With my modest experience as an expert witness in courts, I am against this. It is very difficult to make judges understand, for example, how science works - one can only be so sure, and we go ahead with that, for example, in medicine, as well as cryptography. Judges have a hard time appreciating this nuance.<p>I agree that something stricter should be done, but it should not be about bringing the legal system into play. I see a fundamental issue with bringing science to trial courts, where rhetoric, appeals to emotions, and other different priorities are paramount, not technicalities about overenthusiastic interpretations, data fudging, p-hacking, empirical anomalies and wilful data manipulation.<p>Science works by different norms of truth (I would call this statistical) than the judicial system does (beyond reasonable doubt/preponderance of evidence). I believe an international peer scientific committee ostracising a person from publication for X number of years, or forever, might be a better measure than a criminal trial and punishment in open court.
I couldn't agree more. Here's a hack to do it without any change in legislation. Scientific journals should require the standard boiler-plate of an affidavit form to be added to the article itself and signed and witnessed as part of the publication process. In this manner, every submitted article is also a valid affidavit admissible in court, and knowingly making a statement which is false or misleading in a material particular while under oath is the crime of perjury.<p>As I once said:<p>"Perjury must be a crime. There is only one sin in science, and that sin is faking data, and faking evidence is faking data. Perjury is surely a crime."<p>I'll leave out the unfortunate context in which this needed saying.
I am sympathetic to the fear that this would lead to scientists being thrown in prison for technicalities or even misunderstandings. Nobody wants that.<p>We also don’t want a cottage industry of performative, ladder climbing researchers siphoning funding from real ones.<p>At a minimum, the public could stop funding researchers with faked data or images. It’s unacceptable that the NIH keeps funding known frauds. If you doctor images, you’re done.
Another way of looking at it is what if this was just a legit unintentional scientific error. The scientific process is not suppose to be so fragile that a paper by an individual causes it to go astray for 16 years.
A fair point here is the impact of the crime. As the author puts it: "if it delayed a successful treatment by just 1 year, I estimate that it would have caused the loss of 36 million QALYs (Quality Adjusted Life Years), which is more than the QALYs lost by Americans in World War II."<p>Of course these are just the wildly successful instances of such fraud. Many such fraud probably represent just one more quickly ignore paper. But still, the crime matters because it has this potential to derail a field. The magnitude of the crime is not represented by calling it "fraud".
Fraud is already illegal, and there isn't really any argument made here that scientific fraud should be handled differently from the generic type, indeed most of the author's "response to objections" are that scientific fraud should be treated exactly the same as other fraud.
This would close down whole departments and wouldn't solve the problem of the wrong ranking system by citation.
A more important step towards truth would be to rank research based on accuracy of predictions and independent experiment reproductions.
Science itself would have been considered a form of intellectual misconduct at one point. It would be important to be very careful here with definitions and scope, Eg, limiting it to fake data
The independent scientific misconduct board Denmark implemented seems like a good first step [1].<p>Given it relies on institutions (I assume this includes universities) sending it complaints, I’m not sure how we get over the conflict of interest mentioned.<p>> <i>Sylvain Lesné, the lead author on the Alzheimer’s paper, remains a professor at the University of Minnesota and still receives NIH funding</i><p>Okay, maybe you permit anonymous complaints. If the anonymous complaint results in an adverse finding, the institution is penalised. If not, just the researcher.<p>Penalties should include fines. But also a term during which they are blacklisted from NIH funding.<p>[1] <a href="https://ufm.dk/en/research-and-innovation/councils-and-commissions/The-Danish-Board-on-Research-Misconduct" rel="nofollow">https://ufm.dk/en/research-and-innovation/councils-and-commi...</a>
If one fraudulent paper can derail a field that much, it's only because we have other, larger problems. Such as funding hegemony with the big researchers and big labs.
There are other deterrents possible than this author’s vendetta has<p>Maybe civil charge to pay back the public funding, in the case of there being public funding. Turning researchers into debt serfs may reach better compliance and sustain access to their cognitive abilities on behalf of the public, as opposed to prison
One side effect would probably be, that anti-science rulers would turn this into a weapon against unliked scientific results. Think vaccines. The Damocles sword of endless litigation would hang over every scientific research concerning public policy.<p>Maybe it's overall better to let science sort out science things.
Not knowing much about the jurisprudence here, why would wire fraud not already make scientific misconduct illegal, with the determination about standards of evidence (not necessarily with intent) left to the courtroom?
I do not want this to happen, evem though the scientists were eventually exonerated: <a href="https://www.science.org/content/article/why-italian-earthquake-scientists-were-exonerated" rel="nofollow">https://www.science.org/content/article/why-italian-earthqua...</a>
Publishing a paper isn't necessarily engineering a product, but decisions are made from data, and there are corresponding effects.<p>Reproducibility is the test suite of science. Make experiments reliably, repeatably testable.<p>Wonder if computable research will become a requirement for publication. Will that make a slow process slower?
Imagine if all research had to go through an HR style cover your ass system before any of it could be shared. You would basically shut down all multi-institution research.