George Floyd, Tony Timpa and the dimensions of discourse on the police

 This is going to be one of those posts in which I tell you that everybody on every side is wrong because our political system is blinkered.  I suppose I have telegraphed that.  However, what good is having a blog that nobody reads if I cannot use it to vent my irritation through math?  So here goes.

Eventually.

First, I'm going to tell you a story about a couple of friends of mine from college, and how they got stopped and harassed in a "driving while black" incident.  These stories all have a relatively similar flavor, but the particular joy of this tale of police-imposed woe is that it allows me to bash Texas, because Texas is a shithole country, and fuck Texas.  Seriously.  Fuck Texas.  Secede already.  And take your Texas shoulder-chips with you.  None of us would have a problem with you if it weren't for those shoulder-chips, but those Texas shoulder-chips are really fucking intolerable.  So secede.  Then maybe the rest of us can have a governable country.  So long, Texas, bu-bye.

As you can no doubt infer, this story takes place in Texas, which needs to be fucked.  After being sterilized, because there are too many Texans, and we need fewer Texans.  As an advocate of an accurate Census count, I don't want the state cheated on the numbers, but if they had fewer people to count, wouldn't the literate portion of the country be better off?

Where was I?  Oh, right.  A couple of friends of mine from undergrad decided to take a road trip from the West Coast to Austin for South By Southwest, which is a big music and film festival.  If Texas does secede, America could keep Austin.  We can work out something like with West Berlin in the Cold War.  The rest of Texas really does need to fuck the fuck off.  But Austin?  Whatever.  Anyway, my friends, both African-American, were driving through West Texas.  A hair over the speed limit, but as you probably understand, cops are generally not supposed to pull you over if you are within reason of the speed limit.  You know where this is going, right?  Keep in mind that this is the era before cell phone cameras, 'n stuff.  Anyway, they pull over my friends.  Both remain calm and respectful, as they must, 'cuz, you know.  The Texas State Trooper asks the driver to step out of the car right off the bat.

You know this is going nowhere good, right?

So the driver steps out of the car.  He obeys the cop, calmly and respectfully.  The State Trooper takes the driver out of line of sight and earshot of the passenger.  At this point, we have one witness only, but let's take him at his word.  I do.  Here is his description of what happens.

The State Trooper demands to search the car.  This is clearly unconstitutional, and the driver knows it.  The driver politely asserts his Fourth Amendment rights.  The State Trooper tells him that if he (the driver) does that, the State Trooper will hold the driver and passenger there and call out the dogs.

This, mind you, for our driver going a couple of MPH over the speed limit.  Probable cause?  Not so much.  If the Trooper had that, he wouldn't have "asked" for permission to search the car, under threat of hours under the West Texas desert sun, and dogs.

Driver relents, knowing there are no illicit items in the car.  The Trooper asks the passenger to leave the car.  Passenger politely does so.  Trooper frisks both.

Keep in mind, this is for something that barely counts as a moving violation, and neither the driver nor the passenger did anything to suggest any other crimes.

Trooper then goes through everything in the car, pulling things out willy-nilly, food included.  This is a road trip.  There is fruit, including grapes.  Motherfucker squashes the fucking grapes.  I guess he was worried that grapes might have been used to smuggle, I don't fucking know, he was being a fucking asshole, OK?

Driver and passenger, while harassed and molested, are not assaulted, and are left with their shit on the side of the road, grapes and other shit smashed, and cop goes on his asshole way.

My buddies get to Austin, and run into a couple of their friends at a show.  They made the same road trip.  South By Southwest is a big deal, and a lot of people do it.  It was Spring Break.  Anyway, it turned out that this other pair of African-American young males had the same thing happen to them on the same fucking stretch of fucking back-ass Texas road.

Fucking Texas, m'I right?

So here's what you're thinking.  This is just a thing that happens, right?  Driving while black.  This is why we have a name for it.  We name a thing when it is sufficiently common that it needs a name.

Well, I kinda lied to you.  Yeah, I know I make a big deal about honesty, but it was for your own good.

I was the passenger in that first car.  The driver was my roommate.  We're both white.  So were the two we met in Austin.

Everything else happened exactly as I described it.

Actually, wait...  Were we white?  Sixteen great-grandparents between my old roommate and me.  How many of them were white?  Temporally speaking, zero.  You see, at the time, they weren't considered white.  When they emigrated to the US, they were part of a group that was considered "other," and not white.  It wasn't until some indeterminate point in the mid-20th Century that the my... let's say, "ethnicity" gradually became "white."  And might get un-whited again at some future point, 'cuz race is the dumbest fucking thing ever.

Oh, and while the phrase, "white supremacy," has been used in some very strange ways lately, there are actual people who subscribe to an actual ideology called, "white supremacy."  And while you probably think of me as "white," those people who call themselves "white supremacists" would very much like to murder the fuck out of me, my old roommate, and the whole fucking lot of us.

So you see, what I told you was true.  From a certain point of view.

Where was I?  Oh, right, my driving while black story, except that I'm not black.  I'm white.  Kinda.  Today.  Until enough people go back to not thinking of me as white, and I'm "other" again.

So what was the point of my little story?  Had the four of us been black, this would have been a "driving while black" story.  And notice how cleanly the narrative fits within a "driving while black" framework.  Had I been African-American, I probably would have interpreted the incident through the lens of race, and attributed racial bias to the State Trooper, particularly given... fucking Texas.  Which I just hate anyway.  At the time, I wasn't thinking about it in the context of "driving while black" because... why would I?  The more national dialog shifted to race, the more I thought about how various incidents can be interpreted in different ways, and this kind of incident comes to mind, but that brings me to the basic concept of confirmation bias.  I'll come back to this repeatedly, but if you have a belief about how the world works, you will notice evidence for it, and disregard evidence against it.

Now every empirical analysis I have seen indicates that frivolous, douchey traffic stops are statistically more common for African-Americans than for white people.  But that's a statistical pattern, and finding that out requires doing statistical analysis with proper sampling procedures, and "proper sampling procedures" does not mean just seeing what stories make the news, nor asking your friends and acquaintances.

Douche-y cop behavior can exist independently of race, and sometimes, it's geographic.  Like fucking Texas.  West Texas in particular.  The getting pulled over in West Texas so they can search your car thing?  This is such a thing that songs are written about it.  See, in particular, Ani DiFranco and The String Cheese Incident.  Yeah, the name, but they're an awesome band.  Why does Texas do this?  It's a shithole country.  Weren't you paying attention?  Yet this observation does not mean there are no racial biases.  Every quantitative analysis I have read of police harassment has shown that it is more common for African-Americans.

Were I to base my analysis on my experience in West Fucking Shithole Texas, I might reject such claims, claiming that phrase that the antiscientific postmodernists of the left now love:  lived experience.  And had I lived that experience as an African-American, I might not have bothered reading the literature, taking that lived experience as the only evidence I needed, completely missing the fact that the exact same thing can happen to a bunch of white boys from a snooty, West Coast college on a Spring Break road trip.  Because Texas is a failed experiment in inbreeding.  (Or... a successful experiment in inbreeding?  What counts as success?  I mean, they did manage to inbreed a state...)

And joy of joys, Tony Timpa was killed by fucking Texas pigs.  But it's still going to take me a while to get to him.  Right now, I just want you in a mindset in which you understand that you perceive things in a statistically biased way because statistically, you probably aren't a statistician, so you don't spend your time correcting for perceptual biases the way I do.  Yeah, I'm an arrogant fuck, but I also work harder than basically anyone I know to correct for cognitive biases because I study fucking math.  Math works.  Learn it, live it, know it, plagiarize it.  (Don't criticize it.)

Good thing I've never done any of that shit, and didn't have any of it in the car that day, right?

Anyway, let's talk about why you believe what you believe about the police.  What are your perceptual biases?  When a black person is killed by the cops, it gets covered by the national news.  When a white person gets killed, it doesn't.  Why?  We need to address how journalists construct the news.  This is not a dishonest or conspiratorial process.  OK, Fox, and some other institutions just fucking lie, but there are more subtle problems.  Consider.

Suppose a city with N people has X home break-ins in one week.  If X/N is sufficiently small because N is sufficiently large, then there may be no story any time a break-in occurs, not because X is a "small" number, but because N is a "large" number, all of this being a matter of expectations.  In other words, in a big enough city, we expect home break-ins, so some number, X, may be a nothing-to-see-here-move-along-folks non-story.  The press won't report.  On the other hand, what if the exact same number of break-ins occur, but all along one street?  Statistically, we see the same picture at the city-level, but now there's a story, because now there's a narrative, because not just one neighborhood, but one street?  You've got a story.  At that point, maybe some local paper would actually cover it, if local papers still existed, but oh fuck it, let's just get on FaceBook.  Fuck FaceBook right along with Texas.  They can fuck each other, and produce no offspring, so that's fine.  Fuck them.

Anyway, the point is that whether or not there is a story to a break-in becomes a question of whether or not there is a narrative into which to fit the break-in.  If there is a narrative, cover it.  No narrative, and each one is a disconnected, random event, unworthy of news coverage.

That mentality is a source of news bias, and one about which you likely never think, because unless you heed my advice, you don't spend much time thinking about the stories that don't get covered, even though I rant constantly about what I call "the paradox of news."  If you see it, that's because it is a deviation from the norm, but seeing the story tricks your statistically disinclined brain into thinking that it is the norm. 

So a cop kills a black man.  You have a pre-existing narrative.  Racism.  What happens when a cop kills a white person?  There's no narrative.  So, it's a disconnected, random event.  Hence, unworthy of news.  That is why you never hear about it.  That is why you never hear about Tony Timpa.

Of course, there is another factor.  There is an active interest group pushing news organizations to cover every shooting of an African-American.  There is no group pushing for news coverage of shootings when the person is white.  So watch how this works.  BLM demands that the national news media cover any police killing of an African-American, even if it happens while the African-American is actively engaged in attempted murder, so the shooting occurs to save civilian life.  The media comply.  No one advocates news coverage for police killings of anyone who is white, like Tony Timpa.  This creates a narrative.  The sample of police killings you observe, then, are killings of African-Americans.  The interest group that demands news coverage when African-Americans, but not white people, die, then point to the disparity as evidence of bias.

This is the very definition of a circular argument, yet without an understanding of sampling bias and cognitive bias, those making the circular argument will not even be aware of what they are doing.  They will sincerely believe that they are engaged in a fight against racial killings.  That is how pernicious these logical fallacies are.  Cover George Floyd.  No pressure, on Tony Timpa.  See?  George Floyd!  He's African-American.  I'm not seeing news stories on this happening to white people, are you?  The only possible explanation is racism!  Otherwise, you'd see something like Tony Timpa, and since you don't see anything like that, it must be racism!

There is an answer, of course.  Math!  Statistics!  And this is where it gets very messy, and I ain't solvin' this for you.

Suppose there were no biases in the rate at which police kill civilians.  What proportion of the people killed would be African-American?  Would it be their proportion of the population?  Empirically, that is not what we see.  According to activists, that is absolute, 100% proof-positive of racism, end of discussion.

So try this.  Essentially half the population is male.  What if there were no gender bias in the rate at which police shoot people?  What proportion of people killed by cops would be men?  50%?  It's over 90%.  Is that proof-positive, absolute, 100%, end of discussion that cops are out to kill men because they are misandrists?  Um... no.  That's the exact same argument, though.

Would it be plausible that a cop would be more likely to pull the trigger when faced with a man, in the same circumstances?  It's plausible.  But there are other factors in that statistical discrepancy.  A deviation from the baseline of the population is not, on its own, proof-positive of police bias.

That doesn't stop being true when we move from gender to race.  This brings us to one of the most controversial pieces of research on the topic:  Roland Fryer's "An Empirical Analysis of Racial Differences In Police Use of Force."

Non-lethal force?  Everybody's analysis, that I know, shows a racial discrepancy.  African-Americans are more likely to face non-lethal force than white people.  Where Fryer raises controversy is his argument that the difference between the rate of lethal force used against African-Americans and white people is the result of context.  Is he right?  That's a really hard question, and I'm not going to answer it for you.  And the other problem is that unless you have some understanding of econometrics, crime rates, and the technical materials, you're going to have a hard time assessing his work on your own.

But if you are a lefty, you really want to reject it because he is telling you that you are wrong, in part.  That while non-lethal force is more likely to be used against African-Americans, lethal force is not, and you don't have the technical expertise to evaluate his methods unless you are trained in econometrics, but that claim is going to make you ideologically uncomfortable, while running up against the fact that you observe a lot of George Floyds.

And don't observe any Tony Timpas.

But I've already explained that.

And I am so not telling you what I think of his econometrics.

But here is a consequence.  In your mind, and in many peoples' minds, this will connect the issues of racial bias in policing with the concept of excessive force in policing beyond the data.  These are separate dimensions.  Related, but separate.  They aren't fully orthogonal dimensions, but nor are they as correlated as the current left-wing talking points.  In political science, we love 2-by-2 tables.  So let's have one.  Here.  This should demonstrate the problem.

So consider the table above.  The top row is a little odd.  In the top-left box, I just put the klan.  Yeah, there are people who want the cops to be racist, and think that racist killings are justified, but they're mostly not a part of the discussion.  OK, fine.  Tucker Carlson, but fuck him.  The top-right is a weird box right now.  We've got these idiotic people who want to abolish the police, we've got people who evaluate the justifiability of police actions, not based on context and circumstances and whether or not a violent crime is taking place and needs to be stopped, but on the race of the person being stopped... this has gotten weird.  I'm gonna leave that box blank because filling it in is going to create an issue separate from the purpose of the post.  The bottom row is the interesting row for the purposes of this post.  The common interpretation of George Floyd is that his killing was not just murder, but clearly racist.

So what about Tony Timpa?

If you haven't watched the video, I'd like you to watch it, and in your mind, I'd like you to imagine that he was African-American.  And imagine how you would have reacted had he been African-American.  And imagine how BLM would have reacted to that video had Tony Timpa been African-American.  And imagine how the national media would have reacted had Tony Timpa been African-American.

And if anyone is going through that video, trying to find minor differences between the Tony Timpa death and the George Floyd death to justify the different reactions we have observed, I'd like you to imagine those words coming out of Tucker Carlson's mouth, in reference to a black Tony Timpa.

Have I made my point?

What happened to Tony Timpa was not justified.  We have people going around saying that cops can't even shoot when the perp is actively attempting murder, with a deadly weapon, but we're supposed to accept what happened to Tony Timpa?

And the logical issue created by Timpa is that his death cannot be attributed to racial animus.  Because he was white.

Of course, the point is that it isn't, and shouldn't be about listing anecdotes.  In aggregate, figuring out the role of race requires grappling with the complicated econometrics of Roland Fryer, and no one wants to do that.  Not even me.

All I'm willing to do is to say that it ain't that simple.  And I am deeply bothered by what we are ignoring in the simplification.

And here's what happens in that simplification.  See that 2-by-2 up there?  What happens when we chop off the right-hand column?  It becomes a single column.  A single dimension.  Race and police use of force become one dimension.  Everything collapses onto a single dimension, and the left-right conflict collapses such that evaluations of police use of force become indistinguishable from evaluations of racism among the cops.

To be sure, there are racial biases, but you can't study those biases through press anecdotes for the reasons I have already discussed, nor through your "lived experience," for the reasons I have already discussed.  Those biases must be measured through social science, and the social science is difficult and messy.  Non-lethal force, harassment... those biases are consensus.  Lethal force?

Um...

That's a hard question.  Unless you also think that the cops hate men and are out to kill men because they're all feminazi misandrists.  The math is actually complex, and until the econometrics around Roland Fryer's work are addressed in a rigorous way, we need at least to accept that whatever biases there are, are smaller than what is being portrayed, and might be statistical illusions, while the drive to cover only African-American deaths leaves the Tony Timpas out in the cold.

But I guess he doesn't care about the cold, 'cuz he's dead, and you don't get to call me out on that sick joke, unless you already knew about Tony Timpa.  I'm the one raising attention here, not you, so fuck off.

I'm going to conclude with one more remark about Tony Timpa.  Because I am done treating Ibram X. Kendi like a serious figure.  Remember the logical and moral framework of Ibram X. Kendi:  everything in society is either racist, or anti-racist.  So, when the cops killed Tony Timpa in a way that paralleled George Floyd, were they being racist, or anti-racist?

You can't make a case that it was racism.  And remember how Kendi defined anti-racism:  reducing racial inequality.  If the state of the world is police violence against African-Americans, then by murdering a white guy, they reduced racial inequality.  See?  The murder of Tony Timpa was an act of anti-racism.

So, you can either accept this premise as a defense of the murder of Tony Timpa, or recognize the emptiness of Ibram X. Kendi.  And of course, as long as people subscribe to Kendi and his nonsense, they can never be allowed to know about Tony Timpa and that lower-right hand box.  Otherwise, they would be forced to confront the question of whether his murder was racist or anti-racist, and that's not a question Kendi wants.

Finally, some music that is not typical for this blog.  Yes, some of those that work forces are the same that burn crosses.  It does not logically follow that you will observe racial biases in killings.  One would reasonably hypothesize, but that's a hypothesis, not a logical derivation.  And if you tell me that I must write about these matters in a specific manner, fuck you, I won't do what you tell me.  Academic freedom, motherfucker.  That's what he's saying, right?

Post-script:  Are you bothered by anything I said?  If so... the only place you would be justified would be those totally over-the-top eugenics jokes.  If you got your hackles up by the political analysis while chuckling at the eugenics jokes... got you!  Then again... Texas does pretty much suck, doesn't it?  Heh-heh.  Ain't I a stinker?