Arguments
Recursive Fury: Facts and misrepresentations
Posted on 21 March 2013 by John Cook, Stephan Lewandowsky
Our paper Recursive fury: conspiracist ideation in the blogosphere in response to research on conspiracist ideation has been published. The paper analyzed the public discourse in response to an earlier article by Lewandowsky, Oberauer, and Gignac (LOG12 for short from here on), which has led to some discussion on this blog earlier.
Refreshingly, the journal Frontiers makes all papers available for free with no paywall. Another unique feature of this journal is that readers can post comments directly beneath the abstract. Unfortunately this has led to the posting of a number of misrepresentations of the paper.
In this post, I’ll be addressing some of these misconceptions (but being careful to practise what I preach, will adopt the principles of the Debunking Handbook when I debunk the misconceptions). So here are some key facts about the Recursive Fury paper:
Conspiracy theorists are those who display the characteristics of conspiracy ideation
Yep, just stating the obvious, right? Recursive Fury establishes, from the peer-reviewed literature, the traits of conspiracist ideation, which is the technical term for a cognitive style commonly known as “conspiratorial thinking”. Our paper featured 6 criteria for conspiratorial thinking:
- Nefarious Intent: Assuming that the presumed conspirators have nefarious intentions. For example, if person X assumes that blogger Y colluded with the New York Times to publish a paper damaging to X, then X presumes nefarious intent on the part of Y.
- Persecuted Victim: Self-identifying as the victim of an organised persecution.
- Nihilistic Skepticism: Refusing to believe anything that doesn’t fit into the conspiracy theory. Note that “conspiracy theory” here is a fairly broad term and need not involve a global conspiracy (e.g., that NASA faked the moon landing) but can refer to small-scale events and hypotheses.
- Nothing occurs by Accident: Weaving any small random event into the conspiracy narrative.
- Something Must be Wrong: Switching liberally between different, even contradictory conspiracy theories that have in common only the presumption that there is something wrong in the official account by the alleged conspirators. Thus, people may simultaneously believe that Princess Diana faked her own death and that she was assassinated by MI5.
- Self-Sealing reasoning: Interpreting any evidence against the conspiracy as evidence for the conspiracy. For example, when climate scientists are exonerated of any wrong-doing 9 times over by different investigations, this is reinterpreted to imply that the climate-change conspiracy involves not just the world’s climate scientists but also the investigating bodies and associated governments.
We then went on to identify responses to LOG12 that exhibited these criteria. Our analysis was entirely based on whether or not public statements conformed to the criteria just listed—we made no comment on the merit of any criticism (except in cases where speculations were plain wrong).
A common misrepresentation of Recursive Fury is articulated by one commenter who says “conspiratorial ideation is defined in such a way that any criticism of LOG12, whether true or false, comes under that heading.” Actually, our criteria for conspiracist ideation come from a number of peer-reviewed examinations of conspiratorial thinking and have nothing to do with the substance of any criticism of LOG12. Our objective in Recursive Fury was to demonstrate that some of those criteria arguably applied to the public discourse surrounding LOG12. It does not follow that any criticism of LOG12 involves conspiratorial thinking. Of course not. But if some (not all) critics of a paper on the role of conspiratorial thinking in science denial engage in, well, conspiratorial thinking in response, that’s of scholarly interest.
The criteria for conspiracist ideation are applicable without regard to a statement’s truth or falsity. Recursive Fury is not about defending LOG12. On the contrary, this latest paper puts on the scholarly record many criticisms of LOG12 that had previously been limited to blogs, and it did so without evaluating or rebutting the substance of those criticisms. Some defence!
A few critics have complained that we didn’t include their methodological critiques of LOG12. Such critiques do not fit the conspiracist criteria, which is why they weren’t included. Those critics are welcome to submit rejoinders or comments on LOG12 to the journal in question.
A range of different conspiracy theories are posted in Recursive Fury
Recursive Fury reports and analyzes a number of conspiracy theories regarding LOG12. These range from “global climate activist operation” to “ringleader for conspiratorial activities by the green climate bloggers,” to Stephan Lewandowsky receiving millions of dollars to run The Conversation.
Some folk are able to overlook these many documented instances and insist that “There is no ‘conspiracy’ Mr. Lewandowsky - no matter how many times you try to manufacture one.” Recursive Fury documents a whole spectrum of conspiracy theories. As you get further into the paper, the conspiracy theories become broader and more extreme until you get to my personal favourite – maths professor Kevin Judd being the grand poobah of the “global climate activist operation” at the University of Western Australia. Somehow, those who insist "there are no conspiracies" manage to skip over entire sections of the paper.
It appears that “conspiracy denial” may be another phenomenon associated with climate denial. One blogger cannot see that his claim that climate scientists “colluded with government officials to avoid the law” is conspiratorial. Similarly, another blogger thinks accusing the University of Western Australia of being “a base for this global climate activism operation” is not a conspiratorial hypothesis because he didn’t use the word “conspiracy”.
The Supplementary Material is “raw data”
As well as the Recursive Fury paper, we also published Supplementary Material containing excerpts from blog posts and some comments relevant to the various observed recursive theories. In the paper, we characterise this as “raw data” - all the comments that we encountered that are relevant to the different theories. In contrast, the “processed data” are the excerpted quotes featured in the final paper, where we match the various recursive theories to the conspiracist criteria outlined above.
One misrepresentation of Recursive Fury is that we accuse Professor Richard Betts of the Met Office of being a conspiracy theorist because one of his quotes appears in our raw data. This inclusion of a relevant comment in the raw data of a Supplementary Material document was reported in hyperventilating fashion by one blogger as a spectacular carcrash. However, there is no mention of Professor Betts in our final paper and we are certainly not claiming that he is a conspiracy theorist. To claim otherwise is to ignore what we say about the online supplement in the paper itself. The presence of the comment in the supplementary material just attests to the thoroughness of our daily Google search.
Nevertheless, I can see how this misunderstanding arose. The Supplementary Material features the heading "Excerpt Espousing Conspiracy Theory" referring to the excerpted quotes that we pasted into the spreadsheet. In hindsight, the heading should have been "Excerpt relevant to a recursive theory", because the criterion for inclusion was simply whether or not they referred to one of the hypotheses. The analysis of conspiracist ideation occurred after that, and involved the criteria mentioned at the outset.
In this context, it is important to point out that one reason we made the raw data available is for other scholars to be able to cast an alternative interpretative light on the public discourse relating to LOG12. As we note explicitly in the abstract, it is possible that alternative scholarly interpretations can be put forward, and the peer-reviewed literature is the appropriate forum for such analysis.
LOG12 is in press
The original “Moon Landing” paper (referred to as LOG12) is still in press and due to be published soon. The fact that there was a long delay between acceptance and publication is one of the quirks of the peer-review publication process. Sometimes a paper can move from acceptance to publication with surprising speed (as was the case with Recursive Fury). Sometimes it can take months.
However, this random timing has been over-interpreted by many parties, consistent with the “Nothing occurs by Accident” criteria. For example, one commenter argues that “LOG12 was fundemenatlly [sic] flawed from the start, and throughout. It offered no valuable insight or understanding as a result. It is clear to any rational outside observer it had one purpose - to be used to promote the authors advocacy of catastrophic anthropogenic global warming - and to demean and denigrate those who do not believe as he does. The fact this paper has never been published, as Lewandowsky's repeatedly claims, confirms this finding.” It will be interesting to see whether this commenter resists the “Something Must Be Wrong” urge when LOG12 is published or continue to assert that the research is “a fraud”.
Conclusion
Hindsight is always 20:20 but perhaps we should have anticipated the response to LOG12. The results of LOG12 implied that conspiratorial thinking is linked to climate denial, and hence might emerge in turn to defend climate denial against cognitive analysis – and that’s what happened, as we document in Recursive Fury.
Note: this post was cross-posted on Shaping Tomorrow's World.
Tom Curtis comment 90
You misundertand my position, and therefore misrepresent it. I’d ignore it because it’s not very interesting for others, except that you seem to be asking the moderators to snip me; so I’d better defend myself.
No. I don’t deliberately contradict myself. I have consistently expressed the idea that anonymous on-line surveys like the one on which LOG12 is based are inherently useless for any enquiry more complicated than “do you like/dislike this article?” If you want to find out what someone thinks, believes or feels, the best way is to chat to them. A street interview is a second best, a telephone interview a poor third, and so on.
My reactions to the conspiracy questions are much like many of the readers of the blogs where the survey was publicised. “Don’t know about that”, “Perhaps yes, perhaps not” and so on. I might very well, give different responses on different occasions, just as I might give different answers as to who I’d vote for or my favourite colour. I’ve sometimes expressed that jokingly in comments about Prince Philip killing Lady Di, etc. I tried to make the point seriously once at Climate Audit, that Steve McIntyre was not justified in removing two “scammed” responses. You can’t arbitrarily decide that liars or cunning bastards should be removed from a survey - it takes all sorts. It would be like doing a survey about petty crime and eliminating a respondent because he nicked your tape recorder.
You will note that in my comment 38 replying to Albatross, I don’t deny the reality of the theory of anthropogenic climate change or anthropogenic global warming. I’m therefore a warmist, and not a denier, according to the criteria of the questionnaire in LOG12. The fact that I disagree fervently with everything that Skeptical Science stands for is not a reflection on my state of mind, but on the complexity of a political programme designed to subvert science in the interest of an ideology. You’re not going to get to the bottom of that by asking if people agree or disagree with anything.
geoffchambers says:
But he's not a conspiracy theorist?
Spot on, Sphaerica.
I can attest to the fact that SkS consists on a very wide ideological spectrum of people. In fact, I find that to be true of the broader community of people trying to actively communicate the issues with AGW. I personally find it refreshing that there is such a broad ideological support behind the science of climate change. I don't believe you could say the same about those who reject man-made climate change.
Geoff: ...anonymous on-line surveys like the one on which LOG12 is based are inherently useless...
Just so we don't go too far off the path of facts, it's worth noting that Geoff's remark is a guess and as well is wrong, while in reality the authors of LOG12 used methods shown to work in work performed by other researchers. Indeed, the entire paper was an extension of previous research, employing accepted research tools and with findings paralleling earlier results in terms of connections between ideology and acceptance of particular hazards and risks. The replication of other previous findings by other workers embedded in LOG12 speaks volumes about the validity of the research methods employed in producing LOG12.
If Geoff had by some remarkable circumstance been correct in his guess, we'd then be witnessing the birth of a revolution in social science research, the upheaval of a train of thought extending decades back in time, performed by a vast number of individuals.
@Geoff Chambers and Tom Curtis:
Please stay on topic and stop going off on tangents.
Sphaerica at 02:29 AM on 27 March, 2013
'nuff said.
Phronesis - Given McIntyres propensity for basic math errors, I have been entirely unimpressed by any of his various criticisms of LOG12.
Until, however, you link to something in particular by McIntyre (there are at least 16 different McIntyre blog posts on the subject, but quite notably nothing peer-reviewed), you haven't presented any actual arguments in that regard.
KR, are you saying there are math errors in his rebuttal of LOG12? Or in some other, climate-related work? If the latter, then your conclusion doesn't follow from your premise. Well, it might be rational to not read his arguments if you thought he was too error-prone, but it wouldn't follow to be unimpressed by his arguments based on math errors from some other work. Bias is a pervasive human challenge, so any anti-skeptic or anti-McIntyre scholar would want to go into this issue with some bias-correction algorithms running, otherwise you won't be able to get a clean take on whether he's right or not.
You seem to know where his posts are already (there are 16?), so I'm not sure what good it would do to link you to them. Maybe I can highlight his key arguments sometime this week. LOG12 should never have been published, and I'm surprised it hasn't been retracted at this point -- either by the authors or by Psych Science.
Phronesis - The majority of the Climate Audit posts on LOG12 that I have read (or rather, endured reading, as they are rather nasty) consist of speculative slander about Lewandowskys motives (for example, posts entitled "Anatomy of the Lewandowsky Scam", "Lewandowsky’s Fake Results", or lines like "... Lewandowsky’s tainted methodology – a methodology that relied on fake data to yield fake results"), claims that the data was trashed by scammed responses (no evidence thereof, mind you), and various conspiracy theories regarding how the surveys were distributed.
In his Trying (Unsuccessfully) to Replicate Lewandowsky post, McIntyre makes several errors that are apparently due to his unfamiliarity with exploratory factor analysis (EFA). These include inappropriate eigenvalue selection (using two factors instead of the one significant eigenvector) and use of the default 'R' language rotation that redistributes variance - useful in PCA, but inappropriate in EFA when you want to attribute those components back to the survey questions. This is discussed at some length by Oberauer and Lewandosky, laying out McIntyre's errors.
McIntyre has a history of poor or missing evaluation of principal component significance, as discussed on RealClimate with respect to the Mann et al papers. That's the original topic where you brought up McIntyre's LOG12 discussion, I'll point out.
I would consider this diversion an Argument from Authority on your part to support McIntyre's claims about the "Hockey Stick", and therefore a red herring WRT that discussion. Expertise in one field doesn't support an argument in another, it's a logical fallacy, and it seems quite clear to me that McIntyre has not demonstrated any expertise in EFA either.
Regardless of this side-track, McIntyre's claims about Mann et al have been solidly refuted, most clearly by Wahl and Ammann 2007.
The 'skeptics' have subscribed to, and serially rejected, so many alternative explanations of global warming that their rejection of the flat earth hypothesis ranks as a statistical outlier.
It is unfortunate that Frontier did not conduct enough legal diligence to realize the wannabe plaintiffs would be laughed out of court.
What I have always found fascinating about the "conspiracy of scientists" theorists is the more likely conspiracy they ignore. Getting scientists to agree on things has a lot in common with herding cats, except that the cats are a lot easier to get moving in the same direction and probably better paid... whereas the financial wizards and CEOs funding entities like CEI and Heartland whence the lies about science emanate, the people who control the WSJ and Forbes where they are most commonly published... THOSE people have given us the LIBOR fraud, the FOREX scandal, the Global Economic Crisis, Billion Dollar Bailouts, "Too-Big-To-Jail", and a list of further scandal and corruption far far longer. They are the professional conspiracists and they pay themselves rather well. As someone pointed out, if scientists were in this to make money they're doing it wrong.
Yet that never quite touches their ideation, and that itself is a bit of a puzzle because as a conpiracy theory goes it has a lot more potential. So why doesn't it take off? Why isn't it a MORE common theory than the one that is espoused about climate scientists? ...or is it merely less vocal even though it is more widely held?
There's a semi-relevant question in this because more than one conspiracy theory is available to consider. How is the conspiracy theory chosen by the theorist, why is one more "popular" than the another?
Not going to go into the debate with our visitors.
The difference between the consensus of the climate science and the consensus of the scientist's opinions, is an important one to make, and is missed by many. When one examines the papers to find 97% one is examining the consensus of the science, not the people. It makes the absense of contrary science far far more telling.
Famously (or perhaps infamously), I have disagreed with LOG12 as to whether their results followed from their research. Because my comments on LOG12 have been frequently quoted out of context, and misrepresented, I have felt the need to comment on this matter at a number of sites where I would not normally comment (which need not be a reflection on the site, just my limitted time), and have seen the persistent and viturapetive attacks on Lewandowski first hand. Today I came across the perfect description of that ongoing attack on Lewandowdki. It definitely falls into the "I wish that I said that" category. Here it is:
(Mike Hansen, ATTP, October 23, 2014 at 10:22 am, my emphasis.)