RT list: Is it right to think about RT as a materialising theoretical attempt?

 
Here is MALFET's answer to Dan (it seems that for the moment, my position is to be that of intermediary. I'll try to convince Malfet to post directly here. And, naturally, thank you, Dan, for responding so quickly!).

_________________________________________________

MALFET WROTE: [...] I will start by responding in a more line-by-line fashion than I generally like, so apologies for that, but I think it is necessary to ensure that my responses are maximally and obviously relevant.

Dan Sperber wrote: It seems to me that Malfet's argument (as quoted) applies to cognitive psychology generally and not just to RT.
Indeed.

Dan Sperber wrote: A biological cognitive system is a physical system (brain and so on) that performs computations.
To be sure, that is what a biological cognitive system would be. Whether or not the brain is in fact such a thing remains an open question, and at this time neither Dan Sperber nor I can answer it with any certainty (though I suspect the two of us hold very different hunches on the matter).

Dan Sperber wrote: How a physical system can perform computations is something we understand thanks to Turing and others in a way that was not possible before.
Most certainly.

Dan Sperber wrote: In describing the computations performed by the physical cognitive system, we are radically helped by formal approaches to computation.
Absolutely not. We are not helped by formal approaches to computation. Computation is precisely the expression of a formal ontology in a material domain. Formalism is not an approach to computation; it is entirely the "stuff" of it. (Perhaps D.S. would agree with this and he was just speaking casually in a casual context, but I do want to emphasize that fact for everyone here.)

Dan Sperber wrote: The cognitive system however is not just computational; many of its properties, regarding the flow of energy in the system for instance, are not computational and affect what computations end up being performed. So not just RT, but cognitive psychology in general wants indeed to reap the benefits of our understanding of both formal systems and physical systems. This however is a goal towards which we are moving rather than something we believe we have achieved because ?paying the price? ? which we are all intent on doing ? calls for a lot of work and for more competencies that have been achieved so far.
I have no objection to this in principle, and certainly it is a safe answer, but it casts the whole project in a very strange light. Given the massive architectural uncertainties that Professor Sperber identifies, Relevance Theory seems absurdly and unwarrantedly specific in its formulations. Architecture matters. To postulate categories like "cognitive effects" and "mutual manifestness" before establishing the material constraints and potentials of the system is to free oneself from obligations to the material world. Why should I think there is such a thing as "cognitive effects" except for the fact that such a notion fits intuitively with my understanding of the world? This is not the process of science so much as of folklore.

Dan Sperber wrote: As to the idea that describing cognition or pragmatics as a physical system means forsaking materialism, what weird definition of a physical system and of materialism do you need to arrive at such a conclusion? And what then is its relevance?
I do not know what is particularly weird about my definition. I use the term more or less exactly as it is described here, though in the finer details I am more influenced by Quine than by Fodor: http://en.wikipedia.org/wiki/Materialism [ http://en.wikipedia.org/wiki/Materialism ]

Perhaps Professor Sperber is misconstruing my claim. I am not suggesting that a description of pragmatics as a physical system would not be materialist. Quite the contrary. Rather, I take very much to heart the suggestion that: "La découverte de Turing à elle seule ne donne substance qu[rsquo]à une matérialisation plus modeste: elle permet de comprendre comment les phénomènes mentaux peuvent être matériellement réalisés; elle ne suffire pas à établir comment ils le sont effectivement." ...but with my previously stated reservation that regarding the substance of computation as "plus modeste" seems (on its surface at least) to suggest a dangerous misconstrual of Turing's discoveries.

To the extent that RT asserts pragmatic function to be computational, its materialism is located in formalization. To the extent that RT asserts pragmatic fucntion to be (non-computationally) physical, its materialism is located in "termes neurologiques (ou en des termes réductibles à des termes neurologiques)". Abstractions like "cognitive effects" are materialist only to the extent that they are isomorphic with one of these two organizations. And this is the critical point: Turing's discovery frees us from obligations to a base underlying physical anatomy, but only if the system we are describing is a properly formal one.

What forsakes materialism, therefore, is not simply "describing cognition or pragmatics as a physical system", but rather living in the analytic netherworld between formalism and physicalism. Without these commitments, in what sense can we say that the categories posited by RT meaningfully exist? Why should we have a materialism (any materialism!) if we are unconstrained by the demands of it?

Likewise, what I find "weird" is the claim that a series of analytic categories (such as "cognitive effects"), uncommitted to either physicalist or computationalist ontologies, could be safely considered materialist in any sense other than aspiration.

I don't doubt the incredible descriptive value of Relevance Theory, nor its remarkable achievement in consolidating Grice's maxims into a single governing principle. Furthermore, I find RT's assertion that all communication is framed in meta-communication to be tremendously powerful, something the rest of linguistics would do well to understand. However, all this talk of Turing machines, formalization, and materialism feels as though it is adopted for rhetorical effect more than philosophical motivation. It is either an irrelevant distraction in an otherwise useful informal description, or it is a series of ontological commitments that RT wears as clothing without meaningfully participating in.

My takeaway from Professor Sperber here is that his ontological commitments between physicalism and formalism have yet to be made, but nevertheless somehow his categories of description are already materialist? That, ultimately, is my confusion.
___________________________________________________

José Luis Guijarro
Facultad de Filosofía y Letras
Universidad de Cádiz
11002 Cádiz, España (Spain)
tlf: (34) 956-011.613
fax: (34) 956-015.505
Received on Mon Sep 24 16:23:47 2012