第16章 「非論証的推論」n.1 - トリニティで5年間の講師職
私は、非論証的推論の問題が、予期していたよりもはるかに広い問題であり、はるかに興味深い問題であることに気づいた。そして、それが、大多数の探求において、不当にも帰納法の研究に局限されて来ていることを発見した。 (そうして)私は、帰納的論証(inductive arguments)は、もし常識の範囲内に留めないと、真なる結論よりも偽なる結論に導くことの方がずっと多いという結論に達した。 そして常識によって課せられる制限は、それを感ずることは容易であるが、それをはっきり定式化することは非常に困難である。結局、私は、科学的推論は非論証的な論理(学)以外の原理を必要とするが、帰納法はそういう原理のひとつではない、という結論に達した。 帰納法はもちろんその果たすべき役割をもっているが、それはひとつの前提としてではない。しばらくしてからく(本書で)私はこの問題に戻ることにしよう。 |
Chapter 16: Non-Demonstrative Inference , n.1I returned to England in June 1944 , after three weeks on the Atlantic. Trinity had awarded me a five-years lectureship and I chose as the subject of my annual course, ‘Non-Demonstrative Inference', or N-D.I. for short. I had become increasingly aware of the very limited scope of deductive inference as practised in logic and pure mathematics. I realized that all the inferences used both in common sense and in science are of a different sort from those in deductive logic, and are such that, when the premisses are true and the reasoning correct, the conclusion is only probable. During the first six months after my return from America I had rooms in College and enjoyed a feeling of peacefulness in spite of V1's and V2's. I set to work to investigate probability and the kind of inference which confers probability. I found the subject at first somewhat bewildering as there was a tangle of different problems and each thread had to be separated from every other. The positive outcome appeared in Human Knowledge, but I did not, in that book, mention the various perplexities and tentative hypotheses through which I had arrived at my final conclusions. I now think this was a mistake, as it made the conclusions appear more slap-dash and less solid than, in fact, they were.I found the subject of non-demonstrative inference much larger and much more interesting than I had expected. I found that it had in most discussions been unduly confined to the investigation of induction. I came to the conclusion that inductive arguments, unless they are confined within the limits of common sense, will lead to false conclusions much more often than to true ones. The limitations imposed by common sense are easy to feel but very difficult to formulate. In the end, I came to the conclusion that, although scientific inference needs indemonstrable extra-logical principles, induction is not one of them. It has a part to play, but not as a premiss. I shall return to this subject presently. |