Patterns of New Order Torture: The IndoTorM Dataset (Pohlman et al., 2024)

Pohlman, A., Wibisono, S., Casey, M., Pohlman, C. L., & Louis, W. R. (2024). Patterns of New Order torture: Initial findings from the Indonesia Torture Mapping (IndoTorM) dataset. *Indonesia, 117*, 59–88. https://doi.org/10.1353/ind.2024.a926786 PDF: 02 Secondary Sources/2.2 1965 Violence, Memory and Historiography/Pohlman et al. (2024) - Patterns of New Order Torture (IndoTorM Dataset).pdf PDF PAGES 1 to 31 of 31. Text from the PDF text layer where present, otherwise Tesseract OCR (ind+eng); OCR errors remain. === [PDF page 1] === Patterns of New Order Torture: Initial Findings from the Indonesia Torture Mapping (IndoTorM) Dataset Annie Pohlman, Susilo Wibisono, Mavourneen Casey, Catherine L. Pohlman, Winnifred R. Louis Indonesia, Number 117, April 2024, pp. 59-88 (Article) Published by Cornell University Press DOI: https://doi.org/10.1353/ind.2024.a926786 For additional information about this article https://muse.jhu.edu/article/926786 [128.59.222.107] Project MUSE (2026-01-31 03:49 GMT) Columbia University Libraries === [PDF page 2] === Patterns of New Order Torture Initial Findings from the Indonesia Torture Mapping (IndoTorM) Dataset Annie Pohlman, Susilo Wibisono, Mavourneen Casey, Catherine L. Pohlman, and Winnifred R. Louis [128.59.222.107] Project MUSE (2026-01-31 03:49 GMT) Columbia University Libraries A Problem with Torture Indonesia’s “New Order” regime had a torture problem.1 This military regime, which began with the anti-Communist purges of 1965–66 and ended with the May 1998 riots, perpetrated torture on an unprecedented scale in Indonesia, primarily against its own Annie Pohlman is an historian of modern Indonesia based at The University of Queensland. Susilo Wibisono is a research fellow in the School of Psychology, University of Queensland. Mavourneen Casey is a public health researcher at the Institute of Health and Wellbeing, Federation University Australia. Catherine L. Pohlman is a lecturer in vegetation ecology at Charles Sturt University. Winnifred R. Louis is a professor of psychology at The University of Queensland. Annie Pohlman conceived the IndoTorM project, collected the data, and as the lead author wrote the first full draft and revised the paper to integrate coauthors’ feedback. All other authors provided feedback and contributed to the revision process for the paper. In addition, Susilo Wibisono performed the data analysis and drafted sections of the paper pertaining to the methods and results. Mavourneen Casey collaborated to the design of the original analysis tools and data coding. Catherine L. Pohlman collaborated to design the original dataset and data coding. Winnifred R. Louis collaborated to codesign the analyses, to contribute to the manuscript frame, and to refine the method and results. Indonesia 117 (April 2024) L5315.indd 59 3/28/2024 3:53:55 PM === [PDF page 3] === 60 Pohlman et al. civilian population.2 When looking at the regime’s larger-scale atrocities, hundreds of thousands were tortured by state officials during the anti-Communist purges and tens of thousands during the occupation of East Timor (1975–99), the military operations period in Aceh (DOM, Daerah Operasi Militer, 1989–98), and in West Papua.3 These numbers do not include the hundreds (and sometimes thousands) tortured by state agents during the New Order’s crack downs, or “incidents” (peristiwa) as they are often referred to in Indonesian, such as the “Petrus” (Penembakan Misterius, Mysterious Shootings) killings of alleged criminals in the early-to-mid-1980s, the attacks on civilians in Tanjung Priok in 1984, or in Talangsari in 1989, among many other cases.4 Nor do they include the unknown numbers of victims tortured by state agents during criminal investigations, or those from socially marginalized groups particularly vulnerable to unlawful detention and violence, including drug users, sex workers, and street children.5 The sheer scale of torture during the New Order—that it was used so often against so many by the regime—is something that historians of Indonesia generally agree to be true.6 They further generally agree that, although there were many examples of state- On the military’s involvement in these periods of mass violence, see Jess Melvin, The Army and the Indonesian Genocide: Mechanics of Mass Murder (New York: Routledge, 2018); Tim Relawan untuk Kemanusiaan (TRuK), “Early Documentation No. 3: The Rapes in the Series of Riots: The Climax of an Uncivilized Act of the Nation Life,” July 13, 1998, http://www.our21.com/Indo/atrocities.html; TRuK, Korban Kekerasan Seksual Kerusuhan Mei 1998 (Jakarta: TRuK, 15 October 1998); and Jemma Purdey, Anti-Chinese Violence in Indonesia 1996–1999 (Honolulu: University of Hawai’i Press, 2006), in particular, 147–58. In all these cases, civilians were the majority of victims. Estimates for the number of victims tortured during each period vary. For the anti-Communist purges, John Roosa estimates hundreds of thousands were tortured; see John Roosa, “The Truths of Torture: Victims’ Memories and State Histories in Indonesia,” Indonesia 85 (April 2008): 32. On the frequency of torture use during the occupation of East Timor, see the Commission for Reception, Truth and Reconciliation of East Timor (Comissão de Acolhimento, Verdade e Reconciliação de Timor Leste, CAVR), Chega!: The Report of the Commission for Reception, Truth and Reconciliation (Dili: CAVR, 2005), ch. 7.4, para. 24, 27–29. On torture during the DOM period in Aceh, see Al-Chaidar, Sayed Mudhahar Ahmad, and Yarmen Dinamika, Aceh Bersimbah Darah: Mengungkap Penerapan Status Daerah Operasi Militer (DOM) di Aceh, 1989–1998 (Jakarta: Pustaka Al-Kautsar, 1998). See also the chapter examining torture during the DOM period (for which Annie Pohlman contributed the main analysis) in the forthcoming Final Report of the Aceh Truth and Reconciliation Commission (anticipated 2023). On torture in West Papua, occupied by Indonesia since 1963, see Budi Hernawan, Torture and Peacebuilding in Indonesia: The Case of Papua (London: Routledge, 2017). On the Petrus killings, see Justus M. van der Kroef, “‘Petrus’: Patterns of Prophylactic Murder in Indonesia,” Asian Survey 25, no. 7 (July 1985): 745–59, https://doi.org/ 10.1525/as.1985.25.7.01p0288a. Members of the police fired on a crowd of mostly Muslim protestors in the north Jakarta suburb of Tanjung Priok on September 12, 1984, the incident then leading to extrajudicial executions, torture, and other abuses against at least a further one hundred people. See Subhan S. D. and F. X. Rudy Gunawan, eds., Mereka Bilang di Sini Tidak Ada Tuhan: Suara Korban Tragedi Priok (Jakarta: KontraS and GagasMedia, 2004), https://kontras.org/ wp-content/uploads/2019/09/mereka_disini_tidak_ada_tuhan.pdf. The Talangsari incident occurred in a village of that name in Lampung on February 7, 1989 when members of the military attacked, killing and torturing villagers. The Indonesian National Commission on Human Rights (Komnas HAM) conducted an investigation into the Talangsari incident between 2001 and 2007; see Dian Kuswandini, “Commission Deems Talangsari Incident Gross Rights Violation,” Jakarta Post, September 4, 2008, http://www.thejakartapost.com/news/2008/09/04/commission-deems-talangsari-incident-gross-rights-violation.html. Very little is known about torture of criminals and the socially vulnerable during the New Order beyond anecdotal accounts, an issue currently under investigation by Annie Pohlman. For example, three key collections published shortly after the New Order contemplated the centrality of torture and other forms of state-led violence during the regime, see Benedict R. O’G. Anderson, ed., Violence and the State in Suharto’s Indonesia (Ithaca, NY: Southeast Asia Program, Cornell University, 2001); Ingrid Wessel and Georgia Wimhöfer, Violence in Indonesia (Hamberg: Abera-Verlag, 2001); Freek Colombijn and J. Th. Lindblad, eds., Roots of Violence in Indonesia: Contemporary Violence in Historical Perspective (Leiden: BRILL, 2002). L5315.indd 60 3/28/2024 3:53:55 PM === [PDF page 4] === Patterns of New Order Torture 61 led torture prior to the New Order (during the earlier Sukarno period, the national revolution, the Japanese Occupation and before7), the prevalence of torture increased markedly during the regime, to such an extent that it became “normal.” In 1999, just after the end of the regime, for example, historian Benedict Anderson reflected on this problem of the unprecedented and widespread use of torture by the New Order: If one reads the memoirs of activists who ran afoul of the colonial regime one rarely finds mention of beatings and tortures, let alone electrodes being attached to genitals, and the like. But over the past thirty years these have become “normal” activities of police and military men at the lowest levels. These days, it is normal to beat up someone arrested even before he or she is interrogated; and to execute prisoners on the pretext that they are “attempting to escape.” Some of these things happened in the 1950s and 1960s, but they were not “routine.” That they have become routine means that those who are supposed to uphold the authority of the law in fact violate this law every day with complete impunity.8 To date, historians’ knowledge of the New Order’s torture has been based mostly on the wealth of primary accounts by survivors and witnesses from throughout the regime’s three decades, in addition to the extensive contemporaneous reporting by human rights organizations, all of which have provided ample anecdotal evidence.9 But these many sources are disparate, and cases of torture are reported almost always as individual incidences. These cases are scattered across various reports and studies and have not been systematically analyzed in the academic literature. Scattered as they are, the similarities and differences across these many individual accounts are difficult to see. Certainly, some have examined torture during specific periods (such as in 1965–66) or in particular locations (such as in Aceh). These analyses have shown hints of the systematic and organized nature of the regime’s torture and sometimes shown the similarities of methods within those particular periods or locations.10 A few Indonesian human rights groups have compared small samples of stories of torture perpetrated For one analysis that emphasizes the use of force and violence under the late colonial state, see Henk Schulte Nordholt, “A Genealogy of Violence,” in Roots of Violence in Indonesia: Contemporary Violence in Historical Perspective, edited by Freek Colombijn and J. Thomas Lindblad (Leiden: KITLV Press, 2002), 33–61; see also Marieke Bloembergen, “The Dirty Work of Empire: Modern Policing and Public Order in Surabaya, 1911–1919,” Indonesia 83 (April 2007): 119–50. On torture during the Japanese Occupation period, see, for example, Zenkoku Ken’yūkai Rengōkai Hensan linkai, The Kenpeitai in Java and Sumatra: Selections from The Authentic History of the Kenpeitai (Nihon Kenpei Seishi), trans. Barbara Gifford Shimer and Guy Hobbs (Ithaca, NY: Cornell Modern Indonesia Project, Cornell University, 1986). On the Revolusi period, see the special issue of the Journal of Genocide Research, edited by Bart Luttikhuis and A. Dirk Moses, “Mass Violence and the End of the Dutch Colonial Empire in Indonesia,” Journal of Genocide Research 14, nos. 3–4 (September 2012): 257–76, https://doi.org/10.1080/14623528.2012.719362; and William H. Frederick, “Shadows of an Unseen Hand: Some Patterns of Violence in the Indonesian Revolution, 1945–1949,” in Roots of Violence in Indonesia: Contemporary Violence in Historical Perspective, edited by Freek Colombijn and J. Thomas Lindblad (Leiden: KITLV Press, 2002), 143–72. Benedict R. O’G. Anderson, “Indonesian Nationalism Today and in the Future,” Indonesia 67 (April 1999): 6. These sources are discussed further in the next section. 10 See footnote 2, above. For the 1965 case, for example, the evidence briefs prepared for the Prosecutor for the International People’s Tribunal for 1965 (held in The Netherlands in 2015) showed patterns of similar torture methods across geographical locations. Annie Pohlman was the lead author on the evidence brief for sexual crimes; some of those findings regarding torture are reported in Annie Pohlman, “Sexual Violence as Torture: Crimes Against Humanity during the 1965–66 Killings in Indonesia,” Journal of Genocide Research 19, L5315.indd 61 3/28/2024 3:53:55 PM === [PDF page 5] === 62 Pohlman et al. across different time periods, particularly across the periods of crimes against humanity.11 One such study was done by activists from the group Asia Justice and Rights (AJAR), who collated and compared stories by survivors from 1965 to 1966, Timor Leste and Aceh, to show that there are broad similarities in the forms of torture used across these periods.12 While AJAR continues to collect valuable testimonial evidence, and has likely one of the most comprehensive collections of materials, their focus is (rightly) on assisting survivors rather than analyzing evidence. Their reports thus far are valuable but limited to lists of types of torture seen across periods, and no in-depth analyses of patterns, relationships, or other connections between cases have been attempted. Thus, although Indonesian historians agree that torture was widespread during the New Order, and some analyses have been attempted to elucidate general patterns in how torture was deployed, these attempts have been limited in geographical and temporal scope and have been restricted for the most part to lists of torture types. Without more comprehensive analyses, there is no way to gain insight into how and why the systematic and widespread use of torture developed during the regime. Without looking comparatively at what was done, against whom, and in what circumstances, it is difficult to unpack how the New Order’s problem with torture developed and became normalized. In this article, we outline some of the initial findings from a project that attempts to bring together and synthesize all these data about torture during the New Order and to carry out an in-depth, systematic analysis of these disparate materials.13 This project has produced a first version of the Indonesian Torture Mapping (IndoTorM) dataset, which contains more than four thousand individual case histories of torture perpetrated during the New Order. In this paper, first we introduce the dataset and briefly outline the sources used and information contained therein. We then outline our methods and present initial findings from our first analyses of the dataset. These analyses address our preliminary, broad questions about what these data might show, which were to what extent can patterns in torture be discerned? Is there similarity (homogeneity) or differences (heterogeneity) in patterns across regions (e.g., in Java, Sumatra, or East Timor), or when comparing different types of torture? Are there differences, based on whether the victims were male or female, or whether they were children at the time (those seventeen years old or younger)? As we show below, there were differences by type of torture (physical, sexual, and psychological forms), when and where it was used, just as there were differences in how men, women, and children were tortured. What, though, can these differences and patterns tell us about the New Order’s torture? Furthermore, what do these patterns reveal about how torture became “normal” 11 See, for example, Muhamad Daud Berueh, “Practice of Torture during Soeharto’s Regime,” Article 2, special report, A Façade of Justice for Torture Victims in Indonesia 12, no. 2 (2013): 23–30. KontraS (the Commission for the Disappeared and Victims of Violence) has also produced some reports based on their work with survivors from various periods. See, for example, KontraS, Menyusun Puzzle Pelanggaran HAM 1965: Sebuah Upaya Pendokumentasian (Jakarta: KontraS, 2012). 12 For one published outcome, see Galuh Wandita, Indria Fernida, and Karen Campbell-Nelson, “Mass Torture in 1965–66: A Continuing Legacy,” in The International People’s Tribunal for 1965 and the Indonesian Genocide, edited by Saskia E. Wieringa, Jess Melvin, and Annie Pohlman (London: Routledge, 2019), 60–79. 13 We gratefully acknowledge the main funding for this project from the Australian Research Council (Discovery Early Career Researcher Award, “How Does Torture Become Normal? Indonesia’s New Order Regime, 1965–1998,” DE170100619, Annie Pohlman). L5315.indd 62 3/28/2024 3:53:55 PM === [PDF page 6] === Patterns of New Order Torture 63 during the New Order? After presenting descriptive data, we take a psychological approach to discuss how these findings might help to explain the internalization of “normative” social harmdoing within the New Order’s security services.14 This approach helps us to understand how which harmful acts of torture might become normative (that is, accepted behavior) against particular victims, within groups of military, police, or their auxiliaries.15 Understanding which behaviors become normative within groups is critical, because normative behaviors by definition become widespread and cannot be eliminated by targeting individual “bad apples” (e.g., jailing a single perpetrator of torture). The psychological literature on the development of normative harmdoing helps to explain how individuals within those groups internalize and endorse these behaviors, even when they are harmful, and how the behaviors come to be perceived as “right,” justified, and normal.16 By examining some of the broad patterns in how different types of torture were used, where, and against whom during the New Order, we draw on the existing psychological research and theorizing on normative harmdoing as a lens through which to examine this new torture data and understand torture norms as changing (and unchanging) in this context. The IndoTorM Dataset: Sources and Methods The dataset examined here was compiled between 2010 and 2019 and represents a first attempt to bring together and analyze the existing data on torture during the New Order. This first version of the IndoTorM dataset evolved from what was originally a smaller, qualitative study using oral historical and life narrative approaches conducted by the first author to examine interview data with around two hundred participants who had experienced torture during the mid-to-late 1960s.17 In 2010, she expanded the scope of the study and began compiling information on cases of torture perpetrated throughout the New Order. The sources for this information about torture were predominantly reports published by human rights organizations, such as by the international groups operating during the New Order: Human Rights Watch (then Asia Watch), Amnesty International, and the 14 Specifically, two areas, self-determination and social identity theory. A model that integrates these two areas to understand how harmful normative actions are accepted by members within groups is the MINSOH (the Model of Internalisation of Normative Social Harmdoing); see Catherine E. Amiot, Morgana Lizzio-Wilson, Emma R. Thomas, and Winnifred R. Louis, “Bringing Together Humanistic and Intergroup Perspectives to Build a Model of Internalisation of Normative Social Harmdoing,” European Journal of Social Psychology 50, 15 For the social identity approach’s understanding of norms, see John C. Turner, Social Influence (Milton Keynes: Open University Press, 1991). 16 See, for example, S. Alexander Haslam and Stephen Reicher, “Beyond the Banality of Evil: Three Dynamics of an Interactionalist Social Psychology of Tyranny,” Personality and Social Psychology Bulletin 33, no. 5 (May 2007): 615–22, https://doi.org/ 10.1177/0146167206298570; Alan Page Fiske and Tage Shakti Rai, Virtuous Violence: Hurting and Killing to Create, Sustain, End, and Honor Social Relationships (Cambridge: Cambridge University Press, 2015). Note that the internal cognitive and social processes, and the personal motivations and perceptions of individuals within groups, are beyond the scope of discussion in the present paper. 17 This earlier study was based on interview data collected by Annie Pohlman between 2002 and 2009 and examined mainly women’s experiences of political detention (with a focus on sexual violence and torture) from the mid-1960s through to the late 1970s. For the main outcome from that study, see Annie Pohlman, Women, Sexual Violence, and the Indonesian Killings of 1965–66 (London: Routledge, 2015). L5315.indd 63 3/28/2024 3:53:55 PM === [PDF page 7] === 64 Pohlman et al. UK-based TAPOL.18 Others were reports and documentation produced by Indonesian domestic organizations, particularly during the latter New Order period, such as from legal aid groups (various Legal Aid Foundations, LBH, Lembaga Bantuan Hukum, and the national office, YLBHI, Yayasan Lembaga Bantuan Hukum Indonesia) and human rights groups such as ELSAM (The Institute for Policy Research and Advocacy, Lembaga Studi dan Advokasi Masyarakat) and KontraS (The Commission for the Disappeared and Victims of Violence, Komisi untuk Orang Hilang dan Korban Tindak Kekerasan).19 Other sources included reports and documents produced by Indonesia’s two national human rights commissions, Komnas HAM (National Commission on Human Rights, Komisi Nasional Hak Asasi Manusia) and Komnas Perempuan (National Commission on Violence Against Women, Komisi Nasional Anti-Kekerasan terhadap Perempuan), as well as some contemporaneous newspaper reports and testimonial materials compiled by organizations established by survivors of particular periods of violence (such as the YPKP, the Foundation for Research into the Victims of 1965–1966 Mass Killings, Yayasan Penelitian Korban Pembunuhan 1965/1966). Some further sources used included academic publications, emails, and other communications within and between NGO groups, as well as some letters and other written materials by survivors and witnesses. When taking information from these sources for entry into the dataset, one individual victim of torture was the unit of analysis, though the quantity and specificity of information about that individual’s experience varied significantly across sources, from only a few lines to, in a few cases, hundreds of pages of documentation.20 In total, cases of torture experienced by 4,072 discrete individuals have been incorporated thus far into the current IndoTorM dataset.21 Note that any case examples given in this paper are taken from publicly available sources, though the dataset contains many confidential sources. Table 1, below, gives a breakdown of the types of sources sampled to compile these data. When the scope of the study expanded after 2010 to examine cases of torture from throughout the New Order period, so too did the methods. As the number of cases of individual torture victims grew, this first version of the dataset was developed, and [128.59.222.107] Project MUSE (2026-01-31 03:49 GMT) Columbia University Libraries select, quantitative methods were used to support the classification and organization of information. Information about an individual and their experience of torture was coded for specific types of data using a set of codes. The set of codes was created by the first author in 2010, which she developed originally by open coding around fifty transcripts of interviews with torture survivors to identify types of torture used and to inductively 18 Annie Pohlman was granted access to the Asia Watch materials, held at Columbia University, in April 2015 and to the TAPOL archives in London in September 2018 (“tapol” is the Indonesian word for “political prisoner”). Earlier, she was granted restricted access to some of the Amnesty International materials held at the International Institute of Social History in Amsterdam in 2006. 19 Annie Pohlman collected these materials primarily during several fieldwork trips to Jakarta between 2015 and 2019; further data collection was delayed due to the COVID-19 pandemic from 2020 onward. 20 The main unit of analysis was an individual victim, such that data was captured where possible on one individual’s experience of torture, even if that information were collated from more than one source. Further, all data regarding an individual’s experience of torture were captured within a single line, even when that person may have experienced multiple episodes of torture over months or years. 21 As explained further below, these cases recorded in the dataset are only a minority of the total number of cases found; challenges with coding have delayed incorporation of the remaining cases. L5315.indd 64 3/28/2024 3:53:55 PM === [PDF page 8] === Patterns of New Order Torture 65 Table 1: The number (n) and percent (%) of cases drawn from different categories of information sources for the initial IndoTorM dataset The category of resources n victims % Nongovernment Organization report 2,839 69.7% Personal testimony 296 7.3% United Nations report 200 4.9% Military document 194 4.8% Journalistic report 128 3.1% Government data 112 2.8% Organization’s internal communication 47 1.2% Academic publication 44 1.1% Uncategorized 212 5.2% Total 4,072 100.0% generate major themes.22 For the dataset, she then developed the set of codes to capture information about the types of torture used, in which locations, by whom, and their effects, among other types of information about victims and perpetrators.23 Between 2010 and 2019, information from the sources was coded and entered into the dataset.24 Over the first five years, the codes were refined and expanded a number of times to capture a wider range of forms of violence as more and more variances were uncovered in the source materials. Narrative accounts about torture experiences, whether firsthand or witness accounts, were also added to the dataset in the form of summaries of these accounts and direct quotes from testimonies, where available. As many cases related to deceased individuals, narrative accounts were also included from reports by surviving witnesses. Although the source documents were written mainly in Indonesian and English, all materials in the dataset were translated where necessary into English.25 22 This type of thematic, narrative-based analysis for identifying and categorizing information is common in oral history. See Paul Thompson and Joanna Bornat, The Voice of the Past: Oral History (Oxford: Oxford University Press, 2017), 351–91. 23 On coding for thematic analysis, see Johnny Saldaña, The Coding Manual for Qualitative Researchers, 4th ed. (Thousand Oaks: SAGE, 2021). 24 The coding and data entry was performed by Annie Pohlman, with the assistance of several research interns, supported by The University of Queensland. Interns who worked on the dataset coding between 2011 and 2019 were Jane Ahlstrand, Jennifer Plaistowe, Stephanie Pearson, Melanie Kilby, Daniel Pfeffer, Devina, Bram Adimas Wasito, Amaliah, Matthew Vidler, Katie Valenta, Lindsey Stempke, Neil Renic, and Daria Orlovskaya. 25 Most of the international NGO reports were published in English, while most of the Indonesian domestic reporting was produced in Indonesian. Other languages for materials in the dataset were Tetum, French, and Portuguese. Annie Pohlman is fluent in Indonesian and had advanced knowledge of French and was responsible for most translations for materials in those languages. Research assistance was sought for translations of materials from Tetum and Portuguese (translation support provided by a University of Queensland New Staff Research Fund ([“Torture and the New Order State in Indonesia: Stories from East Timor,” 2012–13] and an Early Career Researcher grant [“Violence and Detention during the New Order Regime in Indonesia: Reading Stories of Torture and Ill-Treatment,” 2015], both held by Annie Pohlman). L5315.indd 65 3/28/2024 3:53:55 PM === [PDF page 9] === 66 Pohlman et al. The dataset was built in Microsoft Excel. In the analysis presented in this paper, we examine five attributes of the individual cases. The demographic codes for gender were male or female (regardless of age), with a further code to denote whether the person was a child at the time they experienced torture (defined as seventeen years old or younger).26 The other attributes were the time period ([1] 1965–66, [2] 1967–69, [3] 1970–75, [4] 1976– 80, [5] 1981–89, [6] 1990–98, and [7] 1999–onward); geographic region ([1] East Timor, [2] Papua/West Papua, [3] Aceh, [4] Java, [5] Sumatra, excluding Aceh, [6] Eastern Indonesia, excluding Papua, [7] Bali, and [8] anywhere else in Indonesia); and the type of torture (three separate categories of codes for physical, sexual, and mental forms). Time refers to when the torture was perpetrated. As the original codes developed by the first author had focused on the mass human rights violations during the anti- Communist purges that began in the mid-1960s and continued through the 1970s, the codes developed for time periods were uneven. Rather than entering a specific year or date (to create a scalar variable), time was coded by the seven periods. If a person were tortured during 1966, then a code of “1” would be registered in the appropriate column and “0” across the remaining columns. Similarly, if a person were tortured in 1966 and in 1970, then “1” was entered for codes 1 and 3. Although some torture cases perpetrated after the end of the New Order were included in the dataset, they are not the focus of the present research. Region refers to the geographical area where the torture occurred. While the region codes generally capture broad areas, the codes are uneven, as they also highlight areas where long-term state-led atrocities occurred under the New Order. Again, if a person were tortured in more than one region—such as a person detained and tortured in East Timor and later transferred to a prison in Java where they again experienced torture—“1” would be entered for codes 1 and 4, to capture multiple regions. The codes for types of torture were divided into three broad categories: physical, sexual, and mental forms. For all types, torture was defined as “severe pain or suffering, whether physical or mental,... intentionally inflicted on a person [for a range of purposes] by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity.”27 Within each of these three broad categories were several subtypes of torture codes. First, for physical forms 26 In this first version of the dataset, there are no demographic codes for other genders, an oversight we plan to remedy in future. There were, however, a handful of cases of torture against those who self-identified as waria (generally, male-to-female transgender persons). For example, one case involved a waria tortured by local policemen in South Jakarta in the early 2000s to try to force them to confess to stealing mobile phones. For a study on gender presentation and sense of self among waria during the New Order period, see Benjamin Hegarty, “Under the Lights, Onto the Stage: Becoming Waria through National Glamour in New Order Indonesia,” Transgender Studies Quarterly 5, no. 3 (August 2018): 355–77, https://doi.org/10.1215/23289252-6900738. 27 This definition comes from Article 1 of the 1984 United Nasions Convention Against Torture and Other Cruel, Inhuman, and Degrading Treatment and Punishment (the “CAT”), the most commonly used contemporary legal definition. Purposes contained therein include obtaining a confession, punishment, intimidation or coercion, or discrimination. See UN General Assembly, Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, December 10, 1984, 1465 U.N.T.S. 85, 113; S. Treaty Doc. 100-20 (1988); 23 I.L.M. 1027 (1984), Art. 1. Note that no analyses regarding type of perpetrator are presented in the present paper. For the great majority of cases examined here, these acts were perpetrated by members of the Indonesian security services (Armed Forces, Police, Intelligence), with some acts by civilian auxiliaries working with these state personnel. As such, all cases meet the CAT definition. L5315.indd 66 3/28/2024 3:53:55 PM === [PDF page 10] === Patterns of New Order Torture 67 of torture, there were codes for (1) beatings, (2) electric shock, (3) submersion/drowning, (4) cutting/penetrative injury, (5) blunt force injury, (6) whippings, (7) other, and (8) burning. Second, for sexual forms, the categories were (1) forced stripping/nudity, (2) rape, (3) sexual assault, (4) threat of rape or sexual assault, and (5) other forms of sexualized violence. Third, for mental and psychological forms of torture, the codes were (1) threats against self, (2) threats against family/others, (3) being forced to witness the torture of others, (4) others tortured in front of/in place of self, and (5) other coercive forms.28 Many individuals experienced more than one type of torture. Initial Results: Patterns of Physical, Sexual, and Mental Torture In this section, we concentrate on the patterns shown when we examined the intersection between two or more of the variables related to types of torture, which we represent using “heat maps” to visualize intensity over regions and time periods. First, however, we present the frequency of observations for each of the variables using bar charts. Figure 1 shows the distribution of sampled cases across regions and time periods. Figure 2 shows the distribution by types of torture: physical, sexual, and mental. Figure 3 shows the distribution of male, female, and child victims. As shown in Figures 1–3, the initial torture data is primarily drawn from men from East Timor, physically tortured in the 1980s and 1990s.29 Drawing on the approach taken by the Ill-Treatment and Torture (ITT) project,30 a descriptive analysis was conducted to associate the types of torture with particular variables (i.e., time period, geographic regions, and gender/age category). To achieve this, we used crosstabulation analysis to map the general types of torture (i.e., physical, sexual, and mental forms) as a column and the historical period as a row. We then converted the frequency of observed torture types into percentages based on the total incidents of torture recorded within each historical period. This procedure allows us to compare the intensities of different types of torture within time periods, across time periods, and among different categories of victims. To visually represent the analysis, we transformed the percentage data representing the ratio of torture into a six-point ordinal measure. To do this, the range of percentages recorded for each category was evenly divided into six “intensity” classes (very low, low, middle-low, middle-high, high, and 28 As discussed at the end of this article, these categories continue to be refined and expanded in the second version of the IndoTorM dataset currently being prepared. 29 Note that the sample within the present IndoTorM dataset is strongly biased in favor of cases from East Timor, as they were the first materials gathered and entered. The sample is not reflective of our estimate for the actual number of torture cases across different regions and periods of the New Order. Author 1 has gathered data on more than ten thousand additional cases, which are currently being incorporated into the revised version of the dataset. These other cases cover, among others, cases from 1965; cases from “incidents,” including Petrus; cases from criminals and socially marginalized groups; and cases from the “DOM” period in Aceh. 30 The ITT Data Collection Project, led by Courtenay R. Conrad and Will H. Moore, used content analysis to measure more than fifteen thousand allegations of torture made by Amnesty International between 1995 to 2005 and used heatmaps to show intensity of different variables. See, for example, Courtney R. Conrad, Jillienne Haglund, and Will H. Moore, “Disaggregating Torture Allegations: Introducing the Ill-Treatment and Torture (ITT) Country-Year Data,” International Studies Perspectives 14, no. 2 (May 2013): 199–220, https://doi.org/10.1111/j.1528-3585.2012.00471.x. L5315.indd 67 3/28/2024 3:53:55 PM === [PDF page 11] === L5315.indd 68 68 Pohlman et al. Figure 1: The number of torture incidents that occurred in different regions (left) and in different time periods (right) recorded in the initial IndoTorM dataset Figure 2: The number of cases of physical torture (left), sexual torture (center), and mental (right) torture recorded in the initial IndoTorM dataset. Each type of torture is subdivided into a number of subtypes 3/28/2024 3:53:58 PM === [PDF page 12] === Patterns of New Order Torture 69 Figure 3: The number of male, female, and child victims included in the initial IndoTorM dataset very high; see Figure 4). We then visualized the results using a heat map to distinguish the six-point ordinal measure (see Figure 4). Cases of Torture by Region and Time First, we explored the dataset to examine how different types of torture were associated [128.59.222.107] Project MUSE (2026-01-31 03:49 GMT) Columbia University Libraries with different regions and different periods during the New Order. To examine this association, we calculated a six-point ordinal measure and created a heat map using the procedure described above (see also Figure 4) for each time period and region. The results are shown in Figure 5. The darker the shade within a box, the higher the intensity of reported cases of torture in that location within that time period. As Figure 5 shows, consistent with Figure 1, above, the majority of cases contained within the present version of the IndoTorM dataset comprise information on torture perpetrated in East Timor between 1970 and 1998. Figure 5 also allows us to identify clusters of torture data in the sample from Java and Sumatra from earlier decades, with the next largest number of cases sampled from Java between 1967 and 1969. Physical Forms of Torture We then considered physical torture in more detail, to examine how the relative frequency of the different subtypes of physical torture varied across time periods, across regions, and according to the gender and age group. As noted above, physical forms of L5315.indd 69 3/28/2024 3:53:59 PM === [PDF page 13] === 70 Pohlman et al. Figure 4: The process followed to transform the relative frequency of different subtypes of torture (within any torture type) into heat maps. Heat maps allow us to compare the relative frequency of different subtypes of torture over time, across regions, or among different victim demographic groups torture were coded across eight main subtypes, from beatings and blunt force violence through to burnings. These categories, however, did not encompass the very broad range of physical torture reported across the more than four thousand cases sampled in the dataset thus far. The category of “other” (7) was used to capture the many forms without a distinct code, such as cases whereby victims were dragged behind a vehicle, or forced to eat spoiled food or harmful objects, or were “dijemur,” an expression that meant that they were forced to stand in the sun for hours on end without drinking, often becoming ill with heatstroke.31 One such case was that of Joseph Sinaga, a fifty-five-year-old farmer, who reported how he was tortured in March 1993 by being “put in the courtyard of the B__ Precinct Police Headquarters to ‘dry’ in the sun all day, naked and without a drop to drink” among a range of other physical, sexualized, and psychological forms of violence 31 In other contexts around the world, this method is sometimes called “planton” or forced standing, often when exposed to the elements; see, for example, how it is defined and coded within the “Human Rights Information and Documentation” (HURIDCOS) system, a documentation system for various forms of human rights abuses originally set up in the early 1980s (discussed further, below). In some of the Indonesian cases documented, the victim also experienced some forced posture and/or confinement while remaining in the sun. In some further cases, victims were forced to perform intense physical activity to exacerbate hyperthermic illness. L5315.indd 70 3/28/2024 3:54:01 PM === [PDF page 14] === Patterns of New Order Torture 71 Figure 5: Changes in the locations of torture over time within the IndoTorM dataset used to force him to confess to a crime he did not commit.32 Given Indonesia’s tropical climate, this method can rapidly induce hyperthermia.33 The “other” code was also applied, however, when there was insufficient detail and/or clarity within the source document to apply any of the other subtype codes.34 This was often the case when the torture was described by a reporting witness, who may have seen the victim at some point and noted injuries, but had no information about how they were inflicted. A report by TAPOL about the murder of Batak church activist Herbert Hutasoit in 1994 in North Sumatra, for example, notes briefly the torture of others: “on 29 May, three villagers were abducted by the [local military] Task Force. When they returned home, their faces were unrecognizable and their bodies were covered in bruises and wounds.”35 In such examples, although there is sufficient information to conclude that these individuals experienced physical torture, no further specific data can be recorded. In some further cases, particular methods of torture described in reports might be coded as more than one subtype. For example, the professor and journalist Adnan Beuransyah was detained and tortured to confess that he was a member of the Free Aceh Movement (GAM: Gerakan Aceh Merdeka) in the early 1990s. He reported a range of 32 ELSAM, Revealing Tortures by Public Officials (Jakarta: ELSAM, 1996), 38. Joseph Sinaga was one of several cases of people detained and tortured documented by ELSAM during a land dispute between locals and officials in the Sei Lepan Local Resettlement Project in Alur Dua village, Babalan, Langkat, North Sumatra. 33 Although not explored in the present paper, this was one form of torture where the reported outcome for the victim was highly age dependent: infants and small children are far more vulnerable to heatstroke, as their body temperatures increase much faster; see Henry F. Krous, Julie M. Nadeau, Richard I. Fukumoto, Brian D. Blackbourne, and Roger W. Byard, “Environmental Hyperthermic Infant and Early Childhood Death: Circumstances, Pathologic Changes, and Manner of Death,” The American Journal of Forensic Medicine and Pathology 22, no. 4 (December 2001): 374–82, https://doi.org/10.1097/00000433-200112000-00008. In these cases, usually as a form of psychological torture against the mother, her baby/child was placed in the sun to be “dijemur.” 34 In such cases, although the victim (or reporting witness) testified that they had been physically tortured, there was insufficient detail about the methods used to code specific forms with a high enough degree of certainty. 35 “Batak Church Activist Murdered in Cold Blood,” TAPOL Bulletin 124 (August 1994), 14, https://vuir.vu.edu.au/26063/1/TAPOL124_compressed.pdf. Note that the report was taken from Jakarta Post, June 19, 1994. L5315.indd 71 3/28/2024 3:54:02 PM === [PDF page 15] === 72 Pohlman et al. Figure 6: Variation in the relative frequency of subtypes of physical torture across time periods in the IndoTorM dataset physical torture methods used, including how “a length of wood was held down over my knees. Another length was placed under my buttocks which was then pumped up and down.... My knees felt as though they would break.”36 This form was coded as both blunt-force torture (as there was sustained/repeated force applied with a flat object) as well as “other” because there was insufficient information to detail the method specifically. We created heatmaps to examine how forms of physical torture changed (or stayed the same) across time periods (Figure 6), among gender and age classes (Figure 7), and across regions (Figure 8). Figure 6 shows the relative frequency/intensity of subtypes of physical torture sampled across the different time periods in the IndoTorM dataset. As shown in Figure 6, two subtypes of physical torture, beatings and those cases classified as “other,” were consistently practiced throughout the entire 1965–post-1999 period, while a third subtype (penetrative injury) was prevalent only during the period 1967–69. When looking at all reported cases within the dataset, this heatmap generally shows that the forms of torture did not change greatly based on the time period in which they were perpetrated. Any observation of consistency for the category of “other” should be interpreted with caution, however, because of the great heterogeneity of cases that this code was used to describe. When the forms of reported physical torture are disaggregated by gender (male/ female) and age group (adult/child), however, other patterns are revealed. Figure 7 shows the breakdown of physical torture subtypes by time period for men, women, and then child victims. When separated by these demographic variables, beatings and “other” forms of physical torture remain the most dominant forms of torture for male and female victims, but with some variations. For torture involving penetrative wounding (cutting into the body with any implement, or cutting off any part of the body), male victims 36 Amnesty International, Power and Impunity: Human Rights under the New Order (London: Amnesty L5315.indd 72 3/28/2024 3:54:03 PM === [PDF page 16] === Patterns of New Order Torture 73 Figure 7: Variation in the relative frequency of subtypes of physical torture across time periods among men, women, and child victims experienced this form of physical torture at a higher rate for the period 1967–69, while women reported this form of torture more from 1999–onward. For the period 1967–69, women victims also reported higher rates of torture involving electric shocks.37 The patterns of physical torture used against children, however, show some differences from those experienced by adults (Figure 7). Children were far less frequently 37 It should be noted that physical torture involving electric shock often coincided with sexualized abuse. In most cases of torture involving electricity, victims were also forced to strip partially or totally naked. Further, while electrical cables (usually run from a car battery) were often attached to extremities such as the ears, fingers, or toes, they were also often attached to the genitals (for both men and women) and, for women, to the breasts. L5315.indd 73 3/28/2024 3:54:04 PM === [PDF page 17] === 74 Pohlman et al. beaten than adults, though they too experienced the many “other” forms of physical torture. Between 1970 and 1975, there is a high frequency of whipping used against child victims, a pattern not seen at all within the experiences of adults. There is also some evidence of techniques of penetrative injuries of children from the 1980s onward. Our final analysis of the physical torture data examined variation in the relative frequency of the different subtypes of physical torture over time across the six different regions included in the dataset. For this analysis, men, women, and children victims were aggregated. Specifically, Figure 8 shows the effect of the interaction of time period and region on the relative frequency of subtypes of physical torture recorded. Examining variation in patterns of physical torture among regions draws attention to the need for further work to elaborate the data from regions and time periods that are less well sampled (e.g., Aceh before the 1980s). Taken at face value, however, the present data show that beatings and “other” forms of physical torture remain consistently the most frequent forms of physical torture sampled. This indicates, generally, that the forms of physical torture employed did not vary substantially in the sample either across regions or time periods. There were some exceptions, however: penetrative injuries showed higher rates of intensity in Java (1965–69), in Papua (1970–75) and in East Timor (1965–66). To a lesser degree, the infliction of burns as part of torture was also more frequent in East Timor (1965–66) and in Java (1976–80). Sexual Forms of Torture The subtypes of sexual torture coded within the IndoTorM dataset, as noted above, covered a range of violent, sexualized acts. As for our analysis of physical torture, we first present a detailed descriptive analysis of the subtypes, and then examine variation by time period, demographic group, and region. These subtypes covered both penetrative forms (rape, whether by a part of the body of the perpetrator or by an implement) and non-penetrative forms of sexual assault, forced stripping and/or nudity, as well as acts of sexualized/gendered coercion and threats and other kinds of sexualized interactions [128.59.222.107] Project MUSE (2026-01-31 03:49 GMT) Columbia University Libraries related to the victim’s experience of torture. The experience and perpetration of torture are both gendered and gendering, and many of these “other” forms of sexual torture documented within the dataset were those that reified or manipulated gender and sexual norms in Indonesia to harm and humiliate victims.38 Examples of “other” sexual violence included forcing victims to undertake demeaning and animalistic behaviors while naked, deliberately altering the appearance of victims (such as by cutting off or 38 Here, we refer to how torture tends to feminize or emasculate victims, how torturers perform aspects of masculinity through violence, and the various social processes that bond groups of (male) perpetrators. See, for example, Darius Rejali, “Torture Makes the Man,” South Central Review 24, no. 1 (Spring 2007): 151–69, https://doi.org/10.1353/scr.2007.0015; and Martha K. Huggins and Mika Haritos-Fatouros, “Bureaucratizing Masculinities among Brazilian Torturers and Murderers,” in Masculinities and Violence, edited by Lee H. Bowker (London: Sage, 1998): 29–54. For previous research findings on the gendered and sexualized forms of torture used in Indonesia, see, during the mid-to-late 1960s period, Pohlman, Women, Sexual Violence, 88–98, 136–51; in East Timor, Hannah Loney, In Women’s Words: Violence and Everyday Life during the Indonesian Occupation of East Timor, 1975–1999 (Eastbourne, UK: Sussex Academic Press, 2018); and in Aceh, see the Final Report of the Aceh Truth and Reconciliation Commission (forthcoming 2023). L5315.indd 74 3/28/2024 3:54:04 PM === [PDF page 18] === L5315.indd 75 Patterns of New Order Torture Figure 8: Variation in the relative frequency of subtypes of physical torture across time periods among 75 the six regions included in the IndoTorM dataset 3/28/2024 3:54:07 PM === [PDF page 19] === 76 Pohlman et al. shaving their hair, particularly for women victims), and coercing sex in exchange for halting of other forms of physical torture or to obtain food.39 Further, sexual forms of torture frequently occurred concurrently with other physical and mental forms of torture, for both men and women. Men and women commonly reported being stripped partially or totally naked prior to interrogation. For both male and female victims, forced nudity often signaled the use of other forms of torture. For female victims, forced nudity more commonly led to other forms of sexual assault. One woman, “Isabel,” for example, was detained and tortured by the Indonesian military in occupied East Timor on three separate occasions during the 1990s. Each time, she explained, she and other women detainees “were forced to sleep with many Indonesian military...” and that, before being raped, “They made us naked and they tortured us with electric shocks on our breasts and eyelids.”40 Figure 9 shows the heatmap for the intensity with which these various subtypes of sexual torture were reported within the dataset. As Figure 9 shows, the most frequently reported form of sexual torture across most time periods was forced stripping/nudity. In contrast, rape shows an increase in torture practice in the 1970s and 1980s, persisting into the 1990s when other sexual assaults are also more commonly reported. When this data is broken down by the victims’ demographic categories, however, there are further observable patterns (see Figure 10, below). For male victims, the dominant type of sexual torture experienced was forced stripping/nudity, while, for female victims, both forced stripping and rape were dominant. For children, no cases are reported of sexual torture prior to the 1970s, something that certainly reflects sampling bias (the comparatively smaller number of children and of cases before the 1970s), but also has the potential to represent protective norms in operation, given that Figure 7 (and below, Figure 13) show child victims experienced physical and mental torture during this period. From the mid-1970s onward, sexual torture is reported, yet the subtypes of sexual torture reported for children were more varied, with the most dominant being stripping and rape. Turning to an analysis of historical continuity and change in forms of sexual torture across different regions, as presented in Figure 11, it must again be acknowledged that the data illustrate the need for more systematic sampling across the whole historical period and regional contexts. For example, the absence of sexual torture data from Eastern Indonesia has the potential to reflect protective norms, but also could arise merely from sampling bias. At face value, the overall data show that stripping was a consistent and dominant type of sexual torture in all regions. In contrast, the practice of rape was more dominant in particular locations and times, such as in East Timor, in which the type of sexual torture was dominated by rape for the end period of occupation. 39 For women and girls, there were also cases whereby their experience of sexual torture occurred concurrently with other sexual crimes over longer periods, such as sexual enslavement. For a discussion of these longer-term sexual crimes within the context of the anti-Communist purges, for example, see Annie Pohlman, “Sexual Slavery, Enforced Prostitution, and Forced Marriage as Crimes Against Humanity during the Indonesia Killings of 1965–1966,” in The International People’s Tribunal for 1965 and the Indonesian Genocide, edited by Saskia E. Wieringa, Jess Melvin, and Annie Pohlman (New York: Routledge, 2019), 96–114. 40 Judicial System Monitoring Programme (JSMP), Torture Survivors: Their Experiences of Violation, Truth and Justice (Dili: JSMP, April 2007), 12. The JSMP was a Timorese legal NGO set up after the independence referendum in 1999 that collated numerous testimonies of abuse during the occupation. L5315.indd 76 3/28/2024 3:54:07 PM === [PDF page 20] === Patterns of New Order Torture 77 Figure 9: Variation in the relative frequency of subtypes of sexual torture across time periods Reports by survivors and various NGOs detail the prevalence of sexual violence used by the security services throughout the occupation of East Timor but many of these reports emphasize its systematic use during the 1990s, consistent with the present data.41 In his 1997 submission to the UN Special Rapporteur on Violence Against Women, for example, activist and scholar George J. Aditjondro detailed the widespread and systemic sexual abuse of women and girls in the territory, stating that “there are substantial patterns of rape, sexual abuse and other types of gender-specific violations being perpetrated against women in East Timor... It is a weapon of the occupying troops, used to subdue the local population.”42 Mental and Psychological Forms of Torture In this section, we first consider the conceptualization and coding of mental torture and then its prevalence across time, by demographic group, and region. It should be acknowledged at the outset that in many of the accounts within the IndoTorM dataset, mental and psychological suffering was described as integral to the experience of torture, including during the physical and sexual forms noted above. These testimonies described intense fear, threats, and coercive and humiliating treatment, among other methods, that caused victims stress and mental pressure. In most cases, these accounts described physical assaults against the victims themselves or acts of some kind that involved 41 For accounts of women’s experiences of sexual and gender-based violence from this period, see Loney, In Women’s Words; Michelle Turner, “Telling” East Timor Personal Testimonies 1942–1992 (Sydney: New South Wales University Press, 1992); Rebecca Winters, ed., Buibere: Voice of East Timorese Women (Darwin: East Timor International Support Center, 1999); and Jude Conway, ed., Step by Step: Women of East Timor, Stories of Resistance and Survival (Darwin: Darwin University Press, 2010). 42 George J. Aditjondro, “Violence by the State against Women in East Timor: A Report to the UN Special Rapporteur on Violence against Women, including Its Causes and Consequences,” East Timor Human Rights Centre, November 7, 1997, 2. L5315.indd 77 3/28/2024 3:54:08 PM === [PDF page 21] === 78 Pohlman et al. Figure 10: Variation in the relative frequency of subtypes of sexual torture reported among men, women, and child victims over time psychological harm. A common example was short- and long-term food deprivation, or being forced to eat rotten food, or to eat or drink something contaminated or harmful. These were methods that involved both physical and mental elements designed to inflict severe suffering on victims.43 Asman, a villager detained with around fifty others during a land dispute in 1993, for example, explained how he and the other detainees were all stripped naked and beaten and then that they “were forced to eat the glass shards of the broken office windows.” He and some other prisoners “were then brought back to 43 See Pau Pérez-Sales, “Hunger: Deprivation and Manipulation of Food as a Torture Method: State of the Art in Research and Ways Forward,” Torture 30, no. 3 (October 2020): 3–19, https://doi.org/10.7146/torture.v32i3.133931. L5315.indd 78 3/28/2024 3:54:09 PM === [PDF page 22] === Patterns of New Order Torture 79 [128.59.222.107] Project MUSE (2026-01-31 03:49 GMT) Columbia University Libraries Figure 11: Variation in the relative frequency of subtypes of sexual torture across time periods among the six regions included in the IndoTorM dataset L5315.indd 79 3/28/2024 3:54:11 PM === [PDF page 23] === 80 Pohlman et al. the back of the [police station] office and thrown in the sewer. It was filled with human excrement. They were forced to eat the feces—eating it themselves and feeding it to their fellow captives.”44 Over the two months that Asman was detained, he described numerous other physical torture methods, some with distinct psychological elements, such as being forced to fight other detainees.45 In some cases, survivors described nonphysical acts carried out against others intended to coerce and intimidate. Commonly, such cases involved the torture of a loved one. These cases involving the torture of a loved one to coerce or intimidate often involved physical or sexual torture, such as the rape of the wife or daughter of a man as an assault against the female victim but also as an act of psychological torture against her relative. What constitutes “psychological torture,” however, covers an amorphous set of classifications and concepts across the legal and medical literature. 46 They cover three broad areas: methods to induce cognitive and emotional suffering (such as threats or humiliation), “no-touch” physical manipulations of the body and environment (sleep/ food deprivation, very cold/hot conditions, solitary confinement), and physical violence with intended psychological or emotional harm (rape, stress positions/forced postures, wet/dry asphyxia).47 Although there is some blurring between what legally constitutes ill-treatment (rather than torture48), some methods of psychological torture have been deliberately miscategorized as somehow less harmful by those who would minimize their resulting injury; examples are terms used by regimes seeking to excuse torture such as “torture-lite” or “white torture,” which denote the less visible harms (in that they tend to leave fewer physical scars) of these methods.49 For psychological methods of torture in the IndoTorM dataset, only those cases that made specific reference to some form of psychological or mental harm or stress were coded. There was great variance in the methods reported and many were coded as “other form of coercion,” often because there were few details within the account beyond the notation that the victim has suffered mental harm. Figure 12 gives the heatmap of the reported cases of mental torture across the different time periods. 44 ELSAM, Revealing Tortures, 36. Asman was also one of those detained in relation to the Sei Lepan land dispute in North Sumatra (see footnote 32, above). 45 ELSAM, Revealing Tortures, 37–38. 46 For example, Joseph El-Khoury, Riwa Haidar, and Andres Barkil-Oteo, “Psychological Torture: Characteristics and Impact on Mental Health,” International Journal of Social Psychiatry 67, no. 5 (August 2021): 500–6, https://doi.org/10.1177/0020764020961800; Ergun Cakal, “Debility, Dependency and Dread: On the Conceptual and Evidentiary Dimensions of Psychological Torture,” Torture 28, no. 2 (August 2018): 15–37, https://doi.org/10.7146/torture.v28i2.106908. 47 See Pau Pérez-Sales, “Psychological Torture,” in Research Handbook on Torture: Legal and Medical Perspectives on Prohibition and Prevention, edited by Malcolm D. Evans and Jens Modvig (Cheltenham: Edward Elgar, 2020), 432–54. 48 See, for example, Metin Başoğlu, Maria Livanou, and Cvetana Crnobarić, “Torture vs Other Cruel, Inhuman, and Degrading Treatment: Is the Distinction Real or Apparent?” Archives of General Psychiatry 64, 49 On the invisibility of such methods and resulting evidentiary challenges, see, for example, Ergun Cakal, “At the Threshold of Justiciable Violence: Configuring and Contesting Torture’s Production,” Law, Culture and the Humanities (forthcoming), https://doi.org/10.1177/17438721211018764. On the evolution of “clean” vs “scarring” techniques in the twentieth century, see Darius Rejali, Torture and Democracy (Princeton: Princeton University Press, 2007). L5315.indd 80 3/28/2024 3:54:11 PM === [PDF page 24] === Patterns of New Order Torture 81 Figure 12: Variation in the relative frequency of subtypes of mental torture across time periods As seen in Figure 12, there were no dominant mental forms of torture reported across all time periods. Forcing prisoners to watch their fellows being tortured was common, as were threats and “other” coercive measures. When these data are separated by demographic group, however, some differences across time may be discerned (Figure 13, below). According to Figure 13, male victims more consistently reported the various “other” forms of coercive forms. The types of psychological torture used against female victims, on the other hand, were more variable. Finally, child victims more commonly experienced threats against themselves and others. When comparing the relative frequency of mental forms of torture by region (Figure 14), it also appears that regional differences in practice are discernible. In Aceh, for example, threats against self and family members were more common. In other regions, however, the forms of mental torture reported were more variable, such as in other parts of Sumatra, East Timor, and Java. Overall, the patterns found across regions and time periods indicate that mental torture was a common way of treating prisoners. In addition, threats against self and/or family were the most common mental torture in all locations. Social Norms for Torture within the New Order’s Apparatus Looking broadly across all regions, these initial results from the IndoTorM dataset paint a picture of both continuity and change over time in the methods of torture used by the New Order’s security apparatus. Some forms were more consistent over time and region: the practices of stripping and beating, for example, as well as the rape of women and the use of threats against children as a form of mental torture. Other forms seemed to show more fluctuation by region and time period: the whipping of children L5315.indd 81 3/28/2024 3:54:12 PM === [PDF page 25] === 82 Pohlman et al. Figure 13: Variation in the relative frequency of subtypes of mental torture across time periods among men, women, and children and penetrative injuries to children, rape and sexual assault against men, and burning. It was also consistent that there were inconsistencies: the catchall categories of “other” physical, sexual, and mental torture were regularly used by coders. Further ambiguities are created by coding difficulties, discussed further below, yet one interpretation of these categories is that it was common for variability to occur across all contexts, with many types of torture emerging infrequently and not becoming normative or dominant. While there are some exceptions, generally, the methods of physical and sexual torture used did not change greatly over time or region. The only consistent differences were those forms more commonly used against male or female victims, or against children. For physical torture, both male and female victims most frequently experienced beatings and “other” forms of physical assault; when it came to child victims, however, “other” physical attacks were the most common. Stripping was a common form of sexual violence for male, female, and child victims, with men most frequently/consistently reporting forced nudity as part of torture. For female and child victims, however, rape L5315.indd 82 3/28/2024 3:54:13 PM === [PDF page 26] === Patterns of New Order Torture 83 Figure 14: The interaction of time periods and locations on the index of the types of mental torture was the most common form of sexual violence. The intensity of these forms of physical and sexual torture rose and fell at different times but were consistently reported across the various regions of Indonesia. Methods of mental or psychological torture, in comparison, showed greater variation for all demographic groups and time periods. What might these continuities (and variations) in the forms tell us about the New Order’s torture? Regarding the consistency of torture methods over time, we argue that L5315.indd 83 3/28/2024 3:54:15 PM === [PDF page 27] === 84 Pohlman et al. these findings suggest two essential points. The first is that they indicate normative action among those performing this torture; that is, the consistency in the types of torture used points to a potential code of behaviors (or social norms) within the security service’s collective harmdoing (with social norms being the rules or standards for behavior within groups50). This code of behaviors, or social norms, may have meant that some forms of torture were more acceptable (normative) than others and that it was more acceptable to harm certain types of victims using some methods over others. When certain behaviors, such as beatings, strippings, the rape of women, and threats against children, prevail across regions and periods of time, from this perspective it gives an indication of the consistent normative support for such behaviors. In contrast, when we see a behavior appear intermittently, or only in a particular region or period of time, such as the rape of men, or whipping of children, it suggests weaker normative support and even the possible operation of conflicting inhibitory norms (such as social stigma) to prevent the practice.51 Thus, the data suggest that for physical torture, beating a man or a woman was more normatively acceptable than beating a child. For sexual violence, while forced nudity was an acceptable form of violence to use against men, women, and children, it was less acceptable to rape a male victim than to rape a female victim or child.52 For mental forms of torture, however, we see no normative types emerging for adults, but rather variability across regions and times. Again, the results must be interpreted with caution in light of sampling concerns, yet at face value they show a stark contrast with the mental torture of children, where there is an apparent normative type (threats) that is consistently deployed. If the consistency of torture methods indicates normative social harmdoing among perpetrators, then we may ask where these norms arise from. The broader psychological literature highlights that norms can come from leaders’ direction (top-down) and modeling by individuals (e.g., charismatic group members, or celebrities and even media and films).53 They may also arise from bottom-up processes of imitation and peer learning.54 Our data do not speak to these processes, although we speculate that these norms were likely inculcated through socialization, peer support, and reinforcement. [128.59.222.107] Project MUSE (2026-01-31 03:49 GMT) Columbia University Libraries 50 See Joanne R. Smith and Winnifred R. Louis, “Group Norms and the Attitude-Behaviour Relationship: Group Norms and Attitude-Behaviour Relations,” Social and Personality Psychology Compass 3, no. 1 (January 51 See Amiot et al., “Bringing Together Humanistic and Intergroup Perspectives,” 485–504. 52 Data on sexual violence against men, particularly penetrative forms, must be interpreted with caution. In Indonesia, as in most parts of the world, reporting of sexual violence by male victims is rare, with this under-reporting attributed to numerous factors. On this topic, see, for example, Pauline Oosterhoff, Prisca Zwanikken, and Evert Ketting, “Sexual Torture of Men in Croatia and Other Conflict Situations: An Open Secret,” Reproductive Health Matters 12, no. 23 (May 2004): 68–77, https://doi.org/10.1016/ S0968-8080(04)23115-9. 53 See, for example, Niklas K. Steffans, S. Alexander Haslam, Michelle K. Ryan, and Thomas Kessler, “Leader Performance and Prototypicality: Their Inter-Relationship and Impact on Leaders’ Identity Entrepreneurship,” European Journal of Social Psychology 43, no. 7 (December 2013): 606–13, https://doi.org/10.1002/ejsp.1985; S, Alexander Haslam, Stephen D. Reicher, and Michael J. Platow, The New Psychology of Leadership: Identity, Influence and Power, 2nd ed. (London: Routledge, 2020). 54 For example, see Tom Postmes, S. Alexander Haslam, and Roderick I. Swaab, “Social Influence in Small Groups: An Interactive Model of Social Identity Formation,” European Review of Social Psychology 16, no. 1 (March 2005): 1–42, https://doi.org/10.1080/10463280440000062; Craig McGarty, Categorization in Social Psychology (London: SAGE Publications Ltd., 1999). L5315.indd 84 3/28/2024 3:54:15 PM === [PDF page 28] === Patterns of New Order Torture 85 By this we mean that there was likely at least some element of training, of instruction in techniques, particularly within an environment where those taught behaviors were reinforced (i.e., within a militarized social structure).55 We qualify this point, however, by drawing attention to the wider research into the dynamics that allow people, usually men, to torture within militarized institutions. Torture by state actors is always a product of factors operating on multiple levels. This wider research points to several enabling factors, ranging from societal (the state’s paranoia and drive to wipe out perceived internal enemies), structural (systemic impunity with little oversight), through to institutional (small units of men who commit violence with the approval of superiors but often without much close supervision), to personal influences (factors that might lead to men self-selecting into violent roles, among a range of potential motivations).56 All these factors need much further research in the Indonesian context,57 but support and reinforcement of taught behaviors is likely to be an enabling factor for which we see evidence in the normative actions indicated in our findings. The lack of change over time in some of the dominant “accepted” methods for torture, and their consistency across regions, also suggest instruction and coordination. Historians of Indonesia’s New Order period, and particularly scholars of the regime’s various military-led operations in different parts of the archipelago, have pointed to the systematic, even routinized, nature of those operations.58 The patterns of militaryled violence in these different regions of Indonesia, and at different times, are all too similar: Indonesian military counterinsurgency strategies look familiar whether viewed during 1965 or in Timor Leste or Aceh or elsewhere. The examples that reveal an essential uniformity are numerous, but they include “fence of legs” (pagar betis) strategies, recruiting and arming militias, multiple levels of intelligence coordination, and terrorizing civilians through “shock” tactics.59 Given the familiar and repetitive 55 On structures for learning torture, see Darius Rejali’s discussion of what he calls the “craft apprenticeship” hypothesis; see Rejali, Torture and Democracy, 20–21, 573–74, and chapters 18–21. 56 For the most detailed study (done with police torturers in Brazil) on how these factors combine to create “atrocity environments,” see Martha K. Huggins, Mika Haritos-Fatouros, and Philip G. Zimbardo, Violence Workers: Police Torturers and Murderers Reconstruct Brazilian Atrocities (Berkeley: University of California Press, 2002); a summary of these main factors is listed on p. xx. See also Jürgen Mackert, “The Secret Society of Torturers: The Social Shaping of Extremely Violent Behaviour,” International Journal of Conflict and Violence 9, no. 1 (January 2015): 107–20, https://doi.org/10.4119/ijcv-3071; Claudia Reyes-Quilodran and Martha Cottam, “Explaining Torture: A Case Study,” International Journal of Criminology and Sociology 7 (July 2018): 206–23, https://doi.org/10.6000/1929-4409.2018.07.15. 57 Research on military structures gives some insight into these questions; see, for example, Douglas Kammen and Siddharth Chandra, A Tour of Duty: Changing Patterns of Military Politics in Indonesia in the 1990s (Ithaca, NY: Cornell Modern Indonesia Project, 1999). Research on perpetrators of state violence, including torture, is extremely rare in Indonesia. Some exceptions include David Jansen, “The Central Role of the Police in Networked Security in Indonesia: A Case Study of Yogyakarta” (PhD diss., The Australian National University, 2010); and filmmaker Joshua Oppenheimer’s work with perpetrators of the 1965–66 mass violence, see Joran ten Brink and Joshua Oppenheimer, eds., Killer Images: Documentary Film, Memory and the Performance of Violence (London: Columbia University Press, 2012). 58 See, for example, Richard Tanter, “Intelligence Agencies and Third World Militarization: A Case Study of Indonesia, 1966–1989, with Special Reference to South Korea, 1961–1989” (PhD diss., Monash University, 1991). See also Geoffrey Robinson, “If You Leave Us Here, We Will Die”: How Genocide was Stopped in East Timor (Princeton: Princeton University Press, 2010), 41–46. 59 On similarities between methods in Aceh and Timor Leste, see Douglas Kammen, “Notes on the Transformation of the East Timor Military Command and Its Implications for Indonesia,” Indonesia 67 (April 1999): 61–76. On Aceh, see Geoffrey Robinson, “Rawan is as Rawan Does: The Origins of Disorder in New Order Aceh,” Indonesia 66 (October 1998): 127–56; Matthew N. Davies, Indonesia’s War over Aceh: Last Stand L5315.indd 85 3/28/2024 3:54:15 PM === [PDF page 29] === 86 Pohlman et al. nature of these other forms of military-led tactics, it makes sense that their forms of torture are similar too.60 Though a specific Indonesian military manual on torture— comparable, say, to the manuals used in the School of the Americas training in Latin America61—has never been made public, certainly other documents have come to light over the years that would suggest the coordinated use of torture (amongst other crimes) by the state security apparatus.62 Our initial findings further support an argument not only that norms around torture developed within Indonesia’s security services but that institutional structures of learning and socialization made them possible. At the same time, the variability of our data suggest some of the limits of normative torture that might be operating in Indonesia: institutional ambivalence about some forms of torture (such as the rape of men) that may in turn be associated with other social norms, such as age and gender roles. Conclusion: Expanding the IndoTorM Dataset The IndoTorM project was designed to bring together and analyze the numerous accounts of torture committed under the New Order, to see what patterns might emerge, or what relationships might exist between the many variables involved, such as type of actor, type of victim, location, and time, among several other factors. The initial results presented in this paper have focused on the patterns in the types of torture used against male, female, and child victims, across time and geographic location. These first analyses have demonstrated that there are clear patterns in what types of torture were used more frequently, based on whether the victims were male or female or children. Many of these patterns remain stable across time and geographic locations. Using a psychological approach to interpret these results, we argue that these findings are suggestive of normative social harmdoing within the Indonesian security services. We further speculate that these norms in the types of torture, which became more “acceptable” for use against particular types of victims, were supported and reinforced through institutional policies and practices. In addition, the variability in types of torture shows what is perhaps not consistently supported (such as the on Mecca’s Porch (London: Routledge, 2006). On Timor Leste, see George J. Aditjondro, “Ninjas, Nanggalas, Monuments, and Mossad Manuals: An Anthropology of Indonesian State Terror in East Timor,” in Death Squad: The Anthropology of State Terror, edited by Jeffrey A. Sluka (Philadelphia: University of Pennsylvania Press, 2000), 158–88; and Robert Cribb, “From Total People’s Defence to Massacre: Explaining Indonesian Military Violence in East Timor,” in Roots of Violence in Indonesia: Contemporary Violence in Historical Perspective, edited by Freek Colombijn and J. Thomas Lindblad (Leiden: KITLV Press, 2002), 227–41. On West Papua during the New Order, see Carmel Budiardjo and Liem Soei Liong, West Papua: The Obliteration of a People, Rev. ed. (Surrey: TAPOL, 1988). 60 While the majority of cases in the current IndoTorM dataset relate to these “larger” cases of mass violence, in the next version, more cases involving torture during incidents and criminal suspects are being integrated, to test whether methods are similar. 61 On these manuals, see, for example, Bill Quigley, “The Case for Closing the School of the Americas,” The Brigham Young University Journal of Public Law 20, no. 1 (Winter 2005): 1–34. 62 Some examples from occupied East Timor come from the Yayasan HAK collection, discussed in Samuel Moore, “The Indonesian Military’s Last Years in East Timor: An Analysis of Its Secret Documents,” Indonesia 72 (October 2001): 9–44. See also the 2007 Kopassus documents from Kotaraja, Papua, published by journalist Allan Nairn; see Allan Nairn, “Breaking News: Secret Files Show Kopassus, Indonesia’s Special Forces, Targets Papuan Churches, Civilians. Documents Leak from Notorious US-Backed United as Obama Lands in Indonesia,” November 9, 2010, https://www.allannairn.org/2010/11/breaking-news-secret-files-show.html. None of these documents give specific instructions on torture. L5315.indd 86 3/28/2024 3:54:15 PM === [PDF page 30] === Patterns of New Order Torture 87 whipping of children) as well as documenting the persistent use of local innovation (the high relative frequency in the “other” categories). Yet these first analyses from the IndoTorM dataset lead to several more questions. There are many, but they include what will wider sampling from across regions and time periods further reveal about the New Order’s torture? Can we unpack the “other” categories to distinguish innovation of practice from unknown/generic descriptions? To answer these questions, we are redesigning the IndoTorM dataset to expand the range of codes to capture information more closely and consistently on the types of torture, where it happened, when, against whom, by whom, and in what circumstances. In our preparation of this manuscript, as noted above, we drew on analyses performed by the ITT project, particularly their heat map visualization techniques. Our interest in being able to undertake comparative analyses with their data and others has prompted us to begin work to make the next version of the dataset more comparable to other systems of coding torture. In particular, we aim to draw on the most comprehensive coding scheme put forward to date, which is the Human Rights Information and Documentation System (HURIDOCS), used internationally mainly by NGOs to document human rights abuses.63 In the next version of the IndoTorM dataset, we have incorporated many of the HURIDOCS codes for describing torture types to facilitate future comparative analyses with datasets from other countries.64 In future research, we also will incorporate data from the many other cases of torture compiled, to sample cases more consistently across regions and time periods. This larger sample will incorporate cases of torture from groups of victims mostly left out of the present dataset, such as those detained and tortured because of their alleged criminal activities and other socially vulnerable groups (e.g., sex workers and drug users). The present IndoTorM dataset findings give us valuable data about the historical patterns of torture during the New Order. However, by expanding the coding and the sample, our hope is that we will be able to further pinpoint patterns and clusters and better answer further questions about how and why torture became normalized during the New Order. The present paper makes five novel contributions to Indonesian studies, to the study of torture, and to the study of normative harmdoing. We provide the first largescale systematic reporting and analysis of New Order torture data in Indonesia, an 63 On the establishment of HURIDOCS, see “Human Rights Information and Documentation System (HURIDOCS) Established,” Commonwealth Law Bulletin 9, no. 1 (January 1983): 289–91. To see the full range of tools, go to www.huridocs.org. 64 Note we also consulted other instruments, such as the Istanbul Protocol—a manual first adopted in 1999 by the UN particularly for medical personnel investigating and documenting torture cases—which outline a range of torture methods more commonly seen across different contexts around the world. See Istanbul Protocol: Manual on the Effective Investigation and Documentation of Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment, UN Doc HR/P/PT/8/Rev. 1 (2004), at para. 145. Part G of the Protocol gives a list of torture methods for investigatory and medical personal as a guide. Chapter V gives instructions on how to interview and take the medical history of victims, as well as a close description of the medical aspects of the most common forms of physical and sexual torture, while chapter VI provides a guide on psychological evidence of torture. See also José Quiroga and Jens Modvig, “Torture Methods and Their Health Impacts,” in Research Handbook on Torture: Legal and Medical Perspectives on Prohibition and Prevention, edited by Malcolm D. Evans and Jens Modvig (Cheltenham: Edward Elgar, 2020), 410–31. See also the Torturing Environmental Scale (TES), developed to measure methods based on which human function is attacked; see Pau Pérez- Sales, Psychological Torture: Definition, Evaluation and Measurement (London: Routledge, 2017). L5315.indd 87 3/28/2024 3:54:15 PM === [PDF page 31] === 88 Pohlman et al. important contribution to understanding this widespread, endemic crime used by that regime. Relatedly, the paper presents what is to our knowledge among very few published analyses of large-scale torture data from anywhere in the world and one of the largest torture datasets ever analyzed by scholars to date. Third, our systematic consideration of stability and change in the types of torture used across various time periods, regions, and targets (by age and gender) is a novel application of the study of normative harmdoing, expanding on previous analyses.65 Finally, the present paper offers an in-depth, scholarly examination of an understudied phenomenon that is literally of life and death importance: the evolution and persistence of violent torture practices over decades. Analyses of torture and mass violence overall are rare, and the present systematic comparison of sexual, physical, and mental torture for men, women, and children is nearly unique, to our knowledge. We hope that it will evoke commentary and research responses from scholars of violence and gender in genocide studies and in gender studies more broadly and to scholars of harmdoing in Indonesia and around the globe. 65 The psychology of normative harmdoing has primarily been addressed with single-event (cross-sectional) experimental and survey methods and has focused on behavior such as discrimination in hiring, see Catherine E. Amiot, Sophie Sansfaçon, Winnifred R. Louis, and Martin Yelle, “Can Intergroup Behaviors Be Emitted Out of Self-Determined Reasons? Testing the Role of Group Norms and Behavioral Congruence in the Internationalization of Discrimination and Parity Behaviours,” Personal and Social Psychology Bulletin 38, no. 1 (January 2012): 63–76, https://doi.org/10.1177/0146167211429804, or harmful behavior in sports, see Catherine E. Amiot, Sophie Sansfaçon and Winnifred R. Louis, “How Normative and Social Identification Processes Predict Self-Determination to Engage in Derogatory Behaviors against Outgroup Hockey Fans,” European Journal of Social Psychology 44, no. 3 (April 2014): 216–30, https://doi.org/10.1002/ejsp.2006). L5315.indd 88 3/28/2024 3:54:15 PM