Incommensurability (philosophy of science)

In the theory of science, incommensurability is the partial or complete untranslatability of the terms of one scientific theory into the terms of another theory. The term plays a role in the discussion of whether and in what way a logical relationship can be established between (rival) theories and in what way they can be compared with one another.

The term was introduced into the philosophy of science in the early 1960s by Thomas S. Kuhn and Paul Feyerabend . The model was the concept of incommensurability used in mathematics . The concrete concepts of incommensurability of Kuhn and Feyerabend and in particular the conclusions that both draw from the thesis they represent that incommensurability plays a role in the history of science differ considerably in some points.

Thomas S. Kuhn

Kuhn first introduced incommensurability in his book The Structure of Scientific Revolutions , where incommensurability is associated with a paradigm shift during scientific revolutions. His concept of incommensurability was subject to a development that ultimately led to his concept of incommensurability being based on Willard Van Orman Quine's thesis that translations are underdetermined. With Kuhn there is no global incommensurability of all concepts of incommensurable theories, but only locally limited incommensurability of some concepts.

Although according to Kuhn no direct "term-to-term" comparison between different incommensurable theories can be made, he also considers incommensurable theories to be comparable in principle. Instead of a term-to-term comparison, a comparison could take place using values that are shared independently of the paradigm, such as accuracy, simplicity, internal and external consistency, etc.

“Most readers […] have supposed that when I spoke of theories as incommensurable, I meant that they could not be compared. But 'incommensurability' is a term borrowed from mathematics, and it there has no such implication. The hypotenuse of an isosceles right triangle is incommensurable with its side, but the two can be compared to any required degree of precision. "

“Most readers […] have assumed that if I spoke of theories as incommensurable, I would mean that they were not comparable. But 'incommensurability' is a term borrowed from mathematics, and there it has no such implication. The hypotenuse of an isosceles and right triangle is incommensurable with its side, but the two can be compared with any precision. "

- TSKuhn : Theory-Change and Structure-Change: Comments on the Sneed Formalism Knowledge, 10, 179–199 (1976)

Paul Feyerabend

Compared to Kuhn, Feyerabend's conception of incommensurability is more radical. According to him, incommensurable theories are not only locally incommensurable, but also globally; that is, no primitive concept of one theory can be translated into the other, there are no bridging laws , and the principles of the conceptual apparatuses of both theories are incompatible. A simple logical-rational comparison between incommensurable theories is not possible for Feyerabend. Nevertheless, this does not automatically mean that the two theories are incomparable. Rather, the comparison of incommensurable theories is possibly difficult and not possible according to simple standards - such as those proposed by critical rationalism - but a comparison is possible in other respects - among other things according to the criterion of predictive success. With Feyerabend, the incommensurability between theories is rather rare. While for Kuhn incommensurability is a side effect of every paradigm shift in scientific revolutions, for Feyerabend, for example, incommensurability is not given in every revolution. For example, he does not regard the Ptolemaic and Copernican worldviews as incommensurable. Only so-called universal theories can be incommensurable if they are interpreted in a certain way.

Feyerabend draws the conclusion from his incommensurability hypothesis that there are no simple universal methods and rules that can ensure a rational approach. The rules by which scientists work are rather complex and contextual. He therefore advocates an anarchist epistemology and socially a democratic relativism. For Feyerabend it is precisely such a way of arranging society that leads to an increase in rationality:

“There is no need to fear that such a way of arranging society will lead to undesireable results. Science itself uses the method of ballot, discussion, vote, though without a clear grasp of its mechanism, and in a heavily biased way. But the rationality of our beliefs will certainly be considerably increased. "

“There is no reason to fear that such a way of arranging society will lead to undesirable results. Science itself uses the methods of polls, discussions and voting without a clear understanding of their mechanisms and in a highly biased manner. But the rationality of our convictions will certainly be increased considerably. "

- P. Feyerabend : Against Method (1975) p. 309

Hans Albert

According to Hans Albert, epistemological positions such as those developed by Kuhn and Feyerabend result in an immunization against criticism of individual theories that he considers methodologically unacceptable.

Incoherence argument

The incoherence argument is an objection to the practical relevance of incommensurability. It is not directed against incommensurability as a theoretical possibility per se, but against the historical examples given by Kuhn and Feyerabend, with which they want to show that incommensurability is not just a more or less exotic theoretical possibility without relevance to actually existing science, but is practically relevant. It basically says that it is inconsistent to claim that earlier worldviews are incommensurable to today's and thus not to be expressed in today's language, but at the same time to claim that they can be analyzed and their incommensurability determined. The argument has been put forward in different versions, for example by Hilary Putnam and Donald Davidson . Putnam thus criticizes the example of Galileo put forward by Feyerabend:

"To tell us that Galileo had 'incommensurable' notion and then go on to describe them at length is totally incoherent."

"Telling us that Galileo had 'incommensurable' terms and then going on to describe them at length is completely incoherent."

- H.Putnam : Reason, Truth and History Cambridge University Press (1981)

According to Howard Sankey, this incoherence can be circumvented by considering both theories embedded in a metalanguage and describing them with this without the terms of one theory being directly translatable into the other. Then the theories, although incommensurable, are also rationally comparable.

Incommensurability and rational comparability

In many receptions, incommensurability is mistakenly equated with the incomparability of two theories. Neither Feyerabend nor Kuhn made such a strict equation.

The weaker assumption that incommensurability, if not necessary, then possibly in some cases can be accompanied by incomparability, has also been criticized from various sides. The objection is that both Kuhn and Feyerabend base their respective justifications for incommensurability on a number of requirements that do not necessarily have to be accepted and for which alternatives exist. Among other things, this concerns the theory of meaning or the perspective adopted from logical empiricism , which understands empirical theories as pure sets of statements. Alternative ways of assigning a meaning to terms of a scientific theory, also allowing a rational comparison in the case of changes in theories and scientific revolutions, are, for example, the causal theory of meaning, which was advocated by Hilary Putnam, or also concepts based on the semantic conception of theories, in which theories are strands and networks of theories play a role in the way concepts are given meaning.

Individual evidence

↑ after I. Hacking (I. Hacking: Introduction to the Philosophy of Natural Sciences , Reclam, 1983, page 118) the term is the result of a conversation that Feyerabend and Kuhn had in 1960 on Telegraph Avenue in Berkley
↑ see his postscript in TS Kuhn: The Structure of Scientific Revolutions. University of Chicago Press, Chicago, 2nd ed.
^ J. Preston: Feyerabend, Philosophy, Science and Society Polity Press (1997) p. 104
↑ See e.g. B. also Paul Feyerabend: About the method. A dialogue. In: Gerard Radnitzky, Gunnar Andersson (eds.): Requirements and limits of science . Mohr, Tübingen 1981. ISBN 3-16-942722-9 . P. 368.
^ J. Preston: Feyerabend, Philosophy, Science and Society Polity Press (1997), p. 117.
↑ P.Feyerabend: Against Method (1975) Appendix. 2
↑ See e.g. B. Hans Albert: Critique of the pure epistemology . Tübingen 1987. pp. 2f.
↑ H.Putnam, Reason, Truth and History Cambridge University Press (1981).
^ D. Davidson: On the very Idea of a Conceptual Scheme. in "Inquiries into Truth and Interpretation" Clarendon Press, Oxford, 183-198 (1984).
^ Howard Sankey: The Incommensurability Thesis Avebury Series in Philosophy of Science (1994).
^ Howard Sankey: The Incommensurability Thesis Avebury Series in Philosophy of Science, (1994).
↑ W. Stegmüller: Problems and Results of the Philosophy of Science and Analytical Philosophy Volume II “Theory and Experience” second part “Theory Structures and Theory Dynamics” .

Web links

Eric Oberheim, Paul Hoyningen-Huene: The Incommensurability of Scientific Theories. In: Edward N. Zalta (Ed.): Stanford Encyclopedia of Philosophy .