Statistics don’t matter

I read the transcript of Trump’s nomination acceptance speech. It was a grisly horror that painted a picture of an America that is a dystopian horror right now, which will be magically and instantaneously transformed on the very day he becomes president, and it was full of lies. The responses (and here’s a typical one) are all about how he twisted statistics dishonestly to make his rhetorical points.

I hate to say this, but the facts don’t matter. You can declare that unemployment rates are down, but those people who are out of work don’t care — and they’re the people Trump is yelling at. You can point out that he’s lying when he says crime is rising, but the people who have been mugged don’t care that they’re a statistic, and they’ll listen to Trump. Worse, people who haven’t been mugged will have been watching the crime stories on their local television station and Fox News, and have the perception that there is all kinds of lawlessness going on around them. Throwing national statistics at people who are counting one, two, many isn’t going to change their minds.

Furthermore, as Trump so ably demonstrates, all too often statistics are used not to identify a truth, but to justify a preconception. I was recently cooly informed that Muslims commit 98% of all terrorist attacks…never mind that the FBI has determined that 94% of all US terror attacks have been by non-Muslims. The guy is convinced that Islam is the source of all evil in the world, and he has a number that reassures him that his opinion is correct.

So here’s our terrifying problem: our little homegrown fascist is tapping into the fear and anxieties about their future of a significant number of people in the country. These fears are partly legitimate, and partly the product of a media that has been stoking them for years. You don’t reassure individuals by telling them that the average person is better off or that the trend lines are all rising — they don’t give a damn about averages when their problem is personal. In fact, waving tables of numbers and graphs at people to tell them their grievances are false is going to be more enraging than reassuring. And meanwhile, Trump will lie about the statistics and validate their gut feelings and pander to every prejudice they’ve got, and guess who they’re going to want to believe?

We’ll counter that by dumping a pile of actuarial tables on them. Yeah, that’ll work.

I’ll also point out that we’re seeing that policy doesn’t matter, either. For years, people have been voting against their own objective self-interest because demagogues have effectively whipped them into a froth of fear over religion, or guns, or abortion. See, for instance, Brownback’s Kansas. Progressive policies are almost always more appealing to the people, when presented without a label…but the media have effectively attached a lot of the hated positions to the Democrats (and rightly so — progressives should support women’s autonomy, minority rights, and oppose war and violence). Witness also Ivanka Trump’s bizarre speech at the RNC yesterday: she basically promoted the Democratic party platform and tried to attach it to her father. You can say he’ll do anything, and the disgruntled voters won’t care.

What are we to do? Data doesn’t matter, policy doesn’t matter. Politics is personal.

I see two strategies (I know there are more). The first necessary step is to recognize that the unhappy people who want Strong Man Trump to cure all their ills actually have legitimate problems — you cannot wave them away with a chart. So you can try to win them over by actually addressing their concerns, which would be ideal, except for the fact that one of their concerns is driven by raw, naked racism. Or you can simply write off that portion of the population as a regressive, deluded mess, and hope that the remainder are sufficiently numerous and motivated to vote, so you can get real political progress despite the unhappiness of that minority (which, if you’re a real progressive, you’ll then try to alleviate).

I think we should definitely be very afraid. The fascism is openly unmasked, and we’re facing a serious risk that it could be victorious in a few months. I dread waking up to newspapers that look like this.


Oh, damn. I did. That’s the front page of the Minneapolis Star Tribune this morning, featuring Great Leader surrounded by flags. And this is my nightmare.

Marked for death

We have this lovely old slippery elm in our backyard — it’s huge and thickly branched and towers over our house. This evening we noticed that sometime during the day, we had a visitor.


That is not a good sign. We’ll have to call the tree doctor tomorrow and get a diagnosis.

Don’t watch this if you’re at all squeamish about blood

I was just reading this post by Shiv about the expectations of femininity, making the point that there is a huge role for perception in how we react to sex-based phenomena — women are supposed to be hyper-emotional, even when they’re not, and we’ve all got this idea that extremely high level cognitive/emotional phenomena can be reduced to a simplistic measure of how much of which steroid you’ve got in your blood.

Here’s the thing: I’ve read the literature attempting to link estrogen to anything but physiology. It’s quite desperate. It’s EvoPsych levels of bad. The problem is that even if you do find a correlation, there’s a million and one moving parts–the biggest problem being “how do you measure levels of emotionality.” In the absence of anything remotely convincing, I remain skeptical of the exact role estrogen supposedly plays on emotional expression. It is far too convenient for these poorly designed experiments to support cultural stereotypes. The problem runs so deep that we’re asking the wrong question–how, exactly, does one measure the null hypothesis? Are you able to reasonably assess stoicism without the gender of the subject prejudicing your measurement?

Men don’t cry, but it’s not because testosterone dries up your tear ducts — it’s because men are mocked fiercely if they show that kind of emotion. Women are supposed to be emotionally expressive, but it’s not because estrogen somehow disinhibits emotional centers of the brain, but because they’re conditioned by years of expectations that girls are supposed to be this way.

But it doesn’t have to be that way, and different cultures have different expectations of gendered behavior, and that is what shapes us most. One of the best discussions I’ve seen of that is in Sarah Hrdy’s Mother Nature: Maternal Instincts and How They Shape the Human Species, in which she takes a cold hard anthropological look at the myth of the woman enslaved by her instinct for mothering, and shows that it’s bunk.

And then I ran across this commercial, which is the most fucking macho thing ever. There’s a lot of blood in it, so you may not want to watch it. It’s an ad for feminine napkins.

Who we are is partly a product of biology (but there’s more commonality between men and women than our sexist society wants to accept), but how we think of ourselves is a matter of attitude.

Mr Rogers says, “Look for the helpers. You will always find people who are helping.”

And then shoot them.

Amanda Marcotte writes about the performative worship of the police at the Republican National Convention. Meanwhile, in Miami, a black caregiver trying to help a severely autistic man who wandered away from a group home, was confronted by police who’d been called by someone who said the man was wielding a gun (it was actually a toy truck). The caregiver ends up lying in the street with his hands up and clearly empty, trying to explain to the police that he was trying to help the man, and please don’t shoot.

So shot him, obvs.

“When I went to the ground, I went to the ground with my hands up,” Kinsey said, “and I am laying there just like this. Telling them again there is no need for firearms. He is autistic. He has a toy truck in his hand.” …

“I’m like this right here, and when he shot me, it was so surprising,” Kinsey said. “I thought it was a mosquito bite, and when it hit me I had my hands in the air, and I’m thinking I just got shot! And I’m saying, ‘Sir, why did you shoot me?’ and his words to me were, ‘I don’t know.’ ”

Then they handcuffed him. Why, I don’t know. I think the new logic has changed from “if you see a criminal, shoot them” to “if you shoot them, you see a criminal.”

So that’s what a gay party is like

I’ve never been to one, and I’ve never even been invited to one, and this is the first account I’ve read of one, and it’s not quite what I would have expected. Hosting the party was Milo Yiannopoulos — perfectly legitimate, since he is openly gay — and…Pam Geller? Geert Wilders? Praise for Trump, excoriation of Democrats and other liberals, condemnation of jihadis, and…bad dancing?

I am so disappointed. That doesn’t sound open and liberal and joyful and fabulous at all.

Quantum paramagnetic looping heme molecules shaped our morphology!


The other day, I was briefly harangued by someone named William Peters on Twitter. It was a strange comment that was simultaneously pro-evolution while trying to imply some peculiar doubts — it was just off a bit. He wasn’t rude or anything, but I was intrigued and curious, so I dug a little deeper. He says many things that are slightly off.

Like having a simple mathematical explanation for the evolution of humans from fish.

The world is spherical, not flat! Cartesian math = 19th C. Polar co-ordinate math describes evolution of fish to man

Uh-oh. I see trouble coming. This is a common attractor in crank biology, the idea that form is only superficially complex, and that underlying it all is an elegant mathematical formula that will explain all, and of course, the author has discovered the secret equation that will cut through all the shallow trivia pursued by his peers. We saw it in Stuart Pivar, in Vincent Fleury, in Erik Andrulis. I can sympathize with the seductive power of geometry, but at some point, you have to look at the data…and the data trumps all your abstract theorizing every time.

But at least Peters isn’t talking about toruses, the usual crank attractor. No, this guy is into vacuum cleaner hoses, magnets, electricity, and…

Quantum biology has arrived #persanguinemnostrum Enjoyed your talk in NZ 2014

We have “quantum”, mission control, I repeat, “quantum”.

If you’re in the mood for some grandeloquent strangeness, check out his website, The opening announcement is this:





[Read more…]

Twitter rule: always punch down

Scalzi has some comments on the banning of Milo, and I particularly like this point.

It’s good that Twitter punted Yiannopoulos, but let’s not pretend that it doesn’t look like Twitter did some celebrity calculus there. Yiannopoulos and pals had a nice long run pointing themselves at all other manner of people they didn’t like, for whatever reason, and essentially Twitter didn’t say “boo” about it. But then they harass a movie star with movie star friends, many of whom are Twitter users with large numbers of followers, and whose complaints about Twitter and the harassment of their friend get play in major news outlets, and Twitter finally boots the ringleader of that shitty little circus.

So the math there at least appears pretty obvious from the outside. You can punch down on Twitter and get away with it, but don’t punch up, and punch up enough to make Twitter look bad, or you’ll get in trouble (after more than a day). Is this actually the way it works? I’m not at Twitter so I can’t say. I can say I do know enough women of all sorts who have gotten all manner of shit by creeps on Twitter, but who weren’t in a movie and had movie star friends or got press play for their harassment. And they basically had to suck it up. So, yeah, from the outside it looks like Twitter made their decision on this based on optics rather than the general well-being of their users.

This is exactly the rule set that fosters bullies, and is going to make the problem worse.