Evidence Status and Pathological Uncertainty: A Working Standard for War Contamination Information

Abstract

Research on war-related environmental contamination operates within heterogeneous information environments where measurements, eyewitness accounts, official statements, archives, reconstructions, inferences, and predictions are mixed. Once incorporated into reports or databases, these materials may acquire the visual form and be treated as equivalent facts.

This paper defines the resulting condition as pathological uncertainty. Unlike ordinary uncertainty concerning measurement, sampling, or model performance, pathological uncertainty arises when sources and types of uncertainty accumulate within an information system and their origins become obscured. More documents may then increase information volume without improving knowledge.

Wartime information requires caution. Strategically interested accounts may exaggerate or conceal losses, alter categories, present incomplete records as comprehensive, or create confirmation through repetition. Archival preservation proves that a statement existed, not that its content was true.

The proposed system distinguishes six evidence statuses: Observed, Documented, Reconstructed, Inferred, Predicted, and Verified. These categories are not an automatic ladder toward truth; verification must specify what was tested and how. Every record should retain its source, date, location, method, transformations, uncertainty, dependencies, contradictions, and missing data. A carefully governed database may therefore provide greater value than a large corpus of poorly classified records.