John Searle is known primarily for his extensive writings in the philosophy of language, but in recent years he has published some celebrated iconoclastic essays in the philosophy of mind. His ‘Minds, Brains, and Programs’, for example, challenged the fundamental assumption in artificial intelligence that cognition can be re-created by the manipulation of physical symbols according to a formal program. That essay provoked a wide variety of responses from champions of AI, responses notable mostly for their unexpected and embarrassing lack of unanimity concerning just where and how Searle’s critique was mistaken.
The Rediscovery of the Mind contains the first book-length exposition of his own positive account of the mind and its place in nature. It advances systematic criticisms not just of classical AI, but of all existing forms of philosophical materialism, and of several research programmes in linguistics and psychology as well. The book will set many people’s teeth on edge. Searle’s text is abrasive, peremptory and occasionally unfair. It is also lucidly written, passionately argued, and sure to provoke controversy. It pulls together Searle’s earlier work in the philosophy of language and integrates it with his mature position in the philosophy of mind. Unexpectedly, this recalls the earlier position of René Descartes.
Like Descartes, Searle is an eloquent and evidently sincere spokesman for what in the 18th century was loosely called ‘The Mechanical Philosophy’ and is now loosely called ‘materialism’. Yet like Descartes, he balks ‘at the very door of the mind’ and declares conscious, intentional phenomena to be wholly real, but distinct from and irreducible to the non-mental physical features of the brain. Like Descartes, Searle has here a profound tension on his hands, and he never entirely succeeds in resolving it. To be sure, his rejection of all forms of reductive materialism concerning the mind is much more circumspect than was Descartes’. Searle wants no part of any dualism of substances. Rather, he makes the bold assertion that mental phenomena are entirely natural and caused by the neurophysiological activities of the brain. He calls this ‘Biological Naturalism’: mental states are natural states of biological organisms.
What distinguishes Searle from other contemporary materialists (identity theorists, functionalists and eliminative materialists), and what unites him with Descartes, is his firm insistence that mental phenomena form an ontologically distinct class of natural phenomena, which are caused by and interact with, but cannot be reduced to, any of the familiar classes of physical phenomena – dynamical, electrical, chemical, biological etc. Here again we may feel the tension that goes unfelt by Searle: how can mental phenomena fail to be reducible to physical phenomena within the brain, if, as Searle asserts, they ultimately arise from nothing other than the complex interactions of those physical phenomena with one another and with the environment?
Searle’s position, however, is no oversight. The supposed ontological gulf between mental and physical phenomena forms the fulcrum of his argument. It provides both the motive and the systematic basis for his deepest criticisms of various recent orthodoxies in artificial intelligence, cognitive psychology, semantic theory, transformational grammar and the philosophy of mind.
A theme familiar from his earlier writings concerns what philosophers call ‘intentionality’ – a technical term for what a non-philosopher would call the ‘meaning’ or ‘content’ of a thought or sentence. Searle insists on making a distinction between the genuine or ‘intrinsic’ intentionality of real mental states and the merely ‘as if’ or ‘derivative’ intentionality of the physical states of various non-mental systems such as thermostats, heliotropic sunflowers, and digital computers.
In the present book, this distinction is claimed by Searle to be tightly connected with a second distinction – between an individual’s conscious states (actual or potential) and the same individual’s deeply or essentially non-conscious states. Intrinsic or genuine intentionality, says Searle, is a property exclusively of states that are a part of someone’s current consciousness (conscious states), or of states that could be brought to consciousness, by memory, prompting, attention, and so forth (‘shallowly unconscious’ states). States that do not meet the disjunctive demands of this Connection Principle are denied anything beyond an ‘as if’ intentionality. Intrinsic meaning, on the one hand, and consciousness, on the other, are thus claimed to be essentially connected with one another.
Since neither thermostats, sunflowers or computers have any conscious states, reasons Searle, their internal states must lack intrinsic intentionality. Accordingly, no form of explanation that presupposes intrinsic intentionality can ever provide more than a metaphorical explanation of their behaviour. Literal explanations must always be drawn instead from physics, biology and the other non-mental sciences. The same applies to the explanation of all aspects of human cognition that fall strictly outside the domain of the actually or potentially conscious.
Searle’s demand that non-conscious phenomena receive their literal explanation in appropriate non-mental terms is mirrored in the complementary demand that genuinely conscious intentional phenomena receive their literal explanation in appropriately mental terms. Just as it is wrong to try to explain the sunflower’s heliotropic behaviour in terms of desires, perceptions and actions, so it is wrong to try to explain a genuinely conscious creature’s perceptions, deliberations and behaviour in terms appropriate solely to the physical substrate that causes conscious activity.
This conviction forms the basis of Searle’s criticism of current forms of materialism, all of which, he believes, have lost sight of the central importance of the phenomenon of consciousness. Or worse, they deliberately downplay its importance by refocusing our attention on the more obviously formal or structural features of cognition such as grammar, logic, problem-solving and learning. It is high time, Searle insists, that the several cognitive sciences rediscover the importance of consciousness, and refocus their explanatory efforts appropriately. Hence the title of his book, which is as much hortatory as descriptive.
It is hard not to be roused by a clarion call to readdress consciousness. Why ever should we resist it? We would all dearly love to understand consciousness better. Yet Searle’s readers are advised to proceed with caution as they follow him in pursuit of this goal. Other major points of view, materialist to the core but no less interested in consciousness, are summarily dismissed. And in the end, Searle himself fails to achieve his goal. Moreover, the interim position he takes is inadequately motivated, doctrinally unstable, and flatly contradicted by every relevant lesson of our scientific history.
Consider first its motivation. The focal issue is the claim that mental phenomena are irreducible to the objective features of the physical brain. The sticking point here, according to Searle, is the subjective character of mental states, as opposed to the objective character of any and all physical states. In the face of this ‘rock-bottom’ divergence, how could mental phenomena possibly be identical with, or somehow constituted from, purely physical phenomena? They are as different as chalk and cheese. The argument is beguiling – which is why it is famous. But Searle is not offering us a new argument: rather, an old one, recently revived by Thomas Nagel and Frank Jackson.
There is also a standard, devastating reply to it which has been in the undergraduate textbooks for a decade. On the most obvious and reasonable interpretation, to say that John’s mental states are subjective in character is just to say that John’s mental states are known-uniquely-to-John-by-introspection. And to say that John’s physical brain states are objective is just to deny that his physical brain states have the hyphenated property at issue. Stated carefully, the argument thus has the following form. 1. John’s mental states are known-uniquely-to-John-by-introspection. 2. John’s physical brain states are not known-uniquely-to-John-by-introspection. Therefore, since they have divergent properties: 3. John’s mental states cannot be identical with any of John’s physical brain states.
Once put in this form, however, the argument is instantly recognised as committing a familiar form of logical fallacy, a fallacy instanced more clearly in the following two examples. 1. Aspirin is known-to-John-as-a-pain-reliever. 2. Acetylsalicylic acid is not known-to-John-as-a-pain-reliever. Therefore, since they have divergent properties: 3. Aspirin cannot be identical with acetylsalicylic acid. Or: 1. The temperature of an object is known-to-John-by-simple-feeling. 2. The mean molecular kinetic energy of an object is not known-to-John-by-simple-feeling. Therefore, since they have divergent properties: 3. Temperature cannot be identical with mean molecular kinetic energy.
In both cases the conclusions are known to be false, despite the presumed truth of all the premises. The problem is that the so-called ‘divergent properties’ consist in nothing more than the item’s being recognised, perceived or known, by somebody, by a specific means and under a specific description. But no such ‘epistemic’ property is an intrinsic feature of the item itself, which might determine its possible identity or non-identity with some candidate thing otherwise apprehended or otherwise described. Indeed, as the two clearly fallacious parallels illustrate, the truth of the argument’s premises need reflect nothing more than John’s overwhelming ignorance of what happens to be identical with what. And as with the parallels, so with the original. Despite its initial appeal, the argument is a non sequitur.
Though he makes no attempt to protect the reader from it, Searle indicates that he is aware of this fallacy, for he briefly insists that he does not intend an ‘epistemic’ construal of subjectivity, which is precisely what the reconstruction of the Subjectivity Argument involves. But beyond this most natural and familiar construal, what other sort is there? Searle intends an ontological construal. The Subjectivity Argument is meant to make ‘a point about what real features exist in the world and not, except derivatively, about how we know about those features’. Fine. Now we need to know which features – beyond the illicit ‘epistemic features’ – are supposed forever to discriminate mental states from brain states.
Searle’s answer is as follows. ‘Suppose we tried to say that pain is really nothing but the patterns of neuron firings. Well, if we tried such an ontological reduction, the essential features of the pain would be left out. No description of the third-person, objective, physiological facts would convey the subjective, first-person character of the pain, simply because the first-person features are different from the third-person features.’ Of this reconstructed version of the Subjectivity Argument, he comments: ‘It is ludicrously simple and quite decisive.’
This last remark is an apt characterisation of any argument that establishes its conclusion by the simple expedient of assuming as its premise (‘the first-person features are different from the third-person features’) a thinly disguised restatement of the very conclusion it aims to establish (‘a pain and its subjective features are not identical with a brain state and its objective features’). Searle’s brief preamble about what certain descriptions can or cannot ‘convey’ is just one more ‘epistemic’ smokescreen. What remains beyond that is a stark example of Begging the Question. Whether or not the subjective mental features we discriminate in introspection are identical with features of our brain that might eventually be discriminated in some objective fashion is exactly what is at issue.
What will determine the answer to this question is not whether our subjective properties seem intuitively to be different from neural properties, but whether cognitive neuroscience eventually succeeds in discovering suitably systematic neural analogues for all the intrinsic and causal properties of mental states. Remember the case of light, to choose one of many historical examples. From the standpoint of uninformed common sense, light and its manifold properties certainly seem to be utterly different from anything so esoteric and alien as orthogonal electric and magnetic fields oscillating at a million billion cycles per second. Yet our strong intuitions of ontological differences notwithstanding, that is exactly what light turns out to be. Who will be so bold as to deny, just as the neuroscientific evidence is starting to pour in, that mental states may meet a similar fate?
John Searle, apparently. And he cites one final consideration in support of his anti-reductionist argument. He points out that in historical cases of the scientific reduction of some objective physical property (such as temperature, sound, colour etc), the reduction always ‘leaves aside’, as something so far unexplained and unreduced, the subjective effects of that objective physical property on conscious experience.
In fact, this is not entirely true, but it is close enough to the truth to merit being dealt with on that assumption. What Searle sees as a symptom of ontological distinction is a reflection of something much simpler: once again, our ignorance. Why is it that Statistical Mechanics does not also account for the subjective effects of temperature on human consciousness? Plainly, because such an account would require in addition an adequate theory of the human brain and its cognitive activities, something we have only recently begun to construct. Similarly, why is it that Wave Mechanics does not also account for the subjective effects of sounds on human consciousness? Plainly, for the same reason. And so on.
All these theories ‘leave aside’, as unexplained and unreduced, a vast variety of other esoteric properties beyond those found in human consciousness. Statistical Mechanics, for example, also leaves aside the effects of heat and temperature on the GNP of Peru, or on bluebird-egg cholesterol levels. Each of these phenomena requires some additional theory beyond SM if it is to be successfully addressed. No one is tempted to insist, on these grounds, that such phenomena must be counted as ontologically distinct, irreducible, non-physical features of reality. The effects of heat and temperature on conscious perception are in exactly the same position.
In fact, Searle’s anti-reductionism has come apart in our hands. The Subjectivity Argument exploits a familiar fallacy or it falls back on a simple begging of the question. And the ‘leaves aside’ argument is a faulty induction from a misapprehended historical pattern. In my judgment, Searle’s position is badly undermotivated. But the situation is darker than this. That position is also unstable. Searle is attempting to embrace both the biologically natural character of mental states and their physical irreducibility. But one or other of these has to go. As he fits these strange bedfellows together, the relation between them is said to be causal: neural phenomena do not constitute mental phenomena, according to Searle, but they do cause them.
The difficulty, however, is that every last one of the many available scientific examples of what he calls ‘micro-to-macro forms of causation’ are also cases where the macro-property at issue is constituted by some feature of the underlying micro-reality. For example, the swift compression of the molecules of a gas into a smaller volume will indeed cause the temperature of the gas to increase; but temperature is constituted by the mean kinetic energy of the molecules. The subtraction of kinetic energy from the molecules of a tray of water will indeed cause the water to become solid (form ice); but the solidity of the ice is constituted by the matrix of positionally stable bonds into which the molecules settle.
Searle’s robust persistence in thinking of mental states as ontologically distinct from, yet causally produced by, brain states reminds me of a comparable persistence in a comparable domain. It appears in the Introduction to Betty Crocker’s Microwave Cooking, a book published soon after microwave ovens began to appear in every American kitchen. (‘Betty Crocker’ is a brand name for various baking products.) Before turning to the recipes, the authors attempt a brief explanation of how such new-fangled devices manage to produce heat in the foodstuffs we put inside them: ‘The magnetron tube converts regular electricity into microwaves ... When [the microwaves] encounter any matter containing moisture – specifically food – they are absorbed into it ... The microwaves agitate and vibrate the moisture molecules at such a great rate that friction is created; the friction, in turn, creates heat and the heat causes the food to cook.’
The decisive failure of comprehension begins to appear halfway through the last sentence. Instead of asserting that the induced motion of the moisture molecules already constitutes heat, and gracefully ending their explanation there, the authors benightedly continue to discuss heat as if it were an ontologically distinct property. They then fall back on their folk understanding of one of the many things that can cause heat: friction. The result is massively misleading to the innocent reader, who is left with the impression that rubbing two molecules together causes heat in the same way that rubbing your two hands together causes heat. In this confusion, the real nature of heat – the motion of the molecules themselves – is left entirely out of the account.
I have always treasured this example, since it illustrates the way in which our folk conceptions can blithely persist, even in the face of established scientific reductions. How much firmer their grip, then, when the relevant reduction is still no more than in prospect. What Searle has written is something not too far from Betty Crocker’s Philosophy of Mind. As a recipe for addressing the true nature of conscious phenomena, it is a bust. What Searle’s book rediscovers is not the mind, but our common-sense, pre-scientific folk-psychology conception of the mind. The aim of science, by contrast, is to discover a new and better conception. In this endeavour, Searle’s book is not likely to help.
Send Letters To:
London Review of Books,
28 Little Russell Street
London, WC1A 2HN
Please include name, address, and a telephone number.