The next important building block of Brandom’s inferentialism is his analysis of deontic status in terms of deontic attitudes – but I’ve sort of already covered that in broad brush strokes, last year on the blog.

Also worth mentioning the priority of judgements in Brandom’s account of meaning: Brandom does not understand judgements in terms of their composition out of smaller meaning-units, but understands smaller meaning-units in terms of the roles they play in judgements. (And of course a judgement, for Brandom, can only be a judgement as part of a larger social system of judgements.)

So those are two further explanatory approaches that Brandom’s project is committed to. I’m not going to spend much time on them here.

In addition, though, for Brandom [and here we start to move into territory I’ve not covered already on the blog; if you see something you regard as a mistaken interpretation, please feel free to flag it as such] we become concept-mongers by entering the space of reasons; by participating in the social practice of asking for and giving reasons.

This is Brandom mobilising Sellars. I’ve read Empiricism and the Philosophy of Mind – I’m not totally clear what other Sellars I should look at to get an adequate sense of the resources Brandom’s making use of / transforming. Recommendations would be welcome.

What is the explanatory order here?

Brandom thinks that concepts need to be understood in inferential terms. That is, roughly – the meaning of a concept is the things we can infer from it. To be very slightly more precise: concepts can for Brandom only be understood in relation to judgements. A judgement is something that can be expressed in a proposition. So the concept of a dog, for example, needs to be understood in terms of the role the concept of a dog plays in propositions about dogs. Sticking at the propositional level, then – a proposition needs to be understood in terms of the inferences we can make from it. Or, to be very slightly more precise still: “Two claims have the same conceptual content if and only if they have the same inferential role.” (MIE 96)

What is an inferential role? Brandom understands inferential role in terms of deontic statuses – which, as we’ve already seen, he in turn understands in terms of deontic attitudes. There are, I think, two fundamental deontic statuses in Brandom’s account: commitment and entitlement. I’ll need to spend a lot longer with these in future posts. But Brandom’s basic idea is that an assertion of a proposition binds one to certain actions (including the endorsement of other propositions), normatively. Those other actions – including those other propositional contents – follow from the assertion – they can be inferred from it – and this binding or inference is itself the content of the original proposition. The meaning of a judgement is that other judgements or actions are implied by it. Content is explained in terms of implications, rather than implications in terms of content.

Where does this binding or this implication come from? What is it?

Brandom’s claim here, if I understand him aright, is that the brute fact that we are animals who ask for and give reasons is the source of the bindingness of commitments etc. – and thus of normativity. That means we are creatures who ask each other for demonstrations of entitlement. This is a brute fact about us. And the content of an entitlement – indeed, the very nature of entitlement itself – needs to be understood in terms of this social practice of challenge and response.

We are creatures who, discursively, ask each other for reasons – that is, challenge social entitlement – and who feel the force of such challenges. That we feel this force is a brute social fact, on Brandom’s account. That we engage in a practice in which challenges of entitlement are responded to by giving reasons for entitlement – this is a brute fact. Brandom says nothing about what counts as a good reason – this is to be hashed out entirely in the social practice of asking for and giving reasons itself. But the basic challenge-response structure of entitlement to assertions is a brute (biologically evolved) fact about the nature of the human organism – and this fact, more than any other, is generative of our capacity to wield conceptual content.

Note that ‘entitlement’ is a deontic status, and is thus to be explained in terms of deontic attitudes – attributions of entitlement. So we’re not positing a mysterious thing called ‘entitlement’ here. We’re saying that as evolved organisms we attribute social statuses to each other, challenge those attributions, and that, when challenged, we characteristically offer reasons (are capable of doing so – this capacity is determining of sapience; the claim is not that we always do offer reasons; still less that those reasons are always good reasons). Entitlement can be explained in terms of behaviour. But the behaviour itself is distinctive – the ‘attribute; challenge; response’ structure of that behaviour is distinctive.

This is very confused as formulated here. But I want to start to work through this aspect of Brandom’s apparatus. I’ll aim to continue with that in my next set of posts.

So – as I say, in this series of posts I am going to push Brandom’s pragmatics into the background for once, and focus on his inferentialist semantics. There may initially be some replication of earlier blog content here, but we’ll be moving into new terrain soon enough.

What is inferentialism?

Brandom contrasts inferentialism, as a philosophical explanatory strategy, with representationalism, which Brandom regards as the dominant modern philosophical tradition. Representationalism takes the concept of representation as explanatorily fundamental, in our accounts of meaning. Inferentialism, by contrast, takes the concept of inference as explanatorily fundamental. What we’re trying to explain here is sapience, or rationality, or ‘concept-mongering’ (all of which come to the same thing, on Brandom’s account – though of course from the point of view of other theoretical perspectives these concepts might not be equivalent).

Of course there’s no intrinsic need to take either representation or inference as derivative of the other – both concepts could be explanatorily fundamental. But Brandom thinks he can show that reference is derivative of inference – thereby going against much of the modern analytic tradition.

So Brandom starts with inference. That’s our first point.

Second point is the idea of material inference. Brandom contrasts the idea of material inference to that of formal inference – and again the question is one of explanatory priority. A formal inference is one that obeys an explicitly formulated rule of inference, which rule is applicable independent of the content of the inference. A material inference, by contrast, is an inference the goodness of which depends on the content of the claims being inferred from and to.

The modern philosophical tradition, as Brandom characterises it, tends to explain apparently material inferences in terms of formal inferences, by suggesting that apparently material inferences rely on suppressed premises that, if made explicit, render the relevant inference context-independent. So, to use one of Brandom’s favourite examples, if I make the inference “It is raining, so it is wet outside”, a formalist would regard me as possessing a hidden premise here – “If it is raining, then it is wet outside” – and the inferential chain here runs “If it is raining, then it is wet outside; it is raining; therefore it is wet outside”. The actual inference them becomes a purely formal logical one – If X then Y; X; therefore Y – and the actual content of the inference is simply plugged in to this formal operation. The inference itself is independent of that content.

Material inference, by contrast, is inference where the content itself matters for the inference itself – where the inferential move should be understood not in terms of a hidden premise that renders the inference a formal logical one, but in terms of one proposition simply implying another, by virtue of the content of the propositions themselves, without any additional mediating operation.

Brandom believes that material inference is explanatorily prior to formal inference – that formal inference should be explained in terms of material inference, not the other way around.

Brandom has arguments for this – notably Wittgenstein’s regress of interpretations, which Brandom takes to show the unacceptability of formalism (or, in his vocabulary, ‘regulism’); a similar argument is made in Lewis Carrol’s ‘What Achilles Said to the Tortoise’, which Brandom also cites. But Brandom’s arguments aren’t specifically my concern here – for now I’m just interested in articulating his position.

More in my next post.

Picking up on Brandom again

February 8, 2012

[Still insufficient time and/or focus to blog in a sustained way – but a few quick remarks on Brandom]

When I left off discussing Brandom, I think I’d basically given a decent account of his normative pragmatics. This pragmatics is the main thing that interests me in Brandom’s work. However, I haven’t yet given here, I don’t think, an adequate account of Brandom’s inferentialist semantics – elaboration of which occupies the bulk of Brandom’s corpus.

The rhetorical and work-allocation situations here are both somewhat delicate.

To recapitulate (apologies to readers who have seen this a thousand times already here…):

My interest in Brandom is in the repurposing of his broad pragmatist and practice-theoretic insights into more overtly social-theoretic intellectual terrain. I want, in the longer run, to make contributions to social scientific research, rather than to semantics or the philosophy of language. I therefore simply am not going to attempt to acquire the mastery of the analytic logistic tradition that would be required to do Brandom’s contribution to that tradition justice. Doing so would require something close to a life-project level of commitment, I think – time measured in decades rather than years. If I were one of those immortals who populate popular culture (some sort of vampire or Highlander figure…) this might be worthwhile; but give the average human life span this project just isn’t compatible with my other intellectual goals.

I also, as regular readers will know, hold an opinion (sufficiently under-justified to possess at present an epistemic status only slightly higher than that of prejudice) that the broad structure of Brandom’s argument could be replicated by an apparatus that does not require Brandom’s specific linguistic-philosophical commitments. [This is a stronger claim than the one I have advanced at length elsewhere on the blog – that Brandom cannot justify the transcendental status he ascribes to some of his truth-claims. I consider this latter claim to be quite strongly warranted; but even if I’m right there, that doesn’t speak to whether an alternative apparatus would in fact be possible.]

[NB: At one level Brandom is comfortable with this. He writes in the Preface to MIE that “Particularly in matters of detail (but by no means there alone), a myriad of choices have had to be made at the cost of spurning attractive, perhaps ultimately superior, alternatives. The approach seldom dictates just one way of doing things. Yet the choice of which large limb to follow off the trunk of the tradition must be made on the basis of the tempting fruit to be seen on the smaller branches it supports. It can only be hoped that where upon closer inspection some of them are found wanting, the fundamental soundness of the tree is not impugned, but only the judgement of the gardener, who pruned the better and nurtured the worse.” I find this thematisation of fallibility to be both moving, and in the spirit of the larger pragmatist and scientific intellectual enterprise.]

This belief is independent of the fact that I don’t myself plan to make use of Brandom’s detailed inferentialist apparatus in my own future work – but it provides another motive for de-emphasising, relatively speaking, Brandom’s philosophy of language, as I deploy that apparatus for non-philosophical purposes. [Non-philosophical here meaning ‘not philosophical’ – nothing to do with Laruelle et al.]

Nevertheless, I want to know what I’m talking about; to possess some degree of competence in Brandom’s philosophy of language. In the coming series of posts I plan to explore Brandom’s inferentialist apparatus, and aim for a fuller – though I’m sure not a comprehensive – understanding of it.