Scott Alexander's SlateStarCodex is one of my favorite blogs. Much of it involves trying to really look at what's going on, ignoring noise and assumptions in favor of data and rational evaluation.
Recently he did this on the topic of whether Donald Trump is more racist than any past Republican candidate for president. His conclusion - argued persuasively and at length - is that no, Trump is not the candidate of white supremacy and the hysterical claims that he is are doing harm. The article is not a defense of Trump, but just a challenge of one particular narrative.
It's a terrific article, but as I read it, I had the vague feeling that at least some of his points could be refuted. But a superficial googling on the article didn't show that anyone had tried to refute any of it. There were just articles lauding the piece.
As I thought about the article there were a few things that began to bother me, and since Alexander turned off comments on the post in order to avoid chaotic flame wars, I'm just going to discuss them here.
Alexander makes a lot of good points. Take, for example, Trump's famous statement "When Mexico sends its people, they’re not sending their best. They’re not sending you. They’re sending people that have lots of problems, and they’re bringing those problems with us. They’re bringing drugs. They’re bringing crime. Their rapists." Alexander points out this is not actually racist because Trump is specifically saying that there are good Mexicans, and that it's just those aren't the ones immigrating to the U.S. It's not actually racist to say, "all the good Mexicans are living in Mexico," even though it's clearly untrue.
Alexander points out that while Trump's obsession with birtherism seems racist, Trump actually believes all sorts of crazy conspiracy theories, many involving white people. He points out that, outside of the liberal bubble, Trump has made many statements praising U.S. diversity, like every other politician. He skillfully takes the "racist Trump" arguments apart piece by piece, leaving you with Trump as an asshole whose racism is closer to that of Robert Dole than Adolph Hitler.
He's probably right. And yet, as I thought about the article later, I began to think of Rodney King.
Rodney King, as you may recall, was videotaped being savagely beaten by the police in 1991.
Watching the video, it was clear that King had been beaten mercilessly for several minutes by many cops while he tried to escape. But when the case went to trial, the defense slowed that video way down and convinced the jury that every time King's arm flailed out as a baton hit him, every time he tried to get on his knees to crawl away, that he was in fact lashing out and trying to attack the cops.
This is the danger of granuality. Anything, when looked at closely enough, can lose its shape. It's like those paintings that look like people in a park from a distance and look like a bunch of dots of paint up close.
Alexander is looking at Trump on a granular level, and on a granular level, you can prove an awful lot.
As much as Alexander is admirably trying to look at all the evidence, he's still just taking a small section of the immense number of things Trump has said and using it to push back on a narrative.
Alexander does, in fact, understand that. In point 17 of his post he shows exactly how all arguments against an insupportable theory can sound, to a believer, "weaselly." If you believe Trump is a racist, you can reject pretty much any arguments to the contrary. And if you're living in the liberal bubble with me, where you basically keep hearing the same five quotes from Trump on an endless loop, it's easy to see the case for racist Trump as a slam dunk.
I'm not saying Alexander's granuality means he's wrong. I'm just saying it's something you need to pay attention to, because he's a very persuasive writer and as anyone familiar with Malcolm Gladwell knows, persuasive writers can persuade you of things that are not entirely true.
As I continued to mull his piece over in my mind, certain other things began to bother me, like this:
13. Doesn’t Trump want to ban (or “extreme vet”, or whatever) Muslims entering the country?Yes, and this is awful.But why do he (and his supporters) want to ban/vet Muslims, and not Hindus or Kenyans, even though most Muslims are white(ish) and most Hindus and Kenyans aren’t? Trump and his supporters are concerned about terrorism, probably since the San Bernardino shooting and Pulse nightclub massacre dominated headlines this election season.
This seemed fine when I first read it, but then it began to seem really, really wrong. Because he is basically saying that racism is tied specifically to skin color. That if Trump is more afraid of "white(ish)" Muslims than darker Kenyans, then somehow he can't be all that racist.
As much as I respect Alexander, this is blatantly stupid. Because racism is not a visual thing. In the past, Americans have been racist against Irish and Italians, among others. My mom had a friend who once said she could deal with her daughter dating a black guy, but would disown her for dating an Asian, even though Asians are generally somewhat lighter (the reason having to do with a hatred of Japanese born during World War II).
And, of course, Hitler hated Jews more than any other race even though Jews were often physically indistinguishable from Germany's non-Jews. I don't know how Hitler felt about Kenyans, but I'm pretty sure he spent less time worrying about them.
And yes, hatred of Muslims has been exacerbated by Muslim terrorists. But racism doesn't have to be based on nothing. It is a matter of taking some aspect of a racial group, whether real or imagined, and applying it to all people of that group. Some African Americans are criminals. Some Jews are greedy bankers. Some Poles are stupid. If you think that subgroup represents the whole, that's racist.
After all, there are tons of white terrorists in America, shooting abortionists or blowing up buildings, but we don't take actions against all white people because some of them are terrible.
I wonder whether a 1930s version of Scott Alexander be able to examine Hitler on a granular level and conclude that he wasn't really more anti-Semitic than other German politicians? After all, the level of anti-Semitism in 1930s politicians was pretty high.
While point 13 is Alexanders worst argument, others are also problematic. For example, the fact that Trump believes in all sorts of conspiracy theories, only some of which are racist, doesn't mean his belief in birtherism isn't still racist; if I believe all drug dealers are black, and I believe moon people have invaded earth, the second belief doesn't make the first one less racist.
And yes, it's true that people have physically attacked Trump supporters, and that's terrible, but there's a difference between attacking someone for how they voted and attacking someone for who they are. If a Muslim is a wearing a Clinton button and someone attacks her you can say it's equivalent, but if she's just wearing a hijab then she's not being attacked for her vote, but for her race and religion. There's a fundamental difference between saying "fuck you, Trump supporter," and "go back to Mexico, wetback."
Alexander is very persuasive, and I am willing to admit there's a pretty good liklihood that Trump isn't the most racist Republican out there and that the actual white-supremist part of his constituency is tiny and is only seeming significant because the media has glommed onto the white-nationalist story and is giving fringe groups way too much attention.
But I'm still nervous, because I'm not convinced that this country hasn't opened the door a crack for racists now. That there is a new acceptance for racism, and sexism, and homophobia that Trump is helping along, and that he really is more dangerous than a typical Republican not just because he's corrupt and incompetent but because he is going to at worst institute racist policies and at best just let racism grow without challenge.
There's an old expression: if it walks like a duck, and it quacks like a duck, it's a duck. For a rationalist like Alexander, this is untrue. If it walks like a duck and quacks like a duck, you can only say for certain that it is something that shares certain characteristics with a duck. It's an important perspective, because the truth is, common sense is not always sensible, few things are truly self-evident, and to assume is, as the saying goes, to make an ass of you and me.
I admire Alexander's attempt to rationally look at that quacking thing to try and figure out if it's really a duck. I just wonder how many data points he would need before finally admitting that, yeah, what we've got here is a duck.