COPElAND | 271
Judge: Yet Christmas is a winter’s day, and I do not think Mr Pickwick would mind the
comparison.
Machine: I don’t think you’re serious. By a winter’s day one means a typical winter’s day,
rather than a special one like Christmas.Scoring the test
How good does the computer’s performance have to be in order for it to pass the test? How
many judges must mistake the computer for a human? Obviously these questions are crucial
when specifying the Turing test, because without answers it’s impossible to say whether a com-
puter has passed or not. Turing gave a very ingenious answer. He said that in order to determine
the answers to these questions, we first need to carry out another set of tests.
These prior tests involve another version of the imitation game. This time none of the players
is a computer. There is a judge and two other human players, one male (A) and one female (B).
The judge’s job is to decide (by chatting) which player is the man, and A’s aim is to trick the
judge into making the wrong choice. Turing uses this man-imitates-woman game as a standard
by which to adjudicate whether the computer passes the computer-imitates-human version of
the imitation game (in which, as Turing said, ‘a machine takes the part of A’).^30 To assess the
computer’s performance, we ask:^31
Will the interrogator decide wrongly as often when the [computer-imitates-human] game is
played . . . as he does when the game is played between a man and a woman?
In other words, if the computer does no worse in the computer-imitates-human game than the
man does in the man-imitates-woman game, then the computer passes the test.
Turing was a little cagey about what would actually be shown if the computer passed his test.
He said that the question ‘Can machines pass the test?’ is ‘not the same as “Do machines think”
’, continuing ‘but it seems near enough for our present purpose, and raises much the same
difficulties’.^32
In one of his philosophical papers, he even cast doubt on the meaningfulness of the question
‘Can machines think?’, saying (rather rashly) that the question is ‘too meaningless to deserve
discussion’.^33 However, he himself indulged in such discussion with gusto. In fact he spoke very
positively about the project of ‘programming a machine to think’ (his words), saying ‘The whole
thinking process is still rather mysterious to us, but I believe that the attempt to make a thinking
machine will help us greatly in finding out how we think ourselves’.^34
what Turing predicted
Turing was also cagey about how long he thought it would be before a computer passed the test.
He predicted (in 1950) that progress in AI during the twentieth century would be sufficiently
rapid that:^35
in about fifty years time it will be possible to programme computers . . . 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.