A Note on Transcendental Arguments

March 13, 2011

I’ve not done my due diligence in the philosophical literature here – there are a heap of texts discussing the nature and validity of transcendental arguments – but I wanted to make a couple of quick observations. The immediate relevance of the topic is the question: in what sense, or to what extent, can Brandom’s arguments in Making It Explicit be regarded as transcendental?

I take it that transcendental arguments are arguments about conditions of possibility. Given an accepted feature of the world or of our experience (x), a transcendental argument operates by claiming that another thing (y) is a necessary condition of the possibility of that thing x. Classically – in Kant – the argument works by finding conditions of possibility of experience in the synthesising activity of the transcendental subject; but the form of the argument is broader.

I want to make a couple of observations about the status of these sorts of arguments.


1) In the classic Kantian form, and in many other versions of ‘transcendental’ arguments that I’ve seen, there’s an implicit or explicit suggestion that the things described as ‘conditions of possibility’ inhabit in some sense a different plane from those with which the argument begins. In (at least a common interpretation of) the Kantian argument, for example, we begin with experience of an empirical world, with its causal connections between objects, etc., and conclude with the synthesising activity of the transcendental subject. The transcendental subject is taken to constitute the empirical world – without such synthesising activity, the empirical world would not manifest to us.

From here philosophers often move to suggesting that the features of the empirical world so synthesised cannot impact on the transcendental subject doing the synthesising. For instance, it is sometimes claimed that the transcendental subject is not impacted by causal laws, because causal laws are in some sense made by the transcendental subject as it synthesises experience. (Alternatively, we have no way of knowing whether causal laws hold in reality as it is in itself, because the empirical world constituted by the transcendental subject is, by virtue of its transcendental constitution, transcendentally ideal.) These moves make the transcendental subject seem like a very spooky non-empirical thing. It’s as if there are two planes of existence – actual empirical existence, and then the conditions of possibility of that existence, which are somehow untouched by the empirical.

The first point I want to make is that this latter claim doesn’t follow from anything that’s gone before. It’s perfectly possible for the subject whose synthesising activity is a condition of possibility of empirical experience also to be an empirical thing. Indeed, there’s no intrinsic reason, given the structure of the argument, to believe it isn’t. It’s perfectly reasonable to ask what the actual empirical features of the world are that generate or are this synthesising activity – there’s no reason why research into this question shouldn’t be an empirical endeavour just like research into any other feature of experience or the world. In that sense, there’s no intrinsic opposition between ‘transcendental’ arguments and ‘naturalised’ epistemology. (And this goes not just for transcendental arguments that derive some kind of synthesising subject, but for any kind of transcendental argument at all.)

In other words, there’s a slippage in a lot of transcendental arguments. Philosophers move from talking about conditions of possibility (what is necessary for x to be true) that are ‘transcendental’ in the technical sense of being necessary for the existence or experience of x – to talking as if the ‘transcendental’ things thereby discovered are somehow also transcendent – or at least different in kind from empirical things that ordinary empirical research might discover. And there’s no justification for this slippage.

Brandom does not participate in this slippage. Thus Making It Explicit does make a set of transcendental arguments about the conditions of possibility of normativity. Brandom believes he can make the case that certain kinds and relations of social practices are necessary in order for normative and conceptual content to be possible at all. This is a transcendental argument in the first sense defined above.

However, Brandom does not then make the move to acting as if these transcendental conditions of possibility are also different in kind from empirical phenomena discoverable by regular empirical research. Brandom ‘black-boxes’ the specific empirical phenomena that generate his classes of normative practice – he isn’t interested or competent to engage in the detailed study of what kinds of biological structures an organism might require in order to participate in the social practices his work discusses, for example. Nevertheless, Brandom does not mean to suggest that the practices his work analyses – and from which his analysis of normativity is derived – are non-empirical. As I’ve already argued, Brandom’s work is naturalistic in this respect.

So this is the first point I want to make: Brandom’s work makes transcendental arguments, but the kinds of transcendental arguments Brandom makes should not be taken as incompatible with – or addressing different kinds of phenomena to – empirical or scientific research programs.

I’m obviously happy with Brandom’s naturalism. However, for myself – and here I’m moving away from exegesis of Brandom into criticism of Brandom – I think that the kind of transcendental argument Brandom deploys still makes claims that are too strong to be justified.


2) Moving back to discussion of transcendental arguments in general, then, the second point I want to make is that transcendental arguments (even of the limited kind deployed by Brandom) rely on unjustifiably strong claims about counterfactual possibilities.

A transcendental argument is of the form: x is the case. y is a condition of possibility of x. Therefore y is the case. I want to suggest that in almost all cases in which arguments of this form are used, we are not justified in claiming that y is a condition of possibility of x (that y is a necessary requirement for the existence or manifestation of x) – simply that y is a mechanism by which x can be produced. I don’t think we generally have the capacity to know that no other set of circumstances could also be capable of producing x. I don’t see on what epistemological basis such a strong claim could be made.

In the case of Brandom, I think Making It Explicit presents a very strong case that the social practices Brandom describes are capable of generating the phenomena of normative and conceptual content with which his argument concludes. I don’t see, however, on what basis Brandom can legitimately claim that only these practices are capable of generating such phenomena. That may be the case – Brandom may be right – (as it happens I don’t think he is) but I don’t see on what epistemological basis such counterfactual claims could possibly be made – the fact that we can’t think of an alternative set of conditions generative of such-and-such a phenomenon does not mean that an alternative set of such conditions is impossible – we may simply be victims of a lack of imagination, reasoning capacity, or information. And I don’t see how this latter set of possibilities could ever be definitively excluded from our reckoning. I therefore don’t see how even ‘weakly’ transcendental arguments, of the kind I take Brandom to be making, can be legitimate. Our claims should be more modest. Not “y is a condition of possibility of x” but rather “y is a way of producing x, and it’s a way we’re familiar with”.


To summarise, then, the kind of ‘transcendental’ argument Brandom is making in Making It Explicit does not conflict with, and indeed is strongly compatible with, naturalistic research and description. Just because y is a transcendental condition of x doesn’t mean that y isn’t also an empirical or causal condition of x. For an argument of the type Brandom’s making, these can in fact be the same kind of claim.

However, I also think Brandom overstates the strength of his argument, and that his argument should not in fact be taken as a transcendental one in any sense (though Brandom certainly thinks it is). All Brandom can show in Making It Explicit is that the practices he describes are one way of generating conceptual and normative content. He cannot – as he seems to wish to – show that these are the only ways of generating conceptual and normative content. Such claims are, in fact, outside our ability to make: as limited creatures without God-like capacities for evaluating empirical counterfactuals, we cannot conclusively know what all else might be possible.


3 Responses to “A Note on Transcendental Arguments”

  1. Dave Says:

    Do you mean that by grounding his arguments empirically he cannot – in a Humean sense – prove their necessity?

  2. duncan Says:

    Yes, I guess so – although my position isn’t really based on the degree to which Brandom grounds his work empirically, so much as on (what I take to be) the impossibility of anyone successfully making transcendental claims at all.

    (I do think people can make claims about necessity, but these are either conventional – as I take logic to be, for example – or open to empirical refutation – as I take a claim about the universality of a physical law to be, for example.)

    So yes, this definitely has a Humean flavour, of the “consign it to the flames” variety.

  3. Dave Says:

    Thanks for the quick response. I think I should have asked ‘Do you mean that by grounding his arguments empirically you show that he cannot prove their necessity?’ My syntactical error there! But your response has still answered my question.

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: