When Good Journalists Get Lost

Prominent and skilled reporter Alec MacGillis of ProPublica recently published a terrible piece of copaganda in the Atlantic that many readers asked me to dissect. Because the article is one of the best examples of copaganda by a very smart but deeply confused progressive person—and because we badly need smart people like MacGillis to change how they report on the urgent problems of the punishment bureaucracy—I decided to write about it.

My newsletter will always be free. It’s mostly me talking about copaganda and subjecting people to my unskilled paintings. However, if you enjoy it and want to donate, you can choose to subscribe for various amounts and the money will be a donation to Civil Rights Corps so that we can continue our important work of fighting against the systems and narratives that support state violence and repression. I don’t personally make any money from this newsletter:

The thesis of the article by MacGillis is that “the cause” of a recent “crime wave” is the fact that jury trials and other pretrial court proceedings are facing a backlog. Before I get to a few of the specific problems with the article, I want to make several general points.

First, there is a scientific consensus that more unequal societies have more violence. By far the most significant way to reduce violence and harm of all kinds in our society would be to invest in the root causes: affordable housing; health care; mental health treatment; early childhood education; systems of social connection and care; nutrition; environmental pollution abatement, etc. Journalism like MacGillis’s piece in the Atlantic is part of a wave of “Think Pieces” that ignore these most important solutions. They distract us from the structural changes our society needs by pushing relatively minor tweaks to the machinery of state violence. This is what we often get in the media: after all, it's not new or sexy to suggest that we could have fewer murders with less poverty, accessible health care, and far better schools.  This Atlantic article, at bottom, is an example of a skilled reporter who doesn't know much about an issue trying to be smart and novel and losing the forest through the trees.

Second, one of the most prominent forms of copaganda involves pointing to various minor policy tweaks within the criminal punishment bureaucracy and claiming that they have significant effects on “crime:” police training, body cameras, small variations in the number or deployment of cops, small fluctuations in sentencing, relatively minor tweaks to bail reform, court backlogs, etc… The reality is that there is no evidence that any of these things play a significant role relative to the root causes listed above in what are huge structural issues of our society.

Third, what these distractions do, though, is take attention away from public urgency to address root causes like inequality. Here is what I wrote in my book Usual Cruelty about this, and it applies perfectly to what MacGillis allowed his police and prosecutor sources to do to him:

Fourth, like other recent pieces in the Atlantic and the New York Times, all of this therefore buys into an anti-science assumption that has pervaded elite thought not unlike climate science denial: punishment or lack of punishment is simply not a main driver of violence.  There are many countries with far lower rates of homicide than the U.S., and no serious person thinks that it is because they have better and more ruthless and efficient mechanisms of human caging. The kind of thinking expressed by MacGillis, even ignoring the unfortunate flaws in his reporting on this particular story, is just such a narrow way of thinking that misses most of human history and most of the contemporary world. It also fails to distinguish between accountability and “consequences,” which MacGillis equates with human caging. It’s a gut punch to the amazing movement of crime survivors and victims of gun violence across the country who have built systems of restorative justice in their communities. I hope MacGillis will reckon with his erasure of them and their inspiring work.

Okay, so to the details. Here is the title:

Note up front a couple important things in the title. First, there is a singular “crime wave.” This is weird, because although crime is near historic lows in the U.S., different forms of “crime” are up or down in different areas around the country. Though the title talks about “crime,” MacGillis’s actual article turns out to be about shootings/murder in just a few cities. This is a very sloppy way to start, but it’s also the fault of Atlantic editors.

Second, MacGillis is not only making a bold claim that he has identified a scientifically “causal” mechanism, but that he has identified “the” cause. As we’ll see below, all of this is not only confused nonsense, but also useful copaganda for powerful interests in the punishment bureaucracy. And MacGillis’s follow-up interview on NPR spewed an even more careless iteration of this misinformation to millions of listeners.

This paragraph describing the thesis belongs in the misinformation hall of fame

First, the paragraph is a good example of an unethical practice I’ve discussed before. MacGillis does what I warned about in my recent essay on how reporters use experts to manipulate people: he grants a weird sort of anonymity to “criminologists” and “many people who work in criminal justice.” By grouping these people in each group all together, he both falsely suggests that these are consensus views among those groups when they are deeply controversial. This tactic also removes accountability from whoever he is talking about. Just who are these “criminologists” and “people who work in criminal justice.” As you read the article, it’s seems clear that he means “pro-police criminologists” and “prosecutors” respectively, but it’s hard to tell since he hides them from us. There is no journalistic reason to allow such controversial claims to be disseminated without the accountability of who is making the assertions. Instead, we are just told that it’s what “experts” think.

Incredibly, two paragraphs later, MacGillis does it again, this time saying that some anonymous consensus of “experts” blames court shutdowns for murder: “Above all, experts say, the shutdowns undermined the promise that crimes would be promptly punished.”

Second, the only assertion supported with a citation in this important paragraph is the supposed connection of rising gun sales to murder. It’s worth noting that this link goes to another Atlantic article blaming the rise in murders not on court closures but on more gun sales! That article, by the notorious former CIA, Palantir, and police contractor who I’ve written about before, was also mostly speculation but it’s pretty hilarious that the Atlantic keeps giving space to these guys to publish stuff that is internally contradictory. Which explanation is it? It’s almost as if they can just spew whatever speculation comes into their heads that happens to be expedient for whatever political point they want to make in the moment.

The article gets worse though. MacGillis attempts to support a bold national, scientific thesis with a bunch of anecdotes and a few misleading stats from Albuquerque and Wichita. If you want to know what it looks like when a reporter has no evidence to support what they are talking about, then take a look at this paragraph from MacGillis, which is a preface to a horrific anecdote about a person being killed by a cellmate in Albuquerque after he had been jailed prior to trial for “nearly 10 months for trying to steal a bottle of tequila with a pocketknife.”

Translation: I can’t give you any evidence for the widespread social phenomemon I’m advocating, but let me tell you a horrific story.

Amazingly, MacGillis’s take away from this story about a presumptively innocent person being murdered in an overcrowded jail cell is that courts need to move faster rather that that the U.S.’s unprecedented rate of wealth-based pretrial detention is horrific and ineffective. It’s really a surprising, cruel take. Was this person killed because the court system isn’t doing jury trials or because he was one of the 500,000 people (mostly illegally) jailed prior to being convicted of anything in this country every night?

Nor does this anecdote provoke any speculation from MacGillis about why it is that U.S. society tolerates such brutal, horrific local jails that separate families and that lead to trauma, violence, mental illness, and isolation that actually increases crime. That, of course, would be an actual expert consensus. He doesn’t even talk about it. Even on his own version of events, MacGillis didn’t have to pitch this as a murder caused by a lack of '“swift conseqences” for murderers; it would be far more natural to describe it as caused by the state of our pretrial detention and jail systems.

Then MacGillis moves to Wichita, and we get the intellectual core of the article

There is so much wrong with this paragraph that it would take a book to do it justice. But it’s clearly written by someone who doesn’t understand how the bail and plea bargaining systems work.

A few quick things: MacGillis’ only support for the bizarre assertion that “many suspects arrested in the shootings were defiant, suggesting that noting would come of the charges” because the court system was shut down is the police chief. But this copaganda makes zero sense. Neither MacGillis nor the police chief points to a single actual example of a shooting suspect being “defiant,” whatever that means. Neither of them points to a single example of a shooting suspect refusing a plea deal because no trials were happening and they thought they could avoid “consequences.”

In reality, the exact opposite has been happening in the cities I’ve been working in: during the pandemic, there are swift and horrific consequences for even being arrested, let alone convicted: people charged with serious crimes are being detained prior to trial, and because few trials are happening, they are languishing in increasingly crowded jails under horrific conditions that deprive exercise, fresh air, nutritious food, medical treatment, and family contact. They are desperate to plead guilty in many cases. Ironically, this is exactly the situation in MacGillis’s New Mexico anecdote. More damningly, MacGillis and the police chief ignore a fact devastating to their argument: what’s happening to all these murder/shooting suspects? Well, I suspect the vast majority of them are being detained prior to trial, which is the norm nationally in cases like that. Is being immediately locked in a cage separated from your family in deadly and grotesque conditions not a “swift consequence”?

How does this fit into MacGillis’s theory that swift consequences and not long punishment deter murder? We have no idea. We are not told, because this article is such an internally inconsistent mess that it doesn’t even bother to see how its own anecdotes and claims contradict its thesis. (Something that would have been consistent with MacGillis’s thesis, for example, is if the police chief had said that police stopped trying to arrest suspects. But that’s precisely not what the chief said: he said they were arresting suspects but that once caught and facing swift consequences of pretrial jailing they were “defiant.”)

I suspect the police chief’s assertion is a fabrication. MacGillis certainly doesn’t do any of the journalism necessary to test it, despite it being the factual claim central to his thesis.

Back in Albuquerque, MacGillis’s main source is the fear-mongering local DA’s policy person. After reciting some meaningless DA office statistics about mostly low level cases (i.e. not the subject of the article), MacGillis allows the DA to spew this nonsense:

Note that this claim is patently false. Even a cursory perusal of Albuquerque’s court system reveals many many many people being punished for many many many crimes. Again, neither the DA nor MacGillis offers a single example of a shooting or murder suspect who didn’t face “consequences.” Instead, we are told that a few dozen fewer jury trials are happening. Are we meant to believe that a supposed increase in shootings was really the result of a couple dozen criminal jury trials, most of which were for lower level crimes anyway?

MacGillis appears to have been hoodwinked. For a journalist who has produced genuinely good, rigorous reporting on companies like Amazon, this lack of skepticism and sophistication is genuinely surprising. The first time I heard this argument was from disgraced, Trumpish former Harris County Sheriff Art Acevedo. It seemed like every time I would talk to him in person or see him in an interview, he was harping on the need to speed up courts, couched in professed concern for “speedy trial rights.” What’s going on here? As I’ve written about, there is a strong undercurrent in the criminal punishment bureaucracy obsessed with “efficiency.” This makes sense: no society in the recorded history of the world has attempted to transfer so many human beings from their families, schools, homes, jobs, churches, and civil society into government run detention facilities. When you combine that with the American legal system’s formal commitment to “process” and the associated nominal protections in the Bill of Rights, you get a system that has to be obsessed with processing quickly with ruthless efficiency.

What’s most incredible about MacGillis’s article is that it doesn’t include any meaningful data to support it’s “causal” argument, not even about Albuquerque or Wichita. MacGillis says that Withita resumed jury trials in July 2020, but “[b]y the end of 2020, homicides were up sharply in Wichita.” How does this fit with his thesis? We are not told.

Later in the article, MacGillis claims that Wichita “bucked” a national trend and say “9 percent” fewer homicides. What MacGillis elides is that this appears to be a decrease of only 5 homicides. MacGillis, a reporter for one of the most rigorous journalistic institutions in the United States (ProPublica) is basing a causal claim regarding an important area of social policy (courts, police, prosecution, murder) in one of the most prominent magazines in the United States (the Atlantic) on a small number of homicides and a few misunderstood cherry-picked statistics from a few handpicked small cities?

More damning, there are lots of other crimes other than murder. In many places, including those with court backlogs, those crimes are down! How does MacGillis’s new scientific theory about court backlogs reducing “swift consequences” explain this? He doesn’t even mention it. Just totally ignored. It’s genuinely hard to understand how editors allowed this. The lack of rigor in this reporting is stunning.

Toward the end of the article, MacGillis does allow a public defender to offer a slightly different perspective. Although MacGillis doesn’t mention this, the basic point from the public defender is largely a consensus view:¹

How MacGillis handles this is very revealing. He gives the last word to the prosecutor, and doesn’t correct the prosecutor’s false statement: this is not an issue “legislators would have to” resolve. The point is that these prosecutors, like prosecutors across the U.S., have created massive court backlogs by continuing to insist on flooding the courts with low level cases. They don’t have to do that. They have complete discretion to decide which of the many crimes to prsoecute and which to ignore. They do it every day when they ignore the crimes of the wealthy or the crimes committed by police and jail guards. The plurality of prosecutions in most major jurisdictions are drug cases, and the vast majority are misdemeanor cases, typically involving unhoused people and people with mental illness.

I’ll stop there because this is getting long. MacGillis is not an outlier. Many smart people and well-meaning reporters have been so thoroughly propagandized by living in a society that cages 5 times its own historical average and 5-10 times other countries, that they accept many aspects of the current system as given and are unable to process that many of its underlying assumptions are flawed. It is my humble hope that engaging with criticism like this can be a small part of them changing.

As the threat of authoritarianism rises, we absolutely must be decreasing the size and power of these bureaucracies, not making them more efficient.

1. For example, officials in Harris County, Texas commissioned a public safety review of what to do with its court backlog in 2020, and the recommendation was clear: dismiss less serious cases and focus on serious ones.