CO-J-02: Three Accounts — The Falsos Positivos Extrajudicial Killings (2002–2010)

Status: DRAFTWords: 10,853

1. Key Takeaways

The falsos positivos ("false positives") were civilians killed by units of the Colombian Army and presented — dressed, armed, photographed, and registered — as guerrillas or other illegal combatants killed in combat (bajas en combate), in order to inflate the operational results by which units and commanders were measured and rewarded during the Democratic Security era. The mechanics were stable across regions and years: victims were typically young men from poor urban districts or contested rural zones, frequently lured with false job offers by civilian recruiters paid per victim, transported to a distant operational area, killed within hours or days, and "legalised" — the perpetrators' own term — with planted weapons, fabricated combat reports, falsified intelligence annexes, and rapid burials, often as unidentified persons (N.N.) in municipal cemeteries. The practice constitutes, by victim count, the gravest atrocity attributed to the Colombian state in its modern history.
The scale is itself the central contested fact, and the corpus must hold the competing figures explicitly. The Fiscalía General de la Nación's consolidated investigation had documented approximately 2,248 victims by the mid-2010s [TBD-VERIFY: the Fiscalía figure cited by the JEP as its comparative baseline — reported as 2,248 victims between 1988 and 2014 — should be confirmed against the JEP's Auto 033 of 2021 and Fiscalía reporting]. On 18 February 2021 the Jurisdicción Especial para la Paz (JEP), in its Case 03 (Asesinatos y desapariciones forzadas presentados como bajas en combate por agentes del Estado), found that at least 6,402 persons were killed and illegitimately presented as combat deaths between 2002 and 2008 — roughly triple the prosecutorial count, and concentrated such that 78 per cent of the historical universe of such killings fell within those six years [TBD-VERIFY: the precise terms of the JEP finding in Auto 033/Auto de priorización of February 2021, including the 78 per cent formulation and the 2002–2008 window]. Higher academic estimates exist (the criminologist and former police lieutenant colonel Omar Rojas Bolaños has argued for a universe approaching 10,000 [TBD-VERIFY]); lower ones are urged by the institutional defence. No serious account any longer disputes that the victims number in the thousands.
The practice was broken open publicly by the Soacha scandal of August–September 2008: at least 19 young men from Soacha, the poor municipality on Bogotá's southern edge, and from southern Bogotá itself [TBD-VERIFY: the canonical Soacha victim count — figures of 16, 19, and 22 appear across judicial and press records depending on the cases aggregated] disappeared after accepting promises of work, and were reported within days as members of illegal armed groups killed in combat by units including Mobile Brigade 15 in Ocaña, Norte de Santander — roughly 650 kilometres away. Among them was Fair Leonardo Porras Bernal, a 26-year-old with a cognitive disability who could not have been a combatant under any description. The geographic absurdity — unemployed young men "killed in combat" across the country within 48 hours of leaving home — made the fabrication undeniable, and the victims' mothers, organised as MAFAPO (Madres de los Falsos Positivos de Soacha y Bogotá), became the case's permanent moral protagonists.
The institutional consequences of Soacha were immediate and remain the hinge of the contested record. On 29 October 2008, after an internal inquiry led by Inspector General of the Armed Forces General Carlos Arturo Suárez, President Álvaro Uribe dismissed 27 army officers and soldiers, including three generals [TBD-VERIFY: the canonical figure of 27 dismissed including three generals and eleven other officers should be confirmed against the 29 October 2008 announcement record]; on 4 November 2008 General Mario Montoya Uribe, commander of the National Army and the most prominent advocate of the body-count standard of operational success, resigned. The government simultaneously framed the dismissals as proof of self-correction and denied any policy dimension — the position it has held, in essentials, for eighteen years.
The incentive architecture is the analytical core of the case. The Ministry of Defence's Directive 029 of 17 November 2005 [TBD-VERIFY: directive number, date, and content — widely reported as a secret directive signed by Defence Minister Camilo Ospina establishing criteria for reward payments for information leading to the death or capture of members of illegal armed groups; its text became public in 2008–2009], together with the informal economy of rewards it sat atop — extra leave passes, medals, promotions, training-course slots, cash payments to informants and recruiters, and command pressure to produce litros de sangre ("litres of blood", a phrase attested in soldier testimony [TBD-VERIFY: attestation context]) — created what economists studying the case have called high-powered incentives on an unverifiable metric. Whether that architecture amounts to a criminal policy (Account 1), a tragically gameable management system exploited by criminal subordinates (Account 2), or the object whose truth could only be fully extracted by conditional-sanction justice (Account 3) is the precise line of division among the three accounts.
This document applies the J-block three-accounts discipline: (1) a systematic-policy / state-crime account, grounded in the JEP's Case 03 findings, the documentation of MAFAPO and the human-rights movement, the UN Special Rapporteur Philip Alston's 2009–2010 findings, Human Rights Watch's 2015 command-responsibility report On Their Watch, and the quantitative body-count-incentive scholarship; (2) a rotten-apples / institutional-defence account, grounded in the military institution's position across four governments, the Uribe-aligned political defence, the contestation of the JEP's figures and of the "systematic" characterisation, and the army's post-2008 reform narrative; and (3) a transitional-justice / restorative account, grounded in the JEP's own design and practice — the acknowledgment track, the Ocaña and Dabeiba hearings at which perpetrators confessed the practice's mechanics before the victims' families, and the restorative-sanction model whose hardest test this case is. No account is adjudicated; each is presented at its strongest, with its weakest points stated.
The killings interlock with the corpus's account of the Uribe-era security turnaround (CO-A-01, CO-A-02) in a way no honest treatment can avoid: the statistical core of the Democratic Security success narrative — rising combat kills, falling guerrilla strength, recovered territorial control — was partly composed, in a proportion that remains disputed, of murdered civilians. The Alston report's much-quoted finding was that the killings were committed "in a more or less systematic fashion by significant elements within the military" while stopping short of finding them official state policy [TBD-VERIFY: verbatim formulation from the 2010 report of the Special Rapporteur on extrajudicial, summary or arbitrary executions, mission to Colombia]. The systematic-policy account treats that line as diplomatic understatement; the institutional defence treats it as vindication that no policy existed; the restorative account treats the question as exactly what the JEP's macro-case was built to answer.
The accountability arc divides into two regimes. Ordinary justice (2008–2018) produced convictions numbering in the high hundreds to above a thousand, overwhelmingly of soldiers and junior officers [TBD-VERIFY: consolidated conviction counts — figures above 800 convicted by 2015 and above 1,700 individuals convicted across the docket appear in Fiscalía and press reporting — against very few colonels and no generals], at sentences of up to 50+ years in the Soacha cases, while command-level cases stalled. The JEP's Case 03 (2018–) inverted the structure: it prioritised regional sub-cases (Catatumbo, Costa Caribe, Antioquia, Casanare, Huila, Meta), extracted confessions from hundreds of military comparecientes in exchange for eligibility for non-custodial restorative sanctions, indicted colonels and at least one general (Brigadier General Paulino Coronado, in the Catatumbo sub-case, July 2021 [TBD-VERIFY: the Catatumbo indictment's roster]), staged the Ocaña (April 2022) and Dabeiba (June 2022) public acknowledgment hearings at which officers and soldiers confessed the system's workings to the mothers of their victims, and in September 2025 issued its first sentence in the case — restorative sanctions of up to eight years, without prison, against ex-members of the Catatumbo units [TBD-VERIFY: the September 2025 first-instance sentence in the Catatumbo sub-case — number sanctioned, sanction content, and appeal status as of mid-2026].
The command-responsibility frontier remains open at the top in both directions. General Montoya appeared before the JEP from 2018 but declined to acknowledge responsibility, pushing his case toward the adversarial track, where conviction after trial carries ordinary prison of up to 20 years [TBD-VERIFY: the procedural status of Montoya's JEP case as of 2026]. The Uribe-knowledge question — what the President knew of the killings his results-pressure system was producing, and when — has never been judicially established: Uribe denies knowledge and policy, points to the 2008 dismissals as his own corrective act, and is shielded by the absence (so far) of documentary evidence of presidential knowledge; critics point to early warnings he is documented to have received (including from his own security adviser and from international officials [TBD-VERIFY: the specific 2006–2008 warnings record]) and to his contemporaneous public dismissals of the allegations as guerrilla-orchestrated lawfare. His separate criminal conviction in 2025 on witness-tampering charges in a paramilitarism case (CO-H-PRES-02) is legally unrelated but politically inseparable from this question.
The case performs heavy political work in the present. The "¿Quién dio la orden?" ("Who gave the order?") mural — listing generals and victim counts, painted over by order after military legal action in 2019 and subsequently protected as legitimate expression by Colombian courts [TBD-VERIFY: the litigation arc of the mural, including the Constitutional/ordinary-court rulings of 2021–2023] — turned the command question into national iconography. In the 2026 electoral cycle (CO-D-09, CO-O-01), the security-nostalgia current that seeks restoration of Uribe-era results confronts, and is confronted by, the now-judicially-established fact that part of those results was fabricated from the bodies of civilians; how each candidate handles the 6,402 is a live test of the country's settlement with its own security state.

2. The Factual Spine: The Practice, the Scandal, the Numbers (2002–2026)

2.1 Antecedents: The Body-Count Metric Before Uribe

The presentation of murdered civilians as combat dead did not begin in 2002. The Colombian armed forces' internal accounting had used combat kills (bajas) as a core performance measure since at least the counterinsurgency expansions of the 1980s and 1990s, and isolated falsos positivos-pattern cases are documented across that period — the Fiscalía's eventual docket reached back to 1988, and the Comisión de la Verdad's final report (2022) treated the phenomenon across 1978–2016, estimating a total universe of approximately 8,208 victims over that span [TBD-VERIFY: the CEV's consolidated estimate and period]. What changed after 2002 was scale, and what produced the change in scale was the conjunction of three things: a presidency that staked its legitimacy on measurable security results; a military command culture, personified by General Mario Montoya, that translated that demand into explicit kill-count expectations down the chain; and a reward architecture that attached concrete, individual benefits to each reported baja.

2.2 The Democratic Security Pressure System (2002–2008)

President Álvaro Uribe took office in August 2002 on the promise of breaking the FARC militarily (CO-A-02). The Democratic Security policy (Seguridad Democrática) made operational results the currency of the entire security establishment: results were reviewed in regular command councils, units were ranked by them, and the President himself was famous for telephoning commanders to demand them. Within that system, bajas en combate held a privileged place over captures and demobilisations as the most legible, most televisable proof of progress. Soldier and officer testimony — eventually given en masse before the JEP — describes brigade and battalion commanders setting kill quotas, threatening transfers and career stagnation for units that failed to "produce", and celebrating high-count units with leave passes (typically five days per baja [TBD-VERIFY: the standard leave increment attested in testimony]), medals, training-course assignments, and public recognition.

The formal apex of the incentive structure was the Ministry of Defence's Directive 029 of 17 November 2005, signed by Minister Camilo Ospina, a classified instrument that systematised reward payments — to informants and cooperating civilians, scaled by the rank and importance of the target — for information leading to the capture or killing of members of illegal armed groups [TBD-VERIFY: the directive's exact mechanism; it is consistently described as paying third parties for results rather than paying soldiers bounties directly, a distinction both Accounts 1 and 2 exploit]. Around the directive grew an informal market: civilian recruiters (reclutadores) who delivered victims to army units for payment; unit slush funds drawn from operational expenses to buy weapons for planting; and standing relationships with demobilised paramilitaries and criminal bands who supplied victims, weapons, or both. The JEP's Catatumbo investigation documented payments to recruiters of roughly COP 200,000–1,000,000 per victim [TBD-VERIFY: payment ranges attested in the Catatumbo sub-case].

The victims were selected for deniability: unemployed young men, casual labourers, the homeless, drug users, people with disabilities, petty offenders, rural men in contested zones — people whose disappearance, the perpetrators calculated, would be slow to be noticed and easy to discredit. The method's forensic signature became standard across regions, a uniformity the JEP later treated as evidence of pattern: victims killed within hours or days of recruitment; bodies dressed in new camouflage or guerrilla-style clothing (sometimes in boots of the wrong size, or clean clothes over fatal wounds — details that became emblematic at trial); weapons planted, frequently old revolvers or unserviceable rifles inconsistent with the reported combat; the scene reported as a combat encounter (combate) with no army casualties; the body registered as an unidentified combatant and buried quickly, often in N.N. graves far from the victim's home.

2.3 Early Warnings (2004–2008)

The practice was reported long before Soacha. Regional human-rights organisations, the Coordinación Colombia–Europa–Estados Unidos (CCEEU) network, and the UN Office of the High Commissioner for Human Rights in Colombia documented rising extrajudicial-execution complaints from 2004 onward; the CCEEU's case database became one of the eventual quantitative bases for the JEP's universe. Internal warnings also existed: the Procuraduría opened disciplinary files; in 2006 the Ministry of Defence received documented alerts [TBD-VERIFY: the specific 2006–2007 internal warnings, including reporting that Defence Minister Juan Manuel Santos and his predecessors received case files from the UN office and from internal inspectors], and in 2007 the Ministry issued Directive 300-28, which formally re-weighted operational evaluation toward demobilisations and captures over kills — a document both sides of the argument cite: Account 2 as proof of self-correction before the scandal, Account 1 as proof that the leadership knew precisely what the kill metric was producing while the killing was at its height. The years 2006–2007 were, by the JEP's count, the practice's statistical peak [TBD-VERIFY: the peak years within the 2002–2008 window per Auto 033].

The government's contemporaneous public posture was denial and counter-attack. President Uribe repeatedly characterised the accusations as a juridical strategy of the guerrillas and their sympathisers to discredit the army; the most-cited formulation is his remark that the young men of Soacha had not been killed while "picking coffee" — that they died in criminal activity [TBD-VERIFY: the verbatim of Uribe's October 2008 statement, widely rendered as the victims having gone "con propósitos delincuenciales" and "no propiamente a recoger café", which the mothers of Soacha treated as a second injury and which Uribe has defended as based on the initial military reporting given him].

2.4 Soacha, the Scandal, and the Purge (2008)

Between January and August 2008, young men began disappearing from Soacha and southern Bogotá after accepting offers of work elsewhere in the country, made by recruiters working a pipeline to army units in Norte de Santander. Their bodies surfaced in Ocaña and surrounding municipalities — registered as members of illegal armed groups killed in combat by Mobile Brigade 15 and the Francisco de Paula Santander Battalion, in some cases within a day or two of leaving home. In September 2008 the matching of missing-person reports against N.N. combat-death registries — driven by the families' insistence and by officials of the Medicina Legal forensic institute and the Personería — broke the story into national media. The geographic and temporal impossibility was the fabrication's undoing, and the profile of victims like Fair Leonardo Porras destroyed the "criminals killed in crime" counter-narrative.

The institutional response came in two acts. On 29 October 2008, following General Carlos Arturo Suárez's rapid internal inquiry, President Uribe announced the dismissal of 27 members of the army — including three generals and a roster of colonels and majors spanning the chain of command relevant to the Soacha-Ocaña pipeline [TBD-VERIFY: the full roster and ranks of the 29 October 2008 dismissals] — for failures of oversight, while insisting the dismissals reflected administrative responsibility, not criminal participation. On 4 November 2008, General Mario Montoya resigned as army commander; he was succeeded by General Óscar González Peña [TBD-VERIFY: successor], and the government reiterated that the institution had been betrayed by criminal conduct within its ranks, not directed by it. In the following months the Ministry of Defence issued a package of reforms — fifteen measures announced in November 2008, the strengthening of operational legal advisers, delegate inspectors in divisions, revised rules of engagement certification, and Directive 208 of 2008 on the reception of demobilised persons [TBD-VERIFY: the precise instruments in the 2008–2011 reform package] — that became the documentary spine of the institutional-defence account.

Reported extrajudicial executions of this type collapsed after 2008 — a fall of more than 90 per cent in the annual complaint rate [TBD-VERIFY: post-2008 incidence statistics], which every account acknowledges and each explains differently: the incentive system was dismantled (Account 1: proving the incentives caused the killings); the institution corrected itself once criminality was exposed (Account 2); accountability pressure changed perpetrator calculus (Account 3).

2.5 Alston, the Prosecutions, and the Long Middle (2009–2018)

The UN Special Rapporteur on extrajudicial executions, Philip Alston, conducted a mission to Colombia in June 2009 and reported in 2010. His findings became the international baseline: the killings were widespread, committed across many units and departments, could not be dismissed as isolated cases, and were carried out "in a more or less systematic fashion by significant elements within the military"; but he found no evidence that they were committed as a matter of official state policy ordered from the top [TBD-VERIFY: verbatim]. He identified the incentive and pressure system, impunity, and the fuero militar (military jurisdiction) as enabling structures, and warned against the transfer of cases to military courts.

Ordinary prosecutions accumulated through the 2010s — the Fiscalía's human-rights unit charged thousands of soldiers, and convictions mounted into the hundreds (the Soacha cases produced sentences exceeding 50 years for soldiers and the recruiters who fed them) — but the docket's structure was bottom-heavy: overwhelmingly soldiers, sergeants, lieutenants, and captains; a handful of lieutenant colonels (the conviction of the Mobile Brigade 15 commander's subordinates among them [TBD-VERIFY: the highest-ranking pre-JEP convictions, including the cases of colonels of the Santander battalion structure]); and no general convicted by the ordinary courts during the entire pre-JEP period. Human Rights Watch's June 2015 report On Their Watch: Evidence of Senior Army Officers' Responsibility for False Positive Killings in Colombia assembled prosecutorial evidence that the commanders of units responsible for hundreds of killings — including officers later promoted to lead the army itself — knew or should have known, and pressed the command-responsibility question that ordinary justice was not reaching. A 2019 controversy under the Duque government (CO-C-01) — the New York Times report that new army orders under commander General Nicacio Martínez again demanded doubled operational results and lowered certainty thresholds — demonstrated the metric's gravitational pull and re-ignited the fear of relapse [TBD-VERIFY: the May 2019 NYT report's specifics and the army's response].

2.6 The JEP's Case 03 and the 6,402 (2018–2026)

The Jurisdicción Especial para la Paz, the transitional tribunal created by the 2016 peace agreement (CO-B-03, CO-I-01 context), opened Case 03 in July 2018 as one of its initial macro-cases. Drawing together the Fiscalía's files, the CCEEU and CINEP databases, Medicina Legal records, and its own compareciente testimony, the JEP's Chamber for the Recognition of Truth issued its universe finding on 18 February 2021: at least 6,402 victims between 2002 and 2008. The number — instantly iconic, painted on walls and worn on shirts — tripled the prosecutorial baseline and relocated the phenomenon from "grave pattern" to one of the largest documented campaigns of extrajudicial killing by a state's regular army in the hemisphere's modern history.

Case 03 proceeded by regional prioritisation: sub-cases on the Catatumbo (Norte de Santander), the Costa Caribe, Antioquia (including the Dabeiba killings), Casanare, Huila, and Meta [TBD-VERIFY: the final sub-case list and any later national phase]. In July 2021 the Chamber indicted (imputó) ten ex-members of the military and one civilian in the Catatumbo sub-case for war crimes and crimes against humanity — among them Brigadier General Paulino Coronado, commander of the 30th Brigade, the highest-ranking officer charged to that point [TBD-VERIFY: the Catatumbo imputation roster, including the battalion commanders Álvaro Diego Tamayo and Santiago Herrera]; further indictments followed in the Costa Caribe and other sub-cases, reaching additional generals [TBD-VERIFY: the Costa Caribe imputation, reported to include General (r) Publio Hernán Mejía and other senior officers of the La Popa battalion structure].

The case's distinctive instruments were the acknowledgment hearings: at Ocaña (26–27 April 2022) and Dabeiba (June 2022), indicted officers and soldiers publicly acknowledged responsibility before victims' families — describing quotas, recruiters, the planting of weapons, the "legalisation" paperwork, and the command climate — in scenes without precedent in Colombian institutional life: a colonel asking the mothers of Soacha and Ocaña for forgiveness; a sergeant explaining the price paid per victim. In September 2025 the JEP's first-instance section issued the case's first sentence, imposing on ex-members of the Catatumbo units the agreement's maximum restorative sanction — up to eight years of effective restriction of liberty without prison, attached to reparative works (search for the disappeared, memorialisation, infrastructure) [TBD-VERIFY: the September 2025 sentence — the number sanctioned (reported as twelve), the sanction architecture (TOARs), victim-party appeals, and the parallel first sentence against the FARC secretariat in Case 01]. As of mid-2026, further sub-case sentences, the adversarial-track cases of non-acknowledging officers (Montoya foremost), and the question of whether the JEP will open a "maximum responsibility" national phase reaching above brigade level remain pending [TBD-VERIFY: Case 03 status as of 2026].

3. Account One: The Systematic-Policy / State-Crime Account

3.1 The Core Claim

The systematic-policy account holds that the falsos positivos were not an accumulation of individual crimes but the predictable, structural output of a system: a results regime that demanded bodies, rewarded bodies, did not verify bodies, and punished their absence. On this account, the 6,402 were killed by an incentive architecture as surely as by the soldiers who pulled triggers; criminal responsibility therefore runs upward through every commander who set quotas, certified fabricated reports, and was promoted on their totals — and the political responsibility runs to the government that built the pressure system, was warned repeatedly what it was producing, and protected it with denial and counter-attack until the fabrications became physically undeniable. The account's strongest formulation: the Democratic Security success narrative (CO-A-01, CO-A-02) was partly manufactured from murdered civilians, and the Colombian state has yet to fully name that fact at the level — presidential — where the demand for results originated.

3.2 The Evidentiary Foundation

The JEP's pattern findings. Case 03's significance for this account is not only the number but the characterisation: the Chamber found the killings to constitute crimes against humanity — a generalised and systematic attack against the civilian population [TBD-VERIFY: the JEP's precise legal characterisation in the 2021 imputations] — and its regional autos reconstructed, unit by unit, an identical machinery: quotas set in command programmes; recruiters on retainer; weapons caches for planting; paperwork chains in which battalion and brigade staffs "legalised" kills they had every means to know were fabricated; and statistical patterns (combat reports with zero army casualties, victims shot at close range, bajas spiking before evaluation periods) inconsistent with combat and consistent with execution. Uniformity across distant regions, this account argues, is the signature of system, not coincidence: the same method in Catatumbo, the Caribbean coast, Antioquia, and Casanare implies transmission through the institution's ordinary channels — training, evaluation, command example — not the spontaneous parallel invention of hundreds of criminal cells.

The incentive paper trail. Directive 029 of 2005 exists; the leave-pass and reward economy is attested by hundreds of confessions; Directive 300-28 of 2007 proves the leadership understood the kill metric's perversity in real time. The quantitative scholarship closed the causal loop: the Fellowship of Reconciliation (FOR) and CCEEU's 2010 study found units receiving certain US assistance and high results pressure associated with elevated execution reports [TBD-VERIFY: the FOR/CCEEU study's design and findings]; and the econometric work of Acemoglu, Fergusson, Robinson, Romero and Vargas — circulated from 2016, published as "The Perils of High-Powered Incentives: Evidence from Colombia's False Positives" — found that units commanded by colonels facing stronger promotion incentives, and brigades under higher kill-rate expectations, produced significantly more false positives, with killings of civilians substituting for combat kills where guerrilla presence was thin [TBD-VERIFY: the paper's precise findings and publication venue]. For this account, the studies establish what the confessions illustrate: the metric did not merely fail to prevent the killings; it generated them.

The victims' documentation. MAFAPO and the broader victims' movement preserved the case through the years of official denial: the mothers of Soacha matched the disappearances to the combat registries, forced the exhumations, testified in every forum, absorbed stigmatisation (including the presidential "no estarían recogiendo café" framing) and threats [TBD-VERIFY: the attested reprisals against Soacha families — including the reported killing of at least one victim's relative who was pressing the cases, widely cited in the MAFAPO record — should be verified and named precisely], and sat opposite the confessing soldiers at Ocaña. The CCEEU/CINEP case databases — built complaint by complaint across the 2000s — supplied the universe against which the Fiscalía's count was shown to be a floor, and were vindicated when the JEP's triangulation landed at 6,402.

International findings. Alston's 2010 report ("more or less systematic"; the incentive system; the impunity structures); the sustained reporting of the UN High Commissioner's Colombia office from 2005 onward; the Inter-American system's case law on aspects of the pattern [TBD-VERIFY: relevant IACtHR/IACHR decisions, including Villamizar Durán et al. v. Colombia (2018) on earlier execution cases]; and HRW's On Their Watch (2015), which used prosecutors' own files to argue that the generals commanding the highest-kill units — including Montoya, and officers subsequently promoted to army command under Santos and Duque — were at minimum wilfully blind, and that the promotion of officers whose units' results were saturated with fabrications is itself evidence of what the institution valued and verified.

3.3 The Command Question: Montoya and the Presidency

The account's apex claims concern the top of the chain. On Montoya: as army commander (2006–2008) he personally embodied the body-count standard — soldier testimony before the JEP describes his command reviews demanding kills, dismissing captures and demobilisations as inferior results, and his slogan-level insistence on bajas [TBD-VERIFY: the attested Montoya formulations in JEP testimony, including demands rendered as wanting "ríos de sangre" or "litros de sangre" attributed in some testimony — attribution contested]; the killings peaked under his command and collapsed after his fall; and his refusal to acknowledge before the JEP, this account holds, marks the boundary where the institution's truth-telling stopped. On Uribe: the account does not claim documentary proof of a presidential order — none has surfaced — but argues command responsibility in its political form: the President built and personally operated the results-pressure system; warnings reached his government from 2005–2006 onward through the UN office, the Procuraduría, and internal channels; his public posture until October 2008 was to attack the accusers; and the constitutional doctrine that the President is commander-in-chief makes the gap between "knew" and "should have known" the central unresolved question of the era. The Soacha dismissals, on this reading, prove the system's correctability — and therefore the prior choice not to correct it while the count was rising.

3.4 The Account's Vulnerabilities

Stated at its strongest, the account carries three burdens. First, the policy inference is an inference: no order to kill civilians has been produced from any level above battalion; Directive 029 paid for results against combatants, not murders; and Alston himself — the account's own witness — declined to find state policy. The account answers that systematicity in international criminal law does not require a written order, only an organisational policy inferable from pattern — but the inference is the argument, and Account 2 attacks it at exactly that joint. Second, the universe's edges are soft: the 6,402 is a triangulated minimum built partly from complaint databases whose individual entries have not all been judicially verified; some fraction of contested cases were genuine combat deaths, and the account's iconography (the number on walls and shirts) carries more certainty than the methodology strictly warrants — a gap the institutional defence exploits. Third, the presidential question outruns the evidence: the move from "the system was his" to "the knowledge was his" is supported by plausibility and warning-records, not by established fact, and the account's political carriers have sometimes asserted as proven what remains unestablished — handing the defence a legitimate charge of overreach.

4. Account Two: The Rotten-Apples / Institutional-Defence Account

4.1 The Core Claim

The institutional-defence account holds that the falsos positivos were crimes — grave, inexcusable, and prosecuted — committed by individuals and corrupted unit networks in violation of policy, doctrine, and law, inside an army of more than 200,000 fighting a real war against real armed organisations whose own atrocities form the context the accusers omit. On this account: the Democratic Security policy ordered the defeat of illegal armed groups within the law; its measurable-results regime was a legitimate, internationally ordinary management instrument that criminal elements gamed; the institution detected, dismissed, and reformed when the Soacha evidence crystallised — faster and more drastically than most armies have ever sanctioned themselves; the post-2008 collapse in cases proves the institution's correction worked; and the subsequent juridical-political construction — the 6,402, the "systematic" characterisation, the "state crime" vocabulary — overcounts the universe, retrofits a policy that never existed, and serves a political project of delegitimising the security forces and the presidency that broke the FARC. Its carriers have been continuous across two decades: the military leadership and its retired-officer associations (ACORE foremost), the Uribe-aligned political bloc (the Centro Democrático after 2013), defence counsel before the JEP, and a strand of legal commentary contesting the JEP's methodology.

4.2 The Evidentiary Foundation

The legality of the formal framework. The account's documentary base is the policy record itself: the Democratic Security policy documents order results against combatants; operational law required target verification; Directive 029/2005, on its face, regulated payments to civilian informants for actionable intelligence — a standard counterinsurgency instrument — and nowhere authorises or prices deaths as such [TBD-VERIFY: the directive's text and the defence reading of it]; and the doctrinal sequence of 2007–2008 (Directive 300-28's re-weighting toward demobilisations and captures; the November 2008 fifteen-measure package; the operational-law advisers embedded in units; Directive 208/2008) shows a leadership that identified the metric's distortion and corrected it. If the killings had been policy, the account asks, why did the same leadership re-engineer the incentive system against them — and why did the killings then stop?

The self-correction record. The 29 October 2008 dismissal of 27 officers including three generals was, the account emphasises, an act of the Uribe government itself, on the findings of the army's own inspector general, before any court had convicted anyone — followed by Montoya's resignation, the transfer of cases to ordinary justice rather than military courts in the majority of instances [TBD-VERIFY: the jurisdictional disposition statistics], and the certification machinery (human-rights vetting, operational legal review, the reglas de encuentro reforms) built in 2009–2011. The post-2008 incidence collapse — acknowledged by every account — is the defence's strongest single datum: institutions whose policy is murder do not stop murdering when the policy's authors remain in office.

The contestation of the 6,402. The account attacks the JEP's universe finding on method: the figure triangulates prosecutorial files with NGO complaint databases (CCEEU, CINEP) whose entries are allegations, not adjudications; it counts cases across a window and attribution standard broader than the criminally established docket; individual entries include deaths later shown to be genuine combat results; and the gap between the Fiscalía's ~2,248 and the JEP's 6,402 was produced by aggregation, not by 4,000 new judicial verifications [TBD-VERIFY: the specific published critiques of the 6,402 methodology, including statements by Centro Democrático figures, retired-officer associations, and the JEP's own explanation of its triangulation in response]. The defence does not, in its serious versions, deny thousands of victims; it denies that the ceiling figure has the evidentiary status its political use implies, and it notes that the JEP itself labels the figure a provisional universe for prioritisation, not a verdict.

The war context. The account insists on the conflict frame the state-crime account brackets: the same years saw FARC mass kidnapping, child recruitment, landmine campaigns, and attacks on civilians; soldiers operated under genuine threat against enemies out of uniform; and a results-driven army at war will contain — as all wartime armies have — criminal networks that exploit operational cover. The institution's honour, on this account, is carried by the hundreds of thousands who fought lawfully and the thousands who died doing so; collective criminalisation of the army libels them and, in the 2020s, degrades the morale and operational confidence of a force still fighting (CO-D-04 context: the Total Peace-era security deterioration is read by this account as partly the fruit of the accusatory decade).

4.3 The Political Defence: Uribe and the Results Presidency

The political stratum of the account, carried by Uribe himself across every forum since 2008, holds: the President demanded results within the law and dismissed those who betrayed it; the demobilisation-and-capture re-weighting and the 2008 purge were his government's acts; no warning he received was concealed or ignored once verified [TBD-VERIFY: Uribe's specific public accounting of the warnings record]; and the accusatory construction serves the left's project of converting a military victory into a moral defeat — the explicit argument of the Centro Democrático's opposition to the JEP's design (CO-B-03) and of its 2026-cycle security platform (CO-D-09). In its harder political versions, the account extends to attacking the JEP's legitimacy as such: a tribunal born of a rejected plebiscite, structurally biased toward equating the state with the insurgency, and imposing "impunity for the FARC, infamy for the soldiers."

4.4 The Account's Weakest Points

The three-accounts discipline requires stating where this account strains, and it strains badly at three joints. First, scale. The rotten-apples frame must explain at least several thousand victims, killed by dozens of units across more than twenty departments over six years with a near-identical method — a distribution that looks like an institution, not like apples. Every increment in the verified count (and the JEP's confessions have verified thousands of cases the complaint databases first recorded) transfers weight from "aberration" to "pattern." Second, the paper trail. The incentive instruments existed and were the institution's own; Directive 300-28 (2007) proves leadership knowledge of the metric's perversity at least a year before Soacha while the killing ran at peak; and the promotion record — commanders of high-fabrication units rising, including to army command — is difficult to square with an institution deceived by its criminals rather than rewarded by their output. Third, the confessions. Since Ocaña and Dabeiba, the defence no longer argues against inference alone: hundreds of the institution's own officers and soldiers have described, under oath and to the victims' faces, quotas, recruiter payments, and command climates — testimony the account can discount as sanction-shopping (confess broadly, escape prison) only at the price of impugning the very soldiers it claims to defend. The serious institutional position as of the mid-2020s has therefore migrated: it concedes pattern at unit-and-brigade level while holding the line at "no policy, no presidency" — a narrower fortress than the one defended in 2008, and one whose remaining wall is the genuinely unresolved evidentiary state of the top-of-chain question.

5. Account Three: The Transitional-Justice / Restorative Account

5.1 The Core Claim

The restorative account is not a third position on what happened — on the facts it stands close to Account 1, whose universe finding is the JEP's own — but a distinct account of what justice for it consists of, and it is contested enough, and load-bearing enough for Colombia's institutional settlement, to require separate statement. Its core claim: a decade of ordinary prosecution produced convictions of triggermen and near-total silence about the system; the JEP's conditional model — full truth and acknowledgment in exchange for non-custodial restorative sanctions, with ordinary prison reserved for those who deny and are convicted — extracted, in four years, the systemic truth that adversarial justice could not reach in fifteen: the universe (6,402), the mechanics (quotas, recruiters, legalisation), the regional command structures (the imputations of colonels and a general), and the public moral reckoning of Ocaña and Dabeiba, where perpetrators named the system to its victims. On this account, the falsos positivos case is the strongest existing vindication of the 2016 agreement's justice architecture — and simultaneously its hardest test, because the model's price is that men who confessed participation in systematic murder will not go to prison, and the account holds that trade defensible only if it is held openly, as a trade, never denied or euphemised.

5.2 The Evidentiary Foundation

The comparative yield. The account's empirical base is a comparison of regimes. Ordinary justice, 2008–2018: convictions in the hundreds-to-low-thousands of mostly low-ranking perpetrators, at an average litigation span of many years per case, with no general convicted, no systemic findings, and a victim experience of fragmentary, adversarial, revictimising process [TBD-VERIFY: consolidated ordinary-justice output statistics]. The JEP, 2018–2025: a universe finding; thousands of comparecientes providing versiones voluntarias; regional pattern reconstructions; imputations reaching brigade command; public acknowledgment before victims; and the first sentences — with the search for the disappeared and exhumations (the Dabeiba cemetery exhumations, which recovered remains of victims "legalised" by units operating in Antioquia [TBD-VERIFY: the Dabeiba exhumation findings from 2019–2021]) materially advanced by perpetrator information that no prosecution had ever pried loose. Truth of this kind, the account argues, is not a consolation prize for punishment foregone; for crimes of system, it is the primary justice good, because it is the only remedy that reaches the system rather than its instruments.

The victims-centred design. The account emphasises that the JEP's procedure made victims procedural protagonists: accredited victim representatives interrogate the versiones, contest insufficient acknowledgments (victim observations forced the expansion of several officers' admissions between their written versiones and the public hearings [TBD-VERIFY]), and shape the reparative content of sanctions. The Ocaña hearing's design — perpetrators facing the Soacha and Catatumbo mothers, acknowledgment evaluated for completeness by the Chamber with victim input — is the account's emblematic artefact: justice as the restoration of the victims' standing to be told the truth to their faces.

The perpetrators' own accounts. The confessions at Ocaña and Dabeiba are this account's distinctive evidence about how the practice worked from inside: soldiers describing the arithmetic of leave passes; non-commissioned officers describing the handoff from recruiter to patrol; officers describing what brigade command asked in results meetings and what everyone understood the asking to mean; the recurring testimony that refusal was career death and, in some units, physically dangerous [TBD-VERIFY: attested cases of soldiers threatened or killed for refusing or reporting, including the documented whistle-blower cases]. This stratum — unavailable in principle to adversarial process, because no defendant testifies against himself at trial — is what converted the statistical pattern into narrated mechanism, and it is the restorative account's answer to both other accounts: to Account 2, that the institution's own members have described the system; to Account 1's harder strand, that the truth now possessed arrived through the door that prison-first justice would have kept shut.

The scholarship. The account draws on the transitional-justice literature on Colombia — the design debates around the 2016 agreement's sanciones propias (restorative sanctions for acknowledgers), the comparative literature on truth-versus-punishment trades, and the constitutional jurisprudence (the Corte Constitucional's review of the JEP framework, CO-I-01) holding that punishment may be conditioned and reduced, but not victims' rights to truth, reparation, and non-repetition [TBD-VERIFY: the relevant Constitutional Court decisions on the Legislative Act 01/2017 framework].

5.3 The Internal Debate: The Impunity Question Held Honestly

The account's integrity depends on stating its own hardest problem without euphemism, and its serious carriers do. The September 2025 Catatumbo sentence imposed a maximum of eight years of non-prison restriction on men who acknowledged participation in scores of murders; victim organisations divided publicly — some MAFAPO members and Catatumbo families accepting the trade as the price of the truth they had received, others, joined by part of the human-rights bar, contesting the sanctions as inadequate to crimes against humanity and appealing aspects of the decision [TBD-VERIFY: the victim-party responses and appeal record of the 2025 sentence]; the Inter-American system and the ICC's Office of the Prosecutor (which closed its Colombia preliminary examination in 2021 on the strength of the JEP's functioning [TBD-VERIFY: the October 2021 ICC–Colombia cooperation agreement and its conditions]) hover as external auditors of whether "effective sanction" is being delivered. Inside the account, the debate runs between a truth-primacy position (the trade is just because the alternative was neither truth nor punishment, but silence with sporadic punishment of subordinates) and a proportionality-anxiety position (the model survives only if sanctions are demonstrably onerous, verified, and reparative — if restricción efectiva is visibly more than a signature regime). Both positions share the account's constitutive commitment: the trade must be named as a trade. The moment the model's defenders deny that anything has been foregone, the account collapses into apologetics; the moment its critics deny that anything has been gained, they collapse into Account 1's hardest strand or Account 2's JEP-delegitimation, and the distinctive third position disappears.

5.4 The Account's Vulnerabilities

Beyond the impunity question it holds internally, the account carries three external vulnerabilities. First, the acknowledgment boundary: the model's yield stops where confession stops — Montoya and the other non-acknowledgers have conceded nothing, their adversarial cases proceed slowly, and if the JEP cannot convict deniers with reasonable speed, the system's incentive logic (confess or face real trial) hollows out; the model is only as credible as its coercive branch [TBD-VERIFY: the state of the adversarial-track (Unidad de Investigación y Acusación) prosecutions as of 2026]. Second, the ceiling: Case 03's imputations reach brigade command; if the macro-case closes without ever adjudicating the division, army-command, and ministerial levels, the restorative model will have traded punishment for a truth that itself stopped at the rank where Account 1 and Account 2 actually disagree. Third, compliance: restorative sanctions are only sanctions if executed and verified over years; the TOAR-execution and monitoring record (works completed, restriction enforced) is thin as of the first sentences, and a compliance failure would retroactively convert the trade into the impunity its critics allege.

6. The Accountability Arc (2008–2026)

6.1 Ordinary Justice: The Bottom-Heavy Decade

The criminal docket opened in earnest only after Soacha. The Fiscalía's National Human Rights Unit concentrated the cases, fighting a sustained jurisdictional battle against transfers to military criminal courts — a battle Alston had flagged as decisive, and which the Constitutional Court's jurisprudence on the fuero militar largely resolved in favour of ordinary jurisdiction for execution allegations [TBD-VERIFY: the jurisprudential line, including the fate of the 2012–2013 fuero-expansion reforms, which the Court struck down in 2013]. The output by the eve of the JEP: investigations touching several thousand members of the armed forces; convictions reported variously above 800 individuals by the mid-2010s and above 1,300–1,800 cumulatively by the late 2010s [TBD-VERIFY: the consolidated conviction counts — published figures differ by counting unit (persons vs. cases) and date, and a verified consolidated number should replace this range]; emblematic Soacha verdicts including the 2011–2013 convictions of soldiers and officers of Mobile Brigade 15 and the Santander battalion structure at sentences exceeding 50 years, and convictions of the civilian recruiters who delivered the victims [TBD-VERIFY: the named Soacha verdicts, including the conviction of Colonel/Major-level defendants and the recruiter Alexander Carretero's cooperation record]. The structural fact about this regime, asserted by Accounts 1 and 3 and not seriously contested by Account 2: no general was convicted by the ordinary courts in the entire pre-JEP period, and command-level cases — including the preliminary investigation the Fiscalía opened against Montoya [TBD-VERIFY: the ordinary-justice Montoya file's status before his JEP submission] — did not reach trial.

6.2 The JEP Machinery

Case 03's procedural architecture became the template for the JEP's macro-case method (CO-B-03 carries the institutional design; this section carries only the case). The sequence: (1) universe construction — the triangulation that produced the 6,402 (Auto 033, 18 February 2021 [TBD-VERIFY: the auto number]); (2) regional prioritisation — six territorial sub-cases selected for concentration of victims and evidence; (3) versiones voluntarias — closed testimonial sessions in which comparecientes (over 3,500 members and ex-members of the forces submitted to the JEP across cases [TBD-VERIFY: compareciente totals for Case 03 specifically]) traded candour for eligibility; (4) imputation — the determination-of-facts autos charging "most responsible" participants with war crimes and crimes against humanity: Catatumbo (July 2021: ten military comparecientes including Brigadier General Paulino Coronado, battalion commanders Álvaro Diego Tamayo and Santiago Herrera Fajardo, and a civilian recruiter [TBD-VERIFY: roster]); Costa Caribe (2023: officers of the Batallón La Popa structure including its former commander [TBD-VERIFY: the La Popa imputation, including Colonel Publio Hernán Mejía, who declined acknowledgment]); Casanare, Antioquia, Huila and Meta autos following [TBD-VERIFY: each sub-case's imputation status as of 2026]; (5) acknowledgment hearings — Ocaña (April 2022) and Dabeiba (June 2022), with further regional hearings after; (6) sentencing — the September 2025 Catatumbo first sentence (restorative sanctions, maximum eight years' effective restriction, attached to search, memorialisation, and reparative works), with the remaining sub-cases' resolutions and the non-acknowledgers' adversarial referrals pending as of mid-2026 [TBD-VERIFY: the full 2025–2026 sentencing and referral record].

6.3 The Command-Responsibility Frontier

Three names mark the frontier's current line. Coronado — the one general imputed in the first wave — acknowledged responsibility in qualified form [TBD-VERIFY: the Chamber's evaluation of Coronado's acknowledgment, which victims contested as incomplete, and his disposition in the 2025 sentence]. Montoya — the era's commanding symbol — submitted to the JEP in 2018, has denied all responsibility across his versiones, was the subject of victim-driven demands for his expulsion to ordinary justice, and stands referred toward the adversarial track, where the Unidad de Investigación y Acusación must build a triable case and where conviction carries up to 20 years' ordinary imprisonment [TBD-VERIFY: Montoya's precise procedural posture as of 2026 — whether formally referred, charged, or still in recognition-phase determination]. Uribe — never under JEP jurisdiction (the JEP does not reach ex-presidents [TBD-VERIFY: the jurisdictional carve-out — sitting/former presidents are subject to the Comisión de Acusaciones of the Cámara, the chronically inert "absolution commission"]) — faces no formal proceeding on the falsos positivos; the knowledge question persists as a historical-political contest, periodically re-ignited (testimony from ex-officers asserting reports reached the Casa de Nariño [TBD-VERIFY: the specific ex-officer assertions, including statements attributed to former army officers before the JEP]; the Comisión de la Verdad's findings on executive responsibility [TBD-VERIFY: the CEV final report's formulation on the presidency]), and structurally untestable in any existing forum — a fact all three accounts register, with opposite affect.

6.4 The Victims' Long Campaign

The accountability arc's motor has been the victims' movement, and the corpus records its instruments. MAFAPO — the mothers (and sisters, wives, daughters) of the Soacha and Bogotá victims — converted private searching into public method: matching disappearances to combat registries, attending every exhumation, sitting in every courtroom, and building an iconography that did the political work the legal system was too slow to do: the photographs of the sons worn on the body; the embroidered shrouds and the costurero (sewing circle) memory practices; the annual commemorations at the Soacha graves; the insistence on the phrase ejecuciones extrajudiciales — and on "¿quién dio la orden?" as the unanswered question. The mural of that name — first painted in Bogotá in October 2019 with generals' faces and a victim count, painted over within hours after military legal pressure, and re-produced in the thousands digitally and on walls across the country precisely because of the suppression attempt; litigation ended with Colombian courts protecting the image as legitimate expression and the campaign (Campaña por la Verdad) updating its count upward with the JEP's findings [TBD-VERIFY: the mural litigation's final disposition and the updated count used after 2021] — is the case's Streisand-effect monument. The 6,402 itself became the movement's number: rendered in crosses on hillsides, in installation art (the 6,402 figures placed in Bogotá's Plaza de Bolívar [TBD-VERIFY: the 2021–2024 memorial installations]), and in the JEP-ordered reparative works that the sentenced comparecientes must now build. The mothers' trajectory — from stigmatised complainants ("no estarían recogiendo café") to the moral centre of nationally televised acknowledgment hearings — is, on any account, the arc's one unambiguous vindication.

7. The Three Accounts in Dialogue

7.1 Convergences

The argument has narrowed since 2008, and the narrowing should be recorded. All three accounts now hold: that civilians numbering at least in the thousands were murdered by members of the army and presented as combat deaths; that the practice was concentrated in 2002–2008 and collapsed after the Soacha exposures and the incentive-system reforms; that the victims were disproportionately poor, marginal, and deliberately selected for deniability; that the recruiters, the planted weapons, and the legalisation paperwork operated as described in the confessions; that the ordinary-justice response was real but bottom-heavy; and that the mothers of Soacha were right and were wronged — first by the killings, then by the stigmatisation. No serious carrier of Account 2 any longer defends the 2008-era position that the phenomenon was marginal or invented. That the floor of consensus now sits at "thousands of civilians murdered by the army under an incentive system the state built" is itself the measure of how far the victims' campaign and the JEP's machinery moved the record.

7.2 Divergences

Three questions structure the remaining disagreement. First, system versus aberration — the policy question. Account 1 reads uniformity-at-scale plus the incentive paper trail as organisational policy in the international-criminal sense; Account 2 concedes pattern at unit level while denying policy at institutional level; the JEP's imputations have so far adjudicated "systematic" within regions and brigades, leaving the institutional apex unadjudicated [TBD-VERIFY: the precise scope of the JEP's systematicity findings]. The question is partly evidentiary (what the unheard versiones and the adversarial cases will yet produce) and partly conceptual (what "policy" requires when no order exists but the reward structure does) — which is why it does not close. Second, the number. 2,248 / 6,402 / ~10,000 are not three estimates of one quantity but outputs of three methods (adjudicated cases; triangulated universe; extrapolation), and each account deploys the figure its theory needs. The corpus convention: cite the JEP's 6,402 as the institutional finding, always with its date and character (a minimum universe for prioritisation), and tag higher and lower figures to their methods. Third, the justice model. Account 1's harder strand and Account 2 converge, from opposite directions, in attacking the JEP — as too lenient toward confessed murderers and as institutionally illegitimate, respectively — while Account 3 defends the trade as the price of the truth both critics now freely cite. The 6,402 itself is JEP-produced: every account that brandishes the number is standing on the model whose legitimacy is in dispute, a dependency Account 3 points out with some relish.

7.3 The Political Work Each Account Performs

The accounts are not free-floating interpretations; each underwrites a present-day position, and corpus users should read them with the deployment visible. Account 1 underwrites the left and human-rights coalition's case against the restoration of uribismo: it makes the Democratic Security era's achievements morally unusable, supplies the Petro-era political language of the Estado criminal [TBD-VERIFY: Petro's specific deployments of the 6,402 in presidential rhetoric], and powers the demand that the 2026 cycle not return the security state's architects. Account 2 underwrites the security-nostalgia platform visible across the 2026 pre-campaign (CO-D-09, CO-O-01): if the killings were betrayals of a sound policy rather than its product, then the policy — order, results, military confidence — can be promised again with corrections rather than repentance; the account's electoral function is precisely to detach "security results" from "false positives" so the former can be campaigned on. The candidates of the right navigate between the account's serious version (crimes occurred; the institution corrected; honour the lawful soldiers) and its vulgar version (the JEP is the enemy; the figures are inflated; the real victims are the soldiers) — and which version prevails in 2026 will matter for whether the JEP's sentences are executed or besieged. Account 3 underwrites the peace-architecture coalition's case that the 2016 settlement (CO-B-02, CO-B-03, CO-K-02) is working and must be protected across alternations: the falsos positivos confessions are Exhibit A whenever the JEP's budget, jurisdiction, or legitimacy is attacked. Each account, note, has an interest in the case remaining open: Account 1 needs the apex unresolved to sustain the demand for more; Account 2 needs it unresolved to sustain the denial; Account 3 needs the process running to sustain the model. The victims' organisations, alone among the carriers, have an unmixed interest in closure — names, remains, sentences executed — and their periodic friction with all three accounts' political users is part of the record.

7.4 Reading Discipline for Corpus Users

Weight the strata by their production conditions: the 2002–2008 official operational record is systematically corrupted at the source (that is what the case is); the 2008–2015 prosecutorial stratum is reliable per-case but unrepresentative of the universe and silent above battalion; the NGO complaint databases are comprehensive but unadjudicated entry-by-entry; the JEP stratum is the richest on mechanism but produced under confession incentives that reward completeness of pattern and may shade individual self-positioning within it; the memoir-and-press stratum on the presidency is politically saturated in both directions. The 6,402 should never be cited in the corpus without its character (JEP minimum universe, Feb 2021); "systematic" should be cited with its adjudicated scope; and claims about Uribe's knowledge should be marked as unestablished in either direction. Cross-reference discipline: the Democratic Security era's security gains are documented at CO-A-01/CO-A-02 and are real; this document establishes that a disputed fraction of their statistical expression was fabricated; both statements are true and the corpus asserts them together.

8. Conclusion

The falsos positivos are the Palace of Justice's successor case in the corpus's contested-legacies block (CO-J-01), and the comparison is instructive in both directions. Like the Palace, the case fixes the pattern in which civilian authority claims the results of military action while disclaiming its methods; unlike the Palace, the atrocity here was not a single catastrophic event but a distributed administrative practice — thousands of small, bureaucratically processed murders, each with paperwork — which is why its exposure required statistics before testimony, and why its decisive evidence is a number. The 6,402 occupies in this case the position the Casa del Florero footage occupies in CO-J-01: the fact around which all interpretation must organise itself.

What can be stated as settled is, as in all J-block documents, narrow and grave: between 2002 and 2008, under a security policy whose results regime attached rewards to combat kills, members of the Colombian Army murdered civilians by the thousands and fabricated the records that presented them as battlefield dead; the state's formal incentive instruments and its command climate made the practice profitable and its falsity invisible to every control that did not want to see it; the practice collapsed when exposure made it costly; ordinary justice punished the bottom of the structure; the transitional system mapped the structure itself, indicted into brigade command, staged the perpetrators' acknowledgment before the victims' mothers, and traded prison for that truth in sentences whose execution is now the test; and no forum has yet adjudicated — in either direction — the responsibility of the army's command or the presidency. Everything beyond that remains the property of the three accounts: a state crime awaiting its apex defendants; an institution's betrayal by its criminals, since corrected; or the hardest case a restorative-justice experiment has yet carried, on whose handling the experiment will be judged.

The document's obligation, as throughout the J-block, is to keep all three accounts in evidence-bearing condition while the record improves — and the record here is still improving on a schedule: pending sub-case sentences, the Montoya adversarial track, the TOAR-compliance record, and the 2026 election's verdict on the political environment in which all of these will be executed. The corpus's one adjudication is the one the consensus floor permits: the mothers of Soacha were telling the truth from the first week, and every institution that took eighteen years to agree with them owes its current findings substantially to their refusal to stop. The 6,402 are the reason the record must keep improving: as of this document's version date, an unknown number of the victims remain unidentified in N.N. graves, and the case is — in the same precise sense as CO-J-01's disappeared — not over.

Sources

Jurisdicción Especial para la Paz, Sala de Reconocimiento de Verdad, Caso 03 (Asesinatos y desapariciones forzadas presentados como bajas en combate por agentes del Estado): Auto de determinación de hechos y conductas, Subcaso Norte de Santander (Catatumbo), Auto 125 de 2021 (julio 2021); Auto 033 de 2021 (18 de febrero de 2021, priorización y universo de 6.402 víctimas) [TBD-VERIFY: auto numbering]; subsequent sub-case autos (Costa Caribe, Casanare, Antioquia, Huila, Meta) and the first-instance sentence of September 2025 (Catatumbo).
Philip Alston, Report of the Special Rapporteur on extrajudicial, summary or arbitrary executions — Mission to Colombia (8–18 June 2009), UN Doc. A/HRC/14/24/Add.2, 2010; and the Rapporteur's June 2009 end-of-mission press statement.
Human Rights Watch, On Their Watch: Evidence of Senior Army Officers' Responsibility for False Positive Killings in Colombia, June 2015.
Ministerio de Defensa Nacional, Directiva Ministerial Permanente 029 de 17 de noviembre de 2005 (criterios para el pago de recompensas); Directiva 300-28 de 2007 (ponderación de desmovilizaciones y capturas sobre muertes en combate); Directiva 208 de 2008; and the fifteen-measure reform package of November 2008.
Fiscalía General de la Nación, Unidad Nacional de Derechos Humanos y DIH, consolidated falsos positivos docket and the Soacha case files (2008–), including the convictions of members of Brigada Móvil 15 and the Batallón Francisco de Paula Santander and of the civilian recruiters.
Comisión de la Verdad (Comisión para el Esclarecimiento de la Verdad, la Convivencia y la No Repetición), Hay futuro si hay verdad — Informe Final, 2022, chapters on extrajudicial executions and the responsibility findings on the security forces.
Oficina en Colombia del Alto Comisionado de las Naciones Unidas para los Derechos Humanos, annual reports on Colombia, 2004–2012 (documentation of execution complaints and the post-2008 reforms).
Coordinación Colombia–Europa–Estados Unidos (CCEEU) and CINEP/Programa por la Paz, extrajudicial-executions case databases and reports, including Ejecuciones extrajudiciales en Colombia 2002–2010 documentation series.
Daron Acemoglu, Leopoldo Fergusson, James Robinson, Dario Romero and Juan F. Vargas, "The Perils of High-Powered Incentives: Evidence from Colombia's False Positives", American Economic Journal: Economic Policy (2020; NBER Working Paper 22617, 2016) [TBD-VERIFY: final publication venue and year].
Fellowship of Reconciliation (FOR) and CCEEU, Military Assistance and Human Rights: Colombia, U.S. Accountability, and Global Implications (John Lindsay-Poland et al.), 2010 [TBD-VERIFY: exact title and authorship].
Omar Eduardo Rojas Bolaños and Fabián Leonardo Benavides Silva, Ejecuciones extrajudiciales en Colombia, 2002–2010: Obediencia ciega en campos de batalla ficticios, Universidad Santo Tomás, 2017.
MAFAPO (Madres de los Falsos Positivos de Soacha y Bogotá), testimonial and memorial record, 2008–2026, including the Ocaña hearing interventions (April 2022) and the costurero memory practices; and the Campaña por la Verdad, "¿Quién dio la orden?" mural and litigation record (2019–2023).
Presidencia de la República, statements of President Álvaro Uribe Vélez on the Soacha cases (October–November 2008), the 29 October 2008 dismissal announcement, and Uribe's subsequent public accounting, including No hay causa perdida (2012) and his testimony record.
Record of the resignation of General Mario Montoya Uribe (4 November 2008) and of his proceedings before the JEP (2018–), including the victim petitions concerning his non-acknowledgment and the adversarial-track referral record.
The New York Times (Nicholas Casey), "Colombia Army's New Kill Orders Send Chills Down Ranks", 18 May 2019; and the Colombian army's and government's responses.
Corte Constitucional de Colombia, jurisprudence on military jurisdiction (fuero penal militar) in extrajudicial-execution cases and on the transitional-justice framework (review of Legislative Act 01 of 2017 and the JEP statutory law) [TBD-VERIFY: principal decisions, including C-674 de 2017 and C-080 de 2018].
International Criminal Court, Office of the Prosecutor, Situation in Colombia — Final Report of the Preliminary Examination and the Cooperation Agreement with the Government of Colombia, October 2021.
Olga Behar and Carolina Ardila Behar, El caso Klein: El origen de los falsos positivos [TBD-VERIFY: relevance and citation]; and the Colombian investigative-press archive on the cases: Semana (including the 2008 Soacha exposures and the 2019 command-pressure investigations), El Espectador, El Tiempo, Verdad Abierta, La Silla Vacía, 2008–2026.
Michael Evans (ed.), National Security Archive, declassified U.S. documents on Colombian military reporting and the false positives, including embassy cables on body-count pressure [TBD-VERIFY: the specific NSA briefing books].
Jurisdicción Especial para la Paz, public acknowledgment hearings record: Ocaña, Norte de Santander (26–27 April 2022) and Dabeiba, Antioquia (June 2022) — transcripts and audiovisual record of the comparecientes' confessions and the victims' interventions.
Unidad de Búsqueda de Personas dadas por Desaparecidas (UBPD) and JEP forensic record: the Dabeiba (Las Mercedes cemetery) exhumations, 2019–2021, and the continuing identification programme for N.N. burials associated with reported combat deaths.
Transitional-justice scholarship on the Colombian model: Rodrigo Uprimny and colleagues (Dejusticia) on restorative sanctions; Kathryn Sikkink et al. on the justice cascade applied to Colombia; comparative assessments of the JEP's Case 03 in the International Journal of Transitional Justice and related literature, 2019–2025 [TBD-VERIFY: specific articles].

CO-J-01: Three Accounts — The 1985 Palace of Justice Siege — the J-block predecessor case; the civil–military pattern (civilian authority claiming results, disclaiming methods) runs from the Palace to the falsos positivos
CO-A-01: Plan Colombia — the US-funded security build-up whose assistance and metrics environment the killings occurred within; the human-rights certification debates
CO-A-02: Álvaro Uribe Presidency (2002–2010) — the Democratic Security policy, the results regime, the Soacha scandal and the 2008 purge in their presidential context
CO-B-02: The 2012–2016 Havana FARC Peace Negotiations — the negotiation that designed the transitional-justice system Case 03 runs on
CO-B-03: 2016 Peace Referendum and Implementation — the JEP's creation, contestation, and survival; the institutional frame of the acknowledgment-vs-adversarial tracks
CO-C-01: Iván Duque Presidency (2018–2022) — the 2019 command-pressure relapse controversy; the Duque-era attempts to modify the JEP; the 6,402 finding's political reception
CO-H-PRES-02: Álvaro Uribe Vélez — A Biography — the Uribe-knowledge question, the 2008 dismissals, and the separate 2024–2025 witness-tampering conviction
CO-I-01: Colombian Constitutional Court and Judicial Architecture — the fuero militar jurisprudence and the constitutional review of the transitional-justice framework
CO-K-02: The Santos 2010–2012 Decision to Negotiate with the FARC — the decision pathway to the justice model that ultimately processed the case
CO-N-01: Colombia in International Perceptions — the falsos positivos in the external human-rights lens; this document closes the gap CO-N-01 flagged
CO-O-01: Colombia Megatrends — the 2026 security-nostalgia dimension and the long-run civil–military settlement question
PE-J-02: Three Accounts — The 2009 Bagua Massacre (El Baguazo)
CO-F-03: back-reference added by symmetry sweep
CO-K-01: back-reference added by symmetry sweep

End of document. Status: DRAFT. Three-accounts discipline applied; no account adjudicated except the narrow consensus floor stated in §8. Sources: 22 references. TBD-VERIFY tags mark victim counts and their methodologies, directive numbers and texts, dismissal rosters, JEP auto numbers and sub-case statuses, the September 2025 sentence's terms, Montoya's procedural posture, and verbatim quotations pending verification against primary records. Symmetry pass for Related Documents pending.