Reasons for Logic, Logic for Reasons: Pragmatics, Semantics, and Conceptual Roles
Sybren Heyndels · The Philosophical Quarterly · 2025
The aim of Reasons for Logic, Logic for Reasons (2024) is to explain the nature of thinking and talking in terms of reason relations. The two fundamental types of reason relations are implication relations (‘reasons for’) and incompatibility relations (‘reasons against’). These, in turn, are understood in terms of four metavocabularies that specify them in different ways. Over the course of the book, Hlobil and Brandom develop these metavocabularies and systematically investigate the relations between them. The result is a ‘metarational functionalist’ (p. 99) account of consciousness and language in terms of the four roles that reason relations can play. Two of these four metavocabularies are (1) a semantic metavocabulary that makes use of alethic modal expressions and (2) a pragmatic metavocabulary that makes use of deontic normative ones. People already familiar with Brandom's work will be familiar with these metavocabularies, as they characterize Brandom's account of the two poles of intentionality. Whereas subjects are understood as taking claims to stand in relations of normative inclusion and exclusion, the rational practice of doing so is understood as the act of representing objective facts as standing in alethic modal relations. This leads to a ‘bimodal hylomorphic conceptual realism’ (p. 167), according to which there is an isomorphism between (1) our normative practice of treating claims as reasons for and against one another and (2) the objective world, which is understood not as a collection of things but as a collection of facts that stand in objective alethic modal relations. Mind and world are said to share a ‘rational form’. Although Brandom has worked out these ideas before, Reasons for Logic, Logic for Reasons (2024) is an effort to spell them out in a much more precise manner. The pragmatic metavocabulary being developed is a normative bilateralist one that includes not only the distinction between the two normative statuses of commitment and entitlement but also the distinction between acceptance and rejection. As a response to Harman's arguments against directly reading off the norms of reasoning from reason relations, it is argued that reason relations indirectly constrain our reasoning practices. They do not directly tell us what to do, but they tell us at least what we are precluded from doing. The preferred semantic metavocabulary is a Finean truth-maker semantics, which explains reason relations in terms of the possibility or impossibility of fusing worldly states. An important technical result is the demonstration in the fourth chapter of the book that there is an isomorphism between the normative bilateralist framework and a version of Finean truth-maker semantics. These two metavocabularies are understood as extrinsic metavocabularies that give explanations of reason relations in either deontic normative or alethic modal ways. They are to be distinguished from intrinsic metavocabularies that do not have an explanatory but rather an explicative role. Although the distinction between explanation and explicitation deserves a more elaborate discussion than the authors provide, the general idea is that an intrinsic metavocabulary allows one to explicitly talk about the reason relations that we implicitly take for granted—an important first step towards possible criticism—while an extrinsic metavocabulary allows one to explain reason relations either in terms of normative constraints on reasoning practices or in terms of possible or impossible fusions of worldly states. The first intrinsic metavocabulary is logical vocabulary [though it is not ‘purely metalinguistic’ (p. 18), as it allows one to convey metalinguistic information while remaining in the object-language], which is characterized as a vocabulary that can be fully elaborated from, and is explicative of, any autonomous base vocabulary. To do justice to the logical expressivist's ideal of logical vocabulary being explicative of any autonomous base vocabulary at all, the challenge becomes to develop a logic that can codify so-called ‘substructural’ base vocabularies that include reason relations that are not ‘topologically closed’ (p. 89) and even reason relations that are not ‘explicitation closed’ (p. 89). Giving up the former entails allowing for reason relations that are non-monotonic and non-transitive. Non-monotonic reason relations are ones that can be defeated by adding further premises (e.g. ‘Tweety is a bird’ is a reason for ‘Tweety flies’, but the conjunction of ‘Tweety is a bird’ and ‘Tweety is a penguin’ is not a reason for ‘Tweety flies’). Failure of monotonicity leads to failure of simple transitivity (e.g. the conjunction of ‘Tweety is a penguin, so Tweety is a bird’ and ‘Tweety is a bird, so Tweety can fly’ is not a reason for ‘Tweety is a penguin, so Tweety can fly’). Abandoning explicitation closure further involves giving up even the principles of Cautious Monotonicity and Cumulative Transitivity. As a result, even a change in the status of a set of sentences as conclusion [‘implicit content’ (p. 84)] to premise [‘explicit content’ (p. 84)] can change the status of a reason relation. The third chapter develops in detail a logic the authors call ‘NMMS’, which is a non-monotonic and multi-succedent logic that can be elaborated from and is explicative of base vocabularies with radically open, substructural reason relations. The second intrinsic metavocabulary is a new implication-space semantics, which is a model-theoretic semantics that captures content in an implicational role on a more abstract level than the pragmatic and semantic metavocabularies discussed earlier: It abstracts away from the ways in which reason relations can be explained in terms of the acts of acceptance and rejection or in terms of the (im)possibility of fusing worldly states. I have two comments about the general story told by Hlobil and Brandom. The first concerns their supposed adherence to what they call a ‘pragmatics-first’ (p. 55, p. 100, p. 303) approach to language. Such an approach is said to argue that ‘all there is to confer meanings or contents on linguistic expressions is the (actual or proper) use of those expressions’ (p. 3) and that ‘the pragmatic attitudes of acceptance and rejection, and the speech acts of assertion and denial that express them, are conceptually and explanatorily prior to the concepts of truth and falsity’ (p. 55). These remarks are reminiscent of the Brandom of Making it Explicit, but it seems to me that the actual story being told in Reasons for Logic, Logic for Reasons is much more congenial to a story in which this methodological commitment to the explanatory priority of acceptance and rejection over truth and falsity is abandoned in favour of a view that sees neither pair of concepts as more fundamental than the other. While the possibility of treating ‘the semantic and pragmatic dimensions of discursive practice as explanatorily coeval’ (p. 32) is mentioned, it is not endorsed. However, in light of their central claim that the bilateralist pragmatic explanations and the semantic truth-maker explanations are isomorphic and to be understood as two ways in which rational forms can be ‘enmattered’ (p. 290), it is unclear in what sense pragmatic explanations still have priority over semantic explanations. If pragmatic and semantic explanations are two sides of the same coin, why continue to treat one or the other as more fundamental? Furthermore, Hlobil and Brandom repeatedly suggest that it is the intrinsic implication-space semantics that is the most fundamental metavocabulary [even calling it their ‘candidate for the language of angels’ (p. 294)], given that it captures reason relations in their purest form by abstracting away from the two ways in which they can be enmattered. A second comment concerns the role of alethic modal vocabulary. Given that alethic modal expressions play a central role in the preferred extrinsic truth-maker semantics, which explains reason relations in terms of the possibility or impossibility of fusions of worldly states, Hlobil and Brandom seem to imply that alethic modal vocabulary is an extrinsic vocabulary with an explanatory function. However, such vocabulary also seems to fit the job description of logical vocabulary, which is interpreted as being an intrinsic vocabulary with an explicative function. After all, alethic modal statements make explicit what remains invariant under a range of subjunctive robustness. Given that ranges of subjunctive robustness are essential to the reason relations that govern autonomous base vocabularies, modal vocabulary seems to be a good example of logical vocabulary—what Hlobil and Brandom call a ‘hybrid or mongrel kind of metavocabulary’, which ‘plays the expressive role of explicating reason relations’ (p. 18). This is a rich and ambitious work that presents the most detailed account yet of some of the main ideas Brandom has been working on over the last few decades. It includes, among many other things, the most elaborate exposition of logical expressivism to date, it develops a new implication-space semantics, and it offers interesting new perspectives on the relation between mind and world. It is a must-read for those interested in the latest developments of Brandom's philosophical journey, and a great display of Hlobil's excellent continuation of this journey.1