So Amber Guyger was found guilty of murder, and I’m still trying to make sense of it, but think I’m probably okay with it.

A Dallas County jury on Tuesday convicted fired police officer Amber Guyger of murder for fatally shooting Botham Jean in his apartment last year.

…

Guyger, 31, fatally shot 26-year-old Jean in his apartment last year. She had said she mistook his apartment for her own and thought Jean was a burglar. She is the first Dallas officer convicted of murder since the 1970s.

I say “probably” because I haven’t seen a good explanation of how this is the crime of murder, so I can only guess.

Guyger admits the shooting, but says she believed she was in her own apartment, which caused her to interpret Jean’s presence as an intrusion, and his failure to submit to her authority as aggression, to which she responded with lethal force in self defense. If that’s actually how it happened, then it’s hard to see this as a murder, because the crime of murder usually requires something like an intent to unlawfully kill someone, and this sounds more like a mistake.

That’s not to say this wasn’t some other kind of serious crime. I don’t know how Texas law works, but in most places it’s a pretty serious crime–often called manslaughter–to kill someone by mistake if your mistake was due to something like negligence or recklessness, which sounds a lot like what Guyger confessed to by admitting she went to the wrong apartment.

(I’m being vague because I don’t know enough about how the laws works in Texas.)

But these laws can be tricky. As the Dallas Morning News explains,

Guyger was initially arrested on a manslaughter charge. But in Texas, manslaughter is a reckless act — like firing a gun into the air and killing someone. A grand jury indicted Guyger on a murder charge because investigators believe she intended to pull the trigger.

To which the defense response is basically that she intended to pull the trigger, but she thought it was necessary to do so because she believed (however mistakenly) that her life was in danger.

I have seen bits of the prosecution argument in the news, and as near as I can tell (for whatever that’s worth) the prosecutor argued that Guyger was angry at the apparent intruder — and she either (1) charged into the apartment intending to kill him, (2) charged into the department when she could have remained safe outside, and/or (3) didn’t do enough to establish that he was a threat to her life before shooting him. This kind of argument nicely takes the question of her mistaken belief out of the picture, because even if it was her apartment, what she did would still have been a crime.

Except…Texas has a castle doctrine law that basically says people are allowed to defend themselves in their homes. As near as I can tell from news reports, it raises the prosecutor’s burden of proof.

If that’s what happened, it answers one of the questions people have been asking, which is how could Guyger’s lawyer raise a castle doctrine argument — an argument about self-defense in one’s own home — when it wasn’t even her home? The answer is that part of the prosecutor’s argument for murder is that what Guyger did would be murder even if she was in her own home, where the castle doctrine applies.

That’s where my guesswork has to end, because deciding whether or not Guyger would have been justified in shooting Jean if he had been in her home depends on the very specific facts of what happened that night. The jury has heard all the testimony about those facts, and I have not, so I’m not going to argue with their decision.

This post by Mark Draughn at Windypundit was originally published at Thinking About Amber Guyger and the Castle Doctrine

My goal with this Python rewrite is not to build a single monolithic criminal justice simulation program. As I’ve emphasized from the start, this project is about learning and exploration, and I haven’t yet learned enough to build an all-in-one solution like that. Rather, I’m writing a collection of Python components — classes in software terminology — that I can assemble and configure in various ways to build a variety of models.

For my purposes, the basic unit of simulation is the Case class, which represents a single criminal case, consisting of a single criminal charge against a single defendant, making its way through the criminal justice system. Each Case entering the system has two basic parameters which describe it:

• prob_convict: A number between 0 and 1 inclusive, representing the probability of a conviction if the case goes to trial.
• sentence_if_convicted: A number indicating the length of the sentence the judge will impose if the defendant is convicted by the trial. This is a unitless value, but I tend to think of it as months.

Cases are grouped together by the Docket class. Each Docket represents a group of cases that are being considered at the same time. The simulation creates one Docket for each iteration of the simulation by calling a DocketFactory class, which itself uses a CaseFactory class to create each of the Cases it puts in the Docket. The CaseFactory class, in turn, uses a random number generation class to generate the prob_convict value for each case. (The sentence_if_convicted value is currently hardcoded at 60 until I implement a better way to generate it.) The DocketFactory class itself uses a random number generator to decide how many Cases to put in each Docket it creates.

(I’ve included three random number generator classes, each providing a different distribution, any of which can be used by a factory class.)

Currently, the simulation is so simple that this grouping of Cases into Dockets has no effect on the outcome, but in the future I hope to use the Docket class to implement the court system’s resource limits and decision making horizon. For example, if there are only enough courtrooms to hold 20 trials per simulation cycle, but 30 cases are on the docket, then the simulated decision makers will have to resolve the remaining 10 cases by some other means, such as plea bargaining or dismissal.

It would have been simpler to have the main simulation code create a Case or a Docket whenever it needed one, but then I’d have to change the main code each time I wanted to run a different simulation with different parameters. Since I’d rather keep the core simulation code stable, I put the code for Case and Docket creation in corresponding CaseFactory and DocketFactory classes that the simulation can invoke when it needs a Case or a Factory. This allows me to change the parameters of the simulation by changing the factory classes without having to change any of the other code in the simulation. This is called a factory pattern, and it’s a good way to write flexible software.

At the top level of the simulation is the Experiment class, which actually runs the simulation engine. Constructing a new Experiment takes three parameters:

• A DocketFactory object which it uses to create dockets and cases,
• a PleaBargainingStrategy object which specifies how plea bargaining decisions are made, and
• a Trial object which simulates the trial outcome.

I’ve already discussed the DocketFactory, but the other two classes are new.

The PleaBargainingStrategy object processes each Docket of cases and figures out which cases will settle with plea bargains and which ones will go on to trial. In the future, I hope to have subclasses that implement a reasonably plausible model of the plea bargain process, but for now all I have is PleaBargainingAtRandom, which implements a dumb algorithm that chooses what to do with each case entirely at random. E.g. if you set its prob_plea parameter to 0.9, the strategy will randomly generate a plea deal 90% of the time. The deal will always be for a sentence that is the mathematically expected length: If prob_convict on a case specifies a 30% chance of conviction, and the sentence if convicted is 60 months (which it always is right now), then the plea deal will be for a sentence of 0.3 * 60 = 18 months.

The Trial class is even simpler: It generates a uniform random number between 0 and 1, and if it’s less than prob_convict the defendant is found guilty and serves the full sentence. Otherwise the defendant is acquitted and does no time.

Those are the pieces that make up a simulation of how cases work through the criminal court system. To show how the pieces come together, let’s walk through a script that runs an experiment.

(If you’re a Python user, you can follow along by installing the crimjustsimpy package in your Python environment. If you are reading this some time after I published it, I may have released newer versions which work differently, so to get the package this post is talking about, make sure to specify version 1.0.5. If you’re interested, the source code for this project is also available on GitHub, and you can specifically view release 1.0.5.)

The sample script starts by defining some parameters at the top:

AVG_CASES_PER_INTERVAL = 100
MAX_CASES_PER_INTERVAL = 300

The number of cases to be generated for each docket is specified by a Poisson distribution, which is a common distribution used to estimate the arrival rate for random independent events, where each one has the same probability of occurring over a given time period. It’s commonly used to simulate things like orders arriving at a warehouse or cars arriving at a gas station. In this case, we’re assuming an average arrival rate of 100 cases per simulation interval.

Although large numbers (relative to the average) would be very rare, the Poisson distribution places no upper limit on the number of cases that can arrive during the interval. To avoid insane outliers, we’re limiting arrivals to a maximum of 300. That’s probably not necessary, because you can see in this histogram of a sample of 100,000 dockets shows, it never gets close to 300:

The next three lines describe the probability distribution from which probabilities of conviction will be drawn.

MIN_PROB_CONVICT = 0.05
MEAN_PROB_CONVICT = 0.4
MAX_PROB_CONVICT = 0.95

In my first attempt, I used a normal distribution with upper and lower limits. This time around, I’m using a beta distribution. I chose it because it has a nice shape, similar to the normal curve, but it also generates values in a strict range of 0 to 1 without the need to impose external cutoffs. As you can see, the configured range only runs from 0.05 to 0.95, so neither a conviction nor an acquittal is ever a certainty. The beta variate is internally scaled to fit withing these limits. Moreover, the beta distribution appears to be a good choice for representing a probability distribution from which to draw probabilities, which is exactly what I’m using it for. 

The resulting histogram of conviction probabilities looks like this:

The next parameter just tells the simulation what probability to use when deciding whether a case will end in a plea deal. This example uses 80%:

PROB_PLEA=0.8

Finally, we specify the number of iterations over which the model will cycle.

ITERATIONS = 240

After that, the script begins constructing and assembling the parts of the experiment. First, it creates a random number generator object that uses the Beta distribution to generate the conviction probabilities for the cases:

convict_gen = cj.RandomScaledBetaProb(
    shape=10.0,
    lower=MIN_PROB_CONVICT,
    middle=MEAN_PROB_CONVICT,
    upper=MAX_PROB_CONVICT)

RandomScaledBetaProb is a wrapper around the Python library function that generates the Beta distribution, specifying the bounds and the mean of the generated values. The shape parameter controls how tightly the numbers cluster around the mean or stretch out to the edges. I just picked 10 because it gives a nice looking curve. It should probably be a parameter in the future.

The next line of the script creates a CaseFactory from the conviction probability generator:

case_factory = cj.CaseFactory(convict_gen=convict_gen)

(If you’re unfamiliar with Python named parameters, that probably looks a little confusing. In “convict_gen=convict_gen“, the first “convict_gen” is the name of the CaseFactory parameter that receives the conviction probability generator, and the second “convict_gen” is a reference to the variable that received the conviction probability generator created in the previous code snippet. This is a common coding convention.)

The next two lines do the same thing for the DocketFactory class. First create the object that generates the arrival rates for the dockets over the Poisson distribution, then create the DocketFactory itself, passing it the CaseFactory we created above and the arrival rate generator created here.

arrival_gen = cj.RandomPoissonBounded(
    mean=AVG_CASES_PER_INTERVAL,
    upper=MAX_CASES_PER_INTERVAL)
docket_factory = cj.DocketFactory(
    case_factory=case_factory,
    arrival_gen=arrival_gen)

The plea bargaining strategy is pretty simple, because it only takes the single value representing the probability of handling the case with a plea deal.

plea_bargaining = cj.PleaBargainingAtRandom(prob_plea=PROB_PLEA)

The trial is even simpler:

trial = cj.Trial()

Now we can assemble all these pieces to build an Experiment object:

experiment = cj.Experiment(
    docket_factory=docket_factory,
    trial=trial,
    plea_bargaining=plea_bargaining)

That brings us to the meat of the simulation: Running the experiment.

data = experiment.run(ITERATIONS)

It’s just one line, but it does all the work of creating the dockets filled with cases, resolving some of them with plea deals, resolving the rest with trial, and gathering up the results. With only 240 dockets and and 100 cases per docket, this runs in less than a second on my computer. In the future, with more cases and more complicated resolution logic, it will probably run a lot longer.

The results of the simulation run are returned in an ExperimentData object that contains all the Docket and Case objects with the results of all the cases. There’s nothing to stop us from traversing the Case objects directly, but there’s a better way:

df = data.casesToDataFrame()

This converts the Case data to a DataFrame, which is a highly-optimized standard Python data structure for handling tabular data. This will make it easier to process the data in the future, because lots of Python code packages understand how to work with DataFrames. For example, the PyCharm development environment I use can render DataFrames as tables:

This shows the output from a simulation run for each case. The fields are defined as follows:

• id: A unique integer ID assigned to each case.
• docketId: A unique integer ID indicating the docket the case belonged to.
• pConvict: The probability of conviction of the case when created.
• maxSentence: The maximum possible sentence if convicted.
• plead: True if the case was handled with a plea deal.
• tried: True if the case was decided with a trial.
• acquitted: True if the defendant was acquitted at trial. Always False if the case didn’t go to trial.
• convicted: True if the defendant was convicted at trial. Always False if the case didn’t go to trial.
• guilty: True if the defendant was convicted at trial or if the defendant took a plea deal.
• sentence: Actual resulting sentence. Always 0 if the defendant was acquitted.

We could plot this data using standard Python graphics packages such as matplotlib and seaborn, but the crimjustsimpy package also includes a Visualization class that collects a handful of convenience methods for creating plots from the DataFrame. For example, this pie chart showing the ratio of cases that plead out vs. going to trial:

This is how the trials turned out:

And this is a breakdown showing all three possible case results: Acquitted at trial, convicted at trial, and plead out:

None of that is particularly surprising. We set up the simulation to take a plea deal 80% of the time, and it did. We set it up for 40% of trials to result in convictions, and they did. We’re just confirming that the simulation ran the way we wanted it to.

It’s slightly more interesting to plot a histogram of the lengths of the resulting sentences for all the cases:

That looks a little weird, but it makes sense. The hump in the middle is showing the sentences from the plea deals, all of which are for the maximum sentence weighted by the probability of conviction, so they follow the same beta distribution hump as the probability of conviction. The bars at the ends are all the cases that went to trial, because they ended with either no sentence or the maximum sentence. The difference between the plea hump and the conviction bar on the right is what defense lawyers call the “trial tax.” It’s the difference between what happens when a defendant admits guilt, and what happens when he makes the government prove it.

That’s about all I’ve got done so far, other than some unit tests and a few sample scripts, which leaves me a lot more to do.

The previous version of this model used an agent-based systems model that tracked the changing state of the cases over time as they worked their way through the system. That was interesting to see, and it gave me the simulated steady-state size of the prison population, which seems like an important metric for studying criminal justice.

I’d like to put that output capability back into the model, but I’m not sure if I should keep a running track of the state of each case like last time, or if I should just run a post-pass to compute time series data for the prison population from the case results.

I’d also like to be able to see intermediate results from the simulation, either by having it generate usable output as it runs, or by being able to run it for a few iterations and then stop to examine what it’s doing. This suggests that keeping a running state will work better than a post-pass analysis.

In addition, if I can stop and start the simulation, I’d also like to be able to change the parameters while it’s stopped, so I can model changes to the system. This suggests I need to break down the Experiment class into at least two parts: The simulation engine, which can be modified along the way, and the simulation data, which persists throughout the simulation run.

Of course, none of that helps me with what I really want to look at, which is the plea bargaining decision. I have some ideas about that, but this post is long enough.

As always, any suggestions would be appreciated.

This post by Mark Draughn at Windypundit was originally published at Model Justice — Starting Over

I’ve identified several possible ways the model could be improved, and based on some feedback from lawyers Mark Bennett and Matt Haiduk, I think I want the new version of the model to eventually implement the following features:

• Allow defendants to have some degree of risk love instead of risk aversion when making plea bargaining decisions.
• Allow risk preferences to vary depending on case quality.
• Allow limits to the extremes of case quality. E.g. Always some chance of acquittal/conviction no matter how strong/weak the case.
• Allow prosecution and defense to have different views of the quality of the case.

In addition, I think I’d like to include a few other ideas:

• Allow mixes of different types of cases. E.g. Different severity of crime.
• Incorporate the way absolute sentencing severity — including collateral consequences — affects risk tolerance.
• Incorporate the effects of defendants being stuck in jail pending trial.

I also want to switch to a different model of plea bargaining — one informed by economic thinking — that I’ll describe in a future post.

Before I can do any of that, I need to make a big change to my approach. The system modeling tool I’ve been using, Insight Maker, is designed to model entities that transition through a series of states. That seemed appropriate tool for modeling a defendant going through various stages of the criminal justice system: Arrest, pre-trial, trial, jail, release, and so on.

However, nearly all of the things I want to change will have an effect at only one point in the process: Plea bargaining. Even things that do not happen until after the trial — prison, collateral consequences — are only part of the model because of how they affect plea bargaining. So the cumulative effect of all these changes will be to greatly complicate the model’s plea bargaining logic.

Unfortunately, while Insight Maker is a fantastic (and free) tool, it’s not really engineered for that kind of complex programming. For what I’m planning, Insight Maker falls short in several ways:

• You can add code to an Insight Maker model to control state transitions, or to execute independently as actions, but there’s no real place to add libraries of code that can be called from other code, which makes it awkward to do anything non-trivial.
• Insight Maker code is using its own private language, invented by the author of Insight Maker. It seems quite competent for a small language, but it doesn’t feel like a full-strength programming language.
• The editing environment for Insight Maker is primitive. It has an editor window implemented in JavaScript, but you can only open one piece of code at a time, and it doesn’t come close to the convenience features available in a modern language-aware integrated development environment.
• When developing software, it’s nice to be able manage multiple versions of the code base so you can try different approaches and easily back out of bad ideas, but there’s no easy way to manage multiple versions of Insight Maker code.
• You can download a model into a file, which you can edit in other tools, but you can’t easily modify or use the model in that format.
• Insight Maker doesn’t have libraries of open source code that you can leverage in your program to add features or speed development.
• Insight Maker graphic output is limited to only certain types of data and certain types of charts.
• Insight Maker doesn’t appear to be thriving. The official Insight Maker forum has only had 15 topics posted in the last three months, and the Insight Maker news stream hasn’t been updated in almost a year and a half.

I considered a number of possible alternatives to Insight Maker. I would love to play around with a professional-strength modeling tool like AnyLogic, Vensim, Powersim, or isee systems’ Stella/iThink packages. AnyLogic in particular can be extended with Java, which would allow me to use a whole world of open-source Java code. On the other hand, these tools are heavily focused on modeling complex business processes, and they are priced as if a corporation was buying them. (Some of them have free or low-cost versions, but those often have serious limitations, not all of which are documented.)

Something like MATLAB/Simulink, in addition to possibly being pricey, is also really intended for modeling physical systems, not system dynamics or the sort of agent-based modeling I was doing. Wolfram SystemModeler is similar, which is a shame, because I already have a home user license.

Speaking of Wolfram, I could probably use Mathematica, because you can do anything in Mathematica, at least according to the folks who make Mathematica. I have a home user license for that too, and I’ve already used it on this project to generate some of the graphics for these posts. The problem is, I get frustrated with Mathematica when I try to do anything complicated, because it’s not like a normal programming language.

But that got me thinking…maybe what I really need is not a simulation tool, but an ordinary programming environment in which I can build a simulation. It wouldn’t have the fancy drag-and-drop block diagrams that all those simulation products provide for building models, but I’m not sure I need that for what I’m doing. I think the connections and flows are simple enough that I can model them with ordinary programming structures.

In that case, the obvious choice for me would be C# and the .NET Framework, since that’s what I work with on my day job. I have access to a ton of development tools, and I’m very familiar with useful code patterns, the development toolchain, and the general approach to developing a C# project. On the other hand, C# is geared towards building production-quality systems, but since I’m still designing the model, I would probably benefit more from a development environment geared toward ad hoc queries and prototyping.

I think I need something that works like R Studio, where I can type in simple commands to change things, run code, and see the results, so I can easily try lots variations and ideas. Or maybe I’d be better with a notebook-style interface, like Mathematica. Both of those also provide data visualization, so you can change something, run it, and see a graphical representation of the result. I don’t actually want to use either of those, however, because I find both R and Mathematica’s “Wolfram Language” to be very annoying for writing anything substantial.

And that brings me to Python. Even though Python has been around for over 25 years, I have never really needed it for anything I was doing, so I’ve pretty much ignored it until now. I tend to prefer heavily-typed languages like C++, Java, and C# that compile down into machine code to run efficiently. Python is an interpreted scripting language that is usually used for writing short programs that stitch together powerful components. (Also, it uses indentation to delimit blocks, which is different from practically everything else in the world these days.)

However, unlike scripting languages such as Perl (or special-purpose tools such as R), Python was actually created from the beginning with full-strength programming features, and it has grown to become one of the most popular computer languages in the world, especially in the data science community. (So if nothing else, learning Python could turn out useful later.)

For my purposes, Python ticks several important boxes:

• Python is a full-strength software system with support for multiple-file software projects and organizational abstractions such as classes, modules, and packages.
• Python is a mature, well-known, and widely used language.
• Python is supported by numerous integrated development environments, including Eclipse (with the PyDev extension), Sublime, Visual Studio, Spyder, and PyCharm (for which I already have a license).
• Python projects are just files in directories, so you can easily edit them in a variety of ways and version control them with git.
• There are tons of usable third-party open source Python libraries at PyPi.
• Python is thriving and growing as a programming language due to its popularity among the data science community, so there are plenty of resources.
• There are development environments for Python that support both an R Studio-style environment (Spyder) and a notebook-style environment (Jupyter).

Of course, to use Python, I’m first going to have to learn Python. Writing tiny programs is easy and quick, but I want to build up a library of modeling components, so I need to learn about things like Python classes, Python modules, and Python packages, as well as figure out a development environment, all before I can start building the model.

I think it’s going to be a while.

This post by Mark Draughn at Windypundit was originally published at Model Justice — Stalled On the Learning Curve

My goal is to build a model of how our justice system handles plea bargains, with an eye toward understanding (1) why a system built around the highly formalized decision-making mechanism of the trial doesn’t actually have many trials, and (2) what we can do about it. As I am neither a professional systems modeler nor a lawyer, this is mostly intended as a learning experience.

I plan to eventually give the model a bunch of parameters such as time spent waiting for trial, case quality, sentencing regime, defendant risk tolerance. If I do a good job building the model, then I should be able to find a set of parameters that make the model produce results that look like our criminal justice system: Lots of plea bargains, very few trials, etc. Then I can experiment with ideas for reform by tweaking the model parameters to see how the model responds.

I’m taking an iterative approach to building the model, starting with a skeleton that has some of the major moving parts of the criminal legal process, and then elaborating on parts of the model to flesh it out into something more realistic.

The most recent model is very simple: There’s only one crime, people are arrested at a fixed rate, they wait in jail for a fixed time, and then they have a trial, after which they may go to jail for a fixed period. I suspect my lawyer readers will be annoyed to learn that the entire trial — the core of their profession — is modeled as a single draw from a random distribution to get the verdict. Plea bargaining, my whole reason for doing this, isn’t even in the model yet.

It’s that last problem that I tried to fix. I’m trying to add a plea bargaining phase somewhere between arrest and trial, and I’d like to tell you about it. This story will not end well.

I started by thinking about the basic decision the defendant has to make during plea bargaining: Take the prosecution’s offer or reject it and go to trial. I think there are two main question to address: What does the defendant know? And what does the defendant prefer?

I’m assuming the defendant has knowledge of a few basic things:

• The amount of time in the plea offer from the prosecution.
• The possible outcomes of the trial, consisting of:
  • The probability of being found guilty.
  • The sentence they will serve if found guilty.

Here you see more of my assumptions: Everybody has the same information, and it’s correct information as far as it goes. Note that I’m not saying anyone knows how the trial will turn out ahead of time — I’m modeling it as a draw from a random variable — but I am assuming that both the prosecutor and defense know what that probability is, and that both can use it to make decisions.

When thinking about a random event, it’s helpful to consider the expected outcome, which is calculated as the sum of the value of each outcome times the probability of that outcome. If you buy a raffle ticket which has a 1-in-10 chance of winning $50, we can calculate the expected value of your ticket as follows: Your 1-in-10 chance is a way of saying p = 0.1 for winning, so we multiply that by the payoff, $50, and get $5. Your 9-in-10 chance of losing has p = 0.9, but you get nothing for that, so the value of losing goes to $0. Add up the expected value for all possible outcomes (5 + 0), and you get the expected value of your ticket: $5.

Similarly, if the defendant knows he will be found guilty half the time (p = 0.5), and he knows the sentence is 10 years, he can calculate the expected sentence as (0.5 x 10 =) 5 years. Once he receives the prosecutor’s plea offer, he knows everything he needs to know (under this model) to make a decision.

The key issue controlling the defendant’s decision is then his attitude toward the risk of going to trial. If the prosecutor’s plea offer is 5 years in jail — exactly equal to the expected sentence calculated above, then you could argue that the defendant should be indifferent to the choice between taking the offer or going to trial, since the expected sentence is 5 years either way.

The problem with that approach is that it ignores risk, and people don’t usually ignore risk. All other things being equal, people usually prefer to reduce their risk. There is a massive amount of evidence for this, including the existence of the entire insurance industry, where customers pay fixed premiums to avoid the risk of having to suffer losses. If this risk aversion extends to plea bargaining decisions, a defendant will choose a certain sentence of 5 years in jail over a 50% chance of spending 10 years in jail, even thought the latter also includes a 50% chance of going free.

But if the defendant prefers a 5-year plea over a trial with a 50% risk of a 10-year sentence, what happens if the prosecutor makes an offer of 5 years and 1 month instead? Well, the insurance industry is pretty profitable, which means that people are paying more in premiums than their insurance companies are paying out to cover losses, which means that (on average) customers lose money when they buy insurance. So I think it’s very likely that a defendant will agree to lose an extra month of life in prison to protect against the risk of losing an extra five years of life in prison. If criminal defendants are risk averse, they would be willing to accept a plea bargain for more time than the mathematically expected sentence.

I am modeling this risk aversion by calculating an adjustment to the probability of guilt that represented how that probability would feel to a risk-averse defendant. E.g. a 50% (p = 0.5) probability of guilt with a 10-year sentence has a mathematically expected sentence of 5 years, but to a risk averse defendant, even a 6-year sentence might be preferable to a 50% risk of a 10 year sentence.

I decided to calculated the adjusted risk-averse factor by calculating an expected probability of an acquittal as 1 – p, then raising it to a power specified as a model parameter, and then converting back to a probability of guilt by subtracting it from 1 again. E.g.

1 – (1 – p)^a

where

p is probability of guilt

a is the risk aversion model parameter

I picked this method of modeling risk aversion because (1) I vaguely remember reading an article about risk aversion that modeled it like this, (2) it has the nice property that p = 0 and p = 1 both map to themselves, so it stays in range, and (3) I like the look of the curve. Here, for example, are the curves with a set to 1, 1.5, and 2 for comparison. Remember a = 1 means no risk aversion.

Now that I had a simple model for defendant choice, it was time to model the prosecution.

One of the fundamental differences between defendants and prosecutors, at least for purposes of my model, is that defendants are individuals who seek their own advantage in a single criminal case, whereas prosecutors work together across hundreds or thousands of trials to achieve a common organizational goal. So while I feel reasonably comfortable in assuming for purposes of the model that most criminal defendants will try to minimize their risk-adjusted prison sentence, I’m not nearly as confident in figuring out what exactly prosecutors are trying to achieve.

To keep the model simple, I’m going to assume that prosecutors want the opposite of what defendants want. That is, if defendants want to minimize the risk-adjusted sentence they expect, prosecutors want to maximize the aggregate expected prison sentence across all defendants.

You’ll notice I didn’t say the prosecution goal was risk-adjusted. That’s because if defendants are like insurance buyers, prosecutors are like insurance companies. For an individual home owner, the loss of their house to fire would be financially devastating without an insurance policy. But to the company that issued the policy, it’s just business. They issued policies for tens or hundreds of thousands of homes, and they are well aware that a bunch of them are going to burn down every year. Meanwhile, they will be collecting premiums on all of them, and very nearly all of them will survive through the year with no loss whatsoever. In investment terms, insurance companies are “diversifying away” the risk.

Prosecutors can do something similar. Using the example from above, a defendant going to trial against a 50% chance of a 10-year sentence has a mathematically expected sentence of 5 years, but he’s not going to serve 5 years. He’s going to serve either zero or 10 years. On the other hand, a prosecution team that deals with 100 of these trials will have a mathematically expected sentence of 5 years per case, which adds up to a aggregate expected sentence across all cases of 500 prisoner-years. And they’re actually going to get about 500 prisoner-years, which is 50% of the theoretical maximum of 1000 prisoner-years. In other words, in the long run, the prosecutor can earn the expected sentence on average.

Furthermore, defendants can only accept or reject the deal that the prosecutor offers them, but the prosecutor’s office can offer whatever deal they want, including dismissal or no deal, to all of the defendants. This means it’s ultimately the prosecutor who gets to decide which defendants get plea bargains.

However, they have to do that within a resource constraint: There’s a limit on the number of trials the justice system can support in any given time period (at least for the short term). So if 100 cases enter the system, there might only be enough capacity in the system for 30 full trials. For the system to keep functioning, the remaining 70 cases will have to be disposed of with a plea bargain.

But how do they choose which defendants get plea deals and which ones go to trial? Well, consider this batch of defendants:

The first three colums show the probability of being found guilty, the fixed 10-year sentence, and the resulting mathematically expected sentence. The fourth column shows the probability of guilt adjusted for risk aversion (using an exponent of 3/4), and then the plea that the defendant would be willing to accept based on that risk adjustment. Finally, the last column shows the difference between the plea and the mathematically expected sentence. I.e. the premium in prison time that the defendant is willing to pay to reduce the risk of a 10-year sentence.

At this point, the prosecution plea bargaining strategy seemed obvious to me: If the goal is to maximize aggregate prison time, then the prosecution can maximize total prison time by offering plea deals to those defendants willing to accept the most additional prison time beyond beyond the mathematically expected sentence. Thus, if they only have resources for 3 trials for the defendants above, they should dismiss the unwinnable case A, plea out B through H, and go to trial on defendants I, J, and K.

That’s when I realized I had screwed up: If the prosecution is trying to maximize aggregate prison time, and if defendants will agree to more than the mathematically expected sentence because they are risk averse, then why would the prosecution ever go to trial on any case? By the assumptions I had made, they could always get more prison time with a plea offer. And yet…trials happen.

There’s a mistake in here somewhere. Possibly more than one. Either I’ve oversimplified my model, or else it’s flat-out wrong. Or both. A few possibilities spring immediately to mind:

• Maybe I was wrong to assume both sides see the same probability of guilt. The sides in each case could have randomly different views of the quality of any given case. For all I know, all trials occur because the prosecution thinks it will be easier to prove guilt than the defense does.
• Defendants may be risk-preferring instead of risk averse. Conventional wisdom is that criminals are more willing to tolerate risk than law abiding people. Or risk-preferring people, criminal or not, may be more likely to attract police attention.
• Prosecutors may be optimizing for something other than aggregate prison sentence, such as conviction rate. Or prison times may combine in a non-linear manner: E.g. five 20-year sentences may be better for a prosecutor’s career than twenty 5-year sentences.
• Maybe collateral (non-prison) effects of a conviction affect defendants’ decisions in ways that matter. E.g. If any conviction, including by plea bargain, could result in deportation, defendants might grasp at even small chances at trial.
• I may need a more complex model of the trial outcome than just a probability of conviction. Maybe there are important phases or types of cases. Or maybe one side has more information and can consistently make better choices.
• Maybe I need to account for interactions between cases in the prosecutors’ decision making. E.g. taking 5 aggravated battery cases to trial may use up resources that could be used on a murder case, which could affect how they make plea bargains.

Time to go back to the drawing board. At the very least, I’ve got some thinking to do. And maybe it’s time for me to actually read something written by people who study this stuff.

This post by Mark Draughn at Windypundit was originally published at Model Justice – Plea Bargaining Blues

What I want to do in the simulation is to assign a probability p to every defendant who enters the model. This will represent the probability of conviction if the case goes to trial. I.e. if p = 0.25 then there is a 25% chance the defendant will be found guilty by the Verdict action when it runs. If p = 0, there’s no chance of a conviction, and if p = 1, there’s no chance of an acquittal. But before I can do that, I need to answer an important question: How should those probabilities be distributed? Or to put it in the language of the problem domain: How does the quality of the criminal cases vary?

The easiest approach is to assume a uniform distribution. That is, to assume that all possible values of p from 0 to 1 are equally likely, so that a 100-bin histogram of the probability of guilt for 100,000 cases would look something like this, with the values spread out more-or-less equally over all the quality levels.

That doesn’t seem realistic to me. I mean, does it seem right that they would bring just as many cases that will succeed 5% of the time as will succeed 50% of the time? I just don’t believe that the prosecution would bring exactly as many cases at each possible probability of success.

This is where it would be great to use research by professional criminologists, or get advice from people with a lot of experience in the criminal justice system, who might be able to advise me on the quality mix of criminal cases. Alternatively, it would be nice to pull these numbers from a sophisticated model of how the police and prosecutors’ offices build cases. Since I don’t have access to either of those right now, I’m just going to wing it.

I’m going to assume that building a case that will persuade a jury involves trying to achieve a whole bunch of smaller subgoals, and that each of these goals would improve the case in a way that pushes the jury toward a guilty vote — the more goals accomplished, the likely the jury is to convict. If we assume that the prosecutor and police officers have a roughly equal probability of accomplishing each of these goals, then the number of goals accomplished will conform to what is known as a binomial distribution.

A good example of a binomial distribution is the probabilities of various outcomes when we flip ten coins and count how many turn up heads. Since each coin has two possible values, and there are 10 coins, there are 2¹⁰ = 1024 possible outcomes. Of those outcomes, the extreme cases of all heads or all tails can each only occur one way:

HHHHHHHHHH
TTTTTTTTTT

So the probability of either of those cases is 1/1024 ≈ 0.000977.

There are 10 possible ways of having one coin come up heads, one for each coin:

HTTTTTTTTT  THTTTTTTTT  TTHTTTTTTT  TTTHTTTTTT  TTTTHTTTTT
TTTTTHTTTT  TTTTTTHTTT  TTTTTTTHTT  TTTTTTTTHT  TTTTTTTTTH

So the probability of exactly one head in a set of flips is 10/1024 ≈ 0.00977. That’s also the probability of exactly one tail in the set of flips.

Now it gets more complicated. For two coins to come up heads, there are a bunch of ways it can happen. We already know there are 10 ways to get 1 head. So assuming we’ve got one head, there are 9 remaining coins, and any one of those can also come up heads. However, that leads to each combination appearing twice. E.g. we can start with TTTHTTTTTT and then pick anther coin to get TTTHTTTHTT. However, that gives the same result as if we had started with TTTTTTTHTT and picked another coin to get TTTHTTTHTT. In other words, the two possible ways of choosing two heads result in identical sets of flips, so we have to divide by 2. That means the number of ways of getting exactly 2 heads is (10 x 9) / 2 = 45 which occurs with a probabiliy of 45/1024 ≈ 0.0439.

There are 10 x 9 x 8 ways to get to 3 heads, divided by 3 x 2 ways to arrange 3 heads identically, leaving 120 possible ways to get exactly 3 heads or 3 tails, with a probability of 0.117. By similar logic, 4 heads or 4 tails come up 210 ways with a probability of 0.205, and 5 of each comes up 252 ways with a probability of 0.246

If we plot the possibility of each possible number of heads, the probabilities look like this:

For Model Justice 1.1, I’m going to be assuming that the quality of cases charged against defendants assume a similar profile. I don’t know how many factors go into building a criminal case, but I assume it’s a lot more than 10. And since I’m just picking a distribution rather than modeling each of the factors, I don’t really have to care. That means that instead of using a binomial distribution, which depends on a fixed number of factors (you can’t flip 8.7 coins), I’m going to use a continuous approximation to the binomial distribution called the normal distribution, which looks something like this:

This is, I must admit, something of a cop-out. The normal distribution is probably the most common distribution ever, at least when discussing real-world phenomena. (Uniform distributions are common in games and gambling.) In fact, scientists often assume data points are normally distributed even when they really aren’t. It’s one of the sins of statistics, usually committed because there are so many more statistical tools available if you can just assume normality.

However, when dealing with data taken from the real world, there are solid statistical reasons for believing that the normal distribution may be close enough. The real world is often messy, with lots of different effects contributing to the thing you are trying to measure. And when a measurement is influenced by lots of small random variables, each of which can nudge it slightly higher or lower, it turns out that the result often does look a lot like a normal distribution.

This can be illustrated with something called a Galton board. That’s a slanted board with pins sticking up out of it in a regular pattern of offset rows. You drop balls into it at the top, and they trickle down. Each time they hit a pin, they can roll off to the left or right, at which point they will hit a pin from the row below, again rolling off to the left or right, and so on. At the bottom of the board, right below the last row of pins, there’s a set of vertical slots to catch and stack the balls. The height of the stack will tell you how many balls ended up there, and if you dump enough balls into the board, the stacks will follow (more or less) a normal curve.

Here’s a decent video with some explanation (skipping the Galton bio at the beginning):

(You can buy the Galton board used in the video here.)

So that’s my justification for using a normal distribution. I can imagine of a number of reasons why the quality of actual criminal cases wouldn’t follow a normal curve but, at least for now, it’s good enough.

I do need to make some adjustments to fit the model. For one thing, the allowable values for a defendants probability of guilt are from 0 to 1, so I might want to drop the normal curve right in the middle of it, something like this:

There are a couple of problems with that idea.

First of all, the normal distribution is a mathematical construct, so despite the nice appearance here, the left and right tails actually extend out to infinity. So what does it mean if we pick a random value out of the normal curve and the probability of guilt is 3.5? My thinking is that the normal curve doesn’t really represent the probability of guilt. Rather, it’s a crude representation of the quality of the case. So if we get values that are greater than 1, it means the case is more than good enough to guarantee (with p = 1) a conviction. Similarly, a case quality less than 0 means there is no way (p = 0) it could lead to a conviction. So if I get a value outside the range 0 to 1, I’ll snap it to 0 or 1.

The second problem is that there’s no reason to believe the average probability of guilt is exactly 0.5. As I said, the normal distribution actually represents the quality of the case, and it would be quite a coincidence if the average case quality just happened to be exactly good enough to produce a perfect split between guilty and not guilty verdicts. It’s likely that strong cases are more work to put together than week cases, so there will be a tendency to produce a lot of weak cases:

Then again, it’s possible that prosecutors avoid charging extremely weak cases that they don’t think they can win, leaving only the strong cases:

The third problem is that cases may deviate from normality in important ways. For example, the practice of discarding bad cases may lead to a different shape of the left-hand side of the distribution curve. Or maybe there’s always a small chance of a not-guilty verdict, from jury excursions or random black swan events, so that even the best cases can sometimes lose. I have no idea, so I’m going to ignore these possibilities for now. (I may add something in later.)

Incorporating this into the model is pretty straightforward, and you can follow along on the new Model Justice 1.1 model.

The shape of a normal distribution curve is described with two values:

• Mean: The average value. Or the value right in the middle of the curve at the highest point. (One of the handy properties of normal distributions is that the mean, median, and mode are all the same.)
• Standard Deviation: This is basically a measurement of how wide the curve is. The greater the standard deviation, the more the curve stretches out to the left and right.

I’ve replaced the Probability of Guilt slider, and the underlying variable, with Guilt Mean and Guilt Standard Deviation sliders that control the shape of the normal distribution. The values shown will cause this model to produce the same ProbGuilty values as version 1.0:

I also modified the code in the Arrest action to generate a random value from the normal distribution. If the value is outside the range [0,1], it will be snapped to the nearest bound. I.e. Values greater than 1 will be snapped to 1 and negative values will be snapped to 0 so they are valid probabilities. Unfortunately, Insight Maker doesn’t give me any way to show the shape of the generated curve. If you want to get some idea what the curves look like, you can use an online normal curve generator, like this one.

(I’ve also added an ID generator that slaps an ID number on each defendant agent, and I’ve added a global Stats object that collects some internal data that I use for debugging.)

Sadly, after all this discussion, the results from the new model are not noticeably different from the results from the old model. It was generating random guilt probabilities around a specified mean, now the new model is generating random guilt probabilities around a mean that is itself randomly generated around a mean. That tends to look like the same thing.

Nevertheless, these changes are a step toward modeling plea bargaining, because they will keep all defendants from making the same decisions.

This post by Mark Draughn at Windypundit was originally published at Model Justice 1.1 – Random Guilt

(If you’re interested, read the whole thread.)

For as long as I’ve been following criminal defense lawyers on the web, they’ve been discussing the way the criminal justice system encourages defendants to accept plea bargains rather than insisting on trials to force the state to prove its case in court. Often it’s some pissed off lawyer who wants to eliminate plea bargaining or end pretrial detention for routine offenses.

I think these kinds of reform proposals raise some interesting questions, and I’d like to explore them. I have some of my own ideas about how plea bargaining works, and why the system is the way it is, and I’ve been thinking of writing a blog post or two about it. But then I got a crazy idea…

For a while now, I’ve been trying to teach myself how to do system dynamics modeling, and these kinds of questions seem like they would make an interesting subject. I’m going to try to build models for some of the strategic issues affecting trials and punishment.

Models of systems are necessarily much simpler than actual systems because the real world is too complex and messy to recreate in a hand-built computer model with perfect detail, and yet we build models because we want them to tell us useful things about the real world. Thus the art and science of model design involves choosing which details the model must capture and which details it can safely ignore.

Before you can do that, however, you have to know how the model will be used. If you are designing a new type of commercial airliner, and you want to know the predicted fuel efficiency under a variety of flight regimes, you may need to build a detailed simulation of the engine and flight control surfaces and connect them to an aerodynamics model parameterized by a weather model of the troposphere and stratosphere. On the other hand, if you work for an airline that is operating the new airplane, you may be able to plan flights around the country using little more than a few tables that relate speed, altitude, and fuel consumption.

A related concern is the accuracy of the model in making predictions. A company planning to build a package sorting facility may do extensive modeling to test the layout and operation of machinery and workstations to achieve optimum efficiency. This has to be good enough to generate quantitative results that will help project managers spot subtle problems, detect conflicting operations, capture transient events, and calculate costs.

At the other end of the accuracy spectrum are relatively simple models that capture broad relationships within a system without attempting to produce mathematically precise predictions. A lot of social science and economic models fall into this category. These are tools of exploration and learning, and we use them to test our intuition and refine our understanding.

That’s what I’m aiming for here. The model I’m building is going to start out extremely simple — it’s not even going to have plea bargaining at first. I hope to improve the model in future posts so it can exhibit more complex behavior including, eventually, plea bargaining. But before I can model plea bargaining, I have to model the circumstances under which plea bargaining arises, and that’s going to take a while.

I’m building this model using a simulation tool called Insight Maker. I chose it because it’s totally free, and because the simulation engine and development environment are written in JavaScript, which means it runs in a web browser. In addition, once I’ve got a model I like, I can share it on the Insight Maker website, and other people can run it directly in their browsers without needing to download and install specialized software.

I’ve decided to use an agent-based model. That means each defendant will be represented in the model by a separate simulation entity which goes through a series of state changes as the model runs. For this initial model, each defendant will progress through three states: Awaiting Trial, On Trial, and (if found guilty) Prison.

I’ve decided to use a simulated time interval of one month between each iteration of the model. That strikes me as a reasonable unit of time for describing the period spent awaiting trial and the period spent in prison. For the sake of some technical convenience, the period spent on trial will be modeled as lasting exactly one month. That’s unrealistic, but I don’t think it’s unrealistic in a way that matters.

I could switch to a shorter unit of simulated time, such as a single day, to make the model more flexible, but I decided to keep it at a month for three reasons: First, I don’t think a finer time resolution will produce a better model. Second, using a small time unit suggests that the model is more accurate than it ever could be. Third, iterating daily instead of monthly would increase the number of iterations by a factor of 30, greatly extending the simulation’s run time.

The complete model, or at least the first version of it, looks like this:

(The actual model is here on the Insight Maker website if you want to follow along or play with it yourself.)

The model has four variable parameters which can be controlled by sliders appearing on the page.

Here’s what they control:

• Arrest Rate: This is the number of defendants arrested each month. Every iteration of the simulation will fire the Arrest action, which will create this many defendant agent entities and add them to the Defendants pool, starting in the Awaiting Trial state.
• Time to Trial: This is the amount of time (in months) that defendants will spend waiting for their trial. In this version of the model, nothing interesting happens during this period, and I could have omitted it. However, I think it will be useful in future versions of the model.
• Probability of Guilt: This is the chance that a defendant will be found guilty at the end of the trial. In this version of the model, every defendant has the same probability of conviction. Future models will allow different probabilities of conviction, as a crude measure of the quality of the case. In preparation for that, each defendant agent in this model has an independent ProbGuilty property, but they are currently all set to the same value when the defendant agent is created, and that value is used in determining the trial outcome.
• MaxSentence: Specifies the maximum sentence for the defendant if convicted. As with probability of guilt, this is currently the same for every defendant, but the agents have independent NominalSentence properties that can be used in the future to adjust sentences on a per-defendant basis.

In addition, each agent has other properties associated with it to contain more detailed state information. I’ve added data items to represent the actual sentence (which is always the maximum sentence in this version of the model) and to help count the sentence down as it is served.

The meat of the model is the Defendant agent and its lifecycle, pictured above. Defendant agents begin their life in the model when the Arrest action fires, at which point they receive their probability of being convicted and their expected sentence. They sit in the Awaiting Trial state for however long Time to Trial specifies, and then they transition to the On Trial state and go through a one-month trial. In the Verdict action, they are found guilty or not guilty with a random draw, based on the probability of guilt assigned to their agent when it was created.

If found not guilty, they are removed from the defendant pool and from the model. If found guilty, they move to the Prison state, and the Serve Time action counts down their sentence with each iteration until their sentence is done, at which point they are removed from the model.

Running the model with the above parameters yields a graph displaying the number of agents in each model state over a simulated period of 10 years:

During the first four months, defendants are added to the model in the Awaiting Trial state (green line) at a rate of 30 per month, for a total population of 120. After that, as each new group of 30 defendants is added to the model, the group of 30 from four months ago stop waiting and enters the Trial state (red line). A month later, those 30 are remove from the Trial state, and a new batch of 30 begins trial. For the rest of the simulation run, these values remain in a steady equilibrium state, with 120 defendants awaiting trial, and 30 defendants on trial.

Starting at the 5th month, 30 defendants per month conclude their trials. Half of them, chosen randomly, are found not guilty and removed from trial. The other half — about 15 per month, on average — go to the Prison state. The prison population continues to build up at that rate until the 41st month, at which point the first batch of defendants is done serving their sentence. Since 15 prisoners per month serve for 36 months, the prison population stabilizes at around (30 x 0.5 x 36 =) 540 prisoners from there on out. Unlike the other populations in the model, the Prison population is influenced by the random events of trial, therefore the Prison population exhibits random rises and falls 

The advantage of using a model is that we can play around with it. If we “get tough on crime” by arresting an extra 10 defendants per month and sentencing the guilty ones to an extra year in prison, the steady state prison population booms to (40 x 0.5 x 48 =) 960 prisoners.

These results are not surprising, nor should they be, with such a simple model, and we didn’t really need an agent-based model to figure out these numbers, as demonstrated by the simple math in the preceding paragraphs. We could easily have modeled these same results with a few spreadsheet formulas. I think, however, that an agent-based model will be more suitable when we want to model more complex situations.

A few limitations or simplifications of this model are worth noting:

• The model assumes there are no resource limits. No matter how many defendants are arrested, there will always be enough prosecutors, defense lawyers, and criminal court rooms to try them all without delay. There’s also plenty of prison space for everyone found guilty.
• Everyone commits the same crime. If you wanted to model multiple crimes, or multiple degrees of severity, you’d have to re-run the model multiple times with different parameter settings and combine the results. Note that this implicitly assumes that cases of one type do not interfere with cases of another type, for which it is necessary, but not sufficient, to assume that there are no resource limits.
• As mentioned, all defendants have the same chance of being found guilty, which implicitly assumes that the cases against all defendants are of equal quality.
• If found guilty, all defendants are hit with the same sentence. There is no uncertainty about the sentence if found guilty.

As I develop future models, I’ll be relaxing some of those constraints.

Finally, you may be wondering why I’m bothering building models. After all, there are people who study this kind of thing for a living. They probably have much better models than I could ever build in my spare time. Why don’t I just use their models? Or better yet, why don’t I just read their criminology papers to find out what they’ve already learned?

The short answer is: Where’s the fun in that? I like exploring new ideas and learning new things.

And if you’re a criminal lawyer, prosecution or defense, who finds this project at all interesting, I could use some help. I want to build more elaborate versions of this model, and I need help figuring out what aspects of the criminal justice system belong in this model, and what aspects can be safely ignored. I need to figure out how to model plea bargain decision making on both sides, and I’m not sure where to begin. I’m a professional software developer, but my interest in the criminal justice system is strictly amateur. Nevertheless, I’d hate to put in so much work into the models — and these posts — only to discover that I’ve been making my readers stupider.

You don’t want that on your conscience, do you? Besides, it might be fun.

This post by Mark Draughn at Windypundit was originally published at Model Justice — Part 1

Political campaigns are mostly free speech, which should be well-protected by the First Amendment, but campaign finance laws attempt to control the money that pays for producing or disseminating this speech. This requires a lot of weird rules, and the results can be bizarre.

For example, the media is exempt, because that’s clearly freedom of the press. So you and I can endorse candidates on our blogs and tell people to vote for them in post after post with little fear of campaign finance laws. However, if you and I buy ads on each other’s blogs saying the same thing, then those are considered independent expenditures, which don’t count as direct campaign contributions, but which are nevertheless required to contain legally mandated disclaimers. (I think they also have to be reported, but I’m not sure because election laws are really complicated.)

There’s an obvious loophole here. If I wanted to donate a million dollars to a candidate, that would be prohibited because it greatly exceeds the limit for campaign contributions. But what if I asked the campaign to tell me how they would spend another million dollars if they had it, and then I went and spent the money myself on exactly what they said they would spend it on? The money never passed through the campaign’s accounts, so it’s not a contribution, right?

Campaign finance regulators have already thought of that trick and headed it off: It’s illegal to spend money to benefit a candidate if you coordinate that expenditure with the campaign. Or to put it another way, if you’re spending money to help a political campaign, it’s illegal to talk about it with the people you’re trying to help. That’s kind of a weird requirement. We’ve gone from free speech to a crime that includes speech as an element.

What if I don’t spend any money? What if I own a bus company and I let the campaign use my buses for free? Campaign finance rules call that an in-kind contribution, and it has to be treated as if it were a financial contribution in the amount of the value of the good or service contributed.

Okay, so what if my best friend Leo is the campaign coordinator, and I pick him up at the airport and then we go have lunch at a restaurant that’s next door to the hall where the candidate is having a rally later that evening? If I’m driving Leo to the restaurant because he’s a campaign staffer, that’s an in-kind contribution. But if I’m driving Leo to the restaurant because I want to have lunch with my friend, that’s not a campaign contribution. It all depends on what reason I have in my head for giving Leo a ride.

That’s pretty weird, right?

It reminds me of a trap in gun laws where it’s legal to buy a gun (if you meet all the qualifications), and it’s legal to sell that gun later (if the buyer meets all the qualifications), but it’s a crime to buy a gun knowing that you will sell it to a specific person later. You can buy the gun intending to keep it for yourself, only to change your mind and sell it later. Or you can buy the gun intending to give it to a specific person as a gift. You can even buy the gun intending to sell it later to some unknown future buyer. But if you know the identity of the person you’re going to sell it to at the time that you buy it, you’re breaking the law. Having that person’s name in your thoughts is an element of the crime.

Cohen paying off Trump’s mistresses leads to similar issues. If he’s paying them off so that Melania doesn’t find out about them, that’s completely legal. But if he’s paying them off to avoid bad press that could harm Trump’s presidential campaign, that’s against the law. It’s easy to imagine Trump and Cohen getting together for one of their regular paying-off-the-mistresses meetings, and everything is going just fine until Trump says, “Hey, this could also help my campaign!” and Cohen says, “Yeah,” and now they’re committing a crime.

There’s nothing unusual about this — that sort of legal logic is well established, and people are prosecuted for it all the time — but you gotta admit it’s kind of a weird way to commit a crime.

Manafort was convicted of more tangible crimes, including five counts of tax fraud and two counts of bank fraud. But he was also found not guilty of 10 other charges. And here’s the part that makes me uneasy: Those 10 acquittals aren’t going to matter much when it comes to sentencing.

That’s because federal sentencing laws allow judges to take a very broad view of the defendant’s conduct when deciding how long to send them to jail. Almost any conduct that led to their appearance before the court can be taken into account for sentencing purposes, including conduct for which a jury has found them not guilty.

That’s one of those bizarre features of that our criminal justice system — civil forfeiture is another — where people who’ve never heard of it tell me I’m stupid because what I’m describing just wouldn’t be allowed. But it is.

So, those 10 acquittals aren’t going to count for much at Manafort’s sentencing because the judge can ignore them and punish Manafort for all the things he was accused of, regardless of what the jury decided.

It ruins my enjoyment of Cohen and Manafort’s downfall when the mechanism used to bring them down seems so petty and corrupt.

This post by Mark Draughn at Windypundit was originally published at Why I Can’t Enjoy the Cohen and Manafort News

One of the confusing things about this horrible incident is that from what we know about him so far, Paddock doesn’t seem like the kind of person who would do something like this. He went through a lot of trouble to kill as many people as he did, so you’d think he must have done something else on a smaller scale first to think this would be worth the effort, and yet according to reports he had no criminal record, no extremist political views, no trouble with his neighbors, and no obvious prior signs of mental illness. That’s strange because you don’t go from being a normal person to being a mass murderer overnight without a period of transition. The idea that people “just snap” is more myth than reality. It takes time to become the kind of person who would commit this kind of atrocity. So why weren’t there any signs?

I suppose one possible explanation is that Stephen Paddock was not the shooter: The real shooter lured him (and his guns) to the hotel, shot him, and then opened fire on the crowd, escaping before the police arrived to find a “convenient” murder-suicide scenario. I don’t actually believe this is what happened. It’s a movie plot, not real life, and unless the perpetrator is a Moriarty-level criminal mastermind, it would also leave a ton of evidence that would be easy for the police to discover. It’s even less likely than just snapping.

(Also, recent reporting about the incident rules it out.)

Another possibility is that Paddock had a brain tumor that caused a sudden, violent change of behavior. No doubt this crossed my mind because I’m familiar with the case of Charles Whitman, who similarly shot at a crowd from a high perch, at the University of Texas (Austin) in 1966, and was discovered at autopsy to have a brain tumor which some neurologists have speculated may have contributed to his behavior. Paddock’s behavior could also be explained by some other form of brain damage, perhaps from a stroke or a traumatic head injury or some other cause.

The truth is, the most likely explanation for the conflict between Paddock’s normal life and its violent end is probably much more prosaic. It’s most likely the same explanation as for every other murderer who is described by acquaintances as “quiet”: We’ve only heard from people who didn’t know him very well.

When something like this happens, it’s easy for the press to find the killer’s neighbors and colleagues, but few of them are likely to know anything relevant. As with most of us, a lot of people may have known him, but few people knew him really well. The media, and maybe the police, just haven’t found people who know, if they even exist.

This post by Mark Draughn at Windypundit was originally published at The Unknown Reason for the Vegas Shooting

I was most struck by the reaction from Mistress Matisse:

Now, this next part is probably the sort of thing that only seems profound at 3:00 in the morning (when I started writing this), but bear with me… What really got me thinking is that Mistress Matisse and Joe Arpaio have something in common: They both like to hurt people.

As you might have guessed from the name, Mistress Matisse is a professional dominatrix. She literally hurts people for a living. And if you follow her Twitter feed (@mistressmatisse), you’ll see she enjoys her work. Oh, I’m sure there are days where she’s like “Oh God, I can’t wait to veg out in front of the TV once I finish sticking needles into this guy’s penis…”, but for the most part, she seems to have found satisfying employment. She hurts people, and she’s good at it.

Nevertheless, Mistress Matisse is pissed about Joe Arpaio, because of course, the difference between Mistress Matisse hurting people and Joe Arpaio hurting people is consent. And consent makes all the difference in the world.

Mistress Matisse posts descriptions and photos of the some of the things she does to her clients and, honestly, I couldn’t take that. If she tried to do some of those things to me without my consent, I would resist, violently if necessary. But her clients not only give their consent, they actually pay her to do those things.

Yet if I’m ever in Seattle and I happen to run into Mistress Matisse, I wouldn’t be the least bit concerned about her pain-causing skills. I don’t pretend to know or understand her very well, but I’ve read enough of her writings to know that she thinks very carefully about issues of consent. She doesn’t want to hurt people against their will.

Sheriff Joe Arpaio, on the other hand, doesn’t give a shit about consent. He’s a thug, and he had an army of deputy thugs working for him. I would be very nervous running into him in a situation he controlled. If he could find a way, he wouldn’t hesitate to have a critic like me thrown in a cage in the 110 degree heat of the Arizona sun.

I don’t have any profound point here. And I certainly hope this is not coming across as “Joe Arpaio is even worse person than a professional dominatrix,” because that would be insulting to the pro dom community. The juxtaposition between Mistress Matisse and Joe Arpaio just struck me as a stark illustration of the importance of consent.

As a someone who leans libertarian, I place a lot of importance in the concept of consent. And I regard the absence of consent as a defining requirement for legitimately designating something as a crime: Sex without consent is rape, commerce without consent is theft, and inflicting pain without consent is torture.

Lack of consent alone is not enough to make something a crime, but if there is full consent, then it should never be a crime. As far as I’m concerned, Mistress Matisse is a successful and valued practitioner of an unusual art. Nothing to worry about.

Joe Arpaio, on the other hand, hurt a lot of people without their consent. Obviously, detaining and punishing criminals without their consent is a necessary part of criminal justice, but the lack of consent also makes it especially important that the criminal justice system is subject to strict requirements and oversight. Arpaio had little of that, and he and his deputies went way beyond what was necessary to keep the peace.

It’s particularly relevant to the contrast with Mistress Matisse that Arpaio jailed consensual sex workers and subjected them to harsh conditions, including one particularly gruesome case in which his guards tortured and killed a sex worker. Yet that’s just one of many deaths and many more prisoners subject to torturous conditions. Arpaio was one of the most monstrous figures in modern American law enforcement.

There’s something very wrong, some kind of complete inversion of the meaning of consent, when we have governments that want to throw people like Mistress Matisse in jails that are run by people like Joe Arpaio.

This post by Mark Draughn at Windypundit was originally published at The Dominatrix and the Sheriff

To make it more concrete, suppose I’m walking down the street, minding my own business, when a stranger confronts me, thrusts a duffle bag into my hands, and runs away. When I open a duffle bag, I find a tightly wrapped kilo of cocaine, a pile of child pornography, and a MAC-10 submachinegun. As I look up, I notice several police officers coming down the street, obviously searching for someone or something. They haven’t noticed me yet. What should I do next?

As a good citizen who has come across evidence of a crime, it seems like I should probably turn all of this in to the police. But if I walk up to one of the cops and show him the bag, I’m essentially confessing to possession of the contraband items and proving it with physical evidence. My possession was unintentional, and that might make for a winning defense at trial, but I’m not so sure it would keep me from being arrested. This seems like a risky choice.

If I had a lot of time to decide what to do, I could contact a criminal defense lawyer to help me. I imagine he could probably make a deal where I don’t get arrested for turning over useful evidence of a crime. Or maybe there’s some other solution.

On the other hand, if I didn’t have the time to talk to a lawyer, I think the safest move would be tossing the duffle bag into the nearest dumpster and hoping that nobody saw me with it. That seems like the simplest way to get out of a sticky situation, but is it? Do I have a legal obligation to report the contents of the duffle bag to the police? After all, I saw the other guy possess it, so I’m a witness to a crime, and I’m also in possession of physical evidence. If I throw away the duffle bag, am I concealing a crime?

My reason for asking this is because of a New York Times opinion piece in which Doctor Sandeep Jauhar describes an incident in which he feels he protected doctor-patient confidentiality:

I once took care of a business executive in the emergency room who had hired call girls during a weekend drug binge. When he saw a police officer outside his room, he quietly handed me an envelope containing a large amount of white powder. I wasn’t sure what to do with it, so I discarded it. For the next several hours the patient eyed me suspiciously, probably wondering whether I had ratted him out. But it never occurred to me to do so.

Professor Steven Lubet from the Pritzker School of Law at Northwestern takes issue:

Hiding evidence of a crime isn’t confidentiality; it’s obstruction of justice. There is nothing about one’s status as a physician — or a lawyer, for that matter — that requires or excuses the possession or concealment of contraband.

Jack Marshall also disagrees:

Confidentiality is one thing, assisting in a crime is another. The Hippocratic Oath says “What I may see or hear in the course of treatment, I will keep to myself.” That only means, however, that doctors who learn about criminal activity a patient may be involved in is bound not to report it (lawyers have the same obligation). Jauhar did more than not report criminal activity; he participated in it. He crossed the line by disposing of contraband.

There’s a side issue here, because an unidentified white powder in an envelope doesn’t prove anything. To this day, Dr. Jauhar has no idea what was really in that envelope. And it doesn’t matter that a cop might have arrested the patient upon seeing that powder without any proof that it was an illegal substance. Nor does it matter that a judge might have held him in jail for weeks before getting a lab report identifying the powder as an illicit substance. The epistemological failings of the criminal justice system have little relevance to medical ethics.

The main question here is what should Doctor Jauhar have done? Marshall says he “crossed the line by disposing of contraband” but then doesn’t explain what he would have done if he had been in Jauhar’s place. Lubet also fails to describe what the doctor should have done. He mentions the problem, but he doesn’t seem to recognize it: “There is nothing about one’s status as a physician — or a lawyer, for that matter — that requires or excuses the possession or concealment of contraband.” When it is illegal to both possess and cease to possess something, that doesn’t seem like very good lawmaking.

Are Lubet and Marshall saying that if they had been in the doctor’s place that they would have walked over to the nearest cop and said, “I just got this from a patient and I thought you should see it,” and then handed him the envelope full of (probably) cocaine, taking it on faith that he wouldn’t get arrested?

It doesn’t work to say they wouldn’t have taken the envelope in the first place, because how could they know what was in it until they looked inside? And please don’t say that the doctor should have handed the envelope back to the patient once he saw what was in it, no unless you want to explain why giving illegal drugs to a patient is more ethical than discarding them.

You might think it’s crazy to believe the police would arrest a doctor for turning over drugs that he found on a patient, but just because the police might not arrest him for it doesn’t mean it’s not a crime. And if it is a crime, they might arrest him for it. It wouldn’t be the first time police decided to arrest someone now and sort out the details later.

I admit that I haven’t heard of any doctors being charged in situations like this, but it does happen to other people. I have read news reports of security guards at music venues taking illegal drugs off concertgoers and then getting arrested for possessing illegal drugs. So unless there’s an explicit exception in the law for doctors in this situation, it seems to me that Dr. Jauhar was in danger of getting arrested, and disposing of the envelope was the prudent thing to do. (This is not legal advice, but it’s very, very practical advice.)

The basic problem I’m having with this whole scenario is that mere possession of contraband is inherently a victimless crime that harms no one. That’s not just my opinion, it’s physics. Granted, there are some substances which can actively endanger innocent people if not contained properly — radioactive materials, unstable explosives, poisonous gases, biosafety level 4 pathogens — but a big pile of cocaine or child pornography or guns does not hurt innocent people merely by existing. And possessing that pile of contraband doesn’t mean you’re hurting people either.

Further proof that crimes of possession are not based on reality comes from the fact that members of the criminal justice system routinely do the same thing as the people they are arresting and prosecuting. If a cop finds a baggie of cocaine in your pocket, he’ll arrest you for possession and then take the baggie of cocaine. That means he’s now in possession of a controlled substance, and yet somehow that’s not a problem. Depending on the nature of the contraband and the standard practices of each court system, the contraband could also end up in the hands of the prosecutor, the defense lawyer, the judge, and even the jury. And apparently none of those people are committing a crime.

(Actually, that might not be the case. Legislatures are notorious for not including exceptions for law enforcement activities in the criminal statutes, meaning that in some states things like police sting operations, SWAT raids, and even death penalty executions are technically crimes. Another way to put that is that police and prosecutors sometimes break the law in the course of their jobs, but nobody charges them with crimes.)

We criminalize the harmless possession of contraband because it’s a useful proxy for other crimes that are more difficult to police. It’s a lot harder to catch somebody smoking pot in their own home than it is to catch them with pot in their pocket as they walk home. It’s a lot harder to catch an armed robber than it is to catch someone in possession of an illegal gun. It’s a lot harder to catch graffiti artists tagging a building than it is to catch them buying spay paint. All other things being equal, police and prosecutors prefer to go after crimes that are easier to prove.

Unfortunately, we also criminalized possession of various contraband goods out of a desire to “send a message” that some related behavior is bad, or because the people enforcing the laws are confused enough to believe that there really is something harmful about possession of contraband.

And so our criminal justice system keeps doing stupid things like arresting people for child pornography images on computers that they possess on behalf of someone else, or arresting people for “residue,” or intercepting packages of drugs in the mail, delivering them to the recipient on the label, and then arresting that person for possessing the drugs that they just gave him.

Possession is often kind of a stupid crime, stupidly enforced. And don’t get me started on crimes based on constructive possession…

This post by Mark Draughn at Windypundit was originally published at Possession Is Such a Strange Crime

And consider what happens if the police bust you for possession of marijuana: The police will confiscate your stash. At that point they are the ones who are in possession of marijuana, aren’t they? Why is it that it was illegal when you did it, but it’s legal when they do it?

Because the law says so, I guess, but why? Cops can’t legally break into someone’s house and steal their jewelry or force women into having sex with them, so why is possession different? What’s the principle here? Clearly, since cops are allowed to do it, possession of marijuana doesn’t actually cause whatever social harm the law is seeking to prevent. So why is possession still illegal?

I think the answer has to be that possession is a proxy for the behavior that causes actual harm. In the case of marijuana, that would be the harm caused by smoking marijuana. (As a libertarian, I don’t see any net social harm from smoking marijuana, but let’s ignore that for the sake of argument.)

The problem — for the drug warriors, anyway — is that it’s hard to catch people smoking marijuana because it only takes a few minutes and they can do it in hiding. So the legislatures have moved backwards on the causal chain to criminalize a precursor act that is easier to detect: People generally possess marijuana for a lot longer than they smoke it, and they often do so in relatively public places where police can observe them, question them, and search them. Thus criminalizing the possession of marijuana serves as an imperfect way to attack the problem of smoking marijuana, at the cost of criminalizing behavior that is not really harmful.

This explains why it’s not a crime for cops to possess marijuana after a bust: They aren’t going to smoke it later.

Other crimes of possession can also be justified by their relation to harmful acts: We criminalize felons possessing guns because we don’t want them murdering people, we criminalize young people possessing cans of spray paint because we don’t want them vandalizing buildings with graffiti, and we criminalize possessing open containers of booze in cars because we don’t want drivers to drink.

The exact conditions under which it’s okay to punish people for harmless behavior in the name of stopping a related harmful act is a moral and ethical question worthy of exploration, but for now I’d like to look at a more practical problem. Because the behavior that is illegal is distinct from the behavior that is harmful, scenarios can arise where the law doesn’t really make sense.

For example, it’s illegal to possess child pornography. Clearly, mere possession is harmless, as proven by the fact that police and prosecutors can possess child pornography in the process of sending people to jail for it. Possession of child porn is nevertheless illegal because we don’t want people to look at it.

Houston criminal defense lawyer Mark Bennett gives an example of how this can lead to a bizarre dilemma:

Suppose that a client comes to you with a problem: he has a computer hard drive full of child pornography, and he wants to know what to do with it. What do you tell him?

It’s illegal for him to continue possessing the images. So you can’t advise him to do nothing (and keep breaking the law).

The smart thing for him to do would be to destroy the hard drive (if I could, I would recommend swisscheesing it with a drill press).

But tampering with evidence is illegal under both Texas and federal law.

This is not just a hypothetical situation. At least one lawyer has been prosecuted for destroying child pornography on a client’s computer.

Lawyers wouldn’t be facing this kind of conundrum if it weren’t for the crime/harm separation that is inherent in possession laws. If possession of child pornography wasn’t a crime, then deleting it from the hard drive wouldn’t be destruction of evidence of a crime. It would arguably even be crime prevention, since deleting it prevents people from looking at it, just as flushing marijuana down the toilet prevents people from smoking it.

(I’ve never heard of someone being charged with destroying evidence of pot possession by smoking it, but people do get charged with tampering with evidence for eating drugs when the cops pull them over.)

On the other hand if the crime and harm of possessing child pornography were linked, if mere possession somehow did cause actual harm, then deleting the pornography from the drive would be stopping a crime in progress. The lawyer would be a crime fighter.

If you have trouble thinking about possession being harmful in that way, try imagining it this way: The lawyer visits the client’s house, but instead of learning that the client has child pornography, he discovers that the client has an actual child, locked in a cage in the basement. Alarmed, the lawyer frees the child, who immediately runs off. The child is confused, however, and is never able to identify his captor or the location he was held. By releasing the child, the lawyer has in a sense obstructed the ability of the police to charge his client for kidnapping, yet no sane person would argue that he shouldn’t have done it.

I’m not going so far as to say that possession should never be a crime, but the fact that possession crimes don’t work that way is a sign that possession is a strange kind of crime, one where stopping the crime is also a crime. I’m not sure how to fix this, but I’m pretty sure it’s broken.

This post by Mark Draughn at Windypundit was originally published at The Crime You Dare Not Stop

News

In most cases of lethal force, not only do we not know what happened, we don’t even have access to anyone who knows what happened. All we get are accounts passed to us by news reporters, editorial writers, bloggers, and acquaintances. As I’ve tried to show, these incidents can get very complicated once you take into account the ethics of lethal force, subjective perceptions, and the possibility that people are lying. Meanwhile, news stories necessarily have to economize on detail, because there’s limited space and limited time to write. Therefore it’s highly unlikely that any of our sources of information will pass along every detail we need to know.

Furthermore, the people telling the story are also subject to all the same distorting influences as the witnesses: They could make mistakes interpreting witness accounts, they could misremember witness accounts, or they could spin the story (or outright lie) to serve their own agenda. Even if they try to keep everything straight, they’re bound to make mistakes, and they’re bound to distort the account based on their own interpretation.

The same thing will happen all over again when people read these stories and pass them on, comment on them, or write blog posts about them (and don’t even get me started on Twitter). It ends up being like a game of telephone, except that instead of passing the message along in a line, from one person to the next, it’s more like something happens to people in the center and information about it radiates outward along multiple interweaving paths. Each person in the information flow receives different stories from different people at different times.

Further, as they receive each piece, they can either adjust their mental model of what happened to accommodate the new information, or they can reject the new information because it seems unlikely to be true, based on the model they have so far. In this way, everybody comes to believe a different version of the story, and they will interpret new information based on the framework established by that story.

Argument

That we all have different ideas of what really happened is a big problem when we start to argue about what it means. It’s hard to have a discussion about right and wrong, crime and punishment, when we don’t agree on the facts. It’s doubly hard when we don’t realize that we’re all working from different mental models of the main incident.

I first noticed this several decades ago, back before the internet was everyone’s source of news, as I watched the story of one such incident dribble out in a series of small revelations. I can’t remember enough details to look it up, so I’m certainly misremembering it, but I’m offering it as an example of a phenomenon, not to discuss the specifics of the incident. The series of revelations, in the order I encountered them, went something like this:

• A drunk guy visiting a friend late at night accidentally went to the wrong house and was yelling and pounding on the back door when the homeowner shot and killed him. To me, it was hard to see how the homeowner could possibly be in reasonable fear for his life, so this sounded like some kind of murder.
• The homeowner said he thought the guy was trying to break in. Burglary is not a reason to kill someone, but a guy breaking into the house while you’re there might justify lethal force. However, it was difficult to agree with the homeowner’s perception, since pounding and yelling aren’t really a break-in.
• Before pounding and yelling on the back door, the guy had rung the front doorbell. Presumably he thought his friend was asleep, so after the doorbell didn’t work, he went around back to try to get his attention. Some people argued that burglars never do that, but I had heard that ringing the doorbell was a great way to find out if a house is unoccupied without doing anything clearly illegal. If the homeowner answers, just say you must have the wrong address and walk away. Or if you’re the violent type, ringing the bell gets the homeowner to open the door for you, so you don’t have to risk attracting attention by breaking in. This information seemed to cut either way in terms of fault.
• The most damning detail to come out was that the homeowner shot the guy through the back door. This seems indefensible. You’re never supposed to shoot without identifying your target, and you have no idea what’s on the other side of the door — what if a neighbor came over to investigate the noise and was standing behind the guy? What if the guy knocking had a child with him? You just don’t know. Shooting through the door was terribly reckless and made it seem more like a murder of some kind.
• At least until it came out that the back door was a screen door. The homeowner heard noises from the back of the house and went to investigate. He didn’t just find someone pounding and kicking the door. He found someone punching and kicking through the screen into the house, while he was standing there with the gun in his hand. No wonder he shot.

This was actually a fairly simple story, and yet many of the people discussing this story, including the news media, weren’t used to thinking about the issues involved in lethal force, and they had missed a bunch of important points, including the final detail that changed everything. And as these points came out, arguments kept erupting over the incident, often between people working from different versions of what happened.

Laura hears that Alice shot Bob after he broke into her house in the middle of the night, and based on that, Laura concludes that Alice justifiably shot Bob in self-defense and should not be charged with a crime. Natalie hears that Alice shot Bob after inviting him over, and based on that, Natalie concludes that Alice murdered Bob and deserves to go to jail. They have both reached reasonable yet incompatible conclusions based on what they know.

It seems like a simple misunderstanding that could be cleared up with a little discussion, but it’s simple human nature that once we have staked out our positions, we become resistant to considering alternatives, even if the evidence for them is pretty good. Confirmation bias and other phenomenon make us resistant to changing out minds and accepting new evidence, even if the evidence from which we formed our initial opinion is proving increasingly shaky. Further, the strength of our belief makes it difficult to consider that other people may have reasonably interpreted the situation very differently.

When these people get into arguments with each other over the incident, the difference in conclusions is simple and obvious, but people’s internal mental models of what happened are hidden, so folks on each side have trouble understanding how people on the other side could reach a different conclusion, and they often misattribute those conclusions to ignorance, prejudice, or anti-social values rather than recognizing that different assumptions are in play.

If Laura and Natalie hear each other’s conclusions without hearing the reasons for them, they’ll conclude bad things about each other. Laura will be appalled that Natalie wants Alice to go to jail for shooting a home invader — she may accuse Natalie of coddling criminals or believing men should be able to attack women without consequences. Natalie, on the other hand, will be appalled that Laura believes that murderers like Alice should go free — she may accuse Laura of coddling criminals or believing that women should be able to gun down men without consequences.

We’ve seen this sort of thing happen in Ferguson, with the shooting of Mike Brown by Darren Wilson. Some people see an unarmed black kid gunned down by a white policeman and conclude that anyone who doesn’t think the cop belongs in jail must be a black-hating racist who thinks it’s okay for cops to gun down black people in the streets. They want justice for Mike Brown.

While there are certainly some racists who feel that way, I think it’s safe to say that many of Wilson’s supporters simply don’t believe that narrative about the encounter. They think Mike Brown viciously attacked Darren Wilson and tried to grab his gun, presumably to kill him, and then when Darren Wilson tried to arrest him, Brown attacked again and Wilson shot to defend himself. They think Brown’s supporters are just white-hating racists making excuses for black violence, or attention-seekers trying to drum up trouble out of ulterior motives. And while there are certainly some people doing just that, I think it’s safe to say that many of Brown’s supporters simply don’t think he attacked Officer Wilson.

It’s hard to understand other people’s arguments when you don’t realize you’re not arguing about the same thing.

Interlude: The Police

Shootings by police officers have been in the news a lot lately, and they’ve touched off a lot of controversy, especially when the police shoot unarmed black males. For the most part, I think we should analyze police shootings the same as any other, but there are a few special considerations.

For one thing, police get a lot of training in how to handle potentially violent encounters, and this training affects their perceptions and the conditions under which they fear for their lives. For example, a police officer may have been trained that an assailant with a knife who is within 20 feet can stab him faster than he can draw a gun, so he’s likely to regard someone 15 feet away with a knife as a deadly threat, and he can likely point to his training as the reason he was in fear for his life.

Also, police follow a departmental use-of-force policy, which specifies when and how officers can use different degrees of force, up to and including deadly force. So officers may escalate their level of violence according to training or departmental policy rather than according to what a reasonable non-police-officer would do.

In addition, we expect police to actively fight crime. This means that we expect them to look for trouble, and when they find it, we expect them to stand their ground rather than try to run away. That would be suspicious or troublesome in an ordinary citizen, but for a police officer, that’s the job. We actually expect police to make arrests. This necessarily involves laying hands on people. It’s a violent act that would be called battery and kidnapping if anyone else did it. As a society, we want police to do this violence to protect us from bad people who would harm us, so acts that would cause other people to lose their right of self defense are often acceptable when done by a police officer.

On the other hand, just because police are carrying out a public service mission in accordance with their societal role, training, and departmental policy doesn’t magically eliminate the social costs of what they do. Having people with guns who are looking for trouble, standing their ground, and grabbing people off the street to lock them up in cages is still a form of harm to the people they are doing it to, even if they are criminals.

When police officers hurt people according to departmental guidelines, that doesn’t mean that it’s okay. It just changes the discussion from a question of personal ethics to a question of public policy. Instead of asking if the cops are bad people, we need to ask if the cops are following bad policies.

Just because the cops were doing what they were told doesn’t mean the policy, the mission, and the training aren’t wrongheaded, unethical, or downright evil. Just because police departments theoretically serve the needs of the society they’re embedded in doesn’t mean they aren’t up to no good. Like all institutions, they will serve their own needs, which may conflict with the public good.

For example, police conduct tens of thousands of drug raids every year, and on some of those raids the officers mistakenly kill an innocent person. The mistakes may be honest mistakes — the police thought they saw a gun, or they thought they were taking fire — so we might conclude that the officer did nothing deserving punishment, nothing that we’d call evil. At the same time, we might also conclude that staging a series of drug raids that carry the risk of killing innocent people is a policy that does more harm than good.

I’m almost done with this subject, but I think I have one more post in me.

This post by Mark Draughn at Windypundit was originally published at Thinking About Lethal Force – Part 3

In important ways, this is wrong. We should believe, as a matter of default, what an accuser says. Ultimately, the costs of wrongly disbelieving a survivor far outweigh the costs of calling someone a rapist.

Scott Greenfield (amoung others, one supposes) has already responded to some of Maxwell’s argument, and he had more about the UVA rape story in an earlier post. (Note that Maxwell’s article was edited to soften it a bit after some of the early responses.) I’d just like to add a few comments.

The accused would have a rough period. He might be suspended from his job; friends might defriend him on Facebook. In the case of Bill Cosby, we might have to stop watching his shows, consuming his books or buying tickets to his traveling stand-up routine. But false accusations are exceedingly rare, and errors can be undone by an investigation that clears the accused, especially if it is done quickly.

The earlier version of Maxwell’s article post seemed to imply that we should abandon the legal presumption of innocence. This more recent version leans more toward a discussion of the moral and procedural issues of campus rape investigations. Nevertheless, as people have been pointing out, we have a presumption of innocence for a reason.

Maxwell seems to be understating the consequences of a rape accusation, at least when the accused isn’t a celebrity or the archetypical rich frat boy of college rape stories. And although some colleges are bastions of progressiveness, I suspect most of them will share the foibles of the society they serve, meaning that a disproportionate number of accusations will be aimed at blacks, Hispanics, and low income students.

And Maxwell’s assumption that if the accusation is ruled false everyone will cheerfully accept the decision of the investigating body seems naive when the news is filled with stories of protests against grand jury decisions not to indict two cops accused of murdering unarmed black men. Grand juries have an even lower standard of proof than college investigations, yet regardless of how you feel about the case, it’s obvious Darren Wilson’s life will not be returning to normal.

In fact (despite various popular myths), the FBI reports that only 2-8 percent of rape allegations turn out to be false, a number that is smaller than the number (10 percent) who lie about car theft.

Yeah, but that’s now. That’s under the system that Zerlina Maxwell wants to change. (And I’m not saying that some change isn’t necessary.)

It comes down to contingent probability. Imagine you’re a woman walking home down a somewhat deserted street when a man you’ve never seen before pulls up next to you and offers to give you a ride. Do you get in? Only a small fraction of men would rape a woman just because she got in their car, so statistically you should be safe, right?

Of course not, because the small percentage of men who are violent rapists are always on the lookout for a chance to get a woman into a position where they can rape her, so they are much more likely than most men to offer to give a woman a ride. Consequently, the group all men has a small percentage of rapists, but because rapists are much more likely to try to get a woman into their car, the group all men who stop to offer a woman a ride has a much higher percentage of rapists. (Or their more common variant, creepers.)

Similarly, only a small percentage of women are likely to make false accusations of rape, and as long as investigators take rape investigation seriously, they won’t be able to make very many of them. But if institutions adopt an always-believe-the-victim methodology, it gives the kind of women who make false accusations a much greater incentive to do so: They will be able to cause a world of hurt for anyone they don’t like. So if we stop protecting the process against false accusations, we can expect to get a lot more of them.

Everything works this way. Most of us have never robbed a bank, but banks still have security to protect against the small percentage who try. Most of us have never tried to enter a stranger’s home at night, but we all lock our doors to protect against the few who do. And most of us would never make a false rape accusation, but it’s still wise to guard against the few who would. Bad actors make things hard for everyone.

Even if Jackie fabricated her account, U-Va. should have taken her word for it during the period while they endeavored to prove or disprove the accusation.

That seems incoherent. If  U-Va is trying to prove or disprove her accusation, how can they take her word for it? Could she mean that U-Va staff should have pretended to take her word for it while investigating behind the scenes?

That actually makes sense, in a way, if you think in terms of different roles people assume in responding to a rape accusation. If a friend tells you she’s been raped, your should help her and support her, which pretty much requires you to believe her. Same thing if you’re a crisis counselor or psychologist: You’re there to help, and accusing the victim of lying or attempting to cross-examine her and check her story is not going to help. If she does turn out to be lying, all that’s happened is that you’ve wasted your time with someone looking for attention.

On the other hand, if your role in responding to the rape accusation is to punish the alleged rapist, the consequences of believing a lie are more severe: You could end up harming an innocent person. Wouldn’t it be best to approach the situation carefully and investigate with an open mind? After all, the falsely accused are victims too. They deserve some belief as well.

This post by Mark Draughn at Windypundit was originally published at Believing Victims

I want to emphasize again that my discussion is neither legal advice nor self defense instruction, because I’m not qualified to give either, so if you read any of this as a how-to guide, you’re making a serious mistake.

Memory

So far in my discussion of the ethics of lethal force, I’ve discussed the basic ethics of self defense and the importance of considering what the participants thought was going on. But unless we are involved in the incident ourselves, we have no direct knowledge of either of those things. We have to rely on what the participants and other witnesses tell us, and that can get tricky.

Human perception is unreliable. We may think we experience the world in continuous high-definition full-motion real-time 3D Technicolor with stereo sound, but decades of research into the physiological and neurological function of our senses tells us that the raw sensory inputs are filled with noise and gaps, and our brains pick bits here and there that seem important, and fill the gaps with plausible guesses (based on our genetics and past experience), presenting the result to our conscious minds so we can react. If we’re under a lot of stress, this process gets streamlined to make it even faster, sacrificing accuracy in favor of quick responses to the most likely threats. We can miss important observations, misidentify objects, confabulate details, and even get confused about the order of events.

This problem appeared in Part 1 as the problem of mistaken beliefs: If Alice is expecting Bob to draw a weapon, and Bob reaches into his pants pocket to withdraw a mobile phone, Alice’s brain may catch the first glimpse of something dark and shiny and her brain may take an urgent shortcut to “Gun!”

If she then shoots Bob, is it fair to say she was careless for shooting before identifying the weapon? Or was it an excusable mistake because she was up against the limits of her sensory system? Does it matter that she could have done more to train her brain to make better decisions? Does it matter if someone trained her incorrectly? Does it matter if she incorrectly believes that people of Bob’s race are more likely to pull a gun?

Whatever we believe about the legitimacy of Alice’s perceptions, the same issues arise again when considering her account of events because what she remembers depends on what she thinks she saw — garbage in, garbage out — so we have to take into consideration the errors in her perception.

Making things worse, not only are people fooled by their perception of reality, but their memory of their perceptions changes over time. Our brains process our memories the same way we process reality, making mistakes and filling in the gaps along the way. We also integrate information from other sources into our memories, believing things we heard from other people, even if they aren’t true: If Dave saw Bob pull out a mobile phone but Alice says she shot him because she saw him draw a gun, Dave may come to believe Alice’s version and remember seeing a gun too.

We try to create a consistent picture of the world, but sometimes that means we replace real memories with invented ones, instead of the other way around. Sometimes the more we think about an event we remember, the more we contaminate our memory with errors and lies.

Interlude: Standing Your Ground, Lying in Wait, and Looking For Trouble

George Zimmerman’s shooting of Trayvon Martin brought a lot of attention to the issue of standing your ground. This enters our analysis in the form of asking whether it’s a moral failure to use lethal force to end a life-threatening situation when the threat could have been averted by some other means. If Alice is sitting in her car, and Bob approaches on foot, wielding a knife and screaming “I’m going to kill you!”, can Alice simply drive away? And if she could, does that mean that shooting him instead is unnecessary and therefore wrong?

What if Alice is walking instead of driving, is she still obligated to try to run? What if she’s trapped in a dead-end alley and can’t run away? What if the alley has a gangway between buildings that she could escape through, but Alice doesn’t see it? How hard should she be required to look for an alternative before shooting?

The details can be confusing. What if Bob is wearing running shoes and Alice is wearing high heels? Is she still obligated to try to escape? If Bob does chase her, how close should she let him get before she stops running and starts shooting? Fifty feet? Ten feet? When he grabs her? Whenever she reasonably believes escape is impossible? Whatever distance her shooting instructor taught her?

What if Alice is in her own home? Is she required to flee her home if she can? If she’s trapped in a room, and the assailant is blocking the exit, is she obligated to try to get out the window? What if she’s on the second floor? What if she’s on the fourth floor? What if she can flee safely, but her child is still in the home somewhere? It’s hard to know where to draw the lines.

Another issue I thought was raised by the Zimmerman case (although it was rarely mentioned) is the problem of someone who’s looking for trouble.

Suppose Alice stops in a rough bar to have a drink, and when she pays, she flashes a large wad of cash in her purse. As she leaves the bar, Bob follows her out into the dark street, pulls a knife, and demands “Your money or your life!” Alice draws her gun and shoots him dead.

That sounds like a pretty clear case of shooting in self defense, but suppose we find out that Alice has been flashing cash in rough bars three times a week for the last six months? Suppose Alice was intentionally trying to draw someone like Bob into attacking her, thus creating a situation in which she would be justified in shooting and killing him?

On the one hand, Bob brought the problem on himself by committing a violent assault. On the other hand, Alice pretty clearly had a premeditated plan to kill someone, which is usually called murder. So while Bob is certainly not innocent, Alice is arguably not innocent either.

Another example might be if Bob breaks into Alice’s place of business after hours, intending to rob the safe, but it turns out Alice is still there, and she confronts Bob and kills him in self defense. If Alice was lying in wait in the store in hope of confronting and killing a burglar, she might not be as innocent as if she had been working late to balance the books.

Standing your ground, lying in wait, looking for trouble — these are murky issue where reasonable people will disagree. I think it ultimately comes down to a conflict of values between Alice’s right to protect herself and society’s preference for reducing violence. We want Alice to be able to shoot to defend her life, but we also want to discourage Alice from entering into situations, deliberately or not, where she is likely to have to do so.

Lies

Up until now, I’ve been discussing the natural difficulties that arise when trying to analyze the morality of a lethal force incident. But now it’s time to address an unnatural difficulty: In addition to the honest perceptual problems that can contaminate witness accounts, we also have to allow for the possibility that witnesses are deliberately lying to hide the truth. In our attempt to understand what happened, we may have opponents who want to fool us.

Until now, we were able to confine our analysis to the particulars of the incident — who was where, what they did, and why they did it — but when we have to consider the possibility that some of the witnesses are lying, it means we have to check their story and evaluate their credibility, and so our analysis becomes explosively complex, consuming everything in the world.

For every single witness, whether they were involved or not, we have to consider the possibility that their story is a lie. We need to check it for internal inconsistencies, and we need to check it against our mental model of how the world works.

If Alice says Bob broke in to her house by smashing the second story window into her bedroom, does that seem like something that really could have happened? Is climbing up to smash a second-story window a common way for attackers to enter a house? Is it a sensible way to enter this house? Or would some other point of entry have been simpler? Does it make sense that Bob could have climbed up there? If he’d need a ladder to make the climb, did police find a ladder somewhere nearby? Is there any way to rule out the possibility that Alice placed the ladder there after shooting Bob? Does the spray of smashed glass comport with Alice’s story of what happened?

We should also consider the motivations of the witnesses. People directly involved have the strongest motivation to lie because they could face punishment for crimes. Alice might shoot Bob in cold blood out of pure malice, and then try to keep herself out of jail (and civil court) by saying she had to defend herself against Bob’s attack. Alternatively, Bob might attack Alice and get shot by her in a legitimate act of self defense, only to try to stay out of jail by claiming from his hospital bed that she shot him without provocation.

We also have to consider how each witness’s story is supported or contradicted by the accounts of the other witnesses. If Alice says Bob broke in, but her neighbor Emily claims to have seen Alice let Bob in the front door, how do we decide whether to believe Alice or Emily? We could analyze each of their stories in detail and try to pick it apart. How did Emily happen to be looking at Alice’s house? Could she see it from where she says she was? How does she know it was Bob? How does she know it was Alice who let him in?

If Carl steps forward as a witness and claims Alice shot in self defense, it’s possible that he is lying for reasons of his own. Perhaps he backs Alice’s story out of a sense of loyalty, even though he saw nothing at all, or perhaps he saw her murder Bob, but he backs her story because he fears her retribution. Perhaps he just hated Bob and doesn’t want to see Alice punished for something he wanted to do himself.

Alternatively, maybe Dave steps forward and says he saw Alice murder Bob. It’s possible that Dave hates Alice and wants to make her life miserable. Or maybe he didn’t actually see anything, but he doesn’t believe Bob would have attacked Alice, so he feels justified in lying to ensure that Alice is punished for the unprovoked murder of Bob. Or maybe Bob was Dave’s best friend, and even though he knows Bob attacked Alice first, he’s still angry enough at Alice to want her to suffer. Or perhaps Bob and Dave both attacked Alice, and now that Alice has won the fight, Dave wishes to blame her for murder to deflect attention away from his own crimes and get vengeance for the death of his friend.

It’s not just the specifics of the incident that matter. It’s also quite reasonable to examine witness stories for peripheral details that turn out not to be true. If Fred says he saw the whole thing while driving home from work, but the location of the incident is not on his route home, then maybe he’s lying about seeing the incident as well. Or maybe the incident is on his route home, but we have reason to believe he was somewhere other than his job that day, so he wouldn’t have been on that route. If he’s lying about the little things, maybe it’s because he’s lying about the big things.

Looking even further afield in our attempt to evaluate the credibility of witness stories, it pays to investigate whether the alleged behavior of the participants in this incident is consistent with their past behavior. Has Bob ever attacked someone before? If Gloria reports being attacked by Bob, it tends to support Alice’s story. On the other hand, if Bob has no history of violence, we might ask whether it’s more likely that Bob did something uncharacteristically violent, or that Alice is making it up.

That in turn might lead us to look into Alice’s past, to see if she has a history of shooting people. We might also then investigate whether Alice has a history of telling lies. If Alice has previously made false accusations that Harold attacked her, it casts doubt on her story that Bob attacked her.

Of course, when we hear either of these stories about the past — Bob’s previous attack on Gloria, Alice’s previous lies about Harold — it opens up two more incidents to analyze. Did Bob really attack Gloria, or is Gloria lying? Did Alice really falsely accuse Harold of attacking her, or is Harold lying? If there are additional witnesses to those incidents, the investigation proceeds to analyze their credibility as well.

The pattern repeats as all kinds of things get pulled in. What kind of person is Bob? Was Bob coming home from church at the time of the incident? Or was he coming home from heavy drinking at a rough bar? Did he get in a fight at the bar? Did he start the fight? Is Bob prejudiced against people like Alice? Is Alice prejudiced against people like Bob? What are the prejudices of every single witness? Do Bob and Alice know each other? Were they coworkers? Did either of them hate the other? Was one of them stalking the other? Were they lovers? Former lovers? Who dumped who, and why?

All of the answers to all of these questions require evidence, and that evidence comes in the form of accounts from other witnesses, all of whom must also be evaluated for plausibility and truthfulness, because people may lie to help their friends and hurt their enemies, or for personal reasons. We have to keep asking, how do we know? And who says? And are they telling the truth?

So now it begins to matter that Alice’s story is backed by Ivan, but maybe Ivan is lying because Jeff thinks Ivan has a thing for Alice because he saw Ivan give her flowers, but Kate says she got flowers from Ivan too, so maybe Ivan just likes giving women flowers, and now Larry is telling us that Jeff is lying about Ivan’s attachment to Alice because he hates her, and, and, and…

We’ve gone from the simple case of Alice shooting Bob to the vagaries of whether Jeff is lying about why Ivan is giving flowers to women. It all seems like a distraction, but there’s also a point about the credibility of a key witness to a shooting, so we can’t ignore it, but we shouldn’t make too much of it either, and so maybe we should get a little more information…

Multiply that by ten, or a hundred, and you can see why discussions of these incidents go so far afield. This is why it may be important that Treyvon Martin had Skittles, or that Michael Brown stole some cigars. This is why it may be important that Brown was going to college, or that officer Wilson had worked for a troubled police department before transferring to Ferguson. These facts aren’t part of the shooting incident, but they may matter because they’re part of how we judge the truthfulness of witnesses.

That’s it for now. In the third and (I think) final chapter, I’ll talk about how we hear about all this, and how we respond to it.

Update: Part 3 is up.

This post by Mark Draughn at Windypundit was originally published at Thinking About Lethal Force – Part 2

I don’t find the fact that a prosecutor could get an indictment probative of whether he should get one, not at all. The special prosecutor in Florida, for example, indicted Zimmerman with nothing that would support a conviction. That’s not serving justice.

Funny you should mention Zimmerman, because prosecutor Angela Corey only charged Zimmerman with a crime after caving in to public pressure to do so. Initially, prosecutors chose not to charge Zimmerman at all. If Bob McCulloch thought Darren Wilson’s shooting of Mike Brown was justified, there was no need to present anything to the grand jury.

If there were conflicting witnesses, it would be lousy practice to merely put on Johnson, about as biased a witness as one could find.

After Darren Wilson. But point taken.

The fact that there were six shots fired isn’t indicative of anything. Police are trained to fire until a subject is down.

I agree with Jack’s point about police tactics. Some time ago, police departments noticed there had been a number of incidents in which an officer fired a shot at an attacker which either missed completely or didn’t stop him quickly enough to prevent him from injuring or killing the officer. So modern police training tends to emphasize shooting until the attacker goes down.

This means that the number of shots fired is not necessarily indicative of maliciousness, anger, or race hatred. It’s just training. However, the number of shots is indicative of intent. One shot could be an accident. Six shots are on purpose.

Or to look at it another way, instead of a white cop shooting an unarmed black teen, imagine that it was just two white dudes from Missouri, Vern and Earl, who ran into each other on the streets one day. Somehow Vern ended up drawing his gun and shooting Earl to death, and it turns out Earl was unarmed, and some of the witnesses describe Vern shooting him in cold blood. In that case, don’t you think the prosecutor would make sure the grand jury heard that Vern shot Earl six times?

Why would a prosecutor want to indict a police officer if the evidence showed that he was in legitimate fear of his life and fired in self defense?

If that’s what the prosecutor thought the evidence showed, why would he present evidence to the grand jury at all?

Why would you assume that if the prosecutor is presenting the evidence fairly, he doesn’t want an indictment?

Because that’s not what prosecutors do when they want an indictment. The grand jury process isn’t “fair” in the same sense that a trial is, and it’s not intended to be. There’s no presumption of innocence, no way for the defense to contest the admission of evidence, no way to impeach a witness. It’s not even an adversarial process.

I will presume, until I see evidence to the contrary, that if the officer isn’t indicted, then there was not sufficient evidence to bring a legitimate indictment. Sure, prosecutors often get grand juries to indict on insufficient evidence. Are you really advocating that? Why? Because the protesters and civil rights groups want it to be a racist cop shooting an unarmed kid, rather than a scared cop shooting a charging thug with more than a 100 pounds on him who already tried to get his gun?

Jack is right that the process Darren Wilson went through is not inherently unjust. And going back to my hypothetical shooting in which Vern killed Earl, I think it’s entirely possible that citizen Vern would receive the same three months of investigation, and the same weighing of evidence, that Officer Wilson did.

The difference is that Vern would be indicted by the grand jury the first day (or just charged directly), and the rest of the process would take place while Vern sat in jail, unless he was able to make whatever really high bail the judge gave him. Also, some of the investigation would probably be done by the defense rather than the prosecutor, and if the prosecutor wasn’t convinced, Vern would have to go to trial or take a plea to a lesser crime.

To illustrate why Darren Wilson’s wonderful treatment by the prosecutor makes people angry, let’s change the hypothetical a bit: Let’s suppose it turns out that Vern is actually a second cousin to former President Bill Clinton. Then suppose the media begins finding witnesses who say that Earl had his hands up when Vern shot him. Now imagine that without interviewing some of those witnesses, and before the forensics are in, the chief of police makes statements of support for the Clinton relative and says he won’t be charged.

However, Earl’s family and friends drum up public criticism, so the county prosecutor — who has family members that work for the Clintons — announces that he will present the case to the grand jury. Unlike most presentations to a grand jury, in which a few witnesses are presented over a couple of days, this case drags on three months and includes 60 witnesses. Unlike most presentations to a grand jury, Vern testifies and tells his side of the story. Unlike most presentations to a grand jury, the prosecutor does not explicitly ask the grand jury for a particular criminal charge. And in the end, unlike most presentations to a grand jury, there’s no true bill, and Clinton’s cousin walks away a free man.

And so my question to Jack: Given your low opinion of Bill Clinton’s ethics, if you found out that a relative of his had been no-billed by a grand jury proceeding unlike any other, run by a prosecutor with close ties to the Clinton family, wouldn’t you be damned suspicious? Wouldn’t that set off your ethics alarms?

Now if you replace Vern with Darren Wilson, Earl with Mike Brown, and “Clinton family” with “police,” that suspicious scenario is pretty close to how the Ferguson grand jury decision feels to a lot of us: A killer with ties to a politically influential group benefited greatly from a one-of-a-kind grand jury proceeding.

Ultimately, the decision may in fact be the right one, but that kind of privilege and special treatment is astoundingly unfair.

(Caveat: I wrote my original post and parts of this one before the grand jury decision was released to the public. I haven’t read the full document release, nor do I plan to — I’ll wait for analysis from some of the web’s smart legal minds. It’s certainly possible that some of the issues I previously viewed as suspicious will turn out to have reasonable explanations once I catch up, but I maintain that I was right to be suspicious at the time.)

This post by Mark Draughn at Windypundit was originally published at The Special Case of Darren Wilson

What’s breaking my heart is the magnitude of the destruction going on during the protests. I paid pretty close attention to the earlier batch of protests, right after Brown was shot, and there was sporadic property damage, some of it severe, but for the most part, the protesters were peaceful. It was neighborhood people out late, milling around, making noise, and occasionally blocking traffic. As a middle-aged white guy, I didn’t see much that would make me feel unsafe to be there.

None of that stopped the racist fuckwads who continued to paint the largely peaceful protesters as looters and and wild animals. And tonight, unfortunately, some of the protesters are doing everything they can to give the racists an “I told you so.”

This time around, the looters are out in force, and a lot of stuff seems to be burning. Just a brief Twitter scan finds mention of a Metro PCS phone store, a Beauty Town beauty supply store, and the outreach office of Alderman Antonio French. Also a New York Grill, a Little Ceasar’s, a Walgreens, a car dealership, an Enterprise car rental, several auto parts stores, a public storage facility, and a church van.

All burning away.

I am reminded of an old joke:

Late at night, a drunk was on his knees beneath a street-light, evidently looking for something. A passer-by, being a good Samaritan, offered to help. “What is it you have lost?” he asked.

”My watch,” replied the drunk. “It fell off when I tripped over the pavement.”

The passer-by joined in the search but after a quarter of an hour, there was still no sign of the watch.

“Where exactly did you trip?” asked the passer-by.

“About half a block up the street,” replied the drunk.

“Then why are you looking for your watch here if you lost it half a block up the street?”

The drunk said: “Because the light’s a lot better here. ”

What we’re seeing in Ferguson is the because-the-light’s-better-here version of sticking-it-to-the-man. The jerks setting fires are angry at the cops and the political institutions they protect, but they know that attacking the establishment is hard. Assaulting the cops and burning down the police station would be one hell of a statement, but the cops are armed and prepared for a fight. Attacking them would require skill and audacity and a willingness to risk getting arrested.

On the other hand, burning down private businesses is much easier. The vandals can tell themselves they’re fighting capitalist oppression or hitting the establishment in pocketbook or whatever, but the truth is that small shops and fast food joints are much softer targets. The light’s a lot better there.

The political leadership of Ferguson, on the other hand, remind me of a (possibly untrue) story I once heard about Iraqi leader Saddam Hussein from the first Gulf War. His forces had invaded Kuwait, and a few months later a coalition lead by the United States counterattacked into eastern Iraq. They moved with breathtaking speed and routed the Iraqi army. Saddam watched the defeat unfold from one of his bunkers, as the coalition forces advanced to a point where they had a clear path to capture Baghdad and Saddam himself.

Before that could happen, however, with Kuwait being freed from Iraqi control, the U.S. declared a unilateral cease-fire. Saddam’s reaction was supposedly a jubilant “We’ve won!” Although twenty thousand Iraqi soldiers had died in the battle, Saddam considered it a victory because he was still standing, and that’s what mattered most.

The police presence available to keep the peace in Ferguson consisted of hundreds of officers, perhaps as many as a thousand according to some sources. The Governor of Missouri declared a state of emergency and dispatched National Guard troops. And yet looters controlled the streets, and a dozen businesses burned out of control for hours. But all the cops are safe and the police station is intact, so I’m guessing they feel they’ve won. Because to the powers-that-be in Ferguson, their safety is what mattered most.

Ah, fuckit. I’m running out of steam here. Maybe it all won’t seem so bad when I wake up in the morning.

Update: Reading back over this, I should point out that even if every fire was started by a different person, which seems unlikely, that’s only a dozen or so arsonists among all those protesters.

I should also add that I’m not saying cops should have taken insane risks with their lives to stop the looting and support a fire suppression effort. On the other hand, police have been comparing themselves to soldiers and saying police work is a battlefield for years, and in response we’ve been equipping them with shields and body armor and armored vehicles. Well, last night the battle came to Ferguson. Did they use all that fine equipment the way we wanted them to?

Finally, as to the grand jury’s decision, a cop shooting an unarmed black man sounds damned suspicious, but these sorts of incidents are always very fact sensitive, and based on the bits and pieces I’ve heard, I can imagine ways in which the shooting of Mike Brown might have been justified. And that’s true for most of these cop-shoots-unarmed-person shootings: Any one of them might be justified. But they can’t all be justified. There are some cops getting away with some shit.

This post by Mark Draughn at Windypundit was originally published at The Drunk Under the Street Lamp Meets Saddam Hussein

On the other hand, if the grand jury decides to indict Wilson, there would be a warrant for his arrest, and that would mean there’s officially an accused murderer on the loose. And we know he’s got access to weapons, training, and an undisputed willingness to kill, right? So how come we haven’t heard anything about the plans to send SWAT teams to capture him?

Obviously, no one expects Darren Wilson to be indicted. The other clue is that it’s taking so long for the grand jury to announce a decision. If anyone in the prosecutor’s office really thought Wilson was a murderer, do you think they would have allowed him to remain on the streets for three months?

As Jeff Gamso points out, this is a slam-dunk indictment if the prosecutor wants it to be. If you or I had shot Michael Brown, the grand jury proceedings would have gone something like this:

• Present witness Dorian Johnson, who was present at the shooting, and have him repeat his statements to the media describing how Darren Wilson shot an unarmed Michael Brown in cold blood.
• Present the medical finding that Brown was shot six times and that he died from the shots.

I’m no lawyer, but I’m pretty sure that’s enough for a murder indictment. There’s no need for cross examination or hearing from the other side. Grand juries proceedings are not a trial, they are the threshold over which an accusation must pass to get to a trial, and it’s a very low threshold. The grand jury just has to think it deserves a trial.

Of course, for them to do that, the prosecutor would have to want them to.

This post by Mark Draughn at Windypundit was originally published at Awaiting the Grand Jury…

It seemed like a good subject for one of my lengthy thinking-out-loud posts, and indeed I was able to pile up a great many words, some of which may even be worth reading.

When I started writing, however, I realized that in order to write about how we discuss lethal force, I was going to have to write about when it’s okay to use lethal force. That runs into the hazard that anything I write about when it’s okay to use lethal force is going to sound an awful lot like legal advice.

So before I get into this, I want you to understand one thing: This is so not legal advice. I’m not a lawyer, and if you ask me when it’s okay for you to shoot someone, I’ll tell you it is never, ever okay for you to shoot someone, because if you’re looking to me for advice on when to shoot someone, you have no business contemplating shooting people.

That said, I’m going to try to explain some basic ideas about lethal force. These ideas are roughly based on the sort of thing you might be taught in a course on self-defense with a firearm, which in turn is based on self defense law. But my discussion is neither legal advice not self-defense instruction (I’m not qualified for either). Rather , this is more like a primer on how to think about stories in the news that involve lethal force.

I’m going to begin with a discussion of the basics of self-defense and then move out from there into increasingly confusing realms of understanding. I fervently hope that nothing I write makes you stupider for having read it, but as always, there are no guarantees.

Self Defense

I guess I should start by asking, when is it okay to kill someone? In a sense, the correct answer is never. Aside from a few special (and very controversial) situations — war, capital punishment, assisted suicide — it is never okay for Alice to intentionally kill Bob.

However, if Bob is attempting to kill Alice — or any other innocent person — it is okay for Alice to try to stop him. She has the right to self defense, and although self defense is generally supposed to be proportional, the fact that he is trying to kill her means that she can be pretty extreme in her attempt to stop him, up to and including doing things that might kill him.

But her goal can never be just to kill him. Shooting and possibly killing Bob can only be justified when it’s necessary to save a life. The way this is commonly put is that Alice can’t shoot to kill, but she can shoot to stop a killing. Sometimes shooting to kill is the only way to stop an attacker, but that’s the only way it can be justified.

(There may or may not be other actions Alice can ethically shoot to stop, such as severe injury, rape, or kidnapping. We can make our own moral judgements about what’s worth killing for, but Alice would also be judged under the laws where she lives. On the other hand, a pacifist may feel it is never okay for Alice to kill Bob, even if Bob is trying to kill her.)

One important point alluded to above is that in order to defend yourself you have to be innocent. If Bob breaks into Alice’s house at night, Alice may be justified in using lethal force to stop him. But even though her shots endanger Bob’s life, he does not then have the right to shoot at her in self-defense. He’s not innocent, he’s a home invader.

If Bob quickly flees the house to get away from Alice, then he’s no longer a threat, so Alice no longer has the right of self-defense and cannot shoot him as he runs away. If Alice decides to follow Bob out of the house and shoot at his back as he flees the scene, then she is the aggressor, and Bob then he arguably acquires the right to defend himself against her, including the use of lethal force. If he shoots back and wounds her and she drops the gun, then Bob loses the right to self-defense and has to stop shooting. If instead he continues shooting, perhaps to finish her off, then Alice is once again the innocent victim, and she arguably has the right shoot Bob to defend her life.

In a complex confrontation between Alice and Bob, the right of self-defense could switch back and forth multiple times. The reasons that happens — Bob fleeing, Alice dropping the gun, and so on — are exactly the kind of details that often gets left out of early news reports, either because the details aren’t known to the reporters, or because the reporters or their editors don’t recognize the significance of the details.

Interlude: Tragedy

Note that it’s possible to believe both that (a) Alice shot Bob in a righteous act of self defense, and that (b) Bob’s death is a tragedy.

For one thing, Bob did not live in this world alone. There are probably people who will miss him and mourn his loss. Bob was somebody’s son, and he may have been somebody’s brother, husband, father, or friend. Even if Bob turns out to have been a violent asshole intent on raping and murdering Alice, it’s not going to change the way that people feel about the Bob they knew. Even if Bob was killed by police while shooting children at a school, Bob’s parents are still going to cry that he’s gone. No matter how much Bob objectively deserved his fate, nobody who knew him is going to instantly incorporate that fact in their emotional response. The heart doesn’t change direction that fast.

But even if no one misses Bob, many people would argue that his death is still a tragedy. Killing Bob may have been necessary to prevent the loss of innocent life, but that doesn’t mean that Bob’s death isn’t a loss as well. Bob’s death destroys both the bad and the good in him, and we can regret the loss of the good parts, even if Bob brought it on himself. No one is as bad as the worst thing they’ve ever done.

That’s not to say that Alice was wrong to kill Bob in self defense. Alice may have killed him to save her life, and we can approve of her doing so, but it’s no criticism of Alice’s actions to say that it would have been even better if Alice had figured out a way to save her life without killing Bob.

There are of course limits to that sentiment — everyone likes to see the bad guys get what’s coming to them — and there’s a lot of subjectivity in our feelings toward criminals. It’s easier to feel the loss of a teenage gangster killed in a gunfight with police if you grew up with people just like him, or if you could have been him in another life. On the other hand, there’s no reason to expect any sympathy from the family of his victims…although that has been known to happen.

Perception and Intent

Self defense is just the first layer of the ethical puzzle, describing the conditions under which Alice is allowed to shoot Bob. There’s a difference, however, between the abstract morality of certain outcomes and the ethics of human action, because people are limited by their perceptions and understanding.

If Alice or Bob honestly misunderstands the facts of a situation in such a way that they would not be doing anything wrong if their understanding was correct, we generally do not consider them to have committed a moral error. We judge people’s ethics by their decisions based on their subjective knowledge at the time of the event, not on our own knowledge in hindsight.

This usually enters the lethal force analysis with regard to the question of whether or not someone’s life was actually endangered. When we say Alice can shoot Bob to stop him from killing her, what we really mean is that Alice can shoot Bob if she reasonably believes it is necessary to stop him from killing her. The usual way to express this is that Alice must be in fear for her life.

Perhaps Bob plays a mean joke on Alice by pointing a fake gun at her and screaming “Die! Die! Die!” just to see her frightened reaction. Since Alice isn’t in on the joke, she might reasonably come to the conclusion that Bob intends to kill her, and so she might believe it is necessary to use lethal force to stop Bob’s “attack” and end up shooting him dead.

Had Bob actually been attempting to kill Alice, we might say this was a righteous shooting in self defense. It would have been the right thing to do, and if in the future Carl attacked Alice the same way, we would want her to shoot him as well. However, since Bob wasn’t actually trying to kill Alice, her shooting him was not the right thing to do, and if Carl performs a similar fake attack on Alice, we ideally want her not to shoot him.

Nevertheless, this does not mean that Alice is a bad person who deserves punishment. We can understand why Alice shot Bob, and although with full knowledge of the situation we realize it was the wrong thing to do, we can also realize that with Alice’s limited knowledge, it probably seemed like the right thing to do. On the other hand, if Alice ignores information — limits her own knowledge, either recklessly or intentionally to avoid the duty to take it into account — we may hold that against her.

Note also that it’s not enough for Alice to be afraid. There has to be some connection to reality. Alice’s fear need not be accurate, but it must be reasonable. If, for example, Alice has an irrational fear of young black males and Bob is a young black male, that’s not a license for Alice to shoot Bob.

In general, the subjective nature of Alice’s perception makes the ethics harder to analyze. If Bob breaks into Alice’s home carrying a gun and charges at her while screaming “Die! Die! Die!”, we’d probably all agree that’s a justified shooting. If Bob was actually playing a prank and didn’t even have bullets in the gun, I think we’d all agree that Alice is not the one at fault.

But what if Bob is just a burglar who intends Alice no harm? What if he breaks into Alice’s house in the night, but he brings no gun and makes no threats, and Alice still shoots him? Shooting Bob is not necessary to save Alice’s life, but does it look that way to Alice? Is she required to see a weapon or evidence of intent to harm before she can shoot him? Or do we say that when Bob breaks into Alice’s house in the middle of the night that she can reasonably presume that he intends to harm her and has the means to do so?

The analysis based on reasonable belief also brings into play factors that might influence Alice’s reasoning. Alice might have had previous encounters with Bob that lead her to believe that he’s dangerous, or she might have previously been attacked by strangers in the same location where she encounters Bob, or police might have been warning area residents of a rapist who breaks into homes in the middle of the night, or Alice might have had self-defense training in which she was told that most intruders who break into occupied homes intend serious harm to the occupants.

It’s even possible for situations to occur in which both sides reasonably believes that they’re in a self-defense situation. Imagine that Alice comes home to find the front door ajar. Furthermore, she can hear someone moving around inside. Concerned, she draws her gun and enters. However, what Alice doesn’t realize is that because she lives in a real estate development where all the houses look the same, she has mistakenly entered Bob’s house. Upon seeing hearing Alice enter the house, Bob becomes concerned and draws his own gun to go investigate. When Alice and Bob encounter each other, each reasonably believes they’ve encountered an armed intruder, and both open fire.

Although Alice is clearly the one who made the mistake, and even though the consequences were profound and tragic, it’s not clear that Alice’s error is the moral equivalent of murder. She certainly wasn’t intending to kill an innocent person. With a little more inventiveness, perhaps involving a malicious third party, we could probably come up with scenarios where it’s not at all obvious that either Bob or Alice is to blame.

This is one of the most contentious areas in the ethics of self defense. On the one hand, we don’t want Alice shooting at anyone who makes her uneasy; on the other hand, we don’t want her to wait until she sees the muzzle flash. There’s a lot of distance between those extremes, and sincere and diligent people can nevertheless have very different ideas of what should reasonably cause someone to fear for their life, or of what mistakes are understandable and excusable.

Making things more complicated, we need to be careful not to confuse our analysis of perceptions and our analysis of reality. It’s not inconsistent to believe that (a) Alice was justified in shooting Bob because she was in fear for her life, and (b) Bob was not actually endangering Alice’s life.

To make that more concrete: Thinking that George Zimmerman and Officer Darren Wilson don’t belong in jail is not the same as thinking that Trayvon Martin and Michael Brown deserved to die, or that it’s open season on young black males. It’s true that there are people who think that Martin and/or Brown were thugs who deserved to die, or that young black males are a threat to public safety, but it would be a mistake to assume that everyone who defends Zimmerman and Wilson is in either of those camps.

By all means, if your opponents are actually violent racists, you should call them out on that. But there’s a difference between wanting to see young black kids killed for no reason and believing Zimmerman and Wilson did the right thing. And there’s a difference between believing Zimmerman and Wilson did the right thing and believing that Zimmerman and Wilson did the wrong thing for understandable reasons. Intellectual honesty demands that we recognize the distinctions and take them seriously, and that when evaluating other people’s opinions, we recognize that they may be making different distinctions than we would.

That’s enough for now. In future posts I’ll explore how we learn about lethal force incidents, and why that adds to the confusion.

Update: Part 2 is up.

This post by Mark Draughn at Windypundit was originally published at Thinking About Lethal Force – Part 1

Consider what life would look like without plea bargaining.  Every defendant would then have to be tried to be convicted, as there would be no incentive to cop out.  Every guilty defendant would then be subject to full freight sentencing, the price that no one in power really expects anyone to pay, which is what allows them to set the MSRP for crime in the stratosphere, tell the groundling during stump speeches about how they’ve saved them from rapists and give headline writers silly numbers for clickbait.

I have no disagreement with this. Criminal defendants would get hammered without the plea bargaining escape hatch. You would not want to be prosecuted for a crime after plea bargaining was eliminated. Nevertheless, Scott misses a point that may be worth discussing (or maybe not, but that never stops me) when he writes this:

The government would need 100 times the existing number of courtrooms, judges, clerks, court officers, prosecutors and jury pools. And public defenders would get stretched a wee bit thinner.  Even so, kids would sit in jail for years awaiting trial, turning that one year potential sentencing into three, until it was their turn in the well.

The government is not going it get what it wants. Increasing the case volume of the justice system 100-fold — or 20-fold if you use the 95% plea figure — is never going to happen. The government can’t afford the cost of holding trials for everyone they charge, and it sure as hell can’t afford the cost of operating prisons for all those people. (Instead of having 1% of our adult population in prison, we’d probably have to hire 1% of our population as guards for our vast prison empire.) If we somehow outlawed plea bargaining tomorrow, there’s no way the government at any level could afford to try and imprison everyone that gets arrested.

Advocates of eliminating plea bargains are counting on this. Without a larger budget, prosecutors would have to make some hard choices about who to prosecute and who to let walk. They’d have to focus their efforts on the people they most want to put in prison and let everyone else go. It seems likely to me that the net effect would be a reduction in aggregate prison time.

On the other hand, it would be devastating for Scott’s clients. The ones found guilty would be pounded into the ground by draconian sentences, much higher than the legislature intended, and these long sentences would be concentrated on a small group of people, which is arguably unfair. (Speculation about this group’s age, race, and gender characteristics are left as an exercise for the reader.)

The other big problem with eliminating plea bargaining is that lots of crimes would go unpunished. While it seems unlikely that every single one of the 95% who currently take plea bargains are guilty, it is probably even less likely that every single one of them is innocent. Many of them would go on to commit more crimes against the public. Eventually, they would be prosecuted and imprisoned, but in the meantime they could wreak havoc on the innocent. It seems likely that the result would be a lot more crime.

Finally, as a libertarian, I’d like to think that if the justice system could only prosecute a small fraction of the crimes it does today, it would abandon enforcement against consensual crimes and focus on crimes that present a real danger to the public. Somehow I doubt that would be the case.

The plea bargaining system could use some reform, but eliminating plea bargains altogether is an extreme measure that would imprison a lot of people for a very long time while allowing crime to run rampant. That doesn’t sound like it would be good for anybody.

This post by Mark Draughn at Windypundit was originally published at What If We Eliminated Plea Bargaining?

Slenderman is an entirely fictional creation of recent vintage:

He is Slenderman, a menacing, faceless specter in a dark suit — sometimes portrayed with octopus-like tentacles — known to haunt children and those who seek to expose him. He was born in 2009 in an online forum for people who enjoy creating fake supernatural images.

[…]

To be clear, the origin story of the monstrous character (sometimes referred to as The Slender Man) in no way urged readers to kill to earn his favor. But Slenderman has undergone hundreds of permutations online in his five-year existence.

If you want to learn more about Slenderman, you can Google him up as easily as I can, but the details aren’t really very important. Although someone studying the psychology of the would-be killers might find it interesting to explore their beliefs about Slenderman in some detail, I doubt that the specifics of the legend are an important part of the cause of this violence.

For one thing, these kinds of killers — serial killers, spree killers, thrill killers — are often a bit confused about the world. Something may have inspired them, but they often seem to get more out of their inspiration than what’s actually there. Back in the late 1980’s a pair of young men named Raymond Belknap and James Vance shot themselves in some sort of suicide pact. Belknap died immediately, but Vance lived for three years afterward, and he claimed they were inspired to suicide by lyrics in some Judas Priest songs. The thing is, he remembered stuff from the lyrics that just wasn’t there. In his book The Gift of Fear, author Gavin de Becker recounts some confusion by Vance when it came to the lyrics:

The group Judas Priest did not create James Vance, of course, but in a sense, he created them. When he was asked about a particular lyric, “They bathed him and clothed him and fed him by hand,” he recited it as “They bathed him and clothed him and fed him a hand.” So he had done more than just react to the songs; he had actually rewritten them, taken a lyric about someone being cared for and turned it into something about cannibalism. Even his admiration was expressed in violent terms. James said he was so enamored of the band that he would do anything for them, “kill many people or shoot the president through the head.” He told lawyers that if the band had said, “Let’s see who can kill the most people,” he would have gone out and done something terrible. In fact, the band said no such thing, and he did something terrible anyway.

Another example is John Hinckley Jr., a disturbed young man who became obsessed with actress Jodie Foster after seeing her in the movie Taxi Driver. After stalking her for a while at Yale University, where he was unable to make any meaningful contact, he eventually decided that his best chance to impress her was to make some spectacular gesture, and on March 30, 1981, he opened fire on President Ronald Reagan and his entourage outside the Hilton Hotel in Washington, D.C., wounding Reagan and three other people. Needless to say, Hinckley’s stated intentions were not grounded in reality, and his assassination attempt did not actually impress Jodie Foster, nor did anyone blame her for his actions. He was just crazy.

A few years before that, serial killer David Berkowitz shot 13 people in New York over a period of several months. When he was captured, he claimed to have killed his women victims because his neighbor’s dog was a demon that had demanded a blood sacrifice. As with Hinckley, the craziness of his motivation was immediately apparent, and nobody actually blamed the dog. Further, some criminal psychologists believed Berkowitz made up the whole “dog made me do it” story to lay the groundwork for an insanity defense, or maybe just to mess with investigators. And depending on how you interpret some of his statements, Berkowitz may also have tried to convince investigators that he was part of a team of murderers, some of whom were still at large. There’s no evidence for that, though.

Ted Kaczynski, the “Unabomber,” carried out a 17-year campaign of mail bombings that claimed 26 victims, three of them fatally. Based on a 35,000-word “manifesto” he sent to newspapers, he appears to have been attacking people involved in technology our of some concern for the dehumanizing effects of industrial society. Like Berkowitz, Kaczynski also seems to have wanted to convince people he had confederates: The manifesto was written in the first-person plural and refers to something called “Freedom Club,” but there is no evidence of involvement by others.

Confessions should generally be taken with a grain of salt. In 1984, a white man named Bernhard Goetz shot four young black males on a commuter train. When captured, he claimed it was self-defense (which held up in court), but he also said that after shooting each man once, he walked over to one of them, told him “You don’t look so bad, here’s another,” and shot him again. At the trial, however, it turned out that each man had been shot only once, and most witnesses said that all the shots were in rapid succession. No witnesses heard him say the “here’s another” line. He had apparently invented a story that made him seem like more of a badass.

Confusion abounds. Some killers claim to be getting back at bullies, but when investigators interview their acquaintances, it turns out they were bullies themselves. Some killers try to manufacture higher purposes for their crimes, such as leading revolutions against real or imagined oppression. Others spin their crimes to make themselves seem less pathetic and more cool and in control. Still other killers try to evade punishment by denying all involvement with the crime or blaming someone else, including the victim. On the other hand, something like 20% of all post-conviction exonerations include false confessions to the crime.

So while it’s important to consider a killer’s statements about his crimes, it’s also important to keep in mind that what he says may be delusional, incoherent, self-aggrandizing, manipulative, or an outright lie. This is obvious when the killer’s explanation is clearly nonsense — involving Slenderman or talking dogs — but just because the killer’s explanation is banal and ordinary doesn’t mean it’s accurate. And just because his explanation is crazy doesn’t mean he’s crazy in the most obvious way.

This post by Mark Draughn at Windypundit was originally published at You Can’t Trust a Killer

Still, a lot of people are talking about it anyway — it’s hard not to, when it’s something so shocking — and I’d like to talk a little bit about what some of them are saying.

The first thing to note is that this is a confusing situation. A lot of people are emotionally upset by what happened, yet the story isn’t really clear yet. That’s a combination that is guaranteed to produce angry misunderstandings. People might say strange things just because they’re working from a different version of the facts than we are, which isn’t their fault (especially if it turns out their facts are correct and ours are wrong). They might also be repeating something they heard without thinking it through or trying to verify the truth of it. So for the first couple of days, if somebody says something that really pisses us off, we might want to cut them some slack and hold off on our angry responses.

And even when we don’t want to cut them any slack, we may still want to hold off on angry responses anyway. Some people are just trolls, saying outrageous stuff just to be outrageous and get a reaction. It’s best not to feed the trolls. (Although in some cases, you will probably want to keep an eye on them in the future. Geez.)

With that in mind, I’d like to talk about some of the discussion of the “cause” of Elliot Rodger’s killing spree. Most of the emerging discussion of his motive centers around a video he posted and a “manifesto” he sent out. I haven’t seen either of them, and I don’t want to, but apparently they’re quite misogynistic, and he literally threatens to kill a bunch of women. This has naturally lead a number of people to make pronouncements about the how Rodger’s killings reflect the misogyny of our society.

One immediate complication to that view is that, as far as I can tell, Elliot Rodger started by killing his male roommates, and he ended up killing more men than women. (He also wounded a lot of people, and the news reports don’t talk much about them, so he may have shot more women than men.) I’m not saying that he wasn’t a misogynist, but it’s probably not a sufficient explanation for his behavior. I’m assuming he has some type of mental disorder.

For that reason alone, it would probably be a mistake to take much of what Rodger says at face value. Furthermore, murderers are often not really reliable sources of information about their motives. Almost none of them say, “I killed because I’m a violent asshole.” They cast blame elsewhere — the victim usually, but also society in general, racial minorities, women, video games, porn, music lyrics…anything other than themselves. Sometimes this is a clear case of a manipulative personality looking for a way out, and sometimes it’s just something they’ve come to believe. They feel lonely, cut off, and rejected by the world, and they blame the world for those feelings, and they seek their revenge.

I’ve heard that Elliot Rodger was associated with the Pick-Up Artist (PUA) community, and some people have tried to use his killing spree as proof of the misogyny in the PUA community. This is spurious reasoning. Spree killers, whatever their motives, are very rare. However they fit into society, whatever communities they are associated with, they are far out at the end of the curve, and they tell us very little about the bulk of the people clustered in the center.

If you want to make judgements about the values of a community, you should base it on the values displayed by the bulk of its members, not by the crazy people on the fringe. Elliot Rodger is not proof of the misogyny in the PUA community. The PUA community is proof of the misogyny in the PUA community.

I’ve also heard that Rodger was a Men’s Rights Activist (MRA), and some people are arguing that his killing spree is proof of the misogyny in that community as well. Again, you can’t judge a group’s values by the behavior of its most extreme members. Judge the misogyny of the MRA community by the misogyny of the MRA community.

There are those who would argue that by portraying Rodger as a lone madman rather than a representative of our misogynistic society, I am minimizing the problem of violence against women. I would argue that the people who are holding Elliot Rodger out as an example of societal misogyny are distracted from the larger social problem.

Spree killers are a small part of the problem of violence against women. The larger problem is less spectacular and more mundane, so it gets less news coverage. Usable statistics are surprisingly hard to find via Google, but according to a study by the Violence Policy Center (based on the FBI UCR Homicide Supplement), in 2009 there were 989 women killed by their husbands or boyfriends. An additional 590 women were killed by other men that they knew. That is far more than are killed by spree killers in any year. Using those numbers as a statistical average, it probably means that on the day Elliot Rodger killed two women in his spree, another 2 or 3 women were killed by their husbands or boyfriends, and an additional 1 or 2 more were killed by some man in their life. And unlike Rodger’s shooting spree, that toll keeps repeating itself day after day after day. It may be a better place to focus our efforts.

(And if you think domestic violence gets little coverage or study, it gets even worse for violence against sex workers, but that’s another matter.)

At the end of the day, I’m pretty confident that Elliot Rodger was some kind of madman. He may have seen himself in a political context, as striking a blow for unloved men everywhere against women and other men, but that doesn’t mean he actually was, or that other men in similar situations are going to follow his example.

And for the record, whatever mental disorder he was suffering from, he’s out on the fringe there as well. People with mental health problems are generally no more violent than everyone else.

This post by Mark Draughn at Windypundit was originally published at Elliot Rodger’s Motive Says Very Little About Anyone Except Elliot Rodger

I’ve heard mixed reports about the effectiveness of the death penalty. Some studies find a deterrent effect, and others do not. I’ve heard that in those studies that have found an effect, much of it goes away if you drop Texas from the data set. On the one hand, that doesn’t prove the study is wrong — of course the results of a study will change if you cherry pick the data — but it’s interesting that it all depends on the state with the most executions. Perhaps the deterrent effect doesn’t show up unless you execute a lot of criminals.

More generally, I believe studies have shown that deterrence effect is not as sensitive to the severity of the punishment as it is to the immediacy and certainty of the punishment. This would be consistent with the idea that the criminal personality includes a high tolerance for risk and that criminals discount the future heavily. So if we want to fight crime, it’s more important to make the punishment swift and sure than it is to make it harsh.

This post by Mark Draughn at Windypundit was originally published at Crime and Incentives

And perhaps because of these changes, an execution in Oklahoma went disturbingly wrong:

McALESTER, Okla. — What was supposed to be the first of two executions here on Tuesday night was halted when the prisoner, Clayton D. Lockett, began to writhe and gasp after he had already been declared unconscious and called out “oh, man,” according to witnesses.

…

A medical technican inserted the IV needle and then the the first drug, a sedative intended to knock the man out and forestall pain, was administered at 6:23 p.m. Ten minutes later, the doctor announced that Mr. Lockett was unconscious, and the team started to administer the next two drugs, a paralytic and one intended to make the heart stop.

At that point, witnesses said, things began to go awry. Mr. Lockett’s body twitched, his foot shook and he mumbled, witnesses said.

At 6:37 p.m., he tried to rise and exhaled loudly. At that point, prison officials pulled a curtain in front of the witnesses and the doctor discovered a “vein failure,” Mr. Patton said.

Without effective sedation, the second two drugs are known to cause agonizing suffocation and pain.

Lockett eventually died, but that didn’t stop the criminal coddling defense bloggers (e.g. Gamso, Gideon, and Greenfield) from wringing their hands and calling for abolition of the death penalty.

Giving up is for losers. What I wanted was someone who could rise to the challenge of killing criminals. Naturally, I turned to the only name in blogging when it comes to sane, measured commentary about the death penalty, Crime and Consequences. I discovered that Kent Scheidegger had this to say:

As I have noted several times on this blog, lethal injection was a mistake from the beginning.  We should have kept the gas chamber and merely used a different gas.  Carbon monoxide, for example, is painless.

We have lethal injection for the time being, though, and we should make it effective.  Congress should act promptly to lift the restrictions on importation of the needed drugs and to outlaw manufacturers’ restrictions on resale of them.

Hmpf. I guess Scheidegger isn’t as much of a conservative as I thought. Not if he wants to limit manufacturers’ ability to set the terms and conditions of sale of their products in the free market. His closing paragraph just confirms his new namby-pamby liberal leanings:

We should do what we can to minimize pain in executions, but we should never forget that even in the worst execution what the murderer suffers is a tiny, tiny fraction of the suffering he chose to inflict on the victim.

“Do what we can to minimize pain in executions”? I visited C&C because I was looking for the hard-boiled law-and-order view, someone who would rain down hell upon the sinners. And he’s worried about minimizing pain? What a pussy.

So I set about searching for another solution. Surely someone among the hundreds of bloggers in my feed reader must have a friggin’ solution for the problem of how to execute people when the old tried-and-true drugs are no longer available. But try as I might, I couldn’t find anything useful.

In a desperate effort, I even visited The Watch, written by noted law enforcement expert Radley Balko, whose wonderful book Warrior Cop was such a loving tribute to the the brave officers who fight the war on drugs. He mentioned the Lockett execution in passing, but he had no solutions either. My heart began to fill with despair.

And then it hit me…

The states don’t need to find a new drug protocol to execute people. There’s another way states can kill people. It’s simple, easy, and effective. Every state already has the mechanisms in place, and they’ve been killing people with them for years. When it’s time for a prisoner to be executed per a judicial order, all they have to do is have a SWAT team stage a no-knock raid on the death chamber!

Just think of it: A few hours before the appointed time, the warden could call one of the local multi-agency drug interdiction task forces and leave an anonymous tip that the condemned individual was selling pot, giving the address of death row and the time when the inmate was expected to be there. The SWAT team would show up, lob flash-bang grenades into the room to stun the witnesses, kick down the door to the chamber, yell at the guy strapped to the gurney to get down on the floor, and when he doesn’t, empty their weapons into his center of mass.

Then the SWAT commander could complete the ritual by pulling a small baggie of crack out of his back pocket and announcing that police had “discovered salable quantities of narcotics at the scene” to reassure spectators that the bad guy got what he deserved.

This post by Mark Draughn at Windypundit was originally published at Death Doesn’t Have to Knock: A Modest Proposal

Also, the SORNA Challenge was using a very broad definition of the social benefits of SORNA, but limiting costs only to those directly incurred by the government, which could make SORNA look much more beneficial than it really is, if people using the study were not careful in interpreting the results.

I decided to check up on how the Challenge was going, and it turns out to be a bust:

We would like to thank everyone who submitted an idea under this Challenge.

We know our Challenge questions are tough, and in this case, no submission met the Challenge, so we will not award a prize.

We will continue to collaborate with the Office of Sex Offender Sentencing, Monitoring, Apprehending, Registering, and Tracking to address this important question.

That’s probably for the best.

This post by Mark Draughn at Windypundit was originally published at SORNA Challenge Update

First up, Scott reminds us of the case of Antwuan Ball, Joseph Jones, and Desmond Thurston who were accused of engaging in a massive drug dealing conspiracy. The case went to a jury trial, and they beat all of the conspiracy charges. The jury only found them guilty of some relatively small-time drug dealing.

When it came time for sentencing, however, the prosecutor argued that they should receive harsh sentences for the crimes they had been convicted of because they committed those crimes as part of a massive drug-dealing conspiracy, even though they had been acquitted of those charges.

This process of sentencing on acquitted conduct turns out to be a pretty common practice. I’m very shaky on the legal reasoning, but I gather the basic principle is that a convicted person can only be sentenced for crimes he’s found guilty of committing, but once he’s convicted, the judge has broad latitude to mete out any sentence up to the maximum spelled out in the criminal code, and in making his sentencing decision, he can take into account factors that have not been proven to the jury. I can’t even pretend to understand the details.

What I do understand, however, is that this is bullshit.

One of the things I’ve learned from my amateur interest in economics is that it is more useful to judge policies by their results, fully accounted for, than by their intent or the mechanism used to enact them. Thus, a legislature that passes price caps on gasoline may intend to make driving more affordable for motorists, but when gas station owners stop selling gas because they cannot do so profitably, the result is going to be a gasoline shortage. The price cap mechanism and the legislature’s intent to ease the life of motorists may be well-intentioned and competently implemented, but any evaluation of the effectiveness of the price cap policy should include the fact that it makes it harder for motorists to fuel their cars.

Most of us think that the verdict in a jury trial is pretty important: You can’t be sent to jail for charges they can’t make stick, right? But in a case like this, where the judge can take into account everything the defendant is accused of (even if he’s acquitted of those crimes by the jury) the prosecution can obtain a hefty sentence — just as if they had succeeded at getting the jury to convict on many charges — as long as they can prove at least one of their accusations and then convince a judge to choose a sentence at the high end of the range based on all the remaining accusations, even if the jury was not convinced. Thus prosecutors can increase a defendant’s sentence (on average) by lodging many accusations, even if few of them can be proven to a jury.

In other words, if we look at the sentencing results produced by the policy of sentencing on acquitted conduct, we can see that the ability of the jury to influence the defendant’s sentence is being diminished by this policy. This is pretty disturbing in a country that is supposed to guarantee a right to a trial by a jury. How real is that right if the jury has so little control over the resulting sentence?

This whole idea angered Jim Caron, a former economist for the U.S. Department of Agriculture, who also happened to be one of the jurors that acquitted Antwuan Ball of so many charges. When he heard about the prosecution’s request, he wrote a scathing letter to the judge. I can’t find a copy of the whole letter, but here’s an excerpt:

As you remember, Judge Roberts, we spent 8 months listening to the evidence, filling countless court-supplied notebooks, making summaries of those notes, and even creating card catalogues to keep track of all the witnesses and their statements. We deliberated for over 2 months, 4 days a week, 8 hours a day. We went over everything in detail. If any of our fellow jurors had a doubt, a question, an idea, or just wanted something repeated, we all stopped and made time. Conspiracy? A crew? With the evidence the prosecutor presented, not one among us could see it. Racketeering? We dismissed that even more quickly. No conspiracy shown but more importantly, where was the money? No big bank accounts. Mostly old cars. Small apartments or living with relatives.

It seems to me a tragedy that one is asked to serve on a jury, serves, but then finds their work may not be given the credit it deserves. We, the jury, all took our charge seriously. We virtually gave up our private lives to devote our time to the cause of justice, and it is a very noble cause as you know, sir. We looked across the table at one another in respect and in sympathy. We listened, we thought, we argued, we got mad and left the room, we broke, we rested that charge until tomorrow, we went on. Eventually, through every hour-long tape of a single drug sale, hundreds of pages of transcripts, ballistics evidence, and photos, we delivered to you our verdicts.

What does it say to our contribution as jurors when we see our verdicts, in my personal view, not given their proper weight. It appears to me that these defendants are being sentenced not on the charges for which they have been found guilty but on the charges for which the District Attorney’s office would have liked them to have been found guilty. Had they shown us hard evidence, that might have been the outcome, but that was not the case. That is how you instructed your jury in this case to perform and for good reason.

This story does not have a happy ending for the defendants. The prosecutors didn’t get quite the sentence they wanted, but the defendants still got tough sentences which were just upheld by the D.C. Circuit Court of Appeals a couple of days ago. Scott has more explanation and outrage.

Scott also posts about an Alabama Law Review article that reveals some disturbing (although not particularly surprising) things about how federal judges react to defendants’ allocution, which is basically their last chance to say something to the court prior to sentencing. I didn’t read the article, but Scott’s summary starts with some hope:

What this suggests isn’t that judges aren’t open to being influenced by allocution, but that the defendant’s statement (as well as defense counsel’s argument) do not sufficiently alter the calculus.  In other words, the opportunity may be there, but it’s either not being used very well or, more likely, there isn’t much that can be said that will change the judge’s views.

But significantly, when allocution does affect the sentence, it serves to lower the sentence, mostly within the guideline range, but closely followed by a below guidelines sentence.  The message here is that it is definitely worthwhile to put in the effort, and get it right.

Unfortunately, that proves to be difficult:

The problem, as is made clear in what follows, is that every judge has an approach that they think is perfect, and they differ markedly.

In their open-ended responses, some judges commented on the value of hearing defendants’ plans for the future, with one judge noting, “Some suggestion that the defendant has a concrete game plan for turning his life around would be helpful.” Many judges commented on the value of hearing the defendant reflect on his or her victims. But as one judge observed, defendants should “resist the powerful urge to whine and blame others.” Also, although defendants might think it wise to ask the court for forgiveness, at least for one judge, it is actually better to ask for leniency instead: It is not a “judge[’]s role to grant forgiveness. Asking for leniency and providing reasons why [a] certain sentence is appropriate works much better.” As this semantic difference demonstrates, defendants must forever be on their toes, navigating the bear-filled woods of each sentencing judge’s preferences—and defense counsel should be their guide.

What struck me about the specific advice is that many judges appear rather petty and trivial in what matters, reflecting a remarkable lack of understanding about the people whose lives are in their hands. Silly aspects, such as semantics, matter greatly to them, as they apparently are unaware that most defendants didn’t go to Harvard Law School and were lucky to have graduated from high school.  Yet, they expect such finely honed allocutions to reflect levels of mitigated speech, delivered with precision and yet sincerity, to appeal to their prep school sensibilities.

One of the the judges in the study advises that lawyers “Don’t let them read these long, prison-written letters. They tend to become maudlin, self-indulgent, and annoying . . . . Some defendants get carried away and start to whine that it wasn’t their fault, etc. That hurts any good that the attorney may have done.”

Scott’s response is moving:

Every lawyer knows the “long, prison-written letters,” put together with the best advice of their jail-house sentencing mavens, which is likely the longest thing they’ve ever written in their lives. They can be enormously proud of their speech, and desperate to deliver it.  They may swear it’s sincere, and demand their right to read it to the judge.  We may try desperately to explain that it’s not as effective as they think it is, to no avail. They want to be sincere, to be real, but it’s not the same sincere and real that judges want to hear.

That’s because, as the paper notes:

After preparation, defendants must come to court and deliver the allocution in a style that connects with the presiding judge. Overwhelmingly, judges in the survey indicated that they want defendants to show genuine remorse and sincerity. One judge bluntly recommended to defense counsel, “If your client cannot be sincere, and that is frequently the case, tell them to shut up.”

The problem with that idea, as the study paper points out, is that judges aren’t very good at detecting sincerity.

In fact, from what I’ve read elsewhere, I think it is safe to say that there is no such thing as a general human ability to detect sincerity. Study after study has shown that when we do correctly identify that someone is lying to us, it is not because of their demeanor, or their body language, or the tone of their voice. We detect their lies primarily through analysis of their statements and our knowledge of the subject matter. We look for internal inconsistencies within their story, we check their statements against facts we can verify, we compare their story to similar stories that we know the truth or falsity of, and we try to nudge them off their story to see if it changes.

Obviously, if the defendant’s allocution conflicts with his earlier testimony, or with other trustworthy evidence from the trial, a judge could conclude that he’s lying. But how can the judge tell if he’s sorry? How can the judge tell if he really wants to be a better person? How can the judge tell what’s in his heart?

Judges do have a lot of experience hearing allocutions, but I don’t think they get too much feedback. If they disbelieve the sincerity of a defendant’s remorse and sentence him to longer than he deserves, how will they ever know if they made a mistake? And if they don’t find out about their mistakes, how will they learn?

Frankly, I’ve always been a little horrified at the idea of “taking responsibility” and related concepts in sentencing. It seems like some kind of Soviet show trial, where the defendant is found guilty and then coerced into confessing his crimes. Because that’s what it is when you threaten to give somebody a longer sentence if they refuse to admit guilt: A coerced confession.

As Scott points out, this leaves defendants who believe they are innocent with a difficult choice:

The most notable, and glaring, omission in this article is what a defendant can do if he maintains his innocence through sentence.  This may suggest that no federal judge believes that any defendant being sentenced is innocent, or that they just don’t want to deal with difficult situations.

Given that they clearly want sincere expressions of remorse, it presents a dilemma for the defendant who maintains that he was wrongly convicted, which means he is unable to gain the advantage of a reduced sentence based on a sincere expression of remorse because he isn’t guilty, or he must give up his position of innocence to feign remorse to appeal to the judge.

Even with defendants who are factually guilty, this approach penalizes people who aren’t well educated or good at explaining things or used to speaking to judges. It’s hard to see what any of those things have to do with how much prison time is necessary, but apparently that’s how the system works.

On the other hand, this is also a system that rewards those who are slick and well-spoken and well-prepared by their attorneys. It rewards those who can acknowledge their bad acts and then look the judge in the eye at just the right point and, with just a hint of real tears, say they’re sorry and they’ve learned their lesson and with God’s help they will do better. It rewards those who are good at figuring out what other people want to hear and then saying it convincingly and with apparent sincerely. It rewards, among others, psychopaths.

Somehow that doesn’t seem like a very good way to fight crime.

This post by Mark Draughn at Windypundit was originally published at Lessons in Allocution and Acquitted Conduct

This isn’t some paranoid, whack-job, cop hating-nonsense. It’s an accurate description of how all criminal laws are enforced.

Even a minor violation such as loitering can have a penalty of several hundred dollars, which is probably an awful lot of money if you’re the kind of person — poor, minority, homeless — who is likely to get hit with a loitering charge. And if you don’t (or can’t) pay the fine, sooner or later someone will get a judge to issue a warrant for your arrest, meaning that men with guns will come to take you away and lock you in a cage. If you’re accused of a more serious crime, they’ll just skip straight to issuing an arrest warrant, and men with guns will come to take you away and lock you in a cage. And in either case, if you resist being taken away and locked in a cage — a perfectly natural (albeit unwise) impulse — the men with guns will overcome your resistance with violence, up to and including killing you.

And all of this can happen before you are ever convicted of any crime.

So for God’s sake, people, think before you legislate! (Or encourage others to do so.) Because whether your law is about selling spray paint to someone younger than 18, buying too much cold medicine, selling raw milk, watering your lawn, or loitering, when you support a new law, you are implicitly saying that you think people who disobey your law should be subject to the violence of arrest and incarceration.

So before you demand a new law, ask yourself: Is what’s bothering you really worth all this trouble?

This post by Mark Draughn at Windypundit was originally published at Law Enforcement

That doesn’t seem like nearly enough. When someone killed Tupelo, Mississippi, police Corporal Kevin Stauffer and injured another officer last month after they responded to a bank robbery, the FBI offered a $50,000 reward, which was matched by the bank. With other donors, the reward grew to $162,500.

Earlier last year, after someone shot and killed a Rite-Aid store manager, the Philadelphia Citizen’s Crime Commission and Rite-Aid offered $20,000. That reminded me of a story I only vaguely remember from around 1970 about a city in which several pharmacists had been killed during robberies. The pharmacists’ association responded by offering a standing $15,000 reward for the killing or wounding of any pharmacist during a robbery. That was so much money at the time (equal to about $90,000 today) that people would turn in their own family members to get it. Very few criminals were willing to risk robbing pharmacies after that.

Then again, I suppose that offering a large reward could create its own problems. After all, in order to collect the $100,000 reward for information about the shooting of a cop, someone has to shoot a cop. I imagine there are people out there who might try to work both sides of that deal by shooting a cop and then planting the gun on someone else so they could turn them in for the reward.

In any case, 1-888-YPO-SHOT was created by Bill Kugelman, whose son was an officer killed in the line of duty, along with the Chicago Police Memorial Foundation. They apparently got the idea from the NYPD’s 1-800-COPSHOT program.

This post by Mark Draughn at Windypundit was originally published at YPO-SHOT

I’m not saying there’s no difference. Punching a stranger for no reason is assault and battery, but punching a home invader is self-defense, which is usually considered justified. But the difference is not that striking a home invader does less harm but that self-defense produces a benefit — defense of the innocent – which we as a society consider to be worth the cost of the harm.

But the punch still hurts.

Jeff Gamso gets this. When discussing state executions of convicted criminals in capital cases, he calls them murders. I suppose that’s wrong as a matter of law (although statutes are sometimes sloppy about these things, so he may have a good argument), but calling it “murder” is his way of emphasizing that the state is doing a terrible thing. To the person being killed, it doesn’t matter that a judge has approved of the killing. It’s no different than being gunned down in the street.

Proponents of capital punishment may well argue that it brings benefits that are worth the cost. However, if they try to downplay the cost by arguing that the condemned person does not deserve to live, then they are trying to cheat: There’s almost no harm you can’t justify if you are allowed to discount the humanity of the victim. An argument for capital punishment must show benefits that offset and therefore justify the killing of a human being. The terrible killing must, at the very least, prevent something even worse.

Ken White gets it too, when discussing the case of David Eckert, who was picked up by cops in New Mexico last January and accused of transporting drugs in his rectum. The cops then took him to some doctors who repeatedly forced their fingers into his anus, gave him several enemas, and eventually sedated him so they could perform a colonoscopy. They did all this to Eckert without his permission and over his objections.

Ken discusses this incident in a post titled, “What Is The Quantum of Proof Necessary for Police to Rape and Torture you in New Mexico?” Given the details of what happened, I applaud his use of the terms “rape” and “torture.” If the accounts and information given Ken White, Scott Greenfield, Chris Ramirez of KOB news, and of course David Eckert and his lawyer are all true, then there is little doubt in my mind that

• Doctors Robert Wilcox and Okay Odocha essentially raped and tortured David Eckert.
• The Gila Regional Medical Center provided the location and equipment for the rape and torture of David Eckert.
• The City of Deming and Police Officers Bobby Orosco, Robert Chavez, and Officer Hernandez may have abetted the rape and torture of David Eckert.
• Hidalgo County and Deputies David Arredondo, Robert Rodriguez, and Patrick Green may have abetted the rape and torture of David Eckert.
• Deputy District Attorney Daniel Dougherty wrote the warrant that led to the rape and torture of David Eckert.
• 6th District Judge Daniel Viramontes signed the warrant that led to the rape and torture of David Eckert.

You’ll notice I haven’t mentioned whether police found the drugs they were looking for. That’s because whether or not they found the drugs doesn’t change the terrible things they did. In point of fact, they didn’t find any drugs, but even if they had, I don’t see how it could be worth the harm they caused.

I wish that when legislators were considering passing new criminal laws, they would stop to think about how those laws will be enforced, and the harm that is inherent in that enforcement. As Molly Crabapple says,

Arrest is always violent. The NYPD may or may not break your ribs, but the process of arrest in America is still a man tying your hands behind your back at gunpoint and locking you in a cage. Holding cells are shit-encrusted boxes, often too crowded to sit down. Police can leave you there for three days; long enough to lose your job. If this seems obvious, I say it because the polite middle classes trivialize arrest. They talk about “keeping people off the streets.” They don’t realize that the constant threat of arrest is traumatic, unless it happens to them or their kids.

Legislators should keep in mind that every arrest is a violent act. Every SWAT raid is a home invasion. Every stop-and-frisk is someone fondling your body against your will. Every interrogation is a stranger asking impertinent questions. Every imprisonment is a kidnapping and every execution is a murder.

I don’t see how you can justify any of that for consensual crimes, since any true harm could be eliminated by withholding consent. And even when the goal is to stop criminals from doing real harm to other people, state-sanctioned harm can only be justified if it is reasonable in proportion to the harm and reasonably likely to actually prevent that harm.

This post by Mark Draughn at Windypundit was originally published at Knowing Harm When We See It

NIJ CHALLENGE: COST-BENEFIT OF SEX OFFENDER REGISTRATION LAW

Are you up for the challenge? Enter NIJ’s first-ever SORNA Challenge! NIJ is seeking innovative ways of developing strategies to measure the implementation costs and public safety benefits of the Sex Offender Notification and Registration Act (SORNA) – part of the Adam Walsh Child Protection and Safety Act of 2006 – by improving the effectiveness of sex offender registration and notification programs in the United States.

…

A cash prize of $50,000 is available. Deadline: Oct. 31.

The announcement also makes this important observation:

Empirical research on sex offenders has grown over the past decade, but no study to date has examined the multifaceted effects of SORNA, specifically the wide range of costs that have been or may be incurred in implementing SORNA, or the public safety benefits achieved with SORNA compliance.

The squawk is a bit upset by that revelation:

This is apparently a new streamlined process of how a bill becomes a law: first enact it, then figure out what it costs and what it does.  What next? Cash prizes and free stuff for figuring out the cost/benefits of the Iraq war?

Given the dreadful aspects of sex offender registration laws — the limits on where registered offenders can live and work, requiring prostitutes and horny teenagers to register as if they were child molesters, the near-impossibility of a return to normal life — I understand his outrage. But to be fair to the NIJ, you can try to predict the effects of legislation in advance (with varying degrees of success) but you can’t actually measure those effects until the legislation is in place.

Even then, measuring the effects of legislation such as SORNA is difficult work. The worst approach — all too common in some circles — is to measure something like the number of offenders registered, or the rate of compliance with in-person check-in requirements, or the number of times the public queries the registration database. The are relatively easy numbers to obtain, but they are the wrong numbers: They are inputs, and we want to measure results. Looking at those numbers would be like measuring the effectiveness of an advertising campaign by counting the number of ads placed instead of the amount of product sold.

So what we want to measure is the effect that SORNA has had on the crime rate. The problem is that while measuring the rise and fall of crime rates is relatively easy (although there are complications), it is much harder to determine why the crime rates rise and fall. There are many causes, all operating at the same time, yet we are only interested in one of them. We need to find a way to untangle the effects of SORNA from all the other things that affect crime rates — demographics, the economy, migration, public safety budgets, police tactics and equipment, and so on. This is a very complex problem, filled with opportunities for error, but social scientists have developed methodologies that sometimes give meaningful results.

I’m no expert, but I think I can describe the general idea behind these methods: The SORNA act works by imposing requirements on local law enforcement agencies, and those agencies have been coming into compliance separately at different times since the law was passed. Furthermore, some of these agencies were already in compliance with parts of SORNA before it even passed. This gives us a kind of natural experimental randomization, since SORNA came into effect in different places at different times, it is unlikely to correspond by coincidence with other factors affecting the crime rate. In one city SORNA might have taken effect in combination with an increase in police manpower, while in another city at another time, SORNA might have taken effect while police manpower was decreasing.

Given a large enough amount of data, it might be possible to detect the effect of SORNA. Perhaps we could look at crime rates in every agency’s territory for, say, the three years before and after SORNA requirements were implemented. On average, if SORNA is providing benefits, we should see a greater reduction in crime (or a smaller increase) for those periods than for comparable periods at agencies that did not implement SORNA.

In addition, some of the confounding factors are measurable, such as the local economy and police budgets — and we can statistically estimate the effects of these factors on the crime rate and then subtract them out of the data to compensate for their effects. This should more clearly bring out the effects of the SORNA legislation.

This kind of study is a complicated undertaking, requiring obtaining SORNA compliance and crime rate data from as many states as possible, and ideally obtaining at least some data on a county-by-county or agency-by-agency basis. Then all the data has to be converted to a single coding system that can be fed into statistical analysis software. There could be lag effects (maybe the benefits of SORNA don’t show up until two years after implementation),and the data might not be robust (in some studies of the deterrent effects of the death penalty, the effects go away if you eliminate Texas from the data). And because I don’t do this sort of thing for a living, there’s probably lots of other issues to worry about that I’m not aware of.

However, in general, that’s the methodology for studying this kind of effect. This work is not easy, and the results are often of poor significance, but the process and its limitations are fairly well understood. It’s a standard approach to these kinds of studies, widely used and accepted by sociologists, economists, and public health statisticians.

Which raises the question…what is it about the accepted methodologies that causes the NIJ to seek “innovative” approaches.

A cynic might suggest that the normal methods are insufficiently rigged in SORNA’s favor. Consider this part of the request for proposal:

NIJ’s SORNA Challenge seeks creative and innovative research strategies for future researchers to use when studying (1) the implementation costs associated with complying with SORNA and/or (2) SORNA’s public safety benefits (examples include, but are not limited to, the Act’s general and specific deterrent effects, its effect on law enforcement’s ability to prevent crime, and its effect on the public’s ability to protect itself).

What stands out immediately is that these two challenge components add up to a highly biased benefit-cost calculation: The study proposal uses a broad definition of the the public safety benefits of SORNA, but limits the cost component only to implementation costs. This totally ignores the costs incurred by anyone outside the government, especially the costs incurred by people whom the government wishes to place on such registries, such as the legal costs of fighting registration, the loss of employment, and the general loss of freedom to live where they want and do what they want. Basically, they’re counting only part of the cost, but they’re including all of the benefits.

Allow me to illustrate the problem with an absurdity: Suppose that instead of SORNA, we were to pass RAAMSA, the Random African-American Male Shooting Act. RAAMSA funds local law enforcement programs in which inner-city police officers annually gun down 10,000 young black men at random. Using the methodology implied by the SORNA Challenge, this is a highly cost-effective program: Assuming $2500 to find, shoot, haul off, and cremate each body, the annual budget is only about $25 million dollars. Given that the lifetime incarceration rates for black men is about 30%, we can therefore assume that RAAMSA takes at least 3000 criminals off the street every year at a cost of less than $8500 each — far below than the cost of imprisoning those criminals for even a single year, and RAAMSA is a permanent (or should we say final) solution.

If the SORNA Challenge study components would miss such a monstrous evil as this, I think it’s safe to say that they could create a misleadingly rosy impression of the benefit-cost qualities of SORNA.

Now it’s entirely possible that no one intends to use these components in a misleading way. Perhaps, years from now, everyone involved with this study will freely and honestly admit its limitations. Perhaps no one will look at the results of these two studies and conclude that the benefits of SORNA outweigh the costs just because the benefits of SORNA outweigh the implementation costs of SORNA. Perhaps no one will issue pronouncements and press releases about the cost-effective wonders of SORNA. Perhaps these results will not be misreported by the media and bloggers.

This post by Mark Draughn at Windypundit was originally published at Suspicions About the SORNA Challenge

(1) These cases are very fact-specific: Guilt or innocence can turn on just a few details, and it will be hard to arrive at a sound conclusion about Zimmerman’s guilt unless you know all the relevant details.

(2) The media usually does not report all the relevant details.

(3) Some of the relevant details are unknowable, except to the participants.

So unless I listen to the entire trial — and maybe not even then because of (3) — I don’t have any opinion about the guilt or innocence of George Zimmerman that I can be confident about. For the most part, neither does anyone else, but that doesn’t seem to stop many of them from talking about it.

This post by Mark Draughn at Windypundit was originally published at Nothing to Say About the Zimmerman Trial