Skip to main content

Who Discovers and Why

A Difficult Birth For Some Ideas

This fascinating article has nothing to do with drug discovery per se, but it has plenty to do with discovery itself. It’s a memoir of the author’s physicist grandfather, who believed while working at his job at Oak Ridge that he might have come across a big result in probability as related to quantum mechanics and physical law. Now, whether this idea is correct or not is the key question, and that’s still unanswered. One reason it’s still unanswered is that his paper trying to explain the idea went unpublished, again and again, a process that consumed a good part of the last years of his career.

The author (Veronique Greenwood) isn’t a physicist herself, but she believes, after having gone through the huge pile of files that her grandfather (Francis Perey) left after his death, that it’s more likely than not that he made a mistake somewhere. If the work had ever been put out there for other physicists and mathematicians to look at, perhaps this could have been resolved, but that never happened. When he sent his manuscript out for review in the early 1980s, he got referee reports like this:

Incomprehensible … if there is any core of actual results anywhere in this incredible work, it is hidden completely from me by the page after page of maddeningly repetitious philosophical froth … No purpose could be served by publishing the work in its present form, because I do not think there is one reader in the world who could make sense of it. . .

OK, then. That wasn’t the only response like that, either. The problem is, densely (and/or poorly) written papers are not necessarily wrong or the work of people who have gone off the rails, although it’s for sure that those people produce work of this exact type. There have been numerous cases across several different fields of people who were right, who were really on to something, but who (in retrospect) just could not make themselves sufficiently understood. The sort of mind that comes up with a wonderful new result is not always the sort of mind that is capable of explaining it to anyone else very well, and for a sufficiently complicated topic, it may take a while no matter who’s explaining it. Kurt Gödel’s Incompleteness Theorem was first announced at a conference roundtable, and apparently the only person who understood it enough to grasp its importance at the time was John von Neumann. And believe me, not every audience has a von Neumann in it.

Many of us have had the experience of getting referee reports back that show that at least one of the people who read the paper did not even grasp what it was supposed to be about. And at times like this, you experience a quick crisis, a volatile mixture of anger and doubt. The first strengthens your resolve that you’re right, because the objections of someone who doesn’t even understand the first thing about your work are worthless. But the doubt hits you because you thought that you had done a clear job of explaining what you were doing – how could anyone miss it? Was reviewer #3 that dense? Or are you just a lot worse at explaining yourself than you thought? You can see these exact struggles as you read the article about Perey’s attempts to get his work published.

The good thing is that all of that work is now available to read, thanks to the magic of the internet. The particular magic of preprint servers is worth thinking about for current ideas: these days, you can at least get your ideas out there. Now, for any individual manuscript that may or may not be a good idea, from either a scientific or career perspective. You might be totally wrong, you might be only partially wrong in a way that could have been fixed if you’d bounced the idea off more people, or you might be completely right but just are doing a hideous job of explaining yourself. It’s a somewhat different situation, but years ago I read that there were a number of editors who saw earlier versions of J. K. Rowling’s first “Harry Potter” book. Which was significantly worse than the version that eventually got published – she took many of the criticisms to heart and went back and worked some more, and what emerged eventually made her wealthy and famous. If she’d just said “Ah, the heck with all of you” and self-published as is, that never would have happened, most likely. Science is different – we’re arguing about matters of truth rather than matters of taste. But it’s not quite as different as we’d like. As Perey’s story shows, not all these problems are easily resolved – at least, not in the lifetimes we have – so there are times when just getting the ideas out there is probably the best thing for the world.

38 comments on “A Difficult Birth For Some Ideas”

  1. Some Dude says:

    The infamous gyre paper is relevant in this context:

    It’s an incomprehensible paper “of maddeningly repetitious philosophical froth” that somehow got through peer review. The consensus seems to be that the author suffered from mental illness, which might the most common origin stories for such papers.

    It is difficult to find the geniuses among the noise produced by people with severe issues.

    1. Derek Lowe says:

      That it is – but the gyre paper seemed immediately recognizable as the product of such. You have to imagine that any handwritten notes for it were in multicolored ink, mostly green, written all around the margins of the page. It’s the “might not be drivel” stuff that’s hard to place.

  2. APAJ says:

    An example from chemistry comes to mind where H. Sachse published a series of papers in 1890-93 on the chair and boat conformations of cyclohexane. His work was completely ignored as the papers were written in dry mathematical drivel. Not until 1918, after the crystal structure of diamond was solved which illustrated Saches propositions, was his theory both recognized (or understood) and proven.

    1. Derek Lowe says:

      Yes! That was an example I was trying to think of, but I couldn’t quite place it. He really did figure that stuff out, but no one could understand him (at least, no one who was doing organic chemistry).

    2. Anonymous says:

      To me, the important point is the difference between PUBLISHED and criticized (and sometimes ridiculed) and then proven to be correct (Sachse – your example; Wegener – tectonic plates; Bretz – Missoula Floods; etc.) and never published at all.

      It is also noted that some discoveries are made independently and published more than once. They are multiples. (calculus: Newton, Liebniz; ribozymes: Cech, Altman; etc.).

      Other discoveries are blocked from publication and remain nulltiples.

      (See: “multiple discovery” at wikipedia, linked to my handle.)

  3. Canada Chemist says:

    An interesting find! Has echoes of a wonderful play by David Auburn called “Proof”

  4. Anonymous says:

    A quick search (google scholar) reveals that F Perey or FG Perey have lots of pubs in the area of nuclear physics and ~7500 cites. Some appear to be single author. His earlier writing might be compared to the unpublished material. Psychologists or forensic handwriting analysts might recognize patterns of change either in the content and writing style or even in his handwriting. (E.g., handwriting analysis of an infamous 20th century dictator showed signs of CNS degeneration.)

  5. Fred says:

    Even Feynman needed a Dyson.

    1. Wavefunction says:

      Not just that, but Feynman’s 1948 talk at the Pocono Conference notoriously bombed because of its novelty, and eminent physicists like Bohr and Dirac thought he was talking nonsense; his approach was simply too novel for them to understand. It also didn’t help that the then-wunderkind Julian Schwinger’s approach was much more rigorous and grounded in familiar techniques, although ultimately Feynman’s approach turned out to be simpler to use. Dyson’s feat was to show the equivalence of the two theories, essentially showing that Feynman was on Schwinger’s more familiar territory.

    2. loupgarous says:

      Speaking of which, Freeman Dyson recalls J. Robert Oppenheimer’s initial reaction to Feynman’s ideas about quantum mechanics as “vituperative”. Oppenheimer’s usual reaction to what he considered fallacies was his trademark dry wit.

      1. Wavefunction says:

        Oppenheimer could be a royal horse’s ass and very nasty when it came to ideas that he either didn’t originate himself or had trouble appreciating: he could be brutal in dismissing people as irrelevant or inferior, and this quality was, in significant part, responsibility for his security troubles later.

  6. Uncle Al says:

    Propose a *testable* incredulity. The universe is only matter (baryogenesis), in crass violation of conservation laws (Sakharov criteria). Vacuum cannot be perfectly achiral isotropic (achiral spacetime curvature, Einstein). Vacuum is doped with chiral spacetime curvature (Einstein, Cartan, Kibble, Sciama). Vacuum is a ppb left foot. Blow a 3:1 R:S mix of calculated extreme chiral-divergent enantiomers (opposite shoes) into a single line ground state transition microwave rotational spectrum,

    Used molecule: 222 lines, ~2 kelvin, MW = 344.45 amu, CHI = 0.239120; different study
    Crafted molecules: 1 line, ~2 kelvin, MW = 171.24 or 214.23 amu, CHI ≈ 0.9

    The opposite shoes’ spectral lines cannot exactly superpose for having divergent embedment energies in left-footed vacuum. A mere 3:1 line asymmetry, one undergrad-day, falsifies 400,000 physical theorist-years.

    … 1) Unequal vacuum embedment energies demand non-identical minimum action vacuum free fall trajectories. The Equivalence Principle is falsified by enantiomeric test masses.
    … 2) Trace chiral anisotropic vacuum, through Noether’s theorem, causes trace non-conservation of angular momentum. It is Milgrom acceleration not dark matter.
    … 3) Add a matter wave grating front end, optically resolved molecular beam, and quantum mechanics is caught in a contradiction

    Who could resist healing multi-$billion physics with a day of analytical chemistry? everybody The worst it can do is succeed.

    1. DrOcto says:

      Case in point.

  7. Bagger Vance says:

    I dunno, for every Godel there’s bound to be a hundred Robert V Gentry’s

    1. Derek Lowe says:

      I’d forgotten about that guy! But I was in college in Arkansas during the McLean v. Arkansas case. That happened after the Republican governor (Frank White, who came out of nowhere and defeated Bill Clinton in 1980 after his two years in office), signed the law mandating Creationism and directly into the courts it all went. . .

  8. MattF says:

    There’s an example of this in physics: Ernst Stueckelberg.

    He discovered several deep and important results in quantum field theory well before Feynman, Schwinger, et. al., but he published in obscure journals and the papers were generally considered incomprehensible.

    1. Anonymous says:

      Another example, from astrophysics: Chandresekhar proposed the eponymous Limit when he was a 20 year old grad student in India. Eddington (a “thought leader” or was opposed to the idea. From wikipedia (linked in my handle):

      “Although Niels Bohr, Fowler, Wolfgang Pauli, and other physicists agreed with Chandrasekhar’s analysis, at the time, owing to Eddington’s status, they were unwilling to publicly support Chandrasekhar. … The drama associated with this disagreement is one of the main themes of Empire of the Stars, Arthur I. Miller’s biography of Chandrasekhar. In Miller’s view: ‘Chandra’s discovery might well have transformed and accelerated developments in both physics and astrophysics in the 1930s. Instead, Eddington’s heavy-handed intervention lent weighty support to the conservative community astrophysicists, who steadfastly refused even to consider the idea that stars might collapse to nothing. As a result, Chandra’s work was almost forgotten.’ ” (From other sources: Chandra stopped working on the problem in order to avoid Eddington’s wrath. Yet, it won him the Nobel Prize in 1983.)

      I guess chemistry and med chem are not the only fields with “thought leaders” to guide or misguide them.

      1. Anonymous says:

        Oops. (1) Factual error: Chandra proposed the C Limit when he was 20 years old and still in India BEFORE leaving for grad school at Cambridge (England). The papers were published in 1931. (2) Typo: “Eddington (a “thought leader”) was opposed to the idea.”

        1. Wavefunction says:

          Actually Chandra worked out the details of the Chandrasekhar limit on the voyage to England, which was pretty cool. Einstein and Eddington are both good examples of young mavericks who became old conservatives. After Chandra’s work, it took other mavericks – most notably Fritz Zwicky, another brilliant scientist who was regarded as a crank by many – to resurrect the idea. Zwicky and his colleague Walter Baade proposed the existence of supernovas. And Lev Landau came up with the idea of neutron stars.

          Finally, it was Oppenheimer and Snyder who worked out the complete evolution of gravitational collapse and came up with black holes. Unfortunately it took another twenty years (it didn’t help that the war intervened) for John Wheeler to resurrect black hole research and turn it into an independent discipline. All this goes to say that even legitimate brilliant ideas may flounder for years in the dark because of a variety of reasons.

        2. dearieme says:

          “who steadfastly refused even to consider the idea that stars might collapse to nothing”: I read once that Eddington had had such an idea much earlier but had then publicly dismissed his own idea as absurd. I wonder how much that mattered.

  9. mjs says:

    Rule for authors of scientific papers: If the Idiot Reviewers have misunderstood your manuscript, that is prima facie evidence that your manuscript can be misunderstood. Revise it.

    1. zero says:

      The third rule of tech: there is always a bigger idiot. No matter how carefully you protect the user, someone will find a way to mess things up. (The first rule is a reboot will fix almost anything. The second is that the user is at fault, and it’s almost always true.)

      I think a similar rule applies to appliances, or to machines in general. There will always be someone willfully ignorant enough to hurt themselves by defeating safety features.

      A hard look at clarity and comprehensibility might be warranted (perhaps with comments from a trusted third party), but make sure you actually need to invest a lot more time improving wording before committing to it.

      1. T says:

        I think perhaps a key element of mjs’s post is the use of “Reviewers” plural. If one reviewer says its incomprehensible, then yes, its possible that they are just the “bigger idiot”, but if several do, then it highly likely that the problem is with your paper.

  10. Emjeff says:

    The late Martin Gardner wrote extensively on the difference between scientific cranks and geniuses out on the fringe of discovery. It’s a pretty fine line, but true scientists are willing to be wrong, . I haven’t made up my mind about this guy yet, but it is a fascinating article, not the least of which is having such a brilliant grandfather – mine only went to the 8th grade…

    1. MrXYZ says:

      The book ‘How the Hippies Saved Physics’ ( covers this topic as well. It’s an odd book but really shows how fine the line is (particularly in Berkeley and the Bay Area in the 1970s).

      1. Wavefunction says:

        Interestingly, Freeman Dyson reviewed that book in the New York Review of Books and expressed sympathy for the principal crank in the book – Jim Carter. In the same review he also expressed sympathy for another notorious crank – Immanuel Velikovsky, who was a friend and neighbor. Dyson did not believe Velikovky’s theories about Venus being ejected from Jupiter and in turn ejecting the moon, but he admired his stubbornness and his refusal to remain silent and said that he despised those scientists who were trying to silence him. He compared Velikovsky to William Blake. Dyson’s point is that people can spout nonsense and still embody greatness (he places Eddington in that category).

        1. Fred says:

          Hah… I actually have hauled around (I think – unpacked in multiple moves) a copy of Velikovsky’s “When Worlds Collide”, obtained and retained for its novelty/amusement value.

          I consider many of Dyson’s ideas, like the Dyson Sphere, unfortunate magical thinking.

          In an earlier observation, I said, “Even Feynman needed a Dyson.” In this context Dyson provided a unique talent for both understanding Feynman’s conceptual framework for QED and also for formulating the conventional mathematical physics in which Feynman was deficient for expressing his concepts of QED.

          1. loupgarous says:

            The “Dyson sphere” is not so obviously “magical thinking” that some astronomers didn’t it seriously as an explanation for periodic fluctuations in the emissions of KIC 8462852 – a “Dyson swarm”.

          2. Wavefunction says:

            What I find most interesting about the Dyson sphere was that it was published in Science in 1960. It would be impossible to imagine a respectable journal like science publishing speculative ideas like this today, although to me that says more about Science’s open-mindedness in 1960 relative to 2018 than anything else. Also worth noting that while Dyson creatively worked out some of the numbers in a Dyson sphere scenario, the idea itself came from the science fiction writer Olaf Stapledon in his “Star Maker”, and Dyson does acknowledge this.

  11. gippgig says:

    Sometimes people just won’t believe a result that contradicts current belief. See amorphous semiconductors.

    1. loupgarous says:

      Thomas Kuhn’s The Structure of Scientific Revolutions is full of examples of how paradigms often die hard in science, Newtonian physics versus Einsteinian physics a prime example.

      R. E. Peierls’s 1960 article on Wolfgang Pauli records his comment on something Einstein said at a colloqium both attended – “You know, what Mr. Einstein said is not so stupid… “

  12. An Old Chemist says:

    I am wondering if the controversey over who first came up with the idea of ” the “Woodward Hoffmann Rules” fits this discusssion. Corey first proposed this idea to Woodward in his office, but Woodward was not very receptive of it. Later, Woodward loved the idea so much so that he passed it on as his own.?

    1. Wavefunction says:

      I don’t think there’s any good evidence that Corey suggested the idea to Woodward in any amount of detail. In fact there was a natural product synthesis paper in which Corey had every opportunity to acknowledge the mechanism before Woodward and Hoffmann, but he didn’t do it. There’s also evidence that Woodward was already thinking about orbital symmetry before Hoffmann came on the scene, most notably inspired by the (sadly short-lived) William Moffit. At best Corey might have had a hazy idea of orbital symmetry (and in that case, there’s a better case to be made that Havinga and Oosterhof came much closer to the concept), but there’s no evidence he had grasped the richness and detail that Woodward and Hoffmann did.

  13. anon says:

    “Science is different – we’re arguing about matters of truth rather than matters of taste.” I’m not so sure about this particular case. I read the linked Perey’s paper and it seemed to be about how one should interpret what is an ill-defined problem. It’s not at all clear to me that there is a single correct interpretation.

  14. JB says:

    Highly recommend every scientist/engineer trying to publish take a course on technical writing. It is of my opinion, but I find more and more Science/Cell/Nature articles are simply unreadable these days. In technical writing, you do not want beautiful prose or a huge vocabulary to show off. You need to be as succinct as possible. Microsoft Word already helps you in this endeavor –all you have to do is turn on the ‘readability’ scoring feature. If you’re regularly scoring above a 15, 16, 17+ in a technical document, there is a good chance you’re losing your audience. Even documents written for the Supreme Court of the US usually do not exceed a 14-16 Flesch-Kincaid readability score. Again, I highly recommend taking a 1 or 2 day course on technical writing. I thought it was well worth it and the course provided a lot of useful tips.

    1. limited says:

      Thanks for that one. I never knew that. Just checked the student’s review and got to 17 after my corrections. Maybe the student can get it to Harry Potter level.

    2. dearieme says:

      Does a technical writing course teach you much beyond what you’d have learned by reading “Plain Words” as a schoolboy?

  15. Guest 5 says:

    It would be interesting to see Cell’s referee comments on this rejected paper:

    Cas9–crRNA ribonucleoprotein complex mediates specific DNA cleavage for adaptive immunity in bacteria

Comments are closed.