2. Inductivism.
The second positive starting point
of behaviorism was the view about
scientific method
which Watson shared with the introspectionists whom
he attacked. Wundt and Titchener, the giants of the
introspective school,
had been vehement in their am-
bition to base
psychology on properly controlled ex-
periments. The general appeal to look into oneself in
order to
decide upon psychological questions, which
one can find, for instance, in
the controversies between
Locke, Berkeley, and Hume, was not good
enough.
Introspective observers had to be carefully trained.
Moreover
Titchener argued that they had to be trained
to distinguish pure experience
as “existences” from the
“meaning” which it has for men in their ordinary
lives.
Unless this could be done psychology would never
arrive at any
pure data on which a science of mind
could be created.
This presupposed a certain view of scientific method,
dating back to Bacon,
which is often called “induc-
tivism” or “observationalism” (Popper,
1962). The
leading idea of this conception can be summed up in
Titchener's own words: “We are agreed, I suppose, that
scientific method may be summed up in the single word
'observation'; the
only way to work in science is to
observe those phenomena which form the
subject-
matter of
science” (Titchener [1908], p. 175). Watson
himself had
basically the same conception of scientific
method. To quote him:
“You will find, then, the behav-
iorist working like any other scientist. His sole object
is to
gather facts about behavior—verify his data—
subject
them both to logic and mathematics (the tool
of every
scientist)” (Watson [1924], p. 7). Watson's
basic objection to
introspectionism was that it was an
attempt to form a science on very
unreliable data about
which experimenters could reach no agreement,
and
which purported to reveal facts about a nonexistent
subject
matter, namely consciousness. If psychologists
were to start from data
provided by rats in mazes they
would at least have a chance of developing a
science
on the basis of publicly observable data.
When it came to making generalizations Watson
again showed his inductivist
allegiance; for the Baco-
nian view was that
generalizations should never go
beyond the data. They should simply record
correla-
tions between observables.
Thus Watson was as unin-
terested in
physiological speculation about intervening
processes as he was hostile to
any recourse to unob-
servable mental
entities or processes to explain what
could be observed. One should, of
course, quantify the
data if possible and operate mathematically on it. But
this was
merely a way of arriving at correlations.
