Steven Pinker is a psychologist who writes on politics. His theories are interesting but are framed too universally to be valid

Psychology is a universal science of human nature, whereas political science is centered on the study of particular historical events and trends. Perhaps it is unsurprising, then, that when a psychologist looks at politics, he presents ideas that are thought-provoking but are too general to quite work. This is fine; political scientists can then take such ideas and try to adapt them more closely to particular circumstances.

The psychologist I’m thinking about here is Steven Pinker, who, in writes the following on the question, “Why Are States So Red and Blue?”:

But why do ideology and geography cluster so predictably? Why, if you know a person’s position on gay marriage, can you predict that he or she will want to increase the military budget and decrease the tax rate . . . there may also be coherent mindsets beneath the diverse opinions that hang together in right-wing and left-wing belief systems. Political philosophers have long known that the ideologies are rooted in different conceptions of human nature — a conflict of visions so fundamental as to align opinions on dozens of issues that would seem to have nothing in common.

This is all fine—except that attitudes on such diverse issues are not so highly correlated. For a quick check, I want to the General Social Survey website and looked up correlations among attitudes on gay marriage (marhomo), military spending (natarms) and upper-income tax rates (tax rich). The correlations are 0.17, 0.09, and 0.05. These correlations aren’t zero (for example, the people who think we spend too little on the military are the most likely to think that taxes on the rich are much too high), but they’re not huge, either. Before developing a theory of why people’s attitudes on such issues are so highly correlated, we should first measure what correlation is actually there.

That said, correlations among issue attitudes are higher among richer, higher-educated, more politically-involved Americans (see Figures 8 and 9 of my American Journal of Sociology paper with Delia Baldassarri), and perhaps that is the group that Pinker is most interested in. Similarly, Pinker talks a lot about the political differences between “red” (Republican-voting) and “blue” (Democratic-voting) states, and these differences, too, are larger among the rich than the poor.

Pinker continues:

While these theories help explain why the seemingly diverse convictions within the right-wing and left-wing mind-sets hang together, they don’t explain why they are tied to geography. The historian David Hackett Fischer traces the divide back to the British settlers of colonial America. The North was largely settled by English farmers, the inland South by Scots-Irish herders. Anthropologists have long noted that societies that herd livestock in rugged terrain tend to develop a “culture of honor.” . . . The psychologist Richard Nisbett has shown that Southerners today continue to manifest a culture of honor which legitimizes violent retaliation. . . . The historian David Courtwright has shown that there is considerable truth to the cinematic clichés of the Wild West and the mountainous South of Davy Crocket, Daniel Boone and the Hatfields and McCoys. . . . The North and coasts are extensions of Europe and continued the government-driven civilizing process that had been gathering momentum since the Middle Ages. The South and West preserved the culture of honor that emerged in the anarchic territories of the growing country, tempered by their own civilizing forces of churches, families and temperance.

This sounds great, but I think it explains too much! If the current red-blue map, or something like it, had persisted for 200 years, then, sure. But the current division between red and blue America pretty much dates from Clinton’s first election in 1992. Remember, as recently as 1988 (OK, not so recent to some of you, but still alive in the memories of oldsters such as Pinker and myself), Michael Dukakis won West Virginia and lost Vermont. And the election map of 1976 looks nothing like recent maps.

In our book, Red State Blue State, we discuss some reasons that states are so red and blue, in particular why upper-income residents of these states are so red and blue. Some of this arises from increasing differences between the parties on political issues. I have no doubt that the psychological and cultural explanations discussed by Pinker are relevant, but I think it’s necessary to put them in a historical context. Otherwise you can end up overexplaining transient political patterns.

31 thoughts on “Steven Pinker is a psychologist who writes on politics. His theories are interesting but are framed too universally to be valid

  1. I think this is a general issue in Psychology. There is an undue tendency to universalize before addressing historical/”contextual” critiques.

  2. I don’t think universalizing is a general issue in Psychology (loooook you’re doing it yourself) more so than anywhere else.
    But Pinker is writing on a blog called the “opinionator” and he likes to talk broadly about broad topics. Then people who know
    more within these topics, disagree.
    http://www.psychologytoday.com/blog/sex-dawn/201103/steven-pinkers-stinker-the-origins-war
    http://www.bookforum.com/inprint/018_04/8575
    http://www.psychologytoday.com/blog/music-matters/201101/was-steven-pinker-right-after-all
    cf. the whole mind modularity debate (which Pinker lost).
    and here a Pinker-endorsement by the famous idiot Satoshi Kanazawa which counts -1
    http://www.psychologytoday.com/blog/the-scientific-fundamentalist/201105/how-is-steven-pinker-not-michael-jordan

    But there’s probably good stuff from Pinker, I don’t want to universalize. He may have jumped the shark here and there though.

  3. Pinker makes me sad. What a loss. With “The Language Instinct” he took a subject he knows a ton about — how we learn language — and explained it really well to a popular audience. It’s a great book. But then he seems to have decided that it’s easier and more fun (or perhaps more profitable, or perhaps it gets him more chicks, I don’t really know his motivation) to write really well about things he doesn’t know much about than it is to become an expert on something. So after The Language Instinct he wrote How The Mind Works, which is a decent book but you can tell he doesn’t have full command of the material. And since then…well, let’s just say I now lump him in with Malcolm Gladwell: He says thought-provoking things and he writes well, but you can pretty much guarantee he is picking his facts and explanations to fit his thesis. It’s not so much that these guys are trying to mislead, it’s more like they don’t much care whether they’re being misleading or not as long as they’re telling a good story. I’m sure they’d rather get things right, but they don’t seem to care enough to put much effort into it.

    You say “before developing a theory of why people’s attitudes on such issues are so highly correlated, we should first measure what correlation is actually there.” The old, good Pinker would probably have agreed with that, but the new, bad Pinker probably thinks that’s stupid. Why would he care what the correlation is? The numbers, whatever they are, wouldn’t make his story any better.

    • Phil:

      I don’t think you should lump Pinker in with Gladwell. For one thing, Pinker and Gladwell have a feud! I think a fairer view of Pinker would be that, through the combination of talent, hard work, and luck that led him to have a high profile, he’d like to take the opportunity to write about all sorts of important things. That’s what being a public intellectual is all about, and indeed that’s what I do (on a much smaller scale) here when I blog on topics such as literature that are only peripherally related to statistics. I do think, though, that before writing on “why are states red and blue” for the New York Times, it would make sense for Pinker to consult with some political scientists. That said, if I were writing on literature for the Times, I might wing it too. So I don’t want to slam Pinker, I’d just like to point him in the direction of people who’ve studied these issues a bit already.

      • Andrew, I restrained myself from saying this in my original comment but now that you have brought it up I can’t resist: I am glad you are not a great writer for a popular audience. (I’d say you’re a great writer for a technical audience and a fair writer for a popular audience.) If you were a better popular writer, Red Fish Blue Fish would have sold hundreds of thousands of copies and you’d be out there writing punditry on literature or history or politics or something, and that would be bad. Just as it is bad that Pinker is doing it. We don’t need more know-nothing pundits, we have a surfeit of them; but we can use many more people who take material that was previously understandable only to a small group of experts and make it accessible to a larger group of people.

        So, yeah, if you were writing on literature in the Times you might wing it too. Which is why I’m glad you are not writing on literature in the Times.

    • I found his most recent book — The Better Angels of our Nature — to be well-written and interesting. I’d classify it as a decent popularization of a broad area of research (not his own, but he looks to have done the work to have brought it together well enough). Have I just been fooled by a cherry-picker?

      Besides, how many popular books can someone write about their specialist research area something without repeating themselves or boring the audience or both?

      • Dan, Pinker’s other writings are well-written and interesting but that doesn’t make them authoritative or even necessarily correct, so I wouldn’t be surprised if TBAOON is well-written, interesting, and misleading. Or, maybe it’s well-written, interesting, and even-handedly informative. I’d bet it’s closer to the former than the latter, but since I haven’t read it I obviously can’t say for certain!

        • There’s a New Yorker review that says “There is much in “The Better Angels of Our Nature” that is confounding. Those developments which might seem to fit into his schema are treated in detail. Yet other episodes that one would think are more relevant to a history of violence are simply glossed over.”

          According to Wikipedia, John Arquilla, in the journal Foreign Policy, had complaints in that vein as well: ‘Arquilla criticized Pinker for using statistics that he said didn’t accurately represent the threats of civilians dying in war: “The problem with the conclusions reached in these studies is their reliance on “battle death” statistics. The pattern of the past century — one recurring in history — is that the deaths of noncombatants due to war has risen, steadily and very dramatically. In World War I, perhaps only 10 percent of the 10 million-plus who died were civilians. The number of noncombatant deaths jumped to as much as 50 percent of the 50 million-plus lives lost in World War II, and the sad toll has kept on rising ever since.”‘

          These critiques fit with my expectations for a Pinker book, but: I still haven’t read the book, so I don’t know for sure.

        • Yeah, it sounds like you have a bias against the guy. Like Stave S says below, he’s done a pretty good job with this book. I’ve not seen any criticism that has the central thesis of this work being even close to wrong — it’s easy to question a few arguments but he makes it clear that it is a body of evidence he is presenting and individual peices may have problems. I think he does a good job of presenting some statistical ideas to a general audience rather well. I’ve only read this book and The Language Instinct so maybe have only seen the good stuff but comparisons to Gladwell (Tipping points! 10000 hours! 0.3 of a second! Igon Values!) are not appropriate.

  4. Why are the geographic correlations so recent? An obvious point is that the south was solidly Democratic from the end of Reconstruction until its one-party system was smashed in the civil rights era, and since then it took some time to realign. Another is that the New Deal era led to some significant aberrations in politics and we are only now back to normal.

    • Wonks:

      1. The south is part of the story but only part. As we discussed in Red State Blue State, these changes occurred in the non-south.

      2. I clicked on the link and read the quote and it makes no sense to me at all. Economic issues remain paramount in general elections and I can’t see how people could think otherwise.

  5. “It is a capital mistake to theorize before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts.”

    – Sherlock Holmes, A Scandal in Bohemia.

  6. I’m not sure the problem is Pinker’s instincts for general theories. It seems more like a lack of attention to empirical details.

    As an aside, is there a tie-in with Dan Kahan’s research on cultural cognition? His work suggests that we can group people into a small set of cultural worldviews, and that group membership is strongly correlated with people’s factual beliefs and value judgments on a range of risk-related policy issues (nuclear power, gun control, abortion, etc.). The amount of variance that his approach explains is relatively modest, but he’s always argued that variance isn’t the proper measure in these contexts. I’m not well versed enough in stats to figure out the merits of this debate, but just wondered if you or anyone here had any thoughts.

    • Brian:

      I don’t know enough about that work to comment, but what I can say is that survey responses are noisy (presumably representing a lot of “nonattitudes” in the population). I think that many individual people feel there is a logic connecting all their political beliefs, but different people have different logics. Pinker gave an example of gay marriage and the military budget. There is a wide range of views on these two topics and the correlation is low, because these two issues can be placed in many different conceptual frameworks.

  7. Apples and oranges Andrew. You’re comparing an op-ed with an academic book and paper. Of course the op-ed is “too general to quite work.” It’s 300 words for an audience of millions! Your academic paper alone is probably 8,000 words for an audience of 8.

    Meanwhile, in his books, Pinker conveys big ideas in a very compelling fashion. My memory of Red State Blue State was that it was a bit of a snoozefest.

    Word for word, Pinker beats you hands down.

    • Nate:

      I’m not competing with Pinker. My problem with the op-ed is not that it simplifies; it’s that it’s factually wrong (in his claim that “if you know a person’s position on gay marriage, can you predict that he or she will want to increase the military budget and decrease the tax rate”) and that his story does not make logical sense (given that it is a universal explanation that only applies to the past 20 or 30 years). Even in 300 words, there’s no reason to make factual errors or advance illogical arguments.

      That said, I think Pinker is great. Everybody makes mistakes, he made one here, I pointed it out and he can do better next time. I make lots of mistakes too, people point them out, etc.

      P.S. If you think only 8 people read our article, you’re dead wrong. Insult me if you want, but please be numerate while you’re doing it! We have some excellent commenters here on this blog, so if you want to poke fun at me, you’ll have to step up your game a bit.

    • “Word for word, Pinker beats you hands down.”

      Nothing to be ashamed of: word for word, Pinker beats every North American generalist intellectual for lucidity, eloquence, and insight. For example, here’s an email interview I conducted with him a decade ago while he was promoting “The Blank Slate:”

      http://www.isteve.com/2002_QA_Steven_Pinker.htm

      You don’t have stuff that well written show up in your inbox every day.

      • Steve:

        Your Pinker interview is fun but I don’t agree with some of it.

        For example, he dismisses the idea that “violence is learned behavior.” Obviously there is innate violence, not all violence is learned, but I don’t see why he thinks it’s so unreasonable to think that much violence is learned.

        Also I don’t buy Pinker’s claim that:

        Sophisticated people sneer at feel-good comedies and saccharine romances in which everyone lives happily ever after. But when it comes to science, these same people say, “Give us schmaltz!” They expect the science of human beings to be a source of emotional uplift and inspirational sermonizing.

        I think the above statement is a classic product of selection bias. Imagine a 2-dimensional space showing artistic sophistication on the x-axis and scientific sophistication on the y-axis. (For the sake of argument, I’ll accept the implication that these dimensions are clearly defined, and I’ll set aside the loaded word “sneer” in the above quote. (“Sneer” is always something that other people do.)) We have people in all four quadrants of this hypothetical graph, and I expect there’s a positive correlation: people who are more sophisticated in one direction are likely to be more sophisticated in the other.

        Now Pinker is taking the subset of people who have high x-values (artistic sophisticates) and noticing that, for these people, they are less scientifically sophisticated. That is, if you select on high values of x, you’ll find y

        • Excellent points. I’m glad you homed in on the “schmaltz” line.

          However, I think Pinker is making an argument about the proper distribution of cultural prestige. Individuals will inevitably differ in their tastes, but it’s not unreasonable for a culture to assign more prestige to some works of art than to others. For example, the prestige of tragedy is an inducement to encourage audiences to invest effort and emotion in difficult works. I’m probably not going to dust off my 35 year old copy of “King Lear” for some bedtime reading tonight, but I am glad I was assigned it way back in high school.

          In particular, Pinker offers the striking suggestion that the high points of tragedy in culture typically involve conflicts of genetic interest (quantified by sociobiologist Robert Trivers in the early 1970s), such as sibling rivalry in the case of Lear’s three daughters:

          “Aristotle was perhaps the first to note that tragic narratives focus on family relations. A story about two strangers who fight to the death, he pointed out, is nowhere near as interesting as a story about two brothers who fight to the death. Cain and Abel, Jacob and Esau, Oedipus and Laius, Michael and Fredo, JR and Bobby, Frasier and Niles, Joseph and his brothers, Lear and his daughters, Hannah and her sisters …”

          By way of analogy, Pinker suggests that while, say, Trivers’ theory of sibling rivalry may make one sad, rejecting it for that reason should be no more culturally prestigious than rejecting, say, The Godfather II because it makes one sadder than, say, Flashdance (which, I must say, I found quite entertaining).

          Similarly, rejecting evolutionary theory or IQ research because it’s not uplifting and schmaltzy enough should not enjoy the cultural tailwind.

  8. http://isteve.blogspot.com/2012/10/pinker-takes-crack-at-explaining-red-v.html

    Here’s my critique of Pinker’s op-ed.

    Basically, us white intellectuals like to talk about subtle differences among whites, such as the difference in crime rates between West Virginia and Minnesota, for which Albion’s Seed provides useful background.

    But, a lot of stuff, like crime rates and voting behavior, are heavily influenced by brute facts of race, which most white intellectuals find tedious and depressing to contemplate.

    The higher homicide rate in South Carolina than in West Virginia, for example, is the reverse of what Fischer’s theory should project, since the Mountaineer state has more ornery Scots-Irish and South Carolina has somewhat less ornery English background. But, South Carolina has a lot more blacks, so it of course has a higher homicide rate. According to the Obama Administration’s Bureau of Justice Statistics, blacks have a homicide offending rate seven times that of whites, a difference that just overwhelms the more subtle differences among different kinds of whites that Pinker talks about.

    Daniel Patrick Moynihan joked in the early 1990s that there is a moderately high correlation (about .53) between school test scores and distance from the Canadian border, so if you want to raise your state’s test scores, all you have to do is drag up closer to the Canadian border.

    Some of this difference is related to Albion’s Seed — northern tier Puritans really were smarter and more academic oriented than the Brits who settled farther South. But a huge fraction of Moynihan’s Law just has to do with northern states having lower percentages of blacks and Hispanics.

  9. Pingback: Kahan on Pinker on politics « Statistical Modeling, Causal Inference, and Social Science

  10. Pingback: More Pinker Pinker Pinker « Statistical Modeling, Causal Inference, and Social Science

Comments are closed.