Autor: Lonergan, Bernard J.F.

Buch: A Third Collection

Titel: A Third Collection

Stichwort: Von der Logik zur Methode; Verifikation ist kein Beweise -> Fehlschluss;

Kurzinhalt: FROM LOGIC TO METHOD; Verification is not proof; to affirm the consequent of an hypothesis, settles nothing about the truth of the antecedent

Textausschnitt: 34/9 The Aristotelian hegemony had been broken, but Aristotelian notions not directly challenged by the new science lived on in quiet possession of the field of common assumptions. Among them was the view that science consisted in true and certain knowledge of causal necessity. Indeed, Newton's deduction of the orbits of the moon and of the planets was regarded as a stunning confirmation ()

35/9 But the logic of the matter is simple. Verification is not proof. For verification is an affirmation of what follows from scientific hypothesis, theory, system. But to affirm the consequent of an hypothesis, settles nothing about the truth of the antecedent from which the consequent follows. A logical conclusion is to be had only when the attempt to verify turns up contrary instances; for then one denies the consequent and from that denial there follows the denial of the antecedent. Accordingly, the principles and laws of an empirical science, no matter how frequently they are verified, may be esteemed ever more probable but may not be considered to be definitively established. (138; Fs) (notabene)

36/9 Moreover, the progress of modern science points in the same direction. Newton was acclaimed because he was considered to have done for mechanics what Euclid had done for geometry. But in the nineteenth century it became clear that Euclidean geometry could no longer be considered the one and only possible geometry. In the twentieth the repeated verification of Einstein's special relativity made it probable that a non-Euclidean geometry was the appropriate conceptualization in physics. (138; Fs) (notabene)

37/9 Similarly, Laplace's determinism was found to have shaky foundations. For Heisenberg's relations of indeterminacy (or uncertainty) reveal a knowledge that is not less but greater than the knowledge offered by classical laws. Formerly, indeed, probability was thought to be no more than a cloak for our ignorance. But now the tables are turned. For classical laws hold only under the blanket proviso, other things being equal. So it is that classical predictions can be notably mistaken because they fail to foresee the interference of some alien factor. But, further, the verification of classical laws is never exact: no more is demanded than that actual measurements fall within the limits set by a theory of probable errors of observation. In brief, classical theory consists of two parts: there is the classical law, and it sets an ideal norm from which actual measurements do not diverge systematically; there is the theory of measurement and it sets the limits within which errors of observation may be considered probable. But, as Patrick Heelan has pointed out, the same two aspects are contained within the single formalism proposed by quantum mechanics. For the single formalism admits two interpretations: one interpretation yields an ideal norm from which actual measurements do not diverge systematically; the other interpretation of the same formalism informs us of the distribution of the divergence from the norm. (138f; Fs)

39/9 There has occurred, then, a transition from logic to method. It has occurred in the field of natural science. It does not, by any means, involve an elimination of logic: for it still is logic that cares for the clarity of terms, the coherence of propositions, the rigor of inferences. But it does involve a shift in the significance of logic. For Aristotle in his Posterior Analytics made his demonstrative syllogism the central piece in his construction both of the nature of science and of the relations between sciences. That construction has turned out to be a procrustean bed on which science cannot lie. So far from providing the key to the whole nature of science, logic has to be content with the task of promoting clarity, coherence, and rigor in the formulation and application of hypotheses and theories. Further, while it is essential that this task be properly performed, still the significance of that performance is measured not by logic itself but by method. For an empirical science is not confined to logical operations with respect to terms, propositions, inferences. It includes observation, description, the formulation of problems, discovery, processes of experimentation, verification, revision. Within the larger whole logic ensures the clarity of terms, the coherence of propositions, the rigor of inferences. And the more successfully it performs that task, the more readily will there come to light not the definitive immutability but the defects of current views and the need to seek more probable opinions. (139f; Fs)