The Chinese Room Revisited
"After the Rain - How the West Lost the East"
Click Here for Information about "Malignant Self Love - Narcissism Revisited" !
Whole forests have been wasted in the effort to refute the Chinese Room Thought Experiment proposed by Searle in 1980 and refined (really derived from axioms) in 1990. The experiment envisages a room in which an English speaker sits, equipped with a book of instructions in English. Through one window messages in Chinese are passed on to him (in the original experiment, two types of messages). He is supposed to follow the instructions and correlate the messages received with other pieces of paper, already in the room, also in Chinese. This collage he passes on to the outside through yet another window. The comparison with a computer is evident. There is input, a processing unit and output. What Searle tried to demonstrate is that there is no need to assume that the central processing unit (the English speaker) understands (or, for that matter, performs any other cognitive or mental function) the input or the output (both in Chinese). Searle generalized and stated that this shows that computers will never be capable of thinking, being conscious, or having other mental states. In his picturesque language “syntax is not a sufficient base for semantics”. Consciousness is not reducible to computations. It takes a certain “stuff” (the brain) to get these results.
Objections to the mode of presentation selected by Searle and to the conclusions that he derived were almost immediately raised. Searle fought back effectively. But throughout these debates a few points seemed to have escaped most of those involved.
First, the English speaker inside the room himself is a conscious entity, replete and complete with mental states, cognition, awareness and emotional powers. Searle went to the extent of introducing himself to the Chinese Room (in his disputation). Whereas Searle would be hard pressed to prove (to himself) that the English speaker in the room is possessed of mental states – this is not the case if he himself were in the room. The Cartesian maxim holds: "Cogito, ergo sum". But this argument – though valid – is not strong. The English speaker (and Searle, for that matter) can easily be replaced in the thought experiment by a Turing machine. His functions are recursive and mechanical.
But there is a much more serious objection. Whomever composed the book of instructions must have been conscious, possessed of mental states and of cognitive processes. Moreover, he must also have had a perfect understanding of Chinese to have authored it. It must have been an entity capable of thinking, analysing, reasoning, theorizing and predicting in the deepest senses of the words. In other words : it must have been intelligent. So, intelligence (we will use it hitherto as a catchphrase for the gamut of mental states) was present in the Chinese Room. It was present in the book of instructions and it was present in the selection of the input of Chinese messages and it was present when the results were deciphered and understood. An intelligent someone must have judged the results to have been coherent and “right”. An intelligent agent must have fed the English speaker with the right input. A very intelligent, conscious, being with a multitude of cognitive mental states must have authored the “program” (the book of instructions). Depending on the content of correlated inputs and outputs, it is conceivable that this intelligent being was also possessed of emotions or an aesthetic attitude as we know it. In the case of real life computers – this would be the programmer.
But it is the computer that Searle is talking about – not its programmer, or some other, external source of intelligence. The computer is devoid of intelligence, the English speaker does not understand Chinese (=”Mentalese”)– not the programmer (or who authored the book of instructions). Yet, is the SOURCE of the intelligence that important ? Shouldn’t we emphasize the LOCUS (site) of the intelligence, where it is stored and used?
Surely, the programmer is the source of any intelligence that a computer possesses. But is this relevant? If the computer were to effectively make use of the intelligence bestowed upon it by the programmer – wouldn't we say that it is intelligent? If tomorrow we will discover that our mental states are induced in us by a supreme intelligence (known to many as God) – should we then say that we are devoid of mental states? If we were to discover in a distant future that what we call “our” intelligence is really a clever program run from a galactic computer centre – will we then feel less entitled to say that we are intelligent? Will our subjective feelings, the way that we experience our selves, change in the wake of this newly acquired knowledge? Will we no longer feel the mental states and the intelligence that we used to feel prior to these discoveries? If Searle were to live in that era – would he have declared himself devoid of mental, cognitive, emotional and intelligent states – just because the source and the mechanism of these phenomena have been found out to be external or remote? Obviously, not. Where the intelligence emanates from, what is its source, how it is conferred, stored, what are the mechanisms of its bestowal – are all irrelevant to the question whether a given entity is intelligent. The only issue relevant is whether the discussed entity is possessed of intelligence, contains intelligence, has intelligent components, stores intelligence and is able to make a dynamic use of it. The locus and its properties (behaviour) matter. If a programmer chose to store intelligence in a computer – then he created an intelligent computer. He conferred his intelligence onto the computer. Intelligence can be replicated endlessly. There is no quantitative law of conservation of mental states. We teach our youngsters – thereby replicating our knowledge and giving them copies of it without “eroding” the original. We shed tears in the movie theatre because the director succeeded to replicate an emotion in us – without losing one bit of original emotion captured on celluloid.
Consciousness, mental states, intelligence are transferable and can be stored and conferred. Pregnancy is a process of conferring intelligence. The book of instructions is stored in our genetic material. We pass on this book to our off spring. The decoding and unfolding of the book are what we call the embryonic phases. Intelligence, therefore, can (and is) passed on (in this case, through the genetic material, in other words: through hardware).
We can identify an emitter (or transmitter) of mental states and a receiver of mental states (equipped with an independent copy of a book of instructions). The receiver can be passive (as television is). In such a case we will not be justified in saying that it is “intelligent” or has a mental life. But – if it possesses the codes and the instructions – it could make independent use of the data, process it, decide upon it, pass it on, mutate it, transform it, react to it. In the latter case we will not be justified in saying that the receiver does NOT possess intelligence or mental states. Again, the source, the trigger of the mental states are irrelevant. What is relevant is to establish that the receiver has a copy of the intelligence or of the other mental states of the agent (the transmitter). If so, then it is intelligent in its own right and has a mental life of its own.
Must the source be point-like, an identifiable unit? Not necessarily. A programmer is a point-like source of intelligence (in the case of a computer). A parent is a point-like source of mental states (in the case of his child). But other sources are conceivable.
For instance, we could think about mental states as emergent. Each part of an entity might not demonstrate them. A neurone cell in the brain has no mental states of it own. But when a population of such parts crosses a quantitatively critical threshold – an epiphenomenon occurs. When many neurones are interlinked – the results are mental states and intelligence. The quantitative critical mass – happens also to be an important qualitative threshold.
Imagine a Chinese Gymnasium instead of a Chinese Room. Instead of one English speaker – there is a multitude of them. Each English speaker is the equivalent of a neurone. Altogether, they constitute a brain. Searle says that if one English speaker does not understand Chinese, it would be ridiculous to assume that a multitude of English speakers would. But reality shows that this is exactly what will happen. A single molecule of gas has no temperature or pressure. A mass of them – does. Where did the temperature and pressure come from? Not from any single molecule – so we are forced to believe that both these qualities emerged. Temperature and pressure (in the case of gas molecules), thinking (in the case of neurones) – are emergent phenomena.
All we can say is that there seems to be an emergent source of mental states. As an embryo develops, it is only when it crosses a certain quantitative threshold (number of differentiated cells) – that he begins to demonstrate mental states. The source is not clear – but the locus is. The residence of the mental states is always known – whether the source is point-like and identifiable, or diffusely emerges as an epiphenomenon.
It is because we can say very little about the source of mental states – and a lot about their locus, that we developed an observer bias. It is much easier to observe mental states in their locus – because they create behaviour. By observing behaviour – we deduce the existence of mental states. The alternative is solipsism (or religious panpsychism, or mere belief). The dichotomy is clear and painful: either we, as observers, cannot recognize mental states, in principle – or, we can recognize them only through their products.
Consider a comatose person. Does he have a mental life going on? Comatose people have been known to have reawakened in the past. So, we know that they are alive in more than the limited physiological sense. But, while still, do they have a mental life of any sort?
We cannot know. This means that in the absence of observables (behaviour, communication) – we cannot be certain that mental states exist. This does not mean that mental states ARE those observables (a common fallacy). This says nothing about the substance of mental states. This statement is confined to our measurements and observations and to their limitations. Yet, the Chinese Room purports to say something about the black box that we call “mental states”. It says that we can know (prove or refute) the existence of a TRUE mental state – as distinct from a simulated one. That, despite appearances, we can tell a “real” mental state apart from its copy. Confusing the source of the intelligence with its locus is at the bottom of this thought experiment. It is conceivable to have an intelligent entity with mental states – that derives (or derived) its intelligence and mental states from a point-like source or acquired these properties in an emergent, epiphenomenal way. The identity of the source and the process through which the mental states were acquired are irrelevant. To say that the entity is not intelligent (the computer, the English speaker) because it got its intelligence from the outside (the programmer) – is like saying that someone is not rich because he got his millions from the national lottery.