For two decades, the chimpanzees of Ngogo moved through Uganda’s Kibale National Park as a single society, more than 200 strong, the largest wild chimpanzee community ever documented. They groomed together, patrolled together, raised offspring in overlapping ranges. Then, beginning around 2015, the bonds started to fray. By 2018, the community had fractured into two rival factions. What followed was not a cold separation. It was war.
A study published in Science in early 2025 lays out the evidence, drawing on daily ranging data archived on Zenodo and covering a full decade of field observations running from 2012 to 2022. The paper reports 24 coordinated lethal attacks between the two subgroups. At least seven adult males and 17 infants were killed. The victims were not strangers from a neighboring territory. They were former groupmates, individuals the attackers had once traveled with, groomed, and defended. (The Zenodo links cited throughout this article point to supplementary datasets deposited by the authors; the Science paper’s own DOI has not yet been independently confirmed by this publication.)
A decade of daily observation
The Ngogo study site has been continuously monitored since the mid-1990s, making it one of the longest-running chimpanzee field projects in the world. Researchers followed parties of chimps on foot every day, recording who traveled with whom, who groomed whom, and where each party ranged. That painstaking record, now archived in publicly available replication datasets, forms the backbone of the Science paper’s analysis.
Using network-science methods, including optimization tools drawn from machine learning, the research team combined grooming, proximity, and party-association data to pinpoint when the community’s social fabric began pulling apart. Change-point detection algorithms flagged 2015 as the onset of polarization, years before the two groups fully separated. By 2018, association patterns had hardened into two discrete clusters that the researchers labeled “western” and “central,” each occupying its own core territory with a shrinking overlap zone between them.
The spatial picture matched the social one. The western subgroup increasingly concentrated its movements in one sector of the Ngogo range; the central subgroup did the same in another. Field teams documented patrol-like behavior along the emerging boundary: males from one faction moving in tight, coordinated parties, scanning and listening for rivals. When they detected vulnerable individuals from the other side, particularly isolated males or mothers with infants, those patrols sometimes escalated into the lethal attacks tallied in the study.
A peer-reviewed account of coalitionary killing at Ngogo details specific incidents, including stalking behavior and the circumstances surrounding individual victims. Research permissions were granted by the Uganda Wildlife Authority and the Uganda National Council for Science and Technology.
Why it matters beyond Ngogo
The last time scientists documented a chimpanzee community splitting and then turning on itself was at Gombe, in Tanzania, where Jane Goodall’s research team watched a smaller group fracture in the early 1970s. The violence that followed, known as the Four-Year War (1974 to 1978), shocked primatologists and reshaped understanding of chimpanzee aggression. Research on that earlier case showed increasing subgrouping in association and grooming networks before the permanent division, a pattern strikingly similar to what the Ngogo data now reveal.
But the parallels have limits. Gombe’s community was far smaller and faced heavy human encroachment. Ngogo’s group was enormous, embedded in a relatively intact forest, and had been cohesive from at least the mid-1990s until roughly 2015, according to the long-term field record described in the Science paper. The fact that both communities followed a similar trajectory, from subtle social polarization to open lethal conflict, suggests the pathway may be a recurring feature of chimpanzee societies rather than a one-off event. It does not prove the same triggers are at work in every population.
Foundational work by Wrangham, Wilson, and colleagues on lethal intergroup aggression established the “imbalance of power” framework, which predicts that larger groups attack smaller ones when the cost of violence is low. The Ngogo fission complicates that model. The aggressors and victims were, until recently, members of the same community. Once the social network fractured, former allies effectively became outsiders, yet they retained detailed knowledge of each other’s ranging patterns and vulnerabilities. That insider knowledge may have made certain individuals, especially isolated males and mothers with newborns, particularly easy targets, blurring the line between classic intergroup warfare and something closer to civil conflict.
What remains uncertain
The study’s ranging data end in 2022. As of May 2026, no published update addresses whether the western and central groups have stabilized, whether further splintering is underway, or whether any reconciliation has occurred. Secondary reporting mentions ongoing monitoring at Ngogo, but specifics are absent. Without updated maps of party movements and encounter rates, it is not yet possible to say whether the violence has tapered off or entered a new phase.
The role of female chimpanzees in the split is another gap. The network analyses focused heavily on male association and grooming patterns, which is standard in chimpanzee field research because males are philopatric, staying in their birth community for life. Females transfer between groups and can influence alliance structures in ways the current data do not fully capture. Whether females actively chose sides, were coerced, or simply followed food resources is an open question.
Conservation implications remain unaddressed by the institutions responsible for Kibale. The Uganda Wildlife Authority approved the research, but no public statement from the agency discusses how the fission and its violence affect management of the Ngogo population. Across East Africa, habitat fragmentation limits where displaced chimpanzees can go. If one subgroup is substantially larger or controls better foraging territory, the smaller faction faces compounding disadvantages: fewer males for defense, reduced access to food, and higher vulnerability to further lethal raids. Earlier Ngogo research by Mitani and colleagues documented how intergroup aggression led to territorial expansion over roughly a decade, with the community occupying areas previously held by rivals it had killed (Mitani et al., 2010, Current Biology). Whether that dynamic is now playing out between two halves of what was once a single group has not been confirmed.
Perhaps the deepest uncertainty involves cause. The authors point to Ngogo’s exceptional size as a likely source of instability, arguing that a community exceeding 200 individuals may have surpassed the social limits for cohesive chimpanzee societies. The Science paper does not specify a single census figure at the moment of fission, and the “more than 200” estimate reflects the community’s approximate size during the study period rather than a precise count at the point of fracture. Alternative explanations have not been ruled out: shifts in fruit availability, disease outbreaks that disrupted dominance hierarchies, or human disturbance at the edges of the range. Without parallel ecological time series, it is difficult to separate demographic pressure from other stressors that might have pushed the community toward fission.
Reading the evidence carefully
The strength of the Ngogo record lies in its observational depth. Ten years of daily follows, combined with network analyses archived in open-access datasets, allow independent researchers to verify the polarization timeline and the spatial separation. This is primary behavioral evidence: direct observation, not inference from genetic samples or camera traps alone. When field teams record the composition of every party they encounter, day after day, they build a record that can be tested from multiple angles, by grooming, by co-traveling, by shared patrols, to determine whether apparent factions are statistical noise or real social units.
Still, readers should distinguish between the counts of lethal events and the interpretation of why they happened. The 24 attacks and the resulting deaths are observational facts recorded by researchers present during or immediately after the incidents. The claim that demographic pressure destabilized the community is an inference, supported by the group’s unusual size but not yet tested against competing explanations. As with many long-term field studies, the behavioral evidence is stronger than the causal narrative built around it.
For now, the Ngogo record stands as both a warning and a resource. It shows that even long-stable chimpanzee communities can tip into violent schism, and it offers a uniquely rich dataset for understanding how social networks fracture, how alliances harden, and how lethal aggression unfolds when yesterday’s partners become today’s enemies. As further field seasons and analyses accumulate, they will either reinforce the current interpretation or reveal a more complex story in which ecology, disease, and human proximity all played a part.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
