Say what?

Evolutionary psychology is criticized a lot for making it too easy to generate hypotheses which are unfalsifiable. You observe a trait x, therefore trait x must be adaptive; here's a story explaining why it's adaptive. For the record, I definitely do not agree that evolutionary psychology is unscientific. It is. It is a subject that studies testable hypotheses. However, it is also being outrageously oversold in the popular media, which are chock-full of unfalsifiable evolutionary just-so stories. For example, Christopher Ryan, author of Sex at Dawn, writes this:

Not only does one sperm compete against millions of your own, but millions of other men. Considering women’s immune system treats sperm as invaders, women select their mates on a cellular level regardless of what their instincts might tell them. (...) not only are women naturally promiscuous, but men are attracted to other men having sex with women (...) This is why women scream during sex--to attract more men to join the fight.

Color me unconvinced.

Rationalizing cruelty, again

James Lewis's essay in which he argues that the ideology of fascism is largely a complex way of rationalizing a desire to see others suffer is both very interesting and incredibly naive. It's interesting because the point that our beliefs should not be taken at face value because they are often designed to be fronts for some unspoken desires isn't argued often enough. It's naive because it grossly underestimates the scope of this very phenomenon. For example, Lewis seems to believe that Americans are immune to the impulse of gratuitous cruelty. Naive doesn't even begin to describe such belief. For another example, Lewis concentrates on fringe or extremist beliefs (fascism, communism, Islamic fundamentalism etc.), not realizing that ideologies designed mainly to rationalize cruelty are (and always have been) very much part of mainstream. For instance, a lot of religion-based morality, or the way we think criminals should be punished, are really just sophisticated ways of allowing people to act on their sadistic impulses while still believing they're fundamentally good people ("C'mon, I'm only stoning him to death because it pleases God, not because I like stoning people to death!")

A graph is worth a thousand words: World War II and the population of Warsaw

(Source: Wikipedia.)

Stunning links

Five evidence-based ways to make yourself happier in the long run.

What's a "bond bubble," and why should we be worried about it?

Complicated mechanisms explained in simple animations. How does a sewing machine manage to knot stitches without releasing the thread? One quick look and you will know.

A moving documentary about Andrew Wiles and the quest to prove Fermat's Last Theorem.

Against banning election betting.

What would happen if businesses stopped maximizing profits and settled for "satisfactory profits" instead?

Who should exist?

Is suicide bad?

A number of months ago I wrote about gun buybacks, arguing that this policy cannot possibly do any good whatsoever. It turns out that I was dead wrong: it certainly can, and it probably does. According to a study done by Australian economists Andrew Leigh and Christine Neill, a nationwide gun buyback program implemented in Australia in 1997, which reduced the stock of firearms by about 20%, has lead to a decrease in the number of firearm suicides of as much as 74%, which translates into about 200 lives per year.

When thinking about the effects of gun buybacks on various outcomes, it never occurred to me to consider suicide--a huge omission on my part. It's clear that there should be an effect, and now we have evidence suggesting that it's there.

OK, so gun buybacks reduce the number of suicides. Is that a good thing? Why should we assume that suicide is always a bad thing?

Actually, we don't have to assume this to believe that this particular consequence of gun buybacks is beneficial. I for one don't think that every act of suicide does more harm than good. Death isn't always the worst thing that can happen. No one knows your life as well as you do, so if you think it's not worth living, you are probably right. (Note: some ethicists argue that suicide need not be a bad thing using the volenti non fit iniuria principle. I don't think this is a good argument. It works as an argument for suicide being legal, not beneficial. It's legal to drink two pints of Jameson a day; few people would argue, however, that it's a good idea to do so.)

I do think that most suicides do more harm than good though (in a social welfare meaning of the phrase), for two reasons. First, suicide victims are often dead wrong when doing their mental cost-benefit analysis trying to figure out if carrying on is worthwhile. Most of the time those mistakes are due to (very much treatable) mental conditions and/or influence of drugs. I cannot find the statistics at the moment, but I do know that in some societies very large number of suicides are directly caused by episodes of alcohol-induced psychosis, for example. Second, a suicide victim does not affect only themselves, but those around them as well, often causing them considerable pain and suffering.

Aside from all this, here's why we can be fairly sure that saving those 200 lives a year is actually a good thing: the decrease in suicides by firearms did not lead to an increase in suicides by other methods. In other words, most potential suicide victims who were all of a sudden cut off from access to guns did not use a substitute. This I think is pretty powerful evidence that many suicide decisions were incorrect (from victims' point of view), and that many acts of self-destruction were carried out only because they were too easy to execute.

Beautiful experiments in social science: Reversing the implication

People tend to make the logical error of reversing the implication, i.e. believing that "If A then B" implies "If B than A." A clever experimental design devised by psychologist Peter Cathcart Wason showed how often that mistake is made. Here's the design:

You are shown a set of four cards placed on a table, each of which has a number on one side and a colored patch on the other side. The visible faces of the cards show 1, 2, red and blue. Which card(s) should you turn over in order to test the truth of the proposition that if a card shows a 2 on one face, then its opposite face is red?

Answer given by more than 90% of participants: 2 and red. Correct answer: 2 and blue. The incorrect answer clearly stems from reversing the implication.

The opposite of truth is just as good as truth

One one of the Polish political blogs I've recently stumbled on a quote from a book written by Viktor Suvorov, a former member of the Soviet secret special-ops forces called Spetsnaz, who defected to the UK:

The most important thing is to confuse the enemy. If the operation is done with lots of Spetsnaz units, you pretend that there's really just a few of them. If the forces you're using are small, you should make the operation look large. (...) Suppose a Spetsnaz unit is to destroy a number of targets, all of them situated more or less on a straight line and in a densely populated area (for example, a pipeline, a power supply line, a highway, a few bridges, etc.). In such situation, the detonators on the explosives placed first are set to long delay, and then that delay should get shorter and shorter on the devices placed later on. In this way, the timing of the explosions will be opposite to the direction along which the unit was marching. If the unit was traveling from East to West, the explosions will start on the West first and then move Eastwards, deceiving the enemy into thinking that the unit was traveling Eastwards as well.

Now if this is actually how Spetsnaz went about deceiving the enemy then they were in dire need of an intro to game theory class. The deception strategy of telling the exact opposite of the truth can only work once; if you repeat it, your enemy will quickly catch on to what you're doing and you might as well tell them the truth outright. How long do you think you'd last in a poker game if you made large bets every time your hand was weak, and small bets every time your hand was strong? The only deception strategy that is viable in the long run is to behave in a way that does not exhibit any discernible patterns. In other words, randomly.

That's right!

(Extremely funny, though probably NSFW.)

Retroregulation

First, a great quote from Eric Falkenstein:

You might remember the SEC in penalizing famous frauds like Madoff, Bayou, and Wood River--after they were exposed as frauds by their own investors. These are the police who show up after you've tackled and hog-tied your intruder, and then take credit for putting him in jail.

In his post, Falkenstein also talks about the "flash order crash" of May 2006, when systems started failing because institutional restrictions on how trades are supposed to be routed were outdated and unprepared to deal with flash orders. The SEC's proposed solution: ban flash orders.

This illustrates two worst features of regulation: that it's always reactive, and that whenever that feature leads to widespread problems, the government answers simply by trying to outlaw the market invention that has exposed regulation's weakness. The first feature is unavoidable. The market will keep inventing things that regulators do not anticipate. But maybe the second one is not. Perhaps a scheme can be designed whereby the regulators are banned from banning things and are instead forced to improve their regulations when those are exposed as inadequate.

Yoga people

I've never taken a yoga class but I see tons people coming in and out of them all the time. Not once did I see a yoga class taker who would be overweight or anywhere close to that. Those of my friends who take yoga classes themselves tell me that this observation holds as well inside one as it does outside. If it's true in general, a likely conclusion is that yoga is much less effective in reducing weight than regular exercise (cardio plus lifting weights).

People work out for two reasons (which aren't mutually exclusive): to lose weight and to stay in shape which improves long-term health. You have to exercise very hard in order to see results in terms of losing weight; if all you want is to stay in shape, on the other hand, an equivalent of 30 minutes of cardio a day is more than enough. People who don't have to lose weight are therefore content with lighter exercise. If you could lose weight doing yoga, you would see people trying to do just that (as you do in regular gyms). The fact that you don't most likely means that you can't, and yoga is therefore chosen only by people who don't have to work out very hard.