Cognition, Metaphysics & Teleology

If you are a high-profile professor of computer science at Yale University, it’s quite probable that the view from your room is peculiar (if not positively distorted). This is especially so if you are also an artist, writer, critic, and provocateur. Savants about town in New Haven often have odd worldviews and David Gelernter is no exception. When Gelernter does computer science, it’s genius. When he does culture wars, it’s coarse.

In this rambling piece over at Commentary, Gelernter sets up and then knocks down several strawmen. I recommend it only because he draws our attention to the reigning scientific metaphor of our times. This metaphor, which equates brains and minds with computers and processing, is a favorite of cognitive science. Gelernter is rightly skeptical. But at the end of his critique, Gelernter goes off the rails. Rather than suggesting that we should be wary of metaphorical mind models, he reverts to metaphysics:

On consciousness and subjectivity, science still has elementary work to do. That work will be done correctly only if researchers understand what subjectivity is, and why it shares the cosmos with objective reality.

Of course the deep and difficult problem of why consciousness exists doesn’t hold for Jews and Christians. Just as God anchors morality, God’s is the viewpoint that knows you are conscious. Knows and cares: Good and evil, sanctity and sin, right and wrong presuppose consciousness. When free will is understood, at last, as an aspect of emotion and not behavior—we are free just insofar as we feel free—it will also be seen to depend on consciousness.

Yes, science still has elementary work to do on brains, minds, and consciousness. The cognitive-computer model has serious flaws. This is not, however, warrant for a reversion to ancient metaphysics.

While there is much in Gelernter’s piece that metaphysicians will find to their liking, Christians will probably stop nodding in agreement near the end, when Gelernter (who is Jewish) claims:

The sanctity of life is what we must affirm against [technology gurus, scientific atheists], and the nightmare of roboticism. Judaism has always preferred the celebration and sanctification of this life in this world to eschatological promises. My guess is that 21st-century Christian thought will move back toward its father and become increasingly Judaized, less focused on death and the afterlife and more on life here today (although my Christian friends will dislike my saying so).

Suffice it to say there is zero evidence that Christianity is moving in this general direction.

Gelernter’s diffuse jeremiad was inspired, at least in part, by the recent dustup over Thomas Nagel’s Mind and Cosmos (2012), with its provocative subtitle: Why the Materialist Neo-Darwinian Conception of Nature Is Almost Certainly False.

Those not familiar with Nagel’s book and reactions it provoked should first read this piece published in May by the Chronicle of Higher Education. Gelernter’s entry in this debate is not particularly helpful; it would not be churlish to characterize it as reactionary, perhaps even primitive.

In stark contrast, Stephen Asma just entered the debate with this brilliant piece on the teleological possibilities for evolutionary biology. For a variety of informed reasons, I’ve always been anti-teleological about these matters. Asma’s piece has forced me to reconsider. At a minimum, it raises serious questions and avenues for investigation. I’m clearly not as informed as I had supposed.


Did you like this? Share it:

6 thoughts on “Cognition, Metaphysics & Teleology

  1. Cris Post author

    I’m not aware of any. Such critiques tend to assume a few forms and not many of these are encouraging. There are of course the critiques done by anti-science types or humanists who just can’t countenance cognitive approaches to brain-mind. I find these useless.

    There are some good, though mild, critiques that come from neuroscientists: people who actually pay mind (pardon the pun) to neural structures, functions, and connectivity. Unlike many cognitivists, they think that our models should pay attention to, and be linked with, biology. I agree. Jaak Panksepp is one such person; I’ve written about his work on the blog and you can find those posts by typing his name in the search box.

    In my estimation, the best form of critique is not (on it face) critique at all: just read a good history of the field. Though it was intended to champion the new cognitive science, I found Howard Gardner’s The Mind’s New Science: A History of the Cognitive Revolution to be helpful in this way. After reading it, I came away with a much deeper understanding of cognitive science. It also made me appreciate the limitations of the field, and what the problems might be. So I would say this was the best “critique” I’ve read, even though it’s history.

    As reader Jayarava and I were just discussing last week, sometimes history or genealogy is the most potent form of critique or debunking. I should also give a shout out here to reader Dominik, who often points out the metaphors that run rampant in cognitive science. He has posted about these on his blog, though I don’t have links to those posts.

  2. Joe Miller

    What are the implications of embodied (and/or extended) cognition for modularity? Does it necessitate any substantive revisions of the model? I’m a novice to these issues, so any pointers are welcome.

  3. Cris Post author

    There are a number of modularity models, so it probably depends on which ones are being discussed. Some of the early models, such as Fodor’s, are fairly generalized and probably able to accommodate embodied and/or extended cognition. Having said that, there is a good deal of difference between embodied and extended cognition.

    The former (when not too radical in its claims) is not that controversial, whereas the latter is the subject of intense debate and doubt. Other modularity models (such as Steven Mithen’s) could probably also accommodate it and may in fact even entail something like embodied cognition.

    But as modularity became more and more popular, modules proliferated like rabbits. The mind was increasingly envisioned as a swiss-army knife with nearly a hundred tools (or “modules”), leaving little or no room for “minor” things like consciousness. Merlin Donald’s model tried to correct this, but I’d say that the more recent evolutionary psychology models (i.e., the dubious ones) probably would not accommodate embodied cognition and certainly not extended cognition.

    But some of these module-happy EP guys might be willing to acknowledge embodied/extended cognition simply by creating new modules for these faculties or operations. That’s how some of them roll. One of them posited a “shamanic module” several years ago. It was pathetic.

    In the end, however, soft modularity (along the lines of Jaak Panksepp’s work and perhaps even Antonio Damasio’s) can certainly accommodate it, and might even require embodied cognition (at a limbic or emotional level).

    Does that help?

  4. Joe Miller

    Could you explain how these models entail embodied cognition?

    The reason I ask is because I’ve become skeptical of representationalism after reading a few of Anthony Chemero’s publications. Many of the cognitivists seem to presuppose the validity of representationalist theory, and I was wondering if cognitivism (and modularity more specifically) required a representationalist foundation.

    I found a pdf of Margaret Boden’s two volume history of cognitive science, so hopefully that can provide the grounds for further exploration.

  5. Cris Post author

    Panksepp and Damasio both place great emphasis on the body, body mapping, drive states, and seeking/aversive behaviors that revolve around body equilibrium or homeostasis. In Damasio’s model, these are primary. In Panksepp’s model, these are less explicit but still tied to body/brain interaction and “emotions” that are homologous in taxa which possess brains. If you read Damasio, I think the entailment should be apparent.

    That’s just my hypothesis; I’m not expert in these matters and have never really given this particular issue much thought. I was simply speaking off the cuff in response to your question. The bottom line is that some cognitive models take the body quite seriously, and in fact depend on body-mapping functions that seem quite congenial to embodied cognition.

Leave a Reply