The Abductive Structure of Scientific Creativity: An Essay on the Ecology of Cognition by Lorenzo Magnani (review)

Historians of philosophy will spill huge amounts of ink scrutinizing the reason why abduction was highlighted so much in the first two decades of the 21st century. Not to mention the numerous scholarly articles on abduction published in logic, cognitive science, artificial intelligence, semiotics, and philosophy of science, several monographs on abduction appeared during this period: Magnani (2001, 2009, 2017), Walton (2004), Gabbay and Woods (2005), Aliseda (2006), to name a few. It looks as if they were responding to Hintikka’s influential paper “What Is Abduction? The Fundamental Problem of Contemporary Epistemology” published in this journal at the end of the last century (Hintikka 1998).

In order to appreciate the value of Lorenzo Magnani’s third book on abduction, it is mandatory to understand the evolution of his views on abduction over the decades. His active interaction with other important figures in the scientific and philosophical community around abduction must be carefully chronicled. One way of summarizing this fascinating story is to contrast the so-called AKM and GW models of abduction, and introduce Magnani’s new EC (Eco-Cognitive) model as a synthesis.¹ The similarities and the differences between Magnani’s and Woods’ evolving views on abduction through their intellectual interaction could be the key for understanding the recent history of the study of abductive reasoning. Another possibility is to consider the extent to which Magnani has been successful in fulfilling his self-assigned aim of going with and beyond Charles S. Peirce (cf. Magnani 2009). Fortunately, these two possible ways of fathoming Magnani’s mind are two sides of the same coin.

With these two possible perspectives in mind, we may divide the eight chapters of Magnani (2017) into three groups: (1) chapters 1 through 4, where Magnani confronts issues about which he and Woods may have disagreements, (2) chapters 5 and 6, where Magnani not only presents his EC model of abduction but also discusses his and Peirce’s reading of Aristotle on abduction, as if he wants to demonstrate that he certainly goes with and beyond Peirce, and (3) chapters 7 and 8, where Magnani hints at how to go more ambitiously beyond Peirce.

Magnani’s aim in the first part, chapters 1 through 4, is to convincingly argue for the knowledge-enhancing character of abduction. Chapter 1 “Enhancing Knowledge: Tracking the External World” enables us to understand why abduction matters by examining its ignorance-preserving nature. As Woods finds in the ignorance-preserving character essential to abduction (while characterizing deduction as truth-preserving and induction as probability-increasing), Magnani’s strenuous effort to emphasize the knowledge-enhancing aspects of abduction seems to be the best strategy for a well-balanced, meticulous examination of (different varieties of) fallibilism.

Since scientific models loom large for Magnani’s purpose, he is bound to be sensitive to an obvious possible objection, the apparently fictional character of scientific models. In Chapter 2 “Distributed Model-Based Science: Scientific Models Are Not Fictions”, Magnani sharply distinguishes between the process of abducing models and the process of abducing fictions. He acknowledges that scientific models can appear fictional to the epistemologist in a static perspective. He claims, however, that “their fictional character disappears if a dynamic perspective is adopted and their possible “constitutive” character (of new knowledge) is admitted”. [p. x]²

How could abduction be knowledge-enhancing, if it is evidentially inert? Magnani tackles this problem in Chapter 3 “Not Everything in Scientific Cognition. Is Evidence-Based. The Epistemology of Evidentially Inert Knowledge Enhancing”. He finds one of the most relevant examples of knowledge-enhancing abduction in model-based reasoning in science. According to him, abduced models, both in science and in human perceptions, cannot be fictions. Even if they are not justified by an empirical evaluation phase, still they play an important role in attaining empirically fecund knowledge. Magnani claims that “[i]t is clear here we are dealing with cases in which abduction is...