is broken down by paper and search term in the Appendix B. 6 Vartanig Vartan, Dow Drops by 2.86; Off by 21.12 for Week; Cities Service Stock Soars, N.Y. TIMES, June 19, 1982, at section 2, page 35 7 The full coding instructions are listed in Appendix A. 8 Ex Sheriff s Deputy Denies Inmate Tortures, N.Y. TIMES, Sept. 8, 1983. 9 Scott Shane, An Elusive Starting Point on Harsh Interrogation, N.Y. TIMES, June 11, 2008
5

5

contained evidence of waterboarding, which international human rights groups and others have denounced as torture.”” 10 ““negative treatment”” –– the technique is described by words with a necessarily negative moral or value judgment attached, such as, ““inhuman”” or ““abusive.”” For example, ““‘‘high value detainees’’ were subjected to ever more barbaric acts, including simulated drowning.”” 11 ““softer treatment”” –– the technique is described using words without any necessarily negative moral or value judgment attached such as descriptions like ““harsh”” or ““controversial.”” For example, ““opinions that allowed the CIA to use aggressive interrogation methods, which included waterboarding.”” 12 ““No treatment”” –– either the procedure is simply dryly explained or there is no elaboration at all. This column was not used if any of the above categories were used. For example, ““some of whom were subjected to waterboarding, an interrogation technique that simulates drowning.”” 13 ““Misc.”” –– a catch all category for every other situation. For example, positive coverage such as ““Capt. Lee Hall …… The water cure, he thought, was no worse in its effect than the native vino.”” 14 If an article fit more than one category we coded both. The exception to this rule was for overlap between the categories ““torture,”” ““implying it’’s torture,”” ““negative treatment,”” and ““softer treatment””; in such cases, only the most severe treatment was coded (““torture”” > ““implying it’’s torture”” > ““negative treatment”” > ““softer treatment””). 15 However, in unusual cases where this hierarchy might exclude valuable information, this was noted in the ““misc.”” column. Notably, the category ““others calling it torture”” is not included in this exception. Thus, overlap with other categories is possible and some articles are in more than one category. ““Others calling it torture”” was left out because while quoting others is an editorial choice of the paper, it is not directly the narrative voice. We therefore allowed overlap to capture both the editorial choices and the narrative voice of the newspaper. Moreover, preliminary examinations

Josh Meyer, Judge reluctant to probe CIA s destruction of tapes, L.A. TIMES, Dec. 22, 2007. Editorial, Looking at America, N.Y. TIMES, Dec. 31, 2007. 12 Evan Perez, Gonzalez Defends Role In Antiterror Policies, WALL ST. J., Dec. 31, 2008. 13 Kevin Johnson, FBI agents objected to interrogation tactics, U.S.A. TODAY, May 5, 2008. 14 ““Water Cure”” and Wine, N.Y. TIMES, May 16, 1902. 15 For example, Scott Shane, An Elusive Starting Point on Harsh Interrogation, N.Y. TIMES, June 11, 2008.
10 11

6

suggested that quoting others was often paired with softer treatment in articles; coding both allowed exploration of this potential overlap. The consistency of coding across researchers was checked by independent verification at two points. The first verification was done by an independent researcher reviewing a random sample of 30 articles, 10 from each of the three coders, and working from the contemporary written coding instructions. For each possible category of treatment, the verifier matched the original coder, at worst, 80% of the time, and generally with accuracies of close to 90%. Furthermore, the coding was most inaccurate in determining whether the article was quoting critics calling the practice torture or others who were not critics calling it torture. Subsequently, and because of the evident difficulty of making such a distinction, these two categories were collapsed into a single category, ““others calling it torture.”” Once this ““error”” was eliminated by merging the terms, the rate of coder reliability increased to, at worst, 90%. The second verification occurred after the coding of all four papers had been completed. Again, an independent researcher who had not coded before, working only from the coding instructions, was asked to code 35 articles——five randomly selected articles from each coder. Again, from each possible category of treatment, the verifier coded the articles the same as the original coder, at worst, 83% of the time. Indeed, the consistency of coding was usually greater than 90%. RESULTS Treatment over Time NY Times News From its first mention of waterboarding in 1901 until 1925, the NY Times rarely described waterboarding as torture, calling it torture or implying the practice was torture in only 11.9% of articles (10 of 84). Most often, waterboarding was not given any treatment (61.9% of articles had no treatment, or 52 of 84). This pattern of treatment changed with the next mention of waterboarding, in 1931, and remained generally consistent until another dramatic shift, in 2004. FIGURE 1: NY Times, Calling/Implying Torture Figure 3, below, in '31-'99 and in '02-'08 illustrates this trend over 100% time. From 1931 to 1999, 80% NY Times journalists Torture/Implied 60% called waterboarding Torture 40% torture or implied that it 20% was torture in 81.5% (44
0% 1931-1999 2002-2008

7

of 54) of the articles. By contrast, from 2002 2008, waterboarding was called torture or implied to be torture in just 2 of 143 articles (1.4%). Notably, of these two articles, one was about waterboarding in Chile and made no mention of the U.S. The decrease in the use of the word torture corresponds to an increase in the use of no treatment FIGURE 2: NY Times, '31-'99 treatment and softer treatment. contrasted with '02-'08 The use of softer 90% 80% treatment increased 70% Torture/Implied from 0% (0 of 54) Torture 60% between 1931 and 2002 Softer 50% Treatment 40% to 45.5% (65 of 143) No Treatment 30% between 2002 and 2008. 20% No treatment use 10% 0% increased from 9.3% of 1931-1999 2002-2008 articles (5 of 54) from 1931 to 1999 to 28.7% (41 of 143) in 2002 2008.

Figure 3: NY Times, News Treatment by Decade 1931-2008
100% 80% 60% 40% 20% 0% 1931-1940 1941-1950 1951-1960 1961-1970 1971-1980 1981-1990 1991-2000 2001-2008 Torture/Implied Torture Softer Treatment No Treatment

Within the period from 2002 to 2008, the paper’’s treatment was generally consistent, without significant trends. NY Times Opinion Opinion pieces were more likely than news pieces to call waterboarding torture during all time periods. Though there were few opinion pieces before 2002, 50% of these articles (7 of 14) said or implied that waterboarding was torture. After 2002, this percent decreased slightly to 49.2% (29 of 59), with 27.1% (16 of 59) of articles giving no treatment and 10.2% (6 of 59) giving negative treatment. Thus as the non opinion pieces increasingly used softer treatment, the opinion pieces continued their use of the word torture.

8

LA Times News The LA Times began reporting the ““water cure,”” as it was known during the Philippine Insurgency, as torture. From 1901 to early May 1902, it was called torture or implied to be torture in 63.6% of articles (7 of 11). However, from the end of May 1902 through 1917, the paper called the practice torture or implied it was torture in only 1 of 32 articles (3.1%). Instead, the paper generally gave the practice no treatment (in 25 of 32 articles, or 78.1% of articles). After 1917, the paper did not mention the practice again until 1935. Figure 4 demonstrates the pattern of treatment after 1935. From 1935 2001, the LA Times called waterboarding torture or implied it was torture in 96.3% of articles (26 of 27). The paper then did not mention waterboarding again until 2006. From 2006 2008, the newspaper called waterboarding torture or implied it was torture in only 4.8% of articles (3 of 63), instead using softer treatment in 58.7% of articles (27 of 63) and giving no treatment in a further 23.8% (15 of 63).
Figure 4: LA Times, News Treatment Over Time
100% 80% 60% 40% 20% 0% 1935-1950 1951-1965 1966-1980 1981-1995 1996-2002 2004-2008 Torture/Im plying Torture Softer Treatment No Treatment

In the period from 2006 to 2008 the paper’’s treatment was generally consistent, without significant trends. LA Times Opinion Only one opinion piece in the LA Times addressed waterboarding before 2003, so it is impossible to establish if there was a change in treatment over time. However, post 2003 opinion articles were more likely than news articles to call the practice torture and, in general, reflect the same pattern found in the NY Times opinion pages. Between 2003 and 2008, the paper’’s opinion pieces called waterboarding torture or implied it was torture in 46.3% of pieces (19 of 41). The remaining opinion pieces applied other categories of treatment: 19.5% of pieces (8 of 41) gave the practice no treatment, 14.6% (6 of 41) gave waterboarding softer treatment, and 7.3% (3 of 41) gave the practice negative treatment. The majority of opinion articles occurred after 2006 (31 of 41) and there was a noticeable shift in treatment between 2007 and 2008. In 2007, only 4 of 15

9

opinion pieces (26.7%) called waterboarding torture or implied the practice was torture. By contrast, in 2008, 10 of 16 (62.5%) opinion pieces did so. WSJ and USA Today The WSJ and USA Today do not have as long a history of reporting on waterboarding as either the LA Times or the NY Times, making it impossible to compare their coverage of the current debate against past practice. USA Today first mentions FIGURE 5: USA Today, News waterboarding in 2004. Out of 18 Treatment, 2006-2008 total news pieces addressing 100% 80% waterboarding after 2004, none 60% 2006called waterboarding torture or 40% 2008 20% implied it was torture. Twelve of 0% the 18 articles (66.7%) gave the Torture Softer No Treatment Treatment practice softer treatment and a further 3 (16.7%) gave waterboarding no treatment. As with the other papers, the USA Today opinion pages are much more likely to call the practice torture. From the first opinion piece on waterboarding in 2006 through 2008, 55.6% of opinion articles (10 of 18) said or implied the practice was torture. Most of the rest of the opinion pieces gave the practice no treatment (27.8% or 5 of 18 pieces). The WSJ published only two news articles that considered waterboarding before 2005: one called FIGURE 6: WSJ, News Treatment, 2006waterboarding torture and one 2008 gave it negative treatment. 60% 50% From 2005 to 2008, only 1 of 63 40% 200630% 2008 articles (1.6%) called 20% 10% waterboarding torture. Notably, 0% Torture Softer No Treatment this one article addressed Treatment waterboarding in East Germany under the Communist regime and did not mention the U.S. In contrast, 55.6% of news articles (35 of 63) gave waterboarding no treatment and 12.7% (8 of 63) gave softer treatment. Within this time period, reporting by the WSJ shifted. In 2007, 85% of articles (17 of 20) gave the practice no treatment. By contrast, in 2008, only 40% of articles (16 of 40) gave no treatment. Instead, the paper quoted others calling the practice torture with increased frequency (18 of 40 articles, or 45%). In contrast to the other papers, the WSJ opinion pieces were as unlikely as their news articles to call waterboarding torture. Between 2005 and 2008, only 1
10

of 38 opinion articles (2.6%) called the practice torture or implied it was torture. By contrast, 52.6% of pieces (20 of 38) gave no treatment and 28.9% (11 of 38) gave the practice softer treatment. Country Responsible for Waterboarding News articles that considered other countries or individuals committing waterboarding were far more likely to classify waterboarding as torture than articles that dealt with the U.S. using waterboarding. In the NY Times, 85.8% of articles (28 of 33) that dealt with a country other than the U.S. using FIGURE 7: NY Times News, Who Was waterboarding against an Waterboarding individual called waterboarding 80% torture or implied it was torture. Torture/ 60% Implying 40% Yet when the U.S. was the Torture 20% perpetrator, only 7.69% (16 of 0% 208) articles said or implied that US Non-US waterboarding was torture. Just 0.8% of the articles (1 of 133) dealing with the War on Terror where the U.S. was the perpetrator said or implied that waterboarding was torture. The LA Times follows a similar pattern of avoiding the label of torture when the U.S. is responsible for FIGURE 8: LA Times News, Who Was Waterboarding using waterboarding. In articles 100% that considered other countries 80% Torture/ using waterboarding, 91.3% of 60% Implying 40% articles (21 of 23) called Torture 20% waterboarding torture or 0% implied the practice was US Non-US torture. When the U.S. was the violator, only 11.4% of articles (9 of 79) used this classification. The WSJ printed just four articles that classified waterboarding as torture; however, three of these articles addressed countries other than the U.S. using waterboarding. The fourth article discussed waterboarding in general, without reference to a specific incidence or specific parties involved. The analysis does not apply in the case of USA Today because all of its articles referred to instances where the U.S. was the perpetrator. None of these articles said or implied that waterboarding was torture.

11

Others Calling It Torture Balance with Softer Treatment All four papers frequently balanced their use of softer treatment by quoting others calling waterboarding torture. Except for a brief spate of articles in 1902 1903 in the NY Times which quoted mostly military officials and senators, almost all of the articles that quote others calling it torture appeared in 2007 and 2008. The NY Times used softer treatment in 66 of 281 total news articles. Of those 66 articles, 30 (45.5%) also quoted others calling the practice torture. In the LA Times, softer treatment was used in 29 of 134 total news articles. Softer treatment overlapped with others calling it torture in 41.4% of the articles (12 of 29). Similarly, in the WSJ, 8 of 66 total news articles used softer treatment. Of those 8, 3 (37.5%) also quoted others calling waterboarding torture. Finally, in USA Today, softer treatment is used in 12 of 18 total news articles. Of those 12 articles, 4 (33.3%) also quoted others calling waterboarding torture. Who Is Being Quoted When quoting others who call waterboarding torture, there is a shift in who the LA Times and the NY Times quoted over time. Before 2007, the NY Times had only scattered articles quoting others. However, beginning in 2007, there is a marked increase in articles quoting others, primarily human rights groups and lawmakers. Human rights representatives predominate during the first half of the year. However, beginning in October, politicians were cited more frequently labeling waterboarding torture. Senator John McCain is the most common source, but other lawmakers also begin to be cited. By 2008, the articles’’ references are more general such as ““by many,”” or ““many legal authorities.”” Stronger phrases such as ““most of the civilized world”” also begin to appear. The LA Times follows a similar pattern. In 2007, this paper mostly quoted human rights groups and Sen. McCain. Beginning in 2008, however, more general references began to be used, such as ““by many”” and ““critics.”” Softer Words Used Each paper had its own words of choice when giving waterboarding softer treatment. These words were consistently used within each paper, though

12

they varied between papers. Before 2004, softer treatment was virtually never used, and so this discussion is limited to 2004 2008. 16 The NY Times overwhelmingly favored the word ““harsh,”” using it in 53 of 65 instances (81.5%) of softer treatment between 2004 and 2008. The LA Times used ““harsh”” (11 of 27, or 40.7%) and ““coercive”” (12 of 27, or 44.4%). USA Today favored ““controversial,”” using it in 50% of articles (6 of 12) giving softer treatment. The WSJ used ““harsh”” (4 of 8, of 50%) and ““aggressive”” (3 of 8, or 37.5%). Opinion pieces in the four papers generally used softer treatment only rarely, making patterns difficult to establish. The WSJ is the exception to this, using softer treatment in 11 pieces. Here, however, there was no favored term: seven different words or phrases were used. Op ed Articles and Editorials Editorials generally treat waterboarding differently than op eds, though the papers are not consistent in this variation. The contrast is revealing because editorial pieces provide the most direct evidence of the views of a paper while the choice of op ed pieces determines the shape of the debate the paper allows on its pages. In the NY Times, as FIGURE 9: NY Times, '05-'08, Editorials v. Figure 9 demonstrates, Op-eds editorial pieces were more 60% 50% likely to call Editorial 40% waterboarding torture or 30% Op-ed to give it negative 20% 10% treatment and less likely 0% to give the practice no Torture/Implying Negative No Treatment Treatment treatment than were op ed articles. From 2005 to 2008, editorial articles called waterboarding torture in 55% of the articles (11 of 20). A further 30% (6 of 20) gave the practice negative treatment. Only 1 of 20 editorials (5%) gave the practice no treatment. In contrast, op ed pieces gave the practice no treatment in 38.5% of cases (15 of 39). No op eds used negative treatment. They said or implied that waterboarding was torture in 46.2% of articles (18 of 39).

16

For the NY Times, 98.45% of the articles describing waterboarding using softer treatment are from the time period 2001 2008. For the LA Times, 93% of the articles that used softer treatment came from the time period 2001 2008.

13

The LA Times showed the reverse relationship between op ed articles and editorials, shown in Figure FIGURE 10: LA Times, '03-'08, Editorials v. 10. In the period 2003 2008, Op-eds 70% editorials said or implied 60% that waterboarding was 50% Editorial 40% torture in 37% of articles Op-ed 30% (10 of 27) and gave no 20% treatment in 22.2% of 10% 0% articles (6 of 27). By Torture/Implying Negative Softer No Treatment contrast, 64.3% of op eds (9 Treatment Treatment of 14) said or implied that waterboarding was torture. The remaining op eds were split among others calling it torture, softer treatment, and no treatment. In the WSJ, in op eds between 2005 and 2008, waterboarding was generally given no treatment (12 of 16 op eds, or 75%). By contrast, editorials from the same time period gave waterboarding no treatment in only 38.1% of articles (8 of 21) and gave softer treatment in 42.9% of articles (9 of 21). In USA Today, with a limited sample size, there was less variation between the treatment offered by the op ed articles and the editorials. Of the editorials, 57.1% (4 of 7) said or implied that waterboarding was torture, while 54.5% of op eds (6 of 11) did so. The rest of the editorials offered no treatment (3 of 7, or 42.9%), while equal numbers of op ed articles used softer treatment and no treatment (2 of 11, or 18.2%) to describe waterboarding. Thus, while there was often a sustained difference in treatment between editorials and op ed articles within the opinion sections of each newspaper, these differences varied from paper to paper. CONCLUSION The results of this study demonstrate that there was a sudden, significant, shift in major print media’’s treatment of waterboarding at the beginning of the 21st century. The media’’s modern coverage of waterboarding did not begin in earnest until 2004, when the first stories about abuses at Abu Ghraib were released. After this point, articles most often used words such as ““harsh”” or ““coercive”” to describe waterboarding or simply gave the practice no treatment,

14

rather than labeling it torture as they had done for the previous seven decades. 17 There is also a significant discrepancy between the point of view offered by news articles and opinion pieces published in these papers. Opinion pieces were much more likely to characterize waterboarding as torture, suggesting that the private opinion of the editors and contributors did not align with the formal face the papers were presenting in their objective reporting. Yet what caused this change in waterboarding’’s treatment over time? Our data does not give any specific reason for this shift, but merely points to the existence of this change in syntax. A piece published by the public editor of The NY Times, Clark Hoyt, 18 suggests that these choices were made deliberately by journalists and their editors, perhaps in an effort to remain neutral in the debate going on in the U.S. If the classification of waterboarding as torture is unclear, Hoyt suggests, then it is irresponsible for journalists to preempt this debate by labeling it as such. The willingness of the newspapers to call the practice torture prior to 2004 seems to refute this claim. According to the data, for almost a century before 2004 there was consensus within the print media that waterboarding was torture. Yet once reports of the use of waterboarding by the CIA and other abuses by the U.S. surfaced, this consensus no longer held, despite the fact that the editors themselves seem to have still been convinced that waterboarding was torture, often labeling it as such in their editorials. The classification of waterboarding is not unclear; the current debate cannot be so divorced from its historical roots. The status quo ante was that

Given the sheer amount of coverage the practice received during this time period, it is possible that the prevalence of no treatment resulted from an assumption that readers would already be informed about the practice. However, on this assumption, the number of articles which give no treatment should increase as time passes. This is not the case. For example, the LA Times gave no treatment in 33.3% of news articles (7 of 21) in 2007 but gave no treatment in 19.5% of articles (8 of 41) in 2008. The NY Times used no treatment in 50% of news articles (5 of 10) in 2005, in 57.1% of articles (8 of 14) in 2006, in 24.5% (13 of 53) in 2007 and in 24.2% (8 of 41) in 2008. Similarly, in the WSJ, no treatment is used in 85% of news articles (17 of 20) in 2007, but only in 40% of articles (16 of 40) in 2008. For these papers, then, instead of increasing as expected, the number of articles with no treatment actually decreased over time. 18 Clark Hoyt, Telling the Brutal Truth, N.Y. TIMES, April 25, 2009, at WK12.
17

15

waterboarding is torture, in American law, 19 international law, 20 and in the newspapers’’ own words. Had the papers not changed their coverage, it would still have been called torture. By straying from that established norm, the newspapers imply disagreement with it, despite their claims to the contrary. In the context of their decades long practice, the newspaper’’s sudden equivocation on waterboarding can hardly be termed neutral.

For example, Court Martial of Major Edwin F. Glenn, Samar, P.I., April 1902 (reprinted in Leon Friedman, THE LAW OF WAR: A DOCUMENTARY HISTORY, 814 (1972)); Case against Masatoshi Sawamura (U.S. Military Commission, Yokohama, 14 29 April, 1947) (Sawamura was convicted of violations of the laws and customs of war for, inter alia, water torture of American prisoners of war, and was sentenced to 30 years hard labor); United States of America v. Hideji Nakamura, Yukio Asano, Seitara Hata, and Takeo Kita (U.S. Military Commission, Yokohama, 1 28 May, 1947. NARA Records, NND 735027 RG 153, Entry 143 Box 1025); Evan Wallach, Drop by Drop: Forgetting the History of Water Torture in U.S. Courts, 45 COLUM. J. TRANSNAT’’L L. 468 (2007). 20 See Convention Against Torture and Other Cruel, Inhuman, or Degrading Treatment or Punishment art. 1, Dec. 10, 1984, S. TREATY DOC. NO. 100 20 (1988), 1465 U.N.T.S. 85; see also International Military Tribunal for the Far East (IMTFE) Record available at National Archives and Diamond Library, Columbia Law School, Treasure, reproduced in facsimile in The Tokyo War Crimes Trial (R. John Pritchard & Sonia Magbanna Zaide eds., Garland Publishing Inc., 1981); Robert D. Sloane, The Cost of Conflation: Preserving the Dualism of Jus Ad Bellum and Jus In Bello in the Contemporary Law of War, 34 YALE J. INT’’L L. 47 (2009); Joby Warrick, Peter Finn & Julie Tate, Red Cross Described ““Torture”” at CIA Jails: Secret Report Implies that U.S. Violated International Law, WASH. POST, Mar. 16, 2009, available at http://www.washingtonpost.com/wp dyn/content/article/2009/03/15/AR2009031502724.html?nav=hcmodule.

19

16

Appendix A Coding Instructions ““Search Term,”” ““Title,”” –– Use the title that appears in the article, not the one that appears in the search results if they’’re different. ““Author,”” –– Please put ““editorial”” if it’’s an editorial and leave a blank if there is no author listed. ““Date,”” –– all in numbers, no month names. ““# of words,”” –– find in abstract in Proquest, on search page in other databases. ““Front Page?,”” –– ““Yes”” or blank ““Link,”” and ““Opinion piece?”” –– ““Yes”” or ““no.”” ““General frame of the article”” –– describe the main thrust of the article as a whole. Is it about Attorney General Mukasey’’s confirmation or investigations into Abu Ghraib? Is it reporting on a POW’’s experiences or a local investigation into a jail? ““Depth of Treatment”” –– describe how much attention waterboarding gets in the article. For this column, please describe the proportion of the article that is devoted to waterboarding, in the form of 1%, 10%, 50%, etc. (If less than 1%, use <1%). If the practice is being included with other practices and discussed, as under the rubric of ““harsh”” methods, include that in the proportion. As long as the article is dealing with the practice, add that to the depth of treatment. The next grouping of columns describes how the term is treated in the article. Enter either ““yes”” or leave it blank unless otherwise specified. The coding can be tricky when waterboarding is grouped with other terms; to count for the following terms, as well as for depth of treatment above, the voice must be talking specifically about waterboarding (or one of its synonyms), OR about interrogation techniques in general IF the article has already elaborated on those techniques and included waterboarding in it. Make sure when the article has included waterboarding in its use of ““techniques””/ ““methods,”” if it refers to techniques later, it’’s talking about the same ones. Pay attention to context; be strict. The possibilities we have are:

17

““Torture”” –– the narrative voice of the article itself explicitly and directly equating it with torture. ““Implying it’’s torture”” –– where the article doesn’’t explicitly say the practice is torture but strongly and directly implies that it is. This will apply to situations where the practice is grouped with others that are torture but the practice itself isn’’t explicitly called torture. ““Others calling it torture”” –– where the article references or quotes someone else calling the practice torture. In this column, just write their name, title, organization, (party) affiliation, whatever is relevant to their stance. ““Negative treatment”” –– when the technique is described by words with a necessarily negative moral/value judgment attached. For example, ““inhuman”” or ““abusive””; check the term sorting below for more examples. Be sure to list exactly what words are used under the column ““negative words used.”” ““Softer treatment”” –– when the technique is described by words without any necessary moral/value judgment attached. For example, softer treatment includes descriptions like ““harsh”” or ““controversial””; check the term sorting below for more examples. Be sure to list exactly what words are used under the column ““softer words used”” ““No treatment”” –– this is when either the procedure is just dryly explained or when there is no elaboration at all; do not use this column if any of the above categories have been used. ““Misc.”” –– a catch all category for every other situation. If you use this category, describe exactly what’’s going on in the article. It is very important that these categories be coded with exactly the same words. If an article gives the practice negative treatment, make sure you put ““yes”” under ““negative treatment”” and then list the specific words under ““negative words used.”” Don’’t do ““yes –– called it barbaric.”” The same holds true for the other columns; consistency is key. If an article fits more than one situation –– has ““others calling it torture”” and also gives it negative treatment, code both. The exception to this is if there is any overlap between the ““torture,”” ““implying it’’s torture,”” ““negative treatment”” and ““softer treatment”” columns; these are in a hierarchy that goes ““torture”” > ““implying it’’s torture”” > ““negative treatment”” > ““softer treatment,”” where you don’’t need to fill in something if anything to its left has been filled in (i.e. don’’t include ““softer treatment”” if they already called it torture). This does not apply to

18

any other columns. If there is an unusual case where you think this might be ignoring valuable information, make a note of it in the Misc. column. The next column asks ““Who was the violator/victim?”” List who performed and who received the waterboarding in that exact violator/victim format. There are several commonly used categories. For articles from the current debate, where the U.S. is waterboarding suspected terrorists or detainees or prisoners of one kind or the other, use ““U.S./prisoners.”” For articles relating to waterboarding Filipinos during the Philippine insurgency at the turn of the century, use the term ““U.S./Philippines.”” For all other instances, if the violator or victim are acting as a representative of their state (e.g. soldier), only put their state (e.g. Japan/China). This is also the case if they’’re citizens in a time of war being targeted as a member of their state. If they are acting as an individual, then be specific (e.g. robbers/deaf mute man) unless they are prisoners, in which case just use ““prisoners”” (e.g. Massachusetts/prisoners; Japan/prisoners). When the article is just talking about waterboarding without reference to a specific instance of waterboarding, use ““general.”” The final two columns are ““Notes”” and ““Relevant Text.”” ““Notes”” is for any notes of interest about the article you may have –– indicate any unusual or uncommon things about the article that you noticed. In ““Relevant Text”” you should copy and paste in the relevant portions of the articles. Don’’t worry about how long they are –– put in everything that is relevant. A final point: do not code book reviews, theater reviews, movie reviews, letters to the editor, or articles that drop the terms as a one line metaphor completely unrelated to the actual discussion (e.g., ““the fall of the stock market was like Chinese water torture.””). Term Sorting: negative: cruel causing or marked by great pain or distress brutal savage; cruel; inhuman; harsh; ferocious inhuman lacking qualities of sympathy, pity, warmth, compassion, or the like; cruel; brutal atrocious extremely or shockingly wicked, cruel, or brutal; dreadful; abominable tormenting to afflict with great bodily or mental suffering; pain

19

degrading to lower in dignity or estimation; bring into contempt; debase; humiliate grisly causing a shudder or feeling of horror; horrible; gruesome barbaric without civilizing influences; uncivilized; primitive; of, like, or befitting barbarians odious deserving or causing hatred; hateful; detestable humiliating lowering the pride, self respect, or dignity of a person; mortifying abusive treating badly or injuriously; mistreating softer: harsh ungentle and unpleasant in action or effect ; grim or unpleasantly severe; stern; cruel; austere objectionable causing or tending to cause an objection, disapproval, or protest; offending good taste, manners, etiquette, propriety, etc.; offensive aggressive characterized by or tending toward unprovoked offensives, attacks, invasions, or the like; militantly forward or menacing coercive use of force or intimidation to obtain compliance improper not in accordance with propriety of behavior, manners severe harsh; unnecessarily extreme; grave; critical; causing discomfort or distress by extreme character or conditions, as weather, cold, or heat; unpleasantly violent, as rain or wind, or a blow or shock; difficult to endure baffling confusing, bewildering, or perplexing controversial subject to controversy; debatable tough vigorous; severe; violent painful affected with, causing, or characterized by pain; laborious; exacting; difficult wrenching To pull at the feelings or emotions of; distress

20

Appendix B Search Term Hits Search term Waterboarding ““water boarding”” ““water board”” Waterboard ““water board”” ““water boarding”” ““water torture”” Submarino ““simulated drowning”” ““mock drowning”” ““near drowning”” Submersion head water Submersion water torture ““form of mock execution”” ““water cure”” ““water treatment”” ““parrot’’s perch”” ““feigned drowning”” ““torture lite”” ““tortura del agua”” ““tormento de toca”” ““punishment of the pump”” ““water detail”” ““asian torture”” NY Times 266 38 1426 12 1426 38 237 116 56 2 189 44 4 0 470 1900 18 5 11 0 0 0 4 0 LA Times 100 63 2174 15 2088 63 231 174 47 8 299 102 104 0 336 1229 21 2 4 0 0 0 10 0
21

Wall Street Journal 102 5 238 6 238 5 91 24 19 0 13 1 1 0 16 1432 2 0 0 0 0 0 3 0

USA Today 47 8 27 3 27 8 48 1 8 0 14 100 100 0 3 100 0 0 0 2 0 0 0 0

““Swedish drink”” ““cold water dash”” ““cold water process”” Total

3 3 4 4808

1 6 12 7089

0 0 0 2196

0 0 0 496

22