THE METAPHYSICS OF AN AUTOMATON.
MR. BABBAGE, in the amusing book which we have noticed elsewhere, enters at some length and very instructively into the intellectual advantages and disadvantages of automatons as compared with men. In many respects, of course, Mr. Babbage assigns them a very great superiority. Of those of a mathematical bent, for instance, it is not so much true that they won't make mistakes a> that they can't. And yet they are perfectly aware of their own needs, and ring the bell quite cheerfully when in want of their human attendants. When one of Mr. Babbage's mathe- matical automatons " wanted a tabular number, say the logarithm of a given numbei., it would ring a bell, and then stop itself. On this the attendant would look at a certain part of the machine, and find that it wanted the logarithm of a given number, say of 2,303. The attendant would then go to the drawer containing the pasteboard cards representing its table of logarithms. From this he would take the required logarithmic card and place it on the machine. Upon this the engine would first ascertain whether the assistant had or had not given him the correct logarithm of the number ; if so, it would use it, and continue its work. But if the engine found the attendant had given him a wrong logarithm, it would then ring a louder bell and stop itself. On the attendant again examining the engine, he would observe the words ' wrong tabular number,' and discover that he really had given the wrong logarithm, and of course be would have to replace a by the right one." This clearly is an automaton of the highest order of mechanical intel- ligence and purpose, and yet it combines with this pertinacity of resolution and discrimination of understanding an absolutely un- erring accuracy in arithmetical operations. Indeed it is more than unerring, incapable of error. Robert Houdin's automatons were of very different and very inferior order to most of Mr. Babbage's and compare rather with Mr. Babbage's "silver lady," who received his guests, than with the great intelligences which the English philosopher called into existence. For instance, Robert Houdin invented a writing man which wrote or drew answers to questions that were put to it, and which was once, in 1848, so fortunate in its guess-work that in drawing a crown as the symbolic answer to a question about the destiny of the present Count de Paris, the pencil broke in its hand, and left the crown a mere unfinished anticipation, almost a prophecy. But, in general, even this automaton's intellect was strictly limited by that of M. Houdin's- giving, for instance, in answer to Louis Philippe's question about the population of Paris, the number according to the old census, without allowance for the subsequent increase. This automaton, therefore, was only an ingenious trick, was scarcely, indeed, a much greater triumph than Vaucanson's automatic duck, which quacked, put out its bill to drink and dabble, swallowed seed, digested it, and passed it by the ordinary channels. Mere imita- tive motions like these are scarcely worthy of the name of automatic. On the other hand, Mr. Babbage's intellectual automatons perform with far greater precision, and on a far more extended scale, operations of which its maker of course fully understood the theory, but in which he is by no means infallible in practice,—in other words, they far outstrip him in the application of his own mathematical principles. The curious and instructive point is, however, to note their specific intellectual disadvantages and difficulties when you compare them with the more fallible intelligence of living men. Mr. Babbage gives us a very interesting illustration of the exact point of divergence between subordinate automatic intelligence and the human intellect which invented it. He proclaims that, in principle at least, all games of skill could be played by automatons constructed on the same general theory as his Calculating Machines, and he has proceeded some way towards the invention of an automaton intended to play at the very simple game called indifferently " naughts and crosses " or " tit-tat-to." There was, however, one not insuperable, but characteristic difficulty. In any case in which it does not matter which of two or three moves should be made by the automaton, it is still absolutely necessary not only to make the automaton select one of them, but to select that one on rule. Now as there is no principle arising out of the intellectual conditions of the game to provide a rule, it become necessary to make an arbitrary rule for the automaton's guidance in this case,—a "cere- monial law," as we may call it, failing an intellectual law. It is nearly a parallel to the case of what we may call the etiquettes of good society. We all know that there is no intrinsic reason why tail-coats are worn in the evening and frock-coats in the morning, and on that very account the rule is more stringent, and its infringement graver, than in the case of rules of social courtesy for which there is good reason. The explanation is, we suppose, that if there were no rule at all, then there would be caprice, and confusion, and anarchy, while in the case of the natural law there would be still something tending to enforce it even if custom were silent. Hence an arbitrary rule is laid down in such cases which soon gets more honour and respect than even rules with a bottom to them. For a very similar reason Mr. Babbage had to find his automaton a principle on which to choose between the two or three equally good moves, and make him put his naughty and crosses in a given place for fear of the anarchical embarrassment which would arise if he had no definite line of action before him. The first arbitrary rule he invented was " to make the machine keep a record of the number of games it had won from the com- mencement of its existence. Whenever two moves, which we may call A and B, were equally conducive to winning the game, the automaton was made to consult the record of the number of games he had won. If that number happened to be even, he was directed to take the course A, if odd, the course B. If there were three moves equally possible, the automaton was directed to divide the number of games he had won by 3. In this case the numbers 0, 1, or 2 might be the remainders, and the machine was directed to take the course A, or B, or C, accordingly." In other words, the automaton gets into difficulties exactly where what metaphysicians call the " liberty of indifference " would come into play. He is obliged to have a rule of conduct when there is no reason why there should be a rule at all, because he, like the victims of society, can have nothing left to his discretion. If he once finds himself without a specific direction as to what he is to do next, he is a lost creature; the law of his being fails him ; he must refer back to his govern- ment for instructions ; he has no power to make a choice. At this meeting of the ways, accordingly, Mr. Babbage has to invent some-
thing equivalent to a binding etiquette for his automaton who can do everything but act without a criterion, but does not care in the least whether that criterion is natural, or artificial and arbitrary. There is a curious parallel to this in the mode adopted by young or superstitious people who cannot decide for themselves what to do or how to believe, and who fix on some arbitrary test which shall be a sign to them what they shall do or believe. Thus an odd number of magpies decides many people to expect misfortune, and an even number success. Or, to take a real instance, Mr. Babbage himself when a child was much exercised with doubts whether the religious truths he had learnt in the Bible were true or not, and he made for himself an arbitrary test by which he determined to brsguided. He said to himself that if on going to a cer- tain room in his home he should find the door open he would believe what he had been taught, but if it should be shut, that he would not. He cannot remember, he tells us, whether in fact he found it open or shut, but he supposesthe former, as his childhood was for many years disturbed by no further doubts on the subject. Now this only differs from the considerations which determined his own automaton's next move in this, that the arbitrary sign was, or at least appeared to be, his own selection, while the automaton's equally arbitrary sign was selected for it by Mr. Babbage.
What, now, is the real difference between the intelligence of the automaton and that of man? Some people will say at once consciousness,—the child is conscious of his calculating power and perhaps of its method, the automaton not. But then it is not the fact that people are conscious of half the mental operations they per- form, and many thinkers now maintain that some of their most wonderful intellectual efforts are done in complete unconscious- ness. It is the favourite explanation of the spirit-drawing and writing phenomena,—so far as they are not trickeries,—that the in- tellect acts in them automatically; that is, uses intellectual tests and .criteria without being conscious that it is using them. We have seen very beautiful drawings made by a lady of the most unques• tionable honour and integrity, who avers not only that she never could draw at all till the "influence" seized her, but that she never at any time knew what the next stroke of her drawing was to be ; nay, that the unfolding of the subject was a greater surprise and interest to her than to those who were watching her, as it grew under her hand. Now, we have always explained this as being unconscious artistic instinct developing itself,—in other words, unconscious intellect. If there be such a phenomenon, and we believe all the physicians who have studied dreams, the acts of somnambulists, and so forth, affirm it most strongly, in what does the unconscious intelligence of the automaton contrived beforehand by Mr. Babbage differ from the unconscious in- telligence of man or woman contrived beforehand by the Creator of man ? If we once admit the absolute un- consciousness of the latter, we doubt whether the acutest metaphysician could find a discriminating criterion. The law of unbroken necessity applies equally to both, for if ever you come 'to a point where courses of action diverge, and there is abso- lutely nothing to determine which course shall be chosen, the auto- matic action of the human intelligence would cease, and either be wakened up into an act of conscious choice, or be foiled as com- pletely as the automaton. In short, what Mr. Babbage's automa- tons teach us is that consciousness is really a defect and a cause of error so far as the mere carrying out of absolutely necessary in- tellectual laws is concerned ; and that either a human intelligence acting automatically, or a machine contrived by human intelli- gence, will carry out all such necessary laws more precisely and rapidly than a mind which reflects upon what it is doing. But they also teach us that where the inexorable chain-work of neces- sary law ends, there the use of freedom and consciousness begins; —of freedom, because an artificial and arbitrary law has to be in- troduced to guide the automaton, simply because it is not equal to that very lowed and simplest of all free acts, the tossing up, as it were, which of two or three equally beneficial courses it shall take, because it cannot determine itself, and in the absence of reason must be determined by a rule ;—of consciousness, because all free choice, all acts of judgment, involve consciousness, and though the intellect can act, so to say, in a lienear way, that is, along a stream of necessary sequences, without awakening, the moment it has to divide itself as it were,—to enter into two different but simultaneous courses, and select between them, it is no longer capable of automatic action, and must take up self-knowledge in the very act of choosing between two alternatives. Mr. Babbage's ingenious experiments constitute a very curious demonstration that the more mechanical, the more automatic, is the action of the intellect, the freer from error will be the operations which it performs ; that the intellectual automaton is for its purposes the superior of the
intellect, because it has neither freedom nor consciousness to disturb its operations ; but that where the automaton gets into difficulties is exactly at that point which the reigning school of philosophy wish to ignore altogether,—the point where freedom and self- consciousness enter together into mental life.
Previous page
