The history of science is the history of rejected ideas (and manuscripts). One example I always come back to is the original work of John Wennberg and colleagues on spatial differences in ‘medical procedures’ and the idea that it is not so much medical need that dictates the number of procedures, but that it is the supply of medical services. Simply put: the more surgeons there are, the more procedures that are carried out1. The deeper implication is that many of these procedures are not medically required — it is just the billing that is needed: surgeons have mortgages and tuition loans to pay off. Wennberg and colleagues at Dartmouth have subsequently shown that a large proportion of the medical procedures or treatments that doctors undertake are unnecessary2.
Wennberg’s original manuscript was rejected by the New England Journal of Medicine (NEJM) but subsequently published in Science. Many of us would rate Science above the NEJM, but there is a lesson here about signal and noise, and how many medical journals in particular obsess over procedure and status at the expense of nurturing originality.
Angus Deaton and Anne Case, two economists, the former with a Nobel Prize to his name, tell a similar story. Their recent work has been on the so-called Deaths of Despair — where mortality rates for subgroups of the US population have increased3. They relate this to educational levels (the effects are largely on those without a college degree) and other social factors. The observation is striking for an advanced economy (although Russia had historically seen increased mortality rates after the collapse of communism).
Coming back to my opening statement, Deaton is quoted in the THE
The work on “deaths of despair” was so important to them that they [Deaton and Case] joined forces again as research collaborators. However, despite their huge excitement about it, their initial paper, sent to medical journals because of its health focus, met with rejections — a tale to warm the heart of any academic whose most cherished research has been knocked back.
When the paper was first submitted it was rejected so quickly that “I thought I had put the wrong email address. You get this ping right back…‘Your paper has been rejected’.” The paper was eventually published in Proceedings of the National Academy of Sciences, to a glowing reception. The editor of the first journal to reject the paper subsequently “took us for a very nice lunch”, adds Deaton.
Another medical journal rejected it within three days with the following justification
The editor, he says, told them: “You’re clearly intrigued by this finding. But you have no causal story for it. And without a causal story this journal has no interest whatsoever.”
(‘no interest whatsoever’ — the arrogance of some editors).
Deaton points out that this is a problem not just for medical journals but in economics journals, too; he thinks the top five economics journals would have rejected the work for the same reason.
“That’s the sort of thing you get in economics all the time,” Deaton goes on, “this sort of causal fetish… I’ve compared that to calling out the fire brigade and saying ‘Our house is on fire, send an engine.’ And they say, ‘Well, what caused the fire? We’re not sending an engine unless you know what caused the fire.’
It is not difficult to see the reasons for the fetish on causality. Science is not just a loose-leaf book of facts about the natural or unnatural world, nor is it just about A/B testing or theory-free RCTs, or even just ‘estimation of effect sizes’. Science is about constructing models of how things work. But sometimes the facts are indeed so bizarre in the light of previous knowledge that you cannot ignore them because without these ‘new facts’ you can’t build subsequent theories. Darwin and much of natural history stands as an example, here, but my personal favourite is that provided by the great biochemist Erwin Chargaff in the late 1940s. Wikipedia describes the first of his ‘rules’.
The first parity rule was that in DNA the number of guanine units is equal to the number of cytosine units, and the number of adenine units is equal to the number of thymine units.
Now, in one sense a simple observation (C=G and A=T), with no causal theory. But run the clock on to Watson and Crick (and others), and see how this ‘fact’ gestated an idea that changed the world.
- The original work was on surgical procedures undertaken by surgeons. Medicine has changed, and now physicians undertake many invasive procedures, and I suspect the same trends would be evident. ↩
- Yes, you can go a lot deeper on this topic and add in more nuance. ↩
- Their book on this topic is Deaths of Despair and the Future of Capitalism published by Princeton Universty Press. ↩