Issue 3 response, to appear also in Tekhnema 6 / Fall 2000


On Getting Turing Wrong Perversely



Justin Leiber


In "What Kind of Turing Test Did Turing Have in Mind,"1 Jean Lassègue apparently makes an important factual mistake which needs correcting; it may indicate that his apparent indifference to what the "Turing test" is widely taken to mean in the cognitive science community is also ground floor ignorance as well. More importantly, his plan to discredit the Turing test as presently understood within the cognitive science community by showing some supposed incongruities in Alan Turing’s original formulation of the Turing test is a blatant instance of what logicians call the genetic fallacy. Further, Lassègue uncharitably and perversely misreads Alan Turing’s own description of what Turing himself modestly called "the imitation game," so that he is then able to say that Turing’s test bizarrely handicaps the woman participant, who will immediately lose both to man and machine. Failing to think that this perverse result might just possibly mean he had misconstrued some ambiguity in Turing’s justly famous paper,2 a misconstrual with no echo in any previous commentary on this much discussed paper, Lassègue then proceeds to "explain" the absurdities he has foisted on Turing by calling him a sexist, scientistic machine-lover, all this opening the way to seeing Turing’s early psycho-sexual development mirrored in his test. I will not follow Lassègue into this latter exercise, for it soon attains what is almost self parody and it has a certain absurdist charm all its own that strikes one more as amusing good fun than as anything reprehensible (except to say that I do not know where he got the idea that Turing’s childhood circumcision might have caused him to become homosexual; he gives no citation and I can find nothing in Turing’s two biographies to support this claim).

Since Lassègue quotes me in his initial description of the Turing test, I feel a special need to correct his initial misreadings. The important factual mistake that Lassègue apparently makes appears thusly,

It was not long afterwards that computer scientists set out to show that the test – whilst only a thought experiment for Turing – could be passed by real machines in the real world. In 1992, the BBC made a documentary about the experiment that took place in Cambridge, USA, suggesting the machine had indeed passed the test with full honors. (p. 2)

It is truth that they – Allen Newell and Herb Simon at Pittsburgh and Marvin Minsky at MIT, etc. – confidently set out to do so in the 1950s. But it has been nothing but minor advances and general failure since and confidence has waned to the point where it is routine for philosophers and scientists discussing the problem to concede that it may take many decades and some enormous breakthroughs, or that it may not ever be possible, to create a Turing test passer. Indeed a huge monetary prize has long been offered for producing such a passer but no one in the field expects to cash in anytime soon. What Professor Lassègue may be referring to is an annual contest in which people who are not experts on the dodges of artificial intelligence try to tell, for example, a human knowledgeable about Shakespearean comedies from a program designed to exhibit the same specialization when all questions are confined to this narrow focus. In these contests, such programs have had some successes. But, as I said, under what has for very long been the consensus in cognitive science about what intelligence a Turing test passer would have to exhibit, under extensive and lengthy interrogation by experts, nothing remotely on the horizon looks likely to succeed. The Professor Robinson, to whose objections Turing replies in his famous 1950 paper, may be pardoned for thinking it "would be a simple thing" to program a computer to pass the test but no one now may so be pardoned. It is an impressive testament to the enormous (and heretofore largely unrecognized) complexity of ordinary human intelligence that passage now seems so difficult, even with computers that almost unthinkably exceed in memory size, speed, and programmability those beginning to become available in Turing’s day. That Lassègue may not know the well understood sense assigned the Turing test within cognitive science is indicated by his willingness to report the absurd claim that it has already been passed and by his toying with various absurd interpretations of it. That he doesn’t want to know or respect what its established sense happens to be is made clear by his complaint that it is suspicious that it isn’t called "Turing’s test," which gives it, he claims, an artificial claim to a sort of objectivity and impersonality that it ought not to have, as he recognizes has occurred as well with the mathematical notion of "Turing machine" (I cannot resist noting that the literature of biochemistry and embryology as well is positively strewn with references to "Turing structures").

But the fact is that "Turing test" does have a consensus meaning within cognitive science, one responsive to the theoretical needs and methodological commitments of the field. Lassègue writes,

[I]s the so-called "Turing test" as objective and scientific as it is claimed to be in the AI and cognitive science community? Do we have to consider it as a threshold one has to cross to be able to enter the realm of a scientific approach to the mind? My answer is "no" because the so-called Turing test [as Turing actually formulated it] says in fact more about Turing’s psychological life than about the science of mind itself. (p. 4)

To argue this way is an elementary example of the genetic fallacy. If it is true that the Cretan, Epimenides, who is credited with originating the liar paradox by saying "All Cretans are liars," did not actually recognize there was anything paradoxical in his pronouncement and he gave in any case a loose version of it, this is hardly grounds for denying that the liar paradox has an objective meaning in logic and mathematics. If the "Turing test" has a clear, useful, and established sense in the cognitive science, it is simply irrelevant to cognitive science whether Turing mis-formulated it originally or expressed it bizarrely through some unconscious, self-revealing, pathological cause. But he didn’t do that either as we will soon see.

After dismissing the question "Can a computing machine think?" as unsuitably vague, Turing proposed to operationalize the issue by substituting what he called the imitation game, the genesis of the Turing test. But he first introduced a preliminary version (game 1) in which a man and a woman compete and then substituted the machine for the man (game 2). I will quote the same lines that Lassègue quotes but I will also insert in italics the material that Lassègue crucially leaves out, replacing them with three dots, thus sanctioning his perverse misreading. Additionally, I have bold-faced the most crucial excision.

It is played between 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 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 [the male], 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 ... a teleprinter will communicate between the two rooms. The object of the game for the third player (B) [the woman] is to help the interrogator. The best strategy for her is probably to give truthful answers ...

?We now ask the question, "What will happen when a machine takes the part of A [the male] in the 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?"

These excisions allow Lassègue to assume, and let his reader assume, that the woman is supposed to have the same reciprocal imitative task (imitating the man) as the man. So he writes, several pages later, finally quoting, out of context, the last line in italic above and its continuation.

Let us go back to Turing’s description of his first game and to the strategies displayed by the players. In fact, the only example given by Turing would not lead to a successful imitation but on the contrary to an unsuccessful one, that is to say to an unreciprocal imitation.

Here is how Turing describes the woman’s strategy:

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.

What can we infer from this description? That the woman will at once be recognized by the interrogator and that the match will end immediately. If the strategy followed by the woman is to tell the truth about herself without imitating the man’s behavior, this truth will become very quickly apparent to the interrogator.

Why should the woman speak the truth? It is obviously a very bad strategy, since it should be characterized, more than anything else, as an absence of strategy ... [I]n Turing’s article the odds are weighed too heavily against the woman and this fact must somehow be explained.(p.9)

But, as Turing very clearly stated, the woman’s task is to help the interrogator (which is exactly parallel to her task in the second game, where a computer is substituted for the man). So, pace Lassègue, the strategy Turing recommends for her is not at all a very bad strategy, one so bizarre that Turing’s recommendation calls out for psychological explanation. To the contrary, it is clearly the best possible strategy and, again in parallel, exactly the right strategy for her (or any human) to follow in the second game. The natural way for the interrogator to start the full blown test is familiar to real professional interrogators: What is your name?, Where were you born and when? ... You’ve told us that you grew up living at 526 Chartres St., New Orleans. Tell us about the buildings near your home then. ... Can you hum "When the saints come marching in"? ... Didn’t you previously say that your boy friend’s name was Xavier? ... What did you say your address was in New Orleans? It is elementary that when you want the authorities to believe that you really are you, you don’t start making things up! This also applies when the interrogator, in the second game, not only asks you personal questions but more general ones. The best strategy for the woman will be to give the most competent answer she can manage rather than giving lots of ad hoc, deliberately stupid answers, for it will be much easier for a computer to simulate dumbness than human competence. It is, for example, child’s play to program a computer to simulate human like mistakes and slowness in mathematical questions but extraordinarily hard to display the talents humans characteristically deploy in ordinary conversation over a range of topics (it certainly hasn’t been done).

Lassègue also tries to suggest another way in which the first test and the second are suspiciously dis-analogous. He points out that if, in the first test, the man pulls off his imitation of a woman, this doesn’t in any way establish that he is a woman, while in the second test, passage by the computer is supposed to establish that the computer is, genuinely, a thinker or intelligent. But this is silly. Turing’s whole point is to distinguish physical traits from mental ones. As Turing emphasizes, "playing fair with the machine" is like blind refereeing or putting an opaque screen between musical performers and their judges. If the man in the first test passes, then it establishes that he can think like (or can cognitively/affectively pass as) a woman. Similarly, just as the man’s passage in the first test does not of course establish that he is a woman, the computer’s passage in the second test does not of course establish that it is a woman (or a human being). And again, in precise analogy, what the computer’s passage of the second test would establish is that it can think like (or can cognitively/affectively pass as) a woman (or presumably, a human being).3 Turing, no human chauvinist he, also makes it amply clear that passage would be a sufficient condition for being intelligent, not at all a necessary one. It might well be, he suggests, that a computer might be produced that really ought to be considered intelligent but would not be able to pass the test (rather like the character Data in TV’s Star Trek: The Next Generation, who is obviously more intelligent and more moral than most humans but doesn’t think/feel like a human in certain respects and is at times baffled by human behavior). It is perhaps also important to add that while Turing speaks of "playing fair with the machine," it would be absurd to think, as Lassègue seems to argue, that Turing identified himself with it or with "pure intelligence." In fact, as his biographer makes clear, Turing felt contempt for the airs put on by upper middle class Englishmen who prided themselves in their capacity for "abstract thought"; thinking machines would take them down a peg.

His perverse misreadings now allow Lassègue to go on, preposterously, to write,

Little by little, a kind of hierarchy in the players’ responses emerges from the text: the woman imitates herself (that is why she is a poor player), the man imitates the woman (he is therefore a better player than the woman). What about the machine which replaces the man in the second game? (p. 10)

Lassègue’s response to his wholly contrived and mis-premissed question is that the computer, when passing, really shows itself, and intelligence, to be male. To the contrary, and to insist on the literal reading of Turing that Lassègue officially demands (but hardly follows), what the computer has to do to pass, to prove it is a thinker, is to be cognitively indistinguishable from a human female. (I am reminded of one of the my daughter Casey’s first sentences, in her twenty-sixth month, "I am a person," which elicited a matronly "Of course you are, my dear" from the next table in a Cotswold restaurant. Baffled, and seeking to understand what this Cartesian announcement could possibly mean, I asked "Is Mommy a person?" and after some repetition eventually got a lisped "yesh." My further inquiry, "Is Daddy a person?" eventually elicited a "No.")

While I said I would not comment upon the rest of Lassègue’s paper, which reads like a manic parody of psychoanalysis, I would like to close by reminding the reader that Alan Turing, through his deciphering work during World War II, which effectively produced the digital electronic computer, has been said to have done more than anyone else to defeat Nazi Germany. Andrew Hodges, Turing’s biographer, gives a full account of this.4 In Brian Johnson’s The Secret War we also find a similar account along with an account of British attempts to confuse German bombers that were guided by radio to their targets at night. We also find there a haunting and premonitory echo of Turing’s imitation game.5

‘. . .They would say, "Das ist eine feind Stimme – an enemy voice, Don’t listen to it!" And we would reply, "Wir sind die rightige Stimme – We are the real voice." This would go on until the pilot became completely confused and didn’t know who was who.’

Though he presents the topic light-heartedly imitation was a serious business for Turing.6


Notes

1 Tekhnema 3, pp. 37-58.
2 A. M. Turing (1950), "Computing Machinery and Intelligence," Mind, vol. LIX, no. 236, pp.433-460.
3 Lest the reader still harbor some suspicion that Turing's assignment of the different roles to the sexes in the particular formulation of the test he gives in "Computing Machinery and Intelligence" might still be very important, I note that Turing had, in his "Intelligent Machinery" report, given a shorter description of the test in which a man takes the "human-thinker" role of the woman and is pitted against a computer program (with the same arrangement of rooms and communications). Turing submitted the "Intelligent Machinery" report to the National Physical Laboratory two years before publication of his Mind paper. See "Intelligent Machinery" in Machine Intelligence 5, Bernard Meltzer and Donald Michie, eds. (New York: American Elsevier Publishing Co., 1970), p. 23.
4 A. Hodges, Alan Turing: The Enigma (New York: Simon & Schuster, 1984). The title is not the one Lassègue gives, viz., "Alan Turing, the Enigma of Intelligence." Doubtless someone abler than I could extract enormous insights from this slip.
5 B. Johnson, The Secret War (London: Methuen, 1978), p. 144.
6 For amusing accounts of a related attack on Turing that fizzled, consult my An Invitation to Cognitive Science (Oxford: Blackwell 1991), p. 127-28. In Ian McEwan's The Imitation Game protagonist Cathy Raine, hoping to do her bit despite the granite assurance of her male chauvinist society that she is a decorative non-entity becomes one of the hundreds of Bletchley women who, in complete ignorance of what they were really doing, endlessly recorded and transcribed the radio signals that fed the fledgling electronic brains that "Turner" and other males in Hut Eight directed. By seducing Turner, Raines manages to learn what is really going on and as "the woman who knows" is locked up for the War's duration. In fact, however, the real Hut Eight contained a female mathematician, like Turing a graduate of Cambridge University, and Turing eventually proposed marriage to her, and the two retained a friendly relationship long after she withdrew from the engagement.