Le progrès d’une science comme la physique détermine dans la logique un mouvement correspondant ; elle aussi est en devenir. Dans le monde atomique en particulier, les relations d’incertitude d’Heisenberg et l’existence de valeurs quantifiées conduisent à la construction d’une logique mieux adaptée à ce genre de recherches, logique trivalente, et de genre doux, c’est-à-dire dans laquelle les couples de propositions peuvent être soit composables, soit incomposables, par rapport à l’opération produit.

SummaryThe following two questions are examined: 1o Do the unobservable parameters possess precise, though unknown, values ? 2o If these unobservable values were known, would it be possible to make precise predictions of the reults of later measurements ?The answer is shown to be negative; the questions, therefore, are not meaningless, being capable of a falsification. The inquiry leads to the establishment of a principle of anomaly, more precisely speaking, of causal anomaly, which is to be added to Heisenberg's principle (...) of indeterminacy. This principle states that the principle of action by contact is violated whenever definite values are assigned to the unobserved quantities, i. e., when an exhaustive interpretation of quantum mechanics is used. The two most important exhaustive interpretations are given by the corpuscle and the wave interpretation; each leads to causal anomalies, though for different places.The causal anomalies can be eliminated by the use of a restrictive interpretation, which separates statements about unobserved quantities, as a third propositional class, from true or false statements. Bohr and Heisenberg have called such statements meaningless, without being able to eliminate them completely; these statements arte thus merely shifted into the metalanguage. In another version of the restrictive interpretation such statements are assigned a third truth value, the value indeterminate, and quantum mechanics is then presented in the form of a three‐valued logic.The inquiry is carried through without the presupposition of any philosophical conception; every interpretation is examined with respect to the consequences to which it leads. ‐ H. R.'. (shrink)

It, is shown that Birkhoff –von Neumann quantum logic (i.e., an orthomodular lattice or poset) possessing an ordering set of probability measures S can be isomorphically represented as a family of fuzzy subsets of S or, equivalently, as a family of propositional functions with arguments ranging over S and belonging to the domain of infinite-valued Łukasiewicz logic. This representation endows BvN quantum logic with a new pair of partially defined binary operations, different from the order-theoretic ones: Łukasiewicz intersection and union (...) of fuzzy sets in the first case and Łukasiewicz conjunction and disjunction in the second. Relations between old and new operations are studied and it is shown that although they coincide whenever new operations are defined, they are not identical in general. The hypothesis that quantum-logical conjunction and disjunction should be represented by Łukasiewicz operations, not by order-theoretic join and meet is formulated and some of its possible consequences are considered. (shrink)