As for Jefferson’s remark that unless a machine could “create concepts and find for itself suitable words in which to express them . . . it would be no cleverer than a parrot,” it provoked a spirited defense of the bird in the paper’s editorial pages that brought an end to the brouhaha, with the author facetiously complaining that
those who have never loved a parrot can hardly appreciate the vehemence of the emotions aroused by these thoughtless words in the breasts of those who have made of this sagacious bird a close and (as far as can be ascertained) devoted companion. . . . Parrots can make things devilish unpleasant if they take a dislike to you, and it would be a prudent as well as a courteous gesture if Professor Jefferson withdrew an observation which has ruffled so many and so well-loved feathers.
So far as the Times was concerned, the call for an apology to the parrot (but not to the scientists) brought the matter to a close. But Turing did not forget what Jefferson had said. If anything, the exchange in the newspaper’s pages only strengthened his interest in machine intelligence. He would soon strike back—and even the parrot would make another appearance.
2.
“Computing Machinery and Intelligence,” Alan Turing’s most famous and in many ways most perverse paper, appeared in Mind in October 1950. Whereas in the NPL report he started with likely objections, here he saved the list of potential objections to computer intelligence for later, and began instead with a clear statement of his intent. “I propose to consider the question, ‘Can machines think?’ This should begin with definitions of the meaning of the terms ‘machine’ and ‘think.’ ” But if these meanings “are to be found by examining how they are commonly used it is difficult to escape the conclusion that the meaning and the answer to the question, ‘Can machines think?’ is to be sought in a statistical survey such as a Gallup poll.” Such an idea, in Turing’s view, was “absurd.”
Instead of offering definitions, Turing recast his question by proposing what he called the imitation game. It would later become known as the Turing test, much as the a-machine of “Computable Numbers” has come to be called a Turing machine. The game, as he explains it,
is played with three people, a man (A), a woman (B), and an interrogator (C) who may be of either sex. The interrogator stays in a room apart from the other two. The object of the game for the interrogator is to determine which of the other two is the man and which is the woman. He knows them by labels X and Y, and at the end of the game he says either “X is A and Y is B” or “X is B and Y is A.” The interrogator is allowed to put questions to A and B thus:
C: Will X please tell me the length of his or her hair?
Now suppose X is actually A, then A must answer. It is A’s object in the game to try and cause C to make the wrong identification. His answer might therefore be
“My hair is shingled, and the longest strands are about nine inches long.”
In order that tones of voice may not help the interrogator the answers should be written, or better still, type-written. The ideal arrangement is to have a teleprinter communicating between the two rooms. Alternatively the question and answers can be repeated by an intermediary. The object of the game for the third player (B) is to help the interrogator. The best strategy for her is probably to give truthful answers. She can add such things as “I am the woman, don’t listen to him!” to her answers, but it will avail nothing as the man can make similar remarks.
We now ask the question, “What will happen when a machine takes the part of A in this game?” Will the interrogator decide wrongly as often when the game is played like this as he does when the game is played between a man and a woman? These questions replace our original, “Can machines think?”
Turing’s proof, in “Computable Numbers,” that the Entschei-dungsproblem was insoluble relied on the ingenious substitution of a complicated question—can a machine decide whether a statement is provable?—with a simpler one: does a certain machine ever print a 0? Along the same lines, in “Computing Machinery and Intelligence” he argued that the complicated question “Can machines think?” could be substituted with the simpler question “Can a machine win the imitation game?” The two, in Turing’s view, were identical, because behavior, as he saw it, was identity. And yet to apply such a mathematically precise notion of identity to the murky matter of what “human” meant was to invite all sorts of objections—and problems.
For example, the ambiguity of Turing’s query “What will happen when the machine takes the part of A?” has occasioned much debate. Does Turing mean to say that instead of being played between a man and a woman, the game should be played between a man and a machine? The rest of the paper would seem to bear out this interpretation. Yet a literal reading of the paragraph suggests a different meaning: that the game should now be played between a man and a computer pretending to be a man pretending to be a woman. Hodges shows little patience for this reading, going so far as to argue that “Turing’s gender-guessing analogy detracts from his own argument. . . .” After all, as he points out, the section that follows the troublesome paragraph is entirely concerned with the ways in which a machine might trick an interrogator into believing that he (or she) was talking to a human being—male or female:
The new problem has the advantage of drawing a fairly sharp line between the physical and the intellectual capacities of a man. No engineer or chemist claims to be able to produce a material which is indistinguishable from the human skin.* It is possible that at some time this might be done, but even supposing this invention available we should feel there was little point in trying to make a “thinking machine” more human by dressing it up in such artificial flesh. The form in which we have set the problem reflects this fact in the condition which prevents the interrogator from seeing or touching the other competitors, or hearing their voices. Some other advantages of the proposed criterion may be shown up by specimen questions and answers. Thus:
Q: Please write me a sonnet on the subject of the Forth Bridge.
A: Count me out on this one. I never could write poetry.
Q: Add 34957 to 76764.
A: (Pause about 30 seconds and then give as answer) 105621.
Q: Do you play chess?
A: Yes.
Q: I have K at my K1, and no other piece. You have only K at K6 and R at R1. What do you play?
A: (After a pause of 15 seconds) R-R8 mate.
Hodges is correct to observe that gender plays no role in the answers given here (including the incorrect addition). And yet to ignore the subtext that Turing’s ambiguity exposes is also to ignore the palpable tone of sexual anxiety that runs all through the paper. For instance, just a few paragraphs after the dialogue quoted above, Turing writes, “It might be urged that when playing the ‘imitation game’ the best strategy for the machine may possibly be something other than imitation of the behaviour of a man. . . . In any case there is no intention to investigate here the theory of the game, and it will be assumed that the best strategy is to try to provide answers that would naturally be given by a man.”* Trying to provide “answers that would naturally be given by a man” would, of course, also be the best strategy for a homosexual to adopt when trying to persuade an interrogator that he is straight; in this alternative version of the imitation game, he would talk about cricket and describe the woman he would like to marry. And though the parallel may be accidental—“a man,” after all, could as easily mean “a human being” as “a male human being”—Turing’s use of the word “naturally” suggests a more heightened awareness of the idea of the “natural” than the situation calls for. Not surprisingly, arguments concerning the naturalness or unnaturalness of homosexuality ran through both antihomosexual diatribes and apologies for homosexuality written in the period, with Oscar Wilde’s championing of the artificial often brandished as an ironic defense of “unnatural” love.*
Turing’s preoccupation with gender recurs several more times during the course of the paper. In section 3, a discussion of exactly what defin
es a “machine” concludes with this rather bizarre proviso that
we wish to exclude from the machines men born in the usual manner.
It is difficult to frame the discussion so as to satisfy [this condition]. One might for instance insist that the team of engineers should be all of one sex, but this would not really be satisfactory, for it is probably possible to rear a complete individual from a single cell of the skin (say) of a man. To do so would be a feat of biological technique deserving of the very highest praise, but we would not be inclined to regard it as a case of “constructing a thinking machine.”
Is the point here that the team of engineers—all of “one sex”—might be able to join together and in a sort of orgy of cloning create a human child? The fantasy is peculiar, using science as a framework for imagining a means by which men without women could generate progeny. Of course, Turing too longed to produce a child of his own—a computer child. It is therefore not surprising that in the paper he soon returns to the metaphor of child rearing and education, employing a “domestic analogy” to describe the ways in which a machine might be taught to obey “not fresh instructions on each repetition, but the same ones over and over again”:
Suppose Mother wants Tommy to call at the cobbler’s every morning on his way to school to see if her shoes are done, she can ask him afresh every morning. Alternatively she can stick up a notice once and for all in the hall which he will see when he leaves for school and which tells him to call for the shoes, and also to destroy the notice when he comes back if he has the shoes with him.
“Tommy” is here the computer, the offspring of a group of engineers who have eschewed cloning in favor of other styles of cooperation—perhaps the sort of cooperation in scientific experiment that Turing so cherished in his friendship with Christopher Morcom. More importantly, Tommy is a digital computer, and in Turing’s estimation only a digital computer—a universal machine—has a shot at ever winning the imitation game.
I believe that in about fifty years’ time* it will be possible to programme computers, with a storage capacity of about 109, to make them play the imitation game so well that an average interrogator will not have more than 70 per cent chance of making the right identification after five minutes of questioning.
The Manchester “Baby” is clearly growing up.
3.
By this point, then, a subtle but distinct strain of anxiety concerning gender, sexual imitation, and even homosexual procreation has come to assert itself within Turing’s “official” argument about machine intelligence. But where does it come from? The answer can be traced back to Sir Geoffrey Jefferson’s Lister Oration, the slightly masculinist tone of which Turing ridicules in the paper, even as he rebuts Jefferson’s “humanist” stance. This is especially evident near the middle of “Computing Machinery and Intelligence,” where Turing takes up once again the strategy of listing—and then refuting—the objections that might be raised to the possibility of a thinking machine. Although Professor Jefferson does not appear by name until the fourth objection—“The Argument from Consciousness”—his spirit is invoked, and mocked, from the very start.
For instance, in his refutation of the first objection—“the theological objection” that “God has given an immortal soul to every man and woman, but not to any other animal or to machines”—Turing questions the implicit superiority of mankind that provided the basis for Jefferson’s diatribe, noting, “I should find the argument more convincing if animals were classed with men, for there is a greater difference, to my mind, between the typical animate and the inanimate than there is between man and the other animals.” Likewise, how are Christians to contend with “the Moslem view that women have no souls”? By invoking the rights not just of women but of animals, Turing allies himself (and his computer) with all the other populations that have suffered at the hand of religions that take the superiority of man (in one case) and mankind (in the other) for granted. Against this he posits his own rather odd theology, which, needless to say, blesses machines, by equating their construction with procreation: “In attempting to construct such machines we should not be irreverently usurping His power of creating souls, any more than we are in the procreation of children: rather we are, in either case, instruments of His will providing mansions for the souls that He creates.”
The assumption of mankind’s innate superiority is challenged even more boldly in Turing’s retort to the second objection, which he calls the “Heads in the Sand” objection and sums up as follows: “The consequences of machines thinking would be too dreadful. Let us hope and believe that they cannot do so.” This, of course, was the very posture to which some of Norbert Wiener’s writings inadvertently appealed, and in responding to it, Turing also responds to Jefferson, noting that the feeling that mankind is “necessarily superior” to the rest of creation “is likely to be quite strong in intellectual people, since they value the power of thinking more highly than others, and are more inclined to base their belief in the superiority of Man on this power.” With his allusions to Shakespeare, Jefferson is exemplary of these “intellectual people” whose tendency to exalt their own species Turing shows so little patience for. It is a point he returns to in his answer to the third objection, the “mathematical objection,” which is essentially the argument (paraphrased in the NPL report) that his own resolution of the Entscheidungsproblem, in conjunction with Gödel’s findings, proves “that there are certain things that . . . a machine cannot do.” Turing was obviously made uncomfortable by the possibility that his solution to the Entscheidungsproblem might be employed in an attack on the machine that the Entscheidungsproblem propelled him to create. In responding to it here, however, he focuses squarely on the psychology of what might be called the natural “superiority complex” of human beings (especially intellectuals), observing shrewdly that when a machine gives a wrong answer to
the appropriate critical question . . . this gives us a certain feeling of superiority. Is this feeling illusory? It is no doubt quite genuine, but I do not think too much importance should be attached to it. We too often give wrong answers to questions ourselves to be justified in being very pleased at such evidence of fallibility on the part of machines. Further, our superiority can only be felt on such an occasion in relation to the one machine over which we have scored our petty triumph. There would be no question of triumphing simultaneously over all machines.
Here Turing seems to be amusing himself, in a rather quiet way, by alluding to Mr. Illtyd Trethowan’s anxiety about “the possible hostility of the machines,” over all of whom we can never hope to triumph. More importantly, this rebuttal gives him the chance to repeat one of his key points—that fallibility is a key ingredient in intelligence.
It is in his refutation of objection 4—“the Argument from Consciousness”—that Turing takes direct aim at Jefferson, whom he begins by quoting and at whom he hurls one of his most memorable and witty retorts:
This argument appears to be a denial of the validity of our test. According to the most extreme form of this view the only way by which one could be sure that a machine thinks is to be the machine and to feel oneself thinking. . . . Likewise according to this view the only way to know that a man thinks is to be that particular man. It is in fact the solipsist point of view. It may be the most logical view to hold but it makes communication of ideas difficult. A is liable to believe “A thinks but B does not” whilst B believes “B thinks but A does not.” Instead of arguing continually over this point it is usual to have the polite convention that everyone thinks.
Rather cleverly, Turing writes that he is “sure that Professor Jefferson does not wish to adopt the extreme and solipsist point of view.” He then compares his own imitation game with a game called viva voce, the purpose of which is “to discover whether some one really understands something or has ‘learned it parrot fashion.’” Notably, the exemplary viva voce that Turing cites is replete with literary references, with the questioner first asking his subject ab
out Shakespeare, then veering into Dickens. The point is that the imitation game also determines whether someone has learned something “parrot fashion”; it differs from viva voce only in that the person being tested is a machine. Nor is it a coincidence that literature plays such a prominent role in this viva voce, the orchestrator of which is presumably a self-proclaimed intellectual such as Jefferson. And surely any self-respecting intellectual would rather abandon the argument from consciousness “than be forced into the solipsist position.”
Having got rid of Jefferson—at least in name—Turing next addresses a whole class of objections that he calls “Arguments from Various Disabilities,” and which he defines as taking the form “I grant you that you can make machines do all the things you have mentioned but you will never be able to make one to do X.” He then offers a rather tongue-in-cheek “selection”:
Be kind, resourceful, beautiful, friendly; have initiative, have a sense of humour, tell right from wrong, make mistakes; fall in love, enjoy strawberries and cream; make some one fall in love with it, learn from experience; use words properly, be the subject of its own thought; have as much diversity of behaviour as a man, do something really new.
As Turing notes, “no support is usually offered for these statements,” most of which are
founded on the principle of scientific induction. . . . The works and customs of mankind do not seem to be very suitable material to which to apply scientific induction. A very large part of space-time must be investigated, if reliable results are to be obtained. Otherwise we may (as most English children do) decide that everybody speaks English, and that it is silly to learn French.
The Man Who Knew Too Much: Alan Turing and the Invention of the Computer (Great Discoveries) Page 21