Skip to navigation – Site map
Book Review

Pietro Salis, Pratiche discorsive razionali: Studi sull’inferenzialismo di Robert Brandom

Milano-Udine, Mimesis, 2016, 332 pages
Massimo Dell’Utri
Bibliographical reference

Pietro Salis, Pratiche discorsive razionali: Studi sull’inferenzialismo di Robert Brandom, Milano-Udine, Mimesis, 2016, 332 pages

Full text

1As the subtitle reveals, this is a book devoted to one of the most distinguished philosophers of our time, Robert Brandom, a thinker whose importance for the debate on topics such as meaning, truth, rationality, objectivity, agreement, holism, compositionality – to name just a few of them – is hardly deniable. Thousands among articles and books have been dedicated in the last two decades to the evaluation of Brandom’s thought, but still considerably few are the works in the Italian panorama dedicated to him. In this respect, Pietro Salis’s book remedies this – and greatly so. Indeed, the book not only gives us a perspicuous illustration of a multi-faceted philosophical system, carefully showing how its many features combine together, but highlights what turned out to be its controversial points, discussing them in full and putting forward original solutions to them. An essential Brandom with improvements, we might say.

2Accordingly, the book divides into two parts. The former lays down the foundations of Brandom’s inferentialist program – based on the idea that the content of thoughts and the meaning of sentences are a result of our inferential abilities and the social practices these abilities engender. Not any inference is relevant in this respect, though. Only “materially good inferences” can serve the purpose of giving content and meaning to our utterances, because what we are talking about here are not just the inferences logically valid – the ones analyzed by logicians on a purely formal ground – but the ones that can be considered good on the basis of the non-logical concepts they involve. Moreover, the inferences at the center of Brandom’s system are nonmonotonic, i.e. inferences in which new premises can affect the validity of an argument involving them. And this is another trait that distances this kind of inferences from the ones of classical logic. Relying then on what he calls “Frege’s pragmatic principle” – according to which when a person asserts a sentence, she commits herself to the truth of the sentence, implicitly acknowledging that, if challenged, she must give a reason in favor of her assertion – Brandom develops his celebrated illustration of the activity known as “giving and asking for reasons.” This is a particular linguistic game in which a speaker, for the sheer fact of making an assertion, takes a commitment before her (actual and potential) interlocutors: the commitment towards the truth of the assertion. It therefore follows that, if the speaker is able to give a justification for what she said, then she gains an entitlement to her assertion. In their turn, commitments and entitlements determine a score that has to be assigned to the assertions in a given discursive practice. Salis stresses how Brandom’s is a systematic conception of the human discursive practice, and how the special linguistic game of giving and asking for reasons is what rationality amounts to. This is what allows Brandom to claim that human rationality possesses an inherent pragmatic character.

3The first part of the book ends with an explanation of what the notions of representation and reference are within Brandom’s inferentialist framework. According to Salis, here resides one of the most original features of the framework, in that it does completely without primitive notions of reference, truth and representation in the account of meaning and discursive practice, thereby setting itself against large part of the received view – which is decidedly representationalist. This is what makes Brandom one of the leading anti-representationalist philosophers nowadays, arranging the stage for a bold alternative explanation of the representational dimension of our assertions: “the intentional directionality of thought and language, in Brandom’s perspective, is clarified thanks to a particular de re conception of propositional attitudes ascriptions” (16). This is particularly important considering the danger inferentialism must avoid of putting human rationality on a mere linguistic basis, quite irrespectively of the world. So, let us briefly see what this de re conception of propositional attitudes ascriptions amounts to, given that it is the central element in Brandom’s task of reconstructing the notions of representation and reference on expressivist grounds.

4The just mentioned danger is quite clear to Salis. Indeed, he stresses how the notion of materially good inference is not enough to dispel the risk of subjectivism and relativism, since this kind of inference is based on the “perspectival character of the individual epistemic and doxastic commitments” (96), and it is just that risk that that perspectival character paves the way for. Relying on the well-known distinction Quine drew between de dicto (to believe that) and de re (to believe of) ascriptions of propositional attitudes, and having clarified that usually we use a de re ascription when we want to make explicit an implicit divergence between our commitments and the ones made by our interlocutors, Brandom suggests “a methodical transformation of the (referentially opaque) de dicto ascriptions into the (referentially transparent) de re ascriptions by means of which differentiate one’s own commitments from the others’ via the reference to factual data” (103). This transformation is particularly useful in cases of misunderstandings and ambiguities: it is in cases like these that the de re ascriptions can make explicit what a speaker’s statement is about, clarifying commitments and entitlements and updating the deontic score. Accordingly, such a transformation is able to guarantee a satisfactory foothold for the evaluation of our arguments, because it would place our assertions within a verifiable – thus debatable – domain. The description of the theoretical link connecting the de re ascriptions of propositional attitudes to the inferentialist conception of meaning – a link centered on the key notion of substitution – is one of the greatest merits of Salis’s analysis.

5The second part of the book contains four essays on some hotly debated topics linked to inferentialism: Brandom’s version of alethic deflationism, the relationship between inferentialism and semantic holism, its relationship to compositionality, and the notion of objectivity. All the essays aim at defending Brandom’s stance, and they manage to do this by means of what proves to be interesting original arguments.

6Just to give an example of how hotly these topics are debated, let us give a quick look at Brandom’s understanding of truth and what a possible qualm may be raised about it. Salis introduces it in the course of his account of the difference which separates Brandom’s inferentialism from Michael Dummett’s justificationism. As is well-known, Dummett advanced a fierce criticism to the traditional non-epistemic conception of truth (and so does Brandom), claiming that it is useless in order to explain how speakers grasp the meaning of the linguistic expression they use. Truth must be something that we epistemically master, according to Dummett, and subscribing to an epistemic conception of truth led him to embrace anti-realism, given that in his perspective what reality amounts to follows from our demonstrative and justificatory abilities. Brandom, however, opts for alethic deflationism, an option opposed to Dummett’s – because alethic deflationism utterly denies that truth can be an “explanatory” concept, thereby denying that, e.g., it can enter in an explanation of the meaning of linguistic expressions. Salis illustrates the features that differentiate Brandom’s from other varieties of alethic deflationism – Brandom’s is a version of the so-called prosentential theory of truth: anaphoric deflationism. Among these features there is “maintaining a dimension that transcends the subjective and intersubjective acknowledgment of what is true” (128): this would allow the prosentential theory of truth to save a typical realist intuition, i.e. the idea according to which the way things stand might be independent from what we know or can in principle know about them. Indeed, philosophers endorsing the prosentential theory deem it “possible to clearly and explicitly distinguish the dimension of truth from that of justification” (128). This is revealed by sentences that have the same circumstances of application, but different consequences of application. It is correct to assert “I will marry Ophelia” in the same circumstances in which it is correct to assert “I believe I will marry Ophelia,” so that both share the same justifications for their assertion, but a priori they do not have the same consequences of application. If Ophelia unfortunately died in a week or so, the former sentence would be false, whereas the latter would still be adequate in the moment it was uttered. And this, according to Salis, shows the transcendental character of “true” in respect of “justified”: “the only justifications supporting an assertion are not sufficient to account for its meaning” (129) and truth – contrary to Dummett.

7Now, I think that the transcendental dimension is a very important feature of truth that, however, the prosententialists are unable to give an account for. This happens because the prosentential theory of truth (and alethic deflationism in general) is neutral toward that transcendental dimension. If Brandom manages to explain this dimension, as he does, it is thanks to his particular version of inferentialism. It is inferentialism that helps making a distinction between the justifications an assertion can currently have and the justifications it can have in principle, as Salis himself beautifully shows elsewhere in the book. The inferential role of a sentence S, according to Brandom, “does not simply boil down to the justifications Js for S, but extends also to its consequences Cs” (124): we may discover in the future that the inferences we are accustomed to use regarding S are different, and find ourselves in the need of changing our inferential use of S. This suffices to speak of a realist character of Brandom’s inferentialism. After all Brandom himself, Salis recalls, applies the term “conceptual realism” to the metaphysical position stemming from his inferentialism: “the idea according to which reality is already conceptually structured, and our knowledge of reality extends so far as our grasp of the concepts that we use in order to describe and understand it finds a space” (124; cf. also 198). So, it is inferentialism, not prosententialism, that possesses the necessary resources to do justice to the “structural gap between truth and justification” (129). Of course, this gap should be recognized at the level of truth as well, and failing to do this might reveal a weak spot in prosententialism.

8However, Salis may have a rejoinder to this. Since truth is – as we may sensibly claim – the most objective concept we have, and since the final chapter of the book is devoted to the task of averting the charge of subjectivism that might be raised toward Brandom’s inferentialism, an argument in favor of the suitability of prosententialism for vindicating the realist character of truth may derive from the arguments of that final chapter. So let us see what Salis says on behalf of Brandom there.

9The challenge facing inferentialism (the one we have been considering above) is something Brandom is quite aware of, and “becomes to say how the commitments and entitlements involved can be understood to have conceptual contents that are objective in the sense of not reducing to what someone or everyone is committed or entitled to” (Brandom, Reply to Bob Hale and Crispin Wright’s “Assertibilist Truth and Objective Content, in B. Weiss & J. Wanderer (eds), Reading Brandom: On Making It Explicit, Milton Park, Routledge, 2010, 361). Meeting this challenge would then suffice to show that inferentialism can intelligibly distinguish between “being right and merely believing one is right” (194), making enough room for a notion of objectivity that does not come down to intersubjectivity – let alone subjectivism. Salis’s strategy is ingenious: he starts admitting that not only the subjectivist danger is facilitated by the perspectival nature of speakers’ commitments, but it is even worsened by the nonmonotonic and counterfactually robust character of material inference itself. Indeed, what this character brings about are strong epistemic asymmetries among participants in the discursive practice, due to the contribute of a specialist nature made by experts within that practice, which gives rise to an imbalance among the commitments participants undertake. Therefore, Salis notices, at first blush Brandom’s deontic scorekeeping model might appear epistemically inadequate, since those asymmetries seem to require additional expressive resources. As a matter of fact, however, this is not the case: on closer inspection, the purported additional expressive resources turn out to be internal to the dynamic of the game of giving and asking for reasons. From this more refined perspective, both the afore-mentioned asymmetries and the imbalance they trigger guarantee an expressive surplus to the discursive interactions in the practice, showing the existence of a transcendental notion of objectivity functioning as a sort of regulative ideal of epistemic progress. But – and this is the chief point made by Salis – since this transcendental idea of objectivity is structurally embedded in the discursive practice, Brandom’s model appears to be “perfectly capable of reaching the very idea of objectivity” (207), defeating once and for all the charge of subjectivism.

10How could all this have an impact on Brandom’s notion of truth? It seems to me that whether or not the idea of objectivity stemming from the discursive practice is attributable to truth itself remains controversial, and so remains the further question regarding whether or not this move would be tantamount to recognize an explicative nature to truth, not only an expressive one. However, what is certain is that Salis’s book is a great contribution to the debate on this and other deep topics currently at the center of the philosophical discussion.

Top of page


Electronic reference

Massimo Dell’Utri, « Pietro Salis, Pratiche discorsive razionali: Studi sull’inferenzialismo di Robert Brandom », European Journal of Pragmatism and American Philosophy [Online], XI-1 | 2019, Online since 19 July 2019, connection on 21 October 2019. URL :

Top of page

About the author

Massimo Dell’Utri

University of Sassari

Top of page


Licence Creative Commons
Author retains copyright and grants the European Journal of Pragmatism and American Philosophy right of first publication with the work simultaneously licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Top of page
  • Logo Associazone culturale Pragma
  • Logo DOAJ – Directory of Open Access Journal
  • Logo École des Hautes Études en Sciences Sociales – EHESS
  • OpenEdition Journals