ISSN 1393-614X Minerva - An Internet Journal of Philosophy Vol. 9 2005.


Truthmaking and Pragmatist Conceptions of Truth and Reality

Sami Pihlström



This paper argues for a rearticulation of the theory of truthmaking within pragmatism. The concept of truthmaking has usually been employed by metaphysical realists (such as D.M. Armstrong), but it can be reinterpreted in a pragmatist manner, following both classical and more recent pragmatists’ (e.g., William James’s and Hilary Putnam’s) ideas on the “making of truth” (or “worldmaking”) as a process within human experience and world-categorization. Thus, a pragmatist criticism of metaphysical realism can be extended to the core areas of realist metaphysics, including the truthmaking theory.



1.                  Introduction

The concept of truthmaking is usually associated with metaphysical projects of a distinctively realist flavor. The truthmaking theorist typically strives to identify those elements of the world – that is, the real, mind- and language-independent world – that make our propositions (insofar as they are true) true. The truthmaking theory is thus also sometimes regarded as a way of explicating the correspondence theory of truth – an explication of that age-old theory more metaphysical than, e.g., Alfred Tarski’s influential semantic conception of truth. Insofar as the correspondence theory of truth and the idea of the world itself as something that makes our truths true are elements of what has been called metaphysical realism, it might even be claimed that the truthmaking theory is essentially committed to metaphysical realism. Such an impression is strengthened, if one takes a look at influential strongly realist formulations of the notion of truthmaking, such as David Armstrong’s (1997, 2004). Armstrong, obviously, should be taken to be committed to metaphysical realism, roughly in Hilary Putnam’s (1981, 1990) well-known sense (see Pihlström 2006b).


In this essay, I shall examine whether the notion of truthmaking might, however, have a role to play in somewhat less strongly realist views, especially in pragmatism or (what some authors call) pragmatic realism (see, e.g., Putnam 1990; Pihlström 1996, 2003; and several essays in Shook 2003). This suggestion is not as odd as it may first seem. There are metaphysicians who think that the truthmaking theory, as such, is metaphysically neutral. For example, Johanna Seibt (2000) draws a sharp distinction between ontology and metaphysics: ontology, for her, is a theory of truthmakers and remains neutral with respect to the realism vs. idealism debate, because different (e.g., realist or idealist) interpretations of the postulated truthmakers (i.e., the ontological categories the theorist commits her/himself to) are possible. These interpretations are, then, “metaphysical valuations” of the structure(s) of the truthmakers one is ontologically committed to.


My purpose is not to endorse Seibt’s position as such, or to elaborate on the specific truthmakers she postulates, but to use her idea of the metaphysical neutrality of truthmaking theory as a starting-point for an investigation of the way(s) in which pragmatism might, especially in its “pragmatically realist” versions, accommodate some plausible insights of the truthmaking theory. I shall begin, however, by explicating, in broad strokes, a metaphysically realist version of the theory, viz., Armstrong’s. My ideas concerning the relations between truthmaking and pragmatist views on truth and reality will emerge against that background. I make no claims to covering the recently lively truthmaking discussion with any historical accuracy. There are plenty of rival theories available – for example, in addition to Armstrong’s writings, the pioneering work by Kevin Mulligan and his collaborators (1984). (See also the papers collected in Lillehammer & Rodriguez-Pereyra 2003.)


2.                  Armstrong’s theory of truthmaking

The most comprehensive source for Armstrong’s sophisticated theory is his recent book, Truth and Truthmakers (2004), although the concept of truthmaking was already centrally employed in a previous work by him, A World of States of Affairs (1997). The more recent book argues for the general philosophical relevance of the concept of truthmaking, for a particular conception of what the truthmaking relation is (including two key theses, “truthmaker necessitarianism” and “truthmaker maximalism”), as well as for a number of carefully laid-out views about what kinds of truthmakers we need in various areas of metaphysical theorizing (e.g., in our treatment of things and their properties, of modalities, of numbers and other abstract objects, and of natural laws).[1]


First, we should note that the term “making” is not used in a causal sense in the theory of truthmaking – either Armstrong’s or anyone else’s. Rather, Armstrong’s (2004) use of this expression intends to capture the idea that there is something in the world – “some existent, some portion of reality, in virtue of which” a given truth is true (p. 5; emphasis added). Armstrong regards propositions as the (primary) truthbearers (see ibid., pp. 12-16). It is not necessary to dwell on the issue of truthbearers here. Armstrong, as a naturalist and materialist, wishes to provide a fully naturalist, this-worldly account of propositions, treating them as possible intentional objects (and analyzing possibilities combinatorially, thus avoiding the postulation of real possible worlds), but of course we need not stick to his analysis. Pragmatists speak more variously about truthbearers. As we shall see, for William James, truths are primarily true beliefs or true “ideas” (whatever that ultimately means); yet, I have nothing against Armstrong’s and many others’ way of speaking about propositions as the content of such beliefs or ideas, provided that an inflationary ontology of propositions as necessarily existing abstract entities is avoided. More recent pragmatists, such as Richard Rorty, have typically regarded linguistic entities (e.g., sentences) as those things that can be said to be true or false. (See also Kirkham 1992.)


Allowing Armstrong’s commitment to propositions, the truthmaking relation, for him, is a relation obtaining between a true proposition and the “portion of reality” in virtue of which it is true, or which makes the proposition true. This relation is a “cross-categorial” one, because it holds between entities belonging to different ontological categories, i.e., between a proposition and the entity or entities that make(s) it true (Armstrong 2004, p. 5). Furthermore, truthmaking is an “internal” relation in the sense that “given just the terms of the relation, the relation between them is necessitated” (ibid., p. 9). Such a relation is “no addition of being”, nothing over and above the terms themselves (i.e., in this case, the proposition which is true and the truthmaker that makes it true), although this, according to Armstrong, is not to say that the relation does not exist.


It is somewhat puzzling, however, that Armstrong (ibid., p. 37) also calls the “truth/truthmaking relation” a “semantic” one. It would be more natural for him to say (as he at least implicitly seems to maintain) that truth is a semantic notion while truthmaking is its metaphysical ground. In his introduction to a volume of essays on realism and antirealism, William Alston (2002, p. 4) explains this idea by saying that “the issue over the nature of truth is logically independent of the metaphysics of facts, of truth makers” – even though a realist metaphysics of facts is the natural choice for someone who holds a realist conception of truth (the “alethic realist”; cf. Alston 1996). While Alston thinks that alethic and metaphysical realisms are independent, René van Woudenberg (2002) argues in the same volume that a realist conception of truth involves metaphysical commitments, in general to mind-independent facts as truthmakers, and more particularly to the existence of a mind-independent, non-relational, and non-dispositional property of being true. (Such a property might be invoked as the truthmaker of the truth of alethic realism, if such a position is true.)


Among Armstrong’s more controversial theses, truthmaker necessitarianism declares that the truthmaking relation is a necessitating relation: the truthmaker necessitates the truth which is true in virtue of it. If the truthmaker exists, then the corresponding truth cannot fail to be true. Of course, necessitation in this metaphysical sense must be carefully distinguished from necessary truth; the truthmaker of a contingent truth exists only contingently, but if it does exist, then that truth must be true. Furthermore, this necessitation of a truth by its truthmaker, because of the cross-categorial nature of the relation, is not to be misconstrued as logical entailment, which could obtain only between propositions (Armstrong 2004, pp. 5-6).


Another central thesis is truthmaker maximalism, according to which every truth has a truthmaker. Armstrong admits that he does not have any direct argument for maximalism, but hopes that “philosophers of realist inclinations will be immediately attracted to the idea that a truth, any truth, should depend for its truth for something ‘outside’ it, in virtue of which it is true” (ibid., p. 7).[2] He does show in great detail how (in his view) plausible truthmakers can be found for truths about various different things in different ontological categories.


Armstrong defends, and relies on, yet another meta-level or methodological thesis, the entailment principle, which says that if a truthmaker makes a certain proposition true, and if that proposition entails another proposition, then the same truthmaker also makes the entailed proposition true (ibid., pp. 10 ff.). This principle plays a key methodological role in Armstrong’s argumentation, but we need not pay further attention to it here. Another central concept that we can bracket here is the concept of a minimal truthmaker – a truthmaker for a proposition such that nothing can be subtracted from the truthmaker in such a way that it would still serve as a truthmaker for that proposition (ibid., p. 20). For ontologically parsimonious truthmaker theorists, the search for a minimal truthmaker of a truth is, understandably, an extremely important task.


On the grounds of his truthmaker theory, presented here only in a barest outline, Armstrong arrives at what he declares to be a realist theory of truth: “p (a proposition) is true if and only if there exists a T (some entity in the world) such that T necessitates that p and p is true in virtue of T” (ibid., p. 17). In the earlier book, Armstrong (1997, p. 128) noted that while the semantics of the truth predicate might be accounted for in terms of a redundancy theory, the correspondence theory must be accepted on a “deeper, ontological level”, that is, as the requirement of providing truthmakers for truths. The realist definition Armstrong gives is not an explicit definition (because, as such, it would be circular), but it is obviously designed to capture the intuitions of realist correspondence theorists who hold that truths must correspond to the way the world is. I think it clearly very well succeeds in this task. The correspondence relation, as such, is often thought to be vacuous or metaphysically mysterious. The truthmaking relation explicates what this relation comes down to, at least for the metaphysical realist who believes that we can make sense of the metaphysical relation of necessitation. Truths should not only correspond to the world, but it should be the world itself – the entities that really exist – that makes those truths true, or that plays the role of something in virtue of which they are true. The direction of determination goes from the largely mind-independent world (the entities there really are) to the truths about it (propositions).[3]


3.                  Pragmatists on truthmaking (and worldmaking)

Turning now to pragmatism, we of course find metaphysical (and anti-metaphysical) views very different from Armstrong’s, who, after all, pays almost no attention at all to pragmatism – not even by criticizing it. Yet, the notion of truthmaking, in some sense at least, can be perceived to be at work in, or at least read into, a number of key pragmatist texts.


In particular, William James (1907, 1909a) famously argued for a dynamic conception of truth according to which truths are “made” rather than found (for more detailed discussions, see Pihlström 1996, 1998, 2004b). More precisely, truths are made in the course of human experience. Truth “happens to an idea” (James 1907, p. 97), instead of being a timeless, abstract, unchanging relation eternally obtaining between a true idea and something that exists independently of it (and which the truth, according to the most naive versions of realism, would simply “copy”). Thus, ideas or beliefs (or other truthbearers) are not eternally true but “become true just in so far as they help us to get into satisfactory relation with other parts of our experience” (ibid., p. 34). “Truth is made, just as health, wealth and strength are made, in the course of experience” (ibid., p. 104). Even the regulative notion of “absolute truth” designates something that “will have to be made, made as a relation incidental to the growth of a mass of verification-experience, to which the half-true ideas are all along contributing their quota” (ibid., p. 107).


In passages like this, the notion of truthmaking is at least metaphorically employed – though of course within an overall philosophical position very different from Armstrong’s metaphysical realism. James, seeking the “cash-value” of truths “in experiential terms” (see ibid., p. 97), has little patience with the realist view, held in James’s time by Bertrand Russell, among others, that truths simply correspond to or are made true by objects or states of affairs that exist ontologically independently of those truths. Instead, the making of truths in and through experience also amounts to the making of reality: the world – as experienced and understood by us humans – is a human construction, at least up to a point (cf. Pihlström 1996, ch. 3; 1998, ch. 1). The direction of determination is not, or not merely, from the world to the true beliefs that are about it; human subjects of (true) belief, with their pragmatic needs and habits of action, partly determine what the world is like. This view was developed in a more radical manner by James’s younger pragmatist contemporary F.C.S. Schiller, whose 1907 collection, Studies in Humanism, contains essays with titles such as “The Making of Truth” and “The Making of Reality” (for some comparisons between James and Schiller, see again Pihlström 1998, ch. 1).


Let us – turning from the concept of truth to the other pole of the truthmaking relation, the concept of reality (or the world) – take a look at a few other key passages in which James and later pragmatists employ the concept of “making”. In his late work, A Pluralistic Universe (1909b, pp. 117-118), James claims, with Henri Bergson, that “[w]hat really exists is not things made but things in the making” and talks about “put[ting] yourself in the making” of things “by a stroke of intuitive sympathy” (emphases in the original). In this sense, reality is something we (help to) create, to which we always inevitably make an “addition” – in a word, something “still in the making” (James 1907, pp. 121-123). Anything we may call a thing is “carved out” by us (ibid., p. 122). According to Schiller, again, the “pragmatically real world”, far from being a ready-made, “original datum” of experience, is “an elaborate construction, made by us, individually and socially, by a purposive selection of the more efficacious, and a rejection of the less efficacious portions of a ‘primary reality’” (Schiller 1907, p. 460). Schiller argued, in a manner close to James’s, that as knowers of reality we inevitably alter, and thus contribute to the making of, reality. The unmade “primary reality” is, as such, nothing for us, but it is turned into a pragmatically meaningful, humanly significant reality possessing a certain structure (only) through the “humanist” process of “making”. (Cf. Schiller 1907, pp. 13, 186-187, 198-202, 216-223, 425-426, 433, 439, 444, 474.)


Here, of course, the relevant “making” relation is not the Armstrongian realist truthmaking relation obtaining between a true proposition and the independent piece of reality which necessitates its truth. Yet, the “making” of reality that James and Schiller talk about seems to be something that the “world” itself does, too, at least insofar as we, as knowers of reality, are ourselves parts of the world or reality that we help to structure. Truth- or worldmaking would not be possible without our contribution to the way(s) the world is, but we are undeniably parts of the world as well. The world can thus be said to make truths true, but only through us, for whom there is a world (or truth) in the first place. There is, as Schiller puts it, a primary reality independently of us, but that reality makes nothing true; it is a mere abstraction of thought, or, in James’s words, something “absolutely dumb and evanescent, the merely ideal limit of our minds” (James 1907, p. 119). Even so, James did not deny the existence of a “standing reality independent of the idea that knows it” (James 1909a, p. 88). Clearly, he wanted to save the minimal rational core of realism (and the correspondence theory), while defending a picture of truth richer and pragmatically more nuanced than the one he found in his realist rivals’ theories.


From these pragmatists’ point of view, one might in any case argue that if Armstrongian metaphysicians regard that kind of a metaphysically primary reality as the source of truthmakers, they erroneously claim something about a reality which has, and can have, no structure (for us) at all. Any structured reality – any reality whose entities could make anything true – must, according to pragmatism, be a humanly categorized reality. All categorization, moreover, arises as a response to specific needs (to be further analyzed along pragmatist lines); accordingly, if the world or reality is to make any truths true, it must already have been “made” (and must be continuously remade) by us world-categorizing concept-users whose categorizing activities are themselves constrained by and oriented toward certain human needs, goals, interests, or purposes. The world which makes our human truths true is itself continuously “in the making”. It is, to use Deweyan rather than Jamesian terms, a reality with a “practical character”.


This kind of experiential and practice-embedded making of truths (and indeed of the world itself) was not emphasized by the father of pragmatism, Charles S. Peirce, who on the whole defended views much more realistic than either James or the other pragmatists of those days (cf. Pihlström 2003, ch. 3). Something similar was, however, emphasized by John Dewey, who even wanted to replace the traditional talk about truth with talk about “warranted assertibility” – something that is certainly dependent on our human points of view to the world. Later pragmatists, including Hilary Putnam and Nelson Goodman (among whom the latter never wanted to describe his views as pragmatist, though), have more explicitly revived the idea of truthmaking, or (in Goodman’s case, more radically) worldmaking. Truths and the world (or, in Goodman’s pluralist and relativist theory, worlds) they are true about are, according to these neopragmatists, human constructions, “made” by us in the course of our experiential and symbolic ways of representing and acting.


Goodman, in particular, is explicit about the metaphor of making – or perhaps it is not even correct to refer to it as a “metaphor” at all. He claims to move from “a world fixed and found”, with its unique truths, to a diversity of right, even conflicting, world-versions and “worlds in the making” (Goodman 1978, p. x). We make worlds by making right versions through our various symbol systems. Truth itself is world- or version-relative: different worlds, different truths. Note that this is not to say that the world, or the particular version of the world at issue, does not make true the truths that hold in that world. But note also that this discussion of the plurality of worlds must, as Goodman explicitly urges, be distinguished from the debates over the plurality of possible worlds (cf., e.g., Lewis 1986): the worlds or versions we make are actual. Yet, our making of worlds or truths is not a creation ex nihilo: making is “remaking” (Goodman 1978, pp. 6-7). But the crucial point is that this is making nonetheless, even in cases in which it seems that we simply “find” what is “out there”, e.g., the stars, which seem to be paradigmatic examples of objects existing independently of our beliefs, perceptions, language-use, theories, or symbol systems. “We have to make what we find, be it the Great Dipper, Sirius, food, fuel, or a stereo system.” (Goodman 1984, p. 36.) As we just saw James expressing these ideas, we must “carve out” whatever we conceive of as an object, or a thing of any kind, even the stars. We are, among other things, “starmakers” (cf. also James 1907, p. 121; McCormick 1996).


While bringing up, particularly in his later writings, some reservations about Goodman’s theory going too far in the direction of antirealism or “irrealism” (Goodman’s own term), sacrificing the empirical reality of objects such as stars (cf. Putnam 1992), Putnam – in an earlier incarnation – also joins in here by claiming that “‘objects’ themselves are as much made as discovered, as much products of our conceptual intervention as of the ‘objective’ factor in experience” (Putnam 1981, p. 54). The reality we live in and about which our statements can be true is a reality with a “human face” (Putnam 1990). Putnam’s more recent writings (e.g., his 1994, 1999) are filled by defenses of commonsense realism and, along with it, the notion of the independence of worldly objects, but he has not given up the pragmatism which already to some extent gave flavor to his (now rejected or at least heavily reinterpreted) internal realism (see also Pihlström 1996, 1998, 2004a).


Several realist critics still read Putnam as an antirealist (cf., e.g., a number of essays in Alston 2002). This can hardly be correct, given Putnam’s recent rethinking of his views on realism. In fact, Putnam has occasionally revived a surprisingly realist-sounding concept of truthmaking (without explicating it in any analytic detail, though), to which we will return in a moment. Although Putnam rarely uses this term, we might read his continuing criticisms of minimalist and disquotational theories of truth, including theories preferring assertibility conditions to truth conditions, as attacks on the tendency of these theories to give up what is sound in the notion of truthmaking (see Putnam 1994, 1995, 1999). Thus, Putnam wants to save the realist conviction that something non-linguistic is needed to account for the truths we are able to formulate in our languages. Given his insistence on the need to invoke something like the truthmaking relation, there is a distinction to be drawn between those pragmatists who find the concept of truthmaking important (in a way or another, though presumably never in Armstrong’s strongly realist way) and those who do not. The former camp includes both Goodman (if we incline to think that worldmaking is a kind of truthmaking) and Putnam, while the latter one includes, arguably, philosophers like Richard Rorty and Donald Davidson (whom Rorty notoriously claims to be a pragmatist of sorts). However, if we interpret Davidson’s (1984, p. 194) statement that nothing in the world “makes” our sentences true simply as a denial of the metaphysically-realist idea that some fixed objects or properties in the non-linguistic world in itself make our true sentences (or propositions) true, or stand in one-to-one correspondence to them, then it is as plausible as the general rejection of metaphysically-realist construals of truthmaking and the correspondence theory of truth is, according to Jamesian or Putnamian pragmatists. Note that Davidson, as a disciple of Quine, speaks about the truth of sentences rather than propositions. Again, I make no claims about what the true truthbearers are. In any case, Davidson (1990, especially pp. 303-304) is very critical of the correspondence theory (see also, however, Niiniluoto 2004, p. 58).


However, if these claims by Davidson (and Rorty) are interpreted in a way which makes pragmatic realism itself seem suspect, they are more vulnerable to Jamesian or Putnamian pragmatist criticism. If there is no relation at all between our sentences or propositions and the world they are supposed to be about (and about which they are, in lucky cases, true), then little remains from even the mildest form of pragmatic realism. Something like truthmaking is needed in any pragmatic realism worth the name, unless one wants to go all the way toward Rortyan antirepresentationalism – a view which most pragmatism scholars regard as an extremely controversial construal of what is essential in pragmatism (cf. Pihlström 1996, 1998).


As already indicated, Putnam, while occasionally speaking metaphorically about “the mind and the world jointly mak[ing] up the mind and the world” (Putnam 1981, p. xi), actually embraces the idea of truthmaking, at least in his relatively recent writings. It is also worth noting that, as he points out, a version of this idea is actually present in Davidson’s seminal 1974 paper attacking conceptual relativism:


[…] in the very last sentence of [“On the Very Idea of a Conceptual Scheme”, p. 198] Davidson speaks of “the familiar objects whose antics make our sentences and opinions true or false.” This remark would seem to suggest that Davidson doesn’t mean to deny that objects, be they near or be they distant from us in space and time, sometimes make our sentences true. (If he had, I should now be quarreling with Davidson, as well as with Rorty.) Rather, Davidson’s point seems to be that we must not think that true sentences correspond one by one to special objects which “make them true,” call them “states of affairs” […]. (Putnam 1994, p. 300; emphasis in the original.)


That is, Putnam should be quarreling with Davidson, had Davidson denied the notion of truthmaking altogether – which, Putnam seems to be saying, he never did (as should be evident to anyone who has read the final sentence of Davidson’s paper). Although Putnam goes on to say that he disagrees with Davidson on the claim that it is “verboten to speak of ‘states of affairs’ at all” (ibid., p. 301), because the category of states of affairs is among the ontological categories we may legitimately use (with some definite pragmatic purposes in mind) within our various ontological structurings of reality, we can perceive here a profound affinity between these two thinkers’ views. The real opposition is, rather, to be found between Putnam and Rorty (including Rorty’s employment of allegedly Davidsonian ideas), because Rorty, unlike Putnam, rejects the very idea of things or the world (or anything, for that matter) making truths true (see ibid., p. 300). Unfortunately, we must leave the details of the Putnam–Rorty comparison aside here (cf. Pihlström 2004a). In a word, for Rorty, there are only causal – flatly physical – relations between language and the world, no semantic or representational (let alone metaphysical) ones. In any event, pace Rorty and Davidson, the pragmatically acceptable idea of truthmaking should be carefully distinguished from the naive view that our sentences or propositions correspond “one-by-one” to objects or states of affairs in the world. Neither metaphysically realist truthmaking theorists like Armstrong, nor their pragmatist rivals, should accept such a notion of correspondence.


In criticizing Rorty and Davidson, Putnam comes surprisingly close to the kind of view Armstrong holds. (Similar commitments to facts or states of affairs as truthmakers can be found, for instance, in Putnam’s 1994 Dewey Lectures, reprinted in Putnam 1999.) However, I think it is clear that Putnam would never join Armstrong in claiming that the world an sich consists of states of affairs or that it could be described as consisting of states of affairs from a privileged (“absolute”, “God’s-Eye”) perspective. Furthermore, Putnam’s qualified agreement with Davidson by no means entails that he would find Davidson’s criticism of conceptual relativism appealing; on the contrary, he has repeatedly defended what he calls conceptual relativity, as well as the doctrine of pluralism partly adopted from the classical pragmatists (cf. Putnam 1990, 1999, 2004; for several recent critical discussions of these ideas, see the essays in Alston 2002; see also Pihlström 2006a). In his treatment of Davidson just cited, Putnam (1994, p. 301) goes on to remind us that we should not ask “whether states of affairs ‘really exist’ or not” (this, apparently, would commit us to metaphysical realism) but “whether notions like ‘state of affairs’ are to be conceived of as having a single determinate meaning or whether they are more happily conceived in terms of an open and ever extendable family of uses”. Putnam’s conceptual relativity would, obviously, suggest the latter. These issues cannot be settled here, though. Among Putnam’s critics, Niiniluoto (2004, p. 70) claims that a realist committed to THE WORLD (“in itself”, as it were) can accept “conceptual pluralism”, but a Putnamian pragmatist or pluralist can hardly find this attempted via media plausible (for more details, see Pihlström 1996, 1998).


Putnam’s views on truthmakers extend to his criticism of not only Davidson and Rorty but of Quine’s “immanent” conception of truth (as something within language) as well:


In my view, whatever authority [ontology or epistemology] had depended entirely on our conceiving of reality and sensations as, respectively, the makers-true and the makers-justified of the sentences we produce – not the makers-true and the makers-justified from within the story, but the things outside the story that hook language onto something outside itself. If Quine is right, then Rorty is right. If the makers-true and the makers-justified have really turned out to be “immanent,” then they have also become “optional.” Quine’s talk of what is “first-class” and what is “second-class” becomes mere rhetoric. (Putnam 1994, p. 348.)


Putnam here clearly implies that, in his view, ontology is, or has traditionally been conceived as, a matter of providing “makers-true” for true sentences – truthmakers that are to be found, in typical cases at least, outside language and not merely (immanently) “within the story”, whether scientific (à la Quine) or non-scientific (à la Rorty). He does not explicitly commit himself to such a project of ontology; on the contrary, his more recent work constitutes a devastating critique of the very idea of ontology (see Putnam 2004; Pihlström 2006a). But he does distance himself from the Quinean-Rortyan attempt to turn ontologizing merely immanent.


Among recent pragmatists, then, Putnam can, with some qualifications, be reinterpreted as a truthmaker theorist of some sort, and it would be interesting to see his explicit reactions to the theory that Armstrong has put forward. On the other hand, Armstrong’s specific views will undoubtedly strike Putnam as too heavily metaphysically realist. A fruitful reinterpretation of those views in terms of pragmatism would be needed to make them fit Putnam’s looser, somewhat metaphorical notion of truthmaking. In attempting such a reinterpretation, we may creatively employ the views of earlier pragmatists like James, Schiller, and Goodman, for whom the metaphor of making is, arguably, more central than it is for Putnam but who are, from the perspective of pragmatic realism, frequently led to unacceptably antirealist (constructivist) ontologies.


4.                  Metaphysically realist truthmaking versus pragmatically realist truthmaking

We have seen how pragmatists can (though of course need not) employ a concept of truthmaking (perhaps in several variations), although such a concept is, in pragmatist theories, usually employed within an overall metaphysics very different from the metaphysical realist’s one – a more processual, dynamic, practice-embedded metaphysics that refuses to draw any principled dichotomies between the world as it is in itself and the world as it is experienced by humans engaging in their unending diversity of practices and habits of action. Such a metaphysics can still be realist, though only pragmatically realist, insofar as it does not reject the claim that there is something “out there” that we never made up out of nothing. Pragmatists, emphasizing our need to struggle with the concrete facts of worldly existence, should be the first to note that truthmaking is not, naively, a “making up” of truths. We live in a concrete, natural world, shaping the world through our experiences, and it is this very world, to some extent shaped by us but irreducible to our contingent and changing structurings of it, that makes true whatever we claim to be true, insofar as our claims to truth are (pragmatically) justified or “right”. Goodman (1978, 1984), among others, finds rightness a more fundamental concept than truth. More generally, pragmatists from James onwards have conceptualized truth as a species of the good, thus linking epistemic and ethical considerations (see also Pihlström 2005).


We should now perceive, further, that pragmatism is not only entitled to the vocabulary of truthmaking but can accommodate much of what Armstrong (2004) says, including the ontological commitments to universals (consider Peirce’s defense of the reality of “generals”) and states of affairs – though only in a reinterpreted sense. These are commitments to the need (a deeply human need) to postulate certain categories within a humanly categorized reality, a world which receives an ontological structure through human practice-embedded categorization (cf. Pihlström 1996, 1998, 2003, 2006a). What we cannot accept, however, is Armstrong’s Russellian views about the need to postulate “totality states of affairs” – except for local totalities or wholes that are never meant to cover the totality of the world as such. There is no “totality” of the world itself, because such a totality is, very simply, never pragmatically structurable or categorizable by us humans. Kant (1781/1787), in his famous Antinomies, attacked precisely the rationalist metaphysicians’ claims to know something about the world as conceived as a totality, e.g., whether the world has spatio-temporal boundaries or not. Putnam’s rejection of “a definite Totality of All Real Objects” (see, e.g., Putnam 1994, p. 303) is equally firm. At best, we may speak about the world as a totality or as a whole as an abstraction, not very different from Kant’s “thing in itself”, that we may pragmatically need as a reminder of the fallibility of our knowledge claims, which are always inevitably presented from local human perspectives.


Armstrong himself (see his 2004, chs. 5-6) believes that totality states of affairs or “general facts” are needed as truthmakers for negative and general truths (among which the latter are a special case of the former). This postulation is motivated, or even required, by his truthmaker necessitarianism. If the world is a contingent totality of states of affairs (let us grant this to Armstrong for the sake of argument), then it seems to be clear that this contingent totality, all by itself, is not sufficient as the truthmaker of the truth that these and only these states of affairs constitute the world. It does not necessitate the truth. This is precisely because the world, as contingent, might have been “smaller” or “bigger”. What is needed in addition to the first-order states of affairs themselves is the totality state of affairs of those states of affairs’ being all the states of affairs there are, a general fact that “collects” them (cf. ibid., p. 59). Among such general facts, which may be more or less local and restricted (e.g., the truthmaker for the truth about there being a definite totality of chairs in this room), there is “a huge general fact (state of affairs) that is the most general state of affairs of all” (ibid.). This “biggest totality state of all, the one embracing all lower-order states of affairs”, is all that is needed to provide truthmakers for lesser totality truths and negative truths (ibid., p. 75; see also Armstrong’s earlier treatment of totality states of affairs in his 1997).


Now, while truthmaker-based metaphysical theories about universals, states of affairs, or even about the way in which possibilities can be constructed out of recombinations of the elements of the actual world (cf. also Armstrong 1997), can be rearticulated in a pragmatist setting, always making it explicit that we are talking about an empirical world whose ontological structuring is not ready-made or given in advance but results from a human practice-embedded categorization, the metaphysics of “allness”, or totality, that Armstrong (2004, ch. 6) is committed to through his account of the truthmakers for negative and general truths cannot be thus rearticulated (or so it seems to me).[4] This metaphysics more directly presupposes not only that it makes sense to speak about the ontological structure of the world as it is in itself but also that there is such a thing as the world as it is in itself, as the given contingent totality it is. This idea contrasts with the Jamesian-Schillerian-Goodmanesque view of the world(s) being constantly “in the making”. According to these pragmatists (and even according to their somewhat more realistically inclined followers, such as Putnam), there is no such thing as the totality of the world, as “ready-made”; moreover, the problem is not just that there, contingently, is no such thing but that we do not know what we are talking about when we make claims about there being such a thing. Our claims to know, or even meaningfully state, something about the totality of the world, about all the objects there are, or about all the states of affairs that obtain, have only the illusion of sense, if we take seriously the pragmatists’ emphasis on the world’s being in the making in and through human experience. This issue would require more detailed attention, but we may here safely rely on Putnam’s (1981, 1990, 1994, 2004) arguments for conceptual relativity (for further discussion and engagement with secondary literature, see again Pihlström 1996, 1998, 2003, 2006a; and Alston 2002).


The pragmatist, I have argued, has a right to the truthmaking vocabulary and even to some specific doctrines one may arrive at through truthmaking considerations (though only in a heavily reinterpreted sense). It is important to distinguish truthmaking from metaphysical realism, thus saving this useful metaphysical concept from scientistic metaphysical realisms like Armstrong’s. Such realists urge that we do not in any sense “make” or “create” the truthmakers that make our propositions true: “They are just there or not there, whatever our opinion about them is.” (Armstrong 2004, p. 33.) Just so – at the empirical level, as we might say. But at the more metaphysically relevant level of the very categorizing of reality (at a transcendental level, as we may call it, if we are not afraid of Kantian notions), things are not so simple. Can we simply claim that, e.g., universals instantiated in particulars (such instantiations thereby forming states of affairs) in the world in itself make our predicative statements (or, rather, the propositions those statements express) true? For a pragmatically inclined thinker, such a claim would be a monstrous example of a commitment to an underlying, ready-made structure of reality as such, a structure in virtue of which, mysteriously, our statements (or the propositions they express) would be true (or false). Inasmuch as pragmatists have found the traditional correspondence relation mysterious or unhelpful, the same charge should be directed against the truthmaking relation – if construed in a metaphysically realist fashion in Armstrong’s manner. In order to be able to make any proposition true, the world, as suggested above, must first be “made” (and “remade”) by us, turned into a meaningfully categorized shape. Only as thus categorized can it make any proposition true. The notion of making has no meaning here unless it is a relation obtaining between identifiable items; any identification, in turn, requires human conceptual categorization.[5]


Moreover, Armstrong’s metaphysically realist theory of truthmaking can be subjected to a self-reflective critical analysis. Suppose his theory (and the metaphysical realism that functions as its base) is true. What is the truthmaker for this truth (in the theory’s own terms)? Perhaps the truthmaker is whatever makes true the modal truth that, necessarily, a truthmaking relation obtains between any truth and the piece of world in virtue of which it is true. Then a truthmaker for this modal truth could be provided in the way Armstrong (2004, ch. 8) proposes. But why should the truthmaking theory be necessarily true? Could we have any legitimate reasons for claiming that it lies in the necessary, unchanging structure of things that truths are made true by truthmakers? This kind of a position would hardly be preferable to a Platonic metaphysics of immutable forms. Rather, as a pervasive feature of the fully contingent natural world (to which even propositions belong, on Armstrong’s deflationary construal), the truthmaking theory should, one might claim, be a contingent truth, or a true generalization, perhaps comparable to laws of nature. Yet, this is hardly plausible, either. The truthmaking theory is a piece of analytic ontology, not a scientific theory. It is hard to see how it could be scientifically evaluated or tested; it has no clear empirical content. On the other hand, it is no mere convention; according to Armstrong, it is supposed to make a great difference what kind of metaphysical views a person holds.


Thus, we observe that the status of Armstrong’s own views is unclear and problematic. It is not at all obvious how his metaphysical findings, including the truthmaking theory itself, are related to our advanced (and fallible) scientific theories (whose truths, he supposes, we should also provide truthmakers for). The same troubles haunt all those philosophers who, like Armstrong, seek to combine metaphysical theorizing about ultimate reality with a naturalistic respect for the advancement of science. On the most general level, the critic may ask what (according to metaphysical realists) makes metaphysical realism itself true. Is it the world itself, or whatever is the case? This answer can hardly be non-question-beggingly defended against the critic of metaphysical realism. Could future science show, for instance, that the theory of truthmaking is false, or that metaphysical realism is false, or that a factualist ontology of states of affairs is false? Are these questions even meaningful, according to an ontologist like Armstrong – and if not, what is the criteria of distinguishing meaningful questions from meaningless ones here (see also Pihlström 2006b)?


In my view, it is much more plausible and straightforward, as well as pragmatically better justifiable, to interpret the truthmaking theory as a human-made truth (if, indeed, we can speak about its truth at all), as a pragmatically efficacious way of analyzing the relation between the elements of the (empirical, cognizable) world and the things we (can) say about that world. This theory is, then, something that our activities, in particular our ontological theorizings, by means of which we categorize the world (always ultimately guided by pragmatic considerations), make true. Without such a pragmatic background, the truthmaking theory (or any general ontological view) is hardly even a candidate for truth or falsity. For the pragmatist, moreover, there is no problem with a vicious regress here: the pragmatic theory of truth, as James (1909a, ch. 8) pointed out, is itself (hopefully) a pragmatically satisfactory account of the concepts of truth and reality. The same can be expected to hold about any pragmatic “theory” of any philosophically significant concept, including the concept of truthmaking, insofar as such a theory can be legitimately called “pragmatic”. This is a further consideration in favor of pragmatism, to be added to the unclarity and sheer implausibility of the metaphysical realist’s postulation of mysterious cross-categorial necessitating relations (etc.) in order to explain our perfectly ordinary concepts of truth and the world. Unlike metaphysical realism, pragmatism is a philosophical interpretation of the concepts of truth and reality that can be readily applied to itself, so as to account for its own truth.


It is, then, the pragmatic efficacy of our beliefs about the world that makes (in a broadly interpreted sense) those beliefs true, by making us able to orientate in the world they are about. This is (pragmatically!) true about our “first-order” beliefs as well as about our “higher-level”, meta-level beliefs about what it means to say that pragmatism is a true account of truth(making). We might consider this proposal a partial vindication of James’s pragmatist conception of truth and reality, without any commitment to his specific (often misunderstood) doctrine of the utility or satisfactoriness of true beliefs.


5.                  Conclusion

I have suggested above that we can retain our ordinary ontological commitments to whatever truthmakers we pragmatically need for our truths – e.g., states of affairs, properties (whether construed as universals, as tropes, or as something different), and possibly even modalities – within a pragmatist ontological framework which carefully avoids the dubious claim that the world as it is in itself, entirely independently of us, is ontologically prestructured so that it could (even in principle) be described from a God’s-Eye-View when described in terms of these (or any other) truthmakers. The notion of truthmaking, as well as the ontological method of providing truthmakers for propositions we take to be true, may itself be a rather extraordinary innovation, remote from our everyday practices, but the ontological commitments our use of this notion (and method) yields may be perfectly ordinary, pragmatically valuable ones. In this manner, even a qualified agreement with an ontological scheme such as Armstrong’s is possible for the pragmatist who insists on the pragmatic core of each and every ontological commitment.


Arguably, pragmatism is even better equipped for truthmaking theorizing than metaphysical realism is, because of pragmatism’s dynamic view of truth as a process within human experience. The metaphilosophical moral to be drawn from all this is that the sharp dichotomies between metaphysics (or ontology) and epistemology one finds in authors like Armstrong – and in many other places as well, e.g., in a number of contributions to Alston (2002) – are unacceptable. Truths are made precisely because we make them in and through our experience, which is itself in and of the natural world we live in. This is a circular picture, of course, but hardly viciously so; more generally speaking, pragmatism cannot avoid circularity, insofar as its own commitments are to be pragmatically evaluated.


As is well known, pragmatism has been put forward as a serious alternative to a number of traditional philosophical positions and methodologies in several different fields of philosophy, e.g., in epistemology, the philosophy of science, ethics, political philosophy, and elsewhere. The success of pragmatism in these various fields seems to be based on its versatility. Philosophers of widely different persuasions, favoring wildly different substantial views, can hold on to pragmatist principles, often modifying them for their own purposes. James (1907, ch. 2) compared pragmatism to a corridor leading to an indefinite number of rooms in which one can find very different intellectual work being done. It remains to be seen whether pragmatism can make its way into the core areas of contemporary (analytic) metaphysics, such as the theory of truthmaking, which may seem to be unpragmatist by definition but which need not be – if the arguments put forward in this paper are on the right track. No detailed pragmatist theory of truthmaking has (yet) been formulated here, but perhaps the groundwork for such a theory has been done by making it clear that the pragmatist need not be afraid of using that concept.





I am grateful to my metaphysically (and realistically!) inclined colleagues (in particular, Prof. Ilkka Niiniluoto and Dr. Heikki J. Koskinen) for several discussions surrounding the topics of this paper, and to Prof. Charbel Niño El-Hani for first suggesting to me that an investigation of a pragmatist theory of truthmaking might be worthwhile.






Alston, William P. 1996. A Realist Conception of Truth. Ithaca, NY & London: Cornell University Press.


Alston, William P. (ed.) 2002. Realism and Antirealism. Ithaca, NY & London: Cornell University Press.


Armstrong, D.M. 1997. A World of States of Affairs. Cambridge: Cambridge University Press.


Armstrong, D.M. 2004. Truth and Truthmakers. Cambridge: Cambridge University Press.


Chrudzimski, Arkadiusz. 2002. “Objects, Properties and States of Affairs: An Aristotelian Ontology of Truthmaking”, Axiomathes 13, 187-215.


Cox, Damian. 1997. “The Trouble with Truth-Makers”, Pacific Philosophical Quarterly 78, 45-62.


Davidson, Donald. 1984. Inquiries into Truth and Interpretation. Oxford: Clarendon Press.


Davidson, Donald. 1990. “The Structure and Content of Truth”, The Journal of Philosophy 87, 279-328.


Goodman, Nelson. 1978. Ways of Worldmaking. Hassocks: Harvester Press, 1992.


Goodman, Nelson. 1984. Of Mind and Other Matters. Cambridge, MA & London: Harvard University Press.


James, William. 1907/1909a. Pragmatism: A New Name for Some Old Ways of Thinking (1907) and The Meaning of Truth: A Sequel to Pragmatism (1909a). One-volume edition. Eds. Frederick H. Burkhardt, Fredson Bowers & Ignas K. Skrupskelis. Cambridge, MA & London: Harvard University Press, 1978.


James, William. 1909b. A Pluralistic Universe. Eds. Frederick H. Burkhardt, Fredson Bowers & Ignas K. Skrupskelis. Cambridge, MA & London: Harvard University Press, 1977.


Kant, Immanuel. 1781/1787. Kritik der reinen Vernunft. Ed. Raymund Schmidt. Hamburg: Felix Meiner, 1990.


Kirkham, Richard L. 1992. Theories of Truth: A Critical Introduction. Cambridge, MA & London: The MIT Press.


Lewis, David. 1986. On the Plurality of Worlds. Oxford: Blackwell.


Lillehammer, Hallvard & Rodriguez-Pereyra, Gonzalo (eds.). 2003. Real Metaphysics: Essays in Honour of D.H. Mellor. London & New York: Routledge.


McCormick, Peter J. (ed.) 1996. Starmaking: Realism, Anti-Realism, and Irrealism. Cambridge, MA & London: The MIT Press.


Mulligan, Kevin, Simons, Peter & Smith, Barry. 1984. “Truth-Makers”, Philosophy and Phenomenological Research 44, 287-321.


Niiniluoto, Ilkka. 1999. Critical Scientific Realism. Oxford: Oxford University Press.


Niiniluoto, Ilkka. 2004. “Tarski’s Definition and Truth-Makers”, Annals of Pure and Applied Logic 126, 57-76.


Parsons, Josh. 1999. “There is No ‘Truthmaker’ Argument against Nominalism”, Australasian Journal of Philosophy 77, 325-334.


Pihlström, Sami. 1996. Structuring the World: The Issue of Realism and the Nature of Ontological Problems in Classical and Contemporary Pragmatism. Acta Philosophica Fennica 59. Helsinki: The Philosophical Society of Finland.


Pihlström, Sami. 1998. Pragmatism and Philosophical Anthropology: Understanding Our Human Life in a Human World. New York: Peter Lang.


Pihlström, Sami. 2003. Naturalizing the Transcendental: A Pragmatic View. Amherst, NY: Prometheus/Humanity Books.


Pihlström, Sami. 2004a. “Putnam and Rorty on Their Pragmatist Heritage: Re-reading James and Dewey.” In Elias L. Khalil (ed.), Dewey, Pragmatism, and Economic Methodology. London & New York: Routledge, pp. 39-61.


Pihlström, Sami. 2004b. “Peirce’s Place in the Pragmatist Tradition.” In Cheryl Misak (ed.), The Cambridge Companion to Peirce. Cambridge: Cambridge University Press, pp. 29-58.


Pihlström, Sami. 2005. Pragmatic Moral Realism: A Transcendental Defense. Amsterdam & New York: Rodopi.


Pihlström, Sami. 2006a. “Putnam’s Conception of Ontology”, forthcoming in Contemporary Pragmatism 3 (Book Symposium on Putnam 2004).


Pihlström, Sami. 2006b. “Armstrong’s Metaphysical Realism”, forthcoming in Markku Keinänen & Tim de Mey (eds.), Problems from Armstrong. Acta Philosophica Fennica. Helsinki: The Philosophical Society of Finland.


Putnam, Hilary. 1981. Reason, Truth and History. Cambridge: Cambridge University Press.


Putnam, Hilary. 1990. Realism with a Human Face. Ed. James Conant. Cambridge, MA & London: Harvard University Press.


Putnam, Hilary. 1992. Renewing Philosophy. Cambridge, MA & London: Harvard University Press.


Putnam, Hilary. 1994. Words and Life. Ed. James Conant. Cambridge, MA & London: Harvard University Press.


Putnam, Hilary. 1995. Pragmatism: An Open Question. Oxford and Cambridge, MA: Blackwell.


Putnam, Hilary. 1999. The Threefold Cord: Mind, Body, and World. New York: Columbia University Press.


Putnam, Hilary. 2004. Ethics without Ontology. Cambridge, MA & London: Harvard University Press.


Schiller, F.C.S. 1907. Studies in Humanism. London: Macmillan.


Seibt, Johanna. 2000. “Ontological Categories: The Explanation of Categorial Inference.” In D. Greimann & C. Peres (eds.), System, Struktur, Wahrheit. Olms Verlag, pp. 272-297.


Shook, John R. (ed.) 2003. Pragmatic Naturalism and Realism. Amherst, NY: Prometheus Books.


Smith, Barry. 1999. “Truthmaker Realism”, Australasian Journal of Philosophy 77, 274-291.


Wittgenstein, Ludwig. 1921. Tractatus logico-philosophicus: Logisch-philosophische Abhandlung. Frankfurt am Main: Suhrkamp, 1964.


van Woudenberg, René. 2002. “On the Metaphysical Implications of Alethic Realism.” In Alston 2002, pp. 119-130.







[1] This section will summarize some main points in Armstrong (2004). For a more metaphilosophically oriented discussion of Armstrong, raising the problem of the relation between his scientific naturalism and apparently first-philosophical metaphysics, see Pihlström (2006b). Armstrong's detailed arguments for his views on the truthmakers of, e.g., modal truths will not be discussed here. See also his earlier (Armstrong 1997, ch. 8) “truthmaker argument” for the postulation of states of affairs. While I will have something to say about states of affairs later, Armstrong's “factualist” ontology (which is partly indebted to Russell and to Wittgenstein's Tractatus) cannot be examined in any detail here. The key idea, very simply, is that states of affairs are needed as the truthmakers of truths of the form a is F. Along these lines, it turns out, according to Armstrong, to be necessary to postulate both particulars and universals as constituents of states of affairs.


[2] For rival views, which do not find the maximalist principle inevitable, see, e.g., Smith (1999), Chrudzimski (2002) and, again, Lillehammer & Rodriguez-Pereyra (2003), especially D.H. Mellor’s contribution. Parsons (1999) distinguishes the (maximalist) “truthmaker principle” (every truth has a truthmaker, in virtue of which that truth is true) from what he calls “truthmaker essentialism” (the thesis that “every truth has a truthmaker, which is essentially that truth’s truthmaker”), arguing (contra Armstrong) that one can hold both the truthmaker principle and nominalism.


[3] On the correspondence theory (explicated along Tarskian lines), see further Niiniluoto (1999) and (2004). In the latter paper, Niiniluoto argues that the Tarskian semantic conception of truth can accommodate the realist idea of truthmaking. Citing Mulligan et al. (1984), he expresses sympathy with the view of tropes as truthmakers, though he goes on to consider facts as truthmakers (Niiniluoto 2004, p. 67). On “alethic realism”, or a realist conception of truth, see also Alston (1996) and several papers collected in Alston (2002).


[4] The need to account for totality states of affairs in truthmaking theory has been emphasized by Cox (1997). Armstrong’s (2004, pp. 78-79) discussion is meant to face the problems raised by Cox: while there may be an infinite regress of truths about totality, no increase in being is required (see also Armstrong 1997, pp. 198-199). Niiniluoto (2004, p. 68) believes the problem of totality is merely epistemic; Armstrong would surely disagree.


[5] This idea summarizes some of the views more thoroughly examined in my as yet unpublished paper, “Transcendental Philosophy as an Ontology”, which defends the idea of metaphysics as investigating (in a Kantian or quasi-Kantian manner) a humanly categorized reality, as distinguished from a reality absolutely independent of us. I should note, however, that there are “truthmaker realists” (who would hardly wish to commit themselves to anything like “transcendental philosophy”) such as Barry Smith (1999, pp. 289-290), who argues that truthmakers should be regarded as “cognition-dependent entities” existing “as a result of certain sorts of cognitively effected demarcations of reality”, that is, as “gerrymandered” “human constructions” in the sense that they “come to be delineated in reality only in virtue of corresponding cognitive practices on the part of human beings”, while being (despite their practice- or cognition-dependence) genuinely “chunks of reality”. Smith’s position differs both from Armstrong’s and from the pragmatists’.






Copyright © 2005 Minerva


All rights are reserved, but fair and good faith use with full attribution may be made of this work for educational or scholarly purposes.




Sami Pihlström is Professor of Philosophy at the University of Tampere, Finland, and Docent of Theoretical Philosophy at the University of Helsinki, Finland.


Mail to: Sami Pihlstrom

Return to Minerva (Volume 9) Main Page                          Go to Top of This Page