Foucault, Lorenzini, and Spinoza on Games of Truth (with some Carnap and Tarski).

Silence or assent seem to be the only legitimate response(s) to a sound argument. I have always found it peculiar that silence or agreement should be the internal goal of philosophy. And I remember with fondness from my graduate school days the manner by which the best logicians would manage to circumvent this: when confronted with a sound argument they couldn’t refute, but evidently didn’t like, they would say with great care, even aplomb, “I don’t find that [argument] interesting.” And lurking there would be an aesthetic sense that could, somehow (not if I tried it, of course) shift the terms of the discussion altogether. (Of course, this could result in a higher order stalemate of conflicting ‘that’s un-interesting.’)

I had to remember those halcyon days when I was (recall also this post) reading Daniele Lorenzini’s (2023) fascinating The Force of Truth: Critique, Genealogy, and Truth-Telling (The University Chicago). This book offers a reading of Foucault on truth where an important distinction between so-called ‘regime(s) of truth’ and ‘game(s) of truth’ is (correctly) ascribed to Foucault. A game of truth has its own rules that regulate ‘true’ and ‘false’ and their ascriptions. According to Lorenzini, “within each game…, truth is index sui and is therefore not “relative” nor reducible to power.” (p. 122) By contrast, regimes of truth do reveal “the necessary interaction between truth and power within the procedures that are used to govern human beings,” (p. 122; some time I hope to understand this necessity in that ‘necessary.’)

To understand this distinction in Foucault, then, between a game and regime of truth, we can go back to my first sentence above. Soundness (validity, truth tables, etc.) is part of the game of truth. The purported legitimacy of how to respond to a sound argument is part of a regime of truth. From the perspective of the game of truth the response is pure surplus.

We can put Foucault’s distinction in quasi-Carnapian terms, the rules that govern the game of truth are internal to or constitute the game; the rules that govern the regime of truth are, in a certain sense, external questions. That there is a regime of truth alongside a game of truth may be inevitable, but the content of the regime of truth is (again to use Carnapian lingo) optative. We could also require a boogey woogy when we encounter a sound argument.

Part of Foucault’s interest is in showing how a regime of truth shapes truth’s social and individual authority over us. Here’s how Lorenzini understands his own reconstruction of Foucault’s project:

What Foucault puts at the core of his critical project, more precisely, is the “therefore” of truth, that is, the often imperceptible procedures that lead us not only to accept certain truth claims, but to submit to them and give them the power to govern our conduct. This is what I have called the problem of the force of truth.—p. 120

The power of truth, then, comes from a submission to allowing it authority, that is, to govern conduct. There is lurking here an unresolved question of what to make of a game of truth that never is enlisted in a regime of truth. But let’s leave that hanging for now.

As my soft-Carnapian re-description of these Foucaultian distinctions hints at I doubt Lorenzini manages to extract Foucault from (his self-ascribed mission to fight [p. 1; p. 106; p. 120]]) the charge of relativism against Foucault in so far as the the plurality of games of truth just is a way to convey many kinds of (ahh) operational frameworks. But it’s possible that Lorenzini is happy to grant that Foucault is a certain kind of coherent relativist.

For, the real work of Lorenzi’s book is to characterize how for Foucault regimes of truth insinuate, as it were, themselves into the authority accorded to games of truth. A key text in Lorenzini’s account is Foucault’s (1979-1980) lecture series translated as On the Government of the Living. As it happens this lecture series is rather important to my own interest in the liberal art of government, so I am grateful to be reading it (about that some other time) thanks to Lorenzini’s fine book.

Below, I have quoted the passage from Foucault that is really the master-key to Lorenzini’s overall argument (it is explicitly at least partially quoted on p. 36; p. 38; p. 39; p. 40, p. 41; p. 44; p. 45; p. 48; p. 51; p. 53; p. 120; p. 121; and probably alluded to more times):

In other words, it is not the truth that so to speak administers its own empire, that judges and sanctions those who obey or disobey it. It is not true that the truth constrains only by truth. To put things very simply, in an almost or completely infantile way, I shall say the following: in the most rigorously constructed arguments imaginable, even in the event of something being recognized as self evident, there is always, and it is always necessary to assume, a certain assertion that does not belong to the logical realm of observation or deduction, in other words, an assertion that does not belong exactly to the realm of the true or false, that is rather a sort of commitment, a sort of profession. In all reasoning there is always this assertion that consists in saying: if it is true, then I will submit; it is true, therefore I submit; it is true, therefore I am bound. But this “therefore” of the “it is true, therefore I submit; it is true, therefore I am bound,” is not a logical “therefore,” it cannot rest on any self-evidence, nor is it univocal moreover. If in a certain number of cases, in a certain number of games of truth, like precisely the logic of the sciences, this “therefore” goes so much without saying that it is as if it is transparent and we do not notice its presence, it nevertheless remains the case that standing back a bit, and when we take science as precisely an historical phenomenon, the “it is true, therefore I submit” becomes much more enigmatic, much more obscure. This “therefore” that links the “it is true” and the “I submit,” or which gives the truth the right to say: you are forced to accept me because I am the truth—in this “therefore,” this “you are forced,” “you are obliged,” “you have to submit,” in this “you have to” of the truth, there is something that does not arise from the truth itself in its structure and content. The “you have to” internal to the truth, immanent to the manifestation of the truth, is a problem that science in itself cannot justify and account for. I think this “you have to” is a fundamental historical-cultural problem.—Michel Foucault, 6 February 1980, On the Government of the Living, translated by Graham Burchell, pp. 96-7 [emphases in original—that italicized ‘therefore’ is underlined in the ms.]

Now, based on this, it is pretty clear why Lorenzini treats regimes of truth as pertaining to the “practical submission” to the truth (see the table in The Force of Truth, p. 38). So far so good. It is not clear to me why in the same table, Lorenzini treats a game of truth as pertaining to epistemic acceptance. I would have though that we’re in the realm of semantic or pragmatic (perhaps even metaphysical) necessity. For Lorenzini’s overall argument not much hinges on this quibble. And I don’t object to what Lorenzini does with the distinction between a regime and game of truth in Foucault. But I don’t think Lorenzini properly characterizes Foucault’s position on how within a game of truth, truth is constituted. Here’s how Lorenzini puts it:

On the one hand, Foucault explains that he does not want— and that, in fact, he never wanted— to deny that the truth is index sui:12 in any game of truth, that is, in any regulated system for the production of truth claims (when it is considered in terms of its formal structure, and not of the individuals who concretely engage with it), “only the truth can legitimately show the true” and establish the distinction between true and false statements. (p—35; emphases in Lorenzini [who is quoting Foucault])

When I first read this passage it puzzled me greatly. Before I get to that. In Lorenzini Note 12 contains a reference to Spinoza’s letter 76 to Burgh (and quotes/translates it): “The truth reveals itself and the false [est enim verum index sui, et falsi].” (p. 142)

Now, in Spinoza the ‘game of truth’ is rather unusual. Rather than conforming to the intuitions shaped by what Tarski (1944) called the ‘classical Aristotelian conception of truth,’ which always involve some relation between the true and a proposition or statement about it,* Spinoza seems to disallow any room between truth and (ahh) the true. This is why he can claim ‘verum index sui.’ (This is why (recall) I tend to call Spinoza’s theory of truth a ‘metaphysical identity theory.’)

However, in most non-Spinozistic games of truth (the ones that involve the clasical conceptoon of truth according to Tarski) that’s not how ‘truth’ operates. Rather it obeys certain rules (as defined in, say, the metalanguage) or certain operations.

So, if Foucault were thinking of Spinoza’s approach to truth as paradigmatic of a game of truth, then it’s not at all clear how he can think there are multiple kinds of games of truth (as he does). In Spinoza, truth is not about epistemic acceptance at all, it is metaphysical necessary. In fact, in Spinoza there is simply no room for skeptical doubt because, to repeat, there is literally no space between the true and truth. Truth is indexed to itself. (I would not translate his latin ‘index’ with reveals,’ but actually with ‘'proof’' or ‘mark.’) This all sounds a bit mysterious, but it is rather central to Spinoza’s account of eternity.

Now let’s look at what Foucault said in his lecture. The passage occurs exactly before the material quoted above and that I said was key to Lorenzini’s approach. Here it is:

However, this objection to the idea of a regime of truth, and against the project of analyzing regimes of truth in general, does not seem entirely satisfactory to me. In actual fact, it seems to me that when we say that it is truth and truth alone that obliges in the truth we are in danger of failing to grasp what I think is an important distinction. We should not confuse two things. On the one hand there is the principle that truth is index sui, that is to say, removing its specifically Spinozist signification, the principle that only the truth can legitimately show the true, that at any rate only the game of truth and falsity can demonstrate what is true. But for all that truth is index sui, this does not mean that the truth is rex sui, that the truth is lex sui, that the truth is judex sui. That is to say, the truth is not creator and holder of the rights it exercises over men, of the obligations the latter have towards it, and of the effects they expect from these obligations when and insofar as they are fulfilled.—Michel Foucault, 6 February 1980, On the Government of the Living, translated by Graham Burchell, p. 96 (emphasis in original)

This passage is explicitly responding to a rather lengthy objection that boils down to the thought that “when it is a question of truth itself there is no need for a regime of truth.” (p. 96)

Now, the way I read Foucault’s passage is as follows: in order to describe the operation of truth (in games of truth), without the way it obliges (in regimes of truth), Foucault adopts Spinozistic terminology (that is, ‘index sui’). According to Foucault this term can be understood as “the principle that only the truth can legitimately show the true.” As I hinted above, I think in Spinoza index sui does not involve ‘legitimacy’ at all. But Foucault’s ‘show’ explains why Lorenzini treats all of this in epistemic terms. (Lorenzini may correctly claim that any disagreement that I may have over Spinoza interpretation is with Foucault and not Lorenzini.) [It is, alas, not a translation error; in French, Foucault was recorded as saying, “Il y a d'une part le principe que le vrai est index, sui c'est-a-dire, en lui otant sa signification proprement spinoziste, que seule la verite peut montrer legitimement le vrai, que seul en tout cas le jeu du vrai et du faux peut demontrer ce qui est vrai.”]

But Foucault immediate adds that he does so by “removing its specifically Spinozist signification.” So, Foucault is using the term ‘index sui’ apart from Spinoza’s apparatus (that is, “the principle the principle that only the truth can legitimately show the true.”) It would be quite odd to use Spinoza’s apparatus to treat of games of truth because this apparatus is not suited to that function.

And then Foucault goes on to say, “only the game of truth and falsity can demonstrate what is true.” So rather than claiming that the truth indicates the true, Foucault distances himself from this claim and implies that this is to be decided by (ahh) the rules that govern “the game of truth and falsity.” Now, why did I add these rules?

Now, I happen to like Lorenzini’s own gloss on a game of truth, that is ‘any regulated system for the production of truth claims.’ To put it in a slogan: despite the terminology, Foucault is not a Spinozist about the game of truth; rather like the logicians of the twentieth century, he thinks of games of truth as rule-governed or as having something like a “grammar” (6 February, p. 99); well maybe not all the game of truth, but at least the ones involved in science(s).

---

*I am careful not to use ‘correspondence’ here because the relation need not be a correspondence theory of truth.