TODAY sociologists file away their facts and findings in
unlabelled cabinets which have no alphabetical index. The
cabinets are also widely dispersed and no single socio-
logist is in a position to be familiar with the location of all of
them. Having largely inherited the filing cabinets from social
sciences which had been established in business a good deal
earlier than sociology, sociology was loath to discard them and
is still clinging to them in a pennywise fashion.

Alas this clinging is a virtue of thrifty beginnings and not of
epoch-making enterprises. The present book certainly does not
claim to be epoch-making. If anything it is a gamble with the
accumulated small change of modern psychological knowledge
in the hope of winning some of the coveted prizes of sociology
and perhaps even of moral philosophy. Yet should my book be
branded as a rash enterprise it may still have been well worth
the adventure of writing it, for, as it is often observed, even
those contribute to the exploration of a maze who come back
only to report the itinerary of their failures. Furthermore it
seems to me that, even if my stake proves unwise, it will be a
gamble of this sort which will eventually rescue social science
from its present predicament.

This book deals centrally with the concepts of 'normal' and
'abnormal' in human personality and behaviour. It seeks to
establish whether one or the other of these can be defined and
thereby an absolute measure of man educed. The string of essays
which follows opens with a discourse on the first principles upon
which the subsequent chapters depend. Here I shall rely on metap-
physical, biological, and psychological information. Following
this I develop my arguments to a stage when the conditions of
defining the human 'norm' or 'abnorm' are within our view.
Thereupon the conclusion I shall reach is that, even though
'normality' is indefinable, the minimum area of
'abnormality' is
already a subject of substantial agreement. Yet if my definition

-v-