Making lemonade

White-on-black killing reminds liberals that harsh punishment is sometimes justified.

White-on-black homicides make liberals aware of the retributive functions of punishment.
I’d rather not have the homicides, but when life hands you a lemon …

Much of the internet chatter about the Michael Dunn case has focused on outrage that the jury hung on the murder charge, convicting the killer instead on three counts of attempted murder and one count of firing a gun. Despite that chatter, no one has found that Dunn was justified in killing Jordan Davis. A hung jury means only that at least one juror didn’t believe that the state had refuted Dunn’s claim of self-defense beyond reasonable doubt, or alternatively that the jurors couldn’t agree as among the alternative charges of first-degree murder, second-degree murder, and voluntary manslaughter.

But now the discussion turns to sentencing (assuming that Dunn is not retried on the murder charge, or is acquitted of it, or that juries keep hanging).

What are the arguments for and against sending the killer away forever?

One of the benefits of a harsh punishment is that it announces and enacts social disapproval of the underlying act. In this case, where some white people seem to think they have a hunting license on young black men, a harsh sentence will help change that opinion, and help shift perceptions of the wrongfulness of the act. (I think we’ve seen this happen with both drunken driving and domestic violence; it’s part of the logic behind “hate crime” enhancements.)

Another benefit of a harsh punishment is that it acts to reaffirm the social value of the victim; again, that seems relevant to this particular crime.

What counts as a harsh punishment depends on the background flow of punishments. For those of us who think that the current sentencing regime is obscenely excessive, that sets up a tension in cases such as this one; a short sentence would send the wrong message about how wrong it is to go hunting for black scalps, while a long one would reinforce the pattern of excessive incarceration. In most European countries, a 15-year sentence would count as ferocious; here, you can get that much for drug-dealing.

In this particular case, where the murder required no physical strength and where it seems to have proceeded from longstanding animus rather than merely momentary passion, there’s also a case to be made for incapacitation; this is someone who might well kill again. Note that he’s already said that his night in jail strengthened his underlying racism, which didn’t really need it.

Say he gets out of prison after 15 years. During that time he will have joined the Aryan Brotherhood, as most white prisoners in bad state prisons find they need to do in self-defense. It’s not easy to see why he would pose less of a risk then than he does today.

So I hope the judge stacks the sentences. Three counts of attempted murder at 20 years each, plus one count of gunfire at 15 years, comes to 75. A 15% discount for good behavior would get the 47-year-old Dunn back on the street sometime in late 2077, at the age of 111. That seems about right to me. It’s the punchline of the old joke:

– “But Your Honor, I can’t do 75 years in prison.”
– “Son, you just do the best you can. That’s all we ask.”

Footnote By the same token, white-collar crime and public corruption are terrible things, but – when they’re prosecuted and lead to incarceration – they help create conservative prison reformers. Hey, it’s not much, but if you do crime-contol policy for a living in this country and don’t develop a sense of humor and the habit of looking on the bright side you’re going to wind up offing yourself.

Second footnote Yes, Dunn was coming from his son’s wedding and had a few drinks inside him. If he’d killed someone under the influence of cocaine, or even cannabis, we’d be hearing about the risks of drug abuse. But when someone kills under the influence of the drug most commonly involved in violence, it barely rates a mention.

Author: Mark Kleiman

Professor of Public Policy at the NYU Marron Institute for Urban Management and editor of the Journal of Drug Policy Analysis. Teaches about the methods of policy analysis about drug abuse control and crime control policy, working out the implications of two principles: that swift and certain sanctions don't have to be severe to be effective, and that well-designed threats usually don't have to be carried out. Books: Drugs and Drug Policy: What Everyone Needs to Know (with Jonathan Caulkins and Angela Hawken) When Brute Force Fails: How to Have Less Crime and Less Punishment (Princeton, 2009; named one of the "books of the year" by The Economist Against Excess: Drug Policy for Results (Basic, 1993) Marijuana: Costs of Abuse, Costs of Control (Greenwood, 1989) UCLA Homepage Curriculum Vitae Contact:

35 thoughts on “Making lemonade”

  1. "White-on-black homicides make liberals aware of the retributive functions of punishment." Yes, I suppose they do.

    Radicals such as myself, on the other hand, find themselves asking whether we're bad people for not wanting to see this man, or any person, put in prison forever, despite the horribleness of his crime and the creepiness of his attitude towards it. As you point out, many state prisons are de facto run by gangs and not administrators. Why do I want to feed someone into such a terrible system if there is any possible alternative to doing so? Mister America is downright punch drunk on retribution. I say it's time to cut him off.

  2. Punishment is not about the offender and it is not about the victim. It is about the rest of us.

    Its purpose is to create an environment that no longer contains the offender.

    1. Change punishment to long-term incarceration and I'm on board.

      Incarceration can serve a punishment purpose, as in Hawai'i's HOPE program . But long-term incarceration isn't useful as punishment. It is supposed to be useful in making our environment safer by keeping the convicted felon away from the rest of us.

      The problem with this theory is that prisons are semi-permeable membranes. Most (not all) convicts will eventually be released on parole or probation or for completing their sentence. The problem for the rest of us is that "penitentiaries" are misnamed: they are about many things, but penitence and atonement aren't very high on the priority list. Instead, they tend to act as finishing schools for criminals. Or, if you prefer Criminal Colleges. They come out more sophisticated about managing their criminal careers. This isn't good for the rest of us.

  3. The rate of conviction in a second trial following a hung jury mistrial is generally high, meaning the chances of a conviction on the murder charge go up if the state reprosecutes. My guess is that the state will not reprosecute if his sentence comes close to life, but not if it is lighter.

    1. You've got selection bias and confusion here about cause-and-effect. If prosecutors think their case is flawed, they don't retry.

  4. Let him serve the entirety of his prison sentence in solitary confinement to ensure that he isn't harmed by potentially irate African-American prisoners.

    Also, let him get the Bradley/Chelsea Manning treatment of being kept naked for the entirety of his sentence to ensure he doesn't use his clothing in order to fashion a means to suicide.

    He should be let out of his cell for no more than one hour per day. Sunlight is for individuals who don't murder over differences in musical taste.

    Once he is secured in such a manner, I hope he lives a long life and spends every waking hour thinking about Jordan Davis what he would be doing if he wasn't cut down before he could plans and projects and opportunities and pitfalls that we call life.

    1. Ugh, this is what happens when you fail to proofread, kids. In the last paragraph above, please read the words "pursue the" after the word "could."

      Also, this is the best argument against capital punishment. For someone like Dunn or Breivik, death is too light a sentence.

  5. "Can't do that much/Do the best you can" is a hoary snide joke and out of place in a serious discussion. My take is that it taxes the wisdom of any sentencing authority — judge, juror, legislator or whomever — to determine who belongs in prison for the near and medium future. There is no such authority who can make a wise and sound decision for a time several decades in the future. The most severe sentence should be something like 20 years, to be followed my a new proceeding to be devised, less than a trial but more than a parole hearing, to determine what if any continuing sentence in justified. And by the way, impossible sentences — such as consecutive life, life-plus, or extending beyond a plausible life expectancy are . . . unintelligent, and make the legal system appear likewise. We should not be eager to make our criminal justice system look unintelligent.

  6. I think harsh sentences can be appropriate for keeping the public safe from a violent person. I agree this seems like such a case. This is also the reason why I actually don't have a moral objection to the very idea of capital punishment. I object on practical grounds: in reality, we get it wrong often enough that the DP is unacceptable, as it is irreversible. The backup option is life in prison. It's perfectly justifiable as societal self-defense.

    As for the booze, the stories about the crime itself mentioned it and I thought the connection was clear.

  7. You sound like a conservative Republican, except they think murder isn't very nice even when it's not white on black.

  8. One thing I think a lot of Americans don't realize is just how bad, say, five years in prison is. We end up spending a ton of money locking everyone up for life, but basically any serious time in prison is severe punishment, very unpleasant.

  9. This is a very sad case. And there are also many unindicted co-conspirators, consisting of every state legislator that voted for that *idiotic* law.

    They should have to go too. Criminal stupidity.

    1. "that *idiotic* law" in no way exonerates the murderous Dunn. He did not "meet force with force." He _initiated_ deadly force and then fled the scene of the crime and ordered a pizza while failing to call the police. There is no punishment too severe for this homicidal coward who aborted the life of an innocent young man less than half his age.

      You could sentence him to be stripped naked and set on fire and I would happily light the match.

    2. How can you say that lawmakers who voted for a *self-defense* law are the equivalent of someone who, unprovoked, pulled a trigger multiple times against an unarmed man? My mind is literally boggled. Literally.

      1. Well, I was a bit wrong b/c the defendant isn't invoking the idiotic law.

        I think though that just having a SYG law on the books makes people less likely to behave sensibly and try to defuse bad situations, even though this jerk isn't using it.

        Now, one *can* make an argument for SYG *in one's actual home.* (Though, I still think if you can safely avoid killing another human being, or heck even a dog or a bear… that is worth doing.) This fool wasn't even at home.

        So, yes, I think the legislators bear some of the blame every time someone in that state gets shot unnecessarily.

  10. I think that Mark is being too much an advocate here, and not enough of an analyst.

    His argument from general deterrence is quite strong: this is a highly publicized case that can show that white men don't have a license to shoot black men, even in Florida. This country doesn't take five year sentences that seriously, although it should. So far, so good.

    His argument from incapacitation is, I think, very weak. Dunn had no criminal record, and only did what he did because he felt a sense of impunity. Ten years or so should remove all sense of impunity. Murder is a low-recidivism crime, anyway.

    I'm never sure about social disapproval as an element of punishment. That would seem to trench on the autonomy of the law from transient public opinion. There is a reason that juries don't sentence.

    His argument for affirming the worth of the victims has some value, especially given the history of black men in America. But then again, I'm always reminded of the sentencing pattern in vehicular homicide: a crime in which the characteristics of the victim are completely unrelated to the nature of the crime. It is thus a favorite of criminologists. For those readers who didn't follow the research, let me put it this way: if you're ever driving drunk and happen to kill somebody in a car, when you leave the car to see what you've done, just pray that the victim you see isn't an attractive young wealthy white woman.

  11. Weird how the media latches on to these man-bites-dog news stories — black-on-white violent crime is 39 times more prevalent than white-on-black violent crime.

    I guess the media finds it boring when whites are murdered, raped, assaulted, etc. by blacks.

    1. From

      In 2011, 193 blacks were murdered by whites and 448 whites were murdered by blacks. 2630 whites were murdered by whites and
      2447 blacks were murdered by blacks.

      Homicide rates are generally thought to be good proxies for violent crime rates. I'm currently thinking the "39 times more prevalent than white-on-black" was produced by a racist who knew he could rely upon his gullible readers to spread his BS around.

      1. "Racism," zzzzzzzzzz.

        Regardless, murder is a *terrible* proxy for violent crime in general — the overwhelming number of murders take place among people who are previously known to each other, which is not necessarily true in other violent crimes. (Despite this, given that Whites outnumber Blacks by a factor of over 6x, black-on-white murder is nearly 15x more prevalent than white-on-black murder, something you are oblivious to, even as you cite the data.)

        Overall, there are 770,000 interracial violent crimes annually in the US between Whites and Blacks. You'd think that since Whites greatly outnumber Blacks, that Whites would be committing the vast majority of these crimes but the *opposite* is true: Blacks commit 85% of them, Whites 15%, making black-on-white crime in fact just under 40 times more prevalent than is white-on-black crime.

        1. Regarding your "murder is a *terrible* proxy" statement: I just ran a regression using crime rate numbers(starting at year 1990) found at R squared was .966679. That's a high correlation.

          Regarding your whites outnumber black statement: A black criminal walking down a city street will see a lot of white people so if they pick their victims randomly, they will end up victimizing a lot of white people. White criminals walking down city streets see mostly white people. If they also pick their victims randomly, they will victimize relatively few blacks.

          Regarding your "15x more prevalent": The numbers were 193 blacks murdered by whites and 448 whites murdered by black.
          So in your world, 193/448 is apparently equal to 15. Most interesting.

          In any case, the propensity for violent crime is highly correlated with economic circumstances. (A middle class black man is not statistically more likely to commit a crime than is a middle class white man.)

          Regarding your 770,000 interracial violent crimes annually statement, according to the FBI, blacks committed 157384 violent crimes in 2011.

          The color of crime website does seem to be extraordinarily unreliable. Perhaps it has something to do with the site owners hating blacks.

          1. I see that the 157384 number is arrests for violent crime so let's disregard that.
   table 42:
            Out of 2,788,600 white victims of a single offender violent crime, the perceived race of the offender was black 15.4% of the time.
            Out of 570,550 black victims of a single offender violent crime, the perceived race of the offender was white 15.9% of the time.

          2. A given population of red and blue people is 100, 86 red people and 14 blue people.

            In a given year, there are 2 murders of blue people by red people and 4 of red people by blue people.

            Question: by what magnitude is the murder of reds by blues more prevalent than that of blues by reds?

          3. Given 98 red people and 2 blue people, one blue murderer who kills 10 people and one red murder who kills 10 people. Also suppose that the targets are determined randomly.

            What is the expected number of red people murdered by blues? What is the expected number of blue people murdered by reds?

          4. Selecting one murder at random, very near the beginning of the two sprees, the odds of it being a blue-killing-a-red would be
            around 10 times higher than they are for a red-killing-a-blue. The relative proportions of the populations in my thought experiment determined that ratio. It had nothing to do with differences in the propensity to kill others of a different color.

            I perhaps should note that the 86% white demographic percentage includes all Hispanics. Hispanics who happen to murdered by blacks are usually lower-class people who happen to live in the same neighborhood as their lower-class murderer.

            As has been pointed frequently in the recent year, the amount of lead consumed by the average child has risen and fallen in concert with crime rates and the correlation is quite good. We know that the consumption of lead is known to cause an increase in aggression and a decrease in impulse control(and a decrease in IQ also), so this apparent connection is very very plausible. We also know that the the amount of lead consumed by the average child is very well correlated with the income of their family. (Poor children tending to have much more exposure to lead.)

      2. Very interesting and useful addition to the data. Are there perpetrator data figures, divided by class/wealth, and by education (say, high school graduation,) as well as by race?

        I hypothesize that majority of homicides are perpetrated by lowerclass, less educated, (yet still young), poor men. Is that true?
        What is the racial distribution in that group, and do the homicide rates conform or contradict the racial distribution?
        I'd be interested in data on the neighborhoods, too, specifically the economic inequality range–are areas with high economic inequality more violent?
        Is cultural environment more determinative of violence than race?
        How could we distinguish between the two, especially if they're economically linked?

        1. The number one variable in murder (which may not be the case with other violent crime) is whether the perpetrator and victim know each other.

          Consequently, the poor end up being murdered by the poor, the rich by the rich, Black by Black, Hispanic by Hispanic, etc. You are much more likely to be murdered by the person sitting next to you right now than by a stranger, or someone outside your family, friends, colleagues, peers, and acquaintances.

          But there is another variable to this, that of instances of black-on-white crime, including murder. Proportionally, far more Blacks commit murder (and other violent crimes) against Whites than do Whites against Blacks.

          It's probably impossible to determine why — that the out-of-wedlock birth rate is 73% among Blacks might be a cause, but who knows.

          But to ignore this altogether, deny it, rationalize it, or "blame the victim" doesn't do justice to the victims at all.

  12. One interesting side note is that in Florida there's a "85% law." That apparently means, much like in the federal system, a convicted criminal will actually serve about 85% of his or her sentence. Sentencing in other states varies considerably.

  13. I wouldn't describe the two goals tht Mark describes as especially important in this case as falling under retributivism. As Ebenezer says, they're really deterrent. Retributivism argues for punishment even if it provides no benefit to society. That's not what Mark is appealing to here.

    1. There was no "stand your ground" defense in this case, if the news stories I've seen can be believed.

      1. Taking into consideration that Dunn did not have a duty to retreat under stand your ground was part of the jury instructions, so yes, it was a factor in the defense.

  14. I'm curious how often somebody waits this late in life before reaching peak violence. Maybe there's a category of privileged, arrogant, somewhat violent underachievers who don't follow the normal pattern of becoming less violent after their 20s but rather become more extreme as they fail at life. If so, that would support a long incarceration period.

    (And I suppose it's possible this isn't his first time, but he seems like a bigmouth so it really might be the first time he pulled the trigger.)

Comments are closed.