|
October 31, 2004 McIntyre to Cotter Dear Dr Cotter, In your original email, you stated that "you will see from our policy
details on www.nature.com/nature/submit/policies
that publicising
confidential material is not permitted without **prior permission** from the
reviewers themselves" and that you would not seek such permission without a
prior request from us. Accordingly we hereby make such a request. Hopefully,
the matter discussed below will accordingly become moot. The responders (defined as the authors of the published contribution that is the subject of the comment, and no-one else) must keep the comment confidential and must not use it for their own research or for any other purpose apart from replying to the comment, nor can they distribute it without first obtaining Nature's permission. http://www.nature.com/nature/submit/gta/index.html It appears that Nature did not take positive steps to ensure compliance with this clause by Dr. Mann, who failed to comply with Nature's confidentiality terms. When breaches of this clause were brought to Nature's attention, Nature took no steps to require compliance. In particular, we refer to the following: 1. Dr. Mann posted a statement at his university website that his undisclosed data transformation prior to PC calculations had no effect on his results, citing the Reply to our Nature submission (then still under review) as a reference, in violation of the above confidentiality policy and in an apparent attempt to pre-empt our submission. We notified Karl Ziemelis of this (referring to the above policy), but he did not respond to us. 2. Nature then posted the statement, to which we had objected, at its own website as part of the Supplementary Information to the Corrigendum by Mann et al. Again, we sent an objection to you. Instead of dealing with the matter, you responded irrelevantly that the statement did not breach Nature's embargo. However, our objection, as was clear from the policy cited in the original email to Karl Ziemelis, was the above confidentiality clause - which is different from the pre-publication embargo. Accordingly, your reply was unresponsive and failed to deal with the substantive confidentiality breach. 3. Additionally, Dr. Mann referred to the pending Reply to our Nature submission in support of another article by Mann et al. on the same topic, then under review at Climatic Change, and even delivered the Reply to the editor of Climatic Change in breach of the above confidentiality clause. This submission to Climatic Change was referred to an article published by Jones and Mann in Reviews of Gephysics. Subsequently, the editor of Climatic Change posted up the Reply on the editor's website <http://stephenschneider.stanford.edu/Publications/PDF_Papers/MannEtAl2004.pdf> while our article was still under review at Nature. There was a breach of confidentiality both in terms of its use as support for the Climatic Change submission and in terms of its distributing the material on the Internet. At this stage, merely withdrawing the article from the Climatic Change website would accomplish little. Secondly, while our submission was under review, Nature published a Corrigendum, which stated that the errors in MBH98 did not "affect their results". This published statement, apparently added during the galley proof stage, was highly misleading and extremely damaging both to us and to our submission. Its publication clearly communicated the presumption of an editorial decision against our paper, at a stage when there were referee reports in our favour and a revised version of our submission was under embargo, thus leaving us unable to respond publicly. It also breached a specific commitment that had been made to us that the Corrigendum would not address matters in controversy. The only way that we could partially mitigate the damages was to reluctantly publicize the process and the anonymous referee reports. In particular, we cite the following: 1. When we learned that Nature was planning a Corrigendum, we expressed our concern that Dr. Mann would use the occasion to make comments about the controversy and asked for the right to reply. Dr. Langenberg promised us that the Corrigendum would contain no mention of the controversy and asked us to preserve confidentiality pending the Corrigendum, which we did. Instead of honouring this commitment, the sentence dismissing the controversy was inserted between the time of the draft Corrigendum in March and the final Corrigendum in July. This sentence is not merely incidental, as it has been quoted as representing Nature's decision on the matter and relied upon in coverage of this matter. 2. Around the time that Nature decided to require a Corrigendum, I was informed that the new SI would not be edited by Nature. If it was not edited, it is a reasonable surmise that the new SI was also not externally peer reviewed, in breach of Nature's stated policy on peer review of SI. Had it been either edited or peer reviewed, the erroneous claim that the PC calculation methodology had no effect on their results might have been caught. 3. The claim that the PC calculation methodology had no effect on their results is manifestly incorrect even on Dr Mann's submissions to you. For example, if you look at MannSuppInfo1.pdf, attached to his Revised Reply, the figure shows a dramatic effect on eigenvalues in the North American tree ring calculations, which has the effect of substantially exaggerating the importance of the bristlecone tree ring signal, exactly consistent with our claims. MannSuppInfo1.pdf also acknowledges that their PC methodology promoted the effect of the bristlecone pines from the PC4 to the PC1. Likewise the Gaspe extrapolation (which was misrepresented in MBH98) did have an effect on results. Then the issue became: what was the impact of this and other erroneously described methods and data on the final Northern Hemisphere temperature calculations. We believe that the effect is considerable; Mann has argued only that they can salvage MBH98-like results with a substantial modification to their model. These issues were very much under consideration at the time that Nature published the pre-emptive claim in the Corrigendum that the errors did not "affect the results". This false and pre-emptive claim adversely affected our ability to get a fair hearing for our submisson, since Nature would look contradictory publishing the claim in the Corrigendum that the errors did not matter in July and then laying open the controversy in August. 4. When we objected to you about the sentence in the Corrigendum, you suggested that we could respond in the pending Communications Arising, a recourse that obviously was not subsequently made available. 5. The referee reports certainly indicate that our reviewers, who were by then quite educated in the matter, were not involved in refereeing the Corrigendum. If they were not involved, we surmise that the Corrigendum itself was not externally peer reviewed. While this may be possible within Nature's policies, it is inconsistent with an undertaking by Karl Ziemelis last November, in which he promised to obtain source materials from Mann et al. and have them externally reviewed, a promise upon which we relied in deciding to submit to Nature in the first place. The source materials in question include the outputs from statistical calculations which were used in the decision not to publish our paper. Recently in September, Dr Ziemelis refused even to request the source materials from Dr. Mann, which he had undertaken to request last November. All in all, it is our view that there has been a complete breakdown within
Nature in its handling of this matter, which has been very damaging to us.
We have sought to partially mitigate by publishing the referee reports,
which remain anonymous. |