Monday, November 1, 2010

Genes for how we think and what we think

Several people have sent us the latest genetics of political science story that has been getting attention all over the web (e.g., here and here).  The paper referred to in these stories just appeared in The Journal of Politics with the title, "Friendships Moderate an Association between a Dopamine Gene Variant and Political Ideology."  People become liberal, the authors claim, if they've got a particular variant of the "novelty seeker gene", DRD4, and had a lot of friends in high school.


This is a follow-up to a 2007 paper by two of the same authors, Fowler and Dawes, reporting two genes that predicted voter turnout in twin studies, although these genes are also moderated by environment, they say; turnout depends on the association between genes and exposures to religious social activity.  They conclude that these findings are important to "how we both model and measures political interactions."


And a follow-up to a 2009 paper reporting the gene for partisanship and joining of political groups (the DRD4 gene again).  Among other papers. 


Now, these are only a few of the authors responsible for the recent large crop of papers reporting genes for voting behavior and political ideology and other behaviors in the realm of political science, but they are particularly visible, and publishing the latest paper just before the November elections -- gee, wonder why -- certainly helps.  The headline in the Fox News story about this work is "Researchers find the liberal gene", and they then go on to say that "liberals can't help themselves, it's in their genes".


OK, let's stick with this for a minute.  Much as we wanted to, we couldn't ignore what's below the headlines, the meat of the piece:

"Ideology is about 40 percent heritable. It's almost half genes and half environment," Fowler told FoxNews.com. What's more, he said, any trait that can be inherited has potentially been with the human race for a long time, meaning political ideology has been a part of us for tens of millions of years.
"If it's really the case that genetic variation is influencing ideology, this isn't something we've been living with for the last ten years," he told FoxNews.com. "These are processes that have been going on for the past million years."
Fowler suggests that it made more sense to be liberal in certain environments at specific points in human history, and in others a conservative ideology was merited. "And this is what made it possible for our species to survive," he said.
"If it made sense for us all to be liberal, natural selection would have made us all liberal."
There's so much wrong here, and if the quotes are accurate, it's Fowler who's wrong, not the journalist.  Fowler may 'know' that DRD4 is a dopamine receptor gene, the receptor for a neurotransmitter, and he may think that political ideology is a result of gene by environment interaction, but the Fox story suggests that he believes he's found a gene 'for' liberalism (in spite of the sentence in the paper itself that rightly cautions that "perhaps the most valuable contribution of this study is not to declare that ‘‘a gene was found’’ for anything, but rather, to provide the first evidence for a possible gene-environment interaction for political ideology").  Emics and etics -- what he says is not what he does.


And,  "...political ideology has been a part of us for tens of millions of years"?  "These are processes that have been going on for the past million years"?  Let's see, Homo sapiens dates from when?  100-200 thousand years ago?  Were our Homo erectus ancestors liberals or Tories?  Being liberal sometimes and conservatives others is what allowed our species to survive?  Who knows what that could even mean.  And of course Fox News picks up on the idea that natural selection didn't make us all liberal, now did it?  No it didn't, because it didn't make sense.


So, if political ideology is genetic, there must be a gene for conservatism too?  In which case, why do we see the kinds of swings in voting patterns in the US that we have seen in the last several decades, and are likely to see this week?  It can't be that it's primarily those who are genetically programmed to be liberal who vote in some elections, and those who are conservative who vote in others.  People actually do change their minds.  Oh, there must be a gene for that, too.


But much more important than these silly quotes, or even than the details of the paper, this paper is an indication of the use and abuse to which genetic data are currently being put.  In a day or two we'll write about the data set that this study was actually based on but for now let's think about what it means to even ask whether there are genes for voting behavior.  (And, by extension, if how you vote is due to gene by environment interaction, and in this case, the 'environment' is the number of friends a teenager has in high school, couldn't it be that there's a gene for the number of friends we have? Well, it won't be a surprise to hear that the answer is yes if you believe studies done with the same National Longitudinal Study of Adolescent Health -- paid for with your tax dollars -- used in the latest Fowler paper.)


It isn't irrelevant to ask who wants to know, who's asking about genes for behavior.  Or why they find these kinds of questions so interesting.  Or whether this kind of science should just be smiled at tolerantly even if it's bogus.  Or what we should think to do, at this stage and before it's too late, if it happens to be correct.


There are many issues here, and they involve the spectrum of things from personal politics (ours, the investigators, and yours), the nature of legitimate science, of criteria for scientific inference, of vested interests, of politics, of ideology and tribalism and much more.


One thing is clear.  Like it or not, this search for genes for traits related to how we behave, or for how well we think and even for what we think, is a wave, perhaps a tidal wave, of a kind of Darwinian determinism sweeping across the political spectrum, led by the United States but (as so often happens) with followers in Europe and Asia (mainly these days, China).  It is not idle, ivory tower science.  Whether it is yet another passing fancy, or something more serious, time will tell.  But for many reasons, this like other aspects of genetic determinism these days is not going to be changed because of the facts: it is a commitment, and taxpayers are going to be paying for it for a long time to come, for better or worse.

8 comments:

Texbrit said...

I would like to hear what a definition of "liberal" might even BE, general enough that it could apply to all people across all cultures across all times. What was liberal caveman?? I guess he believed that gay cavemen could serve in the cave army and...he might have even believed in evolution!

Ken Weiss said...

There is so much to say on this subject that we hope to do a series of posts that will capture what's going on, and in a responsible way. That is itself difficult, because advocates always manoever to defend points of view (and, of course, so do their critics), the science is not simple, and the issues inherently involve politics, not just 'objective' science.

Even showing that something is questionable, if not shoddy science won't persuade those who are committed to a belief that this is opening the "Secrets of the Mind" as the media like to hype this kind of thing.

But certainly those who advocate this kind of thing should be identifying fossils of liberal Neandertals and showing that they held that kind of politics (maybe they were the ones who mated with modern humans!, while their racist brethren, without the 'liberal gene' didn't?)

Anne Buchanan said...

Ah, well, that's an issue, yes, and the subjectivity of definitions, such as what a 'liberal' is, is part of what makes this kind of work such slippery science.

In this case they are talking about 'risk taking', or 'novelty seeking' behavior, as this particular gene variant has been associated with these traits in other studies. So, liberals are people who are more comfortable taking risks. (But only if they had a lot of friends in high school.) Um, so then agreeing that _lowering_ the risk to those without health insurance by supporting health care reform in the US would be considered liberal risk-taking behavior?

But, what if it were 'conservatives' in this study who had been found to have this particular gene variant at higher frequency than (self-described) liberals? It's just as plausible to suggest that it's conservatives, the party of big business, who are more comfortable taking big risks. Who got us into the banking crisis, for example? Financial innovation is nothing if not risky.

So, while the argument positing a liberal/risk association might sound plausible to, say, Fox News -- and many others (though not every media outlet, to their credit) -- even turned on its head it can still _sound_ plausible. That doesn't make it good science.

Texbrit said...

We do know that there was a definite "anti-intellectual" streak amongst cavepeople. They just were not interested in caveleaders who were circumspect, thoughtful, balanced, and based their judgments on past lessons. For example, the few liberal cavepeople advocated hunting in a conservationist way to protect the resource, but a certain candidate for president just said "Hunt Baby Hunt!" to rowdy applause.

James Goetz said...

I wonder if the liberal gene or lack of it had anything to do with the extinction of most Neanderthal populations.

Ken Weiss said...

Note the subtle seriousness of this story. If there are two variants in a gene, as I believe is what is being tested by these authors, then how one labels them can reveal the underlying intention.

Why is it that 'conservative' is considered the baseline and 'liberal' is the labeled-allele? This derives very clearly from the idea of 'normal' and 'abnormal' alleles, and that means that these investigators (and the media that pick up their stuff) are working from a conservative=normal bias, and targeting 'liberal' as the derived, or abnormal variant.

Even if 'liberal' were the allele with lower frequency, this bias is revealing, because it states a social rather than biological asymmetry, the 'liberal' variant need not be the newer one (I don't know if it is or not).

This shows the immediate if not already preceding objectives of this kind of work, and how politics are inherently involved. Of course political biases go both ways, but the point is that this kind of research is neither purely objective nor apolitical.

Holly Dunsworth said...

I expressed the normal allele in the 80s when I begged my parents to vote for Reagan. Somewhere in the 90s I began expressing the mutated liberal version. When I'm old, the odds are that I'll revert back to expressing the conservative allele, at least a little bit.

DRD4 must have a really cool regulation pattern! Epigenetics probably.

And not maternal ones... these are determined by your circle of friends and how big it is.

(because I can't insert sarcastic faces or tone of voice, please imagine those.)

Texbrit said...

"Show me a young Conservative caveman and I'll show you a caveman not invited to the cool cave parties. Show me an old Liberal caveman and I'll show you a caveman with no meat to go with his gruel."

- Winston Churchill