Asymptosis

Another of those convergences: I just joined the Richard Dawkins group on Facebook, and discovered that today is his birthday. (Happy birthday sir!) It’s a convergence because over the last week I’ve been horribly dismayed. After decades of near hero-worship on my part, I’ve discovered that he is not acting as the man I’ve always believed him to be.

The issue is his position on group selection. (Don’t go away: it matters.) The way he has defended that position seems contrary to everything I have always so admired about him.

And I have so admired him, for so long. I have to watch myself constantly to avoid the kind of wild-eyed evangelism that serves only to give aid and comfort to the creationist enemy. The Selfish Gene and The Extended Phenotype provided (some of) the fundamental underpinnings for my understanding of (human) existence, and the belief and value system that’s built on that understanding.

I didn’t really need to read The God Delusion — preaching to the choir — but I did so and greatly enjoyed it purely for the joy of his arguments — the lucidity, the cogency, the logical and rhetorical coherence.

I can’t count the number of times I’ve recounted his anecdote about an aging professor who changes his mind. (“My dear fellow, I wish to thank you. I have been wrong these fifteen years.” . . .  “We clapped our hands red.”) It still brings tears to my eyes when I read it, and epitomizes how science, for all its real-world failings, is fundamentally different from faith. (Here. Start with “It does happen.”)

So, again, I’m nearly teary-eyed at the stance he has taken, and the rhetoric he’s deployed, in response to a body of thinking that has grown over decades and came to something of a culmination in 2007. (I’m late to the party on this one.) That body of evidence and theory contradicts one of his longest- and strongest-held beliefs: that group selection is hooey, that it could not have had any role in the evolution of human altruism.

Remember the stated goal of Dawkins’ seminal book: “My purpose is to examine the biology of selfishness and altruism.”

His basic theory: genes are the units of selection, and organisms are the vehicles of that selection. If a gene causes organisms to have more grandchildren, the gene’s frequency expands in the population.

Based on this, he rightly pooh-poohed warm, mushy, poorly-reasoned notions about genes contributing to “social cohesion” and the like. No altruistic gene could survive in a group if it didn’t provide net benefit for the individual containing that gene — either by helping the individual, helping kin who have the same gene, or through reciprocal payback from other individuals.

But what about the success of groups? Could groups with more altruistic genes have more grandchildren than groups with more purely self-serving genes? Could that group selection effect predominate over individual selection within the group?

It seems plausible, and from the first time I encountered the conundrum, it has always seemed to me to be a purely statistical question.

And that’s how (a damned impressive set of) mid-20th-century evolutionists went at it. They built models, ran the numbers, and determined that no: group selection could not overwhelm the forces of individual selection. If a gene isn’t good for an individual (and/or his kin), it will die out.

That belief achieved an orthodoxy in the political ecology of scientific academe that largely prevented later scientists from even raising the question, and successfully crushed most of the few efforts to re-examine it. It’s agonizingly similar to the despicable response that sociobiology and evolutionary psychology themselves encountered over those same decades, from the likes of Lewontin, Gould, and the “Theory” humanists.

As a result, both professionals and amateurs — including reasonably diligent amateurs like me — have been unthinkingly chanting along with that orthodoxy for years, decades. I don’t know how many times I’ve discredited thinking that seemed rooted in group-selectionist thinking.

And I was wrong. At least, I was too categorical. So I was sometimes/often wrong.

Here’s what makes me so sad: Richard Dawkins has been perhaps the most powerful voice for that orthodoxy, and he seems to be clinging to that idol even when its feet — his feet — are looking resoundingly clay-like.

Cutting to the meat, simplified:

In 2007, David Sloan Wilson and E. O. Wilson (the founder of sociobiology and one of the most brilliant, diligent, and sober evolutionary biologists to ever live, as Dawkins certainly agrees) published a paper (PDF) laying out the cogent, lucid, and compelling case that group selection can indeed predominate over individual selection in the evolution of altruistic genes — that the group can be a vehicle of selection, just as the individual can. (They talk about “multilevel selection.”)

In other words, genes that benefit the group can proliferate in the larger population, even if those genes are disadvantaged within the group. Again, it’s all a matter of models and statistics, and the Wilsons (no relation) deployed and cited damned convincing models and statistics showing that the earlier evolutionists probably got it wrong.

Now if Dawkins had cogent takedowns of those models and statistics, there is nobody I would rather hear them from. But his counterarguments have all been from principles, even when those principles are not thrown into question by Wilson and Wilson — their arguments are based on those principles.

What’s more dismaying is that Dawkins’ few dozen paragraphs in reply (remember, it’s been three years since then) bear all the hallmarks of a religionist who has not a leg to stand on, lashing out in frantic, desperate defense with red herrings, tangents, inapplicable arguments, dodges, weaves, and personal invective. (I’m not a professional in the field, but I know good and bad arguments when I hear them.)

This post is already too long, so I won’t detail everything here. You can see one of Dawkins’ replies here (PDF), and you can read the whole story from D. S. Wilson — including much of Dawkins’ response — here. Wilson’s 19-post blog thread is here in a one PDF.

I’ll just quote one passage from Dawkins to give the flavor of those replies:

…as far as I am concerned, the statement is false: not a semantic confusion; not an exaggeration of a half-truth; not a distortion of a quarter truth; but a total, unmitigated, barefaced lie.

This is not the Richard Dawkins I’ve known and (intellectually) loved for lo these many decades. It is, in fact, the exact opposite of that Richard Dawkins.

I can only quote D.S. Wilson’s words, which precisely echo my most heartfelt feelings:

In my dreams, I imagine him reading my modified haystack model and saying “Well done, David! I have been wrong all these years.”

Richard Dawkins won’t you please come home?

“As debate over the dimensions of Obama’s package snaked through the House…”

Source.

What do you think, is this the same 29% that were still Bush-boosters in 2008?

AFTER HEARING THE ARGUMENTS FOR AND AGAINST THE PROPOSED [FINANCIAL] REFORMS, VOTER SUPPORT FOR LEGISLATION INCREASED

• At the start of the survey, 29% opposed reform, and 40% supported it.
• After details were explained and arguments for and against reform described, opposition stayed at 29% but support rose to over 60%.

Pew survey. The survey gathered demographics on respondents, but I don’t find any cross-tabs on demographics for that 29%. Enquiring minds…

Update: Instead of saying “Probably Not” in the title, I probably should have said “We have no idea.”

Being a Seattle parent with kids in private schools, I’ve been assailed for years by pronouncements and lectures by and about the Seattle-based Gottman Institute (tagline: “Researching and Restoring Relationships”). Their most widely known claim is their ability to predict, after watching a married couple for fifteen minutes, whether they’ll get divorced.

The basic Gottman theory — that facial expressions of contempt during couples’ interactions are predictive of divorce — seems very plausible, intuitively. But there are many intuitively plausible surmises that are just wrong.

And no matter how intuitively plausible it is, the claim always seemed fishy to me. But I never did the research to find out if their predictions were really accurate. Happily, somebody has finally done it for me. Here’s Andrew Gelman, god of all things statistical, blogging about a Slate article excerpted from Laurie Abraham’s Husbands and Wives Club.

Short story, their “predictions” are built on quicksand. Here’s how they do it.

Gather data on a bunch of couples — say, six variables for each couple. Determine which of those couples get divorced. Then run a program that finds an equation correlating the (presumably) predictive variables to the results. (These quite remarkable programs — the realm of ultimate-wonk physicists only a decade or two ago — are now available for free download, or as $49 Excel plug-ins. To quote my buddy Olav, “Isn’t it great living in the future?”)

Here’s what’s wrong: these programs will always find an equation that correlates the variables to the results. (With a greater or lesser “fit” to the data.) Does that mean the equation is predictive? Only if it makes an accurate prediction when applied to a different set of data.

That is what Gottman has not done, at least in his published papers. Every one of them has a new equation that — surprise — “predicts” the divorces in the group with surprising accuracy — the same group that was used to generate the equation.

Now this is true: if the program finds a good data-fitting equation (which Gottman seems to have done — multiple times), there’s a greater chance that the equation will actually be predictive. But there’s only one way to know: use it to predict. If the prediction fails, the predictive ability of the equation is falsified.

Gottman has not (to my knowledge) attempted any falsifiable predictions, so we have no idea if his predictions are true or false.

The Gottman Institute presumably has all the data to hand, and could test past predictions against future results. I’m wondering: will they now do so?

Google doesn’t turn up any hits for “Abraham” on the gottman.com site, so I’m thinking they haven’t responded. One can rather understand why. Abraham says in a comment to the Slate post (and, she says, in a footnote in the book — it’s not search-insideable on Amazon) that she repeatedly requested an interview in May 2009 but Gottman wouldn’t see her until October — too late for her book. I do wish she’d tried again before the excerpt was published, giving him ample benefit of the doubt.

But absent that, I’m quite curiously waiting to see what we hear from The Gottman Institute.

In one of those wonderful confluences, two items just came together for me. I read The Social Atom by Mark Buchanan, and my friend Steve posted a link to an Economist piece on evolution, fairness, markets, and religion.

It all circulates around a central conundrum that evolutionists (including Darwin) have been worrying at since Darwin: why do humans, in all cultures, perform selfless acts? You’d think that natural selection would weed out those fools — that cheaters who take advantage of the selfless would have more grandchildren, making the do-gooders vanish from the population.

The answer seems to be group selection: groups with more selfless types survive better than groups of cheaters. (I’ll be posting soon on the controversy over group selection, particularly Richard Dawkins’ intransigence on the issue; suffice it here to say that it makes sense to me. Can’t wait? Wikipedia.)

Which brings me to The Social Atom. Buchanan looks at systems made of of fairly simple “atoms,” and shows how those atoms’ simple properties can result in very sophisticated and often predictable system behavior. Think (my example) of flocks made up of birds with very simple algorithms — “if there’s a bird on one side of you and it moves away, move closer” — imagine flocks of birds flying, and you get the idea.

Likewise the atoms in a magnet: you can ignore all the insane complexities of sub-atomic particles and just think of those atoms as arrows whose direction affects adjacent atoms’ arrow directions. That single property explains the whole system.

This plays out in human systems too. If you had stock markets where everyone throws darts, the ups and downs of the market would map to a bell curve with a normal distribution — thin tails, with very few days of large ups and downs.

But build a system where individuals switch between strategies in response to market movements and  other individuals’ strategies, and you get the distribution we actually have: a much flatter bell curve with fatter tails — many more days with big up-and-down swings. (This is what brought down Long Term Capital Management. A one-in-five-hundred-year event on a normal bell curve was a one-in-five-year event in the actual distribution of market movements.)

That higher-volatility pattern affects the individuals’ decisions — their strategies — but the pattern remains. Because — this is what’s fascinating — it doesn’t matter what the individual strategies are. The simple fact of individuals adaptively selecting strategies is all it takes for the high-volatility pattern to emerge.

How does this all bear on non-kin altruism? In my words: Once you have a vehicle — language — by which cultural values and mores can be transmitted to others and across generations, cooperation and individual selflessness are naturally emergent properties of that system. They’re inevitable, because groups that don’t transmit and enforce those values don’t survive in competition with ones that do.

And that leads me to the study that Steve and The Economist linked to. It looks at fifteen contemporary, small-scale societies (hunters, fishermen, foragers), and asks how the selflessness that makes small groups prosper could have extended to the kind of global altruism that we see today.

They suggest that “such societies may have required norms and institutions that sustain fairness in ephemeral [one-time] exchanges.” Their findings support that: larger communities “punish” cheaters more, and groups with more market interactions and participation in world religions have more “fair” behavior by individuals.

In my thinking: these social patterns and institutions are naturally emergent properties of a species that has language. (A necessary caveat that I won’t expand on here: periodic genocide is also a naturally emergent property of such a species, for the same reasons of group selection.)

Now I don’t know about you, but my spidey sense detects a strong whiff of axe-grinding in Henrich’s conclusions, demonstrating that those favored children of the Right — markets and religion — are what accounts for fairness. (I think this may be why Steve tweeted it?)

But the study makes me wonder (and wonder why the researchers didn’t wonder): do government-like institutions in those societies also enforce fairness and encourage selflessness? Do equivalents of our three branches — strong leaders, councils of elders, and systems of group adjudication — correlate with more fairness and selflessness in a society?

This dichotomy — between market/religion-based institutions and government institutions — also makes me think again about the Jonathan Haidt research I’ve been blogging recently, showing that Republicans give much more moral weight than liberals to group loyalty. Might it be — since liberals believe more in government as the fairness enforcer — that the two groups just define “group” differently (or that liberals’ support for government is based on reasoned belief instead of “sanctity” or “moral intuition”)?

Steven Pinker has suggested that cooperation with non-kin is one of the three main attributes (along with language and tool/technology use) that distinguish humans from other animals. That cooperation put us at the top of the food chain.

Which leads me to reiterate a thought I expressed recently in the comments, in response to those who champion competition as a great good: competition is a second-order effect; its only merit is that it makes cooperation more efficient. If we’d “all just cooperate” (the woolly headed liberal’s mantra, finger twirling in cheek), we’d all be better off than if we all competed. It’s both obvious and stupid. Given a population of selfish social atoms, competition forces people to cooperate in groups.

But competition (a.k.a. “the market”) is not the only thing that improves cooperation. Henrich’s work shows that religion does as well. And it’s not at all difficult to find other instances  — government, for instance — in which cooperation is improved by . . . cooperation.

Jonathan Haidt asks on his blog:

“Can anyone understand Avatar who lacks all intuitions of purity/sanctity?”

He’s talking about the sanctity of nature, and of spirituality, as against corporate, consumerist, and militarist values.

My answer is “Yes.”

I (a devoted liberal with a “Liberal Purity” score of 1.0–compared to Libs’ 2.7 and Cons’ 2.1) understand it as a brilliantly Machiavellian corporate vehicle to extract cash from those souls who embrace that purist intuition — and from the other groups that are so successfully, simultaneously, pandered to:

1. America haters worldwide who love watching military/corporate America get its ass kicked;

2. Uneducated militarists (who nevertheless love their mommies), and who want to believe that only their like can save the world;

3. Those with contempt for pointy-headed academics who can’t get anything done on their own (at least they’re not depicted as bad — just ineffectual); and

4. Teenage boys who think they’re going to be the next Luke Skywalker, Neo, or Jake Sully: The One.

There’s something for everyone! (Except, yes, Mr. Burns and Colonel Killgore.)

And while extracting that cash, insidiously propagandizing all those groups for the superiority of #2.

Yes, it’s beautifully produced. But re-read The Republic (and The Prince) — that’s the whole point.

I mean cap-C Conservative. Do Conservatives and Republicans value honesty?

I ask in the context of Jonathan Haidt’s research into moral spheres, and which spheres are important to different political groups. (Blogged here and here.)

In response to Haidt’s $1,000 challenge for people to come up with additions to his five spheres, Tim Dean proposes the one that also came immediately to my mind when I first saw the challenge: truth/honesty.

“You should tell the truth” is obviously a widespread or perhaps universal moral intuition. (Though of course it’s not categorical — none of these intuitions is.) And it’s easy to understand how that predilection would have evolved.

I don’t know if truth/honesty merits a place in Haidt’s pantheon — there are complicated issues of interacting and overlapping moral spheres, touched on below. But I am curious about the same thing Tim Dean is:

Another interesting test would be to see how self declared liberals and conservatives respond to issues of truth/honesty. My guess would be that conservatives would rate truth/honesty as being more important than liberals.

My intuition: Conservatives would be shown to rank honesty differently depending on the context of the situation.

• In private, direct, and especially face-to-face situations, I think there’s a 50% probability that Tim is right, that conservatives would care more about honesty than liberals. Significantly more? Very low odds, I think.

• In the public realm — especially the realm of public debate — I think they would be shown to rank honesty very low relative to other spheres, especially group loyalty.

I would suggest that this is a result of conservatives’ greater concern for group loyalty. When the context is groups — promoting them, defending them — both the truth and the fairness realms are downgraded to the benefit of the loyalty realm. In a further (meta) downgrading of honesty, they would be expected to give lip service to honesty while failing to practice it. Think: “fair and balanced.”

This highlights the problematic nature of Haidt’s realms — the interactions between those realms — and the need for research that teases out those interactions. It also highlights the need to distinguish different groups’ moral weightings in different contexts.

Unfortunately, those necessities would/will turn Haidt’s fairly easy-to-understand model into a far more complex field of study — much as genetic interactions and epigenetics have done to genetics.

Will Wilkinson returns me to a subject of fascination to me — the different moral weightings employed by Republicans and Democrats — and points out new findings about the moral weightings of Libertarians.

To recap a previous post on research by Jonathan Haidt, as recounted in an article by Steven Pinker:

Republicans care equally about five spheres of morality: avoiding harm, fairness, group loyalty, respect for authority, and purity.

Democrats mostly care about only two: avoiding harm and fairness.

(I point out in that post that the Democrats’ two favorites basically characterize the gold standard of morality: The Golden Rule. Three of the Republicans’ favored tenets have nothing to do with — are often or mostly antithetical to — that rule.)

Now Haidt’s latest research gives us insights into Libertarians: they care less about all five. Will shares it in his post, and Haidt adds even more in a comment.

Will characterizes these findings by saying “libertarians are liberals who like markets.” I took him to task for his best-possible-light characterization, saying “A libertarian is a liberal without compassion or empathy.” Will quite rightly slapped me down: “I think Jon would insist that ‘less’ means something very different from ‘without.'” To which I — chastened — immediately agreed.

But the “lesser” fact remains–supported in spades by Haidt’s latest work, sneak-peeked in his comment:

Libertarians look much more like liberals than like conservatives on most measures, EXCEPT those that have anything to do with compassion, on which libertarians are lower than liberals AND conservatives.

But here’s where it gets even more interesting (for me at least). A commenter suggests that “libertarianism essentially amounts to is the political expression of autism.” Viewed with best-light beneficience, this is presumably not a pejorative statement but an insight into the autistic cognitive style and its emphasis on rationalism over empathy.

Will responds by suggesting “you should check out Tyler Cowen’s chapter on ‘autistic politics’ in his book Create Your Own Economy.”

Now it just so happens that that book was open on my desk at the time, open to that very chapter. This because I just referred to it in a post the other day. (Quite embarassingly, in fact.)

That book, as I said, is something of a paean to the autistic cognitive style, and that chapter suggests that the world would be a better place–there would be less wars, in particular–if we had more people thinking in that style.

By invoking Tyler’s book Will is essentially importing Tyler’s arguments into the present discussion. So it might serve to directly import some of what Tyler says in the referenced chapter — here with some comments in reply. (I’ve cherry-picked these as springboards for the ensuing discussion. If you want more you can buy the damn book yourself, like I did.) You can skip this section if you want to jump to the meat of the argument.

There is good evidence that people along the autism spectrum are in some measurable ways more objective than non-autistics.

This “in some ways” is I think crucial. It reveals some confusion about one rather paradoxical aspect of the autistic cognitive style: while autistics and some autistic-ish types tend to perceive the world in a “rawer,” less-mentally-mediated form (viz, Temple Grandin’s ability to notice bare facts about a stockyard environment that others miss, and–courtesy of Tyler’s book–Dugdate Stewart’s characterization of Adam Smith as having “a remarkably accurate memory for ‘trifling particulars.'”), there is also a predilection for abstracted structures and rule-based systems, systems that are far removed from those immediate perceptions. Unlike the perceptions, these systems cannot make any a priori claim to “objectivity.”

On the offensively ridiculous and adolescent notion of “objectivism,” I can only point here.

Autistics are attracted to simple and straightforward codes of ethics, applied universally to all human beings.

The implied approbation might be misplaced. Vis-a-vis “a priori,” above, there is nothing to demonstrate that these simplistic codes are more efficacious, or preferable, aside from the usual cop-out “common sense” defense.

Hayek argued that a rich and largely unplanned order can blossom when society is governed by a relatively small set of abstract rules, and, ideally, a constitution; you don’t have to share Hayek’s libertarian and conservative version of this blend to find this an appealing vision.

I would suggest that you do have to do so if you want to carry this maxim to an extreme logical conclusion.

Different kinds of human minds often have difficulty appreciating each other’s virtues, so social arrangements, and personal individual judgments, should be robust to this fact. That is still an argument for social and economic decentralization.

I don’t see a necessary connection between these two statements. There is some presumption suggesting that the latter follows from the former, but I have no idea what that presumption is. It’s easy to come up with reasonable arguments to the contrary.

What has gone wrong in many of the non-free societiies in today’s world is a lack of adherence to abstract rules of behavior and a lack of understanding of such rules as benefiical abstract mechanisms.

Putting aside my knee-jerk annoyance at this kind of “kids these days” class of pontificating: Tyler notably does not suggest that this shortage of autistic-style thinking is what has gone wrong in free societies such as, say, Western Europe.

But he doesn’t take long to imply it:

A list of the most successful societies in the world usually would include the United Kingdom, the Nordic countries, Japan, the United States, Canada, Australia, and New Zealand.

It’s not surprising to find most of the countries of Western Europe missing from this list, even though they are undeniably among “the most successful societies in the world.” (I mean really: Germany?)

Tyler ends the chapter by pointing to Russia as his negative example. It serves nicely as a phony proxy for the Western European countries that otherwise go so conspicuously unmentioned.

It is revealing that in a book centered around autistic characteristics, “empathy” does not appear in the index, and the only instance of “sympathy” refers to his two-page discussion of Adam Smith as possibly being on the autistic spectrum. As Tyler points out, the book Smith was most proud of was his Theory of Moral Sentiments — which book builds a spectacular abstract structure that “starts with the idea of sympathy.” It is in fact obsessed with the idea — the first two chapter titles take sympathy as their subject, and the word appears 182 times in the work. Wrestling with angels?

So what does this all have to do with “lesser empathy” and markets? Haidt speculates that they are related in the way that Will implies:

The lower levels of compassion, and higher levels of need for cognition and tendency to “systemize” rather than empathize, are probably related to the love of markets.

This makes sense to me as well. The abstract concept of markets holds more attraction for the autistic cognitive style than does empathy.

Which leads me to ask several admittedly rhetorical questions:

Does this fondness for abstractions explain Libertarians’ enthusiasm for a form of government based on abstract beliefs that is unexampled among large, thriving, prosperous countries on this planet — an enthusiasm that continues even though no country operating on Libertarian principles has emerged, much less surged ahead of the others?

Does it — combined with the lesser levels of empathy that Haidt demonstrates — explain the movement’s aversion to redistribution, even though every large, thriving, prosperous country engages in massive doses of redistribution? (There are no exceptions.)

Does it explain the continued predictions of disaster for those countries that the more extreme Libertarians have been warning us of for so long — even though those sky-is-falling scenarios have not occurred? (Over the long run — as libertarians will happily point out when it serves their rhetorical turns — things keep getting better.)

Does it, in short, explain an ideology that can only be described as utopian (lacking in any real-world exemplars), but that continues even though its leading proponents are painfully aware of the long, sad history of such utopian belief systems?

Do the lesser quantities of empathy that characterize the autistic and libertarian cognitive and moral styles (almost complete absence, in extreme cases) result in an almost autistic mind-blindness to the reality of successful societies and economies: that all those societies and economies employ policies rooted in empathy, policies that history has demonstrated to be the most economically efficient?

Demonstrably, because those are the societies that have thrived and prospered.

To ask it in terms of abstract theories: since empathy is clearly at the heart of humans’ ability to cooperate, and since humans’ ability to cooperate is what has put us at the top of the food chain (competition just makes that cooperation more efficient, overall), would it be surprising if policies that systematize and efficiently channel that empathy were also successful?

People are obviously unemployed because they want to be.

The Big Picture » Blog Archive » An Epidemic of Laziness?.

Update: More data here.