Rorty, Brandom, and Deleuze are well-known critics of representationalism, a trait they share with the classical American pragmatists. Although they clearly intend slightly different things by the term, they have all, each in their own ways, been deeply concerned to overturn a representationalist approach to meaning or sense, which is to say, an approach that holds that the propositional content of linguistic expressions and mental states is to be explained by a word-world relation: a relation that is generally thought to be causal in some sense, and that tends to involve what Sellars (1997) famously called the “Myth of the Given.”
This essay will examine these three philosophers’ critiques of representationalist semantics and explicate their antirepresentationalist alternatives. 1 But in doing so, my particular concern will be to develop a novel, Deleuzian response to a common objection to semantic antirepresentationalism, namely, that without word-world explainers our talk about the world becomes a “frictionless spinning in a void” of pure conceptuality. 2 The essay will proceed as follows. First of all, I will outline Rorty’s and Brandom’s critiques of the idea that word-world relations explain propositional content and explicate their alternative positions. Rorty’s and Brandom’s responses to charges of the “spinning in the void” type will then be examined, and I will focus in particular on the way in which Brandom, unlike Rorty, hopes to rehabilitate a notion of objectivity within a broadly antirepresentationalist position. I will then look at a final criticism of Brandom’s position on objectivity, namely, that because his notion of objectivity is rendered solely at the linguistic and normative level, it doesn’t include the idea of something “out of our control” constraining thought. Finally, in light of the preceding analyses, I will examine the contribution that Deleuze’s project in Difference and Repetition can make to developing a concept of objectivity, within an antirepresentationalist semantics, that responds to this objection.
In order to discuss the relation between antirepresentationalism and objectivity in Rorty and Brandom in an economical way, Rorty’s position on the issue will be presented through the lens of Brandom’s reading of Rorty in chapters 4 and 5 of the recently published Perspectives on Pragmatism. Here, Brandom examines Rorty’s objection to the representationalist idea that the things making up the world as it is in itself cause us to have meaningful expressions that refer to them, and that represent these things accurately in the case of true representations. He rightly understands Rorty to have derived his objection to this idea from Sellars’s famous critique of the “Myth of the Given” in Empiricism and the Philosophy of Mind. This “Myth” (at least in its perceptual form) is that the world itself, in the form of what is causally given in perception (that is, in the form of perceptible facts), is capable of grounding knowledge claims and explaining propositional contentfulness. In other words, conceptual knowledge is supposed to be built up from preconceptual experience—from facts grasped by the senses—and the meaning of our expressions is supposed to be given by the objective, worldly conditions that render them determinately true or false. In Perspectives on Pragmatism, Brandom identifies two Rortyan responses to this representationalist thesis: one that Rorty effectively takes, and one that Rorty could have taken, and which is taken up by Brandom himself.
Rorty’s effective response to the representationalist thesis that explains propositional content with reference to word-world relations via the perceptual given is that it is incoherent. It is incoherent because the representationalist’s “facts,” grasped perceptually, are simply the wrong kinds of things to exercise authority over how we ought to talk about the world (Brandom 2011, 111). For Sellars, and for Rorty following him, the perceptually given cannot coherently be understood as cognitively significant apart from its role in an inferentially articulated practice of applying empirical concepts. What is cognitively significant, a belief or sentence, is so only in virtue of its relation to other beliefs or sentences and their inferential connections. So any talk about facts could only be talk about something that is already conceptually structured and propositionally contentful—something, in other words, that is the right shape to play a role in our social practices of reason giving or “conversation” (Brandom 2011, 121, 124). For Rorty, the meaning of a sentence is a function of the inferential connections between large numbers of such sentences. 3 It is not a function of the way in which, by virtue of intrinsically meaningful elements of perception, components of sentences hook up with bits of objective reality.
But as Brandom points out, Rorty could equally well have made a different response to the representationalist thesis. For if, as Rorty argues, authority over the correctness and meaning of claims is a “normative status” that is instituted by our social practices, then within our social practices, could we not put this authority where we like? The position that our claims are epistemically and semantically answerable to the things they are objectively about would thus not necessarily be incoherent. It would rather be optional—indeed, as optional as the position that the correctness and contentfulness of our claims is a matter of their relation to the claims endorsed by our conversational peers. But then the question that would need to be answered is how a notion of “answering to what we are objectively talking about” can be rendered at the normative level, entirely within the social practice of giving and asking for reasons. Or again, as Brandom (2011) puts it, the crucial question is
not whether we can invest authority in nonhuman things.… Of course we can. It is rather how we can institute a dimension of assessment of our sayings and doings that is properly understood as granting semantic and epistemic authority over their correctness, to how it is with the things that we… count as thinking and talking about. (113)
This second response and corresponding question are the ones that Bran-dom pursues. I’ll return to outline how he argues for this idea and thus distances himself from Rorty. 4 But before doing so, let us note what Rorty and Brandom do have in common, namely, the idea that knowledge and meaning are not ultimately matters of a relation between words and the world in itself but rather between participants in a conversation, or in a game of giving and asking for reasons. A criticism often leveled at such an antirepresentationalist position, however, is that it leaves open the possibility that our speech might merely be “spinning in the void” of pure conceptuality, or akin to moves in a self-contained game, or not “about” anything real. Antirepresentationalism, in other words, by jettisoning a determinate relation between our speech and the real world “in-itself,” seems to jettison the important category of the objective (objective truth, objective reality) as that which is supposed to exercise authority over how we talk about the world.
Rorty, for his part, is not particularly bothered by this “spinning in the void” criticism. He thinks that we are always in causal contact with the world and are constantly trying to cope with the world’s causal pressure in more and more effective ways. One of the ways we do this is by developing, with our conversational peers, “vocabularies” which encode useful ways of describing the world in relation to various practical projects. But it is one thing to say that the causal pressures of the world lead us to invent ways of talking about it, and quite another thing to say with representationalists that the authority of our vocabularies depends on their representing the way the causal world objectively is, in itself. As is clear from his critique of representationalism, presented above, it is for Rorty a type of category mistake to confuse the causal relation between the world and vocabularies (that is, brute causal pressure), with the normative structure of epistemic justification and propositional content (which is essentially social) (Brandom 2011, 123). Thus rejecting representationalism, it is clear Rorty has no need of the category of objectivity understood as “answering to the world in-itself.” As he concludes in Philosophy and the Mirror of Nature (1979), our only usable notion of “objectivity” is “agreement” with our fellows (337). Or again, as he puts it in works such as Objectivity, Relativism and Truth (1991), the desire for objectivity should be abandoned in favor of the desire for “solidarity” with our fellows (13).
Brandom, as opposed to Rorty, wants to rehabilitate a notion of objectivity from within our social practice of giving and asking for reasons. Like Sellars and Rorty, Brandom suspects that representationalist semantics involves some form of the Myth of the Given and defends a social-practice, inferentialist account of meaning. He is critical of the representationalist semantic strategy of beginning with an explication of the concepts of truth and reference and subsequently proceeding to an explanation of meaning and language use in terms of these more “basic” concepts. However, and quite unlike Rorty, he thinks we can begin with an antirepresentationalist account of meaning and then proceed to give an account of representation, reference, truth, and objectivity (Brandom 1994, xvi, 69–70). In order to understand Brandom’s argument for this in Making It Explicit, we must explicate the relations between three interconnected dimensions of discursivity: normative pragmatics, inferentialist semantics, and a socially articulated account of the representational dimension of discourse.
For Brandom, the meaning of linguistic expressions is a function of the inferential relations in which these expressions are caught up. But these inferential relations must first of all be explained with reference to an underlying normative pragmatics. In other words, the meaning of linguistic expressions is ultimately to be explained in terms of what speakers are doing in using them, and so in terms of the role they play in our norm-governed social practices. For Brandom, the most important social practice in which linguistic expressions play a role is that of attributing and acknowledging “commitments” and “entitlements” to those commitments (1994, 159).
Of course, since meaning and, subsequently, representation, reference, truth, and objectivity, are to be explained with reference to the social practice of attributing and acknowledging commitments and entitlements, none of these notions can be presupposed in giving an account of this social practice. The practice must rather be explained in terms of the norms that implicitly govern it. Brandom thus gives an account of what people do (that is, of their practical attitudes) when they take speech acts to be proper or improper, correct or incorrect. In other words, on this account, commitments and entitlements are ultimately to be understood as “deontic” or social statuses, which are instituted by the practical attitudes of those who attribute and acknowledge such statuses (Brandom 1994, xiv).
Brandom’s account of normative pragmatics can be understood using the model of what he calls “deontic scorekeeping,” which is a “game” of attributing and acknowledging commitments and entitlements (1994, 165–166). 5 The basic “move” in this game is to “undertake” a commitment. By undertaking a commitment, however, one also inherits certain other commitments. These are the “committive consequences” of making the initial commitment and are “attributed” to the game player by her fellow players. These commitments, both undertaken and attributed, constitute one’s “score,” and all players keep track of their own scores, as well as the scores of their fellow players. Importantly, however, in “attributing” a score to him- or herself, each player counts only those committive consequences that he or she “acknowledges.” And because there can be disagreements between players and scorekeepers as to what the committive consequences of undertaking an initial commitment might be, each scorekeeper attributes two scores to the other players, distinguishing between what they think a particular player would acknowledge as their commitments, and what they ought to acknowledge. A crucial consequence of this difference of perspective is thus that, when a player undertakes a commitment, there will invariably be a difference between the committive consequences he or she is attributed and those he or she acknowledges (Brandom 1994, 194).
But now, a scorekeeper also distinguishes between those commitments to which she thinks a player is “entitled” and those to which the player is not so entitled. Like commitments, entitlements have the status they do because of the way that players and scorekeepers treat them in practice: by attributing and acknowledging them. When a player initially undertakes a commitment, he or she is by default entitled to that commitment, that is, until it is “challenged,” and as long as that commitment is not taken to be incompatible with other, preexisting commitments. Moreover, in the same way that scorekeepers take there to be committive consequences to a player’s commitments, they also take there to be “permissive consequences” to a player’s entitlements. In other words, from a given entitlement, other entitlements are inherited (Brandom 1994, 168). And just like commitments, the player will acknowledge some of these further entitlements but not others: a player can thus be treated by scorekeepers as entitled to certain commitments that he or she does not acknowledge.
As mentioned, however, a player’s default entitlement to a commitment can be challenged. 6 This happens when a second player avows a commitment that is taken to be incompatible with it. The first player can respond to this challenge by, for example, avowing another commitment that is taken to be a permissive consequence of the challenged entitlement (Bran-dom 1994, 177–178). But explicating the details of this structure of challenge and response is not as important for our purposes here as noting that different scorekeepers will not necessarily treat the committive avowals and counter-avowals of their fellows in the same way. Indeed, each player-scorekeeper will work out incompatibilities between commitments and attribute entitlements from his or her own perspective, that is to say, as a function of what he or she is personally committed to (his or her “background commitments”). It is clear, then, that there can be no single score for a player’s entitlements. If entitlements are normative statuses, they are not statuses that are instituted once and for all, but only in the socially articulated to-and-fro of undertaking, inheriting, attributing, acknowledging, challenging, avowing, counter-avowing, etc., commitments.
So, for Brandom, it is this game of deontic scorekeeping that models what we practically do when we use linguistic expressions: we collectively undertake, attribute, acknowledge, challenge, and keep track of our commitments and entitlements. But now, in order for this norm-governed social practice to qualify as a specifically discursive practice—that is, in order for a commitment to have the form of a meaningful claim or assertion—Brandom must also account for the propositional content of the linguistic expressions used in this practice. It is at this point that Brandom advances an inferentialist semantics, which builds on his account of normative pragmatics. The core idea is that it is the inferential “articulation” of the relations between attributed and acknowledged commitments and entitlements that generates the meaning of the linguistic expressions used in the game of deontic score-keeping (Brandom 1994, xx–xxi, 141). The relations constitutive of Bran-dom’s normative pragmatics can thus be translated into inferential relations. With regard to commitments, to treat someone who is committed to a first claim as also committed to a second claim is to take it that the content of the first claim “entails” the content of the second. Similarly, for entitlements, to treat someone as entitled to a claim because they are entitled to a second claim is to take the content of the first claim as “supported” by the content of the second. Finally, to challenge someone’s entitlement to a claim because of a commitment to a second claim, which is incompatible with it, is to take the content of the second claim as “excluding” the content of the first. 7
Claims thus acquire meaning by standing in such inferential relations to other claims, and grasping the meaning of a claim is thus, as in Sellars, a type of practical mastery, namely, being able to distinguish in practice what would follow from endorsing it, and what such endorsement would follow from. To grasp what is meant by the claim, “That’s red,” for example, would be to grasp that sentence’s relation to claims such as, “That has a color,” “That’s scarlet,” and “That’s not green.” Speakers mean the same thing by the sentence, “That’s red,” insofar as they recognize the same inferential norms governing the use of that sentence. Moreover, one would be entitled to make such a claim, not because of some reddishness in one’s visual field, but rather because one takes oneself, and is taken by others, to have simultaneously endorsed such other claims and mastered the inferential connections between them (see Brandom 2000, 47–49).
Now, Brandom proceeds to argue that one of the major challenges facing inferentialism is to account for the representational dimension of semantic content, along with the associated notions of truth, reference, and objectivity (Brandom 1994, xvi–xvii, 136–137). Rorty, as has been seen, believes that a philosophy of language suitably purified of the Myth of the Given can dispense with such a task. Brandom follows Rorty in rejecting the idea that the concept of “representation” can play a foundational role in explaining propositional contentfulness. But he does think that representational vocabulary plays an important “expressive” role in discursive practices, namely, that of making explicit the idea that, besides the content of what we say or think, we also talk or think about things (Brandom 2011, 209–215; see also Wanderer 2008, 174–186).
Brandom’s argument here is that the representational aspect of discourse is a feature of the essentially social dimension of communication, that is, of producing and consuming claims and reasons for claims (Brandom 1994, 496). 8 As he puts it, “[t]alk about representation is talk about what it is to secure communication by being able to use one another’s judgments as reasons, as premises in our own inferences, even just hypothetically, to assess their significance in the context of our own collateral commitments” (2000, 167). More specifically, Brandom argues that the deontic scorekeeping distinction of social perspective between the commitments a scorekeeper attributes to a player and those the scorekeeper undertakes in light of his or her own commitments gives us a distinction between what people are saying and what they are saying something about. The technical crux of Bran-dom’s account involves a consideration of the contrast between de dicto and de re ascriptions of propositional attitudes. 9 A de dicto ascription is an attribution of a propositionally contentful commitment to a speaker, which this latter would endorse because of his or her collateral commitments (e.g., “The defense attorney claims that the man who just testified is a credible witness”; or “He claims that the seventh god graces us with his presence”). A de re ascription, on the other hand, codifies a commitment the ascriber undertakes, in relation to his or her own collateral commitments, as to what the ascribee’s dictum is a dictum about (e.g., “The defense attorney claims of [or about] a pathological liar that he is a credible witness”; or “He claims of [or about] the sun that it is shining”). In translating a de dicto ascription into a de re ascription, the scorekeeper-ascriber brings out what, in the scorekeeper’s perspective, the ascribee is talking about. In other words, the representational dimension of discourse is established within the social practice of deontic scorekeeping by players-scorekeepers ascribing commitments de re to each other.
But now, from the distinction of social perspective involved in attributing and undertaking commitments, we also get an account of what people are doing in the game of deontic scorekeeping when they use expressions such as “is true” and “refers.” With regard to the locution “is true,” Brandom’s idea is that for one to employ “is true” in a sentence is simply for one to pick out an antecedent claim and endorse it. 10 To say that what is claimed is true is thus not to attribute a property to the claim; it is to adopt a normative stance toward it. One can, of course, pick out an antecedent claim in a variety of ways: by quoting it, describing it, and so on. For example, to say, “‘Snow is white’ is true,” or even to say, “The standard example of a true sentence in analytic philosophy of language is true,” is just to endorse the claim that snow is white. With respect to the socially articulated game of deontic scorekeeping, for a scorekeeper to take the claim she attributes to another speaker as true is just for the scorekeeper to undertake the same propositional commitment attributed to the speaker (Brandom 1994, 515). The same goes for the use of expressions such as “refers.” For Brandom, when one employs locutions like “refers” in a sentence, this should not be taken as the expression of a word-world relation. Rather, by using “refers” in sentences such as, “The one you referred to as ‘that dullard’ is actually a chess Grandmaster,” one links that sentence to singular terms already introduced by claimings made in a discursive context. In other words, the expression “refers” and its cognates allow for speakers to continue a chain of singular term tokenings in the game of deontic scorekeeping, to inherit inferential content from one another, and to attribute and undertake responsibility for the use of singular terms in making various claims (Brandom 1994, 322). In short, for Brandom, truth and reference are not, as they are for the representationalist, to be understood in terms of word-world relations. Rather, “true” and “refers” are expressively valuable bits of vocabulary “that make explicit essential features of the complex process of coordinating our linguistic scorekeeping” (Brandom 1997a, 154). 11
But now, the normative attitude of taking something to be true cannot, of course, be identified with a claim’s being objectively true of what it is about. For Brandom, what is objectively true, or exactly what a claim refers to, is such that we might all be wrong about it. But how can this idea be cashed out from within the practice of deontic scorekeeping? This is the question raised above—the question not pursued by Rorty: “how can we institute a dimension of assessment of our sayings and doings that is properly understood as granting semantic and epistemic authority over their correctness to how it is with the things that we count as thinking and talking about?” Following Brandom, the answer to this question is already implicit in each scorekeeper’s ability to specify the conceptual content of other people’s commitments in both de dicto and de re form. For, since each scorekeeper takes the claims she endorses to be true or objectively correct, this ability is nothing other than the ability of each scorekeeper to distinguish between claims that are objectively correct and those that are only subjectively taken to be so (Brandom 1994, 595). 12 Indeed, all scorekeepers draw this distinction for themselves, from their own point of view, in keeping track of the commitments undertaken by their fellows. Extrapolating from this idea, Brandom thus argues that objectivity consists precisely in this kind of “perspectival form,” and not in an appeal to nonperspectival contents or to worldly facts as they are in themselves, independently of our social practices. As he puts it, what “is shared by all discursive perspectives is that there is a difference between what is objectively correct in the way of concept application and what is merely taken to be so, not what it is—the structure, not the content” (1994, 600). In other words, objectivity is a function of the inherent perspectivalness of discursivity. Brandom thus concludes that the concept of objectivity need not be abandoned by antirepresentationalists in favor of Rortyan “solidarity” or “agreement,” and criticisms of the “spinning in the void” type can be met from within a social-pragmatist approach to semantics.
However, the criticism has been leveled at Brandom’s rehabilitated, social-pragmatic concept of objectivity that, because it is rendered only at a normative and semantic level, because it is engineered in the game of giving and asking for reasons, it doesn’t include something we would normally associate with objectivity, namely, the idea of something that is “not in our control” constraining our thought. In response to this problem, it has been suggested that Brandom’s inferentialism needs to be supplemented with a pragmatic conception of experience, of the kind endorsed by Dewey and James. Steven Levine (2010), in particular, argues that Brandom’s social-pragmatic conception of objectivity fails to capture an essential feature of objectivity, namely, that what is objective constrains our thought in a way that is beyond our control. A pragmatist need not, however, return to the metaphysical realist’s notion of objectivity. Rather, by elaborating a thick notion of experience in which the rational relations characteristic of the space of reasons are understood as internal to our sensory and bodily responsiveness to the world, pragmatists can formulate a satisfactory notion of “objective constraint from within.” On this account of a Brandomian style inferentialism, which is part and parcel of a thick notion of experience, the rational relations constitutive of the space of reasons will not operate in a “frictionless void.” Rather, they will be understood as sedimenting states of an active body that is always coping with world-generated problems. Moreover, this notion of experience would in no way reinstate some form of the “Myth of the Given,” since what is given “is not the world as it is ‘in itself’ but the world as it is given to a being whose experience is structured by the concepts, skills, and capacities acquired through their previous world coping”—conditions that are themselves plastic and “change due to learning and habituation” (Levine 2010, 587).
Although he does not spell out the detail of this notion of experience, which is shot through with rational relations (apart from a reference to Dewey’s work), 13 Levine’s suggestion offers a promising, pragmatic approach to objectivity that combines insights from both classical and neopragmatism. This suggestion, moreover, finds support in contemporary pragmatist literature. Michael P. Wolf (2008), for example, argues that semantic and epistemological antirepresentationalism ought to be supplemented with a certain understanding of “answering to the facts,” which he understands not as a word-world relation, but as a kind of embodied “responsiveness” to the world, which implies a co-development of bodily and cognitive capacities. 14 While he does not address the “problem” of objectivity, Colin Koopman (2007) takes on board the work of pragmatist critics of both foundationalism and linguisticism and advocates combining—in a form that still avoids the “Myth of the Given”—a classical pragmatist account of experience with the insights of post-linguistic-turn pragmatists like Rorty and Brandom. In particular, Koopman argues that experience must be understood in Deweyan fashion as an “experiment with temporal duration” in which knowledge and meaning develop. Finally, Richard J. Bernstein (2010), while endorsing Brandom’s pragmatic approach to objectivity, nevertheless laments the loss of the classical pragmatist notion of experience (119–123, 125–152). In particular, he thinks that a “linguistic pragmatism” that does not incorporate serious reflection about the role of experience is threatened by a slide into an unsatisfactory and unsustainable “linguistic idealism” (152). In short, while much of the detailed work remains to be carried out, the contemporary pragmatist scene is set for the development of a new “experiential” conception of objectivity, which is consistent with the semantic antirepresentationalism of neopragmatists such as Brandom.
It is at this point that Deleuze’s work can profitably be brought into the conversation. It is the contention of this essay that with his concept of “problematic Ideas” in Difference and Repetition, Deleuze gives us a thick, pragmatic notion of experience that is consistent with an antirepresentationalist—and indeed, a Brandomian—approach to sense or meaning. I will argue for this in what follows. In so doing I will focus on the “Image of Thought” chapter in Difference and Repetition, as this is where Deleuze’s discussion of the dimension of sense is located, and where his particular brand of antirepresentationalist thinking is most prominent.
Let us first of all examine Deleuze’s treatment of the notions of sense, truth, and reference. Like Brandom, Deleuze contends that the representationalist’s privileging of the propositional dimension of designation over sense is deeply flawed. His argument is as follows: we ordinarily distinguish two dimensions of a proposition—“expression,” whereby a proposition says or expresses an idea or something ideal, and “designation,” whereby a proposition indicates the objects to which the expressed sense applies. However, it is also generally agreed that truth and falsity have to do with what words and sentences designate and not with what they express. Sense thus appears as the “condition of the truth” of the proposition, and yet at the same time it must remain indifferent or exterior to what it conditions (Deleuze 1994, 153). 15 In response to this problem, Deleuze asserts that
[t]he relation between a proposition and what it designates must be established within sense itself: the nature of ideal sense is to point beyond itself towards the object designated.… If sense points beyond itself towards the object, the latter can no longer be posited in reality exterior to sense, but only at the limit of its process. Moreover, the proposition’s relation to what it designates, insofar as this relation is established, is constituted within the unity of sense, along with the object which realizes this unity. (1994, 154)
Now, Deleuze’s general claim here is strikingly similar to Brandom’s: truth and reference are not “foundational” but are rather only explicable in terms of a process that generates sense. However, and although this is not immediately evident, Deleuze and Brandom also share a more specific claim, namely, that the process of producing sense involves a social practice of both undertaking commitments and attributing commitments to others. This becomes clear if we consider Deleuze’s implicit uptake in these pages of some of Frege’s (1997) work in On Sinn and Bedeutung. 16 Indeed, as for Frege, Deleuze argues that sense is an ideal entity that is expressed by a declarative sentence, and that is distinct from the words used, the object designated, and the subjective thoughts of the speaker (1994, 154–155). As he then proceeds to argue, alluding to Frege’s discussion of sense and reference in “indirect contexts” (i.e., in ascriptions of propositional attitudes), the sense of a first speaker’s declarative sentence, while not identical with the words used, can always be denoted by a second sentence uttered by another speaker. Moreover, since this second speaker’s utterance expresses a sense that is not identical with the words used, the sense of the second speaker’s utterance can always be taken as the object of a third utterance. The third utterance, in turn, expresses a sense that is not identical with the words used… and so on, ad infinitum (Deleuze 1994, 155; Deleuze 1990, 28–29). While in On Sinn and Bedeutung Frege is primarily interested in resolving a number of logical problems connected with specifying the mode of connection between sign, sense, and reference in indirect discourse, Deleuze is more interested in the “infinite power of language” to speak about itself and in this way to progressively determine the sense of what we say and, through this, the relation between what we say and what we are talking about. This is, I would argue, part of the meaning of Deleuze’s claim cited above, namely, that a “proposition’s relation to what it designates, insofar as this relation is established, is constituted within the unity of sense, along with the object which realizes this unity.” In other words, Deleuze is arguing here that, if a first speaker’s declarative sentence expresses a sense that is (in Frege’s terms) a “mode of presentation” of its referent, when a second speaker’s sentence takes the sense of the first sentence as its object, it expresses a sense that is a “mode of presentation” both of the sense of the first speaker’s sentence, and of the referent of the first speaker’s sentence. If this interpretation of the discussion of sense and reference in Difference and Repetition is correct, this brings Deleuze’s position very close to Brandom’s indeed. In other words, the interpretation being suggested here is that for Deleuze, as for Brandom, the process of making sense of what we are saying is simultaneously the process of semantically determining what it is we are talking about, and this process is a matter of an open-ended and social practice of undertaking commitments and attributing commitments de re to one another.
Deleuze then proceeds to shift the focus of the discussion from the “empirical” operation of sense production—the concrete act of taking the sense of a proposition as the object of a second proposition—to what he calls the “transcendental” operation of sense. He writes:
[S]ense is the veritable loquendum, that which in its empirical operation cannot be said, even though it can be said only in its transcendental operation.… While it is true that we cannot express the sense of what we say, we can at least take the sense of a proposition—in other words, the expressed, as the designated of another proposition—of which in turn we cannot express the sense, and so on to infinity.… However, the inability of empirical consciousness here corresponds to the “nth” power of language and its transcendent repetition to be able to speak infinitely of or about words themselves.… Every proposition of consciousness implies an unconscious of pure thought which constitutes the sphere of sense in which there is infinite regress. (1994, 155)
Deleuze is here arguing that due to an implied infinite regress of sense, sense is not equivalent to what is produced at any particular point in the ongoing and concrete activity of determining the sense what we say and what we are talking about. In other words, “the unity of sense, along with the object which realizes this unity,” is not secured by the empirical operation of sense. Sense rather points beyond itself to a certain “transcendental operation” or “unconsciousness of thought.” Now, we have explored Brandom’s argument that what he calls deontic scorekeeping is a “quasi-transcendental” feature of our communicative practices, which both underlies the production of meaning and explains our notions of truth, reference, and objectivity. The “transcendental operation” that Deleuze understands as subtending the production of sense, truth, and reference, however, is “thicker” than Brandom’s model of deontic scorekeeping, for it involves a number of other elements. Essentially, Deleuze’s position is that true, semantically contentful propositions, along with their objective referents, all emerge as “solutions” through the working out of an underlying “problem” or “problematic Idea” that involves all of the faculties implicated in the production of experience, with language and sociability being only two such faculties (Deleuze 1994, 157).
Deleuze writes that “sense is located in the problem” and that “problems are Ideas” (1994, 157, 162). These problematic Ideas, moreover, “traverse all the faculties” constitutive of what Deleuze calls “real experience” (1994, 146). The main faculties that Deleuze discusses are sensibility, memory, imagination, and thought, as well as language, vitality, sociability, and faculties that have “yet to be discovered” (1994, 143). These faculties are discordant. They “communicate,” but they do not function harmoniously by all bearing on an object supposed the same for each faculty. 17 Rather, each faculty only communicates to another a “difference” or “sign,” the sign being understood simply as “the bearer of a problem” (Deleuze 1994, 140).
For Deleuze, the communication of difference, signs, or problems between faculties commences when the faculty of sensibility encounters something in the world (1994, 139). What is encountered, however, should not be thought of as a particular and discrete worldly entity that causes us to have a perception of how it is. Rather, what is encountered is simply difference: not something that is different, but “free or untamed” difference. As he puts it, what is encountered is “not a sensible being but the being of the sensible… not the given but that by which the given is given” (1994, 140). In other words, contrary to a certain form of the “Myth of the Given,” sensibility does not grasp, perceive, or intuit some worldly fact. It rather grasps that which can only be sensed: an “intensive” difference that is not a particular “something.” This difference, as Deleuze then puts it, “moves the soul, ‘perplexes’ it—in other words, forces it to pose a problem: as though the object of encounter, the sign, were the bearer of a problem—as though it were a problem” (1994, 140). In short, it is such difference that, beginning in sensibility, is communicated from one faculty to another and sets each faculty to work.
So, it is clear that in the discordant communication of difference between the faculties the so-called Myth of the Given is avoided at every level. There is no direct communication between the faculties, which is to say that no particular object or form is transmitted from one faculty to another. The only thing that is communicated is a difference or problem, and confronted with such “free” difference, each faculty, one after the other, is set in motion. 18 Moreover, since no preestablished object or form is imparted from one faculty to another, each faculty must deal with what forces it into activity in the terms which are unique to it. But what this means is that, in confronting difference and being forced to deal with it, each faculty is forced to grasp something that is not so much external to that faculty as its own internal and unique form of difference: something that could only be grasped by that faculty in its own terms, but also something that differs from everything which that faculty effectively grasps once it is set in motion by the confrontation with difference. Thus, for example, in being provoked by sensibility to recollect, the faculty of memory will be forced to recall some “immemorial” source of its recollecting; the imagination, in turn, will phantasize about the essentially “unimaginable” source of the phantasizing it is forced to undertake; and so on (Deleuze 1994, 141–144). Deleuze calls each faculty’s own, unique form of difference its “limit- or transcendent-object” (1994, 146), in order to indicate both that it is something that can only be approached, not reached, and that approaching this paradoxical point is the communication of difference or a problem to another faculty and a charging of this faculty with the ultimately impossible task of resolving it in its own terms.
So, it is through this differential communication between the faculties that we must understand problems or problematic Ideas to traverse all the faculties. But now, what does it mean to say that the production of sense is located in this larger problematic Idea? And what does it mean to say that propositions endowed with sense emerge as solutions to this problem, along with the objects to which propositions refer and which realize their truth?
With regard to the production of sense, we have seen that the faculty of language must confront its loquendum: that “sense” that cannot be said in the empirical exercise of language, but that implies a thicker, transcendental operation. And we now know that this transcendental operation involves the communication of a problem between the other faculties constitutive of experience: sensibility, imagination, memory, thought, and so on. It is thus as a moment in the chain of discordant faculties traversed by the problematic Idea that the production of sense takes place. In other words, the empirical and socially articulated operation of the determination of the sense of a related series of propositions, while nevertheless taking place entirely within the faculty of language, takes place because of the communication of a larger problem from one faculty to another. Or again, provoked by problems flowing in from sensibility, the imagination, and so on, language is forced to formulate words and sentences and to talk about the meaning of these words and sentences.
But now, it is at this point that we can begin to see how a true proposition endowed with sense emerges as a solution to a more fundamental problem, along with the relation of reference between the proposition and that object that realizes its truth. We have seen that for Deleuze, as for Brandom, the empirical and socially articulated operation of making sense of what we are saying is simultaneously the semantic determination of what it is we are talking about. Moreover, as Brandom argues, the distinction between discursive perspectives in this operation gives us a certain conception of objectivity, in that this distinction implies a difference, recognized in each perspective, between “what is objectively correct in the way of concept application and what is merely taken to be so.” However, when Deleuze argues that a true proposition endowed with sense should be understood to be a “solution” to a “problem,” we must understand that the problem in question largely overflows the empirical operation of sense. 19 Indeed, the problem or problematic Idea is said by Deleuze to be a kind of transcendental field. Moreover, it is a transcendental field that is not limited to the normative and linguistic dimension of deontic scorekeeping, but rather concerns all of the faculties constitutive of experience and the discordant relations between them. What, then, does it mean to say that a true, semantically contentful proposition is a solution to a larger problem or problematic Idea? In short, it is to say that this proposition contributes to the resolution of experiential differences and problems flowing in from sensibility, memory, the imagination, and so on. And in this way, as I argue, Deleuze not only combines a Brandomian approach to meaning with a thick notion of experience, he also provides us with an antirepresentationalist conception of objectivity that includes the idea that there is something “out of our control” constraining our talk about things. How exactly is the resolution of difference flowing in from the other faculties achieved? I would suggest that the argument that needs to be made here is in fact a pragmatic one, namely, that the resolution of differences occurs when what a faculty produces in its empirical operation is acted upon in such a way that the experiencer ceases being “shocked” by difference, or successfully “copes” with her environment. With regard to the faculty of language and the empirical operation of sense, to say that language speakers have resolved the differences being communicated to them by sensibility, memory, the imagination, and so on, is to say that, by arming themselves in their practical activity with particular contentful commitments, language speakers have become able to better navigate their way through their physical and cultural world. Here, a certain experiential “responsiveness” in conjunction with a socially articulated process of sense making licenses language speakers to affirm that the objects and states of affairs talked about in their semantically contentful propositions are, for all practical purposes, what they describe them as being. Again, this understanding of experience has nothing to do with the representationalist conception, according to which propositional content is derived from our episodic grasp of real, discrete, and independently existing objects, which cause us to have true representations of them. It is rather a question of practical, pragmatic coping with difference and problems.
Finally, in order to round out the argument that Deleuze combines in his conception of problematic Ideas a “thick” notion of experience with a Brandomian approach to semantics, we can note that, for Deleuze, the “exploration of Ideas” is inseparable from a particular type of practical experience, namely, “an essential apprenticeship or process of learning.” The apprentice, as Deleuze characterizes her, is the one who “constitutes and occupies practical or speculative problems as such” (1994, 164). He writes: “Learning is the appropriate name for the subjective acts carried out when one is confronted with the objecticity [objectité] of a problem (Idea),” and the apprentice is the one who “raises each faculty to the level of its transcendent exercise” (1994, 164–165). It is thus through the ongoing experience of the apprentice that we must understand the emergence of sense, truth, and reference. In other words, the relations constitutive of sense and objective truth and reference are part and parcel of a thick notion of embodied experience. This is, it seems to me, precisely what pragmatists such as Levine are trying to achieve by combining a Brandomian-style inferentialism with a classical pragmatist account of experience. What Deleuze helps us to do is, within an antirepresentationalist position, marry an account of the way in which the empirical and socially articulated operation of sense constrains the way in which we talk about what there objectively is, with the idea that this operation is itself constrained by something that is beyond our control, namely, an involuntary process of learning or apprenticeship. Indeed, Deleuze would likely reject Brandom’s account of the way in which normative and semantic relations alone give us an account of objective truth and reference. What is missing is an account of being constrained in what we say by experience—not an experience whose conception smuggles in some form of the Myth of the Given, but an experience of apprenticeship or learning in which a shock to thought forces us to constitute and resolve a problem. As Deleuze writes,
Concepts only ever designate possibilities. They lack the claws of absolute necessity—in other words, of an original violence inflicted upon thought; the claws of a strangeness or enmity which alone would awaken thought from its natural stupor or eternal possibility: there is only involuntary thought, aroused and constrained within thought, and all the more absolutely necessary for being born, illegitimately, of fortuitousness in the world.… Do not count upon thought to ensure the relative necessity of what it thinks. Rather, count upon the contingency of an encounter with that which forces thought to raise up and educate the absolute necessity of an act of thought. (1994, 139)
In short, and although the word “objectivity” is not one he typically employs, it appears that a suitably pragmatized Deleuze has something to offer contemporary pragmatist attempts to provide a renewed concept of objectivity within semantic and epistemological antirepresentationalism. Objectivity, for Deleuze, has to do with the way in which a complex process of experience constrains us to talk about the world in specific ways. Deleuze in no way leaves our talk “spinning in the void” of pure conceptuality, since what we take to be our semantically significant, true propositions are resolutions of a problem or problematic Idea that is communicated through all of the disjointed faculties of experience, beginning with sensibility’s fundamental encounter with difference.
I would like to acknowledge the helpful feedback I received from a number of people while writing and revising this article. Particular thanks are owed to Simone Bignall, George Duke, Colin Koopman, and Cathy Legg.
1 For an excellent comparison of Rorty’s and Deleuze’s antirepresentationalism, see also Paul Patton’s contribution to this volume.
2 This image and criticism comes, of course, from McDowell (1994).
3 In Philosophy and the Mirror of Nature (1979), Rorty derives this conception from the theory of meaning advanced by Davidson in works such as “Truth and Meaning” (see especially 259–266, 299–305 and 308–311).
4 On the difference between Rorty and Brandom on the importance of “getting things objectively right,” see also Stout (2007).
5 Wanderer gives a very good account of this game-playing model in his monograph on Brandom’s work (2008, 41–48).
6 Entitlement has a “default and challenge structure” (Brandom 1994, 176).
7 A useful summary of the inferential “translation” of these normative relations can be found in Esfeld (1999, 337).
8 For a good analysis of the dependence of Brandom’s theory of reference and objectivity on his theory of communication, see Prien (2010).
9 For the full account, see Brandom (1994, chapter 8). A useful summary of this distinction and its role in Making It Explicit can be found in Brandom (1997b) and (2000, 169–182).
10 On Brandom’s “anaphoric” account of truth and reference, see Brandom (1994, chapter 5) and (1997a).
11 A good discussion of Brandom’s antirepresentationalist approach to truth can also be found in Stout (2002).
12 This argument is clearly explicated in Stout (2007, 24–25), and Prien (2010, 452).
13 The reference in question is to Dewey’s “The Need for a Recovery of Philosophy” (1981). Here, Dewey does indeed talk about experience as being “full of inference” (61). However, the inferences Dewey is interested in are those from “present on-goings” to the future consequences of such processes, insofar as the experiencer actively and intelligently helps to bring these consequences about (69). In other words, for Dewey, to say that experience is shot through with inference is to say that the experiencer deliberately and intentionally participates in the direction of the course of affairs by making “use of natural occurrences for the discovery and determination of consequences” (1981, 69–70). Such inferences, of course, appear markedly different to the types of inference that translate in discursive terms the relations constitutive of Brandom’s normative pragmatics. More work is thus required to spell out how a Brandomian inferentialism can be joined to a Deweyean account of experience. As will be seen, the present essay argues that Deleuze’s account of problematic Ideas in Difference and Repetition effectuates a coupling of this kind. There is perhaps an argument to be made that there exists in Dewey’s work the resources necessary for establishing a coupling of inferentialism and experience analogous to the Deleuzian one presented here. In particular, Dewey’s account of problems and “problematic situations” in chapter 6 of his 1938 Logic—The Theory of Inquiry bears some resemblance to Deleuze’s account of problematic Ideas in Difference and Repetition. But again, the detailed work necessary for developing and defending this thesis has yet to be undertaken. I thank Colin Koopman for raising and discussing this point with me.
14 To be precise, Wolf’s particular target in this essay is Rorty’s antirepresentationalist approach to meaning. Moreover, toward the end of the essay, he argues that Brandom’s work provides us with an important social-pragmatic conception of “answering to the facts.” He leaves it unclear, however, how Brandom’s work gives us a conception of “answering to the facts,” which is akin to the kind of “embodied responsiveness” he calls for in the first part of the essay.
15 For a related criticism, see Brandom (1994, 69).
16 Deleuze does not explicitly refer to Frege in Difference and Repetition. He does, however, reiterate the arguments made in Difference and Repetition with regard to the sense/reference distinction and the “paradox” of the infinite proliferation of sense in The Logic of Sense, published just one year later. In this later text, Frege is explicitly cited, along with Carnap’s criticism of Frege in Meaning and Necessity. See Deleuze (1990, 28–29, 337–338).
17 Deleuze thus rejects the paradigmatic form of the Myth of the Given: when a “sensing” is (illegitimately) held to be already conceptually contentful, the object of sensation is held to be the “same” as the object of the understanding.
18 “There is… a serial connection between the faculties and an order in that series” (Deleuze 1994, 145).
19 “The [problematic] Idea… cannot be reduced to sense, since in turn it is also non-sense” (Deleuze 1994, 155).
Bernstein, Richard J. 2010. The Pragmatic Turn. Malden, Mass.: Polity Press.
Brandom, Robert B. 1994. Making It Explicit: Reasoning, Representing and Discursive Commitment. Cambridge, Mass.: Harvard University Press.
Brandom, Robert B. 1997a. “From Truth to Semantics: A Path through Making It Explicit,” Philosophical Issues, 8: 141–154.
Brandom, Robert B. 1997b. “Replies,” Philosophy and Phenomenological Research, 57(1): 197–201.
Brandom, Robert B. 2000. Articulating Reasons: An Introduction to Inferentialism. Cambridge, Mass.: Harvard University Press.
Brandom, Robert B. 2011. Perspectives on Pragmatism: Classical, Recent and Contemporary. Cambridge, Mass. and London: Harvard University Press.
Deleuze, Gilles. 1990. The Logic of Sense. Translated by Mark Lester with Charles Stivale. Edited by Constantin V. Boundas. New York: Columbia University Press.
Deleuze, Gilles. 1994. Difference and Repetition. Translated by Paul Patton. London: Athlone.
Dewey, John. 1938. Logic—The Theory of Inquiry. New York: Henry Holt and Company.
Dewey, John. 1981. “The Need for a Recovery of Philosophy.” In The Philosophy of John Dewey, edited by John J. McDermott, 58–97. Chicago: University of Chicago Press.
Esfeld, Michael. 1999. “Review of Making It Explicit,” Erkenntnis, 51: 333–346.
Frege, Gottlob. 1997. The Frege Reader. Edited by Michael Beaney. Oxford: Blackwell Publishers.
Koopman, Colin. 2007. “Language Is a Form of Experience: Reconciling Classical Pragmatism and Neopragmatism,” Transactions of the Charles S. Peirce Society, 43(4): 694–727.
Levine, Steven. 2010. “Rehabilitating Objectivity: Rorty, Brandom, and the New Pragmatism,” Canadian Journal of Philosophy, 40(4): 567–589.
McDowell, John. 1994. Mind and World. Cambridge, Mass.: Harvard University Press.
Prien, Bernd. 2010. “Brandom on Communication, Reference and Objectivity,” International Journal of Philosophical Studies, 18(3): 433–458.
Rorty, Richard. 1979. Philosophy and the Mirror of Nature. Princeton NJ: Princeton University Press.
Rorty, Richard. 1991. Objectivity, Relativism, and Truth: Philosophical Papers Vol. 1. Cambridge: Cambridge University Press.
Sellars, Wilfred. 1997. Empiricism and the Philosophy of Mind. Cambridge, Mass. and London: Harvard University Press.
Stout, Jeffrey. 2002. “Radical Interpretation and Pragmatism: Davidson, Rorty, and Brandom on Truth.” In Radical Interpretation in Religion, edited by Nancy K. Frankenberry, 25–52. Port Chester: Cambridge University Press.
Stout, Jeffrey. 2007. “On Our Interest in Getting Things Right: Pragmatism without Narcissism.” In New Pragmatists, edited by Cheryl Misak, 7–31. Oxford: Oxford University Press.
Wanderer, Jeremy. 2008. Robert Brandom. Stocksfield: Acumen.
Wolf, Michael P. 2008. “Language, Mind, and World: Why Can’t We All Just Get Along?” Metaphilosophy, 39(3): 363–380.