¹ In his early writings Peirce called this type of argument hypothesis; later he called it abduction, and sometimes retroduction or presumption. For the sake of simplicity “abduction” will be the term used throughout this paper.

² The year 1891—which is the date at which he left his government post and retired to his home at Milford, Pennsylvania—marks the beginning of a transitional decade dividing the two periods.

³ The distinction drawn in the first period between induction and abduction was kept in a modified form in the later period as the distinction between quantitative and qualitative inductions.

⁴ In the early period Peirce regarded his three categories as a classification of signs rather than of things; any relation or reference to a correlate was a Second, while only meanings or representations were Thirds. In his later period Peirce extended the scope of the categories, interpreting them as possible kinds of being, rather than as kinds of thoughts; thus Secondness becomes the idea of dyadicity in experience, the idea of compulsion and brute force, and Thirdness becomes the category of rationality, objective value, and law.

⁵ What Peirce called logic in his first period he subdivided into phenomenology, esthetics, ethics, and logic in the later period.

⁶ This seems also to have been the purpose of another branch of logic, called by Peirce formal rhetoric, speculative rhetoric, objective logic, and methodeutic.

⁷ Thus Mr. R. B. Braithwaite says that “Peirce only differs from the orthodox account [of scientific procedure] in classifying the thinking of the hypothesis as itself a form of ampliative reasoning, and this difference is, I think, merely a verbal one, namely, that of whether ‘an act of insight …' (5.181) is or is not called reasoning.“

“Review of Collected Papers,” Mind 43 (1934) 510.

⁸ It is worth noting that on this view abduction is no longer to be classed along with induction as over against deduction, for a logic of discovery is as pertinent to mathematics as to science.

⁹ Peirce himself never pointed this connection out, but it is an historical fact that he did not adopt the view that the discovery of an hypothesis is reasoning until he had conceived of reasoning as normative; for his conceptions of logic as a normative science and of abduction as a logic of discovery were both advanced during his later period. See a paper by the author, “Peirce's Conception of Logic as a Normative Science,” The Philosophical Review, 52 (1943) 187–193.

¹⁰ “Logical Machines,” American Journal of Psychology, 1 (1887) 165–170.

¹¹ Ibid., pp. 168–169. Peirce also argues that the memory capacity of such a machine is limited, whereas that of a human reasoner is unlimited, if the possibility of recording on paper is admitted. Though this was true of the machines available in Peirce's time, it is not true of the computing machines built today. See the article by the present writer, “Super Electronic Computing Machine,” Electronic Industries 5 (July, 1946) 62–67.

¹² “Scientific Method,” Vol. II, pp. 500–503, by Peirce and James Mark Baldwin.