Friday, September 17, 2010

Some Comments on Neurophilosophy

I was just re-reading some sections of Neurophilosohy: Toward a Unified Science of the Mind-Brain by Patricia Smith Churchland, in which she advocates eliminative materialism. This is the view that our common ways of thinking about or conceptualizing our cognitive processes -- including "beliefs", "reasoning," and even "thinking" and "conceptualizing" -- are radically mistaken and so must be entirely replaced with neurophysiological processes and terms. The old wine, which they call "folk psychology" to put it on a par with folk religion, cannot fit in the new wineskins of neurological science, and so should be discarded. I encountered some interesting statements at the end of chapter 9 of Neurophilosophy, and so reproduce it here in red font with my comments interspersed. It's kind of like a fisking from days of yore, except I don't intend any disrespect for my subject.

The prospect of transforming folk psychology as we know and love it has prompted objections, some of which I have already covered, but others of which I must consider separately here. One popular objection is that eliminative materialism is self-refuting. In order to state his position, the argument goes, the eliminativist must believe what he says, but what he says is "There really are no such states as beliefs." However, if there are no beliefs, then the eliminativist cannot believe what he says. Or if he believes what he says, then there really are beliefs. The eliminativist can expect to be taken seriously only if his claim cannot, and he thereby refutes himself.

This argument has had multiple expression through the years. C. S. Lewis's Argument from Reason; J. R. Lucas's Gödelian Argument against strong AI; Alvin Plantinga's Evolutionary Argument against Naturalism; etc.

What the eliminativist is fumbling to say is that folk psychology is seriously inadequate as a theory.

As we'll see shortly, claiming that folk psychology is a theory is the linchpin to the eliminativist project; you remove that and the whole structure collapses. For now I'll just point out that it's highly contentious to suggest that folk psychology constitutes a theory. Theories explain data. What the eliminativist calls folk psychology is the data itself: beliefs, concepts, chains of reasoning, etc. We directly experience these things, they are simply given. Any theory, therefore, must explain them. A theory which denied their existence, as eliminativism does, would be completely inadequate. Worse than that, it couldn't even get off the ground, since the data it would purport to explain would be much less secure as data than what the eliminativist denies. In fact, the data it seeks to explain could very well be derived from what the eliminativist denies.

Now within the confines of that very theoretical framework we are bound to describe the eliminativist as believing there are no beliefs; however, this is not because folk psychology is bound to be true, but only because we are confined within the framework the eliminativist wishes to criticize and no alternative framework is available. If the eliminativist is correct in his criticisms,

But how could he be correct in his criticisms if they are expressed in a system that is entirely corrupt? If folk psychology falls, then the specific elements of folk psychology that were used in the formulation of eliminative materialism will fall with it. Isn't this obvious?

and if the old framework is revised and replaced, then by using the new vocabulary the eliminativist's criticisms could be restated with greater sophistication and with no danger of pragmatic contradiction. (For example, the new eliminativist might declare, "I gronkify beliefs," where gronkification is a neuropsychological state defined within the mature new theory.)

This is pretty speculative on her part, but leaving that aside, translation from one system to another can only be done if both are coherent. If one of them is not, then translation would be impossible, there being either nothing to translate from or nothing to translate to. Since the eliminativist's claim is that the old system is entirely corrupt, to talk about translating it into another superior system is simply incoherent.

Churchland suggests we can have a placeholder ("gronkify") to stand for what we usually term "belief." This, allegedly, avoids the problem of saying she believes there are no beliefs without having to go through all the rigamarole of showing exactly what is happening when we say "belief." But "placeholder" is a concept within the system she's denying. I can only make sense of what she's saying if I presuppose the basic validity of folk psychology. This is true regardless of what you think the fundamental unit is: the word, the phrase, the sentence, the concept, whatever. I only know that any unit is valid if the underlying system is valid; and the eliminativist uses these units -- thus presupposes their validity -- in order to deny their validity. This is self-defeating.

The eliminativist may argue that while they're denying that there are beliefs, thoughts, chains of reasoning, etc., they're not denying that something is happening; only that the way we conceive or perceive what's happening is completely invalid. This reminds me of Cratylus who eventually became so convinced that language could not communicate information that he renounced it entirely and stopped speaking. However, when people got in his face and screamed at him, he would wiggle his finger, as if to say, "Something's happening." Yet the eliminative materialist is even less rational than this, since they want to communicate their claim to others, after denying the possibility of communication. It's as if Cratylus wrote books to communicate why he rejected the possibility of communicating.

In response, an eliminativist might provide a parallel, such as how what used to be thought was demonic possession is now recognized as mental illness. The description "demonic possession" was wholly incorrect, but it did pick out an actual feature of the world. Ignoring everything else about this (if it picks out an actual feature, how can it be wholly incorrect?), the problem is that it treats "belief" as if it were an interpretation of what is actually going on. This assumes, again, that folk psychology is a theory. But belief is part of the data; it is not an interpretation of something else, it's the ground level.

At any rate, the eliminativist's claim is not one of translation but of elimination; hence the name. If someone were an eliminativist about language and argued that all languages are completely invalid and must be rejected in favor of a potential future language that we might produce someday, how could he tell this to anyone? Any attempt to describe it could only be done in the very languages he said were wholly corrupt and invalid. Again: if the languages are completely invalid, then those parts of the languages used to express his claims are invalid. In other words, by making this claim about language, the person makes it impossible to tell someone who uses those languages about his theories.

Even so, I think we can just barely comprehend the claim that our language is untrustworthy and a future language will replace it because a) we know there are other languages, and so have a context in which we can put this claim; and b) while any expression of this language-eliminativism would be self-defeating, since it would be in language, it's not immediately evident that we can't think about it. Language and thought have a close and complex relationship, but there are plenty of philosophers who argue that language does not completely circumscribe our thought; we can think beyond the box of whatever language we speak (supposedly). But with the eliminative materialist this option is not available: he can't even think about, contemplate, or get any idea of what eliminative materialism means because he can only do so within the old framework that he pretends to reject.

It would be foolish to suppose folk psychology must be true because at this stage of science to criticize it implies using it. All this shows is that folk psychology is the only theory available now.

If the only available language were completely corrupt, and if people could only think in language, then they could never formulate a new and better language, because they would have no way of doing so. Any attempt could only get underway by presupposing the validity of the language in which they think and speak.

By way of analogy, consider a biologist in the early nineteenth century who wishes to criticize vitalist theory as misconceived. He suggests that there is no such thing as vital spirit and that other accounts must be found to distinguish living things from nonliving things. Consider the following fanciful defense of vitalism:

The anti-vitalist says that there is no such thing as vital spirit. This claim is self-refuting; the speaker can be expected to be taken seriously only if his claim cannot. For if the claim is true, then the speaker does not have vital spirit and must be dead. His claim is meaningful only if it is false. (Patricia S. Churchland 1981d)

The vitalist makes exactly the same mistake here as is made in the foray against the eliminativist.

The reference is to Churchland's short article "Is Determinism Self-Refuting?" in Mind 90 (new series, 1981), a response to Karl Popper's version of the argument that determinism is self-defeating (Popper responded in a short article of his own with the same title in Mind 92, 1983). Churchland's counter-argument that such arguments beg the question has been repeated many times in the literature. Basically, the claim is that the vitalist implicitly presupposes that vitalism is a necessary condition of asserting anything, and then applies this presupposition of vitalism to the anti-vitalist position, discovering unsurprisingly that it renders it self-defeating. Thus he begs the question: he presupposes the truth of his position in order to argue against a contrary position. Essentially, the vitalist ignores the very claim that the anti-vitalist is making, and applies a standard that the anti-vitalist would obviously reject. All the anti-vitalist has to do is deny the implicit presupposition that vitalism is a necessary condition of assertion.

There are several things one could say in response to this, but I'll just focus on three: first, the anti-eliminativist argument and the anti-vitalist argument do not seem parallel. The vitalist is not claiming to directly perceive vital spirits, he perceives that he is alive and posits a vital spirit as an explanation of this datum. The anti-vitalist does not deny the datum, only the interpretation. The eliminativist, however, is claiming that beliefs, concepts, ideas, thoughts, chains of reasoning, etc., are all bogus. Since we directly perceive beliefs, ideas, thoughts, et al., he is therefore not challenging an interpretation but denying the given data. The parallel would be (as Lynne Rudder Baker argued in Saving Belief: A Critique of Physicalism) an anti-vitalist who denied that he is alive. So, contrary to Churchland's claim here, these arguments do not commit "exactly the same mistake."

Now I think the eliminativist would respond by pointing to Quine and arguing that the distinction between data and theory has been shown to be artificial. I would respond that while the border country may be difficult to define clearly, to suggest that there simply is no raw data -- ever -- is absurd. We do directly experience some things, and if your worldview cannot account for that, your worldview is deficient. The point of worldviews, after all, is to account for varying facts. If the only way you can make your worldview work is to deny the most basic and universal facts, it doesn't merely fail: it's unworthy of being taken seriously by anyone.

Eliminativists see folk psychology as being on a par with geocentrism, and their radical-new-conception-that-consists-almost-entirely-of-placeholders as on a par with heliocentrism. But their position is much more extreme than this. A sidereal-eliminativist would deny that the stars exist; more basically, would deny that there are lights in the sky. If people pointed up to the night sky and said, "But I see lights right there," our sidereal-eliminativist would respond, "No, you don't. Once we complete a full stellar theory, we will see that these lights do not exist. It seems strange, but trust me: I'm a philosopher."

Second, question-begging is only a fallacy within the framework of the folk psychology that the eliminativist is claiming to reject. If the whole framework is bogus, why would this particular element be valid? In order for this objection to take hold, the eliminativist has to grant to his opponent the validity of folk psychology; but this obviates his entire claim, that folk psychology is not valid.

Third, it seems to me that Churchland's counter-argument would apply to any claim that something is self-defeating. For example, suppose I scream at the top of my lungs, "I ALWAYS SPEAK SOFTLY!" Someone tells me that my claim is self-defeating, since I did not say it softly. I then respond, "That begs the question. My whole claim, after all, is that I always speak softly. You're ignoring what I'm actually saying and applying a standard to my statement that I reject, namely, the standard that I don't always speak softly. I simply reject that claim, as my statement should make clear." Isn't it obvious that such a response doesn't work? Isn't it obvious that my claim is self-defeating? The person who tells me my claim is self-defeating is not assuming that my asserted claim is false, he's taking it as an actual statement, applying what it says to itself, and then pointing out that it undercuts its own presentation. That's how any claim that something is self-defeating works.

What this shows is that the anti-eliminativist's argument that eliminativism is self-defeating is logically prior to the eliminativist's counter-argument that the anti-eliminativist is begging the question. In order to say that the eliminativist's claim is self-defeating, he has to take what the eliminativist says, apply it to the eliminativist's claim itself, and then show that, if we assume the eliminativist's claims are true or valid or whatever, then the eliminativist's claims are not true or valid or whatever. The counter-response (that he is assuming the eliminativist's claim is false by applying standards to it that it denies) ignores the fact the eliminativist is presupposing the standards he claims to reject. That's the point: his position could only be valid if he hasn't really rejected the framework he says he has rejected. It's built upon the foundation of what he's dismissed as bogus.

He misidentifies the unique availability of a theory with the truth of the theory. I suspect that any grand-scale criticism of a deeply entrenched, broadly encompassing theory will seem to have a self-refuting flavor so long as no replacement theory is available.

Even if there were an available replacement theory, the grand-scale criticism could never be stated within the theory being criticized, since it would necessarily presuppose the categories of the theory being rejected as fallacious.

The reason is this: the available theory specifies not only what counts as an explanation but also the explananda themselves. That is, the phenomena that need explaining are specified in the vocabulary of the available theory (for example, the turning of the crystal spheres, the possession by demons, the transfer of caloric, the nature of consciousness).

Again, we are told that the explananda are not given but are "in the vocabulary of the available theory." That is, they are interpretations, part of an overarching theory which can therefore be dismissed. But if they're actually raw data, the ground level, that needs to be explained, eliminative materialism falls to the ground, since it doesn't explain them; it explains them away.

To tender sweeping criticisms of the entire old theory while still within its framework will therefore typically sound odd. But odd or not, such criticism nonetheless serves an essential role in steering a theory into readiness for revision.

Look, odd claims can be true. Reality is odd. But my conception of what it means to be "odd" is a part of the folk psychology framework that Churchland says she rejects. Outside of that, I can't make any sense of what she's saying; and that's the crux of the issue. If I'm in the throes of a false universal system, I can never get out of it to a better system, because any reason or ground for getting out of it would form a part of the system that I want to reject. Any chain of argument away from it would only be valid within that system; and once that system is rejected, the argument no longer holds, and I no longer have a reason for leaving the system. Moreover, this applies to everyone, including Patricia Churchland. How could evidence, reason, and argument have persuaded her to abandon folk psychology if evidence, reason, and argument are only valid within the confines of folk psychology? She's borrowing folk psychology's tools and mortar in order to construct an edifice that denies the existence of its own foundation and materiel.

Finally, it may be objected that the sentential paradigm will survive, whatever the theoretical revolutions and however thorough our understanding of the brain, because it is useful and natural and forms the nexus of our moral conceptions concerning responsibility, praise, and blame. By way of reply, it should first be mentioned that the issue now concerns a prediction about what will in fact be the social outcome of a theoretical revolution, and my inclination is to back off from making predictions about such matters.

I read this, and then my wife called from the other room, "What's so funny?" because, yes, I was laughing out loud. Churchland doesn't make predictions about such matters? Seriously? I don't mean any disrespect to her, but her whole position is a radical, gratuitous prediction of what science might discover someday, given a very particular (and implausible) set of preconditions. Eliminative materialism assumes a naïve scientific realism and extends into the future (i.e. predicts) what it might lead to. The fact that it would lead us to reject any kind of scientific realism is a small price to pay for ... what exactly? Truth? Logical consistency? Reason? These are all a part of the system she says she rejects.

Nevertheless, it may be useful to consider that objections cut from the same cloth were made on behalf of the geocentric theory of the universe and the creationist theory of man's origins. These theories were defended on grounds that they were useful and natural and were crucial elements in Christian doctrine. If the geocentric theory was wrong, if the creationist story was wrong, then crucial sections of the Bible could not be literally true and man's conception of himself and his place in the universe would be changed.

There's a partial truth here, but as usual, it's being distorted. The geocentric theory of the universe was a crucial element in Aristotelian and Ptolemaic astronomy -- that is, the science of the day -- and Christians looked for passages in the Bible to accommodate it, such as "The world is firmly established; it cannot be moved" (1 Chronicles 16:30; Psalm 93:1; 96:10; cf. Psalm 104:5). But, I'm sorry, these aren't "crucial sections of the Bible"; no doctrine is dependent on them, no theology is called into question by recognizing that their perspective is the surface of the earth rather than somewhere out in space looking down at the earth. As for the geocentric model in general, the center of the universe was considered the least prestigious, the least honorable, location therein; this is why hell was thought to be at the center of the earth, and Satan at the center of hell. Anyone who has studied ancient and medieval cosmology knows this. I wrote about this recently here, and you can read an excellent essay on it by Dennis Danielson, a literary historian, here.

The doctrine of creation, however, is crucial, and modern scientific discoveries have ruled out certain interpretations of Genesis 1 and 2. However, Churchland has erred if she thinks those interpretations exhaust traditional Christian thought on the issue. Historically, Christians have often understood the creation narratives metaphorically. Origen, one of the most important Church Fathers from the second and third centuries, went so far as to ridicule people who thought they referred to actual events (De Principiis 4:1:16). The point being that Christian doctrine has always left plenty of room for such interpretations, and modern scientific discoveries don't touch them. Regarding evolution in particular, I would just point out that many Church Fathers and medieval theologians (such as Augustine, Gregory of Nyssa, Bonaventure, etc.) held the doctrine of rationes seminales according to which God created everything in "seed" form which then developed accordingly. Since such views were widely accepted within Christianity a millennium and a half before Darwin, I just don't see much difficulty here.

At any rate, I think Churchland's reference to these issues reveals something more about her presuppositions and motivations, and those of eliminative materialists in general. She's not just assuming a naïve scientific realism, she's presupposing the conflict thesis, the view that science and religion (or at least science and Christianity) are at war, and science is winning. She's extending into the future (predicting) what the end result of this war will be. The biggest problem with this is that the conflict thesis is almost completely rejected by historians of science. It is a wildly inaccurate view of how science and Christianity have interacted in history. So not only is her position self-defeating, not only is it implausible in the extreme, it's actually based on a discarded theory about the historical interaction between science and Christianity.

The new physics and the new biology each did, in some degree, undermine the power of the Christian church and naturalize man's understanding of himself and the universe.

Well, again, I think there's a partial truth here. The rejection of geocentrism did certainly undermine the power of the Catholic church to some extent. I would reiterate, though, that it failed to undermine any serious doctrinal issue. Traditionally, Christians had accommodated Ptolemy's geocentric model into their hamartiology or doctrine of sin: what was sinful was heavy and fell towards the center; what was light was holy and moved up into the heavens. This is also why they thought the earth didn't move and the heavens did: what was heavy/sinful didn't move and what was light/holy did. When the heliocentric system was presented it was seen as upgrading the earth to the status of a heavenly object and as downgrading the sun (in many ways the source of life, so it could have been perceived as an affront) to the center. Now if this link between geocentrism and the Christian doctrine of sin was thought to be an inextricable link, even mistakenly so, rather than borne out of an attempt to accommodate the science of the day, heliocentrism could have been perceived as contradicting something important to man's self-conception. But as far as I can tell this was not the case. I could certainly be wrong, but what I've read about this focused on the idea that heliocentrism challenged the Catholic church's authority by questioning something it had made a pronouncement on; not to mention the fact that the staunch defenses of geocentrism were made by the scientists on scientific grounds rather than theologians on theological grounds.

But perhaps that was not a bad thing. At a minimum it is worth considering whether transformations in our moral conceptions to adhere more closely to the discovered facts of brain function might be no bad thing as well. Whether this is so will be a complex matter about which I feel ignorant, but it is certainly not a closed matter. It is at least conceivable that our moral and legal institutions will be seen by future generations to be as backward, superstitious, and primitive as we now see the Christian church's doctrine of past centuries concerning the moral significance of disease and the moral property of anesthesia, immunization, and contraception (White 1896).

Yes, she's actually referencing Andrew Dickson White's A History of the Warfare of Science with Theology in Christendom, a work so thoroughly rejected by historians of science that anyone should be embarassed to base any argument on it at all. It has repeatedly been shown to be nothing more than a work of propaganda. "The moral property of anesthesia" refers to the claim that using anesthesia to relieve the pains of childbirth nullified the curse of Eve which was God's punishment on humanity because of the Fall, and therefore set man up against God. Except no such objection was raised; Andrew Dickson White just made it up. Similarly, the moral significance of disease and immunization refers to the claim that Pope Leo XII condemned vaccinations because disease was a righteous punishment from God. Of course, Leo never said or wrote any such thing, nor has any other pope, it's a complete fabrication.

The only claim Churchland refers to that has any validity is contraception, since some Christians reject birth control. However, only the Catholic church does this; Protestant and Orthodox churches generally do not. And the Catholic church's stance is not some knee-jerk reaction, it is a well thought-out position, as exemplified by the 1968 encyclical Humanae Vitae. This doesn't mean we have to agree with it (I don't), but we should at least be thoughtful about it.

Once again we see that the eliminative materialist program is motivated by a demonstrably false theory of the historical relationship between science and Christianity. That Churchland would be so uncritical about Andrew Dickson White, that she would have such blind faith in a discarded theory about the history of science, while simultaneously being hyper-critical of the most basic and obvious things about ourselves and the world, shows that she and her fellow eliminativists are not applying the same standards to their own position that they apply to those they dislike. And how could they? The standards they apply to other positions are basically to completely reject them, turn away, and never look back. It doesn't really matter what the truth is, they'll try to make reality conform to their theory rather than the other way around. Except "truth," "reality," "theory," and everything else are just parts of the washed-up theory of folk psychology. They only have to accept them when they want to.

(cross-posted at Quodlibeta)

1 comment:

Peter said...

LOVED it. I always thought the Churchlands were pulling a fast one with eliminative materialism, but couldn't quite put my finger on how. Wonderful summary, linked to it from luke muelhauser's blog.