By Deep Climate
David Ritson, emeritus professor of physics at Stanford University, has updated Steve McIntyre and the rest of the world on a key controversy concerning the Wegman Report, namely Edward Wegman’s ongoing failure to release supporting material related to the analysis within the Wegman report, more than four years after his promises to do so.
Here is Ritson’s complete comment, addressed to McIntyre at ClimateAudit.
Ritson Posted Oct 23, 2010 at 5:44 PM
Steve,
Your posts on alleged plagiarism by Wegman have been more than exhaustive. Last August I acquainted you with the matter below. Your readers should have been informed by you, but weren’t, that Wegman’s own conduct is at least as reprehensible as that imputed to the “hockey-team”. It is now over four years since the appearance of the Wegman report. At the time I tried to get some confirmations relative to the numbers he had used. RC had a brief posting on the lack of any response to my requests from Wegman et al. At that time you posted at CA on August 31 2006, “Take a Ritalin Dave” chiding me for my impatience. You pointed out that comparable requests from yourself had not, in many instances, been honored for years. Wegman et al were chided by Congressman Waxman’s Office. Finally Wegman replied on Sept 1, 2006 to Waxman. The key paragraph in Wegman’s reply was“I would make the following distinction. The works of Mann et al. we discussed in the report were federally funded, peer-reviewed journal articles. Our report was review of those papers and was not federally funded. Our report called for disclosure of federally funded work. Material based on our report is being prepared for peer review journals at present. It is not clear to me that before the journal peer review process is complete that we have an academic obligation to disclose the details of our methods. Nonetheless, I assure you that as soon as we are functional again, I will create a website that fully discloses all supporting material related to our report to the extent possible. (Some of the code we used was developed by former and current students working at the Naval Surface Warfare Center in Dahlgren, Virginia and may not be disclosed without approval through the Navy’s public release process.)”
This is hardly the openness in scientific communication so eloquently called for by the Wegman report. His letter was not communicated directly to me but was forwarded on by Waxman’s office. Congressman Waxman wrote a personal follow-up letter chiding Wegman and again requesting the items of information. The second request was never responded to. It is now four years since Wegman wrote the above. Wegman never took any one of the steps he had so readily promised. namely publication, a website and public disclosure of all information with the exception of classified code. He never provided either to me or to Waxman any reason for his inaction. Waxman at that time was the ranking minority leader of the congressional oversight committee that had requested the report and he is today the chairman of that committee. As regards congressional credibility Wegman has dug his own grave. The question of plagiarism is trivial compared with the above. [Slightly edited to correct typos.]
The supporting materials related to Michael Mann’s testimony at the July 27, 2006 hearing of U.S. House Committee on Energy and Commerce include a letter from Ritson to Rep. Henry Waxman, and a transcript of email requests sent by Ritson to Wegman.
Here is Ritson’s original request, from the first email sent on July 23, 2006 (nine days after the official release of the Wegman report). Ritson raises a number of very pertinent issues.
Dear Dr. Wegman,
I read with interest you report to the Barton congressional committee. I am very familiar with the work and controversies surrounding the generation of “hockey-sticks” from trend-less red noise. Your Section 4 showed several figures, accompanied by discussion. I have read it carefully, and would appreciate some clarifications as to factual details.1). Which of the figures derive from M&M work and which were independently derived by you?
2). M&M used ARFIMA persistent red-noise throughout their published work. You state that your figure 4.4 results from AR(1) .2 red-noise? If so did you otherwise follow M&M using short-span normalization and 70 member Monte Carlo generated ensembles? Did you use the same AR(1) .2 noise to generate all your figures?
3). If you indeed used similar persistent red-noise to that used by M&M do you believe it to be in accord with real-world proxy-specific noise?
4). Any of my colleagues would have routinely checked their results to see if their derived PC1 (etc) derived from a systematic signal or from random noise. For example for a 70 member population, all that is required is to use the extracted PC1 vector from the 70 members, and apply it to each member to project out its relative sign (and amplitude). For signal dominated results one sign will predominate and for noise dominated results both signs will be roughly equally present. Needless to say when, a couple of years ago, I checked the M&M work, I did just that. The questions raised by your report are clearly of importance, and I would very much appreciate your clarifications of the above,
Sincerely
David Ritson
Deep Climate 
Why am I not surprised that McIntyre buries Ritson’s post deep in his denunciation of Bradley?
Security through obscurity is one of McIntyre’s favorite tricks to hide his decline.
Well, to be fair, it does appear that was where and when Ritson commented, and there’s no evidence that it was “buried”.
re:
“(Some of the code we used was developed by former and current students working at the Naval Surface Warfare Center in Dahlgren, Virginia and may not be disclosed without approval through the Navy’s public release process.)””
People might consult SSWR, section 4 (p.34 of V1.02), which lists interesting issues, of which #23 is:
NSWC, Rigsby, [SAI2008], Were Federal funds used? W.5.6, W.2.3.
In addition, this question clearly applies to the WR.
If John Rigsby (who had done MS @ GMU in 2005, and who I think is currently pursuing PhD) had done the WR work off-hours and/or as part of continuing part-time PhD progress, that’s fine.
If he used NSWC-Dahlgren facilities to do that work, which has no obvious relationship to the missions of Dahlgren (a place I’ve visited), the likelihood of a 4th possible mis-use of funds just went up.
See SSWR, A.7.
It’s certainly a relevant question to wonder if Rigsby was operating on his own time, using his own resources.
But be that as it may, it does appear he used code that was developed for the Navy. Isn’t that a no-no, or at least highly questionable practice? And, come to think of it, could that not in itself bring the Wegman report under Navy research oversight?
Well, pajek wasn’t developed for the Navy. Rigsby certainly used it, as did Sharabati.
It’s not obvious what code in WR section 4 was “developed for the navy.”
Returning to this topic with many unanswered questions:
It’s not clear from Wegman’s response to (former) Congressman Waxman, and as Mashey also noted, what part, if any, of Section 4 of the WR relied on secret Navy code. Was this code used in the supposed analysis of MBH’s work, or was it used for the section on social networks (I’m guessing it was the latter). Why should it be necessary to use secret Navy code to “review” the analysis of MBH’s temperature reconstructions, if the WR was supposed to be a transparent, peer-reviewable study using generally accepted statistical principles (which we know it was not)?
Anyway, it’s four and one-half years and counting, and Dr. Ritson’s request for simple, basic information about what data and methods Wegman used for his analysis is met by crickets chirping from CA and GMU. What might be the appropriate oversight mechanism/entity to look into the possible misuse of federal funds through the Navy in the production of this lie to Congress?
Finally, another “tell” that Wegman had expected his report to be held to academic standards (not that the post-hoc distinction between “report to Congress” vs. “work of scholarship” has any merit) appears in his letter response to Waxman:
If the WR were not a “work of scholarship,” why would Wegman have referred to any “academic obligations” at all? If there is no journal peer review in-process, when do Wegman’s “academic obligations” require him to disclose his methods? If Congress can’t or won’t compel Wegman to take responsibility for his deceptive work, who can and will? Is this the behavior expected of an eminent statistician who Judith Curry so vociferously defends in her noble quest for greater openness and transparency in the production and peer-review of climate research?
By the way, it seems to me that TCO deserves some credit for reviving this issue. IIRC, he contacted Ritson earlier this year to confirm that Wegman had never supplied promised material and information. That may have well led to Ritson’s private contact with McIntyre last August, as described in Ritson’s comment.
I think John Rigsby (and Jeff Solka) may have used Matlab for their earlier SNA-based work.
See for example: http://www.interfacesymposia.org/I03/I2003WebPage/sessions.html
Pingback: The Wegman report sees red (noise) | Deep Climate
That’s terrible, not releasing code and data after 4 years.
“Mr Kettle, there’s a Mr Pot on line 3 for you.”
Seriously though, Wegman should have got around to this ages ago. To set an example to Mann, Briffa, Thompson et al if nothing else.
Seriously though, Wegman should have got around to this ages ago. To set an example to Mann, Briffa, Thompson et al if nothing else.
So, tallbloke, have we earned the right to start hacking servers yet? Or do we have to wait a while longer?
It’s not hiding the code that’s bad, it’s what hiding the code hides. That’s where your equivalence breaks down, as the months ahead will demonstrate. Get some popcorn.
By the way, I’ve started looking at M&M’s original GRL archived code. That too is very revealing in the context of the Wegman report.
Stay tuned. And, yes, stock up on popcorn.
That’s where your equivalence breaks down
I agree there is no equivalence. The actions of Mann, Briffa, Thompson et al are far more heinous.
have we earned the right to start hacking servers yet?
I can’t speak for anyone elses servers, but I always welcome a bit of resilience and hardness testing against mine, so hack away.
We’ll have to agree to disagree on your take on equivalence. McIntyre’s accusations against Mann and Briffa were demonstrated false over and over again. He still refuses to admit, for example, that Mann did not “delete” anything in TAR, but simply used data *as it was supplied*. But this is veering way of topic.
As for Wegman’s code, no need to hack servers (yet another off topic tangent).
In fact, the key to what Wegman actually did, at least for fig 4-1 through 4-4, is all in McIntyre’s original code for the GRL article. No one’s looked carefully enough at that up until now.
Yup, it is terrible that Briffa forwarded McIntyre’s request to the actual data holders, who then supplied McIntyre with the data, who, in turn, kept that little fact quiet for several years, meanwhile continuously claiming Briffa did not want to give him ‘his’ data (which was not ‘his’ to give, and which McIntyre already had anyway).
it’s what hiding the code hides.
How do you know what’s hidden in the code if it’s hidden?
Anyway, Wegman has said he’ll publish the code, and going on Lonnie Thompson’s precedent, he’s got another 16 years to do it in. And counting.
Stay tuned. And, yes, stock up on popcorn.
Stock up on foodstuffs in general, this winter will be a shocker. This thread seems to be big on innuendo, and small on content. I’ll leave you to it until the next thrilling installment.
It’s not what’s in the code – actually the code is just McI’s code anyway (yes, even fig 4-4).
It’s the *act* of hiding the code that is revealing. As GP says, Wegman must know he blew fig 4-4, and generally failed to understand what M&M were doing, but simply can’t admit it.
Deep Climate | October 27, 2010 at 2:04 am | Reply
It’s not what’s in the code – actually the code is just McI’s code anyway (yes, even fig 4-4).
It’s the *act* of hiding the code that is revealing. As GP says, Wegman must know he blew fig 4-4, and generally failed to understand what M&M were doing, but simply can’t admit it.
Well, it kinda blows his whole grandstanding about getting in the experts so you don’t make stupid errors out of the water.
> How do you know what’s hidden in the code if it’s hidden?
Patience my friend, patience. And popcorn 😉
> I’ll leave you to it until the next thrilling installment.
Same here…
tallbloke’s concern about the coming winter may be justified, but perhaps he doesn’t understand that this is also probably a result of the warming in the Arctic:
http://www.enviralment.ca/2010/10/26/warming-arctic-could-mean-colder-winters-for-north-america/
[DC: Sorry, over the line and off topic. ]
Would love to see a bit more on Ritson’s point 4.
A. How the experiment plays out (perhaps graphically).
B. Little more detail on what he does (what does apply mean?) and if he could explain it without (or in addition) to vector-matrices, would appreciate it. (I think Huybers was great in clarifying standard deviation dividing.)
C. Any discussion of this signal versus noise conundrum from other fields. Other previous studies.