[ Pobierz całość w formacie PDF ] a branch of human knowledge to which the science of statistics is not
closely allied, and for the correct understanding, of which the scientific
The Scope and Method of Political Economy/151
marshalling of figures, and observation of aggregate facts, is not more
or less necessary. That the laws deduced from them fall into the ranks of
the branches of knowledge to which they belong when they are fairly
established does not, in my humble judgment, invalidate the scientific
claim of the agency to which they owe their existence. 216
The concluding portion of the above statement would probably meet
with universal acceptance. A method or an agency may, however, be
scientific without thereby becoming itself a science. Statistics, or statis-
tical method, as understood by Dr Mouat, is a very important means
whereby human knowledge is extended; but as such it is to be regarded
as a scientific instrument, rather than as an independent body of doc-
trine constituting a distinct science.
It is, indeed, necessary to recognise a theory of statistics, dealing
with what may be called the technique of the statistical method, that is
to say, the conditions that statistical data must fulfil, the modes in which
they are to be ascertained and collected, the manner of their arrange-
ment and employment for purposes of reasoning, the criteria determin-
ing the validity of arguments based upon them, and the logical character
of the conclusions established by their aid. But all this is really anteced-
ent to the actual use of statistics for any particular purpose. The whole
discussion constitutes, not a separate science, but a special branch or
department of inductive logic or methodology that is. of the science or
art which treats of scientific method in general.
It is, however, in quite a different sense from the above that the
existence of an independent science of statistics is affirmed by the ma-
jority of Continental statisticians, and also by some English writers.217
Statistical science is regarded, not as an abstract science of methods
dealing with phenomena of very various kinds under a distinctive as-
pect, but as a concrete science with a distinctive subject-matter. A dis-
tinction is clearly drawn between statistics as a method and statistics as
a science. It is recognized that the method has a very wide application;
but the science is described as studying exclusively man s social life.
As thus interpreted, statistical science becomes practically equiva-
lent to sociology, with the implication that the sole means whereby so-
ciological, including economic, knowledge can be attained is the sys-
tematic collection and inductive interpretation of social phenomena. There
is the further implication that the data are mainly, if not exclusively,
numerical.
Dr Mayr, taking this view, defines the science of statistics as the
152/John Neville Keynes
systematic statement and explanation of actual events, and of the laws
of man s social life that may be deduced from these, on the basis of the
quantitative observation of aggregates. 218
If it is asked why the quantitative observation of social aggregates
should constitute a distinct science, while no similar claim is made in
regard to the observation of purely physical aggregates, the reply given
is that in the determination of the laws of social life statistical enquiry is
the only possible mode of investigation, and not as in the case of the
physical sciences a merely secondary or supplementary method. So-
cial science and political economy are spoken of as branches or depart-
ments of the science of statistics, a science which studies social and
economic phenomena in the only satisfactory way, namely, by the accu-
mulation of facts and generalization from them. It will be observed that
the doctrine here set forth is even narrower than that which regards
induction as the sole valid method of economic enquiry. For we are now
limited to quantitative induction; qualitative induction, whether histori-
cal or comparative, is out of place as well as the deductive method.
Our grounds for rejecting this view have been given in a previous
chapter, and to pursue the discussion here would merely carry us back
to a class of considerations that have been already sufficiently insisted
upon. It may, however, be added that for the general science of society
we have at any rate another name sociology or social science which
does not beg the question as to method, and which is free from the am-
biguity that at best must attach to the term statistics. For it has to be
allowed that by this term is also meant a method of analysis having an
indefinitely wide range of application outside the science of man in so-
ciety. Thus we speak of moral and intellectual statistics, of vital and
medical statistics, of astronomical and meteorological statistics, of physi-
cal and physiological statistics, as well as of economic and political
statistics. It can hardly be said that there is any concrete department of
enquiry, in which statistics as a method may not find a place; and it is
upon this character of universality that the claim of statistical research
to public recognition and encouragement is frequently based.
If a less extreme view than that above described is taken, and statis-
tics is considered to be a distinct science, but nevertheless not to include
the whole of social science or of economics, then it becomes a part
which is only differentiated from the remainder by the employment of a
particular method. Professor R. Mayo-Smith explicitly recognizes sta-
tistical science as a branch of social science employing a specific method,
The Scope and Method of Political Economy/153
and devoting itself to those problems of life in society which can best be
solved by that method. 219 It seems, however, both unusual and unde-
sirable to differentiate sciences by their method as distinguished from
their subject-matter. We might equally well identify other scientific meth-
ods with those particular sciences in the development of which they
happen to be of special importance. At any rate the question now be-
comes little more than a verbal one. The need be no fundamental dis-
agreement between those who take the view just indicated, and those
[ Pobierz całość w formacie PDF ]zanotowane.pldoc.pisz.plpdf.pisz.plkajaszek.htw.pl
