"Gummo Trotsky" at "Tug Boat Potemkin", perhaps a bit ruffled by obsevations that conservatives are happier, takes results from a Gallup poll showing Republicans are happier than Democrats and combines them with observations by some guy named Adam Elga that "happy" people overestimate certain factors about themselves in order to conclude, if so, then surely:

Republicans are significantly more likely to be self-deluding than either Democrats or Independents.

And, further:

If Republicans are more likely than others to be self-deluding, it's reasonable to suppose that they are also more likely to hang on to beliefs that make them happy, even if the weight of evidence is against their beliefs.

(Could we honestly expect a different conclusion? Dissecting Leftism, who provided the link, notes: "I guess that, as a Leftist, he has to believe that.")

Sadly, while appearing to make a case for the comparative clear-headedness of liberals as compared to deluded, irrational conservatives, he neatly provides disconfirming evidence for his own thesis.

First problem: Gummo gets (deliberately?) confused about what Elga says, apparently reading Elga's piece as saying something akin to "happier people are more self-deluding", from which he concludes that Republicans, if happier, must also be "significantly" more self-deluding.

But that's not what Elga is saying. Elga says two things of note here. First, that normal, healthy (e.g. "non-depressed") people overestimate their abilities and prospects, and that clinicly depressed persons fall closer to the mark. And second, that overestimating one's prospects can make one happier ("There is evidence associating the above sorts of positive illusions with increased happiness").

Gummo's error here is to assume the converse: If overestimating one's lot in life makes one happier, all happiness must come from overestimations. Yeah, sure, and all animals are dogs. He cites no evidence whatsoever that as one's degree of happiness increases, self-overestimation also increases.

The second problem is that Gummo (deliberately?) conflates the terms of the two different pieces when he links them together. The survey uses the terms "Very happy", "Fairly happy", and "Not happy", where Elga deals in "People (non-depressed people, at least)" and "depressed people".

We don't know how many of the "Not happy" people are actually depressed. Perhaps Gummo thinks all of them are. That's naive; depression requires more than mere failure to be happy. And there are plenty of other emotional states than "depression" which could lead a respondant to answer "Not happy", including anger and bitterness.

Using the same kind of leap of faith, we could instead assume that clinical depression is constant among people of all political parties (it is a biological problem, after all) meaning that the "extra" not-happy Democrats must be suffering from some other emotion like rage or bitterness (which fits with some I know).

Or, heck, we could even go the other way, and assume that since Republicans can't possibly be angry (they control the country, right?) that all their non-happy sorts (4%) are suffering depression, and claim it's also "reasonable", based on all the rage we hear from Democrats, to think that non-happy Democrats (7%) must be suffering from rage, not depression -- and then conclude that Democrats are both more miserable AND more self-deluding!

Again, this is no better supported than Gummo's leaps of faith (and I don't believe any of it, myself) but it illustrates that we can conclude just about anything by superimposing our own unsubstantiated yet-oh-so "reasonable" suppositions onto the reported data.

A third problem arises even when we ignore all the former errors and grant Gummo his (tortured) conclusion: Even if Republicans are more likely to overestimate their abilities, the survey only shows 3% of (R) and 7% of (D) to be "not happy", meaning that 97% of Republicans are overestimating their abilities, compared to "only" 93% of Democrats who do the same!

Needless to say, that's like the pot calling the kettle a very, very dark shade of charcoal. :-)

But there's more fun still...

A fourth problems is that Gummo appears to confuse statements like "More people who are X do Y" with statements like "Each person who is X does more of Y".

Even if it were true that happiness equals self-delusion, this survey doesn't show a given Democrat is more clear-headed about their abilities than a given Republican. Instead, it implies that in a room with 200 people, half D and half R, there will be 4 more depressed people, with "more accurate" self assessments, among the Democrats than the Republicans. (Yeah! TWO PERCENT! Whoa!)

He flaunts this profound confusion here:

what it comes down to in the real world of verifiable results is that if you were to ask George W Bush if he was very happy, there's a 62% chance that he would answer yes, whereas if you asked Howard Dean the odds are only 50/50.

No, Gummo, that is precisely what it does not mean. It does not necessarily imply that when given a known, fixed Republican, he will say he is "Very happy" 65% and "Not happy" 3% of the times you might ask. It could equally be true that one particular Republican is (over a given time period) continually "Very happy", and another continually "Not happy", and that there tend to be a lot more of the former types.

A fifth problem is that Gummo (deliberately?) also conflates an accurate estimation of one's abilities and prospects with a breakdown in critical thinking skills in other areas:

If Republicans are more likely than others to be self-deluding, it's reasonable to suppose that they are also more likely to hang on to beliefs that make them happy, even if the weight of evidence is against their beliefs.

But again, to draw this conclusion from the evidence cited is neither "reason" nor "reasonable" but rather wishful thinking.

(a) We do not learn from this Gallup poll why the respondants are happy (or not): Is it their belief system? Their fabulous wealth? Their love life? Their collection of Hummel figurines? We cannot conclude from cited evidence that "beliefs" are the primary component.

(b) Even if are convinced "beliefs" are the primary component of happiness for survey respondants, no evidence is presented indicating these beliefs are of the untrue or irrational kind.

(c) Even if we do assume, further, that these happiness-causing beliefs are all or mostly irrational, there's no evidence present which says that the polled individuals are actually seeing the evidence against them. It may equally be possible that they are simply ignorant of the counter-evidence, so that no "delusion" needs to come into play yet.

(d) In general, there is no evidence presented that an overly positive self-image is linked in any way to general mental impairment, for example in the sciences, mathematics, logic, physics, or other areas in which this supposed "evidence to the contrary" might be located.

(If so, this would be a hilarious result, implying normal, mentally healthy individuals should avoid pursuing the sciences, and that depressed individuals should be conscripted for scientific research.)

(e) Alternately, if non-depressed individuals are not impaired in other types of thinking, and we again accept that these crucial, happiness-causing beliefs are surely false ones and that their adherants know this, it must mean that non-depressed individuals are generally more intellectually dishonest.

(See what fun and interesting conclusions we can draw if we just make a few assumptions unwarranted by the data we're citing!?!)

Instead, I argue (as Elga does also!) an overestimation of one's own abilities doesn't preclude an accurate assessment of others' abilities.

(Indeed, if normal people overestimate their own abilities, conservatism is grounded in a pessimism, not optimism, toward the nature of the human animal and our collective ability to solve all the world's problems. A conservative would say this is a needed corrective for the individual's own errors of overestimation, especially when she is reminded to apply her philosophy to herself.)

So even if we grant, for the sake of argument, that conservatives are more "wrong" about their own abilities, by Elga's own words it doesn't follow that they are making mistakes about other judgements.

A fifth problems is that Gummo (deliberately?) ignores Elga's testimony about the effects of these self mis-estimations. Elga sites "Taylor and Brown, 1988" as saying these estimation errors have good effects: increased care for others, motivation, persistence, and the capacity for creative, productive work.

In contrast (likely because he thinks he's shown mis-estimation is an attribute of US Republicans) Gummo portrays these behaviors as having negative, not beneficial consequences: "Could the human race survive a [sic] the geopolicial [sic] climate that would result ..." from more happiness, and thus also self-delusion (he supposes) resulting from increasing conservativism worldwide?

If we grant that the "Not happy" contingent contains all the depressed people, this still means that if everyone on earth became a Republican, (ignoring the obvious problem of internationalising this statement) depression would only drop a very few percentage points... bounded by the 7% (D) to 4% (R) difference.

Somehow, I think the "human race" could survive the beneficial consequences Elga cites if less than 2% of the population went from "depressed" to "non-depressed".

But then again, I'm not a liberal who fears a positive trait might have been identified in the people I oppose!

Elga's paper warrants it's own set of responses, but that's quite enough for now...

Being right is it's own excuse and justification. One should never have to justify one's motives for being correct about things or telling the truth. But when someone is wrong, the "Why?" questions of motive become worth asking. When one is manifestly, repeatedly, deeply wrong, even more so.

Gummo ponders whether people might reject inconvenient data or engage in muddle-headed thinking in order to feel good about themselves or their views. As far as I can see, he inadvertantly provides us with Exhibit A -- but sadly, probably not located in the party he would have preferred.

If happiness is found to be a conservative thing, then somehow we must find happiness to be bad! Perhaps new slogans could be generated around this idea... Liberals: "We're for increased misery! Because it's good for you!"