TITLE:
On Retroductive Explanation
AUTHORS:
José Luis Rolleri
KEYWORDS:
Abduction, Deduction, Anomaly, Explanatory Hypothesis, Conceptual Innovation
JOURNAL NAME:
Open Journal of Philosophy,
Vol.16 No.4,
November
13,
2026
ABSTRACT: The aim of this paper is to elaborate on the concept of abduction, or retroduction, as a kind of rational thinking, though not logical, for conceiving and holding new physical hypotheses that could explain anomalous phenomena. Retroduction, as Norwood R. Hanson maintains, starts with data about the phenomenon to be explained and advances to a possible explanatory hypothesis, in reverse order to deductive explanation. The justification of explanations, whether retroductive or deductive, resides in the matching between the prediction of a hypothesis with the data about the explanandum, not in the form of the reasoning involved. However, neither of these kinds of explanatory reasoning, by themselves, allows us to find novel, fruitful hypotheses. There are, at least, two significant differences between retroductive and deductive explanations: firstly, in the former the data are a priori to the formulation of a possible explanatory hypothesis whereas in the latter the data are a posterior to a proposed hypothesis to provide an explanation; secondly, often retroductive explanations involves a conceptual change, even a conceptual improvement, while deduction in explanation allow performing the crucial labor of obtain the logical consequence of hypotheses without adding new conceptual content. Thus, successful retroductive explanations can sometimes innovate the conceptual framework of theories, in contrast to deductive explanations.