The Dark Side of the Band of Brothers: Explaining Variance in War Crimes
Bibliographic record
Abstract
On July 25, 1950, an American infantry unit killed a large number of refugees near the South Korean village of No Gun Ri. On December 12, 1948, a British patrol killed twenty-five civilians near the Malayan village of Batang Kali. On March 16, 1993, members of the Canadian Airborne Regiment beat a Somali teenager to death. While each event is horrific, they also represent only one side of the story; many units in these conflicts, facing similar threats, did not kill civilians. This variation raises a critical question: why do some units participate in war crimes while other do not? To answer this question, I tested three explanations: socialization in the laws of war, civilian influence, and unit subcultures. First, I examined the military’s training and enforcement of the laws of war to test whether socialization could explain this variation. Second, I analyzed the influence of civilian leaders. If they exaggerate the importance of a conflict or dehumanize the enemy, units may be more likely to participate in war crimes. Third, I examined the role of unit subcultures. Units may develop beliefs that challenge organizational norms and encourage participation in war crimes. I tested these arguments in case studies of the Korean War, the Malayan Emergency, and the Canadian peacekeeping mission in Somalia. Each conflict provides variation in outcomes: some military units complied with the laws of war and others did not. Based on extensive archival research, I reached three conclusions. First, while the American, British and Canadian militaries as institutions inadequately trained soldiers in the laws of war, junior leaders could compensate and insure compliance with international law. Second, I found that civilian signaling had little effect: soldiers did not trust the statements of civilian leaders. Third, my research revealed that countercultural subcultures may increase the likelihood that units participate in war crimes. These countercultural beliefs have the greatest effect when junior leaders also support them or when junior leaders cannot control the unit. In these circumstances, junior leaders over reliance on punishment fuels the in-group-out-group dynamic that strengthens the subculture.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.020 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.003 | 0.003 |
| Scholarly communication | 0.003 | 0.001 |
| Open science | 0.001 | 0.003 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".