The Intimacy of Terror:
Gender and the Violence of 1965-66 in Bali
Ibu Ari was a new bride in December, 1965 when a group of nationalist paramilitaries entered her family home and took her husband and her younger brother away, never to return. Soon after these two men disappeared, another one came to see her: Bli Made, a neighbour and village PNI leader who was rumored to have had his eye on Ibu Ari for years. No one in the family compound dared to deny Bli Made entrance that afternoon, when he marched in wearing the heavy boots of a soldier, accompanied by half a dozen of his thugs, saying he was there to carry out an 'inspection' [periksa], searching for proof of the family’s communist allegiances. When he ordered Ibu Ari to climb up the ladder to her family’s rice barn, when he followed her up and closed the door behind them, no one, they now say, could move or speak or see for the hour until the door opened again. And when Ibu Ari came down from the rice barn, clutching her kamben across her breasts, she said nothing, and her family never asked. 'We knew that she couldn’t tell us what happened,' says one of her cousins, a woman a few years younger than Ibu Ari. 'How could we speak of it? Death we could speak of; death was different. Even if we were afraid, death was something ordinary [umum]. But ‘inspecting’ women, who could speak of it? We were afraid of the words themselves.'
Ibu Ari still says nothing about that afternoon, only shakes like a tree in a storm if someone mentions Bli Made, who now appears regularly on television after having become a member of the Bali Provincial Legislature from Megawati Soekarnoputri’s PDI-P party in 1999. Ibu Ari doesn’t speak about it, but everyone in the family remembers what no one knows happened or not, so they say nothing when suddenly, in the midst of the daily women’s work of weaving ritual offerings, Ibu Ari will sometimes start speaking to no one they can see or hear, gripping her hands together in front of her chest, closing her eyes and rocking back and forth with the motions often used by women in trance to welcome deities into their bodies. Behind her back, though, some say that Ibu Ari is crazy, the kind of crazy, maybe, that happens when an unquiet history returns to inhabit the present.
Studies of violence and its aftermath have repeatedly shown that traumatic experiences may lead to radical shifts in processes of speaking, meaning-making, remembering and living with memory. The psychic imprints of terror, its destabilisation of social and cultural forms, and the shadows of fear it casts over political landscapes render the languages used to articulate violence ambivalent, shifting and even treacherous. Such insights sit poorly, however, with the frameworks through which traumatic memory is often expected to be expressed in post-conflict public domains. Truth commissions and tribunals, witness and testimony, journalism, fact-finding, history, activism and anthropology all depend—albeit in different ways and to varying extents—on the presumption that straightforward narratives of 'what really happened' are available for communication. The social effects of violence and the realist discourses deployed to address them are thus often discontinuous, with programs of reconciliation or recovery or social repair faced with more complex cultural forms than may be recognised.
This is a problematic with broad implications yet highly local contours. If traumatic events tend to share a capacity to refigure ways of understanding and communicating their import, they do so in specific ways, in particular cultural, historical and political contexts. In post-Soeharto Indonesia, people have recently begun to publicly voice memories of the atrocities that occurred during thirty-two years of a military-backed dictatorship, and to consider models for a Truth and Reconciliation Commission [Komisi Kebenaran dan Rekonsiliasi], authorised by Indonesia’s parliament in 2000 but, as of early 2004, still unimplemented. Yet even as narratives of violence trickle into mainstream discourse, challenging state histories and official silences, the stories that are told offer neither transparent windows onto the past nor coherent representations of a collective memory. Rather, calls for truth-telling have created new spaces of contest, as Indonesians debate interpretations of violent events and ponder which truths are worthy of articulation and attention. Exhortations to come to terms with a troubled past have neither created consensus on what should be spoken or kept silent, nor have they managed to render language a facile instrument of historical recovery or social rapprochement. Emerging from such a cultural landscape, memory itself has been, and remains, a political act.
Speaking about the violence of 1965-66, in which as many as a million Indonesians were killed as part of a military-led offensive against the Indonesian Communist Party (PKI) and other left-wing organisations, has been especially fraught with contest, crosstalk and hesitancy. For decades, Indonesians have feared that any challenges to official state accounts of the violence, which characterised the killings as a rational response to a communist threat to national security, could unleash new state violence and suppression of human rights. These histories authorized by the New Order regime focused exclusively on the 'violence' of the September 30, 1965 coup attempt alleged to have been masterminded by the PKI, treating the anti-communist slaughter and the state's role in directing it as a minor historical footnote. However, with Soeharto’s fall from power in 1998, Indonesian human rights workers, some of them supported by international initiatives for transitional justice or truth and reconciliation, began to gather data on the killings of alleged communists and on the systematic persecution. Survivors of the violence have organised research projects, advocacy programs and support groups, meeting with intellectual, political or therapeutic aims. Members of a national organisation founded in 1999, the Foundation for Research into the Victims of Massacre (YPKP), have undertaken fact-finding missions, and in late 2000 sponsored the excavation of a mass grave in Wonosobo, Central Java, that contained the remains of at least twenty-four people killed in 1965—the first exhumation to publicly offer proof of military complicity in mass murder during 1965. But even with Soeharto’s exit from the presidency, the historical narratives that were used by his regime to create a kind of mythic charter for its existence, justifying its absolute rule as a benevolent protection against communism, seemed to have been invested with lingering powers. Three decades of warnings by the state to be on guard against an ever-present threat of communism—a threat left sinisterly vague—endowed speech about 1965-66 with an almost uncanny ability to evoke fears of violent retribution. Indeed, speaking about the violence has frequently been followed by new violence: in March 2001, a reburial ceremony for the remains found in the Wonosobo grave was attacked by a group proclaiming themselves to be Muslim anti-communists, and a number of outspoken former political prisoners have been subjected to campaigns of harassment and intimidation. Abdurrahman Wahid, president of Indonesia from October 1999 to July 2001, issued a public apology in March 2000 for the role that members of his Nahdlatul Ulama (NU) organisation of 'traditionalist' Muslims played in carrying out the violence. Even Wahid, however, could not succeed in persuading the Indonesian legislature to repeal the 1966 law (TAP XXV/MPRS/1966) banning both the Indonesian Communist Party (PKI) and 'Marxist-Leninist Ideology,' leaving many survivors uncertain as to the very legality of speaking about their experiences. Today, many Indonesians remain seriously concerned about voicing memories of 1965-66, wary of the nationalist government of Megawati Soekarnoputri, which has revived many of the repressive laws that were favoured by the New Order regime in order to crack down on labor activists, student protestors, separatists and the media, and has kept much of Soeharto’s anti-communist legislation intact.
While the dismantling of the New Order regime undoubtedly loosened restrictions on the press and on freedoms of speech and assembly, new openness at the national level has reflected unevenly across those regions affected by the violence. Survivors’ ambivalence about articulating memory indexes not only fears of state repression, but also the forms that the violence took as it became embedded in communities, which were exhorted by the state in 1965 to participate in annihilating communism 'down to its roots' [sampai ke akar-akarnya] by uncovering and destroying the intimate 'enemy under the blanket' [musuh dalam selimut], be they neighbour, spouse, sibling or friend. Especially where those who carried out violent acts have remained in close proximity to those they terrorised, or where the lines between 'perpetrators' and 'victims' became blurred, there has been a reluctance to make public what Veena Das has called 'poisonous knowledge.' Survivors’ desires to speak about their suffering, or to publicly challenge state accounts of their purported complicity in subverting the state, have often been weighed against this implicit understanding that violence does not necessarily create solidarity among victims but rather exposes normally hidden possibilities of betrayals, reprisals and social tensions within families and communities. For example, in Bali, unlike many other places in the archipelago, most of those who lived through the violence have remained in the same communities, tied by customary law [adat] and ritual practice to their villages and temples of origin. Those who carried out violence regularly come face to face with those they terrorised on the streets, in the markets and at communal ceremonies. Speaking about 1965-66 does not, in such contexts, place a narrator in relation only to the state and its dark history, but channels memory through the complex local politics of the present. In Bali, the violence has also entered into an economy of memory, in which particular versions of the past become narratively packaged and standardised as tourism commodities, while other histories are viewed as lacking in value or even undermining the smooth functioning of this market. On an island where up to 80 percent of the population is directly or indirectly dependent upon tourism for their livelihoods, it has been, survivors say, often painfully hard to voice memories that threaten to destabilise the linkage stressed by government officials and the tourism industry between foreign arrivals and the production of images of peace.
This reluctance to publicly express memories of violence has been met by difficulties in hearing certain stories. If, as Anna Tsing and Mary Steedly have argued, power consists in large part in the ability to convene an audience, many survivors of 1965-66 have been disempowered indeed. Some of those who lived through the terror have had their stories recorded by journalists, scholars or activists, and a very few, such as the writer Pramoedya Ananta Toer or the victims’ advocate and former political prisoner, the late Ibu Sulami, have been able to attract international scholarly and media attention. The majority of survivors, however, have lacked a discursive space where their experiences could elicit a sympathetic hearing. Certain memories have been especially unlikely to find form in public narrative. Those that are considered too complex, ambiguous or grounded in local idioms are often difficult for national and international media to digest. Stories that fit uncomfortably into familiar and reassuring genres in which the victims and villains are clearly defined and suffering is redeemed through reconciliation, rehabilitation or revenge are unlikely to gain wide currency. Memories that refuse to take linear narrative form, settling instead into gesture, silence or indirect reference, are often exiled from public representations. And stories that seem tangential to the histories of contesting political parties that have been reproduced by the state and its critics alike in reference to 1965-66, speaking instead to the ways in which the violence entwined with local conflicts that crosscut party allegiances, have tended to be pushed to the edges of attention, described as variants on a national theme rather than constitutive of the very ways in which New Order nationalism was constructed through multiple nodes of violence and hegemonic engagement.
'There was so much in those days that wasn’t spoken,' Ibu Ari says. 'People used to talk about 1965 as the time when ulian raos abuku matemahing pati'—when you could die just because of a word. Spoken words are known in Bali to evoke actions, like the holy mantras of priests or the stories of puppeteers that resonate across the visible [sekala] and invisible [niskala] worlds. The word of a curse, spoken by the powerful, can bring illness or even death, and words can invest the inanimate—a mask, a barong, a jar of holy water—with taksu or charisma. But in 1965, words became new kinds of triggers. Improperly articulated words—an insult never quite forgotten, low Balinese spoken to someone who thought they should have been addressed in high Balinese, flirting exchanges with someone else’s wife—could return from the past to provoke horrifically exaggerated responses. A fifteen-year-old neighbour of Ibu Ari’s who 'talked too much' for some people’s liking was corralled in a wicker cage used to transport pigs and then thrown into the river to drown. A man who witnessed his neighbour helping to burn down someone’s house called out in protest and the next day was dead. A woman food stall vendor whose welcoming small talk was heard as a promise saw her husband killed by her would-be suitor. And one word above all, the word 'communist,' held power to determine who lived and died, a power no one word had ever been known to have before. Uttering the word 'communist,' speakers shifted social assumptions: no longer did the powerful alone utter words of power but the word itself, for those who dared to speak it in accusation, was imagined capable of saving one’s own life and determining others’ destinies. Heady, extraordinary, horrific: language became an unstable weapon in terror’s fantastic arsenal, like a mythical keris dagger, blade loose in the hilt, that could slip and wound its bearer should the flow of battle turn backwards. For as the word 'communist' was wielded, it came to mean far more than one who had pledged to party membership or felt sympathy with the PKI’s aims. As the ambitions of those who spoke it extended beyond the military mandate of 'uprooting the PKI' to staking social claims, exacting revenge or protecting one’s self and family in a treacherously shifting landscape, 'communist' transmuted from a symbol of political affiliation in the narrow sense to an indexical sign pointing to the instability of knowledge itself, to the impossibility of accurately reading another’s signs in an opaque field of highly charged power relations. As another of Ibu Ari’s cousins expressed it: 'Today you call me a communist, tomorrow someone calls you a communist. Anyone could be a communist as long as someone was willing to name them as one.' Even words like sibling or neighbour or friend turned slippery and treacherous, transformed into new hazards like informers, collaborators and provocateurs. And the emotions this speech engendered—the fear Ibu Ari’s cousin speaks of as being 'afraid of the words themselves'—grew so strong as to choke off streams of language and to channel meaning into silent forms.
This new semiotics of terror perhaps explains why when Bli Made came back every few months after his 'inspection' to ask Ibu Ari for money, she didn’t say anything, just sold what jewelry she had to keep up the payments. As a widow marked as 'politically unclean,' with no brother or husband to protect her, she was acutely vulnerable, painfully conscious of what actions a word of hers could evoke from him or what unwanted words from him any action of hers could set loose. But Ibu Ari’s payments to Bli Made were part of an exchange that never quite managed to substitute money for memory—the memories of either party to the transaction or the memories of those who witnessed something, no one was quite sure what, change hands. In the months and years that followed the 'inspection,' Bli Made would sometimes see Ibu Ari at village temple ceremonies or the nearby market, making her way through the crowd. Once she was within shouting distance, he would yell out to her, 'Oh, you want that money I borrowed from you, don’t you?' As the years passed, however, and new young toughs emerged to eclipse Bli Made’s standing in the neighbourhood, and as the rumors multiplied about the number of women—and not only PKI-linked women—he had sexually harassed, abused and threatened, his public calls to Ibu Ari began to sound, people said, more and more like the desperate pleas of a debtor and less like the boasts of an invulnerable assailant. Uttering words that reduced what had transpired between them to a loan of money, Bli Made was met by silence. Ibu Ari never responded as a true woman trader would do, with marketplace banter or aggressive coaxing, and with that absence of language sent out signs that grew all too easy for others to interpret. Ibu Ari took on silence as a barricade, protecting herself from the pain of memory. But even as she erected this wall she opened another door to memory, her own memory of just what karmic debt had been incurred in the rice barn, and the memories of her family and neighbours, which were elaborated from an image of a closed door into an imagination of what lay beyond it during that hour when no one dared to see. Her silence did not preclude semeiosis, involving as it did an awareness of relations of signification on the part of she who does not speak and interpretation on the part of those who do not hear. She was 'muted,' yet her muteness spoke memory.
Women’s stories have been especially subject to ambivalent articulation and reception. The majority of those killed in 1965-66 were men, and characterisation by human rights activists of what happened during this period as a massacre, while politically important and rhetorically powerful, has tended to obscure other important elements of the violence. The majority of historical accounts and political analyses of 1965-66 still fail to recognise the extent to which gender was implicated not only in the months of terror but also in the years of suffering that followed, when those alleged to have had links to the communist party were deprived of their civil rights and subject to strict state and community surveillance. Stories of the sexual assaults that were perpetrated upon women by the military and paramilitaries, of the tortures inflicted upon women political prisoners, of the social marginalisation experienced by women widowed or orphaned in the violence, and of the fear of being labeled 'communist' made it difficult for women to imagine enacting political agency have not become part of mainstream Indonesian public culture nor have they figured prominently in the projects to reevaluate or 'straighten' [meluruskan] Indonesian history that have recently been proposed by Indonesian activists and progressive historians. Despite the important efforts of a handful of scholars and activists to document the gender dimensions of violence in Indonesia, women’s experiences have remained marginal to the discourses of truth and reconciliation and transitional justice that have emerged in Indonesia since 1998. This tangential position of women’s narratives to national debates has often been taken for women’s silence. This, in turn, has encouraged the assumption that women have either been rendered voiceless by traumatic events or that they are no longer concerned with the suffering of those years, having either forgotten or forgiven or never having been directly involved in the elite politics that were said to have sparked the violence to begin with. Muted, reconciled or irrelevant to the master plot: women have been seen as having little to say about or to do with the present politics of the past.
My research on the violence of 1965-66 and the cultural, political and emotional landscape to which it gave rise has attemped to address these issues in Bali, where it has been estimated that some 80,000-100,000 Balinese, or 5-8 percent of the island’s population, were killed between December 1965 and March 1966, and where tens of thousands of women were sexually assaulted, imprisoned without trial, widowed, orphaned or left to survive as they could within severely fragmented communities. In Bali, both the violence and its aftermath were highly gendered, with men and women targeted and affected in particular ways and cultural notions of gender, sexuality, body and family manipulated and reified. Violence became embedded in intimate arenas of social practice, shifting gendered intersubjective relations and frameworks for understanding and articulating them.
During 1965-66 in Bali, the body was obviously a primary locus of terror. The Indonesian military and local paramilitaries aligned with the Indonesian Nationalist Party (PNI), as well as other Balinese motivated by fear, social pressure or desires for personal gain, were responsible for beatings, tortures, rapes and killings. Yet violence worked on and through the body not simply as a material object but as a culturally and politically charged site for expressing, reading and reworking relations of power, drawing on and producing symbolic violence through physical force. Survivors recall alleged communists being publicly beaten or tortured before being killed or abducted, with the injured bodies of those identified as prominent party members sometimes displayed to communities as graphic icons of communist defeat. Bodies of alleged communists were sometimes dismembered, their parts strategically placed in public space in grotesque parodies of Balinese ritual animal sacrifices [caru], evoking familiar cultural forms while exceeding meaning in sense-shattering ways. The symbolic resonances of such violence worked to alienate communists from social order, as if by treating their bodies as one might treat a chicken or a dog their inherent animality could be made apparent. It was not only the division between human and non-human that perpetrators of violence sought to engage, however. More commonly, the violence enacted upon bodies drew upon and reinvigorated densely-inscribed gendered codes of difference to produce terror. Males alleged to be communists were frequently abducted from their homes, transgressing the spatial, social and ritual boundaries that are erected in Bali around the family temple and the core of close agnatic relatives who claim the right to reside near it, as well as the caste hierarchies that determine the proper codes of language and etiquette one should use to enter another’s home. Given the patrilineal, patrilocal ideals of Balinese society, these abductions of men by the military and paramilitaries signaled an emasculating assault on communism by removing the male figures through which political solidarity was assumed to cohere. Removing men from the political order was also a direct assault on the social order, shattering the genealogical lines through which social identity is transmitted. Women, in contrast, were more often specifically targeted for rapes and other sexual torments, which made female bodies distressingly visible, rendering them public political signs of the state’s desire to expose even the most intimate arenas of society to its surveillance and control. The fate of those abducted also tended to follow gendered lines: in the village in which I work, more than half of those men who were abducted never returned home or were formally registered as political detainees, with their whereabouts still unknown to this day (although survivors assume they now lie in one of the scores of mass graves that have been identified as dotting the island). Women were very rarely 'disappeared' in this manner, but were instead placed in detention, some for years, or subject to more informal methods of community oversight. Through these violent practices of disappearance, exposure and containment, a gendered politics of visibility emerged, one which both drew on notions of gender difference to make violence meaningfully excruciating and which reified gender as a binding social force.
The end of most of the physical violence by mid-1967 signaled not an end to survivors’ suffering but the beginning of decades of oppression, as the New Order state elaborated the alleged communist coup attempt into an historical justification for its repressive practices of rule. In the aftermath of the bloodshed, terror settled closely into the space of the family, which became a crucial site for the transmission of fear and the new state ideologies which depended upon it for their maintenance. Not only were families broken apart by deaths and arrests, but the trauma of these losses was compounded by social sanctions against public mourning for the dead, who were demonised by the New Order state as dangerous criminals who deserved their fate. Especially in those cases where the bodies of victims were never recovered and the cremation rituals that would ensure them a place in the pantheon of divine ancestors were never able to be performed, there remain, to this day, ragged gaps in kinship networks. Normally, Balinese, in the village in which I work, are reincarnated back into their extended families, usually within a generation or two of their deaths, and people commonly visit spirit mediums [balian peluasan] to determine who has reincarnated in a child. But since 1965, there have been fewer than a handful of those killed in the violence who have been said to have returned to their families through reincarnation. These painful lingering absences, and the worry that attempts to address them by seeking out victims’ remains and holding proper cremations could provoke the state to punch new holes in the social fabric, encouraged Balinese survivors to enact state scripts of appropriate citizenship with often-exaggerated deference, leading survivors to bitterly cite Bali’s 'successes' at implementing a host of New Order campaigns, from family planning to child immunization to 'love your village' development projects to casting votes for the ruling Golkar party.
Violence did not simply 'unmake' families, however. Rather, it simultaneously ossified ties that had previously been fluid to form fixed units amenable to state surveillance, and strained emotional bonds by inserting suspicion and silence into everyday family life. Post-1965, fragmented Balinese families were perversely knitted back together by the 'clean environment' [bersih lingkungan] policy of the New Order government, which claimed that spouses, parents, siblings, children and even grandchildren of those marked as communists were 'infected' by political 'uncleanliness' and thus were to be barred from participation in the government bureaucracy or civil society organisations. Balinese families, newly corporatised by the use of traditionally flexible and contested kinship relations as tools of political identification, became important sites for social surveillance. Older relatives whose memories of the violence were still strong, monitored the younger generation for actions or utterances that could be interpreted by the state as 'political,' thus risking new repressions on the entire family. Just as survivors of the violence describe the military and paramilitaries’ intrusions into the enclosed space of the family compound as a traumatic violation of normal tenets of sociality, this new insertion of the state into family practice and subjectivity is identified as one of the most disturbing aspects of New Order rule. 'We still spoke to each other,' says one woman, remembering her relations with the several dozen family members with whom she shared both a family compound and a designation as politically 'unclean,' 'but we no longer spoke in the same way. We guarded our words, not knowing who was helping the state guard us.' Extended kinship networks often became fraught with tensions, as 'clean' segments of families grew resentful of being linked to their 'dirty' relatives, and as those who had been terrorised or had experienced the deaths of close family members suspected their more distant relatives of having offered the information that led to their victimisation. These stresses were sometimes compounded by family members who manipulating their relatives’ tenuous positions to claim communally held land as their individual possessions, taking advantage of victims’ fears of the government apparatus to obtain the land ownership certificates the Indonesian state, at the urging of the World Bank, began in the 1970s to promote in the name of order and development.
Despite the common use of concepts like 'collective memory' to refer to the recollections submerged in post-conflict social life, the Balinese families that emerged from the violence were not homogenous repositories of shared understandings of the past. Gender was among the most crucial differences that shaped survivors’ experiences and the limits and possibilities for their enunciation. In families where men had been imprisoned, killed or 'disappeared,' women were often forced to shoulder the burdens of caring for themselves and their children alone or in cooperation with other widows. While some women were lucky enough to be received back into their natal families after the loss of their husbands, many were shunned out of fear of the dangerous political visibility thought to accompany them. The hundreds of Balinese women who were jailed for alleged communist affiliations also faced, upon their release, frequent refusals by their husbands’ families to allow them to reclaim their children, who are considered by Balinese customary law [adat] to belong to the patriline. Not only were former women political prisoners thought to be politically dangerous, they were believed, because of their presumed bitterness and emotional instability, to be more likely to engage in black magic and thus be doubly menacing, even to their own children. Women’s rights as widows under customary law to claim tenure of the lands and possessions of husbands killed or abducted were also easily cast aside by using the stamp of communist. Through such practices and the narrative justifications constructed to account for them, women’s suffering was made to seem a matter of individual destiny with the gendered structural inequalities that shaped it described instead as the extraordinary measures taken to preserve society from the effects of aberrant historical events.
The events of 1965-66 also had a severe impact on possibilities for women’s political participation. In contrast to contemporary tourist representations of Balinese women as passive guardians of tradition, enclosed in an apolitical domestic sphere, in the early 1960s, Balinese members of Gerwani [Gerakan Wanita indonesia or 'Indonesian Women’s Movement'], a leftist group with ties to the PKI, were active in populist art, education, family health and childrearing programs. Although in retrospect many of Gerwani’s programs in Bali seem quite similar to those of the New Order era state-sponsored women’s groups, after September 30 1965, Gerwani members were painted by state rhetoric as sexually degenerate, grotesquely violent and highly dangerous politically. A major strand of the narrative that was constructed by military and civilian propagandists to legitimise the annihilation of the left was the claim that during the attempted coup, Gerwani members in Jakarta had danced naked in front of the generals kidnapped by the PKI conspirators, later castrating them, gouging their eyes out and leaving them to be killed by their male comrades. In Bali, reports were printed in the newspapers that local Gerwani women had prostituted themselves to members of the military in exchange for arms. In these discourses, women’s political agency was equated with an uncontrolled and predatory sexuality, playing upon Balinese patrilineal gender ideologies that see a woman’s sexuality and reproduction as properly under the control of her male relatives. While male communists were painted as evil ideologues challenging the political order, women communists were painted as immoral and unnatural, challenging the very foundations of Balinese tradition and culture.
In early 1966, several months after these stories about Gerwani were broadcast in Bali, thousands of women, most of them young unmarried teenagers whose male relatives had already been killed, imprisoned or threatened as alleged communists, were subjected to organised campaigns of sexual terror. In the village in which I work, PNI paramilitary gangs known as tameng travelled from house to house carrying out strip searches, claiming that they were seeking evidence of women’s communist sympathies in the form of hammer and sickle tattoos on the vagina, lower abdomen or upper leg. Often these searches were carried out in family compounds, in the presence of terrified family members. Sometimes women were ordered to report to government offices in Denpasar for further examinations, which often turned into rape and forced concubinage that extended over weeks, months or even years. Indonesian activists have noted that similar 'searches' took place elsewhere in in the country as well, as a technique of terror that made visible the vulnerability of the by-now-discredited left by exposing women’s bodies to humiliation and abuse, showing up the impotence of their male relatives and comrades to protect them. But while the command to carry out these actions seems likely to have come down from above, in Bali they served to reinforce local notions of the danger of a woman’s sexuality to her family and society, for this telltale hammer and sickle tattoo was said to be found on the same spots on the body where a woman who practices 'left' [pengiwa] or black magic would draw a rerajahan, a magical symbol used to sexually attract or exert power over men. A symbolic resonance was thus constructed between the political left and the magical left and women’s sexuality and men’s disempowerment. Where these searches by the paramilitaries were followed by rapes or forced concubinage, male sexual assault could be read by men not as violence but as a kind of contest of power, with male force triumphing over women’s sexual, magical and political potency. The fear that such actions, and the cultural notions they drew on and magnified, aroused goes much farther, I would argue, toward explaining the relative absence of Balinese women from contemporary politics and activism than does the more common reliance on an ahistorical 'Balinese tradition' to account for women’s apparent lack of public voice.
Even in the post-Soeharto era, as people begin—sometimes tentatively, sometimes more openly—to discuss what happened to them and their families in 1965-66, these histories of gendered violence have not become part of public discourse. For Balinese women who suffered abuses, it has been extraordinarily difficult not only to find social spaces in which to share these stories but to find languages with which to frame such experiences. The woman who describes being 'afraid of the words themselves' when confronted with the knowledge that her cousin had been 'inspected' points to the dependence of traumatic experience on language for its emergence into public memory, and the ability of speech to turn shifty and fraught with threats of betrayal. A periksa or 'inspection,' an Indonesian word reeking of state authority, of efficient, top-down bureaucracy, could enter the intimate space of one’s family home or enact its control on a woman’s body, bringing the state and its subjects into a terrifying new embrace, as men claimed to be guarding the nation against what might be symbolised—literally—on a woman’s vagina. To acknowledge this semiotic slippage and the power behind it was, more often than not, to stifle one’s speech, to make it impossible to acknowledge, even to oneself and one’s closest others, that another word for this experience was rape.
In the decades that followed 1965, as tourism became the cornerstone of state-sponsored development in Bali, Balinese women’s experiences of violence were subject to government and tourism industry attempts at erasure, as the figure of the graceful, gentle, traditional Balinese woman became an icon of exotic, erotic island allure. Similarly, New Order gender ideology—termed 'state ibuism' by Julia Suryakusuma —located women’s subordination in age-old 'tradition,' ignoring women’s previously active participation in politics and the terror that ensured their exit from public culture. Tourism, often unknowingly, reproduced many of the ideals of femininity encoded in state definitions, projecting fantasies of a peaceful retreat from the stresses of the modern world onto Bali, viewing women as guardians of an untroubled, unbroken past. Not only were Balinese women’s political histories and their suffering as victims blotted out by such images, they positioned women in a cramped discursive corner, where their own words threatened to turn back against them. One of the many troubling ironies of 1965-66 is that many survivors of the violence who were marked as being linked to communism and thus were barred from most employment were forced into the informal economic sector. Many of these Balinese who began by selling trinkets to tourists or offering massages on the beach in the early 1970s when mass tourism was first getting underway have ended up becoming deeply invested in the industry, giving them a serious incentive to censor their own memories. Speaking about 1965-66 in ways that counter official history has, in other words, been understood as not only politically dangerous but economically irrational. In this context, Balinese women’s silences need to be understood not as a natural response to trauma, but as the products of very real political, social and economic forces.
There were many things that Ibu Ari did not forget. She thought sometimes about her husband, whom she had never had a chance to grow close to after their arranged marriage, but she thought more often about her younger brother. 'He was the one person in the world I could really talk with,' she remembers. 'We could tell each other everything, even if we didn’t always agree.' Ibu Ari had not, for instance, agreed with her brother’s insistence that Balinese ritual should be simplified to take account of one’s economic condition. This 'Hindu rationalist' movement had grown popular in the early 1960s among the young leftist men of her family, who were high caste but poor in land and the hard currency that came with it. Their thinking had led to conflict among the family, especially after 1964, when Ibu Ari’s uncle died and his PKI-member son and some other young leftist men, including Ibu Ari’s brother, insisted that the family hold a simple cremation ceremony for him, arguing that the essence of the ritual, its practical effects of purifying the dead so that they may take their place among the divine ancestors and later reincarnate into the family, did not require the trappings of social hierarchy represented by a vast and expensive variety of ritual offerings. Most of the women of the family, including Ibu Ari, who were used to devoting their days to making offerings and organising their use for family rituals, were uncomfortable, feeling that such a cremation would not only undermine the value of women’s ritual expertise but would surely evoke curses from their ancestors and shame the family socially. It was a measure of Ibu Ari’s closeness with her brother that they could openly debate such matters of great importance to the family, with no need to gloss their disagreement with careful language and etiquette indicative of a woman’s deference to her male relatives. Indeed it was the language they used—the coarse Balinese ci/ciang for you/me—that Ibu Ari references to remember their intimacy.
This relationship with her beloved brother was cut short by his disappearance, but even then, the tie was not completely severed:
'We were so close, so very close. So close that when he died that afternoon, when he was killed, who knows where, nobody knew the place, that same night he came looking for me. He called out to me three times. I had already fallen asleep over there, next to that small coconut tree. Already he was looking for me. We were so close. He would tell me everything. If he spoke to our older brother once a day, he would speak to me ten times. He had left his watch behind. The day he died, his first son was just forty-two days old, it was the day of his dedinan ceremony. He said to me (about the child), 'Later, when he’s grown, don’t forget about him. It doesn’t matter if you have nothing to eat, you must give him the food from your own mouth, for this child who still lives.' He told me to sell the watch to pay for the dedinan ceremony. Three times he came to me, coming back and forth, telling me, 'Remember, remember, remember.' I was so shocked. I didn’t know that he was dead until the next day, when someone came to tell us he had been killed. They never told us where the place was where he had died, just that he was dead. He told me to remember.'
As the years passed and Soeharto’s New Order state continued its project of history-making, characterising men who died in 1965 as communists who were willing to undermine family, religion and state in pursuit of their evil aims, and erasing from national discourse the sexual assaults on women who were said to be wanton destroyers of society itself, Ibu Ari continued to be visited by her brother. Often he would just greet her and then depart, but sometimes he would give her instructions about family ritual matters. These instructions had little to do with his former stance in favour of simplifying and 'rationalising' religious ritual—a stance that was later glossed by the state as communist 'atheism'—but instead directed Ibu Ari to make additions to the offerings she was preparing to make them more 'complete.' That her brother, who had exhibited little interest while he was alive in the women’s work of offering-making, was now instructing her in ritual procedure was not odd to Ibu Ari; she was aware that once a spirit entered the realms of the dead he or she could change in character. Indeed in the early 1970s, when Ibu Ari was among a group of women visiting a spirit medium to inquire as to who had reincarnated in a child of the family, it was she who was addressed by name through the medium with the voice of her PKI cousin, who before his death in 1965 had caused so much controversy in the family by arguing that his own father should be cremated simply. This cousin, Ibu Ari said, told her that he had changed, that he was now a woman, and exhorted her, like her brother had, to 'remember.'
Perhaps it was the strength of Ibu Ari’s nostalgia for an imagined time before the violence, when she believed words to have meant what they were supposed to, when language was a means of intimacy rather than an implement of social fragmentation, when disagreement need not explode into slaughter, that kept the door open between her and her brother. Perhaps it was Ibu Ari’s desire to be free of the stain of 'communism' with which the state had smeared her family that led her to hear her brother as having been religiously 'rehabilitated,' worthy of a return to history. Perhaps it was her vulnerability as a widow that left her prey to people like Bli Made that evoked in her a desire for protection from her own patriline in the spirit of her brother or, conversely, her struggle to maintain women’s centrality to ritual practice that caused her to voice her brother’s instructions as authoritative. But these are all attempts of the anthropologist to come to terms with the uncanny, to strip it of its mystery. Ibu Ari herself is not interested in such explanations. Whatever the reason—and how, she asks, could the living ever really know what goes on in the realms of the dead?—he still visited Ibu Ari. The last visit she described took place in 2003, when she went with her family to a major ceremony at the Pura Dalem Puri, a temple associated with death rituals near the Besakih temple complex. In the midst of a crowd of hundreds, Ibu Ari felt a pair of hands descend on her shoulders. Not knowing who had touched her, she called out questioningly, 'Bapak?'—the formal Indonesian term of address for a man, the word one might use to speak to a government bureaucrat, a soldier or a stranger. She heard a voice chide her in low Balinese: 'Who are you calling ‘Bapak’? Don’t you [ci] know me [ciang]? Have you forgotten already?' No, Ibu Ari replied, she still remembered.
It would be mistaken, however, to equate recognition of the extraordinary suffering Balinese women have experienced with the assumption that they have been utterly deprived of voice or agency. One crucial way in which women survivors of the violence have challenged state history has been by using their place as family ritual experts to assert memories of the 1965-66 dead in contradiction of their official erasure from national belonging, drawing upon their relations with the realm of the uncanny to ground claims that transcend state authority. In the aftermath of the violence, the 'disappearing' of those accused of communism led to deep social uncertainties about the efficacy of cremation rituals that were forced to substitute effigies [adegan] of shaped dirt and cloth for the bodies of missing victims. Beginning in the 1970s, as the New Order regime assumed greater control over religious practice in Indonesia, the Parisadha Hindu Dharma Indonesia (PHDI), the official state-regulated organisation that claims authority over Hinduism in Indonesia, began to sponsor village nyapuh—'sweeping' or 'cleansing' rituals. These nyapuh were said to act in lieu of cremation for those who remained uncremated due to neglect, lack of financial resources or, implicitly, because the location of their bodies was unknown to their families. PHDI officials insisted that it was the cremation ritual’s purification of the 'soul' [atma], not the presence of the material body of the deceased, that made a cremation effective at returning the dead to the realm of the divine ancestors. Yet women relatives of those killed in 1965-66 often refused to accept this new theological stance, relying on the continuing absence of their disappeared family members as reincarnated in their children to resist official claims about ritual practice and their implicit denial of survivors’ suffering. In their role as leaders of the small family groups who visit spirit mediums [balian peluasan] upon the birth of children to engage in dialogic processes of divination designed to determine which ancestor has returned in the form of a baby, women negotiated narratives of the past. Through their recognition or dismissal of a balian’s clues as to the potential resemblance between ancestor and child, women acted as authors of memory, parsing the flow of possibilities offered by the balian’s speech to construct accounts that referenced both the characteristics of their forebears and the ambiguous status of the 1965-66 dead. If a balian began to describe an ancestor whom the woman client knew to have been killed, she could shift the line of the divination by disagreeing with the description or simply leaving to find another balian whose words she felt were more fitting. By refusing to recognise victims of violence in their children, women do not deny history; rather, they deny that closure has been reached on the past. By asserting the absence of the dead, women keep their memory alive.
Among themselves, women may also tell stories that highlight their ability to manoeuver within severely restrictive contexts and to survive extreme forms of trauma by drawing upon the very discourses that position them as politically marginal reproducers of fixed 'tradition,' 'culture' and 'religion.' In the village in which I work, where one banjar [hamlet organisation] split into a leftist banjar and a nationalist banjar in the 1950s, older women talk about how, when the violence erupted in 1965, they and their mothers were able to secretly travel between the politically divided communities, using back pathways which the paramilitaries did not patrol. Taking advantage of the ambivalence surrounding women’s political power and the liminal space it offered them, the nationalist women were able to carry food and ritual supplies to the leftist women who were afraid to venture out to the market, as well as to pass on information about the actions the local nationalists were planning against leftist families. These women’s networks in fact frustrated an attempt by nationalist paramilitaries to burn down a family compound whose residents were majority PKI, allowing them to defend their home by preparing large tubs of water and untying the flammable thatched roof from its supports in advance of the attack.
These women’s networks were not spontaneous expressions of political or gendered solidarity, but were elaborations of long-standing ties of ritual interdependence and patronage between high caste [brahmana and ksatria] women and their lower caste [sudra] clients that centered on the exchange of ritual labour and materials for specialist ceremonial offerings and expertise. These channels persisted—narrowed and diverted, but still held open—through inter-banjar conflict, the 1965-66 violence and the decades of state oppression and its reproduction in community politics that followed. Recounting these histories, women rarely frame their engagements with each other as political 'resistance,' although the relaying of sensitive information obviously defied the wishes of their menfolk and gave rise to political effects when they passed on information and rumors that undermined anti-communist strategies. Instead, women describe their actions as a maintenance of 'the way things always were,' referencing their responsibilities as women to attend to matters of custom over and against transient political concerns. Women’s reproduction of hegemonic narratives of gender, culture and their own political marginality were not, however, an uncomplicated capitulation to state power, but rather contained a sharp political critique. Not only did they subvert the state’s call to participate in identifying and exterminating communists, they posed what they stressed as women’s inherent concern with and historical knowledge of ritual matters against state propaganda that claimed communists to be antisocial atheists. But if women survivors of the violence, who lost not only family members but also their own sense of security in the world, do not describe themselves in the familiar language of political activism, neither do they see themselves as inherently more 'peaceful' than men, able to 'reconcile' with the past and 'forgive' those in their communities who participated in the state’s project to destroy them. Like male survivors of the violence, Balinese women overwhelmingly wish for justice and restitution—even if they tend to locate such hopes in the ultimate realm of karma rather than in a corrupt and disinterested legal system. What these women’s stories do recognise, however, is what so few narratives of 1965-66 have noted: women’s abilities to create agentive pathways through their social worlds, navigating off the maps drawn by male political elites.
A Semiotics of Suffering and Silence
Some members of the family who have witnessed Ibu Ari speaking to her brother in what appears to them as a state of trance have attempted to push her into a more familiar cultural framework, suggesting that she could perhaps herself become a spirit medium, claiming social significance as one who becomes a conduit through which the living can speak to the dead. They warn that someone who has been given the gift of the medium and refuses to accept it as a social role risks being cursed by the gods with madness, as can a psychic who shows arrogance in his or her personal power at the expense of acknowledging that this gift comes from the divine. Yet Ibu Ari insists that her experience is not the trance of a medium, but rather normal waking consciousness. She denies any agency in initiating this communication: her brother enters her everyday world; she does not purposely try to open a door to the unseen realm where he dwells. She rejects the idea of playing the public role of medium, saying she has no desire for such power. She speaks with her brother, the only person she could ever really speak with, and she has no interest in speaking with others or their dead. After all, she adds, who would consult a psychic who was known to have communist ties? And were her ability to speak with the dead made public, would people start talking again about what she wished them to forget: that she was the widow and the sister of men who had been marked communist? And would they try to force open the closed door of the rice barn, to put into language what had become for her and her family a silence weighted with ambiguous memory? The politics of speaking, even with the dead, are, Ibu Ari knows, treacherous indeed.
Ibu Ari can’t explain why she, out of everyone in her family, now returns the dead to the present, to everyday consciousness, even as official narratives of the violence attempt to exclude them from history and as ritual works to separate them from the mundane work of memory. Nor does she know why the dead come to change the history of their family, instructing them in how to better care for the living. Likewise, her family cannot explain what kind of madness has befallen her, and challenged them. But in what cannot be explained, in the silences that puncture uncanny forms of speech, Ibu Ari still remembers.
My main concerns in this essay have been twofold: first, to place gender at the forefront of our thinking about and working against violence in Indonesia; and second, to suggest a critical approach to women’s narratives of violence that avoids constructing binary poles of 'speech' versus 'silence' or 'victimisation' versus 'agency,' enabling us to recognise the more complex and ambivalent processes of semiosis and social action that have emerged in a climate of severe personal and community trauma and long-standing state repression. Both of these emphases have political and theoretical implications. Recognising the extent to which gender not only marked the forms that violence took, but saturated experiences of living in its aftermath and influenced the social context in which it can now be remembered and articulated, encourages us to move from an exclusive focus on physical force to examine the complex symbolic and structural aspects of violence that have been manifest in women’s lives. It suggests that we attend to the ways in which state-sponsored terror engaged with local gendered relations of inequality to shape patterns of violence and possibilities for recovery that are different for men and women, and which have left long-lasting effects that have persisted even after a change in regime. It asks us to see, next to the haunting figure of one million dead, the uncounted number of those living survivors who still struggle with painful, ambivalent or unsettling memories of violence. It challenges us to not assume that a lack of women’s voices in public culture means that women have been either rendered silent or are no longer concerned with what they experienced, but to learn to listen for their stories in the politically dense semiotic spaces in which they are expressed. For women’s narratives of 1965-66 are not necessarily heroic tales of those who perished in loyalty to belief or prevailed in the face of persecution. To engage in their representation is not to write eulogies for those upon whom victimhood has conferred closure. One day, perhaps, if the generation of survivors passes out of memory, this may indeed be the only genre left available to us. But for now, the stories of those who still live with the unthinkable, stories of survival, of continuing shame, of strength and weakness and remembering and forgetting, of daily tactics and narrow choices and the searching out of space where sanity can be preserved, still remain to be heard.
This paper is based on three years (2000-2003) of field research on 1965-66 and its aftermath in Bali, in collaboration with Degung Santikarma, funded by a John D. and Catherine T. MacArthur Foundation Research and Writing Fellowship and grants from the H.F. Guggenheim Foundation. For very helpful comments and suggestions, I thank Lyn Parker, the two anonymous reviewers for Intersections, Hildred Geertz, Mary Zurbuchen, John MacDougall and Degung Santikarma.
 See, for example, Susan Brison, Aftermath: Violence and the Remaking of a Self, Princeton: Princeton University Press, 2003; Cathy Caruth, Unclaimed Experience: Trauma, Narrative and History, Baltimore: Johns Hopkins University Press, 1996; Valentine Daniel, Charred Lullabies: Chapters in an Anthropography of Violence, Princeton: Princeton University Press, 1996; Veena Das, 'Language and body: transactions in the construction of pain,' in Social Suffering, ed. Arthur Kleinman, Veena Das and Margaret Lock, Berkeley: University of California Press, 1997, pp. 67-92; Elaine Scarry, The Body in Pain: The Making and Unmaking of the World, New York: Oxford University Press, 1987; Michael Taussig, The Nervous System, New York: Routledge, 1992.
 See Agung Putri, 'Evading the truth: will a Truth and Reconciliation Commission ever be formed?' Inside Indonesia 73, Jan/Mar 2003; Mary S. Zurbuchen, 'Looking back to move forward: a truth commission could bring healing for a tragic past,' Inside Indonesia 65, Jan/Mar 2001.
 The exact number of Indonesians killed is unknown and will likely remain so, despite recent efforts at 'fact-finding' by victims’ advocacy groups such as the Yayasan Penelitian Korban Pembantaian [Foundation for Research on the Victims of Massacre]. Estimates have ranged from around 300,000 deaths to as many as three million, with a figure of one million frequently cited in academic and journalistic accounts of the violence. The politics of numbering the dead is, of course, far from straightforward, speaking both to the state’s desire to block access to non-official historical research and to activists’ desires to ground calls for attention to the violence in statistical claims of its significance. It is important to note, however, that while the extent of the suffering wrought by the violence of 1965-66 should be undeniable, survivors often locate its import not in its scope but its intimacy, not in its manageable facticity but in its destabilising incomprehensibility, not in its right to a place in the annals of the twentieth century’s greatest tragedies but in its continuing power to inflect possibilities for living in the present. Gyanendra Pandey discusses a comparable politics of enumerating the deaths that occurred during the partition of British India in 1947, suggesting that such 'extravagant, expandable, unverifiable but credible' statistics function to obscure the social production of history and its qualities of rumor. See Gyanendra Pandey, Remembering Partition, Cambridge: Cambridge University Press, 2001, p. 91.
 While the international media at the time tended to describe the killings as an irrational outburst of primitive emotion, describing 'orgies' of bloodshed and a 'frenzy' of anti-communist fervor (Pulitzer Prize winner John Hughes’s book on 1965, Indonesian Upheaval, New York: David McKay, 1967, recently reissued as The End of Sukarno: A Coup that Misfired: A Purge that Ran Wild, Singapore: Archipelago Press, 2003 offers perhaps the best example of this sensationalist genre), state accounts instead stressed the savage excess of the left, framing military and civilian violence against alleged communists as the careful, calculated and justified enactment of bureaucratic rationality upon those who had forfeited claims to citizenship.
 For more on the events of September 30, 1965 and their political ramifications, see Benedict Anderson and Ruth McVey, A Preliminary Analysis of the 1 October 1965 Coup in Indonesia, Ithaca: Cornell University Press, 1971; Robert Cribb (ed.), The Indonesian Killings of 1965-66: Studies from Java and Bali, Clayton: Monash University Centre of Southeast Asian Studies Papers on Southeast Asia no. 21, 1990; Harold Crouch, The Army and Politics in Indonesia, Ithaca: Cornell University Press, 1978. For an overview of the events in Bali, see Geoffrey Robinson, The Dark Side of Paradise: Political Violence in Bali, Ithaca: Cornell University Press, 1995. For an examination of the cultural and political repercussions of the violence in Bali, see Leslie Dwyer and Degung Santikarma, '’When the world turned to chaos’: 1965 and its aftermath in Bali, Indonesia,' in The Spectre of Genocide: Mass Murder in Historical Perspective, edited by Robert Gellately and Ben Kiernan, Cambridge: Cambridge University Press, 2003. For discussions of the important place that '1965' as history, imaginary and threat has held in state discourse and public culture, see Benedict Anderson, Language and Power: Exploring Political Cultures in Indonesia Ithaca: Cornell University Press, 1994; John Pemberton, On the Subject of 'Java' , Ithaca: Cornell University Press, 1994; James T. Siegel, A New Criminal Type in Jakarta: Counter-Revolution Today, Durham: Duke University Press, 1998; Mary Margaret Steedly, Hanging Without a Rope: Narrative Experience in Colonial and Postcolonial Karoland, Princeton: Princeton University Press, 1993; Saya S. Shiraishi, Young Heroes: The Indonesian Family in Politics, Ithaca: Cornell University Southeast Asia Publications, 1997.
 For discussion of the Wonosobo incident, see Zurbuchen, 'History, memory and the "1965 Incident"'; Femi Adi, n.d., 'Corat-coret Tentang Perkuburan Massal di Hutan dekat Wonosobo' [Notes on the Mass Grave in the Forest near Wonosobo], on the YPKP homepage. In May 2000, the late Ibu Sulami, a former vice secretary of the leftist Indonesian Women’s Movement [Gerwani] and one of the founders of the YPKP, was threatened by members of a group calling itself the 'Anti-Communist Command.' In September 2000, her house, which served as an office for the YPKP, was burned down. See 'YPKP Yayasan Penelitian Korban Pembunuhan 1965/1966, Foundation for the Research of 1965/1966 Massacre'.
 See Robert Hefner, Civil Islam: Muslims and Democratization in Indonesia, Princeton: Princeton University Press, 2001.
 See Amnesty International, 'Indonesia: old laws—new prisoners of conscience,' Report ASA, 21/027, 2003, issued 10 July 2003.
 See Veena Das, 'The act of witnessing: violence, poisonous knowledge and subjectivity,' in Violence and Subjectivity, ed. Veena Das, Arthur Kleinman, Mamphela Ramphele, and Pamela Reynolds, Berkeley: University of California Press, 2000, pp. 205-25.
 See Anna Tsing, 'Gender and performance in Meratus dispute settlement,' in Power and Difference: Gender in Island Southeast Asia, ed. Jane Monnig Atkinson and Shelly Errington, Palo Alto: Stanford University Press, 1990, p. 122. Mary Steedly offers an perceptive analysis of women’s speech and its common misrecognition as silence in her ethnography of narrative experience among the Karo Batak of Sumatra, writing, 'Nowhere in the world are women silent; rarely, I would suspect, have they ever been so. The problem, rather, has been their inability to convene an audience. If we fail to attend to their talking in corners, on the edges, in moments snatched from and cadenced by the demands of everyday labors—and to their speaking at the centers of public discourse, in someone else’s words—then we are not simply recording the exclusion of women’s voices; we are repeating that exclusion' (Steedly, Hanging Without a Rope, p. 198). I would argue for an extension of Steedly’s insight to focus on the non-narrative forms, including ritual speech and practice, through which women may articulate their experience.
 This is not to say that narratives of 1965-66 all presume that 'politics' were the driving force behind the killings; many scholarly analyses and personal testimonies speak to the fact that the military’s exhortation to 'uproot' and 'cleanse' communism from communities provided a pretext for the playing out of long-standing local conflicts or personal antipathies that had little to do with the opposition between the left and the right in the formal political arena. Such narratives run the risk, however, of leaving official historical framings intact, identifying 'real' communists versus 'non-communists' as the primary categories of historical analysis. It is, I would argue, important to note that not all those who were killed or terrorised or socially marginalised were 'communists.' However, it is equally important to recognise how this discourse sets up a 'real' that reproduces the notion of 'communist' as a fixed category of identity without attending to its wide local variations, the sharp contests that have marked it historically, and its deployment as a symbol of the nation and its antithesis. The politics of this are complex: given that state histories of 1965-66 downplayed or even denied the military’s role in carrying out the killings, describing the violence as a natural—if at times somewhat excessive—response by individual Indonesians to the threat of communism to their beloved nation, Indonesian activists have tended to focus most closely on delegitimising this official history, directing their efforts toward research that may have legal uses should tribunals to hold Soeharto regime officials to account for human rights violations be possible at some time in the future. In place of 'popular emotion' as the driving force behind the violence of 1965-66, such work has most often posited a systematic enactment of a New Order state strategy for amassing power and destroying or traumatising potential opposition. While respecting the commitments behind such work, I have been less inclined to view the state as a monolithic force that acted with consistent rationality to orchestrate and carry out killings of those accused of communism. While recognising the state’s role in managing terror in 1965-66, I have been struck by the ways in which terror constantly overflowed its strategic aims and surpassed or even subverted its own goals as it unfolded in specific cultural and political contexts. But this concern with exploring the ways in which violence and its aftermath played out in highly local ways, implicating not only individual social actors but culture itself, has brought its own challenges. It makes it extraordinarily difficult to enforce temporal brackets around 'the 1965 incident,' to unselfconsciously inscribe a 'post-' to 1965, or to unreflexively rely on notions like 'Balinese culture' or 'Balinese belief' to work backward to an understanding of what happened in 1965-66. Not only is what we see today as 'Bali' very much marked—socially, ritually, artistically, politically, economically—by the aftermath of violence, but the experience of participating in state terror has inserted state power into some of the most intimate spaces of Balinese life, making the notion of an external state perpetrator acting on or against a civil society in barbaric fashion at best a simplification of something far more complex and perilous.
 By 1972, the New Order state had grown uneasy with the power that the PNI had gained as a result of their participation in the massacres of the PKI, and it began a process of Golkarisasi in Bali in which the PNI was banned and former communists were, under strict control, cultivated as state supporters.
 cf. Daniel, Charred Lullabies, p. 122
 Saskia Wieringa’s Sexual Politics in Indonesia, New York: Palgrave MacMillan, 2002 offers a groundbreaking account of the history of Gerwani, the New Order state’s demonisation of women’s political participation, and some of the efforts that have begun since Soeharto left office to address the violation of women’s rights that occurred under the New Order. Other works that address women’s experiences of violence through testimonial genres include Carmel Budiarjo, Surviving Indonesia’s Gulag: A Western Woman Tells Her Story, London: Cassell Academic, 2000 and Ibu Marni, 'I am a leaf in a storm: a woman’s political autobiography,' in Inside Indonesia 26, March 1991. Important Indonesian-language discussions of the gender dimensions of 1965-66 include Budiawan, 'Merintis gerakan rekonsiliasi akar rumput berperspectif jender,' in Kompas, March 1, 2004; Ruth Indiah Rahayu, 'Dampak peristiwa 1965: hancurnya perempuan kita!' Sekitar.com Internet Journal, 2004, site accessed 7 June 2004; Sulami, Perempuan, Kebenaran dan Penjara, Jakarta: Cipta Lestari, 1999; Sulami, Merentang Perempuan, Jakarta: Cipta Lestari, 2001; and sections of John Roosa, Ayu Ratih and Hilmar Farid (eds), Tahun yang Tak Pernah Berakhir: Memahami Pengalaman Korban 65, Esai-Esai Sejarah Lisan, Jakarta: ELSAM, 2004.
 In Bali, as elsewhere in Indonesia, these numbers are difficult to estimate with any accuracy. Geoffrey Robinson, The Dark Side of Paradise, cites a figure of 80,000 people killed (p. 1). Adrian Vickers, Bali: A Paradise Created, Berkeley: Periplus, 1989, refers to 100,000 deaths (pp. 170-71). Balinese activists currently working to gather facts about the killings also tend to use a figure of 100,000.
 For a cogent analysis of the relations between physical and symbolic violence as enacted through the body, see Pierre Bourdieu, 'Gender and symbolic violence' in Masculine Domination Stanford: Stanford University Press, 2001, pp. 34-42.
 In the village in which I work, there have continued to be sporadic incidences of violence that residents attribute to the tensions that remain in the wake of 1965-66.
 With few priests willing to officiate at cremation ceremonies for those accused of communism, and with few families willing to risk persecution, imprisonment or even death by holding public rituals, most families of victims were forced to carry out cremation [ngaben]—rites for the disappeared using effigies [adegan] made of dirt shaped to resemble human bodies, which were burned secretly without the customary assistance of the banjar, the hamlet organisation responsible for providing assistance with death rituals. Although these rituals were usually formulaically correct, subjectively they signaled a lack. As one woman whose husband and brother were killed in the violence explained: 'These were proper ngaben ceremonies, with seven kinds of holy water and a complete set of offerings, but after they were over I still didn’t feel 'satisfied' [puas] in my heart.' Many other Balinese refer to the cremations held for the 1965-66 missing as ngaben-ngabenan—'cremation-ish' rituals or 'as if cremations.'
 Hildred and Clifford Geertz’s ground breaking work Kinship in Bali, Chicago: University of Chicago Press, 1975 notes the contested, flexible nature of traditional Balinese kinship relations.
 The politics of land tenure and use in Bali and their relationship to the violence of 1965-66 and the contested figure of 'communism' are far too complex to be addressed here. For more detailed discussion, see Anton Lucas and Carol Warren, 'The state, the people, and their mediators: the struggle over agrarian law reform in post-New Order Indonesia,' in Indonesia 76, 2003, pp. 87-126; Mary Zurbuch and Degung Santikarma, 'Bali after the bombing: land, livelihoods, and legacies of violence,' paper presented at the Yale University Agrarian Studies Colloqium Series, February 2004; and Graeme MacRae, 'The value of land in Bali: land tenure, landreform and commodification,' in Inequality, Crisis and Social Change in Indonesia: The Muted Worlds of Bali, ed. T. Reuter, London: Routledge Curzon, 2002, pp. 143-65.
 Although there were a small number of Balinese women who were senior members of Gerwani, deeply involved in framing and disseminating the ideological tenets of the organisation and in debating the relationship it should have with the PKI, both former Gerwani members and former PNI-affiliated women now describe the majority of Balinese Gerwani members as having been relegated to supporting roles for the PKI, such as organising food and drink for PKI-sponsored events. With little historical information available about Gerwani in Bali and the ways the local organisation may have interpreted policies being set at the central level, it is difficult to know to what extent these contemporary descriptions reflect historical fact. However, Wieringa’s Sexual Politics discusses the gender politics of the PKI, documenting its ambivalence about feminist issues and Gerwani’s subordination of concerns with women’s rights to a focus on class oppression.
 See Robinson, The Dark Side of Paradise, p. 293.
 Indonesia scholars reading my work have pointed out that the Indonesian noun for inspection should be pemeriksaan, not periksa. This is true; however, Balinese do not always speak Indonesian as they 'should.' Grammatically proper or not, Balinese identify periksa—both the word and the events—as emanating from the central Indonesian state.
 Julia Suryakusuma, 'The state and sexuality in New Order Indonesia,' in Fantasizing the Feminine in Indonesia, ed. Laurie Sears, Durham: Duke University Press, 1996, pp. 92-119.
 The history of the various 'Hindu rationalist' movements that have emerged in Bali are far too complex to address here in detail. The important point to note, however, is that the state's characterisation of communist claims about the relations between ritual, labour and class as 'atheism' ignored long-standing theological and political debates in Bali over the form and scale of ritual.
 These nyapuh were especially widespread before large-scale state-sponsored ceremonies, such as the Eka Dasa Rudra of 1979. See David Stuart-Fox, Pura Besakih: Temple, Religion and Society in Bali, Leiden, KITLV Press, 2002. The ritual claims and counter-claims made during the New Order by various factions within and without the PHDI over the issue of cremation are complex indeed and cannot be fully addressed here. Many survivors of 1965-66 believe—and I would concur—that the nyapuh not only reflected shifting Balinese theological and political conerns, but were part of an Indonesia-wide campaign of depoliticisation of village life that sought to remove potential sites of local contestation (including the presence of mass graves) by placing them under state control. For more analysis of discourses of cremation, Hindu 'identity' and modernist reform, see Linda Connor, 'Contestation and transformation of Balinese ritual: the case of ngaben ngirit,' in Being Modern in Bali: Image and Change, New Haven: Yale Southeast Asia Studies Monographs, 1996.