Experimental Wars: Learning and Complexity in Counterinsurgency
Bibliographic record
Abstract
How do policymakers learn to solve complex policy problems? I offer an explanation based on a plurality of policy learning processes. Policymakers learn in multiple ways, thereby grappling with the complexity of their environment. I identify three methods of learning. Learning by paradigm deploys systematic assumptions. Learning by doing employs ad-hoc experiments. Learning by simulation replicates others thoughts and intentions. The three vary in their aptitude to dealing with complexity: learning by doing is most so, and by paradigm least, with simulation falling in between. The most effective approach to complexity will involve combining methods. I apply these to learning in counterinsurgency. I take as cases manuals: documents reflecting what their authors learned. I focus on three such cases. The Hessian officer Johann Ewald served the British for eight years, in the American Revolutionary War. He subsequently wrote of the earliest manuals of irregular war in the European tradition. C. E. Callwell, a prominent Victorian military officer, was the author of the most influential British small wars manual of his generation. He served in the Boer war, among many others. David Galula, a French officer who served in East Asia and in Algeria, is perhaps the most influential counterinsurgency theorist-practitioner on record. I trace their learning processes, as documented in their autobiographical writings, explaining the learning outcomes documented in their manuals. I find those who learn in multiple ways are most adaptive to complexity. Galula was most effective at doing so, and Callwell least so, with Ewald falling in between.
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 | 0.017 | 0.078 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.003 | 0.019 |
| Scholarly communication | 0.006 | 0.009 |
| Open science | 0.002 | 0.004 |
| Research integrity | 0.003 | 0.004 |
| Insufficient payload (model declined to judge) | 0.009 | 0.001 |
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".