Written by Nadia Kim (2026.01.20)
Methodological discussions often emphasize the importance of “new evidence” for theory building. At face value, this emphasis can appear puzzling. Empirical research is cumulative by design: data are generated, interpreted, and mobilized through existing theories, literatures, and debates. In this sense, evidence is rarely epistemically new.
The apparent tension dissolves once “newness” is understood not as the novelty of information, but as novelty of inferential role.
Evidence becomes new when it changes how competing explanations are evaluated. This can occur in several ways:
The key shift is not what the evidence is, but what it does.
From this perspective, the contribution of evidence lies not in revealing unknown facts, but in constraining interpretation. Evidence is informative when it:
Evidence that merely fits an argument is weaker than evidence that forces choices between explanations.
Importantly, this way of reasoning is not unique to any particular methodological framework. It is already embedded in good scholarly practice.
Careful literature reviews implicitly establish which relationships are theoretically plausible, which mechanisms remain contested, and where uncertainty persists. Hypothesis formulation, variable selection, and model specification further translate these assessments into empirical expectations.
What some methodological traditions ask is not that researchers pursue novelty for its own sake, but that they make this inferential structure explicit.