 |
"There are two form of empiricism,
radical and moderate. Characteristic of radical empiricism is the
thesis according to which only empirical sentences may have scientific status.
Moderate empiricism, on the other hand, grants such status not only
to empirical sentences but also to sentences which are valid in virtue of
meaning of expressions occurring in those sentences. [...] The laws of logic
are classfied by moderate empiricism as statements of just this type, i.e.
as statements whose truth is ensured by the meaning of expressions alone"
("Logic and Experience", 1947, in K. Ajdukiewicz, The Scientific
World-Perspective and other Essays (1931-1963), ed. by J. Giedymin,
Reidel, Dordrecht 1978, p. 165).
"Against the principal thesis
of radical empiricism holding that only assertions directly or indirectly
based on experiences have a right to be admitted in science, the following
argument may be put forward. Radical empiricism seems to overlook the fact
that every sentence and therefore also every perceptual sentence is formulated
in a certain language. But every language is bound by certain rules concerning
the acceptnce of sentences. These rules are called by some authors rules
of transformation, rules of language or rules of meaning [see meaning-rules]"
(Id., pp. 172-3).
"The doctrine according
to which apart from empirical also analytical and only analytical sentences
have the right to a place in science might be called moderate empiricism"
(Id., p. 174).
"[...] the view point adopted
by radical empiricists may be also regarded as a programme for doing science.
In that case one must not expect them to prove the truth of their view since
programmes are neither true nor false, they can be only reasonable or unreasonable.
In order to be reasonable they must be purposeful and practicable. The aim
of this paper was to find out whether in principle the programme of radical
empiricism can be put into effect. The result of our considerations
is positive. Whether this programme also serves a purpose can only be shown
by actual scientific practice. However, the actual course of science up
till now does not seem to be consistent with the programme of radical empiricism"
(Id., pp. 180-181).
"In other words, without
prejudging the question of the methodological staus of the axioms of logic
in actual science, I came to the conclusion that there is nothing to prevent
one from constructing science in a language in which the axioms of logic
are not dictated by the rules of language, provided there are in it deductives
rules distinguishing certain forms of inference to be observed on pain of
violating the meaning of expressions. If science were formulated in such
language then the axioms of logic could be regarded as hypotheses indirectly
based on experience through the empirical testing of their consequences
derived in virtue of the deductive rules of the languages. [...]
In order to free scientific knowledge completely from the choice of language,
it was necessary to abandon deductive rules as well. This would mean complete
abandonment of that conception of language which I have recalled here. In
the absence of deductive rules as meaning-rules,
the procedure of testing hypotheses would contain (apart from statements
tested) as an additional element, to be accepted or rejected, the rule of
inferences used in deriving consequences. If conflict arises between the
statement of the theory and direct experience, constistency might be restored
not only by modifying the statements of the theory but also by changing
the rules of inferences and yet without transition to a different language,
i.e. without the change of meaning of expressions"
("The Problem of Empiricism and the Concept of Meaning", 1964,
in op. cit., pp. 316-317).
