Historical Non-Identification From Identical Effects: Why an Effect That Several Histories Would Produce Cannot Say Which One Did
Year: 2026
Type: article
Abstract: An observed effect can be consistent with more than one account of how it arose. When different admissible histories would produce exactly the same available information, that information cannot identify which history occurred. This paper sets out Historical Non-Identification From Identical Effects, a structural and conditional constraint on inference from present evidence to past events. The paper distinguishes identical information from evidence that is merely similar or compatible with several histories. Compatible evidence may still favor one account; the constraint applies when the specified information provides no distinction between them. Its scope is limited to the histories and information under consideration. Further evidence or methods that reveal additional differences can change what is identifiable. The discussion relates this constraint to the identification problem and observational equivalence in econometrics. A reported 2009 Berlin case, in which forensic DNA profiles matched identical twins, illustrates the limit of the available evidence. Separate research using more detailed sequencing illustrates how new methods can reveal distinctions unavailable in earlier profiles. Part of the Relational Conditions and Dynamics series within On Human Modeling, the paper examines how attributions can remain within what their evidence supports. It asks which histories remain admissible, whether the evidence distinguishes them, and what additional information, presumption, or burden of proof supplies a decision when the evidence alone cannot.
Author: Cary, Alexandra
Language: English
Subfield: Law
Field: Social Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W7225639287
FWCI:
Cites:
Cited by:
Related to: