Training, etc.

April 17, 2011

Alright – for those who haven’t seen it and like that sort of thing, there’s a really interesting (imo) debate between myself and Pete Wolfendale in the comments thread to the last post. That’s still ongoing, and throwing up various issues that I’m clearly going to have to pursue in more detail at a later date – but for now I’m going to plough on with the line of thought that the last couple posts were meant to introduce – related to Brandom and ethics.

I’ve said what feels like a thousand times before that Brandom’s core category is the reliable differential responsive disposition. In Between Saying and Doing Brandom also introduces the “Test-Operate-Test-Exit cycle” as a central derivative explanatory concept. I learn from Wikipedia that the “Test-Operate-Test-Exit” concept was introduced by George A. Miller, Eugene Galanter and Karl H. Pribram in their 1960 Plans and the Structure of Behavior. (This is a literature I need to look at.) The function of the concept seems to be basically to make classical behaviourism a little less simplistic, by introducing the concept of an iterative feedback loop that determines how the organism processes stimuli. Brandom’s heavy use of the concept emphasises, I think, the naturalistic (in my sense) lineage and intent of his work.

A T.O.T.E cycle is a necessary but not sufficient condition of sapience, in Brandom’s judgement. This seems obviously right. The (super-simple) point I want to make for now is just that Brandom has, at base, undergirding his complex systemic work, an idea of a) reliable biological responsiveness to stimuli, and b) the ability of the organism to respond to stimuli by changing the way in which it responds to stimuli. The latter concept is (again) obviously necessary but not sufficient for a fully-fledged concept of training. I’m going to jump numerous explanatory stages (on the assumption that we can all imagine how this kind of natural-scientific explanation might in principle work) to talk a little about this set of ideas in a more ordinary-language way.

One way in which the organism’s reaction to stimuli can be transformed (such that one can learn that ‘raven’ is the appropriate word, in English, to pick out a raven, for example) is through overt training. One is taught by people who possess a skill how to manifest that skill. This operates through mimicry, through empathy, and through sanctions (reward and punishment, of whatever kind, for doing things right and wrong). The overt and direct training situation is one in which the trainer has a particular behaviour or capacity in mind, and interacts with the trainee in a manner designed to generate that behaviour or capacity. (The trainer may or may not herself possess the capacity.)

There is also, of course, self-training. The organism is constructed in such a way (perhaps once it has been socialised to a certain degree) that it can respond to experience in a way that is oriented towards responding to future scenarios in a way that the organism currently judges better. I don’t like falling. I find that if I walk a certain way, I fall. I try walking in a different way, and find that I fall less. I adjust my behaviour accordingly.

These are both scenarios in which a certain outcome is held in mind by the trainer – whether the trainer is identical with the trainee or not – and advice is given (externally or internally) based on experience, to attempt to attain the desired behavioural goal.

Behaviour changes can of course be produced in other ways, though. For one thing, a training process (either self-training or training-by-another) can produce behavioural (and, therefore, potentially, given our practice-theoretic framework, perceptual and conceptual) changes that were not intended by the trainer. This can happen because the training is bad – because of a poor set of assumptions about what interactions would produce the desired results. But it will also happen even with the most well-targeted and carefully-thought-through training process. The human organism is a complex thing: the behavioural changes required to produce any given training-outcome will also produce a host of other dispositional shifts – those that are required to enable the desired dispositional shift, and also others that are part of the same complex of capacities and attitudes.

This latter form of behavioural (perceptual / conceptual / attitudinal) shift should be a really central one for any adequate theory of practice. It is closely related, though absolutely not identical, to the concept of ‘pragmatic projection’ which I discussed a couple of posts back. The concept of ‘pragmatic projection’, which I highlighted in a passage from Between Saying and Doing, tells us that practices that have been developed for one purpose can be extended and adapted for another quite different purpose, and that this is (for Brandom) a central feature of linguistic capacity – and (I’m saying, more broadly) of human social practice generally. I am now highlighting a slightly different phenomenon: the fact that the development of a disposition for a specific purpose will involve a complex set of other dispositional shifts as part of the process of socialisation that normalises a behaviour into a disposition, or even just generates the capacity for that behaviour in the first place, and that these shifts will necessarily produce other behaviours and capacities (not just narrowly ‘behavioural’, but also attitudinal, perceptual, conceptual and normative) that enable our intended disposition but that are not identical with it.

[We haven’t exhausted our list of possible ways in which dispositional shifts can be generated, btw – I’ll come back to that in a later post – but I want to pursue this line of thought for now.]

I now want to introduce another concept in this area – the idea that Brandom in Between Saying and Doing calls ‘pragmatic mediation’. I’ll aim to discuss this concept (and BSD generally) in much more detail at a later point. For now suffice to say (too briefly) that Brandom distinguishes between on the one hand drawing a direct semantic connection between meaningful phenomena (e.g. such-and-such a proposition in idiom a can be translated via a given process into such-and-such a proposition in the simpler idiom b), and on the other hand pragmatically mediated semantic connections between meaningful phenomena. In the latter scenario, a proposition in idiom a might not be translatable into a proposition in idiom b, but we might be able to know (Brandom’s argument goes) that the practices required to articulate proposition a also enable the articulation of proposition b. We can therefore draw a connection between the two propositions, in two different idioms, that is not a direct semantic connection, but rather a pragmatically mediated one.

I’ll expand on all this properly at some later date. For now, as is my custom, I want to remove the emphasis on linguistic philosophy from Brandom’s account, and talk about pragmatic mediation more generally as a way of connecting different conceptual/normative contents, that does not operate on an idealist plane (I mean, a plane related to the inner connection of ideas.)

Thus: We may develop a practice that enables the (self-)articulation of a specific conceptual or normative content – or that enables a particular kind of practical skill. Developing this practice may necessarily involve developing the capacity for a heap of other practices, and developing a set of dispositions far broader than the dispositions that compose the initial skill. These additional dispositions might then enable (or indeed make probable) the formulation of other conceptual or normative content, which is not obviously related to our initial conceptual and normative content, and which may belong (at the level of ideas) to inconsistent perceptual schemes or belief systems, etc., but which has a strong pragmatically mediated connection to our initial concepts and/or practices.

I’m emphasising three related ideas here:
1) pragmatic projection
2) [I don’t have a shorthand term for this] the development of dispositions as a necessary (or strongly likely) ‘spandrel’ or offshoot of the development of other dispositions. [call this ‘pragmatic spandrels’ for now.]
3) pragmatic mediation.

These three ideas, within the framework of a naturalistic understanding of the human organism that takes reliable differential responsive dispositions as its basic explanatory unit, and that develops out of this unit, in a manner derived from Making It Explicit’s account of linguistic practice, an account of normative and conceptual content, provide, all together, I believe, the foundation of an extremely robust account of human practice. I’ll obviously say more on all this in future posts.