Neuroscience and Philosophy

Brain, Mind, and Language

Maxwell Bennett, Daniel Dennett, Peter Hacker, John Searle

• Publisher: Columbia University Press
• Year: 2007
• eBook ISBN: 9780231511940
• Topic: Philosophy
• Subtopic: Philosophy History & Theory

In Neuroscience and Philosophy three prominent philosophers and a leading neuroscientist clash over the conceptual presuppositions of cognitive neuroscience. The book begins with an excerpt from Maxwell Bennett and Peter Hacker's Philosophical Foundations of Neuroscience (Blackwell, 2003), which questions the conceptual commitments of cognitive neuroscientists. Their position is then criticized by Daniel Dennett and John Searle, two philosophers who have written extensively on the subject, and Bennett and Hacker in turn respond.

Their impassioned debate encompasses a wide range of central themes: the nature of consciousness, the bearer and location of psychological attributes, the intelligibility of so-called brain maps and representations, the notion of qualia, the coherence of the notion of an intentional stance, and the relationships between mind, brain, and body. Clearly argued and thoroughly engaging, the authors present fundamentally different conceptions of philosophical method, cognitive-neuroscientific explanation, and human nature, and their exchange will appeal to anyone interested in the relation of mind to brain, of psychology to neuroscience, of causal to rational explanation, and of consciousness to self-consciousness.

In his conclusion Daniel Robinson (member of the philosophy faculty at Oxford University and Distinguished Professor Emeritus at Georgetown University) explains why this confrontation is so crucial to the understanding of neuroscientific research. The project of cognitive neuroscience, he asserts, depends on the incorporation of human nature into the framework of science itself. In Robinson's estimation, Dennett and Searle fail to support this undertaking; Bennett and Hacker suggest that the project itself might be based on a conceptual mistake. Exciting and challenging, Neuroscience and Philosophy is an exceptional introduction to the philosophical problems raised by cognitive neuroscience.

THE REBUTTALS

PHILOSOPHY AS NAIVE ANTHROPOLOGY

Comment on Bennett and Hacker

DANIEL DENNETT

Bennett and Hacker’s Philosophical Foundations of Neuroscience (Blackwell, 2003), a collaboration between a philosopher (Hacker) and a neuroscientist (Bennett), is an ambitious attempt to reformulate the research agenda of cognitive neuroscience by demonstrating that cognitive scientists and other theorists, myself among them, have been bewitching one another by misusing language in a systematically “incoherent” and conceptually “confused” way. In both style and substance, the book harks back to Oxford in the early 1960s, when Ordinary Language Philosophy ruled and Ryle and Wittgenstein were the authorities on the meanings of our everyday mentalistic or psychological terms. I myself am a product of that time and place (as is Searle, for that matter), and I find much to agree with in their goals and presuppositions and, before turning to my criticisms, which will be severe, I want to highlight what I think is exactly right in their approach—the oft-forgotten lessons of Ordinary Language Philosophy.

Neuroscientific research … abuts the psychological, and clarity regarding the achievements of brain research presupposes clarity regarding the categories of ordinary psychological description—that is, the categories of sensation and perception, cognition and recollection, cogitation and imagination, emotion and volition. To the extent that neuroscientists fail to grasp the contour lines of the relevant categories, they run the risk not only of asking the wrong questions, but also of misinterpreting their own experimental results. (p. 115)

Just so.¹ When neuroscientists help themselves to the ordinary terms that compose the lore I have dubbed “folk psychology,”² they need to proceed with the utmost caution, since these terms have presuppositions of use that can subvert their purposes and turn otherwise promising empirical theories and models into thinly disguised nonsense. A philosopher—an expert on nuances of meaning that can beguile the theorist’s imagination—is just the right sort of thinker to conduct this important exercise in conceptual hygiene.

I also agree with them (though I would not put it their way) that “the evidential grounds for the ascription of psychological attributes to others are not inductive, but rather criterial; the evidence is logically good evidence” (p. 82). This puts me on their side against, say, Fodor.³

So I agree wholeheartedly with the motivating assumption of their book. I also applaud some of their main themes of criticism, in particular their claim that there are unacknowledged Cartesian leftovers strewn everywhere in cognitive neuroscience and causing substantial mischief. They say, for instance:

Contemporary neuroscientists by and large take colours, sounds, smells and tastes to be “mental constructions created in the brain by sensory processing. They do not exist, as such, outside the brain” [quoting Kandel et al. 1995]. This again differs from Cartesianism only in replacing the mind by the brain. (p. 113)

Here they are criticizing an instance of what I have called “Cartesian materialism” (Consciousness Explained, 1991), and they are right, in my opinion, to see many cognitive neuroscientists as bedazzled by the idea of a place in the brain (which I call the Cartesian Theater) where an inner show of remarkable constructions is put on parade for a (material) res cogitans sitting in the audience.

More particularly, I think they are right to find crippling Cartesianism in Benjamin Libet’s view of intentional action and in some of the theoretical work by Stephen Kosslyn on mental imagery. I also join them in deploring the philosopher’s “technical” term, qualia, a poisoned gift to neuroscience if ever there was one, and I share some of their misgivings about the notorious “what is it like” idiom first explored by Brian Farrell (1950) and made famous by Thomas Nagel (1974). Introspection, they say, is not a form of inner vision; there is no mind’s eye. I agree. And when you have a pain, it isn’t like having a penny; the pain isn’t a thing that is in there. Indeed. Although I don’t agree with everything they say along the paths by which they arrive at all these destinations, I do agree with their conclusions. Or, more accurately, they agree with my conclusions, though they do not mention them.⁴

More surprising to me than their failure to acknowledge these fairly substantial points of agreement is that the core of their book, which is also the core of their quite remarkably insulting attack on me,⁵ is a point I myself initiated and made quite a big deal of back in 1969. Here is what they call the mereological fallacy:

We know what it is for human beings to experience things, to see things, to know, or believe things, to make decisions, to interpret equivocal data, to guess and to form hypotheses. But do we know what it is for a brain to see or hear, for a brain to have experiences, to know or believe something? Do we have any conception of what it would be for a brain to make a decision?

They answer with a ringing NO!

It makes no sense to ascribe psychological predicates (or their negations) to the brain, save metaphorically or metonymically. The resultant combination of words does not say something that is false; rather, it says nothing at all, for it lacks sense. Psychological predicates are predicates that apply essentially to the whole living animal, not to its parts. It is not the eye (let alone the brain) that sees, but we see with our eyes (and we do not see with our brains, although without a brain functioning normally in respect of the visual system, we would not see). (p. 72)

This is at least close kin to the point I made in 1969 when I distinguished the personal and subpersonal levels of explanation. I feel pain; my brain doesn’t. I see things; my eyes don’t. Speaking about pain, for instance, I noted:

An analysis of our ordinary way of speaking about pains shows that no events or processes could be discovered in the brain that would exhibit the characteristics of the putative “mental phenomena” of pain, because talk of pains is essentially non-mechanical, and the events and processes of the brain are essentially mechanical.

(Content and Consciousness, p. 91)

We have so much in common, and yet Bennett and Hacker are utterly dismissive of my work. How can this be explained? As so often in philosophy, it helps to have someone say, resolutely and clearly, what others only hint at or tacitly presuppose. Bennett and Hacker manage to express positions that I have been combating indirectly for forty years but have never before been able to confront head on, for lack of a forthright exponent. Like Jerry Fodor, on whom I have relied for years to blurt out vividly just the points I wish to deny—saving me from attacking a straw man—Bennett and Hacker give me a bold doctrine to criticize. I’ve found the task of marshaling my thoughts on these topics in reaction to their claims to be illuminating to me and, I hope, to others as well.

The Philosophical Background

In this section I am going to speak just of Hacker, leaving his coauthor Bennett out of the discussion, since the points I will be criticizing are clearly Hacker’s contribution. They echo, often in the same words, claims he made in his book, Wittgenstein: Meaning and Mind (Blackwell, 1990), and they are strictly philosophical.

When Hacker lambastes me, over and over, for failing to appreciate the mereological fallacy, this is a case of teaching your grandmother to suck eggs. I am familiar with the point, having pioneered its use. Did I, perhaps, lose my way when I left Oxford? Among the philosophers who have taken my personal level/subpersonal level distinction to heart, at least one—Jennifer Hornsby—has surmised that I might have abandoned it in my later work.⁶ Did I in fact turn my back on this good idea? No.⁷ On this occasion it would be most apt to cite my 1980 criticism of Searle’s defense of the Chinese Room intuition pump:

The systems reply suggests, entirely correctly in my opinion, that Searle has confused different levels of explanation (and attribution). I understand English, my brain doesn’t—nor, more particularly, does the proper part of it (if such can be isolated) that operates to “process” incoming sentences and to execute my speech act intentions.

(Behavioral and Brain Sciences [1980], 3:429)⁸

(This claim of mine was summarily dismissed by Searle, by the way, in his reply in BBS. I’ll be interested to see what he makes of the personal level/subpersonal level distinction in its guise as the mereological fallacy.)⁹

The authoritative text on which Hacker hangs his conviction about the mereological fallacy is a single sentence from St. Ludwig:

It comes to this: Only of a human being and what resembles (behaves like) a living human being can one say: it has sensations; it sees, is blind; hears, is deaf; is conscious or unconscious. (Philosophical Investigations, para. 281)

Right here is where Hacker and I part company. I am happy to cite this passage from Wittgenstein myself; indeed I take myself to be extending Wittgenstein’s position: I see that robots and chess-playing computers and, yes, brains and their parts do “resemble a living human being (by behaving like a human being)”—and this resemblance is sufficient to warrant an adjusted use of psychological vocabulary to characterize that behavior. Hacker does not see this, and he and Bennett call all instances of such usage “incoherent,” insisting again and again that they “do not make sense.” Now who’s right?

Let’s go back to 1969 and see how I put the matter then:

In one respect the distinction between the personal and sub-personal levels of explanation is not new at all. The philosophy of mind initiated by Ryle and Wittgenstein is in large measure an analysis of the concepts we use at the personal level, and the lesson to be learned from Ryle’s attacks on “para-mechanical hypotheses” and Wittgenstein’s often startling insistence that explanations come to an end rather earlier than we had thought is that the personal and sub-personal levels must not be confused. The lesson has occasionally been misconstrued, however, as the lesson that the personal level of explanation is the only level of explanation when the subject matter is human minds and actions. In an important but narrow sense this is true, for as we see in the case of pain, to abandon the personal level is to stop talking about pain. In another important sense it is false, and it is this that is often missed. The recognition that there are two levels of explanation gives birth to the burden of relating them, and this is a task that is not outside the philosopher’s province…. There remains the question of how each bit of the talk about pain is related to neural impulses or talk about neural impulses. This and parallel questions about other phenomena need detailed answers even after it is agreed that there are different sorts of explanation, different levels and categories.

(Content and Consciousness, pp. 95–96)

This passage outlines the task I have set myself during the last thirty-five years. And the boldfaced passages mark the main points of disagreement with Hacker, for my path is not at all the path that he has taken. He gives his reasons, and they are worth careful attention:

[A] Conceptual questions antecede matters of truth and falsehood…. Hence conceptual questions are not amenable to scientific investigation and experimentation or to scientific theorizing. (p. 2)

One can wonder about the first claim. Are not answers to these conceptual questions either true or false? No, according to Hacker:

[B] What truth and falsity is to science, sense and nonsense is to philosophy. (p. 6)

So when philosophers make mistakes they produce nonsense, never falsehoods, and when philosophers do a good job we mustn’t say they get it right or speak the truth but just that they make sense.¹⁰ I am inclined to think that Hacker’s [B] is just plain false, not nonsense, but, be that as it may, Hacker’s second claim in [A], in spite of the “hence,” is a non sequitur. Even if conceptual questions do “antecede” matters of truth and falsity, it might well behoove anybody who wanted to get clear about what the good answers are to investigate the relevant scientific inquiries assiduously. This proposal, which Hacker identifies as Quinian naturalism, he dismisses with an irrelevancy: “we do not think that empirical research can solve any philosophical problems, any more than it can solve problems in mathematics” (p. 414). Well of course not; empirical research doesn’t solve them, it informs them and sometimes adjusts or revises them, and then they sometimes dissolve, and sometimes they can then be solved by further philosophical reflection.

Hacker’s insistence that philosophy is an a priori discipline that has no continuity with empirical science is the chief source of the problems bedeviling this project, as we shall see:

[C] How can one investigate the bounds of sense? Only by examining the use of words. Nonsense is often generated when an expression is used contrary to the rules for its use. The expression in question may be an ordinary, non-technical expression, in which case the rules for its use can be elicited from its standard employment and received explanations of its meaning. Or it may be a technical term of art, in which case the rules for its use must be elicited from the theorist’s introduction of the term and the explanations he offers of its stipulated use. Both kinds of terms can be misused, and when they are, nonsense ensues—a form of words that is excluded from the language. For either nothing has been stipulated as to what the term means in the aberrant context in question, or this form of words is actually excluded by a rule specifying that there is no such thing as (e.g., that there is no such thing as “east of the North Pole”), that this is a form of words that has no use. (p. 6)

This passage is all very reminiscent of 1960 or thereabouts, and I want to remind you of some of the problems with it, which I had thought we had figured out many years ago—but then, we didn’t have this forthright version to use as our target.

How can one investigate the bounds of sense? Only by examining the use of words.

Notice, first, that, no matter what any philosopher may say, examining the use of words is an empirical investigation, which often yields everyday garden-variety truths and falsehoods and is subject to correction by standard observations and objections. Perhaps it was a dim appreciation of this looming contradiction that led Hacker, in his 1990 book, to pronounce as follows:

Grammar is autonomous, not answerable to, but presupposed by, factual propositions. In this sense, unlike means/ends rules, it is arbitrary. But it has a kinship to the non-arbitrary. It is moulded by human nature and the nature of the world around us. (p. 148)

Let grammar be autonomous, whatever that means. One still cannot study it without asking questions—and even if you only ask yourself the questions, you still have to see what you say. The conviction that this method of consulting one’s (grammatical or other) intuitions is entirely distinct from empirical inquiry has a long pedigre...

Table of contents

• Cover 
• Half title
• Title
• Copyright
• Contents 
• Introduction
• The Argument
• Neuroscience and Philosophy
• The Rebuttals
• Reply to the Rebuttals
• Epilogue
• Still Looking: Science And Philosophy In Pursuit of Prince Reason
• Notes