An Expedient Mind: ARGUMENTS

AN EXPEDIENT MIND: ARGUMENTS

Tartan Hen Publications : Home | more books | Contact : feedback@tartanhen.co.uk

SEARLE'S ARGUMENT

This is a critique of an argument presented by John Searle. Searle's argument is contained in an Internet document and it is undated, but I guess it is fairly recent, since it refers to his Chinese room argument and some of the criticisms which have been directed against it. The full text of his article is available on the Internet at

http://www.ecs.soton.ac.uk/~harnad/Papers/Py104/searle.prob.html

Searle's text is copyright, so I cannot reproduce the whole thing here. I will however, make use of extensive quotations, so that I (and the reader) can be quite sure that I am not mis-quoting or paraphrasing his argument unfairly.

My critique will consist of a kind of dialogue - a quotation from Searle, followed immediately by my own comment on that quotation. At the end I will summarise the differences between his position and my own.

QUOTATION (Opening statement, para 1):
The most important scientific discovery of the present era will come when someone -- or some group -- discovers the answer to the following question: How exactly do neurobiological processes in the brain cause consciousness?

MY COMMENT:
My view is that neurological processes do not CAUSE consciousness. These processes ARE consciousness. (ie being conscious is what a particular kind of neurological process does). By stating the "problem" in those terms Searle has excluded from consideration the most plausible theory which challenges his own. This false statement of the "problem" is repeated many times in his essay.

QUOTATION (section 1, para 1):
By `consciousness' I simply mean those subjective states of sentience or awareness that begin when one awakes in the morning from a dreamless sleep and continue throughout the day until one goes to sleep at night or falls into a coma, or dies, or otherwise becomes, as one would say, `unconscious'.

MY COMMENT
A dreamless sleep is a problematic concept. Sleep research suggests that we all dream whenever we sleep (although not all of the time while we are asleep). Dream-sleep is an interesting aspect of consciousness, because while we are asleep and dreaming, we appear to be conscious of the dream. It is only after we are awake that the memory of that conscious experience disappears. This suggests that consciousness is not a monolithic entity, but has various components, some of which may be absent from different varieties of consciousness. I also object to his description of consciousness in terms of concepts such as "sentience" or "awareness". These terms are related to consciousness. The nature of these concepts is dependent upon the explanation we have for consciousness itself. My own view is that sentience, is just a vague general term for consciousness, a nd awareness is one particular component of consciousness. Any explanation of consciousness which is couched in these terms, would, in that view, be self-defining or circular, equivalent to lifting oneself off the ground by the bootstraps.

QUOTATION (section 1 para 2)
It [i.e. consciousness] is sometimes described as that feature of consciousness by way of which there is something that it's like or something that it feels like to be in a certain conscious state. If somebody asks me what it feels like to give a lecture in front of a large audience I can answer that question. But if somebody asks what it feels like to be a shingle or a stone, there is no answer to that question because shingles and stones are not conscious.

MY COMMENT:
This characterisation of consciousness in terms of what "it is like" to be conscious, has often been advanced. But it is only yet another of those circular definition in terms of another form of words which mean exactly the same thing. It is vacuous.

QUOTATION (section 4, para 3)
Just as behavior by itself is not sufficient for consciousness, so computational models of consciousness are not sufficient by themselves for consciousness. The computational model of consciousness stands to consciousness in the same way the computational model of anything stands to the domain being modelled. Nobody supposes that the computational model of rainstorms in London will leave us all wet. But they make the mistake of supposing that the computational model of consciousness is somehow conscious. It is the same mistake in both cases.

MY COMMENT:
It is very definitely not the same mistake in both cases. In his rainstorm model he is suggesting that the model will LEAVE US ALL wet. That is clearly absurd. No one has suggested, however, that an accurate model (or I would claim) functional procedure of consciousness would LEAVE US ALL feeling conscious. Why would the successful operation of an artificial consciousness machine make US ALL (i.e. every one else) feel conscious any more than the presence of my conscious brain mechanism could make John Searle conscious. That is equally absurd. If the operational or functional theory of consciousness is correct, the successful operation of the consciousness procedure would result in only the mechanism itself being conscious. Could that idea apply to the rainstorm model? To answer that we need to examine what the words "being wet", mean in terms of subjective experience. A person who is wet, feels cold, clothes cling to the body, when the person walks, he or she will leave footprints, which look dark at first but (in the right atmospheric conditions, gradually evaporate. A wet person can often feel drips of water running down the back, and so on. All these effects could be built into a sufficiently sophisticated model of a rainstorm. In those terms the model could therefore feel wet. It would experience being wet just as one of us would feel wet, in terms of the sensory signals transmitted from eyes, ears, tactile receptors and so on. I think Searle, by avoiding serious consideration of the terms he is using, and trying instead to distort and therefore to distort the arguments advanced by others, is doing philosophy a disservice.

QUOTATION (section 4, para 4)
Its point is simply this: Computation is defined syntactically. It is defined in terms of the manipulation of symbols. But the syntax by itself can never be sufficient for the sort of contents that characteristically go with conscious thoughts. Just having zeros and ones by themselves is insufficient to guarantee mental content, conscious or unconscious. ......[comment here about his Chinese Room argument, dealt with elsewhere]........................ Syntax by itself is not sufficient for semantic content. In all of the attacks on the Chinese room argument, I have never seen anyone come out baldly and say they think that syntax is sufficient for semantic content.

MY COMMENT
I don't think that Searle understands the nature of syntax and semantics, although these are terms which he uses freely and with great emphasis. He certainly does not understand the procedural view of consciousness. Syntax is a set of rules governing the correct pattern of usage of various symbols in a language. The semantics, or meaning of an expression, is the interpetation made by the procedure involved. The procedure bestows the semantic interpretation on upon the expression (including the pattern in which the individual symbols occur). I do not claim, as he puts it, that "syntax is suffient for semantics". That is a total misrepresentation of the procedural view of consciousness. I do claim that syntax can make a contribution to a semantic interpretation. When a procedure interprets a particular arrangement of symbols, it triggers certain consequences. It is the range and pattern of those consequences which constitute the semantics of a particular arrangement of symbols. In that sense, syntax is a component of semantics. It is the behaviour of the process which determines the semantic interpretation of an expression, not the syntax. Searle's error springs I think from his complete lack of understanding of the complexity of language interpretation. His famous chinese room argument, asks us to suppose that an artificial system based entirely on syntactical relationships, could achieve a level of language processing sufficient to pass muster as a human speaker of a given language. That idea is false. No purely grammatical system could ever achieve such a level of performance. I think Searle has been completely hoodwinked by a facile demonstratioin of the ELIZA machine which did deceive a few naive people into thinking that it was actually equivalent to a human speaker of English. This was, however, equivalent to a stage conjuring trick. It no more proved the adequacy of syntax alone for this purpose that a slick stage conjurer can proved the existence of magic.

If it is still not clear that Searle has got this syntax/semantics argument upside-down and back-to-front, then consider this - A utterance (in a language) is a sequence of words. Every word is a sequence of sounds. Every sound is a vibration of air molecules. So an utterence, at its foundation is nothing but a beating to-and-fro of air molecules. Now air molecules (to adopt Searle's terminology) have "no semantics". So a word can have no semantics, In this way, going back up the chain, we come to the absurd conclusion that an utterence in any language can not have any "semantics". That is, it cannot have any meaning. The reasen we have come to that ludicrous conclusion, is that we assumed (with searle) that "semantics" was some kind of magic juice which inhabits the basic components of the system. In reality, "having semantics" is not an intrinsic property of the components. It is a property bestowed on various arrangements of those components, by the procedure which receives and processes the sequence of tokens - in this case an utterence of a sequence of words. It is entirely possible then, that a single utterence could have two or more (semantic) interpretations if it interpreted by two speakers of two different languages. In its written form, the word "four" in English means the number 4. In French it means a kiln or oven. Could there be a clearer demonstration that a semantic interpretation does not belong to the symbols being interpreted. It belongs to the interpretation process. It pains me to have to point that out to someone who is a professional academic philosopher.

GENERAL COMMENTS
In my book AN EXPEDIENT MIND I made the point that the procedural or functional view of consciousness is that consciousness is not CAUSED by neurological procedures, it IS just those procedures in operation. Another way to put that, is that consciousness supervenes on neurological processes. An useful analogy is the way the temperature of a mass of gas supervenes on the movement of its molecules. This is sometimes loosely described as a causal relationship, but that is a misleading use of the word "cause". A causal relationship implies a time sequence between cause and effect. It also implies that we could in some way break that connection. We could then have a system which has all the necessary procedures in operation without any attendant consciousness. For a supervenience relationship like the one between the movement, or kinetic energy of gas molecules, and the measured temperature of the gas, such a separation is not possible. When we measure the temperature of a gas (or anything else for that matter), what we are measuring is the average kinetic energy of its molecules. The procedural view of consciousness is exactly the same. When we note that a person or an animal is conscious what we are noting is that its brain is performing a particular kind of procedure. You cannot have one without the other, because they are the same thing.

I think, however, that Searle, in this article, is trying to have it both ways. He specifically tells us that his insistance that the neurological processes CAUSE consciousness, does NOT imply that the two phenomena - the processes and the consciousness are two separate entities. I quote

.... does the claim that there is a causal relation between brain and consciousness commit us to a dualism of `physical' things and `mental' things? The answer is a definite no. Brain processes cause consciousness but the consciousness they cause is not some extra substance or entity. It is just a higher level feature of the whole system. The two crucial relationships between consciousness and the brain, then, can be summarized as follows: lower level neuronal processes in the brain cause consciousness and consciousness is simply a higher level feature of the system that is made up of the lower level neuronal elements. (section 2, para 3).

which seems to be in complete agreement with my own position. But stay ... Later, (section 4, para2) he discusses the Turing Test and makes this statement

The Turing test disposes us to make precisely these two mistakes, the mistake of behaviorism and the mistake of computationalism. It leads us to suppose that for a system to be conscious, it is both necessary and sufficient that it has the right computer program or set of programs with the right inputs and outputs. I think you have only to state this position clearly to enable you to see that it must be mistaken. A traditional objection to behaviorism was that behaviorism could not be right because a system could behave as if it were conscious without actually being conscious.

This, it seems to me, implies that the CAUSAL link (claimed by Searle) between neurological procedures and consciousness, can be broken, so that we can have the process without the consciousness. I suspect, however, that Searle's notion of a system which behaves as if it were conscious, is very superficial, in line with his naive acceptance of the ELIZA machine as a model of an artificial language system. The behaviour which I have in mind, however, would be something like the ability to write a treatise or carry on a lengthy discussion on the topic of consciousness.

For these reasons I think Searle's view is self-contradictory. The causal link between process and consciousness, on which he insists, inspite of these protestations, really does imply separability, and therefore a form of dualism. The second entity may not be a non-physical entity, but Searle's position implies that it must be a separate entity. And that is the position which I dispute. The position which I DO support and which is explained fully in my book AN EXPEDIENT MIND, is that the consciousness is a collection of procedures which operate together to produce a specific kind of intelligent, predictive behaviour. And where does the subjective experience come from? That is just the same question, expressed in different words. Having a subjective experience is what these processes do when they operate together. They are the mechanism of subjective experience.

AN OLD JOKE
Searle's argument reminds me of an old joke. You must have heard it, but it will bear repetition. A lunatic consults a psychiatrist. The psychiatrist draws a series of pictures (a square, a circle, a cross, a horizontal line, a dot, a vertical line, a triangle, and so on). He asks the lunatic to name for each the first thing that comes into his mind. And for each the lunatic says just one word 'SEX'. Eventually the psychiatrist puts down his pen and says, "My dear fellow, you have an obsession about sex." "You should talk," replies the lunatic. "You sit there all day drawing nothing but dirty pictures."

What I like about that joke is that there is a kind of logic to the lunatic's position. From his point of view, given his unshakable conviction that every shape or image is related to sex, then he is absolutely correct. The psychiatrist has been drawing dirty pictures.

Now take a look again at Searle's position. He says -

The characteristic mistake in the study of consciousness is to ignore its essential subjectivity and to try to treat it as if it were an objective third person phenomenon. Instead of recognizing that consciousness is essentially a subjective, qualitative phenomenon, many people mistakenly suppose that its essence is that of a control mechanism or a certain kind of set of dispositions to behavior or a computer program. The two most common mistakes about consciousness are to suppose that it can be analysed behavioristically or computationally.

It seems to me that Searle has an unshakable conviction that consciousness cannot be explained in terms of mechanism. He doesn't explain why. He simply asserts that that it is a self-evident fact. A mechanism, in his view, must be a third-party observable entity. He cannot even consider the possibility that any person, who has a personal experience, IS A MECHANISM, or that personal experience can have functional components - like the recording of a memory or access being made to that entry in the memory trace - that that may be what personal experience is, a whole collection of mechanistic operations. Because he insists that subjective experience is non-mechanistic it follows (like the lunatic's low opinion of the psychiatrist) that Searle must regard anyone who tries to explain consciousness in terms of mechanism, of failing to understand that consciousness is a subjective experience (i.e. is not a mechanism, which must, in his view, be only a third-party observable entity). The lunatic sees a CONNECTION between the simple images and SEX. But cannot see that that connection is created in his own mind. Searle sees a DISCONNECTION between mechanism and subjective experience. but cannot see that the DISCONNECTION is something created by his own predispositions.

Tartan Hen Publications : Home | more books | Contact : feedback@tartanhen.co.uk