Redefining Science With Thomas Kuhn

Redefining Science With Thomas Kuhn May 15, 2019

A reread of Kuhn’s essay makes me realize that those who claim to love science still define it all wrong.

I’ve discussed Thomas Kuhn before on my Anti-Science channel, and my colleague PD has posted an essay on the Popper-Kuhn debate. I recently reread Kuhn’s 1962 volume The Structure of Scientific Revolutions, and it still strikes me as an innovative way of looking at scientific endeavor. The view of science that he tried to put to rest is still popular in the science-fan community.

Gimme That Old Time Science

The standard view of scientific inquiry is the model of Baconian induction, where data points are organized by researchers who create a theory to interpret and explain them. The evidence-based theory forms the basis of a framework for further research which tests the theory. New ideas are introduced by additional research, and the theory may have to change or be replaced to accommodate the information. This scenario of scientific inquiry is the one that’s largely accepted by science fans these days, and it stresses the importance of evidence. The self-correcting mechanism of science is the constant testing that all ideas and theories are subject to; scientists are constantly challenging and testing ideas to make sure they’re still reliable in the face of new evidence. Like a house being built brick by brick, science progresses incrementally throughout history and the strength of the evidence gives us confidence that our knowledge is true and corresponds to objective reality.

Kuhn made that scenario look like folklore.

Stuck in the Paradigm with You

In his work, Kuhn described what he called the paradigm, the cognitive and intellectual infrastructure that forms the worldview of a scientific research community. Through education, socialization, and professional development, scientific communities do work that’s defined by and intended to reinforce the paradigm. In contrast to the romantic notion of scientists as mavericks constantly challenging assumptions and theories, Kuhn painted a picture of groupthink and conformism: he said that the vast majority of scientists were involved with normal science, puzzle-solving that makes Nature fit the paradigm.

Like any community biased by its expectations and conditioned to validate its own assumptions, scientists working in a given paradigm emphasize the data that works and ignore the anomalies. In time, however, the anomalies may start to create the awareness of crisis in the paradigm; researchers may try to contextualize the data into a new framework that rivals the old paradigm. If it proves useful enough, like-minded researchers will support the new framework and it will produce a new paradigm when the proponents of the old one either come around to the new way of thinking or (more often) leave the profession and die out.

How Evidence Is Beside the Point

The most striking things about this scenario is the lack of emphasis on evidence. The ease with which scientists can ignore or de-emphasize anomalies is a hallmark of normal science. The very way data is conceptualized and defined under the new paradigm is vastly different than under the old one:

Many of the puzzles of contemporary normal science did not exist until after the most recent scientific revolution. Very few of them can be traced back to the historic beginning of the science within which they now occur. Earlier generations pursued their own problems with their own instruments and their own canons of solution. Nor is it just the problems that have changed. Rather the whole network of fact and theory that the textbook paradigm fits to nature has shifted. Is the constancy of chemical composition, for example, a mere fact of experience that chemists could have discovered by experiment within any one of the worlds within which chemists have practiced? Or is it rather one element—and an indubitable one, at that—in a new fabric of associated fact and theory that Dalton fitted to the earlier chemical experience as a whole, changing that experience in the process? Or by the same token, is the constant acceleration produced by a constant force a mere fact that students of dynamics have always sought, or is it rather the answer to a question that first arose only within Newtonian theory and that that theory could answer from the body of information available before the question was asked?

Different Paradigm, Different World

Most infuriating for the science fans, Kuhn stresses that scientists’ ways of thinking are so theory-laden that researchers in different paradigms are looking at different worlds. That is, they’re seeing the phenomena with such a different basis for understanding it that they have no common language with which to discuss it and make objective theory-choice decisions. “Kuhn’s Gap” is the term philosophers of science use to talk about the incommensurability of competing scientific frameworks; the evidence doesn’t bridge this gap, but rather the notion of cogency in scientific communication.

So Kuhn has done away with the idea of scientific truth having a degree of correspondence to reality; he compares the concept of science progressing toward truth to the outmoded concept of biological evolution progressing toward some goal. A scientific theory or framework is only better if researchers find it more useful in inspiring problem-solving that generates coherent data.

In a book of barely two hundred pages, Kuhn shook the foundations of the way we define scientific inquiry and conceptualize progress, and described the limitations of our knowledge of what’s real. I highly recommend this book to people who want to understand the scientific process in a more nuanced and realistic way.

Who else here has read Kuhn? Do you agree with his ideas? Has his influence been good or bad?

"It’s not like we have evidence that the early Christians decided to pretend myth-Jesus had ..."

No, Conspiracism Has More In Common ..."
"If someone wishes to perpetuate the incorrect concept that the subspecies of Homo Sapiens are ..."

Will Genetic Science End Racism or ..."
"We should not be afraid of science and I refuse to believe that in the ..."

Will Genetic Science End Racism or ..."
"If someone wishes to perpetuate the incorrect concept that the subspecies of Homo Sapiens are ..."

Will Genetic Science End Racism or ..."

Browse Our Archives

Follow Us!


TRENDING AT PATHEOS Nonreligious
What Are Your Thoughts?leave a comment
  • Rann

    Sorry, try again guys…. you don’t even describe the paradigm properly. As someone who has been working in science for 40+ years, I can tell you, Kuhn is battling a strawman.

  • Welcome to Driven to Abstraction, Rann.

    I don’t think Kuhn is “battling” anything, he just describes scientific inquiry as a collective, collaborative, institutional endeavor. Are you saying that what he calls a paradigm doesn’t exist, or that I defined it incorrectly?

  • Milo C

    I’m willing to acknowledge Kuhn’s… statement; I can see scientific endeavor does have collective information gathering, sometimes a few strong theories that develop institutions of like-minded researchers. But I find his conclusions very dubious.
    Does he cite the frequency with which scientists “emphasize the data that works and ignore the anomalies”? Because this would be useful in identifying bias, and that seems to be at the core of the issue. Are scientists working with bias, or are they eliminating it to the best of their ability? Also, let’s hope he isn’t conflating the scientific method with the way it is put into practice by such esteemed bodies as the Discovery Institute.

  • Michael Neville

    I’ve always felt rather dissatisfied by Kuhn. I know that he was a PhD physicist and well acquainted with the philosophy of science, certainly more so than I. However I’ve got some objections.

    Kuhn claimed that much of his approach to the philosophy of science was informed by the simple question “How could Aristotle, one of the world’s greatest philosophers, be so wrong about so much of physics?” His answer to this question that Aristotle was not wrong. Aristotle was simply exposed to knowledge that is different to what we have now and therefore simply perceived the world differently to modern scientists. Indeed, much of what Aristotle believed was perfectly reasonable in terms of evidence at the time and we must beware of judging the past through the lens of today’s knowledge.

    Okay, I can see that. But then Kuhn goes on to assert that different perceptions can be equally valid. There is no right or wrong view. This relativism quickly becomes very problematic for scientists. First, it ignores that Aristotle was famously disinterested in evidence; he believed the ideas of the mind were far superior to any physical observation. More importantly, modern science places great emphasis on the concept of wrong; it is only by comparison with observation that we make progress. We select between theories that describe the world reasonably accurately and theories that don’t. This process of elimination is a fundamentally different starting point to that of Aristotle and it has driven all the major breakthroughs of modern science. There is surprisingly little discussion of this simple point (now known as Popperian falsification) in Kuhn’s book.

    A second problem concerns Kuhn’s idea of the paradigm shift in science. According to him, all of scientific theory and experiment takes place within a given paradigm. From a theory of gravity to particle physics, experiment and theory generally occur within an agreed overarching sets of beliefs. If enough contradictory evidence builds up that cannot fit the paradigm, a new paradigm then arises which replaces the old, so a paradigm shift occurs. So far, this is a perfectly reasonable description of how science is done, if a bit simplistic as it ignores competing models within paradigms.

    But it is what comes next is problematic. According to Kuhn the new paradigm completely replaces the old, rendering the old effectively redundant. Time and again in his book, the new paradigm is portrayed as a new world-view, entirely replacing and invalidating the old. This view amazes me, particularly coming from a physicist. First, it is very difficult for a paradigm to become established in the first place: it has to provide an adequate explanation for hundreds of measurements in different, but related, fields. This skeptical aspect of science should not be understated. For the same reason, it is difficult for a new paradigm to emerge; this is because the old paradigm explained a tremendous amount and is not lightly overthrown. If the old is overthrown, it is only on the emergence of startling new evidence, usually gradually accepted over long periods of time. Even then, there are periods of time during which time the new and old coexist and compete. Pick any revolution you like, from continental drift to quantum physics, the new can only replace the old if it explains all the old did, plus a whole lot more. This is because as new evidence is uncovered, old evidence also remains. This view of science as a cumulative process is criticized by Kuhn, but it matches present-day science very well. For example, it took quantum physics at least 30 years to emerge due to the slow accumulation of experimental observation and theory. Hence, the new is very much an extension, however radical, of the old and the old paradigm is not discarded. We still use non-relativistic and non-quantum physics to this day, where we always used it; where the limitations do not apply.

  • I don’t think Kuhn is implying that scientists deliberately fudge the details of their research, or consciously exclude data that don’t support their hypotheses. The way he describes it, they’re merely working within a professional and institutional structure that’s already defined the aims and boundaries of their activity. They’ve learned what questions they’re likely to generate acceptable answers to, and what sort of problems are in the scope of their work. And the amount of progress that this “normal mode” of science makes appears to validate the assumptions of the paradigm to such an extent that anomalies are dismissed as outliers rather than recognized as flaws.

    It’s not until a rival paradigm reconceptualizes phenomena in a way that assimilates the anomalies that people even realize that there was a problem; however, the researchers invested in the new paradigm make it sound like it’s the answer to questions that scientists had been asking all along.

    It’s just a necessary corrective to the notion that science is all about the evidence. In fact, scientists looking at the same evidence can have fundamental disagreements about how to interpret the evidence and which theoretical framework explains it best.

  • Your use of the term “science fan” strikes me as demeaning, and detracts from any valid points you might want to make.

  • Milo C

    I did not intend to imply Kuhn was accusing anyone of deliberate bias; that was not in my statement at all. I am still not convinced he’s offered a compelling hypothesis, though. Did Kuhn reference or mention the researchers who make it their job to discover unconscious bias in the scientific community? I feel like he’s trying to make a generalized statement questioning the validity of scientific knowledge as we know it without doing his homework.

  • We select between theories that describe the world reasonably accurately and theories that don’t. This process of elimination is a fundamentally different starting point to that of Aristotle and it has driven all the major breakthroughs of modern science.

    This is a really simplistic way of describing scientific progress, and Kuhn’s work did away with this schoolboy positivism half a century ago. What Kuhn calls “the incompleteness and inconsistency of the data-theory fit” is always present to a certain degree, and inspires researchers to define the problems in understanding phenomena in diverse ways.

    It’s easy to look back and say that such-and-such a theory supplanted its predecessor because it “describes the world accurately” whereas its predecessor didn’t. But we forget that the paradigm shift involved a reconceptualization of the phenomena and the problems posed by the prior paradigm’s data-theory fit.

    There is surprisingly little discussion of this simple point (now known as Popperian falsification) in Kuhn’s book.

    It’s true that he doesn’t spend much time discussing Popper’s point. However, there’s a good reason for his dismissal. Kuhn’s basic premise, as I’ve said, is that the data-theory fit is always incomplete and there are always anomalies in the construct. Therefore, the grounds for what Popper calls “falsification”—an observation that appears to disconfirm the theory— is just an expected feature in the nature of scientific inquiry. According to Kuhn, no theory should resist falsification in the strict Popperian sense.

  • Michael Neville

    Aristotle went out of his way not to use simple observations for his “scientific” pronouncements. For instance, he said that women had fewer teeth than men. He never had Mrs. Aristotle open her mouth so he could get a count. He pulled that claim straight out of his rosy red rectum and assumed it was true because he had said it. That was my point about the difference between Aristotle and modern science.

    Kuhn’s “incompleteness and inconsistency of the data-theory fit” is why science never considers anything to be true (unlike Aristotle) and is the basis for science being self-correcting both at the specific claim and general theory levels. That’s all that paradigm shifts are, selt-correction at the theoretical level. This is not to say that science is perfectly self-correcting, because it definitely isn’t. Changes could be made more self-correcting, such as publishing negative results. Scientific journals only publish the researchers whose experiments support the conclusions they were making. There should be a practice of publishing negative results, more widespread publication of experiments that failed to reproduce the result of the original author. That would give a much better idea of where the current consensus needs to be questioned..

  • This is not to say that science is perfectly self-correcting, because it definitely isn’t. Changes could be made more self-correcting, such as publishing negative results. Scientific journals only publish the researchers whose experiments support the conclusions they were making. There should be a practice of publishing negative results, more widespread publication of experiments that failed to reproduce the result of the original author. That would give a much better idea of where the current consensus needs to be questioned.

    That’s true. Publishing only positive results gives us a distorted sense of science’s efficiency. Stephen Jay Gould published an essay called Cordelia’s Dilemma (I read it in the Dinosaur in a Haystack collection) where he argued for the same thing:

    Statistics often get a bum rap in our epithets and editorials. But I am both a champion and a frequent user of statistical procedures, for the science exists largely to identify and root out hopes and misperceptions falsely read into numerical data. Statistics can tell us when published numbers truly point to the probability of a negative result, even though we, in our hopes, have mistakenly conferred a positive interpretation. But statistics cannot rescue us when we hide our nonlights under a bushel (with apologies to Matthew 5:15)–that is, when we only publish positive results and consign our probable negativities to nonscrutiny in our file drawers.

  • Fmr ATrealDonaldTrump

    A la Whitman, I am of many minds about Kuhn.

    I do think he brought some new insight to philosophy of science on how science actually works, and in part in ways Shem notes re induction. That said, while some of those insights were perhaps new to philosophy of science in detail, they weren’t new to philosophy in general. See “Hume, David” on the problem of induction. See “Goodman, Nelson” on the new problem of induction. See a new post at my science and philosophy blog (that’s a clickable link; Disqus isn’t good at highlighting that) for how Goodman ties directly to some criticism of Kuhn philosophically.

    I also think real scientific revolutions, vs nuanced shifts, are rarer than Kuhn claims. And, although I think Errol Morris overstated things, I do think Kuhn could and should have been more self-critiquing. That too and more is at that linked blog post.

  • Fmr ATrealDonaldTrump

    I partially agree with you. Not totally, but partially. I also go beyond you with some other observations. See my own post.

  • Anne Fenwick

    I’ve been reading The Ashtray by Errol Morris. While I basically agree with his conclusions, and of course, with the principle that Kuhn’s behavior towards himm was inappropriate (it’s never acceptable to throw hard objects at your students), I’m rather frustrated that he doesn’t seem bothered to spend more time making the case for his conclusions.

  • Anne Fenwick

    “How could Aristotle, one of the world’s greatest philosophers, be so wrong about so much of physics?”

    That is assuredly the dumbest question I’ve ever come across, on so many levels it hardly seems worth tackling. For a start, Aristotle’s main claim to one of the world’s greatest philosophers is that he covered a lot of ground. That gave him the opportunity to be wrong about almost everything, for the simple reason that he was working, not so much with different information, but with much less information.

  • As I hope this post demonstrates, I’m a big fan of Kuhn. But I also think Errol Morris is a phenomenal talent. I fear that reading The Ashtray is bound to make me feel worse about Morris than about Kuhn, and I don’t want to push my luck.

  • Thanks for your contribution.

  • Jim Jones

    MOO, but science is the study of the repeatable.

  • Fmr ATrealDonaldTrump

    Anne, I addressed that in my main comment, which I think was posted after yours as well. Morris is … interesting, but I think overstates the case. Plus, at my linked blog, I also discuss what others think of Morris’ motivation.

  • Fmr ATrealDonaldTrump

    See my first-tier comment. I don’t think it’s a “strawman,” but … I do think he overrates how many real “paradigm shifts” there are, as in ones that have a major change in POV, are incommensurate, and are controversial.

  • Fmr ATrealDonaldTrump

    While I don’t totally agree with Shem on what Kuhn claimed to have shown, do I also think some people act like science fans rooting for “Team Science”? Hellz yes.

    Do I think some of those people — and some professional scientists — think science should stick its nose in and make pronouncements where it has no business, most branches of philosophy? Hellz yes.

  • Fmr ATrealDonaldTrump

    Well, that person must have said something stupid, as Shem made his polite talk to the hand reply and either they or he then deleted the comment!

  • Fmr ATrealDonaldTrump

    I don’t think Kuhn is talking about that. And, some of this is not new. After all, it was the man who invented quantum theory, Planck, who said that all it took for a major new theory to be accepted was an old generation of scientists to die off.

    I see it as an “ownership” issue, not a bias issue. Scientists engage in motivated reasoning just as much as “movement skeptics” and Gnu Atheists. And even me, maybe.

  • Rann

    I feel he over-exaggerates what he calls shifts. Also, one aspect that he downplays is how, as technology changes and improves its observational ability, it has helped refine most of science.

  • Rann

    I think he overrates and over-exaggerates both the nature and the degree of paradigm shifts into something of a strawman.

  • TXGunner1

    Correct me I am wrong but an anomaly is generally something that occurs so infrequently that it is statistically insignificant. Why would you not ignore that?

  • I feel like he’s trying to make a generalized statement questioning the validity of scientific knowledge as we know it without doing his homework.

    I don’t think Kuhn ever “questions the validity of scientific knowledge.” All he does is point out that scientific inquiry has to be seen in its social and institutional context.

    People who idealize science have a tough time with this, but their umbrage is misplaced. Kuhn isn’t saying science doesn’t work or scientific inquiry isn’t worthwhile. He’s just making us look at the messiness of how our scientific knowledge has developed, and get rid of our illusions about science’s inexorable march toward Truth.

  • Okay. If in your career you haven’t seen the kind of upheavals Kuhn describes, he’d probably say you’ve been involved in the “normal mode” of inquiry, the problem-solving activity within the bounds of an established paradigm.

    You can’t deny, though, that there are examples of pretty big shifts in scientific thinking even in our day. The Big Bang theory, for one, wasn’t just a set of incremental refinements of an existing idea like the Steady State construct; it basically rewrote the history of the universe.

  • Milo C

    Then I agree with Kuhn, social and institutional context should be considered in many fields. A researcher’s work will be influenced by his unconscious bias*, something the scientific community as a whole is beginning to give proper consideration when evaluating that research, but not in all areas equally. And of course this diligence must continue because no lab or researcher can be without fault.
    * = I include here the adherence to schools of thought that can blind researchers to possibilities outside of their paradigm.

  • Thanks for the link to your post. I agree that putting Kuhn on a pedestal is problematic. My admiration for him rests on his driving the final nail into the coffin of Whig Science. It’s just not possible anymore to claim that science is algorithmically progressing toward Truth. His description of the social-institutional context of scientific inquiry makes the idea of “self-correcting” science seem like magical thinking.

    He certainly disowned the relativism that people accuse him of pushing in his work. But that, to me, is one of its strengths. When scientists are looking at the same data but drawing different conclusions, then it’s obvious that scientists have to exercise a lot of ingenuity and persuasion to fill the data-theory gap Kuhn describes.

  • It’s not as if Kuhn thinks that paradigms are a bad thing, or that they’re synonymous with bias. The irony is that commitment to a paradigm’s assumptions are what fuels all the problem-solving activity that generates useful information and applications, as well as genuine scientific progress.

    This goes a long way toward debunking the myth that scientists are “constantly testing their theories.” Most of their activity is validating their theories, because they’re generating meaningful results. The extent to which they’re making reality fit their models only becomes evident when the anomalies pile up enough to make the need for a revision of the paradigm necessary.

  • Rann

    What you see as upheaval as Kuhn describes rarely actually happens in science circles….even the Big Bang isn’t as much an upheaval as most think.

    Continental drift (I’m an earth scientist)was on the backburner for centuries, growing in strength as bits of evidence started showing it was very definitely true. The “unheaval” moment as the public knows it was when Tuzo Wilson was able to show the actual mechanism of plate techtonics due to the then-recent dives to the mid-Atlantic ridge and mapping of geomagnetic mapping of the ocean floor explaining a hypothesis that had been developing for ~60 years.

    The Big Bang Theory (as opposed to the now ending TV series) started in 1922 when Friedmann proposed an expanding universe. 5 years later, Father Lemaître found evidence to show it was expanding.

    When such things are causing upheaval in the general population’s view of the world, they have been well-argued in the scientific community – based on showing the evidence and working the hypothesis until it explains explaining as much as they can. The Steady State was there only because there was a failure of imagination. The theory of continental drift was the same… with people thinking of continents as icebergs floating in open water, rather than a like a ship stuck in pack ice. As I said, I think Kuhn overstates and exaggerates his thesis.

  • Milo C

    I’m not sure “constantly testing their theories” and “validating their theories” are mutually exclusive. There are those that repeat experiments to validate (give a higher level of confidence to) results, and that’s good.
    Most research builds on previous results, meaning their work is not designed to test the previous results again, it is testing related hypotheses. So I think the second statement, “Most of their activity is validating their theories”, is misleading.
    I also think you’ve contradicted yourself. Commitment to a paradigm cannot be responsible for all problem-solving or they would never change. Rejection of a paradigm is often the most dramatic or easily-identifiable point in scientific progress.

  • His description of the social-institutional context of scientific inquiry makes the idea of “self-correcting” science seem like magical thinkinga bit more complicated than originally assumed.

    Hey, I know you’re a fan of Latour, so you surely know that the sense of self-correcting science is recoverable so long as you don’t overly sentimentalize the position of the scientist and understand that they are human and have egos. Sometimes science progresses only with the death of the prior generation, but it is still science, self-correcting.

  • Rann

    Kuhn gets hung up on “established paradigms”.

    The Big Bang didn’t really re-write a paradigm……. unless you believed in the Biblical story of Genesis (which, when you think about it, is only an outline at best). Hubble threw his famous name behind the Steady-State theory so it became semi-popular in the science community and it carried weight in popular culture – replacing Einstein’s static In the astronomy world, Lemaître had already shown the universe was expanding…… what wasn’t known was whether the expansion constant, increasing or decreasing.

    They were operating with the true normal mode of science – exploring and discovering new facts about the universe, wherever they may lead. Sometimes, it is something new, sometimes it’s confirming others’ observations, sometimes it is filling the gaps. The real big happening when a new hypothesis such as the Big Bang comes out is we get a new large collection of questions to answers and predictions to confirm or deny!

  • Fmr ATrealDonaldTrump

    Per an update on my blog post which includes the linked Venn diagram, while Kuhn, IMO, even if he pushed things, did have a number of good ideas to say about philosophy of science in general, he doesn’t have much to offer about scientism, because scientism isn’t really a paradigm within scientific thinking from which to shift. It is to a degree, but, really, it isn’t. https://photos.app.goo.gl/BHUaqkkiBVHfZMj96

  • abb3w

    It’s been a while since I read Kuhn (mainly, his book “The Structure of Scientific Revolutions”). My impression is that his work seems sounder in terms of discussing the psychology of science (or more exactly, scientists) than the philosophy of science. Particularly, I find his claim of incommensurability of competing paradigms to be what seems its most basic flaw. He also seems long-winded.

    I consider Popper’s work better from a philosophical stance, but still significantly flawed; for both, I consider the writings of advocates seems to tend more flawed than the originals.

  • abb3w

    For example, it took quantum physics at least 30 years to emerge due to the slow accumulation of experimental observation and theory.

    Plus a certain number of the old school dying of old age, per Planck’s cynical observation.

  • abb3w

    Parsimony seems a sounder philosophical measure than falsification, particularly as falsification may be classed as a special case of non-parsiomony.

    Popper’s fans seem to focus on falsification, to the neglect of Popper’s observations about parsimony. (Not that I think all his speculation about the philosophical function of parsimony is entirely correct.)

  • abb3w

    It’s just not possible anymore to claim that science is algorithmically progressing toward Truth.

    I think this overstates the difficulties. There clearly are flaws in the human implementation of the “algorithm”, a non-monotonic sense of “progress” required, an underlying limit from finite computational resources that leads to imprecise heuristic approximations in the course of such “progress”, and a final nuisance from the Halting Problem, nevertheless an algorithmic expression can be presented.

    When scientists are looking at the same data but drawing different conclusions, then it’s obvious that scientists have to exercise a lot of ingenuity and persuasion to fill the data-theory gap Kuhn describes.

    I’d term it “substituting handwaving for mathematics”.

  • abb3w

    Correct me I am wrong but an anomaly is generally something that occurs so infrequently that it is statistically insignificant.

    …in one sampling mode. Change the sampling mechanism, you may encounter a different frequency distribution. In isolated environments, samples of Iron-56 tend a lot less subject to self-heating than similar volume samples of Plutonium-238.

  • abb3w

    Repeatability just makes the bookkeeping easier.