1. Introduction

1There are numerous ways to approach the question about the relation between phenomenology, understood as a philosophical approach associated with a tradition that follows the work of Edmund Husserl, and pragmatism, understood widely as inclusive of classic American pragmatism and neopragmatism. Clearly one could propose a straightforward historical account of connections between these two schools of thought. One could also consider specific issues that are addressed in both traditions. One could also focus on thinkers who have attempted to combine both approaches. Each of these strategies could fill the pages of a good-sized article, chapter, or even book. The story I will try to tell will combine these approaches, but I will frame (and thereby limit) my scope by a focus on some contemporary claims about what has been called the “pragmatic turn” in recent embodied-enactive theory. The story has a beginning, a middle and an end, and a plot with two or three turns. I’ll start with the end, and finish with the beginning.

2. The End: The Pragmatic Turn in Enactive Approaches to Cognition

2It’s uncontroversial to say that embodied-enactive approaches to cognition have been influenced by phenomenology. In this regard, for example, we can consider the influence of Merleau-Ponty (and other phenomenologists like Heidegger and Dreyfus) on what is now considered the original statement of the enactive approach, The Embodied Mind, authored by Francisco Varela, Evan Thompson and Elenore Rosch (1991). Later in the 1990s Varela and his colleagues also took up some methodological themes from Husserl (e.g., Varela 1996). I want to consider the claim that at the same time enactivism was also influenced by pragmatism, and specifically, that it introduced a “pragmatic turn” in cognitive science. I won’t spend a lot of time on this, but I will summarize, and expand a little on what I’ve said about it elsewhere (Gallagher 2022).

3There are two points to be made. First, in cognitive science and philosophy of mind pragmatism had already been a strong influence prior to the 1990s and the advent of enactivism. In that sense, enactivism doesn’t introduce a pragmatic turn; there was already something of a pragmatic current running through cognitive science. Second, there is no explicit claim by the initial embodied/enactive theorists that they were offering a form of pragmatism. I say this despite the fact that numerous researchers have associated enactivism with a pragmatic turn in cognitive science (Crippen & Schulkin 2020; Engel, Friston & Kragic 2015; Engel et al. 2013; Johnson 2016; Menary 2015; also see the essays in Madzia & Jung 2016). For example, Engel et al. (2013: 202) state:

In cognitive science, we are currently witnessing a “pragmatic turn,” away from the traditional representation-centered framework towards a paradigm that focuses on understanding cognition as “enactive,” as skillful activity that involves ongoing interaction with the external world.

4It may be true that in 2013 there was a new emphasis on the pragmatic elements of enactive approaches. Moreover, it’s also true that one can see (for example, in Varela and his colleagues) a shift away from internalist, methodological individualistic, in-the-head, representational, formal-syntactical processes (of the sort defended by cognitivist thinkers), to a more externalist, in-the-world, non-representationalist, action-oriented sense-making that highlights the importance of sensory-motor contingencies and social context. Moreover, one could take this to be a certain form of pragmatism (as Engel and others take it to be). But the claim that enactivism represents a pragmatic turn in cognitive science is not correct, or at best, it is an oversimplification.

5In this respect, my first claim is that there already were pragmatic elements at work in cognitive science. Elements of pragmatism influence (implicitly, and sometimes explicitly) debates about cognition from the very beginnings of the classical cognitivist regime – either in its formation or in opposition to that regime. For example, James Gibson’s ecological psychology and his theory of affordances (which is later taken up by enactive cognitive science) was influenced by pragmatism (Gibson 1977; Heft 2001). Of course, it represented a noted opposition to the mainstream. This did not go unnoticed; Fodor and Pylyshyn (1981) defended what they called the establishment view of cognitivism, arguing against what Fodor later called, more generally, the pragmatist view.

As Dewey says, communication by itself does not accomplish anything. In so far as the dialogue between mother and infant succeeds in getting the child to fill his role in exchange… the child is in fact learning not so much a language, as how to proceed in achieving certain ends by the use of language. The input is not a corpus; the output is not a grammar. (Ibid.: 19)

7Bruner, Caudill and Ninio cite Grace De Laguna’s (1927) work on speech, and in so doing they gesture (perhaps unknowingly) to deeper philosophical connections with pragmatism that were hovering in the background, shaping the philosophy of science that was immediately informing the cognitive revolution. For De Laguna, and her co-author (and husband) Theodore De Laguna, the pragmatist view is central to their evolutionary epistemology and to their confirmation holism, sometimes called “concept pragmatism.” This is the idea that “concepts, apart from the conduct which they prompt, mean nothing” (G. & T. De Laguna 1910: 206). If they were not pragmatists in this regard, they were the first neo-pragmatists.

8Bruner’s reference to De Laguna might seem a completely obscure one in this context except that de Laguna had already been informing the work of another quasi- or neo-pragmatist, Willard V. O. Quine (Katzav 2018). Both Grace de Laguna and Quine contributed papers to the same 1950 APA Eastern Division symposium that were subsequently published in The Philosophical Review. Quine’s paper, “Two dogmas of empiricism” (1951), in which, he suggests, there is “a shift towards pragmatism,” was highly influential, indeed, as David Chalmers (2011: 387) suggests “one of the most influential works in 20th-century philosophy.” Like the de Lagunas, Quine defends confirmation holism: no concept or theory can be verified in isolation since it is embedded in a background or web of other concepts and beliefs, including other scientific theories.

9Whether we consider Quine a pragmatist or not (see Haack 2004; Koskinen & Pihlström 2006; Rorty 1980), Quine’s naturalism and empiricism, views widely shared by cognitive science, mean there is no hard line between science and philosophy. Quine (1969: 26) attributes this to the influence of Dewey.

Philosophically I am bound to Dewey by the naturalism that dominated his last three decades. With Dewey I hold that knowledge, mind, and meaning are part of the same world that they have to do with, and that they are to be studied in the same empirical spirit that animates natural science. There is no place for a prior philosophy. (Also see an interview with Quine in Bergström & Føllesdal 1994)

10As Hilary Putnam puts it, “like Quine, the classical pragmatists do not believe that there is a ‘first philosophy’ higher than the practice that we take most seriously when the chips are down” – i.e., the practice of science (Putnam 1994: 154).

11This part of pragmatism is imported directly into the cognitivist camps of cognitive science. Patricia Churchland, for example, follows Quine on this point: “philosophy at its best and properly conceived is continuous with the empirical sciences” (Churchland 1986: 2). This view also leads to debates, which again hover in the deep philosophical background of cognitive science. Starting in the 1970s, for example, a debate between Hilary Putnam and Richard Rorty about the relevance of pragmatism (and neo-pragmatism) ran along simultaneously with their debates about functionalism in the philosophy of cognitive science (Putnam 1975; Rorty 1972, 1982; Richard Bernstein’s essay “The Pragmatic Turn” (2005) and his book, The Pragmatic Turn (2010), focused on Putnam and Rorty; see Legg & Hookway 2021 for a succinct account of Rorty and Putnam’s relation to pragmatism).

13Moreover, and this is my second claim, in contrast to Bruner, Gibson, Quine, Putnam, or Rorty, we do not find in the work of the embodied/enactive theorists like Varela and his colleagues any significant mention of the pragmatism of Dewey or Peirce or James, with perhaps two exceptions (see below), or any explicit claim in the 1990s that they were offering a form of pragmatism. Indeed, there was no mention of pragmatism within the broader embodied-extended-enactivist camp until around 2007 (e.g., Menary 2007; Steiner 2008; Johnson 2008; Chemero 2009; Gallagher 2009). Only then do we find prolonged discussions of the pragmatists – especially Dewey and Peirce – which had not been the case in the enactivist thinkers to that point. And only then do we see emerging a retrospective claim that enactivism had effected a “pragmatic turn” in cognitive science.

14Let me note the exceptions, however. They consist of two passing mentions. Depraz, Varela and Vermersch (2003) do cite Dewey, specifically for his distinction between knowing-how and knowing-that. I think it is significant for our theme that this reference appears in a book that is primarily about phenomenological issues of method, with a more pronounced focus on Husserl. With an even more passing reference to the pragmatism of Peirce, James and Rorty, Depraz, Varela and Vermersch turn directly to engage with phenomenology. Moreover, what the authors of this work say in regard to Dewey is almost verbatim to, again, a passing reference in Varela’s lectures on Ethical Know-How (1999: 19), where it is clear that Varela has been influenced by another phenomenologist, Hubert Dreyfus (specifically Dreyfus’ account of ethical expertise where he cites Dewey for the same distinction between knowing-that and knowing-how, but also includes a broader discussion of Dewey – Dreyfus & Dreyfus 1991). Varela mentions pragmatism twice in this work, and in both cases it is in the phrase “phenomenology and pragmatism” (Varela 1999: 4, 19). I’ll return to this issue below because I think it’s relevant to the general claim that I want to make, namely, that what is being called a “pragmatic turn,” insofar as it is a claim about the advent of enactivism, or the influence of enactivism on cognitive science, is in fact something else.

3. The Middle: The Pragmatic Turn in Phenomenology

15To continue with my non-chronological account, fast forward to 2017 where we find another explicit claim about a pragmatic turn – this time in phenomenology. I’m referencing the collection Pragmatic Perspectives in Phenomenology, edited by Ondřej Švec and Jakub Čapek.

16Švec and Čapek (2017: 1) define “the pragmatic turn in phenomenology: the primacy of practical over theoretical understanding, criticism of the representationalist account of perception and the analysis of language within the context of shared cultural practices.” Still, it is not clear in this context how the phrase “pragmatic turn” is applied to phenomenology. First, a significant amount of discussion in their volume revolves around Heidegger, and it seems that the claim could be that Heidegger takes a pragmatic turn, or creates one in phenomenology.

19This phenomenological pragmatism, and not just the classic pragmatism of James and Dewey, comes into play in pre-enactivist cognitive science, in the psychology of Gibson, who was directly influenced by his reading of Merleau-Ponty (Heft 2001), in Rorty’s Heidegger, and in Dreyfus and those influenced by him. Indeed, one can argue that enactivism, as proposed by Varela, Thompson & Rosch, appropriated this phenomenological pragmatism, especially the pragmatism of Heidegger and Merleau-Ponty. They also favorably cite Dreyfus and Haugeland, although, in varying degrees, they are critical of Rorty and Gibson. Accordingly, this is a complex story with a lot of twists and turns (although not necessarily pragmatic turns).

20To draw a conclusion to these first two (end and middle) parts of the story, we could say that on the one hand, enactivism represents more of a phenomenological turn in cognitive science than a pragmatic turn, although there is certainly a phenomenological pragmatism involved in discussions around the work of Dreyfus and Rorty. On the other hand, one could suggest that since just these various versions of phenomenology were operating in cognitive science, even if they were operating in opposition to the establishment view, enactivism is not really a phenomenological turn. It may be better to say that it continues and builds upon the phenomenology and pragmatism that had been, for some time, informing ongoing debates in this area.

4. The Beginning: Pragmatic Phenomenology

23Spiegelberg examines the correspondence between Peirce and William James, dated 1903, where Peirce tries “to convince James of the necessity of his new phenomenology as outlined in his first two Harvard Lectures on Pragmatism” (Spiegelberg 1956: 168). James was having trouble understanding Peirce’s phenomenological categories, i.e., firstness, secondness, and thirdness (see below). In a follow-up letter in 1904, after James’ essay, “Does consciousness exist?” was published, “Peirce tries to prove to James that his phenomenology is really what James himself was propounding under the new title of radical empiricism” (ibid.). Specifically, Peirce argues that James’ proposal is a phenomenological one, as Peirce understands it.

[T]hat is just the analysis of what kind of constituents there are in our thoughts and lives (whether these be valid or invalid being quite aside from the question). It is a branch of philosophy I am most deeply interested in and which I have worked upon almost as much as I have upon logic. (It has nothing to do with psychology.) […]. [P]henomenology is one science and psychology a very different one […]. Psychology, you may say, observes the same acts as phenomenology does. No. It does not observe the same facts. It looks upon the same world and the same world that the astronomer looks at but what it observes in that world is different. Psychology of all sciences stands most in need of the discoveries of the logician, which he makes by the aid of the phenomenologist. (Peirce, “Letter to James October 3, 1904”; cited in Spiegelberg 1956: 168)

24Spiegelberg, who was a former student of Husserl’s, provides the following gloss:

There is hardly one sentence in this statement with which Husserl could not fully agree. What would have had Husserl’s particular approval is the disregard for the question of validity or invalidity, the emphasis on the radical difference between phenomenology and psychology, the affirmation that phenomenology is a science, a point particularly important to Husserl, and as such the foundation not only of philosophy, but even of logic. Nor would this exhaust the list of possible agreements. (Ibid.: 168)

25The disregard for validity is not a dismissal of logic for either of these logicians, but what some might consider a form of phenomenological epoché (e.g., Stjernfelt 2007: 142). Indeed, both thinkers would agree that logic should be founded on phenomenology (of the transcendental sort for Husserl). As Peirce put it: Phenomenology has no right to appeal to logic, except to deductive logic. On the contrary, logic must be founded on phenomenology (CP 8.297). And in a letter to Dewey, Peirce writes: “There are three sciences according to me to which Logic ought to appeal for principles, because they do not depend upon Logic. They are Mathematics, Phenomenology, and Ethics.” (Letter to John Dewey, June 1904, CP 8.242).

26Spiegelberg goes on to show, however, that there are some important differences between Husserl and Peirce on the issue of phenomenology. For my purposes these differences are not so important, and indeed, one might argue that the differences between Peirce and Husserl were not as great as the differences between Husserl and Heidegger. For example, Spiegelberg points to one difference which seems less of a difference and more of a shared concept. Peirce’s phenomenological category of “firstness” seems equivalent (or close) to Husserl’s notion of hyletic data, in the context of a phenomenology of perception. Firstness consists in qualities or feelings of “myriad-fold variety,” including “redness, an odor, ‘an infinite dead ache’ […]” (Spiegelberg 1956: 169; Peirce CP 5.44). For Husserl (1982: 203), hyletic data consist of “color-data, touch-data and tone-data, and the like […] sensuous pleasure, pain, and tickle sensations, and so forth, and no doubt also sensuous moments belonging to the sphere of ‘drives’ [Triebe].”

Firstness precedes all synthesis and all differentiation; it has no unity and no parts. It cannot be articulately thought; assert it, and it has already lost its innocence; for assertion always implies the denial of something else […]. Remember that every description of it must be false to it. (Peirce CP 1.357)

27Likewise, for Husserl, the hyletic element of consciousness “has in itself nothing pertaining to intentionality” (1982: 203). Although it becomes part of and becomes informed (or synthesized) by the intentionality of consciousness, hyle, on its own, offers no meaning since it is “irrational stuff without any sense, though, of course, accessible to rationalization” (ibid.: 208).

28Isabel Stearns (1952), who assisted Paul Weiss and Charles Hartshorne (who studied with Husserl in the 1920s) in editing Peirce’s Collected Papers, called firstness “the most elusive” of the phenomenological categories. Similar things have been said about Husserl’s notion of hyle, which, for later phenomenologists like Sartre, Gurwitsch and Merleau-Ponty becomes a contentious issue (see Smith 1977; Gallagher 1986).

29Peirce defines secondness as “the mode of being of that which is such as it is, with respect to a second but regardless of any third” (CP 8.328). More concretely, he characterizes it as “brute actions of one subject or substance on another” (CP 5.469), or “the experience of an effort” or “opposition” (CP 8.330). Although Spiegelberg can see no comparable element to secondness in Husserl’s phenomenology, one can find in Husserl’s analyses of active and passive synthesis the idea of a tension, if not what Peirce calls a “struggle,” between egoic and non-egoic processes. This would certainly count as a form of secondness for Peirce, who talks of secondness as a “double-sided consciousness” (Peirce 1999: 268). One might also consider the relation involved in the Husserl’s schema: apprehension – content of apprehension (Auffassung – Auffassungsinhalt), where the content is hyletic content (or firstness).

30And with thirdness, which Peirce characterizes as “the mental or quasi-mental influence of one subject on another relatively to a third” (CP 5.469), but also as “habit” (CP 1.536), “cognition” (CP 1.536-37), or “representation” (CP 5.66), we can think of intentionality, meaning, and the connection with semiotics which would take us back to Stjernfelt’s work.

31For my purposes, it is Peirce’s 1903 Harvard lectures on pragmatism that provide the best evidence of what he took to be the important phenomenological part of pragmatism. There he characterizes phenomenology as follows:

I will so far follow Hegel as to call this science Phenomenology although I will not restrict it to the observation and analysis of experience but extend it to describing all the features that are common to whatever is experienced or might conceivably be experienced or become an object of study in any way direct or indirect. (Harvard Lectures on Pragmatism, 1903, CP 5.37)

32In the lectures he also stressed the need for the phenomenologist to develop three qualities.

1. Seeing what stares one in the face, just as it presents itself, unreplaced by any interpretation, unsophisticated by any allowance for this or that modifying circumstance;

2. Resolute discrimination, which fastens itself like a bulldog upon the particular features that we are studying; and

3. The generalizing power of the mathematician who produces the abstract formula that comprehends the very essence of the feature under examination purified from all admixture of extraneous and irrelevant accompaniments. (CP 5.42)

33Spiegelberg does not, and I think anyone familiar with Husserl’s phenomenology would not see much daylight between these statements and Husserl’s own methods. I am not suggesting, of course, that the phenomenologies of Husserl and Peirce are identical (see Sonesson 2013 for some important differences). It is sufficient for my purposes to say that some form of phenomenology (perhaps a close cousin to Husserl’s) was an important part of pragmatism early on.

35Finally, one could also make the case with respect to Dewey, who in some of his main works (Experience and Nature, and Art as Experience) does not hesitate to provide insights into experience, even if not using a phenomenological method. All of this phenomenological pragmatism, by the way, would qualify any claim that pragmatism remains third-person in contrast to the phenomenological first-person (Švec & Čapek 2017: 8; Crowell 2013). Specifically with respect to Dewey, we see not only how compatible his early work is with Merleau-Ponty’s phenomenology of perception, but we can see how easily his work on aesthetic experience can be integrated with recent phenomenological investigations of performance and embodied experience (e.g., Høffding & Sanchez 2020; Gallagher 2021).

5. Conclusion

36In contrast to any claim about embodied-enactive approaches representing a pragmatic turn in cognitive science, I would suggest that pragmatism had already been informing debates in cognitive science from the very beginning. And in contrast to any claim about a pragmatic turn in phenomenology, it’s easy to see that there have been elements of pragmatism operating in phenomenology all along. And finally, in regard to the relation between phenomenology and pragmatism, Peirce himself made an early turn to something akin to phenomenology; and just as he saw the same thing in James’s radical empiricism, Husserl recognized phenomenological motifs in James’ earlier work. Perhaps, in this case, we should talk about phenomenological turns, rather than pragmatic turns. What motivates the exceptional, but only passing mentions of pragmatism in Varela and colleagues (Varela 1999; Depraz, Varela & Vermersch 2003) is Dreyfus’s pragmatic interpretation of phenomenology, which was one variation of pragmatism that was operating in and around the cognitive sciences.

37I will so far follow Hegel as to say that in this history there is something like a dialectical Bildungsroman of self-consciousness similar to the one developed in Phenomenology of Spirit. Two consistent, albeit not identical, philosophies continue on lines of development more or less simultaneously, without any significant measure of recognition, one for the other. At various points in recent times some proponents of these approaches wake up to that fact and self-consciously come to recognize that there are clear connections between these approaches that can be further developed in the realm of ideas. I’m inclined to follow Merleau-Ponty here and say that there is no complete synthesis or convergence to follow; but surely some productive and interesting intertwinings between phenomenology and pragmatism are still possible.