Hacker News Reader

I understand the issue is the loose definition of "innocent" - if you take it to mean someone shot by police who is absolutely innocent of any charge - or if you define it as someone who was doing something criminal or suspect but the situation did not initially warrant their shooting. Hence the emphasis on deescalation training.

Consider a hypothetical: two separate scenarios of kids shoplifting, one white the other black: the white kid gets called out by an officer, is pursued and apprehended without arms drawn, the police officer writes-up a citation and the kid spends the night in jail as a warning before being released. A black kid does the same thing, but the officer decides to draw their sidearm and orders them to freeze - the kid turns around to look and is immediately shot by the officer (search online for "shot for turning around police"). The officer can (at present, justifiably) say they feared for their life because they thought they were about to be charged or expected their target to draw their own firearm. Or take the same scenario but both kids do actually fight back somehow - I believe the average white police officer would still be more hesitant to use lethal force against a white assailant than a black one.

I won't deny the usual statistics ("blacks commit more crimes per-capita", etc) but don't think that's where I believe the problem is. I'm not content with the sentiment that "black lives matter" - I'd much rather it be "criminals' lives matter" (I'm aware of the straightforward racist implication - that's why I dare not say it too loudly).

Yeah, I was purposefully vague about "innocent" because I don't know what the most interesting, precise definition would be. I agree with you that comparing results to crimes is important in assessing the role race plays in these encounters.

> I believe the average white police officer would still be more hesitant to use lethal force against a white assailant than a black one.

I thought I heard about studies into this, but I can't seem to dig them up. I'll edit or reply again if/when I find them.

> I won't deny the usual statistics ("blacks commit more crimes per-capita", etc) but don't think that's where I believe the problem is. I'm not content with the sentiment that "black lives matter" - I'd much rather it be "criminals' lives matter"

If the discriminant is criminal behavior and not race, then (by definition), the issue isn't one of race (at least not in any meaningful way). It would be good to identify if this is the case or not, so we can focus our national debate on how to appropriately deal with crime and not on racism.

This is not what the statistic or the study the OP pointed says.

The study is clear in its conclusions: in the hypothetical situation you just presented it is the white kid that has the higher probability of getting shot/killed by the police.

In a nutshell what the study based on the data says is: After you get stop by the police, you are more likely to get shot/killed if you are white than if you are black.

You didn't research very hard.

From the Washington Post[0]:

> White people make up roughly 62 percent of the U.S. population but only about 49 percent of those who are killed by police officers. African Americans, however, account for 24 percent of those fatally shot and killed by the police despite being just 13 percent of the U.S. population. As The Post noted in a new analysis published last week, that means black Americans are 2.5 times as likely as white Americans to be shot and killed by police officers.

If you look at fatal police shootings so far this year[1], the numbers are the same.

You should provide a link to the Harvard study you mentioned.

Here[2] is one from Yale which someone once touted to me, which supports your claims. But they only used data from a limited number of police departments in Houston, TX. I'm not sure why anyone would look at that and not the national numbers. In the paper, of their own methods they say, "In essence, this is equivalent to analyzing labor market discrimination on a set of firms willing to supply a researcher with their Human Resources data!".

There are many places to find this information online. It is a fact: black people are shot and killed more often by police than white people.

[0] https://www.washingtonpost.com/news/post-nation/wp/2016/07/1...

[1] https://www.washingtonpost.com/graphics/national/police-shoo...

[2] https://law.yale.edu/system/files/area/workshop/leo/leo16_fr...

I believe The Guardian also ran a project to count people murdered by law enforcement because there are no strict requirements to record this across the country:

https://www.theguardian.com/us-news/ng-interactive/2015/jun/...

Obligatory: "Murdered" != Killed.

I think that was the same impetus as the WashPo dataset. I do wish the United States made it mandatory for agencies to submit their data to the FBI. I also wish Eric Holder's Justice Dept didn't withhold offender race data from the NCVS. Interestingly, the FBI's Unified Crime Report seems to cover about 50% of police-involved shootings, making it a pretty good sample (and the UCR is generally consistent with the NCVS which is a survey of victims). Leaving data collection up to the media is sketchy, notably because the media are more likely to report officer race if the suspect is black and even more so if the officer is white.

That's not the right comparison, statistically. You'd want to compare the number of police encounters vs the number of police encounters that result in death, broken down by race. The statistic that you quoted would skew if one race has less encounters with police than the other group (which if I recall is the case here. I think NYT and WSJ both ran articles several months ago about this.)

Not that this changes the rough idea here which is that someone changing a license plate on a car shouldn't have a gun pointed at her and have to depend more on her de-escalation skills than the officer's.

> You didn't research very hard. From the Washington Post:

Actually, that very quote was what inspired my research. I wanted to know if the quote was misleading in favor of one narrative, so I filtered out cases where the suspect was armed and otherwise viably threatening.

Here's the link to the harvard study: http://www.nber.org/papers/w22399

> It is a fact: black people are shot and killed more often by police than white people.

I don't dispute that. I'm skeptical that black people are unjustifiably shot more frequently than white people, or at least with the frequency the media would have me believe.

Then in that case, I suppose it all depends on who is doing the justification. Nearly every police shooting in America is "justified" — be it by a police review board, a prosecutor, a grand jury, or a trial jury.

I agree, and unfortunately I don't have a good idea about how to audit the entire justice system. My gut says it's unlikely that an entire diverse industry is prejudiced against one particular racial group, and that thousands of laypeople are in on it. Hopefully we can get better data if it doesn't already exist.

You think it unlikely that there is a systemic bias? Bearing in mind that systemic bias isn't usually the result of an overt plan or even hidden conspiracy but instead the simple average of the bias of all the parts of the system.

Something like that. I guess I don't think bias is as responsible or pervasive as we're lead to believe. Frequently when we look at a disparity, I find myself thinking "Why didn't they check if X, Y, or Z were responsible instead of bias?". One example is the gender wage gap; we now know that women make less than men because they elect into different careers, because they have different priorities. Different groups have different distributions of any given behavior, and not all behaviors have the same outcome. This is all just a hunch, so I won't defend it too much; it just seems more plausible to me than the current "everyone is subconsciously biased against one racial group" theory.

I sort of see what you're getting at, but considering there's been a clear systemic bias against women and black people in our very recent past, and considering that it took a lot of effort (even a civil war, sort of, in the case of black people) to reduce these biases, I find it entirely reasonable to assume that they're still there, to a degree, and that we'll have to expend quite a bit of effort to reduce them further.

Those were conscious biases though. Most people in our society try very hard to avoid even unconscious bias, often by being consciously biased against minority groups (this might range from small acts of kindness toward a minority group to nonprofits that serve minority groups to corporate inclusivity policies to government programs and affirmative action policies). Given all that, I have a hard time believing bias plays a large role in explaining disparities.

>My gut says it's unlikely that an entire diverse industry is prejudiced against one particular racial group, and that thousands of laypeople are in on it.

Is that a general statement? Because the 20th century would like a word.

If we really managed to eliminate institutional racism in the justice system, we should probably try and work out the date that occurred, and make it a national holiday.

https://en.wikipedia.org/wiki/Institutional_racism

>White people make up roughly 62 percent of the U.S. population but only about 49 percent of those who are killed by police officers. African Americans, however, account for 24 percent of those fatally shot and killed by the police despite being just 13 percent of the U.S. population.

Which is meaningless, since black people commit more serious crimes than white people.

If you break it down by people suspected of having committed a felony, you're actually more likely to be shot as a white guy.

> black people commit more serious crimes than white people.

Convicted? Yes. Arrested? Sure. Commit? I don't know, maybe, but I bet if you factored in economic status there'd be a lot of evening out; i.e. poor people commit more serious crimes, black people are disproportionately poor, therefore more black people commit more serious crimes.

As I understand it, there's no real question about whether or not black Americans commit more crime than other races (the data are all consistent, including that collected from victims and witnesses), but SES alone doesn't explain the gap IIRC. I suspect it has to do with the relative concentration of poverty, but that's just a hunch and I doubt it's all down to one simple factor anyway.

Commit, yes. For serious crimes like murder and A & B there's a victim.

The causes are irrelevant from the standpoint of police officers. Cops have to deal with the situation in which they find themselves.

That's assuming there's an officer involved at all. I also didn't have a specific definition of "serious crimes" in mind, whether it would only include violent crimes or also non-violent ones like burglary or fraud (of significant value).

By "serious" I meant violent. The statistics line up for non-violent crimes, too, but cops don't get amped up when they arrest shoplifters, and I thought I would preempt the whole "but cops don't arrest white people for drugs" line of argument.

Okay. I return to my earlier point, "black people commit more serious crimes" suggests there's a genetic or cultural explanation while economic status is likely more relevant; i.e. poor people commit more serious crimes.

>I return to my earlier point, "black people commit more serious crimes" suggests there's a genetic or cultural explanation...

Does it? Not that I can see, particularly in response to the assertion black people are shot disproportionately by police.

I don't understand the relationships between SES, culture, and crime, but I'm baffled that we pretend that veneration of gang life in inner city communities can't possibly factor into the equation even in a small way. To head off the stupid "white people listen to rap too!" false equivalence, crime worship can't be reduced to music selection, it's (by definition of "worship") a way of life. I guess I don't think we should rule out a cultural cause; it's not like we need to say "every black person worships criminals" or "no white people worship criminals"--if culture plays a role, I hope we can be mature enough to recognize that it's not a binary proposition, and can engage it earnestly and compassionately.

Even if you believe that, the two core points of the experience stand:

> My behavior is how everyone should act in those situations: comply, survive and complain later.

> I don’t have de-escalation training. I’m the one being held at gunpoint. I’m the one thinking my life could end if he panics. Yet, I’m the one who must remain calm. The legal system is asking untrained civilians to de-escalate panicky cops.

He acknowledges that in his comment

> The story and its interest doesn't hinge on this (in)accuracy ...

I agree that the media coverage on this issue is pretty much garbage. The vast majority of cases, even those where the person killed was not committing the crime of which they were suspected, are much more complicated than "they were black." This belief puts me at odds with many (perhaps all) in the "racial justice" movement.

All of that said, there is a core of truth here that I believe should not be overlooked, and that's that dark-skinned people are treated differently than light-skinned people. There is a great deal of scientific literature that supports this belief - generally speaking, people of all races view others with darker skin to be stronger, more aggressive, less innocent, and often older than they really are.

Maybe, but I'm skeptical. The data often point in both directions, and the literature often fails to look at obvious alternative explanations (prejudice is the null hypothesis). Besides, the field of sociology is politically homogeneous; I don't know how much their political ideologues contribute to confirmation bias, but I suspect the literature would look different if this weren't the case.

Does it particularly matter? Even if the majority of the panic is due to the media fanning the flames of racism for profit (and I think it is), it still effects the victim's state of mind.

It doesn't matter if the victim's narrative was that she was likely to get shot because she is black, but yet the reality is that cops are equal opportunity murderers. Either way, the victim is ultimately panicked because they're being assaulted with a deadly weapon.

I was very clear that the story doesn't depend on this. Why are you confused?

I'm not confused and it seems we're close to being in violent agreement. But I'm pointing out that where one puts the emphasis is important.

If OP has fallen into the narrative that she is especially at risk being black [0], then there is a fine line between trying to extract her from that possibly nonconstructive narrative and shouting down her entire story. This is the general problem with the identity politics tarpit - if one objects too hard at a certain collectivist narrative (in this case, racism), then they are easily pigeonholed into being the "other" in that narrative, fueling the unproductive fire.

[0] Which given the location, is likely true regardless of the overriding distribution.

Good point, well said, and I apologize for my misunderstanding.

Last year I trolled the 2015 Washington Post police shooting dataset

In case this wasn't a simple typo, just FYI: It's "trawled".

They're largely interchangeable in this context:http://blog.oxforddictionaries.com/2015/05/trolling-trawling...

They're synonymous, but I edited my comment because I can't unsee the modern meaning of "troll". Thanks for that. ;)