Truthmaking. Are Facts Still Really Indispensable?

2.1 Categorial Facts as True Propositions

First, facts can be taken to be true propositions. For example, the fact that the rose is red is simply identical to the proposition that the rose is red.

The father of treating categorial facts as true propositions is Bertrand Russell. In On the Nature of Truth (1907) he considered facts as true propositions treated as non-mental complexes of objects. A similar view is found in Frege, who wrote about facts in his essay Thought (1956 [1918]). Propositions were, for him, senses (intensions) of sentences. Sentences express propositions belonging to the realm of immaterial objects, and facts, for Frege, were true propositions. Both Frege and Russell accepted facts as abstract propositions, but Frege assumed that they did not contain constituents of the world, and Russell assumed that they did (see Rami 2004). Ramsey also held a similar view of facts, but identified them not with true propositions, but with true judgments, including the mental component (Ramsey and Moore 1927).

Treating facts as true propositions seems to have been more popular in the first half of the twentieth century. However, this concept still tempts some contemporary philosophers such as Lenart (2021, p. 2158), Rosen (2010, p. 114), Skiles (2015, p. 720) or Woods (2018, p. 633). For example, Rosen (2010) claims that facts are true Russellian propositions built from worldly items:

[F]acts are structured entities built up from worldly items—objects, relations, connectives, quantifiers, etc.—in roughly the sense in which sentences are built up from words. For my purposes, facts might be identified with true Russellian propositions (…). (Rosen 2010, p. 114)

In what follows, I take Rosen’s view as expressing this type of categorial facts, because he is one of the few proponents of using them as truthmakers (2010, p. 114 n3).^[2]

2.2 Propositional Categorial Facts

Propositional facts are described as corresponding to true propositions (Betti 2015, p. 167). They are not identical with true propositions, but are derived from them as the truth of propositions, so they could be called “truths” (Fine 1982, p. 52). They are ideal (extra-worldly) entities (see Betti 2015, p. 170; Fine 1982, p. 53). In contrast to facts as true propositions, propositional facts (and compositional facts as well) operate not at the level of sense, but at the level of reference (see Betti 2015, pp. 23–24). Propositional facts are named by certain that-clauses or by nominals of the form “the fact that p” (Betti 2015, p. xv). Some authors accept that they are about something (have the property “aboutness”) and unstructured entities (not composed of the objects they are about) (Betti 2015, p. 172). For example, the fact that the rose is red is, according to this propositional type of categorial facts, the truth that the rose is red, which corresponds to the true proposition that the rose is red.

The father of the propositional view of categorial facts is George E. Moore. According to him, the propositional view is a perspective that he previously endorsed (1953, p. 261). However, in the same work, Moore claims that truth is not a property of beliefs, but of objects of belief, or propositions. Propositions are expressed in beliefs. Whether a proposition is true or false, it is a fact that the proposition exists. This fact is simply identical with the proposition itself. However, there is a second fact that is true only of true propositions, namely that they have the property “truth.” True beliefs refer to facts that possessions of the property of being true by propositions. As he points out, “the truth of that particular proposition [‘that lions exist’] is a fact which has to the belief a relation which no other fact has to it” (1953, p. 261).

The propositional view does not have many contemporary proponents, nor has anyone yet come up with the idea of using truths as truthmakers for truths. Of the more contemporary authors, the most prominent proponent of propositional facts is Strawson (see Betti 2015, p. xv).^[3] He holds that truthbearers are statements, which are speech episodes about objects. Truthbearers state facts, although facts do not exist in the world. Facts are pseudo-material correlates of statements. Facts are what true statements state and not what statements are about. Saying “it is a fact” is only one possible way to say “it is true” (Strawson 2001 [1950]). Strawson (2001) does not determine whether facts are identical to true statements (as Searle [1998] observes). In Reply to John Strawson (1998) claims that facts are being true of propositions, which means that they are ideal truths corresponding to propositions:

A proposition, an intensional abstract item, may have many properties (…) It may also have the property of being true; and then it may properly be called ‘a truth’; and this is what a fact is—a truth: just as much an intensional abstract entity as the proposition which ‘fits’ or ‘corresponds to’ it. (Strawson 1998, p. 403).

2.3 Worldly Categorial Facts

The worldly approach to facts, or “compositional” as Betti (2015) calls it, is the most common view of categorial facts. It is also the one that has been used most often in truthmaking so far, so I will focus most on it. Although both propositional and worldly facts are at the reference level, there is a profound difference between them. The identity of worldly facts is not explained in terms of propositions, but rather they are primitive or independently explained. They are structured entities or complexes. There is a correspondence between the structure of worldly facts and the structure of propositions (given a structural understanding of propositions) (Fine 1982, p. 52). Worldly facts are concrete (non-qualitative objects that are not possessed by anything) (Betti 2015, pp. 20, 27). They are part of the furniture of the world, have non-mereological composition (Betti 2015, p. 30), and are causally efficacious (Betti 2015, p. 168). Betti suggests that worldly facts are spatiotemporal because they satisfy the empirical criterion of identity (2015, p. 169). Worldly facts are “semantically idle” in that they do not refer to, mean, or be about anything (Betti 2015, pp. 24, 33). As the most representative view of worldly categorial facts, I take Armstrong’s (1997) idea that facts are spatiotemporal instantiations of universals by particulars. For example, the fact that the rose is red is an instantiation of the universal of redness by the rose.

Since the worldly understanding of categorial facts has been most significant in the development of the concept of truthmaking, I would like to mention in a concise way three stages of the development of this concept with emphasize on the truthmaking aspect.

The first stage of the development of the notion of worldly fact is discussed by the father of the approach, i.e. David Hume (1994 [1758]), followed by Husserl (1913). They treated facts as similar to events, emphasizing their contingency. According to Hume, matters of fact, as he calls facts, are objects of human reason. Facts, for Hume, have the characteristics of contingent events because they are not governed by the rule of non-contradiction, nor does causation regulate their occurrences. This idea of contingent facts was also taken up by Husserl (1913) in Ideas, where he wrote about matters of fact (Tatsachen) as contingent individual entities. For him “Contingency” ^[4] means “factualness.” Facts, according to Husserl, are real (i.e., independent of consciousness in the phenomenological vocabulary) individuals and spatiotemporal beings. They are objects of experiential sciences (1983, p. 7). All in all, worldly facts, in this first stage of theoretical development, were treated apart from the perspective of truthmaking and confused with contingent events.

The second stage in the research on worldly facts is represented by Husserl (1901), Russell (2001 [1912], 1919) and Wittgenstein (1922). They treated facts as a unique category of entities, clearly differentiating the categories of facts and events. Russell and Wittgenstein suggested that facts are complex entities that establish the truth of true propositions and Husserl (1901), also seemed to support treating facts as truthmakers. Therefore, they anticipated later work on truthmaking. However, these authors did not concern themselves with the specific nature of facts. Husserl in his Logical Investigations described states of affairs (Sachverhalten) as objects of judgment and other acts (like presentations, wishes, and doubts) (1901, p. 378). He was one of the first to use the term “truthmaking,” when he wrote about “verifying state of affairs” (Husserl 1976 [1901], p. 767). It was translated by J. N. Findlay in this way, but the German expression wahrmachenden Sachverhalt (1901, p. 596) can be also translated as “state of affairs making true.”^[5] For Russell (1919, 2001 facts were complexes with further constituents. In Truth and Falsehood (2001), he maintained that to establish the nature of truth it is necessary that something exists outside the belief on which truth or falsity ontologically depends. This view implies that truth is a correspondence between facts and beliefs. Russell also uses the notion of truthmaking, when claiming that “what makes a belief true is a fact” (2001, p. 24, his italics). In his later work he included a similar doctrine of worldly facts. In Russell (1919) he held that facts are anything that is complex (that has constituents). He claimed that facts refer to the features of the world that make assertions true or false. This description of the structured facts implies that every constituent of a fact has a position (or positions) in it. Facts are differentiated either by their constituents or by the position of their constituents. However, Wittgenstein’s Tractatus (1922) represents the most significant contribution to the second stage of development regarding the concept of worldly facts. According to Wittgenstein facts are complex beings, which combine, in a definite way, simple objects (entities, things). According to Ploudre (2016), Wittgenstein’s concept of fact allows for three interpretations, all of which support the role of facts as truthmakers. To sum up, all the proponents of the second stage of development of the notion of worldly facts considered them a separate category of entities unlike the first stage, that confused facts and events. Moreover, in contrast to the first stage facts are, often complex, truthmakers. However, the authors do not elaborate more on the intrinsic nature of facts as entities, as in the next stage of theoretical development of the notion.

The third stage began with Armstrong’s concept of facts as states of affairs (Armstrong 1997, 2004). I take Armstrong’s account to be the most detailed exposition of the concept of worldly categorial facts and, as such, a paradigmatic for it.^[6] Armstrong’s work developed intuitions originally found in Wittgenstein and Russell. For Armstrong, facts are not only a separate category of entities understood as complex truthmakers (as in the second stage of the concept’s development), but, their nature for him is defined as spatiotemporal instantiations of universals (properties or relations) by particulars. For Armstrong, the argument for the existence of states of affairs is the famous “Truthmaker Argument.” Accordingly, only a state of affairs (the term Armstrong uses to call facts) and not mere universals, particulars, or their pairs could account for the truth of the proposition:

I conclude that we can accept the truthmaker argument for states of affairs. (…) In particular, there is no call to bind together the constituents of a state of affairs by anything beyond the state of affairs itself. The instantiation of universals by particulars is just the state of affairs itself. (Armstrong 1997, p. 119).

3.1 Categorial Facts as True Propositions as Truthmakers

First of all, if we identify facts with true propositions, then the question of whether facts can be truthmakers becomes a question of whether true propositions can be truthmakers. In other words, then facts are not some special entity, but a label for true propositions. So the question of whether facts are still indispensable seems inappropriate. If one identifies them with true propositions, one has, in a sense, already abandoned the notion of facts. This is precisely one of Betti’s (2015) strategies. She argues that propositional facts reduce to true propositions, and thus recognizes that we can get rid of such a notion of propositional facts in our metaphysical reservoir.

Even if one assumes that the identification of facts with true propositions is not an attempt to get rid of facts, but rather serves the function of describing their nature, it is still difficult to accept facts understood in this way. In truthmaking theory, a truthmaker is indicated precisely for true propositions (true truthbearers). Suggesting facts as true propositions as truthmakers for true propositions is circular and only postpones the search for a truthmaker. For example, if the truthmaker for the true proposition that the rose is red is the true proposition that the rose is red, it postpones the question of what is the truthmaker for that true proposition.

This conclusion is supported by the discussion of so-called deflationary truthmaking. Proponents of deflationary truthmaking (Hornsby 2005; Merricks 2007; Perrine 2015; Schnieder 2006) claim that we can provide the truth of propositions at a lower cost than proponents of substantive truthmaking, who use various entities as truthmakers (tropes, individuals, states of affairs). Deflationists reject truthmakers as well as the substantive truthmaking relation as necessitation. They stop just at saying that to ensure truth it is sufficient to say that <p> is true because p. This is an asymmetric analog of Tarski’s deflationary equivalence scheme (see Asay and Baron 2020). Significantly, in the truthmaking deflationist schema (called the “B-schema” from “because”) we have a proposition on one side of the schema and its name in metalanguage on the other (Asay 2018, p. 908; see Asay and Baron 2020, p. 18). The B-schema thus connects two propositional entities, and so asserts that it is sufficient for ensuring the truth of propositions. It seems, that those who believe that facts as true propositions make propositions true do the same.

Nevertheless, Asay and Baron have recently raised significant criticisms of the deflationary approach on three grounds (2020, pp. 12–20). First, deflationary truthmaking only pretends to be non-substantive on the basis of the asymmetric equivalent of the Tarskian schema. In reality, however, it still presupposes the ontological dependence of truth on being, which it doesn’t deflate. Secondly, even if one assumes that deflationists can maintain a deflationary model of truthmaking while maintaining the ontological dependence of truth and being, they still have to explain how their B-schema does the work of substantive truthmaking. This leads deflationists to rely on an epistemic rather than an ontic theory of truth explanation. But this leads to a denial of necessitation, since in a world without epistemic agents nothing would depend on nothing ontologically. Third, deflationary truthmaking amounts to the claim that the truth of <<p> is true> depends on the truth of <p>. But this only postpones the question of what is the truthmaker for the truth of <p>. As I see it, these three objections to deflationary truthmaking can be reduced to the claim that the deflationist is proposing a cheaper way of ensuring truth. However, this is merely a change of terminology (from the full-blooded truthmaking relation of e.g. necessitation to the B-schema). It only postpones the question of truthmakers and does not abandon the assumption of the ontological dependence of truth on being.

An analogous argument to that proposed by Asay and Baron against deflationary truthmaking can be made against using facts identical to true propositions as truthmakers. This is analogous to the deflationary B-schema. In fact, it is even more radical than the deflationist approach, which on the right side of the B-schema does not repeat the true proposition, but only its name in meta-language. Thus, the proponent of identifying facts with true propositions de facto claims that true propositions make themselves true. But this does not eliminate the question of what makes this proposition true, and how the ontological dependence of truth on being is to be understood (see Asay and Baron 2020, p. 15, 20). This problem seems to be recognized by those who argue that the B-schema, while providing the closest explanation in the chain of explanations, does not exclude further, more substantive ones (Mcgrath 2003; Schnieder 2006; see also Schulte 2011). Of course, there are some rudimentary cases in which a true proposition can be truthmaker. Such a case is a situation in which the truthmaker to the true proposition that there is at least one proposition is that very true proposition (see David 2008, p. 153). In other cases, one should look for truthmakers, which will belong to a different category of entities than true propositions. Therefore, facts as true propositions seem to be ineffective truthmakers.

Notwithstanding the argument against using facts as true propositions as truthmakers in general, an argument can be made against facts as Russellian or Fregean propositions as truthmakers. Some analogies to the criticism of the deflationary theory of truth can be applied to this. Namely, I argue that one can make analogous arguments on the grounds of truthmaking theory as in Rami’s argument against Horwich’s (1998) appeal to facts as true propositions, both understood in Russellian and Fregean terms.

Accepting facts as Russellian propositions with worldly constituents runs into the problem of negative facts. Rami notes in the context of truth theory that for the fact that snow is not yellow there is no Russellian proposition that can be identified with it (2004, p. 85). I think that this problem is particularly relevant to facts understood in a worldly sense (Rami 2004, p. 84), so I will address it in one of the next sections. Moreover, if one assumes that facts are true propositions with constituents, it is difficult to distinguish between true and false propositions (see Rami 2004, pp. 83–84).

As far as Fregean propositions are concerned, whether they are true or false propositions, they are complexes of concepts (senses). When we ask what fact explains the difference in logical value between these two propositions, we cannot refer to some of their conceptual constituents. This is because the logical value of propositions does not depend on their constituents, since the proposition has the same constituents regardless of whether it is true or false. Therefore, the difference in logical value must be explained by something extrinsic to these propositions (Rami 2004, pp. 85–86). Although Rami is not talking about truthmaking, but about truth theory, one can continue this argument up to the level of truthmaking theory. True Fregean propositions are incapable of being truthmakers for truths about the difference in logical value of propositions. Since they are unable to make this true (without reference to something external to them), they are unable to make any truths true at all.

3.2 Propositional Categorial Facts as Truthmakers

Propositional facts have not usually been treated as truthmakers. However, I’d like to consider this possibility as well, and show what might have motivated their not being chosen as truthmakers, or what should warn potential proponents of this approach. First, some, like Betti (2015), assume that propositional facts reduce to true propositions. In this case, propositional facts face the same problems as facts that are identical to true propositions, and thus share with facts as true propositions a low effectiveness as truthmakers.^[7]

On the other hand, even if we take propositional facts to be a different category of entities from facts as true propositions, some constraints on propositional facts remain in the truthmaking perspective. This is especially the case because propositional facts are understood as the truth of propositions.

Treating the truth of propositions as truthmakers leads to circular reasoning. We ask about the truthmaker for truth p by defining it as truth p (or any other truth q relevant to the truth in question). But this only postpones the question about the truthmaker for that truth, analogous to the case of facts understood as true propositions. This is similar to Fumerton’s (1995) argument against the coherence theory of truth. Fumerton notes that the coherentist claims that <p> is made true by being coherent with a set of other believed propositions. Nevertheless, the coherentist’s facts about beliefs are actually truths about beliefs (1995, p. 138). He goes on to argue that the coherentist falls into the conceptual regress of truthmaking because every time one tries to understand what makes a particular proposition true, one is necessarily led to ask what makes another proposition true. The coherentist claims that what makes P true is that it coheres with the set of propositions Q. At the end of this inquiry, however, we can ask what makes Q true, to which the coherentist replies that it is made true by that it coheres with the set of propositions R, and so on ad infinitum (1995, p. 140).^[8] As Fumerton notes, “(…) it is an almost comical error, therefore, to suppose that one can think of a fact [understood as truth] as a truthmaker (1995, p. 138).” Fumerton’s argument in the context of problems for truthmakers within the coherence theory of truth can be translated into the theory of truthmaking. Truthmakers cannot be truths because they generate the vicious regress of the search for further truthmakers. We end up having to appeal (in a large number of cases) to something non-propositional.

This can also be expressed in the way that truth cannot be a truthmaker, because truthmaking is not concerned with truths, but with their content. In other words, it is concerned with what propositions are about.

This is echoed in the recent discussion of whether truthmaking is an explanation. Kitamura (2022) argues against Asay (2020), who maintains that truthmaking is not an explanation of truth. Kitamura notes that proponents of truthmaking as explanation are not fundamentally interested in truth itself. Although we ask the question why <φ> is true, what we are really asking is why φ. Thus, proponents of truthmaking as explanation do not accept deflationary truthmaking, which asserts via the B-schema that <φ> is true because φ. Kitamura notes that the B-schema merely repeats the original focus of the question it seeks to answer. Truthmaking, properly understood, is not concerned with truth, but with things about which there are truths. Thus, it explains subject matter of truths, not truth of propositions (2022, p. 3). Kitamura’s argument within the controversy concerning truthmaking as explanation shows well that the truth of propositions cannot be a truthmaker. It is only the starting point for the search for truthmakers. Thus, facts understood as truths of propositions are not effective truthmakers.^[9]

Because of the abstract nature of facts as truths in the case of many propositions about concrete objects, they can at most be intermediaries or proxies between truthbearers and entities to which they are related. In this situation, however, we will need additional truthmakers for the truth that concrete entities are related in a certain way to propositional facts, and for the truth that through propositional facts they are related to the truthbearers representing them. Therefore, for the sake of economy, it is better to assume that facts as truths are usually not truthmakers.

3.3 Worldly Categorial Facts as Truthmakers

The question of worldly facts is the most important, because it is this understanding of facts that has been most often accepted as truthmakers. However, this conception of facts has also received the strongest criticism so far because of the so-called Unity Problem. The Unity Problem has been described at length in the debate, so I will briefly recall it at the beginning of this section. Then I will present two further arguments against this notion of facts as truthmakers, with particular reference to the context of deflationary truthmaking.

The Unity Problem, though not directly related to the treatment of facts as truthmakers, is very much related to it. Connecting particulars by means of a relation led to the question of what, in turn, connects this relation to the particulars. This led to the introduction of further relations connecting constituents to previous relations, and consequently to what is known as Bradley’s relation regress and the Unity Problem. The main reason for Armstrong’s acceptance of states as truthmakers was precisely to stop Bradley’s regress. Armstrong’s famous Truthmaker Argument was designed to break this regress and save the unity of the constituents by not assuming an additional relation, but by assuming that it is the state of affairs itself that unifies the constituents (Armstrong 1997, pp. 113–119). It turns out, however, that even Armstrong’s concept of states of affairs is susceptible to the Unity Problem, as the arguments of Vallicella (2000) and Betti (2015) show. In addition, Dodd argues against the Truthmaker Argument, claiming that the supposed unity of states of affairs cannot be explained (Dodd 1999, p. 152).^[10] The Unity Problem prevents states of affairs from playing the role of truthmakers, since states of affairs can only play the role of truthmakers insofar as they play the role of unifiers, as Betti notes:

To be sure, once facts are rejected as unifiers of the world (because we have better alternatives) Armstrong’s truthmaker argument will also be rejected as an argument in favor of facts as truthmakers, for the latter role depends on the former. But his does not mean that my arguments dismiss the need for truthmakers altogether. I am only arguing against the effort to make [worldly] facts play that role. Other entities can play the role of truthmaker (if we think that role must be played) (Betti 2015, pp. 49–50).

I do not want to repeat here the entire argumentation of Betti (2015, pp. 39–106),^[11] Vallicella (2000) and Dodd (1999), to which I refer. What is important is to emphasize that those who would accept the worldly facts as truthmakers must show how the Unity Problem can be avoided.

Independently of the Unity Problem, two other arguments can be made against compositional facts. Unlike the Unity Problem, they strike directly at the use of worldly facts as truthmakers.

First, there is the problem of using worldly facts as truthmakers for negative truths, such as there are no unicorns. For such a truth, it is difficult to identify a state of affairs that makes it true, since it is not identical to any state of affairs in the actual world (Rami 2004, p. 84). Moreover, one can argue, as Dodd (2007) does. Accordingly negative truths say that there are no things of a certain kind and that something lacks a certain property. However, if an object lacks a property, then there is no sense in which object and property can be brought together into a single state of affairs. Moreover, the individual himself cannot make sentences such as “The liquid is odorless” true, because then there could be a possible world in which this liquid exists and is not odorless (2007, pp. 386–387). For Dodd: “A state of affairs, after all, is something’s having a property, not an object’s lacking one (or, worse, the absence of things of a certain kind); and, equally, neither the non-existence of a kind of thing nor the particularised failure of an object to have a property are themselves particularised qualities.” (2007, p. 386)

The problem of negative truths was one of the motivations for the rise of the deflationary truthmaking trend, within which a third argument against the use of world facts as truthmakers can be made. Deflationists note that truth can be ensured in a cheaper way than by appealing to the complex entities like worldly facts. To this end they invoke the B-schema (<p> is true because p). In this way they do not need worldly facts, which share the Unity Problem and the problem of the status of negative facts. Among those who refer to the B-schema, Schnieder notes that the claim that Socrates is bald because there is a fact of Socrates’ baldness that expresses truthmaking by worldly facts is incorrect (2006, p. 39–41). It attempts to explain with the complex concept of worldly fact the more primitive concept that expresses the constituents of that fact (Socrates, baldness). Therefore it assumes an explanatory relation that does not exist. Thus, according to Schnieder (and other deflationists who refer to the B-schema), we should reject truthmaking by worldly facts and stick to the B-schema (see Schnieder 2006, pp. 36–37, see also Schulte 2011, p. 419 n 20).

On the other hand, as I noted in the previous section, deflationary truthmaking has its own problems. So maybe we can’t do without worldly facts after all? Nevertheless, even among deflationists themselves, opinions are divided. In addition to the group of deflationists who appeal to the B-schema, there are also moderate deflationists who appeal to a more parsimonious ontology than worldly facts. Instead of the worldly facts and tropes traditionally used in truthmaking,^[12] they appeal either to individuals (Melia 2005) or to objects and properties that are not combined into world facts (Rychter 2014, Dodd 2007, see also Dodd 1999, p. 154). Moreover, Betti (2015, p. xxii) herself, who criticizes worldly facts because of the Unity Problem, claims that we can assume that mereological complexes of relata-specific relations and their relata are sufficient for truthmaking. All in all, worldly facts face three major problems: the Unity Problem, the problem of negative facts, and the problem of explaining simple concepts by complex concepts. Moreover, as the debate shows, we do not need to refer to them at all to ensure truth.

5.1 The Reductive Approach

The defender of deflationary facts must explain the relation that connects them to other categories of entities that serve as truthmakers. To do this, he can simply claim that they are identical to the entities of those categories that serve as truthmakers. However, such a strategy of identifying deflationary facts with entities of other categories leads to their reduction to those categories. The consequence is the abandonment of the concept of facts, which become unnecessary because they are nothing over and above the other categories of entities. In other words, as a result of the reductive theory of deflationary facts, there is such a significant weakening of the content of the concept of fact that it loses its reason for being.

An analogous objection was made by Plantinga (2003) against Lewis (1986). Lewis speaks of various entities as possible worlds, properties and propositions (Lewis 1986). Although Lewis speaks of possible worlds, properties and propositions, he adopts an ontology that consists only of concrete possibilia and sets (Plantinga 2003). According to Plantinga’s (2003) interpretation, Lewis actually models modality in set-theoretic constructions on concrete individuals. Possible worlds, universals and propositions exist by him only as concrete individuals and set-theoretic constructions. Plantinga claims that Lewis as an antirealist “holds that there are no such things [e.g. universals], and may add that the role said by some to be played by them is, in fact, played by entities of some other sort” (2003, p. 192), i.e. by concrete individuals and sets. Hence, according to Plantinga, entities expressing modality in Lewis (like possible worlds) actually reduce to other entities, like concrete individuals or sets.^[16]

Similar to how Lewis reduces possible worlds to concrete individuals and their sets in accounting for modality, one could reduce deflationary facts to categories of entities in accounting for truthmakers. One could maintain a reductionist position with respect to deflationary facts in the same way that Lewis is an reductionist with respect to possible worlds. As a result of this reductive approach, there would be no need to speak of deflationary facts, since there would be only entities of different categories as truthmakers. Alternatively, deflationary facts would have a representational status (intermediary between entities and propositions), by which they would face the problems of facts as true propositions and propositional facts.

Smith and Simon (2007) make a similar argument. They argue that proponents of a truthmaking-driven definition of truth cannot appeal to deflationary facts (which they call plebeian facts), because such an approach is unable to provide adequate truthmakers. The purpose of truthmaking-driven definitions of truth is to emphasize the way in which true claims turn out to be equivalent to ontological claims. Nevertheless, the deflationary account of facts is intended to be ontologically modest or neutral. The fact-talk of advocates of deflationary facts seems reducible to some non-fact-involving talk, so it has no exploratory function in our metaphysics. Thus, any definition given in terms of plebeian facts is nothing more than an abbreviation of a definition formulated in other terms. In other words, deflationary facts are just façon de parler about something else (2007, p. 87).^[17]

Moreover, truthmaking theory was established to counter reductive theories such as behaviorism or phenomenalism, which accept some true propositions but are unable to identify truthmakers for them. A similar problem seems to apply to the reductive theory of deflationary facts. If one accepts that deflationary facts are reducible to other entities, then one cannot identify separate truthmakers for true propositions about deflationary facts (e.g., “deflationary facts exist”). However, logically independent basic propositions must have distinct truthmakers (Maurin 2005, p. 141). Thus propositions about deflationary facts are false, including propositions about their existence. Since propositions about deflationary facts do not have truthmakers (and the entities to which they are reduced do), deflationary facts do not exist. Hence, the strategy that is one of the main motivations of truthmaking theory lends itself to use against deflationary facts.

In addition, the reduction of deflationary facts to other categories of entities provokes the question of whether deflationary facts can be considered minimal truthmakers, the provision of which is favored by some proponents of truthmaking. It seems that due to their reduction to other entities, they are not minimal truthmakers, but at most their representations.

Moreover, deflationary facts cannot provide a basis for maximalism (for those who support it). If one wants to adopt maximalism, one cannot base it on entities that reduce themselves to other entities, since they are deprived of their own content. The adoption of deflationary facts as truthmakers resembles the strategy of saving maximalism through so-called totalizer fact. This is the approach criticized by Smith and Simon. They note that the totalizer fact is a deflationary fact. As they claim “the totalizer fact exists if and only if the collection of individuals with which it is associated constitutes all and only the entities satisfying the given condition” and it actually is “nothing more than a façon de parler about something else” (2007, p. 87, their italics). They suggest that totalizer fact has no function other than to stop maximalism, so its postulation is ad hoc. Similarly, in the case of deflationary facts as truthmakers, it is not clear what the reason for their existence is other than an attempt to forcefully save the concept of facts in residual form. Moreover, their postulation does not solve the problem of truthmakers for negative truths. Since they are reduced to other entities, the proponent of maximalism still has to explain how existing entities make true propositions about the non-existence of certain portions of the world.

5.2 The Non-Reductive Approach

Of course, an advocate of deflationary facts can appeal to a non-reductive account of the relation of deflationary facts to entities of different categories. For this purpose, he can invoke the concept of multiple realizability, which has already found wide application outside the debate within philosophy of mind.^[18] Multiple realizability is a term derived from the debate over the body-mind problem within philosophy of mind. It was introduced to justify a non-reductionist view called functionalism. Functionalism purported to be the opposite of reductive body-mind identity theory, which proclaimed the identity of physical states with mental states. Multiply realizability within functionalism is related to the issue of the difference between token identity and type identity (see Bickle 2020).

By Fodor (1974) we find the difference between type identity and token identity in distinguishing between token and type physicalism. According to him, although type physicalism (type identity) entails token physicalism (token identity), token physicalism does not entail type physicalism. It means there can be token identity without type identity. From the fact that token physical events are token mental events, it does not follow that physical natural kinds are co-extensive with mental natural kinds.

Although the idea of multiple realizability is rooted in the philosophy of mind, it can also be applied to other areas of philosophy As Polger and Shapiro note:

The idea is just that for some entities—properties, states, kinds, objects – being that entity is a matter of having a certain function. Whatever performs that function thereby realizes the entity of which that function is characteristic. (…) [T]he conception of realization we favor does not, prima facie and in itself, put any limitation on which items may be the relata of realization. In our view, the relata of realization relations are whatever things can have and perform functions, in a broad sense (Polger and Shapiro 2016, p. 22).

Applying the concept of multiple realizability to this issue of the relationship between deflationary facts and entities of different categories, we can say that being a deflationary fact is a matter of having the function of truthmaking. So entity of any ontological category that performs that function thereby realizes deflationary facts. To put it in terms of token-identity and type-nonidentity: entities of different ontological categories that can ensure the truth of truthbearers are token-identical with each other as truthmakers (e.g., the trope of the redness of the rose and the instantiation of the universals of redness by the rose for the true proposition that the rose is red). However entities of different categories are not type-identical with deflationary facts. Stąd deflacyjne fakty nie redukują się do jednostek różnych kategorii.

At first glance, this seems to effectively avoid the problem of reducing deflationary facts to other categories of truthmakers. However, this seems to be an ad hoc argument, since reductive accounts of multiple realizability can also be identified. It is not clear why an advocate of deflationary facts would prefer a non-reductive rather than a reductive approach, other than that he wants to avoid reducing deflationary facts. This is reminiscent of Betti’s argument against Armstrong’s postulation of facts as an ad hoc solution to Bradley’s regress problem (if they can only stop regress and there is no other reason to accept them) (Betti 2015, pp. 82–83). A reductive account of multiple realizability in truthmaking theory can be found in Schulte (2011, p. 421). He argues that substantial truthmaker explanations are about explaining a higher-level fact by a lower-level fact. He puts it this way: higher-level facts are multiple realizable, where the explanation of higher-level facts by lower-level facts is a reductive explanation. In other words, higher-level facts are reducible to lower-level facts that realize them.

Thus, a defender of deflationary facts would have to provide additional arguments in order to maintain a non-reductive account of multiple realizability. But apart from the mere fact that he does not want to adopt the reductive account, it is in vain to find it in him. Moreover, he would have to explain why it is necessary to choose exactly such an entity and not another for the realization of a given deflationary fact. Thus, it seems that deflationary facts are nothing more than entities of other categories that are used as truthmakers. Instead of saying “(deflationary) fact,” it is better to simply say “truthmaker.” Deflationary facts have become just a label, unfortunately they are not needed for truthmaking.

The deflationary notion of facts leads to a denial of the basic intuition of truthmaking (Mulligan et al.), which is to point to ontological grounds for truths. Instead, it leads to an indifferent notion of the truthmaker and, consequently, to the deflationary theory of truth from which the fathers of truthmaking distanced themselves. Wright and Pedersen propose something similar in the analogous case of adopting a deflationary notion of facts in a correspondence theory of truth. They note that the deflationary notion of facts is unacceptable to correspondence theorists because it makes them indistinguishable from other theories of truth, such as deflationist minimalism, disquotationalism, modest identity theory, coherence theories, and so on (2010, p. 215). So, to paraphrase Wright and Pedersen (who express this phrase in the context of the correspondence theory of truth), a fact-based theory of truthmaking that cannot make good on a substantive conception of facts is not a fact-based theory of truthmaking worth having (see 2010, p. 215).