Devoir de Philosophie

Encyclopedia of Philosophy: Ajdukiewicz, Kazimierz

Publié le 11/01/2010

Extrait du document

Radical conventionalism is closely related to Ajdukiewicz’s theory of language and meaning. The meanings of

expressions in a language generate rules for accepting sentences of L. Ajdukiewicz singles out three kinds of

meaning-rules: axiomatic (they require the unconditional acceptance of certain sentences, for example ‘A is A’),

deductive (for example, B follows from ‘if A then B’ and A), and empirical (the sentence ‘snow is white’ is

asserted in a situation in which a person asserting this sentence perceives that snow is white).

« non-translatable.

The acceptance or rejection of sentences is always related to a definite language L.

If L is closed and connected, empirical situations do not force us either to accept or reject any sentence, because we canalways change our conceptual apparatus.

This is an essential strengthening of usual conventionalism.

For Poincaré and Duhem , we are free to change our theoretical principles, because they are hidden conventions.

For Ajdukiewicz, experiential reports are also closely related to conceptual apparatus, and since every conceptual apparatusproduces a world-perspective, we can say that theories and observational reports are accepted not absolutely but relative to world-perspectives.

This is why Ajdukiewicz called his conventionalism ‘radical', contrary to the moderate view of the Frenchmen (see Conventionalism ). 3 Semantic epistemology In the middle 1930s Ajdukiewicz rejected radical conventionalism, because he came to the view that his idea of connected and closed languages was a ‘paper fiction'.

The change was also strongly motivated by the work of Tarski which convinced many philosophers that semantics has important applications in philosophy.

When he was a radical conventionalist, Ajdukiewicz did not draw any ontological theses from his epistemological considerations; but his semantic epistemology is an attempt to bring together epistemology and ontology.

If we speak about the world, we use an object-language.

Since epistemology intends to say something about the world and our knowledge of it, an epistemologist must use a meta-language in order to capture knowledge and its object. Ajdukiewicz, employing metalogic and semantics, gave a rigorous analysis of Rickert's transcendental idealism and Berkeley's subjective idealism (see Berkeley, G. ).

For Ajdukiewicz, both kinds of idealism are incorrect, because they neglect basic results of metalogic and semantics.

Ajdukiewicz rejects Rickert's idealism, because truth, contrary to Rickert, cannot be established exclusively by purely deductive procedures; the incompleteness of arithmetic is an essential premise of Ajdukiewicz's argument.

Berkeley's thesis that ordinary objects are complexes of our ideas is rejected, because it conflates syntax and semantics.

According to Ajdukiewicz, Berkeley uses a language which is very similar to the language of syntax and offers a syntactic-like definition of existence. However, since existence is basically a semantic concept and semantics is not fully definable in syntax, Berkeley's argument fails.

Thus, semantic epistemology leads to a realist account of existence. 4 Philosophy of science In addition to his discussions of radical conventionalism, which implies that there is no absolute gap between theories and experiential reports, Ajdukiewicz also worked on concrete problems in the philosophy of science.

In particular, he was interested in the logic of fallible inferences.

His approach was based on concepts borrowed from decision theory.

In general, acceptances (rejections) of sentences are actions which are associated with profitsand losses.

Assume that A is a sentence to be accepted and that Z is the minimal acceptable profit for the agent, when A is true, and S is the minimal acceptable loss, when A is false.

According to Ajdukiewicz the ratio S=(S + Z) expresses the degree of certainty which an agent accepting A can ascribe to this sentence.

This relates degrees of certainty to mathematical probabilities.

Having this framework, Ajdukiewicz tried to establish the degree of conclusiveness of a fallible scheme of inference.

Assume that K is background knowledge.

We are interested in the. »

↓↓↓ APERÇU DU DOCUMENT ↓↓↓

Liens utiles