During the 12-hour interval that Steig deigned to permit comments on his “Tutorial” about principal components, he made the following backhanded criticism of blogger communications with him:
[Response: Ryan: Unlike most of your fellow bloggers, you have been very gracious in your communications with me. They could learn something from you.
I don’t know how many bloggers Steig was in communications with, but I was one of them and I regard my communications with Steig as being entirely proper. Given that Steig has tarred a number of people with the above accusation, it would be nice if he acknowledged that my communications, like Ryan’s, were courteous. Below I reproduce my correspondence with Steig to demonstrate this:
On Jan 23, 2009, I emailed Steig as follows:
Dear Dr Steig,
In your article you refer to the development of “50-year-long, spatially complete estimate of monthly Antarctic temperature anomalies.” Could I please have a digital copy of this data. Also I presume that this output was derived from corresponding monthly gridded versions of T_IR and T_B and I would appreciate a copy of these (or equivalent) input data as well. Regards. Steve McIntyre
Steig promptly replied the same day as follows:
Steve
I have always intended to provide all the material on line; I wasn’t allowed to do this before the paper was published. I would have done it already but have been busy answering emails. I should have these up on line next week.
Eric
To which I responded:
Thanks.
“all the material” – It would also be an excellent idea to put your source code up. Using statistical techniques that are not well understood to derive newsy applied results is a bit risky and you should err on the side of caution in making your code available to independent analysis.It would also serve to defuse people who are instantly suspicious of Mann’s RegEM. This is an application where it seems much more plausible to me than where it’s used to justify bristlecones. I’ve noted this at my blog as a caveat to instantly suspicious readers.
Regards, Steve McIntyre
A couple of days later, Steig broke off communications with the following comment at CA:
If you happen to find any legitimate errors in our work, when you try to duplicate it, I will be delighted to discuss it, provided it is through the venue of the peer-reviewed literature. I will not further respond to queries from Steve McIntyre by email, nor via this blog. I have always given you the benefit of the doubt, but your thinly-veiled accusations of scientific fraud (by association!) have crossed an important ethical line. Shame on you.
Over at RC on Feb 3, he posted the following “gracious” comment”, along with other comments evidencing similar “graciousness”:
To continue with the analogy with financial auditing, let me very clear on what I mean by legitimate: In the business world, auditors 1) don’t publicly accuse a company of withholding data prior to requesting said data; 2) are not self-appointed; 3) have to demonstrate integrity and competence; 4) are regulated. On this point, if you are suggesting that Steve McIntyre be regulated by an oversight committee, and have his auditor’s license revoked when he breaks ethical rules, then we may have something we can agree on.–eric]
I don’t know how many “bloggers” were in email correspondence with Steig, but I certainly do not think that there is anything inappropriate in the language of my email correspondence with Steig.
It’s interesting to read Steig’s peroration to the commentary on his paper on Jan 27 when the ink was barely dry on the paper, data was still unavailable (the AVHRR data only became available in late March and the Comiso data is still unavailable):
All in all, the critical commentary about this paper has been remarkably weak….The poor level of their response is not surprising, but it does exemplify the tactics of the whole ‘bury ones head in the sand” movement – they’d much rather make noise than actually work out what is happening. It would be nice if this demonstration of intellectual bankruptcy got some media attention itself.
http://www.realclimate.org/index.php/archives/2009/01/warm-reception-to-antarctic-warming-story/
I’m not sure how Steig expected people to be able to “work out” his methodology within a few days of publication without data then being available, without any available of source code or a comprehensive and detailed methodological description.
Climate Audit 
150 Comments
Steig:
This comment always baffled me. These people have a ridiculously thin skin when it comes to criticism, and what they think you imply in asking for the source code to remove doubt about the actual methodology used.
I wrote before that asking scientists to provide source code and data was akin to an academic Rorshach Test revealing how confident they really were about their results.
I couldn’t imagine Einstein or Feynman reacting like this? Could you?
Steve,
Just to be clear, is this the all the communications you have had with Dr Steig?
Steve: I have had no other email communications with Steig this year. I haven’t checked prior years, but I definitely don’t recall any email correspondence in prior years that could remotely give rise to accusations of lack of graciousness.
I sometimes think there are two Dr. Steigs: one in his normal, helpful mode, and one just after he’s had a chat with his co-authors, or something.
Steve, I am a longtime reader of your blog – though I have never commented before. I have a degree in English from Cambridge (though my career in finance involved a lot of time series modeling using a canonical variate analysis based method) where we spent much time analyzing tone. I must honestly say that I have long thought that your communications with other professionals were unnecessarily snarky. For instance in the email you quote above you say:
“all the material” – It would also be an excellent idea to put your source code up. Using statistical techniques that are not well understood to derive newsy applied results is a bit risky and you should err on the side of caution in making your code available to independent analysis.”
I am frankly not surprised Dr Steig was irritated. Talking about “newsy applied results” implies that he has derived these results with the intention of making news – rather than of making scientific progress. This may well be true but it is definitely not “gracious” to say so as you cannot possibly have hard evidence as to his motives.
A more gracious way to phrase the email might have been:
Dear Dr Steig, May I suggest that when you post your data you post your code also. As I’m sure you are aware there has been some controversy regarding the use by Mann’s of RegEM. Personally I feel that its use in this context is plausible but if the code was posted it will make it much easier to overcome any residual skepticism that some might have. Kind regards, Steve McIntyre
Re: Mark (#4),
Mark
The only problem is that your supplied language contains “hot button” words which could easily offend some climate scientists.
I’m not talking about “controversy” or “plausible”. I’m talking about words like “Kind”,”regards”,”Steve” and especially “McIntyre”. Those words are so loaded with meaning that some climate scientists refuse to say them even once lest their “peer-reviewed work” gets microscopically analyzed by the lead investigator of the long running reality series “CSI(Climate Science Investigations):Toronto”.
Re: Mark (#4),
I agree with this comment. I think the term “newsy” suggests ulterior motives, regardless of whether they are present or not. Also, Steve, just don’t say “Thanks,” in a perfunctory manner. You should say, instead, “Thank you, Eric.”
There are ways to use the language that calm those you are corresponding with and make them want to cooperate with you. I’ve noticed that many of your emails to those you are auditing are curt, less than gracious, and while not threatening are, given your present occupation and resulting notoriety, not re-assuring. Your emails sometimes remind me the type of inquisitorial questioning used regularly on 60 minutes when they had the goods on someone.
Or to put it another way using television analogies, why beat em up like Jack Bauer when you can be like Columbo and gradually get the information you need?
Re: theduke (#11),
I must admit that I found myself in complete agreement with Mark (and “theduke”) as I read their comments. Steig’s (and The Team’s) sensitivity and hostile treatment of skeptics is not justification (whether true in the present situation or not) to send subtly offensive emails, as some commentaters on this blog imply. Ironically enough, by suggesting such, they deny the very premise of your blog.
I have thought in the past (and in the above email samples) that there is a certain disdain, however subtle, displayed for Steig and other “opponents.” While there is justification for some irritability and even anger at times, it still defeats the purpose of the correspondence–to get the data.
If the purpose of the emails is to return “snarkiness” for snarkiness in order to make clear their (imho) intellectually dishonest approach, then by all means, include the (justified) prods. But if your purpose is to actually get the data, following Mark and theduke’s suggestions seem entirely appropriate. Curtness, without regard to any other characteristic of an email, can, in and of itself, be interpreted as antipathy. It’s just better to avoid it completely.
BTW, I’ve been as irritated and angry at the way The Hockey Team members show their contempt for anyone who would dare suggest alternative theses to their dogma. But I also feel that you can make them dig their graves even deeper by being over-courteous, if necessary, in response to their tantrums. The contrast can only make more stark their emotional attachment to global warming theism–the antithesis of science.
Re: theduke (#11),
I agree to an extent. Steve has baggage, for good or ill. And I’m sure Mann and Schmidt have fully briefed Dr. Steig.
I’m not familiar with normal scientific discourse between professionals, but Steve’s emails to Dr. Steig seemed to assume something not in evidence, and stated so. A little editorializing, so to speak.
The same goes with Steve’s posts. He tends to slip in a fair amount of snark, and that can’t help his situation. You get what you pay for.
Re: Mark (#4),
And people wonder why we have a financial crisis …
Re: Mark (#4),
I have to disagree. While it’s certainly true that any form of text only based communications is easily misinterpreted, I think it’s fair to say that it is also encumbent upon the reciever to take this into account – if there are multiple interpretations of the wording, the reciever should not automatically assume the worst of the sender. If you are unsure exactly what is meant, then ask for clarification, pointing out your interpretation, and perhaps adding something like “… I trust that this is not what you meant to convey.” After all, what’s sauce for the goose…
I too perceived a little residual snark in your email to Steig. Perhaps it was the association with your arch nemesis? Still, his response was a bit over the top.
Your comment:
“I don’t know how many “bloggers” were in email correspondence with Steig, but I certainly do not think that there is anything inappropriate in the language of my email correspondence with Steig.”
Your error was mentioning the holy Mann.
Emails (and blog commnents, no doubt) are notoriously subject to misinterpretation of tone. The sender can have been thinking one thing and the recipient suspects underlying motives that are not there. I am inclined to agree with Mark that your emails could be seen as somewhat brusque, even if you didn’t mean them that way. On the other hand, Steig needn’t have reacted the way he did.
As far as I am aware, it is not Mann’s RegEm anyway is it? Didn’t they use someone elses Matlab RegEm script for the analyses?
Steve shouldn’t have to genuflect to get the data. And there is a LOT of history here with “the community” and their disdain for Steve’s analysis.
I guess Eric needs to remember that until something is verified, it’s not fact. The more difficult he makes verification, then it just opens the door to all sorts of speculation as to motives, however unfair.
Newsy?… I took that to be a typo and read it as ‘newly’… Therefor, “Using statistical techniques that are not well understood to derive newly applied results is a bit risky and…”
Did diddums chuck a wobbly over a typo???
I too agree with Mark#4 and theduke#11; your comments come over to me as hostile and deliberately offensive. I’m well aware of the background here, but whether or not you consider that an excuse, this sort of approach simply isn’t helpful.
I’ve been in business for many years, written and received lots of emails, I don’t think anything in your emails to Steig would be perceived as hostile or worthy of breaking off a correspondence in a huff in business. It is clear that you have reservations, if you like, you are on different sides in a discussion of a business case.
But if this were a proposal of the sort I have often participated in and led discussions on, to go into a new venture, bid for a license, buy a company, your tone would be the normal cut and thrust of exchanges between people with different views of the matter.
What’s more to the point, the idea of enthusiastically proposing one side of such a venture based on studies you have done, whose methodology and algorithms you refuse to reveal, well, do this once and your reputation in a company will be shot. Do it a few times, and your career there will be over.
Re: anon (#14),
Having lived in both the academic and business worlds I agree with anon with respect to business communication. In business the great leveler is money, and robust exchange generally leads to the best deal for everyone, so no-one takes anything particularly personally. (A long winded way of saying “it’s just business”).
In academia the currency is reputation, not money, so comments that would be brushed off in a business context can cause greater offense.
Re: Jonathan Baxter (#43), “In academia the currency is reputation, not money…”
Among literary theorists, maybe, but not among scientists. The currency of science is ideas. Those who have the best ones rise to prominence. That some scientists are arrogant grand-standers is beside the point.
Someone who has a good idea, one that explains a lot of data and especially that predicts unexpected results, is eager to have the idea widely examined. The wider the play, the more successful the idea, the greater number of other researchers using the idea, the richer the scientist in the currency of science.
Those who are reticent in communication, obscurantist regarding method, suspicious of examination, hostile to replication, and uncooperative with requests for clarity are behaving in a manner entirely contrary to scientists who have discovered something new or important.
The latter want to clarion their results and methods in order that they have the widest possible currency. The satisfactions of having one’s ideas widely used is the be-all and end-all of science.
Obscurantism in science is a glaring aberration, and consistent only with something to hide. It’s my view that, for such people, the particular result is all and it’s by hook or by crook that they get it. And once gotten, challenge will not be brooked. Hence the hostility and obscurantism. It’s not science these people are doing, regardless of the math and/or physics salted into their methods.
Re: theduke, Jonathan:
Tone?
In regard to open and honest communication in science I can’t go past the late Richard Feynman’s description of Cargo Cult Science. That is the sort of science that is secretive, one that is missing “a kind of scientific integrity, a principle of scientific thought that corresponds to a kind of utter honesty”. As Feynman said, “The first principle (of science) is that you must not fool yourself—and you are the easiest person to fool.”
Association with whom? Mann? And is an accusation of association with Mann already considered an accusation of fraud? Scientific fraud, no less!
And what does Steig consider “Scientific” fraud? E.g. withholding data and code? And why does he go on do it then?
I am getting more and more agitated as I write this, I should stop before I start calling names and get snipped…
“some controversy regarding the use by Mann’s of RegEM”
———————
Re: Mark (#4), Dr Steig would be pissed off that you dare to suggest “some controversy” about his co-author Dr Mann. All in all, you’ve rashed his skin so please, don’t communicate with him anymore (copyright Santer).
People who don’t frequent blogs very much can have difficulty drawing distinctions between what the blogger says and what his commenters say, or even what supporters of the blogger say on other blogs. I recall that Dr Steig showed great annoyance that someone on this blog (not Steve) had mispelled his name, so we are dealing with a certain level of sensitivity here.
Dr Steig is not impressed with the amount of critique his work has attracted. Since it now has more holes than a collander maybe he is having second thoughts or a conscience attack?? Who knows.
Pulling the “high moral ground” is a stunt.
regards
Oh my gawd, some precious snags have surfaced. Steve, there was nothing wrong with what you wrote. The blog, by common use, has a bit of rough and tumble.
This blog is not about “tone” or old school tie mannerisms. it’s about getting to the best understanding of complex matters od maths and science. What do you charmers want? “e=mc^2, with sincere thanks”?
I suspect Steig is running scared and pulling this stunt is a convenient way of not having to face the criticisms of his work, it goes along with the “peer reviewed” silliness as well.
Now if I were him, I’d be on this site and JeffID’s everyday defending my work.
Your emails are brief but there is nothing in them to suggest rudeness or insult.
Oh and it may be worth comparing Ryan O’s responses to criticisms and comments and Steig’s.
Sorry, Steve, but I have to go with Mark & duke also. I’m a scientist (luckily, not a climate scientist) and I’d feel a bit put out by a blunt request like yours for data. The thing is, even if you publish a paper that is a stinking pile of horse…, you’ve still put a lot of work into it, and requests for data ought to acknowledge that, even if only implicitly.
In addition, you’ve got a serious reputation (in both a good way and a bad way), and if you ever asked _me_ for my data, I’d be quaking in my boots. If you’re going to eviscerate someone, you should at least try to smile.
I am one of those “fellow bloggers” who communicated with Dr. Steig. On March 19th, I sent Dr.Steig the following e-mail:
In response, I received an automated “out-of-office” reply. The next day, I wrote to Dr.Comiso:
Dr. Comiso replied promptly and politely (with a copy to Eric Steig) that he preferred that Dr. Steig provide the information and that i would get “an accurate and complete answer to your inquiries about the paper”. Shortly thereafter some new data did appear on the website. It was not the observed satellite data which could possibly be used to determine how many PCs it takes to replace a thermometer. Moreover, to this day, I have not received a personal response to my presumably “ungracious” request.
Re: RomanM (#26),
This is the type of email I would have written. I might even have added congratulations for having an important paper published in a prestigious magazine.
That said, I prefer the traditional greeting that Steve used. The generic salutation, “Dear _____,” is gracious for a reason. If you are contacting someone for the first time and seeking coooperation, it’s the beginning of a process that could prove enlightening and beneficial for both parties. The graciousness of the salutation needs to be extended into the letter at least until the hard questions and requests for information begin.
Was such a relation between Dr. Steig and SM possible in this case? Not likely. Dr. S was probably looking for something to allow him to throw off his calm veneer and go on the attack in defense of his and his friends’ professional integrity. But the more gracious the initial requests are, the more irrational his intemperate response will look.
William’s point in #25 is well-taken. If you have what could prove to be a “stinking pile of horse… ” for a paper, when that first email from Steve McIntyre comes in, it must put you into a cold sweat. In the future, a little friendly chitchat in the initial stages of communication with those who are writing papers of this sort might help the science to advance more quickly and less contentiously.
Re: theduke (#60),
My salutation “Hello!” is a personal affectation that I have developed in writing e-mails because “Dear xxx” simply sounds odd to me as the start of a dialogue and I rarly use it. It is not intended to have any other emotional content other than as an introduction and it has never occurred to me that it might be considered “ungracious”. It is only after some previous (amicable) contact that I might consider saying “Hi xxx” or some such. Perhaps I should reconsider its use in the future.
Re: John S. (#64),
IMHO, the rise and proliferation of “novel” techniques in scientific disciplines tends to spring from an insular mentality where procedures may originate in the long ago from genuine statistical rersearch but are then altered and enhanced (without theoretical justification) as they are passed along within the discipline. The importance of the basics such as you mention are either forgotten or ignored. “Proofs” of the efficacy of these changes in the methods are often a single application to a contrived data set.
The same researchers using these convoluted methods are then surprised when someone with some statistical understanding questions the results in a way that a good consultant should have done before the paper was published. The wagons are circled and a protectionist mentality takes over. Reasonable dialogue becomes difficult and we end up with threads like this one.
Re: RomanM (#70),
Agreed. For scientific “peerdom” to retain any validity, firm planks are required to direct misguided “advances” into the deep. Alas, for many novel claims about climate, nothing less than a pier would suffice.
Re: RomanM (#26),
Your sly reference to “how many PCs it takes to replace a thermometer” is utterly priceless! Especially in light of the granularity with which Steig produced his trend map.
It’s astonishing how many in “climate science” rely upon a technique designed to compact multiple samples of reliable data over some well-defined natural domain to try to cure various deficiencies–or total lack of data–in a single sample over an abitrarily chosen domain. Most have never heard of Shannon’s Sampling Theorem, which applies in the spatial domain as well as in the temporal. And the concept of consistent estimators, i.e., those that converge upon the ensemble or population charateristics, rather than upon the sample properties, likewise seems to be terra incognita. RegEM, with its recursive imputations, provides an all-too-convenient deus ex machina by which to claim information that is simply not available about the real world. I’m glad to see the curtain gradually being pulled on this academic conceit.
Oh come on, people.
Steig is a co-author at RealClimate and has known and corresponded before with Steve. There’s no way that Steig could possibly have fooled himself into believing that Steve was not going to ask him for both data and source code in order to replicate the results – unless Steig is an idiot, a possibility I refuse to entertain.
Steig must have known that the request would come. I suspect that Mann flew into a rage at the thought of “He who must not be named” investigating yet another one of Mann’s analyses and forbade Steig from providing anything. The effusion that Steig produced was a smokescreen and nothing else.
We’ve seen before how requesting data and source code has seen thin-skinned skittish behaviour from climate scientists that would make a prima donna ballerina blush with embarrassment.
Witness Phil Jones to Warwick Hughes’ entirely reasonable and amiable request for data which carried on for months with Jones saying that when he gets back to the office, he’ll provide and then suddenly out of the blue producing the fabulous line “Why should I provide my data to you, when all you’re going to do is find something wrong with it?”
If Steig or Mann thought that the analysis was statistically and scientifically robust, they would have published the lot and demanded that Steve do his worst.
Re: John A (#27),
Si to John A. It appears that Steig is looking for excuses not to reply. Like the Team.
We now know that the choice of 3 PCs gave the optimal warming result. The Team was quite aware that should their methodology be made public, they would be subjected to review and criticism by the climate sleuths. Hence, stonewall all requests for the evidence — Steig’s response was just a diversionary tactic. Buy more time, and maybe the issue will fade away.
Steve,
I agree with the point you were making to Steig. However, I also agree with several of the posters above that you could have said the same thing with more neutral words. I know that this is difficult considering the way that the team has responded to your very reasonable requests in the past.
John A,
You have a good point. If Steig is offended by being compared to Mann, then perhaps “the team” does not hold him in as high a regard as we might have thought.
Re: Brooks Hurd (#29),
> perhaps “the team” does not hold him [Mann] in as high a regard as we might have thought.
We are told that the way to know if a paper is good quality and well received in the scientific community is whether it is cited by later papers. How often has Mann08 been cited (or is it too early to tell)?
Re: AndyL (#31), you ask How often has Mann08 been cited (or is it too early to tell)?
His paper Proxy-based reconstructions of hemispheric and global surface temperature variations over the past two millennia has been cited 5 times so far: once in McIntyre & McKitrick’s comment, once by Mann in the reply to the comment, and three times by independent groups. This seems quite high to me, but it is really too early to tell.
I have to go with Mark, JohnM and the duke on this one. Steve hasn’t helped himself over the years by the tone he uses. He’d be far more effective if he was more polite and professional. And published.
RomanM, if I received a request like yours, I’d not bother to reply either.
Re: DaveR (#30),
So what is it that you find objectionable?
I did in fact identify my professional academic stature (which is equivalent to Dr. Steig’s) and university connections in the signature of the e-mails.
Hate to pile on, but I (sort of) side with the thoughts of #30. I’ve felt many times over the years that the personal emails have provided ‘just enough’ room for the recipient(s) to take offense…and as pointed out these folks are predisposed to take offense if at all possible. I’m a huge supporter (and have contributed funds multiple times in the past) of this blog and Steve M’s work, but sometimes I wish he would almost bend over backwards so as to *not* ruffle feathers of those he audits. I just think in the long run he would get more cooperation and we would get more insight into the actual arguments being made.
Thanks for this post Steve, Dr. Steig was reasonably cordial yesterday and it was different from my own experience. As about everyone probably knows my one post on WUWT comment thread got pretty thoroughly repeated by several people at RC. For my own experience with the good Doc, did not start out anywhere within reason. First several polite absolutely non-critical requests for data at RC were snipped, the data I was politely requesting behind the scenes was said to be fully available to other questioners without mention of my requests, then when a comment got through I was told to take matlab classes and the code was available. It went on for a bit with some questions on Dr. Steig’s email which I haven’t discussed publicly, followed by more snipping at RC.
In the process it got so bad, I even explained to gavin at one point, that if he cooperated with me and I found warming was correct for the Antarctic, the whole feud type atmosphere would settle down. – snip with no reply. PLENTY of genuflecting. Finally over a month later, after the Dr. came back from the Antarctic, he put the link up for the data we’d been asking for since the beginning. The link was password protected comments were sent to RC requesting the link to be released — SNIP. I sent another email about the password and got my FIRST cordial professioinal reply which I then pointed out was cordial and professional not mentioning any of the items above in my post and the password protection was removed.
I don’t believe that RC didn’t notice the hours of people trying to get the data and making comments in between about the link not working. I fully believe they sat back and laughed.
You know I’ve worked on this thing for months, it has been interesting in the same way a good crossword can be interesting. At the same time I was ‘very’ frustrated with the response at RC and by Steig, I’m new to climate science. To say the least it has been quite un-similar to my experiences with physicists, mathematicians, geologists and engineers.
Dr. Steig’s tone was so much different in the thread yesterday it was like Jeckyl and Hyde. If Dr. Steig continues to cooperate in this fashion we could get so much more done. The whole discussion would change color ‘maybe’ even my Irish temper wouldn’t come out like it did after re-explaining to a WUWT poster for the fourth time why our results are different.
Anyway, thanks again Steve (and others)for giving me a starting point. I’ve learned a lot from it and will continue. There are plenty of other Steig et al details to work out – like what the heck is a simple PCA reconstruction? I’d ask but the thread is closed.
Oh yeah I should mention also that from my own understanding of the paper, Ryan’s reply was absolutely devastating to the entire post. After sitting in moderation from a few minutes after the post went up for 12 hours it disappeared from moderation, AS MINE had done. Soon mine reappeared in moderation and then was posted. Ryan’s disappeared for an extended time and was initially posted at the end of the thread with formatting removed. Two comments added and the thread was closed.
First, I’ve had a long history of asking for data. Personally I don’t believe that the tone of the request has ever been material to a refusal decision.
In passing, it seems that there is a bit of cultural difference between academics and the ROW in correspondence style. For example, one reader suggests:
Look, I’m a bit terse in email correspondence. If I want to write ironic emails, I can style them up as though Steig were my pal. “Thank you, Eric”. But I don’t know him.
Another reader suggests:
I see no reason to say one way or the other whether “its use in this context is plausible”.
Another reader writes:
Yep.
“I certainly do not think that there is anything inappropriate in the language of my email correspondence with Steig.” Oh yes there is – there’s a missing question mark after the first “data”. Hmph.
OR, to be serious, your e-mail seems to me to have the brusqueness that might pass muster in Scotland but would be felt to be excessively direct in England. It’s my impression that Yanks expect even more sucking up to than Englishmen. The problem is with your roots, perhaps?
One can catch more flies with honey than with vinegar…
After reading Stieg’s reply to Steve, my first reply was “Huh, what did I miss?” and went back to read Steve’s last message. I took your mentioning of Mann more as a friendly bit of advice that if you don’t want to painted the same as Mann, then be open with the data and code, and protect your reputation. This, to me anyway, was more a bit of friendly advice rather than any indirect claim of fraud. But then again, Stieg is a college professor who is used to having students venerate him, and apparently he expects other professionals to give him the same treatment.
I have noticed that, generally speaking, the people who respond to these blogs are generally professionals who either work in the real world or deal with the real world on a regular basis, and are not sequested in an academic environment. This tends to make the communication, especially among scientists and engineers, a bit more direct. To me, Steve’s comments weren’t sugary sweet, but they certainly were not insulting.
I would say that your comments were within the bell curve for tone. Certainly, the request was reasonable. But the subtext might have inadvertently conveyed a sense of superiority.
The problem is that Stieg and his colleagues come with a certain defensiveness towards you because you are critiquing their life’s work. That’s not your fault; it just is what it is.
And so the phrases “it would be an excellent idea if …” and “is a bit risky…” and “you should …” convey, to the mind of one already defensive, a sense that you are his advisor, qualified to tell him how to manage his affairs.
This interpretation appears to account for the abrupt change in tone from his first response to his second.
The problem is that your advice is actually good advice, so a reasonable person might just swallow hard and do it anyway. But defensive people are not reasonable people, and there it is.
One of the main goals of science is reproducibility by someone other than yourself. The fact that methods and data are hard to get, or in some cases, impossible leads me to believe that some work in the climate field is too weak to stand a serious level of scrutiny. I don’t consider 2 or 3 reviewers a serious level of scrutiny, especially if they don’t have access to all the methods and data used in the paper.
Science is becoming more accessible to non-scientists everyday and it shouldn’t matter who is asking for the data. The data should be available as soon as the paper is published.
Re: Karl B. (#42),
I wouldn’t consider 2 or 3 reviewers a serious level of scrutiny either. Consider Dr Steig’s own response to Jeff Id at RC
I think it is interesting that the reviewer was concerned about this. To me, this is an indication that he/she did not fully understand all of what was going on in the paper. And clearly, there were no detailed algorithm descriptions or (gasp!) code available at that time.
I might be wrong of course.
The RC post began with a discussion of the dangers of extrapolation. However, the data presented in the paper was temporally interpolated, not extrapolated. Comments 27 & 35 pointed this out. In my opinion, the responses didn’t address the comments.
The choice of 3 PCs was based on a “rule of thumb” from the North paper. The RC post discusses selecting a truncation point based on “overlapping error bars”. Here’s what I don’t understand… the weather station and satellite error bars in the RC figure all have equal width. The errors bars in Fig. 4 of North decrease as the true eigenvalue magnitude drops. Does anyone understand why the error bars look different in the North paper and in the RC figure? This seems important since the PC truncation point was selected based on these widths.
(Please ignore my 44)
People,
what we are talking about here? About scientist who didn’t provide source code, data and method for independent parties to replicate his results (what is most basic requirement for anything to qualify as “scientific result”). And we are arguing back and forth if the request from one blogger was “gracious” enough or not, to “justify” refusal of scientist not only to disclose the data and method, but to break communication with him? Don’t we somehow assume that Steig has moral right to refuse disclosing the data without dire consequences for his scientific credibility? Is it really possible that nuances in Steve’s tone are more important that Steig’s refusal to provides the details of his analysis? I am afraid with such an approach we would get nowhere. L- snip
Look at Steig’s cynicism, he requests regulation of climate auditing! And he gives to Steve “benefit of the doubt”! (Dear God, imagine Roy Spencer refusing to provide the method for replication of UAH data, claiming that although he gave “benefit of the doubt” to third party, third party disappointed him eventually and now he is not going to give the data now – what would happen to his career in minutes, with support of all of you who now dissect nuances in Steve’s tone, and try to justify Steig’s – snip- behavior)! But, Mr Steig is the one who spends public money on his research and not Mr McIntyre.- snip
Some of the comments in this thread make me want to laugh and sigh at the same time.
There is no doubt in my mind that, if a scientist is really interested in seeking the truth and in doing so realizes that legitimate criticism and a discussion of that criticism can be helpful and that requests for more information for analyzing their works is part of that truth seeking process, the scientist (and particularly one whose has an academic position that must make him sensitive to diverse personalities and cultures – or at least should) would keep the scientific task at hand in the forefront and forego getting wrapped in his own underwear over what might or might not motivate a responder.
To defend the behavior of a scientist by parsing the details of how that scientist is addressed seems almost to put that scientist in the position of royalty with handed-down protocols on how to interact with his highness. I would think that a scientist, who might have baser instincts and react by telling off those seeking information or making a critique his works or by the scientist imputing motivations of those critiquing him, in the end would do what a true scientist would do in seeking the truth. We would all then merely point to his personality and disposition in passing and get on with the business of analyzing science papers.
What really sticks in my craw over many of these “conflicts” is that the scientist under review sometimes appears to be looking for a reason not to cooperate and, of course, can usually find one. Along come the defenders of that attitude of scientist-with-a-chip-on-his-shoulder and we have all manner of reasons for not cooperating in the search for the truth.
I found early in my career as a fix it guy around the house to go to those grumpy old men in the hardware store who, when I would ask them if such and such a procedure would fix my problem, would look at me in disbelief and tell me I did not know what I was doing – and then tell me, with enthusiasm, how to fix the problem. I came to seek those types out and rather discount some of the advice I would get from those with a sweet and polite disposition and seemingly a bit too much in a hurry to agree with me.
Well, we’re talking SCIENCE, aren’t we? Any alleged tone of one’s request is (supposed to be) immaterial. Scientists (supposedly) do research to discover the truth of the natural world, so they don’t “own” the data and code. People who act as though they have something to hide probably do have something to hide. Steig’s behavior, as quoted in this post, reminds me dearly of Richard III: “I do the wrong, and first begin to brawl.”
Where do people get the notion that if you “beg nicely”, an author will immediately open up their database for your perusal? Will the author think that you want the data so you can frame it and hang it on the wall? I can’t think of a single person who would “welcome” a request to examine their data in regard to one of their publications, particularly from a quarter such as CA which has good analysis resources and does not share in the consensus.
Such requests are probably viewed in the same light as those from tax auditors. Even if you did not intentionally cook the books, you are afraid that maybe your interpretation of the tax laws may not be the same as those of the IRS (e.g. your application of a rule for determining the number of PCs you should use may not apply in your case because your data does not come close to satisfying the necessary assumptions under which that rule is derived).
My approach has been to be up front with the fact that I would critically (and hopefully, fairly) look at the results and share that information with others involved in the “auditing” process. With a reasonable response from an author, I would be more than willing to discuss the material with them first before posting anything. Unfortunately, when someone chooses not to respond at all to what I perceive as a reasonable request, as with Steve, the snark imp can sometimes overwhelm my tolerance levels as well.
The questionable methods, lack of transparency, stonewalling and outright resisting of requests for data and code, moderation filters and spin at RC, are the hallmarks of the “team” and predates (correct me if I’m wrong) any snark on the part of Steve. The frustrations of Steve putting up with all this stuff comes through in his posts. Personally, I think it is a bit hypocritical for the team to point fingers at Steve for how he conducts his blog when you consider this background. If RC put their moderation energy into filtering out the ad-homs and started dealing straight up and snark free with people who want to check their work then there would be a basis to expect the same.
There are many scientists who do not agree with Steve who come to participate in discussions here. How does Steve treat them? With snark and sarcasm? From what I have seen, these exchanges are cordial and professional. Views are exchanged and maybe even mutual understanding evolves. These are not people comming in here softening their language and tip-toeing around trying to make sure that Steve is not offended.
Steve rightly draws attention to behaviour scientific practicies that should concern any reasonable person.
You’re going to the extreme. No one said “beg”, just be respectful. Don’t assume something that hasn’t been shown yet (like “newsy applied results”).
I see nothing at all objectionable in the tone of Steve M’s correspondence. It is a bit “terse” but having to deal with a lot of emails in my job has instilled in me an appreciation of terseness as opposed to wordy and flowery prose. Get to the point, let me know what you need or want to say with as few words as possible, and let’s get on with it.
IMO, the only way to take offense at something like this is to be deliberately looking for offense to serve as a justification to not have to respond to the request. In my life, I’ve noticed that (particularly in online forums and correspondence) when someone starts complaining about the “tone” of the discussion, it’s because they know they are wrong or do not have the facts to completely back up their position. It’s used deliberately as a way to distract and move the debate side-ways.
If Dr. Steig thins this our Host’s correspondence is harsh and insulting, he must have led a very, very sheltered life. I’ve seen such interchanges that would make his hair turn white, all between people who nevertheless kept the debate going and responded to requests for information freely.
I am consistently amazed by the attitudes of some involved in Climate Science. Maybe it’s a case of the old academic saw that states “Why are the battles in academia so vicious? Because the stakes are so small.” Unfortunately in cases like this, the stakes, for science, it’s reputation, and for people who will be impacted greatly by proposed government actions based on such research, are anything but small.
Hmm. I too read “newsy” as a typo of newly. I find it most odd that an English major would expect that it wasn’t, given that the sentence structure reads very oddly if that were the intended word, but perfectly fine if it’s been mistyped.
I’m also in agreement the normal terseness of business communication probably doesn’t fly too well with those who don’t venture off-campus or outside the echo chamber into normal society.
If I may suggest, Steve, in future perhaps something like this:
Oh great munificence, upon whom the sun rises and sets, kissing the air around his glory.
If it pleases you, instruct this, your humble servant in those mysteries which are yours solely to comprehend.
I understand that one will never become so wise and noble as your radiance at the arts, but, if it behooves thee, one hopes to gain understanding of some small portion of those ways and means.
After all, with one so thin-skinned as to take offense at the “tone” and “implications (however thinly veiled)” of an email, flattery will likely get you everywhere. And a trowel is the appropriate tool with these folks methinks.
I think Stieg is irritated because his attempt to cherry pick data to support a pre-ordained result was busted.
Dissembling off into how unruly and cheeky the proles are is a not unexpected response.
I think this discussion is missing the point. Steve was polite enough not to be tossed out the door, even I was at the beginning. Most of what Steig said was likely directed toward my own and other comments in the thread at WUWT but even that isn’t the problem.
If someone publishes a scientific paper they should welcome questioning. In hindsight my early polite requests for code and data clipped from the RC thread indicated a bit too much fear and defensiveness of response to make me comfortable. Now after reviewing the paper, the faults it has are apparent to everyone and it’s difficult for the author to change them after the fact. Also, I believe I attributed some of the attitude of the constant attack dogs at RC to the hosts – something I of all people should understand better. The defensiveness from the beginning has a different look for me now that we’re nearing the end.
I don’t believe this or the false superiority complex had to do with anyone’s tone but rather a lack of certainty of the conclusion right from the beginning. When Steve M discovered the mysterious patterns in autocorrelated data combined with correlation vs distance smearing, the paper was in very big trouble.
So then the paper is replicated and everything builds until Ryan’s post on tAV, CA and then WUWT (which was a bit stronger than they expected) resulted in a request for reply from Dr. Steig’s collegues. He chose to do an unfortunate post on overfitting and improper verification. We’d already looked at these details of course which left him very open. The post yesterday was like Christmas morning.
My own thought was to wait a bit and ask a few questions, Ryan’s reply was very strong. So strong in fact that it was removed from the thread, discussed behind the scenes, accidentally reformatted and initially re-posted without reply. Later comments were added and the thread was ended with about the nicest comment I’ve ever read from an RC scientist to a person assigned the title of ‘skeptic’ – not that I read there often b/c I’m not typically allowed to post.
I guess what I’m saying is the good doc was unsure of himself yesterday and likely from the beginning. My guess from his recent explanation of the math is that Mann was the mathematician. The early reaction had nothing to do with tone but rather that Dr. Steig was unsure of the result. After Ryan’s post and reply, his tone became reasonable (even to me) completely the opposite to what I encountered from the beginning. I feel a little bad for his current situation, but am happy with politeness. So in return I’m going to forget the past and make a new effort to be extra polite. Imagine how much better things would be if the situation changed between RC and those with real questions and it went both ways.
Re: Jeff Id (#57),
This is evidence of a lack of confidence in his results. Shutting the thread down, too. Note that Steve doesn’t shut a thread down as soon as someone shows his mistakes (what few there are), rather, he accepts the information and moves on with a revised view of what he was doing.
Mark
Re: Jeff Id (#57),
But how often does that really happen? Again, I’m no scientist (as if that wasn’t blatantly obvious!), but one hears of heated discussion where one scientist calls the ideas of another “preposterous”, among other things. It’s against human nature to want to be questioned about your work. Some folks can rise above that, at least on the surface, but the vast majority do not.
As a soon-to-be author of a paper that involved using positive matrix factorization (similar but superior to PCA for source apportionment if you ask me)to determine source profiles, I knew my statistical analytical abilities were weak and relied on a statistician friend to help. He had no investment in the results one way or the other and provided great assistance. I am sure if anyone cared about the results of this paper (no one outside of aerosol chemists will) flaws could be found in the analysis that would not doubt reinforce said persons position. If I get an email from Steve I will surely be transparent as I have nothing to hide and no hidden motivation, but the amount of time a back-and-forth takes up might not be worth it. I don’t know how Steve finds the time to do everything he does.
I think having co-authors that were statisticians would be valuable if you could insure their neutrality, which would be near-impossible with regards to anything related to global warming. But I can dream can’t I?
I’m not a scientist but it’s my opinion that your request was judgemental and Stieg’s natural reaction was defensive and to ignore it.
You can sometimes get something out of people by “killing them with kindness”. However, I’m doubtful that the employment of all the principles in Dale Carnegie’s “How to Win Friends and Influence People” would have gotten you the information you were requesting.
Your reputation now preceeds you and short of a FOIA request, I would not expect “team scientists” or other Climate Scientists to provide you information. There is too great a risk of embarassment for them to do so and no compulsion that requires them to do it.
In my short reads of science history, those on opposing sides of the aisle do not always “play fairly”.
Re: Edward (#61),
Well, yes there is, actually. If they don’t, then it isn’t science.
Re: Edward (#61),
You mean, like, if they’re wrong? The whole point of the scientific method is to try find fault with your hypothesis… you understand that, right? It’s only embarrassing if you refuse to let others make that attempt.
snip –
Mark
I agree with anon (#14), that SteveM’s emails are fine. I too have written such terse letters and emails to clients, partners and colleagues, with no subtext intended; the tone is brevity, not confrontation. But I’ve also communicated with government agencies (eg. GSC, OGS, NRCan, CCRA) and learned that they tend to respond more directly and in a more timely manner when handled with kid gloves. Steig and other climate scientists that SteveM attempts to deal with are essentially public employees who seem to expect certain niceties that the ROW doesn’t necessarily require.
Regardless, with the well documented history of stonewalling by Steig and his colleagues, its probably too late to treat them with kid gloves. Their hackles are firmly in place and unlikely to be lowered. Not particularily professional, IMO, but it is what it is.
I think Steve has been too civil. All my advisers over the years would have dragged me through hot coals. And I would have expected no less from complete strangers. Goodness, has any member of the team given a seminar before?
We are having a discussion about materials and methods. This is something I would discuss the first day of my general and organic chemistry labs. The students would then be graded during the year on how well they listened.
If I had ever overlooked some detail in a paper, I would have apologized to the questioner and immediately sent a letter to the journal with a correction to materials and methods.
The fact that members of the scientific community can withhold information necessary for repeating an experiment, with no word from their institution or journal, does irreparable harm to scientific inquiry. I often wonder what the National Academy of Sciences and the American Association for the Advancement of Science do, other than publish their respective journals. When they write about the need for good science education, what are they referring to?
There are scientists (faculty, post-docs, graduate students, and undergraduate assistants) who might not appreciate the reference to freshman general chemistry or sophomore organic chemistry. However, I wonder how all these people can sit by and let one of the basic rules of scientific research be ignored.
Walter
One point not yet drawn out is the responsibility that Nature has in not insisting that its policies regarding disclosure of data and methods be observed.
No doubt Nature is following this controversy, or at least it should be. The detailed discussion at several sites is bringing their failure to require proper archiving into the public eye.
If they had followed their policies, we wouldn’t be having this discussion.
As a further point on data: I would much prefer that climate scientists comply with a “best practices” standard e.g. that of leading econometric journals, in which the journals require them to archive functioning code and data as used as a condition of review. This entire song and dance engaging in quasi-litigation for data is a total waste of time.
And for readers who think that the refusals have something to do with tone or my somehow missing the magic word – don’t kid yourselves. The refusals have nothing to do with the tone of the request. I could have dressed the requests in the flowery language of an 18th century courtier at the court of Louis XVIII and it would have got the same result.
Part of the refusals are defensiveness. However, I think that some of the refusals are simply litigation strategy. The more time that I have to spend fighting to get the data, the less time I have to actually analyse the data.
As long as the conversation is limited to the journals, they can get away with this, because (1) the journals either do not have or do not enforce proper policies; (2) the journals will reject comments referring to data issues as “non-scientific” e.g. the reviews of this part of our Santer comment.
The popularity of climate blogs makes this strategy less effective. Every time that one of these guys refuses a data request, I’ll publicize it. It gets the frustration off my chest and places the refusal in the sunlight. The refusers hate the exposure. The journals hate the exposure as well.
The public judges the refusals entirely differently than the “peers” and whenever this sort of issue arises, there’s a lot of bad publicity for the scientist in question and a lot of piling on at the blogs. The scientist invariably blames me for making an issue of the data refusal, rather than looking in the mirror and asking whether he ought to have just archived the data in the first place or, at a minimum, when asked.
Sometimes, the scientists lose track of what actually happened. In the case at hand, I obviously sent an email to Steig requesting data (and code). And yet only a few days later, Steig demanded that I be disciplined for criticizing the failure to provide data without even asking for it – a patently untrue allegation.
Re: Steve McIntyre (#71),
I generally agree with you, Steve. That doesn’t mean the “tone” doesn’t make a difference.
Obviously the AGW/Skeptic situation is the culprit here, even though, as you’ve stated in the past, you don’t disagree with the general theory of AGW. So in reality you’re on the same “side” as the RC folks in that respect, you just don’t agree with their practices, or lack thereof. You demand robust results, or at least robust methods, and they don’t seem to like that. I get that. And I don’t think it’s going to change. We’ve had congressional hearings, published refutations, etc, and the juggernaut is still steaming along.
Re: Steve McIntyre (#71),
Re: Anthony Watts (#92),
.
.
As I indicated in my previous comment, I completely agree with your (Steve’s) characterization of how defensive The Team is and, imo, know their results are being legitimately criticized and, hence, seek almost any excuse to disengage from any scientific give-and-take. The purpose is to give as little publicity (and concomitant credence inseparable from it) to anyone that might suggest the debate isn’t over.
.
In my experience, they are too often condescending, personal, and deliberately belittling and dismissive of anyone who might credibly undermine what they perceive as their standing in the eyes of their climatology community and, especially, the mainstream media (not to mention the funding stream their being viewed as “the experts” requires). Real Climate’s censorship of legitimate comments is as crude a propaganda technique that one can employ and I think they understand exactly what they are doing.
.
In other words, I agree with your characterization of their approach, Steve, but I think a larger issue is being missed: Given what we know are the Team’s SOPs, any interactions we have need to be based on that understanding and appropriate “strategies” developed for presenting the “outside world” the most trenchant arguments for beginning, by the simple act of questioning, to put pressure for a change in behavior of the offending parties (and journals).
.
Where I do disagree with you, however, is in assuming their behavior is immovable. I have long participated in “discussions” with the most insulting individuals on a range of topics in, for example, USENET newsgroup forums. While I know the discussion will be futile in changing a particular individual’s mind, I also understand that there are “lurkers” who are watching how the debate unfolds. My posts are for them, not the purveyors of the vitriol and personal attacks. They are persuadable and lurking for the very reason that they have not developed strong views on the matter.
.
And even with the most incorrigible of “opponents,” I notice that a strictly neutral, rational statement of where I stand, even immediately following a post/comment made from animus where I still act as if they presented a neutral argument (at least I treat it as such), actually can change the way the other individual responds. You’d be surprised at the change in attitude that often takes place–at least in the tone of the discussion. In a sense, they are being shamed into proper behavior in order to avoid the very contrast the impersonal, neutral, rational response to them would highlight. In fact, they might be downright disturbed by the lack of response to their incivility .
.
It’s this reduction in paranoia, if I can use the term, that the conversation (or data, in your case) isn’t being pursued for the purpose of humiliation (not your fault or a rational perception of the way you correspond with them), that I think over the long run can elicit additional cooperation (even if simple begrudging acknowledgment of valid points).
.
Heck, if you can invest them in your own research (e.g., suggesting cooperation on a paper with them to “extend” their results rather than to debunk them, asking their opinion on a point you’re considering) might, with time (and depending on the individual–Gavin might be a lost cause ) produce some conciliatory gestures. If they perceive your work as adversarial, they’re bound to treat you as the adversary.
.
What I suggest above can be a time-consuming process–if merely in the careful construction of correspondence or remarks. But I think it’s a process that has at least the possibility of bearing some fruit. If not, the outside observer will be presented with the very contrast that will convince them of the validity of your approach–and that is the ultimate goal of any approach one chooses.
.
Didn’t mean to be long-winded (though I guess I actually did, since I wrote it all down ), but I tend to err on being comprehensive versus being misunderstood–a fault I impose on others who have to wade through the material.
Re: JohnM (#102),
JohnM, I would have liked to have settled some of these issues long ago. There are lots of things that I’m interested, but which are hard to get to when even the simplest thing is contested.
While there aren’t many overt concessions, as I mentioned before, some young climate scientists at AGU mentioned privately (and ABSOLUTELY not for attribution) that they felt that I’d proved that Team-style multivariate methods could not do the job on existing data and that the only route out of the impasse was to get new and better data, which might take 10 years or more.
Von Storch, who’s vied for the minor credit of breaking Mannian methods, has said that he had trouble publishing critical analyses prior to our papers. After our papers, the methodological issues were firmly on the table.
I think that there is increased caution in the media as well. I can’t help but think that Andy Revkin’s take on Mann et al 2008 would have been different without Climate Audit.
Re: Steve McIntyre (#107),
I agree completely on the influence CA has had on the greater work of the scientific community and issues that would never have been approached by them otherwise. I can only express a hearty appreciation that your site exists. I absolutely love the scientific/technical detail (along with Jeff Id’s site and the guest postings by RyanO and others).
I recently came across this anonymous referee #2 critique of a paper by Burger and Cubasch on Mannian methodology (your papers are cited along with others)–specifically targeting the CE & RE verification scores (see here).
As I read the vitriolic and utterly condescending tone of referee #2 (regardless of the merit of his points), visions of RC blog antics popped off the page immediately. I literally could not distinguish the respective tones and assume that it comes directly out of the mouth of a RC Team member. It just absolutely astonishes me that someone lacking such rudimentary professional collegiality is allowed to even sit as a referee on a paper–not to mention one that, it appears clear, they have a personal stake in.
Someone so emotionally invested in a particular theory or outcome has crossed the line from science to religion. Once the emotional attachment has become this apparent (as is so abundantly attested on RC), it is impossible for the concerned party to do good science, which relies on objectivity and open-mindededness to new possibilities that are the lifeblood of the scientific method.
Re: JohnM (#119),
We discussed this review at the time. Reviewer #2 is obviously Mann himself, both from the distinctive vocabulary and writing style and from the condemnation of the article for not considering the findings of still unpublished Mann submissions. I was asked to review Bürger’s resubmission; I don’t like reviewing papers as I actually do it quite thoroughly. So poor old Bürger ended up with both me and Mann as reviewers. Oddly my review of his resubmission is not online at CPD, though Bürger’s response to my review is (he characterizes me as a thorough reviewer) . Go figure.
This is such old hat.
I’ve witnessed the identical response in exchanges with others on a wide range of issues. By bureaucrats where I have requested public information too.
Over and over again.
The old hat part is the ultimate refusal to provide what is requested.
That’s always the convenient result which follows the exagerated pretense of personal offense.
Never mind the often poor tone from the recipient of the request, it takes very little offense to trigger the denial and exit from the exhange.
Dr. Steig could have easily “scolded” McIntyre in any number of ways and conveyed his offense.
But as is often the case the preferred default choice is to obfuscate, obstruct and refuse cooperation. Why? Because they can. It’s essentially the middle finger.
Re: Howard S. (#73),
I have too.
I used to work at night in a Network Operations Center (NOC). I am/was fairly PC-literate, so was the de-facto person to fix on-the-spot PC software and hardware issues. So invariably I would end up opening tickets with LAN support to fix problems I couldn’t take care of. Because I wasn’t there when the support guys were, my tickets were often left untouched or glossed over. Many of the issues didn’t get fixed until I got belligerent. Consequently I would get in trouble for getting belligerent, while the support guys weren’t held to task. It just seems to be the way a lot of people are.
Has anyone looked at these studies that address the statistical methodologies?
Christiansen et al (2009)
The underestimation of the amplitude of the low frequency variability demonstrated for all of the seven methods discourage the use of reconstructions to estimate the rareness of the recent warming. That this underestimation is found for all the reconstruction methods is rather depressing and strongly suggests that this point should be investigated further before any real improvements in the reconstruction methods can be made.
and
Riedwyl et al (2008)
This paper presents a comparison of principal component (PC) regression and regularized expectation maximization (RegEM) to reconstruct European summer and winter surface air temperature over the past millennium. … For the specific predictor network given in this paper, both techniques underestimate the target temperature variations to an increasing extent as more noise is added to the signal, albeit RegEM less than with PC regression.
It appears that the methodologies do not capture the low frequency varibility.
Re: Vernon (#74),
Thanks for the abstracts on low-frequency variability. Record-length is invariably the ultimate constraint on our knowledge of climate. The Antarctic records are all simply too short (and too often gapped) to capture the full variability. They no more show the range of temperatures than, say, a half-hour wave record shows the range of ocean surface elevation, in which tides figure prominently. And, unlike the tides, there are no sound means of extrapolating the broad-band, low-frequency components seen in the longest available temperature records. Kalman-Bucy predictive filters might do a good job for a few years, but not for decades. Moreover, it is precisely the low-frequencies (below the Hale cycle) that are most subject to corruption by UHI effects (particularly strong in cold climates), instrument changes and drifts due to deteriorating shelters, etc. It’s supremely ironic that Steig would choose to reconstruct a 50-yr trend for a continent whereon the average record is considerably shorter. But such is the nature of agenda-driven “science.”
Just for the record, are the detailed methods and code for Steig 2009 archived and available to the public?
Re: Clark (#75),
No.
It’s still not, another reason for my sometimes too critical tone.
Disciplined? A spanking perhaps or maybe random bits dropped from your blog posts. Seriously, this guy needs to grow a pair.
Steve,
May I suggest than any future correspondence with Steig should be formatted in the following way – so as to avoid any misunderstanding or hurt on his part:
>>Thanks. 😉
“all the material” – It would also be an excellent idea to put your source code up. 🙂 Using statistical techniques that are not well understood 😮 to derive newsy applied results is a bit risky 😕 and you should err on the side of caution in making your code available to independent analysis. 😉
It would also serve to defuse people who are instantly suspicious of Mann’s RegEM. 🙂 This is an application where it seems much more plausible to me than where it’s used to justify bristlecones. 😐 I’ve noted this at my blog as a caveat to instantly suspicious readers. 🙂
Regards, Steve McIntyre 😀 😀 :-D<<
Winston Churchill ended his December 8, 1941 letter to the Japanese Ambassador, declaring that a state of war now existed between the United Kingdom and Japan, with the courtly flourish “I have the honour to be, with high consideration, Sir, Your obedient servant”.
Some people did not like this ceremonious style.
Churchill’s response:
“But after all when you have to kill a man it costs nothing to be polite.”
If I ended an email to Steig with the salutation:
I don’t think that it would help.
I agree that it costs nothing to be polite. I try hard not to get angry and encourage readers to be polite. I fail to see anything impolite in my email. It’s not mannered or fawning, but neither is it impolite.
Re: Steve McIntyre (#83), It’s all a matter of perspective.
Re: Steve McIntyre (#83),
Now THAT would be funny! 🙂
Maybe the next time you request data you should post your draft request for “auditing” by the masses here before you send it.
Who knows? Maybe if you sent a Churchillian email it might break the ice…
Nah, never happen.
Look what I found — my first requests for data and code at RC.
Cut from moderation. I tried again.
Cut again.
Undeterred I tried again.
Eventually part of a comment was let through in edited form on RC – requesting code and DATA.
I like the new tone better than the old.
snip – you’re using prohibited language
Or “Climate Journal Audit.”
To be honest, I think this blog of Steve’s is doing what you describe above very effectively.
But what the hell: let a thousand flowers bloom.
If you want to call yourself a scientist, do your work openly. If you want to impress while hiding your method, be a magician. If you want to try to get away with doing both, lose the thin skin.
BTW Steve, was it ‘newsy’ or ‘newly’?
RomanM, #70: “Perhaps I should reconsider its use in the future.”
In a letter in which you are requesting crucial and hard-won data, it might be best to stick with established formalities. Other than that, I thought your letter was very good.
But Steve is right. Hard-core members of the Team this time probably had no intention of cooperating in an effort to get the science right.
I’m wondering if the fact that you smart guys were able to penetrate and free Steig’s code for this paper has the Team in a state shock. I’m also wondering how it will effect their future work.
You did a great job. It was exciting to follow your progress.
But it is not a business proposal, so the analogy fails — I’ve participated in hundreds of business exchanges as well. One key difference is that in a business proposal, even when tensions run high, there is a vested interest on both sides to work together toward a larger goal and get something done. The situation here is completely different. Indeed, there is a vested interest on the part of one side to avoid communication and involvement altogether. Thus, the context of the communication is completely different. It is this different context that I believe calls for a more subtle tone.
I’m a huge fan of Steve’s tremendous efforts and contributions, but given the wording of the “all the material” paragraph in his last email, I’m not sure he should expect Steig to acknowledge that his “communications, like Ryan’s, were courteous.”
Two comments:
1) Email by definition, is often a non-gracious medium given it’s often hurried composition. Email often results in serious misunderstandings. Been there done that.
2) For the record, since WUWT is mentioned both here and on RC, I was not in communication with Dr. Steig. I’d post the same thing on RC but Dr. Steig decided he didn’t want any more input and closed comments.
But I agree with Steve, even if the data request was delivered in person with Roses and a box of chocolates, I doubt Dr. Steig or any other RC team members would provide the data and code, no matter the framing. It is not in their best interest to do so and replication reviews are so darned inconvenient.
Time to take it to the next level then.
Oh forgot to mention. I’ll graciously apologize for my stance if Dr. Steig, knowing full well now the intense interest in data and code, produces it.
Steig’s analogy with auditors is quite interesting. The entire scientific enterprise is built on unofficial audits. If you claim you can synthesize an important hormone or something, people will try to duplicate it. If you claim cold fusion, ditto. No claim about subatomic particles is considered valid until more than one team can duplicate the observations. The reason most work is NOT checked is simply that it does not rise to the level of importance that other scientists examine it critically. And yet, Steig does not like people looking under the hood. Curious.
Re: Craig Loehle (#95),
What is really striking is the seemingly total lack of interest in the climate science community to replicate such studies as MBH, Steig, Santer etc. (Instead they make new studies and ‘move on’.)
I agree with many here who don’t think Steve’s ‘tone’ has anything to do with it at all. I would rather think that those nuances ‘don’t matter’.
After all, isn’t all this really just noise? But the need for transparency and full disclosure of data and code shouldn’t get lost in that noise.
Apparently you can look under the hood, if and only if you are wearing the appropriate purple robe.
Re: jeez (#96),
That’s known as the “Guild of Mannian Mechanics”.
Mr. Goodwrench, they are not.
Re: Anthony Watts (#98),
In fact, they should form a band: “Mann and the Mechanics” 😉 and play only digitally synthesized music.
Some suggested song titles might be:
“Mea Culpa”
“All I need is a miracle”
OK I’ll stop now.
I am an associate editor for Ecological Modelling and as such review about a dozen papers a year for the journal. One explicit question for reviewers is whether the model equations are described in sufficient detail that a reader can duplicate the work. If not, it gets rejected.
I was in a workshop (around a table discussing biodiversity) and the biggest cheese in the room, a charming but unstoppable famous person, was putting his new index up on the board. I pointed out that it could become negative, which is nonsensical. He said “impossible” and it was just then time to go to dinner. The next morning he said first thing that I was right and he presented a fixed index. That’s how it should be. Conjecture and refutation. If not it is P H D (piled higher and deeper)…
I think it’s a bit late to talk about “gracious” communications. This is war.
You chaps are on the bleeding edge of Science 2.0 where a lot of science is going to get done on blogs and by guys in pajamas running R in their basements.
The Team just don’t like this at all. And why would it? In the end, Science 2.0 is going to put the Team out of business.
Maybe what we need is a Code Duello.
Perhaps a good example of the strictly neutral, scrubbed clean of any “hooks” for offense approach, I think Steig’s reaction to RyanO demonstrates that the treat-with-kid-gloves (but direct) approach can elicit useful discussion. And when Steig cuts off the comments given the well considered and scholarly approach of RyanO, the reasons become obvious. No “I find your last remarks to be offensive” retorts–only the silent retreat of a scolded dog back into its doghouse (metaphorically, of course–I wouldn’t want to offend anyone ).
On a separate front, I’ve always found the if-it’s-not-peer-reviewed-I’m-not-going-to-discuss-it camouflage to be a blatant example of the logical fallacy of appeal-to-authority. Being or not being an authority has no bearing on the validity of a rational argument.
This episode followed a typical Team script:
1) Make a half-hearted attempt at cooperation
2) Take extraordinary offense at some minor slight – real or perceived
3) Storm off in high dudgeon while claiming “I will not respond to Steve M…etc. etc.”
They follow this script so that they can claim that they tried to cooperate but were abused by mean old Steve M.
For someone who been on the receiving end of so much slander and vilification, Steve is rather restrained and should be entitled to a little snark.
Tom C. #105 summarizes the situation admirably. Steve’s requests are absolutely straight forward; everyone is an experienced professional; his “snark” is tiny compared to the bluffing, rebuffing, and acting in high dudgeon of E. Steig re necessary verification of scientific research claims. ‘Fess up everyone; Science today needs Auditing rather than Peer Reviewing.
We’ll go lite for a minute: Is it supposed to be “newsy”, or that new contender “newly”? No answer yet! Ah, the perils of navigating in the English language! Somehow this reminds me of Dorothy Parker’s great comment about horticulture: “You can lead a horticulture, but you can’t make her think.” (No matter how OT this comment is, I’m betting that that Parker quote is just too good to get snipped.)
In regard to the tutorial provided by Real Climate, I must comment on the level of comments between that blog and this one. Real Climate purports to be the blog which provides comments from real scientists. However, even a cursory examination of their comment section would suggest otherwise. Compare the comments received here on Steig’s PCA analysis and those on Real Climate. Here one can read learned and considered commentary on the mathematics. There one reads the exclamations of cheer leaders. There is next to no scientific commentary there. None of their regular commenters seem to perform any independent inquiry. Steig’s comments on blog were much more appropriate for Real Climate that Climate Audit.
Re: TAG (#111),
They’d probably get edited out if the did.
Misunderstandings happen. Check out the Halsey-Nimitz misunderstanding during the Battle of Leyte Gulf: “The world wonders”. On the other hand, neither Halsey nor Nimitz were withholding data, code or procedures.
…and you can lead a horse to water but a pencil must be lead 🙂
Shane
Wow, all the comments saying Steve writes with hostility and snarkiness, I just don’t get it. I suppose, my having NOT been educated at Cambridge, means I never learned how to write with the humility of a peon, but then again, I did go to US public schools, so you’d think they’d want me trained to communicate thusly with the bureaucratic powers that be and be a good little cog in the machine. I don’t know. Steve’s communications with Steig seem very collegial and polite, to me.
Re: Mike Lorrey (#115),
as an Oxford graduate, I’m relieved to learn that Cambridge students learn “how to write with the humility of a peon”. This obviously avoids much confusion in polite society. One wonders whether this aspect of the Cambridge curriculum might be profitably incorporated into the curriculum of climate science classes in provincial American universities. At Oxford, needless to say, students did not receive instructions in this technique, as it was obviously not relevant to them.
Re: Steve McIntyre (#124),
That reply sounds more English than Canadian to me.
Steve: quite so. At Oxford, they used to call Cambridge the “original redbrick” university. I don’t know whether anyone at Oxford in 1970-71 had ever heard of the University of East Anglia. Hey, it was then.
Re: Steve McIntyre (#124),
Given Mr Lorrey’s self-confessed lack of experience with a Cambridge education, I do not think that he can be regarded as a reliable witness with respect to the content of that education. Accordingly, Steve really should not be taking Mr Lorrey’s word for it without checking the accuracy of his assertions.
Of course, we at Cambridge had heard of Oxford. It was the place you would pop over to for a boozy weekend with the thickos from school who could only do artsy subjects.
Tra-la !
Steve: This nicely illustrates the tone of my prior comment for Ken Fritsch – Oxford and Cambridge comments from one to another are a tony sort of “Yo mama’s so fat that…..” Ken, both my comment and fFreddy’s are in that spirit – don;t take them too literally.
Re: fFreddy (#127),
Steve M, I get it. I’d like to think that the well thought out and properly delivered barb/prodding or even an insult has not only more sting, if that is what was intended, but can provide some entertaining and competitive verbalizations, if that is what was intended.
I refined my techniques from watching British based shows on PBS after first learning them as an under-aged bar room participant in college who needed a way (or the satisfaction) of swapping insults with those of age and at the same time avoiding a physical conflict and potentially being kicked out of school.
Steve, Jeff Id and Co don’t “adjust your tone” to please sanctimonious utterings from anywhere. We would all be the poorer for it.
Let’s not kid ourselves, the hurt feelings ploy is transparent. It’s a delaying tactic until they can conjure up a suitable defense.
Good luck to them…
Real Climate has no leg to stand on with regard to this issue of civility.
The best solution seems to be , as you said, to expose their regular refusals to share data, one by one, as they occur—shine the light.
snip – you used a red-letter prohibited word, prohibited regardless of the context.
snip
Steve: Blog policies prohibit speculation about motives.
Re: Edward (#121),
snip – c’mon, Mark – let’s not debate this sort of stuff
Fine..
Edward, please re-read my post. You completely missed my point. I did not state, nor imply, that AGW is a scam.
Mark
Whenever I see exchanges like the one Steve is detailing here I always try to look at what is happening from the point of view of all the parties to see who is not being reasonable.
Whilst I am personally supportive of much of what Steve is doing I have had reason to criticise him on a couple of occasions. This is primarily attributable to where I consider he has escalated matters via this blog without allowing private email correspondence to reach its full conclusion.
To a lesser degree I believe that applies in the instance described here although I have much greater issues with how Eric Steig has handled matters than with what Steve has done.
Firstly I do not have any particular issue with the email exchange that Steve has detailed. However I do not believe that the emails are what Eric was complaining about. The complaints are attributable to a post that Steve made on Jan 21st titled Antarctic RegEM. That post raised issues with data availability prior to the email exchange between Steve and Eric. I would have preferred that the email exchange had taken place prior to the blog post.
Eric expressed 3 main complaints about that post.
1. Steve mentioned that some data was not available. Personally as later events established Steve was entirely justified in raising the issue so I have no sympathy with Eric on how he handled the data availability.
2. Steve made the following statement “Matches to information at the BAS are complicated by Mannian spelling errors and pointless scribal variations (things like D_10 vs D10, which can be matched manually, but why are the variations there in the first place??)”. The reference to Mannian spelling errors is a bit snarky but Steve is identifying quality issues which are entirely legitimate to highlight. Steve has also been on the receiving end of much worse snark from RC and others so I have little sympathy with this issue either. However I would prefer that Steve keeps the snark low key and humerous so he can claim the high ground in such exchanges.
3. Eric was complaining about accusations of fraud by association. Based on the exchange between Erik Ramberg and Steve at post 28 then I assume that some posters were guilty of this type of accusation and as I no longer see them in the thread I expect Steve deleted them in line with his blog policy. In my opinion Steve does a much better job of killing such accusations than the RC team (which includes Eric) does.
I do have an issue with Steve reraising the matter in this thread but only because it is a distraction from discussing what Eric has recently posted. It also wont help to build bridges not that I believe there is much chance of them being built whilst the current hostile attitude continues from most parties involved in the debate.
I do find the way that Eric has handled the post and subsequent discussion at RC to be rather ill advised. Personally I regard the final exhange with Ryan as an agreement to disagree rather than a concession. To me this demonstrates a continuing inability of the RC team to properly engage on technical discussions. I am also assuming that Eric wrote this to rebutt the work by Ryan O in particular and as such represents the extent of the issues that Eric has with Ryans recent posts at the Air Vent. Hopefully Ryan can now revisit his own work to ensure the issues that Eric has raised are fully addressed. That will increase confidence that Ryans work is valid.
Re: clivere (#126),
thanks for the thoughtful consideration of this. I won’t pick off all the issues but you say:
One of my standards for an article is whether the author has archived his data. Readers shouldn’t have to ask the author for his data – nicely or otherwise. From time to time, I’ve complimented authors who’ve had their ducks in a row and their data is online concurrent with their article. Jud Partin and David Black are two that come to mind and there are others. Mann made a good try for Mann et al 2008 though it was marred a bit by cock-ups. But he tried and I recognized the effort at the time. So I was disappointed that Steig hadn’t made an equivalent effort and reported it. I would like climate science to achieve a standard where this sort of clerical archiving is done automatically as a condition of publishing. Until then, it’s an item that I check off in a review.
The author’s response to a data request is a different thing altogether. And unfortunately, it was like pulling teeth to get the AVHRR data and, to my knowledge, the relevant cloudmasking data as used remains unavailable (Ryan O is up to speed on that.)
Re: Steve McIntyre (#128), Yep. Still unavailable. No ETA, either.
snip – the person in question was venting about the iniquities of climate scientists. He knows which word was prohibited.
snip – please stop this sort of complaining.
Re:#125
Funny you should mention it Steve, but at about that time I found myself with offers from Oxford and East Anglia to do my doctorate. Guess which I accepted?
Re 129:
I’m afraid I don’t know—I’m very puzzled about this.
I’ve read and re-read my post—and the only possible prohibited word is the one I quoted in [129] as having been used to describe me—-not one I used to describe anyone else.
If it’s not that, I don’t know for the life of me what the word can be.
As far as ‘venting on the iniquities of climate scientists’ is concerned—– what I was doing was describing the treatment meted out to any dissenter or questioner at RC—and commenting that surely on this matter that is going to affect all our lives, we have a democratic right to ask questions.
What is Stieg’s problem, anyway? We scientists are a tough lot, and don’t have to whine about an email’s “tone.” Stieg should have grown out of that foolishness when he graduated from kindergarten. Sad to see scientists guarding their cherry picked data when they should be trying to discover the truth.
Here’s another example of gracious communicating from the master himself (Mike). From a comment on RC…
Re: BKC (#136),
Too funny. The thread was open for 12 hours. I didn’t even know the post was up until after it had closed.
Re: Steve McIntyre (#138),
I guess I must be an attacker – since it’s plural, so since I was allowed a single question I used it for a single purpose. I don’t recall saying that upside down thermometers was a reasonable methodology. I also don’t recall any response from RC to the fact that Ryan checked the verification against BOTH datasets corrected and uncorrected with almost the exact same result. Mike seems to have missed that little itsy bitsy detail.
What is an upside down temperature anomaly anyway? Is it an anomalosity?
I can understand the rational somewhat, the thermometers are on the bottom of the earth so they are already upside down but I get confused as to why they don’t invert all the data. Every day I start to understand more how climatology is way beyond my simple self.
Re: jeff Id (#139),
Jeff, don;’t forget Mann et al PNAS 2009 (reply to MM):
The corollary to this is that Antarctic RegEM is “insensitive” to whether the station data is used upside down or not.
Re: jeff Id (#139), So, you have a fringe website, eh? Is it that leather fringe on the sleeves? That is out of fashion anyway…
Re: Craig Loehle (#141),
On the fringe of Climatology, thank god. It’s pretty cool being recognized by famous people – no matter what they’re famous for. I wonder if I could get an autograph on my Easton.
I’ve been pretty hard on the guy but of everything I’ve ever seen, he’s earned it.
Re: Steve McIntyre (#140),
I doubt I’ll forget that one until well after I’m gone. There are no words I can repeat here for the response or the (So surreal that I’m not sure if he means it) opinion. In my mind the upside down proxy isn’t nearly as bad as an upside down thermometer simply because nobody really knew if they were temp anyway. How this tiny detail can be assumed unverified is way way beyond me. But if they know the assumed mechanism for relationship how the [real snip] can they just flip it. Ok, so even beyond that they flipped it from the early recon to the late recon in M08 how idiotic is that.
At least they had the claim that it was a temperature proxy so who can say for sure what other influences are involved and it likely didn’t have a huge effect because there are hundreds of proxies. In the Antarctic recon they are THERMOMETERS for god sakes. Sure they are on the backside of the earth but even there they measure temp don’t they? And they only used 34 thermometers flipping 5 of them upside down!
If they don’t mind flipping a temp anomaly over I can think of some new modeling techniques which could help GISS get more in line with satellites.
Re: jeff Id (#142),
Jeff, here is an excellent commentary on the use of obscure multivariate methods to flip data (referenced on an earlier occasion.) It is also a convincing answer as to whether it “matters” whether you flip data.
Re: Steve McIntyre (#144),
That is hilarious, I made it a post.
I’ve got some smart people on tAV who despite the obvious nature of the problem trying to rationalize how it might be reasonable to read thermometers upside down. It’s a funny world we live in which I’m unlikely to ever understand.
Re: jeff Id (#146), that was originally pointed out by a reader on the Tiljander thread.
I really don’t see a distinction between the Youtube cartoon and Mann’s reply at PNAS.
Re: Steve McIntyre (#148),
I’m not familiar with the process of the criticism and reply. Does someone review the quality of the comment and reply or is it just added to the record?
I mean, what he said makes no sense in the physical world under any form of possible interpretations of science, so is there somebody who’s supposed to say—Dear Michael Mann, I’ve reviewed the situation carefully. Please explain what the heck are you talking about, Love boss guy?
I’d like to send one about inverted thermometers to Nature without going after all the endless detail. It’s too simple with only 34 thermometers being used.
Re: jeff Id (#146),
Obviously, some of the thermometers are inversely teleconnected to the continental temperature. That is apparently one of the features of REG-EM. It knows which series need to be inverted to find the “true” trend, similar to the Mann-o-Matic.
I understand a new, advanced algorethm is being developed that can not only invert individual series, but adjust series slopes to come up with the “true” trend. It’s going the be called the FIX-(s)-EN.
Mann is lucky for a thing known as tenure.
Mark
Steve – I recognise that you are rather bemused by the premature closure at RC but it is also worth trying to figure out why Eric made the post in the first place and what he was trying to achieve in constructing it the way he did.
If you read past the condescending tone then it is obvious that he felt that something somewhere needed rebutting. However at no point did he identify clearly what it was. There were some footnotes added later that specifically mention Ryan O and link to the 2 posts at the Air Vent one of which contains content that Anthony reposted. Whilst I suspect it is the second post they were most concerned with the footnotes imply that Eric was not aware of it until he had already started work on his post.
It is clear from some of the comments by Eric and Mike Mann that the claim that the paper had been debunked was a source of irritation. This may be attributable to Anthony using falsified for his post title.
However that does not in itself explain the need to do the lecture and follow that with the technical justification for PC retention. As a result I perceive they do have significant concerns with what Ryan has produced. I am still not clear if Eric made the post because PC retention is the root cause of their concern or is the target for rebutting because he feels they have a viable argument here. Given that they made the effort to do the post then I would have expected them to raise any significant issues with Ryan’s work that they were aware of
Re: clivere (#143),
They were only showing that they responded to criticism. Then they ended it “before” too many questions got to the point. I say “before” because they deleted postings even though they will not admit it.
This a a real joke. The last time I attempted posting, I was not even criticizing. I expect it was automatic. Spelling needs checking as well.
Jeff, here are weights of the long temperature “proxies” in MBH98 AD1820 network. At the end of the linear algebra, 3 of 11 series have been flipped over. Thus, cold Leningrad temperatures in the late 18th century are interpreted as evidence of warmth. Perhaps Napoleon used Mann et al as forecasters in 1812.
The Team riposte is that it doesn’t “matter” – they can still “get” a HS without the flipped over temperature series and, indeed, nothing much “matters” except the bristlecones. (And you know the anguish and gnashing of teeth if you do a recon without bristlecones.)
From a mathematical point, it certainly evidenced to me that there was something wrong with the method when it resulted in such nonsensical results – a concept that thus far has eluded the Community.
It is commonly accepted that intellectuals tend to be thin skinned, but Steig’s ballastic reaction must set new a new benchmark for becomming unmoored. No doubt he took Steves statement that “…you should err on the side of caution in making your code available to independent analysis” to be a professional admonishment, and it touched off a barrage of self-serving defensiveness and charges.
The over the top wailing of being wronged usually suggests that someone is engaging in a deliberate descent into irrationality – a way of using denial, acting out, and delusional projection to defend the sense of moral self. Steig’s “I do not desire to act in a professionally unethical way, you do” is a way of placing his own motives on another person.
Obviously the fellow needs how to use mature defenses of the self, such as a mild humor.