“If a paper’s major conclusions are shown to be wrong we will retract the paper”: PLoS

One of the issues that comes up again and again on Retraction Watch is when it’s appropriate to retract a paper. There are varying opinions. Some commenters have suggested, given the stigma attached, retraction should be reserved for fraud, while many more say error — even unintentional — is enough to merit withdrawal. Some others, however, say retraction is appropriate when a paper is later proven wrong, even in the absence of misconduct or mistakes.

Today, apparently prompted by a retraction that fits into that last category and was, by some accounts, a surprise to the paper’s authors, Public Library of Science (PLoS) medicine editorial director Virginia Barbour and PLoS Pathogens editor-in-chief Kasturi Haldar take the issue head-on. Barbour — who is also chair of the Committee on Publication Ethics, which of course has retraction guidelines — and Haldar write:

We firmly believe that acceleration also requires being open about correcting the literature as needed so that research can be built on a solid foundation. Hence as editors and as a publisher we encourage the publication of studies that replicate or refute work we have previously published. We work with authors (through communication with the corresponding author) to publish corrections if we find parts of articles to be inaccurate. If a paper’s major conclusions are shown to be wrong we will retract the paper. By doing so, and by being open about our motives, we hope to clarify once and for all that there is no shame in correcting the literature. Despite the best of efforts, errors occur and their timely and effective remedy should be considered the mark of responsible authors, editors and publishers.

There’s a lot we applaud in this editorial, particularly that “there is no shame in correcting the literature.” We suspect that the “[i]f a paper’s major conclusions are shown to be wrong we will retract the paper” will be more controversial. We’d urge you to read the entire editorial, and of course let us know what you think.

69 thoughts on ““If a paper’s major conclusions are shown to be wrong we will retract the paper”: PLoS”

  1. I like the sentiment, but seems like a dangerous erasure of scientific history. If major conclusions are supported by the data in the original paper, and there was not misconduct or error, even if those conclusions are later found to not be broadly supported, it would a mistake to remove them from the literature.

    1. Nothing is being erased from the literature. You can visit the link and see that the retracted PLoS article remains on the web site. It’s now accompanied by the retraction.

      A retraction is a public statement, not an erasure or removal. (If journals start deleting articles for problems, as opposed to retracting them, maybe Ivan will start Resection Watch.)

      1. In fact most if not all Brazilian journals, in the VERY rare event of a retraction, completely remove the papers from public view. This is under my view shoving the dirt under the carpet.

  2. I have problems with any call to retract any paper that is proven wrong at a later date. It seems to me that its’ the responsibility of the researcher to keep up with his/her field, and know when previous work has been shown to be faulty. If we’re going to start retracting every erroneous paper, then a significant fraction of all publications are going to fall by the wayside. Retraction should be reserved for fraud or because of some error that should have been recognized by the authors that invalidate the entire paper.

    Promiscuous retraction suggests an assumption that once a paper is published, it is sanctioned as correct by ‘science.’ This is a naive understanding of the scientific publishing process. Publication is the BEGINNING of criticism, not the end. Peer review is a minimum standard, a gatekeeper that simply lets the critical process begin. Peer review – and publication – should not be seen as a stamp of approval, requiring retraction when faults are later found.

    Whether you work in immunology or plant ecology, it is your responsibility to keep up with the literature of your field. If new work supplants old, so be it. Let’s keep the big gun of retraction for extreme cases – the literature self-corrects, with no need of official retraction.

    1. “Promiscuous retraction suggests an assumption that once a paper is published, it is sanctioned as correct by ‘science.’ This is a naive understanding of the scientific publishing process. Publication is the BEGINNING of criticism, not the end. ” – Well stated! I completely agree!

    2. In reply to markbul September 25, 2012 at 5:15 pm

      “significant fraction of all publications are going to fall by the wayside” is the point.
      Science, not stamp collecting.

      “should have been” It is not like that.

      I am confused by the second paragraph. If there is no retraction then publication is the stamp of approval.

      “the literature self-corrects, with no need of official retraction” How is that possible?

      1. It’s possible because the scientific literature provides context. Papers are not published in vacuo, but follow and are followed by other papers. If a paper is not followed by other papers, it is not important, and s thus irrelevant. If it is, the opportunity for correction — or, more often, stepwise amendment — arises in the subsequent papers. It is not at all difficult to think of classic papers that were only partially correct, but which ignited major and important lines of inquiry. It would be absurd to retract these landmark publications. The problem with this policy is that it appears to call for absolutist remedies in situations that will in most cases be shot through with subtleties of interpretation and methodology. What is more, it will slow scientific progress by penalizing those who take scientific risks or suggest new hypotheses.

        Papers should be retracted for faulty (non-reproducible) data, not for faulty interpretations.

      2. ” If there is no retraction then publication is the stamp of approval.”

        No, publication is not an official stamp of approval – as in stamp of correctness. Any graduate student should spend many, many hours reading papers and tearing them apart. A significant part of what gets published every year is crap, and a basic part of becoming a scientist is learning how to determine the wheat from the chaff. Publication is not the END of the process – it is the beginning. Publication does not determine correctness, it begins the process of criticism.

        ““the literature self-corrects, with no need of official retraction” How is that possible?”

        Honestly, I can’t believe you’re asking that question. The present case is a classic example. People read the original paper, thought they saw flaws, tried to replicate the results, failed to do so, and published their results.

        The fact is that retraction is not needed to ‘correct the literature.’ The literature – the sum total of the work of the relevant scientific community – has already corrected itself. That’s why we’re discussing this now. Every single person working in this field knows the whole story, from A to Z. Retraction, in this case, can only serve the interests of the editors of the journal. Retraction allows them to wash their hands of the mess.

        The case for retraction due to fraud is perfectly clear. For ‘correcting the record?’ Not so much. I seem to recall a survey of papers that garnered significant publicity in Science and Nature a while ago. After five years, many of the ‘breakthroughs’ had been proven false. Are journals really going to keep track and retract old papers every month? Where do you draw the line? What was a bright line becomes an epic headache.

  3. How could this even work?
    If B proves A to be wrong, then A’s work will be retracted. But now C comes along and his/here results are going to be slightly different from either two and then what happens? B gets retracted and A gets re-instated? This has ‘Big Mess’ written all over it.

    1. Totally agree… especially if there are subtle differences between the experiments. Its always helpful to have everything in the literature. I think it would be better if the papers were marked as “in contradiction to the findings in …”. Who is to say the second is better. And it also makes PLOS an easy target. Want to get something out of the literature thats in PLOS, refute it somewhere else. Want to get something taken out of somewhere else… good luck!

      1. In reply to Dave Bridges (@dave_bridges) September 25, 2012 at 5:38 pm

        I understand that the people from PLoS said:
        “If a paper’s major conclusions are shown to be wrong we will retract the paper”
        I don’t see how “subtle differences between the experiments” impinge on that.

        Sorry for being blunt, but you wrote “Its always helpful to have everything in the literature”.
        Does that include the stuff from the Ancients that was incorrect and held people back?

      2. David H. wrote: “I don’t see how “subtle differences between the experiments” impinge on that.”

        If you don’t see how that could be, it’s difficult to imagine that you have any meaningful familiarity with how scientific evidence accrues and is interpreted.

      3. Hint: quite often, things that initially seem to be “subtle differences” turn out to be immensely important.

        People who work on cell signaling, for example, know that much of our knowledge of heterotrimeric G proteins derives from early experiments where one set of researchers used plastic test tubes and detergent A, while another set of researchers used glass test tubes and detergent B. These “subtle differences” initially suggested that some published papers were “wrong” and others “right.” But over time it became apparent that all the studies were right; the difference was the presence of trace AlF salts that serendipitously locked the proteins into an activated transition state.

        Premature retraction of one or the other set of papers would likely have slowed the search for the meaningful difference, and could have retarded progress in this medically important field by a decade or more.

  4. Makes sense. However, perhaps retractions need qualifying, because although it is indeed true that “there is no shame in correcting the literature.”, we do need to distinguish this sort of retraction from the ones due to fraud. Moreover, “retraction” means to take back, so it also has a negative feel to it. Perhaps we should have “Honourable Retraction” (cum laude!) and “Dishonourable Retraction”.

  5. (Repost from the Plos blog)

    I think you are in danger of making a fundamental mistake here. The guidelines call for retraction in the case of fraud/misconduct or honest error that has lead to an unreliable result – and I think the contamination issue for the XMRV paper might count (though this needs to weighed against the benefit to the community of the paper trial of this story – but that is another issue).

    But there is a huge leap to stating that papers should be retracted “if a paper’s major conclusions are shown to be wrong”. Should the first estimates of the mass of the electron be retracted because later estimates converged on a different number? Should papers that showed potentially significant relationships in small samples that didn’t hold up in larger trials be retracted – even if potentially that data was used as part of the meta-study? This cannot be justified.

    The literature is a record of the progress of science – and that progress is not linear. By erasing by-ways and diversions that end up not contributing directly to the ‘final’ answer (whatever that is), you are imposing a view of science that is not true to itself and is perhaps sanitised beyond the point of usefulness.

    If you erase from scientific history the merely mistaken (as opposed to the fraudulent or compromised) you are doing a disservice to the notion of science itself.

    1. “If you erase from scientific history the merely mistaken (as opposed to the fraudulent or compromised) you are doing a disservice to the notion of science itself.”

      Sums it up perfectly. Well said.

      1. Indeed, indeed, indeed. Very well expressed. Publications document a historical process and preserve what was believed in the past. We should neither erase or denigrate the past even if we think we now have a more correct interpretation. Past observations and interpretations may turn out to be valuable in various ways. Scientific revolutions may still occur. Editor’s notations like “contradicted by …” or “believed to be incorrect starting in 20xx” may be helpful to future researchers, but any attempt to relegate papers to the pile of things that we label as wrong may come back to bite us.

  6. I agree with the earlier commenters — this seems like a recipe for disaster. It replaces peer review with editorial opinion. Of course the present editors don’t have any intention of bias, or personal dislike, but will the next editors be completely unbiased? Where do they draw the line for “major conclusions” and “wrong”? What if every figure is absolutely correct, but the authors speculate a little too wildly in the discussion? What if all the data are perfectly fine, the authors speculate completely reasonably, but five years later new data suggests their speculations were wrong?

    How much proof does there have to be that the “major conclusion” is wrong? One paper? Two? Five? If the editors just don’t agree with the speculation and are looking for an excuse to retract it, is it two, but if they like it, is it five? What if the editor publishes a paper that disagrees — is that sufficient proof?

    This seems like a really, really horrible idea.

    1. In reply to Ian September 25, 2012 at 5:42 pm

      “How much proof does there have to be that the “major conclusion” is wrong?”
      One good experiment.

      1. Wow. Have you ever actually done science? Good experiments disagree with each other all the time.

        That’s such a scientifically naive statement it leaves me speechless.

  7. It seems to me that there’s been a false dichotomy set up between “Retract the paper” and “don’t retract the paper.” If those were the only options, the argument for retracting quality papers that happened to be later contradicted might be a little stronger. But there are also additions/corrections, and it seems to me that this is the appropriate place to indicate that later results are not consistent with initial findings.

  8. This particular question is only a symptom of a larger problem: retraction is a blunt weapon for fixing problems, but it seems to be almost the only one journals have.

    Consequently, papers can be retracted for academic misconduct (data deliberately wrong), error (data unintentionally wrong or misleading), failure to be replicated (data superseded), or some procedural problem, like ethical review (data and conclusions are not challenged).

    The saying is that when you only have a hammer, every problem looks like a nail. Instead of worrying if these papers should be hit by the hammer of retraction, let’s give journals more tools in the toolbelt.

      1. I am still thinking about this. But my first thought is to create a wider set of “editorial concerns”. for lack of a better term.

        The word “retraction” sort of suggests was the original point of retractions, to expunge something from the record. But now, a retracted paper is not, in any meaningful sense, unavailable. Retracted papers still accumulate citations, as has been documented here on Retraction Watch.

        The situation now is that “retraction” is a cryptic, inconsistently applied marker for “a bad problem.” Why not just make it explicit and descriptive?

        “Contains fabricated data.”

        “Experiment not supported by replication.”

        “Incorrect statistical analyses.”

        “Correct data sharing agreements not signed as required by law.”

        People sometimes talk about “the scientific record” as though is should be some sort of pristine, distilled essence of very important knowledge. (For instance, you hear people talking about results they don’t think are important “cluttering up the record”.) It isn’t, and never has been, so trying to purify the “scientific record” is a fool’s errand. Leave it all it, but work on better follow-up, tracking, and commentary.

  9. (Adding later as more thoughts strike me) Making honest and valid experiment disappear from the scientific universe just because their conclusions is wrong, misses an important point: Probably 90% of the time I look up a paper, it isn’t because of the “major conclusion”; it’s because I want to understand the details of a procedure, or the expected result of a particular experiment.

    These experiments were not wrong; they simply didn’t show what the authors thought they did.

    Now the XMRV paper has disappeared: If we wanted to replicate the study, how would we do so?

    This retroactive scrubbing of history is going to bias the literature, prevent replication, and force scientists to re-invent techniques all over.

      1. I doesn’t matter whether the retraction “erases” the article or not—it does so effectively.

        If you want to create a record of progress on a particular topic, you write an editorial.

        This is idiotic.

  10. Let’s not lose track of a critical point here: The editors retracted this paper unilaterally, without bothering to hear from the authors. They sent one email to one author, waited a couple of weeks (including a major US holiday) and then when they didn’t hear back — retracted the paper.

    Even people who agree with this new definition of “retraction” must agree that this is frantically fast. The paper had been out for years — would waiting for another week have further contaminated the scientific literature? Would making it 342 weeks instead of 340, by sending a registered letter to the authors, have led to irreparable harm to the scientific world?

    The message here is that the editors were so excited to try out their new retraction hammer that they couldn’t be bothered with courtesy or respect.

    There’s no way PLoS comes out of this looking good.

  11. Usually PLoS Journals charge article processing fee. If a paper is retracted what happens to the fee paid – will the authors get back the money they have paid….

  12. No comments are showing on the PLOS blog yet. Here’s mine:

    If a paper’s major conclusions are shown to be wrong we will retract the paper.

    This is a terrible policy and I hope you will retract (!) it right away.

    As you yourselves point out, there’s no shame in being wrong — that’s how science works! There is, however, enormous shame in e.g. fabricating data, so if you are going to start retracting articles for reasons other than fraud, the onus is on you to distinguish very clearly between “kinds” of retractions.

    You seem to be arguing that “retraction is not a dirty word”, but that’s simply not the way it’s understood by the research community. If you want to flag a paper as having had its major conclusions overturned by later work, add a comment and link to the relevant literature — an erratum or an editorial. There is no need for the blunt instrument overkill of a retraction.

    Regardless of the above, how are you going to decide that a paper’s major conclusions are wrong? What happens when new evidence shows up to indicate that the paper was right all along? Will you start retracting retractions? Of retracted retractions? Etc. The growth of experimental knowledge is a slow, messy process — it’s just not possible to put neat little “right” and “wrong” labels on it.

    Finally, it simply beggars belief that you would take such drastic action without adequate consultation with the authors. The only thing that could possibly give you as editors that right is clear evidence of misconduct.

    Tell me: as an author, why would I submit my work to a journal wherein I might at any time find it emblazoned with the dreaded scarlet R by unilateral editorial fiat?

  13. What happens when a new paradigm replaces an old one? Newton retracted by Einstein?
    Retractions have a useful, if partially fuzzy function. Let’s not blow them out of proportion.

  14. I agree with Bill and Bob, Retractions should be for blatant violations of ethics, not because somebody found a flaw although the authors did not make the mistake intentionally. This will be an overkill kind, and can only hinder progress of new scientific ideas. Every new scientific idea proposed earlier have found to be flawed partially and people start chiseling it to make it perfect inch by inch in later papers. So do you wanna scrap all those?

  15. I am quite sceptical about this.

    What will be the ‘selection criteria’ for a ‘wrong conclusion’ ?. As said by others, retractation should be for clear violations of ethics (fraud/misconduct/fabricated data) or in extreme cases like the XMVR virus or the arsenic loving bacteria.

    but again, what is a ‘wrong conclusion’ ? Just unreproducible by other labs ? How many times do you read in a paper something like : ‘previous studies, resulting in several compelling, but somewhat contradictory findings’, ‘our methods are slightly different from….’, ‘cell-specifc effect….’.

    Overall conclusions can very well be “wrong” just because that’s how science works.

    Correcting the literature by publishing contradictory papers happens everyday. When papers which seemingly contradict each other, which one should be retracted ???

  16. Ivan, I am arguing near and far for the correction or retraction of the high-profile “Australian Paradox” study, and its spectacularly false conclusion of “an inverse relationship” between the consumption of added sugar and obesity.

    The University’s response so far has featured pure intellectual arrogance: “If you wish to engage me again on this topic, please do so by sending me a reprint of a publication that you have had accepted by a peer reviewed journal”.

    That’s rather a rather silly and circular position: that is, I have documented clearly that University’s scientists have published an extraordinarily faulty paper – via a peer-review process that was an extraordinary failure – only to be told by the University that it will not consider my concerns about the disturbing failure of the peer-review process until my claim is peer reviewed and published! What a disgrace.

    The University of Sydney even has taken to pretending that quality control involved “internationally accepted standard practice”, when it must know that the lead author and the Guest Editor of the journal are the same person! Is that really standard practice? Should we have confidence in a failed quality-control process that seems to boil down to the paper being self-assessed as excellent by its famous lead author?

    Check out my email interaction with the University of Sydney at: http://www.australianparadox.com/pdf/Sept2012-Conversations.pdf

    Please – anyone – hit me hard below if you I’m the one being unreasonable.

    1. Few points to consider:
      1. The University of Sydney has financial problems.
      2. There are suggestions for a new tax on soft-drinks, which would affect the consumption.
      3. Who would benefit from the conclusion that “The findings challenge the implicit assumption that taxes and other measures to reduce intake of soft drinks will be an effective strategy in global efforts to reduce obesity.”
      Now, connect the dots and get the real picture!

      I wonder: What’s the difference between this case and getting funds for “research” from Tobacco industry?

      1. Thanks YKBOA. I see your point but I have assumed simple negligence. There certainly seems to be plenty of negligence involved. My favourite example was when the “scientists” – under scrutiny from an economist and a high-profile journalist – claimed falsely in their rebuttal that cars not humans have been consuming up to 14kg pp pa of the available sugar via the rapid growth of ethanol production. Yet a 10-minute Internet search revealed the true answer to be zero, as raw sugar is not used in ethanol production in Australia. (These are two of Australia’s highest-profile scientists!) But instead of conceding the point, the authors simply deleted the made-up false claim and rushed to publish “Australian Paradox Revisited”, which largely just restated the original false conclusions rather than dealing with the highlighted serious errors (it’s all documented on Slides 8-10, 17, 38-44 at http://www.australianparadox.com/pdf/AUSTRALIAN-PARADOX-101-SLIDESHOW.pdf ).

        But, yes, it is notable that the scientists’ spectacularly false conclusion – “an inverse relationship” between added sugar consumption and obesity – tends to support the University of Sydney’s food-stamping enterprise, whose customers include big sellers of sugar and sugary products. One’s eyebrows are raised further on finding (separately) that at least two of the authors’ big-selling diet books contain the further (false) claim that “There is absolute consensus that sugar in food does not cause diabetes”. So based on nothing solid, sugar is exonerated in the development of both obesity and diabetes, despite growing concerns elsewhere that added sugar (specifically added fructose) may be the single-biggest driver of global obesity and diabetes: http://www.nytimes.com/2011/04/17/magazine/mag-17Sugar-t.html?pagewanted=all .

        Slides 51 and 52 summarise the extraordinary Australian Paradox episode. Who has a reasonable explanation for all that? Until I hear one, I‘ll be arguing near and far that Australian Paradox and Australian Paradox Revisited should be corrected by the authors, the journal and/or the University of Sydney, starting with the misspelling of “Roberston”, that silly misreading of the authors’ Figure 5A – my Slide 9, claiming “decreased by 10%” rather than “increased by 30%”! – and that preferred sugar series that is based on an official dataset discontinued as unreliable by the Australian Statistician more than a decade before the negligent – and possibly fraudulent – paper in dispute was published (Slides 17 and 38-44).

        Full retraction of Australian Paradox may be the only reasonable response in the circumstances.

        What do others think?

        In my opinion, the Australian Paradox fiasco has become an academic and scientific disgrace. I have done my best to document the facts as I see them, and urge a serious investigation into this disturbing matter: http://www.australianparadox.com/pdf/Sept2012-Conversations.pdf .

  17. Let’s also not forget that Editors have their own theories and their own conflict of interests. This could be dangerous.

  18. This seems pretty silly. One states how one performed the experiment and reports the observations. The interpretation is always contextual and subject to change.

    I tell my students that we should always be able to stand by our data collection but we might not stand by our conclusions, which need to take into account later evidence or experimental limitations.

    There should never be any need to “retract” one’s observations unless there is an experimental flaw.

  19. RE: “If a paper’s major conclusions are shown to be wrong we will retract the paper”

    Well, YES and NO.
    Science is a process of creating new theories which refute the old theories.

    Popper (1957) investigates the role of theory in testing a hypothesis, explaining observations and predicting outcomes. He points out that the initial conditions influenced by the force of universal laws described in a theoretical framework result in outcomes. When the initial conditions can not be explained or the outcomes differ from predictions, the theoretical framework is falsified, i.e. refuted, in what he calls “error elimination” process which plays the same role as natural selection for biological evolution. Therefore, he concludes, theoretical frameworks that better “survive” the process of refutation are more applicable to the problem situations in real life, and the scientific knowledge advances through the evolution of theories.

    I am asking: How can we have evolution of theories is we erase the old ones?
    The only thing that can happen for sure is REPEATING of mistakes from the past.

    A paper should be retracted when it is proved to be fraudulent or compromised, and not because a later paper shows that the earlier conclusions are wrong. Having Frameworks to deal with misconduct is meaningless if these are not applied properly. Transparency Index has the potential to show whether the editor/publisher/institution really Do-the-Right-Thing (i.e. retract the paper) when evidence for fraudulent or compromised paper is presented to them.

    Popper K., 1957, “The poverty of historicism”, 2nd edn. London: Routledge & Kegan Paul, 130–143.

    1. What do you do if *old* papers show the methodology in the new paper is wrong?

      Or if the paper itself contains clear contradictions that are left unmentioned/discussed?

      Or if the paper draws a conclusion that clearly violates basic chemistry – but leaves that violation undiscussed?

      I have 350+ papers with such errors in a database – some published in PLoS One.

      For those who love a puzzle, what is wrong in this paper published in PLoS One?
      I will give one hint: compare Tables 1 and 4 and try and explain the differences.

      Note that any attempts at correcting the errors in this paper likely lead to a completely different paper with several changes in the conclusions. Also, several of the errors are rather basic and include the use of an incorrect equation (or rather, incorrectly interpreted equation, of which the incorrect use has already been discussed in 2000 and again in 2011)..

      1. Its important to know when a result has been SUPERSEDED, but that is very different from RETRACTION which suggests that Ms. has some serious fraud or error that the authors could have known about at the time of submission. I suggest a NEW category that cold be attached to any Ms., namely SUPERSEDED, this should be clearly visible on the contents page and in the metadata attached to the reference. This way honestly conducted work stays on record for historical reasons, and indeed because some part of the methodology might remain useful. furthermore, superseding can take place when authors are no longer available to retract, dead for example

      2. Well, yes, this is indeed PLoS One, but I have in my database also papers in Biomacromolecules, J Phys Chem B and C, Chem Comm, and even one in JACS. Not quite PLoS One…

  20. I suspect there may also be a simple utilitarian motive at play here also. CFS affects many people, people who have discovered the joys of pubmed and are naturally eager for a cause and a cure for their distress.

    I have a feeling that many people rather liked the idea of being infected by an artificial mouse viral jumping from the test tube into humans – and it has a rather salutary moral dimension to it also – to the point that they might be rather unwilling to give up the idea and this in turn might create difficulties for medical specialists in the field.

    This retraction is an attempt to try and knock the whole thing on the head, as much for the benefit of sufferers: no really, there is no xeno-mouse frankenstein virus out to get you, please believe us, pretty please. I suppose it must have seemed like a good idea for the researchers who introduced the concept at the time.

  21. If there is something fundamentally wrong with the “meat” of the paper (experiments, observations, methods, models, etc.) – retract it. If the “spin” (discussion, conclusions, interpretations, etc.) is faulty, let the paper live. Hypothetically, I discover an unprecedented cosmic body in the Universe and conclude that it is composed of pure gold. Somebody else comes long, takes another look at the object I discovered, and proves beyond doubt that it is composed of pure silver. It would be silly to retract my paper. After all, it was me who discovered this thing. Without my discovery, the other guy would have never taken a look at it. My paper was a stepping stone for him. It did not lead him astray.

  22. “Letters to the Editor” are already linked to the original article in PubMed. This would be a perfectly good venue for providing a more officious expression of concern.

  23. Our blog yesterday has generated a huge amount of comment on the issue of retractions – an unexpected but great outcome if only because it raises the issue of retractions and their function. We’ve pulled out a few issues (of the many that were raised):

    • The phrase “If a paper’s major conclusions are shown to be wrong we will retract the paper” has been pulled out of the blog and over/misinterpreted. Obviously, we have no intention of attempting (even if we could) to change how science works –by replication, by building on one’s own and the work of others, by testing and refining theories. Anyone who knows PLOS would hopefully understand our intention here. Our intention is not that retractions are used indiscriminately but to make it clear (as has been said many times before by us and others) that retractions are a useful tool in correcting the literature and do have a place when a piece of research is so unreliable that readers need to be alerted. To refer back to the COPE guidelines:
    “ Journal editors should consider retracting a publication if: they have clear evidence that the findings are unreliable, either as a result of misconduct (e.g. data fabrication) or honest error (e.g. miscalculation or experimental error)”

    • Authors clearly hate the word retraction– no matter what the intention, we don’t seem to be able to get away from the idea that somehow there is shame in the name. Do we need another term? “Correction” is not enough and “Expression of Concern” should be used when there is a degree of uncertainty (again see the COPE guidelines). There is clearly a need for a term that marks a paper as not to be relied on. Maybe we need to come up with another name, but we can’t back away from the idea.

    • Can editors act unilaterally? Yes, though obviously they should not if possible. Editors have a key part to play in correcting the literature and though ideally everyone will agree with a course of action, sometimes they won’t. The COPE guidelines say this: “Who should issue the retraction? “Articles may be retracted by their author(s) or by the journal editor… responsibility for the journal’s content rests with the editor s/he should always have the final decision about retracting material. Journal editors may retract publications (or issue expressions of concern) even if all or some of the authors refuse to retract the publication themselves.”

    • The role of the “Corresponding author” as the major conduit for communication between authors and the journal about issues such as this is important and deserves discussion and clarification.

    It’s fantastic that the issue of the reliability of the scientific record and the mechanisms for correcting it has raised such a passionate response – let’s continue the debate about that.

      1. Words acquire negative connotations through the context in which they are used. Cretin, moron, imbecile, idiot – these used to be perfectly neutral words. The word ‘retraction’ is in essence a neutral word. The stigma attached to it has its roots in the context in which this word operates. You could just as well refer to ‘retraction’ as ‘dancing in the rain’ or ‘candy floss’ – it would not take long for these phrases to acquire negative connotations as well.

    1. ginnybarbour (in your COPE role, great to see you here)

      Would COPE guidelines say anything about a case involving a publisher involved with COPE:
      a) Two of three Editors in Chief wrote two articles (2009 and 2011) that were mostly plagiarized, well-documented.

      b) The publisher allowed them (2012) to simply revise the online articles, with no comment on the plagiarism.

      c) The publisher claimed to have investigated (but no report), after email/s/FAXes to Board and execs and said all was well.

      d) The two editors silently dropped off masthead,

    2. OK, here we go. Point by point rebuttal to your insane comments:

      1) “….an unexpected but great outcome if only because it raises the issue of retractions and their function”

      —–Disagree completely, especially if the discussion is senseless and serves only to further taint the image of science in the public. Discussions about retractions need to be taken down a notch and everyone needs to take a breather and jump off the bandwagon for a few minutes. Not every discussion about retractions is warranted, interesting, valid and/or timely. Most of them these days are sensationalized and driven by bloggers trying to stir up debate and draw hits to their sites.

      1) How is “If a paper’s major conclusions are shown to be wrong we will retract the paper” any different from “Journal editors should consider retracting a publication if: they have clear evidence that the findings are unreliable”? In fact, the latter is even worse since now you have relaxed the criteria to “unreliable” from plain “wrong”. In some way almost all papers have some inbuilt unreliability, especially if the research is highly novel and cutting-edge.

      2) “Our intention is not that retractions are used indiscriminately but to make it clear (as has been said many times before by us and others) that retractions are a useful tool in correcting the literature and do have a place when a piece of research is so unreliable that readers need to be alerted”

      —– I certainly do not need an editor to tell me when a paper is “unreliable” or not. I can figure that out. It is not an editors job to play god with the scientific literature. Get over yourselves.

      3) “There is clearly a need for a term that marks a paper as not to be relied on”

      —–Based on what? Your analysis of what? The very idea of this is ridiculous. I agree that fraudulent papers need retracting, sure, but this is ridiculous. If I think a paper is crap or unreliable, i wont cite it and wont use it in any way when I am formulating ideas.

    3. Dear Ms Barbour,

      This month makes one year since I have informed COPE (case Ref.: Benach & Muntaner) about duplicate publication (in Gaceta Sanitaria) combined with copyright irregularities (Elsevier Reference: 120325-000342 and Reference: 120324-000411).

      After intensive email communication in which I have provided more than enough evidence, on 24 July 2012 in e-mail to me Natalie Ridgeway noted:
      “Your complaint has been passed to the Chair of COPE (Dr Virginia Barbour) for review”

      Could you, please, advise on the progress of this matter?

      Thank you in advance.

  24. “[i]f a paper’s major conclusions are shown to be wrong we will retract the paper” will be more controversial.

    This is not controversial, it is a load of rubbish. After reading the post by Ginny Barbour my conclusion is that PLoS editors agree with this view (that it is a load of rubbish) and that they would have preferred to leave it out.

    I work with membrane proteins and have been trying to make conclusions about structure from functional experiments. Now that we have X-ray diffraction structures for many of these it is clear that some of our conclusions were wrong. Do we have to retract the papers? Of course not! Reintrepretation of the data, when possible, is in order but retraction?…..go to hell!!

    I would recommend the PLoS Editors responsible for the blunder to retract the sentence. It is dangerous particularly if cited out of context. I think this is also the opinion of the overwhelming majority of those replying here.

  25. Fucking ridiculous. This article deserves no attention whatsoever and PLoS have lost their bloody minds if they think retracting papers that are “wrong” is acceptable. We may as well go and delete 95% of all papers published since…..well, since papers were published. Absurd.

    1. editorial processing itself is not straightforward…retractions are the only way to go them. PLoS needs to re-think about the credentials of the editorial board members (especially PLoS One may be others as well). This is serious…

  26. In reply to rory robertson (former fattie), September 26, 2012 at 4:25 pm

    Rory, you think that the paper will be retracted because you have proved unequivocally that the authors are wrong. Let me tell you that I’m waiting one year now for sorting out an obvious and very straightforward case of multiple misconducts combined with copyright irregularities, and the editor is in a state of denial, the publisher does not bother to reply, the institutions (two universities) refuse to acknowledge the misconduct (even when it is in contravention of their own Framework) and COPE in continuously deflecting the issue. All of them think that, as the proverb says: “Every wonder is for 3 days”, i.e. you will be soon forgotten, while their paper will remain (and ONLY this matters to them).

    Look here, where one is asking me “Why I keep on pushing???” http://www.retractionwatch.com/2012/09/27/left-and-right-apparently-agree-that-gmo-studies-should-be-retracted-but-theyre-talking-about-different-papers/#comment-25979

    From now on you’ll face: Ignorance, Denial, Arrogance, and even Intimidation.
    I do not know about you, but I will NOT stop until they rectify the multiple misconducts.

    1. Hi YKBOL. Good on you, and best wishes. Me? No grand plans. It’s summer and time to be at the beach with the kids and on the golf course! But I’ll keep just chipping away, doing what I can reasonably do. Any major developments will be recorded on the LHS of http://www.australianparadox.com/

      I think the University has been very unwise to defend its “shonky sugar study” rather than defend the need for competence and integrity in publicly funded research. My recent E-mail conversations with the University’s Deputy Vice Chancellor, Research have been widely read – http://www.australianparadox.com/pdf/Sept2012-Conversations.pdf – and feedback about the University of Sydney’s disingenuous defence has been very negative.

      I’m guessing plenty of Oz academics and scientists enjoy having a coffee while reading the comments online in response to colleagues’ opinion pieces, including this one: http://theconversation.edu.au/scientific-data-should-be-shared-an-open-letter-to-the-arc-9458

      Indeed, Universities and scientists outside the Australia’s prestigious Group of Eight may long enjoy “dining out” on the Australian Paradox fiasco. The University of Sydney’s “shonky sugar study” for many will become Exhibit A when B-grade universities and scientists in Australia and offshore argue that our Group of Eight universities and scientists are not as clever as advertised.

      I assume there’s an influential individual or two out there who ultimately will decide that the University of Sydney needs to do the right thing by the taxpayers who fund it. After all, this clownish “study” with the obviously false conclusion has become a menace to public health (see Slide 46 at http://www.australianparadox.com/pdf/AUSTRALIAN-PARADOX-101-SLIDESHOW.pdf ).

      Or maybe not. Anyway, it’s a beautiful blue-sky day in Sydney so I’m for the beach. How about those Sydney Swans yesterday. Best-ever Grand Final?

    2. Relating back to the topic of this thread, I think this case- like the Ulrich Lichtenthaler one- illustrates that sometimes it *is* better to retract a paper because its conclusions are ‘wrong’, in the absence of fraud or methodological errors. There is a grey area between results and conclusions, and inconsistencies can occur within this grey area – a graph shows an upward slope, but the authors conclude that it goes down; regression coefficients are clearly insignificant but the author claims they are significant. The problem here seems not to lie in the results as such, but in the authors’ interpretation of these results (i.e., conclusions). Yet these interpretations are so interwoven with the results that a reader who merely skims a paper (as busy academics tend to do) can easily overlook the discrepancy between them. It is much easier to spot a discrepancy between, e.g., insignificant coefficients labelled as such in the results section, and conclusions assuming significant effects.

  27. This is indeed an interesting debate. If an effect in the central nervous system has been classically attributed to the effect of neurotransmitter X, yet it is later discovered that the co-released transmitter Y was responsible all along? Should all the classical studies be retracted? I don’t think so. This simply reflects the auto-correcting nature of science, the emergence of better techniques and greater biological knowledge. On the other hand, if a viral DNA is found in patients with disease A, and the lab finds out soon after that it’s because the PCR water was inadvertently contaminated with viral DNA, should the work be retracted? It seems like it should, To avoid the stigma – possibly career damaging – of retracting a study, in the absence of evidence of misconduct, a special term could be devised to describe such retractions. Perhaps “technical retraction”? Or something else? Other cases are less clear. If the phenotype of a mouse model that people has been used for years is suddenly found to be partly/totally explained by a co-segregating mutation outside of the mutated gene that was thought to be causing the phenotype, what should we do? There are actually examples of such situations.

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.