HEADING up into the mountains of Morazán, in the bright, clear air near the Honduran border, you cross the Torola River, the wooden slats of the one-lane bridge clattering beneath your wheels, and enter what was the fiercest of El Salvador’s zonas rojas — or “red zones,” as the military officers knew them during a decade of civil war — and after climbing for some time you take leave of the worn blacktop to follow for several miles a bone-jarring dirt track that hugs a mountainside, and soon you will find, among ruined towns and long-abandoned villages that are coming slowly, painfully back to life, a tiny hamlet, by now little more than a scattering of ruins, that is being rapidly reclaimed by the earth, its broken adobe walls cracking and crumbling and giving way before an onslaught of weeds, which are fuelled by the rain that beats down each afternoon and by the fog that settles heavily at night in the valleys. Nearby, in the long-depopulated villages, you can see stirrings of life: even in Arambala, a mile or so away, with its broad grassy plaza bordered by collapsed buildings and dominated, where once a fine church stood, by a shell-pocked bell tower and a jagged adobe arch looming against the sky — even here, a boy leads a brown cow by a rope, a man in a billed cap and bluejeans trudges along bearing lengths of lumber on his shoulder, three little girls stand on tiptoe at a porch railing, waving and giggling at a passing car.

 

In a remote corner of El Salvador, investigators uncovered the remains of a horrible crime — a crime that Washington had long denied. The villagers of El Mozote had the misfortune to find themselves in the path of the Salvadoran Army’s anti-Communist crusade. The story of the massacre at El Mozote — how it came about, and hy it had to be denied — stands as a central parable of the Cold War.

But follow the stony dirt track, which turns and twists through the woodland, and in a few minutes you enter a large clearing, and here all is quiet. No one has returned to El Mozote. Empty as it is, shot through with sunlight, the place remains — as a young guerrilla who had patrolled here during the war told me with a shiver — espantoso: spooky, scary, dreadful. After a moment’s gaze, half a dozen battered structures — roofless, doorless, windowless, half engulfed by underbrush — resolve themselves into a semblance of pattern: four ruins off to the right must have marked the main street, and a fifth the beginning of a side lane, while an open area opposite looks to have been a common, though no church can be seen — only a ragged knoll, a sort of earthen platform nearly invisible beneath a great tangle of weeds and brush.

Into this quiet clearing, in mid-October last year, a convoy of four-wheel drives and pickup trucks rumbled, disgorging into the center of El Mozote a score of outsiders. Some of these men and women — most of them young, and casually dressed in T-shirts and jeans and work pants — began dumping out into the dust a glinting clutter of machetes, picks, and hoes. Others gathered around the hillock, consulted clipboards and notebooks and maps, poked around in the man-high brush. Finally, they took up machetes and began to hack at the weeds, being careful not to pull any, lest the movement of the roots disturb what lay beneath. Chopping and hacking in the morning sun, they uncovered, bit by bit, a mass of red-brown soil, and before long they had revealed an earthen mound protruding several feet from the ground, like a lopsided bluff, and barely contained at its base by a low stone wall.

---

 

They pounded stakes into the ground and marked off the mound with bright-yellow tape; they stretched lengths of twine this way and that to divide it into quadrangles; they brought out tape measures and rulers and levels to record its dimensions and map its contours. And then they began to dig. At first, they loosened the earth with hoes, took it up in shovels, dumped it into plastic pails, and poured it onto a screen large enough to require several people to shake it back and forth. As they dug deeper, they exchanged these tools for smaller, more precise ones: hand shovels, trowels, brushes, dustpans, screens. Slowly, painstakingly, they dug and sifted, making their way through the several feet of earth and crumbled adobe — remnants of a building’s walls — and, by the end of the second day, reaching wood-beam splinters and tile shards, many now blackened by fire, that had formed the building’s roof. Then, late on the afternoon of the third day, as they crouched low over the ground and stroked with tiny brushes to draw away bits of reddish dust, darkened forms began to emerge from the earth, taking shape in the soil like fossils embedded in stone; and soon they knew that they had begun to find, in the northeast corner of the ruined sacristy of the church of Santa Catarina of El Mozote, the skulls of those who had once worshipped there. By the next afternoon, the workers had uncovered twenty-five of them, and all but two were the skulls of children.

Later that afternoon, the leaders of the team — four young experts from the Argentine Forensic Anthropology Unit, who had gained a worldwide reputation for having exhumed sites of massacres in Guatemala and Bolivia and Panama and Iraq, as well as in their own country — piled into their white four-wheel-drive vehicle and followed the bumpy, stony road out of El Mozote (the Thistle). Slowly, they drove through Arambala, waving to the smiling little girls standing on their porch, and out onto the calle negra — the “black road” — which traced its way up the spine of the red zone, stretching north from the city of San Francisco Gotera to the mountain town of Perqu?n, not far from the Honduran border. At the black road, the Argentines turned left, as they did each evening, heading down to Gotera, but this time they stopped in front of a small house — a hut, really, made of scrap wood and sheet metal and set among banana trees some fifteen yards from the road. Getting out of the car, they climbed through the barbed wire and called out, and soon there appeared at the door a middle-aged woman, heavyset, with high cheekbones, strong features, and a powerful air of dignity. In some excitement, the Argentines told her what they had found that day. The woman listened silently, and when they had finished she paused, then spoke.

 

“No les dije?” she asked. (“Didn’t I tell you?”) “Si sólo se oí­a aquella gran gritazón.” (“All you could hear was that enormous screaming.”)

For eleven years, Rufina Amaya Mí­rquez had served the world as the most eloquent witness of what had happened at El Mozote, but though she had told her story again and again, much of the world had refused to believe her. In the polarized and brutal world of wartime El Salvador, the newspapers and radio stations simply ignored what Rufina had to say, as they habitually ignored unpalatable accounts of how the government was prosecuting the war against the leftist rebels.

In the United States, however, Rufina’s account of what had happened at El Mozote appeared on the front pages of the Washington Post and the New York Times, at the very moment when members of Congress were bitterly debating whether they should cut off aid to a Salvadoran regime so desperate that it had apparently resorted to the most savage methods of war. El Mozote seemed to epitomize those methods, and in Washington the story heralded what became perhaps the classic debate of the late Cold War: between those who argued that, given the geopolitical stakes in Central America, the United States had no choice but to go on supporting a “friendly” regime, however disreputable it might seem, because the alternative — the possibility of another Communist victory in the region — was clearly worse, and those who insisted that the country must be willing to wash its hands of what had become a morally corrupting struggle. Rufina’s story came to Washington just when the country’s paramount Cold War national-security concerns were clashing — as loudly and unambiguously as they ever would during four decades — with its professed high-minded respect for human rights.

In the United States, the free press was not to be denied: El Mozote was reported; Rufina’s story was told; the angry debate in Congress intensified. But then the Republican Administration, burdened as it was with the heavy duties of national security, denied that any credible evidence existed that a massacre had taken place; and the Democratic Congress, after denouncing, yet again, the murderous abuses of the Salvadoran regime, in the end accepted the Administration’s “certification” that its ally was nonetheless making a “significant effort to comply with internationally recognized human rights.” The flow of aid went on, and soon increased.

By early 1992, when a peace agreement between the government and the guerrillas was finally signed, Americans had spent more than four billion dollars funding a civil war that had lasted twelve years and left seventy-five thousand Salvadorans dead. By then, of course, the bitter fight over El Mozote had largely been forgotten; Washington had turned its gaze to other places and other things. For most Americans, El Salvador had long since slipped back into obscurity. But El Mozote may well have been the largest massacre in modern Latin-American history. That in the United States it came to be known, that it was exposed to the light and then allowed to fall back into the dark, makes the story of El Mozote — how it came to happen and how it came to be denied — a central parable of the Cold War.

In the weeks that followed the discovery of the skulls of the children, as each day’s work at El Mozote yielded up a fresh harvest, the initial numbers came to seem small. But in San Salvador, five hours by road to the west, where President Alfredo Cristiani and the generals and the guerrillas-turned-politicians were struggling with one another about how to put in place, or not put in place, a purge of the officer corps, which was proving to be the most difficult provision of the ten-month-old peace accord — struggling, that is, over what kind of “reconciliation” would come to pass in El Salvador after more than a decade of savage war — the first skulls of the children were enough to provoke a poisonous controversy. Those twenty-three skulls, and the nearly one hundred more that were uncovered in the succeeding days, were accommodated by the nascent Salvadoran body politic in two ways. Members of human-rights groups (those members, that is, who had survived the war), along with the politicians of the left (many of whom had been guerrillas shortly before), hailed the discovery as definitive evidence that a matanza — a great killing — had taken place in Morazán, and that what they had been saying for eleven years had at last been proved true. Members of the government, on the other hand, and various military officers found themselves forced to concede that something had indeed happened in Morazán, but they insisted that the situation was more complicated than it appeared. Dr. Juan Mateu Llort, the director of El Salvador’s Institute of Forensic Medicine, declared that the skulls themselves proved nothing, for “there were an abundance of armed children in the guerrillas.” El Diario de Hoy, an influential right-wing daily, published a reconstruction according to which guerrillas had “barricaded themselves in what seemed to have been a religious center and from there opened fire on the troops, making the deaths of children, women and old people possible.” President Cristiani’s government, already under attack for stalling in the dismissal of senior officers, maintained its position that no records existed of any Army operation in Morazán in early December of 1981.

And yet on the ninth day of that month any reader of La Prensa Gráfica, one of San Salvador’s major dailies, could have learned that “all the highways with access to Gotera and the other towns in the Department of Morazán are under strict military control … No vehicles or individuals are permitted to enter the zones of conflict in order to avoid accidents or misunderstandings … Neither was the entry of journalists or individuals permitted.” The Department of Morazán had been sealed off from the rest of the country. Four thousand men, drawn from the security forces — the National Guard and the Treasury Police — and from regular units of the Salvadoran Army, were hard at work. The area north of the Torola River, the heart of the red zone, was alive with the thud of mortars, the clatter of small-arms fire, and the intermittent roar of helicopters. Two days before, Operaci?n Rescate — Operation Rescue — had begun.

Many of the towns and villages were already empty; during and after Army operations of the previous spring and fall, thousands of peasants had left their homes and begun a long trek over the mountains to the Honduran border and refugee camps beyond. Of those who remained, many made it a practice, at the first sign of any Army approach, to leave their villages and hide in the caves and ravines and gullies that honeycombed the mountainous region. But El Mozote was crowded; in the days before Operation Rescue, people from the outlying areas had flooded into the hamlet.

“Many people were passing by the house, saying, ‘Come on, let’s go to El Mozote,’ ” an old peasant named Sebastiano Luna told me as he stood behind the yellow tape, watching the experts bent low over the brown earth of the sacristy of Santa Catarina. Between their feet lay an expanse of dark rubble, a miniature landscape of hills and ridges and valleys in every shade of brown. It took a moment or two to distinguish, among the dirty-brown hillocks, the skulls and parts of skulls, each marked with a bit of red tape and a number; and, beneath the skulls and skull fragments and the earthen rubble, scores of small brown bundles, heaped one on top of another, twisted together, the material so impregnated with blood and soil that it could no longer be recognized as clothing.

Amid the rubble in the northeast corner of the tiny room that had been called el convento (though it was really a kind of combined sacristy and parish house, in which an itinerant priest, when he visited the hamlet, would vest himself, and sometimes, perhaps, stay the night), a dark-haired young woman in denim overalls was kneeling. She slowly drew a small bundle toward her — it had beenlabelled No. 59 — and began, with almost agonizing gentleness, extracting the brown bits and placing them on a sheet of cardboard. “Left tibia, fragments only,” she sang out in a low monotone. “Vertebrae, one, two, three … six of them … Tibia, left, I think … Metacarpals …”

Now she disentangled the bits of ruined fabric: “Belt of brown leather, metal buckle … Pants, light in color, with patches of blue and green color in the posterior part … In the pants pocket … ah … um …” The strong voice took an odd slide downward and stopped. Over her shoulder, I saw her staring at something in her palm, then heard her swear in a low voice: “Hijo de puta!” She turned and opened her hand to reveal a tiny figure: a little horse of bright-orange plastic. No. 59 had been a lucky child, had had a family prosperous enough to provide a lucky toy.

After a moment, the anthropologist Mercedes Doretti said, “Ordinarily, we could use this for identification. I mean, even after eleven years, any mother would recognize this as her kid’s, you know?” She looked back at No. 59 and then at the brown rubble. “But here, here they killed all the mothers, too.”

Behind the yellow tape, Sebastiano Luna and his wife, Alba Ignacia del Cid, stood silent amid a knot of peasants, watching. They had walked from their small house, several miles outside El Mozote, where the dirt track joins the black road. Eleven years before, in early December, scores of people were passing by their house, pulling their children along by the hand, laboring under the weight of their belongings. “Come with us!” they had called out to the old couple. “Come with us to El Mozote!”

The afternoon before, the people of El Mozote had gathered, some fifty yards from the church, in front of the general store of Marcos Dí­az, the richest man in town. He had summoned the townspeople, neighbors and customers all, and when they had assembled — perhaps a couple of hundred of them, the men in caps and straw hats and the women in bright-colored skirts, holding children in their arms — Marcos Dí­az addressed them from his doorway. He had just come up the mountain from his regular buying trip to San Miguel, he said, and as he was waiting at the checkpoint in Gotera, at the entrance to the red zone, an officer in the town had greeted him — Marcos Dí­az, an important man, had friends among the officers — and then pulled him aside for a little talk. Dí­az would do well to stock up, the officer said, for soon the Army would launch a large operation in Morazán, and “nothing and no one” would be permitted to enter or leave the zone. But his friend Dí­az needn’t worry, the officer assured him. The people of El Mozote would have no problems — provided they stayed where they were.

In the street that day, these words of Marcos Dí­az’s set off a debate. Some townspeople wanted to head for the mountains immediately, for the war had lately been coming closer to the hamlet; only the week before, a plane had dropped two bombs near El Mozote, damaging its one-room school, and though no one had been hurt, the people had been terrified. “A lot of people wanted to leave — there was a lot of fear,” Rufina Amaya said when I visited her a year ago. “And a few people did leave. My godfather left, with his family. My children were crying. They said, ‘Mama, let’s go.’ ” But Marcos Dí­az, a man of influence, had put his prestige on the line, and he insisted that his neighbors would be safe only if they stayed in their homes — that if they left the hamlet they and their families risked being caught up in the operation. “That was the lie,” Rufina Amaya told me. “That was the betrayal. Otherwise, people would have left.” In the end, Marcos Dí­az’s prestige decided the issue. Though the debate went on that afternoon and into the following morning, most of the people of El Mozote finally accepted his assurances.

They had seen soldiers before, after all; soldiers often passed through on patrol and sometimes bought supplies in El Mozote. Only the month before, soldiers had come during an operation and occupied El Chingo and La Cruz, two hills overlooking the town, and though the people of El Mozote could hear mortars and scattered shooting in the distance, the soldiers had not bothered them. In the crazy-quilt map of northern Morazán in 1981, where villages “belonged” to the government or to the guerrillas or to neither or both, where the officers saw the towns and hamlets in varying shades of pink and red, El Mozote had not been known as a guerrilla town. “The Army spent a lot of time around here,” Rufina told me. “We all sold them food. If the soldiers were looking to find guerrillas, that was fine with us, because we didn’t have anything to do with them. And the guerrillas knew about our relations with the Army.”

The guerrillas knew, the soldiers knew: northern Morazán during the early eighties was a very small world, in which identity, or the perception of identity, often meant the difference between living and dying. That El Mozote in late 1981 was not a guerrilla town is a fact central to Rufina’s story and lies at the heart of the mystery of what happened there; and though it is a fact — one that almost everyone from the zone affirms — it seems to have nonetheless been a slightly more complicated fact than Rufina makes out. As in many other communities in northern Morazán, the people of El Mozote were struggling to keep their balance in the middle of the perilously shifting ground of a brutal war — were working hard to remain on friendly terms with the soldiers while fearing to alienate the guerrillas. Joaqu?n Villalobos, who was the leading comandante of the People’s Revolutionary Army (or E.R.P.), the dominant guerrilla group in Morazán, told me flatly during an interview last year in his headquarters in San Salvador that the people of El Mozote “would not support us” — only to concede twenty minutes later that his fighters had, at least on some occasions, bought supplies in the hamlet. “They had the lowest level of relationship with us — only the very slightest commercial one,” Villalobos said. Licho, a rebel commander who had grown up in Jocoaitique, a few miles from El Mozote, acknowledged to me during an interview in Perqu?n that in the late seventies “some from El Mozote had been our supporters,” but that long before 1981 these supporters “had come along with us, they were with us.” He added quickly, “The people who were still in El Mozote were afraid of us.”

But the reason, apparently, was not only their fear — the frank terror that many villagers in the zone felt about exposing themselves to Army retribution — but their ideology. The guerrillas’ support in Morazán had grown largely in soil made fertile by the work of Catholic liberation theology, but El Mozote had been uniquely unreceptive to such blandishments, for the hamlet was a stronghold of the Protestant evangelical movement. People had begun to convert as early as the mid-sixties, and by 1980 it is likely that half or more of the people in El Mozote considered themselves born-again Christians; the evangelicals had their own chapel and their own pastor, and they were known — as were born-again Christians throughout Central America — for their anti-Communism. “Everyone knew there were many evangelicals in El Mozote, and these people wouldn’t support us,” Licho told me. “Sometimes they sold us things, yes, but they didn’t want anything to do with us.”

So, unlike many other hamlets of Morazán, El Mozote was a place where the guerrillas had learned not to look for recruits; instead, a delicate coexistence had been forged — an unstated agreement by both parties to look the other way. The guerrillas passed by El Mozote only at night, and when they did, Rufina says, “the people would hear the dogs barking and they’d be afraid.” She remembers seeing guerrillas only once in daylight: a few ragged young people, unarmed and wearing civilian clothes, had come into the hamlet and tried to hold a meeting in the tiny church of Santa Catarina. Rufina didn’t attend, nor did most of the other townspeople. “I remember people saying, ‘Don’t get involved. Let’s just live and work and not get involved.’ People just didn’t want anything to do with it. I had four children to look after. You’re worrying about feeding your family, and you try not to pay attention to these other things.”

And so when Marcos Dí­az brought his news from Gotera, when he conveyed the strong words of the officer and presented the choice as one of leaving the town and risking “getting involved” in the operation or of staying put and remaining safe, there was never much doubt about what the people of El Mozote would in the end decide. That very afternoon, at Marcos Dí­az’s urging, people began fanning out from the hamlet into the outlying districts to spread the word that one and all should come to El Mozote, and quickly, for only there would they be protected. Marcos Dí­az helped matters along by letting it be known that he would offer on credit as much food and other supplies as the newcomers needed. Peasants poured into the hamlet, occupying every bit of space. “All the rooms in Marcos Dí­az’s house were filled with people,” Rufina recalls. “Every house had people staying there from outside.” Even the common in front of the church was crowded with people, for the few houses could not accommodate them all.

” ‘Come to El Mozote’ — that’s what everyone was saying,” the old peasant Sebastiano Luna told me. He and Alba Ignacia del Cid had stood in front of their house, had watched the people pass. But they had decided not to go. “I had half an idea something bad might happen,” Sebastiano said. “So I told her” — nodding to his wife — ” ‘You, you go if you want to. I’m staying.’ ” “And I,” said Alba, “I said, ‘No, no, I won’t go without you, because they’ll ask me where my husband is. They’ll say he’s not here because he’s a guerrilla and then they’ll kill me. Either we both go or we both stay.’ ” So Sebastiano and Alba hid in the mountains above their house. They saw soldiers pass by, and saw a helicopter hover and descend. And later they saw thick columns of smoke rising from El Mozote, and smelled the odor of what seemed like tons of roasting meat.

Four miles south of El Mozote, outside the hamlet of La Guacamaya, the guerrillas of the People’s Revolutionary Army also awaited the soldiers. From their agents in the capital, they knew that large shipments of American munitions had been arriving at Ilopango Airport, and that truckloads of troops had begun moving along the Pan-American Highway toward Morazán. On December 1st, Jonás, the most powerful comandante in the zone, had pulled aside Santiago, the director of the E.R.P.’s clandestine Radio Venceremos, and informed him that “an operation of great breadth, named Yunque y Martillo” — Hammer and Anvil — was being planned. Santiago recalls that “intelligence sources within the Army itself” had passed on a report of a key meeting at the High Command. According to the source’s reconstruction, the Minister of Defense, Colonel José Guillermo Garcí­a, declared to his officers that Operation Rescue must “wrest the offensive from the F.M.L.N.” — the guerrilla umbrella group, of which the E.R.P. was one of five members. His Vice-Minister, Colonel Francisco Adolfo Castillo, added that the troops “must advance no matter what the cost until we reach the command post and Radio Venceremos.” Then Lieutenant Colonel Domingo Monterrosa Barrios, the dynamic commander of the élite Atlacatl Battalion, broke in, agreeing wholeheartedly that “so long as we don’t finish off this Radio Venceremos, we’ll always have a scorpion up our ass.”

Colonel Monterrosa, who at the time was the most celebrated field commander in the Salvadoran Army, was well known to have an obsession with Radio Venceremos. He was not alone: the station, which specialized in ideological propaganda, acerbic commentary, and pointed ridicule of the government, infuriated most officers, for its every broadcast reminded the world of the Army’s impotence in much of Morazán. Even worse, the radio managed to be funny. “They actually acted out this daytime serial, like a soap opera, with Ambassador Hinton in it,” a United States defense attaché of the time recalls. “They’d call the Ambassador ‘this gringo who is marrying a Salvadoran woman’ ” — Deane Hinton was about to marry a woman from one of the country’s wealthy families — “and at the end they’d say, ‘Tune in again tomorrow.’ And you couldn’t do anything about it. Most people at the Embassy, including the Ambassador, wanted to hear it.” The mortified Salvadoran officers maintained that the broadcasts originated in Nicaragua or Honduras.

Colonel Monterrosa was mortified by Radio Venceremos as well, but, unlike his colleagues, he had determined, in his rage and frustration, to do something about it. For Monterrosa, as American military advisers had come to realize, was a very different kind of Salvadoran officer. By late 1981, with Congress and the American public having shown themselves resolutely opposed to dispatching American combat forces to Central America, it had become quite clear that the only way to prevent “another Nicaragua” was somehow to “reform” the Salvadoran Army. “We were on our last legs,” an American military adviser who was in the country at the time told me. “We had to reform or we were going to lose. And it wasn’t because the guerrillas were so good; it was because the Army was so bad.” Salvadoran troops were sent into the field virtually untrained, soldiers rarely left the barracks after five o’clock in the afternoon, and officers rarely left the barracks at all. “The institution simply did not support people being good commanders,” this adviser said. “I mean, who ever got relieved? You could surrender with eighty-five men and nothing at all would happen to you.”

As the Americans soon realized, however, “reform” meant remaking an officer corps that had developed its own, very special criteria for advancement and reward. These had to do not with military competence but with politics: with showing unstinting loyalty to “the institution” and, above all, to one’s military-academy class — one’s tanda, as it was called. A hundred teen-age boys might enter the Gerardo Barrios Military Academy, and from their number perhaps twenty toughened, hardened men would emerge four years later; throughout the next quarter century, these men would be promoted together, would become rich together, and would gradually gain power together. If among them there proved to be embarrassing incompetents, not to mention murderers and rapists and thieves, then these men were shielded by their classmates, and defended ferociously. Finally, perhaps two decades after graduation, one or two from the tanda — those who had stood out early on as presidenciables, as destined to become leaders of the country — would lobby within the officer corps to become the President of El Salvador.

Monterrosa had graduated in 1963, and though the records show him fourth in his class of nineteen, it is a testament to the respect he inspired that many officers now remember him as first. In the academy, he was a magnetic figure, charismatic from the start. Short, with the simple face and large nose of a Salvadoran peasant, he walked with the peasant’s long, loping stride, which made his distinctly nonmartial figure recognizable from far off. General Adolfo Blandón, a former chief of staff, who was in his last year in the academy when Monterrosa was in his first, recalls that the young man “established himself immediately as the best in his class — the top rank in studies, physical conditioning, knowledge of the concepts of war.”

Normally, of course, such prestige, such respect from his colleagues, would brand him presidenciable. But, unlike his fellows, Monterrosa was, as Blandón puts it, “that rare thing: a pure, one- hundred-per-cent soldier, a natural leader, a born military man with the rare quality of being able to instill loyalty in his men.”

In the years after his graduation, Monterrosa taught at the academy, took courses from the Americans in Panama, travelled to Taiwan to study anti-Communist counter-insurgency tactics, and served in the paratroops as part of El Salvador’s first free-fall team. After the controversial elections of 1972, in which a hard-line faction of the military stole the ballot from what looked to be a winning Christian Democratic ticket, led by San Salvador’s Mayor José Napoleón Duarte, Monterrosa grew close to the new military President, Colonel Arturo Molina.

In the Army at this time, the key focus was on politics, and the struggle over El Salvador’s stunted political development increasingly split the country, and the officer corps. By the late seventies, after Molina had given place to General Carlos Humberto Romero, in another dubious election, the situation had become even more polarized. On the far left, several tiny guerrilla groups were kidnapping businessmen, robbing banks, and, on occasion, assassinating prominent rightist leaders. Activists on the moderate left, having been denied an electoral path to the Presidential Palace by the Army’s habitual ballot tampering, joined populist forces in organizing vast demonstrations, and managed to bring hundreds of thousands of people into the streets. The security forces generally responded to these demonstrations with unflinching violence, shooting down scores, and sometimes hundreds, of Salvadorans.

Within the Salvadoran officer corps, the country’s political crisis had reopened a political fault line that had spread apart periodically throughout the century. Back in 1960, a faction of “progressive” officers had staged a coup, but it had been quickly reversed by a conservative counter-coup; in 1972, when Duarte’s victory was stolen by conservative officers, the progressives attempted another, with the same result. Finally, in October of 1979, with at least tacit American support, a group of young “reformists” who called themselves the juventud militar — the “military youth” — overthrew General Romero and set in his place a “progressive” junta, which included politicians of the left. As had happened two decades before, however, the conservatives in the Army almost immediately regained the upper hand, and now, under cover of a more internationally acceptable “reformist” government, they felt free to combat the “Communist agitation” in their own particular way — by intensifying the “dirty war” against the left.

The most visible signs of the “dirty war” were mutilated corpses that each morning littered the streets of El Salvador’s cities. Sometimes the bodies were headless, or faceless, their features having been obliterated with a shotgun blast or an application of battery acid; sometimes limbs were missing, or hands or feet chopped off, or eyes gouged out; women’s genitals were torn and bloody, bespeaking repeated rape; men’s were often found severed and stuffed into their mouths. And cut into the flesh of a corpse’s back or chest was likely to be the signature of one or another of the “death squads” that had done the work, the most notorious of which were the Union of White Warriors and the Maximiliano Hernández Martí­nez Brigade.

The latter was named for a general who had taken over the country in 1931, during a time of rising leftist agitation among the peasantry, and had responded the following year with a campaign of repression so ferocious that it came to be known simply as La Matanza. Throughout the western part of the country, where an abortive rebellion had been centered, members of the National Guard, along with civilian irregulars, lined peasants up against a wall and shot them. Before the purge was over, they had murdered well over ten thousand people.

Now rightist officers who proudly counted themselves heirs of Martí­nez determined to root out this new leftist infection with equal thoroughness. Drawing on money from wealthy businessmen who had moved to Miami to avoid kidnapping or assassination, and benefitting from the theoretical guidance of ideological compatriots in neighboring Guatemala, the officers organized and unleashed an efficient campaign of terror in the cities. The campaign intensified dramatically after the “progressive” coup of October, 1979. By the end of the year, monthly estimates of the dead ranged as high as eight hundred.

Against the urban infrastructure of the left — the network of political organizers, labor leaders, human-rights workers, teachers, and activists of all progressive stripes which had put together the enormous demonstrations of the late seventies — this technique proved devastating. “These people weren’t organized militarily, which is what made them so easy to kill,” William Stanley, a professor of political science at the University of New Mexico, told me in an interview in San Salvador. As the repression went on, month after month, it became less and less discriminating. “By the end, the killing basically outran the intelligence capability of the Army and the security services, and they began killing according to very crude profiles,” Stanley said. “I remember, for example, hearing that a big pile of corpses was discovered one morning, and almost all of them turned out to be young women wearing jeans and tennis shoes. Apparently, one of the intelligence people had decided that this ‘profile’ — you know, young women who dressed in that way — made it easy to separate out ‘leftists,’ and so that became one of the profiles that they used to round up so-called subversives.”

Some civilians were certainly involved, particularly on the funding end, but there can be no doubt that the “dirty war” was basically organized and directed by Salvadoran Army officers — and no doubt, either, that the American Embassy was well aware of it. “There was no secret about who was doing the killing,” Howard Lane, the public-affairs officer in the Embassy from 1980 to 1982, told me in an interview. “I mean, you formed that view within forty-eight hours after arriving in the country, and there was no secret at all about it — except, maybe, in the White House.” In public, the fiction was resolutely maintained that the identity of the killers was a mystery — that the corpses were the work of “rightist vigilantes.” This campaign of lies was designed in part to accommodate the squeamishness of the Administration in Washington, which had to deal with growing concern in Congress about “human-rights violations,” particularly after several notorious cases, including the murder, in March of 1980, of Archbishop Oscar Romero while he said Mass; the rape and murder, the following December, of four American churchwomen; and the assassination, in January of 1981, of the head of the Salvadoran land-reform agency and two of his American advisers.

On the evening of December 1, 1981, Santiago, the director of Radio Venceremos, after learning from Jonás, the comandante, about the coming operation, set out on foot from the guerrilla base at La Guacamaya, four miles south of El Mozote. As darkness fell, Santiago hiked east over the hills and through the gullies, crossed the Rí­o Sapo, and climbed down into a heavily forested ravine at El Zapotal. Here, dug into a rock niche half a dozen feet underground, was the “studio” of Radio Venceremos, which consisted of a small transmitter, an unwieldy gasoline generator, assorted tape recorders, microphones, and other paraphernalia, and a flexible antenna that snaked its way up through a forest of brush. Santiago gathered his handful of young staff members, and soon news of the coming operation was broadcast throughout the zone.

Back in La Guacamaya, in a rough encampment in the open air, perhaps two hundred young men and women, outfitted in a motley combination of peasant clothing and camouflage garb, were making preparations. Some cleaned their weapons — mostly old M1s and Mausers, along with a few captured American M16s. Many of the women bent over smooth flat stones, grinding corn, making the meal that would serve as the company’s fuel during the days ahead — for, confronted with the arrival of thousands of troops, the guerrillas of the E.R.P. were preparing not to fight but to flee.

Mobility and quickness had always been central to the guerrillas’ strength, along with their familiarity with the mountain terrain. Like El Salvador’s other radical groups, the People’s Revolutionary Army had been the brainchild of young urban intellectuals, who had founded the organization in Mexico City in 1972, funded it during the mid-seventies largely by robbing banks and by seizing and ransoming wealthy businessmen, and battled among themselves for its leadership, using high-toned abstract arguments of the left (which more than once deteriorated into violent schism).

“The revolutionary process started in Morazán around 1977 or ’78 with the consciousness-raising of Christian ‘base communities’ led by radical priests,” said Licho, the rebel commander, whose parents were campesinos living on the other side of the black road from El Mozote. “We young people would get together and read the Bible and apply it to our own situation, and gradually we became more politically aware.” When the young men came of age, the guerrilla leaders often urged them to join the Army — they had urged Licho to do so — in order to receive military training and gain firsthand knowledge of the enemy while providing useful intelligence until they could return to their home provinces to take up arms.

By 1980, small groups of young guerrillas were operating throughout northern Morazán, drawing food and support from sympathetic peasants, and launching raids from time to time against the National Guard posts in the towns. They would attack suddenly, kill a few Guardsmen and capture their weapons, then fade back into the bush. After the posts had been reinforced, the Guardsmen responded, as they had done for years, by beating or killing peasants they suspected of having been “infected” with Communist sympathies. This quickened the flow of able-bodied men and women into the mountains. Soon some villages were inhabited almost entirely by old people and mothers and their children. The Guardsmen abandoned some towns completely — in effect, ceding them to the control of the guerrillas. And the people abandoned other towns, either fleeing to the refugee camps beyond the Honduran border or joining the guerrillas, and thus forming, as time went on, a quasi-permanent baggage train of masas, or civilian supporters. “The people who supported us moved around as our rear guard, providing food and other help,” Licho told me. “In some areas, our supporters were in the majority, in others not.” The distinction between combatants and noncombatants, never very clear in this guerrilla war, was growing cloudier still.

The Salvadoran High Command had become increasingly alarmed by the situation in Morazán. “The military view the situation in the countryside as critical,” the United States Ambassador, Frank Devine, wrote in a 1980 cable. “Many feel there are ‘liberated’ areas where they dare not operate due to the concentration of leftist-terrorist strength.”

In January of 1981, the F.M.L.N. proclaimed a “final offensive” — the badly equipped guerrillas hoped to provoke a popular insurrection, as the Sandinistas had done in Nicaragua eighteen months before, and to do it in the days just before Ronald Reagan took power in Washington — but the people did not rise up, and the offensive ended in a costly defeat. After the collapse, hundreds of fighters streamed out of the cities and headed for the mountains. Having failed to overthrow the government, and having seen many of their civilian sympathizers liquidated in the past months by death squads, the guerrillas decided to focus their forces on a full-scale rural insurgency rooted in the northern mountains.

By November, General Fred F. Woerner, whom a worried Pentagon had sent to assess the Salvadoran war, was concluding in a secret report that the situation on the ground had so deteriorated that a primary aim of the Salvadoran Army had now become to “prevent the establishment of an insurgent ‘liberated’ zone in the Department of Morazán, which could lead to international recognition of the insurgents as a belligerent force.” (Three months before, France and Mexico had recognized the F.M.L.N. as “a representative political force.”) If the guerrillas were not dislodged from Morazán, the Salvadoran officers feared, they would risk seeing their country split in two.

On December 1, 1981, after Radio Venceremos broadcast word that the Army was coming, people throughout northern Morazán began talking among themselves, arguing, and coming to decisions about what to do next. Hundreds of people assembled outside the guerrilla camp at La Guacamaya, having packed up what tortillas and beans they had, and gathered their children, ready for the hard trek ahead.

On Monday, December 7th, the young men and women of Radio Venceremos began doing what they had practiced many times: rapidly dismantling the components, loading the generator aboard a mule, and hoisting the transmitter, the antenna, and the other equipment on one another’s backs. Then they hiked off to join the fighters at La Guacamaya.

Around this time, according to Joaquí­n Villalobos, representatives of the guerrillas approached El Mozote and attempted to warn the campesinos there. “We always had rear-guard people, political people, behind the lines,” he says. “So when the fighting was beginning in the south they advised people in the north to leave the zone.” But the people in El Mozote had already made their decision. “Because they had little relation to us, and because they were evangelical, they decided they had little to fear from the Army,” Villalobos says. More likely, they had decided, after listening to the words of Marcos Dí­az, that the danger would be greater outside the hamlet than within.

“We told them what might happen,” Licho says. “But they didn’t believe that the Army would do anything to them.” Perhaps they regarded the guerrillas’ warnings — those who heard them, that is (Rufina, for one, heard nothing) — as attempts at recruitment. As the people of El Mozote well knew, in the view of the Salvadoran Army, to go with the guerrillas was to be a guerrilla.

By Tuesday morning, December 8th, the guerrillas at La Guacamaya could hear the sounds of battle, of mortars and small-arms fire, coming, it seemed, from all directions; they knew by now that perhaps four thousand soldiers had entered the zone, that troops had crossed the Torola and were moving toward them from the south, that others were approaching the Sapo from the east. The only way clear had seemed to be to the north, toward the Honduran border; but, even as the Radio Venceremos announcers were putting out their last broadcast, urging the people of the zone to join the guerrilla columns, the guerrillas heard the helicopters approach and saw them pass overhead, carrying the troops of Domingo Monterrosa’s Atlacatl Battalion northward, to the mountain town of Perquí­n.

To reach Perquí­n from El Mozote, you turn right on the black road and begin to climb. Soon the grade grows steeper, the tropical brush gives place to mountain pine, and the air lightens and grows fresh. Here and there, a bit of sorghum or corn or maguey pokes out from among the trees, but, increasingly, from the red soil of the mountainside only great white rocks grow. The overpowering fragrance of freshly cut pine announces the hamlet of La Tejera and its sawmill, a low building of unstripped logs surrounded by stacks of new planks. Finally, a sign announces Perquí­n; the road tilts sharply upward and becomes a street of large cobblestones; and, after a few moments’ jolting, the traveller comes upon a dramatically uneven town square, which, despite blasted buildings and damaged streets, remains an oddly beautiful, vaguely otherworldly place. At its heart is a bizarre park, which accommodates many wildly slanting levels of green grass, like lopsided terraces on a cultivated but dilapidated hillside. Bordering the park are a yellow-painted clinic, a rough-hewn little hut, and a remarkable church crowned by a bulbous steeple.

When Colonel Monterrosa set his helicopter down here in December of 1981, he found a town in government hands, but barely. Only four months earlier, in mid-August, the guerrillas had swept out of the surrounding hills and overwhelmed the local National Guard post, killing four men and capturing five. “There were many young ones, but some really old ones, too,” children in Perquí­n told Alma Guillermoprieto, then a stringer for the Washington Post. “There were eight women. Some of them were in uniforms, but most of them wore raggedy clothes, like us. We knew some of them; they were from this town.” The guerrillas had spent a week and a half digging defensive trenches, buying corn from the local coí¶perative, and marching about the streets shouting “Pueblo libre!” and other slogans. When the Air Force began bombing the city, ten days later, the guerrillas swiftly vanished, fading into the mountains and ravines they knew so well, and leaving behind the four dead men, buried in a bomb crater, and also the civilians who had been there all along — the civilians who, after playing host to the guerrillas for ten days, now gazed with all innocence into the faces of the National Guardsmen who had taken the places of their dead comrades.

Colonel Monterrosa had thought long and hard about civilians and guerrilla war, about the necessity of counterinsurgency, about the frustrations of the odd and bloody conflict that the overextended Salvadoran Army had been fighting and losing. When the men of his Atlacatl Battalion touched down in Perquí­n that Tuesday morning in December, storming from helicopters in a crouch, gripping their helmets tightly against the backwash from the rotors, the officers had in their pockets lists of names to hand to the National Guardsmen. While the Atlacatl captains mustered their troops, the Guardsmen marched off through the town and pounded on doors. They were big men, well fed, and they looked even bigger than they were, outfitted in high black boots and uniforms of heavy greenish-brown cloth, with automatic rifles on their backs, and razor-sharp machetes hanging at their belts.

“In those days, if they came to your house to ask you to come with them to ‘do something,’ you’d end up dead,” a Perquí­n man whom the Guardsmen visited that morning told me. When he heard the pounding and pulled open the door to find the Guardsmen there glowering down at him — they always glowered, for their business was, and had been since the early days of the century, to induce fear in the countryside and to stamp out rebellion from the moment it revealed itself as a lessening of fear in a campesino’s eyes — this man could only try to control his terror as the Guardsmen stared for a moment, then barked, “Hey, we have work to do! Come with us and help us do it!” The man came outside, watched as one of the Guardsmen ran his finger down the list that Monterrosa’s men had handed him, then looked up, exchanged glances with his partner, and murmured, “Ya vamos dándole.” (“Now let’s get started.”) The Perquí­n man knew what that meant — the killing was to begin — and, in a panic, he began to protest, digging an identification card out of his pocket and begging the Guardsmen to look at it carefully. Finally, after a terrible few minutes, he succeeded in convincing these impassive men that the name on the list was not his — that one of the surnames was different.

Nonetheless, the Guardsmen hustled him along the streets with them, and as they moved through town they pounded on other doors and collected other frightened men. Those men numbered ten by the time they reached a field in front of the clinic, which was a blur of unaccustomed activity: helicopters landing and hovering and departing, and, amid the blast and the roar from the rotors, hundreds of men in green moving about, checking weapons, cinching the straps on their packs, and talking among themselves as officers marched back and forth shouting orders. By then, several hundred of the Atlacatl soldiers had stormed off the helicopters, most of them in olive green, and a few in camouflage garb above black jungle boots. On the shoulders of their uniforms they bore, in white or yellow, the figure of an Indian and the word “Atlacatl” (the name of a legendary Indian warrior who had led the fight against the conquistadores). To a practiced eye, they seemed a somewhat different breed from most Salvadoran soldiers — more businesslike, grimmer even — and their equipment was better: they had the latest American M16s, plenty of M60 machine guns, 90-mm. recoilless rifles, and 60- and 81-mm. mortars.

But it wasn’t their equipment that made them “the élite, American-trained Atlacatl Battalion” (as press accounts invariably identified them). It was their aggressiveness, their willingness to “do the job”: a willingness that the rest of the badly led and badly trained Army generally lacked. In part, perhaps, this aggressiveness was instilled by American trainers — Special Forces personnel, who, beginning in March, had been coming over from Southern Command, in Panama, to show the Salvadoran recruits how to shoot and how to seize positions. Mostly, though, it came from Monterrosa. Among senior field commanders who in many cases, as one lieutenant put it to me, “don’t even own fatigues,” Monterrosa seemed a soldier of the classic type: aggressive, charismatic, a man who liked nothing better than to get out in the field and fight alongside his troops. The Salvadoran grunts — mostly unlettered peasant boys, many of whom had been pulled from buses or off country roads and pressed into service, having received little training and less regard from their officers — loved Monterrosa for his willingness to get down in the dirt with them and fight. The press loved him, too: not only was he a natural story but he was only too happy to invite reporters to come along with him in his helicopter. And, of course, the Americans loved him as well: Colonel John Cash, a United States military attaché, speaks of “a hot-shot strategist like Monterrosa, whom I’d put up against any American hot shot.”

As the war moved decisively to the countryside, the American government was no longer able to deny that it had a major problem on its hands. The Salvadoran officers were showing themselves utterly incapable of fighting a war of rural counter-insurgency. Not only was the Army, with a total of thirteen thousand men facing perhaps a third that many guerrillas, terribly overstretched, but its officer corps was burdened by a byzantine political structure and a perverse system of anti-incentives. The most important commands from the military point of view were from the point of view of most Salvadoran officers the least desirable, and the result was that those posts tended to be assigned to the politically least powerful, and often least talented, members of the officer corps. “The guys in the real combat commands tended to be the total incompetents,” Todd Greentree, who was a junior reporting officer in the United States Embassy at the time, told me. “These guys would be sent out there to the end of the line, and they’d spend their days drinking in the cuartel.”

Embassy officials recommended, cajoled, and finally urged reassignments, but changes, when they came at all, came only after enormous effort. The explanation was not just the superior political and economic power of the right wing of the officer corps but the fact that the tanda system, in which classmates, no matter what their failings, were fiercely protected, appeared nearly impervious to outside pressure — including pressure from the Americans, who were now pouring hundreds of millions of dollars into the country. As the officers understood only too quickly, the ultimate sanction that the Americans could brandish — turning off the aid spigot — threatened to hurt the Americans themselves as much as it would hurt the Salvadorans, since the American fear of a Communist El Salvador taking its place alongside Sandinista Nicaragua had become overriding. Even during the final months of the Carter Administration, this underlying reality became embarrassingly evident, when President Carter, after cutting off aid in response to the murder of the American churchwomen, rushed to restore it only a few weeks later, in the face of the rebels’ “final offensive.”

Ronald Reagan did not suffer from the same ambivalence. By the fall of 1981, Reagan had removed the outspoken American Ambassador, Robert White; had vowed, through Secretary of State Alexander Haig, to “draw the line” in El Salvador against Communist subversion in the hemisphere; had almost doubled economic aid for El Salvador, to a hundred and forty-four million dollars, and increased military aid, from twenty-six million dollars to more than thirty-five million; and, in November, had begun funding the Nicaraguan Contra fighters as a proxy force against the Sandinista government. By late 1981, the priorities of American policy in El Salvador had become unmistakable.

The Americans had stepped forward to fund the war, but they were unwilling to fight it; it would be left to the Salvadorans to defeat the guerrillas. “The guerrilla always carries his masas into battle with him” was a famous Army saying of the era, a piece of received wisdom from that darkest period of the Salvadoran civil war, and its author was Colonel Monterrosa himself. It was intended not only as a statement of fact but as a general affirmation of principle: in this bloody war, in the red zones, there was really no such thing as a civilian.

A large professional Army would have reoccupied territory and sent out aggressive patrols, all the while doing “political work” in the countryside to regain the loyalty of the people. Indeed, that was part of the rationale behind the search-and-destroy operations. “There are a lot of different names for counter-guerrilla fighting,” Colonel Castillo, then the Vice-Minister of Defense, told me in an interview. “Whether they call it Hammer and Anvil, or the Piston, or something else, it’s all the same idea — to try to expel the guerrillas from the zone. After we managed to expel them, they would lose the support of all the people they had indoctrinated.”

But in those days, Castillo conceded, the Army “didn’t have enough equipment or forces to maintain operations there for a long enough time.” The result was that the Army would enter a zone in force; the guerrillas, after a few minor engagements, would flee; and the soldiers, after killing a number of supposed “subversives” (civilians who may or may not have been guerrilla supporters but hadn’t been quick enough, or smart enough, to get out of the way), would evacuate the zone, leaving a token force behind — which the guerrillas, when they flowed back in a few days later, would maul and expel.

The Army’s tactic was not effective, and it made for great frustration. “When I arrived here, in June of 1982, the Salvadoran officers used to brag to me that they didn’t take prisoners,” Colonel Cash, the military attaché, said. “They said, ‘We don’t want to dignify them by taking prisoners.’ They wouldn’t even call them prisoners, or guerrillas. They called them terroristas — delincuentes terroristas.” (General Blandón, the former chief of staff, told me, “Before 1983, we never took prisoners of war.”) As the guerrillas were reduced to the status of terrorist delinquents, all civilians in certain zones were reduced to the status of masas, guerrilla supporters, and thus became legitimate targets. North of the Torola, for example, it was believed that the civilians and the guerrillas were all mixed together, and were indistinguishable.

By late 1980, the Army had begun the tactic that William Stanley, the political-science professor, refers to as “killing by zone.” One of the first such operations took place in October, and began with a staff meeting in Perquí­n. “Colonel Castillo explained that it was necessary to stop the Communist revolution — that it was necessary to make an example of this place, so we wouldn’t have the same problems in other parts of the country,” an officer who had been present at the meeting told me. “He said we must take into account that the great majority of the people here are guerrillas. So the idea was to surround them all, to create this ‘hammer and anvil’ thing, push all the people down to Villa El Rosario, where a huge artillery barrage would be unleashed. The city would be totally destroyed. We were going to make an example of these people.” The brutality of this operation provoked the first major exodus from Morazán, as entire populations fled their villages for the refugee camps in Gotera, or for the camps over the border in Honduras.

Despite the Army’s success in taking away the water, however, the fish continued to multiply and grow stronger. In November of 1980, a month after the Villa El Rosario operation, the guerrillas began to receive the first of a number of shipments of small arms from the Sandinista regime in Nicaragua — “a mixture of fals, M16s, and Uzis,” according to Stanley. After the collapse of the “final offensive,” in January, the guerrillas also benefitted from a fresh infusion of manpower, including not only the fighters who had fled the cities but a number of important deserters from the Army.

In front of the Perquí­n health clinic that Tuesday in early December, amid the backwash from the helicopters, the men of the Atlacatl mustered and made ready. The National Guardsmen, who by this time had collected the ten villagers, pushed their reluctant charges forward through the troops until they reached a tall, green-eyed officer in combat fatigues, who was striding about amid the commotion, pointing here and there and issuing orders. One of the Perquí­n men, who had served in the Army several years before, recognized the officer as Major Natividad de Jesús Cáceres Cabrera, a legendary figure: sixth in his academy class, a born-again Christian, a fanatical anti-Communist, and now the executive officer of the Atlacatl Battalion. (Later, his legend grew: as a colonel in command of Chalatenango in 1986, he forced all the residents of that substantial city to “express their desire for peace … their purity, their soul, and also their cleanliness” by painting the entire city white; and in 1989, on a Salvadoran highway, Cáceres ordered his men to block the convoy of the American Ambassador, William Walker, and, when the Ambassador refused to emerge and offer proof of his identity, threatened to blow up his limousine with antitank weapons.)

On that Tuesday, Major Cáceres looked over the ten men and gestured to five captains who were organizing the companies under their command. “He put two of us with each company,” one of the Perquí­n men told me, “and he said, ‘We want you to come with us, to show us the area.’ ” They had been brought there to serve as guides for the Atlacatl.

Major Cáceres gathered the captains together, gave them pseudonyms to be used over the radio during the operation — he himself would be known as Charlie — and issued a few orders. Then the five companies of the Atlacatl moved out, down the mountainside. Everywhere, above the roar of the helicopters, could be heard the thud of mortars and the booming of artillery. “It was a huge operation,” the guide from Perquí­n told me. “There were helicopters and planes and heavy equipment and troops all through the mountains, and they even had animals to cart along some of the guns and ammunition.”

As the Atlacatl men set off south from Perquí­n, hundreds of other soldiers were moving steadily north. Having been deployed as a blocking force along the Torola and Sapo Rivers, to the south and east, and along the black road, to the west, they were now tightening the circle. These units, the hammer of the operation, were meant to push all the guerrillas in the zone up toward the anvil of the Atlacatl and crush them against the best troops the Army had to offer. But, as a lieutenant involved in the operation remarked to me, “you take troops from all over the country and move them up to Morazán in about ninety truckloads, right along the Pan-American Highway — I mean, you think somebody might notice?”

As Monterrosa’s men circled the hills below Perquí­n, the guerrillas of the People’s Revolutionary Army, far to the south, at La Guacamaya, completed their preparations. Confronted with a heavy force blocking the river to the south, and the Atlacatl moving down from the north, the guerrillas would break straight west, punching their way through the military’s lines at the black road. That night, some of their train started the trek: long columns of peasants, their belongings, food, and young children bundled on their backs, trudged single file through the mountains, flowing in a vast nocturnal exodus that would carry them over the mountains to the Honduran border.

On the morning of Wednesday, December 9th, while thick mist still carpeted the valleys, the men of the Third Company of the Atlacatl rose in their encampment on a hill called El Gigante, broke camp, and circled back toward the black road. In the hamlet of La Tejera that afternoon, they seized three civilians, two youths and an old man of eighty or more, hustled them along to a field not far from the sawmill, and began interrogating them “very strongly, very brutally,” according to the guide from Perquí­n. The officers accused the men of being guerrillas, demanded to be given the names of their comrades, to be told where they had hidden their weapons. When the men denied the charges, Major Cáceres declared that they would be executed; the killing, he said, would begin here. But then a farmer from the area came forward. The two youths worked for him, he told the Major, and he protested vigorously that they had nothing to do with the guerrillas. One of the guides vouched for them as well, and after a prolonged dispute the men were spared.

This argument over identity, over who was a guerrilla and who wasn’t and what constituted evidence one way or the other, would recur during the next two days. Already in La Tejera, officers disagreed about whether the men should have been spared; according to the guide, Captain Walter Oswaldo Salazar, the company commander, reacted angrily when he was told of a comment from another officer that the local people should be treated with respect unless there was evidence that they were guerrillas. “Salazar said, ‘No, these are all guerrillas,’ ” the guide said. “He said the soldiers could go ahead and kill any of them, or all of them.” Later that day, according to the guide, Captain Salazar let slip his suspicion that the other officer was in fact a guerrilla himself, and vowed to assassinate him.

This wasn’t simply paranoia. “We had tremendous infiltration in the Army at that time,” the lieutenant involved in the operation told me. “We knew that certain sales of arms were going to these people, that information was being leaked — all our operations, all our movements, were being leaked.” The overwhelming suspicion that this engendered, together with the growing panic among the officers about the deterioration in the government position, gave the hardest-line officers a decisive upper hand.

“The hard-core guys there really did believe that it was a virus, an infection,” Todd Greentree said. “They’d always say ‘a cancer’ — you know, ‘Communism is a cancer.’ And so if you’re a guerrilla they don’t just kill you, they kill your cousin, you know, everybody in the family, to make sure the cancer is cut out.”

These officers, of course, had Salvadoran history on their side. “They had a ‘kill the seed’ mentality,” Professor Stanley told me. “After all, what happened in 1932? To this day, when someone wants to make a threat here, why do they invoke the name of Martí­nez?” — the author of the Matanza. “Because he is an icon, that’s why. The idea of going out to the zones and killing everyone is not a new idea. It’s a proved idea.”

Putting that proved idea into practice would become the mission of the Atlacatl Battalion. Hoping to insure that at least one unit of the Salvadoran Army was adequately prepared to fight, the Americans sent Special Forces instructors in early 1981 to train the first recruits of the new Immediate Reaction Infantry Battalion (biri). Yet, as the American advisers well knew, the epithet of “élite, American-trained” that was hung on the Atlacatl by the press was a bit of a joke. “They had no specialized training,” one of the original Special Forces trainers told me. “They had basic individualized training — you know, basic shooting, marksmanship, squad tactics. I mean, the difference was that the Salvadorans basically had no trained units in the country, so this was going to be a unit that would be trained.”

Some officials in the Embassy and the Pentagon had wanted the entire unit to be trained in the United States — and, indeed, later in the year recruits for the second of the biris, the Belloso, would be flown en masse to Fort Bragg, North Carolina. But the Atlacatl had something the Belloso didn’t: it had Monterrosa. “That the battalion wasn’t sent to the United States but was trained by Monterrosa here was in large part a testament to his authority,” a contemporary of Monterrosa’s told me. “The High Command had been preparing him, grooming him. He had taken all the courses the Americans offered, including those for the paratroopers and the commandos. His ambition became very concrete around the time the Americans decided to direct a major counter-insurgency effort here. When the Atlacatl came along, he jumped at it.”

From the beginning, Monterrosa worked to give his new force a mí­stica — a mystique. “They shot animals and smeared the blood all over their faces, they slit open the animals’ bellies and drank the blood,” a lieutenant in another unit told me. “They were a hell of a raunchy unit. They had no discipline of fire, none at all. I mean, they saw something moving out there, they shot it — deer, pigs, whatever. You’d be out there in the field trying to sleep, and all night those assholes would keep shooting at things.” According to one reporter, the men of the Atlacatl celebrated their graduation by collecting all the dead animals they could find off the roads — dogs, vultures, anything — boiling them together into a bloody soup, and chugging it down. Then they stood at rigid attention and sang, full-throated, the unit’s theme song, “Somos Guerreros”:

We are warriors!
Warriors all!
We are going forth to kill A mountain of terrorists!

By the fall of 1981, the Atlacatl was well on its way to building that mountain. The pattern of its operations had become well known: units of the regular Army and the security forces would move into place along the border of one of the red zones, walling it off, with the help, very often, of a natural barrier, like a river or a mountain range. Then a blocking force would invade the zone, pushing before it everyone and everything living. Finally, the helicopters would sweep in, and the men of the Atlacatl would storm out, bombard all whom the trap had snared with artillery and mortar fire, and then with small arms.

It was the strategy of “draining the sea,” or, as Monterrosa was heard to describe it on occasion, of La Limpieza — the Cleanup. Those parts of El Salvador “infected” by Communism were being ruthlessly scrubbed; the cancer would be cut out, even if healthy flesh had to be lost, too. “El Mozote was in a place, in a zone, that was one hundred per cent controlled by the guerrillas,” one of the original American advisers with the Atlacatl told me. “You try to dry those areas up. You know you’re not going to be able to work with the civilian population up there, you’re never going to get a permanent base there. So you just decide to kill everybody. That’ll scare everybody else out of the zone. It’s done more out of frustration than anything else.”

Joaquí­n Villalobos, the E.R.P. comandante, freely conceded to me in an interview that in a number of the most notorious operations, both before and after El Mozote, many of the civilians killed were in fact sympathetic to the guerrillas. “In San Vicente in 1982, for example, the massacre at El Calabozo that involved more than two hundred people,” he said. “This was a situation where the Army was stronger, where our guerrilla force was too weak to protect our followers. We simply weren’t able to provide those people sufficient military protection. It was the same in 1980 at the Sumpul River in Chalatenango, where a group of our sympathizers were fleeing, trying to cross the river.” The guerrillas, benefitting from very good intelligence and excellent mobility, generally managed to escape from the zones ahead of the Army; it was their supporters, and any other civilians who happened to be there, who took the punishment.

In the case of many of the massacres during the early eighties, then, the Salvadoran Army was managing to do what it set out to do: killing Salvadorans who were sympathetic to the insurgents. However blatantly this behavior violated the rules of war — however infamous it was to murder men, women, and children en masse, without trial or investigation, simply because of the political sympathies of some of their number — the strategy did at least have some rationale. Even against this grim background, El Mozote stands out. “El Mozote was a town that was not militant,” Villalobos said. “That’s why what happened at El Mozote was special.”

Sometime during the incident at La Tejera that Wednesday afternoon, word came over the radio that the First Company of the Atlacatl had engaged the guerrillas. “There was an exchange of fire, an armed confrontation,” the guide says. But, like so much else in this story, the battle — its intensity, even its precise location — has become a matter of fierce dispute. From the start, the Salvadoran military claimed that the fighting took place at El Mozote itself. On December 17th, a C.I.A. officer cabled from San Salvador that “the heaviest fighting had occurred at El Mozote … where 30 to 35 insurgents and four Salvadoran soldiers were killed.”

It is impossible to know for sure, but from the context of the cable it seems very probable that the C.I.A. man’s information came, one way or another, from the Salvadoran Army. The guide, on the other hand, who was a few miles away and heard the report on the fighting as it came over the Atlacatl radio, places it “around Arambala. It was a little skirmish,” he told me, “and it happened at El Portillón, near Arambala — a little over a mile from El Mozote.”

Villalobos, who appears to remember the operation in great detail, also insists that the fighting took place at Arambala, which “was in effect our rear guard,” he said. “Although most of the serious fighting took place south of us, along the Torola, there was a minor level of fighting, including maybe a little mortar fire, near Arambala.” He went on to say, “It’s normal when you displace a large force to leave small units to protect the retreat and keep up resistance.” Guerrilla squads around Arambala, north of La Guacamaya, were in a perfect position to protect the flank of the main guerrilla force as it retreated west.

Santiago, who was still in La Guacamaya, readying his Radio Venceremos crew for that night’s retreat, describes how “the pressure of the enemy was growing in his north-south advance.” On that day, he writes in his memoirs, “the comrades of the Fourth Section took by assault a position of the Atlacatl Battalion and captured two rifles” — a plausible number in view of the four dead that the Salvadoran Army apparently acknowledged. But Santiago makes no mention of the “30 to 35 insurgents” killed that are claimed in the C.I.A. cable, and neither, so far as I know, does any other guerrilla memoir. This would have been a very large number of dead; the fact that no one mentions them, and the fact that, in the wake of this fighting, the guerrillas did indeed manage, as Santiago recounts, to “maintain the lines of fire and organize the movement to break the circle and make a joke of Monterrosa’s hammer blow” — these two facts lead one to wonder whether the officers, in providing their reports to their own superiors (and possibly to the C.I.A.), had created a victory at El Mozote from what was in fact a defeat at Arambala.

The officers would have been especially reluctant to admit a defeat at the hands of the Fourth Section. An élite guerrilla unit, it had been trained, in large part, by Captain Francisco Emilio Mena Sandoval, an Army officer who had deserted to the guerrillas the previous January. Salvadoran officers had developed a deep hatred for Mena Sandoval, regarding him and others like him as much more despicable forms of life than, say, Villalobos: in their eyes, the latter was merely a delinquent terrorist, whereas officers like Mena Sandoval were traitors. And, as it happened, the officers and men of the Atlacatl had a special reason not only to hate Mena Sandoval but to remember with the greatest distaste the town of Arambala and also the hamlet of El Mozote, just down the road.

It was near Arambala, eight months earlier, that the first unit of the brash new Atlacatl had ventured forth to show the guerrillas, and the rest of the Army, what it was made of; and it was there that, to the embarrassment of its officers and men, the highly touted new unit suffered a humiliating defeat — in large part because Captain Mena Sandoval had had the foresight to steal an Army radio when he came over to the guerrillas. Thanks to the radio and Mena Sandoval’s knowledge of the enemy’s codes, the rebels were able to keep one crucial step ahead of their opponents. “We defended one line on the outskirts of El Mozote, which the enemy was unable to take for many days,” Mena Sandoval writes in his memoirs. “Their cost in casualties kept growing, as did our morale. It had been twelve days of combat and we had almost no casualties.”

Finally, after twenty-two days of intense fighting, the guerrillas slipped away across the black road under cover of night. As for the Atlacatl, news of its poor performance spread quickly through the Army. Soon officers and soldiers began passing on a little joke. The Atlacatl’s designation as a biri, they said, stood not for Immediate Reaction Infantry Battalion, as everyone had thought, but for Immediate Retreat Infantry Battalion. This kind of needling would likely have assured that, eight months later, many officers and soldiers in the Atlacatl would have retained vivid memories of Arambala and El Mozote.

Now, after the initial engagement on Wednesday, according to the guide, “we heard by radio that the other company killed people there.” Under the gaze of Major Cáceres, who was then with the First Company, the troops entered the town of Arambala, brought out the people who had remained there, and assembled them in the plaza. They led the women and children to the church and locked them inside. Then the troops ordered the men to lie face down on the ground, whereupon they bound them, blindfolded them, and began to beat them, demanding information about the guerrillas. A number of men — the guide believes as many as twenty (and his estimate agrees with the figure given in a detailed analysis of the operation in and around El Mozote by Tutela Legal, the San Salvador Archbishopric’s human-rights office, in November, 1991), though other estimates range as low as three — were taken from the assembly, led away, and executed.

In Arambala, the officers still relied on their lists to select who would die. However, by the following afternoon, Thursday, the lists had run out, and at some point — perhaps that day, perhaps late the day before — the officers made a decision about the direction the operation was to take. For, despite Rufina Amaya’s bitter conviction that there had been a “betrayal,” that the officer who had spoken to Marcos Dí­az as he left Gotera had taken part in a nefarious plot to make sure that the people of El Mozote stayed in their homes to await the fate that had been planned for them, an equally likely explanation — and, in a way, a more horrible one — is that the officer was in fact trying to do his friend Dí­az and the people of the hamlet a favor, for at that point nothing whatever may have been planned for them.

Whenever the officers made the decision, it is clear that by the time they reached El Mozote they had ordered a change in tactics. “They had lists from Perquí­n south to Arambala,” the guide told me. “But farther down, there were no lists. Farther down, they killed everything down to the ground. Farther down was scorched earth.”

Just after midnight on Wednesday, as the men of the Atlacatl settled down to sleep, a long column wound its way out of La Guacamaya and snaked slowly through the ravines and gullies, heading west toward the black road. The guerrillas and their entourage travelled quietly: the only sound in the tense darkness was that of hundreds of moving feet. The fighters came first, lugging their rifles and ammunition and whatever other supplies they could manage. Then came the civilian followers, loaded down with their bundles of clothing and sacks of tortillas and coffee, and nervously hushing their children. And at the rear came the men and women of Radio Venceremos, bent under the weight of the transmitter and the generator and the other equipment that formed the station’s heart.

In the end, it was these burdens which betrayed them: the weight slowed them, so that, as they finally came within sight of the black road, struggling along in increasing panic, the darkness thinned and faded, dawn broke behind them, and they could see, as they gazed upward from their hiding place — a ravine full of prickly maguey — the men of the Atlacatl rising and stretching there on the highway. One soldier was swirling his poncho around him to free it of moisture, and the first rays of sunlight glinted off the droplets. The guerrillas had been caught, but turning back was out of the question; there was nothing for it but to run.

“Advance!” Jonás ordered. No one moved. “Advance, I say!”

A handful of guerrillas broke from their cover, zigzagging in a wild, desperate sprint toward the road, staggering under the weight of their equipment. A moment passed before they heard the shouts of the soldiers, and a moment more before the bullets started to come. They took cover and returned fire, then again ran, took cover, and fired; but they were badly exposed, and by the time they had managed to cross a hundred and fifty yards of open country three men had been hit. One of them, Toni, had been carrying the transmitter, and as he collapsed his precious burden slipped from his back and tumbled down, end over end, into another ravine. His comrades gathered around him. Toni was dying; the bullets kept coming; there could be no question of retrieving the transmitter. Monterrosa would have his war prize.

Late that Thursday afternoon, the men of the Atlacatl trudged into El Mozote. They found the streets deserted. For the last two days, the thud of the mortars, the firecracker staccato of the small arms, and the roar of the aircraft had been coming steadily closer, and that morning helicopters and planes of the Salvadoran Air Force had strafed and bombed the area around the hamlet, terrifying the inhabitants. “Everything was closer every day, louder every day,” Rufina Amaya told me, “and finally, by that day, the people were hiding in their houses.”

The strafing ceased not long before the men of the Atlacatl entered the hamlet, dragging with them civilians they had found hiding along the way. Tired and impatient, the soldiers swarmed about the houses of El Mozote and pounded on the doors with the butts of their M16s. “Salgan!” they shouted angrily. “Get out here! Get out here now!”

Hesitantly, the people came out into the twilight, frightened, bewildered, unsure of what was happening. The soldiers, cursing and yelling, pulled them forward, hustled them along with the butts of their rifles, herded everyone into the center of the street. Rufina and her husband, Domingo Claros, emerged with their four children: he was carrying three-year-old Marta Lilián and leading Cristino, nine years old, while Rufina had five-year-old Marí­a Dolores by the hand and carried at her breast Marí­a Isabel, eight months old. “They told us all to lie down in the street, boca abajo” — literally, “mouth down” — “and they began pushing some of us down,” Rufina says. “As my husband was setting the little girl down, a soldier pushed him to the ground. The girl started to cry. By then, all the children were crying.”

The entire town lay like that, perhaps four hundred people face down in the dirt, as darkness fell. Between the wailing of at least a hundred children and the shouting of the soldiers — hundreds had entered the hamlet by now — the din must have been unbearable. The soldiers marched up and down the lines of people, kicking one here and there, striking another with a rifle butt, and all the while keeping up a steady rain of shouted insults and demands. As Rufina tells it, a soldier would stop next to a man or a woman, kick the prone body, and bark out a question: Who were the guerrillas? Where were they? Where did they hide their guns? The men and women of El Mozote insisted that there were no guerrillas there, that they knew nothing of guerrillas or weapons. “If you want to find guerrillas,” one woman shouted tearfully, raising her head from the ground, “go out there” — she waved toward the hills — “outside town. But here, here we’re not guerrillas.”

This only made the soldiers angrier. “All you sons of bitches are collaborators,” an officer said. “You’re going to have to pay for those bastards.”

At one point, as Rufina tells it, the wealthy and influential Marcos Dí­az, lying in the street beside his wife and their sons and daughters, raised his head. “Wait!” he pleaded. “They promised me nothing would happen to the people here. The officer told me so.”

At that, the Atlacatl officer laughed and said, “No, motherfucker, you all have to pay. Now, get your face back in the ground.” And he raised his black boot and pushed Marcos Dí­az’s head down into the dirt.

“They were very abusive,” Rufina says. “We couldn’t do anything. They had all these guns. We had to obey.” Some of the soldiers took down names as others went along the lines demanding to see people’s hands and pulling from their fingers any rings they saw, then ordering them to turn over their jewelry and crucifixes and anything else that might have some value.

The people of El Mozote lay there in the street, their faces in the dust, the children sobbing, for a long time. The soldiers yelled, strode back and forth, aimed their weapons at them. “We thought that they were going to kill us all — that we were sentenced to die right there,” Rufina says.

But finally the soldiers ordered them to get up. As the people of El Mozote climbed unsteadily to their feet, the soldiers barked out an order: they were to go back into their houses, the soldiers said, and not let “even so much as their noses” poke out the door.

The people, terrified, grateful to be alive, hurried into their houses — crowded into them, for virtually every room in the hamlet held extra people. Now the wailing of children made the houses seem smaller still. No one slept. Outside, the men of the Atlacatl shouted and laughed and sang songs, punctuating the hilarity with celebratory bursts of gunfire. Rufina and her husband, packed into a house with two other families, struggled to calm their children. “They were hungry, and we had no food to give them,” she says. “We were going to kill a chicken to feed them, but as soon as we lit a candle the soldiers yelled at us from the street to put out the light. Our children were scared, and hungry, and the littlest ones were messing all over themselves, and we couldn’t even take them outside to clean them.”

So they huddled inside in the darkness, listening anxiously to the laughter, starting up each time it was cut by a burst of automatic fire, and all the while trying to soothe the children. “The saddest thing was that the children were crying and we could do nothing for them,” Rufina says. Soon everything would be all right, their parents assured them — soon they would be safe.

Perhaps the parents began to believe it themselves. After the terror of that evening, after feeling the earth against their faces and the gun muzzles at their necks, Rufina and her husband prayed that they had seen the worst, that the soldiers would leave the next day. “We were thinking that because they hadn’t killed us yet, maybe they wouldn’t,” Rufina says. After all, no one had really been harmed, and, even if the promises of Marcos Dí­az’s officer friend had been worthless — well, the people here had never had trouble with the Army. The people knew that they weren’t guerrillas, and the soldiers, despite their angry shouting, must know it, too.

As the people of El Mozote huddled in their dark houses, down at Osicala, the base camp of the operation, south of the Torola River, the officers were taking stock. The first stage of the operation — the convergence of the Atlacatl companies on El Mozote, the capture of the hamlet and its people — had gone well.

“The first phase was over,” an officer involved in the operation told me. “All the unit commanders came to Osicala to talk it over. I was heading for the mess hall, and I bumped into” — he named a major who at that time was a key figure in military intelligence — “and he said to me, ‘Look, we might need you tomorrow. Be ready.’ ” Then the major gave the younger officer a rundown of the situation. “He said, ‘You know, the first phase is over, the units have gone through and done what they’ve had to do, and now it’s just a question of going in there and interrogating those people’ — you know, like P.O.W.s. I asked him if there had been any guerrillas there, and he said, ‘No, they’re gone. But we might need you. We have people to interrogate. We have maybe six hundred people altogether.’ ”

That was a lot of people to interrogate. “If I had gone in there,” the officer told me, “I would have expected to spend two or three days, considering all the people they had.”

The two men stood there for four or five minutes while the major briefed the younger man on the sort of information they wanted to get out of the prisoners. “Basically, we were looking for the guerrillas’ means of support — how they were getting their food, and so on. We’d stopped a lot of their communications, but we wanted to know their logistics, how they were getting their supplies, what their routes were, and so on. Especially, we wanted to know who it was they’d infiltrated” — into the Army itself — “and who was selling them arms. We had evidence that there was considerable selling of arms from the Army at that time — I mean, you could ask three and four times what a weapon was worth, and these people would pay it, and many of the soldiers couldn’t resist that. There was selling of information as well. All our operations were being leaked. Everyone wanted to make a buck — that was the game.”

Other officers passed by as the two men talked. That the first phase had been completed, that the Atlacatl had seized El Mozote and now held its population prisoner — that much was widely known among the officers at Osicala that night. “My impression was that the plan was to spend the next day interrogating these people,” the officer told me. “And apparently that was the major’s impression as well. But the next day he never called me. And by that night everyone knew that something had happened.”

While it was still dark, the soldiers came to Rufina’s door and began pounding on it with the butts of their rifles. “Salgan!” they shouted once again. “Get out here!” The families were hustled outside. “We wanted to give our children food,” Rufina says, “but they said, ‘No, get out to the plaza.’ ”

All around, the people were emerging from the houses; the soldiers pushed them along roughly, and in the darkness they stumbled over the ground and bumped against one another. “Form lines!” the soldiers shouted. “Men and older boys over here! Women and children over there!”

Soon all the people of El Mozote were lined up in the plaza. The soldiers ordered them not to move. They stood for hours. The children, having had no food and no rest, sobbed and fussed; the mothers tried to quiet them. The soldiers, unlike the evening before, said little. “They just marched up and down the lines looking real mean and ugly, not saying anything,” Rufina says. And so the sun rose on the people of El Mozote that Friday.

Around seven, they heard the sound of a helicopter approaching. As it hovered overhead, the soldiers began herding the people from the plaza. The men were ordered into the church, a small whitewashed building adjacent to the even smaller sacristy; the women and children were crowded into the house of Alfredo Márquez, a small building on the main street a few feet from the larger house of Marcos Dí­az and directly opposite the church and the sacristy.

Looking out a window of the tightly crowded house — well over a hundred women and children had been forced into a space meant for perhaps a dozen — Rufina saw the helicopter touch down in the plaza and half a dozen officers climb out. She saw several of them, accompanied by soldiers of the Atlacatl, stride to the church, where the men were being held. The others came marching to the house where she was, and pushed through the door into the packed, noisy room.

“They had bayonets on their guns, and they used them to push the women back,” Rufina says. “They said we were collaborators. They were angry. They kept asking us where our pistols were, where the men had hidden our guns, and when we kept saying, again and again, that we didn’t have any, they’d push at us with the bayonets. Then they’d say, ‘Shut up, old woman, what are you crying about?’ They said they’d kill us if we didn’t tell them.”

After only a few minutes, the officers strode out, leaving soldiers to guard the door. Around this time, the helicopter lifted off, taking at least some of the officers along.

Now the women began to hear shouting from the church. “We could hear them yelling — the men,” Rufina says. “They were screaming, ‘No! No! Don’t do this to us! Don’t kill us!’ ”

When she heard the screams, Rufina, who together with her children had been sitting on a bench with her back to the front wall of the house — the wall facing the church — climbed up on the bench so that she could look out a small window high up in that wall. Through the window she saw soldiers leading groups of men from the little whitewashed church — blindfolded men whose hands were bound behind them. Each pair of soldiers led five or six men past the house of Alfredo Márquez and took them out of the hamlet in various directions. After a time, she saw her husband in one group, and as she watched, along with young Cristino, who had climbed up next to her, eager to see what was happening, they both saw him — Domingo Claros, twenty-nine-year-old woodcutter, husband of Rufina and father of Cristino, Marí­a Dolores, Marta Lilián, and Marí­a Isabel — bolt forward, together with another man, in a desperate effort to escape the soldiers. But there was nowhere to run. The men of the Atlacatl levelled their M16s and brought both men down with short bursts of fire. Then the soldiers strode forward to where the men lay gasping on the ground, and, unsheathing their machetes, they bent over them, grasped their hair, jerked their heads back sharply, and beheaded them with strong blows to the backs of their necks.

“I got down from the bench and I hugged my children to me,” Rufina says. “My son was crying and saying over and over, ‘They killed my father.’ I was crying. I knew then that they were all being taken away to be killed. I just hugged my children to me and cried.”

While the officers had been questioning the women, other officers and soldiers were interrogating the men in the church. “Many of the men were bound, blindfolded, and forced to lie face down on the ground while they were interrogated,” according to the Tutela Legal report (which was evidently compiled with the coí¶peration of at least one soldier who had been present). “The soldiers would step on their backs and pull their heads back by their hair until they screamed in pain.” For all their brutality, however, the interrogations of the men appear to have been almost as perfunctory as those of the women. The officers devoted scarcely an hour to questioning the hundreds of supposed collaborators, which makes it difficult to believe that they really expected to acquire useful intelligence from the people of El Mozote.

At about eight o’clock, “various of the men who had been gathered in the church were lifted off the ground and decapitated with machetes by soldiers,” according to the Tutela report. “The soldiers dragged the bodies and the heads of the decapitated victims to the convent of the church, where they were piled together.” It must have been at this point that the women in the house across the street began to hear the men screaming.

Decapitation is tiring work, and slow, and more than a hundred men were crammed into that small building. After the initial beheadings — it is unclear how many died inside the church — the soldiers began bringing the men out in groups, and it was from one of the first of the groups that Domingo Claros had attempted to escape.

While Rufina huddled with her children in the crowded house, mourning her husband, other women climbed up on the bench beside her and peered out the small window. From here, they, too, saw the soldiers taking groups of men from the church and marching them off in different directions.

Outside the hamlet, on a hill known as El Pinalito, the guides from Perquí­n waited in the company of several corporals — the officers had ordered them to stay there, lest they become confused with the townspeople during the operation — and throughout the morning the guides watched the soldiers pass. “I saw them marching along groups of maybe ten each,” one guide told me. “They were all blindfolded, and they had their hands tied behind their backs. Then we would hear the shots, the bursts from the rifles.” Out in the forest, the soldiers forced the men to the ground and ordered them to lie flat, with their faces against the earth, as they had lain, with their families, the evening before. Then the soldiers lowered their M16s and fired bursts into each man’s brain.

“All morning, you could hear the shots, the crying and the screaming,” Rufina says. In the house of Alfredo Márquez, some of the children had become hysterical, and no one knew how to calm them. Cristino begged his mother tearfully to take them out of the house, lest they be killed, as he had seen his father killed. Rufina could do nothing but point helplessly to the guards and try to calm him. None of the women had any idea what would happen next. “We just cried and hugged one another.”

Around midday, a group of soldiers came into the house. “Now it’s your turn, women,” one of the soldiers said. They were going to take the women out now in groups, the soldier explained, and then, he said, the women would be free to go to their homes, or down to Gotera, or wherever they liked.

With that, the soldiers began picking out, one by one, the younger women and the girls, and pulling them toward the door. “The girls would hang on to their mothers, and the soldiers would come in and just grab them from their mothers,” Rufina says. “There was a lot of screaming and shouting. Everyone was screaming, ‘No! No! Don’t do this!’ But the soldiers would hit the mothers with the butts of their rifles, and they would reach behind and grab the girls and pull them along with them.”

From the house of Alfredo Márquez, the soldiers marched the group of young women and girls — some of them as young as ten years old — out of the hamlet and up onto the hills known as El Chingo and La Cruz. Before long, the women in the house could hear screams coming from the hills.

The guides, on El Pinalito, nearby, also heard the screaming. “We could hear the women being raped on the hills,” the Perquí­n man told me. “And then, you know, the soldiers would pass by, coming from there, and they’d talk about it. You know, they were talking and joking, saying how much they liked the twelve-year-olds.”

In the midst of this, one or perhaps two helicopters — accounts differ, as they do about many details of the story — touched down in the plaza in front of the church, and a number of officers climbed out. From his vantage point on the hill, the guide says, he recognized the distinctive figure of an officer he had seen several times before: Colonel Jaime Ernesto Flores Grijalba, the commander of the Third Brigade, in San Miguel, who was widely known as El Gordo (the Fat Man). Among the officers accompanying Colonel Flores was one famous figure, a small but charismatic man whom the soldiers of the Atlacatl proudly pointed out to the guide: Lieutenant Colonel Domingo Monterrosa, their beloved commander.

The officers, having been received at their helicopter by Major Cáceres and the company captains, were escorted to a house not far from the church, and disappeared inside. After some time, during which the killings went on around El Mozote — and also in the adjacent hamlet of Tierra Colorada, where patrolling Atlacatl troops had begun shooting people they found hiding in the houses — the officers strolled out onto the common, climbed back into their helicopter, and lifted off from El Mozote.

Around this time, the soldiers returned to the house of Alfredo Márquez. “I was still sitting on the bench with my kids,” Rufina says. “When they came back, they began separating the women from their kids. They pulled the mothers away, leaving the children there crying. They took one group of women and then in a while they came back and took another. That was the saddest thing — little by little, the mothers disappeared, and the house became filled mostly with crying children.”

Rufina found herself in one of the last groups. “It must have been five o’clock. There were maybe twenty of us. I was crying and struggling with the soldiers, because I had my baby on my chest. It took two soldiers to pull the baby from me. So when I came outside into the street, I was the last in the group. I was crying and miserable, and begging God to help me.”

The soldiers marched the women down the main street. They passed the house of Marcos Dí­az on the right and, on the left, that of Ambrosiano Claros, where Rufina and her family had spent the previous night. Ambrosiano Claros’s house was in flames. “I saw other houses burning, and I saw blood on the ground. We turned the corner and walked toward the house of Israel Márquez. Then the woman at the head of the line — we were in single file — began to scream. She had looked through the door and seen the people in the house.”

What the woman had seen was thick pools of blood covering the floor and, farther inside, piles of bloody corpses — the bodies of the women who only minutes before had been sitting in the house with them, waiting.

“The first woman screamed, ‘There are dead people! They’re killing people!’ and everyone began screaming. All down the line, the women began resisting, hugging one another, begging the soldiers not to kill them. The soldiers were struggling with them, trying to push the first women into the house. One soldier said, ‘Don’t cry, women. Here comes the Devil to take you.’ ”

Rufina, still at the end of the line, fell to her knees. “I was crying and begging God to forgive my sins,” she says. “Though I was almost at the feet of the soldiers, I wasn’t begging them — I was begging God. Where I was kneeling, I was between a crab-apple and a pine tree. Maybe that was what saved me. In all the yelling and commotion, they didn’t see me there. The soldier behind me had gone up front to help with the first women. They didn’t see me when I crawled between the trees.”

The crab-apple tree — which still stands, next to the ruin of Israel Márquez’s house, as gnarled and twisted an old crab apple as one can imagine — was within about fifteen feet of the house. “I couldn’t move, couldn’t even cry,” Rufina says. “I had to remain absolutely still and silent. The whole group was still outside the house — the women grabbing one another and hugging one another and trying to resist. Soon, though, the soldiers pushed some of them into the house. I couldn’t see inside, but I started hearing shots and screams.”

Finally, when the screams and the gunfire had stopped, some of the soldiers went off. A few minutes later, they returned, pushing along the last group of women, and now Rufina heard the sequence — the cries of terror, the screaming, the begging, and the shooting — all over again. After a time, those sounds ceased. In the sudden silence, scattered shooting and fainter screams could be heard echoing from the hills. A few feet from where Rufina lay hidden behind the tree, nine or ten soldiers laid down their guns and collapsed wearily to the ground.

“Well, all these old bastards are dead,” one said to somebody farther off. “Go ahead and burn the house.”

It was growing dark, and soon flames were rising from the house of Israel Márquez, highlighting the soldiers’ faces and the trunk of the tree. It grew so hot that Rufina began to fear that the tree would catch and she would be forced to run. She had remained perfectly still, hardly daring to breathe, and her legs had begun to fall asleep. And the soldiers, still close enough to touch, remained where they were, smoking cigarettes and watching the fire.

“We’ll just stay here and wait for the witches of Mozote to come out of that fire,” one said.

The soldiers watched the fire and talked, and Rufina, frozen in her terror a few feet away, listened. “Well, we’ve killed all the old men and women,” one said. “But there’s still a lot of kids down there. You know, a lot of those kids are really good-looking, really cute. I wouldn’t want to kill all of them. Maybe we can keep some of them, you know — take them with us.”

“What are you talking about?” another soldier answered roughly. “We have to finish everyone, you know that. That’s the colonel’s order. This is an operativo de tierra arrasada here” — a scorched-earth operation — “and we have to kill the kids as well, or we’ll get it ourselves.”

“Listen, I don’t want to kill kids,” the first soldier said.

“Look,” another said. “We have orders to finish everyone and we have to complete our orders. That’s it.”

At about this time, up on the hill known as El Pinalito, Captain Salazar was shrugging off a guide’s timid plea for the children’s lives. “If we don’t kill them now,” he said angrily, “they’ll just grow up to be guerrillas. We have to take care of the job now.”

Meanwhile, the soldiers sat and gazed at the burning house. Finally, one stood up. “Well, no witches came out,” he said. “There are no witches. Let’s go see what kind of food they have in that store.”

With that, the other men got to their feet, picked up their rifles, and trudged off. A few minutes later, Rufina could hear, from the store of Marcos Dí­az, “bottles clinking — you know, as if they were drinking sodas.”

The fire was still burning furiously, but the big crab-apple tree, which some miracle had kept from igniting, shielded Rufina from the heat. Over the crackling of the fire she could still hear, coming from the hill called La Cruz, the screams of the girls. Now and again, she heard a burst of gunfire.

After a time, when the soldiers seemed to have finished drinking their sodas, Rufina heard crying and screaming begin from the house of Alfredo Márquez: the screaming of the children. “They were crying, ‘Mommy! Mommy! They’re hurting us! Help us! They’re cutting us! They’re choking us! Help us!’

“Then I heard one of my children crying. My son, Cristino, was crying, ‘Mama Rufina, help me! They’re killing me! They killed my sister! They’re killing me! Help me!’ I didn’t know what to do. They were killing my children. I knew that if I went back there to help my children I would be cut to pieces. But I couldn’t stand to hear it, I couldn’t bear it. I was afraid that I would cry out, that I would scream, that I would go crazy. I couldn’t stand it, and I prayed to God to help me. I promised God that if He helped me I would tell the world what happened here.

“Then I tied my hair up and tied my skirt between my legs and I crawled on my belly out from behind the tree. There were animals there, cows and a dog, and they saw me, and I was afraid they would make a noise, but God made them stay quiet as I crawled among them. I crawled across the road and under the barbed wire and into the maguey on the other side. I crawled a little farther through the thorns, and I dug a little hole with my hands and put my face in the hole so I could cry without anyone hearing. I could hear the children screaming still, and I lay there with my face against the earth and cried.”

Rufina could not see the children; she could only hear their cries as the soldiers waded into them, slashing some with their machetes, crushing the skulls of others with the butts of their rifles. Many others — the youngest children, most below the age of twelve — the soldiers herded from the house of Alfredo Márquez across the street to the sacristy, pushing them, crying and screaming, into the dark tiny room. There the soldiers raised their M16s and emptied their magazines into the roomful of children.

Not all the children of El Mozote died at the sacristy. A young man now known as Chepe Mozote told me that when the townspeople were forced to assemble on the plaza that evening he and his little brother had been left behind in their house, on the outskirts of the hamlet, near the school. By the next morning, Chepe had heard plenty of shooting; his mother had not returned. “About six o’clock, around ten soldiers in camouflage uniforms came to the house,” Chepe says. “They asked me where my mother was. I told them she had gone to the plaza the night before. I asked them if I could see my mother, and they said I couldn’t but I should come with them to the playing field” — near the school. “They said when we got there they would explain where my mother was.”

Carrying his little brother, Chepe went with the soldiers and walked along with them as they searched house to house. “We found maybe fifteen kids,” he says, “and then they took us all to the playing field. On the way, I heard shooting and I saw some dead bodies, maybe five old people.” When they reached the playing field, “there were maybe thirty children,” he says. “The soldiers were putting ropes on the trees. I was seven years old, and I didn’t really understand what was happening until I saw one of the soldiers take a kid he had been carrying — the kid was maybe three years old — throw him in the air, and stab him with a bayonet.

“They slit some of the kids’ throats, and many they hanged from the tree. All of us were crying now, but we were their prisoners — there was nothing we could do. The soldiers kept telling us, ‘You are guerrillas and this is justice. This is justice.’ Finally, there were only three of us left. I watched them hang my brother. He was two years old. I could see I was going to be killed soon, and I thought it would be better to die running, so I ran. I slipped through the soldiers and dived into the bushes. They fired into the bushes, but none of their bullets hit me.”

Lying amid the maguey that night, Rufina Amaya heard the chorus of screams dwindle to a few voices, and those grew weaker and weaker and finally ceased. She heard the officers order that fire be put to the house of Alfredo Márquez and the church and the sacristy, and from the maguey she saw the flames rise and then she heard faint cries start up again inside the buildings and the short bursts of gunfire finishing off a few wounded, who had been forced by the flames to reveal that they were still alive.

Soon the only sounds were those which trickled down from the hills — laughter, intermittent screams, a few shots. On La Cruz, soldiers were raping the young girls who were left. On El Chingo and El Pinalito, other soldiers busied themselves making camp. Down in the hamlet, a few troops walked about here and there, patrolling. Not far from the still burning house of Israel Márquez, two soldiers halted suddenly, and one of them pointed to the patch of maguey. He lowered his rifle and fired, and after a moment his companion fired, too. In the patch of brush, the stream of bullets sent a dark-green rain of maguey shreds fluttering to the earth. Then the soldiers charged forward and began poking among the weeds.

“She was right here,” one said, pulling at some maguey. “I saw her, I know it.”

Up on the hills, the soldiers listened to the shots, exchanged glances, and waited. Then they went on with what they had been doing: watching the flames rise from the burning houses and talking quietly among themselves, telling tales of the day’s work.

They spoke wonderingly about the evangelicals, those people whose faith seemed to grant them a strange power.

“They said maybe some of the people believed in God so strongly that they just delivered themselves up, they didn’t resist,” the guide told me. “They said some of the people were singing even as they were killed.”

There was one in particular the soldiers talked about that evening (she is mentioned in the Tutela Legal report as well): a girl on La Cruz whom they had raped many times during the course of the afternoon, and through it all, while the other women of El Mozote had screamed and cried as if they had never had a man, this girl had sung hymns, strange evangelical songs, and she had kept right on singing, too, even after they had done what had to be done, and shot her in the chest. She had lain there on La Cruz with the blood flowing from her chest, and had kept on singing — a bit weaker than before, but still singing. And the soldiers, stupefied, had watched and pointed. Then they had grown tired of the game and shot her again, and she sang still, and their wonder began to turn to fear — until finally they had unsheathed their machetes and hacked through her neck, and at last the singing had stopped.

Now the soldiers argued about this. Some declared that the girl’s strange power proved that God existed. And that brought them back to the killing of the children. “There were a lot of differences among the soldiers about whether this had been a good thing or whether they shouldn’t have done it,” the guide told me.

As the soldiers related it now, the guide said, there had been a disagreement outside the schoolhouse, where a number of children were being held. Some of the men had hesitated, saying they didn’t want to kill the children, and the others had ridiculed them.

According to one account, a soldier had called the commanding officer. “Hey, Major!” he had shouted. “Someone says he won’t kill children!”

“Which son of a bitch says that?” the Major had shouted back angrily, striding over. The Major had not hesitated to do what an officer does in such situations: show leadership. He’d pushed into the group of children, seized a little boy, thrown him in the air, and impaled him as he fell. That had put an end to the discussion.

Now, up on the hills, the soldiers talked and argued and watched the burning houses, while the two men down below still searched among the maguey, cursing at the sharp thorns.

“I know she was here,” the first soldier said. “I saw her. She was right here.”

“No, no,” his companion finally said. “There’s no one here. You’re just seeing the dead. You’re seeing ghosts. The ghosts of the people you killed are frightening you.” With that, the soldiers looked at each other, then turned and trotted back to the center of the hamlet. Amid the maguey, Rufina Amaya closed her eyes, remained motionless. After a time, she reached out a hand and began groping about in the weeds, slowly pulling the thorny strips to her, gathering them into a pile and heaping them over her body.

She lay there still when the stars began to disappear from the lightening sky. She heard sounds of movement from the hills, rising voices as the men woke, urinated, ate, prepared their equipment. Shots echoed here and there, interspersed with the barking and howling of dogs and the lowing of cows as the soldiers killed the animals one by one. From up on La Cruz came a burst of high-pitched screaming and begging, followed by a prolonged chorus of gunfire, and, at last, silence. And then the men of the Atlacatl, having completed the operation in El Mozote, moved out.

Hours earlier, when the chill of the night came on, Rufina Amaya had shivered, for the maguey had badly ripped her blouse and skirt. The thorns had torn the flesh of her arms and legs, but at the time she hadn’t noticed. Now she could feel the cuts, swelling and throbbing, and the blood, dried and prickly, on her limbs. And as she lay sobbing amid the thorns, listening to the soldiers pass, her breasts ached with the milk that had gathered there to feed her youngest child.

Marching past the church, which was burning still, past the carcasses of cows and dogs, and out of El Mozote, the men of the Atlacatl did not see the dark shape in the maguey patch, the heap of dark-green leaves. Their minds were on their work, which on that Saturday morning in December lay ahead in the hamlet of Los Toriles.

In Los Toriles, “the soldiers pulled people from their houses and hustled them into the square,” the guide told me, “and went down the line taking money and anything of value out of people’s pockets. Then they just lined the people up against a wall and shot them with machine guns. The people fell like trees falling.”

Even so, the killing in Los Toriles took much of the day. Some of the residents, having seen the columns of smoke rising the afternoon before from El Mozote, had fled their homes and hidden in caves above the hamlet. But most had stayed, wanting to protect their homes: they remembered that on a previous operation soldiers had set fire to houses they found empty, claiming that they belonged to guerrillas.

By afternoon, the streets of Los Toriles were filled with corpses. “It was so terrible that we had to jump over the dead so as not to step on them,” the guide told me. “There were dogs and cows and other animals, and people of all ages, from newborn to very old. I saw them shoot an old woman, and they had to hold her up to shoot her. I was filled with pity. I wished we had gone out and fought guerrillas, because to see all those dead children filled me with sadness.”

As night fell, the soldiers walked through the town setting fire to the houses. It was dark by the time they left Los Toriles, to march south toward the guerrilla stronghold of La Guacamaya. They made camp in open country, rose at dawn, and, as they prepared to move out again, Captain Salazar motioned them over. The men of the Atlacatl gathered in a circle, sitting cross-legged on the ground as he stood and addressed them.

“Señores!” the Captain said angrily. “What we did yesterday, and the day before, this is called war. This is what war is. War is hell. And, goddammit, if I order you to kill your mother, that is just what you’re going to do. Now, I don’t want to hear that, afterward, while you’re out drinking and bullshitting among yourselves, you’re whining and complaining about this, about how terrible it was. I don’t want to hear that. Because what we did yesterday, what we’ve been doing on this operation — this is war, gentlemen. This is what war is.” And for perhaps half an hour the Captain went on speaking in his angry voice, and the men shifted uneasily.

“There had been a lot of talk about whether it was right,” the guide said, “and this had clearly got back to the Captain.” Finally, the tirade over, the men got to their feet. Soon they were marching south again.

Late that afternoon, they reached La Guacamaya. They found nothing there but dead animals; the guerrillas had long since departed. The soldiers spent two nights there, resting and cleaning their equipment. Helicopters landed, bringing Colonel Flores and other top officers, who met with the Atlacatl officers for “evaluation and coí¶rdination.” The operation was now winding down.

“It was a walk-through by then, a joke,” an officer in another unit told me. “The guerrillas were long gone, and everybody knew it.”

On the second morning, the men of the Atlacatl marched west, heading for the black road. On their way, they passed the hamlet of La Joya. “Everything was dead there — animals and people all mixed together,” the guide said. “Vultures were everywhere. You couldn’t stand to be there, because of the stink.”

Above the hamlet, in the caves and ravines and wooded gullies, those who had managed to escape the troops shivered and waited, and tried to keep their children still. Some had left their homes before the soldiers came; others had managed to flee when men from the Atlacatl, on the day some of their comrades were “cleansing” El Mozote, stormed La Joya. “Suddenly, there was shooting and explosions all over,” Andrea Márquez, who had been twenty years old at the time, said. “We didn’t even see the soldiers at first. There were bullets flying everywhere. I grabbed my little girl — she was one and a half — and put her on my back, and we started crawling through the brush with bullets flying and explosions all around.” She showed me an ugly scar from a shrapnel wound on her knee. “We crawled and then we ran and ran, and after a while my baby made sounds as if she were thirsty, and I pulled her around and then I saw there was a wound in her head, and I realized I was covered with blood.”

No one else was around — the people had scattered at the soldiers’ assault — and Andrea Márquez was too terrified to go back toward La Joya. Holding her child in her arms, she climbed higher into the mountains, found a cave, and tried to care for her daughter’s wound with leaves and with water from a stream. Eight days later, she found a stick and dug a hole and buried her little girl. Then, delirious with grief and shock and terror, she wandered high into the northern mountains.

Months later, the surviving villagers, those few who remained in Morazán, began to murmur fearfully to one another that a witch had come to haunt the mountains — a savage woman, who could be glimpsed from time to time late at night by moonlight, naked but for her waist-length hair, as she crouched by a stream and stripped the flesh from a wriggling fish with long, sharp fingernails. The villagers were frightened of her, for they knew that it was after the matanza, the great killing of El Mozote, that the witch had come to haunt the mountains.

As the tide of soldiers ebbed from northern Morazán, the guer- rillas flowed back in. “We knew there would be killing, but we never expected what we found,” said Licho, who was with one of the first units to return. “It was desolation, total desolation — not a person alive, not an animal alive, not a house that hadn’t been burned. There were bodies in the houses, bodies in the fields, bodies in the wells.”

The guerrillas immediately sent reports of the killing to their commanders; but there was a problem. “The comandancia didn’t believe us — they didn’t believe the numbers,” Licho said. “So we began to count. We sent units all over looking for bodies. A lot of them were not in the houses — they were lying out in the grass, in the fields, in the woods. We sent three reports up to the comandancia, and finally they sent other people down to the zone, because they still couldn’t believe the numbers.”

Survivors were straggling back from the caves and mountain gullies to find the plazas of their ruined villages so thick with vultures that, in the words of one man I talked with, “they seemed covered by a moving black carpet.” People wept, mourned, and, when they could, buried their dead. Pedro Chicas, who had hidden in a cave above La Joya, returned to the hamlet to find “everything burned, everything dead — corpses everywhere in the street,” he said. “Everything was dead — cows, horses, chickens, pigs. We couldn’t do anything with the badly charred people, but the others we buried.”

As the survivors returned to the hamlets around El Mozote, people from the zone were making contact with representatives of Socorro Jurí­dico (which was then the human-rights organization of the Archbishopric of San Salvador). Roberto Cuellar, a Socorro Jurí­dico official at the time, remembers hearing from “members of church groups, and just people there, you know, neighbors.” Within days — it is unclear how many days: limited and sometimes conflicting memories make this sequence particularly difficult to reconstruct — Cuellar telephoned the Reverend William L. Wipfler, who was the director of the human-rights office of the National Council of Churches, in New York. “Cuellar said the Atlacatl Brigade had committed a massacre in a town called El Mozote and in another called La Joya, and that he thought there might be hundreds killed, and nothing had been cleaned up yet,” Wipfler says. “Socorro had an eyewitness account — it had got two people into El Mozote.”

Wipfler immediately sent a telegram to Ambassador Hinton, asking “confirmation or otherwise” of “reliable reports received here [that] indicate that between December 10 and 13 joint military and security forces operation took place in Morazán Department which resulted in over 900 civilian deaths.” He also telephoned officials at Amnesty International and other leading human-rights agencies in New York and Washington, and left a message for Raymond Bonner, at the New York Times’ bureau in Mexico City. As Wipfler remembers it, Cuellar’s call came no later than December 20th and probably earlier. (The telegram to Hinton, sent under the name of the Reverend Eugene Stockwell, Wipfler’s boss, has been dated December 15th, only four days after the massacre, but there is a possibility that it was actually sent a few days later.)

On December 17th, an exhausted Santiago staggered into a guerrilla camp at Jucuarán — a town fifty miles south of La Guacamaya — along with the other Venceremos people, and there, he says, he found waiting for him a radio message from Morazán: “The Atlacatl Battalion massacred a thousand peasants in various hamlets and villages.” If Santiago’s memory of the date is accurate, then this number, arrived at less than a week after the killing, could only have been a very rough estimate; soldiers still occupied La Guacamaya and the area of El Mozote, and the guerrillas could have made no precise count. But after days of counting in some of the stinking hamlets, and the compilation, with the help of survivors, of partial lists of names, the comandantes had finally been forced to believe that many hundreds had died, and they had apparently settled on the round number of a thousand. And now they wanted Santiago and the others back, with a new transmitter that had been awaiting them in Jucuarán, so they could begin to make the world believe as well.

After five days of all-night marches, the small Venceremos crew trudged into the ravine at El Zapotal. It was noon on December 24th. On Christmas Eve, according to his memoirs, Santiago was able to take to the airwaves and tell the world that Radio Venceremos had been reborn — and to announce that during its two weeks of silence a great killing had taken place in northern Morazán. It was the inauguration of an ambitious propaganda campaign, which gathered steam steadily through December and January, and into February. The propaganda was based on truth, which is supposedly the most effective kind, but the Salvadoran government and, later, the American government would skillfully use the fact that it was propaganda — and particularly the fact that the number of dead seemed to increase with each broadcast — to undermine its truth.

On December 29th, the guerrillas stormed the Army detachments that had been left to occupy some of the hamlets in the zone, including at La Guacamaya and at or near El Mozote itself. “The attack on La Guacamaya became a major bloodbath,” Villalobos, the guerrilla comandante, told me. “The lieutenant in charge knew he was in a hopeless position, but he refused to surrender — probably because he knew what had happened at El Mozote, and feared reprisals. We annihilated his position, and he died in combat. We buried him in his uniform to honor him.”

The guerrillas took seven prisoners in the operation, and two of them were made use of a few days later in what Santiago calls the “information battle to denounce the genocide.” Each prisoner described what he had seen in Cerro Pando, a village three miles south of El Mozote. “I expected to see dead, because we had talked to troops who had already been out and they said they had killed many guerrillas,” a sergeant said. “Then we looked in the houses … and I realized that it wasn’t the way they said, because I saw dead children, and the mother was hugging one of her children — I think it was her youngest son.”

Santiago himself now visited El Mozote with his “mobile unit,” and broadcast a description of the devastation, saying that “it looks as if a cyclone had passed through”; that from the ruins of the sacristy came “a penetrating odor that indicated that beneath … were to be found who knows how many cadavers of the people of El Mozote”; and that in the shattered building he could see “macabre scenes, hunks of human hair, and fingers amid the rubble.”

Late in December, the F.M.L.N. got in touch with Raymond Bonner, of the Times, and informed him that his long-standing request to visit guerrilla-held territory would be approved; he would be welcome to come to Morazán in early January. Also around that time, a guerrilla patrol stumbled upon some campesinos cowering in a ravine, and discovered among them a near-hysterical woman of thirty-eight, whose legs and arms and face were scored with cuts. The peasants said that they had come upon her near a river — found her crouched there nearly naked, her limbs and body smeared with blood and covered with thorns. “I could hardly speak,” Rufina Amaya recalls. “I talked and cried, talked and cried — couldn’t eat, couldn’t drink, just babbled and cried and talked to God.”

Now the guerrillas had found her, and they rejoiced when they realized who she was. “They were all happy that there was at least one survivor,” Rufina says. “They all came up around me and hugged me. I didn’t know what was going on, who they were, what they wanted.” She was taken to El Zapotal and interviewed, and before long the voice of Rufina Amaya, telling in careful detail the story of what had happened in El Mozote, was broadcast throughout El Salvador.

On December 31st, the General Command of the Morazán front of the F.M.L.N. issued “a call to the International Red Cross, the O.A.S. Human Rights Commission, and the international press to verify the genocide of more than nine hundred Salvadorans” in El Mozote and the surrounding hamlets. “We ask these organizations to be the eyes of the world’s conscience,” the comandantes said.

That night, at the same time that Radio Venceremos was broadcasting a Mass “in memory of the thousand massacred,” El Salvador’s provisional President, José Napoleón Duarte, felt obliged to take to the airwaves and deny the accusations personally. The entire massacre story, he said, was “a guerrilla trick” meant to smear his government at the very moment when the United States Congress was considering aid to El Salvador.

Duarte was right in at least one respect: though the El Mozote controversy appeared to center on what had happened in a handful of hamlets in a remote region of El Salvador, the real point of focus had shifted to Washington — and, in particular, to Congress, which was perceived as the weak spot in the armor of the Salvadoran government. It was Congress that voted the money that paid for the American guns and helicopters and military advisers; and in recent years, as the atrocities had grown ever more frequent, Congress had done so with increasing reluctance. Two days before Duarte’s speech, Reagan had signed Congress’s amendment of the Foreign Assistance Act of 1961, which required the President to “certify” that the Salvadoran government “is making a concerted and significant effort to comply with internationally recognized human rights” and “is achieving substantial control over all elements of its own armed forces, so as to bring to an end the indiscriminate torture and murder of Salvadoran citizens by these forces.” If such a certification was not delivered to Congress by January 29th, and convincingly defended, all funds and assistance for El Salvador would be immediately suspended.

Now all sides prepared for the debate over certification, which would provide concerned congressmen, church leaders, heads of human-rights groups, and others with a new opportunity to document the abuses committed by the Salvadoran government in prosecuting the war. Administration officials, meanwhile, both in Washington and in the Embassy in San Salvador, prepared to defend the government and demonstrate that, despite appearances, the Salvadorans were improving in their respect for human rights. Many of these officials viewed the certification requirement with singular contempt.

“If Congress felt so strongly about human-rights abuses, it could have simply cut off aid,” Elliott Abrams, who had just been sworn in as Assistant Secretary of State for Human Rights and Humanitarian Affairs, told me. “But Congress didn’t cut off aid, because it didn’t want to risk being blamed, if the guerrillas won as a result, for ‘losing’ El Salvador. Instead, they required certification — which is to say, they agreed to fund the war while reserving the right to call us Fascists.” Deane Hinton, the United States Ambassador at the time, later told an interviewer that he viewed certification “as a way for the Congress … to be for and against something at the same time.” He went on to say that congressmen “didn’t want to take the responsibility to deny resources to the government of El Salvador and on the other hand they didn’t want to endorse it, so they created a certification procedure and made the rest of us jump through the hoop, and the President had to certify it,” and added, “It is a political cop-out by a lot of congressmen.”

At the root of this “political cop-out,” in the unspoken view of many Administration officials, was a simple truth: that when it came down to either supporting the Salvadoran government, however unseemly its methods, or allowing a victory by the guerrillas the choice was clear — and the only difference between the people in the Administration and the hypocrites in Congress was that the Reagan officials were not afraid to say it straight out. Abrams told me, “I used to say to people, ‘I mean, I can see arguing for an F.M.L.N. victory on political grounds or economic grounds — but on human-rights grounds? I mean, that’s crazy.’ ” Abrams stood the human-rights argument on its head, contending that to argue for an aid cutoff was, in effect, to argue for a guerrilla victory, and that at the end of the day, however badly the Salvadoran government behaved, those collective atrocities could never approach the general disaster for human rights that an F.M.L.N. victory would represent.

Abrams was not alone in taking this line of argument, which appears to have been aimed at persuading those — conservative Democrats most prominently — who, however much they deplored human-rights abuses in El Salvador, nonetheless worried about taking the blame for any advance of Communism in the hemisphere. The day after the certification was delivered to Congress, the State Department sent out a cable, over Secretary of State Haig’s name, urging American diplomats to describe the El Salvador policy as “a grit-your-teeth policy: to support a reformist junta, with a lot of bad eggs in and around it, in order to avoid a Somoza-Sandinista choice. For critics to narrow their focus to the teeth-gritting without considering the policy’s larger aims is shallow and unfair.” For those who “can’t take” the current Salvadoran government, Haig wrote, “the honest response is not to say the junta is — surprise — beset and flawed, but rather to make the case that it’s acceptable to the United States if El Salvador goes the Cuban way.”

It was against this background that Ambassador Hinton and the State Department began receiving reports about a massacre in Morazán. “Coming on top of everything else, El Mozote, if true, might have destroyed the entire effort,” Thomas O. Enders, the Assistant Secretary of State for Inter-American Affairs, told me. “Who knows? I certainly thought that when I first heard about it.”

The Embassy began a counterattack, following a pattern that it held to throughout: undermining the reports not by investigating the facts but by casting doubt on their source. In response to the request from the Reverend Wipfler, of the National Council of Churches, Hinton cabled back, on January 8th, that he did “not know what your sources are but the only sources that I have seen alleging something like this are clandestine Radio Venceremos reports.” The Ambassador then quoted in full a heavy-handed Venceremos text from January 2nd, in which El Mozote was compared repeatedly to My Lai; added that he “found it interesting” that a guerrilla communiqué two days earlier had not mentioned El Mozote at all; and concluded that he did not “consider Radio Venceremos to be a reliable source.”

Hinton — who declined to be interviewed for this account — must have known that the National Council of Churches could not have got its information from Radio Venceremos. (Among other things, Wipfler’s cable was sent before the station had resumed broadcasting.) He himself had probably already received reports from sources of his own that something had happened in Morazán; after all, no fewer than ten American advisers were working with the Atlacatl at the time. According to one of them, members of the Milgroup — the Military Advisory Group at the Embassy — had telephoned the Atlacatl base in La Libertad within a few days of the massacre. “They called up and talked to the Special Forces people and told them they wanted Monterrosa to come in — they wanted to talk to him about something that had happened during the operation,” this adviser told me. “But Monterrosa just climbed into his helicopter and said, ‘If they want to talk to me, I’ll be out with my troops.’ He wasn’t going to go in and talk to those guys. He said, ‘If I go in and let them talk to me about this thing, I’ll never be able to get anyone to go out there and fight for me again.’ And then he got into his helicopter and took off” — heading back to Morazán.

How had the Milgroup officers heard so quickly that “something had happened” in Morazán? Although the adviser believes it was the guerrillas who got word to the Embassy, a number of highly placed Salvadorans, including one prominent politician of the time who had many friends among senior officers, claim that two American advisers were actually observing the operation from the base camp at Osicala. On its face, the charge is not entirely implausible — American advisers had been known to violate the prohibition against accompanying their charges into the field — but it is impossible to confirm. Colonel Moody Hayes, who was then the Milgroup commander, refused to discuss El Mozote with me, explaining that he didn’t know “what might still be classified,” while officers from the defense attaché’s office and from Milgroup who were willing to talk generally dismissed the charge as unfounded. State Department officials, however, were clearly worried about the possibility. “Certainly, one of the issues I remember raising between us and the Embassy was, Were there any American advisers on this sortie?” Enders said. “The Embassy made a great effort to talk to advisers who were with the Atlacatl to try to find out the truth.”

Of course, had the truth been that Americans were at Osicala, it would have been a very hard truth to make public — or, for that matter, to confide to a superior. The officers involved would surely have known, as Enders conceded, that admitting such an unfortunate misjudgment “would have ruined those guys’ careers — they would have been cashiered. So no one’s going to volunteer, ‘Hey, I was up there.’ ” By the same token, Embassy officials would have been acutely aware of the effect such a revelation would have had on the entire American effort in El Salvador. “It would have been devastating,” Enders said. “American advisers with a unit that committed an atrocity? Devastating. Can you imagine anything more corrosive of the entire military effort?” Coming at such a time, it might well have made the Salvadoran war, in Enders’ expression, “unfundable.” Enders insists, however, that “given the small numbers of advisers involved, if they’d been there we would have known about it.”

Sometime in mid-December of 1981, “contacts of great confidence on the left” approached Todd Greentree, the junior reporting officer, and told him that a massacre had taken place in Morazán. Greentree knew of Opera- tion Rescue from the defense attaché’s reports but knew nothing about a massacre. “I first heard about it from the left,” Greentree, who is now the Nicaragua desk officer at the State Department, says. “The most important thing was that they offered me a special safe-conduct to go up there and see it for myself. Obviously, a decision had been taken very high up in the F.M.L.N. to do this for propaganda purposes. I knew the guerrillas would never have masqueraded something like this, would never have fabricated it, if they were offering safe-conduct. I was convinced that something had gone on, and that it was bad. I mean, it was pretty clear, if they were going to do this, that something must have happened.”

Greentree conveyed this message from the left to Ambassador Hinton. A meeting was held. “His response was ‘No, you can’t do this under guerrilla escort,’ ” Greentree says. ” ‘That would be too risky, and you would just be playing into their hands.’ I mean, I should emphasize that I never got the feeling that they just wanted this to go away. But there were political and military constraints that we were operating under.”

Kenneth Bleakley, who was the deputy chief of mission, told me “Todd was a very courageous young officer, but it was just too much of a risk to send somebody out there.”

The decision was clearly the Ambassador’s to make. Peter Romero, who was an El Salvador specialist at the State Department, says, “However much we might have wanted more information, no one in State was going to make that call. It was clearly the Ambassador’s call. And at the time, basically, the Embassy staff down there were targets — they were targeted by the F.M.L.N.”

Greentree was unable to accept the guerrillas’ offer to visit El Mozote and have a look for himself. But, as he soon learned, two other Americans were about to do just that.

Late on the evening of January 3rd, in the mountains near the Salvadoran border, a dusty car pulled to a stop and disgorged into the barren Honduran landscape two Americans in hiking boots. They slung their backpacks on the ground, stretched, and after a few moments of searching found a boy who had been waiting for them — their F.M.L.N. contact. The boy led them into the quiet darkness, heading down a rocky trail to the bank of a river. In the moonlight, the three stripped and, holding their clothing and their packs above their heads, picked their way unsteadily through the rushing cold water until they reached the far shore — and the border of the guerrilla-held Department of Morazán.

“I was scared,” Raymond Bonner said. “All I could think was, The military, what if the military ambushes us?”

It was not an idle fear; Bonner, of the New York Times, and the photographer Susan Meiselas — together with Alma Guillermoprieto, of the Washington Post, who followed them into Morazán ten days later — would be the first members of the American establishment press to report on the Salvadoran war from the guerrilla side.

Bonner and Guillermoprieto had both been working hard for months to arrange a trip in, lobbying through F.M.L.N. contacts in Mexico and New York as well as in El Salvador. In early December, they had finally seen their trips confirmed, only to have them cancelled after the start of Operation Rescue. Later that month, Bonner’s contacts had informed him that the trip was on again. “To the charge that the guerrillas took us in to report on the massacre, I’d say now, ‘You’re damn right they did,’ ” he said. “But at the time I didn’t really know about the massacre.”

Bonner telephoned Meiselas, who was in New York, and, in a magnanimous gesture, before flying to Tegucigalpa also put in a call to Guillermoprieto, in Mexico City, who immediately resumed her own “desperate, intense, round-the-clock phone lobbying” with her F.M.L.N. contacts. Ten days later, after rendezvousing with a guerrilla contact in the Tegucigalpa market, she found herself being deposited “under a bush in the middle of the night” near the Honduran border, along with a pile of supplies.

Bonner and Meiselas, and Guillermoprieto, describe the trip in the same way: hiking all night through the moonlit mountains, and at dawn coming upon the first guerrilla camp — a scattering of tents, under pine trees, that held twenty-five or thirty people. By dawn on the third day — January 6th — Bonner and Meiselas had reached the area of El Mozote. “There were bodies and parts of bodies,” Meiselas said. “We saw about twenty-five houses destroyed around Arambala and El Mozote. My strongest memory was this grouping of evangelicals, fourteen of them, who had come together thinking their faith would protect them. They were strewn across the earth next to a cornfield, and you could see on their faces the horror of what had happened to them.”

At a burial near El Zapotal, they were introduced to Rufina Amaya, and Bonner interviewed her at length. A few days later, the guerrillas gave him a handwritten list, which they said contained the names of those who had died at El Mozote and in the surrounding hamlets. “I did the tally, came up with the number seven hundred, tried to get the number of men, women, and children, got a sample of names,” Bonner said.

A few days later, Bonner and Meiselas began the hike back to Honduras. At the middle camp, they met a battered Guillermoprieto, one of whose legs was swollen from an accident involving a rock and a mule. At just about the time Bonner reached Mexico City and began to file his stories, Guillermoprieto was nearing El Mozote.

“We started smelling it from Arambala,” she said. “These kids started leading me down paths and pointing to houses and saying again and again, ‘Aquí­ hay muertos, aquí­ hay muertos.’ The most traumatizing thing was looking at these little houses where whole families had been blown away — these recognizable human beings, in their little dresses, just lying there mummifying in the sun. We kept walking, got to El Mozote. We walked down these charming and beautiful roads, then to the center of town, where there was this kind of rubbly place” — the sacristy — “and, in it, a stupefying number of bones. There was a charred wooden beam lying on top of the bodies, and there were bones sticking up, and pieces of flesh. You could see vertebrae and femurs sticking out. No attempt had been made to bury the bodies.”

In some shock, she was led to La Guacamaya. “Everyone there had lost someone in his family — everyone — and everyone was in a state of controlled hysteria.”

The great exodus that had begun with the offensive in mid-December was still under way. “It was that massacre, the most horrible, that really caused the glass of water to overflow,” Licho told me. “People flowed out of the zone, either toward Honduras or south toward Gotera or into the guerrillas. A lot of people joined us as combatants then.”

At the urging of Jonás, the guerrilla commander, Guillermoprieto saw Rufina the next day. Later, she spoke to two young men who had seen their families murdered in La Joya. Then, thinking of Bonner and his head start, she scribbled her story in her notebook, folded up the pages, and hid them in a plastic film cannister. She found a guerrilla courier and persuaded him, with some difficulty, to carry the precious cargo to Tegucigalpa and deliver it to a colleague, who could telephone the story in to the Post.

On January 26th, the day Guillermoprieto got back to Tegucigalpa, the Times ran Bonner’s first story from Morazán, headlined “with salvador’s rebels in combat zone.” Guillermoprieto had already been on the telephone to the Post’s foreign editor, and they managed to get her El Mozote story, along with a Meiselas photograph of the rubble of the sacristy, onto the front page of the first edition of the next day’s paper — eighteen hundred words, headlined “salvadoran peasants describe mass killing; woman tells of children’s death.” Editors in the Times’ Washington bureau, seeing the piece in the Post’s early edition, telephoned New York, where Bonner’s El Mozote story had been awaiting editing at the foreign desk. Craig Whitney, then the deputy foreign editor, and the deskmen managed to rush Bonner’s slightly shorter article, headlined “massacre of hundreds reported in salvador village,” into the paper’s late edition. Six weeks after the massacre, El Mozote had made it onto the front pages of America’s two most important newspapers.

The following day, Ronald Reagan sent to Congress the Administration’s certification that the government of El Salvador was “making a concerted and significant effort to comply with internationally recognized human rights.”

Two days later, on January 30th, Todd Greentree drove out to Ilopango Airport, climbed into a Salvadoran Army Alouette helicopter, and in a few minutes was sweeping over green volcanic landscape toward the mountains of Morazán. At his side was Major John McKay, of the defense attaché’s office. A one-eyed marine (he had been wounded in Vietnam), McKay was known to have the best contacts among the Salvadoran officers of any American in the country. The two men were headed for El Mozote to have a look for themselves.

It was not the most propitious time. The Army was tense; three days before, guerrilla commandos had stormed Ilopango in a daring raid and had succeeded in destroying a large part of El Salvador’s Air Force as it sat on the tarmac. The raid — which the guerrillas named Operation Martyrs of Heroic Morazán, in honor of those killed in December — would not look good in Washington. The congressional debate loomed large in the minds of those in the United States Embassy. “It was in the middle of a phenomenally packed, intense period down there,” Greentree recently told me by telephone. “We had the investigation of the murders of the nuns, we had the Constituent Assembly elections coming up, and, of course, we had the certification” — which only intensified the pressure from “the political microscope in the States,” as Greentree called it. “The primary policy objective at the time was to get the certification through,” he said, and the spectacular reports of the massacre threatened the certification. “From the Embassy’s point of view, the guerrillas were trying to make us look as bad as possible. They wanted to shut the whole thing down.”

The Americans landed at the brigade command in San Miguel to refuel and to receive their first briefing. “The brigade commander was expecting us,” Greentree said. “In San Miguel, that was Flores” — Colonel Jaime Ernesto Flores Grijalba, the over-all commander of Operation Rescue. Also present, Greentree believes — he is not absolutely certain — was Domingo Monterrosa. The officers gave the Americans “a sort of after-action report, saying which units were where,” Greentree said. “As I recall, the Atlacatl was the main combat unit, and they talked about this hammer-and-anvil nonsense. We were dismayed, because the Atlacatl was supposed to have developed new tactics, but now they were back to the same old shit — you know, insert a blocking force and then carry out a sweep.” The message about El Mozote — the version that the Salvadoran Army had presumably already provided the defense attaché’s office — was, in effect, that the Army had fought hard to dislodge a large company of guerrillas from the town, and though perhaps a few civilians had been killed in the crossfire, soldiers certainly had not carried out a massacre.

Colonel Flores was not particularly happy to see the Americans, and it was clear that his attitude was shared by the other officers they encountered that day. As McKay — who is now a colonel attached to nato headquarters in Brussels, and was given permission to speak publicly about the events at El Mozote by the Defense Department — told me, “In general, we had very little coí¶peration when we went to Morazán.”

They left San Miguel and flew over the Torola toward El Mozote. “You could see there had been a combat sweep through the area,” Greentree said. “You could tell El Mozote had been pretty much destroyed. Roofs were collapsed, buildings were destroyed, and the place was pretty much abandoned.”

As they flew over El Mozote, Greentree went on, he could see signs of battle. “There was an escarpment close to the town, an obvious line of defense, and you could see trench lines there. There were definitely fortifications in the vicinity.” When I pressed him for details, he said that the fortifications might have been closer to Arambala, a mile or so away.

They made several passes at a couple of hundred feet, then circled around for a better look. “As we lost altitude and got within range, we got shot at,” Greentree said. “That was pretty standard stuff out there. It was definitely not a landing situation.”

They headed to Gotera, touched down at the barracks, and received another briefing. “The purpose of the briefing was to impress on us that this was a war zone out there,” said Bleakley, the deputy chief of mission, who had come to Gotera on another helicopter and met Greentree and McKay there. The officers’ point was that “not only were they not out there killing civilians but they were fighting for their lives in that very dangerous war zone to protect the civilians from guerrilla atrocities.”

The Americans said they’d like to have a look, talk to some people in and around the town. “It was extremely tense,” McKay told me. “The Army was clearly not happy with our presence there.”

Accompanied by a squad of soldiers, McKay, Greentree, and Bleakley set off for the refugee camp outside Gotera. “We literally went up and down the streets, saying, ‘Hey, do you know anyone from El Mozote?’ ” Bleakley said. “The impression you got from people was that this was a conflict zone, that the people still up there were camp followers, you know, involved in the conflict.”

And yet, as McKay acknowledged, the presence of the soldiers made the task of conducting what would, in any case, have been difficult interviews almost impossible. “You had a bunch of very intimidated, scared people, and now the Army presence further intimidated them,” McKay said. “I mean, the Atlacatl had supposedly done something horrible, and now these gringos show up under this pretense of investigating it, but in the presence of these soldiers. It was probably the worst thing you could do. I mean, you didn’t have to be a rocket scientist to know what the Army people were there for.”

Greentree managed to speak to a number of people — including a mayor from one of the towns near El Mozote and several peasants who had lived near the hamlet — out of the soldiers’ hearing. “McKay would work the military and keep them distracted while I went out and around and talked to people,” Greentree said.

The three Americans agreed that the information they gathered in the refugee camp was not explicit. As Greentree put it, “I did not get any direct eyewitness accounts of what had taken place, of the type that Ray Bonner and Alma Guillermoprieto reported. It was more sort of the way people were talking and the way the kids around were still looking as if they’d been through hell, and people saying, ‘Yes, my wife was killed’ — that sort of thing.”

Sometime during these interviews, he and McKay became convinced that something had happened in El Mozote. “You could observe and feel this tremendous fear,” McKay said. “I was in Vietnam, and I recognized the ambience. The fear was overriding, and we sensed it.”

“People were freaked out and pretty scared about talking, and stuff,” Greentree said. Nonetheless, the interviews in the refugee camp “convinced me that there probably had been a massacre, that they had lined people up and shot them.”

Bleakley, however (who, as deputy chief of mission, was the senior officer of the three), told me that though “it was clear people had been killed, some of them civilians, what we couldn’t answer was the fundamental question — you know, the difference between subduing a town and pulling out the civilians, My Lai style, and massacring them.”

Still, Greentree said, “each person I talked to confirmed the impression that something bad had happened, but nobody was willing to go ahead and give the exact story.” He drew this conclusion “from things they said, their general manner — and their general unwillingness to talk. And that includes the soldiers as well. I mean, you talk to a soldier who thinks he’s taken part in some heroic operation — and a Latin soldier, I mean — you can’t get him to shut up. But these soldiers would say nothing. There was something there.”

Travelling with the squad of soldiers, McKay and Greentree left the refugee camp (Bleakley, who had business in the camp, stayed at Gotera), climbed into a military jeep, and headed up the black road. “We went to five villages,” McKay said, including Jocoaitique, within a few miles of El Mozote. “We talked to a priest who gave us oblique information that something horrible had happened, and that it was committed by the Army.”

Now the two men, accompanied by the soldiers, set out for El Mozote to see for themselves. “Between five and seven clicks south of Jocoaitique, we were going to turn off the road toward the hamlet and head there cross-country,” McKay said. But the soldiers had begun to grow quiet. “There began to be complaints. They were already sensitive about the civilian with me. Now they were getting more and more sullen. You know, they’d look at the ground, mumble something about being out of radio contact.” Finally, the group reached the place where they’d have to leave the black road for El Mozote. At that point, the soldiers just stopped. “The sergeant said, ‘We’re not going any farther, we’re not going to help you.’ It was made very clear that we would get no more coí¶peration.”

They had come very close to El Mozote. In less than an hour, they could have seen for themselves the burned buildings, the ruined sacristy, and the bodies. But, with the soldiers’ refusal to go on, the Americans faced the choice of heading on across open country — guerrilla-controlled country — without protection or turning back. “You want to know what made me decide?” McKay said. “Well, I’d been on that helicopter over there, and we’d received fire, and, the month before, the guerrillas had wiped out a whole company up there. What made me decide — me, the big tough marine? I was scared shitless.”

The choice was clear. The Americans, with their soldier escort, turned around and trooped back to Gotera, and from there the helicopter carried them back to the capital. The investigation was over.

At the Embassy, Greentree sat down and began to write, and by the following day, after consultations with Bleakley and review by others in the Embassy, including Ambassador Hinton, a lengthy cable, over the Ambassador’s name, was dispatched to Washington — a cable that provided the basis for what Assistant Secretary of State Enders told Congress two days later. This cable, which was originally obtained in 1983 by a Washington research group called the National Security Archives under a Freedom of Information Act request, is a remarkable document. Its opening paragraph — the all-important “summary” that heads diplomatic cables — reads (with emphasis added) as follows:

Embassy investigation of reported massacre at El Mozote including visit to the area by assistant [defense attaché] and [Embassy officer] concludes following: Although it is not possible to prove or disprove excesses of violence against the civilian population of El Mozote by Government troops, it is certain that the guerrilla forces who established defensive positions in El Mozote did nothing to remove them from the path of battle which they were aware was coming and had prepared for, nor is there any evidence that those who remained attempted to leave. Civilians did die during Operación Rescate but no evidence could be found to confirm that Government forces systematically massacred civilians in the operation zone, nor that the number of civilians killed even remotely approached number being cited in other reports circulating internationally. We are still pursuing question as to which Army units were present in El Mozote. End Summary.

In the entire summary, only one point is considered solid enough to be dubbed “certain” — that “the guerrilla forces who established defensive positions in El Mozote did nothing to remove [civilians] from the path of battle.” And yet, as Greentree conceded in our conversation, the descriptions of fighting in El Mozote, and of the “defensive positions” there, came largely, if not exclusively, from the Army briefings. “The information that we had presented to us as concrete was, of course, from the Army side, about the conduct of the combat operation,” he said. The slender version of what happened in El Mozote seems to be a mixture of Army briefings and, at best, inferences by Greentree and Bleakley.

How could the investigators be certain that the guerrillas did nothing to remove civilians “from the path of battle”? The day before the trip to Morazán, another junior reporting officer had sent Bleakley a memorandum passing along the report of a source — the name is effaced in the version released by the government this fall — who, while skeptical of the reported numbers of dead, said that “the military did undertake a sweep (‘limpieza’) of the area, that residents of the area were given time to leave it, that most did and that among the unknown number of victims of the operation were some (unspecified) evangelicals who unwisely chose to stay behind.” (Bleakley does not recall the document, but he did say that it “conforms with my memory of the time that there were people who were part of this new evangelical movement in El Salvador who would live in guerrilla areas and manage to stay above the conflict.”) It may be that some of the odd language of the summary (“nor is there any evidence that those who remained attempted to leave”) was influenced by this memorandum.

In any event, the assertion that guerrillas “did nothing to remove” civilians is actually contradicted later in the cable, when the authors describe an “aged couple” who said that guerrillas “told them to leave in early December.” According to the cable, this “aged couple” returned to El Mozote after “the fighting had ended and soldiers were in control.” What did they find? “They claimed they saw dozens of bodies.” This “claim” is simply quoted, without comment, as is the remark, in the next paragraph, by a man who “knew of violent fighting in El Mozote and other nearby cantons” but was “unwilling to discuss comportment of government forces saying ‘This is something one should talk about in another time, in another country.’ ”

These quotations, together with the flat statements to me from Greentree and McKay that it was clear to them at the time that “something horrible” had happened at El Mozote, that “there probably had been a massacre,” make the cable’s summary puzzling, to say the least. Read now, the circumspect locutions that dominate the summary take on the aspect of shields — judicious phrases by which the investigators deflected the burden of explicitly recounting what they strongly suspected had happened. What is curious is how, instead of building on their observations, inferences, and conclusions to present the best version possible of what probably happened, they emphasize the gap between what could be definitively proved to have happened — which, of course, wasn’t much, given the reticence of the people and the constraints on the investigators’ movements — and what the newspapers and the guerrillas were claiming had happened. It is a peculiar way of reasoning, built, as it is, on the assumption that in the absence of definitive proof nothing at all can really be said to be known. In effect, officials made active use of the obstacles to finding out the truth — and formidable obstacles certainly existed in El Salvador in 1982 — to avoid saying clearly and honestly what they knew and what they suspected.

McKay, at least, seems to have been troubled by this at the time. “We could not have said, ‘My God, there’s been a massacre,’ ” he told me. “But, truth be known, the ambiguity of the cable that went out — in my own conscience I began to question it. And then when I saw the New York Times piece, and the picture, that really got me to thinking. Bonner and I had gone to Quantico together, went to Vietnam together.” McKay finally sent off another cable — “through my own channels,” presumably a military or an intelligence circuit — “and though I can’t say categorically that I actually wrote ‘something horrible happened,’ what I said was to the effect that something had occurred, because of the fear we had detected from the people there.”

McKay, of course, had reviewed the State Department cable before it was sent, but he was not its author; Greentree was. Though he was only twenty-eight years old, Greentree had already earned the respect of his Foreign Service colleagues and — what was much rarer in El Salvador — was considered a competent, trustworthy official by many in the press corps. Indeed, even a decade later, in his understanding of what had happened in El Salvador he seemed to me the most perceptive of the American officials I interviewed. It was Greentree who embodied the United States government in the closest contact it would make to the massacre at El Mozote, and yet it was Greentree who provided the reporting that would enable the government to deny that the massacre had happened. It is tempting to conclude that he simply suppressed what was inconvenient, but the truth of what happened in the writing of the cable, like most of the United States’ dealings with the issue of “human rights” in El Salvador, is rather more interesting than that.

Greentree’s recollection, during a series of telephone interviews, of the writing of the cable and of its contents followed a fascinating progression. “As I recall,” he told me, “I gave the military account the benefit of the doubt, but I probably put in the summary more ambiguity about what I felt.” He went on to say, “There were probably a few lines in there that emphasized that, hey, we infer from some of the information we picked up that something happened, and so on.” When he was told no such ambiguity could be found in the summary — that, in effect, the only ambiguity in the cable was the conflict, wholly unacknowledged, between its conclusions and some of the observations in the body — he said he “imagined that in the clearing process that got taken out.”

The “clearing process,” in which the cable made the rounds of officials in the Embassy for review, centered on Kenneth Bleakley, and his recollection of the trip, alone among the recollections of the three, coincides with the conclusions drawn in the cable. Nonetheless, Greentree insisted to me that he “did not feel that what went out distorted beyond acceptability” what he had written. In a later comment, he stated emphatically, “At no time during my tour in El Salvador was a report that I had anything to do with ever distorted by the Embassy. Because those are the standards that Hinton set.” Like many in the Embassy, and throughout the Foreign Service, Greentree had great respect for Hinton. He describes Hinton as “a totally credible person” and, in writing what he wrote, he clearly felt the pressure to conform to the older man’s standards. Yet it is hard not to suspect that Greentree’s strong belief that the cable contained more “ambiguity” than in fact it did reflects a lingering unease with the final product — a conflict that persists, even after twelve years, between what he wrote and what he felt he should have written.

“I had been in the Foreign Service for only a couple of years at that time,” Greentree told me, “and we had a very strong Ambassador, and our instructions were to be clear and clean — to not distort. You write it down, and then that becomes the eyes and ears of the United States government. And this was especially important because the journalists reporting in El Salvador were thought to be biased. So if I had said everyone was crying, and everything — well, that wouldn’t have had any credibility, either. We reported what we saw, and the main requirement was to distinguish between what you saw and what other people said, and, even more than the standards of journalism, to keep your ‘slant’ out of what you were reporting.”

Had he not been operating under the constraints of politics in Washington, what would he have written differently?

“Well, I would have put in more strongly the impression that abuses against the civilian population probably took place in El Mozote and the surrounding areas during that operation.” But he repeated, “It was just an impression. There was no direct corroborating evidence.”

Yet this was his strongest impression, and since the limitations caused it to be omitted, didn’t they feel rather artificial, at the least?

“That’s right,” he said. “But that’s where, I guess, political judgment came into it. And it was not the judgment that you would think — that, you know, the Ambassador’s got to make sure that the information is politically correct. It was that, for the rest of the report to have credibility among people who were far away and whose priorities were — you know, we’re talking about people like Tom Enders — whose priorities were definitely not necessarily about getting at exactly what happened: in order for the report to have credibility, all those things have to be kept to a minimum.”

At that point, one begins to understand the pressures on the Embassy, and the effect that the great game of politics being played in Washington had on those who were supposedly acting, within El Salvador, as “the eyes and ears of the United States government.” Ambassador Hinton was “the guy who sets the standards,” Greentree said. “So, of course, since I was a junior officer, my eyes were not on the policy. They were being very affected by the things I was seeing and encountering out there. From the Ambassador’s perspective, he had to keep his eye on where we were supposed to be going in the country, and he had to put where the ‘truth’ was in the context of that. In other words, the possibility that the guerrillas were making a major propaganda ploy over a massacre that might or might not have occurred in El Mozote, and were doing so for the purpose of derailing U.S. policy — well, what the Embassy had to say about that event had to be very, very carefully phrased and controlled, to get as close as possible to what happened and as far away as possible from propaganda on either side, regardless of what might then happen to it once the report got to Washington and was one way or another translated into testimony before Congress.”

In reality, then, the admonition to be “clear and clean,” to be “professional” and “not distort,” served as an excuse to exclude from the cable the very things that had most impressed the men who actually ventured into the war zone. The emphasis on “clean” reporting permitted the blinding and deafening of the government, and served to remove from its field of perception what might have proved to be, in the Washington of early 1982, a very inconvenient fact.

In place of McKay’s clear impression that “something horrible happened,” and of Greentree’s conviction “that there probably had been a massacre, that they had lined people up and shot them,” the cable supplied to officials in the State Department a number of arguments that they might find useful in impeaching the press accounts of El Mozote — deeply misleading arguments that would form the basis of the government’s effort to discredit the reports of the massacre. After citing the numbers of dead that had appeared in the Times and the Washington Post, the cable noted, “It is estimated that no more than 300 people were in the entire canton prior to December 1981” — ignoring the fact that both newspapers had made it quite clear that the massacre took place in El Mozote and in a number of hamlets around it. As for the names of the dead “subsequently reported in the U.S. press,” the cable suggested that those “may well have been extracts in whole or part from the civil registries … stolen from Jocoaitique by subversives,” though it offered no evidence whatever for this assertion. “I don’t recall thinking it was what happened,” Greentree said when he was asked about the Jocoaitique claim, “but I thought it was a possibility.” And yet he might have learned from Bonner (with whom Greentree was in frequent contact) that the guerrillas had shown the reporter the list several days before they attacked and captured Jocoaitique, so the “possibility” that the names were actually drawn from captured civil registries from Jocoaitique — a charge that an Assistant Secretary of State would soon be repeating to the Congress of the United States — was not a possibility at all.

“El Mozote is in the heart of guerrilla territory,” one reads on page 2 of the cable, “and its inhabitants have spent most of the past three years willingly or unwillingly cooperating with insurgents” — an odd locution, particularly since the next sentence notes the fact that “Government forces” were last posted in El Mozote in August of 1981, just four months before Operation Rescue. The observation about “willingly or unwillingly cooperating with the insurgents” echoes the attitude of the Salvadoran Army, in which anyone living north of the Torola must be, a priori, a guerrilla follower — and was thus, in the officers’ view, fair game. And yet Greentree clearly understood that the reality was more complex.

“Most of these people didn’t want anything to do with any of this stuff,” he told me. “They just wanted people to leave them alone … They were victims of this whole thing … If they could get away by giving guerrillas some corn and chickens, and still live on their farms, that’s what they would do. At the same time, if the people had to get by by giving corn and chickens to the half a dozen Guardia Nacional who were living in their town, then they would do that — whatever it took to enable them to live.”

It is an eloquent and concise statement of what the civil war had done to many of the people of Morazán by 1981, but, unfortunately, nothing near such depth of understanding is allowed to come through in the cable.

The cable concludes by noting that the defense attaché’s office “is attempting to determine which Army units were present in El Mozote during and after the operation.” Of course, if the Embassy wanted to discover what had happened in Morazán this should have been the other path of inquiry: putting the question directly to the American-funded and American-trained Army. And yet six weeks after the events were alleged to have taken place the Embassy reported that it had not managed to discover which units were in El Mozote — this although at least ten American advisers were assigned to the Atlacatl, the unit accused in all the press reports.

As several recently released cables confirm, however, matters were a bit more complicated. On the day Greentree and McKay made their trip to Morazán, Ambassador Hinton had a discussion with Salvadoran Defense Minister Garcí­a — “on margin of dinner,” as he puts it in his cable — about El Mozote. The General (Garcí­a had been promoted on January 2nd) was about to make a trip to Washington to attend “a Congressional prayer breakfast,” and the Ambassador warned him that he should “be ready to respond to Morazán massacre story.” General Garcí­a, Hinton writes, “was his usual cocky self. ‘I’ll deny it and prove it fabricated.’ I wished him well and added he would have to explain away details provided by correspondents. It might be possible — we were investigating and were grateful for his help — but he should bear in mind that something had gone wrong. Who did it, when, and in what circumstances was something else.”

Two days later, on the afternoon of February 1st, an American officer from the defense attaché’s office travelled out to the Atlacatl headquarters and met with, among others, Major Cáceres, Major José Armando Azmitia Melara, and Lieutenant Colonel Monterrosa. The officer’s mission was “to specifically determine if the battalion, or elements thereof, were involved in the fighting around and in El Mozote.” After “greetings and pleasantries were exchanged,” the American put his question to the Colonel. “Monterrosa remained distantly courteous, but he firmly told RO [Reporting Officer] that he was not in a position to discuss these specific subjects, and that RO had better obtain permission from the General Staff of the Armed Forces before he came with such inquiries to his (LTC Monterrosa’s) Battalion.” Things had begun badly, the American reports: “Quite frankly, [I] felt that the interview, albeit short, was thus to be terminated.” The American hastened to make “the obligatory apologies for what the Colonel may have interpreted as impertinence. RO also pointed out that candid answers to the questions posed would facilitate countering recent press releases which were less than complimentary to the Armed Forces of El Salvador.”

Major Azmitia now spoke in “what can only be described as a parable,” explaining that “the unit that had fought at El Mozote had had a tough time of it” and that “because of the intensity and duration of the battle … there were undoubtedly casualties among non-combatants.” Monterrosa put in that “the unit involved had had to fight through fixed enemy positions, then, once in the town, fire was received from the houses in the town.” Here, according to the American officer, “Monterrosa now utilized the first person: ‘I do not have X-ray vision, and I cannot see inside the house from which someone is shooting at me; nor in those type of circumstances am I very disposed to waste time trying to find out who else might be in the house.’ … At this point LTC Monterrosa said … he was only speaking in general, not specific, terms about what had occurred at El Mozote.”

The Salvadorans, it seems clear, are giving the American a conspiratorial wink; but the American doesn’t seem to get it, and he ventures to ask Monterrosa if any prisoners had been taken. At this, the Colonel again “assumed an adamant demeanor, and suggested that RO consult with the general staff, or get permission from the general staff to ask him such questions.” Major Azmitia, as he escorted the American officer to his vehicle, “acted apologetic about the fact that they could not provide more facts about El Mozote, but he was sure that RO understood what LTC Monterrosa had been talking about.”

In his summary the American notes that “the two hours spent with these officers was interesting, to say the least. The nuances, subtleties and indirect comparisons used by LTC Monterrosa and Maj Azmitia were intriguing. Yet the central questions remain without definitive answers.” Here the officer ventures a “personal opinion” that the “Atlacatl battalion or elements thereof participated in the attack on El Mozote,” but he adds that “precluding permission from General Garcí­a … definitive answers … may never be forthcoming.”

Ambassador Hinton, clearly a bit frustrated, now asked his Milgroup commander if “it were possible High Command did not know where and when their field forces operated. No, it was not, he told me.” The Ambassador sent this officer to see the Salvadoran chief of staff, who informed the American that the “Defense Minister wanted no one other than himself to deal with that question.” Hinton, having been rebuffed on his subordinate’s level, now paid a personal visit to General Garcí­a. “We joshed a bit as is our wont,” reports Hinton, describing how the General complimented him “on my Washington Post interview which he said put things exactly right.” Hinton reminds the General that “Tom [Enders] had today gone to Congress to defend the additional 55 million in military assistance” and that “in this connection … reports published in the Washington Post and the New York Times about alleged Morazán massacre … caused great concern.”

Garcí­a replied, according to Hinton, that “the Morazán business was a ‘novela,’ pure Marxist propaganda devoid of foundation. I said it was clearly propaganda that its timing had been carefully calculated but there were so many details that it was difficult to deal with the stories.” The Ambassador now asks the General about some of those details, including the identity, among others, of Major Cáceres. The General replies that Cáceres is “a straightforward, honorable soldier who would never have killed women and children as described in the story.” After some discussion, General Garcí­a acknowledges that “the Atlacatl Battalion had been at El Mozote during the December sweep,” but then he “reiterated that the story was a pack of lies.” Garcí­a does promise the Ambassador that he will look into the matter further. “He asked me to leave with him the stories and I did so adding as a sweetener the Washington Post editorial of January 29 supporting our common policies.”

One gets a vivid sense from these cables of the frustrating position the Americans had placed themselves in in their dealings with the Salvadoran military. The Salvadorans behave with an arrogance that bespeaks their awareness of their own power. Washington was behind them and they knew it: Why should they comply with these local officials, except in those cases where they absolutely had to?

In the case of El Mozote, it was already clear that they didn’t have to. Greentree remembers thinking as he sat in the helicopter on the way back to the capital from Morazán, “If we’re really going to get to the bottom of this, there’s going to have to be a decision to put a tremendous amount of energy into it, to carry out a more formal investigation, like the ones conducted for the Americans” — the four churchwomen. “I remember feeling frustrated and dissatisfied with what we came back with. But, if we’d wanted to go any further with it, it would have taken a decision to expend a tremendous amount of effort.” No such decision was ever made.

Two days after Greentree’s cable arrived at the State Department, Assistant Secretary Thomas O. Enders went up to Capitol Hill. Sitting before the House Subcommittee on Inter-American Affairs, he set out to defend the President’s certification that the Salvadoran government was making a “concerted and significant effort to comply with internationally recognized human rights.”

Secretary Enders told Congress that he would make “a coherent attempt to answer the question that you have raised … are we getting some results.” “Results” he would interpret to mean improvement. Thus he would be arguing, in essence, that, however horrendous “the human-rights situation” might now be in El Salvador, the last year had in fact been less horrendous than the year before. The effect of this argument was to shift the ground of the debate. “Previously, it had been ‘We think this human-rights thing is important and you don’t think it’s that important,’ ” Aryeh Neier, then the director of Americas Watch, told me. “What the Reagan Administration did was embrace the principle of human rights and then conduct warfare over the facts. The fight over El Mozote exemplified this.”

The human-rights groups had geared up to fight this new war; Americas Watch, for example, which had been founded only the summer before, issued a book-length study, “Report on Human Rights in El Salvador,” on January 26th, two days before the certification was sent up to Congress. For the human-rights groups and for leading Democratic congressmen, as well as for the Administration officials, the fight would center on information and how it was gathered.

“Accurate information,” Secretary Enders began. “I think we all have found out that is very hard to establish. The responsibility for the overwhelming number of deaths is never legally determined nor usually accounted for by clear or coherent evidence. Seventy per cent of the political murders known to our Embassy were committed by unknown assailants.” As in the cable, the fact that the killers’ identities could not be definitively known, though in most cases few doubted who the killers were, was used as a shield — an excuse to ignore what was known. In the absence of conclusive, undeniable proof, the government would feel free to assert that all was darkness.

“We sent two Embassy officers down to investigate the reports … of the massacre in Mozote,” the Secretary went on. “It is clear from the report that they gave that there has been a confrontation between the guerrillas occupying Mozote and attacking government forces last December. There is no evidence to confirm that government forces systematically massacred civilians in the operations zone, or that the number of civilians remotely approached the seven hundred and thirty-three or nine hundred and twenty-six victims cited in the press.” Echoing the strategy suggested in Greentree’s cable, Enders went on, “I note they asked how many people there were in that canton and were told probably not more than three hundred in December, and there are many survivors including refugees, now. So we have to be very careful about trying to adduce evidence to the certification. We try, our Embassy tries, to investigate every report we receive.”

Six days later, Elliott Abrams, the Assistant Secretary of State for Human Rights and Humanitarian Affairs, remarked to the Senate Foreign Relations Committee that the El Mozote case “is a very interesting one in a sense, because we found, for example, that the numbers, first of all, were not credible, because as Secretary Enders notes, our information was that there were only three hundred people in the canton.”

The argument about numbers is, of course, deeply misleading — no one who read the Times and the Post articles could have missed the fact that the killing had taken place in several hamlets; two of the three survivors Guillermoprieto quoted, for example, were from La Joya, not El Mozote. But the argument exemplifies a pattern. Claiming to have investigated “the facts” and to have found “no evidence” of a massacre, American officials then seized on aspects of the charges that, they said, reveal them to be propaganda. “We find … that it is an event that happened in mid-December [but it] is then publicized when the certification comes forward to the committee,” Abrams told the Senate. “So, it appears to be an incident which is at least being significantly misused, at the very best, by the guerrillas.” In an interview more than a decade later, Abrams made the same argument. He pointed out that the massacre had “supposedly” taken place in December, and asked, “If it had really been a massacre and not a firefight, why didn’t we hear right off from the F.M.L.N.? I mean, we didn’t start hearing about it until a month later.”

As has been noted, the guerrillas first “publicized” the massacre about two weeks after the event — as soon as they had got Radio Venceremos back on the air. All the same, it is indisputable that the volume of reporting about El Mozote from Venceremos, from human-rights groups, and from the international press grew steadily throughout January, and reached a crescendo the day before Reagan’s certification, with the front-page stories in the Post and the Times. Certainly a significant part of this publicity — it is impossible to say how much — was owing, directly and indirectly, to the efforts of those, beginning with the guerrillas and their international propaganda apparatus, who had a strong interest in derailing the Administration’s policy in El Salvador. But Administration officials focussed obsessively on this unsurprising reality, as if the very fact that the El Mozote story was being used as propaganda — that it was, as Abrams put it, “significantly misused … by the guerrillas” — in itself constituted proof that the massacre hadn’t taken place.

To many in the Administration, the importance of the massacre was that it had such propaganda value, and that the propaganda, coming at a crucial time, posed a threat to American aid. Preserving the Salvadoran government and helping it win the war were paramount; “improving human rights” naturally took a back seat since, as the Administration liked to put it, by far the worst disaster that could befall human rights in El Salvador was a Communist victory. This attitude was no mystery to the Salvadoran leaders; despite the periodic brouhahas over certain atrocities, they could see the bottom line quite clearly, which was, as Abrams phrased it, that “whatever you think of us from a human-rights point of view, what you think of us from a security point of view is determinative.”

To say the least, this attitude did not encourage anyone in the State Department to make any additional effort to find out what had happened at El Mozote. As far as the Department officials were concerned, Greentree’s cable was the end of the matter. The cable had come from Hinton’s Embassy, and Hinton had a great deal of prestige in the Department. By now, however, Hinton himself had taken a rather different view. “I would be grateful if Department would use extreme care in describing my views on alleged massacre,” he cabled on February 1st. Apparently, Washington had sent out cables saying that the Ambassador, in his reply to the National Council of Churches, had denied the massacre had taken place. “My letter did not ‘deny’ incident: it reported that at that time I had no confirmation and … had no reason to believe Venceremos reports. I still don’t believe Venceremos version but additional evidence strongly suggests that something happened that should not have happened and that it is quite possible Salvadoran military did commit excesses.”

Not only McKay and Greentree but now Hinton himself had come to the conclusion that “something happened” at El Mozote — and Hinton had now told the State Department so. To this, he added a frank appraisal of the Salvadoran officers’ credibility. “I find Garcí­a’s assertion … ‘we have absolutely no information on military actions in El Mozote’ to be stonewalling without credibility. I have tried to warn him re need to face up to problem, but my impression is he thinks categoric denial is way to handle question. Department officers may wish to discuss matter with him … before U.S. press gets to him.”

As it happened, however, “Department officers” seem to have agreed with General Garcí­a. They had the Greentree cable, and they would make use of it. After all, the question would come down to — as Abrams put it to me — “Do you believe the Embassy, an agency of the United States government, or Americas Watch?” Americas Watch and other human-rights organizations, Abrams said, “did not have a great deal of credibility with us,” for, in his view, they had ranged themselves on the side of those who argued, in effect, for an F.M.L.N. victory, and thus they served as willing tools of the hypocrites in Congress who now forced Administration officials to undergo a meaningless certification exercise. “Certification was this political game they were playing,” Howard Lane, the Embassy press officer, told me. “I mean, everybody knew, Congress knew, what they” — the Salvadoran government — “were doing down there. By then, they had to know, unless they refused to see it. So they beat their breasts, and tore their hair, and yelled about human rights, and made us jump through this hoop called certification. If any Ambassador wanted to keep his job, he had to jump, which meant essentially saying the half-empty glass was really half full. It was a game. I mean, ‘improvement’ — what’s improvement, anyway? You kill eight hundred and it goes down to two hundred, that’s improvement. The whole thing was an exercise in the absurd.”

Ever the good soldier, Enders on Capitol Hill attacked the numbers from the human-rights groups, put forward the Administration’s numbers, explained how, despite all appearances, the Salvadoran government was “making progress.” He testified, “The results are slow in coming. I would agree with you on that. But they are coming … The figures show it. We have September, October, November, December figures for 1980, which show something on the order of eight hundred, seven hundred and seventy-nine, five hundred and seventy-five, and six hundred and sixty-five political murders. That is for 1980. We have the same figures for this year which show September, a hundred and seventy-one, October, a hundred and sixty-one, November, three hundred and two. It shows December, two hundred. Our returns are showing markedly different numbers on the same methodology.”

That methodology, as anyone who had looked into it knew, was very obviously flawed. The Administration’s numbers, drawn from the Embassy’s weekly “grim gram,” were based on reports in the Salvadoran newspapers, all of which not only ranged from conservative to unabashedly right-wing but weighted their reporting toward the cities. In 1981, fewer people were being killed in the cities, because fewer activists were there to be killed; most of those who had not been liquidated in late 1979 or 1980 had moved to the mountains. And the killings in the mountains, in the isolated hamlets and villages, rarely reached the pages of newspapers in the capital.

“Let me be clear this is not a complete report,” Enders told Congress. “Nobody has a complete report … But, nonetheless, it is a coherent attempt to answer the question that you have raised … are we getting some results. This is the indication that I submit to you that we are.”

To this statement a number of congressmen responded with outraged eloquence. Gerry Studds, Democrat of Massachusetts, told Enders, “If there is anything left of the English language in this city … it is gone now, because the President has just certified that up is down and in is out and black is white. I anticipate him telling us that war is peace at any moment.” It was an irresistible quote, and it made for great television. But it didn’t make any difference. Enders had supplied a “coherent attempt to answer the question” that Congress had posed, and though Democratic congressmen would not spare their voices, or their sarcasm, in noting “the Orwellian tones of this certification,” as Steven Solarz, Democrat of New York, put it — though congressmen attacked the numbers and the methodology, and the hearings became contentious and angry — it was clear that, come what may, there would not be the votes to cut off aid to El Salvador, for that, as everybody knew, would mean “losing” the country to the Communists. At root, nearly everyone tacitly agreed (the Democrats — whose purported “loss” of China three decades before was still a painful Party memory — no less than the Republican Administration and its allies) that that eventuality was too intolerable even to contemplate, and that in the end the Salvadoran government, by whatever means, had to win the war, or the country’s security would be unacceptably threatened. And so, because of this underlying agreement, the entire debate, loud and angry as it appeared at first glance, was not a debate. It was an exercise for the cameras.

As for El Mozote, since the Salvadoran newspapers said nothing about it, those who had died there merited no place in the numbers Secretary Enders brought to Congress. Had the massacre somehow been “proved” to the State Department’s satisfaction — had it been, somehow, impossible for the Administration to deny — El Mozote would have had an ugly effect on the Administration’s numbers: political murders would have shown an increase in December from six hundred and sixty-five to well over a thousand, rather than the sharp decline he claimed. Would this have led Congress to reject the certification and cut off aid? Reading the record now, feeling once again the fear in Washington of an F.M.L.N. victory and of the blame such a victory might impose on American politicians, the question seems, sadly, difficult to answer. Aid might have been reduced, true, but, at most, Congress might have managed to cut off aid temporarily, only to restore it again in a panic — as Carter had done — at the first new guerrilla onslaught.

But this is speculation. In the event, the dead of El Mozote did not really come into the discussion at all.

On February 10th, the Wall Street Journal published a lengthy editorial headed “The Media’s War,” in which it noted that the public’s “perceptions are badly confused” on the war in El Salvador, and attributed much of that confusion to “the way the struggle is being covered by the U.S. press.” Most notable were several paragraphs that took up the question of El Mozote:

Take the recent controversy over charges of a “massacre” by an elite battalion of the El Salvadoran army. On January 27, Raymond Bonner of the New York Times and Alma Guillermoprieto of the Washington Post simultaneously reported on a visit to rebel territory, repeating interviews in which they were told that hundreds of civilians were killed in the village of Mozote in December. Thomas O. Enders, assistant secretary of state for Inter-American affairs, later cast doubt on the reports. There had been a military operation but no systematic killing of civilians, he said, and anyway the population of the village was only 300 before the attack in which 926 people supposedly died.

When a correspondent is offered a chance to tour rebel territory, he certainly ought to accept, and to report what he sees and hears. But there is such a thing as being overly credulous. Mr. Bonner reported “it is clear” the massacre happened, while Miss Guillermoprieto took pains to say that reporters had been “taken to tour” the site by guerrillas with the purpose of showing their control and providing evidence of the massacre. In other words, whatever the mixture of truth or fabrication, this was a propaganda exercise.

Realistically, neither the press nor the State Department has the power to establish conclusively what happened at Mozote in December, and we’re sure the sophisticated editors of the Times recognize as much. Yet as an institution, their paper has closed ranks behind a reporter out on a limb, waging a little campaign to bolster his position by impugning his critics. A “news analysis” charged the government of sowing confusion by questioning press reports “without presenting detailed evidence to support its position.” The analysis posed the question of “how American diplomats gather information abroad,” but not the same question about American reporters.

Oddly missing from these paragraphs, and from the rest of that very long editorial, was any acknowledgment that the two reporters had actually seen corpses — in Guillermoprieto’s case, at least, dozens of corpses — and that Meiselas had taken photographs of those corpses. Instead, the editorial said that the two journalists “repeat interviews in which they were told that hundreds of civilians were killed in the village of Mozote,” and then said immediately afterward that Enders “later cast doubt on the reports” — as if Enders, or his representatives, had actually made it to the village, as if the kind of evidence he was purveying were no different from what were, after all, two eyewitness accounts, if not of the events themselves, then of their aftermath. The reporting done by the journalists and by the Embassy officials is repeatedly yoked together, as if the two parties had visited the same sites, seen the same evidence, talked to the same people, and merely drawn different conclusions. Neither party, the editorial declared, “has the power to establish conclusively what happened at Mozote” — the implication being, as the Administration itself had argued repeatedly in its defense of its Salvadoran allies, that, since there is no “conclusive” account, nothing can be truly known. The idea that much of a journalist’s business consists of a studied sifting of what is said and what is observed, of a careful wrestling with gradations of evidence, and a continual judgment of the credibility of witnesses — this notion is nowhere present in the sixteen paragraphs of the Journal’s editorial.

Seven days after the Journal’s attack on the Times’ “overly credulous” reporter, the State Department received a cable over the name of the Ambassador to Honduras, John Negroponte, reporting on a visit by an Embassy official and a House Foreign Affairs Committee staff member to the refugee camp at Colomoncagua, to which many of the refugees from Morazán had fled two months before. According to the cable, the refugees described to the American diplomat “a military sweep in Morazán December 7 to 17 which they claim resulted in large numbers of civilian casualties and physical destruction, leading to their exodus.” The cable went on to say that “names of villages cited coincide with New York Times article of January 28 same subject.” The reporting officer added that the refugees’ “decision to flee at this time when in the past they had remained during sweeps … lends credibility to reportedly greater magnitude and intensity of … military operations in Northern Morazán.” This information was not made public.

Six months after the Journal’s attack on him, Raymond Bonner was gone from Central America. Since the El Mozote story and the controversy surrounding it, Bonner had been under great pressure, enduring a steady fusillade of criticism from the Embassy and the State Department, as well as from various right-wing American publications for whom Bonner had come to symbolize the supposed “leftward tilt” of reporting in Central America. In August, 1982, Bonner received a telephone call in his Managua hotel room informing him that he should report to the Metro desk in New York.

The Times’ decision to remove a correspondent who had been the focus of an aggressive campaign of Administration criticism no doubt had a significant effect on reporting from El Salvador. The New York Times editors appeared to have “caved” to government pressure, and the Administration seemed to have succeeded in its campaign to have a troublesome reporter — the most dogged and influential in El Salvador — pulled off the beat.

The public position of A. M. Rosenthal, then the executive editor of the Times, has always been, as he told me by telephone, that “at no time did anybody in the United States government suggest to me, directly or indirectly, that I remove Mr. Bonner,” and, further, that “anyone who would approach the New York Times and suggest to me that I remove or punish a correspondent would have to be an idiot. To imply that a man who devoted himself to journalism would remove a reporter because of the U.S. government or the C.I.A., or whatever, is ridiculous, naí¯ve, cruel, and slanderous.”

According to Rosenthal, Bonner was removed because he had never been fully trained in the Times’ particular methods. Bonner, he said, “didn’t know the techniques of weaving a story together … I brought him back because it seemed terribly unfair to leave him there without training.” Bonner had been trained as a lawyer, had been an assistant district attorney and a Nader’s Raider, and had joined the Times as a stringer in Central America. Seymour Topping, then the managing editor, told me that “because we were considerably pressed at the time in getting people into the field in Salvador, we short-circuited what would be our normal process of training people on Metro to learn the style and methods of the Times.” Bonner, Topping went on, “had done a first-class job of investigative journalism, and there was never any question that he had come up with the facts — that his stories were true. But, if he had been more experienced, the way he had written his stories — qualified them, etc. — would have left him much less open to criticism.”

But “training” was not the only issue — for that matter, as Bonner pointed out to me, he had spent a good part of 1981 on the Metro desk — and, at least in Rosenthal’s case, the question of Bonner’s “journalistic technique” seems to have been inextricably bound up with what the executive editor came to perceive as the reporter’s left-wing sympathies. “If anybody ever asked me to withdraw him, he’d still be there,” Rosenthal told me, and certainly the idea that the government simply pressured the Times into withdrawing Bonner is wrong. Rosenthal suggests that others have promoted this version of the story because “I was an agent of change in the Times, and a lot of people didn’t like my politics”; but conversations with a number of Times reporters and editors, former and current, persuaded me that the campaign against Bonner was more effective than it might have been because of Rosenthal’s own politics. Several people told me that Rosenthal had made no secret that he was unhappy with Bonner, because the reporter, as one characterized the editor’s view, “was too willing to accept the Communist side of the story. He was very vocal that Bonner was sympathetic to the Communist side in Central America.” The criticism from the right — led by the Wall Street Journal editorial on El Mozote — “resonated with Abe, because it reinforced his own suspicions about Bonner. There seemed to be a growing audience out there that agreed with Abe.” Several current and former Times employees (none of whom would speak for attribution) pointed to a scene in a Georgetown restaurant a few weeks after the El Mozote story ran — it was the evening of the annual Gridiron dinner — in which Rosenthal criticized Bonner and angrily described the sufferings that Communist regimes inflict on their people.

(Bonner finally left the Times in 1984; in 1987, he began writing for The New Yorker — as did, two years later, Alma Guillermoprieto. He left the magazine in 1992; he is now writing special assignments for the Times.)

El Mozote represented the climax of the era of the great massacres. It was not the last of them — most notably, in August of 1982 the Atlacatl, in an operation similar to that in El Mozote, killed some two hundred people at El Calabozo, in the Department of San Vicente — but after El Mozote the Army relied less and less on search-and-destroy operations that entailed large-scale killing of civilians. It may be that the guerrillas’ use of El Mozote for propaganda and the controversy that followed in the United States led senior officers to begin to realize the potential cost of such slaughter. It may be that the highly visible denunciations in Congress finally lent the Embassy’s habitual scoldings a bit more credibility. (Even someone as firmly contemptuous of congressional pressure as Elliott Abrams acknowledges that “the good-cop, bad-cop routine with Congress was very effective” and that “there was some positive impact there in reducing the killing.”) It may be that the officers realized that lesser massacres — of forty people or fewer, say — could accomplish as much without attracting so much attention.

More important, the key Salvadoran officers no doubt realized that El Mozote had accomplished its purpose. It was not only that in much of north- ern Morazán the civilians had fled beyond the border — that in several key areas the water had been taken from the fish. It was what El Mozote had meant — what it had said — to those who remained. For El Mozote was, above all, a statement. By doing what it did in El Mozote, the Army had proclaimed loudly and unmistakably to the people of Morazán, and to the peasants in surrounding areas as well, a simple message: Whatever the circumstances, the guerrillas can’t protect you, and we, the officers and the soldiers, are willing to do absolutely anything to avoid losing this war — we are willing to do whatever it takes.

By late 1982, the tide had begun to turn in Morazán, which is to say not that the Army had begun to win but that it had become less than certain that it would lose. The preceding March, the elections for the Constituent Assembly, on which the Reagan Administration had set much store, had been a huge political success for Administration policy, with a much higher turnout than had been expected. By exerting enormous pressure, the Administration had succeeded in blocking Roberto d’Aubuisson, the best known of the ultra-rightists, from becoming provisional President. Instead, the officers and party leaders and the Americans had agreed upon Alvaro Magaña Borja, a wealthy aristocrat and international banker with many old friends in the officer corps, as a compromise.

The successful elections and the consequent emergence of the highly presentable, English-speaking Magaña helped the Administration placate Congress. (By the July certification report, the Administration had altered its language from “no evidence to confirm” to “no evidence to support” allegations of “large-scale massacres allegedly committed by government forces” — in direct contradiction of what Hinton, and even Greentree, had reported.) Congress more than doubled military aid, from thirty-five million dollars to eighty-two million, and increased economic aid to more than twice that. Not only were the Americans sending new, top-of-the-line equipment and plenty of ammunition but they were expanding the Army — training hundreds of officers and soldiers in the States. Most important, Colonel Jaime Flores, apparently because of rather too blatant irregularities in his payroll in San Miguel, incurred the wrath of Magaña, and was consequently “promoted” from command of the all-important Third Brigade to command of the less important First Brigade, and, finally, to that of San Salvador’s Fire Department. To replace Flores in San Miguel, Magaña drew on the obvious — the inevitable — choice: Lieutenant Colonel Domingo Monterrosa.

Monterrosa thus became the military commander of the entire eastern zone of El Salvador, and entered upon the period of his greatest renown. Very often, the Army publicity people or the American press people steered reporters straight to the dynamic colonel. “He was a phenomenon,” Lucia Annunziata, who travelled frequently with Monterrosa as a correspondent for La Repubblica, told me. “The Americans were always telling us that here he was, here was the new breed of officer they were always promising. He had embraced completely the anti-Communist ideology of the Americans. By then, he talked not like some kind of butcher but like an American. He was completely full of this idea of conquering hearts and minds.”

In 1978, Monterrosa had attended the Political Warfare Cadres Academy, in Peitou, Taiwan, and had been trained there in what he described to an interviewer as “war of the masses” and “Communism of this side.” He’d returned to El Salvador “very enthusiastic” about the skills he had learned — “how to project ourselves to the civilian population and win them over” — but found to his dismay that senior officers weren’t very interested. Now he began to apply what he’d learned.

“He was always tactically very good,” Licho, the rebel commander, told me. “Then he began using much more intelligent methods. You know, whenever he would take a village he would come in personally and do political work himself.” His soldiers, usually helicopter-borne, would storm a town, flushing out the armed guerrillas, and then Monterrosa would arrive and gather the people together. “He would make a speech there in the plaza,” Annunziata said. “He would ask, ‘Who is sick? Who needs help?’ Then he would say, ‘Do you know these people?’ — that is, the guerrillas. And, of course, no one would answer. And he would say, in this soft voice, ‘Are you sure? Are you sure you don’t have a cousin with them?’ ”

By this time, people all over the countryside recognized the famous figure of Monterrosa. He was short — stooped, even — with a slight paunch. “He was completely nonmartial,” Annunziata said. “He always wore this tattered, sweat-stained camouflage-green bandanna on his head, and he had a real Indian face — big nose, receding chin. With that bandanna, he looked like an old aunt. He was a bit of a fop, a bit dandified. He had this young boy always with him, a beautiful young boy of ten or twelve, who took care of his things. He was always touching his soldiers — very physical, you know. At night, he would get in his red hammock and put on blue gloves and cover his face with a blue towel. He was a real dandy.

“It was late in the afternoon, and we were outside the town of Carolina, on a hill above it. Monterrosa was sitting on a low stone wall, with his feet dangling over the side. He got on the radiophone and he called, ‘Charlie, Charlie’ — that was his code name — ‘to Orange,’ and he gave the coí¶rdinates, and the planes came and bombed and all the while he was directing the planes with the radio. We looked down, and we could see another Army unit entering the town and then the guerrillas leaving from the other side.

“The next morning, the people came out of the town in a long column. You could see them winding their way up the hill in a long line, moving up to where Monterrosa was sitting on the same wall, leaning back, looking halfway between a king and a hero. And, one by one, the peasants passed in front of him, and each of them had an offering. One of them would give him an egg, another some tortillas, another would push forward a young boy to sign up. And Monterrosa would motion to an aide, as he reclined there like a Roman emperor. I remember a father carrying a little boy who had his head covered with a white handkerchief, and then when he came in front of Monterrosa the father unveiled the kid’s head and you could see he had this big growth on his face. And Monterrosa nodded to an aide. The aide grabbed the radio and called in the helicopter to take the kid to the hospital in the city.”

By 1983, Monterrosa’s new tactics had begun to show some success. “He changed the way he related to the local population, and he was less arrogant in his military stance toward us,” Villalobos, the E.R.P. comandante, told me. “There was this first stage, I think, in which he executed the massacres not only because it formed part of his military training and it was tactically approved by the High Command but also because he didn’t think it would become a political problem. Then, later, he realized that this sort of tactic didn’t work. It did not produce a quick military victory.”

Annunziata agreed. “He was not bloodthirsty, but he was so neurotically driven — he wanted at all costs to win the war,” she said. “The point was to create a turning point, a watershed, to turn the tide, and to do it by scaring the hell out of the enemy. It was a deliberate demonstration of cruelty to show them that the guerrillas couldn’t protect them. And he understood that you do this as cruelly, as brutally as possible; you rape, impale, whatever, to show them the cost.”

To most of the reporters who covered him now — few of whom had been in the country in 1981 — El Mozote was just a distant rumor, a dark echo from the past. “He was the press-corps officer, you know, very personable,” Jon Lee Anderson, who was reporting for Time magazine, said, “but there was always this buzz that he was responsible for El Mozote, and, of course, he always denied it.” By this time, Monterrosa had a mistress in the press corps — a beautiful young Salvadoran woman who worked for an American television network. Annunziata recalls, “He would helicopter in to the Camino Real” — the San Salvador hotel favored by the international press — “to visit her, and he would burst through the door of the press offices in his combat fatigues and come over and look over your shoulder at what you were writing and say, ‘Have you written about me today?’ ” Monterrosa’s girlfriend let her colleagues know — speaking in all confidence, of course — that there had been “a problem” with the El Mozote operation, and although, for understandable reasons, she wasn’t free to go into details, all one had to know was that on that particular day the Colonel had unfortunately “lost radio contact” with his men — with regrettable consequences.

The guerrillas did not find this story very convincing. “He was well known to all the guerrillas as the man who had ordered the massacre,” Licho said. “Everybody wanted to kill him in combat.” Now, however, their adversary had begun doing what they themselves knew was the most effective thing to do in order to win the war: “political work” in the countryside. “He started learning; he began to play football with the people, help their families. We realized that for someone as militarily talented as he was to start to do real political work could be very dangerous. I think it was at the beginning of 1983 that we started making plans to kill him.”

“Villalobos and Monterrosa were obsessed with each other’s psychology,” Annunziata said. “For Monterrosa, it was like looking in a mirror. He had this obsession with the guerrillas — with knowing them, understanding them. He had studied all the different groups, and claimed he could always tell which one had staged an operation. He felt he was the alter ego of the guerrillas. Every night, out in the field, he would listen to the radio, first to the BBC and then to Radio Venceremos, listening to what they said he’d done that day. Every night, you could hear, coming from his hammock, the ‘Internationale’ playing over Radio Venceremos.”

As it happened, Monterrosa’s fascination with Radio Venceremos — his capture of the transmitter, after all, had been the high point of Operation Rescue — had not escaped the notice of his alter ego. “A basic principle of warfare is to study the psychology of the enemy commanders,” Villalobos said. “Monterrosa was obsessed with war trophies. He got personally involved in combat situations when his men captured something — to such an extent that at times he lost the ability to coí¶rdinate troop movements. Once, he arrived personally to take charge of video records that they had captured from us. Another time, it was a scale model that we had used to plan an attack. Each time, he came himself. And he was desperate to stop Radio Venceremos. I mean, any cassette or tape recorder he found was turned into a great victory.”

Increasingly, in late 1983 and on into 1984, Monterrosa had victories to celebrate. His “beans and bullets” campaign was making progress in Morazán. The area under F.M.L.N. control was gradually shrinking, and so, even more critically, was the guerrillas’ manpower base. By the summer of 1984, reports had begun circulating that the guerrillas were reduced to conscripting civilians into their ranks.

That September, the Americans gave the Army ten new Huey helicopters. “When I saw that news about the helicopters, I told a friend, ‘Monterrosa will be coming after us,’ ” Villalobos said. ” ‘He will use those helicopters to attack the command post.’ ” The delivery of the helicopters, not coincidentally, came at the time of a major diplomatic initiative by the Salvadoran government. On October 15th, President José Napoleón Duarte — he had won an election earlier that year, and replaced Magaña — joined Minister of Defense Carlos Eugenio Vides Casanova and other government representatives in a meeting with guerrilla leaders (among them Guillermo Ungo, who had been Duarte’s Vice-Presidential running mate in the stolen elections of 1972), in La Palma, about eighty miles west of Morazán.

Three days after the meeting, Monterrosa launched a major offensive in Morazán, a six-thousand-man sweep called Torola IV. “The war goes on,” he told James LeMoyne, of the New York Times, as they stood at the base at Osicala, watching the men of the Atlacatl board the new Huey helicopters and lift off into the northern sky. “There are times when you have to make war to gain peace.”

Villalobos would be coí¶rdinating the response to Torola IV, a campaign that, in its broad outline, appeared much like Operation Rescue, three years before: soldiers from regular Army units were storming north across the Torola, and the helicopter-borne men of the Atlacatl were moving down from Perquí­n and other mountain towns. This time, however, the guerrillas’ response would be somewhat different.

Villalobos and his staff were hard at work planning an ambush for the town of Joateca, a few miles east of El Mozote. It was a well-planned ambush — they had devoted many hours to its preparation — but an unusual one: the guerrillas were planning to have the Army ambush them and thus “capture” a prize that they very much wanted Monterrosa to claim.

On October 22nd, Monterrosa helicoptered in to Joateca. With him was, among others, Jon Lee Anderson, of Time. “It was real air-mobile ops,” Anderson told me in an interview. “Flying around from one place to another, inserting troops, choppering around, moving several times a day.” In Joateca, he said, an advance platoon had flushed out the guerrillas the day before. Now the people were gathered there, waiting for Monterrosa. “It was this turfy plaza in this ramshackle old hamlet — you know, cobblestones, shaded front porches — and he gathered the townspeople around and gave them this hearts-and-minds sort of speech. He was sitting at a table with a microphone in his hand, and he had a woman social worker and a civilian psychologist there beside him.”

Anderson quotes Monterrosa as telling the peasants, “We are your true brothers. We’re not the caretakers of the rich. Do you see any rich among us? We give our blood to the soil, but it’s up to you to make it fertile.”

Around that time, not far from the plaza where Monterrosa was speaking, his men had pounced on a group of hapless guerrillas. “We sent a column of our fighters to fall into an ambush,” Villalobos said, “and then they were supposed to leave the transmitter, as if, you know, they’d had to abandon it” — as they’d had to do three years before. “But it didn’t work out that way. We weren’t able to get the transmitter up to where the combat took place. We were upset — we thought we had blown the operation. I mean, they should have been suspicious.”

The rebels had left the transmitter near a graveyard on the outskirts of Joateca. Not far away, Villalobos and his men were waiting tensely, listening intently to their radios. Suddenly, they heard soldiers begin to talk excitedly to one another. “As soon as they found the transmitter, there was a big celebration,” Villalobos said. “We could hear them talking about all the prizes they would get, and so on.” The soldiers began to congratulate one another, speculating on how happy the Colonel would be when his men brought him this priceless treasure. There was not a hint of suspicion. “Just as vanity blinded Monterrosa, it blinded his soldiers as well,” Villalobos said. “We just had to wait for his personal psychology to play itself out.”

Late in the afternoon, Jon Lee Anderson sat down to interview the Colonel. “He disappeared for a while, and then he came back very excited,” Anderson said. “He sat down next to me on the stoop of this old peasant house, and he confided to me that he thought he’d found the transmitter. It was in this graveyard, in a cemetery at the edge of this little hamlet. This was somewhat far from where things were happening, it’s true, but the town had definitely been theirs — I mean, there were graffiti everywhere.”

Anderson seized that moment to ask Monterrosa about the rumors that still clung to him about what had happened at El Mozote. “It was late, and we were sitting there, just the two of us, and I said, ‘Colonel, qué pasó en El Mozote?’ And there was this long pause, and he looked away, and finally he said, ‘No es como dicen’ — ‘It’s not like they say.’ ”

Monterrosa would say no more, but Anderson took his answer as a tacit confirmation that Monterrosa had been involved in the massacre. Shortly before, James LeMoyne had asked Monterrosa the same question, and, according to LeMoyne, the Colonel, in the aftermath of a long and exhausting day of combat, had answered more bluntly. “He shrugged and said, ‘Yeah, we did it. We carried out a limpieza there. We killed everyone,’ ” LeMoyne told me. “He said, ‘In those days, I thought that was what we had to do to win the war. And I was wrong.’ ”

Late that evening, Anderson and his photographer left, somewhat regretfully, for the capital. They needed to file their stories, but they intended to rejoin Monterrosa in his chopper the following day. The next day, three senior officers and a three-man Army television crew arrived in Joateca. Along with a local priest and sacristan, they planned to accompany the victorious Colonel as he carried his prize back to the capital. It was to be a triumphal entrance. The capture of the transmitter was an enormous propaganda victory, and Monterrosa wanted to film it, record it, publicize it — to milk it for all it was worth.

The men climbed aboard the helicopter and took their seats, and as the rotors roared overhead soldiers began loading the equipment aboard. Sitting in the place of honor beside Monterrosa, as it happened, was Todd Greentree, of the United States Embassy. “We were sitting together,” Greentree said. “He was buckling in, and people were stowing aboard all these duffelbags that belonged to different soldiers — you know, ‘Take this back to my wife in San Salvador.’ The transmitter must have been in one of those. Then a soldier came over to Monterrosa to tell him he had a radio call, and he got off to take it.”

Greentree was in a great hurry to get back to the capital — he has long since forgotten why. “I saw that another helicopter was getting ready to take off, and I was in such a hurry to get back that I got off and climbed aboard.”

On a hill northwest of the town, the guerrillas of the E.R.P. watched excitedly as the Huey slowly rose above the tree line. They waited until it had reached its apogee, pointed a remote-control device in a direct line of sight, and pressed the button. Nothing happened. “We didn’t know what had gone wrong,” Villalobos said. “We thought we had a malfunction. Then we heard his press conference” — Monterrosa was apparently being interviewed by radio, announcing his destruction of Radio Venceremos — “and we realized that it was the wrong helicopter.”

They sat tensely on the hill deep into the afternoon, until at last, after what must have seemed an interminable wait, a second helicopter climbed above the treetops and lofted into space. The big aircraft rose high over Joateca, turned, and began to head west, toward the Sapo River — toward the tiny hamlet of El Mozote. Poised high in the blue sky, it caught the sun. Far below, a man from Perquí­n gazed upward, squinted, and then saw the machine of war — he had seen such machines so many times over Morazán — suddenly blossom into a great orange-and-black fireball; and then he was deafened by the explosion.

The man, who had been forced to guide Monterrosa’s men on their limpieza three years before, said, “I remember thinking, If only he had gone a few minutes more, his blood would have been mixed with the soil of El Mozote.”

Monterrosa was five years dead before the exiles returned to Morazán. Crowded into the trucks and buses that came over the mountains from the Honduran refugee camps, they flooded back into the deserted villages and hamlets of the red zone. The Salvadoran government could do nothing to stop them, for it was November, 1989, and across the country the guerrillas had unleashed a general offensive that, in the political shock it provoked, would turn out to be the Salvadoran equivalent of Tet: it would put an end to the long civil war.

The fighting was especially brutal in San Salvador, where guerrillas dug themselves in in the crowded slums, and the military managed to extract them only by bombing and strafing civilian neighborhoods. But the turning point of the offensive, and of the war itself, came during the early hours of November 16th, when commandos scaled the back wall of the shady campus of the University of Central America, roused five Jesuit priests from sleep, ordered them to lie with their faces against the ground, and emptied automatic weapons into their brains. Before they departed, the soldiers killed a sixth priest, the Jesuits’ cook, and her fifteen-year-old daughter. The scene they left behind — the obliterated skulls of the priests, the green lawn soaked in blood and brains, the fantastically redundant number of spent cartridges — was one of spectacular carnage. And though the soldiers made a halfhearted attempt to scrawl a few leftist slogans, it would very shortly become clear that those who had done this work were the men of the Atlacatl.

It was an enormous political blunder, for it said to the world, and especially to the Americans in Congress, that after the billions and billions of dollars and all the fine words about “training” and “reform,” at bottom the Salvadoran Army remained what it had been at El Mozote. But by now Ronald Reagan had gone, and so had the ideological threat he had so feared. The time had come to bring the war to an end.

In the mountains of Morazán, in what was still the red zone, the refugees rebuilt their community. In the Honduran camps, they had made friends among the international aid workers, and now, with help from the European Community and other agencies, they raised up new buildings of straight brown planks: a shoe factory, a handicraft shop, a nursery to hold the children during the day when the people went to work. And they named their community Segundo Montes, after one of the fallen Jesuits.

On October 26, 1990, Pedro Chicas Romero, of La Joya, who had hidden in a cave above the hamlet as the soldiers killed his relatives and his neighbors, went down to San Francisco Gotera and filed a criminal complaint with the Court of the First Instance, accusing the Atlacatl Battalion of responsibility for the killings in El Mozote and the villages around it, and asking that Judge Federico Ernesto Portillo Campos investigate and punish those responsible. Among the first witnesses to give testimony in the case was Rufina Amaya Márquez.

The investigation proceeded haltingly, and although Tutela Legal, among other human-rights organizations, tried to push the Judge forward — by publishing, in November, 1991, the first full investigation of the El Mozote massacre, including the names of seven hundred and ninety-four dead — it is hard to know what might have come of it had not the government of President Alfredo Cristiani and the comandantes of the F.M.L.N. come together, in Mexico City in January, 1992, and signed an agreement to end the twelve-year-old war. Among other things, the agreement provided that the Army be purged of “known human rights violators” and reduced by half; that the guerrillas disarm and some of their number join a new civilian police force; and that the Atlacatl and the other rapid-reaction battalions be disbanded. The agreement also provided for a “Truth Commission” that would take on “the task of investigating serious acts of violence that have occurred since 1980 and whose impact on society urgently demands that the public should know the truth.”

The experts from the Argentine Forensic Anthropology Unit entered the country in February, and although the investigation was repeatedly stalled, the people of Morazán helped it along (among other things, by staging a boisterous demonstration in front of the Gotera courthouse in April), and so did the three Truth Commissioners when they arrived, in June.

Finally, in October, the experts began to dig. And there, on the third day, in the silence of the ruined hamlet of El Mozote, all the words and claims and counterclaims that had been loudly made for nearly eleven years abruptly gave way before the mute force of material fact. The bones were there, the cartridges were there; the sleeping reality of El Mozote had finally been awoken.

They dug and sifted and charted for thirty-five days, and soon the cartridges and the clothing and the bones and bone fragments, all labelled and packed away in bright manila envelopes and fresh new cartons, would depart El Mozote and travel by car to a laboratory in San Salvador, where the experts worked away into December. The following March, when the United Nations made public the Truth Commission’s report, entitled “From Madness to Hope: The 12-Year War in El Salvador,” the analysis of the evidence was there, laid out for the reader in clear, precise language, each successive sentence demolishing one or another of the myths put forward during the previous twelve years. Of the hundred and forty-three skulls found, all “were deposited during the same temporal event,” which is “unlikely to have occurred later than 1981.” El Mozote could not have been a guerrilla graveyard, as some had claimed, especially since all but twelve of the one hundred and forty-three remains identified turned out to be those of children under twelve years of age, including at least one fetus, found between the pelvic bones of one of the adults.

The cartridges recovered in the sacristy showed that “at least twenty-four people participated in the shooting,” and the distribution of the shells indicated that they fired “from within the house, from the doorway, and probably through a window to the right of the door.” Finally, of the two hundred and forty-five cartridge cases that were studied — all but one from American M16 rifles — “184 had discernible headstamps, identifying the ammunition as having been manufactured for the United States Government at Lake City, Missouri.”

From this evidence and from a wealth of testimony, the Truth Commission would conclude that “more than 500 identified victims perished at El Mozote and in the other villages. Many other victims have not been identified.” To identify them would likely require more exhumations — at other sites in El Mozote, as well as in La Joya and in the other hamlets where the killing took place. But the Truth Commission has finished its report, and, five days after the report was published, the Salvadoran legislature pushed through a blanket amnesty that would bar from prosecution those responsible for El Mozote and other atrocities of the civil war. In view of this, Judge Portillo, after allowing two American anthropologists to work in the hamlet for several weeks with inconclusive results, in effect closed down his investigation. The other victims of El Mozote will continue to lie undisturbed in the soil of Morazán.

Last July, the Secretary of State’s Panel on El Salvador, created in the wake of the Truth Commission report, concluded that the Department’s handling of the massacre investigation “undermined the Department’s credibility with its critics — and probably with the Salvadorans — in a serious way that has not healed.” The panel concluded its review by noting that “a massacre had indeed occurred and the U.S. statements on the case were wrong. On December 11, 1992, two Embassy officers went to El Mozote to attend a ceremony honoring those who had died in the massacre.” Only the Wall Street Journal remained more circumspect; in February, in a report from El Mozote on its editorial page, entitled “The War’s Over, but El Salvador Still Fights Propaganda Battle,” the Journal conceded that while “it appears that a massacre of some kind took place, questions remain,” including, the Journal said, “Who were the true perpetrators of this awful crime?”

If you drive out from San Salvador today, along the highway toward Morazán, passing the barracks of the Domingo Monterrosa Third Brigade, and crossing the narrow bridge on the Torola, its wooden planks clattering beneath your wheels, you will find, amid the sorghum and the corn and the tufts of maguey, the clean new buildings of Segundo Montes, housing the boot factory and the handicraft shop and the other factories brought back from the refugee camps. In one of the buildings, you will find the woman who fled La Joya in 1981, was forced to bury her wounded child in the mountains, went mad, and became the witch of El Mozote that the villagers came to fear. Andrea Márquez works in the nursery, caring for the children of Segundo Montes. Farther up the black road, if you step through the barbed wire you will find Rufina Amaya living in a small house with her little girl, Marta, who is now four years old. And if you head up the black road to Perquí­n, with its battered central square and its mural of the slain Archbishop Romero, you will come to Radio Venceremos, which has graduated from its various holes in the ground to an actual building on a nearby hill; concrete, single story, and small, it is a museum now, a gallery to exhibit pictures of the station’s former subterranean quarters. Out in front, beside a well-preserved bomb crater with a carefully tended stone-and-flower border, and behind a brass plaque, you will find a dramatically twisted and burned torso of steel. As the people there will tell you, it is what remains of a helicopter that was blown from the sky one fine day, and it happens to be the most cherished monument in all Morazán.