The philosopher P. Achinstein (Speculation, p. 16ff) developed a helpful concept of evidence. Four concepts are presented: potential, veridical, ES, and subjective.
(A) Some fact e is potential evidence that a hypothesis h holds, iff the probability of there being an explanatory connection between e and h, p(E(e,h)|e) > 1/2. (with e := true, e ⊬ h).
* Notes to potential evidence
- the inequality probably means that p(E) ≠ 1/2, i.e. not indifferent or know-nothing situation, because it'd be nice to have a 0% or 100%.
- The "explanatory connection" between e and h seems to be given by an additional list of assumptions (p. 194-195) which include correlations grounded in empirical data, etc.
- The "explanatory connection" E(e,h)|e explained in terms of a threshold probability of 50%, is maybe related to a statistical type of explanation (i.e. a la Wesley Salmon)?
No comments:
Post a Comment