Selectivism Without Minimalism

I suggest the pessimism expressed by Vickers is an artifact of particular selective emphases in SVV's works—especially regarding content minimalism, individual derivations taken by themselves, and their conditioning realist commitment to just predictive success.

Still, Vickers' worry about minimalism is on target: The equilibrium point of interpretive minimalism seems just too close to antirealism (Constructive Empiri­cism, Ramsey sentence non-realism, and such). But Vickers' specific complaints rest on global rather than specific misgivings: Subjecting a theory to the selectivist strategy he endorses strengthens the theory's epistemological credentials, but— Vickers urges—it still leaves open the possibility that some of what remains may be idle, and the minimalist stance behind the strategy pushes towards antirealism. Lamenting the lack of certainty and the vulnerability of theoretical claims in empirical science are old complaints, however. Non-skeptical responses to them are also old. In 1581, confronted with general doubts about the status of astronomy, Christopher Clavius judiciously argued that it is not enough merely to speculate that there may be some other way or method of accounting for the celestial appearances than the one then accepted. For a challenge to have any force, he maintained, an opponent must produce a specific alternative; only then can we profitably decide whether there is a reason to worry about a given option^[80]. A related reaction occurs in Newton's Fourth Rule of Reasoning:

In experimental philosophy we are to look upon propositions inferred by general induction from phenomena as accurately or very nearly true, notwithstanding any contrary hypotheses that may be imagined, till such time as other phenomena occur by which they may either be made more accurate, or liable to exceptions. (My emphasis).

Applying these suggestions to the selectivist approach, then, we should not trim content that is presently both successful and free of compelling specific doubts against it—leave in place content that faces merely possible doubts. This recom­mendation runs against minimalisms that declare inflationary all content that remains open to “possible” doubts, but the point is that those minimalisms lack compelling reason in their support. Unless explicit reasons are given for either keeping or weakening content, the resulting theory-parts will rest on “preferences” and feel rationally vulnerable. This fear of subjectivism is a factor behind Vickers' gloomy conclusion. The good news is that the needed reasons are frequently available, at least in disciplines that thrive in predictive power, or so I will argue.

In the rest of this paper, I propose a naturalistic generalization of the SVV strategy that maintains the focus on inferential analysis of predictions and content-reduction, but lowers the commitment to content minimalism. One general difficulty with SVV's concentration on minimalism and individual derivations is the way this concentration calls attention away from significant realist resources, par­ticularly confirmational relations that become unavailable when one goes for minimalist interpretation or takes derivations in isolation. Particularly pertinent among these resources are representative diachronic records of the empirical and explanatory difference that specific parts make to success as a theory plays the field.

The proposal outlined in the remaining sections moves in two complementary directions: (a) It makes both predictive power and some explanatory power nec­essary conditions for realist commitment to theory-parts. (b) The proposed approach sides with naturalism in trying to bring the assessment of theory-parts more in line with leading confirmational criteria of success and trustworthiness found in scientific practice.

The starting point is a clarification of what makes a theory-part crucial, super­fluous, or toxic for advancing the scientific study of a domain.