Chapter One Discovering S-21 On 7 January 1979, a bright, breezy day in Cambodia's cool season, heavily armed Vietnamese forces, accompanied by lightly armed Cambodian allies, reached the outskirts of Phnom Penh after a blitzkrieg campaign that had begun on Christmas Day. For over a year, Vietnam had been at war with the Maoist-inspired regime of Democratic Kampuchea (DK), known in the West as the Khmer Rouge. Their invading force of over one hundred thousand troops, including armored units, was reinforced by a sustained aerial bombardment. The rapidity of the Vietnamese success took their commanders by surprise. After barely two weeks of fighting, Cambodia cracked open like an egg. The leaders of DK, most of their army, and tens of thousands of their followers fled or were herded out of the city. The invaders were welcomed by nearly everyone who stayed behind. These were people terrorized and exhausted by nearly four years of undernourishment, back-breaking labor, and widespread executions. A similar welcome, tragically misplaced, had greeted the Khmer Rouge themselves when they had occupied Phnom Penh in April 1975 and ordered its population into the countryside to become agricultural workers. In both cases, people were longing desperately for peace. By late afternoon the Vietnamese forces had occupied the city. Aside from a few hundred prisoners of war and other people--including some of the workers at S-21--who were in hiding, waiting to escape, Phnom Penh was empty. After the Khmer Rouge had emptied the city in 1975, Phnom Penh had remained the country's capital, but it never regained its status as an urban center. The bureaucrats, soldiers, and factory workers quartered there probably never numbered more than fifty thousand. During the DK era, the country had no stores, markets, schools, temples, or public facilities, except for a warehouse in the capital serving the diplomatic community. In Phnom Penh, barbed-wire fences enclosed factories, workshops, barracks, and government offices. Street signs were painted over, and barbed-wire entanglements blocked many streets to traffic. Banana trees were planted in vacant lots. Automobiles abandoned in 1975 were rusted in piles along with refrigerators, washing machines, television sets, and typewriters. Scraps of paper in the gutters included prerevolutionary currency, worthless under the Khmer Rouge. On 7 January 1979, no people or animals could be seen. As in 1975, the central government, such as it was, had disappeared. Once again, Cambodians were being made to start at zero. The effect of the desolation on the newcomers was phantasmagoric. Chey Saphon, for example, was a forty-seven-year-old Cambodian Communist who had fought against the French in the 1950s. He had lived in Vietnam since 1955 and had been trained as a journalist. On 7 January he was thrilled to be returning home with the Vietnamese troops. He was so unnerved by what he saw, however, that years later he recalled that he "spent the whole afternoon in tears." Over the next few days Vietnamese troops fanned out across Phnom Penh. On 8 January, in the southern sector of Tuol Svay Prey, two Vietnamese photojournalists who had accompanied the invasion were drawn toward a particular compound by the smell of decomposing bodies. The silent, malodorous site was surrounded by a corrugated tin fence topped with coils of barbed wire. Over the gate was a red placard inscribed in yellow with a Khmer slogan: "Fortify the spirit of the revolution! Be on your guard against the strategy and tactics of the enemy so as to defend the country, the people and the Party." The place carried no other identification. Pushing inside, the two photographers found themselves on the grounds of what appeared once to have been a high school. The spacious, dilapidated compound measured roughly four hundred meters from east to west and six hundred meters from north to south (see illustrations). It consisted of four whitewashed concrete buildings, each three stories high, with balcony corridors running along each upper story. A fifth, single-story wooden building, facing west, split the compound into two identical grassy spaces. To the rear of each of these, one of the taller buildings faced east, toward the entrance. Similar buildings marked off the northern and southern boundaries of the compound. The purpose of the compound was unclear to the two men, although the single-story building, littered with papers and office equipment, had obviously been used for some sort of administration. In rooms on the ground floor of the southernmost building, the two Vietnamese came across the corpses of several recently murdered men. Some of the bodies were chained to iron beds. The prisoners' throats had been cut. The blood on the floors was still wet. Altogether the bodies of fourteen people were discovered in the compound, apparently killed only a couple of days before. In large classrooms on the upper floors of the western buildings, the patrol found heaps of shackles, handcuffs, whips, and lengths of chain. Other rooms on the upper floors had been divided by clumsily bricked partitions into small cells where each prisoner's foot had been manacled, as William Shawcross later wrote, "to a shackle large enough to take a ship's anchor." Ammunition boxes in some of the cells contained human feces. On the third floor were slightly larger, more elaborately constructed cells with wooden walls and doors. The two intruders took photographs of all the rooms in the facility, adding photos of the corpses. They then "informed the Vietnamese authorities" of what they had found. That evening the corpses were burnt "as a sanitary measure." Some of the photographs taken at that time now hang in the rooms where the bodies were found. Over the next few days the Vietnamese and their Cambodian assistants discovered in nearby houses thousands of documents in Khmer, thousands of mug-shot photographs and undeveloped negatives, hundreds of cadre notebooks, and stacks of DK publications. In a workshop near the front gate they found several recently completed, oversized concrete busts of the DK prime minister, Pol Pot, a concrete mold for the statues, and some portraits of him, apparently painted from photographs. The Vietnamese had stumbled into a vicious and important Khmer Rouge facility. Documents found at the site soon revealed that it had been designated in the DK era by the code name S-21. The "S," it seemed, stood for sala, or "hall," while "21" was the code number assigned to santebal, a Khmer compound term that combined the words santisuk (security) and nokorbal (police). "S-21," and santebal, were names for DK's security police, or special branch. Over the next few weeks the history of the site was pieced together. In the early 1960s, when Cambodia had been ruled by Prince Norodom Sihanouk, it had been a high school. It was named after Ponhea Yat, a semilegendary Cambodian king associated with the foundation of Phnom Penh. After Sihanouk was overthrown in 1970--the event that sparked Cambodia's civil war--the school had taken the name of the surrounding district, Tuol Svay Prey (hillock of the wild mango). An adjoining primary school was called Tuol Sleng (hillock of the sleng tree). This name was used to designate the entire compound after it became the Museum of Genocidal Crimes in 1980, perhaps because the sleng tree bears poisonous fruit. The code name S-21 began to appear on Khmer Rouge documents in September 1975. For the next nine months, until the facility came into operation in May or June 1976, the security service's work was spread among several units in Phnom Penh, the southern suburb of Ta Khmau, and in Sector 25, north of the capital. By the end of 1975, according to a former guard, Kok Sros, interviewed in 1997, santebal coalesced under the command of Kang Keck Ieu (alias Duch), a former schoolteacher who had been in charge of security in the so-called special zone north of the capital during the civil war. Duch became the director of the Tuol Sleng facility in June 1976. He remained in command until the day the Vietnamese arrived. Sensing the historical importance and the propaganda value of their discovery, the Vietnamese closed off the site, cleaned it up, and began, with Cambodian help, to examine its voluminous archive. On 25 January 1979, a group of journalists from socialist countries was invited to Cambodia by the Vietnamese to report on and celebrate the installation of the new Cambodian government, known as the People's Republic of Kampuchea (PRK). The journalists were the first official visitors to see Tuol Sleng. Chey Saphon accompanied them to the site. One of the journalists, the Cuban Miguel Rivero, wrote later that "there were still traces of blood on the floor. The smell was even more penetrating. There were thousands of green flies circling the room." Rivero added that he saw documents "written in Sanskrit" and "several" copies of Mao Zedong's Little Red Book at the "Dantesque" site. Soon afterwards, in February or March 1979 (his own memory is uncertain), Mai Lam, a Vietnamese colonel who was fluent in Khmer and had extensive experience in legal studies and museology, arrived in Phnom Penh. He was given the task of organizing the documents found at S-21 into an archive and transforming the facility into what David Hawk has called "a museum of the Cambodian nightmare." The first aspect of Mai Lam's work was more urgent than the second. It was hoped that documents found at the prison could be introduced as evidence in the trials of Pol Pot and Ieng Sary, DK's minister of foreign affairs, on charges of genocide. These took place in Phnom Penh in August 1979. Although valuable information about S-21 was produced at the trials, none of the documents in the archive provided the smoking gun that the Vietnamese and PRK officials probably hoped to find. No document linking either Pol Pot or Ieng Sary directly with orders to eliminate people at S-21 has ever been discovered, although the lines of authority linking S-21 with the Party Center (mochhim pak) have been established beyond doubt. Because of his penchant for history, his experience with museums (he had organized the Museum of American War Crimes in Ho Chi Minh City), and the criminality of what had happened at S-21, Mai Lam approached his work with enthusiasm and pride. His genuine, somewhat patronizing fondness for Cambodia and its people, based on his experiences in Cambodia in the first Indochina war, also inspired him. "In order to understand the crimes of Pol Pot-Ieng Sary," he told interviewers in 1994, "first you should understand Cambodians, both the people and the country." In turning S-21 into a museum of genocide, Mai Lam wanted to arrange Cambodia's recent past to fit the requirements of the PRK and its Vietnamese mentors as well as the long-term needs, as he saw them, of the Cambodian people. Because numbers of the "Pol Pot-Ieng Sary genocidal clique," as the Vietnamese labeled them, had been Cambodians themselves, the message that Mai Lam was trying to deliver was different from the one that he had hoped to convey in the Museum of American War Crimes, but it was just as harsh. The history that he constructed in the exhibits at S-21 denied the leaders of the CPK any socialist credentials and encouraged viewers to make connections between the DK regime and Tuol Sleng on the one hand, and Nazi Germany and what Serge Thion has called the "sinister charisma" of Auschwitz on the other. The comparisons were fitting insofar as S-21, like the Nazi death camps, was a secret facility where all the inmates were condemned to death, but any more explicit links between Nazism and DK, although seductive, were inexact. A Cambodian survivor of S-21, Ung Pech, became the director of the museum when it opened in 1980. He held the position for several years and traveled with Mai Lam to France, the USSR, and Eastern Europe in the early 1980s to visit museums and exhibits memorializing the Holocaust. Although Mai Lam remained in Cambodia until 1988, working at Tuol Sleng much of the time, he concealed his "specialist-consultant" role from outsiders, creating the impression that the initiatives for the museum and its design had come from the Cambodian victims rather than from the Vietnamese--an impression that he was eager to correct in his interviews in the 1990s. Over the next few months, people working at the prison constructed a rough history of the facility, drawing on entry and execution records, memoranda by prison officials, and the memories of survivors. Between April 1975 and the first week of 1979, they discovered, at least fourteen thousand men, women, and children had been held by S-21. Because the entry records for several months of 1978 were incomplete, the true number of prisoners was undoubtedly higher. Of the documented prisoners, all but a dozen specially exempted ones, including Ung Pech, had been put to death. Since 1979, seven of these survivors have come forward. Their memories, corroborated by those of former workers at the prison, have been invaluable for this study. The records from S-21 also showed that most of the lower-ranking prisoners had been held for a few days or weeks, whereas more important ones, and lesser figures suspected of grave offenses, had been incarcerated for several months. Thousands of the prisoners, regardless of their importance, had undergone interrogation, prepared "answers" (chomlaoy) or confessions admitting counterrevolutionary crimes, and submitted lists of their associates, titled "strings" or "networks of traitors" (khsae kbot), that sometimes ran to several hundred names. The texts range from a single page to several hundred pages. Roughly 4,300 of them have so far come to light, including those of nearly all the important DK figures known to have been purged. Confession texts, survivors' memories, and the grisly instruments discovered at the site made it clear that torture was widely inflicted at S-21. Tortured or threatened with torture, few prisoners maintained their innocence for long. Considered guilty from the moment they arrived--the traditional Cambodian phrase for prisoner, neak thos, translates literally as "guilty person"--thousands of these men and women were expected to confess their guilt in writing before they were taken off to be killed. This bizarre procedure drew some of its inspiration from the notion of revolutionary justice enshrined in the Reign of Terror in eighteenth-century France and enacted in the Moscow show trials in the 1930s and also from the land reform and "reeducation" campaigns in China in the 1940s and in Vietnam a decade later. In spite or perhaps because of these manifold influences, no precise or overriding foreign model for S-21 can be identified. Moreover, the severity of practices at S-21 and the literalness with which interrogators went about their business also reflected prerevolutionary Cambodian punitive traditions, by which prisoners were never considered innocent and crimes of lèse-majesté were mercilessly punished. Although DK's economic and social policies do not fit into a fascist framework, the resemblances between S-21 and Nazi death camps are striking. Works discussing the Holocaust provide insights into the psychology of torturers, administrators, and victims at the prison, as do more recent works that deal with torturers in the "dirty war" in Argentina in the 1970s and early 1980s. The list of materials that I have found useful for comparative purposes could easily be extended. The most striking difference between the German and Cambodian cases lies in the extent of the documentation produced at S-21. Prisoners both under the Nazis and in DK were removed from any semblance of legal protection; but whereas those in the Nazi death camps were simply exploited for physical labor while awaiting execution, those in S-21 were treated almost as if they were subject to a judicial system and their confessions were to provide evidence for a court of law. In this respect they resemble the alleged counterrevolutionaries who went on "trial" in the Soviet Union in large numbers in the 1930s. In Nazi Germany, political prisoners were kept in separate camps from those targeted for execution and were somewhat better treated. At S-21, all were charged with political offenses, and all were to be killed. Like the Nazi extermination camps and the Argentine torture facilities, S-21 was a secret facility, and the need for secrecy influenced much of what happened inside its walls. The prison's existence was known only to those who worked or were imprisoned there and to a handful of high-ranking cadres, known as the Party Center, who reviewed the documents emerging from S-21 and selected the individuals and the military and other units to be purged. Interrogators, clerks, photographers, guards, and cooks at the prison were forbidden to mingle with workers elsewhere, and the compound soon earned an eerie reputation. A factory worker in a nearby compound, interviewed in 1989, referred to S-21 as "the place where people went in but never came out." The factory workers were uncertain about what went on inside its walls but were ready to think the worst. Party leaders never referred to S-21 by name. In 1997, when questioned by the journalist Nate Thayer, Pol Pot denied any knowledge of "Tuol Sleng," hinting that the museum and its archive were Vietnamese concoctions. "I was at the top," he said: I made only big decisions on big issues. I want to tell you--Tuol Sleng was a Vietnamese exhibition. A journalist wrote that. People talk about Tuol Sleng, Tuol Sleng, Tuol Sleng.... When I first heard about Tuol Sleng it was on the Voice of America. I listened twice. Guided tours of S-21 were first organized in March 1979, but for over a year, as the museum took shape, only foreigners were admitted because, as a PRK Ministry of Culture, Information, and Propaganda document from 1980 asserted, the site was intended primarily "to show ... international guests the cruel torture committed by the traitors to the Khmer people." In the meantime, Mai Lam and his associates were slowly transforming the site into a museum. In July 1980 the ban on Cambodian visitors was lifted, and tens of thousands visited S-21, many of them seeking information about relatives who had disappeared. They consulted hundreds of enlarged mug shots of prisoners on view on the ground floor of the prison, which formed a major component of the museum display. As Judy Ledgerwood has written, many of the visitors were also "searching for meaning, for some explanation of what had happened. A visit would not have been an easy task; people who went through the museum in the first year said that the stench of the place was overwhelming." Some thirty-two thousand people visited the museum in the first week it was open to the public. By October 1980, over three hundred thousand Cambodians and eleven thousand foreigners had passed through the facility. Mai Lam always had high ambitions for S-21. He wanted to establish a museum and organize an archive that would be useful to the Cambodian people and would prevent them from forgetting what had happened under "the contemptible Pot" (a-Pot) . One of his more melodramatic exhibits was a large map of Cambodia, composed of skulls with the rivers shown in blood red. In the early 1980s, after S-21's killing field at Choeung Ek, west of the capital, had been excavated under his direction, he supervised the exhumation of thousands of bodies and ordered the construction of memorial stupa at the site, fronted with glass and filled with skulls. Talking to Sara Colm in 1995, Mai Lam said: For seven years I studied ... to build up the Museum ... for the Cambodian people to help them study the war and the many aspects of war crimes.... For the regular people who cannot understand, the museum can help them. Even though they suffered from the regime, as a researcher I want them to go [to the museum]. Even though it makes them cry.... The Cambodian people who suffered the war could not understand the war--and the new generation also cannot understand. For many Cambodians, as Ledgerwood points out, there are problems of "authenticity" in a museum established by foreigners to press home fortuitous parallels between the "genocidal regime" of DK and Hitler's Germany. At the same time, for the survivors the vast and seemingly random cruelties of the DK regime easily became encapsulated in the museum's displays. Nazism seemed as good a label as any other for the horrors that the survivors of the regime had undergone. The indifference of DK officials to their victims, exhibited in room after room, recurs in the memories of many survivors. Cambodians' interpretations of the Pol Pot era slip easily into Manichean frameworks that make poor history but are emotionally satisfying and consistent with much of what they remember. This point has been driven home by the French psychiatrists Jean-Pierre Hiegel and Colette Landrac, who worked in Khmer Rouge refugee camps in Thailand in the 1980s: It is always more comfortable to have a Manichean vision of the world, for that allows us not to ask too many questions or at least to have the answer readily at hand. In this fashion, representing the Khmer Rouge as an homogenous group of indoctrinated fanatics, the incarnation of absolute evil, responsible for all the unhappiness of the Khmer people, is a reductive vision of a complex phenomenon but one which a good many people find satisfying. Within just such a Manichean framework, the PRK regime worked hard to focus people's anger onto the "genocidal clique" that had governed Cambodia between April 1975 and January 1979. While the new government based its legitimacy on the fact that it had come to power by toppling the Khmer Rouge, it was in no position to condemn the entire movement, since so many prominent PRK figures had been Khmer Rouge themselves until they defected to Vietnam in 1977 and 1978. The continuing existence of DK's leaders and their armed followers on the Thai-Cambodian border, however, gave the Vietnamese a rationale for keeping their troops in the country and allowed the PRK to label its political opponents as Khmer Rouge. Like their predecessors in other Cambodian regimes, PRK spokesmen arranged history to suit their day-to-day requirements. In their formulations, the Cambodian Communist movement had been an authentic revolutionary one, up to and including the liberation of Phnom Penh in 1975, when the movement had suddenly and inexplicably spun out of control. This contorted narrative enabled the PRK to celebrate the socialist "triumph" of 17 April 1975 while condemning the people who brought it about. PRK historiography also stressed a long-standing official friendship between Khmer and Vietnamese movements and regimes that was hard to locate in the historical record. These tangled readings of the past made sense to Party faithful. After all, only an authentic revolutionary movement could have defeated the United States; and, once the wheel of history had turned, no such movement could have been so cruel to ordinary people or could have opposed the genuinely revolutionary Vietnamese. These ex post facto explanations, however, were of little interest to most nonrevolutionary Khmer. They found it easier to focus their memories on "the contemptible Pot," whose bizarre, unpardonable crime was not that he had been a Communist (or a "fascist") but that he had presided over the deaths of so many of his own people. If the Vietnamese wanted to call Pol Pot a fascist, people would go along with it, without knowing much about the subject. Thus, talking with Lionel Vairon in 1995, the S-21 survivor Pha Thachan, by then a general in the Cambodian army, stated: Yes, what happened under Pol Pot was "Communism," but it was of a "fascist" kind, and it surpassed fascism. In fascism the Germans never killed their own people, they only killed foreigners. They killed French and Poles and so on. Pol Pot on the other hand killed his own people, three million of them. The fascists never did this. In annual "days of hate," the government mobilized public opinion and refreshed people's memories of the DK period. On these occasions, anti-Pol Pot demonstrations were organized for school children, PRK officials made speeches condemning the DK, and Vann Nath and other survivors of S-21 were called on to recite their experiences at the prison. By the mid-1980s visitors to the archive, relocated to the second floor of one of the western buildings at Tuol Sleng, were impressed by the mass of documentation collected there. Many of the dossiers were over a foot thick. Hundreds were typed and duplicated. Glass-fronted cabinets in the archive were stuffed with cadre notebooks recording political meetings, military seminars, and sessions of paramedical training. Stapled "summaries" of confessions, stacked in piles, sometimes ran to hundreds of pages. Journalists and scholars were encouraged to photocopy confession texts and other materials. In the 1980s, the human rights activist David Hawk assembled a daunting collection of materials from Tuol Sleng that provided ample evidence of the extrajudicial crimes of the DK regime. His efforts and those of others to bring DK's leaders to justice then and later were stymied by Thai intransigence and political considerations. Thoughts about bringing the Khmer Rouge leaders to trial gathered steam again in the 1990s as the Khmer Rouge movement lost momentum. After 1989, when the Vietnamese withdrew their troops from Cambodia, the fate of the archive, and even of the museum, looked uncertain. In 1991 Cornell University, noted for the richness of its Southeast Asian library holdings, proposed to catalogue and microfilm the S-21 archive, which came under the Ministry of Culture's jurisdiction. The work was completed in two years. A full set of the microfilms was deposited in the Cornell University library; another set was retained by the museum. The microfilmed materials cover 210 reels of film, including eleven reels of retakes. The reels contain all the confession texts discovered at the site, including those by foreign prisoners (filmed on separate reels). Foreigners' confessions were primarily those of Vietnamese prisoners of war but also included statements from Thai and Vietnamese fishermen, some Vietnamese civilians, and a handful of American, British, and Australian sailors who were arrested when their boats strayed too near the Cambodian coast. Among the most revealing confessions in the archive are those of seventy-nine former workers at the prison. Twenty-four of these prisoners had been interrogators, and twelve had been document workers. Most of the others had been guards. These texts provide valuable biographical data about the young men working at the prison. They are also helpful in documenting work patterns at S-21, the style of interrogations, and the practice of torture (tearunikam) there. The microfilmed reels also reproduce a range of nonconfessional materials that were discovered at the prison. These include entry and execution records, typed summaries of confessions broken down by region, and military unit and government office and study notebooks that cover such diverse subjects as politics, aircraft identification, mathematics, medicine, artillery, and small arms. The nonconfessional materials also include copies of the CPK's statutes, speeches, and directives from the Party Center and copies of DK's theoretical journals, Revolutionary Flags (Tung Padevat) and Revolutionary Youth, distributed to Party members. Materials stemming from the prison itself include a report for the first three months of 1977, written notes from interrogators reporting the torture of prisoners, rules for guards, and notes by Duch, the prison director, on a range of issues, including his analysis of the confession texts in a handwritten document probably written in early 1978, titled "The Last Plan." Probably the most revealing nonconfessional text microfilmed by Cornell is a fifty-five-page study notebook compiled by an interrogator, prepared in 1976. This text is discussed in detail in chapter 5. When the microfilming at S-21 was completed in 1993, it was thought that the reels included all the significant surviving material from the prison. In 1995, however, another S-21 archive, held in the Cambodian Ministry of the Interior, was presented to the Documentation Center-Cambodia (DC-Cam), an affiliate of the Cambodia Genocide Program established by Yale University, with a grant from the U.S. Department of State, with a view to gathering documentation of the DK regime. The new material, held by DC-Cam in Phnom Penh, seems to be have been drawn from the archives of the DK minister of defense and national security, Son Sen, who also oversaw the operations of the prison. Over fifty confession texts in this collection contain annotations in Son Sen's writing. Many of the confession cover sheets also bear handwritten annotations in Vietnamese, suggesting that they had been reviewed under the Vietnamese protectorate of Cambodia in the 1980s. The newly discovered materials included dozens of confessions that had not survived in the S-21 archive, as well as valuable administrative materials, such as notes from self-critical study sessions held for cadres at the prison and notebooks compiled by senior interrogators. DK materials unrelated to S-21 are also housed in DC-Cam and will undoubtedly be of interest to scholars of the regime. Several hundred documents from S-21 itself that were not microfilmed have also found their way into the DC-Cam collections since 1996. These include miscellaneous, fragmentary interrogation schedules, lists of prisoners who were ill, documents transmitted with prisoners to S-21, and over two hundred additional study notebooks. Two that are of special interest were prepared in 1977 and 1978 by the chief interrogator, Mam Nay (alias Chan) and, in a shared notebook, by two senior interrogators known by their pseudonyms Tuy and Pon (henceforth the Tuy-Pon notebook). Discovering S-21, in other words, is a process that began in January 1979 and is still under way. The mass of material now available seems sufficient to support a detailed study of the prison. The Yugoslav writer Milovan Djilas has observed that "the way prisons are run and their inmates are treated gives a faithful picture of society, especially of the ideas and methods of those who dominate the society"--a remark that seems particularly appropriate to S-21 and Democratic Kampuchea. As we pore through the materials and listen to the voices of so many people living under extreme conditions, we may also learn something about ourselves. I first visited Tuol Sleng for less than an hour in August 1981. Since 1990, I have returned to the museum many times. In spite or perhaps because of the courtesy and friendliness of the staff, I am always disoriented by the place. On every visit, I've been struck by the contrast between the peaceful, sun-soaked compound and the horrific exhibits on display, between the whitewashed classrooms with their yellow and white tile floors and the instruments of torture they contain, between the children at play outside the buildings and the mug shots of other children en route to being killed. In the museum, the eyes of the mounted mug shots, and especially those of the women and children, seem to follow me. Knowing as we do, and as they did not, that every one of them was facing death when the photographs were taken gives the photos an unnerving quality that is more affecting, for me at least, than the photographs of dead prisoners or the grisly portrayals of torture painted after 1979 by the S-21 survivor, Vann Nath, that are also included in the display. On most of my visits mynah birds have hopped along the overgrown paths. Roosters have crowed around the neighborhood, the sound competing with the hum of traffic on Monivong Boulevard to the east or, in the dry season, with music broadcast over loudspeakers from Buddhist wedding celebrations nearby. The noises in the 1970s were different. Almost every night in the pitch-dark, silent city, workers at the prison who were quartered on the boulevard heard the screams of people being tortured. Indeed, all the survivors and people who worked at the prison share the memory of hearing people crying out in pain at night. Moving through the museum, absorbing its archive and listening to survivors and to people who worked at the prison, we can still hear many of these ghostly voices. They control the narrative that follows. Copyright © 1999 Regents of the University of California. All rights reserved.
