Thursday, September 06, 2007

Solomon, Greenberg and Pyszczynski on

"Worldview Defense"

By John J. Ray (M.A.; Ph.D.) -- September, 2007

I have had a cursory look at the "Worldview Defense" research and found much to amuse in it. I don't rule out there being some useful kernel of truth in it but to firm that up one way or another I would have to do a detailed critique of at least some of the papers basic to it -- which is not an inviting prospect. I have decided therefore simply to gather together below my blog posts on the subject


A critique of research by Sheldon Solomon, Jeff Greenberg, and Tom Pyszczynski

That old Leftist, John B. Judis, has an article in The New Republic that summarizes a stream of psychological research into fear of death that goes by the name of "worldview defense". The idea is that if you are reminded of your own mortality, you become more conservative.

In one way, that is all fair enough. The old saying "A conservative is a liberal who was mugged last night" embodies a similar idea and represents a claim that conservatives often make: That they are more realistic and that Leftists are dreamers who need to be brought down to earth. Being vividly reminded of your own forthcoming death (which is what the psychological experiments concerned do) should invoke a similar burst of realism and disable dreamy views of life.

So an interpretations of the findings congenial to conservatives is more than possible. Judis and those he quotes, however, strain to find a more elaborate interpretation that is some way derisory of conservatives. But in doing so Judis falls into a trap common among psychologists and other Leftists: He lives in such a self-protective Leftist bubble that he basically just does not know what conservatism is or what conservatives think. For decades now, psychologists have been devising questionnaires that allegedly "measure" conservatism but which in fact give no prediction of vote at all! The ludicrous Bob Altemeyer is the most recent example of that. The highpoint of such ignorance, however, would have to be the 2003 "Berkeley" study which classified various Communist leaders as conservatives. That Communists and conservatives have radically different views about the world had apparently not penetrated the ivory towers of UCB!

So we have the following remarkable comment from Judis: Also central to worldview defense is the protection of tradition against social experimentation, of community values against individual prerogatives". And you thought it was conservatives who stood for individual liberties! Not so, according to Judis. Conservatives stand for "community values". So Hillary Clinton, with her quotation of an old African saying that "It takes a village to raise a child" must be a conservative!

So I don't think we really need to say much more about such profound ignorance. As this stuff falls squarely within my own field of professional exspertise, however, I will make one more comment: Answers to questions that are obtained from young college students (which is mainly what Judis is referring to) often tell you very little about the real world. The very first piece of psychological research that I ever did was based on responses from students and I found a most gratifying correlation of .808 between the two variables concerned. Being a born skeptic, however, I then did something that psychologists almost never do: I repeated the research among a group much more representative of the general population. And I found NO correlation between my two variables on that group.

And so it seems also to be with the research by Pyszczynski and friends that Judis quotes. Using student responses, Pyszczynski et al. found a correlation between awareness of death and what they (in their confused way) define as conservatism but I carried out long ago a piece of research into much the same question. I looked at the correlation between attitude to death and conservatism among a general population sample. And I used a measure of conservatism that DID closely reflect the political divisions of the day. So what did I find? I found that there was NO correlation between attititude to death and conservatism whatsoever. Nor was there any connection between anxiety generally and conservatism. So the Pyszczynski/Judis claims fail a more rigorous test. What they think happens, does NOT happen in the real world.

And I carried out that piece of research in collaboration with the head of our local Sociology department -- an impeccably mainstream Jewish Leftist. So the ad hominem attacks that one expects from Leftists would be more than usually implausible in this case.



I referred recently to the Solomon, Greenberg & Pyszczynski research summarized by Judis. The interpretation of the research findings concerned was highly speculative and unparsimonious in that the effect of priming people to think about death was somehow magnified into telling us something about "worldview defense". The point of the theoretical extravaganza was of course to portray conservatives in a bad light. I suggested a more straightforward interpretation of the findings which was rather supportive of a conservative view of the world.

In the end, however, I concluded that the research concerned really told us nothing at all about anything because it was not based on any kind of representative sampling. Solomon, Greenberg & Pyszczynski obviously disagree with me on that, however, so I have found another study based on their kind of "sampling" that should interest them. A study by Shariff & Norenzayan also looked at the effect of "priming" people's perceptions. Where the Solomon, Greenberg & Pyszczynski research showed that priming people to think about death caused them to become more conservative, however, the Shariff et al. research showed that priming people to think about God caused them to become more altruistic and kinder towards others! It tends to show that Christians are nicer people, in other words. I wonder what Pyszczynski and friends think about that? One thing they CANNOT consistently do is dismiss the finding on the grounds of unrepresentative sampling!


An amusing response from Pyszczynski

Who says that conservatism is produced by a fear of death

Pyszczynski did respond promptly to the criticism of his research that I posted yesterday. After the burst of personal abuse that one expects from Leftists he did have two actual points to make:

He said that the "priming" experiment that he does has now been repeated worldwide to a total of about 350 times and always gives a similar outcome. He seems to think that doing unrepresentative sampling 350 times is in some way as good as doing one really representative sample. No logic there at all, of course. So I replied simply that doing a silly thing 350 times does not make it right.

His second point concerned the "God" research that I mentioned. He commented quite reasonably that the outcome you get would depend on the concept of God involved. That the usual Christian concept of God as a God of Love might have been behind the finding I mentioned -- that thinking about God makes you more altruistic -- he did not address, however. He preferred abstract argument to research results. But it is consistent with his argument that the usual Christian view of God makes Christians better people. How embarrassing for him!

But as in the story by Conan Doyle ("Silver Blaze") about the dog that did not bark, the most interesting thing was what Pyszczynski did NOT mention. His theory claims that worrying about death is behind a lot of conservatism. But my research in a general population sample has shown that conservatives worry about death and worry in general no more than do others. So his theory is disconfirmed at the bottom line. No matter what groups of tame students do when confronted by Pyszczynski's little experiment, the conclusion from those experiments does NOT generalize to the population at large. So he is talking about something of no real-world significance. He had no reply to that whatever. I guess he didn't reply because he couldn't!

He also did not reply to my point that he was overinterpreting his data. In science the most parsimonious (simplest) explanation for a finding is always preferred and I offered an explanation for his basic finding that was a lot simpler than the fanciful edifice he has erected. I guess he couldn't reply to that either! He has devoted decades of work to his little theory, however, so I don't expect that mere evidence and logic will cause him to drop or even modify it. Jim Sidanius had the grace to go into a major backdown when I confronted him with evidence and logic about his "Social Dominance Orientation" theory but there is no sign of that from Pyszczynski -- at least so far.



Further to my post of yesterday, I reproduce below the latest email from Prof. Pyszczynski. As far as I can see the first and fourth sentences contradict one-another!

We have NEVER claimed that conservatives have any more fear of death than anyone. That would be completely at odds with terror management theory or the research that has been done to test hypotheses derived from it. This is something anyone who has read our work would know. The point is that conservative ideology is a type of worldview that people use to protect themselves against a fear of death that is a natural consequence of wanting to live and knowing that you must die. Just as liberal ideology is. Just as science and religion are. When people are reminded of death, they cling more to the aspects of their worldviews that protect them from this fear. Conservatives and liberals do it -- but birds and educated fleas do not. There is some evidence that conservative ideology might be especially useful in providing protection but the jury is out as to whether that is inherent in conservative ideology or particular to the particulars of today's culture.

The only sense I can make out of it is a claim that people become more extreme in their views (whatever those views may be) when they think about death. That is a fairly humdrum proposal but does not correspond to how other researchers see the Pyszczynski work. Note this quote about the Pyszczynski work, for instance:

"Terror management research has typically found that people respond harshly toward offending others when reminded of their mortality".

Clearly, Pyszczynski has been claiming that death anxiety moves you towards a SPECIFIC view, not simply a more extreme view. And among Leftist psychologists, punitiveness has long been held to be a feature of conservatism -- which is why I have previously written on that subject (I found that impunitive people are the oddballs!).

At any event, Pyszczynski has clearly repudiated the summary of his work by Judis -- which is rather fun. I am a bit disappointed that he didn't deflate under attack as rapidly as Gilbert Harman but Harman is a genuinely acute thinker.


"Truthers" and worldview defense

I had a bit of a laugh above about the "worldview defense" research of Pyszczynski and friends. I could see no good evidence for his claim that conservative views arise out of a need to defend a worldview.

The evidence Prof. P. offers for his claims concerns attitudes. What students say about their views is the foundation of his theory. I replied by pointing out (among other things) that proper general population survey research into conservative attitudes does not reveal the correlations to be expected from Prof. P.'s theory.

There is however abroad these days one very strange set of attitudes: The attitudes of the "truthers" -- people who believe that the dumb but strangely clever George Bush conspired with the neocons, the Jews and various others to blow up the twin towers on 9/11. Apparently the twin towers were deliberately detonated by GWB rather than being knocked out by bin Laden's henchmen flying hijacked airliners. Some of the claims of the truthers have a superficial plausibility but all have often been debunked (e.g. here).

The interesting thing about truther theory is that, for all its vast implausibility, it is widely believed. According to a general population poll carried out by Zogby 42.6% of Democrat voters believe some version of it versus 19.2% of Republican voters. Nearly half of Left-leaning voters say they believe in a total absurdity!

So why do they do that? A simple answer: It is BDS (Bush Derangement Syndrome). Bush is so hated by many on the Left that they cannot accept any evidence that might support his policies. Their worldview requires that Bush be (as they often say) akin to Hitler and the embodiment of all evil. That he might have had reasonable grounds for his overthrow of two aggressive Islamic regimes (in Afghansistan and Iraq) just cannot be accepted. Rather than accept that it is preferable to believe a total absurdity.

So if that is not worldview defense I would like to know what would be. It is worldview defense carried to the extent of an extreme mental pathology. So, as I have often previously remarked, Leftists are great projectors -- they attribute to others things that are really true of themselves. So to find out what is true of them, just listen to what they say about conservatives. Prof. P. was so alive to the phenomenon of worldview defense precisely because it fills the heads of many of his fellow Leftists -- and maybe even his own head.


More data on those "death-fearing" conservatives

I have received the following report from a reader. It gives us some more of that pesky general population data -- the sort of data that almost all psychologists avoid like the plague. Students are SO much better at giving responses that accord with Leftist stereotypes:

Reading John Ray's September 4, 2007 critique of the Pyszczynski et al. claim that fear of death is behind conservatism, I couldn't help but analyze data from the NORC General Social Survey, one of the most respected databases of U.S. public opinion in existence, to further explore the matter. In fact, the GSS data completely support Dr. Ray's conclusion that there are no meaningful differences whatsoever in how liberals and conservatives view death. Again, the received "wisdom" concerning the psychology of conservatism espoused by Pyszczynski and academic social science in general, is shown to be speculative and fails to survive even moderate scrutiny.


For decades, the GSS has asked respondents about their political orientations under the variable name POLVIEWS and whether they believe in life after death under the variable name POSTLIFE. I recoded the variable POLVIEWS into the following categories:

* Extremely liberal/Liberal =1,
* Slightly Liberal = 2,
* Moderate/Middle of the Road = 3,
* Slightly Conservative = 4, and
* and Conservative/Extremely Conservative = 5.

POSTLIFE was recoded into the following categories:

* Believe in life after death =1,
* Unsure = 2, and
* Don=t believe in life after death = 3.


Conservatives were much more likely than liberals to believe in life after death, with approximately 80% of conservatives/extreme conservatives reporting that they believe in life after death and only 66% of liberals/extremely liberals reporting the same. (All percentages are rounded upwards.) While a person's belief in life after death is not necessarily predictive of whether he fears death, the results are suggestive and weaken Pyszczynski's paradigm. The Pyszczynski et al theories would seem to imply that belief in life after death is a cognitive defensive mechanism but a sizable proportion of individuals with liberal orientations clearly exhibit the same phenomenon, so it is as least not a peculiarly conservative defence mechanism. And in any event, the number of people who absolutely reject the possibility of an afterlife is very small, only about 9% in this database.

To further explore whether conservatives might harbor more fear than liberals regarding death, I analyzed how respondents answered an additional set of questions asked between 1983-1987. Each respondent who reported a belief in life after death was asked a further series of questions, prefaced as follows:

Of course, no one knows exactly what life after death would be like, but here are some ideas people have had. How likely do you feel each possibility is? Very likely, somewhat likely, not too likely, or not likely at all?

Among the various alternatives were the following descriptions of an afterlife: (1) A life of peace and tranquility (POSTLF1) and (2) A life like the one here on earth, only better (POSTLF3). There was very little meaningful difference between liberals and conservatives regarding their views on the matter. With respect to the first variable, POSTLF1, eighty-five percent of liberals/extreme liberals thought it somewhat or very likely that an afterlife would be a life of peace and tranquility, while an even greater 95% of conservatives/extreme conservatives held similarly positive views of life after death. Regarding the second variable, POSTLF3, 57% of liberals/extreme liberals thought it somewhat or very likely that an afterlife would be better than life on earth, and 59% of conservatives/extreme conservatives reported it somewhat or very likely that an afterlife would be better than life on earth. Again, all percentages are rounded.

Insofar as there is a discernable difference between liberals and conservatives in how they view death, if anything conservatives might appear to face the prospect with less fear and anxiety than those who lean leftward politically. Note that similar results are obtained when comparing respondents who are slightly liberal versus those who are slightly conservative.

If in fact conservatives have a more positive view of existence after death than that held by liberals, perhaps it explains why they seem to be at least as happy and as satisfied with life as liberals. The GSS asks respondents, "Taken all together, how would you say things are these days: Would you say that you are very happy, pretty happy, or not too happy?" (HAPPY). Eighty-eight percent of liberals/extreme liberals reported being very happy or pretty happy, and approximately 90% of conservatives/extreme conservatives reported being very happy or pretty happy. Again, the same trend is found when comparing slightly liberal respondents with slightly conservative respondents.

And on the prospects of future generations, conservatives appear to be at least as optimistic, if not more so, than liberals. The GSS asks respondents whether they agree or disagree with the following statement: "It's hardly fair to bring a child into the world with the way things look for the future." Forty percent (40%) of liberals/extreme liberals and 34% of slightly liberal respondents agreed with the statement, while 36% of conservatives/extreme conservatives and 33% of slightly conservative respondents agreed with the statement.

These findings represent yet another piece of evidence that conservatives suffer from no greater existential anxiety or dread than do liberals. The claims of a relationship between psychological maladjustment and social/political conservatism, posited by classical authoritarianism theory, are simply inaccurate.


No comments: