Kuhn distinguishes two meanings for ‘paradigm’:

  • in a narrow sense, (1) the examples a discipline uses to articulate its assumptions that set the template for solving puzzles that arise in the discipline during a period of normal science,

  • and in a broad sense, (2) the key theories, instruments, values, and metaphysical assumptions that are made prior to puzzle-solving which constitute the disciplinary matrix of normal science.

Count two paradigms ‘incommensurable’ when the new paradigm is a radical alteration of the matrix of measurements, observations, and language adopted by the scientific community not for good reasons but by a non-rational kind of religious conversion to a new world that is different than, and discontinuous with, the one represented by the former paradigm.

In The Structure of Scientific Revolutions, Thomas S. Kuhn presents a post-modern, antifoundationalist philosophy of science that rejects the modern premise of science as cumulative progress, and instead proposes that science proceeds in a series of paradigm shifts with each new paradigm incommensurable to the previous. Kuhn shows the dramatic character of paradigm shifts in pointing to the historical cases of scientific revolution: from Newtonian mechanics to Einstein’s theory of relativity, Priestley’s dephlogisticated air to Lavoisier’s oxygen, and Ptolemy’s system to Copernicus’s system of planetary astronomy. My interest in this article will be to suggest the idea of incommensurable paradigms is unintelligible, and to sketch the upshots of this for the philosophy of science.

I proceed in stages. After reviewing Kuhn’s idea of incommensurable paradigms (Section I), I show that the idea of incommensurable paradigms is unintelligible (Section II). Then I consider the upshots of this view, namely that in order to be meaningful, Kuhn’s theory, even by Kuhn’s own lights, ought to be interpreted in a soft sense as having metaphorical meaning, rather than in a hard sense as having literal meaning (Section III). Finally, I argue that the logic of incommensurable paradigms depends on conscious, not self-conscious statements (Section IV), and suggest against his intentions that this leads his theory of science to be really useful as a social scientific, not philosophical theory of science (Section V).

I. Incommensurable Paradigms

In The Structure of Scientific Revolutions, Kuhn famously argues that science proceeds in a series of paradigm shifts. In a paradigm shift, the pre-paradigmatic period is chronologically prior to the paradigmatic period. The idea of chronology may seem to indicate that Kuhn argues for progress; however, since Kuhn restricts the achievement of knowledge to the paradigm itself, such progress can only be made within a paradigm, not across paradigms. This reflects Kuhn’s central argument that paradigms simply replace one another without progressing to an ultimately true description of nature. Instead of building off the previous paradigm, paradigms take the form of a rigid, deterministic normal science that consists in puzzles and rules set by an “established viewpoint” which a scientific community assumes when conducting science.1 Specifically, from this viewpoint, the scientific community finds resources from within a paradigm in order to complete its central task of “puzzle-solving.”2 Predetermined by commitments to the puzzles and rules set by the paradigm, the scientist is then relegated to the status of a servile puzzle-solver who uncreatively completes puzzles. Puzzle-solving therefore reifies the paradigm.

In observing the environment of normal science as self-reifying, Kuhn implies that the scientific community has a non-rational character. The scientific community experiences, he writes, “professionalization” and develops “esoteric vocabulary and skills” which cause an “immense restriction of the scientist’s vision” and “considerable resistance to paradigm change,”3 making them “narrow-minded.”4 In other words, the paradigm precedes the scientific community such that the scientist is constrained by paradigmatic methodological rules that are followed in research. It follows that paradigms are without the need for “full interpretation or rationalization,” simply because the scientists are already in dogmatic, full agreement.5 Furthermore, in full agreement, members of the scientific community can devise, Kuhn says, an “ad hoc modification of their theory in order to eliminate any apparent conflict.”6 So, from Kuhn’s external point of view, the scientific community tacitly behaves non-rationally in the laboratory, contrary to how they first-personally seem to behave rationally.7

The idea that scientists tacitly behave non-rationally is consistent with the broader idea that paradigm shifts occur non-rationally rather than rationally. To be sure, Kuhn admits that the scientists’ puzzle-solving activity within the paradigm of a normal science is rationally cumulative, but he explains that cumulative progress does not exist at the external level of observing paradigm shifts.8 From an external point of view, we find the absence of cumulative progress during paradigm shifts from the fact that paradigm shifts are what he phrases: “destructive as well as constructive.”9 This is the case in virtue of how paradigm shifts occur during a period of “crisis,” which is evoked from an “anomaly” in the accepted paradigm.10 The anomaly questions fundamental generalizations of the normal science, and blurs the paradigm to the point that its rules are loosened and the paradigm capitulates.11 During the period of crisis, the scientific community responds by critically theorizing candidate replacement paradigms in sparring debates just as in the pre-paradigmatic period, thus coming full circle. Despite the apparent rational behavior of the scientific community in doing critical inquiry during a crisis, Kuhn argues that a “gestalt switch” provides a good analogy for the paradigm shift whereby the scientist’s reality radically changes all at once to a different framework with a new methodology, goal, and view of the field.12 The scientific community experiences a kind of religious conversion and enters a “different world” with a new set of measurements, observational data, and language that is discontinuous, or incommensurable with the set of the former paradigm.13 In this sense, paradigm shifts are determined not by reason, but rather by tacit, dogmatic conversion outside our rational control. Summing up his theory, we can construct Kuhn’s premise:

(A) Paradigms are incommensurable in the sense that they occupy different worlds such that each paradigm has a different content or meaning that is discontinuous with the former paradigm.14

II. The Intelligibility of Incommensurable Paradigms

These observations invite criticism. Namely, what does it mean for two incommensurable paradigms to represent ‘two different worlds’? If Donald Davidson’s argument in “On the Very Idea of a Conceptual Scheme” is right as I think it is, Kuhn’s idea of incommensurable paradigms becomes unintelligible. As Davidson points out, the fact that we can translate a paradigm into our own language – where only a “Webster’s dictionary”15 separates the ostensibly ‘two different worlds’ – shows the idea of incommensurable paradigms is unintelligible. For, if there were two completely different worlds with no means of translation between each other (i.e., no means of translating ‘Twater’ on Twin Earth to ‘Water’ on Earth in Putnam’s Twin Earth thought experiment), then there would be no way of making sense of the other world. Furthermore, the idea of simply labeling an entity that we are unable to make sense of as an ‘incommensurable paradigm’ is arbitrary and intellectually dishonest, since we don’t even know what the entity is. Indeed, to us, such an entity is unintelligible. Thus, in order to be intellectually honest, we should instead admit the entity as such: unintelligible.

Though it is perhaps a banal fact that two worlds are literally discontinuous, underlying the point is Davidson’s central premise, which we can identify as:

(B) A paradigm is a language, meaning that a paradigm’s content – i.e., its theory, methodology, goal, measurements, observational data, and language – is not separate from the basic language used by the paradigm.

Importantly, for Davidson, language is not meant in the way Kuhn means it, being in the sense of vocabulary or theory-laden propositions, but rather in the sense of language simpliciter, that is, referents, words.16 For example, two scientific communities ostensibly operating in incommensurable paradigms speak, say, English, and are therefore able to communicate to each other by translating their theory-laden propositions. In first agreeing to communicate, they have ipso facto also agreed to be charitable to each other and come to an understanding. Davidson’s point is the fact of translatability between the two paradigms presupposes a shared foundation, or point of minimal contact that makes the two paradigms continuous with each other.17 Two paradigms being essentially continuous thereby disproves any idea of incommensurable paradigms or ‘two different worlds’ as the two worlds are, in fact, “only words apart.”18 Hypothetically speaking, the two worlds are able to be bridged into one world once they are in full translation, meaning any idea of partial translatability, or point of minimum contact between paradigms is no different than full translation in view of intelligibility. Altogether, referencing Quine’s “Two Dogmas of Empiricism,” Davidson calls the foregoing proof a proof against the third dogma of empiricism: the idea of scheme-content dualism.19 Evidently, Kuhn commits the third dogma of empiricism.

III. Incommensurable Paradigms as a Metaphor

The view that the idea of incommensurable paradigms is unintelligible has upshots. Even by Kuhn’s own lights, we can interpret ‘two different worlds’ to in a soft sense have metaphorical meaning, rather than to in a hard sense have literal meaning. The metaphor becomes evident when we understand Kuhn to be attempting to make sense of a phenomenon in the history of science that strikes him and many others in wonder. That is, wonder at how two different schools of thought in the same discipline, operating in different periods, can use the same words and refer to the same things, yet mean those words in radically different ways. It is precisely in response to this phenomenon in the history of science that Kuhn derives his theory. As already explained, Kuhn conjectures that each school of thought works with a unique set of measurements, observational data, and language that constructs its own paradigm – its own world. The idea that paradigms are incommensurable, or constitute ‘two different worlds,’ then, is not the product of conceptual analysis of science but an empirical generalization of science as a phenomenon. It is a metaphor used to describe the phenomenon he experiences when observing the history of science.

In his Postscript, Kuhn doubles down on the idea that incommensurable paradigms are literally ‘two different worlds.’ This is not surprising given the stakes, but his clarification of ‘two different worlds’ in view of translatability is very revealing of how he metaphorically, not literally grounds his theory in his own experience. Particularly, in his response to charges of non-rationality, he stubbornly concedes that in a paradigm shift, the deliberative procedure of “theory-choice” by scientists in crisis does not resemble a logical proof.20 He explains that this is because theory-choice is a procedure of one party converting the other to their paradigm by persuasion. The activity of conversion, by the use of words and persuasive argument, thus makes Kuhnian paradigm shifts not as mysterious nor ineffable as the ‘gestalt switch’ suggests. Where Kuhn makes the bolder claim, however, is in the necessary condition he takes conversion through persuasion to occur. That is, that the persuasive reasons for choosing a theory be a function of values, not logic.21 With the two parties separated by paradigmatic values, not inner logic, the two parties hence translate each other’s paradigms not into one coherent world, but into their own languages – their own different worlds.22

The function of values as the ultimate grounds of a paradigm shift in Kuhn’s theory still finds the idea of incommensurable paradigms unintelligible. Working from the premise of scheme-content dualism (A), Kuhn explains that the experiences of persuasion and conversion are “not the same.”23 He asserts that persuasion may occur, but is not necessary for a paradigm shift; rather, conversion is necessary for a paradigm shift.24 In the experience of persuasion, two paradigmatic language communities may translate the other community’s language to theirs, but this translation requires each community beginning, he writes, “vicariously to understand how a statement previously opaque could seem an explanation to members of the opposing group;” that is, “to translate a theory or worldview into one’s own language is not to make it one’s own.”25 Instead of acts of persuasion being essential to a paradigm shift, Kuhn concludes they are “points of entry” for conversion, serving as unknowable motives for conversion.26 Thus, leaving the cause of conversion altogether unknowable, conversion is grounded in nothing more than Kuhn’s own allusive sense of wonder about how one paradigmatic language community can somehow flip to becoming a different language community. He conjectures the idea of the mysteriously ‘gestalt’ conversion experience of a paradigm shift to explain this puzzling phenomenon. But, familiar with Davidson, Kuhn’s attempt here to explain the inexplicable makes the conjecture unintelligible. Nonetheless, being charitable to Kuhn, his theory can be rationalized in a soft sense as a metaphor that poignantly describes his observation of the history of science. It may, like literature, speak to others who share the same observation, particularly those with post-modern intuitions.

IV. The Logic of Incommensurable Paradigms

Under this interpretation of Kuhn, we are now in a position to develop a firm understanding of where Kuhn’s wonder leads him astray, logically speaking. Namely, to grasp the meaning of incommensurable paradigms as a phenomenon, Wittgenstein’s analysis of the rabbit-duck illusion helps us in distinguishing perception from interpretation, or conscious from self-conscious statements, namely, “seeing that” versus “seeing as.”27 The basic thought is if you only see a rabbit when looking at the illusion, you would say “this is a rabbit,” but once you become aware of the duality you would say “now I see it as a rabbit.”28 You might also say “it’s a rabbit-duck,” which, for Wittgenstein, is a perceptual report.28 In light of Kuhn, the rabbit-duck illusion shows how one’s perception of ‘two different worlds’ becomes unintelligible once they reflect that “it’s a rabbit-duck,” since by realizing “it’s a rabbit-duck,” the perceiver has successfully translated the two perceptions into one coherent synthesis, that is, one world. It thus follows Kuhn’s theory only in a soft sense provides a metaphorical interpretation of what he perceives in the history of science. The idea of incommensurable paradigms in science is a formulation of the perceptual report “it’s a rabbit-duck” such that upon reading Kuhn’s book, we as readers have the guidance to flip between two incommensurable paradigms – i.e., Aristotle’s and Galileo’s – and report outwardly to the library or inwardly to ourselves, “it’s a paradigm shift!”

To be sure, Kuhn is careful to reject the rabbit-duck illusion as an analogy to paradigm shifts. He contends that the illusion is an artificial psychological experiment contrived for producing a gestalt switch. The contrived experience simply does not reflect how scientists actually think when a paradigm shift occurs. That is, Kuhn’s main point is that when a paradigm shift occurs, the entire world of the science is changed to where, unlike the observer of the rabbit-duck illusion, there is no way of reverting back.29 But, as aforementioned, Kuhn’s theory expressed in the book constitutes his rabbit-duck illusion. Kuhn has contrived the metaphor of incommensurable paradigms, and has written it into the artefact of a book with which the reader reads and is entertained by how Aristotle and Galileo seem to operate in two different worlds. The reader, not the scientist, flips between pages and ideas just as how the observer flips between rabbit and duck after realizing it’s a rabbit-duck. Kuhn’s objection to the rabbit-duck illusion as an analogy to his theory, then, is only a wholehearted attempt to evade the fact that his theory is a conjecture that inevitably needs to contend with the reader, not his imagined scientist. Indeed, looking to sell his theory, he operates merely at the conscious, unreflective level of the scientist’s “seeing that” statements, not at the self-conscious, reflective level of the reader’s “seeing as” statements. Of course, operating at the conscious level makes the “seeing that” statements of two different language communities seem as though they are incommensurable.

V. The Utility of Kuhn’s Theory

Despite the unintelligibility of incommensurable paradigms, Kuhn’s theory, as I have interpreted it, in a soft sense can be a metaphor that speaks to many readers who share a similar observation of, and post-modern intuition about, the history of science (as it has for the last half-century!). Because of its seminal achievement in this respect, Kuhn’s theory clearly ought to be taken seriously. But, in light of the criticism this article levels, how ought we do so? My answer: Kuhn’s theory is valid as a social scientific, not philosophical theory of science.

This answer derives from looking at what exactly the theory is useful for. To this end, we find the premise, appended to Davidson’s:

(B*) Paradigms are an artefactual category useful for social scientists in describing scientific practices.30

Placed in the social sciences, Kuhn’s theory then is a conjecture to be tested by other social scientists who read the work. As a social theory, the theory passes the test depending on whether other social scientists critically accept the theory or not. Upon accepting the theory, the social scientists apply the artefactual category to their field work, and perhaps, upon experiment, come to reject the theory. The point here is that Kuhn’s theory is certainly useful for social scientists in describing scientific practices that seem radically different, though as a social theory finds its worth not in conceptual analysis, as in the case of philosophy, but in its critical acceptance in the social scientific community.31

The distinguishing feature of (B*) is that it shows that paradigms are not useful for philosophers who are concerned with meaningful conceptual analysis, consisting in precise, valid, and modest propositions which by the conclusion hang together in an intelligible picture. Simply put, the idea of incommensurable paradigms is unintelligible as a proposition, and as a result fails to explain anything conceptually about the meaning or character of science. Among philosophers of science whom I believe succeed in meaningful conceptual analysis is Karl Popper through his attempt to explain the character of science by taking on the scientist’s first-personal point of view, which he calls the “critical attitude.”32 By contrast, on my reading, we see Kuhn not to do conceptual analysis, but instead to categorize movements in science as a social scientist. There is no doubt that Kuhn’s theory is important for making sense of how we understand previous movements in science from the perspective of a post-modern point of view. Philosophers, too, can appreciate its attempt to show unity in the history of science which humbles us to appreciate our engagement with science as a modest, Sisyphean fact of the human condition, not a teleological goal pursued in bad faith. Commendably, the theory brings attention to the social aspect of science, bringing us outside the laboratories, or for us non-scientists, our caves in which we dogmatically idolize science. Popper also shares this virtue.

Conclusion

In this article, I hope to have shown that Kuhn’s idea of incommensurable paradigms, which is core to his theory, is in a hard sense unintelligible but in a soft sense nonetheless useful as an artefact for social scientists. I made my case by reviewing Kuhn’s premise of scheme-content dualism (A) that grounds his theory, locating the consideration of translatability between paradigms in Davidson’s premise (B), articulating the upshots of (B), and noting the utility of Kuhn’s theory in light of my criticisms with my premise (B*).

About the Author

Joseph F. Diller is a master’s student of philosophy at the University of Chicago; formerly an undergraduate student of philosophy and history at the University of Chicago. He has broad primary interests in moral, political, and legal philosophy, as well as secondary interests in epistemology and the philosophy of science. Reach him via email at jfdiller@uchicago.edu or on Twitter @Dillerjf.

Bibliography

  • Davidson, Donald. “On the Very Idea of a Conceptual Scheme.” Proceedings and Addresses of the American Philosophical Association 47 (1973): 5–20.

  • Kuhn, Thomas S. The Structure of Scientific Revolutions. Introductory Essay by Ian Hacking. 4th ed. The University of Chicago Press, 2012.

  • Popper, Karl. Conjectures and Refutations. 3rd ed. Routledge Classics, 2004.

  • Wittgenstein, Ludwig. Philosophical Investigations. Translated by G. E. M. Anscombe. 3rd ed. The Macmillan Company, 1969.


  1. Kuhn, The Structure of Scientific Revolutions, 39. [return]
  2. Ibid., 42. [return]
  3. Ibid., 64. [return]
  4. Ibid., 42. [return]
  5. Ibid., 44. [return]
  6. Ibid., 78. [return]
  7. For reference, critical rationalism is an epistemology advanced by Karl Popper that proposes statements be made, and examined, under the condition that they are logically falsifiable. Kuhn’s scientist does not behave with such a form of rationality, even if they claim to be ‘rational.’ [return]
  8. Kuhn, The Structure of Scientific Revolutions, 52. [return]
  9. Ibid., 66. [return]
  10. Ibid., 82. [return]
  11. Ibid., 84. [return]
  12. Ibid., 85. [return]
  13. Ibid., 147. [return]
  14. This is otherwise known as the idea of scheme-content dualism. [return]
  15. Davidson, “On the Very Idea of a Conceptual Scheme,” 11. [return]
  16. We can understand the source of their difference in how they mean ‘language’ to lie in Davidson’s known externalism about meaning and Kuhn’s implicit internalism about meaning. I accuse Kuhn of being an internalist since he justifies meaning with the scientist’s internal mental states, that is, the scientist’s subjective beliefs, intentions, and attitudes about a word. Being an internalist allows Kuhn to demarcate one set of subjective beliefs about words from another, i.e., one paradigm incommensurate with another. [return]
  17. By “point of minimal contact,” I mean the idea that two language users of two different languages could easily translate each other’s languages, and thereby make one world, without being perfect translators who know the translation of every word and phrase, but only some. With a point of minimal contact, they have a starting point to work upwards from. For example, there existed a point of minimal contact in the case of the Rosetta Stone. Linguists were able to translate previously untranslatable, and therefore unintelligible, languages on the basis of fragments of each language that meant the same thing (i.e., same content) but were said by a different language (i.e., different scheme). Indeed, as the stone had Ancient Greek, Ancient Egyptian Hieroglyphs, and Demotic Script, and as the linguists knew Ancient Greek, linguists had a point of minimal contact to translate what was on the stone of the Ancient Egyptian Hieroglyphs and Demotic Script. From there, they were able to easily infer the rest of the languages. [return]
  18. Davidson, “On the Very Idea of a Conceptual Scheme,” 11 (my emphasis). Here, we can see the significance of the idea of a point of minimum contact. Without a point of minimum contact, the interpreter of another text that is untranslatable is discontinuous with the interpreter’s paradigm, making the other text unidentifiable. Very much like with UFOs, we can create science fiction speculating the world of the text or object, but it is only a fantasy world, not another intelligible world. Indeed, in order to not look foolishly unscientific, the interpreter must designate the text, or object that is without a point of minimum contact as unidentifiable. An example of interpreters looking foolish is interpretations of the “Voynich Manuscript.” Dated 1400-1430 A.D, the manuscript has experienced many efforts of translation, especially after 1912, when it was purchased by a Polish book dealer. Altogether, many cryptographers have failed to decode it; though more recently, a television writer and a historian claimed to have interpreted it as a woman’s health guide (see Gibbs, Nicholas. “Voynich Manuscript: The Solution.” TLS. September 8, 2017. Available at: https://www.the-tls.co.uk/articles/voynich-manuscript-solution/). They translated two sentences and claimed it was written in abbreviations of Latin whereby each letter was really an abbreviation of a Latin word. However, the two sentences they supposedly translated made no grammatical sense. Clearly, there was not enough of a minimum point of contact to make the manuscript intelligible. Their interpretation proved to be an unwarranted leap. The manuscript is still unintelligible, and could for all we know be a farce intended to drive translators mad. [return]
  19. Ibid., 11. [return]
  20. Kuhn, The Structure of Scientific Revolutions, 198. [return]
  21. Ibid. [return]
  22. Ibid., 202. [return]
  23. Ibid., 201. [return]
  24. Ibid. [return]
  25. Ibid., 202. [return]
  26. Ibid., 203. [return]
  27. Wittgenstein, Philosophical Investigations, 194-197. [return]
  28. Ibid., 195. [return]
  29. Kuhn, The Structure of Scientific Revolutions, 114-115. [return]
  30. By ‘artefactual category,’ I mean a conceptual tool, prone to evolution, made intentionally by social scientists to accomplish some purpose in the social sciences. Another example, among others, of an artefactual category in the social sciences is Marxism’s labor theory of value. The theory analyzes the economic value of a product as an embodiment of the worker’s labor. Under conditions of capitalism, where the working class is alienated from their labor by the ruling class, the analysis goes, products are not evaluated according to their proper value, but their exchange value, i.e., price. Therefore, the working class’s only viable path to liberation from the problem of alienation is revolution against the ruling class. Just as in the case of Kuhn’s theory, the labor theory of value is a theory testable only by other evaluating social scientists. The ‘test’ is only a matter of whether there are enough social scientists who accept the theory and think it’s useful to analyze economic value in such a way. [return]
  31. This is a reference to Karl Popper’s philosophy of science as a series of conjectures of refutations. (See, Conjectures and Refutations by Karl Popper.) I take this to be the cardinal philosophy of science, as it demarcates science from pseudo-science. Here, I apply Popper’s theory to Kuhn’s, showing that Kuhn’s can be contained by Popper’s. [return]
  32. Popper, Conjectures and Refutations, 64-65. [return]