Was introduced to Thomas Kuhn's philosophy of science and failed to understand how and why history of science can serve for more than descriptive purposes. Further, it seems that Kuhn believes that it is possible to describe history of science in a neutral manner and then derive from it the philosophy that aligns with it. But isn't it true that every historical description is philosophy-laden? I mean, isn't it true that history from the outset cannot be at all described unless assuming ahead philosophical guidelines (for example - with regard to growth of knowledge)?

  • 2
    Can you ever describe something without some sort of assumptions ? – Mauro ALLEGRANZA Dec 8 '15 at 12:47
  • 4
    We can say that T.K. starts from a consideration of some relevant episodes in the history of science to "derive" a model suitable to explain how science changes during time. I think that it is correct to say that he implicitly assumes that history shows a "trend" of growth of knowledge. Has his model succeeded into explain how this growth is possible ? It is debatable ... – Mauro ALLEGRANZA Dec 8 '15 at 13:59
  • 4
    Thanks for these comments; and thanks for answers. My trouble with Kuhn's reliance on history is that he seems to describe history in virtue of implicit philosophical view he holds and then he argues that history sides with his philosophy. In which case he is circular (since of course 'history' sides with his philosophy if he described the history in light of his philosophical view which he assumed ahead). – L.M. Student Dec 8 '15 at 16:17
  • 4
    In part I agree with you : we can say that he selected "suitable" case histories that "fit" with his model. It is hard to say that all "changes" in the growth of scientific knowledge support his theories. But this is nothing new with regards to "philosophical" theories... – Mauro ALLEGRANZA Dec 8 '15 at 16:36
  • 4
    I think that Paul Feyerabend argued something similar to what you say: he roughly said that since for every historical description there exits a case it does not cover then it follows there is no rule governing growth of scientific knowledge and "anything goes". But his reasoning dismisses science all the same. It is why I question the reliance on history which Kuhn seemed to popularize. Although my professor praises Kuhn I cannot but doubt his historical approach which seems circular... ; -- Thanks again. – L.M. Student Dec 8 '15 at 17:03

There is an inherent and well-known ambiguity in TSSR (Kuhn, 1962), where it is often unclear if Kuhn is simply describing how Science has been done (or is being done) or if he is offering a prescription on how Science should be done. This is for instance expressed by Feyerabend (1970):

Whenever I read Kuhn, I am troubeled by the following question: are we here presented with methodological prescriptions which tell the scientist how to proceed; or are we given a description, void of any evaluative element, of those activities which are generally called 'scientific'? Kuhn's writings, it seems to me, do not lead to a straitforward answer. They are ambiguous in the sense that they are compatible with both interpretations.

Similar thoughts are also given in other chapters of Criticism and the growth of knowledge (1970), which is an excellent companion reading to TSSR. For instance, Masterman (1970) is describing the many different - and sometimes conflicting - ways in which Kuhn is using his paradigms. These ambiguities are clearly related to your question, since the "alignment" between history events and an underlying philosophy is more critical if you are using TSSR for prescriptive purposes.


Feyerabend. 1970. Consolidations for the specialist. In: Criticism and the growth of knowledge. eds: Lakatos & Musgrave. Cambridge Uni Press. Cambridge, UK.

Masterman. 1970. The Nature of a Paradigm. In: Criticism and the growth of knowledge. eds: Lakatos & Musgrave. Cambridge Uni Press. Cambridge, UK.

  • 4
    Thanks for bringing forth this descriptive-prescriptive dichotomy which indeed lies behind my question. To me Kuhn seems to have used history to explain scientists (sociology) rather than science (episteme); he seems to have been doing historical (descriptive) rather than philosophical (prescriptive) work, although he himself defended his view (in response to Feyerabend) by claiming that his work should be read from both descriptive and prescriptive points of view. – L.M. Student Dec 9 '15 at 11:50
  • 4
    p.s. Obviously there are many narratives which could describe science history; but Kuhn, as I see it, celebrates his narrative while ignoring the fact that other narratives could be equally plausible as his. And if so, he cannot use his historical narrative as 'evidence' to the plausibility of his alleged prescriptive (namely - philosophical) view of science. But he nevertheless does. -- Thanks again. – L.M. Student Dec 9 '15 at 11:50
  • 3
    True about alternative narratives with regards to historical "data", which in a way is a sociological version of the inverse problem in science. And from what I've read by Kuhn he sometimes seems annoyed that people see his writings as ambiguous, but sometimes seems to embrace the ambiguity (so he's ambiguous in the responses as well....). – fileunderwater Dec 9 '15 at 11:57

These are a lot of questions. I will highlight some notes:

  • Kuhn tries to analyze science from a different perspective and in fact tries to reveal the mechanisms that underly the evolution of science by analyzing the history of science itself. An analogy is perhaps the following: these days a lot of work is done in data-analysis. And while you can perform statistical tests in order to describe your data, you can also use statistics in order to make predictions for the future. In this way, 'historical' data, can be used for more than just descriptive matters.

  • It is true that according to Kuhn you cannot describe science without having some rules or language which is determined by the paradigm that you are working in. However, it is possible to talk about science even though you are working in a different paradigm; but in this case you will a different set of rules and language. An example of this is the notion of 'mass'. The concept of 'mass' used in Newtonian physics is a different concept than the concept of 'mass' used in Einsteinian physics. It is possible for both scientists to talk about this concept, however to be able to talk to each other both must do a much greater effort, because the language first must be clear. This is what Kuhn calls incommensurability.


A fundamental premise advanced by Kuhn is that established frameworks of scientific understanding (his paradigms) are subject to change. There are many examples, but the simplest and best is the Copernican revolution.

Historians look at events and trends to distill meaning, and Kuhn actually uses a multitude of historical examples in Structure (Kuhn, 1962) to show that history may work against 'normal' scientific endeavors, because of the tendencies of scientists to work within and further advance accepted theories, rather working to challenge those theories where fundamental anomolies are encountered. He makes the case that what we think we know may in fact be wrong. A paradigm shift is thus very difficult to achieve, which, per Kuhn, is amply evidenced in the history of science.

Philosophers share the historians' search for meaning. Both share the scientists' pursuit of understanding, and ultimate advancement of knowledge, though the methods may be quite different. When viewed as overlapping Venn diagrams, however, it is clear that Kuhn (and other philosophers of science) would be unable to work in a vaccuum. So, in the case of scientific evolution (or revolution), yes, philosophers and historians of sciance need to collaborate closely. Whether they would or do depends on whether they have an integrative view of humanities, or a siloed view. Politics and personal bias also tend to enter into it.

What philosophy can provide are the cognitive tools and frameworks (call them epistemologies) that bring logic and consistency to scientific studies. The breakthroughs of Bacon/Newton that founded empirical science are grounded in such notions, though they pre-dated Kant's categorical framework. But Kant clearly established his C.F. as a way to "sketch the complete plan for the whole of a science" (CPR(A), Transcendental Logic Sec III, 11).

If philosophy is defined nominally as how we come to understand things, then perhaps history is in fact deeply connected with philosophical principles. Both the means and the ends of these communities look fundamentally different at first blush. But I believe the seach for understanding has deep roots, providing a common basis for thinking that advances all academic branches in the search for knowledge.

Suggested reading: Kuhn, TS. Structure of Scientific Revolutions (1962), U.of Chicago Press.


I too have an introductory level understanding of Kuhn’s philosophy. I have only just completed an introductory course in the philosophy of science and we only touched briefly on Kuhn's ideas, so you may wish to take what I say with a grain of salt.

Kuhn’s philosophy deals specifically with the subject of theory choice. Those scientific theories which have come to be the orthodoxy of different historical periods (i.e., Kuhn’s paradigms) are objective facts of history. However, as Kuhn argues, there is no algorithm for theory choice and history shows us that the methods by which we chose our scientific theories contain various subjective elements such as social factors and prejudices.

When you say that you “failed to understand how and why history of science can serve for more than descriptive purposes” you appear to be ignoring these objective, historical facts. Kuhn’s philosophy appears (to me) to undermine our faith in the correctness (and permanence) of our current scientific theories by suggesting the presence of a degree of epistemic relativism rooted in our social framework. As our social norms change over time, so too do our scientific theories. I believe that Kuhn denied being an epistemic relativist, but I also believe that most people read his work in this way.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.