I’ve been trying to write on Brandom, relating his project to the problem of the threatened antinomy between autonomy and external obligations that I’ve been trying to develop. The somewhat lengthy results can be found below the fold. I’m not sure that I’ve nailed all the details yet, especially with respect to some of the more technical aspects of Brandom’s system, but I think the gist is clear.
Also, on the topic of Brandom, N Pepperell and L Magee’s excellent conference paper on him and Habermas can be found here.
I – The problem restated
So far, we have been introduced to an antinomy that arises from primitive but seemingly intuitively compelling conceptions of freedom and objectivity. The problem was articulated in terms of spontaneity and receptivity since characterising freedom as something implying self-direction and objectivity as implying external direction in normatively assessable matters appeared to invite incompatibilities at this level. Attributing authority to an agent to contribute to the determination of the propriety of its thought and action was one way of specifying what the agent’s (normative) freedom consisted in. However, leaving this authority unqualified came with the risk of neglecting the receptivity of the agent, understood as the role that things other than the agent play in determining the standards by which its activity is legitimately assessed and which the agent must be attentive to if they are to act appropriately. We are meant to care about receptivity, and so not to be able to rest content with a story that places all the emphasis on the spontaneous activity of agents to determine what it is to act rightly, because receptivity seems to be a condition of objectivity. Conversely, only attending to receptivity, without a spontaneous role for the agent that shows how they are actively involved in determining what they ought to do, risks failing to do justice to the full extent of our freedom.
So, if freedom implies some spontaneous contribution from the agent and objectivity implies some receptive constraint upon the agent, we ought to be able to tell a story that neatly reconciles them both. The goal is to be able to demonstrate two main things. Firstly, that the authority that the agent has over themselves is not problematically curtailed by the authority exercised over it by things other than itself. Secondly, that the rational constraint upon the agent imposed by others and the world at large is not threatened by the special role that the agent has in determining the normative standards it is beholden to. In short, we want to show how it is not contradictory to suppose that the agent is both genuinely self-directing and genuinely externally directed.
Thus, we have a relatively abstract problem concerning the compatibility of spontaneity and receptivity that is motivated by a story about freedom and objectivity. The focus upon freedom and objectivity is also a way of focusing and particularising an otherwise incredibly complex issue, which as Robert Pippin puts it is:
understand[ing] at the highest level of generality the relation between active or spontaneous thought and our receptive and corporeal sensibility and bodily embodiment … a problem so inclusive as to be common to theoretical and practical philosophy.[1]
By concentrating simply on freedom and objectivity, hopefully it will be easier to get some traction on a part of this general problem.
II – Brandom’s account
Brandom’s work is of interest because he takes himself to be addressing this ambitious general problem too.[2] At the same time, it is possible to focus in on a set of themes in his work that also addresses our more particular problem, since he takes seriously the need to account for freedom and objectivity.
In this section I shall reconstruct a Brandomian account of the connection between freedom and objectivity. To do so adequately will require some discussion of his philosophical project as a whole, with which we will start. Following on from this, we shall examine his account of autonomy and will see how it links up with his social model of conceptual normativity. Finally, in turn, his explanation of objectivity will be related to this social model in such a way that its relation to the autonomy of the individual will be made clear. Upon completing this task, in the succeeding section it will be possible to identify some areas to be questioned with respect to Brandom’s position.
Concepts
Put most simply, Brandom’s central concern is with explaining what someone must do in order to count as saying something. Even this basic formulation of his project indicates some of his methodological presuppositions with respect to explanatory ordering: prioritising pragmatics over semantics, use over meaning, and linguistic practice over appeal to mental states. But why does such a pragmatist approach to language have anything to say about freedom or objectivity? The answer is that explaining how linguistic utterances come to be endowed with propositional content requires Brandom to give an account of the role of concepts, where this exploration of conceptuality lies at the heart of his work.[3] We do not need to go into all the details of this story here. Instead, we can simply note a few features of Brandom’s understanding of concepts before going on to explain what their connection is to freedom and objectivity.
For Brandom, concepts are a species of norm — ones which determine the propriety of making certain moves, such as linguistic utterances, in a reason-governed game. As such, to possess a concept is to have an implicit practical capacity (a form of know-how) to discriminate between appropriate and inappropriate activities in certain contexts. Or put somewhat differently, it is to grasp an inferential role because it constitutes the capacity to determine whether something constitutes a reason for something else, and thereby to some degree to trace what someone (such as oneself) is rationally entitled and committed to do.
Framed in this way, and following Kant, discursivity is at the heart of subjectivity, since the two fundamental activities of the subject, namely judging and acting, can be seen to be ineliminably conceptual.[4] This is because judgement and action is intentional, which means that it possesses a content that we have committed ourselves to taking as true or making true respectively. What-it-is that one is committed to taking or making true – the content of judgement or action – is something that can be or not be fulfilled, and thus someone who commits themselves in such a way can either meet or (crucially) fail to meet the commitment that they have undertaken (e.g. by judging incorrectly, making an erroneous inference or failing to do properly what they intended). This means that this intentional content is specifiable in terms of norms. These norms, which are the determinants of content and articulate what it is to fulfil the commitments undertaken in action and judgement, are what Brandom means by ‘concepts’.[5] So, while Brandom does not think all norms are conceptual, his account gives pride of place to conceptual normativity in understanding our cognitive and practical relation to the world.
To recapitulate, Brandom wants to tell a story that explains what contentful language is in terms of the practical capacity of agents to follow conceptual norms. However, this is not simply a narrowly linguistic matter because these conceptual norms determine what it is to judge and act correctly (in co-operation with higher-level conceptual norms that they themselves are subject to). While Brandom’s account remains on a formal level, in that it does not address what he calls the messy ‘retail’ content of particular norms and simply describes their abstract structural relation to semantic content in general, it nonetheless develops an understanding of many of the general features of these norms. As such, his work has a wide application to issues relating to conceptual normativity in the round. As we shall go on to see, this is why it bears upon freedom and objectivity.
Autonomy and Sociality
We have seen that, for Brandom, in all intentional action the agent undertakes a responsibility determined and specifiable by conceptual norms. It is here that freedom enters his picture. This is because the authority of norms that we are thereby accountable to is always conditional upon us endorsing them, and this is something we must freely choose to do and is thus never an independent fact. So, for Brandom, rather than freedom being opposed to constraint, it consists in constraint of a special kind: rational constraint by self-imposed norms.[6]
What Brandom thinks make his conception of freedom an autonomous one is that there are limits upon the agent such that they are not the final arbiter of what they are rationally committed to. This is because Brandom separates the force of conceptual norms from their content. So, whilst it is the agent who binds themselves by endorsing norms and thus decides whether they are bound, it is not the agent who determines what it is that they are so bound to. For Brandom, the content of conceptual norms is socially administered, such that it is other agents who determine this through a process of negotiation between the agent who has bound themselves and those who take the agent to be bound. So, for example, it is up to me what move that I make in a certain game, such as saying ‘Kierkegaard was naïve,’ but the normative significance of this move – what other moves it rationally commits, entitles and prohibits me from making – is not up to me. This administered content is paradigmatically determined by how my fellows use concepts like naïvety and those inferentially related to it and thus how they will hold me responsible for my usage of it.[7]
We have seen that in this model there is a division of conceptual labour, so to speak, between the agent as the source of the force of the conceptual norms that they are responsible to and their fellows as the administrators of its content. The reason for this is that Brandom aims to do justice to both main aspects of the self-binding metaphor: that it is we who bind ourselves, and that we can be genuinely bound nonetheless. He thinks that attributing the force and the content of conceptual norms (both whether we are and to what we are committed) to us as individuals would leave an account open to a purportedly Wittgensteinian worry. The worry is that when “whatever is going to seem right to me is right … that only means that here we can’t talk about ‘right’.”[8] As such, he thinks that there must be must be more to being autonomously committed than merely taking oneself to be committed.[9] In Brandom’s terms, this means that our normative attitudes are not simply the same as our normative status. It is here, in accounting for the separablity of the commitments we take ourselves to have from those we really do have, that sociality is called upon to do some explanatory work.
One way of ascribing a role to sociality here that would fund a distinction between the individual’s normative attitudes and their actual normative status would be to attribute incorrigibility to communal consensus, majority opinion or the role of certain experts. In this case, whatever one of the aforementioned groups took to be the content of a conceptual norm just would be its content. This would be a way of identifying a criterion that put some distance between what the individual took the content of a commitment to be and the actual content of the commitment. So too, it would do so in a way that we might think preserved autonomy, since it would still be up to the individual whether they would use the concept in question, but just not always what significance that usage was going to have. However, as we will now see, Brandom rejects accounts of this general type since he claims that they do not do justice to objectivity.
Objectivity
Brandom’s complaint is directed towards accounts which make ‘the community’ (either in part or as whole) the highest authority which determines conceptual content and thus the propriety of occasions of concept-use. The problem is that they rescue the notion of autonomy from subjectivism only by failing to fully respect the objective commitments that our use of concepts imposes upon us. While various cultural and social norms appear to be unintelligible unless we refer them to such a communal model, Brandom claims that conceptual norms as a whole should not be understood to be of this sort. So, for example, we might think that it does not make sense to suppose that we all might be wrong about who is married or what the rules of chess are. But we need to allow for cases where this is not so. To take Brandom’s example:
our use of the term ‘mass’ is such that the facts settle whether the mass of the universe is large enough that it will eventually suffer gravitational collapse, independently of what we, even all of us and forever, take those facts to be. We could all be wrong in our assessment of this claim, could all be treating as a correct application of the concepts involved what is objectively an incorrect one. [10]
It is because concepts have a representational purport – and thereby the possibility of picking out objects – that they allow the world to ‘react back’ and determine that our use of them is mistaken. It is this feature that would appear to get lost in anti-representationalist accounts that, unlike Brandom’s, do not reconstruct some notion of representation, even though they reject representations as fundamental explanatory items.
As has been suggested above, Brandom wants to retain sociality as a crucial element in his account, though for the reasons just outlined this is not in order to set up either part or whole of a community as an incorrigible authority. Again, the force of conceptual norms is dependent on the individual whereas their content is dependent upon the activity of others. But whereas the former model understood the social on an ‘I-we’ basis, where the individual was related to a group perspective that determined the content of their commitments, Brandom understands sociality on an ‘I-thou’ or second-person basis. For Brandom, there is no such thing as ‘the group perspective’, but rather, following Davidson, simply individuals each interpreting the behaviour of their fellows.
Each individual is said to keep score on those they interact with (including themselves), tracking what the scoree takes themselves to be committed and entitled to do as well as what they are committed and entitled to by the scorekeeper’s lights. The details are fairly complex, but the central idea is that the scorekeeper can attribute both de dicto and de re attitudes to the scoree. De dicto ascriptions, attached to that-clauses, track what by the scorekeeper’s lights the scoree takes themselves to be committed and entitled to. De re ascriptions, attached to of-clauses, track what the scorekeeper takes the scoree to be committed to based upon the commitments and entitlements that the scorekeeper endorses and takes himself (the scorekeeper) to be committed and entitled to. So, scorekeeper and scoree may agree that the scoree has committed herself to the proposition that the man in the brown hat is a spy. This provides both scorekeeper and scoree with a shared level of content such that they can both enter into the game of giving and asking for reasons; such a game can get going so that the participants can interact with a common reference point. However, the scorekeeper and scoree may disagree in their auxiliary assumptions (e.g. with the scorekeeper committing himself to the proposition that the only person wearing a brown hat would be Orcutt, whereas the scoree commits herself to the proposition that Orcutt is another city). In this case, the scorekeeper can ascribe the de re attitude to the scoree that she believes of Orcutt that he is a spy, although she does not believe that Orcutt is a spy.
The potential asymmetry between de re and de dicto attitude ascriptions that Brandom’s deontic scorekeeping model allows is then used to explain the notion of objectivity. This is because the scorekeeper’s de re attitude ascriptions fix the objects that he takes the scoree’s commitments and entitlements (their truthmakers) and these can be different from his de dicto attitude ascriptions. The possibility of this difference explains the possibility of drawing a distinction between what is objectively true (by the scorekeeper’s lights) and what merely seems true to the scoree. What is more, the possibility of drawing such a distinction is hardwired into the process by which all semantic content arises and is thus available to all rational agents by virtue of them being able to play the game of giving and asking for reasons. Thus, as Brandom puts it, “objectivity is a structural aspect of the social-perspectival form of conceptual contents.”[11] Although, of course, we must always remain at the level of a scorekeeping perspective here, since this reference to this model cannot be used to determine what is actually correct. This would be the job of the messy business of inquiry itself.
Brief Summary
So as to get a firmer grip on what the connection between freedom and objectivity is in this story, it may be helpful to summarise some of the steps. Brandom wants us to understand freedom in general as a form of rational self-constraint. His model is one whereby we are autonomously bound by norms by actively undertaking a commitment to them. This account of autonomy is used to explain how conceptual norms come to bind us, since in using concepts by making moves in reason-governed games we choose to undertake the commitments that specify the content of those concepts. However, he thinks that it is only the normative force that we introduce as individuals by binding ourselves. The content – to what rather than whether we are bound – is not up to us though. This content is determined through a complex process of negotiations with one’s fellows and is tracked by the deontic scorekeeping that agents must be able to engage in so as to enter the game of giving and asking for reasons. Features of this deontic scorekeeping also explain objectivity, or how our normative attitudes of taking certain uses of concepts to be correct are answerable to attitude-transcendent facts. This is because it introduces the permanent possibility of making a distinction between ascriptions of what one takes to be the case and ascriptions specifying the objects that determine whether one is correct.
III – Response
We might think that this Brandomian account will allow us to avoid the problem of the compatibility of freedom and objectivity. The reasons for this will be set out next, followed by some concerns with such a proposed solution to our worries.
If we recall, our problem was to do justice to both normative self-direction and external direction in such a way that also reconciled these two notions at the level of the individual agent. Brandom’s account looks like it may be able to do this, since it grounds all normative authority on the autonomy of the agent, but in a way that tempers this role for spontaneity by moments of receptivity that make this authority also conditional upon social practices and, ultimately, independent facts in the world. Since all normative authority must be autonomously endorsed, the individual agent would seem to be self-directing in a way that does not appear to be problematically curtailed by things other than itself. However, the agent is still genuinely rationally constrained by others and the world of large, since they are not the ultimate arbiters of what commitments they have or whether they have met those commitments. Thus, Brandom’s account seems to be able to solve our problem.
However, some reservations might be expressed about Brandom’s position. I shall conclude by identifying three of these here that seem most relevant to the specific problem that we are concerned with.
(1) Firstly, we may think that Brandom’s account fails to provide a plausible model of autonomy or objectivity, since it presents us with a forced choice of sorts. One of the things that made Brandom’s account attractive was that it seemed to be able to show how the authority of objects over our use of concepts was not a threat to our autonomy. The way it did that was to make that authority conditional upon the prior endorsement of the agent. However, if this is the case then it can seem as if we have not done justice to any worldly authority at all and thus have a notion of objectivity that is too weak. This is because there is no explanation of how we learn from the world in such a way that it is the world that dictates how our practices should change and not a prior choice to acknowledge an authority that then in turn gives the world a significance such that it compels us. There is always the mediation of our endorsement of a norm as binding upon us that we can refuse to carry through.
Does this objection end up begging the question against broadly constructivist accounts of normativity like Brandom’s? Perhaps. There is a sense in which one could escape the objective commitments that concept-use commits someone to on Brandom’s account, which may threaten to weaken the degree to which such commitments are objective ones. However, although addressed to a slightly different, Brandom seems to have a reply to worries of this general sort:
“So in the conceptual normativity implicit in linguistic practice we have a model of a kind of constraint-loss of negative freedom-that is repaid many times over in a bonanza of positive freedom. Anyone who was in a position to consider the trade-off rationally would consider it a once-in-a-lifetime bargain. Of course, one need not be a creature like us. As Sellars says, one always could simply not speak-but only at the price of having nothing to say. And non-sapient sentients are hardly in a position to weigh the pros and cons involved. But the fact remains that there is an argument that shows that at least this sort of normative constraint is rational from the point of view of the individual-that it pays off by opening up a dimension of positive expressive freedom that is a pearl without price, available in no other way.” [12]
A reply of this sort points to the otherwise inaccessible possibilities that self-constraint opens up and thus seemed to demonstrate the unattractiveness of a situation in which we renounce engagement in practices that would institute objective commitments. Still, we may nevertheless choose to do so, even if this is not rationally prudent. Moreover, this can make autonomy look like a forced choice – one in which we ‘willingly’ endorse a set of conceptual norms, but only under the threat of the paralysis in our activities as agents.
(2) Secondly, we may be concerned about the sort of conceptual division of labour that Brandom sets up that separates the normative force from the content of commitments such that individuals determine whether they are committed but not what they are committed to. How far these two elements can come apart may be cause for concern since it seems that someone could autonomously undertake an objective commitment through their use of concepts but find that its content is different, and perhaps radically so, from the one that they take themselves to have undertaken. If this is possible, again we may then wonder whether Brandom has done justice to the notion of autonomy that is meant to ameliorate objective commitments in such a way that they do not conflict with the individual’s self-direction.
We might expect Brandom to reassure us by means of some minimal conditions of normative commitment that suppose that the force and content of conceptual norms cannot become diremptive – that the autonomous agent must somehow be able to recognise what they have done as undertaking a commitment that others get the final say over what counts as fulfilling it. Far from it though, since Brandom uses the example of the ‘Queen’s schilling’ to illustrate his account.[13] This was a nineteenth-century practice whereby illiterates could demonstrate that they had accepted enlistment in the navy by accepting a schilling instead of having to sign a contract. However, recruiters would often simply trick drunkards into accepting such a schilling so as to enlist them. Brandom takes this a helpful example of the division between undertaking a commitment whose normative significance one only fully discovers at a later time. However, if this is what autonomous commitment amounts to then it seems very cheap indeed, for it seems very strange to suppose that the drunkards in question could be said to have autonomously bound themselves by accepting the Queen’s schilling. This suggests that, at least as it stands, Brandom’s account of autonomous obligation is problematic.
(3) Thirdly, and perhaps most seriously, we may think that Brandom’s account falls prey to the so-called Kantian paradox. As we have seen, Brandom is attendant to the dangers of supposing that all legislative activity takes place at the level of the individual where all binding oneself would consist in would be taking oneself to be bound. However, it seems that Brandom may have problems when it comes to the status of rational norms themselves. If we try to construe these on a self-legislative model whereby we autonomously adopt norms, then we are faced with two options. Either this autonomous endorsement is subject to rational norms or it is not. If it is so subject, then we will want to know how these norms came to be authoritative, where, if the answer is more self-legislation, then the same question can be reiterated at higher levels. If it is not so subject, then it seems that such legislation is arbitrary – where no simply random act would seem to be able to command the rational authority that it is supposed to institute.
A Brandomian might object to the stark way in which we have setup the institution of rational norms. Instead, they may say that there is no single heroic act that moves us from irrationality to rationality. Rather, rational norms would be something that arose from countless concrete practices of treating something as being a reason for something else – this being a diffuse and distributed sense of rational constraint that it is absurd to suppose has an easily identifiable starting-point. If this response is to work however, we will need to see a much more fleshed out account of it that is lacking in Brandom as yet.
[1] ‘McDowell’s Germans’, manuscript version, p.2.[2] See MIE, p. 614-8, inter alia.[3] We can gather this from the very first line of Articulating Reasons: “This is a book about the use and content of concepts.” (2000: p.1, emphasis in original)
[4] In fact, using concepts is in some sense all a Kantian subject can really be said to do, since judgement is synthesising representations according to concepts and acting is undertaking behaviour on the basis of a concept of law (i.e. a maxim).
[5] “The key to the conceptual is […] the special sort of authority one becomes subject to in applying concepts-the way in which conceptually articulated acts are liable to assessments of correctness and incorrectness and incorrectness according to the concepts they involve.” MIE: p.8.
[6] Cf. MIE, p.50-2 and TMD, p.219.
[7] We need not give a full account of this negotiation model here. See MIE at 8.6.2 where seeds of it can be found; however, it is only elaborated at length in Brandom interpretation of Hegel in §3-4 of ‘Some Pragmatist Themes in Hegel’s Idealism’ in TMD. Perhaps the fullest treatment of it is given in the third lecture, ‘History, Reason and Reality’, of his very recent Woodbridge Lectures, ‘Animating Ideas of Idealism: Semantic Sonata in Kant and Hegel’.
[8] PI, §258
[9] “Endorsing a rule gives it a grip on us. Part of that grip is that the rule does not mean just what we might later take it to mean.” MIE, p.52.
[10] MIE, p.53. See also, p.137: “actual practical attitudes of taking or treating as correct institute the normative statuses of materially correct inferences, and these material proprieties of inference in turn confer conceptual content-that content nonetheless involves objective proprieties to which the practical attitudes underlying the meanings themselves answer.” Or again: “Our cognitive attitudes must ultimately answer to … attitude-transcendent facts.” (ibid.)
[11] MIE, p.597.
[12] ‘Autonomy, Community and Freedom’, second Woodbridge Lecture, manuscript, p. 27-8.
[13] MIE, p. 163.
Grundlegung 
Pingback: Roughtheory.org » Grundlegung on Brandom
Nice discussion. A quick question. In summarizing how deontic scorekeeping explains objectivity, you write: “This is because it introduces the permanent possibility of making a distinction between ascriptions of what one takes to be the case and ascriptions specifying the objects that determine whether one is correct.” But on Brandom’s account, isn’t true to say that the “ascriptions specifying the objects that determine whether one is correct” are actually just the de re attributions of what someone else takes to be the case? And aren’t these ultimately two different things – since what someone else takes to be correct may not be what is objectively correct?
Hi Andrew — thanks for your comment.
Just to clarify (although apologies if I mischaracterise what you are saying), I don’t think that the de re ascriptions are just what the scorekeeper takes to be the case. After all, they are the ascription of an attitude to the scoree. Rather, what the scorekeeper takes to be the case constitutes a set of auxillary assumptions in de re ascription that specify different inferential relations of commitment and entitlement in relation to the particular commitment in question than the ones that the scoree may recognise. This seems to be very different from saying that de re ascriptions are ‘attributions of what someone else takes to be the case’ alone, since they are employed with relation to a commitment of the scoree’s — one that the scorekeeper may not share.
As I understand it (which is not yet as well as I’d like), the de re ascriptions use the scorekeeper’s set of commitments and entitlements to specify objects that the commitments of the scoree are answerable to by the scorekeeper’s lights. The de dicto ascriptions do not specify such objects, using the scoree’s commitments simply to specify the proposition that the scorekeeper thinks that the scoree takes themselves to be committed to. So, the contrast is not between what the scorekeeper takes to be true and what the scoree does (or what the scorekeeper takes the scoree to take as true). Rather, it is, all from the scorekeeper’s persepctive, the contrast between what the scorekeeper takes the scoree to take themselves to be committed to and what objects settle whether the scoree is correct to do so. With this structure in place, it is possible to generate clashes such that someone can be believe-of Orcutt that he is spy but believe-that he is not a spy, etc.
Now, having said all that, the worry that what a scorekeeper takes to be true may not be true might still be thought to return. This is because the auxilliary assumptions of the scorekeeper might be wrong, and thus the de re ascriptions may specify the wrong objects to which attributed commitments are answerable to. But the solitary scorekeeper is not the final determinant of the conceptual content of the commitments of all those whose scores they track. They simply track commitments and entitlements in such a way that the game of giving and asking for reasons can get going, where the negotiation model seems to come into play as the determinant of content (although I’m not sure how distinct scorekeeping and negotiation are really). As such, a degree of fallibilism is allowable here.
Thanks Tom, that is very helpful. I’ll let you know if I have any other thoughts.
I’m not sure about your second reservation. It doesn’t seem Brandomian to include a condition where an agent can recognize a norm binding them just by it being a norm. In the case of the queen’s shilling, the agent will not recognize the commitment he has made, much like one won’t realize all the commitments one takes on in virtue of saying, say, that copper melts at 1084C. One may not know the further commitments because there are further things that determine what those commitments are but nonetheless entitle people that know them to update your score in their books accordingly. The drunk who takes the queen’s shilling will have a different set of books on his commitments than the officer that gave it to him. The minimal condition that you offer to Brandom seems to be something that he’d reject because of this.
The queen’s shilling case was offered as an example of a commitment defined in terms changes in status and what sort of things one needs to do in order to count as undertaking such a change. The bit about learning the significance of the commitment later is not unique to that. Most commitments undertaken, at least theoretical ones, will have consequences that we only learn about, and can only learn about, at a later date.
I think the question raised in the first reservation is really interesting, but I have a question about it. You say, “This is because there is no explanation of how we learn from the world in such a way that it is the world that dictates how our practices should change and not a prior choice to acknowledge an authority that then in turn gives the world a significance such that it compels us.”
How we learn from the world is difficult, and I’m not sure that Brandom has a lot of answer there. The world is supposed to get connected to our concepts via perception and the RDRDs. I don’t have a great grip on that story, but that doesn’t seem to be what you are worried about. The worry is more like the rule-following worry about reinterpreting a rule, right? When you say, ” There is always the mediation of our endorsement of a norm as binding upon us that we can refuse to carry through,” it sounds a little more like the rule-following worry. What do you mean by “the mediation of our endorsement of a norm”? There are a couple of ways that Brandom would, I think, flesh out that idea. One is avowing the norm in speech. This isn’t enough to count as following it, although the other agents might temporarily adjust your score according to that assertion. Another way is doing something that accords with the norm. This is supposed to get around the rule-following problems, as far as I understand it, because one is not making the norm explicit in a propositional form. It remains implicit in the practice, to be followed, and since it is implicit does not stand in need of an interpretation or, possibly, an endorsement. It seems like you might be leaning more towards the former, making the norm explicit, which is not the route Brandom wants to go.
Sorry about the delay in getting back to you…
As for your first worry, I don’t mean to deny that for Brandom one can be committed to something without knowing that one is committed, and you are certainly right to point out that many if not most committments will have consequences that the person undertaking that committment will not be aware of.
Maybe one way of reapproaching the problem is by first identifying three ‘moments’ that Brandom wants to do justice to: phenomenalism, realism and autonomy. The phenomenalist moment is part of the explanatory methodology and refers all aspects of the account told to the mechanics of the practice of people taking themselves and others as committed. As such, the only explanatory resources available are what we as individuals can discern from our immanent perspective, thus excluding robust metaphysical appeals of any quasi-platonist variety. The more realist moment of the story (the one at least meant to assuage some realist intutions) is the fact that the propriety of the ground level takings-as — that is, whether a purported committment of someone’s ought to have been undertaken — are ultimately determined by the objects specified in their commitments. Finally, the autonomous moment is provided by the condition that whether someone undertakes a committment is up to them, though not exactly what they are committed to (i.e. the inferential consequences they are then subject to). The issue is thus whether Brandom can hold onto all three such moments.
Emphasising the phenomenalism too much, such as making someone’s actual commitments just what many or most scorekeepers say they are, runs the risk of not doing justice to autonomy — something Brandom evidently wants to do. However, I’m not sure that it good enough to simply say that in the Queen’s schilling case there will just be two different sets of books. There will be, of course — but a further element seems to come into play. This is because I take it that, while still under the auspices of the phenomenalist condition, the autonomy condition will govern the propriety of attributing commitments to oneself and others.
So, the Queen’s schilling case seems to show that, other things being equal, Brandom takes the recruiters to be justified in attributing the commitment to the drunkard. If he is being consistent, this suggests that the autonomy condition must be very weak, being something of the form ‘Doing something that makes it appropriate for others to attribute a commitment to you’, where what is ‘appropriate’ is not something for the individual to decide. Whilst I take it that we need to avoid some form of ‘Humpty Dumptyism’ (i.e. “My words mean whatever I want them to mean”), and thereby allowing that I can have commitments that I do not take myself to have, this Brandomian model seems to go too far. For when it comes to originally deciding to do something such that I undertake a new commitment (rather than inherit a commitment inferentially from others I am already committed to), it does not seem to be an autonomous undertaking if I have no inkling at all of the normative significance of the act I perform. Thus, on the question of autonomy, saying ‘This is copper’ without being a metallurgist, and therby undertaking a bunch of commitments I am unaware of, seems very different to accepting a schilling and thereby autonomously undertaking a commitment to join the navy.
As for the other worry (that is, my first reservation, your second comment), it wasn’t the rule-following considerations that I had in mind here though I see why you bring them up and how they might be used against my sketchy objection. My thought was simply that Brandom might be said to make the authority exerted by empirical objects second-rate because they can only normatively compel us if we first somehow agree to let them. The problem that you suggest seems to make a parallel between the sort of mediation that takes place in endorsing a norm, e.g. by using a concept, and the (purported) mediation of an interpretation of a rule that stands between the rule and its application. And if that was what I was doing, I might be open to the charge of misconstruing Brandom as some sort of regulist about normativity, who takes all norms to be explicit rules, whereas in fact he thinks that explicit rule-following is always conditional upon implicit practices of rule-following. There might be further problems with such an analogy too, such as the rule-following considerations concerning what a rule demands rather than whether it is in force. But the original parallel was not one I was trying to make, though I’ll have a think about it. (Of course, Brandom misreads Wittgenstein in his discussion of regulism, etc. in Ch.1 of MIE and I think draws some misguided conclusions from all the rule-following material.)
My point was instead a rather cruder one, that whether or not ‘endorsement’ is explicit or implicit in practice in a non-propositional way, normative authority is conditional in a way that we might think threatens ‘objectivity’. As a robust objection it is too weak and, as I say, risks begging the question in favour of some tubthumping directed at the anti-realist tendencies in Brandom. The more sophisticated form of it is reservation 3, and it is that line of thought in general which I think forms the most serious objection to Brandom.
I think I see what you are saying with regards to the queen’s schilling. I’m not sure there is much of a distinction drawn in Brandom’s view between practical commitments, such as what you get from the queen’s schilling, and theoretical commitments, such as what you get from assertions. It seems like you don’t have a problem with taking on (inheriting) theoretical commitments of which you are unaware but you do with undertaking practical commitments whose normative force you are unaware of. I agree that Brandom, at best, imposes a fairly weak condition on autonomy. In the case of the queen’s schilling, it doesn’t seem like he quite ends up with Humpty-Dumptyism though. The schilling practice was set up already and allowed some people to do more things autonomously. It would seem like Humpty-Dumptyism if the schilling case were set up such that the recruiters taking someone to have accepted the coin under whatever conditions counted as that person undertaking the commitments associated with joining the army.
I think you are right that the third objection is the big one for Brandom. There is a forthcoming paper by Rowland Stout called “Being Subject to the Rule That Tells You to Do What the Rules Tell You to Do”, which I think is in agreement with you, though I haven’t read it. I’m not sure what sort of response Brandom would offer to your third objection, but the meta-problem you raise is probably going to cause trouble for him.
There are probably some problems that come out of Brandom’s take on Wittgenstein and the rule-following discussion in chapter 1. It burdens the book with some odd commitments and an odd trajectory for responding to certain problems.
Perhaps I wasn’t clear enough by what I meant by Humpty-Dumptyism. This was a charge that I think Brandom can and does avoid: the idea that what the speaker means, with all the inferential commitments that they thus take on by saying something, is only up to the speaker. I think Brandom is right to try and formulate a position in which the normative significance of an utterance is not entirely down to the speaker; however, he seems to go too far in how he handles the details of an account that makes the normative significance of an utterance something that others determine in large part.
(I think we’re using the term ‘Humpty-Dumptyism’ the other way round from each other, whereby I took it mean the first situation and you the second. Maybe there’s a Brandomian meta-debate to be had about what it ‘really’ means… :o) )
If Brandom’s autonomy-preserving distinction between whether and to what we are committed comes down to saying that whether we make an utterance or some more implicit game-move is up to us, whereas what the significance (i.e. commitments it brings) of that act is will be up to others, then in that case his autonomy condition seems too weak. I take this to be the case whether it is practical or theoretical commitments at stake.
For example, it seems too much to me to say that someone can correctly attribute a commitment to joining the navy to the drunkard simply because he accepts the coin, but equally were I to say ‘That tree is dying’ and for others to be able to correctly take me to be commited to some thesis about the behaviour of cats because they play games in which a token of that type can have that inferential significance. But I am not objecting to the idea that promising to help someone move house might mean that I undertake lots of commitments that I am unaware of, and that if I had known them at the time I would not have promised; so too, in the case of saying ‘this is copper’ and being commitmed to the object having a melting point of 1084C where I am not aware that this is an inferential consequence of making the original utterance. In short, I think there has to be a tighter connection between whether and to what one is commited than Brandom allows, where one ought to be able to have some grasp of what sort of thing they might be getting into, while still allowing that the two sides do not simply collapse into each another.
Perhaps I am missing the signifcance of the distinction that you seem to draw between the recruiters attributing a commitment and the drunkard themselves undertaking one. I take it that if the recruiters correctly attribute a commitment (as Brandom seems to imply) then the drunkard must have themselves undertaken that self-same commitment, even if the drunkard did not recognise the full signifcance of their act. I am construing the autonomy-condition as a criterion for what it is to undertake a commitment and therefore a condition governing the propriety of other’s scorekeeping attributions too. But perhaps that is a misconstrual of its role, perhaps in light of there being no robust conditions on the propriety of scorekeeping at all — no higher authority to appeal to that can determine correct scorekeeping. I took Brandom’s claim that it is norms ‘all the way down’ to exclude such a reading, but perhaps that was too hasty of me. At least, something along these lines might explain our different conclusions in this case.
Thanks for the link to the Stout paper — I shall certainly give it a read.
I had thought you were using Humpty-Dumptyism in an extended sense. I was trying to use it in that sense when talking about the Queen’s schilling. Namely, the soldiers interpreting the drunkard a certain way made that interpretation correct, sort of like meaning by one’s words whatever one chooses. Since that wasn’t the worry, it probably isn’t important to get into.
I think the autonomy condition you say is too weak is one that Brandom is happy to allow in some, maybe even many, cases. It seems to be there for a lot of theoretical and practical commitments. It still seems like accepting that something else can fix the correct consequences of your commitments in the theoretical case, e.g. copper, carries over in the same way way to the Queen’s schilling case. I’m not sure the “that tree is dying” example undermines it. In the schilling case, the drunkard is in a position to be playing the game of accepting the schilling by being a subject of the queen. A foreigner can’t be correctly taken to enlist since he’s not in the right sort of social position. If the drunkard is a foreigner, accepting the coin can’t count as enlisting. The question for the tree case is whether or not you’re in the right sort of social position to be playing the game where your statement can be interpreted as being about cats. If you can’t play the cat game, it doesn’t seem to undermine the schilling example; you would be like the foreign drunkard. If you can play the game, then it seems like it wouldn’t be that unreasonable for them to take you to be talking about cats. Brandom would likely allow that it is ambiguous (indeterminate?) what language-game you are playing, depending on the score-keepers involved (in the same vein as Davidson allowing that it is indeterminate which language one is speaking). If this isn’t an option that Brandom can take for some reason, then I’d agree with your take on the tree case.
Was there somewhere in MIE or Articulating Reasons where Brandom endorsed a stronger version of autonomy than the one described as too weak? The idea seems like it should fit into the picture of MIE but I don’t have it situated in my map of the book.
(Sorry for taking so long to reply. The holidays kept me busy then the new semester started.)
After giving it some more thought, I’m not entirely happy with my response to the tree case. I’m not sure if you meant that the community took your tree statement to have consequences involving cats (e.g. because cats kill trees around there) or to itself be about cats and so have cat consequences (e.g. because they treat “tree” type utterances as being about furry little critters that meow). I’m guessing the latter. It seems like Brandom would probably agree that it would be too much to say that the utterance about a tree is taken to be about cats because of how a community treats tokens of that type. This is because he would probably appeal to the representational aspect of the content of your utterance. This would make it such that they wouldn’t be correct in attributing those cat commitments to you. This mode of response isn’t available in the case of the queen’s schilling since the practical commitments attributed do not have the sort of representational aspect that the theoretical commitments stemming from your claim do. This is a disanalogy between the two cases. The worry is seeming bigger though.
Pingback: Brandom’s master strategy « Grundlegung
For what seems to me a slightly stronger account of autonomy, I was thinking of material in TMD, like the discussion towards the end of section II and then section III of ‘Some Pragmatist Themes In Hegel’s Idealism’. There, Brandom discusses the processes of negotiation of the content of commitments and connects this with what he thinks is a Hegelian account of recognitive community. Perhaps in light of this discussion, you are right to point to the fact of shared communal bonds that the drunkard has with the recruiters that will be to some degree absent with foreigners. If the drunkard can somehow opt out (or fail to opt in) to the set of social practices that make it appropriate to confer the social status of an enlisted man, then Brandom seems to be on marginally better ground. (Perhaps some appeal might be made here to the idea that we must identify with some community, analogous to the case of the necessity of binding ourselves in some ways if we are to be rational at all.)
I’m still worried about some of the details of this position though.
In some senses the discussion in this article reinforces my worries about a lack of robust autonomy for agents since so much still seems to hinge on others’ treatment of tokenings, like so:
As I’ve said previously, I am quite happy to accept a social account of autonomy whereby that autonomy is grounded on a dependence upon others, but this way of spelling it out, dividing the labour between my choice of tokenings and the significance that tokening has, makes me uneasy. I suspect this might stem from lacking a sufficient grasp of Davidson on radical interpretation, etc., and an attendent hesistation over whether I ought to accept all the consequences of such a Davidsonian position.
On the other hand, I am worried that in some others ways the autonomy conditions are too strong in connection with the model of community membership. Here there is a requirement of reciprocity such that recognition by the agent and members of the community are necessary and only jointly sufficient for membership. Your example of Camus and being an existentialist seems to illustrate some of the difficulties with this. So too, it is not clear to me exactly how this recognition is supposed to work in light of other things Brandom says. In particular, I’m not sure if it is meant to be a normal case of commitment such that I choose the piece I play and others determine its significance. It seems there might be problems with taking this to be case *before* someone is a member of said community (although some sort of retroactive model akin to the common law case might work here — that the true signifcance of what I do only becomes apparent later in terms of how my actions are subsequently treated).
Sorry that was a bit of a mess of ideas; I think I need to put some sustained thought into all of this!