Where’s the line between scientific post-publication peer review critiques and libel?

Mann4
Michael Mann, via Wikimedia http://en.wikipedia.org/wiki/File:Mann4.jpg

One of the issues that comes up frequently when we’re moderating comments here on Retraction Watch is the distinction between “I think these images look strange” and “this researcher committed fraud.” That’s a pretty important distinction, because potentially actionable cases of libel live somewhere in between, probably closer to the latter — as Paul Brookes found out the hard way last year when Science-Fraud.org was shuttered by legal threats.

We may have some new legal precedent to follow on the matter soon, it turns out. Climate scientist Michael Mann sued The National Review in 2012 after the conservative magazine published a Competitive Enterprise Institute statement as follows:

Mann could be said to be the Jerry Sandusky of climate science, except that instead of molesting children, he has molested and tortured data in the service of politicized science that could have dire economic consequences for the nation and planet.

The case has twisted and turned, as Mother Jones reports, but last week, DC Superior Court Judge Frederick Weisberg rejected a motion to dismiss by the magazine. In that decision, Weisberg wrote:

Opinions and rhetorical hyperbole are protected speech under the First Amendment. Arguably, several of defendants’ statements fall into these protected categories. Some of defendants’ statements, however, contain what could reasonably be understood as assertions of fact. Accusing a scientist of conducting his research fraudulently, manipulating his data to achieve a predetermined or political outcome, or purposefully distorting the scientific truth are factual allegations. They go to the heart of scientific integrity. They can be proven true or false. If false, they are defamatory. If made with actual malice, they are actionable.

We’ll be keeping an eye on this case.

Hat tip: LabRigger

42 thoughts on “Where’s the line between scientific post-publication peer review critiques and libel?”

  1. It may be of interest to understand the extent to which actual findings of misconduct can become the matter for a court and hence subject to other laws than those of science. Or are they by definition always in agreement?

  2. More “interesting” if you are UK based and doubtless even more so in other jurisdictions! While it does get complex, drawing comparisons between possible data manipulation and child molesters (as in the case you highlight) or between scientific critiques of data and cyber bullying (as in a recent news item in Science http://www.sciencemag.org/content/343/6169/358.full) is not useful. It trivialises restricted, but very serious social problems and, at least in the former case, lays one open to charges of defamation.

    So we should stick to the data. The problem here is that the response from institutions (journals and employers) is weak to non-existent. Out of frustration many a rash word is uttered.

  3. I think that there is an important distinction that needs to be made here.

    Scientists have an obligation to correct the scientific record particularly in their own field of study. If a paper appears with what appears to have an “excessively manipulated image,” or other such evidence of POTENTIAL misconduct (aka fraud), then a scientist has an obligation to report this to the journal, ethics departments and even ORI. As long as the communication is with those “authorities” and NOT made public, then libel will be a very high bar to reach. The accusation must be totally false AND there must be malice, the latter all but ruled out for a scientist fulfilling her obligation to her field in good faith.

    The National Review has no such obligation to protect the scientific record per se. AND, their accusations were PUBLIC.

    Paul Brookes posted PUBLIC accusations. He did good, but unfortunately fell under a different bar.

    Threat of lawsuits (and actual lawsuits) have no place in SCIENTIFIC discourse.

    The whistleblower laws are real, but not intended to protect generic and very public libel.

  4. It’s the reference to Jerry Sandusky that makes this case special. The judgment seems to understand that the comparison to child molestation is an indicator of ‘actual malice’.

    If I say you have faked a gel, that is a statement that could be either true or false. If I say you are an academic fraud, that may be defamation. But if I say you are the Jerry Sandusky of Science, that would be malice.

    I’m actually surprised that the National review is defending the case. A quick apology might have saved them a lot of money. They are at high risk if this goes to a jury trial.

    1. If the case proceeds to discovery, National Review and the author Mark Steyn will have the opportunity to dig into Mann’s research and communications and attempt to show that he knowingly committed shenanigans with his research. (I have no idea if this is true, although I note for the record that the majority of AGW researchers tend to refuse demands that their data be archived for public inspection and replication.) If Mann did knowingly commit shenanigans, then “truth is an absolute defense.”

      1. Can you provie any evidence that the majority of AW researchers tend to refuse demands hat their data be archived for public inspection and replication?

        As far as I have understood from peope knowledgable in the field of data sharing, climate science is actually well ahead of many, many other fields in terms of data sharing.

      2. I don’t think that is true. Discovery cannot be used as a fishing expedition to retroactively find something to justify libel. They will have to state specifically what information they need for their defense. They have said he manipulated data. To defend that they must have at least one example already in mind. They will then have to specify which data set, and what sort of manipulation. They cannot say “give us all your data and then we will look for something”.

        1. @Marco,
          Only anecdotal reports, nothing systematic. It could be true that most climate researchers are ahead of the curve as far as data and code archiving is concerned. But there are instances of the “big name” researchers refusing to comply with (or just ignoring) requests to archive their data sets or computer code. (I do think things are better in the last few years, maybe because of pressure from the gadflies.)

          @Dan,
          Well, if a person has a reasonable belief that X is true, and states it, and gets sued on the grounds that X is untrue, one of the defenses is to prove that X is in fact true. (Part of the glorious first amendment is that a person generally does not need ironclad proof of all relevant facts before stating an opinion.)

          If it turns out that X is in fact untrue, then the case will turn to matters such as, what constitutes reasonable belief, or did the person know that X was false at the time it was asserted as true, and so on.

    2. I agree with Dan. By adding Mann and Sandusky in the same sentence immediately alludes to Mann being a child molester (without actually explicitly stating this, but simply by word association). I would go further to state that this comment exceeds hyperbole: “could have dire economic consequences for the nation and planet”. This is a massive claim to make. A clear case of sensationalist journalism here. So, in this case, I agree that Mann should contest the statement’s validity. However, are court trials and juries needed? What happened to people just talking among themselves and then issuing an apology and perhaps a corrigendum? I am of the opinion that if errors are reported through the correct channels, namely to the editor boards, the authors and the journal’s publisher, and if no action is taken within a reasonable amount of time, then yes, we have the duty to contact other academics and to contact higher authorities and make the case public, for example on blogs, provided that all documents, web-links etc. can be provided. How else are poor and increasingly penniless scientists (with o legal or other recourse) supposed to show that there is abuse taking place in science publishing? Paul Brookes did find out the hard way, most likely because of a poor choice of words, specifically one, “fraud”. I believe that Brooks was used to set an example to others of what to expect if we were to publically expose scientific misconduct. But his case has inspired many to continue the battle in their fields of study to achieve justice and to correct the scientific record. It is easy, as ferniglab correctly states “Out of frustration many a rash word is uttered.” However, if we can show the exact error, misconduct (duplication, plagiarism or self-plagiarism), even better, to quantify it, then public exposure should not be considered some sort of act of libelous malice. Instead, it should be hailed as a great step towards transparency in the scientific process. Being able to do so freely, and responsibly, would add pressure to ensure that the literature is corrected. I guess there are all levels of whistle-blowing or ratting, or whatever other negative connotation has been bestowed upon such a “noble act” by authorities. The truth is if an unethical act has been performed, and published, that fact remains, no matter what language we use, or how euphemistic we may want to represent the fact. I am curious, how did the Judge in this case prove malice?

      1. It is extremely unlikely that this case is based around the mentioning of Sandusky – as per the judge’s own words, the crux of the matter are the use of the word “fraudulent”, and the statement “has molested and tortured data in the service of politicized science”.

        This is going to be a very tough sell for the defendants – whatever the process of discovery is going to unearth, the only reasonable way to demonstrate fraud is if Mann has confessed it himself to anybody in a written form on a discoverable record. That’s as improbable as it gets.

        1. I think the Sandusky comment is key. The plaintiff will address this point directly to the jury. Imagine you were compared to Sandusky in your professional capacity. Anyone will feel libeled by that. No hope for the defense.

          And as you say, proving fraud on the part of Mann is near impossible unless there are records that we are not aware of. He’s unlikely to have written “…and now I commit fraud…” in his notebooks.

    3. The judge in his ruling made it pretty clear that the problem was not the over the top simile, but the accusations of fraud

      . . .plaintiff alleges that CEI published, and National Review republished, the following defamatory statement: “Mann could be said to be the Jerry Sandusky of climate science, except that instead of molesting children, he has molested and tortured data in the service of politicized science that could have dire economic consequences for the nation and planet.” The allegedly defamatory aspect of this sentence is the statement that plaintiff “molested and tortured data,” not the rhetorically hyperbolic comparison to convicted child molester Jerry Sandusky.

      1. “The judge in his ruling made it pretty clear that the problem was not the over the top simile, but the accusations of fraud”

        I don’t think the judge said quite that. If I understand the case, the NR was claiming that their accusation of fraud was a mere opinion, and thus not actionable. The judge said no way. (BTW, they were using the anti-SLAPP law to defend this suit, which is ludicrous.)

        The comparison to Sandusky is important because it shows “actual malice”. There’s no way that anyone could consider that in any light except malice. That’s what makes the case for the plaintiff, and that’s what’s going to bump the damages up into 6 or even 7 figures.

        A quick apology and statement that such language does not reflect the editorial standards… blah, blah, blah… could have saved them a lot of money and aggravation.

  5. See also Mann vs Steyn Lurches Forward and Steyn Doesn’t Understand the Picture.
    Steyn is a dream defendant … for plaintiffs. It is not really a great idea to insult judges and have your legal team run.

    The Sandusky comparison may be relevant for malice, but may just be icing on the cake. Recall that the case is not just against Steyn & NR, but Rand Simberg and Competitive Enterprise Institute …
    and the latter has a long history of organizing attacks (and maybe funding some) against Mann, going back to 2001. See Strange Scholarship in the Wegman Report (Sept 2010), PDF pp.24-32.

    In 2010, it was unclear how Ross McKitrick got hooked up with CEI in late 2001, but I later found that the well-connected & opportunistic-recruiter Fred Singer had given a talk in Canada for Christopher Essex (author with McKitrick) in early 2001, explaining why then-obscure Canadians were repeatedly brought to Washington, given speaking slots, introduced to Senator Inhofe, coached, promoted via the WSJ, etc, etc, and eventually gave the May 11, 2005 talk for CEI and the George Marshall Institute … that became the “blueprint” for the Congressional show trial called the Wegman Report.

    Anyway, there is plenty of well-documented history supporting CEI malice, if any is needed. Might Mann’s lawyers be aware of these details? Possibly, given that the above report is quite frequently cited in detail in Mann’s book 🙂

    If they think they can use McIntyre & McKitrick(2005) to claim fraud on the part of Mann, Bradley, Hughes(1999), they have a problem. McIntyre’s R code used an explicit 100:1 cherry-pick to manufacture the desired results from an already-flawed statistical analysis, as per Deep Climate’s fine sleuthing in Due Diligence… and Nick Stokes’ Effects of selection in the Wegman Report.

    Complex statistical arguments seem unlikely to work in court, but code for a 100:1 cherry-pick cannot be an accident.

    If they want to invoke Wegman, that would be delightful, given FOIA Facts series.

    1. Steyn is a dream defendant … for plaintiffs.

      I’ll say: “Mr. Steyn is currently representing himself.”

      Mr. Steyn did not disavow the assertion of fact that Dr. Mann had “molested and tortured data,” and he added insult to injury by describing Dr. Mann as “the man behind the fraudulent climate change ‘hockey-stick’ graph.” Am. Compl. ¶ 28. In context, calling Dr. Mann’s work “fraudulent” is itself defamatory and parallels Mr. Simberg’s claim that Dr. Mann “molested and tortured data.”

      The interesting part of this, to me at least, as I haven’t been following it, is that the failed motion to dismiss basically was the anti-SLAPP, which was the first major hurdle for Mann. And that the defendants are no longer shielded from discovery, which is somewhat amusing, given NRO’s thinly veiled threat to abuse the process if given the opportunity.

  6. Honest to heaven, I think it’s just completely off their rockers for the National Review to use/publish the Sandusky reference on Mann. There are tons of other ways to work in a characterisation (try Mel Brooks for size, perhaps), so this is not an example of sound editorial/copydesk judgment. I can’t believe some living, breathing human being actually cleared that kind of language for publication.

  7. The line is pretty clear – if it’s true, it’s not libel. Threateners of libel suits have a tendency to deflect the argument toward the hyperbole, the specifics of the delivery, while failing to realize that simply addressing the underlying facts would make the problem go away. If someone criticizes your work, instead of getting hung up on what names they called you while doing so, concentrate on the data (that is, of course, if you actually have a case of any substance based on real data).

    Real scientists, when challenged, reach out for the lab-book, not to the lawyer or the image-management consultant.

    1. When my integrity (and therefore my livelihood) is challenged, I’m attacking back on every possible angle. The hyperbolic rhetoric and snide editorializing on your blog was ineloquent and needless; if you want people to just stick to the facts, you need to hold yourself to the same standard. If you expect people to hand over their lab notebooks, perhaps you could’ve started by making your accusations not under an anonymous pseudonym. I find that hypocritical and naive — even as I do the same.

      I could ask you to upload all your lab notebooks beginning from your graduate work up through present day (including any work you have supervised) and post a link to it here?

      1. That’s a huge undertaking (and something I am actually working on in the coming year, within reason). In the mean-time, if you email me and ask for the original data for any figure in a paper where I’m senior author in the last 6 years (US ORI statute of limitations), I’ll get it to you within a reasonable time frame (days).** Even if it’s older we can probably find those data too because almost every single piece of primary data from my lab’ since 2003 is stored on a shared drive, backed up to tape every night, and accessible to administrators (e.g. Dept. vice chair for research) and everyone in the lab’. No-one is allowed to store data on their own PC, laptop or USB sticks, only on the shared drive. This is a strong disincentive to data manipulation because if there are any shenanigans then someone WILL find out. The same drive has all our methods – no secret voodoo recipes buried on post-it notes here.

        In contrast, I’ve made numerous requests to other scientists for original data and in most cases I’m still waiting after years or more. I know of one lab where a person had to sneak in at night to copy a page from someone else’s note-book because they wouldn’t share a method. In the case of my blog, of the 275 papers noted, the original data files were posted for only one. Together, these events suggest that many labs do not pay enough attention to sharing or keeping their methods and data available. This is supported by a recent study… http://www.cell.com/current-biology/abstract/S0960-9822%2813%2901400-0

        **At risk of inviting a barrage of data requests (a la DDS attack), anyone wishing to question the integrity of someone’s data has to be precise about exactly what they’re criticizing; this is what my blog did in great detail. If you want my data, state clearly what you want it for and ask specific questions, perhaps even including evidence that led you to suspect a problem, and I’ll be happy to engage. Regarding my grad’ school work… coming right up… (buys plane ticket to go to parent’s attic 4000 miles away and extract 40 MB hard disk from 1992 Mac Powerbook 100).

        1. That’s good, and I applaud you’re chasing of fraud, a thankless task….
          but re the topic of this post, I’d suggest reading the {CEI, GMI,Fred Singer, Myron Ebell, Steve McIntyre, Ross McKitrick, Joe Barton, Jim Inhofe, Joe Barton, Wall Street Journal, National Review, etc} 10+-year attack vs Mann before forming an opinion on this one. Much of that is in Strange Scholarship in the Wegman Report.

          The primary source of claims against Mann was Steve McIntyre, who demonstrably used a 100:1 cherry-pick atop bad statistical parameters to generate misleading graphs, which Ed Wegman then used, promised to provide Henry Waxman with the code behind a Congressional show trial for Mann … and never did. About a third of the Wegman Report was plagiarized, in some cases, plagiarized text was then falsified to let non-experts invert expert conclusions they did not want. All that’s been documented in excruciating detail, reports linked in earlier posts, for quick look, try FOIA Facts 1. Much of the work was done by an psot-doc on an alcohol-study fellowship, and both she and Wegman claimed the Wegman Report on Federal grants, his from the Army Research Office. He and the post-doc later lost their WIREs:CS Editor-in_Chief jobs: miinor problem, they wrote two mostly-plagiarized papers for their own journal and got caught.

          Unlike lab experiments that may or may not be easily replicable, paleoclimate temperature reconstructions are done from data not usually collected by people doing the analysis, but by others who put them in generally-available databases. many different groups have analyzed that same data in different ways to extra signal from the noise, and any credible, comparable results generally fall within the error bars of MBH99. People need to be very careful about the different fields of science, especially when one of them is subject to well-funded persistent political attacks. Some of the same groups did this to Ben Santer earlier, and then tried for Jon Overpeck, but that didn’t work. Bu Mann has been the main target.

          1. John Mashey – are you suggesting people like McIntyre have been part of a (coordinated?) “well-funded” political attack against Mann? If so could you please provide evidence for it.

            As I said there is an unbound irony in all these accusations or innuendo against specific people, surfacing exactly in comments to this post.

          2. I wrote: ‘People need to be very careful about the different fields of science, especially when one of them is subject to well-funded persistent political attacks.’

            In the thousand or so pages I’ve written on the Wegman affair, related attacks on the hockey stick, and the well-funded, well-coordinated thinktanks that organize such things, I have never once said that *McIntyre* himself was well-funded, although CEI/GMI did pay for at least one of his trips to Washington. Actually, I thought that McIntyre’s payment was primairly in attention, like getting front-page articles in Wall Street Jorunal and getting introduced to US Senators, and getting coaching from thinktanks who used to care about him.. But that is just opinion.
            It is fact that McKitrick(2001) and McIntyre(2003) were well wired into Washington thintktanks GMI and CEI, especially via CEI’s Ebell, a key figure … and recall that this suit involved CEI, and there is a lot of well-documented history. Emails via discovery might be more fun.

            I already posted link to SSWR, but see especially pp.27-32. It references Crescendo to Climategae Cacophony, see pp.2, 10, 15, 19-20, 92-106, 166 for examples of earlier history.

          3. Thank you. I find “payment in attention” rather ambiguous – what evidence does one have on McIntyre’s motives? (Just like with Mann’s, any form of malice on the part of the scientist requires extraordinary evidence).

            You also mentioned other people in the “attack” group -politicians aside do we have evidence of any of them being funded by or through such a group?

    2. As I understand from my knowledgable colleagues, Mann’s work has been challenged and checked long ago; that is, people have used the same data (which were not his, but data collected by others) and the same methods and found the same result.

      At a later point, multiple investigations cleared him of any wrongdoing.

      Now, years later, certain people attack him with the same already debunked claims.

      It would seem that reaching out for the lab-book will not help in setting these people straight, considering that that has already been done.

        1. So the rebuttal to my statement that other scientists using the same data and the same methods got the same results (Wahl & Amman, dx.doi.org/10.1007/s10584-006-9105-7) is a lengthy opinion piece filled with conspiracies, published itself on a blog (hockeyschtick) maintained by someone who thinks the greenhouse effect does not exist.

          In other words, I am not mistaken. Regardless of how many times you debunk something with your data, certain people will continue to make the same claims.

          1. Yesterday Professor Sir Brian Hoskins, Director of the Grantham Institute for Climate Change, IPCC contributor and many other things, spoke at a hearing of a UK parliamentary committee.

            When asked a question explicitly mentioning the HS, Sir Brian replied by talking of “mistakes…in some cases” and “maybe an accentuation”. He could but did not tell a large number of members of parliament that the HS had been verified.

            A transcript of the relevant part of the exchange has been made available by commenter ‘geoffchambers’ at the Bishop Hill blog:

            http://www.bishop-hill.net/blog/2014/1/28/ar5-hearings.html

            Yeo: Do you have any anxiety that controversies that arose from previous reports – take the hockeystick graph which seems to be referred to quite frequently – do you have any anxieties that those sort of controversies may cast doubt on the conclusions reached in AR5?

            Hoskins: I think there is a concern always that mistakes have been made, as in some cases, or maybe an accentuation of one view, and it’s a group of people and mistakes will be made and that should not reflect on anything in the future for that body, and of course we should all be sceptical and we are all sceptical the whole time, so probing it must go on, and it’s not going to be taken as the Bible, but it should be taken as the view of a large group of scientists of a diverse range of where scientists come from, and this is the consensus view given by them.

          2. I don’t know what this is supposed to mean in the context of this discussion. Looks to me like Hoskins comes with general remarks on the IPCC reports and supposed controversies, not anything that says “yes, there has been one grand conspiracy to disappear the MWP”, as the link “Realist” provided essentially claims.

          3. I do not speak for ‘hockeyshtick’ or anybody else but me. I actually find it useless to argue a point by linking to partisan blog posts, of whatever disposition .

            As for Hoskins, it is my opinion that if he really believes the HS to be a multiple-verified scientific fact, he should have said so, instead of referring to mistakes and “accentuations” when asked about it.

          4. I guess it matters how the question was understood. To me it sounds like the question is about the IPCC reports, not about the hockeystick. If Hoskins understood it the same, the “mistakes” and “accentuations” in Hoskins’ response refer to the IPCC report, not to the hockeystick. Throughout the whole session it is clear that the scientists are discussing the general aspects of the IPCC process, and steer clear of discussing individual cases.

            Whether the hockeystick can be called a “multiple-verified scientific fact” will probably be in the eye of the beholder. Others have verified that you get the results reported in Mann et al when using the same proxies and methods (and even some other methods, as reported in the paper I linked to earlier). In that sense it has been verified that the results reported in Mann et al are accurately reported. Whether you consider other reconstructions with newer and more methods and/or data as scientific verification or falsification of the Mann et al reconstruction does not matter in this discussion. Fact remains there is no evidence of falsification or deliberate inappropriate handling of data in Mann et al.

      1. I am not well enough informed to argue specific cases, but there are *claims* that Mann’s results have been manipulated by cherry picking data points, analytical methods, and so on. If he were forthcoming with his data sets and codes, then his analyses could be re-run by potentially hundreds of statisticians and evaluated. Deposit in a public archive at time of publication, not years later and only when forced (such as is required now for gene array and similar data). Part of the Wegman committee controversy seems to involve refusal to share or incomplete code and data sharing.

        1. Well, going through some of the links on this website (http://www.desmogblog.com/ed-wegman-promised-data-rep-henry-waxman-six-years-ago-where-it), it looks like the Wegman ‘committee’ is actually the one that should be accused of not sharing code. As I understand it, this may be because it would show Wegman just reran a code written by Stephen McIntyre, including the same cherrypick of the top 1% of most extreme hockeystick shapes.

          Complaints about Mann’s code in the Wegman report are the usual problems with code-sharing: it’s written for your application, your computer, etc, so yes, not easy to use by others.

          1. It is ironic but not unexpected that the comments about a post on the line between critiquing and libeling of a scientist, are descending into attacks against this or that scientist, with links to deeply partisan external blog posts.

      2. Agreed.

        This is not really about science. This is about the politics of climate change.

        Personal attacks have no place in scientific discourse.

        Attacking the science (and by science i mean the data,manuscript, or even press
        release) should always be fair game.

        That’s what the whistleblower laws are about. Those laws protect scientists who make a good faith concern about the science.

        Bad faith is a pretty high bar. Personal feud no problem if the science is the least bit suspect. Reasonable exlamation for the questionable science (and the acussed is eventionally “exonerated”) is no problem for the accusor as it is her right and obligation to raise the question.

        And, science by press release. I can’t imagine any questioning if this that could be libel by a legit scientist.

        Have you noticed that those who conduct science by press conference are most likely to employ libel lawyers?

        1. Retraction Watch can’t possibly be serious. You think this smear of Michael Mann possibly straddles the line between peer review and libel? I. Don’t. Think. So.

          1. In these comments, John Mashey said:

            McIntyre’s R code used an explicit 100:1 cherry-pick to manufacture the desired results from an already-flawed statistical analysis, as per Deep Climate’s fine sleuthing in Due Diligence… and Nick Stokes’ Effects of selection in the Wegman Report.

            Complex statistical arguments seem unlikely to work in court, but code for a 100:1 cherry-pick cannot be an accident.

            The National Review wrote that Mann “could be said to be the Jerry Sandusky of climate science, except that instead of molesting children, he has molested and tortured data in the service of politicized science that could have dire economic consequences for the nation and planet.”

            I wonder what Mann would have said if NR had written about the hockey stick what Mashey wrote about McIntyre, namely that the HS was the product of cherry-picking and bad statistics used to generate misleading graphs in what cannot have been an accident.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.