by JOHN ROOSA
IN A SCENE IN the genre-bending documentary film The Act of Killing (2012), a journalist,
Soaduon Siregar, claims to have a sudden insight while standing in a film studio.
Having just watched two executioners reenact their deeds of 1965–66 before a
camera, he realizes why he had never known that they had garroted hundreds of detainees
when he was working with them in the same building. His old friends had been so
“smooth” (he used the English word) that he had not noticed them running a human abattoir
upstairs. One of the executioners standing in the studio with crude makeup plastered
on his face, Adi Zulkardy, is incredulous and insists Siregar must have known
since even the neighbors heard the screams of the victims: “It was an open secret
(rahasia umum).” When the director of the film, Joshua Oppenheimer, interjects,
noting that Siregar’s longtime boss at the local newspaper had already admitted to
being an interrogator who decided which detainees were to die, Siregar becomes only
more adamant in denying any knowledge of the killing.
This exchange, ending in the abject embarrassment of Siregar, neatly exemplifies the
confusing dynamics of an open secret. These executioners in the North Sumatran city of
Medan knew at the time that their work was meant to be partly secret. They received
batches of tied-up captives from the army, murdered them in the back part of a building,
placed the corpses into sacks, trucked the sacks to a nearby bridge, and threw them into
the river below. The victims just disappeared. There were no midday public spectacles in
the central square of the city. No information was given to the relatives as to what had
happened to their loved ones. Yet these perpetrators did not fully conceal their work
either. Their building stood on a busy commercial street in the city’s center. Neighbors
and passers-by could see detainees being brought there, hear the screams, and catch
glimpses of the sacks being taken out. The perpetrators wanted to be discreet but saw
no need for absolute secrecy. After all, they wanted to demonstrate their power and terrorize
the public. As members of the youth wing of the army’s own political party, they
prided themselves on being “free men” (preman), able to do whatever they wished.
The film reveals, however, their lack of freedom in narrating their acts of killing. The pathetic
Siregar, in steadfastly keeping the secret, was only doing what he thought someone
complicit in mass murder should be doing in front of a filmmaker, especially a foreign
filmmaker. It was startling for him to find Zulkardy and the other executioner, the flamboyant
star of the film, Anwar Congo, flouting the conventional silence about the
executions. The film is, in one sense, a narrative of Anwar Congo, the “free man,” the
small-time mafia boss aspiring to eternal celluloid glory, claiming the freedom to
narrate events that were never supposed to be narrated in front of cameras—and then
regretting at the end that he had tried.
As an open secret, the slaughter of 1965–66 has been a difficult event to understand.
So much information has remained hidden that basic questions have not been able to be
answered with any precision or confidence. Consider the question: who were the perpetrators?
The Act of Killing clearly shows that, in the case of Medan, civilian militiamen
were the perpetrators. But the film’s opening lines of text mention that the army was responsible
for the anti-communist slaughter. The film, with a focus on the militiamen’s
memory of the event rather than the event itself, does not explicate their exact relationship
with the army. Some viewers have come away from the film with the mistaken impression
that the army was not responsible, though Oppenheimer thought that the
references to the army in the film were sufficient to convey the idea that the army had
“outsourced” the killing to the militiamen (Melvin 2013).
The scholarly literature is marked by an ambiguity about the identity of the perpetrators.
Historians have had a hard time understanding the respective roles of the army
and the civilian organizations during the anti-communist violence. Did the army decide to
launch the slaughter and then order about the militiamen like pawns or did it capitulate in
the face of popular pressure from below? As Robert Cribb (2001, 235) has put it, “the
most intractable difficulty” in understanding the killing has been the determination of
“the relative importance of army initiative and local tension.” The answer to this question
is important since it determines whether to categorize the event as a case of spontaneous,
horizontal violence, of “neighbor killing neighbor,” or as a case of bureaucratic, vertical
violence, of the state massacring its own citizens. Was this a case of chaotic violence,
of people “running amok,” or was it case of a well-organized political genocide, or something
in between? Historians have tended to argue that the violence was committed by a
combination of army personnel and civilian militias, with the role of each varying by
region; in some regions of the archipelago, the army took the lead; in other regions,
the militias took the lead. This argument, what can be called the dualistic thesis,
denies an overarching national pattern to the killings and rejects the idea that the killings
were the responsibility of a single person, group of people, or institution.
Since the collapse of the thirty-two-year-old Suharto dictatorship in 1998, the mass
killings have been more widely discussed. More secrets have been opened up. While
there have been no trials of the perpetrators and no truth and reconciliation commission,
there have been many unofficial investigations and public forums, such as book launchings,
film screenings, academic conferences, public lectures, television talk shows, and a
few exhumations and markings of mass graves. Researchers in a variety of provinces have
uncovered new information. Journalists have been publishing many articles about the killings.
The editors of the country’s main weekly news magazine, Tempo, after watching The
Act of Killing, organized a team of journalists to interview civilian perpetrators, the equivalents
of Anwar Congo, all around the country. Tempo’s special issue in October 2012 sold
out immediately, and photocopies were sold at a premium on the streets (Tempo 2012). It
was the first time the stories of the executioners had appeared in the magazine, after
forty-one years of publication.
The new research has only scratched the surface of a complicated set of events that
spanned many months across a sprawling archipelagic nation of some 100 million people
in the mid-1960s. Compared with the literature on other cases of genocide in the twentieth
century, the literature on this political genocide in Indonesia is still remarkably
sparse and underdeveloped (Cribb 2009). The new research, however, has allowed
some patterns to emerge more clearly, such as the pattern of the perpetrators hiding
the traces of the massacres. There is more evidence now explaining why there has
been an absence of evidence. Many of the victims, from Aceh in the west to Flores in
the east, were disappeared in remarkably similar ways. The new research suggests that
the conventional wisdom of the dualistic thesis needs to be rethought. One faction
of the army high command in Jakarta, Suharto and his clique of officers, appear to
have had much greater agency than previously assumed in organizing the killings.
THE OFFICIAL LINE: SPONTANEOUS VIOLENCE
Primary sources about the killings have been scarce, and researchers have had to sift
through evidence that has been anecdotal, fragmentary, and unreliable when not downright
fraudulent (Roosa 2013). The Indonesian-language newspapers, the proverbial location
for the first draft of history, did not report on the executions even when, as
revealed in the case of Medan, some journalists were serving as executioners and had
first-hand knowledge. There must have been thousands of massacres across the
country, yet not a single photograph of them is known to exist (Strassler 2005). The
trail of official documentation is frustratingly meager; only a few Indonesian government
documents pertaining to the slaughter have surfaced (Indonesia 1986). Most of the
primary sources are reports of foreign journalists and foreign diplomats who were in Indonesia
at the time (Schaefer and Wardaya 2013; Simpson 2008).
The regime of General Suharto, from 1966 to 1998, preferred to keep quiet about
the anti-communist murders. It never commissioned a book to explain, justify, or deny
them. Instead, it treated them as nonevents, as if they had never happened or were
not worth mentioning. The state-sponsored, six-volume The History of Indonesia,
written by the country’s leading historians of the 1970s and 1980s, contains just one sentence
about the mass killings—a sentence that is factually incorrect and hopelessly ambiguous:
“Only in East Java and Bali arose the chaos of abductions and killings, which
were successfully brought to order again” (Poesponegoro and Notosusanto 1990,
6:403–4). The sentence avoids identifying both the victims and perpetrators and
implies civilians spontaneously attacked each other until the state intervened to stop
them. All the Suharto regime’s narratives of history celebrated the “crushing” (penumpasan)
of the Communist Party of Indonesia (Partai Komunis Indonesia or PKI) while carefully
avoiding mention of bloodshed (Roosa 2012). History textbooks used in the schools,
still today, do not even mention that there were “abductions and killings,” much less massacres,
in the course of “crushing” the PKI (Leksana 2009).
State propaganda since late 1965 has focused on a single event: the September 30th
Movement. Allegedly, the PKI masterminded that movement of military personnel who
abducted and executed six army generals in Jakarta in early October 1965. Suharto’s army
held every member of the party, and every member of organizations loosely affiliated with
the party, to be responsible for that brief, quickly suppressed uprising against the army
high command (Roosa 2006). One history lesson that every Indonesian schoolchild has
learned is that the PKI needed to be “crushed” because it was responsible for a horrific,
treasonous act of barbarism. The September 30th Movement killed twelve people (four
of them unintended targets) and lasted only twenty-four hours in Jakarta. But the magic
narrative arts of psychological warfare specialists transformed it into a massive, nationwide
revolt and the worst evil ever to befall Indonesia. Schoolchildren learn a lot
about the September 30th Movement but nothing about the hundreds of thousands of
people murdered in the name of suppressing the movement.
It is obvious, given the vicious anti-communist propaganda the army issued every day
in the press after October 1, 1965, that the army instigated people to take action against
the PKI. The army controlled the newspapers and the radio and kept up a steady stream
of largely fabricated stories about PKI brutality and treachery. The army urged people to
join the campaign to “crush” and “annihilate” (menumpas, mengganyang, membasmi) the
PKI. But the propaganda did not explicitly call upon people to kill communists. By itself,
the propaganda does not prove the army was responsible for the mass killings. One could
argue that Suharto’s army, intending only to arrest and detain large numbers of PKI supporters,
lost control of the operation; civilians, outside of army direction, took advantage
of that time of crisis to kill people who had been threatening the security of their property
and challenging their authority.
At the time of the killings, the Indonesian army generals tended to disavow any responsibility
for them. Even when reporting to President Sukarno, their
commander-in-chief, they claimed that they were doing their best to stop the killings,
not commit them. The army ensured that Sukarno’s Fact Finding Commission (FFC),
created in late December 1965 to investigate the killings, was perfunctory and exculpatory.
US Ambassador Marshall Green reported to the State Department that the army
would arrange for the FFC to die a “natural death” (Green 1965). The FFC did not
die but it might as well have: its brief final report, dated January 10, 1966, contained
very few facts. The report presented a grand historical narrative structured in a tripartite
formula of prologue-event-epilogue. According to this narrative, the communists had antagonized
good, patriotic, god-fearing Indonesians for decades (“the prologue”), and
then, by launching a revolt called the September 30th Movement (“the event”), had provoked
them into a righteous fury (“the epilogue”). The mass killings were simply the
natural, inevitable response to the PKI’s revolt. According to the FFC, the army tried
to stop the killing but could not contain the masses who were determined to release
their pent-up rage and exact vengeance.1
While blaming overly emotional civilians for the killings, the FFC report included
one telling qualification. It admitted, without citing specific cases, that some state officials
had carried out forced disappearances: “[T]here were executions of September 30th
Movement individuals by government agencies that were not officially announced to
their family members.” It then proceeded to justify the tactic of disappearing detainees:
“[S]uch executions can be considered normal in security operations done in the context of
1Photocopies of the FFC report circulated among researchers after 1966. To my knowledge, it was
first published in 1995 as an appendix to the autobiography of Oei Tjoe Tat, a minister in Sukarno’s
cabinet who had served on the commission (Oei Tjoe Tat 1995, 348–66). suppressing a rebellion.” But the report added a reservation. Not making an “official announcement”
made it seem as if the executions were “wild” (liar) actions that were not
under the state’s control (Oei Tjoe Tat 1995, 355). Such a reservation was not serious
enough to convince the army to stop the practice.
By not dismissing the commander of the army, Major General Suharto, who had so
obviously failed in his assigned task of “restoring order,” Sukarno implicitly sanctioned
Suharto’s narrative—that the army was doing its best to stop an overwhelming wave of
popular violence. He neither discredited that narrative nor articulated an alternative
one of his own. Sukarno had been the source of all truth in Indonesian politics since
the start of his one-man polity, Guided Democracy, in 1959. Politicians competed
with one another to prove their loyalty to him. He was “the extension of the tongue
of the people” (penambung lidah rakjat) whose annual Independence Day speeches articulated
the existential meaning of the Indonesian nation—where it had come from,
where it was, and where it was going. But in the face of the massacres he was strangely
mute; he did not offer the people a way to understand the terror going on all
Sukarno never released the FFC report to the public. In an address before his
cabinet and the press five days after receiving the FFC report, he announced only the
report’s estimate on the number of people killed. He turned to the press corps: “[H]ey
journalists, you have to be careful what you write now.…[T]he Fact Finding Commission
found that 87,000—men and women—have been killed!” (Setiyono and Triyana 2003,
362). That was the figure that newspapers around the world reported (New York
Times 1966). It hardly matters that Sukarno carelessly misread the number. The FFC’s
actual figure, 78,500, was wholly unreliable; it was not based on a detailed accounting.
Sukarno may have been careless with the number because he knew it was inaccurate;
he had been informed by his one loyalist in the FFC that the real number of victims
was probably six times higher (Oei Tjoe Tat 1995, 192).
Sukarno confessed at the same meeting on January 15, 1966, that he had been frequently
crying and lost in introspection during the previous weeks, wondering if he had
done anything wrong. He asked rhetorically: “Do we have to redeem the killing of a
number of generals [by the September 30th Movement] with the killing of tens of thousands,
maybe even hundreds of thousands, of people of our own nation?” (Setiyono and
Triyana 2003, 362). The only way he could make sense of the killing was as a civil war that
was being instigated by foreign imperial powers intent on overthrowing him. But he did
not identify the Indonesians collaborating with those foreign powers. He did not accuse
Suharto or any other army general of organizing massacres. In his final Independence
Day speech on August 17, 1966, by which time Suharto had already reduced him to a
figurehead president, unable to control the appointment of cabinet ministers, he did
not mention the killings—an odd omission for a speech titled “Never Abandon
History” (Sukarno 1966).
The officials of Suharto’s new regime, while not publishing the FFC report or any
other report about the killings, occasionally mentioned the killings in offhanded, impromptu
comments in later years. The narrative of spontaneous popular violence to suppress
a PKI revolt was the official line even if it was never explicitly proclaimed as such.
Since 1998, human rights organizations and victims’ organizations have demanded a truth
and reconciliation commission and an official apology (Komnas Perempuan 2007). The
government’s Human Rights Commission has called for the perpetrators to be prosecuted
(Komnas HAM 2012). Most state officials respond to such demands by reverting to
the old narrative, insisting that whatever violence was committed was justifiable, either
as an unstoppable expression of the people’s hatred of the PKI or as a necessary part
of the military campaign to suppress the PKI’s revolt. The PKI started the violence,
and the people and the army needed to use violence to stop the PKI. Officials do not acknowledge
that there were any human rights violations for which the government today
has to apologize. As Vice-President Jusuf Kalla has stated, “The first victims were our generals.
They [the killers of the generals] should ask forgiveness from us” (Kompas 2015).
THE HISTORIANS’ DUALISTIC THESIS
The most frequently cited primary source, Indonesian Upheaval, a book written by
the British-American journalist John Hughes who was reporting from Jakarta in 1965–66,
initiated the dualistic thesis. Hughes claimed that the army was responsible for the killings
in some areas while the civilian militias were responsible in other areas. The army in
Central Java “seems to have exerted broad control over the bloodbath, although many
civilians were also recruited to kill Communists” (Hughes  2002, 165–66). In
East Java, the executions were ordered by the army but were “largely handed over to civilians”
to carry out (166). In the area around Medan in North Sumatra, the army commander,
Brigadier General Kemal Idris, was instructed by his superiors in Jakarta to
“clean up” the communists: “Nobody needed to spell out what ‘cleaning up’ meant”
(154). Bali was an exception to this pattern of army control over the anti-PKI violence.
There, Hughes claimed, the civilians embarked on a “frenzy” of indiscriminate killing
until the army intervened to force them to be more selective in their targets (183–92).
The army “had to call the people off” in Bali (197).
Historians have often resorted to the use of the passive voice and middle voice to
avoid assigning agency. Merle Ricklefs (1993, 287) wrote in his textbook A History of
Modern Indonesia Since c. 1300: “In October 1965 the killings started. Violence
against people associated with the PKI took place across the country, but the worst massacres
were in Java and Bali.” Killings started, violence took place, and massacres happened.
Ricklefs did venture two sentences with active verbs: in Central Java, the army
“assisted youths in finding Communists” but stood by in Bali as “upper caste PNI landlords
took the lead in urging the extermination of PKI members” (288). One will notice
the lack of an active verb to express the act of killing; the army “triggered” and “assisted”
the killings, and the Balinese civilians merely “urged” that the PKI be killed. The reader is
left wondering who actually did the killing and how they did it. The victims just happened
to die: “PKI supporters fell” (288). To the extent that Ricklefs identified the respective
roles of the army and the militiamen, he presented the army as merely instigating and
assisting the anti-communist civilians to commit murders that they already wanted to
commit. In Java as a whole, the killing was supposedly a result of a much older conflict
between devout Muslims (santri) and nominal Muslims (abangan). The army “supported
and encouraged zealots from the santri side of Javanese society in finding PKI targets
among their abangan brothers” (288).
The idea of the army leading the killing in some areas and the civilians leading it in
other areas has been accepted by a variety of scholars. An early analysis of the killing in
the late 1970s by Harold Crouch, an expert on the Indonesian army’s rise to power, presented
the killing in Central Java as having “remained under army control” (Crouch
 1988, 151). But in East Java, Crouch, contradicting Hughes, thought that civilians,
not the army, “took the lead in attacking the PKI” (147). In Bali too, “army officers gave at
least tacit approval to the plans of the PKI’s civilian opponents, but the massacre quickly
ran out of hand” (152). In Aceh, the anti-communist civilians, supposedly animated by a
religious fervor against atheism, moved against the PKI before the army swung into
action (142–43). Robert Cribb concludes from several local studies written in the
1990s that there was considerable variation: in some districts, “the military did most of
the killing,” in others “local men of violence” did (Cribb 2001, 237). He suggested that
in some districts, the civilian participation was so extensive that “the killing was truly collective”
(Cribb 2010, 452–53). Anthropologists conducting fieldwork in specific locales
have tended to uncover cases of army-organized massacres and disappearances (Beatty
2009, 50–53; Bowen 1991, 119–22; Hefner 1990, 209–15) but such case studies have
been easily accommodated by the dualistic thesis. Adrian Vickers, in his masterful textbook
on Indonesian history, has summed up the common wisdom: “In many areas the
army guided civilian groups as to who should be targeted or sponsored local militias,
but in other places vigilante action preceded the army” (Vickers 2005, 157–58).
A recent study based on a thorough reading of the European-language literature by a
nonspecialist of Indonesian history arrives at the same conclusion: the mass killing was
committed by a “coalition” of army personnel and civilians, with the relative importance
of either side varying by region. Christian Gerlach, a German researcher, has written a
profusely footnoted analysis that presents the army’s role as “indispensable” but
limited. He emphasizes the “participatory character” of the violence, citing Aceh, East
Java, and Bali as regions where anti-communist civilians proceeded on their own to
attack communist civilians. His main thesis is that the violence largely welled up from
the conflicts within Indonesian society rather than being imposed from above by the
army (Gerlach 2010, 30–31). Indonesian society is one of a number of “extremely
violent societies”—as the title of Gerlach’s book puts it.
Indonesia’s leading historians affirm this dualistic thesis. Taufik Abdullah, with
funding from the Ministry of Education starting in 2005, led a large research project
on the events of 1965–66 and published three hefty volumes containing contributions
from a variety of historians. Abdullah and his fellow editors argue that there was no national
pattern to the killing; the timing and intensity of the violence varied from province
to province. The only conflicts they found worth studying were “local conflicts,” and those
were largely driven by struggles between civilian groups over power and resources. The
army’s role was subsidiary: it provided the anti-PKI civilian groups some logistical support
and a license to kill, but only rarely did the army organize and carry out the killing itself
(Abdullah, Abdurrachman, and Gunawan 2012, xii–xxix).
Civilian militias have tended to be confused in later years as to how much responsibility
they should assume for the executions. The leaders of Nahdlatul Ulama (NU), an
Islamic organization deeply involved in the massacres in East Java, have alternated
between presenting themselves as loyal servants of the state, dutifully following the
army’s orders, and as being the heroic warriors who single-handedly defeated the
villainous PKI on their own. One of the NU’s first book-length accounts of its role was
meant to prove that its militia, Banser, had put itself at the service of the army: “[T]he
crushing actions that involved Banser were always under the chain of command, the
orders, of the army” (Sunyoto 1996, i). That was in 1996 when the newly formed left-wing
party, the People’s Democratic Party, was denouncing the Suharto regime for the
1965–66 massacres. NU leaders were then worried that army officers were washing
their hands of the matter and shifting blame onto the NU. The organization’s official
line changed in 2013, when The Act of Killing and the special edition of Tempo were
being widely discussed. Then the NU leaders were concerned that the stories being
told by people who had committed or witnessed executions in East Java were portraying
Banser members as thoughtless thugs at the beck and call of the army. In 2013, the NU
leaders wanted to emphasize that the NU acted on its own, that it was not “used” (diperalat)
by the army: “[E]ven without being ordered by the military, NU by itself confronted
and fought against the PKI, because the PKI was an enemy with which it had been in
conflict for decades” (Mun’im DZ 2013, 19–20).
From these two NU books of 1996 and 2013, it is hard to know how to characterize
the relationship between the NU and the army. The former, unlike the latter, at least attempts
to provide some factual information about the violence, complete with dates,
place names, and the names of the individual men providing the information. The
book acknowledges that members of the NU paramilitary group, Banser, attacked PKI
offices and homes and killed PKI members they found there (Sunyoto 1996, 110–36).
It also describes how Banser members received batches of detainees from the army
for them to execute. For instance, one Banser member from a district near Kediri
claimed that his group was involved in capturing about 6,000 people (a number that
seems exaggerated), herding them to a makeshift camp in an open field, taking away
batches of thirty to forty detainees to a forest every night, executing them and then
burying the corpses in a mass grave (155–56). All five of the Banser members whose
stories about the executions appear in the book mention that the orders to execute detainees
came from army officers (153–60). The book is not entirely reliable; it implausibly
claims that these massacres only occurred after March 1966 when Suharto banned the
PKI, as if the banning of the party provided some kind of legal justification for extrajudicial
Two Australian scholars who have attempted to determine the NU’s role conclude
that the NU did not play a “subservient or subsidiary role in the killings” (Fealy and
McGregor 2012, 129). But then the pattern of killing they describe is one in which the
army determined whether or not detainees were to be massacred (120, 124–25).
While NU members enthusiastically supported the repression of the PKI (the mass
round-ups, the attacks on offices and houses, the beating and killing of individuals), it
is not clear that they were the ones who initiated and organized the disappearances of
The respective roles of the army and the civilian militias are discussed in Joshua
Oppenheimer’s second film about the social memory of the massacres in North
Sumatra, The Look of Silence (2014). In one scene, viewers watch two civilians in the
plantation belt around Medan reenact their butchery by a riverbank. One of them,
Amir Hasan, is a schoolteacher and amateur artist. He and his friend Inong describe
how the army delivered tied-up and blindfolded detainees by the truckload for them
to chop up with machetes and then push into the Snake River. They understood they
were used by the army as a façade: “the world would be angry,” they explain to Oppenheimer,
if it was clear that the army was responsible for the massacres. They admit their
role was to provide some plausibility to the army’s narrative about spontaneous violence
by volatile masses. Still, they seem boastful about their deeds, as if they proved their
manhood and gained greater power in the local community through such violence.
The film ends with Hasan’s widow and sons meeting a relative of one of his victims
and desperately denying they knew anything about Hasan’s work as an executioner,
even when they had talked about it with Oppenheimer before. In that particular time
and place, they sensed the open secret needed to be secret, though they knew it was
already out in the open.
CASE STUDIES: ACEH AND BALI
Within the literature on the 1965–66 killings, there is a consensus on the dualistic
thesis: in some regions the army took the lead, and in other regions the civilian militias
did. This consensus does not mean that the dualistic thesis is correct. The consensus
may exist simply because the evidence is still too scarce and inconclusive to allow for a
more refined and accurate analysis. One way to test its validity is to examine the killing
in the two provinces where the killings, according to nearly all writers, were largely committed
by civilians: Aceh and Bali. The Muslims of Aceh and the Hindus of Bali were supposedly
infuriated by the PKI’s lack of respect for their venerated religious traditions and
were unable to control their fury. More recent studies of Aceh and Bali suggest that this
characterization of the killing is entirely incorrect.
The research of Jessica Melvin in Aceh convincingly demonstrates that the army
commander for Aceh, Brigadier General Ishak Djuarsa, initiated the killings by touring
the province in early October 1965 and ordering his subordinates to arrange the
murder of people who were affiliated with the PKI.2 Melvin discovered army records
being held in the archive of the provincial government and interviewed over seventy individuals.
The army records clearly show army commanders pressuring the noncommunist
organizations in the province to join the campaign to murder suspected communists
(Melvin 2015). Djuarsa may not have had direct orders from his superiors in Jakarta to
organize this campaign, but he clearly collaborated with his immediate superior, Lieutenant
General Ahmad Mokoginta, who headed a command covering the entire island of
Sumatra and frequently communicated with fellow senior generals in Jakarta.
One important contribution of Melvin’s research is its focus on what she calls the
“mechanics of mass murder”—the precise forms of violence. She distinguishes
between two phases of killing. The first phase was when the civilian militias, after
being mobilized by the army, rampaged through cities and villages, killing individuals associated
with the PKI and leaving the corpses in the streets. The army in Aceh kept track
of the number of dead in what it called “public killings”; its final count was 1,941. The
2Melvin’s book, Mechanics of Mass Murder: How the Indonesian Military Initiated and Implemented
the Indonesian Genocide, the Case of Aceh, based on her PhD dissertation at Melbourne University,
Second phase, however, resulted in a much higher death toll: the army ordered the massacre
of detainees. In this phase, large groups were taken out of prisons and detention
camps by the truckload and executed in remote areas where there were few or no witnesses.
The victims disappeared without a trace. It was this kind of killing that one
sees dramatized in the film by Garin Nugroho, Puisi tak terkuburkan (Unburied
poetry, 2000), about an imprisoned Acehnese poet, Ibrahim Kadir, who watched as his
fellow detainees were taken out at night to be executed.
As in Aceh, the killing in Bali was also an army operation. Geoffrey Robinson’s research
in the 1990s revealed the crucial role of the para-commandos of the Resimen
Para Komando Angkatan Darat (RPKAD) in initiating the massacres in Bali in early
December 1965 (Robinson 1995, 273–303). He showed how the governor of Bali,
A. A. B. Sutedja, and the army commander based in Bali, Brigadier General Sjafiuddin,
both loyal to President Sukarno, prevented a murderous anti-communist campaign in
October and November. They temporarily banned the PKI and ordered members to regularly
report to local authorities. But they refused to permit the kind of mass killing that
had already occurred in other provinces, such as Aceh, Central Java, and East Java. They
were doing what the Sukarnoist army commander for West Java, Major General Ibrahim
Adjie, was doing at the same time (Herlina 2012). Unlike Adjie, they were unable to
persist in fending off intervention from Jakarta. After the killing of a police officer and
two militiamen in a village named Tegalbadeng on November 30, the local anticommunist
forces, including those within the officer corps stationed on Bali, claimed
that the PKI was beginning an offensive and needed to be suppressed more thoroughly.
The central army command in Jakarta dispatched RPKAD troops to Bali on December
7–9 to buttress those in Bali calling for more violent actions against the PKI (Kammen
and Jenkins 2012).
My own research on Bali builds on Robinson’s by investigating particular massacres.3
The best-known massacre was of the top leaders of the PKI’s provincial committee and
prominent Balinese who had supported the PKI. That massacre was in the village of
Kapal, about 12 kilometers from Denpasar. I have not yet discovered any documents
about it. I have gleaned the facts about it from oral interviews with seven eyewitnesses
(five of whom were members of an anti-communist militia) and five relatives of the
victims. Since the victims of the Kapal massacre were among the most prominent
public figures in Bali, news of this massacre traveled all around the island until it
became the stuff of rumor and legend. After sifting through the different stories, I
have determined that it occurred on the night of December 16 and claimed the lives
of thirty-five people.
It is telling that the PKI leaders were still alive by mid-December. Many of those
killed in Kapal had been under the protection of the police in Denpasar. The RPKAD
troops grabbed them from their places of detention, trucked them tied up and blindfolded
to the cremation grounds in Kapal, and then mowed them down with machine gun
fire. The role of the local anti-communist militia in Kapal was to prepare the execution
site; dig the mass grave; and murder one of the victims, the man considered to be the
3The first results of this research will be forthcoming in 2016 in an Indonesian-language biography
of a Balinese businessman, parliamentarian, and veteran of the anti-colonial struggle, I Gede Puger,
who was executed in the Kapal massacre. ringleader of the PKI in Bali, at the insistence of the RPKAD. None of the families of the victims received official notification of their deaths. They only heard rumors. One of the
men of Kapal who dug the mass grave during the day and then shoveled the dirt back into
it at night to cover the bodies recalled that the army officer who was supervising the operation
told them once they had finished their backbreaking labor, “If anyone comes by
here asking about this, just tell them ‘I don’t know.’”4
In Bali, there was no distinct phase of “public killings” as in Aceh. Nearly all the killings
were massacres of detainees. While the anti-communists killed some people in the
week between the Tegalbadeng incident and the arrival of the RPKAD, most of the massacres
occurred after the RPKAD came. One document I came across was an account of
the meetings between the main anti-communist political party in Bali, the Indonesian
Nationalist Party (PNI), and the army in December 1965.5 It was written by a PNI
record-keeper. It notes that a local army commander gathered the leaders of the PNI
and two other anti-communist parties, in the army command (the Kodim) for the area
around Denpasar on December 10, right after the RPKAD troops had arrived, to
discuss how they would “crush the September 30th Movement.” At the meeting,
they agreed that the killing would be extensive—every member of the PKI was a fair
target—but it would be done discreetly: “Don’t be demonstrative. It is not to be seen
by children and women and people who don’t need to know.”
The massacres in Bali were the result of an army operation, initiated by an RPKAD
officer, Major Djasmin, who created and controlled a network of anti-PKI civilians, the
Unified Coordinating Body of Actions for the Destruction of the Counterrevolutionary
September 30th Movement (Badan Koordinasi Kesatuan Aksi Pengganjangan Kontrev
Gestapu, known by its former members as KOKAP), which was formed on December
9 at his behest.6 The RPKAD troops under Djasmin were not sent because of a shortage
of military personnel in Bali; the number of soldiers already stationed there was more
than enough to carry out an operation to kill unarmed civilians. The purpose of the
RPKAD was to overawe the military officers, policemen, and state officials in Bali who
did not agree with the operation. Not surprisingly, Djasmin disavowed any responsibility
for the massacres in later years. He told one researcher that his role in Bali was to stop the
killings being committed by Balinese civilians (Conboy 2003, 149).
That the killing in Aceh and Bali became recorded in history books as the work of
frenzied, bloodthirsty masses represents the army’s success in concealing and obfuscating
its role. The army at the time did not want to be in the forefront, taking credit for massacres.
Those responsible for atrocities rarely wish their actions to be carefully documented and widely publicized. One can see in Aceh and Bali the same pattern of killing that one finds elsewhere in Indonesia. Researchers who have studied massacres in North Sumatra, Central Java, and Flores have found that the army led operations to
disappear large numbers of people accused as communists (Elsam 2012, 2013;McGregor
2012 Oppenheimer and Uwemedimo 2009; Prior 2011; Tri Hasworo 2004; van Klinken
2013). Tempo’s interviews with civilian perpetrators from East Java, Palembang, and Palu
also reveal the same pattern of forced disappearances (Tempo 2012, 56–72, 100, 106–7).
The dualistic thesis does not grasp the striking uniformity in these descriptions in such
widely dispersed locales. For all the diversity in the anti-communist violence, one finds
a remarkable consistency across the provinces in the practice of disappearing people
who had already been taken captive. One finds army personnel organizing the civilians,
administrating the detention camps, and arranging the trucks to transport the detainees
to the execution sites.
4Personal communication with Wayan (pseudonym), Kapal, August 2, 2013.
5The document is titled “Catatan dari Rapat Dewan Daerah, PNI” (Notes on meetings of the Provincial
Council of the PNI) and dated December 22, 1965. It has been reproduced in Hidayat
6The formation of the militia was announced in the local newspaper of Denpasar (Suara Indonesia
It is hard to believe that army personnel and militiamen independently adopted the
same method of disappearing detainees in all of these far-flung areas. There must have
been instructions from Suharto and his allied generals in Jakarta to their subordinates
about organizing the killings in this particular way. Researchers, on the basis of the
local studies that have been done so far, are now in a position to start working back up
the chain of command and examining the decision-making with the army high
command in Jakarta. If the internal army documents do ever enter the public record,
in the same way US government records about Indonesia in 1965–66 have, one can
expect to find further evidence against the dualistic approach.
The repression of the PKI could have been achieved, as it was in West Java, without
widespread massacres. The questions that can now be posed are: When and why did the
army high command under Suharto decide upon a policy of mass disappearances? How
did they overcome resistance within the army, police, and civilian administration to
ensure that the policy was implemented? Many unwritten stories of that time concern
the noncommunists, from the village level up to the presidential palace, who tried to
prevent the army from carrying out massacres. The diversity in the national pattern
may have been due precisely to this resistance. In provinces where the leaders of the
army and civilian government were united behind Suharto’s policy, as in Aceh, the massacres
began early; where they were not, as in Bali, the massacres only began once the
resistance could be overcome. While President Sukarno did not dismiss Suharto or
expose his support for the massacres, he called for calm and urged all state officials to
prevent killing. All regional army commanders understood by late October 1965 that
they faced two different orders: one from the commander of the army, Suharto,
stating that they should kill, and one from the president, Sukarno, stating that they
should not kill. They had to guess which leader was likely to prevail. Some of the massacres,
such as the February–April 1966 massacre in Flores, where the PKI was small and
presented no threat (Prior 2011; van Klinken 2013), appear to have been organized by
army officers for the sake of their own career advancement. They proved to their superiors
that they were heroic warriors of the anti-PKI crusade who harbored no sympathies
The recent research, by analyzing the precise forms of violence by which so many
people were killed, has begun to reveal the full horror of the massacres. It has been possible
to expose the stories of the perpetrators as self-serving myths—the PKI was on the
offensive, the PKI had created a situation of “kill or be killed,” it was a time of war, and so
on. Secretive, cold-blooded executions of large groups of tied-up detainees are impossible
to justify.7 If the perpetrators insist on claiming it was indeed a war, then they will have
to admit that their actions must be defined as war crimes, as the killing of prisoners of war.
By documenting disappearances, researchers are now able to write about the silent suffering
of the families of the victims who were left wondering if their loved ones were still
alive; some spent years fruitlessly searching for them (Wiludiharto 2004).
More Indonesians are now embracing a humanitarian perspective on the mass killings
of 1965–66, recognizing that regardless of what the victims did before 1965, nothing
can justify the kind of violence that was inflicted upon them. That violence was objectionable
not just because it was often indiscriminate, but because it targeted unarmed people
who had already been taken captive. The well-known Jesuit scholar in Jakarta, Franz
Magnis-Suseno, who was part of an intensely anti-communist Catholic youth group in
Central Java in 1965–66, has been advocating this perspective: “[T]hat the PKI was a
hated and feared enemy cannot justify the systematic killing and destruction of millions
of people who had been attracted to the PKI” (Magnis-Suseno 2015). Some of the
younger members of the NU, such as those grouped in the nongovernmental organization
Syarikat, have argued this point as well (Budiawan 2004). It would be a revolution
in the public discourse of Indonesia if many more citizens could share the simple
moral outrage of the late Firman Lubis, a medical doctor who wrote three autobiographical
volumes detailing his daily life from the 1950s to the 1970s. Living in Jakarta, he did
not hear about the massacres. He only learned about them when he visited Central Java
in 1966: “Even though I did not like communism, the arbitrary and lawless actions
blessed by the military authorities at that time disgusted me.… Probably the instinctive
doctrine of war—kill or be killed—influenced them. But even still, their actions cannot
be justified. The PKI that they opposed was not a military power. It was a just a legal
political party consisting of civilians. The situation cannot be equated with a war”
The disappearances in Indonesia were on a much larger scale than those in the later,
better-researched cases of Chile, where about 2,000 were disappeared under Pinochet,
1973–89 (Stern 2006, 392–94); and Argentina, where about 10,000 were disappeared
under the military junta, 1976–83 (Robben 2005, 323). Suharto managed to stay in
power for much longer than Pinochet and the Argentinian generals. He left behind institutions
that have been committed to enforcing a silence on the atrocities of 1965–66. Researchers
continue to face harassment and intimidation from state agencies and civil
society organizations that insist that the political genocide be treated as a nonevent. An
elderly Indonesian man, for instance, returning from decades of exile in Sweden to his
ancestral home in 2015, was promptly deported after taking pictures of the mass grave
in which he believes his father lies. The men in the village who had been complicit in
the massacre reported him to local army personnel (Tempo 2015). Efforts, such as
7The 2010 Convention for the Protection of All Persons Against Enforced Disappearance, which
the Indonesian government has signed but not ratified, removes any excuse for the practice:
“No exceptional circumstances whatsoever, whether a state of war or a threat of war, internal political
instability or any other public emergency, may be invoked as a justification for enforced
disappearance.” those described in this essay, to pry open an open secret confront a dense web of state,
quasi-state, and non-state actors who want the history of 1965–66 to be as invisible as the
victims buried underground in the unmarked mass graves.
I would like to thank Joseph Nevins, Gerry van Klinken, Douglas Kammen, Ayu
Ratih, and the three anonymous reviewers of this journal for their incisive comments.
List of References
ABDULLAH, TAUFIK, SUKRI ABDURRACHMAN, and RESTU GUNAWAN, eds. 2012. Malam
bencana 1965 dalam belitan krisis nasional, bagian II: Konflik lokal [The disastrous
night of 1965 in the coils of the national crisis, part II: Local conflicts]. Jakarta: Obor.
BEATTY, ANDREW. 2009. A Shadow Falls in the Heart of Java. London: Faber and Faber.
BOWEN, JOHN. 1991. Sumatran Politics and Poetics: Gayo History, 1900–1989. New
Haven, Conn.: Yale University Press.
BUDIAWAN. 2004. Mematahkan pewarisan ingatan: Wacana anti-komunis dan politik
rekonsiliasi pasca-Suharto [Breaking the inheritance of memory: Anti-communist
discourse and politics of reconciliation after Suharto]. Jakarta: Elsam.
CONBOY, KEN. 2003. Kopassus: Inside Indonesia’s Special Forces. Jakarta: Equinox
CRIBB, ROBERT. 2001. “Genocide in Indonesia, 1965–66.” Journal of Genocide Research 3
——. 2009. “The Indonesian Massacres.” In Century of Genocide: Critical Essays and
Eyewitness Accounts, eds. Samuel Totten and William Parsons, 193–217.
New York: Routledge.
——. 2010. “Political Genocides in Postcolonial Asia.” In The Oxford Handbook of Genocide
Studies, eds. Donald Bloxham and A. Dirk Moses, 445–64. Oxford: Oxford University
CROUCH, HAROLD.  1988. The Army and Politics in Indonesia. Rev. ed. Ithaca, N.Y.:
Cornell University Press.
ELSAM. 2012. Pulangkan mereka! Merangkai ingatan penghilangan paksa di Indonesia
[Bring them home! Connecting the memories of disappearances in Indonesia].
——. 2013. Jembatan Bacem [Bacem Bridge]. Jakarta: Elsam. Film on DVD.
FEALY, GREG, and KATHERINE MCGREGOR. 2012. “East Java and the Role of the Nahdlatul
Ulama in the 1965–66 Anti-Communist Violence.” In The Contours of Mass Violence
in Indonesia, 1965–68, eds. Douglas Kammen and Katherine McGregor,
104–30. Singapore: NUS Press.
GERLACH, CHRISTIAN. 2010. Extremely Violent Societies: Mass Violence in the Twentieth-
Century World. Cambridge: Cambridge University Press.
GREEN, MARSHALL. 1965. Telegram from US Embassy in Jakarta to Secretary of State,
“Joint Sitep no. 70,” December 29. National Security Archive, Indonesia and East
Timor Documentation Project, File 16, p. 16.
14 John Roosa
HEFNER, ROBERT. 1990. The Political Economy of Mountain Java: An Interpretative
History. Berkeley: University of California Press.
HERLINA, NINA. 2012. “‘Tatar Sunda’ digoncang konflik sosial politik” [The “Sunda realm”
shaken by social-political conflict]. In Malam bencana 1965 dalam belitan krisis
nasional, bagian II: Konflik lokal [The disastrous night of 1965 in the coils of the
national crisis, part II: Local conflicts], eds. Taufik Abdullah, Sukri Abdurrachman,
and Restu Gunawan, 51–78. Jakarta: Obor.
HIDAYAT, MANSUR. 1999. Biografi politik I Gusti Putu Merta (1913–1992) [A political
biography of I Gusti Putu Merta (1913–1992)]. BA thesis, Udayana University,
Faculty of Arts.
HUGHES, JOHN.  2002. The End of Sukarno: A Coup that Misfired: A Purge that Ran
Wild. Singapore: Archipelago Press. First published as Indonesian Upheaval.
INDONESIA. 1986. “Report from East Java.” Translated by Benedict Anderson. 41.
KAMMEN, DOUGLAS, and DAVID JENKINS. 2012. “The Army Para-commando Regiment and
the Reign of Terror in Central Java and Bali.” In The Contours of Mass Violence in
Indonesia, 1965–68, eds. Douglas Kammen and Katherine McGregor, 75–103. Singapore:
KOMNAS HAM [NATIONAL HUMAN RIGHTS COMMISSION]. 2012. Pernyataan Komnas HAM
tentang hasil penyelidikan pelanggaran HAM yang berat peristiwa 1965–1966
[Statement by the National Human Rights Commission on the results of the investigation
into gross human rights violations during the 1965–1966 event]. Jakarta:
KOMNAS PEREMPUAN [NATIONAL COMMISSION ON VIOLENCE AGAINST WOMEN]. 2007. Mendengarkan
suara korban perempuan peristiwa 1965 [Listening to the voices of
the women victimized in the 1965 event]. Jakarta: Komnas Perempuan.
KOMPAS. 2015. “Kalla: Pemerintah tidak akan minta maaf untuk kasus HAM 1965” [Kalla:
The government will not apologize for human rights cases in 1965]. November 11.
Akan.Minta.Maaf.untuk.Kasus.HAM.1965 (accessed March 12, 2016).
LEKSANA, GRACE. 2009. “Reconciliation through History Education: Reconstructing the
Social Memory of the 1965–66 Violence in Indonesia.” In Reconciling Indonesia:
Grassroots Agency for Peace, ed. Birgit Bräuchler, 175–91. New York: Routledge.
LUBIS, FIRMAN. 2008. Jakarta 1960an: Kenangan semasa mahasiswa [Jakarta in the 1960s:
Reminiscences of my student days]. Jakarta: Musup Jakarta.
MAGNIS-SUSENO, FRANZ. 2015. “Membersihkan dosa kolektif G30S” [Cleansing the collective
sin of the September 30th Movement]. Kompas. September 29. http://nasional.
page=all (accessed March 12, 2016).
MCGREGOR, KATE. 2012. “Mass Graves and the Memories of the 1965 Indonesian Killings.”
In The Contours of Mass Violence in Indonesia, 1965–68, eds. Douglas
Kammen and Katherine McGregor, 234–62. Singapore: NUS Press.
MELVIN, JESSICA. 2013. “Interview with Joshua Oppenheimer.” Inside Indonesia 112.
March 13, 2016).
——. 2015. “Documenting Genocide.” Inside Indonesia 122. http://www.insideindonesia.
org/documenting-genocide (accessed March 13, 2016).
——. Forthcoming. Mechanics of Mass Murder: How the Indonesian Military Initiated
and Implemented the Indonesian Genocide, the Case of Aceh. Honolulu: University
of Hawai‘i Press.
The State of Knowledge about an Open Secret 15
MUN’IM DZ, H. ABDUL. 2013. Benturan NU-PKI, 1948–1965 [The NU-PKI clash: 1948–
1965]. Depok: Langgar swadaya nusantara.
NEW YORK TIMES. 1966. “Sukarno Reports Killing of 87,000.” January 16.
OEI TJOE TAT. 1995. Memoar Oei Tjoe Tat: Pembantu Presiden Soekarno [The memoir of
Oei Tjoe Tat: President Soekarno’s assistant]. Jakarta: Hasta mitra.
OPPENHEIMER, JOSHUA, and MICHAEL UWEMEDIMO. 2009. “Show of Force: A Cinema-
Séance of Power and Violence in Sumatra’s Plantation Belt.” Critical Quarterly
POESPONEGORO, MARWATI DJOENED, and NUGROHO NOTOSUSANTO. 1990. Sejarah nasional
Indonesia [The national history of Indonesia]. 6 vols. 6th ed. Jakarta: Balai Pustaka.
PRIOR, JOHN MANSFORD. 2011. “The Silent Scream of a Silenced History: Part One: The
Maumere Massacre of 1966.” Exchange 40:117–43.
RICKLEFS, MERLE. 1993. A History of Modern Indonesia since c. 1300. 2nd ed. Stanford,
Calif.: Stanford University Press.
ROBBEN, ANTONIUS C. G. M. 2005. Political Violence and Trauma in Argentina. Philadelphia:
University of Pennsylvania Press.
ROBINSON, GEOFFREY. 1995. The Dark Side of Paradise: Political Violence in Bali. Ithaca,
N.Y.: Cornell University Press.
ROOSA, JOHN. 2006. Pretext for Mass Murder: The September 30th Movement and Suharto’s
Coup d’État in Indonesia. Madison: University of Wisconsin Press.
——. 2012. “The September 30th Movement: The Aporias of the Official Narratives.” In
The Contours of Mass Violence in Indonesia, 1965–68, eds. Douglas Kammen and
Katherine McGregor, 25–49. Singapore: NUS Press.
——. 2013. “Who Knows? Oral History Methods in the Study of the Massacres of
1965–66 in Indonesia.” Oral History Forum d’histoire orale 33:1–29.
SCHAEFER, BERND, and BASKARA T. WARDAYA, eds. 2013. 1965: Indonesia and the World.
SETIYONO, BUDI, and BONNIE TRIYANA, eds. 2003. Revolusi belum selesai: Kumpulan pidato
Presiden Soekarno 30 September 1965—Pelengkap Nawaksara [The revolution is
not yet finished: A collection of the speeches of President Sukarno from September
30, 1965 to the supplement to the Nawaksara speech]. Semarang: Mesiass.
SIMPSON, BRADLEY. 2008. Economists with Guns: Authoritarian Development and
U.S.-Indonesian Relations, 1960–1968. Stanford, Calif.: Stanford University Press.
STERN, STEVE. 2006. Battling for Hearts and Minds: Memory Struggles in Pinochet’s
Chile, 1973–1988. Durham, N.C.: Duke University Press.
STRASSLER, KAREN. 2005. “Material Witnesses: Photographs and the Making of Reformasi
Memory.” In Beginning to Remember: The Past in the Indonesian Present, ed. Mary
Zurbuchen, 278–311. Singapore: NUS Press.
SUARA INDONESIA. 1965. “Badan Koordinasi Kesatuan Aksi Pengganjangan Kontrev
Gestapu” [Unified Coordinating Body of Actions for the Destruction of the Counterrevolutionary
September 30th Movement]. December 10.
SUKARNO. 1966. Djangan sekali-kali meninggalkan sedjarah [Never abandon history].
SUNYOTO, AGUS. 1996. Banser berjihad: Menumpas PKI [Banser’s holy war: Crushing the
PKI]. Tulungagung: Pesulukan thoriqoh agung.
TEMPO. 2012. “Liputan khusus: Pengakuan algojo” [Special report: The confessions of the
executioners]. October 1–7, 50–125.
——. 2015. “Kisah Tom Iljas, diusir dari Indonesia karena ziarah ke makam orang tua”
[The story of Tom Ilyas, expelled from Indonesia for making the pilgrimage to
16 John Roosa
the grave of his father]. October 18. http://nasional.tempo.co/read/news/2015/10/18/
TRI HASWORO, RINTO. 2004. “Penangkapan dan pembunuhan di Jawa tengah setelah
G-30-S” [Arrests and murders in Central Java after the September 30th Movement].
In Tahun yang tak pernah berakhir: Memahami pengalaman korban ’65: Esai-esai
sejarah lisan [The year that never ended: Understanding the experience of the
victims of ’65: Oral history essays], eds. John Roosa, Ayu Ratih, and Hilmar Farid,
25–60. Jakarta: Elsam.
VAN KLINKEN, GERRY. 2013. Murder in Maumere: Postcolonial Citizenship. Inaugural
lecture 477. Amsterdam: University of Amsterdam.
VICKERS, ADRIAN. 2005. A History of Modern Indonesia. Cambridge: Cambridge University
WILUDIHARTO, YAYAN. 2004. “Penantian panjang di jalan penuh batas” [A long wait on a
road full of obstacles]. In Tahun yang tak pernah berakhir: Memahami pengalaman
korban ’65: Esai-esai sejarah lisan [The year that never ended: Understanding the
experience of the victims of ’65: Oral history essays], eds. John Roosa, Ayu Ratih,
and Hilmar Farid, 61–86. Jakarta: Elsam.
This post is also available in: English