Entry into a discoverer's culture often proves acutely uncomfortable, especially for scientists, and sophisticated resistance to such entry ordinarily begins with the discoverer's own retrospects and continues in perpetuity....
Systematic distortions of memory, both the discoverer's memory and the memory of many of his contemporaries, are a first manifestation of resistance. Another, regularly found among members of later generations, is the attribution of real or supposed anomalies in the discoverer's behavior to "confusion."...
The famous paper that announced the Bohr model of the hydrogen atom was submitted from Copenhagen on 6 March 1913 and published the following July, the first installment of a three-part series. I first read it during the fall of 1962 in preparation for interviews with its author. Not surprisingly, the paper includes a full description of the quantized Bohr model for the hydrogen atom as it would be taught in an elementary physics course today. But it also includes a number of phrases incompatible with that model. In particular, Bohr sometimes wrote as though the hydrogen spectrum were emitted by an electron falling into the ground state from outside the atom and strumming all the stationary states that it passed along the way.
These anomalous remarks, together with Bohr's repeated assertion that he had not known the Balmer formula until February 1913, suggested an unexpected hypothesis, subsequently fully confirmed by the discovery of an unpublished manuscript. Many months before he attempted an explanation of spectra, Bohr had developed a quantized version of the Rutherford atom for chemical applications of the sort made familiar by J.J. Thompson. That model, which I was quite sure had had only a ground state, provided the basis for the second and third installments of the 1913 series. The first, which developed the Bohr model for hydrogen and derived the Balmer formula from it, was a last-minute insertion.
My first few interviews with Bohr dealt with the background for his atomic model, and I asked what sorts of connections he had made between the Rutherford atom and the quantum during the period before his attention was directed to the Balmer formula. He replied that he could not have had developed ideas on the subject before turning to spectra, and his assistant later reported to me that, after I had left the room, Bohr shook his head and said of our exchange, "Stupid question. Stupid question."
All that occurred at our first interview. For the next one, I included a similar question in a list submitted to Bohr in advance, and it was received in much the same way as the original. One last attempt to retrieve memories of an early quantized Rutherford atom occurred late in the third interview. This time, however, when Bohr said again that there could have been no concrete model without the Balmer formula in hand, I for the first time showed him the passages in his famous paper that led me to enquire. He looked them over and then muttered to himself, "Perhaps it was a mistake to put the paper into print so fast. Perhaps I should have waited until I had it right." Then, he went over to his personal collection of reprints, took from it a paper he had presented to the Danish Academy of Sciences six months after the publication of his original paper, and handed it to me with the words, "It's alright there, isn't it?" About the earlier model not a word was ever forthcoming.
...Faced with apparent anomalies in the work of the discoverer, scientists and at least an occasional historian protect their version of the discovery by invoking the discoverer's "confusion" during the early stages of its emergence. It is only because he was confused, they explain, that his words fail to fit their story.
These appeals to confusion are damaging, but not because discoverers are never confused. Typically, they are, and Bohr's discovery of the Bohr atom is a clear example. When he wrote the paper announcing his discovery, he had two incompatible models in mind, and he occasionally confused them, mixed the two up. No reading of his first reports on his invention will eliminate the resulting contradictions, and those contradictions, which testify to his confusion, provide essential clues to the reconstruction of his route to the discovery. The standard appeal to confusion dismisses those clues, rejects them as challenges to historical reconstruction, and permits the attribution of confusion to stand as the end of the story. That is the first part of the damage.
For the second, more serious part, compare the case of Planck. Again there are anomalies in the early papers; again they provide clues to an unsuspected state of mind; and, again, dismissing them discards evidence essential for historical reconstruction. Thus far the damage is the same. But in Planck's case, unlike Bohr's, the anomalies do not take the form of internal contradictions, and they therefore provide no reason to suppose that Planck himself felt or had reason to feel confused. If it is nevertheless appropriate to apply the term to him, that is by virtue of the second standard use of the word "confused," one independent of the state of mind of the person to whom it is applied.
Consider, for example, the case of a student who, having read a textbook derivation of the black-body distribution law, then wrote it up in a way like the one found in Planck's early papers. That student would be confused, not in the sense of being pulled about by conflicting elements in his thought, but in the sense of having seen only dimly or confusedly the structure of the derivation that had been set before him by the text. That, I believe, is the sense of "confused" in the minds of people, mostly scientists, who complain, for example, that I try too hard to make the thought of a Planck or a Boltzmann logical and coherent. Why, I am repeatedly asked, can I not simply acknowledge that they were confused?
That way of talking about a discoverer makes no sense. Taken literally, it suggests that the discovery, of which its author is to have had only a confused view, had already been made, was somehow already there, in the discoverer's mind. Occasionally that implication is explicit. The discoverer, I am then told, was relying on intuition; his view of his discovery was still so clouded that he could only grope his way to it; that is why he described what he had in mind in such odd and inconsistent ways, appeared so much a sleepwalker as he proceeded towards his discovery.
Doubtless, few of those who explain anomaly by resort to confusion would go quite so far, but all must encounter the identical difficulty. What licensed our calling the student confused was our knowledge of the concepts he brought to the text and of the proper way to fit the two together. If only he had clearly seen that much himself, he would not, any more than we, have described the derivation as he did. When, in the absence of internal contradiction, we apply the label "confused" to Planck, we are again using ourselves as the measure. We assume that Planck brought to his problem the same concepts we do, and we explain his anomalous behavior as we would explain similar behavior of our own. But the concepts we bring to the black-body problem are themselves products of the discovery Planck had not yet made. To claim for them a role in the emergence of his discovery is again to make him a sleepwalker or else clairvoyant. That is an incoherent notion of discovery - one that makes discovery dependent on prior grasp of what is to be discovered. No other result of the resort to confusion is so damaging.--Thomas Kuhn, "Revisiting Planck," HSPS (14: 231-52), 1984, reprinted as the Afterword to the second edition (i.e., U. of Chicago rather than Oxford) of Black-Body Theory and the Quantum Discontinuity, 1894-1912.
― Frank Kogan (Frank Kogan), Tuesday, 5 August 2003 21:42 (eighteen years ago) link
― mark s (mark s), Tuesday, 5 August 2003 21:46 (eighteen years ago) link
― adam (adam), Wednesday, 6 August 2003 20:56 (eighteen years ago) link
― Frank Kogan (Frank Kogan), Wednesday, 6 August 2003 21:33 (eighteen years ago) link
But just as much of the error derives from semi-digested assumptions about what battles are MEANT to be like. The observing experts - who comment on the action in another room, then come in to show what really happened - pointed out (scathingly) that the tactics adopted suited troops with muskets facing cavalry, not legionaries facing tribesmen in chariots: the participants half-knew too much about later-hence-'superior' warfare, and - after being goaded out into the open and completely surrounded by vastly larger numbers - decided that savages would all just run away if their leader was killed.
― Frank Kogan (Frank Kogan), Monday, 15 September 2003 17:36 (eighteen years ago) link
but back to kuhn. another fun thing to talk about is his idea of "normal science" which seems like the biggest easiest thing to contest.
kuhn claims that "normal science" in fields which are ALREADY paradigmatic and NOT in paradigmatic crisis is essentially puzzle-solving -- i.e. finding anomolies and demonstrating that they are not anomolies. i.e. not only within, but FOR the paradigm. kuhn also claims that generally paradigms don't shift within generations but between them and a key reason they are ABLE to shift is young ppl. come in and accept the new ones and the old foax eventually go away.
this is a strong claim. it applies only if ppl. are NOT reflexive about paradigms, which is dodgy. furthermore it applies only if scientists really get to set problems for themselves and if they really choose problems based on puzzle-solving to bolster existing paradigms.
so take the fact say that physics still doesn't have a grand unified theory but instead two incommensurable accounts (quantum and relative) that apply at different scales. physicists acknowledge this a problem, accept it, some work to solve it but many don't. so posit a unified theory is worked out -- this is a paradigm shift, obv. but would scientists die out before young ones accept it!? does this instead mean that physics is in a period of protracted crisis? if so we would expect MOST research, by kuhn's account, to go towards resolving this crisis. but most research does NOT. instead "normal science" proceeds to do different things largely NOT driven by a desire to resolve this massive anomoly but to extend othere discrete areas of knowledge. what areas? is this puzzle-solving of smaller issues? again, largely no. it's investigating the specific properties of certain elements and interactions. but *not* chosen by their key status in resolving anomolies, but rather b/c these elements and interactions are impt. to those giving the $, most of whom even if they're funding pure science are interested in particular applied ramifications. (i.e. if you're the dept of defense, how to make things go boom -- if yr. doing physics related to chemicals in dna, maybe how to understand cancer)
so we have to conclude that kuhn's story either doesn't hold or that "normal science" is nonparadigmatic or that it held in the special cases he examined but not today. which then means we have to ask what made it hold more for say concepts of physical motion from the greeks thru newton than now, which then means we have to say maybe this is because kuhn wasn't constructivist ENOUGH and so he can actually describe an activity called "doing science" that the greeks and newton both did, and this abstracted "doing science" contains in it the seeds of how kuhn can describe what "doing science" involves. which is to say we have to ask what about these historical actors kuhn left out b/c it wasn't "doing science."
another thing we can do with kuhn is ask if his NORMATIVE perscriptions were correct, which argued that a paradigm-driven "normal science" that he described was the "best" type of science to do b/c the loss in freedom of topics to explore was made up by a wealth of specifics.
which can also be asked by positing that paradigm-driven normal science does exist and does provide a specific set of anomoly-related questions to answer and then posing the question -- if one IS to accumulate a specialized body of data is the best way to accumulate this data and theory to choose the specialized areas of research based on anomoly-resolution ANYWAY?
we can also ask do scientists THINK they are doing "normal science" involving anomoly-driven puzzle-solving and do they think this more now that kuhn is widespread than before he was, and does kuhn's dissemination have a role in why ppl. think this is what they are doing more now. and does kuhn's dissemination also mean that they are more reflexive towards their paradigms than before.
this also reminds me that i have to read Leviathan and the Air-Pump.
― Sterling Clover (s_clover), Tuesday, 26 July 2005 05:21 (sixteen years ago) link
― Sterling Clover (s_clover), Friday, 29 July 2005 22:33 (sixteen years ago) link
I've read through your post several times without seeing where Kuhn's story is wrong or why you conclude that physics as it's currently practiced is not paradigmatic. I wonder what you think "paradigmatic" means. The phrase "anomaly-driven problem solving" isn't relevant to the question - it's a straw man, since Kuhn never said that the problem solving in normal science is mostly devoted to explaining anomalies. Whether it is or not is beside the point, at least in regard to the concept "paradigm," given that according to Kuhn all problems within a normal science - not just the anomaly-driven ones - must be approached within the framework of the accepted paradigms, and most use exemplars. The question of whether there is life on Mars is not anomaly-driven, but nonetheless is within the paradigms of modern chemistry and biology. For what it's worth, where Kuhn in The Structure of Scientific Revolutions discusses anomalies (chapter 6), they are not the bread and butter of normal science but rather the precursors to revolutionary science (albeit the result of normal science).
A quick example that illustrates both normal and revolutionary science (and I don't pretend to understand the physics that I'm summarizing): When Planck back in the 1890s and early '00s, doing normal science, tried to derive a distribution law for black-body radiation, he modeled his derivation after Boltzmann's derivation of the entropy and velocity distribution of a gas. So Planck wasn't trying to fix an anomaly, but rather to solve problem B by following a strategy used earlier in solving a similar-seeming problem A. This is an example of what Kuhn means by normal science, and Boltzmann's derivation is an example of what Kuhn means by an exemplar. Again, what makes it normal science of its day is that it uses an accepted model and is within the framework of classical physics. What eventually takes it out of normal science and makes it revolutionary is that one of the adjustments that has to be made to Boltzmann's method in order for it to work on the black-body problem is that energy levels have to be posited as discontinuous; this adjustment blows a hole in classical physics. (By the way, according to Kuhn, Planck didn't recognize at first that discontinuity was a requirement - he had in fact botched his derivation, though this was only discovered later by Einstein, who'd already figured out in 1906 (as had Ehrenfest) that discontinuity was required. Planck himself didn't come around to this view until 1908 (according to Kuhn).)
Your point about money is as irrelevant here is it was upthread. Why shouldn't problem solving go where the money is? If you want to say that Kuhn's account of normal science is wrong, you have to show that, whoever chooses the problems, and however the research is paid for, (1) the science conducted does not model itself upon strategies derived from previous puzzles, and (2) the science is not conducted within the framework of accepted paradigms.
"Pure science" versus some other sort of science (that paid for by the Pentagon, etc.) is your bugaboo, or Fuller's, but it has nothing one way or another to do with Kuhn, who never uses the phrase "pure science" or distinguishes between pure and impure. The question as to whether the science is "pure" or not, and who pays for it, simply has no bearing on the concepts "paradigm" and "normal science."
Mind you, I don't understand modern physics well enough to speak with any competence about whether it's paradigmatic or not, or whether it's going through a protracted crisis or not. Physicists don't seem to be acting as if it's in crisis, at least not in accounts read by this layman. Whether it's in crisis or not, relativity and quantum physics don't seem to be on the verge of being abandoned, and physics hasn't broken into two competing schools, those who support relativity versus those who support quantum physics, has it? Therefore, I don't see how you can say that physicists don't share the same disciplinary matrix/paradigm.
Also, is the lack of a unified field theory an anomaly? I suppose this depends on how one defines "anomaly." My understanding or - misunderstanding - of modern physics is that relativity can't be reduced to quantum terms and vice versa. However, if no one's challenging the terms themselves - i.e., relativity isn't making quantum physics wrong, and quantum physics isn't making relativity wrong - then there's no urgency, and no anomaly. Even if you think there should be urgency, the lack of urgency doesn't in any way undermine the concepts "normal science" and "paradigm," and Kuhn never said that crises must lead to revolutionary science. And - conversely - in the postscript that he added in the second edition of Structure in 1969, he said that he'd overemphasized the role of crisis in provoking a paradigm shift and that scientific revolutions need not invariably be preceded by crisis.
(I'm once again talking in ignorance, but it seems to me that an example of a potential anomaly in physics would be if "dark energy" was determined not to exist, making the accelerating expansion of the universe a true anomaly, i.e., something not yet explicable by the theory of relativity.)
kuhn also claims that generally paradigms don't shift within generations but between them and a key reason they are ABLE to shift is young ppl. come in and accept the new ones and the old foax eventually go away.
Well, first off, even if Kuhn had said this, it wouldn't be an integral part of his ideas. But in fact it's not what he said. He said the shifts sometimes take a generation, implying it's because oldsters are less likely to make the shift than are young 'uns, but he didn't say that no one in the old generation accepts a shift, which would be absurd, since sometimes the guy who starts the shift is in the old generation himself, e.g. Copernicus, Planck. Kuhn's own writing shows examples of adherents of an old theory going over to the new: Lorentz and Jeans, for instance, in regard to the quantum (and Planck too, for that matter). Whether the scientists are old or young, a significant number of scientists in a field must be converted from the old paradigm to the new one, and old people dying off won't have an impact on whether young 'uns convert if the young 'uns don't have good reason to. People aren't born into a new paradigm; they have to invent it, or embrace it.
So your word "key" and your capitalization of "ABLE" are very wrong. The key reason that a paradigm is able to shift is not that old people die off but that the new paradigm does stuff that the old one can't. That is what chapter 12 of Structure is about. The idea that old people have trouble accepting a new theory was old hat (Kuhn: "these facts and others like them are too commonly known to need further emphasis"). Kuhn's reason for bringing it up was to emphasize that shifting one's allegiance to a new paradigm was like a conversion experience, but also that people who resisted the new paradigm were not being irrational - that the functioning of science in its normal state demands that a lot of scientists assume that the paradigm they've got "will ultimately solve all its problems, that nature can be shoved into the box the paradigm provides." In any event, look at pp 150-159 of the second edition of Structure, and you'll see that Kuhn is chiefly concerned not with what prevents people from adopting a new paradigm, but rather with what persuades them to.
Also, a reason that paradigm shifts may take a generation or longer is that it takes a while for the new paradigm to be worked out. That it took until 1928 for quantum physics to pull itself together into a paradigm wasn't due to obstruction by fogies but rather due to the whippersnappers taking that long to come up with a worked-out model. So when we talk of "competing" paradigms, we're being somewhat misleading, since the "new" paradigm doesn't fully exist yet: It's under construction. Even if everyone has abandoned the old paradigm, the revolution isn't over until the construction is substantially complete.
Here's a thought (based on what Kuhn himself said): What's a crisis for one person might not be a crisis for others, hence what's a reason for overhaul for one person won't be for another. Was astronomy really in a crisis in the early 1500s due to the need for calendar reform and to the imprecision of Ptolemy's system? Hadn't these "problems" been known for years, without anyone revamping cosmology due to them? Maybe it was a crisis for Copernicus but wasn't a crisis for astronomy and physics until the Copernican ideas began to be plausible to more than a few intellectuals; it was then that you had to come up with a new concept of motion. So you could say that Copernicus caused the crisis more than that he responded to one. To return to the quantum discontinuity, in Kuhn's account it wasn't introduced in response to any big crisis but just because in 1906 Einstein and Ehrenfest realized you couldn't derive Planck's distribution law without it. This is what caused all the excitement and led to rethinking a bunch of other stuff as well.
Again, the pieces by Kuhn that I recommend most are "Revisiting Planck," which is reprinted as the Afterword to the second edition (i.e., U. of Chicago rather than Oxford) of Black-Body Theory and the Quantum Discontinuity, 1894-1912, and "What Are Scientific Revolutions?" which is the first article in The Road Since Structure. I still feel that you and I are talking passed each other, and that you don't have Kuhn's ideas fully in sight. Some preconception or need is blocking your vision, or distorting it. Kuhn's most crucial ideas - his opposition to teleology, his belief that scientific knowledge does not accumulate incrementally, his quasi-Darwinian account of how different scientists vary in their response to anomalies, his idea that scientists use models rather than following rules, and his understanding that a scientific revolution doesn't simply involve adjusting theories to facts, since the new theories create different "facts," which is what makes them "incommensurable" with the old - haven't yet piqued your interest. One thing you need to do is to set aside for the time being questions relating to mainstreams and margins and whether good stuff gets shunted aside when paradigms dominate and whether power relations as they reveal themselves in science are fair or not. You might even want to set aside for the moment the question of whether Kuhn is right or not and rather make an effort to get inside of Kuhn's head - become Kuhnian, as it were, in relation to Kuhn. Study Kuhn's writing from "the viewpoint that gives [his] opinions the maximum internal coherence and the closest possible fit to nature." Assume that, within his framework, he knew what he was about. E.g., he lived until 1996, and he certainly would have noticed that physics hadn't come up with a unified field theory, and he would have known what postwar physicists spent their time working on - he'd been trained as a physicist, after all. So you have to ask yourself why it didn't, from his point of view, jump out at him that his story either doesn't hold or that normal science is nonparadigmatic or that it held only in the special cases he examined but not today. Assume that - in fact - from his point of view, his story does hold today and that normal modern physics is paradigmatic. Read him with an eye towards finding out what that point of view is.
― Frank Kogan (Frank Kogan), Sunday, 7 August 2005 03:10 (sixteen years ago) link
― ulrich schnauss, Sunday, 7 August 2005 03:16 (sixteen years ago) link
I can't make sense of this statement. It's written in some sort of code that I was never taught.
Anyway, for the sake of argument, let's say that, except where I explicitly dispute Kuhn, every idea that I attribute to Kuhn on this thread can also be assigned to me. And of course any of the ideas on this thread that I seem to be presenting as my own can also be assigned to me, even if I stole them. So now let's rewrite your sentence as follows:
kogan relativizes the cartesian divide without overturning it, and therefore renders it general.
So, now, explain and support this contention, using the actual words on this thread as your evidence. Not from anywhere else; just this thread. If you think an idea or an assumption is Kuhn's/Kogan's, check the idea against the words on this thread. And if a sentence or two on this thread seems to point towards a particular belief being Kuhn's/Kogan's, check to see if other Kuhn/Kogan words on the thread are compatible with that belief.
A brief thought, from Kogan if not Kuhn: Descartes' mind-matter divide has nothing to do with the functioning of science one way or the other, including Descartes' science. ("Nothing to do with" being a bit of an exaggeration. Descartes and others never put the divide into effect, but some of them used the divide in their philosophy, some of which was used to justify, explain, and protect science. But it was never an integral part of science.)
Another brief thought from Kogan: I haven't read Descartes in 30 years, and don't remember much of what I did read.
― Frank Kogan (Frank Kogan), Sunday, 7 August 2005 04:13 (sixteen years ago) link
― Frank Kogan (Frank Kogan), Sunday, 7 August 2005 04:15 (sixteen years ago) link
― kingfish (Kingfish), Sunday, 7 August 2005 07:55 (sixteen years ago) link
― Atom Heart - Dots, Sunday, 7 August 2005 08:30 (sixteen years ago) link
― latebloomer (latebloomer), Sunday, 7 August 2005 13:09 (sixteen years ago) link
― Momus (Momus), Sunday, 7 August 2005 14:42 (sixteen years ago) link
― dahlin (dahlin), Sunday, 7 August 2005 14:44 (sixteen years ago) link
I can't give a full answer to yr. post right now Kogan, which would involve the "cartesian divide" thing and also pulling objectionable quotes from Kuhn on "normal science." I will though, I promise. In the meantime, the points you consider Kuhn's main ones are interesting, and are useful, but aren't all that contestable. Opposition to teleology and belief that scientific knowledge doesn't accumulate incrementally are key, yeah, but once you accept them then.. what. I want to know what you think Kuhn thinks "normal science" mainly involves.
One interesting extension of Kuhn was by Lakatos who coined the idea of "Kuhn loss" which encompasses the *destructive* element of scientific revolutions -- those things which had been theoretically encompassed prior, but which a new paradigm fails to account for.
(v/v the cartesian divide thing, a small explanation for now is that Kuhn holds that every paradigm is as "true" as every other -- hence relativizes -- but, he holds that there is a divide between those who CREATE the paradigms and those who do the "puzzle solving" of "normal science" -- hence between the theorizers and the "doers" and so rather than bridging the gap between "theory" and "reality" [i.e. mind and body, knowledge and substance, etc] he simply creates many theory-reality pairs. a monist account of science would recognize that the difference between paradigm-innovative science and "normal" science is in how they are recieved nore than how they are conducted, and thus render them part of a single process rather than two distinct entities)
― Secundus Covarient (s_clover), Sunday, 7 August 2005 23:23 (sixteen years ago) link
― don, Monday, 8 August 2005 03:59 (sixteen years ago) link
― don, Monday, 8 August 2005 04:06 (sixteen years ago) link
― don, Monday, 8 August 2005 04:28 (sixteen years ago) link
attn kogan: http://lareviewofbooks.org/article.php?type=&id=564&fulltext=1
― s.clover, Friday, 20 April 2012 20:32 (ten years ago) link