THE LOGICAL STATUS OF MENTAL FACTORS 117

That difficulties, and even fallacies, may be involved in
applying simple arithmetic to living personalities, I readily
admit. But the objections as stated rest on a very narrow
notion of what mental measurement entails; and, since none
of the writers explicitly enunciates the presumable require-
ments, I shall attempt to set them forth here.

To measure is " to estimate (an immaterial thing, person's
character, etc.) by some standard or rule." l If * etc.' may
be held to cover anything that belongs to a definable class
and possesses the necessary properties, whether or not it is
* immaterial/ then this definition will serve for measurement
in physical science as well as in psychological. Every
measurement calls for an experiment. Nevertheless,
measuring as such even in physical science is a mental pro-
cess not a physical: it consists, not in adding the weights,
but in reading the scale where the pointer points. Thus,
contrary to the implications of the argument just quoted,
the mere circumstance that instruments have to be used for
the more exact forms of measurement does not turn it into
a physical process, any more than the use of a telescope
turns observation into a physical process. " C'est toujours
avec nos sens que nous nous servons de nos instru-

Psychologie, I, pp. 260 f. The reader may also refer to the recent Report
of the British Association Committee on c Quantitative Estimates of Sensory
Events' (Dundee Meeting, 1939) and the admirable Presidential Address of
Dr. R. J. Bartlett at the same meeting. I should add that the above was in
type before I had an opportunity of reading either of these suggestive papers,
and must therefore confine my comments to footnotes.

1 Oxford Dictionary, s.v. The physicist adopts a slightly narrower
definition. " Measurement is the process of assigning numbers to represent
qualities: the object of measurement is to enable the powerful weapon of
mathematical analysis to be applied to the subject-matter of a science "
(Campbell, Physics : The Elements, p. 267). I have a mild preference for the
broader definition given by the dictionary, first, because I do not hold that
mathematical analysis is limited to numbers; and, secondly, because I hold
that the assignment of non-numerical marks (e.g. the ' alpha minus,' * beta
plus,' etc., of the Arts examiner) is also a form of measurement. The fact
that his method is inexact does not prevent it from being a form of measure-
ment, though the fact that it is usually very inexact makes it usually a bad
form of measurement. But " in the sense understood by the man in the
street, exactness has almost disappeared from the subject-matter of modern
science " (Jeffreys, loc. cit., p. 214).