DISCUSSION 
need help. Whom do you see coming out of church but old 
ladies whose husbands have just died? 
Trowell: One of the real problems which I have seen in this 
rapidly developing technological and scientific age, speaking as 
someone who has taught medicine for some 30 years, is that we 
soon get out of date, yet we remain in power. ‘This I think presents 
the world with quite a new facet of human progress. 
I would also like to add, speaking now as a parson, how much 
I have valued the views of some of our younger colleagues: I 
think we may begin to see that perhaps there are certain areas 
of life which science cannot explore and certain value judg- 
ments which need the philosopher, or even perhaps the leaders 
of religion. Certainly my village congregation is not composed 
largely of bereaved widows! 
Lederberg: I donot think the problem of MacKay’s unbiased 
computer is soluble, but I do not think it is important either. It 
is impossible to say which is the truer judgment of what the 
prediction ought to have been, but the fact of the situation is 
the final result. There are innumerable variables and one of 
these is the prediction that was made during the course of the 
election. Some people, like MacKay, say that we biased the 
prediction and we ought to have a counter-counter-suggestion 
for it. I think that it is simply an unavoidable feature of social 
life that there are innumerable contributions to the ultimate 
outcome of democratic decisions. 
I should like to put another question to MacKay without, I 
hope, obscuring my very general concurrence with his discus- 
sion. I feel rather suspicious of one kind of machine that says 
to another: ‘“‘Well, you’re doing very well, but I also have 
judgment and my decisions are wiser than yours. I am unable 
to specify to you the processes by which I reached these con- 
clusions, but | insist that they are nevertheless right”’. What is 
the validity ofan unjustifiable judgment ? Agreed, it may often be 
uneconomical to search through all the processes in an attempt 
to anticipate the final judgment. For example, in order to 
anticipate all the possible outcomes of an automated experiment 
to detect life on Mars, I would have to read all that I knew about 
178 
