1The book under review tackles one of the most puzzling aspects of Charles S. Peirce’s philosophy, namely his conception of the relationship between psychology and logic. As Jean-Marie Chevalier notes from the very first pages of the introduction to his book, Peirce’s entire work is traversed by two strands of thought that apparently contradict each other. On the one hand, Peirce repeatedly insisted upon the autonomy of logic from psychology. Viewed from this angle, he was one of the most significant anti-psychologistic thinkers of the period spanning the end of the nineteenth and beginning of the twentieth centuries, along with philosophers such as Gottlob Frege and Edmund Husserl. On the other hand – and unlike these two other thinkers – Peirce on several occasions defended a view of the mutual influence between logic and psychology that might seem to relapse fully into the psychologistic error he elsewhere criticized. Chevalier opens his book with a list of propositions that, at first glance, present a radically psychologistic position (e.g., the idea that logical inferences are “associations caused by an irritation of the nervous ganglia,” 7), and yet are all propositions that Peirce defended at some point in his long intellectual career. How might we explain these contradictions?

2Needless to say, this question has already been the target of a rich body of secondary literature. Broadly speaking, commentators have attempted to ease the contradiction by adopting one of two possible strategies. The first strategy is to describe Peirce’s work as the theater of a conflict, or a creative tension, between two different impulses: transcendentalism and naturalism, for instance; or a metaphysical impulse and a scientistic impulse. According to this interpretive strategy, the contradiction between psychologistic and anti-psychologistic tendencies with which Chevalier opens his book cannot be denied, but neither does it necessarily represent a mortal sin on the part of the philosopher. Indeed, it is precisely the conflict between different tendencies that makes Peirce’s philosophy so fertile and creative. A second strategy, on the contrary, sees the contradiction as something that we ought to solve through analytical effort. This strategy seeks to introduce a number of conceptual nuances, by noting, for instance, that both “logic” and “psychology” are complex and multi-layered concepts in Peirce’s work. Therefore, we cannot answer the question of whether and how psychology influences logic with a simple “yes” or “no.” The proper answer will depend on the specific sense of “logic” and “psychology” that we are considering.

3While influenced by both strategies, Chevalier ultimately proposes a third approach. He argues that the tensions inherent in Peirce’s thought are to be understood as an attempt to articulate a new philosophical position, one that is neither logical nor psychological, but rather points to a field of study that is yet to be “invented.” This field is epistemology or the theory of knowledge.

4It is a stimulating thesis, which deserves credit for emphasizing the epochal significance of Peirce’s thought in the history of modern philosophy. However, it is a thesis that faces an immediate difficulty. As Chevalier himself points out, Peirce made only limited use of the concept of epistemology – although with an important exception that I shall discuss below. What is more, it is not immediately clear in what sense we can take epistemology to originate with Peirce. At least in the broad sense of the term, epistemology has been part and parcel of philosophy since Plato, although it took center stage only in the modern era. If there is, therefore, an inventor of epistemology in this broad sense, it should probably be identified as one of the founding figures of modern philosophy such as Kant or Descartes, rather than Peirce. In a narrower sense of the term, both the word “epistemology” and the equivalent German term “Erkenntnislehre” were coined in the nineteenth century, but independently of Peirce. So in what sense exactly can we speak of Peirce as the inventor of epistemology? I do not think that Chevalier is arguing that Peirce’s work is the causal point of origin of this nineteenth-century branch of philosophy. Rather, I take him to be suggesting that Peirce was a pioneer and a particularly acute representative of the philosophical project that we normally associate with the term “epistemology,” namely the project of investigating “the modes of knowing” in a manner that transcends those traditional dichotomies of modern philosophy upon which the whole question of psychologism was erected (303). These are, for instance, the dichotomy between the normative and the empirical, or between the transcendental and the natural.

5Chevalier’s book is thus a new interpretation of the relationship between logic and psychology which pivots on Peirce’s pioneering role in shaping the field that we commonly call epistemology. But it is also much more than this. It is a painstaking reconstruction of the development of Peirce’s philosophy from the juvenilia to the final phase of his career. In this respect, the book is heir to some of the milestones of Peirce’s studies, such as the works of Karl-Otto Apel, Murray Murphey, and Max H. Fisch, who have long convinced us of the necessity of adopting a philological and diachronic perspective in interpreting the sheer mass of texts and manuscripts that Peirce left behind. And, indeed, Chevalier’s meticulous reconstruction of the various stages of Peirce’s career adds much value to his attempt to solve the puzzle of psychologism, for it provides the reader with a key to better understand the status of the problem and appreciate its central position among the intellectual urgencies that characterize Peirce’s entire oeuvre.

6Following a standard partitioning of Peirce’s intellectual career, the book is divided into four parts, each of which deals with a period of 10-15 years. Part One covers the first decade of Peirce’s work (1857-1867) and deals mainly with his early semiotic writings, his study of Boolean algebra, and his ground-breaking paper “On a New List of Categories” (Peirce 1867/1984). This paper is particularly relevant to the book’s main theme on account of the specific variety of transcendental philosophy that it proposes. Peirce developed a theory of logical and metaphysical categories that takes its cue from Kant but makes at least two fundamental changes. The first change is a correction to the logical basis of Kantian philosophy (27). The second is a departure from the transcendental deduction of categories. This departure, however, according to Chevalier, should not be read as an abdication of the transcendental perspective tout court. What Peirce found problematic about Kant is not so much the idea that deduction should be transcendental, but that it should be… a deduction. In the “New List of Categories,” Peirce retained the idea of a transcendental justification of knowledge but suggested that this kind of justification had to incorporate the other two forms of inference on which his logical theory was based, namely abduction and induction (29-30).

7The second part of the book covers the years 1868-1884. In that period, Peirce wrote two seminal series of essays: the “anti-Cartesian” essays of 1868-1869 (so called because they aimed at refuting a private, intuitionistic conception of the mind) and the essays on pragmatism of 1877-1878. The anti-Cartesian essays represent, in my view, the cornerstone of Chevalier’s interpretation, because it is precisely in reference to these essays that Peirce characterized his work as pertaining to a branch of inquiry that is neither logic nor psychology, but rather the doctrine of knowledge or Erkenntnislehre (65). In using this category, Peirce probably wanted to account for the fact that these essays make heavy use of psychological concepts but at the same time broaden the scope of those concepts to the point that they actually stop being psychological concepts in the proper sense of the word and become logical concepts. A notable illustration of this approach is the concept of emotion. Emotions, for Peirce, are not so much mental states, feelings, or even bodily attitudes (as they are to his fellow pragmatist James). Rather, they are a type of inference, in the sense that they serve the function of “grouping into a general formula the various particular judgments concerning a situation” (71). And we could say the same about other key concepts of Peirce’s epistemology such as habit, doubt, or belief.

8Chevalier next focuses on Peirce’s output during the few years he spent at Johns Hopkins University, in the first half of the 1880s. These were crucial years for Peirce’s study of the relationship between logic and psychology, primarily because he carried out pioneering work in experimental psychology. (On this topic see also Cristalli 2017.) However, Peirce’s foray into experimental psychology remained a relatively isolated case, probably because of the premature termination of his employment at Johns Hopkins.

9Thus, in the 1890s (a period covered by the book’s third part), Peirce returned to more decidedly metaphysical interests. His ambitious philosophy of evolution revolves around the idea of a law of mind. This idea allowed Peirce to apply his evolutionist outlook not only to the realm of nature, but also to the realm of culture, because it provided an account of the propagation and growth of ideas that was analogous to the growth and evolution of living species. Once again we can recognize in this move the tendency, discussed above, to subject the concepts of psychology to a much broader philosophical analysis (180). In Chevalier’s interpretation, it is precisely this tendency that defines the distinctly epistemological strain of Peircean philosophy.

10In the fourth part, devoted to the final 15 years of Peirce’s career, Chevalier focuses on yet another crucial segment of Peirce’s philosophy, namely the project of a classification of the sciences. Broadly speaking, the classification of the sciences is a systematization of the various scientific disciplines and their relations to one another. Logic and psychology are, of course, part of this systematization. Thus, one might think that the classification of the sciences might proffer us the ultimate truth about Peirce’s ideas on the issue of psychologism and psychology’s relevance to logic. However, things are not that simple, because Peirce’s actual ideas about the relationship between the various scientific disciplines are much more fluid and dynamic than one might think when looking only at the classification of the sciences. For this reason, Chevalier does not allow the classification of the sciences alone to adjudicate the discussion of how logic and psychology interact, nor does he consider the absence of epistemology as a separate branch in the classification structure to be a decisive objection to his interpretation of Peirce as the “inventor” of epistemology (241). Rather, he uses the classification of the sciences primarily to highlight a point upon which Peirce insisted a great deal in the last years of his production, namely the idea that logic is a normative science, along with aesthetics and ethics. It is this normative character (in other words, its dealing with issues of right and wrong) that sets logic apart from other, more descriptive, approaches to the study of mind.

11The increasing emphasis on the normative dimension of logic explains in part why the division between psychology and logic, which had become more blurred in the 1880s and 1890s, tends to acquire a stronger value in the latter period of Peirce’s career. Other factors contributed to the same tendency. Chevalier notes the disappointment that Peirce began to harbor toward the “unfulfilled promises” of experimental psychology (217), as well as his growing interest in phenomenology, which represents a radically different way of observing the world from that of the empirical sciences. Finally, one should make mention of Peirce’s growing interest in diagrammatic logic, which increasingly appeared to him as the appropriate tool for capturing the “essence” of our cognitive procedures, thereby strengthening his faith in the possibility of articulating logical truths that are independent of empirical observations. (See Chevalier’s “Interlude” on formal logic on 135-45, as well as the book’s concluding chapter “Diagrams and Dialogue,” 289-301.)

12This review cannot aspire to touch on all the topics explored in this book, which will be very informative both for the experienced reader and for those seeking an introduction to Peirce’s thought. But given that I have just discussed formal logic, I would like to close with a question on that matter. In the course of his survey, Chevalier quite appropriately focuses on two major innovations in Peircean logic: the logic of relatives and the logic of diagrams. He says somewhat less about a third innovation, the logic of abduction. Yet it might be worth exploring this category in more detail, because abduction seems to offer specific opportunities for a dialogue between logic and psychology. Abduction is that type of inference in which the conclusion does not necessarily follow from the premises (as in deduction), nor is it supported by probabilistic reasoning (as in induction). Rather, it is reached by dint of an irreducibly creative operation. Developing normative criteria that help us determine which abductions are correct and which are not is therefore a particularly challenging task, because making these normative criteria derive from any formal procedure does not seem possible. (It is precisely in response to this difficulty that Peirce insisted on the importance of the observational and embodied dimension of diagrammatic reasoning; see 144.) Peirce was indeed inclined to turn to the history of science – to empirical observations of how humans actually reasoned in the past – in order to derive “practical lessons” about how scientists should generate their hypotheses. Can we say the same about psychology? That is, is there a sense in which empirical investigation of cognitive processes can help us formulate “practical lessons” about how we might come up with the right kind of abductions about scientific facts? This question strikes me as a fruitful way to continue researching Peircean epistemology on the basis of the ideas contained in Chevalier’s instructive book.