14 August 2009

Real Climate Responds - Updated

UPDATE: Eric Steig writes to let me know that Prof. Huston McCulloch has withdrawn his accusation of plagiarism and has asked me to announce this here, which I am happy to do. Steig says that a test of my integrity is whether I will mention this prominently here. I offered to Steig to post his complete email to me under its own thread. Haven't heard back, but hopefully that will pass the test.

McCulloch is a bit incredulous, as he explains here. McCulloch writes that Steig indicates that "none of the 6 authors learned of the error from my post" at the time that they submitted the correction to Nature, despite the fact that McCulloch sent them an email at the time. And so there is little point in further debating this issue here, as it has left the realm of the empirical. You either believe Steig et al. or you don't, to which McCulloch says "I can only take him at his word." Accordingly, I'll close comments on this thread.

Real Climate has responded to my criticism that several of its authors had plagiarized Hu McCulloch's identification of an error in a paper about temperature trends in Antarctica. The facts of the matter are not in dispute, specifically, McCulloch's analysis was correct, they were aware that he had publicized it and he was i fact the first to publicly expose the error.

Consequently, their defense rests on three claims:

1. They had already thought of the error.

2. Their error was in the application of statistics, not basic statistical principles.

3. Publication on a blog does not count, and thus presumably, is ripe for appropriation. It is first to the peer reviewed literature that matters.

Here is the Real Climate defense in their words:
In this case, McCulloch’s comment on the paper were perfectly valid, but he chose to avoid the context of normal scientific exchange — instead posting his comments on ClimateAudit.org — and then playing a game of ‘gotcha’ by claiming plagiarism when he wasn’t cited.

McCulloch accuses Steig et al. of appropriating his ‘finding’ that Steig et al. did not account for autocorrelation when calculating the significance of trends. While the published version of the paper didn’t include such a correction, it is obvious that the authors were aware of the need to do so, since in the text of the paper it is stated that this correction was made. The corrected calculations were done using well-known methods, the details of which are available in myriad statistics textbooks and journal articles. There can therefore be no claim on Dr. McCulloch’s part of any originality either for the idea of making such a correction, nor for the methods for doing so, all of which were discussed in the original paper. Had Dr. McCulloch been the first person to make Steig et al. aware of the error in the paper, or had he written directly to Nature at any time prior to the submission of the Corrigendum, it would have been appropriate to acknowledge him and the authors would have been happy to do so. Lest there be any confusion about this, we note that, as discussed in the Corrigendum, the error has no impact on the main conclusions in the paper.

The reply to this is obvious.

1. In academia it is not who thinks of an idea first that matters. It is who publicizes it first, whether in a talk, at a conference, in a draft paper, in a published paper, or yes, on a blog. An approach based only on first to the peer reviewed literature makes a pretty sad statement about the morality of modern academia.

2. This is just a silly claim that adds words but no substance to their defense. McCulloch was not claiming originality in either statistics or the need to apply an autocorrleation adjustment, simply that the authors has done so incorrectly. This is a red herring.

3. Real Climate authors presumably wouldn't steal an idea presented at a departmental seminar, and ideas presented on popular blogs should be treated no differently. I completely reject the implication from Real Climate that stealing ideas from blogs is acceptable because they are not made "context of normal scientific exchange." Sorry guys, it 2009, and blogs are part of the "context of normal scientific exchange."

Overall, a pretty poor set of excuses for not doing the right thing in the first place.

16 comments:

  1. The problem with this, Roger, is that you have no (proveable)idea when Steig et al. thought this up or where they got the idea from--all you have is a "he said, she said."

    As I said over at RealClimate, if you want scientific "credit" for something, publish it in the literature. Otherwise, you are left with the situation where are have to try to convince folks that you thought of it first and somebody else stole it from you--a very difficult thing to clearly establish.

    If I have plans on publishing something, I am VERY guarded about where I discuss the topic--knowing full well, that if I put it out in public, it is fair game.

    While scruples certainly may come in to play (i.e., an offer to co-author or acknowledge if appropriate), it is not the only factor.

    -Chip

    ReplyDelete
  2. The sad thing is that I and, I would guess, many others are not surprised by the pusillaminous CYA exercise. It is all blustering after being caught with your hand in the cookie jar.

    In business, one thing you are told never to do is p*** off the auditors. Only bad things can happen. I strongly suspect that it will increase the motivation of Hu and others to dissect statistically driven papers whose conclusions are "too convenient" and "too definitive".

    Mann et al's new paper on TCs will get seriously audited.

    ReplyDelete
  3. -1-Chip

    Thanks, but I disagree, McCulloch did publish the idea and it has a unambiguous time stamp. Steig acknowledges being alerted to it and its substance.

    Let's assume that Steig had already discovered the error independently before receiving cCulloch's email. He could/should have easily announced his discovery of the error in a comment on his own very popular blog, and had he done so this discussion would be moot. He chose not to, so McCulloch got there first based on the evidence.

    If I publish the solution to Fermat's Last Theorem here on my blog and I send it to you, and you are aware of it, and then you submit the same analysis to the Journal of Famous Math Problems without any attribution whatsoever, then you have plagiarized my work -- whether you thought of it first or not. (The history of science literature is full of examples of scholars who refused to read each other's work for exactly this reason.)

    Plagiarism does not have a different definition in science than it does in the rest of the world.

    ReplyDelete
  4. Roger:
    You may be right about the plagiarism issue, but for me there is a more fundamental issue of good manners and bad manners. I suspect that Chip would agree that the lack of any mention of Hu McCulloch at all is an example of bad manners.

    In addition, it is my understanding that it is appropriate in an article to acknowledge anyone, especially reviewers, whose formal or informal comments made a substantive contribution. Failure to do so would soon result in the loss of insightful reviewers. Again this is simply good manners.

    ReplyDelete
  5. Does this mean we can ignore RealClimate because, as a blog, it's not the proper way to challenge results? Chip must think his blog material next to worthless.

    ReplyDelete
  6. Roger,

    I agree that there appears to be enough evidence to suggest that Steig et al. should probably at least have acknowledged McCulloch.

    If I go to a conference and see that someone else is presenting something akin to what I have been working on and have plans to publish, I will most certainly cite their presentation in my write-up. If I didn't go to the conference, and have no idea what went on there, then obviously, I wouldn't make reference to them.

    In the case at hand, I don't think it involved a proof of Fermat's Last Theorem, but rather the application of a rather generic statistical procedure. One, which, according to Steig, he had already realized was appropriate. That there seemed to be an interest in the results of the application of that procedure, likely led him to publish a correction.

    Granted, had I received an email from someone pointing out something I had done was in error (something that I already knew about), I would have at least taken the time to have thanked them and told them that I was already on top of it and explained whether it mattered or not, and if so, what steps I may be taking to do something about it. But I probably wouldn’t acknowledge them in an official correction. Had Steig taken the time to have done at least that, then probably this would have been avoided.

    Based on what I have gathered, I think Eric did handle numerous inquiries into the details of this particular work (including several from me). Whether this grew tiresome, or whether some fell through the cracks, I don’t know. But apparently, he did not respond to Hu’s email.

    It seems a better case could be made for being discourteous, than for having committed plagiarism.

    Clearly there is a lot of animosity between the circles of RealClimate and ClimateAudit. This situation seems to have provided an (or rather, another) opportunity to lessen that animosity a bit, but obviously, it has been used to further heighten it. This is a shame.

    -Chip

    ReplyDelete
  7. -6-Chip

    You write:

    "It seems a better case could be made for being discourteous, than for having committed plagiarism."

    This is a fair statement. However, my view is that once the issue involves publication in a major journal, the boundary between discourtesy and plagiarism becomes much sharper.

    As I wrote in the original post, I don't think that this involved an intentional effort to plagiarize, but merely an intentional effort to be discourteous.

    Wherever one comes out on this, it is clear that simply avoiding the discourtesy would have avoided all of this.

    ReplyDelete
  8. Do they realize that they are jeopardizing their credibility.

    In such a matter, the ability to prove when one got idea is crucial.

    1) Prof Mccullogh post on climate audit is time stamped.

    2)Steig was notified of it via email.

    3)To date, there is no proof that Steig ever mentioned to Mccullogh that he was aware of the problem.

    They clearly plagiarized prof. McCullogh and they don't even deny it when they say that blog don't count (Steig student will be happy to hear this).

    Their excuse are poor and "still unconvincing" to use the title of one of their latest post.

    Maybe they have passed so much time with computer model tha they live in their own reality.

    ReplyDelete
  9. "1. In academia it is not who thinks of an idea first that matters. It is who publicizes it first, whether in a talk, at a conference, in a draft paper, in a published paper, or yes, on a blog. An approach based only on first to the peer reviewed literature makes a pretty sad statement about the morality of modern academia."

    Is this written somewhere? Otherwise it sounds like you guys are whining about trivia.

    ReplyDelete
  10. 7-Roger: Agreed.

    5-David: I hope that my blog posts are thought-provoking and reflect appropriate science. They serve as a timely and informal way to make comments on issues and reach a different audience than a much more time-consuming, formal and thorough handling of the subject as would be required for peer-reviewed publication. Are they worthless? I hope not. I rely on material presented at RealClimate (and other blogs) in formulating my opinions about many matters scientific. Peer-review serves as the meat on the blog-constructed bones. Not all cases require fleshing out, but in many instances, it would really would make for a better beast.

    -Chip

    ReplyDelete
  11. -bigcitylib

    20 seconds on Google took me to Carlton College where it explains when you need to use a citation to other's work, on its webpage on academic honesty. It says:

    "When do I need to cite?

    Whenever you borrow words or ideas, you need to acknowledge their source. The following situations almost always require citation:

    * Whenever you use quotes
    * Whenever you paraphrase
    * Whenever you use an idea that someone else has already expressed
    * Whenever you make specific reference to the work of another
    * Whenever someone else’s work has been critical in developing your own ideas."

    To emphasize:

    ***Whenever you use an idea that someone else has already expressed***

    http://apps.carleton.edu/campus/doc/honesty/intext/

    It also says:

    "Most cases of plagiarism can be avoided, however, by citing sources."

    None of this is news to academics, who take such issues pretty seriously.

    ReplyDelete
  12. Roger, McCulloch sent his email to all of the Steig coauthors, not just Steig.

    Contrary to the statement by Steig to which you link, an answerback received by Jeff Id at the time from an email to Steig did not say that he had no email - it said that he had limited email, text files only and less than 30K. Hu's message was only 6K.

    Other recipients were D Schneider, Mann, Shindell, Comiso, Rutherford.

    ReplyDelete
  13. According to Steig, McCulloch has withdrawn the charge/complaint after hearing the description of events from Steig.

    Are you still pursuing this even if the supposed aggrieved party is not?

    ReplyDelete
  14. I agree with Chip. There's no issue of plagiarism here, only courtesy. Roger settled that when he said "McCulloch was not claiming originality...". If there's no claim of originality, there's nothing to plagiarize.

    Suppose Dr S had said "Dr Hu has pointed out an omission (thanks), and I have recalculated thus...". That should make everyone happy. It corrects the discourtesy, but not any supposed plagiarism. The corrigendum would still be his own.

    ReplyDelete
  15. The "had he written to Nature" bit is interesting. He emailed them personally, and they have no responsibility to acknowledge him. He publishes on a public forum, and they have no responsibility to acknowledge him. But had he written to the Gods on High at Nature, then they would have acknowledged him. Interesting logic, that.

    ReplyDelete
  16. Why not require all science journals to have an electronic comments section on their online pubs? that was people who find errors can simply post them with their names.- so the attribution is there for all to see. Needing to run scientific dialogue through the power structure of Nature seems silly and unnecessary in this day and age- and harkens back to science as a good old boys' network rather than a search for truth. In my opinion.

    ReplyDelete