Transactions of the Charles S. Peirce Society: A Quarterly Journal in American Philosophy

Peirce presented early formulations of abductive inference nearly 150 years ago. Since then new interpretations have been presented (some of [End Page 252] them quite independently of Peirce), but there is still a close link to Peirce’s formulations. Lorenzo Magnani has written a new book on abduction where the main theme is not Peirce’s interpretation of abduction as such, but the book is very interesting from the Peircean perspective. In this review I shall focus on some of the themes from this point of view.

Magnani presents a broad view on abduction instead of presenting only one specific interpretation of abduction. This reflects the history of abduction. Peirce’s writings themselves give elements for various interpretations. It must be remembered that Peirce developed abduction (under various names) during a time period of approximately 50 years (from the 1860s until 1913).¹ It is debatable how much Peirce changed his views on abduction. Usually two main interpretations of Peirce’s views are discussed; early on Peirce emphasized abduction as an evidencing process with a syllogistic interpretation, and the later Peirce emphasized a methodological process where abduction is especially related to the first stages of inquiry.² A related feature, which is not so often noticed, was the relationship of abduction to a “guessing instinct.” In his early writings Peirce already had argued that human beings have an instinct for guessing fertile hypotheses but he explicitly argued that this cannot be a basis for the validity of inference in general, nor of hypothetic inference.³ In his later writings, this kind of instinct was more prominently suggested as a basis for abduction.⁴ After Peirce, abduction was very marginally analyzed for a long time. An interest in abduction was awakened by a new interest in discovery in the 1980s. N. R. Hanson was an early proponent already in the 1950s and 1960s. Abduction is still today closely linked to discussions of discovery (as can also be seen in Magnani’s book) but not necessarily. Quite independently of Peirce, Gilbert Harman formulated an inference-to-the-best-explanation (IBE) model, also during the 1960s, which is different but so close to Peirce’s formulations that some came to align it with ‘abduction’.

Magnani does not stick to one kind of an interpretation of abduction. Nowadays abduction is developed and used in many areas of research, for example: logic, methodology, semiotics, artificial intelligence, and philosophy. It is then no surprise that there are quite different interpretations of abduction. Magnani clearly does not want to emphasize (potential) conflicts among various interpretations but aims at seeing a variety of uses of them. For example, when analyzing the role of instinctual aspects compared to “inferential” aspects of abduction, he maintains that there is no real opposition between them. Many, me included, see this relationship as more problematic. Still I think that Magnani’s view is quite similar to what the later Peirce was suggesting. If we aim at understanding how human cognition works at its best when using or generating hypotheses, there is no sharp line between inferential and instinctual aspects; rather, human beings are using all possible kinds of [End Page 253] elements as aids. Magnani does not want to understand only abductive reasoning but more broadly abductive cognition. Peirce provides a good basis for this with a general theory of signs and representations. Magnani also nicely shows how there are similarities in Peirce’s ideas on the role of an “abductive” instinct of animals (like chickens to peck food) and in Darwin’s arguments on “simple” animals like worms having already some rudimentary degree of intelligence.

Even if admitting this continuum from instinctual to inferential aspects, I would argue for a clearer analytic separation of these aspects (in line with Peirce’s early views on this matter). One argument for this kind of a separation is that the instinctual basis for abduction quite easily stops short of any further explanation of the processes in question. If...