120 THE FACTORS OF THE MIND

sets out to show that certain items or certain individuals
obey the three postulates just stated, he finds several
troublesome obstacles and entanglements appearing in his
path. To begin with, mental qualities, as named and
described in popular parlance, are highly complex. Con-
sequently, when we try to classify or grade things according
to mental qualities, the experiment may break down at the
outset. Widely different results are reported by different
observers, and even by the same observer on different
occasions. Hence the hasty conclusion has often been
drawn that mental phenomena by their very nature are not
amenable to measurement.

For example, in laying down the principles of scientific measure-
ment, Campbell begins by announcing that, although we can
measure the number, weight, density, and possibly hardness of a
group of physical objects, we cannot measure their colour or their
beauty.1 Let us consider his two examples, since they are precisely
the things the factorist would like to measure. Let us take colour
first, and attempt the necessary experiment to see how far they con-
form to our three relational postulates. All observers will agree
that crimson lake (#) and gamboge (y) are both less blue than ultra-
marine (z). Accordingly, let us compare the first two colours in
terms of the same relation. At once the difficulties begin. One
observer declares that crimson lake is less blue than gamboge,
because " gamboge has a touch of green in it, and therefore of
blue " ; another, that the gamboge is less blue than crimson lake
because crimson lake " contains a little purple " ; a third, that the
two are " interchangeable, at least as far as blueness is concerned."
The vast majority insist that the question put is meaningless.
Must we then conclude with Campbell that " there is no natural order
of the colours," and that " the assignment of numerals to colours is
arbitrary, because it is not dictated by judgements which form part
of the subject-matter of science and for which universal and imper-
sonal assent can be obtained ? "

If so, we shall hardly think it worth while to collect similar
observations in regard to relative beauty. We shall be inclined to
acquiesce forthwith in Campbell's view that any attempt to measure
beauty would be " fantastic." " No agreement," he asserts, " can
be obtained for judgements concerning it," The end of the argu-
ment is easy to foresee. Only judgements regarding length,

1 Loc* cit., p. 268.