As I write this the fuss about our Frontiers article, “What about Building 7?” A social psychological study of online discussion of 9/11 conspiracy theories, has mostly died down, so now seems like a good time to do a bit of a postmortem and take a look at some of the issues that have come up since the whole ruckus kicked off with Kevin Barrett’s awful PressTV article and its infinite copy-pastes around the blogspamosphere.
I’ve been contacted for a couple of interviews since this whole business started: a Google Hangouts interview with the Renegade Variety Hour, and an audio podcast interview with Greg Moffitt of Legalise Freedom. These two shows come from quite different perspectives – Renegade Variety Hour was very conventionalist while Legalise Freedom had more of a conspiracist slant to it – so it was an interesting experience and I was glad to have the opportunity to correct some of the misconceptions that have been circulating around. I was also invited to be on Coast to Coast AM, but that got pushed back so it might not happen. We’ll see.
On the whole it’s been a pretty interesting two weeks. There’s a bit more to talk about, including a new counter-rebuttal from Barrett, below the fold.
So in response to my earlier post trying to clear up the misconceptions about our article , Barrett recently wrote a counter-reply that briefly appeared on PressTV before being taken down for some reason and reposted on Veterans Today a day later. In it he makes a few main points:
- The findings really do show that conventionalists are more hostile and less open-minded than conspiracists, so what’s the problem?
- I’m obviously trying to backtrack because I’m afraid of being blackballed by the academy
- I have no sense of humour about this whole business and should be grateful for the publicity, and he was being ironic anyway, geez, calm down buddy
I’ll just deal with these points very briefly because there’s more important stuff to get to. First, I talk about the hostility finding a few paragraphs down – briefly, the comments were more hostile, but there’s reason to believe this doesn’t generalise to the commenters themselves, and there’s actually some evidence the other way on this one. Second, I didn’t mention open-mindedness anywhere in the paper; that’s Barrett’s own interpretation of a finding that could equally be taken a different way. If these were the only misinterpretations that Barrett had made it’d be no big deal and I probably wouldn’t have even taken the time to reply. The issue was the “conspiracists are more sane” claim, which has no basis in the research and which is conspicuously missing from the new article. I guess that was a joke? Barrett’s readers took it pretty seriously judging from some of the emails Karen and I got, so as a joke it seems a bit tone-deaf.
Third, well, at least he’s not claiming I got a 3AM call from George Soros demanding that I go back on my findings. There’s no profit motive here, because any future employers would probably not take their cues from smear job articles on Iranian state media. I just don’t like people abusing my work to spread self-serving misinformation.
Finally I’m not interested in publicity if it means the science is misrepresented. (I would like to thank Barrett for the promotion to Professor, though. Does it come with a dental plan?)
Hostility and “Sanity”
Anyway, on to more substantive matters. The idea that the study demonstrates that conspiracists are more “sane” than conventionalists (which some people are still pushing, even though Barrett himself seems to have abandoned it) essentially rests on four major points: that conspiracist comments outnumbered conventionalist comments in our sample; that conspiracists are more likely to take historical context into account; that conventionalists are “fanatically attached” to their own interpretations; and that conventionalists tend to be more hostile. The first point is a non-sequitur and ignores consistent trends in public opinion polls (some of which we cited in the paper) showing that 9/11 conspiracism remains a minority view; the second point is simply false; and the third is baseless self-serving speculation without anything to back it up in the research. This was all spelled out in the previous post so I’m not going to recapitulate it here.
The one thing Barrett got mostly right, and which he’s still going on about, was the hostility finding: we did find that conventionalist comments were more hostile on average than conspiracist comments. Some people have looked at this finding and concluded that it constitutes evidence that conspiracists really are more mentally well-adjusted than conventionalists, that the rest of Barrett’s misrepresentations don’t really matter, and, for good measure, that They Got To Me (if you look closely, during the Renegade Variety Hour interview you can see that I’m blinking out SOS in Morse Code with my eyes).
The problem with this interpretation lies in extrapolating from “conventionalists are more hostile in these sorts of discussions” to “conventionalists are generally more hostile” and “hostile people are totally cray-cray.” I think a lot of this is a case of people either not reading the paper or just scanning through it to find the bits that support their argument. First and foremost, the explanation we present in the paper for conventionalists being more hostile is that they’re essentially trying to enforce conformity to the majority view – it’s a prediction derived from social influence theory. Essentially, when you have a majority and a minority with conflicting views on some subject, the minority has to be united, civil, and informative in order to effect opinion change in the majority. The majority, on the other hand, has a bit more leeway and doesn’t have to rely on the same tactics in order to convince the minority to abandon its position. People holding the majority position being somewhat more hostile on average in discussions with the minority makes perfect sense in this context – it’s both unnecessary and scientifically irresponsible to conclude on the basis of this kind of discussion that one group is more well-adjusted than the other based entirely on behaviour in this one particular context, because the situation is asymmetrical. In other words, there’s no basis to conclude that the difference in hostility is due to something about the person rather than the situation.
Secondly, there’s been some research showing that interpersonal hostility on a dispositional level is positively correlated with attitudes toward the existence of conspiracies. Abalakina-Paap, Stephan, Craig, and Gregory (1999), in a questionnaire study using a well-validated measure of traitlike hostility, showed that the more hostile someone is, the more positive their attitude toward the existence of conspiracies. This difference even persisted when taking into account a range of other relevant psychological variables such as trust, locus of control, and authoritarianism. In other words, when you measure the hostility of people (as a relatively stable personality trait) rather than comments (as captured in the context of an asymmetrical argument in a news website comments section), you get the opposite result. This doesn’t contradict our finding, though, at least not directly, because the two studies asked different questions. Abalakina-Paap et al. looked at the question of whether conspiracism is associated with hostility as a personality trait, while we examined the rhetoric used by conspiracists and conventionalists within a particular, highly specific social situation.
Before anyone jumps to the opposite conclusion to the one put forward by Barrett, though, there’s another problem: hostility is a completely bogus proxy for “sanity” anyway. The idea that “more hostile = less sane” is fraught with unjustified assumptions: in order for this to hold, one of the following must be true. Either:
- Hostility is always a sign of mental instability, such that more hostile people are always “less sane;” or
- There’s an acceptably “sane” range of hostility outside of which conventionalists fall, or some kind of sanity/hostility curve with a downward slope at the point of difference between the two groups.
I don’t know anyone who would argue the first point explicitly – most people would probably agree that some baseline level of hostility or aggression or confrontation is necessary to function properly in society, much less in persuasive debate. The second simply has no evidence for it, certainly not from this study.
So why are people defending Barrett’s article on the basis of the hostility result? I can’t fault anyone for not knowing about the Abalakina-Paap study, but surely if they were that invested in the hostility finding they could have read the paper and seen the bit about social influence or at least thought through the “more hostile = crazy” reasoning. The fact is, people just like to see themselves complimented and their enemies insulted, and they’ll do whatever rhetorical backflips they need to hang onto that even if the reasoning turns out to have been wrong all along – and that’s pretty much true no matter what they think about 9/11.
By the way, you – yes you – you look really nice today. I know you’re maybe not so confident about your hair right at the moment, but it totally suits you. And that shirt? Daaaamn. Looking good.