W/O Comment
Since when does “adapted” mean “redrawn or drawn from the data”? If I had the data that went into the graph, then I would have plotted up a new graph—but what’s the difference if I plot a new graph the way I want or alter some other graph so it plots what I want to show?Paul Krugman
I go over to the CRU site and get the latest temperature data and plot it up—If I want to plot it up for 1901-2000 I can, if I want to plot it up for 1851-2010 I can. I am not bound to use only the plots contained in Brohan et al., but I could use those plots if wanted to. Had Brohan, for space reasons or other reasons, plotted the NH, the SH, and global combination as a single figure, and I’d have no compunction against erasing the NH and SH if I wanted to concentrate on the global results. If Brohan had plotted the HadCRUT2 and HadCRUT3 on a single figure (sorta like their Fig. 13), and I only wanted to show the HadCRUT3 data (after all, that is what the paper was about), I’d have no problem erasing the HadCRUT2 data from their chart. Had Gillett et al. plotted their “improved” 1851-2010 results and the presumably non-improved 1901-2000 results on the same chart and I only wanted to show the improved results (after all, that was the paper was about), I would erase the 1901-2000 results—oh yeah, I did do that! Had I asked Nathan Gillett for the data that went into that chart, I seriously doubt that he would have provided it with restrictions as to how I had to plot it. If Gillett et al. had only used 1901-2000 data, the paper probably would have never been produced or accepted—one of the major novelties of the research was using the full temperature record. Gillett et al. considered the results using 1851-2010 an improvement over previous work using 1900-1999 temperatures. So why oh why oh why is everyone so fired up over us focusing on the new improved results—those that were clearly preferred by the authors themselves?
If you all want to dredge up the Congressional testimony from 1998—then there seems to be a bit more room to argue as to which Scenario the original author preferred for the 1988-1997 period (of course, this has already been argued to death)—but in the case of Gillett et al. it is pretty darn clear the preferences of the authors.
Honestly, I am at a loss to understand the outrage over our handling of Gillett et al. (or Schmittner et al. for that matter—another case where we presented concentrated on the main results of the paper).
Let me instead go meta; this is an example of why policy debate is so frustrating, and why I’m not polite. The key thing about how the conservative movement handles debate is that it never gives up an argument, no matter how often and how thoroughly it has been refuted. Oh, there will be more sophisticated arguments made too; but the zombie lies will be rolled out again and again, with little or no pushback from the “respectable” wing of the movement.
In comments and elsewhere I fairly often encounter the pearl-clutchers, who want to know why I can’t politely disagree, since we’re all arguing in good faith, right? Wrong.
Posts
28 comments:
Eh, of the various Michaels & Chip tricks, I don't find this one to be so egregious. I mean, yes, they clearly wanted to delete the 1900-2000 because it didn't help their case, but it is the sort of manipulation which can also be a legitimate communication tool (eg, simplifying graphs).
On the other hand, claiming "Which means that results of Gillet et al. are in direct accordance with the results of Michaels et al. published 10 years prior" or that Michaels 2002 had "similar results" is a worse sin, in my opinion, when Michaels et al. 2002 focused on spurious black carbon arguments (but net aerosol forcing is better constrained than the forcing of individual components, so an increased BC estimate is often paired with an increased SO2 estimate), often-rebutted Lindzen iris arguments, some random climate feedback arguments, an argument that CO2 concentration should only increase at 1.5 ppm per year ("there is no indication that the current trend is breaking down" - promptly disproved by the next 10 years of CO2 concentration data), and, finally, the closest parallel to Gillett, which basically tries to claim that warming should be linear and so the trend of the past 25 years can just be extrapolated to 2100 (this being effectively the same assumption they made for CO2 concentrations, and almost as poorly justified).
-MMM
My education says that taking a graph like this without permission is a breach of copyright law pure and simple.
In Universities the copyright officer has to continually beat academics over the head to convince them that the law is the what the law says it is and NOT what they would prefer it to be.
Seems Chip has the same syndrome.
The climate ferret.
The dumbest thing about this whole kerfuffle is that even the "adapted" Gillett graph throws a stick into the spokes of Michaels & Knappenberger's research, if accepted at face value. Their favored value of sensitivity agrees with Lindzen (see Michaels et. al. 2002), which is roughly 1° C for CO2 doubling. Yet the adapted graph consistently shows results of about 2° for CO2 doubling. So if you believe Gillett, and apparently M&C do, then M&C have been getting it horribly wrong for years.
I bow to your wisdom, Chip: you were wrong.
Eli, thanks.
Krugman hit the nail on the head.
Albatross
You've gotta love Paul Krugman.
Just to attempt a pop-up comment:
Paul Krugman usually is incisively correct!
Even if we ascribe the most simon pure of motives to the Pat and Mike Show, the editing of a figure without approval of the figure's author is at the very least unwise.
There is no way to ensure that the new figure retains a representative subset of the information conveyed in the old figure. At an absolute minimum, one should obtain the data from which the old figure was developed, perform a documented meta-analysis and then, as a courtesy and to protect one's own tuckus, pass the new figure by the original author for comment (though not necessarily approval).
I do not think the current episode sufficient to infer dishonesty, but that can't be ruled out as inconsistent with the facts.
Climate ferret, if we watch the silly policy "debates" playing out in the US right now, it does seem that the law is whatever you want it to be, as long as you're rich and powerful or you have rich and powerful backers. Plus, the law doesn't matter anyway because ooh look, JOBS JOBS JOBS!!!!!!!!
As for this particular case? Unfortunately I think Chip's use of the graph might well fall under "fair use". Granted it's dishonest, but it may still be legal (under a sane interpretation of the law, at least).
-- frank
Yep, Krugman nailed it. As I read Krugman, many diverse examples merged in my mind and distilled into a single paradigm.
Thanks Eli for presenting this.
Jason Miller
(A Southern Cottontail)
Should we be more concerned by Chip's willingness to misrepresent another's work, or by his seeming inability to understand that this verges on scientific misconduct?
The former we can dismiss as questionable, while the latter raises the question of whether he is really capable of doing science?
"The key thing about how the conservative movement handles debate is that it never gives up an argument, no matter how often and how thoroughly it has been refuted. "
Among those on the losing side at the Diet of Worms this became known as the Doctrine of Infallibility. Five centuries later it was taken up by Dominionists, who rechristened it The Heartland Institutel.
W, Churchvole
frank -- "I think Chip's use of the graph might well fall under "fair use""
Not so sure about that. Modification and altering usually needs to have permission sought for. The original authors also have exclusive rights to create derivatives of the work. If the work is no longer recognisable after modification then that can be seen as fair use, which I don't think applies here, somehow.
As a blogger who regularly uses graphics from science papers I have to say that there is only one way to proceed when you want to draw a conclusion from part of a graphic.
You post the whole graphic, explain in the text what part of it you're talking about, and what conclusion you draw. Then leave it to your readers to assess your conclusion on it's own merits with the benefit of all of the evidence.
This principle extends to the use of papers: I always try to locate a paywall free copy of the paper in question so readers can go to source and see if I'm BSing them.
Snipping, graphics or quotes, runs the risk of being seen as dishonest, so should not be done at any time. When it's done it should ring alarm bells. I've not the time to pursue this issue to see the impact on the issue at hand - but my BS alarm is ringing.
Since the original M&M paper attempting to rebut Mann, there's been a lot of argument by innuendo. That paper had the famous graph which had spit all to do with the minor statistical flaws that M&M based their paper on.
IIRC, M&M also responded with wounded innocence.
> I think Chip's use of the graph might well fall under "fair use"
I'm pretty sure that it does. Copying for the purpose of commenting is fine. Even annotating the graph with big red arrows or whatever is fine. Wiping out part without telling -- hmmm. Hardly a legal problem though.
But then copyright law was never meant to enforce intellectual honesty.
KAP, to call what Chip et al. do "research" is IMHO a distortion of the term.
Their entire stock in trade is just that: Distortion (of research by others). A few years back I visited WCR and read through the most recent ten or so articles, looking for the porky within. Sure enough, everyone single one of them contained a major intentional distortion. It's what they do. Even more to the point, it's what they're paid to do.
But Chip is unfailingly polite!
You have to remember he does _advocacy_.
"the natural conclusion to be drawn from the omission of a fact is that the fact did not occur."
Modern trial advocacy: analysis and practice -- By Steven Lubet
Dr. Jay Cadbury, phd.
What a credible source to site, Paul Krugman. If you want to know how to destroy an economy, this is the guy. This is the same moron that kept writing columns about how we needed to spend more money on bailouts! Someone here please tell me, how much more money does Warren Buffet need? We just gave him $95 billion, apparently that isn't enough.
I know you guys love spreading the wealth around but this guy has no clue how to do it.
Dr. Jay Cadbury, phd.
Also again, I call on Eli to better explain the graph he claims was manipulated, I've looked at and can't see the difference.
Jay, Jay, Jay,
By now you have had a chance to review the description of this hack job at Skeptical Science.
It could be said to belong to the category of mendacious crap as described by Edward Carr.
Dr. Jay Cadbury, phd.
@Lionel A
your link is the first time I have viewed the supposed misrepresentation at SkS. Now here is what I need to know, what is the temperature change in the original graph? Why did Mr. Cook not post it, so we could compare it with Michaels 0.45?
Jay, Jay, Jay,
Now I am sure if you read carefully through the articles and responses on that article at Skeptical Science and Eli's coverage in Chip Clipps you should be able to answer your own questions. Most everybody else has grasped the implications here.
You people never like indicating specifics do you as demonstrated by the ambiguity in 'original graph' and the enigma of 'Michaels 0.45'
"What a credible source to site, Paul Krugman"
What a credible person to challenge someone else's credibility ... not. You're a stupid know-nothing, Jay.
"Also again, I call on Eli to better explain the graph he claims was manipulated, I've looked at and can't see the difference. "
That's because you're blind as well as stupid, Jay.
@Lionel A
"You people never like indicating specifics do you as demonstrated by the ambiguity in 'original graph' and the enigma of 'Michaels 0.45"
What! Lionel, I'm trying to be specific here. Okay, so by original graph, I meant Hansen's graphic. All I am asking is why Cook didn't post a temperature change on it like he did with Michael's graph.
My suspicion as to why Cook didn't post it is because the value is almost identical and Cook and Eli are making a big deal about nothing.
@mk
okay so you think we should give Warren Buffet more money I suppose. The question is, how much more money should we give him, mk? Please be specific. I bet most people here would disagree with you.
Jay,
Imagine a blockquote here
'I'm trying to be specific here. Okay, so by original graph, I meant Hansen's graphic.'
Now that was not clear from the context now was it? There were all manner of graphs (and even graphics) amongst all relevant posts. So is it a graph or a graphic and in what.
But then you go on to to muddy the waters further:
Imagine a another blockquote here
'All I am asking is why Cook didn't post a temperature change on it like he did with Michael's graph.'
You are still not making any sense.
Have you really read through all pertinent articles and comments?
Whatever you seem to be trying to pick nits so as to miss the main point. Which is .....?
All together now!
'Dr' Jay apparently thinks Cook put the "0.45" on the graph not Michaels.
'Dr' Jay should work to improve his reading comprehension skills.
Anonyspilospylla
(verification hermotl - is Mrs Lubos involved?)
Dr. Jay Cadbury, phd.
Okay just forget the whole thing, you guys are right and I'm wrong. Although I will point out to Lionel A, the first graph in his link does say "ORIGINAL" right above it, so I don't see how my labeling it with thet exact same term created confusion. Either way, neither of you can point out any difference between the graphs. Instead, you keep telling me I don't get it, while providing 0 explanation.
Jay wrote,
'Although I will point out to Lionel A, the first graph in his link does say "ORIGINAL" right above it...'
Ah! Now you are being specific but when you wrote, and I quote:
'...what is the temperature change in the original graph?'
you were anything but specific as is clear to see. You may have known what you meant but anybody reading that question would not have been sure as to which original graph you were referring to.
phd in ambiguity, perhaps?
Post a Comment