Texas-Style Justice, According to The Atlantic

There's coverage last week by Juleyka Lantigua-Williams in The Atlantic of the excellent Texas Justice Initiative dataset.

The TJI data is great research, on the part of Amanda Woog, Jennifer Laurin, Kali Nicole Gross, Michele Deitch and those who collected this data via the Texas Office of the Attorney General from law enforcement agencies, local jails, and the Texas Department of Criminal Justice (TDCJ). The project is important and excellent.

The article? Well, it's done with an irresponsible, or at the very least uninformed, spin. At first, judging by the headline, it really looks as if Texas killed SEVEN THOUSAND PEOPLE. Texas did not.

For example, to say in the lede, "Many died of natural causes" is not quite the same as saying in the lede, "Seventy percent, or 4,870 of the 6913, died of natural causes, and eight percent, or 573, died as the result of a "justifiable homicide" - which most often means that they were killed while attempting to kill a police officer, civilian, guard or another prisoner. Four percent, or 275 people, voluntarily used alcohol or drugs in such quantities that they died of ingesting these substances, and 772, or 11 percent, committed suicide."

It says, "deaths in custody," but doesn't really describe the difference between deaths in prison of prisoners (4,684 of the 6913) and deaths of people in jail or in contacts with police officers. It does not mention that 96% - 4,522 - of the 4,684 prisoner deaths were of natural causes (4,223) or suicide (299).

I consider it somewhat dishonest to not point out that more than eight out of ten cases of all in-custody deaths were natural causes or suicide, and that nine in ten were those plus voluntary substance abuse or justified homicide.

Yet the article talks only about "Texas-style justice." That is both gratuitous and misleading.

I don't dispute Lantigua-Williams ultimately put these figures (somewhat) later in the article - she mentions those figures six paragraphs - or five hundred and eleven words - into the story that is ostensibly about the data. But it is the Lede through which most people take the tone and timbre of a piece; it helps influence the opinion of most readers.

Then the article highlights this nugget from the website, without context: "Black people comprised 12% of Texas’s population in 2010 and 30% of custodial deaths in 2005-2015. White people comprised 45% of Texas’s population in 2010 and 42% of custodial deaths in 2005-2015. Latinos/as comprised 38% of Texas’s population in 2010 and 28% of custodial deaths in 2005-2015."

So, what are we, the readers, to make of the stated fact that 30% of prisoners are black, while only 12% of Texans are black? It's a straightforward question.

Is the article suggesting, by highlighting this sole passage from the TJI, and based on two unrelated numbers (racial composition of prisoners versus racial composition of a diverse state that takes 12 hours to drive across) that Texas black people are being unfairly railroaded in wholesale quantities by the entire Texas criminal justice system? That would be truly amazing, because judges and courts in Texas are run entirely separately from police and corrections. The reader is left indelibly with that exact impression.

Let's compare the data on these issues from another state: California. I know that the article's headline called the TJI data "A first-ever database,It's important to note that the headline and sub-hed may well have been written by an editor, not Lantiqua-Williams, so we cannot point the finger in this case." but as I wrote in February, the State of California's Department of Justice had already released similarly detailed, similarly visualized data on similar issues, and updated it this year. Many other open data projects already do, or are already seeking to, provide this kind of data (the raw data is available here).

California had a similar number of deaths in custody (6,837) over the same period (2005-2015). Like Texas, the majority of California's deaths in custody were natural causes (61.4%) or suicide (10.5%), or accidental (8.4%). A slightly higher number were homicides by law enforcement (14.4%) but California also breaks out homicide by other inmates (2.8%), and "other".

Interestingly, California-Style Justice also has (by the criteria of the article) a "disproportionate" number of deaths of African Americans relative to their statewide proportion - African Americans comprise just 5.8% of Californians, but a whopping 23.7% of inmates. Rather than leaving it there, though, California took it the next step: comparing prison populations to arrests. They found that Blacks comprise 16.7% of arrests, 23.7% of incarcerations, and 24.2% of deaths - meaning that blacks comprise an appropriate number of inmates based on arrests.

White people in California comprise 41% of the population, 34.9% of arrests, 28.8% of inmates, and 41.6% of deaths - a disproportionately high number compared to both arrests and inmate populations. Thoughts?

Even if we were to ignore the really inappropriate and gratuitously mean remark about Texas-style justice from the article, what we see here is that in fact, based solely on comparing state-to-prison populations, California is just a little "worse" than Texas in terms of representation of African Americans. This almost certainly means one of two things: the relationship between state population and state prisoner population doesn't mean what the article seems to imply it means, or, maybe, the entire United States Criminal Justice System is similarly racist.

To make that point, we would have to suggest that this same wholesale, unchecked systemic bias against black and Hispanic (white and black) prisoners exists in equal measure across the United States in separately run police, courts and corrections, because these Texas prisoner population numbers match really closely the national averages of criminality by race. Such a conclusion has not been proved by any study.

According to the excellent Bureau of Justice Statistics, the national incarceration rate combined for federal and state prisoners for white males is about 466 per 100,000, and for black males it's about 2,805 per 100,000. If I were disingenuous or bad at math, I would say that nationally, black males are incarcerated at a rate, of nearly six times that of white males.

But that would leave out many factors, and be both inaccurate and unfair.

What I will say is, having reviewed the data, Texas is not really much of an outlier in terms of racial composition of prisoner populations, except for the fact that Texas is open enough to provide these data to researchers like the TJI. The TJI didn't uncover this data - it as gathered and provided by the state - what TJI did was make it accessible to scholars and activists and those curious enough to dig in to it. It's a wonderful public service!

I would suggest perhaps the issue involves something else other than race?

As it turns out, poverty, not race, drives and is the best indicator and predictor of criminality in the United States. The racial composition of those living in poverty match almost exactly the racial composition of those incarcerated in the United States. It's really quite uncanny.

What the article also left out was the fact that, as I recently wrote in The Washington Post, African Americans in Texas make just under half the money of white Texans (2009 Texas median household income for whites was $59,836; while for blacks it was $35,438, and for Hispanics $35,628).

This is surely not the fault of the police, the courts or the corrections department.

That the racial composition of the prisoner population is what it is, in Texas and across the country, is not merely a criminal justice issue. It is a societal issue, a property-tax issue (no one wants to pay higher property taxes for better pre-K education, outdoor playgrounds, school recreation programs, school breakfast and lunch programs, music classes and the hundred other things sociologists have been saying since 1960 drives success in early education).

It is, I submit, a political issue at the state and national level.

Yet in her article, Lantigua-Williams leaves cops and judges and corrections officials holding the entire bag based on some, frankly, really slapdash demographic research.

At the same time, an article ostensibly written to celebrate the availability of data, unfairly undermines the very value of this wonderful and open data. In the hands of those who want to find what the data shows, as opposed to finding what we seek in the data, open data like this can be extremely valuable. Unmentioned in this article (despite erroneously giving it credit for being first) is that Texas - yes, Texas - is a national leader in making available racial profiling detection programs and data, officer-involved killing data, and open data for law enforcement.

When it comes to data, don't mess with Texas.

It would have been really great to have an article in The Atlantic that recognizes that the State of Texas is actually trying to help, as opposed to sucker-punching it for "Texas-style justice," and implying its entire criminal justice system is run by racists.

That is a position that refuses to acknowledge whatever the role of criminals in the criminal justice system.


Nick Selby proudly serves as a Texas law enforcement officer. He is the author of In Context: Understanding Police Killings of Unarmed Civilians, and The Context Blog


James, Lois, Stephen M. James, and Bryan J. Vila. “The Reverse Racism Effect.” Criminology & Public Policy (2016).

Get Notified Of New Posts
(We won't ever sell, rent, or lend your email; nor will we ever spam you)

Texas Style Justice - August 1, 2016 -