Full Idea
Were every probable inference less certain than its premises, science, which piles inference upon inference, often quite deeply, would soon be in a bad way.
Gist of Idea
If each inference slightly reduced our certainty, science would soon be in trouble
Source
Charles Sanders Peirce (Reasoning and the Logic of Things [1898], IV)
Book Reference
Peirce,Charles Sanders: 'Reasoning and the Logic of Things', ed/tr. Ketner,K.L. [Harvard 1992], p.165
A Reaction
This seems to endorse Aristotle's picture of demonstration about scientific and practical things as being a form of precise logic, rather than progressive probabilities. Our generalisations may be more certain than the particulars they rely on.