III.07 / INTELLIGENCE
Human-source intelligence as public methodological history
How did historical investigators assess a human account, and which independent records led them to revise it?
Start with the question.
Human-source information reaches an analyst through people with different access, memories, incentives, and vocabularies. Public retrospectives such as Kent's account of the Cuba estimate help expose how judgments can fail when expectations shape the interpretation of reports. The historical task is to reconstruct what was knowable at the time and to distinguish direct observation from hearsay or retrospective explanation.
Map the dependence between three reports.
Suppose three published recollections repeat the same unusual detail. Trace their citations and interview dates. If all three ultimately depend on one witness, the detail has one evidential origin even though it appears in three publications. A source graph makes that dependence clearer than a raw count of agreeing accounts.
What would you inspect?
- The witness's opportunity to observe and the interval before the account was recorded.
- Shared sources, copied phrasing, and later revisions to a recollection.
Take the next step.
Analyze one public historical dispute by separating independent observations, inherited claims, and the later narrative built around them.
Put part of the question to work.
These models isolate a mechanism you can vary. Their assumptions define the part of this topic they can illuminate.
Try connectivity and removal.
Change one parameter, compare the numerical output, and export the result. The experiment loads when you open it.
Watch the quantities change.
Connectivity and removal
Compare local connections with the paths that hold a network together.
Read the explanation and method
Forty-eight points are drawn from one fixed random seed. A link joins two points whenever their distance is within the displayed radius.
The radius grows from 0.05 to 0.4 of the unit square. Existing links remain while new links merge previously separate connected groups.
Orange marks the largest connected group. Connectivity here describes paths through this constructed graph; it does not measure information quality, influence, or causality.
Method. Two nodes share an undirected edge when their Euclidean distance is at most the connection radius. After removal, a depth-first traversal assigns every active node to a connected component. Groups are ranked by size; ties use their lowest node identifier.
Read the supporting material.
- A Crucial Estimate Relived
Sherman Kent / Studies in Intelligence · A participant's retrospective on the 1962 Cuba estimate; useful for examining source evaluation and analytic failure.
