Ti6 THE FACTORS OF THE MIND

rent controversies, it is particularly desirable not to shirk
this issue. I shall, therefore, endeavour to state what this
transformation from a qualitative classification to a quantita-
tive grading seems to presuppose. " Attributes," says
Stephenson, " cannot be converted into variables in the
twinkling of an eye." " To calculate a correlation coeffi-
cient," says Thomson, " is to assume that the marks [cor-
related] are in some sense commensurable." And both
writers regard the requirements of valid measurement or
marking as fatal to certain proposals, recently made, for
applying factor-analysis to new types of psychological
material, and in particular to certain modes of correlating
persons. Representatives of the c intuitionist * school go
further ; they maintain that every form of factor-analysis
is vitiated from the start, since the very attempt to apply
quantitative measurement to' living personalities' " implies
views that are not merely false but meaningless." " No one
can safely assign a figure to any mental quality or mental
product."

Their arguments, couched more or less in their own words,
may be summarized as follows. " Lengths, weights, and
times," they say, " can be legitimately measured in scientific
units; they are unidimensional variables. The factorist
proposes to measure traits and persons in the same way.
But the variables he has to factorize do not differ simply in
numerical magnitude. They are qualities, not a set of
quantities: we cannot add units of intelligence as we add
twelve inches to make a foot; nor can we put samples of
behaviour into a balance, and subtract and multiply them
as we subtract and multiply ounces and pounds. It fol-
lows, therefore, that to take over the mathematical methods
of the physical sciences and use them for the purposes of
psychological description and analysis will commit us to a
fundamental flaw." l

1 Cf. * Measurement versus Intuition in Applied Psychology,' Character
and Personality, VI, pp. 114-31 and refs. In general psychology (as distinct
from applied) the dispute over * measurement versus intuition' is at least as
old as Leibniz, who maintained that, since mental phenomena could not be
represented as continuous variables, any mathematical treatment of psycho-
logical problems was doomed to fail; cf. Miinsterberg, Grundzilge d.