Introduction

To claim that civil wars are “exceptionally brutal”^[i] is an empirically proven cliché, given that atrocities feature prominently between insurgencies and counterinsurgencies.^[ii] As an analytical category in the study of war, “atrocity” encapsulates excess: it refers to moments where extreme violence is applied, in which cruelty and shock value serve as primary motivations for perpetrators.^[iii] To understand why and against whom they occur, scholarship converges on a set of functionalist and psychosocial explanations that treat atrocities as instrumental to war-making. Understanding what motivates such violence and how neighbours could come to kill one another has implications for legal accountability, and ethical standards for warfare.^[iv] Nevertheless, this produces what Dwyer and Ryan (2012) called “the concealment of the massacre among the perpetrators,”^[v] in which atrocities’ victims are rarely assessed as complex subjects.

Drawing on the internal conflict between the Peruvian state and Sendero Luminoso (the Shining Path) from 1980 to 2000, I argue that atrocities can be worldmaking, moulding subjectivities and erasing boundaries between victims, spectators, and perpetrators. “Worldmaking” is a concept introduced by American philosopher Nelson Goodman (1978) to describe the ways in which humans use language, symbols, rituals, and science to construct multiple parallel worlds through which life can be experienced.^[vi] Rather than asking whether atrocities are planned or aberrational, this essay investigates the role atrocities play within insurgent projects. First, I review the functionalist and micro-political schools of atrocity scholarship, and later describe atrocities committed in Peru. The essay finds that ritualised atrocities can function as a mechanism of subject formation and permeability between categories of political belonging.

Atrocity Literature: Between Utility and Aberration

Traditional atrocity scholarship examines acts of war that are deemed “excessive,” “cruel,” and “irrational.”^[vii] Indeed, Mitton (2012) defines atrocities as “the deliberate infliction of cruel physical harm on groups or individuals.”^[viii] By focusing on “hot-blooded” and “cold-blooded” acts,^[ix] atrocities are framed as either planned, or spontaneously driven by indiscipline or faulty intelligence.^[x]

Functionalist scholarship treats atrocities as instruments whose use is shaped by strategic incentives and organisational characteristics. Some scholars emphasise coercion as a viable mode of population control. Heuser (2012) demonstrated that the historical prevalence of indiscriminate violence runs contrary to “hearts and minds” theories,^[xi] while Hazelton (2017) argued that successful counterinsurgencies secure backing from elites and employ brute force, rather than cultivating popular legitimacy.^[xii] Other analyses link violence to structural and organisational incentives. Collier and Hoeffler (2004) treat ideology as epiphenomenal to economic motivations for belligerency,^[xiii] and Weinstein (2006) argues that the degree to which insurgents depend on noncombatants for resources indicates whether their relations are likely to be peaceful or violent.^[xiv] Kalyvas (2006) points out that rebels and state actors share the problem of accurately distinguishing their enemies from the wider population, and the impetus to disrupt support mechanisms between communities and belligerents.^[xv] These authors agree that atrocities become likely where insurgents and counterinsurgents are more insecure.^[xvi] The functionalist perspective does not, however, treat atrocities as politically and socially transformative to the identities of those involved. An underlying assumption is that victims, perpetrators, and bystanders are to some extent fixed categories,^[xvii] before, during, and after atrocities take place.

Where functionalists explain when and where atrocities occur, micro-political perspectives investigate how individuals are compelled to inflict cruel suffering.^[xviii] For Fuji (2009), violence between people in close spatial, cultural, social, political, and economic proximity “destroys not just bodies, but bonds.”^[xix] Mitton’s (2012) study of child soldiers in Sierra Leone reveals that participating in atrocities inoculates perpetrators from the social units and moral structures within which they previously existed.^[xx] If committing atrocities can both express and affect an armed group’s embeddedness in society, then these authors echo Sanín and Wood’s (2014) argument that building collective ideology conditions insurgents’ use of violence, and bridges heterogeneous motivations for participating in an armed group in the first place.^[xxi] Scholarship has thus attended to widespread calls for greater attention into psychological factors at the individual level, which explain why atrocities are committed.^[xxii] However, this shift comes at the expense of a more agential interpretation of the role and status of victims, rarely treating them as active political subjects.

Both the functionalist and micro-political views leave victims as analytically static objects. Recentering them entails understanding their connection (or lack thereof) to perpetrators, and reframing atrocities as essential moments of interaction. Atrocities enable perpetrators to communicate “through myth, memory, or both.”^[xxiii] Fundamentally, this process is underpinned by an “othering” of victims which enables any harm levied against them to communicate notions of status, subjugation, and a correction of historical grievances.^[xxiv] If, per Frankfurter (2025), atrocities can “manifest… a world turned upside down,” it follows that they must be witnessed, and convey a set of underlying ideas.^[xxv] Through mutilation and cruelty, Dwyer and Ryan (2012)  claim that victims’ bodies often become a collective “stage on which suffering is inflicted.”^[xxvi] But Fuji (2009) complicates this understanding by suggesting that one day’s perpetrator or bystander may be the next day’s victim.^[xxvii] These authors inherit Foucault’s (1995) argument that justice triumphs in the spectacle of punishments applied.^[xxviii] Atrocities can therefore shape political subjects through performance and blurring the boundaries between categories of responsibility, spectatorship, and victimhood. To understand why a given atrocity happened, understanding their forms and subliminal social relations becomes as important as assessing strategy, organisations, and psychology.

Peru’s Internal War and the Shining Path

Peru’s internal conflict offers a useful window into atrocity as a worldmaking phenomenon, given the ideological absolutism and systematic employment of coerced participation in their violence. The Shining Path emerged in 1962 as a Maoist left-wing political party, out of the reopened University of Huamanga in the Andean highland region of Ayacucho, finding leaders in faculty and members among students.^[xxix] In their majority, students were of local Indigenous descent, speaking Quechua in their communities as well as Spanish.^[xxx] Its leader, Abimael Guzmán Reynoso, taught philosophy and bridged debates related to the implementation of Marxism and Maoism.^[xxxi] By 1978, Senderistas led the University’s ambitious “extension services” agenda across highland communities in Ayacucho, such as practice farms and education for mature students.^[xxxii] These programs gained prominence as public welfare and international development programs shuttered during a retrenchment of American foreign aid and the 1968 military coup.^[xxxiii] Furthermore, this role embedded the Shining Path as an institution within local communities, even as its leadership remained culturally and linguistically distinct. The majority of leaders did not hail from poor, rural Andean highland communities, as they were mestizos from wealthier, urbanized coastal regions.^[xxxiv] Through labour, marriage, and deep participation in the quotidian of rural communities, Palmer (1986) declared that the Shining Path grew to “[see] themselves as the true vanguard of the peasant proletariat.”^[xxxv] The University of Huamanga thus afforded the Shining Path a vehicle to build popular legitimacy without violence. As a result, rural communities in Ayacucho provided a social base that would later supply the movement with resources and recruits.

The Senderistas began their armed struggle in 1980, breaking with left-wing parties that participated in the first general elections after 12 years of military rule.^[xxxvi] The Sendero ought to tip the scales of local and national legitimacy in their favour by “destroying local power, whether state or traditional, and constructing ‘people’s support bases.’”^[xxxvii] Until 1982 they largely sabotaged infrastructure, before escalating to direct confrontations with local police, prison breakouts, and assassinating Ayacucho’s political and business elites.^[xxxviii] President Belaúnde declared a state of emergency across Ayacucho by 1983, enacting military oversight over much of the region.^[xxxix] The military’s aggressive deployment drove the Shining Path to infiltrate other regions and adapt methods toward securing “support bases,” escalating the conflict to a national scale.^[xl] Between 1989 and 1992, Senderistas controlled broad swathes of central and southern Peru, and state collapse was perceived to be imminent.^[xli] Guzmán called for intensified attacks against the capital of Lima, including a bombing that killed 25 and injured over 150.^[xlii] After dissolving parliament in 1992, President Fujimori executively strengthened the immunity and independence of state security services, increasing the role of state-backed death squads and counter-terror units against the Sendero.^[xliii] Guzmán was captured in Lima by the year’s end, disrupting decision-making and leading the Shining Path to significantly deescalate violence.^[xliv] In 1993, an incarcerated Guzmán publicly appealed for demobilisation, overriding leaders who continued advocating for “people’s” war.^[xlv] As such, by the year 2000, Senderistas were reduced to remote pockets of the Peruvian Amazon.^[xlvi]

The Shining Path’s peasant membership and its emphasis on popular legitimacy did not impede them from committing atrocities. In these cases, Andean peasants were often victimised, but could also be coerced to participate in violence. The most famous atrocity, the 1983 Lucanamarca Massacre, saw a column of Senderistas publicly torture, execute, and mutilate 69 peasants, some of which were their own family members.^[xlvii]Guzmán advocated for a strategy of “counter reestablishment” in which Sendero cells would operate in towns close to military bases, inducing state-backed abuses that would build antipathy to the regime.^[xlviii] “Forced retreats” relocated entire communities to labour in nomadic Sendero camps; this gave Senderistas opportunities to enforce their ideology, obliging all people to reside under one roof, hosting indoctrination programs for youth, and redistributing assets expropriated from captured communities.^[xlix] In “people’s trials,” Senderistas punished transgressions with summary executions that the victim’s relatives and community would be forced to celebrate with song, dance, and drink.^[l] Townspeople generally advocated for physical punishments short of death, while militants preferred conducting extremely ‘intimate’ executions and mutilations.^[li] After slaying a businessman in the town of Chungui, Senderistas forced the townsfolk to accept their redistribution of the victim’s wealth, properties, and assets. In larger urban centres like Lima, the Shining Path commandeered political, business, academic and trade institutions with threats, and often through targeted assassinations and terror attacks.^[lii] Among indigenous communities in the Peruvian Amazon, insurgents attacked non-indigenous settlers, miners, missionaries, and criminals.^[liii] The Sendero thus adjusted their use of violence in accordance with local political contexts.

It is important to recognise that the Sendero, armed forces, and state-backed militias waged a war marked by massacres and human rights abuses. Atrocities were not one-sided nor were they concentrated squarely in Senderista forces, despite their primary authorship throughout the conflict.^[liv] Degregori (1998) claims that reciprocal atrocities committed by rondas, or peasant self-defence forces, prevented the Shining Path from expanding to northern highland regions.^[lv] Military units were accused of coercing the public to attend humiliation events for suspected “subversives,” carrying out summary executions of suspected collaborators, forcibly mobilising rondas, and obliging them to carry out acts of torture.^[lvi] In this sense, atrocities committed by both the Shining Path and state-backed forces enforced social and political allegiances, and appeared to blur the boundaries between victims, perpetrators, and spectators.

Atrocity as Performance

Atrocities committed by the Shining Path and documented above, including the Lucanamarca Massacre and “people’s trials,” reordered social subjectivities by ritualising their performance, and the forced participation of victims. These attempts proved self-defeating, as the forced involvement of civilians in indiscriminate violence hollowed the Sendero’s rural “support bases,” mobilized rondas, and positioned the state to “win hearts and minds” by delivering extension services and public infrastructure.^[lvii]Shining Path atrocities certainly align with functionalist and psychosocial perspectives, since they served as instruments of population control, and to unify perpetrators in their ideological mission. However, the ritualistic and participatory character of Senderista violence against non-combatants reveals an additional element of worldmaking: an active and creative process of constructing, shaping, and negotiating the reality experienced by individuals and their communities.

This logic is arguably strongest in insurgencies that converge ideological absolutism with social infiltration, instrumentalising violence to produce subjects as well as to defeat enemies. Per Gorriti (1999), Senderistas were conditioned to kill and die for the “immense work of moral and material destruction” upon which Guzmán’s theory of victory was predicated.^[lviii] As such, Degregori highlighted their use of violence for “purification,” capable of subverting social systems by being applied in absolutist terms.^[lix] It was a doctrine which viewed militants and peasant communities as a unified force whose “only will was to kill and die for the party.”^[lx] The outward manifestation of these ideas was to torture, kill, and mutilate detractors regardless of the gravity of their transgression, as described in “people’s trials.”^[lxi] Frankfurter and Foucault would characterize these as both punitive and pedagogical spectacles^[lxii] which clashed against conflicting moral and institutional hierarchies, like the family unit. Contrary to Weinstein’s correlation between resource-dependency on civilians and nonviolent relations,^[lxiii] the Shining Path broadly implemented this approach while still largely dependent on local communities for shelter, supplies, intelligence, recruits, and any number of resources necessary for a successful insurgency. The Sendero’s violence caused, beyond what Manrique (1998) assessed as disrupted political expression, a reshaping of the social and cultural “space” that victims resided within.^[lxiv] Atrocities therefore not only punished deviance, but redefined acceptable conduct and a multidimensional sense of belonging.

Another consequence of performativity was the blurring of distinctions between victims and perpetrators. “Counter reestablishment” strategies converged strategic and performative logics by carrying out attacks and atrocities in settlements adjacent to military bases.^[lxv] As discussed, this occurred in order to “provoke blind, excessive reactions from the state.”^[lxvi] Kalyvas’ idea that atrocities are committed in the spaces where armed groups feel most insecure (e.g., nearby military or counter-insurgency staging grounds) helps identify the strategic setting in which civilians may be victimised,^[lxvii] but does less to explain the diversity in the forms atrocities took, including methods of violence, victims, and the interplay of roles between perpetrators and bystanders. Episodes such as the Lucanamarca Massacre, “people’s trials,” and the Chungui killing exemplify the link between performativity and categorical limits: with perpetrators and victims belonging to the same family, by helping determine punishments, and by forcibly celebrating and accepting redistributed assets, civilians and Senderistas fluidly exchanged roles as victims, perpetrators, and spectators.^[lxviii] Ritual violence and coerced participation thus sought to enable peasants’ “strategic identification” with an insurgent worldview, while rendering neutrality infeasible.^[lxix] With extensive mandates afforded to state security services from 1983 onwards, state reactions were guaranteed to be obtuse.^[lxx] Between militant excess and state-backed reprisals, civilians were, together, compelled to violence and the Shining Path’s ideological project. This supports Mitton’s (2012) argument that atrocities can sever perpetrators from preexisting moral and social accountability frameworks,^[lxxi] but extends it by demonstrating how spectators and coerced participants could be similarly transformed. In this way, the Shining Path atrocities functioned as attempts to produce a revolutionary social order.

Conclusion

Atrocities frequently take place during rebellions because violence against civilians can appear useful: community support can be coerced, group cohesion can be strengthened, and population control can erode the social architecture supporting an actor’s enemies. With intelligence asymmetries and moral and ideological conditioning in irregular war, groups and individuals can be compelled into premeditated or spontaneous acts of cruelty. Both functionalist and micro-political accounts therefore rightly emphasise the role of strategic incentives and the material and ideological backgrounds of armed groups as reasons why atrocities occur. However, the case of Peru’s Shining Path indicates that atrocities can also be attempts to transform social relations and proliferate a transgressive political subjectivity. They can be a political technology that erases the boundaries between those who commit, suffer, and witness atrocities. Senderista atrocities were, in short, an exercise in worldmaking. Absolutist violence nevertheless drove resistance against the Shining Path and repudiation by Peruvian society, and ultimately their defeat by the state.

In ideologically motivated insurgencies, atrocities can labour toward a broad objective of social mobilisation. Taken together, these observations extend, rather than contradict, established literature. Atrocities in the context of rebellions cannot be interpreted solely as instruments serving the goals of their perpetrators, or as expressions of armed groups’ resource endowments and organisational structures. Instead, research should treat violence and cruelty against civilians as constitutive practices that may actively produce an insurgent social order. Because violence bridged responsibility and victimhood, research should also account for the forms atrocities can take, and how victims are constituted as subjects. In this way, victims long “concealed” by a fascination with their killers, can be analytically valuable. They can, in some small way, overcome the indignities that history and academia may have inflicted upon them, by being treated as much more than passive recipients of cruelty.

Bibliography

• Collier, Paul, and Anke Hoeffler. “Greed and Grievance in Civil War.” Oxford Economic Papers 56, no. 4 (2004): 563–95. https://doi.org/10.1093/oep/gpf064.
• Degregori, Carlos Iván. “Harvesting Storms: Peasant Rondas and the Defeat of Sendero Luminoso in Ayacucho.” In Shining and Other Paths, edited by Steve J. Stern. War and Society in Peru, 1980-1995. Duke University Press, 1998. https://doi.org/10.2307/j.ctv1168cn5.11.
• Dwyer, Philip G., and Lyndall Ryan. “Introduction: The Massacre and History.” In Theatres Of Violence, 1st ed., vol. 11. Massacre, Mass Killing and Atrocity throughout History. Berghahn Books, 2012. https://doi.org/10.2307/j.ctt9qcw1p.5.
• Foucault, Michel. Discipline and Punish: The Birth of the Prison. 2nd ed. Translated by Alan Sheridan. Vintage Books, 1995.
• Frankfurter, David. “Ritual and the Enemy Body: A New Approach to Modern Atrocity.” Comparative Studies in Society and History 67, no. 4 (2025): 735–60. https://doi.org/10.1017/S0010417525000076.
• Fujii, Lee Ann. “Introduction: Genocide among Neighbors.” In Killing Neighbors. Webs of Violence in Rwanda. Cornell University Press, 2009. https://www.jstor.org/stable/10.7591/j.ctt7z7s5.6.
• Goodman, Nelson. Ways of Worldmaking. Harvester Studies in Philosophy 5. Harvester Press, 1978.
• Gorriti, Gustavo. The Shining Path: A History of the Millenarian War in Peru. New ed. Translated by Robin Kirk. Vol. 1. Latin America in Translation. The University of North Carolina Press, 1999.
• Hatun Willakuy: Abbreviated Version of the Final Report of the Truth and Reconciliation Commission. Truth and Reconciliation Commission, 2004. https://idehpucp.pucp.edu.pe/images/publicaciones/hatun-willakuy-english-version.pdf.
• Hazelton, Jacqueline L. “The ‘Hearts and Minds’ Fallacy: Violence, Coercion, and Success in Counterinsurgency Warfare.” International Security 42, no. 1 (2017): 80–113. https://www.jstor.org/stable/26777810.
• Heuser, Beatrice. “Atrocities in Theory and Practice: An Introduction.” Civil Wars 14, no. 1 (2012): 2–28. https://doi.org/10.1080/13698249.2012.654677.
• Kalyvas, Stathis N. “Barbarism.” In The Logic of Violence in Civil War, 1st ed. Cambridge University Press, 2006. https://doi.org/10.1017/CBO9780511818462.
• Manrique, Nelson. “The War for the Central Sierra.” In Shining and Other Paths, edited by Steve J. Stern. War and Society in Peru, 1980-1995. Duke University Press, 1998. https://doi.org/10.2307/j.ctv1168cn5.13.
• Mitton, Kieran. “Irrational Actors and the Process of Brutalisation: Understanding Atrocity in the Sierra Leonean Conflict (1991–2002).” Civil Wars 14, no. 1 (2012): 104–22. https://doi.org/10.1080/13698249.2012.654691.
• Palmer, David Scott. “Rebellion in Rural Peru: The Origins and Evolution of Sendero Luminoso.” Comparative Politics 18, no. 2 (1986): 127–46. https://doi.org/10.2307/421840.
• Pino H., Ponciano del. “Family, Culture, and ‘Revolution’: Everyday Life with Sendero Luminoso.” In Shining and Other Paths, edited by Steve J. Stern. War and Society in Peru, 1980-1995. Duke University Press, 1998. https://doi.org/10.2307/j.ctv1168cn5.12.
• Ron, James. “Ideology in Context: Explaining Sendero Luminoso’s Tactical Escalation.” Journal of Peace Research 38, no. 5 (2001): 569–92. https://doi.org/10.1177/0022343301038005002.
• Sanín, Francisco Gutiérrez, and Elisabeth Jean Wood. “Ideology in Civil War: Instrumental Adoption and Beyond.” Journal of Peace Research 51, no. 2 (2014): 213–26. https://www.jstor.org/stable/24557417.
• Weinstein, Jeremy M. Inside Rebellion: The Politics of Insurgent Violence. 1st ed. Cambridge University Press, 2006. https://doi.org/10.1017/CBO9780511808654.

---

^[i] Kalyvas (2006, 53)

^[ii] Heuser (2012, 2)

^[iii] Mitton (2012); Fuji (2009)

^[iv] Ibid; Heuser (2012); Fuji (2009); Kalyvas (2006)

^[v] Dwyer and Ryan (2012, xiv)

^[vi] Goodman (1978)

^[vii] Mitton (2012, 105)

^[viii] Ibid, 105

^[ix] Dwyer and Ryan (2012, xv)

^[x] Heuser (2012, 23)

^[xi] Ibid, 3

^[xii] Hazelton (2017, 81)

^[xiii] Collier and Hoeffler (2004)

^[xiv] Weinstein (2006, 339)

^[xv] Kalyvas (2006, 69)

^[xvi] Ibid, 84

^[xvii] Fuji (2009, 8)

^[xviii] Sanín and Wood (2014, 222)

^[xix] Fuji (2009, 3)

^[xx] Mitton (2012, 3)

^[xxi] Sanín and Wood (2014, 221)

^[xxii] Mitton (2012)

^[xxiii] Fuji (2009, 8)

^[xxiv] Dwyer and Ryan (2012, xvii)

^[xxv] Frankfurter (2025, 740)

^[xxvi] Dwyer and Ryan (2012, xviii)

^[xxvii] Fuji (2009, 8)

^[xxviii] Foucault (1995, 34)

^[xxix] Truth and Reconciliation Commission (2004, 127)

^[xxx] Ibid, 127

^[xxxi] Palmer (1986, 128)

^[xxxii] Truth and Reconciliation Commission (2004, 128)

^[xxxiii] Palmer, (1986, 136)

^[xxxiv] Ibid, 137

^[xxxv] Ibid, 138

^[xxxvi] Ron (2001, 569)

^[xxxvii] Truth and Reconciliation Commission (2004, 37)

^[xxxviii] Manrique (1998, 197)

^[xxxix] Palmer (1986, 129)

^[xl] Truth and Reconciliation Commission (2004, 138)

^[xli] Ibid, 68

^[xlii] Ibid, 70

^[xliii] Ibid, 70

^[xliv] Ibid, 70

^[xlv] Ibid, 71

^[xlvi] Ibid, 72

^[xlvii] Truth and Reconciliation Commission (2004, 129)

^[xlviii] Ibid, 133

^[xlix] Ibid, 108

^[l] Ibid, 130

^[li] Degregori (1998, 137)

^[lii] Ibid, 139

^[liii] Ibid, 140

^[liv] Ibid, 18

^[lv] Degregori (1998, 136)

^[lvi] Truth and Reconciliation Commission (2004, 116)

^[lvii] Degregori (1998, 147)

^[lviii] Gorriti (1999, 135)

^[lix] Degregori (1998, 136)

^[lx] Pino (1998, 159)

^[lxi] Truth and Reconciliation Commission (2004)

^[lxii] Frankfurter (2025, 740); Foucault (1995, 34)

^[lxiii] Weinstein (2006, 339)

^[lxiv] Manrique (1998, 199)

^[lxv] Truth and Reconciliation Commission (2004, 133)

^[lxvi] Gorriti (1999, 135)

^[lxvii] Kalyvas (2006, 84)

^[lxviii] Truth and Reconciliation Commission (2004, 106; 129)

^[lxix] Degregori (1998, 132)

^[lxx] Gorriti (1999, 173)

^[lxxi] Mitton (2012, 12)