A prominent sleep researcher whose work has come under intense scrutiny has lost a paper for duplication, aka self-plagiarism.
Matthew Walker, of UC Berkeley, is the author of Why We Sleep, a bestselling treatise on the many woes of fatigue. Instantly popular, it was touted everywhere, from Bill Gates to The New York Times, which enthused:
“Why We Sleep” mounts a persuasive, exuberant case for addressing our societal sleep deficit and for the virtues of sleep itself. It is recommended night-table reading in the most pragmatic sense.
But among the many criticisms of Walker’s book are that he failed to accurately cite his sources; that he misused at least one figure; and that much of his analysis is just plain screwy. (Andrew Gelman even suggested that Walker’s misuse of the figure might meet the National Institutes of Health’s definition of research misconduct.)
Last August, Walker — who says sleep is our “superpower” — published a paper in Neuron titled “A societal sleep prescription,” in which he argued that:
We are suffering a global sleep-loss epidemic. The health consequences within an individual are well characterized. But does society suffer just as much? Here, I discuss how insufficient sleep erodes our societal fabric as much as it does our biological fabric, and offer some prescriptive remedies.
The article sounded familiar to readers, for good reason. One reader in particular — Alexey Guzey, the author of a critique of Walker’s work — noticed that Walker had published a largely identical version of the essay in The Lancet the year before.
Evidently, the editors of Neuron were, um, asleep at the switch. According to the retraction notice:
It has come to the author’s attention that the degree of overlap between the contents of the NeuroView that he wrote for Neuron (Neuron 103 (2019) 559–562, https://doi.org/10.1016/j.neuron.2019.06.015) and a brief article that he had previously written for The Lancet (Lancet 391 (2018) 2598–2599, https://doi.org/10.1016/S0140-6736(18)31316-3) is not acceptable according to Neuron’s policy. The 2018 piece focused on medicine, on the effect of lack of sleep on doctors and patients, while the 2019 piece discussed this as well as other effects on society, businesses, and children. Because of the reuse of several verbatim paragraphs, the Neuron article must be retracted, given Neuron’s instructions for authors and his disclosure at the time of publication.
Apologies are offered to readers of the journal that this was not detected during the submission process.
An email to Walker was met with:
Thanks for your message. I’m on professional leave 2019-2020, and unavailable. I will respond to your email soon after.
Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].
I have detected and brought to the attention of the appropriate editors dozens of examples of textual duplication no less extensive. In most cases, no action was taken. In a few, a correction was issued. In the case of Vannini I, Fanini F, Muller Fabbri. MicroRNAs as lung cancer biomarkers and key players in lung carcinogenesis [published correction appears in Clin Biochem. 2019 Jan;63:162] [published correction appears in Clin Biochem. 2019 Jun;68:58]. Clin Biochem. 2013;46(10-11):918-925. https://pubmed.ncbi.nlm.nih.gov/23396164/ where the issue was extensive overlap with the text of uncited articles of other authors two corrections were published.
Thank you, Dr. Sanders, for your work.
I can also commiserate: I have reported many textual duplication example in my field to editors. Most editors do not care to reply. Escalating the issue from editor to publisher rarely helps — there are no real consequence from breaking copyright laws in academic publishing, so publishers care even less. They got paid, that’s all that matters.
Anyways, to highlight some bad actors. When I pointed out that a paper is plagiarizing figures from a textbook, EJP Plus issues an “erratum” as if it is not a big deal: https://link.springer.com/article/10.1140%2Fepjp%2Fi2019-12589-x
Meanwhile, when I pointed out to the pay-to-play journal Sci. Rep. that a paper has a guest author (author from by-line who wasn’t in the author contributions or in authors list of the supplementary material), they assured me this is perfectly OK at Sci. Rep. They whitewashed the problem away with an “erratum” as well: https://www.nature.com/articles/srep46975
So it goes.
I am not saying that this is a good thing to do, but it’s not a crime. Instead, it’s worth reflecting why exactly duplication is frowned upon. The copyright was infringed, but this is a legal issue, not a moral one. He did not plagiarize anybody. The author’s words belong arguably to him, and the journals blackmail authors into ceding copyright for free. Who is the scoundrel then? Both Lancet and Neuron are paywalled journals, hence I would welcome a triplication into a OA journal, so that I can at least read it. On a peripheral note, this post oozes scorn and malevolence. I do not know the author, but why is it relevant that Bill Gates quoted him? The remark that “much of his analysis is just plain screwy” is not even backed by any argument! It seems that Mr. Marcus is mainly targeting him because of his perceived celebrity status. I do not know Walker’s work and cannot judge it. But I can judge Marcus’ post and that is tabloid trash.
KP, extensive text recycling becomes a moral issue when authors intentionally deceive readers, and especially editors of journals with explicit policies about the originality of textual material of submissions, into believing that the recycled text is new when, in fact, it is not.
Academic publication is the coin of the realm. Textual duplication is a form of counterfeiting. “But I am only making a copy of a $100 bill that is mine,” you say. It’s still counterfeiting. Please see https://www.sciencedirect.com/science/article/pii/S0009912019312019 where the issues of copyright, plagiarism, and textual duplication are discussed.
I actually agree with KP that there are more henious crimes out there in the publishing world. After all, the original thoughts and writing came from the author himself.
I think the lines are increasingly blurred between say an “obvious fraud” and “copyright issues”, and we should not conflate the two. To give another example, although duplicate publication is universally scorned, a recent study that reviewed journal policies (for 209 medical journals) found that “policies on salami publication are inconsistent and lack specific definitions of inappropriate divisions of papers.”
I know for a fact that a lot of my colleagues (and even senior academics) routinely advise their students to divide bigger studies that generate a lot of data into multiple, separate publications. A simple search would also find many examples of studies with closely related outcomes and utilising the same study population, published by the same research team.
You should check out Matt Walker’s use and “ understanding“ of p-values in some of his sleep and learning papers! In one review, he multiplied several together to claim a result is super-significant!
I’ve known Dr. Walker for many years (as a colleague in the field) and this is all very disheartening to read. I think the textual duplication is the least of the problems outlined here and in the accompanying links. It’s pretty hard to write on the same topic multiple times without falling back on things you’ve said before. Of course, flat out cutting and pasting is not acceptable. Dr. Walker is prone to hyperbole, which may have gotten the better of him. I also listened to the Joe Rogan podcast, and did chuckle a bit when I heard some of the ‘over the top’ comments–which I knew to be (minimally) pushing the bounds of the actual empirical findings. He knows not to do this (as much or at all) at scientific meetings, where he can be challenged. Having said all that, his own scientific studies on sleep and learning have not been challenged in terms of fraud, etc., as far as I know. One can argue (as many do) about whether his results are replicable or even meaningful, but that falls within the realm of the scientific process. He also has been an articulate and popular advocate of expanding sleep research. And that may have gotten him into trouble.
Also, please check out Dr. Walker’s response to some of the criticism discussed here and in adjoining links:
https://sleepdiplomat.wordpress.com/2019/12/19/why-we-sleep-responses-to-questions-from-readers/#CDC_WHO
It is worth noting that since the publication of this article, Dr. Bik seems to have found another article of Dr. Walker’s with a large overlap with previous work (no retraction or EoC):
https://pubpeer.com/publications/10F58BA89C8E961A2F61E75DD8C91C#2