Sunday, October 4, 2020

DISHONESTY IN ACADEMIA: THE DEAFENING SILENCE OF THE ROYAL SOCIETY OPEN SCIENCE JOURNAL ON AN ACCEPTED PAPER THAT FAILED THE PEER REVIEW PROCESS





More than two years ago, on February 26th 2018, I was contacted by the Royal Society Open Science Journal to referee a submitted manuscript. Two prior referees had accepted the paper and two had rejected it, and I was the tiebreaker. The manuscript, Quantum Correlations are Weaved by the Spinors of the Euclidean Primitives by Joy Christian, basically claims that Bell’s theorem is incorrect. If true, this would be a game changer in the foundation of quantum mechanics. Bell’s theorem shows that it is impossible to construct a local realistic model of the theory.

Bell’s result is an impossibility proof; it attracts such passion as the impossibility of perpetual motion machines that were so popular some 100 years ago. A manuscript claiming the invention of a working perpetual motion device, proof that Earth is flat (yes, there is such a thing as an annual conference of Flat-Earth-ers), or that the sun circles Earth would be rejected by any respectable journal right away.

So, what if someone managed to “disprove” Bell’s theorem and, better yet, to publish that “discovery”? This would create lots of debates and excitement – certainly, notoriety and free publicity for the journal who published your claim. In other words, good business.

But who is claiming to have “disproven” Bell theorem? Enter Joy Christian, who has been asserting this claim for 13 years. It was debunked by many scientists and scientific panels over the years, yet Christian is not having any of it. Basically, he alleges to have found a method for obtaining the quantum correlation of a Bell pair of particles, by using a Bell “loophole”. In the no-mans-land at the intersection of physics, mathematics, and philosophy, experts in all three fields are scarce. Christian’s ‘method’ is based on a mathematical error, which is ultimately adding apples and oranges, but the error is hard to spot if you are not a genuine expert in geometric algebra. Add to this the language and structure of a well-written physics paper and you might convince an unsuspecting referee to approve your manuscript.

I had found Christian’s mistake again in the manuscript and I recommended to reject the paper. Certain that it would never be published, I went about my daily business. Imagine my surprise when I heard Christian had somehow managed to publish his nonsense. I thought this impossible; the vote had been 3 to 2 for rejection. I checked and found that indeed, the paper had gotten accepted after submitting a revision. However, I was not contacted by the journal to review the revision. I started contacting colleagues who had to deal with Joy’s claims before, and together with Philippe Grangier, Richard Gill, Howard Wiseman, Brukner Časlav, Gregor Weihs, and Scott Aaronson, in a letter to the journal on July 28th 2018, we asked that the article be withdrawn:

Dear Editor-in-Chief,

We are writing to you about the publication of the paper “Quantum Correlations are weaved by the spinors of the Euclidean primitives” by Joy Christian in your journal on May 30 2018 http://rsos.royalsocietypublishing.org/content/5/5/180526

The result of this paper conflicts with an established scientific fact (Bell’s theorem) well known in the foundations of quantum physics and a basis of modern quantum information science; moreover, the subject of recent high-profile experiments (“loophole free tests of Bell’s theorem”). The paper contains numerous errors in elementary algebra, calculus, and logic. The manuscript was rejected by three of the five reviewers, but the editorial process as stated to the reviewers by your journal was not followed: the manuscript was accepted without informing the reviewers and giving them a chance to rebut the misleading statements made by the author (see review history on the link above).

The claims made by the author are well known from 2007 and they were disproven in the past (https://fqxi.org/community/forum/topic/1577 ). From time to time Joy Christian attempts to publish his faulty claims and recently a similar paper was withdrawn by Annals of Physics https://www.sciencedirect.com/science/article/pii/S0003491616300975

The journal did extend an invitation to write a rebuttal paper but stated that Joy Christian would be a reviewer to the rebuttal. This is not an acceptable course of action from an ethical point of view because it legitimizes scientific dishonesty on behalf of Joy Christian who is well aware of the issues with his arguments for more than 10 years and yet continues to obfuscate the truth.

Considering this, we are respectfully asking your journal to withdraw the paper.



Sincerely,

Florin Moldoveanu - George Mason University (reviewer 5)

Richard Gill – Leiden University

Howard Wiseman – Griffith University (reviewer 3)

Scott Aaronson - University of Texas

Philippe Grangier - Institute of Optics, Charles Fabry Laboratory

Brukner Caslav - IQOQI - Institute for Quantum Optics and Quantum Information Vienna

Gregor Weihs – Innsbruck University




This was about two years ago. We kept asking for updates, and when not stonewalling us, the journal kept pushing one roadblock after another.

The Royal Society Open Science Journal had more than two years to get their act together. By now, their silence speaks louder than words.

It is unconscionable that instead of putting extra checks in place for authors with a history of inaccurate publications, the journal violated their own peer review policy and chose to maintain a faulty paper instead of withdrawing it.

We gave the journal the benefit of the doubt for two years. The passing of time made it clear that the decision to maintain the faulty paper is no accident and no mistake.

Perhaps this is a symptom of a larger systemic problem with open journals who are paid by the authors to get their papers (usually rejected elsewhere) published. When your salary and livelihood depend on the people you are supposed to enforce rules upon, the temptation to bend those rules is high.

I grew up in a former communist country of the eastern bloc. At the time of communism, a rule supposed to be enforced by the traffic police was that if you pay for a traffic ticket on the spot, you will be charged with half the fine. You might guess that most officers pocketed that money. The rule only solidified endemic corruption.

In our case, the author is well known for making the same incorrect argument over and over again. However, the root of the problem seems to be with the journal. After all, had they followed their own policy, the problem would not have arisen in the first place. And in case the mistake was genuine (as sometimes mistakes do happen), is two years enough time to get the record straight? It makes me wonder: just how often did the editors turn a blind eye to publication issues to secure revenue? Is it really a good idea that those in charge of rule enforcement are financially dependent on the rule violators?

3 comments:

  1. Isn't it the very point of "Open Science" journals that they mostly publish garbage like Joy Christian's?

    ReplyDelete
    Replies
    1. Indeed! There is a similar paper by the same author in IEEE Access. These two journals are the RS and IEEE's answer to predatory publishing. They can charge more, because of the name on the label, but the business model is the same. There is a third article in a serious pure maths journal: Communications in Algebra. How hat slipped through the usual serious refereeing is a mystery.

      Delete
  2. See also https://link.springer.com/article/10.1007/s00006-020-1046-0 "A 1d Up Approach to Conformal Geometric Algebra: Applications in Line Fitting and Quantum Mechanics", Anthony N. Lasenby, Advances in Applied Clifford Algebras volume 30, Article number: 22 (2020); and J Chrisian,"Eight-dimensional octonion-like but associative normed division algebra", Communications in Algebra (2020), doi: 10.1080/00927872.2020.1791899

    ReplyDelete