• 1 This narrative is especially clear in Mounce 1997, and Misak 2013, and see Rorty 1979, 1982, 1991 [ (...)

1Contemporary Pragmatism is often understood as taking either a “subjectivist” form (typically associated with Richard Rorty), or an “objectivist” form (most often associated with Hilary Putnam).1 However, while most are happy to list C. S. Peirce as the originator of the comparatively objective form, William James’s position in this narrative is harder to place.

• 2 For the most part, I’ll be sticking with “temporary” and “absolute” in what follows.

2This should not be surprising, since the division between these objective and subjective versions of pragmatism tends to be most easily cashed out in terms of competing views about the nature of truth, and James’s talk of truth is notoriously, and deliberately, ambiguous. In particular, James takes our use of “truth” to pick out both “temporary” (or “subjective,” “relative” or “half”) truth, and “absolute” (or “objective or “ultimate”) truth.2 As he put it in his 1908 Harvard seminar:

• 4 See Rorty (1991 [1986]: 128). Something like Rorty’s view of the role of absolute truth in James is (...)
• 5 See Putnam 1987, 2005.
• 6 See the discussion of James (1975 [1909]: 143) in Putnam (2017 [1987]: 173), and Putnam (2017 [2005 (...)
• 7 Putnam (2017 [2005]: 200).

3Roughly, a claim is “temporarily” true if we are warranted in asserting it at the time, and “absolutely” true if it would continue to have such warrant even if inquiry were extended indefinitely. Depending on what sort of truth you focus on, James’s pragmatism will seem very subjectivistic or comparatively objectivistic. So, for instance, Rorty takes James to be primarily interested in temporary truth, with absolute truth only serving as an unrealized (and possibly unrealizable) ideal that plays a role similar to his own “cautionary” use of “true” – a mere reminder that the truths we take to be warranted now may be rejected by us in the future.4 Putnam, by contrast, argues for an objectivist reading of James, one where he takes absolute truth to be the primary sense of the term.5 Putnam argues that the ideal of absolute truth isn’t merely regulative for James, and that not only did James believe that some of our beliefs (such as his own belief in the pragmatic theory of truth)6 actually were absolutely true: he also took the best explanation of why some of our beliefs were warranted (i.e.: “temporarily true”) to be that they were absolutely true as well.7

• 8 For an early instance of this complaint, see Russell (1910: 111).
• 9 See, for instance, James (1975 [1907]: 34, 36-7), and James (1975 [1909]: 132).
• 10 James (2004: 400).

5Of course, most philosophers don’t see such an ambiguity in the word “truth.” On the contrary, they assume that we already have two words that pick out what James calls “temporary belief” and “ultimate truth,” namely “belief” and “truth.” The “unfortunate” thing, according to such philosophers, is not that “truth” is used ambiguously in English, but rather that James uses the normally unambiguous word in such an ambiguous fashion.8 However, James’s position comes from his treating a theory of truth as starting with those things that are taken to be true (the “temporary truths”), and generalizing from those to get an account of the meaning of “truth” (at least in its “temporary” sense).9 This is just a general instance of James’s “method of attacking problems by asking what their terms are ‘known as’.”10 He thinks that looking at our actual use of even a conceptually central terms like “true” can lead us to insights about it that mere a priori reflection on the concept cannot.

• 11 Putnam (2017 [1987]: 185).

6That said, even if James is entitled to suppose that there are two uses of “true,” it can be very hard to tell which of the two types of truth he is talking about at any given point. Putnam claims that “James quite freely equates ‘true’ and ‘absolutely true’; it is ‘half-true’ that always takes the qualifier.”11 However, an examination of James’s talk of truth doesn’t really bear this out. So, for instance, when James writes about belief revision he freely switches from the “previous mass of opinions” to the “older stock of truths,” and claims that “the greatest enemy of any one of our truths may be the rest of our truths” (James 1975 [1907]: 34). He clearly seems to have temporary truths in mind here, since the incompatibility that produces such enmity between truths cannot exist among absolute truths. In much the same way, his claim that “we have to live today by what truth we can get today, and be ready tomorrow to call if falsehood” (ibid.: 107) seems to require that “truth” be read as temporary rather than absolute, especially since the examples he gives right after (Ptolemaic astronomy, Euclidian space, Aristotelian logic, Scholastic metaphysics) are all examples of temporary truths for James.

• 12 See also James (1975 [1909]: 100).

8There are, of course, bare uses of “true” in James that seem more in line with the absolute reading as well, and one reason for the seemingly negligent way James has of flagging which type of truth he is talking about may be his view that “The pragmatist definition of truth applies to both” (ibid.: 434).12 James may feel that there is no need to disambiguate his truth talk, since much of what he says is meant to work equally well on both readings. However, if James really thinks that “truth” is ambiguous, how could he claim that the same definition applies to both uses of the term?

• 13 And one should note (with Putnam 2017 [1987]: 173) that James’s talk of “definitions” here is prett (...)

9Well, while James does take there to be two uses of “true,” he certainly does not take the two meanings are completely independent of each other. In particular, his claim that the definition applies to both meanings of “true” in spite of the fact that these are different meanings can be explained by the fact that “absolute” truth is an idealization of our “temporary” truths.13 That is, it is what we get if we indefinitely extend the norms that govern our movements from one temporary “truth” to another.

• 14 See, for instance, “The Sentiment of Rationality” in James (1979 [1897]).

10The respective roles of absolute truth and temporary truth would thus be very much like those traditionally seen between truth and belief, but absolute truth and temporary truth are not independent of each other in the way that belief and truth were traditionally taken to be. Constraints that determine our temporary truths are still there when we idealize, and so “subjective” factors that contribute to temporary truth can still be there when we get to absolute truths.14

• 15 For a discussion of this, see Jackman 1998.

12This is partially because James takes the process of inquiry to ultimately determine the referents of the expressions involved (rather than just determining the truth-value of independently meaningful propositions), and it is their constitutive contribution to what our expressions mean that allows our subjective preferences to contribute to even the “absolute” truth of our thoughts and utterances.15

• 16 James (1975 [1907]: 107).
• 17 James (1975 [1907]: 107).

15But even if subjective factors contribute to what is ultimately “absolutely” true, why shouldn’t we still simply (as Putnam suggests) tie the word “true” to the ideal associated with it? (Especially given that it is an ideal that we are very attached to.) This question leads to a major difference between the Jamesian and the standard “intellectualist” position, which ties us back to the earlier point that while the intellectualists think of truth as something out there waiting to be discovered, for James “the absolute truth will have to be made.”16 In particular, while we certainly succeed in making our temporary truths, there will be no guarantee that we can make absolute truths, and thus no way of being certain that any of our claims are (absolutely) true or false. This brings a degree of fallibilism about the existence of absolute truth that isn’t as visible on most accounts. Indeed, James at times seems to display not only fallibilism, but also a good deal of pessimism about the existence of absolute truth. Remember that in our initial quotation from James’s 1908 seminar, he described absolute truth as “a purely abstract thing that nobody may, perhaps, ever be in possession of,” and famously claims that it “runs on all fours with the perfectly wise man, and with the absolutely complete experience; and, if these ideals are ever realized, they will all be realized together.”17 It often isn’t sufficiently stressed just how pessimistic a thing this is to say. The perfectly wise man and the absolutely complete experience are ideals we do not expect to ever be realized, and if absolute truth needs to be realized with these two if it is to be realized at all, then we can be quite confident that no claim of ours will ever be “absolutely” true.

• 18 Referring to his earlier indicative understanding of it as being a “grievous error” (Misak 2016: 14 (...)
• 19 Putnam (2017 [1987]: 181).
• 20 As Putnam puts it, “James’s ‘radical empiricism’ has no room for Peirce’s ‘Thirdness’” (Putnam 2017 (...)

17The first of these is James’s purported unwillingness to understand the relation between absolute truth and inquiry in counterfactual terms. Peirce originally characterized truth in “How to make our ideas clear” as “The opinion which is fated to be ultimately agreed to by all who investigate” (Peirce 1992 [1878]: 139), but he famously moved from an indicative to a subjunctive understanding of the conditional involved in the pragmatist connection between truth and inquiry. He thus moves his theory to a focus on how extended inquiry would rather than will go,18 but he took James to have a “nominalistic” commitment to staying with the indicative conditional on this matter. Putnam agrees that James would be unwilling to modify his account of truth in this way,19 but while it may be the case that James is more wary of bare possibilities than Peirce,20 one should not overstate this difference. While James does often tie truth to actual verification, he also insists that the “workings” by which we understand truth can be “actual or possible” (James 1975 [1909]: 7), and that most of our truths stay at the more “virtual” stage. For James, verifiability “is as good as verification” and “[i]ndirectly or only potentially verifying processes may thus be true as well as full verification-processes” (James 1975 [1907]: 99, 100). Consequently, even if James isn’t simply willing to adopt Peirce’s robustly realistic attitude towards possibilities, he could still understand absolute truth in something more like Peirce’s “subjunctive” sense, where something is “virtually” absolutely true if it is what we would settle on if inquiry went to its ideal limit.

• 21 An assumption that would have seemed more natural in James’s time as the influence of the sort of “ (...)
• 22 Indeed, it has been suggested that formal reasons relating to Gödel’s incompleteness proof entail t (...)
• 23 Lamberth (1999: 218; 2005: 228).
• 24 See particularly Putnam (2017 [1987]: 173; 2017 [2005]: 193, 199 n.29), where he talks about James’ (...)
• 25 Peircian truth thus comes out as something closer to Crispin Wright’s notion of “superassertability (...)

18Another reason why one might think that James could be skeptical about absolute truth is that he often seemed to think of it as being unified in a particular way.21 That is, if we only get absolute truth when we have reached the end of inquiry for every question, then we might think that we will never have it.22 James’s talk of absolute truth as running on all fours with the perfectly wise man and absolutely complete experience might encourage this interpretation, and Lamberth certainly takes it as one of the main reasons for thinking that James was skeptical about absolute truth.23 On this reading, when James talks of his account of truth being “an account of truths in the plural” (James 1975 [1907]: 104), these must be temporary truths, since absolute truth is always “one.” Putnam, however, notes that there are some occasions where James seems to talk about individual claims or theories being absolutely true,24 and while Peirce’s own conception of truth has often been understood in terms of something like a global end of inquiry, it can also be understood in a more “piecemeal” fashion, so that an individual belief would be absolutely true if no amount of further inquiry would lead us to revise it.25 Such an account would allow one to understand the truth of individual sentences on their own without any need to presuppose any “global” end of inquiry where all questions were settled. Thus, according to Putnam, even if James has an indicative understanding of the relation between truth and inquiry, he does not need to be a skeptic about absolute truth, because there are many questions for which we have reached the end of inquiry (that is to say, further investigation will not change our current answers to those questions).

• 27 In this respect, I think that Levine 2013 understates how radical James’s views in this area are by (...)

23Note that the quotation above ends with the question of whether any of the competing schemes are truer absolutely, not of which one is.²⁷ Inquiry into a question may never produce a stable answer, since there may be no stable framework for inquiry, and when we adopt, say, a scientific framework, many claims that were previously endorsed will be denied because their ontological presuppositions will be rejected (so even a “piecemeal” truth like “There is a chair in my office” will turn out not to be absolutely true, because in some contexts we will be working with a conceptual framework where there are not “commonsense” objects like chairs at all). If we are stuck moving between such “incomplete” conceptual schemes, our claims will end up being, as James put it above, only “relatively true, or true within those borders of experience” (James 1975 [1907]: 107).

• 28 These consequences are given a more prominent place in these later works, but the basic picture of (...)

24Of course, one might think that this is only a temporary state, and that we should expect that eventually we would be able to find a single conceptual system in which we can explain everything. However, James seems at times pessimistic about the status quo changing, and his reasons for being so become clearer in the series of lecture he delivered a year after Pragmatism, lectures subsequently collected as A Pluralistic Universe. In that book (as well as his posthumously published Some Problems of Philosophy), James presents a picture of how our concepts work that might lead one to doubt that any claim that made use of them could ever turn out to be “absolutely” true.28

• 29 James (1970 [1909]: 109). See also James (1970 [1909]: 96-7), James (1988 [1908]: 438).
• 30 For instance, those concepts discussed in Pragmatism’s chapter on “Common Sense” such as “space,” “ (...)

25According to James, concepts emerged “in the interests of practice essentially and only subordinately in the interests of theory,”29 and our most fundamental concepts30 evolved to serve the most basic of these ends. However, a conceptual system that emerged this way may not be well suited to provide the kind of consistent theoretical account of reality that absolute truth requires. Indeed, James’s infamous rejection of the “logic of identity” in A Pluralistic Universe is best understood as a rejection not so much of logic itself, but of the assumption that the inferential structure of our basic concepts matches the structure of reality at a more “global” level.

• 31 Indeed, one might think this about the concept of truth itself, and that worries about the liar par (...)
• 32 For further discussion, see Jackman 2018.

26James argues that, if extended beyond the practical contexts in which they emerged, our concepts can eventually misrepresent the realities they normally help us cope with, and this problem is not limited to the concepts of common sense.31 James seems to suspect that it will be a problem with any conceptual system, since conceptualization itself misrepresents the “continuous” nature of reality.32 Concepts (at least as James understands them) require sharp boundaries, and while the imposition of models of the world where things are sharply defined has tremendous practical value, James takes it to inevitably misrepresent the richness of reality, and thus such models may be unable to get to a point of absolute truth. The pinch will always be felt, and thus no stable rest to inquiry will ever be reached.

• 33 Putnam (2017 [2005]: 191).
• 34 Putnam (2017 [2005]: 192). Note that Putnam is in some sense aware of this, as seen in his discussi (...)

28Putnam quotes this passage approvingly,33 but there is a serious problem with the metaphor, one that becomes clear when Putnam starts to explain it with the claim that “the ‘beans’ are bits of pure experience.”34 The cast of beans certainly does represent experience, but for James, experience is expressly not broken up into “bits” at all, and its continuous nature is something that he consistently stressed from his critiques of the atomistic nature of associationist psychology in his Principles of Psychology (James 1981 [1890]), to his last writings on percepts and concepts (James 1979 [1911]).

• 35 Putnam (2017 [1987]: 176), see also Putnam 1996, 2001.

29Putnam credits James with being the “first post-cartesian philosopher to completely reject the idea the perception requires intermediaries,”35 but this divergence between percepts and concepts, with the former being continuous and the latter discrete, is one reason why James would have trouble extending the sort of “direct realism” that he has about the perceptual realm to the conceptual. Putnam sees James’s failure to do so as mistake, and thinks that just as “James decisively rejected the interface conception of perception” (Putnam 2017 [1987]: 186-7), he should have posited a more “direct” connection between conception and reality. However, James takes percepts and reality to both have a “continuous” structure, so the identification of the two makes sense for James, but since he sees concepts as having a completely different sort of (“discrete”) structure, the identification is much harder to make.

• 36 See, for instance, James (1975 [1909]: 106-7), and for a discussion of the relevance of this aspect (...)
• 37 Putnam (2017 [2005]: 200).

30It is thus not surprising that James, unlike Peirce, draws a fairly sharp distinction between truth and reality.36 This gives us reason to doubt Putnam’s claim that absolute truth has an “explanatory” role for James in that a belief’s being absolutely true is often the best explanation for it being temporarily true. According to Putnam, the fact that it is (absolutely) true that, say, Franklin Roosevelt was president of the United States in 1940 is no different from the fact that Franklin Roosevelt was president of the United States in 1940, and that fact is the best explanation of all the evidence we have that he was, evidence which in turn explains why we believe that he was (and thus why the belief that he was president in 1940 is temporarily true).37 Now it may be the case that if reality really can be broken up into “facts” that mirror the structure of our concepts, then those facts (and the absolute truths that express them) would best explain why many of our beliefs are warranted. However, it is just this conception of reality that James questions in his later work, and while a belief of our’s may be warranted because it fits reality better than any other, that “fit” may still not be tight enough for what it is fitting to be considered a fact that makes it “absolutely” true.

• 38 I should note here that while some have questioned whether James could endorse the possibility of c (...)
• 39 This won’t take away from the importance of absolute truth as an ideal that regulates our behavior, (...)

31So what should we say if the “pessimistic” conclusion that James imagines actually turned out to be the case, and the regulative ideal that our use of “true” aspired to was unsatisfiable?38 Of course it would then follow that nothing was “absolutely” true, but it seems less clear that we should conclude that nothing was true. After all, we should, with James, consider the possibility that the failure of absolute truth should lead us instead to conclude that the “absolute” interpretation is not the best account of what we mean by “true.”39

• 40 Peirce (1958 [1908]: 398).
• 41 See Strawson 1962.

33In the pessimistic scenario where absolute truth is unattainable, much the same can be said for “true.” It may be, ultimately, subjective factors that would determine whether we give up on the existence of truth altogether. Some might feel that the absolute conception is so central to our notion of truth that if there were no absolute truth then we should say that there is no truth at all. Indeed, Peirce seems committed to a position something like this when he claims, “I do not say that it is infallibly true that there is any belief to which a person would come if he were to carry his inquiries far enough. I only say that that alone is what I call Truth. I cannot infallibly know that there is any truth.”40 Others might feel that truth-talk is central enough for us that if the ideal of absolute truth proved to be unrealizable, then we would simply need to understand “truth” in a way that doesn’t make absolutist presupposition. I think that James’s willingness to understand truth as “temporary truth” puts him firmly in this latter camp. If you think that absolute truth is a realizable ideal, then it will be natural to think that temporary truth is a pretty shabby candidate for what most of us mean by “true.” However, if, like James, you think that this ideal may fail to be realized, even in principle, then the shabby candidate can begin to look like what we really have been talking about with the term all along. (In this sense, true might be like “free,” where the concept is so central to our practices that we might not want to take on any sort of error theory about its application).41

• 42 See, for instance, Putnam 1981, Rorty 1998.
• 43 Putnam (2017 [2005]: 194).
• 44 For James’s conception of a “live” hypotheses, see James (1979 [1897]: 14).

34Given Putnam and Rorty’s extended debate about just “how far” pragmatism should go,42 it is unsurprising that Putnam would like to argue that both of Pragmatism’s fathers were on his side, and while I’ve argued here that James’s writings about absolute truth hardly put him in Putnam’s camp, James’s position still isn’t quite as radical as Rorty’s suggestion that absolute truth is just an ideal. For James, it is still an empirical question whether inquiry could develop in a way that leads to absolute truth. While he may have presented some reasons for doubting that this sort of consensus can be reached, those considerations are hardly conclusive, and if it did turn out that absolute truth was attainable, it would be natural for that to be the primary way to understand what truth was. Putnam ultimately claims that James takes his account of truth to be a hypothesis,43 and I would argue that both the absolute and temporary readings of truth are very much “live” hypotheses for James.44 How inquiry will ultimately pan out, and thus how truth should ultimately be understood, is something for future inquiries to decide.