The complications of assessing racism in police departments

June 26, 2020 • 9:45 am

In a few previous posts (e.g., here and here), prompted by claims of African-American linguist John McWhorter, I examined the various biases and difficulties that plague attempts to see if police kill black suspects at a higher rate than whites. This new article in FiveThirtyEight, though not providing an answer to the problem, shows further complications in the attempt to get answers, so that at present we have no idea if there are racial disparities in who becomes the object of police violence.

Click on the screenshot to read:

There are three ways to compare racial disparities:

1.) Proportion of racial populations who are victims of police violence.  As is well known, about 0.096% of black men and boys will be killed by the cops during their lifetime. That compares to 0.039%  among whites: a 2.5-fold difference. But this doesn’t mean that blacks are more likely to be killed in police encounters, because they may encounter police more often, even if the “kill rate” is the same among races. Which brings us to the second calculation.:

2.) Proportion of racial populations who are victims of police violence, normalized by the proportion of encounters with police. As McWhorter pointed out, blacks encounter police more than whites, so even if both races experience violence at the same rate per encounter, there could still be a differential mortality of blacks at the hands of the cops, but one that wouldn’t necessarily denote racism. There are some data on this, as the article notes:

One example of an encounter denominator approach is a 2019 study by Roland Fryer, an economist at Harvard. He found that police shoot white, Black and Hispanic Americans whom they’ve stopped at equal rates.3 At first blush, that would seem like evidence that the police are not racially biased — every demographic is being treated equally, after all.

As I pointed out, though, this statistic is not perfect because the types of encounters and their dynamics may differ among races, and you need to control for that, which hasn’t been done But the FiveThirtyEight article adds another complication:

3.) Members of racial groups might be stopped at different rates because of racial bias itself. This is called “collider bias.” There  are in fact data suggesting that although the frequency deaths per encounter may not differ among races, blacks and whites may be stopped at different rates. And that, in fact, seems to be the case, as we knew from traffic tickets, a disparity (blacks stopped more often) that disappears after dark when race becomes less evident. And, as the article says, it seems as if blacks are stopped when there is less evidence for stopping them than there is for whites, a difference that would reflect racism:

But we know that police officers are more likely to stop Black and Hispanic people than white ones — and that more of those stops are unfounded. Researchers measure this with something called the “hit rate,” or the rate at which contraband is actually found on the people who were stopped. A lower hit rate implies bias because it means that the decision to search someone was made with less evidence. White people stopped in New York City, for example, were more likely to be carrying a weapon than Black and Hispanic people who were stopped. White drivers stopped by the police were more likely to have contraband than Black and Hispanic drivers nationally.

As political scientists Knox, Will Lowe and Jonathan Mummolo, among others, have pointed out, that complicates Fryer’s findings. All of a sudden, what at first appeared to be equal treatment actually suggests unequal treatment. Because of the initial discrimination in who gets stopped, the sample of stopped people isn’t the same across races. The different hit rate indicates that stopped white people are actually more likely to have contraband, on average, than stopped Black people. In other words, in a world without discrimination in who was stopped — if the Black and white people who were stopped were equally likely to be engaged in criminal activity — you’d see an even bigger disparity in outcomes.

In other words, black people stopped by cops may be less likely to be engaged in criminal activity. To control for this, we need the data about the proportion of blacks versus whites who were observed and stopped or not stopped by police. To get this data seems impossible (you can be stopped for acting suspiciously, or just because you’re black), but the differential “hit rate” suggests that cops are targeting blacks at a higher rate than whites without good reason. And that suggests racial bias.  Curiously, I think this is pretty evident when explained in words, but FiveThirtyEight presents the same results graphically—to my mind, not clarifying matters much.

And even if bias doesn’t emerge in the rate of killings, it does come out in other aspects of the justice system, for an encounter with a cop is only the first step in a long chain of events that might culminate in jail. The article gives evidence of bias in the subsequent steps of the process:

Across the U.S., demonstrators have spent the past few weeks protesting against racial disparities in the country’s criminal justice system. There’s plenty of data to back them up: Black and Hispanic people are stopped more frequentlyincluding traffic stops, and are more likely to be arrested. Once stopped, police are more likely to use force against, shoot and kill Black citizens. And then once in jail, Black defendants are more likely to be denied bail, which in turn makes conviction more likely. And when convicted, sentencing is also biased against Black defendants, with Black defendants more likely to be incarcerated.

I haven’t read the links in the previous paragraph, which of course could be themselves biased (for example, is race the only reason why black defendants are more likely to be denied bail, or are there other factors like criminal records or the nature of the crime?). But I think there are enough data to conclude one thing: there is evidence for racism in police practices and in the judicial system, and this must be remedied.

On the main issue of whether black deaths at the hands of police reflect racism, at least in part, we don’t yet know the answer. To rephrase Hitchens’s Razor, what can be asserted without evidence must be buttressed with evidence before it can be accepted.

h/t: Ken

133 thoughts on “The complications of assessing racism in police departments

  1. Apparently, if you are a police officer you are 18.5 times more likely to be killed by a black perpetrator than the opposite.

      1. I knew this was coming. Heather Macdonald wrote an article that i posted on a related topic. No one commented. That is telling.

            1. That article is just as uncited as your post. Again, you need to proved a citation for the 18.5x claim. I am not saying it’s wrong – just that you have NOT defended it by linking to another comment where it was not cited either.

              1. I did not write the article so it is not my job to provide citations. It is author’s job. I have yet to see any rebuttal or push back against the author. Furthermore i have yet to see any stats that claim the opposite.

              2. Cool. So we can simply dismiss your claims as unevidenced nonsense. I’m ok with that.

              3. @dabertini

                You’re passing bullshit around and using other bullshit as support for your “argument”. It simply won’t do to deny responsibility with a glib “I did not write it.” That’s nothing but dishonest redirection.

                In your words again, “That is telling.”

              4. “I did not write the article so it is not my job to provide citations. It is author’s job.”

                Reminiscent of Donald Trump’s recurrent comeback when he’s called for citing some BS statistic (such as that 3 million illegals voted for Hillary in 2016): “Well, that’s just what I heard.”

                True, it’s the author’s job to provide reliable citations. It’s up to you to cite reliable articles when making contentious assertions.

              1. Thanks. That is the first rebuttal that i have read on the heather macdonald piece.

              2. I don’t find the rebuttal convincing. One of the main points made (to rebut the argument) is that WHITE people are much more likely to kill a cop than be killed by a cop. That’s true. The figures for cop killings (by both black and white perpetrators) should put the claim about cops being homicidal maniacs to rest. The cops come out as losers in this conflict.
                Perhaps their families should be marching on the streets.

        1. I don’t think you can find any evidence because they do not keep the stats on the shootings and killings by the police. This is why doing this is in the legislation now in the house of representatives. They simply do no keep the stats. The Washington post gives us the only accurate info on how many police killings per year that there is in this country.

        2. I googled “police officer 18.5 times more likely” and one place where the claim is published is in a piece on which referred to “Afro-Simians”!

      2. The question of how much more likely an officer is to be killed by a black than a black is to be killed by an officer depends a lot on how it’s defined. For example, are we talking about per capita, or per encounter, or what? Are we only talking about killings of innocent blacks by police, or are we including justified killings of black criminals?

        If we want to look at it per capita, then over the past 30 years or so, an average of 64 cops were feloniously killed per year. There are roughly 675,000 police in the US. So about 9.5 per 100,000 police are feloniously killed per year on average. Blacks kill police at the highest rate, about 3.8x the rate of whites. Of the 64 cops killed per year, about 23.7 are killed by blacks, or a rate of 3.5 per 100,000. At the same time, about 0.77 blacks per 100,000 were killed by police (assuming 250 per year and taking the average black populations over the same time period). That gives a ratio of about 4.5:1 cops to blacks, judged per capita. But we don’t want to compare unjustified killings of police with justified killings of criminals, so I’d say 0.03 blacks per 100,000 are unjustly killed (using the figure I’ve heard bandied about that 96% of them are justified – you can check the Washington Post and Guardian databases if you want to see for yourself). That gives a ratio of about 117:1 cops to innocent blacks, judged per capita.

        What about per encounter? Over the same period, there were about 3.7 million arrests of blacks per year, or 6.7 blacks killed per 100,000 arrests. So, I’d estimate 0.27 unjustified killings of blacks per 100,000 arrests. And again we have 23.7 cops killed by blacks per 3.7 million arrests, or 0.63 per arrest. That gives a ratio of 2.35 cops killed by blacks for each innocent black killed, judged per arrest attempt.

        It seems like any way you slice it, innocent cops are more likely to be killed by blacks than innocent blacks are to be killed by cops. But I don’t know where the 18.5x figure comes from. Maybe it’s valid. Maybe it’s not.

        1. Your analysis seems pretty fair. The risk may be small but American cops face a far bigger risk than cops in other developed countries (guns). The unarmed UK police suffer about 1 fatality per year due to criminal action (they do get assaulted a lot). That would be about 6 per year in terms of the US population. So the US rate for cops killed by criminals is about 10x the British rate.
          Its mistifying that this burden of danger faced by the US police is rarely considered.

            1. Which is a completely unrealistic conversation. And a waste of time when there are actual, realistic things that can be done.

              1. Slavery? You’re expecting a civil war to outlaw guns? Probably right, that’s what it would take.

    1. Once, out of curiosity, I did a survey at the police officer memorial site and found that killer ethnicity approximately followed proportions in the general population. The data I used was just one year, so the sample size is very, very small. Only about 40 police officers are murdered yearly. Being a police officer is actually a pretty safe profession, compared, for example, with sanitation or construction workers or a boat fishermen.

          1. Ah yes. Idiot drivers and trash compactors are indeed dangerous. But I agree with dabertini’s consternation about the list. Why supervisors of mechanics? I know a number of mechanics and they seem like decent, amiable folks (if a bit greasy), not in the least dangerous. Or at least no more so than any one else. So I’m mystified.

            1. “First-line supervisors of mechanics, installers and repairers” — could be they spend a lot of time moving in and around dangerous spaces, including ones involving ladders and scaffolding, sites the layout of which may be subject to rapid change.

              1. I was kidding about dangerous mechanics. I bet you’re right. Everyday -every single day- I am humbled about how very little I actually know. Alas, that’s never stopped me from pontificating, a lesson I keep having to learn. I read somewhere that even planaria – a single cell metazoan- can learn. You can teach them how to follow a maze and it never forgets. You can even cut it in two and the daughter critters will complete the maze. I wonder what my problem is.


              2. Hell’s bells, Edward, if pontification were a dangerous occupation, a lot of us around here would never get off the disabled list.

                I know I’d be suffering from repetitive stress injury. 🙂

      1. These FBI statistics on felonious killings of officers are also relevant. At least according to these, the killers’ ethnicities does not follow their proportion in the population. (For example, blacks kill police at 3.8x the rate of whites (including Hispanic whites) and 13x the rate of Asians/Pacific Islanders.)

        It is a pretty safe job, at least if you just consider the chance of being murdered.

      2. Well, this is certainly a statistical rabbit hole.
        Wading through the data, it appears to me that one reason that the rate of police fatalities might be lower than expected is that the job “police officer” seems to include a great many people who do not engage in risky police-type activities.
        Also, the type of fatal event is not always what one would imagine. Vehicular accident is a big cause of fatalities over a number of professions.
        Cashiers and sales persons who die on the job are likely to be victims of homicide.
        As far as actual deaths on the job, it seems like big rig truckers have the highest raw numbers by far.

        But none of the listed occupations have very high risk levels, compared to historical norms.
        My Dad, as an example, flew “wild weasels” over North Vietnam. When he was doing that, there was a 60% chance of being killed or shot down and captured before completing the required tour of 100 combat missions. He did two of those tours.

        Living in this country at the present time is fairly safe, no matter what you do, unless you choose a career as a fentanyl addict or the like.

  2. I think another important variable is the race of the police – or perhaps more precisely but more complicated- the genotype.

    That is, what differences can be observed comparing a table that might contain:

    black cop vs. white citizen
    white cop vs. black citizen

    1. Good point. Might help determine the extent to which police shootings are racially motivated. My uninformed guess is that police shootings are pretty much the same whether the cops are white or black, but I’d like to see the numbers.

  3. The rot in our culture goes far deeper than pulling over people because they are the wrong color in the wrong place (and then killing them, or abusing them as they go through a system designed to brutalize). It is also everyday, interactions with cops. I see it myself. Although I’ve never had a pleasant experience even talking to a cop (seriously; there is something about that profession which brings out the very worst in people – or perhaps it attracts the very worst people) my experience pales in comparison to the daily lives of many black people. How does one go through their daily lives knowing that they have a target on their backs all the time? The presumption that you are a criminal or doing something illegal just by being there. It isn’t just cops – society as a whole does this to them too.

    I agree with those saying we need to rebuild police culture and I also think that can’t be done unless it is dismantled first. There are many aspects of the wider culture where this needs to happen as well. I so not know how to do either of those things but it begins by waking up to it (sorry about the allusion to “woke” but the term is apt here). We all see it and we all need to acknowledge it is real and needs to change. That is a necessary first step, without it no progress is possible.

  4. It is a.long and arduous journey changing peoples peoples attitudes. This issue will not be resolved (changed) in our lifetimes.

  5. My tablet does me in at least once a week. I wonder if I can change that in my lifetime. Probably technology will get there first.

  6. Alternative plan:

    Pass two edicts down from the President (now or later if a Dem gets in) down through state governors and mayors regarding law enforcement:

    1) Do not hire any bigots. There are screening tests for this. Apply screening (or face the obvious) for bigots already on board, and take their guns away, give them desk jobs. Double the salary ($34,000 is a joke) in order to attract better character, which will also be buoyed by #2 below …

    2) Assert that civil law and order is to be firmly and universally asserted as Job1, with no tolerance for tolerance of thugs.

    Who cares about putative “racism” let alone “systemic” racism if we get all bigots out of enforcement.

    Other: live stream body cams; advanced training in better protocols; more powerful “stoppers” than tasers.

    1. “Double the salary ($34,000 is a joke) in order to attract better character . . . .”

      I agree, $34K is a joke for police officers.
      (Is it also a joke for, say. K-12 public school teachers? I’ve heard more than once of a U.S. representative or senator, making $175K/yr, upon leaving office saying to-the-effect, “I’m going to go out and make some REAL money.”)

      Just congenially curious, what is the basis for an alleged correlation (let alone causation) between increased money and increased character? A reasonable case can be made for a link between increased money and increased lack of character.

      1. The case can be made without trying to correlate money, intelligence, and character in individuals. If cops are paid more, it will attract more applicants. This results in the police being able to be more selective in who they hire. It probably also aids in retention since police officers will be less likely to look for other sources of income (eg, writing novels about life on the street) or switching to a more lucrative profession. They can treat it more as a serious career choice.

      1. A little googling gets starting salary somewhere around $46k per year but it often depends on the area of course. It’s still pretty low. It is also reasonable to consider that the working lifetime of an officer is not as long as most civilians. Some will become upper management, of course, but most will retire much before 65. They might become security guards after but I’m sure that pays even less. So, not $34k but not great either.

  7. A little surprised PCCE hasn’t posted/referred to Sam Harris’ recent long monologue about US police-violence & race.
    Unusually, Sam has made a transcript of that podcast available on his website too.
    In it, he says:
    “The police used more deadly force against white people — both in absolute numbers, and in terms of their contribution to crime and violence in our society. But the public perception is, of course, completely different.”
    “The best data we have suggest that for, whatever reason, whites are more likely to be killed by cops once an arrest is attempted.”
    Sam links to studies and sources for his claims,
    Chris G

    1. I’m sure it’s mentioned but that doesn’t rule out police brutality against “blacks” deriving from prejudice.

      1. You mean: despite police killing more whites in both absolute numbers and as % of police-encounters with whites, police may kill those whites for wholly different reasons to killing blacks?
        And the killing of blacks is primarily because of racism?
        A hypothesis based on what?

        1. A. I didn’t claim I am proposing a hypothetical explanation for anything.

          B. My claim of prejudice among police is weak. I thought that was obvious. Knowing something about what “prejudice” should be requisite for discussion here.

          C. A substantial discussion stemming from the Sam Harris discussion would, most effectively be done separately from this original post which was about McWhorter.

    2. Interestingly, a 22 June Spiked piece ‘Five myths peddled by Black Lives Matter’

      “Very few people of any race are shot or killed directly by police in Britain, and so the focus here is on the disproportionate number of deaths in police custody. Of the 163 people who were registered as having died in or following police custody in England and Wales in the decade up to 2018-19, 13 were black – just under eight per cent. But looking at the proportions of those arrested who die in custody, rather than of the wider population, arrested white people are actually more likely to die in such circumstances than arrested black people. Black people in the decade up to 2018-19 made up nine per cent of arrests, but eight per cent of deaths in custody, whereas white people were 79 per cent of arrests and 85 per cent of deaths in custody.”

    3. Hm. The data I’m able to find do not support Sam’s claim. From statista, in the US, whits are 74% of the population, blacks are 13%. In 2019, 370 whites were killed, 235 blacks. According to these numbers, the rate at which blacks are killed is higher than it is for whites. For the rates to be equal, per capita, 2019 killings should have been 370 whites to 63 blacks.

      1. You’ve not understood the correlation Sam summarises.
        Rather than look at the per-capita rates, Sam looks at the ‘per-arrest/police-encounter’ rates i.e. blacks encounter police-attention disproportionally more often, but whites suffer more police-killings per encounter.
        And: look at the rates in relation to rates of crime.

        1. Well then, I’d like to see Sam’s definition of “encounter”. If he was only counting encounters with people who are actually dangerous and actually engaged in a crime, then his point is irrelevant. That’s not what the protests are about. The issue is (perhaps not completely, but largely) about the killing of black people who were minding their own business until the police showed up.

          1. I think you’re clutching at straws here.
            George Floyd’s encounter with the police was due to an accusation of a crime.
            Other notable cases involved stopping a car because of a broken light; the selling of cigarettes outside a store; a suspect matching description of a shop-robber.
            Do you have data to support your claim that police are “killing black people who were minding their own business”?

            1. Unwittingly passing a counterfeit 20 does not rise to the level of “crime” imo. You’ve likely passed counterfeit bills yourself and not known it. Besides which, I don’t think it’s clear the 20 actually even was fake. So, the list of black people minding their own business would in fact start with Floyd. Then Breona Taylor. Then Tamir Rice. Eric Garner (selling cigarettes does not make one a dangerous criminal). Philando Castile. Stephon Clark. David McAtee. These are just off the top of my head. The reason they spark outrage is because of how undeserved and unnecessary their deaths were. The violence that caused their deaths was not part of a standoff with dangerous criminals.

              1. The 911 call was made after he allegedly passed the bill, but the caller expressed concern that he
                “he’s sitting on his car cause he is awfully drunk and he’s not in control of himself” and had “started to go, drive the car”

                It is reasonable to expect the police to respond to a call about someone attempting to drive while clearly intoxicated.

                Nobody is saying that the cops acted correctly, but the narrative of “executed for passing a fake bill” does not match the facts, either.

              2. The whole point of Jerry’s post (and of Sam Harris’ podcast) was to address the issue of police killings, and question the widely held belief that police are more likely to kill black people motivated by racism.
                Despite there being growing data/evidence that the claims are not true, many (most? most whites?) people continue to side with the claims: you appear to be one of them.

                Your claim that police are “killing black people who were minding their own business” is a sweeping statement. And implies that police do not kill white people who are also just ‘minding their own business’, yet more whites are killed when police encounter them.

                When challenged about the potential criminal behaviour that may have prompted police interest, you make further excuses e.g. “selling cigarettes does not make one a dangerous criminal” – so police should only take interest in dangerous criminals?

                For every horrific unnecessary unjustifiable police killing of a black person, there are equivalent killings of a white person.
                As Sam says in his podcast, he can’t name them (despite Sam having just read an article my McWhorter listing them) because only black killings have videos/articles that go viral.

                It seems essential, now more than ever, we don’t allow biased reporting, emotional reaction, sweeping generalisations, erroneous presumptions, and failure to look carefully at the data, get in the way of us getting to the truth of what’s going on.

              3. I am not making sweeping statements. I am saying that there are cases of police killing blacks who were basically minding their own business. That’s simply a matter of fact. I have searched for similar instances of police killing white people, ie, instances in which, had the person been left alone, nothing bad would have happened. I was able to find two: Tony Timpa and Daniel Shaver. Why are there dozens of these kinds of killings of black people but only two of whites? Yes, I know, my google searches aren’t conclusive; but it damn sure is suggestive.

                Also, black people “encountering” the police so much more frequently than white people doesn’t seem to me to be exculpatory. It’s yet another indication of racism. Or are black people simply more criminal than whites?

              4. “…the narrative of “executed for passing a fake bill” does not match the facts…”

                Indeed. Something like that would be awful. Execution for the crime of being intoxicated, however, is completely understandable.

          2. “The issue is …”

            My take would be :

            One large problem is police officers deciding for themselves on the spot on their own which innocent citizens to revoke the right to a fair trial from – proceeding directly to death sentence in the process. An associated problem is the immunity- when in doubt, feel free to put a knee on the citizen’s neck for 8 minutes and 46 seconds.

  8. While I think the hit rate disparity is very real, police aren’t the only cause, and possibly not even the main cause. Local governments set policies that determine who gets heavily policed and who doesn’t.

    The laws themselves can also be part of the problem. As Ken Kukec and I mentioned in another thread, the penalties for different forms of cocaine are suspiciously correlated to the skin colors of their users.

    1. No disrespect to you because you seem to be the type of person who wouldn’t know better, but this is one of the more naive arguments I regularly hear repeated.

      For better or for worse, I DO know better. Don’t ask how, but I’ll just say it was a long time ago. Comparing powder and crack cocaine is like comparing a joint to 2 hits of acid.

      This seems like the more obvious reason why the attitude towards these drugs would be different.

      1. Crack is a more concentrated form of the same controlled substance contained in powder cocaine. As such, it merits some difference in its treatment for sentencing purposes.

        The current ratio is 18:1; originally, the disparity was 100:1 — meaning that five grams of rock (an amount easily consistent with personal use) was treated for sentencing purposes the same as half a kilo of powder (a quantity consistent with at least mid-level distribution) — and, thus, subject to the same five-year minimum-mandatory (and potential 20-year maximum) sentences. (Fifty grams of crack, was subject to the same 10-year minimum-mandatory sentence at five kilos of powder. In cases where the offender had any type of prior felony drug conviction, both the minimum-mandatory, and the potential maximum, sentences were doubled — meaning that someone with a prior offense for felony marijuana possession then convicted of possessing 10 grams of crack would receive a mandatory term of 20 years imprisonment, and up to 40 years.)

        Are you saying — based upon whatever experience you claim (but decline to share) that makes you “know better” than others — that such sentencing disparities were just and appropriate? Are you aware of the grossly disparate impact these sentencing laws had on minority offenders?

        1. I’m just saying if part of the objective of sentencing is deterrence and to generally express societal disapproval, then it would make sense that crack is treated differently than cocaine.

          You can talk chemistry all day long but having smoked crack and snorted powder I can tell you that the phenomenological experience of these two drugs is drastically different, as is their potential for personal and collateral damage. Powder use can (and has) certainly wrecked people’s lives, but regular crack use can turn into an unimaginable train wreck. In other words, they have different potentials for self and other harm, so if the idea is to more strongly deter and disapprove of the more damaging substance, crack should be treated differently.

          Sometimes peoples lived experience trumps theory, particularly in the area of drug use. You can talk about LSD all day long but until you’ve taken it you really have no idea.

        2. Let me just add… those numbers you tossed off seem absolutely absurd to me. I have no idea what sort of time people typically do in actuality, but there is no way anyone should be doing anywhere close to that amount of time for using ANY sort of drug. I’m not convinced decriminalization of all drugs is the answer, but the idea of someone spending 20 years in jail for drug use is horrifying.

          That sort of brings up another point… absurd sentencing ranges. Presumably to raise the stakes so high that people won’t actually go to court and instead take a plea deal.

          Regardless… all I’m saying is it would make sense to treat crack more harshly than powder, and both more harshly than pot (which is rapidly being decriminalized all together, as it probably should be).

  9. Of course a strong argument can be made that the police need to be reformed without reference to racial bias. They are killing innocent citizens! (Well, if not completely innocent, they definitely didn’t deserve to die.)

    1. Yes. Whether or not racism is the cause, the fact that US police kill citizens at much higher rates than any other industrialized country is something well worth tackling.

      1. To both of you: please expand on your use of the word “kill.”

        Law enforcement commits homicide on around 1000 individuals every year in the course of duty. Only a few — sometimes one or two per year — are not justified.

        When you say “kill” do you mean unjustified homicide? Or?

          1. Exactly. Precisely. What in the US passes for justified is a much lower standard than many other nations.

            What is justified is the actual problem. The “what is justified” has gotten so out of whack that we need to make a serious adjustment. Citing the rule book, or training norms, or the ruling of an internal investigation, or the ruling of a police union, and claiming, see?, Justified, is ludicrous.

            As usual we have managed allow the entirely wrong incentives to evolve and grow. The system incentivizes, damn near requires, excessive use of violence among police.

            1. I’m no expert but my impression is that justification often amounts to just a measurement of fear on the part of the officer. There are problems with that:

              1. Anger and fear are often confused and present simultaneously.

              2. Some officers are obviously more fearful than others.

              3. It is too subjective and doesn’t give a rule governing behavior that is independent of an officer’s mental state.

              4. It naturally leads to conclusions that the officer is virtually always right and the victim wrong.

              5. The officer who gives in to anger always has an easy out when his/her behavior is called into question.

              1. My sense on the matter is that if a normal civilian couldn’t get away with it, then a cop shouldn’t be able to get away with it. (I.e. the self-defense standards should be the same.) I think it’s pretty clear that cops are treated much more leniently than us regular folks.

              2. I agree that the police should only break the law that non-police must abide by in certain situations. For example, they can’t speed unless they use their lights and sirens and they can’t use those except under circumstances that the community specifies. If police break such rules, they are punished in exactly the same way a civilian would. And the police are required to arrest each other on that basis.

                I’m sure our police would absolutely hate such a scheme but it would put into law the idea that their ability to go outside the law that everyone else must followed must be justified or it is illegal. Just as seeing Trump without a mask leads others to feel the right to do the same, police breaking the rules because they feel like it encourages bad behavior on the part of citizens and engenders disrespect.

      2. Remember, 1000 homicides per year …

        I challenge all of you who just posted opinions claiming vast injustice in the “justifications,” to cite ten homicides by law enforcement in the USA in the past year that were deemed “justified” by the law enforcement profession, legal authorities, and courts, yet should not have been – because in your opinion the parameters of justification was wrongful.

        That would be 1% bad shoots that should have been prosecuted as murder in your opinion, but were not.

        Stipulated: the Floyd case can be your #1. What about nine more?

        1. That sounds so easy why bother. Right now we’re really only hearing about the incidents that involve a death. I am sure there are many abuses every single day involving severe injury. Surely you don’t imagine finding 10 would be difficult. There are probably 10 a week at least.

          Rather than make everyone jump through hoops, distracting from the real issue, why not state clearly what you believe to be the case and with which you think most others here would not agree? Cut to the chase!

          1. Avoiding.

            It’s your claim. All authority says 1000 are justified. You claim vast injustice in the parameters which let law enforcement get away with murder. Please point to ten worse cases.

            Or…keep avoiding and say I’m distracting from the real issue, as if unjust homicide is not the real issue! If it isn’t, Mr. Topping, what is?

            1. What a shocker, the police departments say all police homicides are justified. I wonder why it is that the only ones that look unjustified are when a third party manages to get it on video. Now there’s a coincidence.

              1. Police departments, the judiciary, the Supreme Court, and the grateful citizenry of the United States say 1000 homicides are justified in the carrying out of law enforcement. Do you seriously claim criminals stealing and rampaging, armed and threatening, should not be stopped by officers?

                And guess what? So do Antifa and BLM. They only scream when there actually IS a bad shoot. They are all over the rest of the 1000 and they know they are not murder. Can you imagine the rioting if indeed unjustified murders were occurring all the time?

              2. Oh, please, you’re telling me the Supreme Court has looked into 1000 homicides a year? And did I say anything like that about criminals? You’re just ranting and rambling to try and support your agenda.

            2. As I’ve already said, it is the definition of “justified” that is at the heart of the discussion. What is an officer justified in doing (and required to do) based on the situation, including the behavior of the citizen in question. To make it all about murder deflects from this core issue. “Did the officer murder the citizen?” presumes a certain definition of murder that is usually applied to citizens, thereby confusing the issue.

            3. Nothing more to say to people blind to the necessary exigencies of law enforcement responsible for countering deadly threats to the lives and property of citizens by criminals.

              1. Were the Louisville police who used a battering ram to knock down Breonna Taylor’s door with a no-knock warrant at 1:00AM, spray 10 rounds, 8 into her, some into a neighboring apartment, was that responsible law enforcement? They suspected someone there had received drugs, though no drugs were found. On their incident report, they reported no injuries, though she had 8 bullets in her.

                None of the three officers have been charged, though one was just fired, three months later, so that’s something. I suppose that will go on your list as another “justified” death by police.

            4. Unfortunately, I don’t have numbers to support or refute your claim. I know what I have lived, read and seen from a variety of news agencies (not just one favored outlet.)

              I am fortunate in not having known a White person killed by police, but I know of several recent killings of Marshallese teenagers. Islanders, not Black, but minority. And I’ve known numerous non-White people who have been treated by police in ways that probably wouldn’t have happened to them if they were White.

              As I’ve mentioned before, as a White, I lived in Southern California with first hand knowledge of what was being done to Blacks and Hispanics by area police. And, discrimination against dark skinned people by White landlords and neighbors.

              There have been any number of news media reports over the years in the U.S. about police corruption and withholding of information by police departments around the country. New York. Los Angeles. Chicago. Portland. Seattle. A number of Southern Police Departments. Etc. Police Unions have too much power. Police can get away with behaviors that would probably not be forgiven in corporate America. Police Chiefs are replaced. Information is gathered. Citizen task forces make suggestions. Why do we still have these problems? How are we going to get change to take place? Why should citizens have to take video to prove that members of our police forces are unnecessarily brutal and intentionally or accidentally injuring or killing people?

              The only reason a police-person should kill is to save his/her or someone else’s life.
              There’s no valid excuse for killing people committing minor crimes or non-crimes, proven or not (it’s not up to the police-person to determine that.)

              1. Rowena, can you point to a specific incident, other than the Floyd case, where a law enforcement officer killed a person committing a minor crime?

              2. I’ll give you a list from my own town, John.

                Dontre Hamilton His crime was suffering from mental illness and sleeping on a bench.
                Larry Jenkins died while sitting in the passenger seat of a car, unarmed. Cop thought somehow he was carjacking. He wasn’t.
                Justin Fields was shot in the back.
                Sylville Smith shot running away following traffic stop
                Terry Williams shot for evading traffic stop.

                There are others. And there are white victims, too. But if you are thinking that Floyd’s death represents some kind of rare outlier then you can crank up that google machine and look around.

              3. I agree. The statistics do matter as to which problems are really the worst and which reforms will work best but, if the intent is to dismiss the charges, forget about it.

              4. Most of these are from Wiki or news sources on my iPhone. If you are interested, you can find more about these people and/or additional references:

                Lahquan McDonald, 10-20-14, Chicago, IL, Behaving erratically on street. Had a knife in pocket, Shot and killed by police.

                Eric Garner, 7-17-14, Staten Island, NY,, Accused of selling single cigarettes w/o tax stamp. Chokehold used by police, shot, killed.

                Michael Brown, 8-9-14, Ferguson, MO, Unknown precipitating facts. Brown is reported to have tried to gain control of policeman’s gun while in police car. Brown and friend fled. Officer shot and killed Brown. Discrepancy in accounts.

                Tamir Rice, 11-24-14, Cleveland, OH, 12 year old boy had toy gun that looked like real gun. Shot and killed by police.

                Walter Scott, 4-4-15, Charleston, NC, stopped for nonfunctional brake light. Scott had a knife and ran. Police shot him in the back and he died.

                Freddie Gray, 4-12-15, Baltimore, MD, Killed by police while in transit in police car. Spinal cord injury.

                Sandra Bland, 7-13-15, Waller Co, TX, Stopped for minor traffic violation (pretextual traffic stop). Hanged in jail cell. Death ruled suicide.

                Philando Castile, 7-6-16, St. Anthony, MN, Traffic stop. Castle told officer he had a gun. He was licensed to carry. Officers thought he was pulling out his gun and shot and killed him.

                Alton Sterling, 7-15-16, Baton Rouge, LA, Selling CDs. Had a gun. Shot and killed by police as they attempted to control his arms. Castile was not the person that had been called about.

                Korryn Gaines, 8-1-16, Randallstown, MD, Police serving warrant for traffic violation. She wouldn’t get out of car, went home, had shot gun, standoff, shot and killed by police.

                Botham Jean, 9-6-18, Dallas, TX, Policewoman enters what she thinks is her apartment but it is her neighbor’s, she shot and killed the man there.

                Vicente Vállela, 2- -19, Arrested on burglary charge, suffocated by police who were charged with homicide,

                Elijah McClain, 8-24-19, Aurora, CO, Man reported as wearing mask, acting suspiciously, but no threat. Police forced him to the ground, used a chokehold. He was administered Ketamine by paramedic, died of cardiac arrest.

                Breonna Taylor, 3-13-20, Louisville, KY, police entered wrong apartment with no-knock search warrant. Breonna’s boyfriend shot at police. Police shot and killed Breonna.

                Carlos Ingram-Lopez, 4-21-20, Tucson, AZ, reported by relatives as drunk and running around the house naked (drug use), pinned to ground by police for 12 minutes, couldn’t breathe, died.

                George Floyd, 5-25-20, Minneapolis, MN, counterfeit $20.00 bill, police killed with chokehold.

                Rayshard Brooks, 6-12-20, Atlanta, GA, Asleep in car at drive-in drive-through. Blood alcohol level too high to drive. Scuffle. Taser. Brooks punched by cop. Brooks ran. Shot and killed by police.

  10. I’m not clear on why one would qualify the numbers of blacks/whites killed by police with the numbers of blacks/whites stopped by police.

    It seems to me, an admitted statistical noob, that it would be like saying I eat more donuts than you, but that’s only because I stop at the donut store more often. If we controlled for stops at the donut store, then we’d eat the same number of donuts. Well, yeah, but that’s not the reality of it. The reality is that I’m getting fatter because I’m eating more donuts, counterfactual scenarios be damned. Where’s my mistake?

    1. Eating doughnuts is not a crime!
      Getting fat is not a crime!

      To be killed by a cop, you need to first have been stopped by a cop.
      So seems reasonable to compare % of police-encounters to how many of those turn lethal.
      As to why higher % of blacks encounter police attention, that may very well directly relate to % of crimes committed, but may also include police-prejudice.

      1. My point is that, like stopping at the donut shop more frequently is inextricably linked to my eating more donuts, police stopping black people at a higher rate than whites is inextricably linked to the rate at which black people are killed by police. It doesn’t makes sense to me to say “but IF blacks were stopped at a lower rate, more comparable to the rate at which whites are stopped, then the killings would also be more comparable.” Well, ok, but that’s not the way it is! The point is that stopping for donuts so frequently is something I need to fix, just as stopping black people at a higher rate than whites is something police need to fix.

        1. If you eat twice as many doughnuts as me, because you shop at the doughnut shop twice as often, then there’s no interesting insight there.
          If I eat as many doughnuts as you despite shopping half as frequently, that’s significant.
          If blacks are less likely to be killed per police-arrest/encounter, it doesn’t matter whether they’re stopped more of less often per-capita.
          You are more likely to be killed, per arrest, if you are white.

  11. Sam Harris has claimed : “The best data we have suggest that for, whatever reason, whites are more likely to be killed by cops once an arrest is attempted.”
    Others have used similar data. There are lots of problems with drawing a meaningful conclusion.
    For example, this could result because blacks are more often stopped and arrested unfairly for minor offenses where confrontations are less likely. Even attempting to find a definitive answer to the question of racial basis for killings is impossible without a lot of more detailed data on the reasons for stops and arrests.

    1. Not sure there would necessarily be a significant-correlation between ‘level of offence being investigated’ and ‘confrontation/lethality-of-force’ use by police.
      But I think it’s highly likely (as Sam discusses in recent podcast) there’s a correlation between ‘resisting arrest’/’not complying with police instructions’ and force used.
      Attitudes towards the police may highly influence behaviour during arrest, whether on suspicion of being in possession of minor-drugs, or suspicion of murder; and these attitudes appear to differ between black and white suspects.
      So yes, more detailed data would help on this.

  12. Like most things I think our current law enforcement crisis is from a combination of factors: racism, militarization, unchecked authority, and a culture of violence.

  13. Unsurprisingly, the article never discusses trade-offs. Would crime rates be lower if police officers ignored their work experience and did not take different base rates between white and black criminality into account? About everyone accepts the extreme sexism of police departments, because in practice there is no alternative to assuming that men are more criminally inclined.

    I’d also be interesting in a comparison that was less crude and included more demographic groups. Why are East Asians missing? And how do Blacks with an ADOS background compare to recent African immigrants?

    I understand only too well that such studies are unlikely to be done. No amount of evidence could possibly discard the assumption that police officers are racist, because researchers who doubt this will never be promoted and results showing no racial bias will be discarded. But maybe there will be a woke take on how discussing publication bias in academic studies is racist or ableist.

    1. No amount of evidence could possibly discard the assumption that police officers are racist, because researchers who doubt this will never be promoted and results showing no racial bias will be discarded.

      You know this how?

      1. That sentence was hyperbole, of course. But the idea of structural/systemic/whatever racism is not and was never intended to be falsifiable.

        1. Could you enlarge on what you suppose structural or systemic racism to be, and then give reasons why it ‘was never intended to be falsifiable’?

          1. Just to help you in your researches, institutional racism was defined by Sir William Macpherson in the UK’s Lawrence report (1999) as: “The collective failure of an organization to provide an appropriate and professional service to people because of their colour, culture, or ethnic origin. It can be seen or detected in processes, attitudes and behaviour that amount to discrimination through prejudice, ignorance, thoughtlessness and racist stereotyping which disadvantage minority ethnic people.” You may google ‘Stephen Lawrence’ and find out about his murder in 1993(not by the police) and its aftermath, which included the police’s attempt to smear the poor young man’s parents and the reluctance on part of both the police and the Crown Prosecution Service to make proper investigations. Two men, juveniles at the time of the crime, were eventually convicted of the murder in 2012.

            There has also been more recently the Windrush scandal, which involved deporting black British citizens who had lived in Britain for years or been born in Britain. If you want to learn what institutional racism is like, I suggest you read the harrowing ‘The Windrush Betrayal’ by Amelia Gentleman (who is, incidentally, the wife of Boris Johnson’s younger brother). Even now, the Tory government is dragging its feet over giving reparations to people whose lives were ruined by the affair. Comfortable white people I know in Britain don’t really want to hear about it, because they would prefer not to be challenged by injustice. As the great Judith Shklar wrote in ‘The Faces of Injustice’: ‘But then willed ignorance is at the heart of passive injustice.’

            It is curious to me how many white people become quite extraordinarily defensive when the issue of racism is aroused, and start banging on loudly on about ‘order’ (as opposed to justice), or try by means of interpreting statistics(or misinterpreting or falsifying them, as in the case of Helen MacDonald) to show that institutional racism does not really exist or that it is not so bad as it is claimed to be, particularly by its victims, or that racism can only be entertained by individuals and not by society (I suppose on the presumption that, in Mrs Thatcher’s immortal words, ‘society does not exist’ – I know very well that she went on to qualify this, but those are the words that have either been embraced or criticised).

            1. And the Flint, Michigan, water crisis, the Grenfell Tower fire (for which, again, victims – those that did not die – have not yet been properly compensated), and the death of Trayvon Martin and the subsequent acquittal of his murderer are all very good examples of institutional racism and, in the case of the Grenfell Tower, institutional neglect of the ‘unworthy’ poor. Regarding the Grenfell Tower fire, the Conservative Party’s Jacob Rees-Mogg said on television that nobody had really intended it, and therefore there were no evil intentions and no blame to be put upon the Kensington Council who had allowed inflammable cladding to be used illegally and not bothered to to respond to the number of concerns that been raised about the safety of the tower block, or upon anyone else. He went on to intimate that the people who lived there were too unintelligent to try to escape, as he would have done – the Fire Service had of course requested people to stay in their flats and not attempt to flee.

  14. I think that concentrating on the racial issues is a mistake even though they certainly exist.

    Police misconduct disproportionally affects blacks. Effective police reform will disproportionally help blacks even if they have no racial factors. You cannot force a police officer not be a racist but you can work to preventing their abuse of people.

  15. Racism can be difficult to pin down. Black people, Hispanics, Muslims, Jews, and probably others, are regularly lectured on what racism is and isn’t, when and where it is or isn’t taking place, all by those who have never experienced it. It can be frustrating to listen to.

    1. White people too. You forgot to include white people in your list of who gets lectured to about what racism is.

      1. And why not? Broaden your mind! The book has received good reviews in Britain, and I have ordered it and intend to read it. The author is, I suspect, no James Baldwin (someone who made elderly white men like J. Edgar Hoover sit up and take notice), but she appears to have something worth saying, so I intend to read it, which does not of course mean that I shall necessarily agree with everything she says.

  16. I think only you and Sam Harris (from the long dead sceptic/atheist community) seriously examine this question.
    To not accept the woke assertions on police violence is regarded as racist.

    1. I really think that you should stop worrying so much about the ‘woke’, stop assuming that it is only the ‘woke’ who grow exercised about police violence, and pay attention to the many serious people who have written on racism and the part it plays in the American polity.

      1. As for Sam Harris, I find that sort of approach in all honesty frivolous. A black man, and many more than one black man, as well as a black child of 12, have been killed by the police or vicious vigilantes, and in many cases have got away with it,and I think outrage is the proper response. I am afraid I neither understand nor sympathise with the attitude that we must put our outrage aside and apply our intellects to trying to show that things are not really as bad as all that. However bad they are, they are very bad, and surely we should be trying to create a fairer and less violent society. Why shouldn’t people be outraged? Why is it only Black Lives Matter that gets all this not very relevant attention, and not, say, the Dylan Roofs of this world and their links with, or membership of, the varieties of the white supremacist movement?

        1. “Why is it only Black Lives Matter that gets all this not very relevant attention . . . .?

          Just congenially curious, how much attention do BLM – and you – think Calvin Munerlyn should get? (

          (Or am I supposed to keep my mouth shut about this incident?)

          1. Congenially or congenitally curious? Black Lives Matter, as you must know, is not an organisation. As for myself, I find the shooting of the security guard appalling, and hope that the perpetrators are caught. Does that satisfy you?

            1. And why should you ask me if I suppose you should keep your mouth shut about it? I don’t suppose that you, or anyone, should keep their mouths shut about it.

            2. “Congenially or congenitally curious?”

              I acknowledge your view that curiosity is somehow congenital.

              “Black Lives Matter, as you must know, is not an organisation.”

              How do you define an “organization”? By what moniker would you have BLM identified – an “entity”? An “association” of like-minded people?

              As for myself, I find the shooting of the security guard appalling, and hope that the perpetrators are caught. Does that satisfy you?

              Yes, it does satisfy me that you – prompted – acknowledge, however grudgingly, that incident.

              1. Do you attribute that incident to BLM? Or is every criminal act a black person attributable to BLM.

              2. “Do you attribute that incident to BLM? Or is every criminal act a black person attributable to BLM.”

                Where have I said anything which can be reasonably construed as attributing this incident to BLM?

                I am interested in what any person – who claims an association or identification with BLM – has to say – especially unprompted – about the incident.

              3. And I wonder why this incident is so important to you? Is it because it was a criminal act perpetrated by a black person? Because that is what it seems like.

              4. “And I wonder why this incident is so important to you? Is it because it was a criminal act perpetrated by a black person? Because that is what it seems like.”

                For whom, if anyone, should it not be (“so”) important?

  17. As for the study involving before dark and after dark police stops…

    1) Could age have something to do with it? I know I got pulled over a lot more when I was younger, even though I don’t think I drive much better. I think white demographics are shifted to older people.

    2) For that matter, how about economic bracket? Perhaps the police have figured out that tickets “stick” more for those from a lower economic bracket?

    3) Is it possible people are getting pulled over because they fit suspect descriptions?

    4) Here’s a somewhat wacky idea, but it shouldn’t be ruled out… what if it’s actually black officers that are more likely to pull over black drivers?

    Anyway, I’m just throwing stuff out there, mainly because it seems like “racism” is the go to explanation for any discrepancy these days, and it doesn’t seem like alternative explanations get much attention (or if they do, they are actively shouted down).

    1. I have been stopped quite a lot by the police at one part of my life because I matched an offender profile. While this never resulted in an arrest, I could understand the situation and would have acted in the same way if I was a policeman (details will not be provided, sorry).

      The dilemma that a police officer faces is that he can only deal with a small amount of crime directly while having little certain knowledge about who is an offender and who is not.

      He needs an algorithm to maximize his effectiveness. It will in no small part be influenced by his experience.

      Suppose he no sexist and that 90% of the worst offenders are men, who make up 5% of the male population. He applies equal standards to everyone, controlling 50 men and 50 women. The result: 2.75 offenders caught.

      Another officer does not think that women can commit crimes and does not bother to stop them. He manages to catch 5 criminals after stopping 100 men.

      Yet another officer takes multiple factors into account and achieves a better success rate than his blatantly sexist colleague. But how could he ever prove that he is not sexist, especially if he uses criteria that are highly correlated with sex?

  18. Another example of the fine police attitude that people keep disparaging. “NYPD union calls on officers to ‘defy’ elected city government to do ‘what is right’.”

    “DEFY DeBlasio lead the NYPD don’t be afraid to be fired! Doing what is right and not popular is never easy, DO what is RIGHT & the men and women of the NYPD will walk through the FIRES of hell with you. Show NO FEAR, DeBlasio is weak. Defend the city honor your OATH.”

    The oath NYC police take when they join begins “I do solemnly swear that I will support the constitution of the US, and the constitution of the State of NY…”

    Funny, it doesn’t say anything about “Defend the city honor.”

  19. One issue that does not get brought up enough in these discussions is that in a significant sense, a large function of the police is to protect accused criminals from the citizenry.

    Regular people leave enforcement to the police and punishment to the courts only as long as the police and courts are perceived to do their jobs effectively.

    The alternative is horrible to contemplate.

    1. When people are apologetic about police brutality, it’s often because they do not think that the court system is reliable.

  20. I feel that there is some missing of the point in many of this discusions.
    Some of the questions I would like answered are.

    How is socio-economic status related rates of violent interaction with the police?
    It seems reasonable to expect that they are.

    Are African Americans more likley to be of lower socio-economic status?
    My understanding of the US is that this is so.

    If so are African Americans more likely to be of lower social-economic status due to past and present injustice, racism etc?
    And would find it hard to argue that they are not.

    A black person might be murdered by a police officer not simply because that police officer is racist but because a history of systematic racism has put them both into that situation.

Leave a Reply