Parrots and Concepts

In one of his favorite examples, Robert Brandom points out that while a parrot may very well respond differentially to colors, and even say “red” when presented with a red swatch, the parrot is nonetheless responding much as a thermometer does when it detects temperature changes and responds appropriately by turning on the heater. What is missing in both cases, according to Brandom, is the ‘practical mastery of the inferential articulation in which grasp of conceptual content consists.’ (Articulating Reasons 162). In other words, although the parrot can identify the swatch as red she cannot then go on and use this as a reason for inferring that it is colored, that it is not green, a squirrel, etc. A parrot cannot participate in the game of giving and asking for reasons, and thus they lack the use of concepts.

Integral to Brandom’s understanding of concepts is the further claim that each concept is necessarily related to other concepts, to the inferential web of premises and conclusions that connect one concept with another. As Brandom puts it, when it comes to the inferential connections that constitute conceptual content, a natural consequence is that ‘one must grasp many such contents in order to grasp any.’ To be able to use a concept one must already possess a host of other concepts, which leads to the question: from where do we get our use of concepts? How can we ever acquire the use of concepts if we must already possess them to use them? The answer for Brandom is that conceptual contents are ‘social achievements.’ The force of reasons that binds us to the inferential connections between recognizing and judging that a swatch is red and further judging that it is colored, not green, etc., is itself the result of rendering explicit the authority of socially sanctioned activities, an authority that legitimizes our uses as appropriate (à la Wittgenstein). ‘The conceptual faculty,’ Brandom thus claims, ‘is the faculty of grasping rules, of appreciating the distinction between correct and incorrect application they determine,’ and these determinations of correct and incorrect application derive, in the end, from the authority of socially approved practice, or forms of life as Wittgenstein put it. ‘This is why Brandom cites with approval John Haugeland’s claim, in the context of Haugeland’s discussion of Heidegger, that ‘all transcendental constitution is social institution.’ Brandom is thus a holist, as he avows repeatedly in his Tales of the Mighty Dead, and the whole that is most crucial for his project is the social whole; or, as Brandom tellingly titles the final section of his key work, Making it Explicit, ‘we have met the norms, and they are ours.’

It is clear that Brandom would have no issues with Hume’s example of the missing shade of blue. Since the application of concepts involves inferential ties to other concepts and judgments, and presupposes a holism as well, then the lateral relations of these inferential ties can provide the resources that would enable one to conceptualize the missing shade. It was precisely this holism and the lateral relations they entail that David Pears, in his critique of Hume, claims Hume needed to resolve adequately the problem of the missing shade of blue. Brandom would also no doubt argue that a parrot, since it lacks the use of concepts, would be unable to come up with the idea of the missing shade of blue. But Hume, as I discuss in earlier posts (see here and here), also does not think we would have a problem with the missing shade. There is evidence Hume may well have supported a form of social sanctioning to account for this. For instance, Hume remarked about how well the blind from birth poet, Thomas Blacklock, was able to describe things using colors. Hume theorizes that ‘Tis certain we always think in some language, viz. in that which is most familiar to us; and ‘tis but too frequent to substitute words instead of ideas.’ Rather than an actual grasp of the colors, Blacklock used the words as he had heard them used, in their socially sanctioned way, and was thus able to ‘substitute words instead of ideas.’ But it would be a mistake to read Hume as a holist, and Brandom is certainly correct not to do so; however, I think it is equally mistaken to claim Hume is an atomist. On my reading of Hume as a hyper-realist, what is to be stressed is the multiplicity of impressions and ideas rather than the identity of impressions and ideas that would be in line with the one-multiple dualism (following Deleuze here on this point). The very identity of an impression as identifiable, such as a shade of blue, thus presupposes a reality that exceeds its identification and is irreducible to the holism of the social. A Humean hyper-realism does not found the normative force of the inferential articulation of concepts upon the holism of the social, but rather it is the reality that needs to be filtered and selected in order to construct a whole. This is why Deleuze, in reference to Hume, argues that ‘one can only invent a whole, since the only invention possible is that of the whole.’ From a hyper-realist Humean perspective, therefore, the social is not the whole that is the transcendental condition for the normative force of judgments (recall Haugeland: ‘all transcendental constitution is social institution’), but the transcendental condition is invention itself, including the invention of social institutions (which is key to understanding Deleuze’s reading of Hume’s political writings), and this invention is irreducible to any whole or to any predetermining identities. It should be noted that Timothy Williamson (in The Philosophy of Philosophy) takes Brandom to task for an analogous reason, though for a very different end and purpose, when he challenges the understanding-assent link that is crucial to inferentialism. If I understand that this is blue, then I must also assent to the judgment that it is colored, that it is not a squirrel, etc. Williamson claims that the case for such understanding-assent links, and for the social holism upon which they are founded, does not stand up to closer scrutiny. It would take another post to detail these arguments, but the Humean critique of Brandom that I have made can be found elsewhere in different forms.

7 thoughts on “Parrots and Concepts

  1. Pingback: Spinoza, appetites, and inferentialism | Aberrant Monism

  2. Hi Jeff,

    This comment is a little bit late, but better late than never. I have a few issues with the reading of Brandom you’re putting forward here. Specifically, I can’t see the strong connection you’re drawing between Brandom’s semantic holism and the importance of the social institution of inferential norms, i.e., I don’t know what you’re talking about when you talk about ‘social holism’. This alarms me because you use it in your next post to contrast Brandom’s approach with a kind of Deleuzian metaphysics. But this would seem to be comparing apples and oranges. Brandom’s is a semantic, not a metaphysical holism. All it says is that the content of concepts consists in their inferential relations to other concepts, and that it is not possible to separate out a privileged set of such relations that are content-constitutive (i.e., to follow Quine in rejecting the analytic/synthetic distinction). This tells us nothing about the metaphysical status of the particulars that fall under these concepts, whether they are all bound up in some metaphysical whole (e.g., substance, the absolute, or a Deleuzian plane of immanence), or not.

    The social does play a very important role in making this kind of radical semantic holism possible, insofar as it is the social differences in scorekeeping perspective (and our ability to navigate across them) that makes possible anything like objective assessment and correction of the inferential connections between our concepts – it is the social nature of representation that puts our holistic web of commitments in contact with the world. However, I don’t see how this amounts to a ‘holism of the social’, or what the consequences of this would be.

    Could you clarify?

    • Hi Pete, Thanks for your comment. I know you have spent much more time thinking through Brandom’s texts than I have so I appreciate your thoughts. In this post and the few that follow I do tend to be more critical of Brandom, but there is much in his work that I think is indispensable. In particular, I agree with his nonreductive, nonnaturalist approach; or more precisely, I agree with the distinctive place Brandom gives to philosophy in contradistinction to the natural sciences. Thus for Brandom, as he puts it,

      …semantics is not just one more special science. It is (also) a philosophical discipline. Among its tasks is precisely the understanding if need be criticism of the concepts employed by other disciplines. (“Modality, Normativity, and Intentionality,” p. 599).

      This distinction between philosophy and the sciences, and what the implications of this are, has been a subject I have been working through of late and on this point Brandom and I are singing from the same hymnal. What I am doing in this post, and to get to your comment, has been trying to point to where I see my approach differing from Brandom’s. In short, I think Brandom is what you might call a Hegelian anti-realist. In The Tales of the Mighty Dead, for example, when addressing the problem of acquiring concepts, a problem he claims was insoluble ‘within the framework of pre-Kantian rationalism,’ Brandom claims that concepts are not just their inferential connections to one another, but, as he puts it,

      for Hegel the content of a commitment depends functionaly not only on its inferential connections and role in an expressive developmental sequence, but also on the commitments acknowledged and attributed by other members of the same community…This social dimension of Hegel’s functionalism, and the holism that inevitably goes with it, is picked up both by early Heidegger and the later Wittgenstein.’ (TMD p. 32, emphasis mine).

      So where my critique of holism comes in then is not so much because I think Brandom is committed to a metaphysical whole, such as substance or the absolute (and I’m not sure I would put the plane of immanence in there as you do but that’s another matter), but rather I find that for Brandom our conceptualizations of reality have their contents determined by the role they play within a Hegelian functionalism and the holism that goes with it; and this, for me, limits and predetermines what it is to be real, and hence is a form of anti-realism, or what you call deflationary realism. In fact, I think we largely agree on this criticism. In your “Transcendental Realism” essay, for example, you attribute, and correctly I believe, a ‘thin notion of reality’ to Brandom’s semantics. My critique of Brandom’s holism is thus a critique of this thin notion of reality in favor of a thick notion. I have detailed many of my arguments in my forthcoming essay for Deleuze Studies, “Between Realism and Anti-realism,” which you can read here if you are interested. In any event, I’m at work on a post at the moment where I defend Brandom’s general approach, especially his call to differentiate the task of philosophy from the natural sciences. Thanks again for your comment.

  3. Pingback: From Normative to Problematizing Semantics | Aberrant Monism

  4. Hi Jeff,

    Thanks for the essay (and thanks for referencing ‘deflationary realism’). I have a better idea of where you’re coming from now, but I still have some issues. I’ll try to number them and address them briefly:-

    1) In the essay you assimilate what I call ‘deflationary realism’ to anti-realism or correlationism, and I think this is a big mistake. As you correctly note in the essay, correlationism is a matter of anti-absolutism, but deflationary realism is not anti-absolutist in this way. The deflationist is happy for their to be absolute answers regarding the structure of the world, he simply thinks that these answers are provided by the semantics of our thought and talk about the world, not by anything resembling traditional metaphysical inquiry.

    2) In your response above you imply that there is an important link between Brandom’s semantic holism and his deflationary realism. Now, it’s certainly the case that many semantic holists (e.g., Wittgenstein, Quine, Davidson, Brandom) do adopt a deflationary realist position, and it may even be that such a holism does lend itself to that position, but I don’t think it’s the case that it implies it. Indeed, my own approach not only endorses Brandom’s semantic holism along with a more full-blooded metaphysical realism, but actually tries to use this holism in articulating this realism (i.e., in articulating a thick notion of reality). Now, it’s of course possible that my approach is flawed, but I still don’t see how this semantic holism and its related social dimension “limits and predetermines what it is to be real”.

    3) I recognise that Deleuze’s plane of immanence is quite different from Hegel’s absolute idea (and from Spinoza’s substance for that matter), but I do believe that it is still a whole. Badiou is right in claiming that the whole plays an important role in Deleuze, he’s just wrong in assimilating it to the kind of role that the absolute idea or Spinozan substance plays. I’ve written a bit about this elsewhere, but I don’t think I’ve provided a completely adequate account of it.

    4) In the essay you use Meillassoux’s appeal to the Cantorian transfinite to counter Lewis’ arguments about undermining futures. I’m not going to endorse Lewis’ arguments, but I will say that this whole move on Meillassoux’s part is pretty suspect, and you should be more careful with it. This is because Meillassoux never tells us precisely why the fact that there can’t be a set of all sets means there can’t be a set of all possibilities. Yes, there is an infinite hierarchy of infinite cardinalities, and thus no largest cardinality, but why can’t the set of possible worlds have a specific infinite cardinality? This is something that Lewis himself addresses in ‘On the Plurality of Worlds’. He doesn’t say precisely what cardinality the set of possible worlds has, but he does think it has a definite one, based on an explicit principle (called the principle of recombination) that provides him with guidelines for how the set of possibilities is constructed. If Meillassoux’s argument is to go through he needs to demonstrate something like the claim that the power set of any set of possibilities is itself a set of possibilities. Not only is this not obviously true, but demonstrating it would require some guidelines for constructing possible worlds analogous to Lewis’ principle of recombination.

    That’s all I can think of right now!

    • Hi Pete,
      Just a quick response to your numbered points:
      1) If the answers to the structure of the world are provided by the semantics of our thought and talk about the world, whether taken as absolutes or not, that is still anti-realist as I use the term in that reality is not affirmed as anything other than that which can be correlated with the structures of access to this reality. Perhaps this is a confusion in the use of terms, which I freely grant I may be committing, but I don’t see anything here that undermines my basic arguments.
      2) As for your attempt to embrace a holism that affirms a thick notion of reality, I am with you with respect to what you seek to show, though I’m not yet convinced that a semantic holism is up to the task. I’ll not argue that your position is flawed, but I’ll simply say I’m not persuaded based on what I’ve read as of yet (and with respect to your arguments that is very little, so this is not saying anything about you).
      3) I do think that the notion of The plane of immanence is a little discussed concept of Deleuze’s. There is plenty of space as a result for disagreement on this issue simply due to the fact that Deleuze provides little to resolve these potential debates. It’s possible that it is one of the issues prone to the intellectual mitosis I discussed in an earlier post. I do think that the significance of the nondenumerable in Deleuze’s thought works against a more traidtional conception of the whole, which is how you read the concept.
      4) I agree with you almost completely on this point. I’m presenting Meillassoux’s point, not endorsing it. In fact, I support the principle of sufficient reason and thus will naturally be unpersuaded by Meillassoux’s arguments. On a related note, I also find Meillassoux’s arguments regarding the meaninglessness of mathematical symbols as a justification for his arguments for the necessity of contingency rather unconvincing. I suppose as a cultural habit there is the need for the hot new French philosopher, and Meillassoux is the nom du jour. That said, his arguments are provocative and often instigate thought. And from my perspective, that is what philosophy is all about.
      As usual, Pete, your comments are much appreciated.

  5. Pingback: conceptual automata | Aberrant Monism

Leave a Reply

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

You are commenting using your 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 )

Connecting to %s