Embodied Norms

March 26, 2011

I’ve said (repeatedly) before that Brandom’s pragmatist apparatus has greater scope than Brandom’s own emphasis on linguistic philosophy might imply. One consequence of Brandom’s emphasis on the linguistic is that it tends to obscure the extent to which Brandom’s work theorises the embodied nature of the normative. As I’ve said before, one of the core categories of Brandom’s philosophy is reliable differential responsive dispositions (RDRDs) – that is, dispositions for an entity to respond to different given stimuli or (more broadly) circumstances in reliably different ways. This is, at least potentially, a naturalistic category. The task Brandom sets himself, in Making It Explicit, is to explain how one can start with RDRDs and end up with normative and conceptual content – how to ‘bake a normative cake’ out of (potentially) natural ingredients.

When I first discussed RDRDs on the blog, I wasn’t in a position to give an adequate account of how Brandom aims to achieve this. It is not uncommon, in the secondary literature on Brandom, for interpreters to believe that Brandom has essentially fiated himself normativity. MIE builds its account of normative and conceptual content out of the base unit of “normative practices” – that is, practices that can be done right or wrong – and it therefore seems that Brandom is bringing in normativity at the ground floor, evading any serious explanatory task.

What this common interpretation fails to understand is that MIE‘s argument as a whole allows Brandom to explain the basis on which normative practices become normative. For Brandom the social property of being normative can be entirely explained in terms of the (naturalistically describable) social practice of taking as normative. Now, a practice is only normative if it is properly taken as normative – the taking-as practice that institutes normativity is itself a normative practice that can be done right and wrong. It may therefore seem that we have a vicious explanatory regress. But as I explained in my posts on Joseph Heath’s interpretation of Brandom, the potential regresses involved here are not the same as either a) an actual regress of justification, for any given justificatory project – because any given community will have axiomatic (doxic) shared presuppositions in which the chain of justification can be grounded (even if those and any presuppositions can be challenged from the perspective of another social-perspectival location); or b) a metatheoretical regress of explanation, since the entire set of practices that generate the possible challenges to any given norm can themselves all (in principle) be understood naturalistically [which, as I said in my post on Brandom’s definition of ‘naturalism’, is not the same as permitting a scientistic metalanguage to so describe them]. ‘Original intentionality’ (or original normativity) therefore resides, for Brandom, not intrinsically with any given practice, but with a community of practice – normativity can only exist as an emergent property of many complexly related social practices, some of which will then rightly be regarded as normative from within a set of perspectives made possible (as normative) by a more extensive set.

These remarks are meant to be recapitulatory – I’m not aiming to make the argument here, but rather to summarise the conclusion of my early long series of Brandom posts, to which I refer baffled readers. The point I’m making, for purposes of this post, is that although Brandom is widely regarded as an anti-naturalistic philosopher, who regards normative practices as ontologically fundamental for the purposes of his system, it is in fact legitimate to see Brandom as treating reliable differential responsive dispositions as fundamental, and explaining the social process of rightly taking RDRDs as normative in terms of RDRDs themselves. [And this is not an ‘eliminativist’ position because normativity remains fundamental (and real!) for us.]

So Brandom’s account of normativity is based on RDRDs – reliable differential responsive dispositions. The principal dispositions in question are of course dispositions of the human organism. (I’ve argued against Brandom’s human exceptionalism, and I’m also sympathetic to ‘extended mind’ accounts of cognition – but for purposes of this discussion at least, I’m assuming we’re basically talking about humans.) This means, of course, that Brandom’s theory of normativity is also in some sense a theory of embodied cognition. It’s this aspect of Bradom’s thought that I want briefly to explore.

That’s the first thing. The second thing is that Brandom has a theory of implicit norms. Normative practices can be normative even if we are not capable of articulating the norms that they implicitly contain – in fact, this is the default condition for normative practices. Again, I’ve discussed this (though in somewhat less detail) in earlier posts on the blog. I need to go into this issue in some greater depth at a later date – but for now the important point is that we don’t need to be aware, at any given moment or indeed at all, of what norm is contained in a normative practice for that practice to indeed be normative (the relevant taking-as doesn’t have to be a conscious assessment of the norm as norm). (We need to be capable of in principle articulating the norm, but that’s a separate issue.) Most norms that guide actions are implicit nearly all of the time. We do what we do through biological predisposition and social habituation. What we do is still normative – but it need not be consciously thematised to be so normative. Indeed, normativity itself would be impossible if our conscious thematisations of norms were not built on the solid bedrock of habitual and practically-invariant biological dispositions (although of course we can change the bedrock through conscious deliberation on habit or indeed at times biological dispositions, but again that’s a separate issue.)

Brandom’s theoretical and ethical inclination is towards explicitation: Brandom is a fully signed-up member of the Enlightenment project of bringing unconsidered habit into the gaze of wakeful theoretical consciousness, in order to assess the legitimacy of that habit, and modify our behaviour accordingly. It is this theoretical and ethical inclination that motivates Making It Explicit – the book’s theorisation and explanation of the process of explicitation is driven by the desire to make explicit how we make norms explicit, as an important element of this ethico-theoretical task. It’s important to note, however, that this theoretical-ethical orientation is not itself demanded by the apparatus Brandom ends up with. Making It Explicit simply tells us how (in Brandom’s opinion) the generation of normativity and (as part of that) the capacity to make implicit norms explicit to ourselves functions – MIE is silent on the question of what we should do with this theoretical knowledge (even if the work’s own practice serves as a model for a particular theoretico-ethical orientation.)

Indeed, it is a core aspect of Making It Explicit‘s theoretical account that we can’t make everything explicit – at least not all at once. The activities of theoretical consciousness are built out of practical capacities, for Brandom, and any intellectual endeavour, even the most Enlightenment-oriented, must rely on a vast array of habitual activities and premises held constant to enable its articulation. This doesn’t mean that those same premises and habitual activities can’t, at some other time, be thematised and contested (or justified) in their own right. But the very capacity for wakeful self-examination is grounded in a presently-unexamined set of wholly implicit norms.

This isn’t a paradoxical result, about the blindness of reason to itself – as I say, Brandom draws no such conclusion, nor should he, and he remains a fully signed up member of the classically-articulated Enlightenment project of testing our beliefs and actions in the light of reason. But it means that, for someone with this basic ethico-theoretical orientiation, Brandom is unusually sensitive to, and willing to theorise, the non-conscious, implicit aspects of human cognition and behaviour.

Which brings me to my point: that norms are, in the first place, for Brandom, implicit and embodied. The extent to which this is so can be overlooked because Brandom himself is so fixated on a specific class of normative practice – linguistic behaviour. But if we loosen the hold of linguistic philosophy over Brandom’s pragmatist apparatus for a minute, we can see that a lot of what Brandom is saying about normativity would not be incompatible with, for example, Pierre Bourdieu’s discussions of habitus. It is this aspect of Brandom’s work that I want to focus on for a short while.

Advertisements

33 Responses to “Embodied Norms”

  1. deontologistics Says:

    A great post here. I think the differences between our interpretations of Brandom are starting to be whittled down, as there is nothing here that I disagree with as far as I can see (beyond your sustained commitment to the non-centrality of linguistic practice). However, I think you still need to come to terms with the transcendental side of Brandom’s project. Yes, the institution of rationality is embodied in certain specific practices, but there is *something* general which is embodied in this specific way, and for want of a better word this is transcendental. Just because there have been many poor attempts to make transcendental arguments does not mean that all transcendental arguments are doomed to failure, or that they all must depend upon failures of imagination.

  2. duncan Says:

    Hi Pete – thanks for this – I’m glad we’re basically on the same page. On the two points of difference you mention:

    1) W/r/t the non-centrality of linguistic practice, there are two different things here that I’m probably not being careful enough to differentiate in these posts. The first is just that Brandom himself is interested in analysing linguistic practice, and puts the overwhelming majority of his attention into this area. There is, however, of course, nothing to stop one analysing non-linguistic practice within a fully Brandomian framework. Brandom obviously doesn’t deny that non-linguistic practices can have normative practice status, so for most practical purposes all I’m urging is that it’s possible to broaden the scope of Brandom-influenced research beyond Brandom’s personal research interests, which I can’t imagine Brandom would have a problem with. In the passage from Between Saying and Doing that I quoted a couple posts back, for example, Brandom refers to the phenomenon of “pragmatic projection”. He calls it (interpreting Wittgenstein) “an absolutely fundamental discursive phenomenon” and gives (Wittgensteinian) examples from linguistic practice. I agree with all of this – but “pragmatic projection” is of course also a phenomenon that is present all over the place in non-linguistic practice. I don’t take myself to be disagreeing with Brandom in any way when I say this – Brandom is aware of the scope of the category – I’m just saying that Brandom’s own research interests lead him to emphasise a certain set of consequences of his metatheoretical apparatus, and I’m interested in emphasising others. This can be done in full compatibility with both the letter and the spirit of Brandom’s work, I think.

    For most of the things I want to use Brandom’s apparatus for, that’s all I need. However, there’s also a separate issue, which is that (as you know), I’m dubious about Brandom’s claims that only linguistic practice can get him his derivation of normative and conceptual content. This is a substantive disagreement, though it’s one that I may never try to make good on, as it were. The issue here is over what kinds of practice a community possessing ‘original intentionality’ would have to participate in. Brandom thinks that ‘original intentionality’ can only emerge from a community engaged in linguistic practice, and that in explanatory terms the taking of non-linguistic practice as normative is a derivative phenomenon. I’m sceptical about this – I think Brandom underestimates the possibility for non-linguistic (empathic) inhabiting of alternative subject-positions, as an alternative mechanism by which multiple sets of books can be kept in a social process of deontic scorekeeping (which is the basis for normative and conceptual content, on Brandom’s account). However, the fact of the matter is that for almost all practical analytic purposes this disagreement is also incidental. In fact we are linguistic creatures, participating in a community of linguistic practice – our own attribution of normative statuses often does involve linguistic practice – and since, for Brandom, once ‘original intentionality’ is up and running, we can then attribute intentionality and normativity to any kind of practice at all (we may do so improperly, of course, but that’s a downstream issue not determined by Brandom’s metatheory) it actually doesn’t matter for a Brandomian analysis of real non-linguistic practices where we locate the constitution of ‘original intentionality / normativity’, provided we’ve got it (which obviously we do). Since the work I want to be doing with Brandom’s metatheory is not itself ultimately going to be operating at a metatheoretical level, it should be possible for people to ignore this disagreement, if they’re committed to the full Brandomian apparatus, without this having much in the way of consequences for other claims I’m making. In that sense, the emphasis I’m placing on my disagreement with Brandom over the status of linguistic philosophy is probably unwarranted – I really only keep mentioning it because I find it helpful to keep in mind when working through something as complex as MIE (and perhaps in hopes that I’ll eventually be able to cash out my intuition here in more detailed terms – but it’s actually not important for my broader project that I be able to do so).

    2) Something similar applies on the transcendental issue. I have my own set of intuitions here: I find it genuinely plausible that something like rationality couldn’t exist without some sort of deontic scorekeeping – this seems to me to be a credible candidate for a transhistorical invariant, common to all sapient creatures. Contrariwise I (again) find it implausible that the linguistic practices Brandom catalogues in MIE are the only way in which such a scorekeeping practice could be instituted – I think Brandom’s placing far too much weight on specific human institutions. I could, of course, be wrong about either of these things. However, my objection to the transcendental nature of Brandom’s argument is more fundamental – even if Brandom is right about the practices that are necessary for the institution of rationality, I don’t see how he could possibly know he’s right. My objection is to the “warranted” side of things, rather than, in the first instance, to the “true belief” side. I don’t know on what epistemological basis Brandom could demonstrate that there is of necessity no alternative set of mechanisms by which similar effects could be achieved – this seems to me to be just an unknowable thing. The best we can do, by my lights, is to say that every known instance of rationality participates in social practices of this kind, and that we can’t think of an alternative set of social practices that could achieve the same result. We might be right in inferring, from this, that no other set of social practices could achieve the same result – but I don’t think we can make this claim with anything like the strength, or on the kind of theoretical basis, that Brandom seems to want to.

    I know almost nothing about the field, so this is a hazardous comparison to make, and should therefore be treated with extreme caution – but an analogy might be drawn with scientific attempts to define the conditions necessary for something to count as a living organism. One can provide a definition of a ‘living organism’ that operates at such a high level of generality that it’s hard to imagine something that wouldn’t fall under the definition that we’d also be willing to call “alive” (if only for reasons of definitional tautology). But definitions of this kind operate at such a high plain of abstraction that they’re not generally, as I understand things, very descriptively (or analytically) useful. [An analogy with Brandom might be defining rationality as involving conceptual content – fine, probably (if only because we could treat such a statement as definitional-tautologous), but (for that very reason) it doesn’t tell us much.] Alternatively, one can try to talk about specific biological processes that might be required in order for life to be established [just as Brandom talks about specific social practices required for the institution of rationality] – but then one pretty quickly runs into the risk of simply being proved empirically wrong, as has happened with such definitions of life in the past. Of course, any given postulated requirement for the establishment of life (the presence of amino acids, say) might indeed be necessary (it’s a very plausible candidate). But it seems worth regarding such claims as extremely plausible hypotheses based on the observed evidence, rather than as proven necessity – because there’s always the possibility of running into some counterexample – in fact that possibility is a condition of the claim’s scientific status. Brandom’s work would be more scientific if it followed this methodological principle. I’m of course a fan of science, and I don’t regard the qualifications involved in classifying a claim as a plausible hypothesis compatible with all available evidence, rather than a logical necessity, as involving a practically-significant loss in strength of assertion. So there are really (again) two issues here, for me: on the one hand I regard certain candidates for ‘transcendental’ status that Brandom proposes implausible (though I may never cash out that critique in detail, by trying to elaborate potential counter-examples); on the other hand I just regard ‘transcendental’ status as too strong a category of claim for Brandom to be filing his own metatheoretical proposals in, however accurate they are.

    All this said, however, the same thing applies as with linguistic practice – for the purposes for which I aim ultimately to be deploying Brandomian resources, it probably doesn’t matter whether we take his categories as transcendental (in the weak sense elaborated in my earlier post on transcendental arguments) or not. Humans do indeed engage in linguistic practice, so whether we believe that’s because this is the way in which rationality happens to be instituted for us, or because we have to behave this way in order to count as sapient entities at all, is somewhat academic when it comes to the analysis of human society. So again, while I think there’s insufficient justification for us to commit oneself to the stronger set of claims, one can probably do everything I want to do while retaining the full Brandomian apparatus.

    I don’t know if that makes you less uneasy about the direction I’m rowing in here. We probably do just disagree on the transcendental issue, but I’m uncertain how much of practical (or theoretical) consequence the disagreement involves us in.

  3. deontologistics Says:

    Hi Duncan,

    I can see things from your point of view, and I see how your problems with Brandom’s linguistic rationalism and his (understated) transcendentalism don’t necessarily conflict with the project of using his ideas about norms for analysing social structures. I have some further issues with what you’ve said on both fronts, but first I think I should recommend Ruth Millikan’s work, who, although I don’t know her very well, criticises Brandom in much the same way you do, and tries to locate the structures of communication in biological functions as opposed to (or rather as the basis for) social norms.

    On to my issues:-

    1) On the topic of language, you say:-

    “The issue here is over what kinds of practice a community possessing ‘original intentionality’ would have to participate in. Brandom thinks that ‘original intentionality’ can only emerge from a community engaged in linguistic practice, and that in explanatory terms the taking of non-linguistic practice as normative is a derivative phenomenon. I’m sceptical about this – I think Brandom underestimates the possibility for non-linguistic (empathic) inhabiting of alternative subject-positions, as an alternative mechanism by which multiple sets of books can be kept in a social process of deontic scorekeeping (which is the basis for normative and conceptual content, on Brandom’s account).”

    I still think you’re unclear about precisely what you mean by linguistic, when Brandom is very precise about what he means. For Brandom, anyone who can play the game of giving and asking for reasons is a language user. It doesn’t matter what kinds of things they use as tokens, its the role that those tokens play in the game that makes them linguistics tokens (i.e., sentences). Here you suggest that there can be something like scorekeeping without token transmission between individuals. How, precisely, are these non-linguistic scorekeepers keeping score? In order for them to be genuinely tracking inferential relations between commitments and entitlements, they need to have certain ways of individuating commitments and entitlements, i.e., they need private tokens even if there are no public tokens. The question is then what this gap between private and public tokens is, and whether it can be sustained. If some outward display of behaviour is reliably linked with the ascription of a certain token then its hard to see how this wouldn’t then count as a tokening. Moreover, it’s hard to see how it’d be possible to keep score on the basis of ’empathy’ unless one did draw such reliable connections between the outward behaviour of those one keeps score on and the normative statuses one ascribes to them. In short, I think you’re trying to hold open a distinction (between language use and scorekeeping) which cannot but collapse.

    2) On the topic of the transcendental, you say:-

    “Contrariwise I (again) find it implausible that the linguistic practices Brandom catalogues in MIE are the only way in which such a scorekeeping practice could be instituted – I think Brandom’s placing far too much weight on specific human institutions. I could, of course, be wrong about either of these things. However, my objection to the transcendental nature of Brandom’s argument is more fundamental – even if Brandom is right about the practices that are necessary for the institution of rationality, I don’t see how he could possibly know he’s right.”

    The answer is that he can know he’s right (or be warranted in thinking that he is) if he has a good argument that justifies his position. Your challenge is thus that you don’t think that there can be such transcendental arguments distinct from empirical arguments about observable phenomena. I think you need to back this point up by actually looking at the structure of the transcendental arguments Brandom does supply and seeing where (or rather, if) they fall down due to making elicit empirical assumptions. However, I think that there are perfectly good examples of these kinds of non-empirical argument, and they correspond to non-empirical forms of possibility. To take a limit-case, mathematical arguments can quite obviously demonstrate necessary conditions without the need for empirical supplementation. The question is whether there is a way of talking about the practices of rationality that is completely independent of the way they are empirically instantiated that would then be sufficient to license similar claims to necessity. This essentially means talking about causal systems at a level of maximal abstraction.

    I happen to think that this is entirely possible, and that Brandom demonstrates it in BSD in his abstract discussion of abilities and the ways they can be algorithmically elaborated. I’m trying to develop this further at the moment into a more complete abstract model of rational systems. The crucial insight here is that it’s perfectly legitimate to use the *empirical categories* of causality, event, etc., in describing abstract causal systems as long as one doesn’t use *empirical concepts* such as synapse pathway, vocal resonance, and the like. Anyway, there’s much more work to be done on this front before any of this will see the light of day. For now, I’ll simply ask you to pay attention to the form of the arguments Brandom gives in BSD (and perhaps also in chapter 6 of MIE) and ask yourself whether they genuinely are dependent upon empirical assumptions, rather than simply assuming that they must be.

  4. duncan Says:

    Thanks Pete – I’ll check out Millikan’s work. Probably not for a while, since I’ve got other obligations floating around at the moment – but I take your reading suggestions seriously, and will work my way round to them eventually. Of course I like Brandom’s emphasis on the social – I’m just more interested than he is in non-linguistic sociality.

    On which topic! (And in the same order…)

    1) On the linguistic.

    I accept there’s some fuzziness about my definition of the linguistic – and I’m not too invested in any particular drawing of the line, so I’m relatively happy to shift usage according to the conventions and preferences of the people I’m interested in interacting with. So, for example, do a shepherd’s whistles, which communicate with a sheepdog, count as linguistic tokens? For some people they would absolutely count as linguistic (by virtue of the fact that they are tokens communicating commands / requests / information). For other people they would absolutely not count as linguistic (because there’s no grammar to such communicative acts, for instance.) I don’t really care which definition of the linguistic we’re working with, as long as we’re clear on what we mean. So I’m happy to operate with a broad definition of the linguistic, which would include such tokens (or indeed be defined by their presence) – this seems sensible.

    Now, I should emphasise that my ideas in this area are underdeveloped, so there’s an exploratory quality to these remarks. But I think there are two fairly different directions of response to your comments here. (I think the key thing you say in describing and contesting my position is: Here you suggest that there can be something like scorekeeping without token transmission between individuals.)

    First up, although I’m myself interested in social behaviour in the ordinary-language sense, I actually think Brandom’s official definition of the ‘social’ is much broader than the term’s ordinary-language meaning. Strictly speaking, for Brandom, a ‘linguistic community’ could exist within a single empirical individual, where the communicative tokens in question are things like memory traces, and the different social-perspectival locations are things like different attitudes adopted within the same consciousness. If I’ve got the right end of the stick (and it’s possible I don’t), Brandom says this somewhere in a ‘response to critics’ piece – I’d need to track it down. This may be an unwelcome consequence of Brandom’s theoretical apparatus, from Brandom’s point of view – I’m not sure – but I think it’s there.

    So, to take a rather overly speculative example, I don’t find it out-of-line implausible to think that a non-‘social’ creature (say a shark or something) might be capable of possessing something like ‘conceptual content’ despite not participating in a communicative community of mutual recognition. The relevant ‘social’ practices could in principle happen entirely internally to the individual organism. Further, I’d be inclined to think that a creature could inhabit the social-perspectival location of another creature (say, its prey), without the second (preyed upon) creature aiming to communicate (and, perhaps, actively wanting not to communicate) its own thoughts’ and feelings’ conceptual and normative content. Being able imaginatively to inhabit another’s subject-position, while keeping track of both a) what the other thinks they ought to do to avoid being eaten, and b) what the other actually ought to do to avoid being eaten, seems like a useful skill-set for a hungry predator. This is scorekeeping, with the keeping of multiple sets of books crucial to Brandom’s account, but it exists without a real community of communication as I understand it. Now one could argue that the expressive bodily behaviour that allows one individual to infer the normative content of another’s commitments counts as ‘linguistic’ – and if that’s a definition that enough people would accept, I’d go along with it – but it seems like a real stretch from the meanings that are generally operative in discussions of this kind. The analysis of the relevant practices would certainly look very different from the analysis of, e.g., sub-sentential units in MIE – which leads me to think that this is not a phenomenon that Brandom would comfortably accept as a linguistic one. I don’t think there’s anything in the prey’s expressive (inadvertent) communication of conceptual or normative content that could plausibly fill the role of a ‘sentence’, for example.

    Moving along from this, I’m also inclined to think (though again open to being persuaded otherwise) that various non-human primate communities engage in deontic scorekeeping. I was talking with ktismatics about related issues the other week, and I was saying that I really need to spend more time with the ethological literature if I’m going to make these sorts of claims. Nevertheless, my underinformed sense of things is that gorilla communities (for example) keep track of individuals’ social status in a manner plausibly describable as deontic scorekeeping. There’s communication involved in this, to be sure – and of course the communication involves tokens, in the sense of sounds and gestures etc. – so if that’s our definition of the linguistic, we’re dealing with a linguistic community. But I don’t think Brandom’s intricate analysis of sentence-structure etc. has much bearing on this kind of communication – so my sense, again, is that this would fall outside MIE‘s own operative sense of the linguistic.

    The issue, I suppose, is that while I agree that it’s very difficult to imagine conceptual content without token-transmission in some sense, Brandom’s account isn’t just talking about token-transmission – it’s talking about very specific structures of token-transmission and manipulation [his detailed analysis of anaphora, sub-sentential units, etc.], and my feeling is that the structures and practices Brandom analyses are simply too specifically tied to our own linguistic practices to be plausible candidates for necessary features of token-transmission in general (whether we regard the latter as definitionally linguistic or not). So I do think there’s a plausible gap between language use and scorekeeping, as those terms appear to be being used in Brandom’s work.

    2) On the transcendental issue. You’re right of course that empirical evidence is not the only kind of reason Brandom (or anyone) can offer for finding claims compelling – Brandom (or anyone) can also offer non-empirically grounded argument. Which reasons we take as good reasons and which as bad is, of course, a fraught topic – but any position that suggests that only empirical evidence counts as a reason is (in both our opinions) a pretty implausible / unsustainable one.

    To take the limit-case you mention, mathematical claims and arguments are of course non-empirical, but can still be true and valid. My view here, however (as I guess you might suspect) is a conventionalist one. I think that ‘mathematical necessity’ is a legitimate and useful concept, but that what it in fact denotes (whether or not it is taken as denoting this by most people who use it) is an extremely high strength of commitment to a specific set of discursive practices by members of a given community. This doesn’t, to my mind, reduce the legitimacy of the concept – nor does it mean that mathematical claims can’t tell us important and at times uncanny things about the real world. But I think the concept of mathematical necessity can be, so to speak, swallowed up by an inferentialist account like Brandom’s, and therefore cashed out in terms of social practice. Since, however, I regard social practices as empirical phenomena, I don’t think an analogy with mathematical necessity can be used to bolster an argument that such practices need not be discussed empirically.

    [It might be worth mentioning in passing that my understanding of ‘strong conventionalism’ isn’t actually incompatible with mathematical Platonism [I was part of a conversation about this last night, is why it’s on my mind] – just as (to my mind) the embedding of reference in an inferentialist account of meaning doesn’t mean there aren’t real things referred to. This is by-the-by, but it bugs me when conventionalism is taken as a form of anti-realism.]

    So yes – my claim is that I don’t think there can be legitimate transcendental arguments even of the weak kind that Brandom aims to make, because what the concept of ‘necessity’ names is a community’s commitment to maintaining a proposition’s positive truth-value come what may [this is basically Quine’s point about analytic statements, obviously], and this is in the first instance a fact about social practice, rather than in any immediate way about the world. Now of course Brandom’s whole project is to show that being a fact about social practice isn’t incompatible (and can in fact be constitutive) of being a fact about the world – so this doesn’t mean that the propositions to which the community is so committed are problematic or wrong – but it means that their ability to refer to (or usefully model) a trans-communal reality can only be tested empirically and pragmatically (assuming we adopt a broadly secular pro-scientific perspective, which I think we should).

    Now that last point could seriously stand expansion – I realise there’s a degree of question-begging in my concluding parenthetical. But I’m pushing against the limits of the day here. I guess I’ll conclude for now, and return to this set of issues when I have more time and headspace.

  5. duncan Says:

    Sorry – can’t quite let this go before sleep – in the jargon of BSD (which I’m still reading, I should emphasise), let’s say we believe we’ve isolated a set of practices that are PV-necessary for any vocabulary at all. What kind of claim is this? I accept that Brandom thinks it’s a very general claim – he’s talking about conditions of possibility of the discursive, there’s no question – but I also think that Brandom’s own apparatus gives an account of the status of this metatheoretical assertion that makes it in principle vulnerable to empirical contestation in a way that transcendental arguments classically aren’t supposed to be. And I think the scientific best-practice understanding of the status of such a claim (which I advocate adopting) is as best-hypothesis, rather than necessary truth. (It may be our best hypothesis as to what might be a necessary truth – but the point is it doesn’t have the same status as the synthetic a priori, which I take not to be meant to be vulnerable to empirical contestation.) Of course the question as what counts as ’empirical’ versus ‘conventional’ is hugely complicated by Brandom’s own metatheoretical apparatus, so this could really really do with expansion. But not now!

  6. deontologistics Says:

    Hi Duncan,

    I’m sorry if I drop too many references on you. I know what it’s like to have a towering pile of books to read that new ones keep getting added to. Millikan has just recently been added to mine, which is why I mentioned her. I recently read her paper on the relation between Sellars, Brandom and herself (The Father, the Son and the Daughter). It strikes me that this would be a great jumping in point for you (I can email you it if you like).

    On to my retorts:-

    1) You’re right that Brandom accepts that a discursive community of differing social perspectives can exist within a single individual, at least insofar as that individual can track the differences between its commitments over time (i.e., its perspectives at different times), but this doesn’t get you out of the importance of the tokens it keeps score with baring a certain structure.

    You’re obviously right that there are a bunch of fuzzy boundaries here, but I think that (as ever) we should strive to make them precise wherever possible. Personally I try to distinguish between language and communication. The latter can be understood in terms of information transmission between individuals, and the way these informational signals are tied into regularities of behaviour. Communication is more or less complex, and it’s certainly the case that whatever language is it’s a species of communication, and so emerges out of forms of communication that are less complex (in certain specific ways). The important point is that communicative behaviours that are proto-linguistic (as might be the case with primates) is that we understand them as privative forms of linguistic behaviours, which means that we need to understand what’s essential to language that they lack. Brandom’s answer is that language must involve a very specific kind of information processing, namely, the tracking of inferential relations between the commitments undertaken by the use of tokens in transmitting information that can thereby be construed as genuinely conceptual (and thus representational) content. A group of animals (or humans for that matter) might share a number of the abilities *necessary* to do this without thereby having abilities that are *sufficient* to do it. We can thus understand the relations between their proto-linguistic forms of communication and our own, but only on the basis of a characterisation of what these abilities are, which I take to be a transcendental matter (to be approached in a similar way to the way Brandom approaches it in BSD).

    So, in your example of the shark, we can say that it has certain abilities to *simulate* the actions of its prey on the basis of an intuitive grasp of the way it responds to various situations that the shark also has some grasp of, and we can accept that this is a complex form of information processing that is quite possibly necessary for genuine scorekeeping, but I don’t think it counts as scorekeeping. To put it in somewhat Kantian terms, the imagination (intuitive simulation) is no substitute for understanding (the ability to keep track of separate concepts) and reason (the ability to grasp their relations and deploy them within inferences).

    The point about subsentential units is especially apt here, as I think it’s possible to make a stronger argument than Brandom does for the necessity of subsentential structure, insofar as only subsentential structure allows for compositional projection of new sentences into new contexts, and for the acquisition of new concepts (and the new sentences one can construct with them) and the alteration of old concepts (and the changes this makes to the inferential relations between the old sentences they govern). Without this capacity one’s concepts (and thus one’s scorekeeping practices for keeping track of whole tokens) can’t adapt to the world properly, and thus can’t track it (or represent it) properly. Again, something that had hardcoded ability to keep track of the relations between tokens without discriminating their internal structure would count as displaying necessary but not sufficient abilities for full blown language use (which is representational, not merely informational).

    When it comes to the primate communities that seem to track eachother’s deontic statuses, I again think we’re talking about privative forms of the requisite abilities (we could go further into privation and talk about similar abilities in colonies of ants, that do discriminate on the basis of social status). The important point is that without abilities to scorekeep on inferential relations these communities don’t have the ability to keep track of the content of practical commitments, and they thus can’t be counted as instituting commitments that can be extended beyond familiar contexts (i.e., commitments that have ‘objective’ content in Brandom’s terminology). This isn’t to diminish the social achievement of primates, or to say that there is a huge empirical gap between their behaviours and our own, the difference is that we have abilities to adopt a completely different attitude to these practices, and this produces a normative difference in kind out of an empirical difference in degree. Again, the important thing is to get clear about the transcendental question of what the generic abilities that are required to institute deontic statuses are, and then relate these to empirical questions of who precisely has these abilities and how they are configured.

    Finally, the ultimate point (which I think people like Millikan and Dennett sometimes miss) is that Brandom is not trying to provide a universal account of communication (i.e., information), and thus doesn’t need to be sensitive to many of the empirical issues they are so concerned with, but a universal account of language (i.e., the conceptual) that could be situated in relation to such an account, but can equally be developed with a certain level of independence from it. This is the significance of Brandom’s point about explaining what the trick is, rather than how it is done.

    I’ll put up my thoughts on point (2) when I don’t have a train to catch!

  7. deontologistics Says:

    Aha! Free wifi on the train. Here’s my additional thoughts:-

    2) I think that your response isn’t just question begging but downright circular. Picking up on your conventionalist claims about mathematical necessity, you’re essentially saying that it’s indexed to what the community *does* hold fixed, rather than what they *should* hold fixed. There are of course all sorts of examples of cases where mathematicians show our previous intuitive grasp of some mathematical notion to be a limited form of a more general structure, such as the invention of non-euclidean geometries by way of the suspension of the parallel postulate, but these aren cases where we uncover deeper necessities which are actually present. This is indicated by the fact that in these cases we’re confronted with choices regarding the actual content of our original concept (e.g., the concept of geometric space or manifold), where we could decide to retain the term ‘manifold’ for the more limited phenomena and produce a new term for the more general one of which it is a case (as Lakatos shows most perceptively). This kind of broadening of horizons that is characteristic of mathematical reason is very different from the discovery that a proof is *wrong*. It amounts to the discovery of implicit premises contained in the content of the intuitive mathematical concepts we’re working with, which we then explicate and decide whether or not to rearrange. All of this is done at the level of the *ideal*, whereas only errors in mathematical reasoning (and the histories of the contingent arrangements we choose) need be understood in terms of the *real* behaviours in which it is instantiated.

    Yes, mathematical necessity can be understood in terms of empirically describable practices, but these are practices that instantiate a certain form of attitude transcendence. One hasn’t thereby reduced particular mathematical necessities to any regularities of behaviour, but merely described mathematical necessity *as such* in terms of the particular social perspectival *form* that those behaviours implement. It’s got nothing to do with what we’re *representing* when engaging in mathematical discourse, nor whether these things are *real* or not. It’s got everything to do with the normative structures in question. This means that we return to the question of whether or not these forms are properly transcendental, i.e., historically invariant conditions that must be met in order to count as a rational agent (or in this case, a mathematical reasoner). Hence, we’ve moved in a circle.

    One might charge that my position is then moving in a similar circle, insofar as transcendental forms of argument (which are non-empirical) depend for their validity upon the possibility of transcendental norms governing those forms of argument. There is indeed such a circularity there, but I believe it is virtuous, and that it is the point of my (still developing) method of fundamental deontology to elaborate it in a way that makes this virtuousness explicit. However, in the absence of a complete elaboration of my project, I can point out an important point of conflict between my circularity and your own. This is that my position is fully consistent with Brandom’s account of the ‘objectivity’ (again what I call objectivity is stronger than what Brandom calls it) of conceptual content, as it is manifest in its various forms (which I distinguish as various forms of truth), whereas your position implicitly contradicts it. You seem to want to endorse Brandom’s account of the objectivity of conceptual content in social-perspectival terms, but then to be able to step outside of it in order to say that ‘really’ the norms that are determined in this way are indeterminate and historically contingent. The problem with this isn’t that that perspective is illegitimate. Brandoms himself acknowledges it as involved within the very process of the historical determination of norms itself (at least, to some extent), but it’s only one perspective. Of course, from the empirical perspective, norms aren’t fully determinate, because the reasons for choosing which way to further determine them aren’t empirical reasons (except in the case of the norms that constitute empirical conceptual content), but the empirical perspective is not the only one we can occupy. Indeed, it’s not possible for us to just occupy that perspective without treating the norms of discourse themselves as indeterminate and thereby undermining the possibility of discourse (this is the basis of what I call the *primary bind*).

    In short, I think you’re problem is that you can’t see the conditions of the possibility of the empirical perspective (the game of giving and asking for reasons as restricted to empirical content), because you refuse to countenance the possibility of a non-empirical perspective within which one could articulate such conditions. However, at the very lest, logic, mathematics, and computer science are testimonies to the fact that such perspectives are available, and that they actually provide claims that are not only useful within the empirical domain, but can actually play a role in *constituting* empirical content in virtue of the relations between the practices in which these perspectives consist (this is what my hierarchy of types of truth/discourse in the TR Essay is meant to convey).

    As I’ll be trying to explain in my paper at Dundee (from a very different direction), we’ve got no choice but to do transcendental philosophy. The trick is just to move from the more or less implicit and intuitive forms of it found in Kant and his successors to a well defined method analogous to the transformation of mathematical practice that occurred at the beginning of the 20th century (and continued throughout it). If nothing else, Husserl should be valorised for trying to do this, even if his way of going about it is entirely wrong headed (focusing upon ‘givenness’ as opposed to ‘bindingness’). Hegel, Habermas, and Brandom have all in their own ways provided us with clues for how such a method must be constructed, but it these need to be synthesised into a systematic and foundational approach. Some people are allergic to this kind of philosophising, as they think it lacks humility. My response to this is quite simply this: fuck false humility. Philosophy should be done however it demands it must be done. Once we work out how philosophy must be done, our task is to do it that way. It is better to fail to meet a profound challenge than to succeed by compromising it’s profundity.

  8. duncan Says:

    Hi Pete – I’m super busy, and there’s a lot to respond to here, so I’ll respond very selectively for now, and circle back round when I have more time. (“Super busy” is not code for “don’t want to have this conversation” – I’m just super busy)

    Picking up on your conventionalist claims about mathematical necessity, you’re essentially saying that it’s indexed to what the community *does* hold fixed, rather than what they *should* hold fixed.

    No – I mean to be relying on (/ shorthanding) the discussion in the original post. So I’m not meaning to index necessity directly to communal practices, though I realise it probably sounds that way. I’m saying (to expand) that an inferential link possesses mathematical necessity if it is properly taken to possess mathematical necessity. What it means to properly take something as possessing mathematical necessity, on a Brandomian account, however, can in turn be understood in terms of actual acts of takings-as [from a given interpretive standpoint – which is itself an act of taking-as, of course] – while (at the same time) not being identifiable [except via an act of interpretation / taking-as that can itself be undertaken properly or improperly, the propriety or impropriety of which can in turn be explained in a precisely parallel way] with any given empirical act or acts of taking-as. The argument is formally parallel to (in fact is the same argument as) the one related to empirical reference – and I don’t think it devolves into regularism.

    There is a privileging of the empirical in my analysis, but that’s because I take it (and one could take a different route here) that the practices that one can use – the RDRDs – as foundational for the Brandomian apparatus should be understood as empirical practices. One could of course argue that the relevant foundational RDRDs are not empirical things (they could, in principle, be algorithmic, for instance). But that would be a different kind of objection, I think?

    More when I have time…

  9. duncan Says:

    You seem to want to endorse Brandom’s account of the objectivity of conceptual content in social-perspectival terms, but then to be able to step outside of it in order to say that ‘really’ the norms that are determined in this way are indeterminate and historically contingent.

    Yes, I guess that’s right – I think that the norms that are determined in this way probably really are historically contingent – that’s my position, and it’s one that I think is true. Of course, I don’t take “historically contingent” and “true” to be incompatible properties – I think the Brandomian apparatus can give a good account of why that’s so, and of how one can “step outside” a social perspective to the degree of making general reference- and truth-claims while still acknowledging the social-perspectival constitution of those claims.

    I want to quibble on “indeterminate”, however – I’ll have a think about how best to articulate it.

  10. duncan Says:

    This kind of broadening of horizons that is characteristic of mathematical reason is very different from the discovery that a proof is *wrong*. It amounts to the discovery of implicit premises contained in the content of the intuitive mathematical concepts we’re working with, which we then explicate and decide whether or not to rearrange.

    Yes, something like Brandom’s ‘common law’ model. But the point is that the implicit premises ‘discovered’ can be fully explained in terms of the interpretive acts of the community as a whole, even though no individual member of the community has sovereignty over the (non-empirical [but emergent from empirical practices, given a specific social-perspectival location]) content of the ‘discovered’ content. So the content is both genuinely discovered (by the individual) and wholly made (by the community, of which the individual is a member), without this meaning regularism. So in a certain sense it’s true to say that –

    All of this is done at the level of the *ideal*

    – but it’s certainly not true (imo) to conclude that –

    only errors in mathematical reasoning (and the histories of the contingent arrangements we choose) need be understood in terms of the *real* behaviours in which it is instantiated.

    It can all be understood in terms of the real behaviours in which it is ‘instatiated’ (/from which it is produced). It just that it can’t be so understood from a non-normative standpoint. The normative act of taking-as that enables this standpoint is also a real behaviour, though. In this sense, Brandom’s analysis is ‘symmetrical’.

  11. duncan Says:

    Ok, sorry, I’m going to stop there for now – I’ll definitely try to pick all this up when I have a spare minute. Hope the Dundee conference goes well…


  12. […] there’s Duncan Law’s recent post on Brandom – Embodied Norms – where we’ve been having a cracking good discussion about our different perspectives […]

  13. deontologistics Says:

    I’ll wait to respond until you’ve had a chance to sit down and work things out in more detail. I’ve linked to the post (and ensuing discussion) on my blog as I think we’re really hashing out some good stuff here. Always a pleasure to spar with you 🙂

  14. duncan Says:

    Ok – sorry to leave this hanging for so long.

    On sharks and primates and so forth –

    The important point is that without abilities to scorekeep on inferential relations these communities don’t have the ability to keep track of the content of practical commitments, and they thus can’t be counted as instituting commitments that can be extended beyond familiar contexts (i.e., commitments that have ‘objective’ content in Brandom’s terminology).

    The issue is that I don’t see on what basis we can confidently claim that the scorekeeping such non-human or non-linguistic communities engage in is non-inferential, and thus non-conceptual. If we understand the capacity to scorekeep on inferential relations as dependent on the practical ability to manipulate communicative tokens in the kind of sentential structures analysed in MIE, I can see why we’d conclude that non-human primates (for example) are incapable of possessing and communicating conceptual content. But this is the point at issue. I’m suggesting that it at least seems possible (I find it acceptably plausible) that alternative mechanisms of scorekeeping, that don’t rely on these kinds of token-structures, but that nonetheless are capable of tracking multiple sets of books of practical commitments, are achievable, and indeed may well be present in non-human (and indeed human) social interactions.

    I want to separate out the in-principle possibility of this situation from the question of whether actual non-human non-linguistic communities do exhibit this behaviour – the former is all I need to make my case. There is empirical work that at least claims to have located behaviour that would require scorekeeping capacity in non-human primates (e.g. this piece – Sarah F. Brosnan and Frans B. M. de Waal, Monkeys Reject Unequal Pay, Nature 425, 297-299 (18 September 2003) – which has been (compellingly) criticised all over the shop e.g. here – Roma, P. G. P. G., Silberberg, A. A., Ruggiero, A. M. A. M., & Suomi, S. J. S. J. (2006). Capuchin monkeys, inequity aversion, and the frustration effect. Journal of Comparative Psychology (Washington, D.C.: 1983), 120(1), 67-73. In general I’m just unconvinced that there’s enough good empirical research in this area to form any firm view either way with much scientific legitimacy – the conclusion of the second piece I just linked to (that “prudence dictates withholding endorsement of the existence of inequity aversion in nonhuman primates”) seems absolutely sound to me. (Of course my claim w/r/t non-linguistic normative behaviour would be weaker than Brosnan and Waal’s – I’m talking about scorekeeping in a very abstract sense, without making any claims as to what commitments (e.g. innate dislike of inequity) are regarded as proper and improper by the creatures keeping score. But even so.))

    The issue, though, is why we would regard it as implausible that non-human or non-linguistic creatures are incapable of keeping score in a manner that would enable inference in Brandom’s sense. As a counterfactual thought experiment taking its bearings from the claims of the paper above, let’s say you’ve got a capuchin monkey, or any creature at all, that’s in a social group a member of which is disbursing food. Monkey keeps track of the social status of other monkeys, including its sense of their entitlement to various foodstuffs, given the social dynamics of the group. It also keeps track of the behaviour of the food-disbursing creature, and tracks what it (our monkey) regards it (the food-disbursing creature) as committed to, given food-disburser’s previous behaviour and its (the monkey’s) sense of the dynamics of the group (deontic statuses, in other words). This is a normative judgement about counterfactual possibilities, and involves inferences in Brandom’s sense. (If food-disbursing creature has given grapes to monkey B, the creature has committed itself (according to monkey A’s sense of the normative statuses and proprieties governing its behaviour) to also giving grapes to monkey A – this is an inferentially and normatively contentful judgement of deontic status. It’s also an sense of status that can be pretty easily non-linguistically communicated, given the context, by monkey A getting pissed off when it doesn’t get grapes.)

    Why should we regard this behaviour as not involving or creating normative statuses or conceptual content? We can, of course, fiat (if we wish) that the kinds of normative attitudes that grant conceptual content must be linguistic if the outcome is to count as sapience – but then we are simply defining sapience as coterminous with language use, and the point becomes tautologous. I think a less tautologous way of drawing the (fuzzy) line is required. I think Brandom’s work can plausibly provide such a criterion: a community (including in principle a ‘community’ limited to the internal states of a single organism) counts as exhibiting sapience if it participates in practices of deontic scorekeeping and asking for and giving reasons (very broadly defined), including the keeping of multiple sets of books from different social-perspectival locations. [I’d be happy to entertain alternative accounts of sapience, obviously, but I find it plausible at least that any community that exhibits this set of properties ought to count as sapient by the meaning of that term we generally deploy.] I think that Brandom’s linguistic philosophy successfully (as far as I can tell at present state of knowledge) elaborates one set of social practices that can definitely produce the necessary building blocks of this account – normative deontic attitudes – and both precipitate them out of and elaborate them into (so to speak) a full set of sapience-enabling communal practices. I don’t think, however, that Brandom does or could demonstrate that language is the only kind of social practice capable of instituting these normative deontic attitudes. As we’ve discussed before, I don’t see what kind of (legitimate) authority such a claim could be based on.

    Now Brandom sometimes uses Popper’s falsification stuff when he makes bold claims of these sorts – he justifies his own theoretical orientation by the methodological rule of thumb that one should try to make the strongest claim compatible with the evidence, since we will learn more through the attempt to refute this claim than we would by making weaker claims. I’m not super-fond of that Popperian principle (which is often used as a license to make wild claims under the banner of methodological rigour, imo) – but I accept that there’s a challenge laid down by Brandom’s work on the linguistic. Brandom has laid out in enormous detail an account of a set of (linguistic) social practices that can institute normative and conceptual content. Those of us (like myself) who feel Brandom places undue weight on the linguistic are thereby (implicitly) challenged to provide an alternative, comparably detailed account that wouldn’t involve specifically linguistic practices. It may be that no such account could be produced. I don’t see anything about Brandom’s own account that could settle the matter one way or the other – if we like, we could regard Brandom’s account as open to falsification, but providing the best theory yet available to explain the observable facts. There’s a difference between this attitude to Brandom’s work, however, and the idea that Brandom’s work has shown, somehow, that language is necessary for sapience. I just don’t see how to get that claim out of the work’s actual arguments (even though everything that Brandom says implies he believes it).

    Anyway, I sense that this won’t be satisfactory to you, and there’s also a fair bit in the above that I’ve already said several times before. Part of the impasse here – if impasse there be – may, I think, be due to the fact that I have yet to elaborate / work through on the blog my sense of the core Brandomian categories we’re talking about – I mean principally the concepts of deontic attitudes, the game of asking for and giving reasons, and material inference. My strong suspicion is that we have subtly but significantly different understandings of at least the last two categories – I suspect I understand ‘the game of asking for and giving reasons’ and ‘material inference’ as capturing a broader swathe of phenomena that you’re inclined to, and I think this is driving a fair bit of the other philosophical disagreements here. Our disagreements about the role of the linguistic may (my hunch goes) come more clearly into view once I start working through my sense of those categories in detail. And doing so would of course also be a necessary preliminary to elaborating a non-linguistically-oriented explanatory endeavour parallel to (and otherwise more or less identical with) Brandom’s. So that would be useful too – though again, I emphasise that I’m not actually planning to attempt such a thing in detail. Future posts here may usefully clarify our disagreements to both of us, anyway.

    I have lots more to say in response to your other points, and obviously there’s some overlap of subject-matter (e.g. the question of the transcendental), so some of that will continue to address issues relevant here – but I’m going to call a halt again for now, and quite likely vanish away for weeks again. I’m definitely going to respond to your other comments when I have time though.

    Cheers…

  15. duncan Says:

    Now, w/r/t mathematical necessity, etc. I’ve already responded in a scattered way, and I’ll try not to repeat myself – but to expand and elaborate somewhat:

    It’s got nothing to do with what we’re *representing* when engaging in mathematical discourse, nor whether these things are *real* or not. It’s got everything to do with the normative structures in question.

    I agree with this (my remark about mathematical platonism was meant as an aside, rather than to bear argumentative weight). But this position is, of course, compatible with conventionalism. My view (as I said in one of my shorter responses) is that mathematical necessity gains its necessity from the strength with which we are committed to upholding specific norms (mathematical principles). This doesn’t mean, however, that the practices with which we commit ourselves to these norms can’t also (or instead!) commit us to norms that we are (individually or collectively) unaware we have committed ourselves to. It’s therefore perfectly possible to be a strong mathematical conventionalist while still believing that the proprieties of mathematical enquiry are attitude-transcendent, in the sense that a whole community can be wrong about the proprieties associated with their own conventions.

    This account of course places an appropriately heavy emphasis on the phenomenon that you highlight:

    It amounts to the discovery of implicit premises contained in the content of the intuitive mathematical concepts we’re working with, which we then explicate and decide whether or not to rearrange.

    But those mathematical concepts are themselves based in what we do – what we collectively take to be the rules of the relevant game.

    There’s an important difference between the attitude-transcendence associated with empirical claims, and those associated with conventional claims like those of mathematics: the basis on which one can legitimately make a truth-claim on empirical matters involves somewhere along the line the granting of a social role to objects such that a non-human event or set of events takes over part of the social role of legitimation of claims. (And indeed one would imagine that this is a condition of possibility of experience, though one can I guess imagine sci-fi scenarios in which sapient entities interact with nothing but each other – where there is no non-overtly-social ‘world’). Things are granted the power of judgement over our words and deeds. This’ll be true to some extent with mathematical claims as well, since if one builds a bridge and finds it doesn’t span the width of the river, one might decide to reassess the calculations one made in ordering the wooden timbers. But there’s a reasonably strong sense in which mathematical claims need not be empirically-contestable in anything like this sense. The arbiter here is basically other mathematicians – but of course because the community of mathematicians possesses the qualities Brandom describes when deriving reference from deontic attitudes, it’s possible to generate a concept of mathematical objectivity such that mathematical truth is not tied to any empirical practice, not even the practice of the community as a whole, while still understanding the truth-value of mathematical claims in entirely conventionalist terms.

    I’ve got more to say in response to your second long comment, but this seems like a natural pause-point, so I’m again going to call a halt for now, and hope I can return to say more before too very long.

  16. duncan Says:

    Okay –

    you can’t see the conditions of the possibility of the empirical perspective (the game of giving and asking for reasons as restricted to empirical content), because you refuse to countenance the possibility of a non-empirical perspective within which one could articulate such conditions

    It depends what you mean by ’empirical perspective’ here, but I think I reject this. I agree with the distinction Brandom draws between naturalistic and normative perspectives – I just use the term ‘naturalistic’ more broadly than Brandom does. I agree with the point Brandom’s making, however, that a normative framework cannot be translated into a set of descriptive claims (naturalistic or otherwise). In that sense I certainly think that there are perspectives available beyond the empirical. Normative claims are not empirical claims – they are not claims about what is, but claims about what should be. And I agree with Brandom that normative commitments are a condition of possibility of empirical description, and in that sense ‘constitute’ the possibility of reference. I certainly don’t think that the game of giving and asking for reasons is restricted to empirical content: a thoroughly non-empirical reason (e.g. “torture is evil”) can be a reason in every relevant sense.

    On the other hand (and to use the terminology of Between Saying and Doing) I do think that the practices that institute normative commitments can be thoroughly specified in empirical terms (at least that’s the hypothesis that structures my research-program), and that such specification, if achieved, could be a full explanation of the emergence of normative content. Thus my position is not a ‘naturalistic’ one in the sense in which Brandom uses that term, because I do not think a naturalistic vocabulary (e.g. that of physical science) is sufficient to articulate the content of normative vocabulary; I do, however, think that a naturalistic vocabulary is sufficient to specify the practices participation in which would be sufficient to institute a normative vocabulary. The claim is about ‘pragmatic mediation’ in BSD‘s sense.

    All that’s by way of saying that while I’m inclined to think that there’s something we disagree on here, I’m not sure you’ve quite hit it when you say that I refuse to countenance the possibility of a non-empirical perspective. My claim is just that the emergence of non-empirical perspectives can be analysed (in a ‘pragmatically mediated’, rather than a classically reductive, sense) in empirical terms.

  17. duncan Says:

    More associatively – I think it’s worthwhile differentiating two things. There are two senses in which a norm can be ‘historically invariant’ – a weak sense and a strong sense. The weak sense is that you or I or anyone can take our norms to be transhistorically applicable. Thus I might be of the opinion that torture is monstrous, in any and all circumstances. It’s always wrong to torture people; it was always wrong to torture people; it will always be wrong to torture people. This is a transhistorically applicable (and applied) ethical demand.

    That’s the weak sense. It’s weak because it doesn’t imply any intrinsic reason why this transhistorical ethical demand (as I take it to be) should be recognised as such (even implicitly) by any given rational agent. So – it has always been wrong to torture people. But (at the same time) it has not always been recognised that it has always been wrong to torture people, even implicitly. It is perfectly possible (in principle) for there to exist agents who have no commitment (not even the most implicit) to the idea that torture is wrong.

    The ethical obligation not to torture is thus (I judge) transhistorical in application but not transhistorical in source – i.e. not transcendental. I emphasise this because I think that some of the time objections to regarding norms as “historically contingent” involve the idea that giving up one kind of transhistoricity involves giving up the other – the idea is that if norms are to be generally applicable, they must be experienced as binding in some sense (even if only implicitly) by all rational agents. I believe we can unproblematically give up this latter claim without giving up the former. (That’s a separate issue from whether there are in fact transcendental norms, of course.)

    More immediately relevantly, on this –

    Some people are allergic to this kind of philosophising, as they think it lacks humility. My response to this is quite simply this: fuck false humility… It is better to fail to meet a profound challenge than to succeed by compromising it’s profundity.

    My position, of course, is that my humility is not false: it’s the real deal, bigger and better than ever. 😛 I’m obviously unconvinced that transcendental philosophy in anything much like the Kantian sense is in fact legitimate. My inclination is to see a lot of philosophical claims that have an aura of profundity as in fact lacking rational warrant. I also have the (‘therapeutic’ in the Wittgensteinian sense) inclination to think that a fair bit of the time when philosophical profundity is lost or compromised, what’s actually being lost or compromised are misconceptions about the nature of the claims being made. I don’t, for instance, think that the brand of epistemic humility I promote can involve any real loss in legitimate strength of epistemic claims. I suppose also at base my own intellectual project is sufficiently megalomaniacal that I’m relatively untroubled by the charge of humility :-P.

    But I’m starting to ramble now, and I think I’ve covered most of the ground I meant to. Apologies again for leaving this conversation so long suspended. Schedule will be fierce again now, so apologies also in advance if I leave any response unanswered for a comparable length of time. It’s very interesting to discuss this stuff – and I think a lot of these issues could be addressed profitably at greater length at a later date. I’ve not really touched on the issue of conterfactuals and possible worlds and modal vocabulary, for instance. But enough for now. Hope all is well with you… Cheers…

  18. Christopher Eddy Says:

    You say, Duncan, that “It is perfectly possible (in principle) for there to exist agents who have no commitment (not even the most implicit) to the idea that torture is wrong.” I want to question this.

    It is fundamental to Brandom’s project that every interlocutor is committed, though maybe only implicitly, to observing the laws of non-contradiction and excluded middle, otherwise there could be no possibility of the incompatibility between one commitment or entitlement and another on which score-keeping relies, so the implicit commitment to avoiding absurdity must be taken for granted in all interlocutors.

    It is fundamental to Brandom’s project also that whoever proposes any course of action is thereby implicitly committed to justifying his proposal if asked to do so. Justifying, of course, is a matter of giving reasons, but there are two quite different kinds of reason-giving that need to be distinguished: (a) There is the rationalizing kind which involves the speaker’s adopting the intentional stance towards himself, whereby he interprets himself as making his proposal for the reasons he offers; and (b) there is the justifying kind, and I want to say that all speakers are implicitly committed to recognising that “justifying” is different from “rationalizing”, in that to count as “justifying”, the reasons offered must be reasons not merely for the speaker to make his proposal, but for others, – particularly those others who have demanded the justification, – to consent to it. To make a proposal is implicitly to ask others to consent to it or at least not to oppose it.

    If this is so, then there seems to be a logical constraint on the kind of proposals that could be justified, because no interlocutor could ever without absurdity be asked, for any reason whatsoever, to consent to be subjected to any type of action without his consent; e.g., to a law which provided that, if certain conditions were met, he might be “tortured”, i.e., that pain might be inflicted on him deliberately, either for coercion or for punishment, even if, at the time of that infliction, he had explicitly withdrawn his consent to it.

    I’m not sure that you can rely on natural empathy to make torture wrong because it may come into conflict with an equally natural desire for retribution, but I do think Brandom’s system offers us the basis for saying that every interlocutor is implicitly committed to the wrongness of torture.

    I don’t feel the need to characterize this argument as either “naturalistic”, “non-naturalistic” or “transcendental”: those terms strike me as merely ideological. I just take the argument at face value, and it seems to work. (See my answer to Galen Strawson in the LRB, Letters, 24.10.2013)

    One final point: I have for many years been looking without much success for a way of publishing a philosophical kind of verse for which the current poetry scene has no place, and the kind of conversation I find on this blog provides the perfect context: I couldn’t wish for a better audience, so, gentlemen, please accept the following as a small gift from a grateful follower:

    MIRACLE

    Two Commandments would suffice,
    so Jesus thought, for Paradise,-
    an economy of mnemonic labour:
    Love God! and Love your neighbour!
    Obviously, when things get rough,
    it all depends what you mean by “love”,
    and, if you’re shaped by evolution
    to take delight in retribution,
    don’t rely on being good-natured
    to neutralize that natural hatred.
    Reason’s mind-compelling force
    alone prohibits that fateful course:
    in its miraculous power we trust
    to help us love and so be just.

  19. duncan Says:

    Christopher – I finally found an evening to write up my response to your comment! Apologies once again for the incredible delay on this. So as not to lose track of things, I replicated your comment in blockquotes, and I’ve responded more or less point by point. I’ve not read this back, so I hope it makes sense. There was more that I’d planned to say – but I’m just going to go ahead and post, for fear of this all vanishing away.

    ~~

    It is fundamental to Brandom’s project that every interlocutor is committed, though maybe only implicitly, to observing the laws of non-contradiction and excluded middle, otherwise there could be no possibility of the incompatibility between one commitment or entitlement and another on which score-keeping relies, so the implicit commitment to avoiding absurdity must be taken for granted in all interlocutors.

    Yes, I think I agree with this. [Reasons why I only think I agree, rather than am certain 😛 : 1) I’m still a little too far outside this headspace to feel confident in my judgements here; 2) I think it’s also integral to Brandom’s project that even if we’re all committed to a given principle, there can still exist a lot of disagreement about what commitment to that principle entails, so what it means to be committed to observing the laws of non-contradiction and excluded middle isn’t in itself fully determined by the bare commitment; 3) Brandom’s first book (coauthored with Nicholas Rescher) is on paraconsistent logics. I’ve only glanced at it (apart from the still-unpublished Hegel book, it’s the only book of Brandom’s that I haven’t read through), but in the words of the online Stanford Encyclopedia of Philosophy:

    Rescher and Brandom introduced a modal semantics including, besides ordinary possible worlds (taken as maximally consistent collections of states of affairs), also non-standard worlds that are locally inconsistent (that is, such that, for some A, both A and ¬A hold at them), and incomplete (that is, such that for some A, neither A nor ¬A hold at them).

    I don’t have a view as to how this project connects to the project of Making It Explicit, but unless or until I work that question through properly, I’m uneasy about stating with full confidence the exact role that the laws of non-contradiction and excluded middle play in the Brandomian apparatus. That said, I do think you’re right here.]

    It is fundamental to Brandom’s project also that whoever proposes any course of action is thereby implicitly committed to justifying his proposal if asked to do so.

    Ye-e-e-es – I definitely agree with this – but one needs to be careful, because there’s potential for confusion around the sense of “justify” that applies here. There’s a narrow and a broad sense of ‘justify’, just as there’s a narrow and a broad sense of ‘rationalism’ at play in Brandom: Brandom’s narrow slimline rationalism provides the resources with which one can assemble the more substantive (and discursively/ethically/politically impactful) rationalism to which Brandom is also committed, but the one does not lead intrinsically to the other; Brandom’s slimline rationalism would also be compatible with many though not all forms of substantive irrationalism.

    To clarify/unpack that: when Brandom says that one is implicitly committed to justifying one’s proposals if asked to do so, this just means that one ought to be able to offer a proposition that could serve as a premise in an inference that results in your proposal. Of course, as you know, the fact that one is so committed doesn’t mean that one in practice has to honor the commitment (one could just ignore this commitment; or one could have an independent commitment to not offer reasons for this proposal – such double-binds are compatible with Brandom’s rationalism). But, more to the point, the reason one in fact offers may be a rejection of the need for a substantive reason to ground this proposal altogether. So, if you ask me why we should do X, and I respond “I don’t need to give my fucking reasons; if I tell you to do something, you should just do it!”, this proposition is a reason in the formal, slimline sense in which Brandom is here using the term. Premise A: If I tell you to do something, you should do it, regardless of reasons. Premise B: I’m telling you to do something. Conclusion: You should do it. This is a perfectly coherent inference, even though it doesn’t offer any meaningful reasons in most substantive senses of the term.

    So, while I agree that it is fundamental to Brandom’s project that we’re committed to justifying ourselves, that “justifying” is very narrow in scope. It doesn’t mean our justifications have to be good, or even contentful or coherent – they just need to be capable of occupying a formal position in discursive space. We therefore can’t leap from this slimline rationalism to a more substantive, impactful (useful and virtuous) rationalism without a lot more theoretical legwork (which legwork I don’t think Making It Explicit does or could provide – such matters need to be hashed out ‘downstream’ of the very abstract metatheory Brandom is concerned with in that work).

    Right. All of the above is probably a little beside the point. I think with this next quote from your comment, and my response below, we’re getting closer to the heart of the matter.

    Justifying, of course, is a matter of giving reasons, but there are two quite different kinds of reason-giving that need to be distinguished: (a) There is the rationalizing kind which involves the speaker’s adopting the intentional stance towards himself, whereby he interprets himself as making his proposal for the reasons he offers; and (b) there is the justifying kind, and I want to say that all speakers are implicitly committed to recognising that “justifying” is different from “rationalizing”, in that to count as “justifying”, the reasons offered must be reasons not merely for the speaker to make his proposal, but for others, – particularly those others who have demanded the justification, – to consent to it.

    Now I agree with this, but: I think here it is very important to draw the distinction between what people should take as good, persuasive reasons, and what they do take as good, persuasive reasons.

    As we know, one of the principal imports of Brandom’s work is its ability to differentiate ought from is – to differentiate objective norms from actually-existing social sanctions, even though its account of the former is built up exclusively from the materials of the latter. One of the ‘outputs’ of Brandom’s work (at least on my reading of it) is its ability to explain why our norms may rightly differ from those of the social milieu we inhabit, even though our norms are fully explicable in terms of the social practices of that milieu. This ‘magic trick’ – this apparent creation of something out of nothing – the appearance of (apparent) normative transcendence, from a fully immanent, deflationary, pragmatist, social-scientific explanatory framework, is Brandom’s great achievement (imo).

    The fact that Brandom’s apparatus, taken all together, is capable of acheiving this result, also (apparently, but not actually, paradoxically) allows Brandom to make use of this achieved result at earlier stages of his argument.

    Thus, Brandom can state that although normative objectivity (correctness versus incorrectness in relation to an apparently ‘external’ norm) is instituted by the attribution of correctness or incorrectness in deontic attitudes, the objective norm is not explicable simply by the actually-existing attribution of it, but by the correct attribution of it: the status of objectivity is reverse-engineered into the normative practices that are themselves used to institute objectivity. (Part of how one reacts to Brandom’s whole project, I think, depends on whether one regards this explanatory move as a vicious or a virtuous circle; I regard it as a virtuous circle, but many of Brandom’s critics are obviously coming from the ‘vicious circle’ place.)

    The reason I mention this, is that when you write “the reasons offered must be reasons not merely for the speaker to make his proposal, but for others, – particularly those who have demanded the justification, – to consent to it”… I absolutely agree, but we are (as you may already agree! – but I just want to be really explicit about the premises here) thoroughly in the space of “should”. If I offer a reason for my action, and it is a good reason, then those who have asked for my reason should agree with it (consent to it)… for some value of ‘should’. If I am right (we believe) then they themselves should agree with/consent to my reason, if they are right. This is all good. But of course it doesn’t have any bearing, in practice, on whether the person who has demanded these reasons does consent to them… that is an entirely different question.

    Maybe I seem to be overstressing this point – apologies if so. But I think it’s important to differentiate the counter-factual should, of consent in rational public discourse, with the demand to actually in practice persuade people. The former can sometimes slip into the latter, but they are always in principle differentiable; it is always possible that even though nobody is persuaded by your reasons, the reasons are still good ones and the conclusion that they lead to still correct.

    Ok.

    To make a proposal is implicitly to ask others to consent to it or at least not to oppose it.

    Per my response above, I would say this differently. “To make a proposal is implicitly to suggest that others should consent to it, or at least should not oppose it.” This is a different thing from asking others to actually consent or actually not oppose.

    And it is different in at least two ways. Or, to put it another way, there are at least two ways in which these two (the ‘should’ and the ‘does’) can diverge here. I’m shifting registers a bit here – getting more concrete – so I wouldn’t want the following remarks to be taken to operate at the same level of abstraction as the rest of the Brandomian apparatus; but I think getting a little more concrete may help clarify the line of thought…

    Two ways in which the ‘should in principle agree’ and the ‘will actually agree’ can diverge – crassly put. (This way of phrasing things is almost certainly not the best – apologies – I’m extemporising a bit here).

    1) Error.

    On this perspective, fundamental premises and fundamental goals are shared, but there is a disagreement about how best to pursue these goals. In this scenario, when I say “you should agree with this” I’m saying “if you were to think through your commitments more carefully, you would see that the logic of contradiction and incompatibility would eventually eliminate the commitments that lead you to believe that we are at odds on this issue. Your fundamental commitments are the same as mine; the commitments on which we differ are the result of unresolved tensions in your views, which I can help you to resolve; and when they are resolved we will have reached consensus.” When I say “You should think this” I am saying “If you were to adhere consistently to your own basic commitments, you would see that you already implicitly agree with me.”

    2) Interests.

    On this perspective, there is a more fundamental difference of subjectivity, between the stances of the two interlocutors. The interlocutors want fundamentally different things; or they have fundamentally different values. Here a working through of inconsistencies in the interlocutors position will not resolve difference into consensus; it will simply clarify the lines of difference. In this scenario, when I say “you should think this” I am not saying “if you worked though your own commitments you would see that you are already implicitly committed to this”; rather, I am saying “you should be a different kind of person”. I am saying “you should think differently, at base”; I am saying “your very subject-position is wrong, in some way”. In this scenario, consensus in a global sense is simply not on the table. We differ; we will always differ unless one of us changes commitments in a very fundamental way. We are at ‘a differend’.

    From my point of view, actual social interactions are just as commonly locatable in category (2) as in category (1). And what this means is that for many proposals, while I can wish people to consent to them, and I can hope that they change their fundamental commitments sufficiently to allow them to consent to them, I cannot realistically expect them to actually consent to them. I can ask – but I cannot expect an answer in the affirmative without also expecting a transformation of self (on the part of my interlocutor) sufficiently substantial to result in a radically different self on the other side of this exchange.

    Ok. That’s my view about how much of social life operates. But having sketched that view, it’s I guess obvious that the concept of consent is very very difficult to ‘bake in’ to our fundamental ‘social ontology’, as it were. To put things in terms of Mannheim’s schema of ideologies, I’m saying that we need to take account of the ‘socialist/communist’ vision of the battle of interests, as well as the ‘liberal-democratic’ vision of rational discursive orientation towards consensus. (I’m aware that I’m shifting levels of abstraction here – apologies for that – but I think my basic point could be stated (more carefully) in a way that doesn’t collapse downstream politics into fundamental practice theory.)

    So, to continue:

    If this is so, then there seems to be a logical constraint on the kind of proposals that could be justified, because no interlocutor could ever without absurdity be asked, for any reason whatsoever, to consent to be subjected to any type of action without his consent

    I don’t think I agree that interlocutors cannot consent to be subjected to any type of action without consent – or at least I think that one needs to be cautious about where one locates consent in such commitments. I take it that one of the elements of Brandom’s apparatus is the view that we can commit to things that we do not recognise ourselves as having committed ourselves to – and that we remain committed to these things, even once they are made explicit to us. “[U]ndertaking a commitment can be understood as authorizing, licensing, or entitling those who attribute that commitment to sanction nonperformance.” (MIE p. 163) – and this can be the case even if one does not endorse the sanction at the time it is performed.

    More fundamentally, though, I don’t think Brandom’s apparatus warrants giving such a central role to individual consent (at the metatheoretical level: giving a central role to individual consent may still be ethically or politically right – I’m just saying that this fact, if it is a fact, is not grounded in Brandom’s metatheory.) Hopefully I’ve unpacked enough already to give some sense of why I think this. In a word: – although the Brandomian apparatus tethers normative objectivity to the collective attribution of normative objectivity in a community, it does so in a way that is capable of severing actual normative objectivity from any individual’s judgement of normative objectivity: to say that something is correct is not to say that someone will agree to it, or even that they should agree to it while remaining the person that they are. My claim that something is a correct action may be based in my view that all people ‘of the right sort’ would agree; but this individual in front of me – sinful, slothful, one of the accursed lesser breeds, hardly capable of rational thought or speech, half devil and half child – this person is incapable of agreeing, though of course they would if they (impossibly) were saved, redeemed, or civilised. Or – in a different vein – as the person in front of me raises their weapon to murder my children, I do not need their consent to assault them first, because (in most scenarios one can imagine) that consent would never be forthcoming; if they were a different person, no doubt, they would agree with my action – but they are not and will not, in this world we inhabit. Sometimes actions are taken without consent – and their rightness or wrongness is not necessarily connected to whether consent is or could be granted. (Whether we regard the actions as right or wrong – and whether they are right or wrong – is connect to other normative criteria.)

    I’m aware, of course, that positions like that one I’m advocating here can very very easily lead to intensely ugly commitments and actions. There is always the risk of a ‘slippery slope’ any time one emphasise the unimportance of consent – even if (as I want to here) one also emphasises that this importance can be reinstated at a different level of theoretical articulation. However, as I say, I don’t believe the positions I’m advocating here must lead to such commitments; I don’t believe they should lead to such commitments; and I don’t believe that we can ‘bake in’ our disavowal of such commitments to this most abstract layer of our metatheoretical apparatus. Our disavowal of coercive violence has to come from a different place: our practical politics and ethics.

    So to return to the rest of your sentence of which I only quoted the first segment above:

    If this is so, then there seems to be a logical constraint on the kind of proposals that could be justified, because no interlocutor could ever without absurdity be asked, for any reason whatsoever, to consent to be subjected to any type of action without his consent; e.g., to a law which provided that, if certain conditions were met, he might be “tortured”, i.e., that pain might be inflicted on him deliberately, either for coercion or for punishment, even if, at the time of that infliction, he had explicitly withdrawn his consent to it.

    For reasons hopefully clear from the above, I don’t think this conclusion can be drawn from Brandom’s metatheory. Just as Brandom’s metatheory can’t give us a substantive rationalism, but only a (very) slimline rationalism, that can be used as the basis for assembling a more ethically and politcally impactful rationalism, ‘downstream’… so Brandom’s normative framework can’t be used to underwrite norms as specific as the rejection of torture; though it can explain the social basis of our ability to hold ourselves to such a norm.

    All that said – and apologies for disagreeing so much, when I think it is likely that in practice we disagree on very little: I do not believe that anything in what I’ve said above should stand at odds with the sentiments expressed in your poem… for which thank you.


  20. Duncan,

    Thank you very much for the time and care you took in responding to my remarks. I have to admit that my understanding of MIE is limited, and that much of Part Two is beyond me, so it is entirely possible that I have been too keen to take Brandom as giving us a more substantial rationalism than he actually does; also I have to confess to a certain impatience with Brandom that he shows himself so little inclined to proceed “downstream” at a brisker pace, – in which respect I found his Williams interview (for which many thanks) very revealing: his insistence on being a “professional philosopher”, not a “public intellectual”; but let’s cut to the chase.

    Firstly, I fully accept the point that reasons do not have to be good ones in order to function as reasons, and of course we know that even very good reasons cannot compel compliance, since not all goods are mutually compatible. “I’m telling you, so do it!” is, as you say, a form of reason-giving. What I now realise is that my presentation was misleading and drew you into a great deal of what may have been unnecessary labour on the nature of reasons and for this I apologise.

    I do not think we disagree about where we want to go, ethically and politically, but only about how we can legitimately get there. My central point is one with which you very reluctantly disagree, i.e., that “there seems to be a logical constraint on the kind of proposals that could be justified because no interlocutor could ever without absurdity be asked, for any reason whatsoever, to consent to be subjected to any type of action without his consent”, and you give four kinds of reason for doing so:

    (1) You say “We can commit to things that we do not recognise ourselves as having committed ourselves to”, – a revelation not original with Brandom, of course, since it is one with which Plato’s Socrates so often stuns his interlocutors; but we remain committed to Q as the unforeseen implication of P only until we retract P, which we can do at any time because it would surely be absurd to maintain that asserting P now commits us to asserting it forever after. There is no requirement in Brandom’s holistic system for diachronic consistency, only for synchronic coherence.

    (2) You quote Brandom: “Undertaking a commitment can be understood as authorizing … those who attribute that commitment to sanction non-performance”, and of course I agree, but you claim that “this can be the case even if one does not endorse the sanction at the time it is performed”, and this is precisely the point I contest. As I see it, one cannot be responsible without being committed, one cannot be committed without authorizing, one cannot authorize without consenting and one cannot without absurdity consent to being compelled, so one cannot be committed to being subject to sanction by compulsion.

    (3) You say “I don’t think Brandom’s apparatus warrants giving such a central role to individual consent”, and it is true that Brandom himself does not deploy “consent” as one of his central terms, but I do not see how we can make sense of responsibility, authority and commitment without consent being implied. Brandom rejects communalistic relativism and insists that commitment is always an “I-thou” relation, so the individual is always a party to the conversation: “we” are never in a position to determine (by talking, as Utilitarians do, over the heads of those whose fates they presume to decide) what an individual is committed to regardless of his own explicit declarations at the time.

    (4) Finally, you appeal to defensive action as a case where it is justifiable that “actions are taken without consent” in the sense that the assailant who threatens your children doesn’t consent to your assaulting him in their defence; but, as I interpret Brandom, an authoritative and responsible agent who attacks your children thereby implicitly endorses the claim “that one may attack another” and thereby licenses you to attack him for so long as he continues to attack you and yours. The attack implies consent to the defensive actions it provokes, which neatly illustrates the difference between a semantic and a psychologistic or naturalistic understanding of action.

    But our original difference was on the subject of torture and my point is that it is only by the most extreme abuse of language that one could speak of using acts of torture “in self-defense”, for the obvious reason that, for you to be able to torture someone, it is necessary that he be helpless and at your mercy and therefore incapable, by definition, of posing a threat to you. The same is obviously true of the death penalty: the executioner is not threatened by the convict who stands before him bound hand and foot.

    What I have said about the impossibility of consenting to be compelled entails the impossibility of justifying any kind of compulsory punishment, but this does not spell the end of social order because it allows legal penalties to be accepted by convicts as penances are by penitents, and, when those penalties are not accepted by convicts, it allows the state, in self-defence, to treat the convict as it would a p.o.w., interning him until the cessation of hostilities. In a naturalistic perspective, internment under the Geneva Convention might not seem distinguishable from imprisonment, but it is the kind of distinction for which some people, e.g., Bobby Sands, have been prepared to die.

    Please do not feel under any pressure to reply quickly, but I hope you will tell me if and where you feel I have evaded any of the challenges you posed for me.

  21. duncan Says:

    Hi Christopher – many thanks for these remarks, which again I will reply to! (I don’t think it will take me quite so long, this time, either, though I guess I shouldn’t promise… :-/ ) The immediate prompt for commenting here, though, is to apologise that I’ve gone ahead and written a new post rather than try to get back into quite this thought space again. Not particularly relevant to our conversation – but not totally irrelevant either.

    As I say though: apologies for my sluggishness here – I will reply properly and on point eventually…

  22. duncan Says:

    Hi Christopher – this may be a little rushed, so apologies if either content or tone suffer from that. But taking your numbered points in order:

    1)

    we remain committed to Q as the unforeseen implication of P only until we retract P, which we can do at any time because it would surely be absurd to maintain that asserting P now commits us to asserting it forever after. There is no requirement in Brandom’s holistic system for diachronic consistency, only for synchronic coherence.

    This is an interesting and illuminating way of putting it – and I agree, it’s one of the obligations of reason that we sacrifice diachronic consistency to synchronic coherence, as incompatibilties are revealed among our commitments by the process of explicitation. However, I think we need to be careful not to be too dismissive of the obligations of diachronic consistency, because on Brandom’s account, as I understand it, the obligations of diachronic consistency are what give our concepts content. Brandom’s explanatory apparatus is, as he puts it, sanctions-based. Those sanctions need not be coercive – Brandom is at pains to point out that sanctions can be fully normative – though I want to say more on this important point, eventually. However, one of the core elements of Brandom’s apparatus is his idea that what gives our commitments their content is the entitlement of others to sanction us for actions incompatible with those commitments; and reversing our commitments does not, on Brandom’s account, automatically eliminate those entitlements to hold us to earlier commitments. There may of course be social mechanisms for voiding those entitlements, but this is a separate issue from the fact – which I think is core to Brandom’s system – that those entitlements can in principle persist whatever our own actions subsequent to taking on the relevant commitment.

    The above may, however, just be restating the point you’re contesting in your (2). I’ll label the next two sentences I quote, for ease of reference in a moment:

    [A] You quote Brandom: “Undertaking a commitment can be understood as authorizing … those who attribute that commitment to sanction non-performance”, and of course I agree, but you claim that “this can be the case even if one does not endorse the sanction at the time it is performed”, and this is precisely the point I contest.

    Then:

    [B] As I see it, one cannot be responsible without being committed, one cannot be committed without authorizing, one cannot authorize without consenting and one cannot without absurdity consent to being compelled, so one cannot be committed to being subject to sanction by compulsion.

    To try to get a more fine-grained sense of our disagreement here, I want to separate out ‘sanction’ and ‘sanction by compulsion’. I take it to be unambiguous in MIE that for Brandom we can legitimately be sanctioned for commitments to which we do not take ourselves in fact to be committed. Without this element of Brandom’s apparatus, the overall argument – the distinction between de dicto and de re, and therefore the whole ‘double book’ scorekeeping apparatus – wouldn’t work, so I don’t think this point can be given up without giving up on most of the rest of the whole. I therefore think I disagree with the final clause of your [A].

    This doesn’t mean, however, that Brandom’s apparatus requires “sanction by compulsion” as an essential building block of his system – Brandom is clear that he thinks this isn’t the case. And I take it that your core claim is “one cannot without absurdity consent to being compelled”, rather than the broader point about the whether it is ever possible to legitimately sanction actors for commitments they do not take themselves to hold.

    Of course, to say that Brandom’s system doesn’t require sanction by compulsion is a massively weaker claim than saying it prohibits it. But issues around coercion and consent may be best addressed by moving on into your (3):

    I do not see how we can make sense of responsibility, authority and commitment without consent being implied. Brandom rejects communalistic relativism and insists that commitment is always an “I-thou” relation, so the individual is always a party to the conversation: “we” are never in a position to determine (by talking, as Utilitarians do, over the heads of those whose fates they presume to decide) what an individual is committed to regardless of his own explicit declarations at the time.

    This is the heart of our discussion, I think. But regrettably I’m out of time to reply on this now! I’ll pick this up again in my next comment – apologies for leaving this dangling, but better perhaps to post these incomplete remarks than to put the conversation entirely on hold again…

  23. duncan Says:

    To re-articulate my response to your point (1) perhaps a little more clearly: if I promise to do something, and then I change my mind at the last moment and “un-promise”, retracting my original commitment, the person to whom I made my initial commitment can (at least in many circumstances) legitimately regard me as having broken my promise, even if I insist that there was really no promise to break at the point (moments after my retraction) at which it would have needed to be fulfilled. So not all commitments can be retracted on the decision of the person who made those commitments; nor can all the sanctions associated with those commitments be evaded in this manner; and this fact is one of the ‘constitutive possibilities’ of the concept of commitment.


  24. Duncan, it seems to me perfectly possible to break a promise, admit (a la Nick Clegg) to having broken it and to being at fault, – not for having broken the promise, but for having made a promise I ought not to have made. I must then endure the obloquy of infidelity, of course, but what else I must endure will depend on the substance of the case, and this is where that distinction between my property and my body becomes important.

    If it is a matter of property, then I have no grounds for regarding as unjust whatever loss of property follows by law from my breach of promise, because I can claim as my property only what the laws of property recognise as such; but I cannot justly be required, upon breach of a contract, to surrender a quantity of my flesh (e.g., a kidney), – even if that is what I had previously promised to do, – because human bodies and their parts cannot justly be treated by law as items of property.

    Why not? Because it is only as a participant in discourse that I can be committed, and it is only via my body that I can participate in discourse, so that to surrender my body would be to surrender my capacity for commitment, and I do not see how I could meaningfully commit myself to that (except, perhaps, where I ask to be helped to die because my capacity for meaningful commitment is already hopelessly compromised by my physical state).

    Likewise, I want to argue that there are certain types of proposal to which, even when I purport to consent to them, I cannot meaningfully be taken as having consented, e.g., that I be compelled to suffer X if convicted of Y. See M.I.E., p.75: “it is no use asking what it is to express a proposition or other content (to purport to represent), without asking what it is to grasp or understand such purport.”

    It seems to me that Brandom goes beyond the requirements of his own system in interpreting certain ways of responding explicitly as “sanctions” when all that is required is the idea of responses that are accepted as appropriate. If I break a promise to you, you may not trust me in the future, and that may be a consequence I have to accept, but that, – however disadvantageous it may be to me, – is not a “sanction”, merely an appropriate response.

    Indeed, one of the difficulties that one faces in reading Brandom is the lack of detailed applications to precisely the kind of ethical and legal problems we are discussing. M.I.E., of course, is already long enough, and bringing in this kind of detail might have made it a great deal longer, but he is not averse to exploring practical implications in casuistic detail in other contexts, e.g., in the discussion of Red Barn Facades in “Articulating Reasons”. Perhaps it’s his desire to avoid the role of the public intellectual that explains it, but why then does he use a word with all the baggage of “sanctions”?

  25. duncan Says:

    Postponing response to almost all of this until I’ve finished responding to your earlier comment – but just to say that I hear ‘sanctions’ as a very general category. It doesn’t seem troubling, to me, to call a loss of reputation a ‘sanction’, and disapproval or distrust can in practice be a very powerful social sanction. And further to that – and at the risk of muddying the waters by introducing yet another set of considerations – one of the reasons I don’t think Brandom can maintain the ‘firewall’ between normative and material sanctions (or appropriate response versus sanction, in the alternative vocabulary you suggest) that he is at pains to draw in the first chapter of MIE, is that even the mildest form of appropriate response – e.g. the diminishment of the trust you have in me – may have very substantial material consequences down the line. (Because I broke my promise to you, you don’t trust me as much as you did; so you don’t write as glowing a reference as you would otherwise have done when I leave your employment; I therefore fail to get another job, which leaves me unable to pay my rent and I find myself on the street, where my heart ailment reappears and I die young and alone, etc.) So I don’t think material consequences can be easily separated from even the mildest normative approval or disapproval: society as a whole can impose the most severe material consequences even if no individual is doing so, or at least intentionally and directly doing so. This takes us some distance still further afield though – I’ll reply more relevantly to your comments at greater length.


  26. Hi, Duncan,

    This is not to hurry you, but just to let you know I’m still here. A quick response to the above is that there is no action which can’t have harmful consequences; e.g., my not writing you a favourable reference might save you from a job which would otherwise have offered you prospects that could have tempted you into corrupt ways that might eventually lead to an outcome as miserable as the one you have sketched. Likewise, if I beat you up, you might experience that not as a sanction on you, but as shame on me.

    There can be no material, but only a normative difference between a sanction and a reward: a sanction has to be imposed as such in order to be experienced as such, and can function as a sanction even if it has beneficial consequences; indeed, in Gorgias, Plato’s Socrates argues that just penalties are to be embraced for their meliorative effects.

    The summer vacation will no doubt give you time for a fully considered response and I would much rather have that than the product of a few moments snatched from other commitments.

    Chris


  27. Hi, Duncan,

    A further thought about reasons: Telling me “Do x or I’ll punish you” will work as a reason for me to do x only if you have the power to punish me, i.e., its hidden premise is a social structure in which power is unequally distributed.

    Chris


  28. Hi, Duncan,

    The summer vacation was when I imagine you would, if ever, have had the time to produce a comprehensive refutation of the arguments I offered you, but the vacation is now past and no refutation has appeared, so I am anxious to know whether you are still committed to providing a fully considered response.

    No one who has confidently dismissed my arguments seems to me to have understood the logical points at stake, but your reaction, though unconvinced, has not been dismissive: you recognised the power of my position and your blog over the years more than adequately demonstrates your logical prowess: indeed, I am bound to recognise that your ability both to deploy highly complex abstractions and to analyse them very swiftly, – as in your expert demolition of Zizek’s sinister rhetoric, – far exceeds my own.

    Furthermore, I have not wrapped my case in cloudy abstractions, but have made every effort to reduce the arguments to their simplest forms so that, if they were false, it should not be a difficult matter to refute them. I tell myself that, if I am wrong, you should be able to point out the error; but equally, if I could obtain your public endorsement, I should have good reason to feel more confident of my position.

    I also believe that, if you could be convinced, you would be better able and better placed than I to gain recognition for the case at the academic level, i.e., to set the arguments in their wider philosophical context, and I have begun to entertain the hope of a jointly authored publication. I think you will agree that, if my arguments are not wrong, they provide the basis for a new direction in moral and political philosophy and in jurisprudence.

    Chris

  29. duncan Says:

    I am anxious to know whether you are still committed to providing a fully considered response.

    Yes – I am, and indeed I have a 1/3rd (??) completed draft reply sitting in my Google Drive – I’m just horribly busy 😦 My reply will, I’m sure, be utterly inadequate given the staggering time it’ll have taken me to produce it, but barring accident/injury it will happen one day 😦 [Short interim version: I remain unpersuaded, I’m afraid, but I probably need to think more about the issues you raise around imprisonment.]


  30. I’m sure you’ll understand if I say how pleased I am that the task of refuting me is proving so time-consuming. I shall possess my soul henceforth in patient expectation of an even more pleasing outcome.

  31. duncan Says:

    Well my goodness this is totally inadequate, given how much time it’s taken me to get round to writing it. Honestly though I think I should just go ahead and post it given that I’ve found time to write it at all… sorry about the extreme time / content differential, but so it goes :-/ For what it’s worth..

    – I shouldn’t have brought in the self-defense situation: I wasn’t meaning to do any work with self-defense as a category, and it unnecessarily muddied the issue – apologies.

    – I think there may be a difference between us in, as it were, our degree of pessimism about possible social institutions. That is to say: my working assumption as to one of the sources of our disagreement is that I’m more pessimistic on this score than you are. Re: your comment about fascism: I think that fascism in the narrow sense (the form of fascism exemplified by Mussolini’s Italy and Hitler’s Germany, however one wants to specifically define it) is an escapable phenomenon, that should be fought against. Understanding fascism more broadly, though, as something like “arbitrary imposition of coercive power by agents of a state”… I think ‘fascism’ in this sense is basically coeval with states, and inseparable from them. This, to me, isn’t an argument against states in general – it’s a fact about what we have to deal with, if we have states, or something like them. Even more broadly – I’m sceptical that *any* form of governance is possible that doesn’t involve coercive power by agents of a governance structure. This isn’t a ‘theoretical’ claim, as it were – if you like, it’s an assumption about ‘human nature’ (malleable and varied though that is).

    To put it another way: once upon a time I was a big fan of Derrida. I’m not any more, but a phrase from his essay on Levinas has stuck with me: he writes there something to the effect that our task is to achieve “the lesser violence in an economy of violence”. This still sums up my baseline attitude on these issues: violent coercion is here to stay; it’s in the grain; the institutional problem is – as it were – using coercion to minimise coercion. I regard this as, in a way, the ‘liberal problematic’ – and I think it’s the right problematic. The institution-design problem is how to disperse and organise power in such a way that some of the coercions we end up with can serve as checks and balances on the abuse of power. So my critique of coercion is not a critique of coercion ‘as such’ – it’s more limited and contextual.

    – Your distinction between compulsory punishment of criminals versus treating opponents of the state as prisoners of war is an interesting one – and to be honest I don’t have a good or clear response on this. I guess I’m unclear what purpose is served by drawing the distinction this way. I understand in the case of Bobby Sands why this distinction matters: the political point at issue was whether Sands was a member of the unofficial army of a subjugated state fighting against an occupying power, or a criminal subject to a single sovereign state’s legitimate authority, so the argument makes sense to me in this context. But I don’t see what is gained by generalising the ‘prisoner of war’ relationship to the entire policing, judicial and prison system of all (potentially legitimate) states [or similar governance apparatuses] (excepting cases of voluntary penance – which imo would be few and far between?)… this strikes me a stretching the category of prisoner of war beyond its reasonable limits. If the goal is a more humane prison system, or a less egregiously discriminatory and oppressive policing system, aren’t these goals better fought for directly? But I feel the goal from your pov is more ‘in principle’ than that: legitimating confinement as the self defense of the state or its citizens, as opposed to the imposition of coercive punishment of any kind. I can see why it might be better to think of the prison system in terms of protection of others, rather than punishment of the criminals – I think I can get behind that; but I don’t easily think of that in terms of the reframing you suggest. Partly I guess that’s because, while actually-existing prisons are typically barbaric, I do think there’s a role for prison as disincentive, not just as immediate self-defense. (Disincentive could, of course, be conceptualised as ‘self defense in aggregate’ – but doing so would the link with individual consent that I take to be your motivating interest?) That may be a quite basic difference between us. All that said (albeit in muddled fashion), I’m conscious that the analysis of the institutions of policing, law and prisons – the whole internal coercive population-management apparatus of the modern state – is a large terrain that I haven’t given nearly enough thought / attention to – so all these comments are rather provisional and hazy.

    – That confusedly said, I think I now have a clearer sense of our differences, which if you’ll forgive me I’ll try to re-present in my own words. I appreciate that one never captures the nuances (even the broad brush nuances :-P) of another’s position when one attempts these paraphrases – but even so… very crassly: as I see it, your interest in Brandom is motivated in part by how you understand Brandom’s shift from the ‘I-We’ to the ‘I-Thou’ understanding of responsibility, obligation, norms, etc. Putting this in my own words: the classic ‘I-We’ model of the relationship between the individual and society places norms at the level of the ‘We’ and has the individual subject to those norms, which impinge upon the individual from outside them. Durkheim would be an example of this; Lukacs would be an example; fascist conceptions of the ‘Volk’ would be an example. What ‘social ontology’ and what politics we end up with will depend on how we understand the ‘we’, but still. Often, on this approach, the ‘ontological’ status of the We seems a little unclear… It’s the Geist, or it’s the Volk, or it’s class consciousness understood as some sort of quasi-metaphysical totality, or it’s a mysterious national character, etc. etc. Or it’s a ‘form of life’, indeed, at least on some interpretations of Wittgenstein. [Not that some of these categories couldn’t be understood in other terms, as well, to be clear.]

    What are the problems with this conception of the social basis of norms? Let’s say there are two problems.

    First problem is ontological: what *is* this ‘we’? It’s hard to give an account of the ‘we’ that doesn’t either reduce it to the aggregate behaviour of individuals, hypostatise it, naturalise it (as the essence of the human, say), or evade the question.

    Second problem is political: making norms reside in some supra-personal ‘we’ of dubious ontological status makes it very easy to justify the subjugation of the individual to this purported social(/metaphysical) entity. If norms ultimately reside in the ‘we’, any given individual can in principle be cut off from those norms, untouched by them – at least at the level of their conscious or stated commitments… and if this is the case, well, we can do anything to that individual that we like: to the extent that they depart from the norms determined by this ‘we’ they are outside the space of ethical considerations (or the space of reasons…)

    So… as I understand where you’re coming from, you’re partly motivated by the goal of rejecting this possibility. That is: while there may be norms that guide your actions that another individual doesn’t acknowledge, you can’t legitimately coerce the other on the basis of those norms to the extent that the other doesn’t accept them. So the ‘I-Thou’ relation solves the political problem presented by the concept of the We that stands above and potentially against the I or the You – it ensures that such a displacement or nullification of consent isn’t possible.

    Our disagreement, then, is that I don’t think Brandom’s ‘I-Thou’ account does this. With some qualifications and nuances, I think Brandom’s shift from ‘I-We’ to ‘I-Thou’ is basically intended to solve the ‘ontological’ problem of what kind of things norms are – where they reside, where they come from. The ‘I-Thou’ account Brandom gives provides, if you like, a appropriately microfounded account of what the ‘We’ in ‘I-We’ is (such that we can do without that vocabulary if we wish). That’s the point of the account, more or less, from my point of view. I don’t think Brandom’s account pins legitimacy of apparently coercive action to the consent of the individual with whom one is interacting; I don’t think that’s what it’s for, or how the account works.

    On my read, Brandom explains how the norms we appeal to emerge out of lots of individual interactions, that can be understood as individual interactions – he doesn’t require an appeal to a ‘we’ at any ontological level. But the point of ‘I-Thou’ talk is *not*, on my read, to place consent front and center; I think Brandom’s metatheoretical system is entirely compatible with a paternalism (or indeed a stronger understanding of the other as enemy) that acts in a coercive way on an individual in ways that contradict their commitments and lack their consent but that are grounded in norms that have their source elsewhere.

    Now – there is a *weak* sense in which Brandom’s metatheory will incline one to a more ‘consent-oriented’ politics: Brandom’s system gives no basis for an ontological or metaphysical differentiation between people whose views, commitments, and consent matters, and people whose views, commitments and consent can be ignored. So Brandom’s metaphysical system is incompatible with (say) a metaphysical understanding of the Volk. If people whose politics excludes the political participation of a class of individuals wish to provide a theoretical justification for that exclusion, they will need to do so in a ‘downstream’ fashion (assuming they for some odd reason wish that exclusion to be compatible with Brandom’s metatheory). [If I were doing my due diligence properly here I’d add a very important set of qualifications around the issue of sentient entities without language-capacity, which is often a flashpoint in debates around Brandom and subject-hood and consent, etc.; but I’ve addressed that question somewhere or other on the blog I’m almost sure, and I’ll just gesture here to those earlier remarks.]

    Now to be clear, I don’t think the use you’re suggesting Brandom should be put to is an indefensible one: I think your use of Brandom to provide a strong basis for the role of consent in political justification [assuming I’ve understood you right, which of course I may not have!] could be made with fewer departures from Brandom’s stated commitments than I make when I try to extend his scorekeeping apparatus to non-linguistic scorekeeping (for example). But don’t think Brandom’s apparatus as it currently stands provides intrinsic warrant for the commitments I take you to be aiming to ground with it, and it’s not a direction I myself want to push his work in.

    Well that felt like a lot of noise without much clarity – obviously I’ve just outlined my preferences/commitments above rather than made any sort of proper argument to defend those preferences/commitments – I *think* arguments for what I say above are floating around in the Brandom content on this blog – but drawing them together in the service of this point seems like a bigger task than I have time for now… and of course this whole response may be missing the point or askew from what you’re trying to push on. :-/ Still – there it is, for what it’s worth…


  32. Duncan,

    I realise attempting to engage you in further argument would be futile, but I don’t intend to leave you with the last word. My main point is not, of course, about an interpretation of Brandom: it’s about that most fundamental of all philosophical questions, “How ought we to live?”

    I am arguing that normativity (or what ordinary people call “responsibility”: the recognition of being subject to judgement and therefore necessarily committed to doing only what can be justified) is implicit in relations between interlocutors, that a contradiction is entailed in asking any speaker to consent to the proposal that he be disposed of without his consent and that no reason can ever justify a self-contradictory proposal, from which it follows that no speaker can ever responsibly propose that another be insulted, enslaved, raped, tortured or murdered – even with a court order. This conclusion creates a problem for the arbitrary exercise of power which the insidious fictions of social contract theory were specifically devised to obfuscate.

    To claim that any individual or institution could be justified in compelling one under its control to endure torture or the death penalty is to repudiate two central principles of the liberal order: (1) the statement in the Declaration of Independence that the authority of government derives from “the consent of the governed” because it is simply absurd to pretend that someone who begs for clemency has “consented” to be executed (or is that just my being too literal-minded?); and (2) the Golden Rule, since it is absurd to pretend that, if you were a convict begging for clemency, you would still “really” wish to be executed and thus, in Rousseau’s exquisite phrase, “forced to be free”.

    My argument explains, without appealing to Scripture, how and why the “consent of the governed” is fundamental and why the Golden Rule has the absolute force it does. “Do unto others as you would have them do unto you” is another way of saying “Love your neighbour as yourself”, where “loving” someone means having his interests at heart. That your neighbour can suddenly become your enemy is only too obvious from so many modern instances,(Rwanda, the Balkans) so that the extension of the command to “Love your enemies” shouldn’t come as a surprise. Even when defending yourself against your enemies, you don’t have carte blanche: you have an absolute entitlement only to what is necessary and proportionate for your defence, even if it results in your enemy’s death: what you cannot justify is deliberately harming someone who, whatever he has done, is now in your power and can’t therefore be a threat to you.

    The big problem for liberal society is that it has separated itself from Traditional Religion but has not yet found a way of grounding the absolutism of the Golden Rule in something other than Scripture and therefore has no coherent way of teaching its children the morality it relies on for its survival, which leaves it vulnerable to the likes of Dick Cheney. I think this is the most important problem for philosophers at this moment and that my solution is a serious contender.

  33. duncan Says:

    Thanks Christoper – I think it’s apt that this comment of yours, above, be the last substantive contribution to the conversation. Apologies again that my comments here have been so very slow in coming.

    Best regards…


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: