Tuesday, September 16, 2014

The Fodorgraph

A fodorgraph is an explicit representation which is what is left when you take a literal physical image, subtract the spatial array of colored marks, and then throwaway the paper.
The Philosophical Lexicon





Hello Again,

In this post, the PB team brings you this interview featuring both a thoroughly novice interviewer and the fabulous Jerry Fodor.

During this phone conversation, we chatted about Fodor's recent heretical forays into the debate about the notion of natural selection among other such light topics as Bayesian & probabilistic models of mind, modularity, nativism, generative grammar, & neuroscience.

I'd like to highlight three aspects of the interview that stood out for me.

First, the words modularity & nativism aren't part of google's legitimate word list. At present they are appearing on my screen with blaring, red squiggly lines beneath them (perhaps Peter Norvig had some say in this state of affairs?). Simultaneously exerting a centripetal and centrifugal force on people in the field of linguistics, these terms are strongly associated with Fodor. And rightly so: he has spent a lifetime working out and defending a representational theory of mind that crucially presupposes them. For what it's worth I think that they have been given a bad rap in some of the recent literature, and are due for a second fitting, which is exactly why I thought it relevant to invoke them during this interview.

I won't go into a full blown defence of either, save to say if you haven't directly read Fodor discuss these topics you are missing out. The style of argumentation is in and of itself so illuminating about what a good exposition of a set of arguments looks like that the effort of reading him is worth it on those grounds alone.

Second, Fodor thinks that neuroscience has achieved virtually zilch with respect to furthering our understanding of the mind. This may sound like an exotic viewpoint, but it does echo a stirring in linguistics with respect to what the proper orientation is between neuroscience & cognitive science should be. This concern is, to varying degrees, noticeable in the work of people like Norbert Hornstein, Randy Gallistel, and Noam Chomsky. For an example of something in print on this topic that readily comes to mind see this or this.

Third, a comment about history – which features heavily, though not by any means primarily, in Fodor's recent argument about natural selection (as the prototypical adaptationist thinks of the notion). History is not a scientific level of explanation. Why?
If I follow Fodor's line of thinking, he claims something to the effect that it is impossible to specify a materialist-deterministic mechanism which, given a particular constellation of factors, determines why a person or thing ended up where they did and not some other place. And for better or for worse, materialistic-deterministic stories are the only ones we presently accept in the sciences.

For instance, consider the notion of natural selection and behaviourist learning theory. As mentioned in the interview, both of these strands of thought have something deeply in common. In essence, if the only thing we've got in these narratives is a history of the selective pressures on an organism then they will fail to be predictive for the same reasons as narrow behaviourist narratives fail to be anything but post hoc. That is, they present a history of, figuratively speaking, selective pressures. Fodor's argument on this topic runs quite a bit further. In any event, I won't spoil your fun in teasing apart the whole argument as it exists in print. Or, if you desire, in video-debate. The interesting idea that caught my attention was the claim in his book What Darwin Got Wrong that in the 19th century history was widely considered an adequate level of explanation. Intuitively, this contextualizes some of the explanatory inadequacies in both Darwin, Marx, and the neogrammarians. More importantly, if you buy the argument, it militates against attempted revivals of neogrammerian doctrines (see for example Evolutionary Phonology by Juliette Blevins). For a specific treatment of the problem of dealing in historical explanations in linguistics see Purnell (2009) in this book (chapter 17).

This post has become unwieldy so I think I had better stop here. But at the risk of wearing out my welcome I think I had better say one last thing. Why does all this jazz about natural selection matter? The answer is that inquiry into the evolution of language has become very popular these days. And also very polemical (for example see here; particularly see the comments). I think a lot of the work in evolution of language is trying to get a free lunch from genetics, adaptationist lit, and evolutionary biology. I also think that this is likely to lead to a lot of bruised egos and not much of a lunch.

The End.

No comments:

Post a Comment