Evidence; The Basis of Knowledge

  • Joseph C. Pitt
Part of the The University of Western Ontario Series in Philosophy of Science book series (WONS, volume 50)


There are fundamentally two types of evidential situations: (1) situations in which data is used to formulate novel hypotheses and which constitute the basis for the articulation of new theories, and (2) situations in which data is used to justify the acceptance of a hypothesis or theory. These two types fall on either side of a traditional philosophical distinction between the contexts of discovery and justification.70 More generally, the two types of evidential situations involve different types of reasoning. Thus, in the context of discovery, where one is trying to generate new Information, inductive or ampliative reasoning is tolerated and sometimes even sanctioned, whereas in the contexts of justification, the ideal is deductive reasoning. The goal in this second case is to establish as solid a link as possible between evidence and hypothesis.


Chapter Versus Rival Theory Ordinary Experience Mathematical Demonstration Aristotelian Account 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 70.
    cf. Reichenbach [1951, p. 231] and Popper [1959, p. 31].Google Scholar
  2. 71.
    For some this move may seem unwarranted since Galileo only reduces the two notions of “things” to one single kind and allows the distinction between different types of motion to remain. He does not abandon the Aristotelian notion of uniform circular motion as the only kind appropriate to celestial activity. On the other hand, he does insist on the move through which all objects are to be considered in terms of terrestrial analogues.Google Scholar
  3. 72.
    cf. Kuhn [1958] for his account of this conceptualization of pre-Copernican cosmology.Google Scholar
  4. 73.
    See Grant [1984] for an interesting discussion of the variety of views within the Aristotelian camp on the necessity of maintaining the distinction between the two domains. His argument also sup-ports my earlier objections to categorizing individual philosophers by schools. See also Schmitt [1983] for further support of this idea. Schmitt argues against the attempt to characterize medieval Aristotelianism as monolithic.Google Scholar
  5. 74.
    In some cases it is not clear that Galileo needed the telescopic evidence over which he made such a fuss. One particular example of this concerns the phases of Venus. Roger Ariew [1987] has shown that Galileo did not need the telescopic findings since he had the geometry to show why Venus ought to exhibit phases and the theory that dictated such observations can be traced back to Albert of Saxony. On the other hand, what Galileo needed to prove his case and what he required from a propaganda point of view ought not to be confused.Google Scholar
  6. 75.
    See Quine [1960, Chapter 2] on the Gavagai example.Google Scholar
  7. 76.
    This requirement clearly raises a host of issues traditionally asso-ciated with the problem of incommensurability. More important, it seems to ignore them. But this need not be troublesome. What is going on here is rather straightforward. Alleged Kuhnian paradigm shifts notwithstanding, the terminology of older theories does not disappear when new rival theories take over. Their language remains part of the baggage of science. It is only by a complicated process that it either disappears or is assimilated. In the course of this process terms are redefined, problems are recast, some issues come to be seen as red herrings, etc. None of this argues against the philosophical point of incommensurability. Rather, in its place I am suggesting we acknowledge the historical process and make allow-ance for it when considering general evidential questions. History is not as authoritarian as some philosophical considerations would have us believe.Google Scholar
  8. 77.
    Galileo might have gone on to add “quintessence.” For most of the arguments about the crystalline substance are designed to show that while there is a word for it, one cannot make conceptual sense of what it is supposed to be; cf. Drake [1967, pp. 68–69]; Opere (VII, 93–94).Google Scholar
  9. 78.
    This need not be taken as an offensive sexist comment on Galileo’s part. But if it is, there is no reason why Galileo should be con-demned, or the offending language eliminated. His comments reflect the society in which he lived. If we now see some of the assumptions of that society and culture as offensive, then we should take steps to ensure that we do not repeat their errors. But that does not mean that we should ignore or attempt to cover up those past mistakes.Google Scholar
  10. 79.
    See Drake [1976] for evidence of Galileo’s early attacks on official Aristotelian philosophy.Google Scholar
  11. 80.
    For a discussion of the controversy over Galileo’s experiments see Chapter 4, Footnote 43.Google Scholar
  12. 81.
    Drake [ 1981 ] has woven his translation of this piece into a fictitious dialogue among Galileo’s three favorite discussants. The dialogue contains Drake’s views on Galileo’s use of experiments and the role of causes in explanations. The text of Galileo’s work is set off typo-graphically, making the entire piece read like a scholastic commen-tary. All in all it is a delightful exercise, füll of insights and historical information.Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 1992

Authors and Affiliations

  • Joseph C. Pitt
    • 1
  1. 1.Virginia Polytechnic Institute and State UniversityUSA

Personalised recommendations