Dewey and Peirce shared a common focus on the elaboration of a model of inquiry that seeks to remove doubt concerning the answer to some question by identifying potential answers to the question, ascertaining the evidence available for evaluating the candidacy of such answers as solutions to the problem posed, conducting experiments to acquire more evidence and deciding on the basis of the available evidence which of the potential answers to add to the stock of knowledge. My own proposals concerning how to model well-conducted inquiry depart in several respects from the proposals of both Peirce and Dewey. But these two great philosophers gave classical expression to the ideas that inspired the projects I have undertaken. In this essay, I shall comment on some aspects of Dewey's vision of the logic of inquiry, pointing to important respects in which I depart from his approach. Because I shall be arguing with Dewey, I wish to emphasize here and now that I am arguing not to dismiss him or his ideas but to sharpen some of my ideas by confrontation with one of the points of view that inspired them.

The Main Task of Epistemology: Justifying Changes in Full Belief

I would bet my bottom dollar that William Henry Seward was Secretary of State in the Cabinets of Presidents Abraham Lincoln and Andrew Johnson. I am equally confident that while holding that office under Johnson he negotiated the purchase of Alaska by the United States of America from Russia. I am sure that the purchase was in 1867 give or take a year or so. (I am fairly confident that the year was 1867 but not absolutely certain.) However, I am absolutely certain that Seward had his permanent residence in Auburn, New York, for most of his life.

Confident assertion of matters of fact strikes many as out of place. Epistemic modesty requires that I refrain from expressions of absolute certainty not only about the date of the purchase of Alaska but concerning all the other claims made in the previous paragraph.

A public claim of absolute certainty even when one is sincere in the claim is not the way to win friends and influence people. But champions of epistemic modesty are concerned with more than conversational etiquette or political correctness. Not only should we avoid expressing absolute certainty, we should not be absolutely certain about questions of fact.

Many philosophers would concede that we should all be certain about the truth of the theses of logic and, perhaps, concerning other theses lurking in the cores of our conceptual frameworks.

PROBABLE DEDUCTION

Peirce wrote extensively on deduction, induction, and hypothesis beginning with the Harvard Lectures of 1865 and Lowell Lectures of 1866. The ideas that he examined in those early discussions were reworked over nearly two decades until the comprehensive statement of his view contained in “A Theory of Probable Inference” of 1883 that was included in the Studies in Logic, by the Members of the Johns Hopkins University and is reprinted in W 4, 408-450. This remarkable paper developed a version of the Neyman-Pearson account of confidence interval estimation that incorporated the main elements of the rationale offered for its adoption in the early 1930s and presented it as an account of inductive inference.

In his retrospective reflection on the question of induction in 1902 (CP 2.102), Peirce revealed satisfaction with the views on induction advanced in 1883 and this attitude is confirmed in other remarks from that period. However, Peirce did express dissatisfaction concerning his notion of “Hypothetic Inference.” Although Peirce called it Hypothetic Inference or Hypothesis from 1865 to 1883 and later, in 1902, Peirce replaced the term “Hypothesis” with “Abduction.”

TIES FOR OPTIMALITY

Rational agents do the best they can. That is to say, they choose an option from those available to them that is the best, all things considered, among those available options. This maximizing recipe creates no difficulty in case there is exactly one option that is best among the options available to agent X. agent X should choose the optimal option.

Suppose, however, that like Buridan's Ass, X faces a predicament where two or more options are optimal. Which option should X choose? It does no good to say that all optimal options are equally optimal. X knows that. But X may still want guidance as to what to do.

To insist that rational agents never face decision problems where more than one available option is optimal is to disallow the possibility that decision makers can judge distinct options to be equally valuable. If A and B are equally valuable, all things considered, and belong to set of feasible options S, then the subset of S consisting of the pair {A, B} could coherently be a set of available options. Both A and B are equally optimal relative to that set of available options.

Buridan's Ass X need not be frozen in indecision, leading to doom. X will choose for the best no matter which optimal option X chooses. X may not feel the need for guidance between equally optimal options.