jump to navigation

Kuhn’s Demon, or: The Iconoclastic Tradition in Science Criticism January 21, 2013

Posted by Will Thomas in Ideology of Science.
Tags: , , , , , , , , , , , , , , , , , , , , , , , , , , , ,

The message that scientists are human and that science is messy is crossing the ether once again. Of late there has been a certain degree of excitement percolating among historians and science-studies scholars over the #OverlyHonestMethods hashtag on Twitter. It seems that scientists are sending out tweets about their work that happen to nicely coincide with the images of science that historians would like to present to the world. Twitter is, of course, the world hub for “trending” topics, and this particular trend is already simmering down. But, before the moment is completely gone, it might be worthwhile to reflect briefly on the history of the ideas 1) that there exists a dominant image of science as a pristine and rigidly ordered activity, and 2) that the negation of this image would be broadly beneficial to scientists and society.


Is it idol-smashing time again already?

Michael Bycroft has already posted on the OverlyHonestMethods issue, discussing how the trend, far from presenting a newly accurate view of science, is actually a very partial image of how science works, and is, therefore, misleading in its own way. This point is fairly easy for our colleagues to take, because the point of highlighting the OverlyHonestMethods trend in the first place was that it serves as a corrective to some more dominant point of view, and, therefore, was never intended to be a full picture. Indeed, Bycroft’s post is itself labeled a “corrective” to what was already a corrective. The thing about debates over the value of correctives is that they are not really debates about (in this case) the nature of science at all—or, if they are, they are very bad ones—but about what sort of images of science ought to be presented to various audiences so as to do them the greatest possible service.

If we follow these conversations, it swiftly becomes clear that nobody has any firmly grounded notions about what sorts of images scientists and the public hold or need, if indeed they need any at all. Instead, it becomes apparent that commentators’ perceptions of the need for certain kinds of images depends not only on their particular view of the world and its problems, but, more importantly I would argue, upon what moral they see themselves as qualified and called upon to convey to whomever will listen.

In this post, I would like to offer a criss-crossing trail of historical bread crumbs indicating some of the avenues the tradition of offering “correctives” to a dominant conception of “science” has gone down over the past 50-odd years. The subject demands more rigorous treatment, but this post will do for now for our purposes.

polanyiThe earliest clear expression of this tradition that I’ve seen is in physical-chemist-turned-sociologist Michael Polanyi’s (1891-1976) 1958 work, Personal Knowledge (based on his 1951-52 Gifford Lectures at the University of Aberdeen), which stressed the role of things like craft knowledge, suspicion, and perception in scientific work. Polanyi made his iconoclastic intensions immediately clear in his preface (vii):

I start by rejecting the ideal of scientific detachment. In the exact sciences, this false ideal is perhaps harmless, for it is in fact disregarded there by scientists. But we shall see that it exercises a destructive influence in biology, psychology and sociology, and falsified our whole outlook far beyond the domain of science. I want to establish an alternative ideal of knowledge, quite generally.

Polanyi was a well-known opponent of the Marxist movement to “plan” science (itself a much misunderstood topic which we should discuss another time), and his conception of the importance of subjectivity in science was central to his conviction that science should not be planned. Polanyi’s book was also heavily influenced by psychological concepts, which problematized the relations between perception and belief. For more on Polanyi, have a look at Mary Jo Nye’s Michael Polanyi and His Generation: Origins of the Social Construction of Science (2011).

As Nye recommends, and as the reference to psychology demands, we can turn swiftly to Thomas Kuhn (1922-1996), who made rather legendary use of Gestalt psychology in the formulation of his concept of the “paradigm”. And it so happens that on page 1 of The Structure of Scientific Revolutions (1962), Kuhn defined an explicitly iconoclastic “role for history”:

History, if viewed as a repository for more than anecdote or chronology, could produce a decisive transformation in the image of science by which we are now possessed. That image has previously been drawn, even by scientists themselves, mainly from the study of finished scientific achievements as these are recorded in the classics and, more recently, in the textbooks from which each new scientific generation learns to practice its trade.

Kuhn’s casting of himself as an exorcist of the textbook-spawned demon by which “we” are evidently “possessed” deserves notation—and it makes for a cool blog-post title—but, we can also follow his more specific “won’t somebody please think of the children” line into the smutty ’70s and Stephen Brush’s 1974 Science article, “Should the History of Science Be Rated X?”1 In his piece Brush suggested that a role for science education might be to instill students with pristine ideals, but that, ultimately, it might be better to let history tell them the facts of life before they hear them from someone else.



Backtracking, Kuhn also mentioned scientific papers, which leads us directly to Peter Medawar’s 1963 BBC broadcast-turned-essay, “Is the Scientific Paper Fraudulent?” Medawar (1915-1987), a Nobel Prize-winner and a well known science writer represents a figure we used to see more often than we do now in these conversations: the renegade scientist who dares Speak the Truth. Of course, for Medawar, scientific papers weren’t really fraudulent; rather:

The scientific paper is a fraud in the sense that it does give a totally misleading narrative of the processes of thought that go into the making of scientific discoveries. The inductive format of the scientific paper should be discarded. The discussion which in the traditional scientific paper goes last should surely come at the beginning. The scientific facts and scientific acts should follow the discussion, and scientists should not be ashamed to admit, as many of them apparently are ashamed to admit, that hypotheses appear in their minds along uncharted by-ways of thought; that they are imaginative and inspirational in character; that they are indeed adventures of the mind.

Probably the most famous renegade scientist is James Watson (b. 1928). On this blog, I’ve already discussed the widespread media stir caused by his 1968 memoir, The Double Helix, as well as sociologist Robert Merton’s rush to declare that the “human” and competitive image of science stressed in discussions of the book were not pathologies of twentieth-century science, but rather a constant feature stemming from the sociological structure of scientific inquiry, which he reckoned has been hidden by “pious biographers” who wrote in “sapless prose” and turned scientists into—quoting mathematician and Newton biographer2 Augustus de Morgan (1806-1871)— “monsters of perfection”.

Pretty-decent historian of science R. A. Fisher in 1924

Pretty-decent historian of science R. A. Fisher in 1924

But getting back to scientific papers and the idea of fraudulent scientific presentation, we might also take our story all the way back to statistician and genetic theorist R.A. Fisher’s (1890-1962) 1936 paper in Annals of Science, “Has Mendel’s Work Been Rediscovered?” which tackled, among other things, a footnote written by genetics pioneer William Bateson (1861-1926) suggesting that Mendel rigged his reports of his experiments. Among his many intellectual peculiarities, Fisher was actually a fairly sensitive historian of scientific reasoning and practice, and his original paper is worth reading.3 One of the points he made was—this will be familiar—that data may be presented in a certain way for expository and pedagogical purposes (119):

Mendel was an experienced and successful teacher, and might well have adopted a style of presentation suitable for the lecture-room without feeling under any obligation to complicate his story by unessential details. The style of didactic presentation, with its conventional simplifications, represents, as is well known, a tradition far more ancient among scientific writers than the more literal narratives in which experiments are now habitually presented. [?] Models of the former would certainly be more readily accessible to Mendel than of the latter.

Effectively, those pesky textbooks again. Fisher ultimately exonerated Mendel, and presented the case as an example of what careful historical scrutiny can say about how scientific experimentation and reasoning works. Similar points would be made by historians Gerald Holton in 1978 and Allan Franklin in 1981 concerning Robert Millikan’s arguments and handling of data in his famous oil-drop experiments.


However, historical questions surrounding Mendel’s and Millikan’s treatment of data became case examples in journalists William Broad and Nicholas Wade’s 1982 book, Betrayers of the Truth: Fraud and Deceit in the Halls of Science. The book—again—sought to overturn a “flawed ideal” (the title of their first chapter) in order to present a picture of how “science really works,” as opposed to the “conventional wisdom,” which held that “science is a strictly logical process, objectivity is the essence of the scientist’s attitude to his work, and scientific claims are rigorously checked by peer scrutiny and the replication of experiments” (7). Read carefully, Broad and Wade’s treatment of their cases was somewhat more subtle than the book’s inflammatory (and, indeed, deceitful) language might suggest. Nevertheless, the idea that Mendel and Millikan committed scientific malpractice by cooking their data swiftly entered the broader lexicon of readers of popular science.

Trained historians have generally been more careful. In The Private Science of Louis Pasteur (1995), Gerald Geison used the “discrepancies” between the published record of Louis Pasteur’s work and the “private science” recovered from Pasteur’s notebooks as an opportunity to analyze the social and intellectual relations between the private and public in science, and explicitly denied that “private documents somehow permit direct access to the ‘real’ work of the scientist” (15). At the same time, Geison did set up the usual foils: the book would show “how much the standard Pastorian legend needs to be qualified, even transformed” (4); how great scientists like Pasteur were not great in that they were “the ‘first’ to advance concepts that look ‘right’ in the light of current knowledge, nor insofar as they adhered to the precepts of an allegedly clear-cut Scientific Method that their lessers and rivals presumably violated” (10); indeed, it would “gradually become clear that some of Pasteur’s most important work often failed to conform to ordinary notions of Scientific Method” (16).

If the iconoclastic tradition was a rhetorical maneuver that scientists, sociologists, historians, and journalists could use to provide an aura of cogency about their commentary, it was also a weapon that could be turned inward. As we have seen on this blog, in the 1970s the sociologist Michael Mulkay accused Mertonian sociology of reinforcing a rhetorical “ideology of science,” which did not resemble the actual practice of science, in order to enhance its authority. (Sociologist Steve Fuller has more recently suggested that Thomas Kuhn was engaged in a similar pursuit.)

Proponents of the sociology of scientific knowledge would make great use of the iconoclastic tradition. In Knowledge and Social Imagery (1976), David Bloor would accuse prior sociologists of lacking the iconoclastic “nerve and will” (4) to apply their tools to the domain of scientific knowledge. Meanwhile, where history had been traditionally cast as a tool used to destroy false images, in the late 1970s Barry Barnes and Steven Shapin suggested that historians’ work had itself been systematically corrupted by a false image of science, specifically a presumption of its “individualistic epistemology”. It would, instead, be the adoption of insights from SSK that would allow historians to create proper, “naturalistic” images of science henceforth.


A critique turned into history turned back into critique

In Leviathan and the Air-Pump (1985), Shapin, with historian Simon Schaffer, would identify the seventeenth-century rise of experimental philosophy as a particular historical moment when a false image of science arose, which corresponded to a denial of knowledge’s “conventional and artifactual status,” and which was only in the “late twentieth century” being “called into serious question” (344). In The Pasteurization of France (1988), Bruno Latour seized on this periodization to identify it as the moment where “we … created, in a single movement, politics on one side and science or technoscience on the other” (5). Recast as a false separation of “nature” from “culture,” in 1993 Latour would take this false image of the world to define “modern” thought, and thus his own work as tantamount to an era-defining intellectual revolution.

So, I suppose the point of all this is that we probably should be aware of a solid half-century of critical tradition weighing down upon our heads before we trust any instincts we might have that what the world genuinely needs from historians is an Honest Discussion about the nature of scientific work, and that it might be about to happen any time now. That’s what Polanyi thought.

1I always feel guilty wheeling this article out whenever the subject of the “role” of history of science comes up, because it’s the only time I ever mention Brush on this blog; I really ought to find an opportunity to talk about his work in the history of physics….

2Iconoclasm and Newton is a subject all its own, involving, among others, economist John Maynard Keynes, who in 1946 published an essay entitled “Newton the Man”. On this subject I will pass the ball off to Rebekah Higgitt’s blog post, “Newton and Alchemy: A Constant Surprise?”

3Among other interesting points in the paper, Fisher discusses the “polemical use” of the concept of “rediscovery” of genetics, which makes him sound bizarrely like a recent-vintage professional historian of science.


1. William Burns - January 21, 2013

What a fun post which I really enjoyed reading.

2. Bycroft demonstrates how to respond to #overlyhonestmethods | The Bubble Chamber - January 21, 2013

[…] science-in-practice and science-in-publication see Will Thomas’s “Kuhn’s Demon, or: The Iconoclastic Tradition in Science Criticism” at Etherwave […]

3. Michael Bycroft - January 29, 2013

And that’s just the (second half of) the twentieth century! Here are two earlier examples:

1. In the article you refer to by Medawar, the author explicitly defended Whewell’s view of scientific reasoning against that of John Stuart Mill. So it is no surprise that Whewell had himself written against what he saw as Mill’s simplistic, logic-chopping, history-blind account of scientific discovery (see Whewell’s “Of Induction, with Especial Reference to M. J. Stuart Mill’s ‘System of Logic,'” 1849. The Mill-Whewell debate is nicely covered in Laura Snyder, “Reforming Philosophy,” Chicago UP, 2006).

2. To go back to the eighteenth century, Joseph Priestley in his “History and Present State of Electricity” (1767) criticised scientists who in their published works exaggerated the ease by which they reached their conclusions. Priestley thought this made their conclusions seem like the consequence of the mysterious workings of genius, and he worried that this would discourage ordinary mortals from taking up science. Here’s what he wrote about the form of Newton’s “Opticks”, for example:

“And were it possible to see in what manner he [Newton] was first led to those speculations [in the “Opticks”], the very steps by which he pursued them, the time that he spent in making experiments, and all the unsuccessful and insignificant ones that he made in the course of them; as our pleasure of one kind would be increased, our admiration [for Newton] would probably decrease” (vol. II, 168).

It is interesting to see that the iconoclasm of one age is the orthodoxy of another. Priestley was attacking what he called the “synthetic” method of telling a history, in which inquiry consists in finding proofs for a hypothesis formulated in advance. With Whewell the situation was reversed: he criticised Mill for failing to see the formative role of hypothesis in induction. In the twentieth century, Medawar criticised scientific papers on the same grounds, invoking Popper as well as Whewell. But now things have gone full circle, and *Popper* is the target of iconoclasts: nowadays we join Priestley in criticising those who think that science consists in the testing of well-defined hypotheses formed in advance.

To be fair on the iconoclasts, this does not mean that they are always just repeating the orthodoxy of the age that preceded them, or that the iconoclasts of one age contradict the iconoclasts of the next. Perhaps the orthodoxy keeps oscillating between two extremes–roughly, the inductivists and the deductivists–and both of those extremes are simplistic and bare-boned; the iconoclasts of any given age simply attack the prevailing orthodoxy.

Still, the moral of your post stands: there’s nothing new about saying that science is messier, and scientists more human, than we might infer from the writings of scientists or philosophers of science.

Will Thomas - February 1, 2013

Michael, thanks for the comment, and sorry for the tardy reply. In addition to being mad busy, I’ve been mulling over how the nature of the iconoclastic tradition changes through time: who is the audience for iconoclastic rhetoric, what form do the idols take (scientific papers, textbooks, public imagery…), what specific danger does idolatry risk, and so forth.

I am tempted to say that earlier manifestations tend to focus more on debates between philosophers over what constitutes proper epistemology, but, then, Priestley (among others) quite clearly connected his ideas about natural philosophical procedure to more general problems of legitimacy (as Simon Schaffer and Jan Golinski have shown), and, as Whewell and Mill were both political philosophers, I would not be surprised if they do the same in some part of their writings I’m not familiar with. (I happen to have a copy of Snyder’s book in my office, but I haven’t dug into it.)

I often wonder about “oscillating” discourses, whether they really oscillate, or if we just happen to notice one side rather than another of various criss-crossing “corrective” efforts through history. While poking around for this post, I definitely noticed a lot of concern over inductivism in the 1950s and ’60s when making note of the more specific targets of iconoclasm, but, then, there’s always a lot of anti-deductivism going around, too.

In some ways, this is similar to the (no doubt related) discourse about “science” and its trouble fitting in with “society” or “culture”. William Burns and I were just going over Schaffer’s recent radio discussion of Huxley vs. Arnold, and its different, more secular echoes in Snow vs. Leavis, and, then, what I take to be its further echoes in science studies.

At any rate, I don’t have any firm conclusions to draw at the moment, but, again, thanks for bringing up the long history.

Michael Bycroft - February 4, 2013

Thanks for this reply. Snyder certainly goes beyond the debate between Whewell and Mill about scientific method. From the introduction to her book:

“The differences between Mill and Whewell over induction cannot be understood without setting them in the broader context of their conflicts over the proper way to reform society. Both men believed that proper scientific method could aid in renovating not only science but also morality and politics (4).”

I’m not sure if this counts as “general problems of legimitacy,” but it certainly brought their methodological debate to a broad audience.

The other resemblance between Whewell and Priestley on the one hand, and present-day iconoclasts on the other, is of course that their idol-bashing evidence comes in large part from the history of science.

Michael Bycroft - February 5, 2013

PS. You might be interested in this article by Larry Laudan, if you have not across it already:

Laudan, Larry. “Why Was the Logic of Discovery Abandoned?” In Thomas Nickles (ed.). Scientific Discovery, Logic and Rationality. Boston Studies in the Philosophy of Science, 49 (Boston: D. Reidel, 1980) pp.173-183.

Laudan argues that since the seventeenth century there has been an oscillation between believers in “generative” and “consequentialist” modes of inquiry. In the former the theory comes last, being “generated” from prior facts; in the latter the theory comes first, and is assessed by its consequences. Laudan has an interesting and ambitious theory according to which consequentialist logics prevail when scientists are interested in unobservable phenomena (atoms and molecules and the like) whereas generative logics prevail when scientists restrict themselves to descriptions of observable phenomena.

Laudan treats this purely as an oscillation within the philosophy of science, without reference to the political and historiographical aspects of the debate.

Will Thomas - February 5, 2013

Thanks very much for both comments, Michael. I’m pretty well convinced now that the tradition traced here from Polanyi is continuous with the same one we can trace from Priestley. (From Priestley to Polanyi would be a great review article…) The Laudan looks interesting, and I have not read it. A forthcoming post deals with the tension between Galison and Kent Staley on disunity vs. unity of science, and will include a discussion of Galison’s invocation of rational reconstruction as historiographical anathema. Part of that discussion will address Lakatos’s prefatory discussion of the disjuncture in the relationship between philosophy and history of science. It will also jibe fairly nicely with my two-parter on Fisher.

Michael Bycroft - February 5, 2013

Looking forward to it!

4. 10 links for January 2013 — News from Somewhere - February 1, 2013

[…] 1872-73 Epizootic or the Canadian Horse Distemper. Discovery Institute: Darwin and Eugenics Again. Norovirus and the reporting of epidemics through history. Water stories. The Science Museum is pants. The Popularisation of Science: 1890-1914. How To Make a Victorian Villain (or the Tale of Isaac Baker Brown) Part 1 and part 2. History of Science on Stamps. History Meets Biology at the AHA and Highlights from the 2013 Annual Meeting of the American Historical Association. Kuhn’s Demon, or: The Iconoclastic Tradition in Science Criticism. […]

5. Joachim - February 9, 2013

The most fundamental iconoclasm against method still seems to be “Against Method,” indeed, by Feyerabend.

Will Thomas - February 9, 2013

Unquestionably. It was almost sort of a private challenge to myself to write this post without referring to Feyerabend! I almost mentioned that he provided one of the approving blurbs for Broad and Wade’s book, but the post was too long and digressive already.

6. The Giants’ Shoulders #56 | The Dispersal of Darwin - February 16, 2013

[…] Will Thomas, “Kuhn’s Demon, or: The Iconoclastic Tradition in Science Criticism,” Ether Wave Propaganda, January 21, 2013, https://etherwave.wordpress.com/2013/01/21/kuhns-demon-or-the-iconoclastic-tradition-in-science-criti…. […]

7. Will Thomas - February 24, 2013

For a while I’ve had a mind to create posts that will serve as repositories for examples of certain historical or historiographical phenomena. So, here is an example of Kuhn’s demon I’ve just run across, complete with his educational corollary.

Joseph Agassi, “The Rationality of Discovery” in Thomas Nickles (ed.), Scientific Discovery, Logic, and Rationality Boston Studies in the Philosophy of Science 56 (D. Reidel, 1980)

“Both these theories [that rationality equals algorism, and that it equals quasi-algorism] do untold damage, especially in high school and college education… But, first and foremost, the view of unanimity in science, the view that in science at least, one need not struggle in order to get a hearing for one’s discovery, should be recognized as empirically refuted, and its refutation should be made public knowledge.”

“The descriptive phrase of my title, ‘the rationality of discovery’, sounds distinctly odd … because we are all turned to consider discovery as scientific and hence as obviously rational….”

“The present essay is an attempted criticism of this popular answer [that scientists unfailing recognize any rational discovery almost as soon as it is made] — or rather of the complacency behind it, which, I think, is a form of insensitivity to much human suffering, such as the suffering of those who for years try to bring some discovery or another to the public eye.”

“The most common idea of a rational procedure is the idea of algorismic solutions to given problems. This idea is one of the most harmful potent myths of the Western world. How school children acquire it is a very interesting question which I have not sufficiently studied to discuss here and now…. By and large, school children accept algorisms that they do not understand, such as those of multiplying large numbers by large numbers…. Let me report that many students were surprised to discover in my presence, empirically and by introspection imposed by my interrupting their calculating activities, that they guess while solving division problems; and and at times the discovery came with a shock.”

“How true is the myth of scientific unanimity or consensus? If it is true, can we hope that the discovery of educational defects and of ways of curing them will become unanimously accepted overnight? If not, what then?

Here we see the force of Kuhn’s version (1962) of the doctrine of unanimity or consensus. He says some fields are scientific and have a consensus, or a paradigm, perhaps two camps with consensus in each, or a bi-paradigm… What to do about a misconceived doctrine that is endorsed by the consensus? Make a scientific revolution and change it. But how? Who makes the revolution? The leadership, of course! Who is a leader? One who is accepted by consensus, of course! Does this tell me how to act rationally? No. Michael Polanyi, whose ideas Kuhn espouses (Polanyi, 1958), gives an example….”

8. Will Thomas - March 5, 2013

More-or-less the same thing as Broad & Wade, using many of the same case examples: Alexander Kohn, False Prophets: Fraud and Error in Science and Medicine (Oxford: Blackwell, 1986).

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: