"J. Hirsch. [...] engaged in unscrupulous practices, including falsifying analyses and selectively presenting data to support unfounded claims. [...] Hirsch's tactics include manipulation of public opinion, personal attacks on our team members, and threats and complaints to our management and funding agencies." - Mikhail Eremets, the single most highly regarded high pressure experimentalist today.
Jorge E. Hirsch, the Californian physicist who once invented the h-index and who these days fights superconducting fraud, has been trying in vain to get access to raw data in a paper from a russian lab in Germany. What Hirsch instead got, was a witch hunt, a smear pamphlet and police threats.
You can read about Hirsch’s past successes of exposing the room-temperature superconductor fraud of Ranga Dias here:
Hirsch’s current target is a 2022 paper in Nature Communications from the lab of Mikhail Eremets, 75 year old Belarus-born and Moscow-trained superconductor researcher at the Max Planck Institute for Chemistry in Mainz, Germany. Eremets is a bigwig in his field, in October 2022 the New York Timessummed up his much celebrated breakthroughs when excitingly reporting about Dias’s room-temperature superconductor (which later was exposed by Hirsch and others as totally fake):
“In 2015, Mikhail Eremets, […] reported that hydrogen sulfide — a molecule consisting of two hydrogen atoms and one sulfur atom — turned superconducting at minus 94 degrees Fahrenheit when squeezed to about 22 million pounds per square inch. That was a record warm temperature for a superconductor at the time.
Dr. Eremets and other scientists subsequently discovered that lanthanum hydride — a compound containing hydrogen and lanthanum — reached a superconducting temperature of minus 10 degrees Fahrenheit at ultrahigh pressures.”
Eremets received many prizes and awards, in 2015 he was one of Nature’s 10 top scientists, in 2020 he was named the “Falling Walls Breakthrough of the Year in Physical Sciences“, nominated by his employer, the Max Planck Society. After the Dias double-meltdown with two retracted Nature papers, Eremets became the superconductor community’s last hope. University of Buffalo professor Eva Zurek said in an interview from May 2024:
“What the press should realize is that there have been also a number of successes. So, for example, I think that the work of Mikhail Eremets on H3S that was shown in numerous experiments and also in theoretical calculations to have particular superconducting properties — that should be hailed as a success. There has been work on LaH10 by Eremets [and Russell] Hemley.”
This paper was presumably meant to further cement Eremets’s claim to the future Nobel Prize:
“The data that support the findings of this study are available from the corresponding authors upon reasonable request.”
Nature Publishing Group indeed has a clear rule in this regard:
“After publication, readers who encounter refusal by the authors to comply with these policies should contact the chief editor of the journal. In cases where editors are unable to resolve a complaint, the journal may refer the matter to the authors’ funding institution and/or publish a formal statement of correction, attached online to the publication, stating that readers have been unable to obtain necessary materials to replicate the findings.”
“Open access contributions include original scientific research results, raw data and metadata, source materials, digital representations of pictorial and graphical materials and scholarly multimedia material.”
Apparently this all is just meaningless, like the human rights paragraphs in the russian constitution. Hirsch wrote on PubPeer in August 2023:
“I requested the underlying data for Figs. 3a, 3e, 3b, 3f of this paper on Jan 11, 2023. This is because the published data for Figs. 3a and 3e, as well as for Figs. 3b and 3f, are nominally the same but incompatible with each other, and I would like to understand why that is. I asked the authors to explain, but they did not provide an explanation. Neither did they supply the data. The journal told me that it had received the data from the authors but will not share them with me because they are “confidential”. I requested that the journal posts an Editor Note informing readers that data are unavailable to readers. The journal responded that because data were share with editors they “cannot write an editorial note on the published article stating the data is unavailable as this would be factually incorrect.”
Instead, the journal published something else, on 1 September 2023. An Author Correction, which, among other things added a previously “omitted Supplementary Fig. S12.”
Maarten van Kampen suspected “severe ‘processing’ of the data to obtain picture-perfect and convincing results” and summarized the correction’s lenghty text as such:
“the published data has been somewhat massaged (vertical shift and a very unclear ‘we performed additional linear transformations…’).
As one member of the community told me:
“A straightforward verification if the data in Figs. 3a and 3e of [Author Correction for Minkov et al 2022] are related by a linear transformation, is to make a parametric plot of the moment Ma(Hj) versus Me(Hj). This gives the following plot.
If the two data sets were related by a linear transformation, all points would fall on a straight line. The non-linearity in the graph may be caused by a mere mixup of data files. “
Meaning, the correction is indeed dodgy and at least admits errors in data handling. On 6 March 2023. the Eremets paper received this:
“Editor’s Note: The Editorial Staff at Nature Communications have been alerted to potential problems in the manner in which the raw data have been processed. Nature Communications is working with the authors to resolve the matter and assess whether these will have an impact on the conclusions that can be reliably drawn.”
You can watch Hirsch’s concerns in this video he posted on YouTube:
The retired Oxford professor and open data activist Dorothy Bishop commented in a blog post about the behaviour of Eremets group:
“…asking for data is not tantamount to accusing researchers of fraud: it should be accepted as normal scientific practice to make data available in order that others can check the reproducibility of findings. If someone treats such a request as an accusation, or deems it “unreasonable”, then I’m afraid it just makes me suspicious.”
Falling Walls
Indeed, the correction and the editorial not suggest other people are suspicious also. Maarten van Kampen is used to deconstructing dodgy papers and sometimes even causing their retractions in top journals. Not just by Ranga Dias or some Iranian papermill fraudsters. Even by German researchers. Even by Nobelists. Even when they scream at him. Ask Thomas Südhof.
Maarten also wrote to the Max Planck Society to complain about Eremets’ refusal to share data, despite signed agreements to do so. Their Internal Audit Department at the Administrative Headquarters in Munich replied and announced to investigate.
In between, Maarten used his technical skills to extract some of the raw data from the published figures! He then posted his detailed analysis on PubPeer. TL;DR: maybe Eremets’ superconductor discovery is not quite what it’s claimed to be. More research is needed, you know.
The basis was Hirsch’s analysis, as Maarten writes:
“Being denied the numerical data Hirsch had to resorts to tricks to show the issue. In for example his paper Can linear transformations bend a straight line? he draws a red line between the 30 mT and 100 mT datapoints of both 100 K curves in Fig. 3(a) and 3(e):
In Fig. 3(a) the cyan 100 K points sit below the red straight line (yellow highlight). But in Fig. 3(e) the blue 100 K points are scattered around the line. And there is simply no way that a linear transformation / linear background subtraction can make that happen.
The above indeed appears to be a critical issue: the authors use Fig. 3(a) to find at what field there datapoints deviate from linear to find a critical field. But at the resolution of Fig. 3(e) its 100K points just seem to sit on a straight line.”
Maarten then used the figures from the published PDF file of the Minkov et al 2022 paper, which are conveniently stored in vector format. He then extracted the numerical data from the figures, and even validated his method using Fig. S12 which was added in the Correction.
“The curves in the above figure should correspond to the red 160 K curves in Figures 3(a) and 3(e). Panels (a) and (b) of Fig. S12 show the as-measured data and hence should correspond to Fig. 3(e). Fig. S12(c) shows the effect of the background subtraction procedure. The first bit indicated by a red line and the label Hp should correspond to the red curve plotted in Fig. 3(a).”
Maarten writes that he extracted the field and magnetization values of all data points shown in Fig. 3(a), (b), (e), and (f) with a rather high accuracy. Here is his validation, where Fig. 3(e) data is plotted on top of the Fig. S12(b) and (c) curves:
“The green points in the left panel above are directly digitized from the Fig. 3(e) 160K curve (-0.1…0.1 T). The fit is perfect. This is somewhat odd, though. In Fig. 2(e) it is written that the 160 K curve is offset by 2e-8 Am^2. Apparently the raw data shown in Fig. S12(a) and (b) have that exact same offset.”
And now, the comparison of Fig. 3(a) with a background-corrected version of Fig. 3(e):
“The left panel shows in red the recovered 160K Fig. 3(a) data and in green the recovered-and-corrected Fig. 3(e) data. Up to ~35 mT the two curves are identical, but above that large deviations occur. The green ‘Fig. 3(e)’ curve may show the same trend, but has much larger noise. The right panel shows the difference between the two curves. Below 35 mT that difference lies below ±3e-11 Am^2. This is as perfect as it gets when the Fig. 3(e) data recovery resolution is ±2e-11 Am^2. Above that the errors explode.”
Maarten then repeats Hirsch’ initial analysis of the Fig. 3(a/e) 100 K data using the numerical data he extracted:
“Again the 3(a) and 3(e) curves are identical (±3e-11 Am^2) up to 30 mT, but beyond that the 3(e) curve shows large noise and even takes a different path. I have added two red lines connecting the 30 and 100 mT points as Hirsch did in his Can linear transformations bend a straight line? paper. And that obviously leads to the same conclusion: the 3(e) data is near-linear in the 30-100 mT range, whilst all 3(a) datapoints sit below the 30-100 mT line.”
It gets worse. Minkov et al used two different materials in Fig 3: H3S and LaH10. Maarten also plotted the (b)-(f) panel pair:
“The conclusion is the same: below 30 mT the datasets match to the ±3e-11 Am^2 level, above 30 mT the raw data largely deviates:”
Maarten’s summary of his findings:
“The above analysis shows that for all four datasets we could check the processed curves in panels (a/b) do not correspond to the raw data in panels (e/f). However, below 30 mT the curves are exactly (to within our accuracy) identical. This precludes that the authors are displaying different datasets, for example using a carefully measured M(H) curve for Fig. 3(a) and showing a noisy one as raw data in 3(e).
For the same reason the curves in Fig. 3(a) cannot be the result of averaging a large number of noisy curves: the <30 mT data does exactly match with the (single) raw measurement shown. Alternatively, one can also estimate that the signal-to-noise should improve by a factor of ~10 to obtain the smooth curves displayed in panels (a/b). To make this gain one would have to repeat the measurement 100 times. For each measurement one would have to heat the sample above Tc and then again cool down without field to measure the virgin magnetization curve. This would take an exceedingly large amount of time. Note that the measurements themselves already show signs of ‘time-optimization’. Up to 30 mT the magnetization data is taken in 1 mT increments, above that with 5 mT steps. In hindsight this may not have been an optimal choice as the author-derived Hp values (deviation from linear M(H) behavior) sit mostly above 30 mT.
The authors determine a field Hp from the smooth data in Figs. 3(a) and (b). The first part of the virgin M(H) curve is linear in field. When the magnetization starts to deviate from linear (by an unspecified amount) the authors say the field Hp is exceeded. This field is plotted in Figs. 3(c) and (d):
In panel (c) the authors find an Hp of 65.5 ± 3.5 mT for the 100 K point. This may be feasible with the smooth curves from Fig. 3(a). But one can certainly not derive that value from the ‘raw data’ curve shown above in Comparison of Fig. 3(a) and 3(e).”
And then it got even worse.
Maarten learned that the Eremets group used a S700X SQUID magnetometer by Cryogenic Limited which for technical reasons apparently can’t correctly measure what they wanted to measure.
The device has two current supplies, one for a low, ±50 mT range, the other for the full ±7000 mT range. It seems, the authors measured their M(H) curves up to 30 mT with the low-noise current source, then switched to the other current source, which introduced massive noise. Maarten explains that the data was truthfully reported n Figs. 3(a) and (b) up to 30 mT, but after that “some sort of smoothing procedure” was applied , which was “not disclosed in the publication, nor in its correction“, and not even after “1.5 year of increasing peer pressure“.
It is the validity of this smoothing procedure which Maarten says is very hard to justofy.
“100 K Fig. 3(a/e) data extracted from the as-published figures […] The blue points then are the smoothed M(H) data and the green points the actually measured data. Using PPT and human intelligence I added one possible smooth curve in red.”
Maarten’s verdict:
“We are looking for a smoothing procedure that puts its smoothed points below 13 of the 14 measured points in the 30-100 mT range.”
And that’s why the raw data is unavailable for scrutiny. Maarten’s raw data however, unlike that of Eremets, is ready available via this link (also posted on PubPeer).
Despite the extensive correction, conclusions of the Minkov et al 2022 study seem to be affected a bit, worse than one initially thought, some conclusions may be poised to fly out of the window. Maybe this is why Eremets refuses to share the raw data? Maybe Eremets should retract this paper, repeat some experiments, and then try again? Maybe the superconductor field is cursed? Just like the stem cell research field?
“After the huge box-office success of “Nature 2020: Room-temperature superconductivity in CSH” this March our Nature studios released a sequel with the same star-studded cast: “Nature 2023: Near-ambient superconductivity in N-doped LuHx”. – Maarten van Kampen
But the superconductor community has better things to do than navel-gazing.
The Witch Hunt
I was included in a lenghty email exchange (which originally concerned the APS March Meeting 2024) between Hirsch, Eremets, his first author and group member Vasily Minkov, and other peers.
In one email Minkov, a native russian, shares 2 preprints by him and Eremets. Talantsev et al 2023a and Talantsev et al 2023b, both announce to refute Hirsch’s and Frank Marsiglio‘s criticism of Eremets’s superconductor claims. In that email Minkov then progresses to accuse someone – obviously Hirsch – of all possible things:
“In these papers, we meticulously document instances of data manipulation, cherry-picking, and misleading analysis techniques. Specifically, we highlight instances where certain conclusions were drawn based on flawed models, selective data manipulation, and the disregard of crucial reference measurements. These practices undermine the integrity of scientific inquiry and pose significant challenges to the advancement of our understanding in the field of superconductivity.
It is important to note that despite our repeated requests, certain individuals have failed to provide adequate details regarding their data processing methods. This lack of transparency raises serious concerns about the validity of their findings and adherence to ethical scientific practices.
Moving forward, it is imperative that we maintain a commitment to rigorous methodology and transparent reporting to uphold the integrity of our scientific endeavors.”
Evgeny Talantsev, the first author of those two preprints, used a russian institutional email address to equal Hirsch to flat-earth science denialists when replying to Hirsch’s invitation to watch his talk:
“I will wait to watch the video, until International Conference on Particle Physics and Cosmology (COSMO) invites Professors from The Flat Earth Society to discuss problems in quantum gravity. “
In other emails, he also accused other people of unethical practices, possibly their own criticism and refusal to gang up on Hirsch made Talansev angry. I will refrain from commenting on why russians are ALWAYS the victims and accuse others of things which actually, ah never mind.
Simon Kimber, one of those who helped expose the Dias fraud, replied to Talantsev:
“With respect to hydride superconductivity, I have always been of the opinion that fields get the researchers that they deserve. You guys and girls (irrespective of your own merits and opinions) have ended up with Ranga Dias. I’d think a bit about that personally. That’s the elephant in the room, not details related to other people’s publications.”
I think he is right!
In an email from 11 May 2024 addressed to dozens of peers, Eremets, co-signing with Minkov, referred to their “serendipitous discovery” of the 200 K sulfur hydride superconductor (SC) as a breakthrough “of enormous importance“. Eremets also wrote:
“However, there are non-scientific problems. The Dias affair cast a shadow of doubt over the field. Another problem is J. Hirsch. He questions this SC from the beginning because it contradicts his “hole SC” theory. To support it, he has engaged in unscrupulous practices, including falsifying analyses and selectively presenting data to support unfounded claims. […] In addition to publication, Hirsch’s tactics include manipulation of public opinion, personal attacks on our team members, and threats and complaints to our management and funding agencies. This has paralyzed the work of our group, put a lot of stresses.
About sharing our data. Hirsch’s activities force us to be more cautious about sharing our data.”
“Threats”? “Falsifying analyses”? I know how those things work in academia, and am pretty sure this is NOT what Hirsch ever did, no matter how annoyingly he may have been when demanding the raw data. Never mind. The Harvard professor Subir Sachdev replied to Eremets’ tirade:
“I take this opportunity to congratulate you on your remarkable scientific achievements, and look forward to more path-breaking results from your group with the full support of our community.”
From “analysis and conclusion of our paper remain valid” via drafted correction to “the authors retract this publication”. A guest post by Maarten van Kampen.
“Can we just please see the data, the full data please? What is the reason for Max Planck to withhold it? Please post the full original raw data from all experiments on Zenodo. It takes 30 minutes.”
Indeed. Especially since even the journal’s editors publicly admit that the data was, uhm, beautified. Victor Galitski of University of Maryland, joined Frolov’s demand:
“Dr. Minkov and all, There was a request to you from Dr. Frolov and others to provide full data. Can you provide it, ‘yes’ or ‘no’? I am one of the moderators @arXiv in superconductivity and it would be useful information for our team. I am also going to forward your reply, if any, and the correspondence below to the Nature Editors handling superconductivity content. If there is no response within a week (by close of business 5PM CET April 2), I will assume the answer is ‘no.’ “
“MIT’s receipt and review of allegations of possible research misconduct by my office are treated as confidential under MIT and, to the extent applicable, federal policies. MIT does not intend to disclose its receipt and review of these allegations to others. By the same token, you therefore may not disclose to others that you brought…
I contacted the parties involved in Hirsch’s email exchange for comments. Silence back.
But Warren E. Pickett, emeritus professor at University of California Davis, replied. He described Eremets as “the single most highly regarded high pressure experimentalist today“, had only scorn for Hirsch (“don’t believe nor publish what Hirsch says or writes“), accused him of “threats and harrassment“, raged against “this (government sponsored) “panacea” of data sharing“, and instructed me to “talk to several leading scientist in the experimental high pressure science field.” One of those he listed was Russel Hemley of University of Illinois. To clarify: I was instructed to learn from the dude who collaborated with the superconductor fraudster Ranga Dias and then tried to save him with a preprint of his own, “independently” claiming that Dias’s fake superconductor was real. And who was appointed by University of Rochester as investigator of Dias, which of course led to his total whitewash (read June 2024 Shorts and April 2024 Shorts).
Pickett also eventually shared his views about For Better Science:
“Sensationtilist reposting, without any regard to tracking down what is happening, to actual journalism. There seem to be many of these web sites“
What can you expect from someone who authored together with Dias, Hemley and Eremets “The 2021 room-temperature superconductivity roadmap” in March 2022. The roadmap failed to predict TWO Nature retractions and fraud findings for Dias, as well as the Eremets circus above, so maybe don’t trust these roadmappers as your guides?
Two sets of events for Women in STEM: the theoretical physicist Alessandro Strumia, soon likely ex-CERN affiliated, decried feminist conspiracies and the discrimination against males like himself, in a workshop talk on gender. Right after, the Nobel Prize for physics was finally after 55 years given to a woman. Thing is: one of the other…
Especially because of this. For a year, “A community letter recounting demeaning, intimidating, and threatening misbehavior” has been circulated to anyone who matters. It accused Hirsch of all possible crimes – “Professional misconduct in handling of data“, “Inflammatory Language“, “Degradation of leaders of superconductivity theory“, “Bullying and threatening conference organizers“, “Threatening employers of scientists“, “Intimidating students“, “Abuse of Seniority“, “Aggressive, harassing action” and even denial of “Scientific truth“, i.e. of the holy dogmas of superconductivity. The main distributors of this 19-page long pamphlet were Ranga Dias and his associate Ashkan Salamat. Yet the latter referred to it as “The Pickett letter”, one wonders why.
The anonymous letter in perfect native-speaker English also accuses Hirsch of harassing “early career faculty“, most obviously the assistant professor Ranga Dias is the victim:
“Hirsch has contacted university leadership of early career faculty who have reported recent discoveries which, in Hirsch’s view, violates his theory. He has sent several emails to Department Chairs and to higher officials such as Deans, University Provosts, and University Presidents, stating their culpability in scientific fraud if they do not take action against individuals with whom Hirsch is having a scientific difference. An investigation concluded that there had been no fraudulent behavior. […] Hirsch’s aggressive behavior continues to the present time (October 2022), now trying to intimidate students of his scientific foes into changing their stated declaration regarding an editorial decision on their paper. The apparent reason, implied but not stated, is that they could be subject to a charge of scientific misconduct.”
Later, Dias’ students and even his collaborator and business partner Salamat all turned on him, accused Dias of fraud, and requested retraction (read September 2023 Shorts).
And this is why this pamphlet matters also for the Eremets case of denied data sharing:
“Hirsch reanalyzed data without knowledge of the data collection process and then published his reanalysis without consent or concurrence of the methodology by the author who provided the data. These activities appear to constitute misconduct in the handling of data of others.”
Wow. Publishing unreliable science is NOT misconduct, but reading such papers without permission is.
“The business of selling authorships and citations needs a steady supply of paper-shaped vehicles. It is most efficient to produce these in assembly lines that focus on a narrow topic.” – Maarten van Kampen
There is even a bullet-point list at the end, with Hirsch’s “serious infractions”:
violating accepted treatment of the data of others
intimidation of the students of other PIs
harassing several leaders of the community
attempting to damage career opportunities of physicists The serious but still unethical infractions include
defaming leaders in theoretical superconductivity
attempts to damage the entire superconducting community
frequently posting (sometimes publishing) unscientific and inflammatory writings, unfit for a professional physicist
Psychologists call this “projection”, which is apparently what the author(s) of the letter suffer(s) from. Here is this pamphlet, in a version commented and debunked (in red) by Dirk Van Der Marel (who himself is falsely accused of plagiarism there, on page 11):
I wrote to Pickett, asking him to deny that he indeed wrote that awful anonymous pamphlet. Pickett replied and didn’t deny being the author, even when I thanked him for confirming his authorship. He only meekly stated that there was no reference to Dias in that letter except in Hirsch’s own emails in addendum. Which is obviously not correct, see above.
Sad and pathetic, don’t you agree? I almost start missing Ruggiero Santilli.
EU Commission gives €6 million to an obscure German start-up, promising to convert sewage to synthetic fuels. Internet sources suggest behind all this is “Professor” Ruggiero Santilli, the litigious “Florida Genius”, eternally self-appointed Nobel Prize candidate and sock- puppeteering businessman. Exactly the kind of “fringe scientist” Smut Clyde likes to write about!
As a scientist, one doesn’t have to love Hirsch, but one at least should love science and not confuse it with a fairy-telling ego-trip. This is where Pickett and the superconductor physics community utterly failed. Instead of fighting against massive fraud and bunk in the midst, they fell upon the whistleblower Hirsch and everyone who dared to collaborate with him.
Call the Police!
The irony is that the open data activist Frolov (who never replied to me) also did something which was not OK. He was the organiser of the International Conference on Reproducibility in Condensed Matter Physics in Pittsburgh, which took place 9-11 May 2024. Marsiglio was invited as speaker, Hirsch was admitted in January 2024, but was instructed that he won’t be allowed give a talk. After Hirsch protested, Frolov informed him on 9 April 2024: “Bro you no longer coming“. But Hirsch already paid for his hotel stay, so he was determined to attend the conference. On 7 May 2024, Frolov wrote to Hirsch, in a chain of emails:
“I informed Pitt Police” “Please do not attempt to approach the event or me, I will ask the police to interfere” “I am forwarding all communications to Pitt police. Please do not attempt to show up at this event. You will not be allowed to attend” “I ask you not to attempt attending the event. Campus police are on notice.”
This is unusual even for academia, and yet nobody in the email chain protested, except Israel Felner of Hebrew University, who said: “Calling the police to prevent attendance at a scientific conference is suitable for totalitarian and/ or dark regimes“. Pickett however explained to me in this regard that the organisers must “know some very pertinent facts that led to their exclusion of Hirsch.” Now, this insinuating slander is very typical for academia.
Cold Fusion is back, and EU Commission now funds it with €10 million. One project specifically builds on Fleischmann and Pons, the other is run by Italy’s most notorious Cold Fusion loon, Francesco Celani.
What next? I don’t think Eremets paper will survive now. But the superconductor community will continue blaming Hirsch for everything.
Maarten van Kampen predicts this future retraction notice for Minkov et al 2022:
“We have now established that some key data processing steps—namely, the smoothing procedure applied to the raw data used to generate the M(H) plots in Fig. 3a and Fig. 3b—used a non-standard, user-defined procedure. The details of the procedure were not specified in the paper and the validity of the smoothing procedure has subsequently been called into question”
Update 27.05.2024
New things came to light!
Turned out, already 8 months ago Jorge Hirsch submitted a Letter to Editor inNature Communications, which is similar to his December 2023 preprint “On the Author Correction to “Magnetic field screening in hydride superconductors“. The Letter to Editor hasn’t been published yet.
At around that same time, on 29 September 2023, Hirsch received this letter from Springer Nature’s Research Integrity Director Chris Graf:
“The journal contacted the authors regarding your request and the authors explained why they chose not to share their data with you. I acknowledge that one might disagree with the reasons provided, but the authors have sufficiently explained why they have not shared their data with you. As Springer Nature considers the correspondence with the authors confidential, we cannot share with you any further details. In addition, you also raised a number of concerns with the article itself. The authors have shared the data with the editor and after further consideration, the editor decided that a correction of the article is appropriate. […] we recognize the right of the authors to not share the data with you, in line with the authors’ chosen data availability statement.”
Obviously, at Springer Nature there is an exception to the data sharing agreement if you don’t like the person asking for the data.
Speaking of. Below is a certain Max Planck researcher warning his audience of the dangers of witches: Hirsch and Marsiglio. And calling for a witch hunt, maybe?
Reader, this is a real unedited photo from a seminar on superconductor physics.
One-Time
Monthly
I thank all my donors for supporting my journalism. You can be one of them! Make a one-time donation:
I thank all my donors for supporting my journalism. You can be one of them! Make a monthly donation:
Demanding raw data? OMFG! You guys need to learn how to think like people in finance:
This was quoted at me so many times back then when I had to work with such brillian minds. Every now and then I find myself humming it to a tune.
‘Camille Samuels, a managing director at Versant Ventures, cautioned that putting researchers on the spot to prove their data could set a bad tone for subsequent relations with a company’s founders. “I think the best way to prevent yourself from funding biotechs that have a faulty scientific basis is to develop a trusting relationship with the scientific founders,” she told SciBX. “I think that starting a productive, long-term business relationship is hard to do if you use a ‘guilty before proven innocent’ approach.’
Look, there’s a lot of controversy going on. I can’t really comment on the Hirsch vs Eremets stuff since I don’t know the story, aside from some of Hirsch’s code doesn’t handle floating point errors very well. The retraction of Dias’ work is probably justified because of the weird ways he handled the data, but insinuating reproduction of results in a preprint by a different group with different students and different equipment is faked, by putting “independent” in quotes, is just slanderous and conspiratorial.
> I can’t really comment on the Hirsch vs Eremets stuff since I don’t know the story
That was kind of the topic of the first bit of this post. The shorter version is available on PubPeer, here and here.
I believe you refer to the preprint of Hemley, reproducing the LuH superconductivity reported by Dias. I have a half-finished FBS post on the NSF report on Dias. It mentions that Dias was consistently unavailable for interviews, in one case (April last year I believe) because he was supporting Hemley with experiments. The NSF reports dissects Dias LuH results as a combination of fabrication and cherry picking noise from a loose contact. The Hemley preprint data look exactly like Dias’ loose contact data.
Hemley appears to have been external investigator in the 2nd URochester investigation. The NSF was able to see that Dias served the 2nd group of investigators a different story than the 1st. And that these were incompatible. I cannot belief Hemley did not see this.
I honestly have no words for such completely blind and unquestioning support for a rotten apple in the superconducting community.
My 2 cents: it is fully justified to put “independent” in quotes here. Hemley has proven to not to be the most critical thinker in the superconductivity field.
Just re-read your comment Adam. And noticed you stating (emphasis mine):
The retraction of Dias’ work is probably justified because of the weird ways he handled the data, …
There is really no reason I can think of why someone would write ‘probably’ here. Or ‘weird ways he handled the data’. Dias a a serial fabulist who fabricated data in many papers. He build his career on a string of lies, starting from his partially plagiarized PhD thesis (as a minor point) to his latest Nature 2023 retraction. Many of his papers are now retracted, with more to come. The NSF report meticulously details how Dias fabricated data and bullied his PhD students in complying with him. The latter is also detailed in this chilling article. Dias is as bad as it gets when it comes to scientific fraud. He has done irreparable damage to people and a whole field of science.
His retractions have nothing to do with a “weird way” of handling data. In some instances he literally took a measurement(?) from his thesis, multiplied it by a polynomial, and passed it off as a new superconducting material. The NSF investigation regularly speaks of fabrication.
Can you please read the NSF front to back? And then For Better Science? And then do some soul searching as to why you used ‘probably justified’ and ‘weird ways he handled the data’?
Of course it’s weird, we should be even satisfied with this label (am not). It would be blasphemous to name it fraudulent or even manipulative, all weird practitioners should be very careful cause they may be next ones to attract the spotlight on themselves.
Adam, read the UR investigation report. Please. You obviously have no idea how absolutely damning the report is if you say the retraction is “probably justified” and characterize the blatant fabrication as “weird ways of handling the data”. Dias was caught absolutely red-handed. It’s all there in the report for you to see if you care to become better informed.
Furthermore, noting that Hemley’s work was not truly independent is neither slanderous nor conspiratorial. It is a simple fact – Hemley and Dias were co-PI’s on the same DOE grant supporting work on superconductivity. That on top of the additional facts Maarten laid out in his reply to you.
A personal experience – asked authors of ARPES experiments for raw data for calculation comparison, to no avail. Had to resort to graphical comparison and even a reviewer criticized us for doing it. Papers in this topic more and more resemble stories with beautiful pictures and plots and their authors are not that inventive in hiding the data, they simply do not respond to a request, instead of stating something of the sort “My PhD student’s dog ate it”. 🙂
AV, it seems the process has been somewhat refined since I dealt with raw data acquisition about 15 years ago. This is in the context of the pharma industry attempting to do ‘target validation,’ i.e., reproducing interesting claims in in-house labs to determine if ‘bringing it to the clinic’ is viable.
Approximately 70% of the requested data was either never provided, incomplete, so sloppily (mis)labeled that we instantly lost interest, or so bloated that it was almost impossible to find the relevant parts.
Seems so, J. Journals from at least the natural fields of science require now Data Availability Statements or something similar for publication, however we see that these are void, or just for the looks. Same goes for the Ethical statements, of course. And the editors vigorously defend the authors that have been revealed violating these statements.
Science has turned into a publication-citation rat race, where “analyzing papers and asking for raw data is research misconduct” to cite Leonid here from slightly above. The h-index is the Lord, the citations count is the Messiah. There is almost no way to remain in this race performing proper scientific research.
“Dear Prof. Hirsch,
Our ins3tute fully supports the work of Mikhail Eremets and his team.”
In November 2023, Lelienveld doubled down on Hirsch:
“As emphasized earlier, our institute fully supports Mikhail Eremets.
Rather than trying to discredit his experimental data, you may want to consider performing your own experiments and contribu1ng the data to the scien1fic debate.”
Claudia Felser, VP of Max Planck Society, replied to Hirsch in October 2023 and basically explained that the raw data exists but is not available for external scrutiny, especially to Hirsch whom she seems to suspect of being a dangerous psychopath and a research fraudster:
“First of all, may I point out that the data on which the publication is based are available at the Institute. At the same time, I can confirm that, in response to your inquiry, it has been verified that the data support the results presented in the publication.
The inaccuracies that came to light during the review of the publication were corrected on 01 September 2023 in a corrigendum agreed with the peer reviewers.
In conclusion, it is therefore clear to me that there was no scientific misconduct on the part of our employee and that further action on the part of the Max Planck Society is not indicated.
I do not consider myself responsible for your request for access to the raw data. This is a matter for the authors, who have, however, pointed out in their statement to me that access to data can be refused if there is a fear that one’s own research will be impaired or if there are justified concerns about possible misuse of the data.
By way of explanation, the authors refer to a dispute they had with you about the use of altered raw experimental data in your publication Supercond. Nov.”
Hirsch wrote to the President of Max Planck Society Patrick Cramer who then directed him back to Felser. Who then wrote to Hirsch in January 2024:
“The President has asked me, as the responsible Vice-President of the Sec<on, to give you a final answer on this matter. We have looked into your request again and s<ll see no reason to take further action against Mr Eremets.
We had already informed you that Mr Eremets had expressed concerns about making his research data available. We believe that Mr Eremets has sufficiently justified these concerns. Given that the editors of Nature
Communications, and therefore independent third parties, have also considered your concerns and have not granted you access to the data, we see no need to appoint a further panel of external experts to assess
the situation.
With the corrigendum published by Mr Eremets, the case is closed for us, and we would welcome it if the discussion could be conducted on a purely scientific level in the future.
In the absence of concrete evidence of misconduct, we urge you, in the spirit of fair science, to refrain from making allegations that the authors
have transformed and manipulated the measured data, or even fabricated data, without due process of law.
Please accept this letter as final. Unless I become aware of actual misconduct, I will not pursue this matter further."
They obviously expect for Eremets to get a Nobel prize. Good luck.
“Dear Dr. Hirsch,
following up your request, we have reinspected our records: We had received a detailled report from our DPG Ombudspersons Prof. Hardo Bruhns and Prof. Roser Valenti, presented by the DPG Chief Executive Officer in the last meeting of the Board (Vorstand).
This report is based on an indepth analysis of the case and extensive exchange letters with you. With that in mind the Board (Vorstand) observed no indication of misconduct such that the matter is considered resolved in accordance with Article 12 of the Statutes of the German Physical Society.
Upon your request, in addition, I had a personal and intense exchange with Prof. Roser Valenti last Saturday, 13.04.24, again supporting that observation. Since there have been no new arguments since then, including the ones you provided to me, we confirm again that the case is closed.
Sincerely, Joachim Ullrich”
What a “surprise”, have met Valenti at a Cost Action conference. Several years later as an editorial board member she made us a “favour” of justifying a rejection of our manuscript even from a review in Phys. Rev. B.
The absurdity has gone beyond any limits and surely it will continue to dig deeper and deeper.
We will not make the data available, you should keep scientific level of discussion and case closed. The party is always right, even when it’s wrong at least sounds a little bit better…
Here are some legal notes on “verbal expression offenses” under the federal German Criminal Code (StGB). These legal provisions probably also exist in other countries, they are taken from the prosecution of intellectual property and copyright law. Accusing a person (e.g. Dr. EMERETS) of having disclosed untrue intellectual property to the scientific public, and thus having gained an advantage such as highly appreciated acknowledgement, can have criminal consequences if the injured party (e.g.Dr. EMERETS) files a criminal complaint for defamation with the public prosecutor’s office (StGB §187) against the accuser (e.g.HIRSCH). Then, however, there is a great danger that the injured party (HIRSCH) will file a “negative declaratory action” (ZPO §256) before the civil court to verify the accuracy of his allegation. Here he then has the right to present all his inconvenient factual allegations about the metrological tracking of the intellectual property of Dr. EMERET to be filed where this defamation was first made (place of the crime). This is regulated in the EU. Its understood, this legal action only works,if the injured party (Dr EMERETS) has taken criminal action against the infringer. Here some literature references from high court decision in such cases. The plaintiff must be represented by a lawyer in the Federal Republic of Germany if he files his lawsuit there; this may be different in other EU countries, depending on the EU regulations on the place of jurisdiction or place of the “unlawful act” (BGH NJW 1986, S 2507; ZÖLLER, Zivilprozessordnung, 25. Auflage, ZPO §256, Rn. 7 u. 14a); BGH Urteil vom 2.3.1993, VI ZR 74/92- Frankfurt, LKW Unfall = NJW 1993 S. 1716-1717),
………one of the first court decisions in a similar case in the FRG is shown in “BGH Urteil 12.01.1960, Aktz: I ZR 30/58- OLG Zelle ‘Plagiatsvorwurf’, in GRUR 1960, Bl 503”. At that time, the Copyright Act did not yet exist in the form it does today.
Germany has a criminal offense of Ùble Nachrede (malicious slander). For example if you keep calling someone a fraudster without proof. Not for saying their research is irreproducible or otherwise dodgy. Civil law has more opportunities for slander lawsuits, especially in USA. I am sure the reason why Eremets never sued Hirsch is not due to Eremets lacking the money.
I found particularly disturbing defining the Max -Plank group led by Eremets “a Russian lab in Germany”. Eremets left Russia in 1991. He worked in Paris, Tokyo, Osaka, Washington and Oxford. Since 2001, he is at Max Planck Institute for Chemistry in Mainz. He can not be associated to the Russian war against Ukraine. What is the link between the place of birth and studies of Eremets and the reliability of the data published in his paper?
And maybe you will notice that the names of his lab members and collaborators sre , well, russian?
I have no idea where Eremets’ political views lie, but his connection to russia and russian countrymen is undeniable.
It’s full of russists in Europe in the field of superconductivity, Germany, France, Switzerland, Belgium, Sweden. Some of the have already gone in retirement but their faithful puppets have replaced them.
According to you Eremets should have blame the two Russian media? Anyway this is not the point. There is a scientific dispute on a paper. Biasing a scientific discussion with sentences like “Russian data sharing”, or “why russians are ALWAYS the victims and accuse others of things which actually, ah never mind” is to me inappropriate and misleading. Let’s be clear, I am horrified by the criminal war of Russia against Ukraine and I appreciate a lot the effort of “For Better Science” to promote integrity and research reproducibility in science. Scientific integrity implies avoiding fraudulent but also explicit/implicit bias in the analysis of facts.
It is not just these two articles and if you care to look at Eremets papers you will see a surprisingly overwhelming presence of russian names, certainly for a scientist in Germany.
Talantsev is in russia, and I think collaborating with scientists in russian universities is morally wrong.
Again you miss my point: your position “collaborating with scientists in russian universities is morally wrong” should not interfere with the analysis on Erements’s paper on Magnetic field screening in H3S.
For Better Science 
Demanding raw data? OMFG!
You guys need to learn how to think like people in finance:
This was quoted at me so many times back then when I had to work with such brillian minds.
Every now and then I find myself humming it to a tune.
‘Camille Samuels, a managing director at Versant Ventures, cautioned
that putting researchers on the spot to prove their data could set a bad
tone for subsequent relations with a company’s founders.
“I think the best way to prevent yourself from funding biotechs that
have a faulty scientific basis is to develop a trusting relationship with the
scientific founders,” she told SciBX. “I think that starting a productive,
long-term business relationship is hard to do if you use a ‘guilty before
proven innocent’ approach.’
Hedging against academic risk
https://link.springer.com/article/10.1038/scibx.2011.416#preview
Anyways, thanks MvK, LS et al.
LikeLike
Look, there’s a lot of controversy going on. I can’t really comment on the Hirsch vs Eremets stuff since I don’t know the story, aside from some of Hirsch’s code doesn’t handle floating point errors very well. The retraction of Dias’ work is probably justified because of the weird ways he handled the data, but insinuating reproduction of results in a preprint by a different group with different students and different equipment is faked, by putting “independent” in quotes, is just slanderous and conspiratorial.
LikeLike
> I can’t really comment on the Hirsch vs Eremets stuff since I don’t know the story
That was kind of the topic of the first bit of this post. The shorter version is available on PubPeer, here and here.
I believe you refer to the preprint of Hemley, reproducing the LuH superconductivity reported by Dias. I have a half-finished FBS post on the NSF report on Dias. It mentions that Dias was consistently unavailable for interviews, in one case (April last year I believe) because he was supporting Hemley with experiments. The NSF reports dissects Dias LuH results as a combination of fabrication and cherry picking noise from a loose contact. The Hemley preprint data look exactly like Dias’ loose contact data.
Hemley appears to have been external investigator in the 2nd URochester investigation. The NSF was able to see that Dias served the 2nd group of investigators a different story than the 1st. And that these were incompatible. I cannot belief Hemley did not see this.
To top it off, Hemley co-authored the paper Observation of Conventional Near Room Temperature Superconductivity in Carbonaceous Sulfur Hydride with Dias, 2023-02. Hemley had then been lied to by Dias in his role of 2nd-wave investigator. The Nature CSH paper was already retracted. The fraud in it was there for all to see.
I honestly have no words for such completely blind and unquestioning support for a rotten apple in the superconducting community.
My 2 cents: it is fully justified to put “independent” in quotes here. Hemley has proven to not to be the most critical thinker in the superconductivity field.
LikeLike
No mate, faking data and then defending it is “slanderous and conspiratorial”.
Get a bloody dictionary if you don’t know what these words mean.
LikeLike
Just re-read your comment Adam. And noticed you stating (emphasis mine):
There is really no reason I can think of why someone would write ‘probably’ here. Or ‘weird ways he handled the data’. Dias a a serial fabulist who fabricated data in many papers. He build his career on a string of lies, starting from his partially plagiarized PhD thesis (as a minor point) to his latest Nature 2023 retraction. Many of his papers are now retracted, with more to come. The NSF report meticulously details how Dias fabricated data and bullied his PhD students in complying with him. The latter is also detailed in this chilling article. Dias is as bad as it gets when it comes to scientific fraud. He has done irreparable damage to people and a whole field of science.
His retractions have nothing to do with a “weird way” of handling data. In some instances he literally took a measurement(?) from his thesis, multiplied it by a polynomial, and passed it off as a new superconducting material. The NSF investigation regularly speaks of fabrication.
Can you please read the NSF front to back? And then For Better Science? And then do some soul searching as to why you used ‘probably justified’ and ‘weird ways he handled the data’?
LikeLike
Of course it’s weird, we should be even satisfied with this label (am not). It would be blasphemous to name it fraudulent or even manipulative, all weird practitioners should be very careful cause they may be next ones to attract the spotlight on themselves.
LikeLike
Adam, read the UR investigation report. Please. You obviously have no idea how absolutely damning the report is if you say the retraction is “probably justified” and characterize the blatant fabrication as “weird ways of handling the data”. Dias was caught absolutely red-handed. It’s all there in the report for you to see if you care to become better informed.
Furthermore, noting that Hemley’s work was not truly independent is neither slanderous nor conspiratorial. It is a simple fact – Hemley and Dias were co-PI’s on the same DOE grant supporting work on superconductivity. That on top of the additional facts Maarten laid out in his reply to you.
LikeLike
Adam, are you sure you are not Russel Hemley or his associate? As your IP address suggests. Please check and report
LikeLike
A personal experience – asked authors of ARPES experiments for raw data for calculation comparison, to no avail. Had to resort to graphical comparison and even a reviewer criticized us for doing it. Papers in this topic more and more resemble stories with beautiful pictures and plots and their authors are not that inventive in hiding the data, they simply do not respond to a request, instead of stating something of the sort “My PhD student’s dog ate it”. 🙂
LikeLike
Asking people for raw data or analyzing their papers without permission is research misconduct. Read the Pickett letter.
LikeLike
Nah, I want to play the didn’t know card in court. 😛
LikeLike
AV, it seems the process has been somewhat refined since I dealt with raw data acquisition about 15 years ago. This is in the context of the pharma industry attempting to do ‘target validation,’ i.e., reproducing interesting claims in in-house labs to determine if ‘bringing it to the clinic’ is viable.
Approximately 70% of the requested data was either never provided, incomplete, so sloppily (mis)labeled that we instantly lost interest, or so bloated that it was almost impossible to find the relevant parts.
LikeLike
Seems so, J. Journals from at least the natural fields of science require now Data Availability Statements or something similar for publication, however we see that these are void, or just for the looks. Same goes for the Ethical statements, of course. And the editors vigorously defend the authors that have been revealed violating these statements.
Science has turned into a publication-citation rat race, where “analyzing papers and asking for raw data is research misconduct” to cite Leonid here from slightly above. The h-index is the Lord, the citations count is the Messiah. There is almost no way to remain in this race performing proper scientific research.
LikeLike
J. Hirsch just shared his email communications on PubPeer.
Here are “Requests to the Max Planck Society for the data and their response”
Key quotes.
Jos Lelieveld, scientific director of the Max Planck Institute for Chemistry in Mainz, wrote to Hirsch in September 2023:
In November 2023, Lelienveld doubled down on Hirsch:
Claudia Felser, VP of Max Planck Society, replied to Hirsch in October 2023 and basically explained that the raw data exists but is not available for external scrutiny, especially to Hirsch whom she seems to suspect of being a dangerous psychopath and a research fraudster:
Hirsch wrote to the President of Max Planck Society Patrick Cramer who then directed him back to Felser. Who then wrote to Hirsch in January 2024:
They obviously expect for Eremets to get a Nobel prize. Good luck.
LikeLike
Here is the “Request to the DPG (German Physical Society) to look into this and their response “.
DPG President wrote to Hirsch in April 2024:
LikeLike
What a “surprise”, have met Valenti at a Cost Action conference. Several years later as an editorial board member she made us a “favour” of justifying a rejection of our manuscript even from a review in Phys. Rev. B.
LikeLike
The absurdity has gone beyond any limits and surely it will continue to dig deeper and deeper.
We will not make the data available, you should keep scientific level of discussion and case closed. The party is always right, even when it’s wrong at least sounds a little bit better…
LikeLiked by 1 person
Here are some legal notes on “verbal expression offenses” under the federal German Criminal Code (StGB). These legal provisions probably also exist in other countries, they are taken from the prosecution of intellectual property and copyright law. Accusing a person (e.g. Dr. EMERETS) of having disclosed untrue intellectual property to the scientific public, and thus having gained an advantage such as highly appreciated acknowledgement, can have criminal consequences if the injured party (e.g.Dr. EMERETS) files a criminal complaint for defamation with the public prosecutor’s office (StGB §187) against the accuser (e.g.HIRSCH). Then, however, there is a great danger that the injured party (HIRSCH) will file a “negative declaratory action” (ZPO §256) before the civil court to verify the accuracy of his allegation. Here he then has the right to present all his inconvenient factual allegations about the metrological tracking of the intellectual property of Dr. EMERET to be filed where this defamation was first made (place of the crime). This is regulated in the EU. Its understood, this legal action only works,if the injured party (Dr EMERETS) has taken criminal action against the infringer. Here some literature references from high court decision in such cases. The plaintiff must be represented by a lawyer in the Federal Republic of Germany if he files his lawsuit there; this may be different in other EU countries, depending on the EU regulations on the place of jurisdiction or place of the “unlawful act” (BGH NJW 1986, S 2507; ZÖLLER, Zivilprozessordnung, 25. Auflage, ZPO §256, Rn. 7 u. 14a); BGH Urteil vom 2.3.1993, VI ZR 74/92- Frankfurt, LKW Unfall = NJW 1993 S. 1716-1717),
Translated with DeepL.com (free version)
LikeLike
………one of the first court decisions in a similar case in the FRG is shown in “BGH Urteil 12.01.1960, Aktz: I ZR 30/58- OLG Zelle ‘Plagiatsvorwurf’, in GRUR 1960, Bl 503”. At that time, the Copyright Act did not yet exist in the form it does today.
LikeLike
This is far from my field or expertise. Are you arguing that one way to get raw data is by getting yourself sued by the party unwilling to share?
I did like how the NSF report become public because of Dias suing Rochester University. That was an unexpected bit of openness.
LikeLike
Germany has a criminal offense of Ùble Nachrede (malicious slander). For example if you keep calling someone a fraudster without proof. Not for saying their research is irreproducible or otherwise dodgy. Civil law has more opportunities for slander lawsuits, especially in USA. I am sure the reason why Eremets never sued Hirsch is not due to Eremets lacking the money.
LikeLike
I found particularly disturbing defining the Max -Plank group led by Eremets “a Russian lab in Germany”. Eremets left Russia in 1991. He worked in Paris, Tokyo, Osaka, Washington and Oxford. Since 2001, he is at Max Planck Institute for Chemistry in Mainz. He can not be associated to the Russian war against Ukraine. What is the link between the place of birth and studies of Eremets and the reliability of the data published in his paper?
LikeLike
Funny, Eremets doesn’t seem to mind that russian media portrays him as a “russian physicist”.
For example
https://ria.ru/20151218/1344287882.html
https://nauka.tass.ru/nauka/6817722?utm_source=google.com&utm_medium=organic&utm_campaign=google.com&utm_referrer=google.com
And maybe you will notice that the names of his lab members and collaborators sre , well, russian?
I have no idea where Eremets’ political views lie, but his connection to russia and russian countrymen is undeniable.
LikeLike
It’s full of russists in Europe in the field of superconductivity, Germany, France, Switzerland, Belgium, Sweden. Some of the have already gone in retirement but their faithful puppets have replaced them.
LikeLiked by 1 person
According to you Eremets should have blame the two Russian media? Anyway this is not the point. There is a scientific dispute on a paper. Biasing a scientific discussion with sentences like “Russian data sharing”, or “why russians are ALWAYS the victims and accuse others of things which actually, ah never mind” is to me inappropriate and misleading. Let’s be clear, I am horrified by the criminal war of Russia against Ukraine and I appreciate a lot the effort of “For Better Science” to promote integrity and research reproducibility in science. Scientific integrity implies avoiding fraudulent but also explicit/implicit bias in the analysis of facts.
LikeLike
It is not just these two articles and if you care to look at Eremets papers you will see a surprisingly overwhelming presence of russian names, certainly for a scientist in Germany.
Talantsev is in russia, and I think collaborating with scientists in russian universities is morally wrong.
LikeLike
Again you miss my point: your position “collaborating with scientists in russian universities is morally wrong” should not interfere with the analysis on Erements’s paper on Magnetic field screening in H3S.
LikeLike
It doesn’t interfere because it’s not even me who is doing the analysis.
LikeLike