In March, Thomas Harvey sat in on a talk at the American Physical Society’s summit in Denver on a new model of the universe that claimed to eliminate the need for dark matter and dark energy. William George, a fluid dynamicist who spent his career studying turbulence and an emeritus professor at Chalmers University of Technology in Sweden, gave the presentation.
“It was very rambling,” said Harvey, a postdoc working in high energy theory at MIT, “and it was very clearly wrong.”
After the talk, Harvey found George’s paper on the topic in Physics of Fluids. He decided to flag the problems he saw on PubPeer under a pseudonym, including errors in the math and misapplied principles of general relativity, and then sent a “polite email” to the editors.
“That’s where it went very wrong,” Harvey said.
Harvey’s email about the paper went to Savvas Hatzikiriakos, one of the journal’s two co-editors in chief, expecting it would remain confidential. “I just kind of assumed that surely any sensible journal would operate that way,” he told Retraction Watch. AIP Publishing, which publishes the journal, has guidelines for editors that make it their responsibility to “protect the anonymity of individuals raising concerns.”
Instead, the editors shared Harvey’s concerns with the authors — as well as his name, his institution and that he was a postdoc. In their reply, George and coauthor Thomas Johansson, also retired from Chalmers, acknowledged the paper contained a math error but insisted the conclusions were correct. The authors also suggested Harvey be “encouraged to withdraw” his criticisms, and wrote if the criticisms were published, “we will surely do our best to make them public and take our chances in the public arena.”
The authors’ response also recounted a “similarly aggressive young post-doc” who had criticized an earlier paper, resulting in it not being published in Physics of Fluids. They made it a point to present the postdoc’s comments “in every forum for several years,” and the postdoc “never managed to live down the stupidity and arrogance of his stupid remarks.”
George told Retraction Watch he wrote this example to again make the point that what goes in print stays there. “Harvey was correct in recognizing we had made an error. And that was useful. But he handled it stupidly,” George told us in an email. “If our theory proves to be correct, he (and many others) will look rather stupid.”
Harvey said he isn’t worried, since the authors are retired and work in a different field. But added “it felt a bit like a threat” to his career.
George rejected that interpretation in an email to Retraction Watch. “Certainly not a threat. But an attempt to make him aware that things you say in print last forever,” he wrote. “And when you say dumb things and especially say them maliciously (which he did in the original comment), they tend to live for a very long time.”
Harvey learned his identity had been shared when Hatzikiriakos sent him the authors’ reply. Harvey wrote back to him and his co-editor Gerald Fuller, concerned that his identity had been shared without his consent. “A researcher should be able to flag a mathematical error to a journal without having their identity disclosed and without receiving threats in return,” he wrote. In a call with Retraction Watch, he described the situation as “basically editorial malpractice.”
Hatzikiriakos apologized to Harvey for disclosing his identity, calling it an “oversight,” and said Harvey’s original email didn’t indicate his comment was submitted in confidence. Hatzikiriakos suggested he submit a formal comment to the journal, which would be reviewed alongside a response from the authors.
“This is the only formal process available to address and potentially correct fundamentally flawed published work, thereby preventing the dissemination of incorrect conclusions,” Hatzikiriakos wrote.
“I didn’t really want to do this,” Harvey told Retraction Watch, “but as they had already leaked my identity, I felt I had little to lose.”
AIP Publishing said it is aware of concerns about the paper and about how the criticisms were handled, and is reviewing the case. “We are unable to comment further while the review is underway,” said media relations specialist Hannah Daniel.
There are “many things wrong” with the paper itself, according to Harvey. One is an algebraic error in one of the paper’s central equations. “They’re trying to do cosmology, and they’re left with the universe that is empty,” Harvey said. With their equation, “there’s nothing in the universe.” He called this error, and others he pointed out, “elementary.”
George acknowledged the math error. “Obviously we screwed up,” he wrote on PubPeer, and promised to issue an erratum, but maintains the rest of the analysis is still valid. In the original response to Harvey’s criticisms, the authors argued that physicists are “hopelessly biased against our findings,” calling their critics “apostles of the status quo” which is “precisely WHY our results are published in a fluids journal.”
According to George, arXiv told him and his coauthor they were not qualified to submit a paper on cosmology or astrophysics. (Disclosure: Retraction Watch cofounder Ivan Oransky is chair of arXiv’s board of directors but is not involved in day-to-day management or editorial decisions.) He said he pitched the paper to the Physics of Fluids editor-in-chief instead, and supplied a list of potential referees. Physics of Fluids has been his go-to journal for controversial ideas since his first paper there in 1973, George said.
“We are surely not withdrawing the original paper,” he told Retraction Watch.
Harvey is now waiting on the journal to publish his comment, as well as the authors’ response. Eight referees reviewed his comment, and none of them defended the original paper, according to the documents seen by Retraction Watch. Several recommended the article be retracted, and some stated that the authors’ response did not resolve the issues.
Hatzikiriakos, the journal’s editor, told Harvey that publishing a comment and response would be of greater benefit to the scientific community, per the referees’ suggestions — though this suggestion did not appear in the referee comments seen by Retraction Watch. Hatzikiriakos did not respond to our requests for comment.
In George and Johansson’s draft formal response, shared with Retraction Watch, they stand by their results, with the “single exception” of correcting the equation, “however upsetting they may be to the [General Relativity] and Astronomy communities.”
Another paper rejected from arXiv is now in edits at Physical Review Fluids, an APS journal, according to George. “It will probably cause similar outrage, since it too shows Dark Matter is not necessary,” he said.
After Retraction Watch wrote to AIP, its research integrity and ethics manager emailed Harvey promising a “thorough review.”
“If I see something that is clearly incorrect, it’s published in a prestigious, apparently well-established journal … and is out of scope,” Harvey said, “probably, I have some kind of moral obligation to try and point this out.”
The next time he finds a problem with a paper, he will make it clear he expects anonymity, Harvey told Retraction Watch. “I thought I was doing them a favor,” he said, “but it didn’t seem to be viewed that way.”
Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on X or Bluesky, like us on Facebook, follow us on LinkedIn, 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].
That’s what a journal gets when it publishes out of scope. Physics of Fluids most likely had no idea who competent referees on the topic of the manuscript might be, and none of the options in such a situation (AI, suggestions by the authors, random pick,…) is a good one.
The paper is really bad and should never have been published—both because it contains serious errors and because it makes no sense for the journal. Moreover, apart from the technical problems, it was clearly written by people with very limited familiarity with the scientific literature on relativity.
And those technical problems are pretty glaring. Section II is titled: “AN ALTERNATIVE APPROACHTOTHEEINSTEIN FIELD EQUATIONS.” However, the authors appear to have no idea what would actually constitute a solution of the Einstein field equations. Their proposed metric is very obviously not a solution, as the authors have explicitly abandoned local energy-momentum conservation (which is an almost trivial consequence of the Einstein field equations, coming straight from the Bianchi identity). Their posited metric may have the properties that they describe in later sections, but it is not consistent with general relativity.
Did Hatzikiriakos explicitly identify Harvey as a postdoc to the authors? If so, was the disclosure really an “oversight”?
He did, and he acknowledged he’d screwed up in a response to me
Harvey: don’t worry because there are institutions and hiring committees who, contrary to dubious editors and old cranks, value taking a stance.
This is a mostly accurate summary except for three things. There was nothing polite about Harvey’s initial letter to Physics of Fluids. It was insulting to the journal, the editors and entire field of fluid dynamics. We responded appropriately. Second, he did not request anonymity. He identified himself. Third, we only learned he was a post doc from his website, not from Physics of Fluids. My `words of advice’ were most surely not intended as a threat. I’ve spent the 60 years of my career as professor and scientist building people up in theirs, not tearing them down — even those with whom I disagreed strongly. And I have never requested anonymity — I even sign my reviews so the authors know it was me. Or even more important, when it was not me. Given the nature of this discussion we have asked Physics of Fluids to make available on-line ALL of the correspondence, reviews and the discussion. It might make interesting reading in 50 years, and even now.
Surely you can understand that you, an emeritus professor, stating that if published you intend to present what you consider rude/inaccurate/insulting/etc. comments from Harvey, a postdoc, “in every forum for years” and that the last time you did this the individual “never managed to live down the stupidity and arrogance of his stupid remarks.” would reasonably be interpreted as a threat to his career and reputation?
You may have found it justified (we haven’t seen the full email thread and can’t make that assessment) but that makes it no less threatening.
As an emeritus professor myself, if I found myself publicly on the wrong side of a conversation like that, I hope my family would intervene at some point. As you say, it stays in print. And may attract more than its fair share of attention.
+1 🙂
I don’t know about Sweden, but in my country emeritus status is a privilege rather than a right and it is reviewed every few years. Such behavior would likely be ground for its non-renewal.
This is a huge cop out William. No reasonable person could read what you were quoted as saying in this article as anything but a threat.
I throw stones, occasionally, since my student days, and even now. But I was not bothered to hide my hands because I have to take responsibility for what I write/say. Even if one prefers to hide their hands (for whatever reason(s)), there are many ways for anyone to leak any confidential information ‘unofficially’ (from anyone: staff/editors/referees). If the editor leaked it, then so be it. Harvey should say, “yeah, I am the man, catch me if you can”. Unlike robots, humans are always political with feelings, so, all should take it like men. Anyway, there are 2 reasons the authors could be furious, hence the explosive responses from the authors-(i) Harvey correctly found the mistakes, which is serious, and the authors felt trapped, or (ii) the mistake pointed out was minor that can be corrected, but the comments made by Harvey made it seem like the author’s work is completely wrong. Let the correct scientific arguments decide which theory can survive the scientific scrutiny, not because the new proposal is against the known theory and therefore it should be rejected. One should also not resort to democratic process to decide which theory can or cannot be published.
He wanted the authority of flagging a problem and not the responsibility that goes with the authority. What’s new?
The responsibility is on the original authors to not make mistakes, and furthermore not to lash out and make menace of retaliation when someone makes corrections.
To say we have pure rampant fraud leading in our life causes is easily the biggest understatement of millennium.
Per Ivan C Christov, Physics of Fluids might be – to put it mildly – publishing nonsense even in the journal’s nominally core topic.
https://pubpeer.com/search?q=PoF_2026_Similarity_Error_Drop
Thanks for sharing! The PubPeer list will grow. It is only a very small sampling of a very pernicious problem.
PoF’s behavior has been shocking the fluid mechanics community for years. This new foray into bad cosmology is an unsurprising next step, a consequence of the lack of leadership at PoF and, ultimately, at AIP.
Some other nuggets of information:
https://retractionwatch.com/2025/10/22/exclusive-american-heart-association-reviewing-award-to-rocket-scientist-with-eight-retractions/#comment-2355518
https://retractionwatch.com/2025/06/10/physics-of-fluids-aip-publishing-sanal-kumar-journal-retraction-thermodynamics/#comment-2339804
https://retractionwatch.com/2024/04/24/controversial-rocket-scientist-in-india-threatens-legal-action-after-journals-pull-papers/ (the whole comments section)
Your comment also made me think about this part of the story: “Hatzikiriakos, the journal’s editor, told Harvey that publishing a comment and response would be of greater benefit to the scientific community,…”
But, the comment-and-response route has a demonstrably poor track record at Phys. Fluids. In 2021, the journal published a Comment by Frigaard & Poole (Phys. Fluids 33, 129101) explaining that reducing the Casson model to a constant factor (1+1/β) just describes a Newtonian fluid with a higher viscosity, with no yield stress at all. Yet, Phys. Fluids has kept publishing erroneous papers featuring exactly this mistake: multiple since 2025, four of them in the 2026 volumes alone, all flagged at the PubPeer link above; but here’s a more specific search: https://pubpeer.com/search?q=%22PoF_2026_Similarity_Error_Drop%22+AND+%225.0075898%22 . A published comment doesn’t correct much if the same error keeps passing “review” under the named EiC’s tenure…
When arrogant authors stubbornly want to publish in a field they know nothing about, here are the consequences.
Physics of Fluids is now becoming more of a business avenue and their editorial and review standards are abysmally poor. They don’t encourage comments and they have been publishing absolutely wrong physics for sometime now. Their “Faster and Fairer” is simply a motto on paper, and the journal is simply leaving on borrowed legacy established people who genuinely cared for fluid dynamics in the past (and moved on to form PRF in 2015).
Errors and rhetoric aside, I just don’t understand why Harvey did not reach out to George first? Is this not how you should do things? Make the other person aware of a potential problem? They then have the chance to rectify the problem. If it is ignored or brushed under the carpet, then sure contact the journal with your concerns.
The article says Harvey posted on PubPeer before contacting the journal. George did respond there, but his opening reply called the comments “gratuitous” and critics “apostles of the status quo”. That hardly suggests a warm reception. The APS abstract also claims to dispense with dark matter, dark energy and inflation, and explain the missing antimatter through backwards time propagation. Faced with claims of that scope, I can understand preferring public scrutiny to a private exchange.
You have a rosy view of the world. If I see a publication without merit in my field, I have no urge to preach to the authors. When you know your stuff and the other side doesn’t, you might as well be talking to the walls. It is not a technical or academic interesting conversation. What I do have is an urge to tell the journal they screwed up as that is relevant to me.
When notified of a legitimate error in a published work the responsible author(s) need to suck it up and deal with it. How they happen to be notified of the error is irrelevant.
Anonymous criticisms -> PubPeer
Public criticisms -> Letter to the editor
If someone found an error in my paper, I would thank them and correct it. Finding errors helps the authors more than anyone else. Too bad the reviewers didn’t find it.
By starting with a non-Newtonian gravity can one logically obviate the dark sector and avoid overlooked problems of modelling gravity with a scalar field as a non-force.
Without going into a lot of detail, wrongly supposing “dark matter” can serve as a fair “placeholder” phrase ends up ruling out non-Newtonian gravity, whether or not the GR-based replacement happens with e.g. source-wise-stationarily-rippled vector gravity fields supporting gravitationally-extended structures at both galactic-scale and cluster scale. This much should be apparent to all those in the gravity field in general and shows a systematic bias.
The paper under discussion is highly irregular, not only because of its content but also because it does not follow the established standards of publication-level research. Did the production staff, editors or reviewers read the paper?
Of the 65 references, roughly a third are not scholarly sources: three YouTube videos (Refs. 9, 11, 27/35), four Wikipedia entries (Refs. 19, 32, 38, 51), a Medium post (Ref. 12, the source for the “worst prediction in the history of physics” framing), an academia.edu upload (Ref. 43, the sole support for dropping the (1+z) factor from the supernova data, which is the paper’s central empirical claim), a hyperphysics page (Ref. 25), three private or personal communications (Refs. 28, 36, 37, one from 1965), a Wolfram Community thread (Ref. 36), and six citations to the authors’ own website or unpublished APS/DFD meeting slides (Refs. 10, 13–16, 22). Two distinct references, Ref. 27 (“P. Dirac speaking to F. Hund on symmetry, 1982”) and Ref. 35 (“R. Smethurst, Dr. Becky: Crisis in cosmology”), carry the same YouTube ID, Et8-gg6XNDY; checked on Oct 15, 2026, that video is gone (“removed following a copyright removal request by Technische Informationsbibliothek (TIB)”), and since TIB holds the IWF scientific-film archive the footage was almost certainly the Dirac interview, so Ref. 27 is right and Ref. 35 points to the wrong video. The Dirac quote that opens Sec. II and is invoked again in Sec. XI therefore has no reachable source in the paper. Ref. 30 (Guth) is labeled “Youtube.com” but links to MIT OCW.
Physics of Fluids at one time was a good journal. That seems no longer the case. A telling sign is the number of papers they publish each year. Before 2018 this was roughly around 500 papers a year. In 2020 they exceeded 1000 papers for the first time. In 2021 they hit 2000. In 2023 they hit 3000. In 2024 they hit 5000. That looks like a paper mill to me. AIP should be ashamed. Evidently they like collecting those author-publication fees.
As many others have said here, PoF has gone from the second best journal in the field to an absolute joke. They will publish -anything-: papers that would have been rejected from a tier 3 journal previously, reviews written by groups who have never worked on the topic, and completely non-physical nonsense. I used to joke that fluids had never featured on retraction watch, but PoF has now made it several times.