his system; and these papers, which include several systematic articles
written before the book's publication, should enable the reader to
glimpse, with Tolman, the nature of the changes which take place
in a systematic purposive behaviorism as it is modified to encompass
the data of an expanding field.

This collection of papers is not only as a significant scientific
publication, but also a revealing human document. Tolman is one
of those rare beings among system builders who has a sense of humor
about himself and his theorizing. System building for him is not a
grim business. It is a happy, gay, creative activity, and his papers
express all of this to the full. No matter what the subject, how ab-
stract the treatment, his wit, humor, magnanimity, and tolerance
are written into each analysis. He is constitutionally incapable of
writing dogmatically or of publishing a polemic. The papers in the
present book not only trace explicitly the history of significant
ideas but also portray a person with grace.

Tolman's system is characterized by two major attributes: the
first of these is the breadth and all-inclusiveness of his psychology.
Above all else, he has insisted that behavior is multidetermined
and that an adequate system must encompass all psychological data.
He believes that a theorist's job is to try to describe and account
for the entire field which lies within his discipline, and not to
restrict it arbitrarily to the more amenable areas. Tolman has never
been interested in writing a "small scientific theoretical system" --
he has always sought the complete formulation. This has meant the
witting rejection of attempting finalistic formulation at this stage
of the science, and the characteristically cheerful acceptance of a
programmatic role.

The second major characteristic of Tolman's system building -- his
use of intervening variables and other hypothetical constructs -- has
been decisive in forcing theorizing in psychology out of periph-
eralism and into centralism. The genius of his theorizing lies in his
clear understanding that attempts to deal with correlations between
stimulus field and resulting behavior can be understood only in
terms of postulated intervening variables, hypothetical constructs,
dynamic central processes. The problem of science, for him, is not
that of seeking correlations but of seeking the systematic meaning
of the observed correlations.

-vi-