Thursday 1/30

The History of Cognition III: Contemporary Philosophy






We opened our discussion by reviewing Descartes and his solution to the mind-body problem before turning to Hume's approach.

Humean philosophy of mind stands in stark contrast to Cartesian philosophy of mind reminiscent of the way in which Aristotelian philosophy of mind stands to the Platonic view. Where Descartes held that knowledge is possible (indeed, only possible, foundationally speaking) without reference to experience, Hume held that knowledge is only possible given experience. Hume's conjectures on the nature of the mind are a direct result of his basic assumptions about knowledge.

Hume proposes a kind of neutral monism in which the underlying substance of the universe is neither mental nor physical, but some fundamental, unifying substance. The mind, to use his terminology, is a bundle of perceptions. Hume's account of the mind is ultimately grounded in his empiricism--the thesis that all knowledge comes from experience. Thus for Hume perceptions can either be impressions--the result of sensory experience--or they can be ideas, or copies of impressions. Ideas are related to one other in one of three regular ways: by resemblance, contiguity, or cause and effect.

Reflect for a moment on Hume's point. Resemblance, contiguity, and cause-and-effect are relations between ideas, not things or events themselves. This is a startling assertion. In the case of cause-and-effect, Hume is claiming that we cannot establish the causes and effects of events in the world beyond the mere constant conjunction of our impressions of them. Yet if all we have is the constant conjunction of impressions, there is no reason to expect that one event we ordinarily think of as an effect will necessarily follow the event we ordinarily construe as its cause. That is, we think there are law-like regularities in nature, but since the causal relation is a relationship between ideas, there is really no reason to think that the world will be anything tomorrow like it was today. Humean empiricism, to wit, leads us to Humean Skepticism.

Whatever we may think of their answers, Descartes and Hume (and their many contemporaries) do us a tremendous service by sharpening our questions. In Philosophy, at any rate, it is often more difficult to get a good handle on the question than the answer. Hence the joke in Douglas Adams "Life, the Universe, and Everything": The answer to the question of the life, the Universe, and everything turns out, after the millennial cogitations of a god-like artificial intelligence, to be 42. The AI is not, however, up to figuring out what the question is to this answer, which requires creating an even greater artificial intelligence to find the question to the answer to the question of life, the Universe, and everything.

Philosophical jokes aside, several students have commented that this course is rather challenging. Let me say first of all that I endorse that sentiment wholeheartedly. The Mind is an enormously challenging topic. Some of the best minds (ba-dum-ching) have spent their lives trying to understand it. The readings are difficult, if not at times opaque. There is a constant and uncomfortable feeling in working through this material of not quite grasping it. For those who "just want the answer!" it can be particularly distressing.

If you happen to feel that way, rest assured that it's normal; you're not the only one. Not by a longshot.

Is it worth it?

I think so.

There is something intrinsically fascinating about the mind, and the deeper our understanding of it the better. Comments on course evaluations from students who have stuck with the course suggest the same. They report almost without exception that this is one of the most challenging courses they've had, yet it has also they've discovered been one of the most rewarding.

Please bear in mind that I'm always available to meet outside of class and help talk you through the arguments. If you can't make my office hours, TR 11:00 - 2:00, then by all means email me and we'll set up an appointment.

In the end, though, your understanding of the material is determined not by hearing me talk about it but by writing about it. Philosophy is like that. It's why the problem sets, for example, are so important. For it's only when you have to explain an issue yourself that you begin to get the sense that you've mastered the issue.

As one of my students put it when she visited long after having graduated, "Berkich, I don't remember a word you said, but I remember every word I wrote."

Or as Wittgenstein put it, "Philosophy is an activity!" It is, that is, something you do. It is not something you can do passively listening. You have to be engaged in it yourself. Philosophy is in this respect unique among all disciplines: To learn philosophy you have to do philosophy. (And, one might hasten to add, doing philosophy is always and forever learning philosophy.

So, be patient. Don't expect every reading to go smoothly. Be prepared to feel intellectually woozy, disoriented, and, at times, abysmally lost.

Set aside about three hours a week outside of class to read and reflect on the authors, my notes and handouts, and these synopses. Set aside an additional three hours every week to work on each problem set--but not all at once! Nobody can think that hard that long. Spread it out a bit. Write drafts and revise. You'll get comments back from me, and hopefully they will help you get your footing with the material. Please note that, unlike the Term Paper project, I do not as a rule review drafts of problem sets. I am happy to discuss specific questions you might have or clarify the questions I'm asking on the problem sets, but it's simply not feasible to review those drafts.

Bear in mind as you work something athletes well understand but which tends to get lost in other contexts: It is only when we're faced with genuine challenges that we strengthen our skills, broaden our understanding, explore our potential, and develop our mastery.

I'm not sure if, all that said, you're feeling encouraged or discouraged. I hope not the latter. But do let me know. We can talk about it.

Now, having discussed Hume, we turned to the contemporary philosophical literature. What I am attempting to do is give you a sense of how philosophical progress proceeds. That is, by following one thread, we have a bit of taste for the philosophical dialogue that occurs as one philosopher criticizes another philosopher and charts new intellectual terrain by articulating and defending a new solution.

Before getting on to our thread of interest, it is absolutely crucial to be clear on the Mind-Body problem. If, in answer to the question,

"What is the Mind-Body Problem?",

you say

"It is the problem of understanding how two different substances, Mind and Body, can causally interact,"

then you haven't answered the question. What you've described is a problem for Cartesian Dualism (aka "Interactionism"), which is thus a problem for one solution to the Mind-Body problem. It is not the Mind-Body problem. There are, as we saw, many solutions to the Mind-Body problem, and those solutions have additional problems. For example, Descartes (Interactionism) has the problem of explaining how two substances can nevertheless interact causally, while Leibnitz (Parallelism) has the problem of explaining how mental events and physical events can be so perfectly coordinated.

The Mind-Body problem is more fundamental than the further problems a given solution to it might have. Here is one admittedly pithy way to cast the problem:

What can be the mind and the body be in light of the fact that none of the properties we ascribe to minds can be ascribed to bodies and none of the properties we ascribe to bodies can be ascribe, except perhaps metaphorically, to minds?

Moving on, today we saw Ryle's objection to Cartesian Dualism (remember the Category Mistake!) and his proposal of Logical Behaviorism, which we carefully distinguished from Methodological (Skinnerian) Behaviorism. We then took up Putnam's rather devastating critique of Logical Behaviorism: Remember the SuperDuperSpartans! The thought experiment of the SuperDuperSpartans shows us that mental states can differ without changes in behavioral dispositions, which blocks the Logical Behaviorist from reducing the former to the latter. A possible response to the demise of the Logical Behaviorist, who tries to make progress by giving the mind the boot from our discussions altogether, embraces the mind, but identifies it with the body thusly: Every mental event of some type is a physical (think neural) event of some type. This position is called Type Physicalism. What it holds is that, for example, pains just are (type) identical to c-fibre nerve firings. To be in the mental state of experiencing pains just is to have c-fibre nerves firing, and nothing more. This certainly seems more promising than Logical Behaviorism, yet it, too, runs afoul of Putnam's skilled analysis.

To recap, I suggested that we are in the process of tracing one thread of a centuries-old conversation between philosophers. This is important, because it helps us understand how the science of mind got where it is now and why it makes the assumptions it does. Exploring positions we ultimately reject is thus crucial to grasping currently accepted positions.

Thus we saw Ryle's objection to Cartesian Dualism and his proposal of Logical Behaviorism, which we carefully distinguished from Methodological (Skinnerian) Behaviorism. We also saw Putnam's objections to Logical Behaviorism, which led us to the proposal of Type-Physicalism. Putnam's Multiple-Realizability argument against Type-Physicalism, finally, led us to Machine Functionalism, and here is where I pause to more fully explain the intuition behind Machine Functionalism.

That is, our operating assumption (by which I mean, "let us see how much progress we can make in understanding the mind in Dretske's sense under this assumption") for the semester is Machine Functionalism, so it is worthwhile spending some time understanding the thesis and arguments, like the Multiple-Realizability argument, for it.

Aristotle's view of the soul has a decidedly functionalist flavor to it. Or, better, functionalism has a decidedly Aristotelian flavor. Aristotle described and classified souls in terms of the various functions they might exhibit. Vegetative souls only have the capacity or function of nutrition, growth, and decay. Animal souls have further capacities, including perception and locomotion. Human souls are unique in also having a rational capacity. We can think of each of these capacities as functions of the soul.

A bit of background may prove useful: Aristotle described what has to be given to have a complete explanation of a thing. For example, suppose we are to explain a statue of a hero. Then we must describe its four causes:

  1. The material cause of the statue, or the bronze out of which it is made;
  2. The formal cause of the statue, or the hero's particular physical characteristics the statue represents;
  3. The efficient cause of the statue, or the sculptor who creates the statue; and
  4. The final cause of the statue, or the purpose for which it was made--perhaps in this case to commemorate the hero's great deeds.

Such explanations are called teleological explanations (from the Greek 'telos', translated as 'purpose' or 'end'.) They are also sometimes called functional explanations, and some sciences trade in them. Consider, for example, the ordinary explanation of the kidney in terms of its function in filtering blood.

Functionalism in the philosophy of mind seeks to explain mental states in terms of their functional roles. For example, pain states satisfy a certain functional role in an organism, and even organisms with remarkably different underlying neurophysiologies may, echoing the Multiple-Realizability argument, experience pain. Analytic Functionalism, a descendant of Logical Behaviourism, sees talk of mental states as reducible to--and ultimately eliminable in favor of--talk of an organism's functional states. Machine Functionalism, which we now call computationalism, sees mental states as resulting from the (mechanistic) functioning of basic physical elements like switches or neurons.

Computationalism suggests various more or less crude descriptors:

  • The Mind is to the Body as the Software is to the Computer Hardware.
  • The brain is the computer, the mind its states.
  • The brain is wetware.
  • The mind is a function of the brain.
  • Mental states are to brain states as telling time is to the gears of a watch.
  • The mind is what the brain does.

Next time we take up an altogether different thread of intellectual history to try to make more rigorous sense of computationalism.