“This guy is pretty much scalped over here. This guy, he’s breathing but I couldn’t get a pulse on him. I think he’s going,” a Border Patrol agent is heard saying on Doña Ana County Sheriff’s Office body cam footage from August 3, 2021. Early that morning, Border Patrol agents pursued a vehicle at high speeds on a narrow road, causing a crash that killed two people and injured eight others.

One of those who died following the August 3 crash was Erik Molix, a U.S. citizen whose parent is now our client. Between 2019 and 2021, the number of deaths resulting from Border Patrol vehicle pursuits jumped 11-fold, to 23 deaths last year. The body count for such chases continues to increase this year. The agency’s deadly actions raise urgent questions about how these pursuits are investigated and what measures are taken to ensure public safety and accountability.

Of particular concern has been the involvement of Border Patrol Critical Incident Teams (CITs). For the first time ever, we were able to obtain a copy of the incident report produced by a CIT following a fatal incident. That report, on the August 3 crash, is littered with inconsistencies and inaccuracies, revealing how CITs operate in a way that could obfuscate the facts of the incident and protect the Border Patrol and its agents from accountability.

This report proves that ending CITs, as U.S. Customs and Border Protection (CBP) has announced it will do at the end of this fiscal year, is not enough. The agency must preserve all past records created by the CITs and initiate an independent review of cases impacted by CIT involvement.

Customs and Border Protection, Border Patrol’s parent agency, has a long history of allowing personnel to commit abuses with impunity. The agency’s disciplinary system has long failed to secure appropriate outcomes for misconduct. At the center of the internal system to review agents’ misconduct and issue disciplinary consequences is the CBP Office of Professional Responsibility (OPR). While OPR is tasked with conducting independent and impartial investigations, confidence in their results is wholly undermined by the involvement of Border Patrol’s CITs as investigators.

Documents uncovered by the Southern Border Communities Coalition last year revealed the existence of CITs, whose stated mission is the “mitigation of civil liability” for Border Patrol agents who might face lawsuits for misconduct. CITs report to Border Patrol leadership at the sector level and are tasked with conducting investigations of “any traffic collision” and any Border Patrol conduct that “results in death, serious bodily injury, significant property damage, or other exposure to significant civil liability.” These teams, which do not operate independently or impartially, have a clear bias and conflict of interest when they are charged with investigating incidents where agency personnel may prove to be liable for misconduct.

Credit: Sharon Chischilly

Troublingly, CITs are not designated by the U.S. Office of Personnel Management as criminal investigators and are not authorized to undertake any such investigations. All other federal agencies tasked with the investigation of potential criminal misconduct operate under explicit statutory authority. The U.S. Department of Justice’s own standards and guidelines for internal investigations further instruct that only allegations of minor misconduct, such as “discourtesy or rudeness,” should be relegated to the “unit level,” with oversight by the authorized internal affairs office.

James Tomsheck, head of CBP Internal Affairs from 2006 to 2014, stated in court documents published in 2021 that Border Patrol “had no authority to investigate, but it nonetheless consistently tried to assert investigative authority, and very frequently interfered with legitimate investigations,” adding that, “It was standard practice for Border Patrol to defend incidents in use of force, to always make it appear that it was justified.”

Little is known about CITs’ activities in cases involving deadly vehicle pursuits, beyond their apparent mandate within the agency to investigate such incidents. We were only able to confirm the direct involvement of an El Paso Sector’s CIT in investigating the recent crash that killed our client’s son by way of state-level public records requests — and the CIT report for this crash raises more questions than it answers.

The CIT report for the August 3 incident is the first of its kind obtained by advocates investigating CBP accountability. It provides us with unprecedented insight into how CIT investigators process a scene and the quality of the reporting that they generate, upon which OPR and other investigators rely.

The 162-page “Report of Investigation” indicates that the CIT played the central role in investigating the deadly crash. The CIT responded immediately, deploying investigators to the scene of the crash as well as the regional hospitals where victims had been transported. They photographed the crash, collected physical evidence, obtained video recordings and radio communications, interviewed and photographed the victims and Border Patrol agents involved, obtained reports from New Mexico State Police and the El Paso County Office of the Medical Examiner, and analyzed the evidence gathered from the scene. Their investigators reconstructed the crash and calculated the vehicle’s estimated speed.

Troublingly, the CIT report is riddled with errors, gaps, and inconsistencies that fundamentally undermine the investigation.

The report contains multiple inconsistent narratives that purport to describe when and why a Border Patrol agent initially began to follow the victims’ vehicle as it approached a Border Patrol checkpoint on NMSR 185, in rural southern New Mexico, and the CIT investigators make no effort in the report to identify or reconcile those discrepancies. The report’s summary says that the agent initially followed the vehicle “at a distance,” but when the vehicle passed the checkpoint the Border Patrol agent was, according to the CIT investigators’ summaries of video recordings, only 0.04 seconds behind the vehicle.

The report is unclear about when the Border Patrol agent following the vehicle activated his emergency equipment, which is a critical data point because CBP policies regarding vehicle pursuits only apply when an agent activates their lights and sirens. The agent reportedly did not activate his emergency equipment until after the vehicle bypassed the checkpoint, even though he was already following the vehicle extremely closely. Some of the vehicle’s passengers reportedly only saw emergency lights but did not hear sirens, but the investigators make no mention of whether recordings indicate if both lights and sirens were in fact activated. Both are mandatory under CBP’s vehicle pursuit policy.

Additionally, summaries of video recordings in the report indicate that Border Patrol agents at the checkpoint set out “spike strips” to attempt to stop the vehicle, but the E-STAR Incident Report states that no such devices were deployed. Again, the CIT investigators make no note of this discrepancy.

Credit: Sharon Chischilly

The circumstances of the crash are likewise unclear and the report does not provide any critical analysis of gaps and inconsistencies in the record. The last radio transmission by the agent who initiated the pursuit was at mile marker 27, but the crash did not occur until approximately mile marker 28.5. What transpired in the intervening mile and a half is not clearly accounted for. The CIT investigators’ summary of the incident states that the “pursuit” lasted about 3.2 miles, but the Border Patrol checkpoint is only approximately 2.5 miles from the crash site. The Significant Incident Report provides yet another inconsistency without explanation, stating that the pursuit only lasted about 1.5 miles. Remarkably, the latitudinal and longitudinal coordinates in the E-STAR Incident Report indicate that the pursuit started at the Border Patrol checkpoint and ended in a location off a country road that is 38.6 miles away. The same Incident Report states that the pursuit lasted 2 minutes and traversed 2.5 miles. The crash is attributed varyingly to high speeds on a curve in the road and to the vehicle “being overloaded.” The report alternates between characterizing the incident as a “collision” and as a rollover.

The report fails to scrutinize statements addressing how promptly the Border Patrol agents called for emergency medical services. The CIT investigators’ summary states that agents “immediately” requested EMS, but time-stamped radio logs in the Significant Incident Report indicate that EMS and medical airlifts were not requested until 10 minutes after the crash. The report’s transcription of radio communications, which lacks timestamps, indicates the urgency of the situation, with an agent on the scene stating that “most” of the vehicle’s occupants were “nonresponsive” and saying, “Give me as many ambulances uh that you can send.” Injuries ranged from brain damage and skull fractures to spinal fractures, hematomas, and lacerations. One passenger’s scalp was ripped off, multiple individuals had to be intubated, and two ultimately died.

The report makes clear that CIT investigators were among those first notified of the incident. Within 20 minutes of the crash, Border Patrol personnel reached out to the CIT, and a CIT investigator was en route to the scene from El Paso about half an hour later. Another CIT investigator went directly to the regional hospital to meet the air ambulance there. It is not apparent from the report when OPR was first notified. The report indicates that one of the CIT investigators met with an OPR Special Agent in Las Cruces on the day of the incident and OPR conducted some interviews. The report does not contain information indicating whether CIT investigators conducted interviews alongside OPR or otherwise.

The photographs that CIT investigators took of the incident are wholly inadequate in providing clarity about the crash. The photographs at the scene of the crash do not have a timestamp. The photographer was standing at a distance from the vehicles parked on the roadway, such that no license plate numbers or unit numbers are visible, other than for the vehicle that crashed. The report does not include any photographs of the places on the roadside where people who were ejected from the vehicle were located. Subsequent photographs taken of the vehicle’s driver and passengers, many still hospitalized, do not have names, locations, timestamps, or other identifying information.

The photographs of the Border Patrol vehicles involved in the pursuit are, inexplicably, not from the scene of the crash. They were taken at night, in a parking lot with poor lighting, and one vehicle is partially covered by the shadow of the photographer. The Border Patrol vehicles appear to have been recently cleaned. One of the Border Patrol vehicles was also photographed in what appears to be an automotive repair shop on a lift. These, too, lack any timestamps, although the report indicates that CIT investigators did not request photographs of the Border Patrol vehicles involved in the incident until August 9, six days after the crash. No explanation for the delay is provided and raises questions about any potential contact between the Border Patrol and victims’ vehicles.

Other inconsistencies and gaps in the report further call into question its reliability. The report does not indicate a chain of custody for any evidence collected but confirms that the photographs, diagrams, audio recordings, and video recordings of the incident are all archived by the El Paso CIT. A checklist of required notifications does not indicate that any photographs were collected, although the report itself contains photos taken at the scene. That same checklist does not indicate that any administrative claim forms for damages, injury or death were distributed, despite the multiple deaths and severe injuries incurred by the vehicle’s occupants. One part of the report says that the U.S. citizen driver was arrested at the time of the crash, but another part of the report says that he was not arrested or taken into custody. The report erroneously states that the driver died of his injuries on August 16, when in fact he was pronounced deceased on August 15.

The report does not provide detailed information about the custody determinations and immigration processing for the vehicle’s passengers, but it does indicate that at least two individuals were rapidly expelled from the United States under Title 42. Those expulsions took place only a few days after the incident, while the CIT investigation into the crash was ongoing. Notably, the CIT report includes a copy of a newspaper article regarding a letter sent by the ACLUs of Texas and New Mexico to CBP calling for an independent investigation into this incident. When victims and witnesses are expelled by CBP, the same agency conducting the investigation into an incident involving its own agents, it raises serious questions as to whether those individuals were denied access to legal protections or remedies that may have been available to them as a result of CBP misconduct.

Fundamentally, the CIT report of the August 3 vehicle pursuit and crash demonstrates the need for the agency to preserve all past records created by the CITs and initiate an independent review of cases impacted by CIT involvement. If this report is indicative of the standards at which CITs perform, OPR has jeopardized the independence of past investigations by relying on documentation of such dubious accuracy for serious incidents involving injuries and deaths.

Correction: A previous version of this article misstated the origin of the body cam footage from the crash scene. The footage is from a Doña Ana County Sheriff’s Office’s body camera.

On June 8, the State of Arizona is scheduled to execute Frank Atwood. Atwood, whose mother escaped Nazi persecution in Vienna, faces a perverse choice: suffer execution by lethal injection, or by hydrogen cyanide gas, a version of which was used by the Nazis to murder more than one million Jewish people at Auschwitz. Arizona last used cyanide gas in 1999 to execute Walter LaGrand. The state killed him violently, in all of our names, through a painfully slow asphyxiation that dragged on for an unfathomable 18 minutes.

Arizona’s commitment to this hideous execution method should shock the conscience of all Americans. It is particularly appalling to me as the board chair of the Jewish Community Relations Council (JCRC) of Greater Phoenix, and to my broader community. Earlier this year, we sued the state of Arizona to stop it from using this barbaric execution tool, infamous as the primary chemical deployed in Nazi death camps during the Holocaust.

This is the first time the JCRC has ever been a litigant. We hope it is the last.

The killing of another human being through a practice known to cause pain and suffering undermines everything the Jewish people stand for. Judaism introduced to the world the idea that every life is imbued with infinite value. Long before the constitutions of Arizona or the United States banned cruel and unusual penal punishments, the Jewish people opposed them, as they diminish the humanity and dignity of everyone involved, the punished and those who inflict the punishment alike.

For thousands of years, Jewish teachings have inveighed against practices common in the cities, nations, and empires where we lived: mutilation, burning at the stake, and throwing the condemned into a funeral pyre, to name only a few. Those same moral and ethical values require us to stand against a practice that we recognize, from our own tragic history, as highly likely to cause severe pain and suffering.

We are shocked that Arizona is proposing to use the very same chemical compound used by Nazis — often referred to as Zyklon B. Indeed, the resumption of this practice risks, for Holocaust survivors, reliving a trauma they have spent their lifetimes trying to forget.

As Jews, we aspire to make our society a light unto the nations. On this matter, Arizona stands out for its darkness. Nearly the entire civilized world — and most of the United States — has abandoned the use of Zyklon B. No state has used it to execute for more than 20 years. There is no question that killing a person with hydrogen cyanide is both cruel and unusual. In fact, Frank Atwood’s lawyers argue that forcing him to choose between cyanide gas and lethal injection violates his rights.

The state has raised many defenses to our suit, most of them hyper-technical, legal ones that, if adopted, would prevent any challenge to a state’s use of any method of execution, no matter how cruel or unusual. What is absent from the state’s arguments is any argument that killing another human being with hydrogen cyanide is anything but cruel or unusual.

As June 8 approaches, we continue to hold out hope that the state will confess explicitly what it concedes implicitly — that killing a person with Zyklon B is torture and thus that we, following every other state and the community of civilized nations, should bar its potential use. This hideous Nazi weapon of the Holocaust should be banished to history, not resurrected to wield violence in the name of Arizonans.

Throughout the U.S., judges, prosecutors, and parole boards are given algorithms to guide life-altering decisions about the liberty of the people before them, based mainly on perceived risks to “public safety.” At the same time, people accused and convicted of crimes are given little support. With underfunded public defense in most of these contexts, and no right to counsel in others (e.g., in parole decisions), the system is stacked against them. We wanted to find out what would happen if we flipped the script and used algorithms to benefit people entangled in the legal system, rather than those who wield power against them.

In a recent peer-reviewed study, the ACLU and collaborators at Carnegie Mellon and the University of Pennsylvania asked a simple question: Can one predict the risk of the criminal justice system to the people accused by it, instead of the risks posed by the people themselves?

The answer seems to be yes, and the process of creating a tool like this helps lay bare broader issues in the logic of existing risk assessment tools. While traditional risk assessment tools consider risks to the public such as the likelihood of reoffending, the criminal legal system itself poses a host of risks to the people ensnared in it, many of which extend to their families and communities and have long-term repercussions. These include being denied pretrial release, receiving a sentence disproportionately lengthy for the given conviction, being wrongfully convicted, being saddled with a record that makes it impossible to obtain housing or employment, and more.

The prototype risk assessment instrument we created predicts a person accused of a federal crime’s risk of receiving an especially lengthy sentence based on factors that should be legally irrelevant in determining the sentence length, like the accused person’s race or the party of the president who appointed the judge. The instrument performs comparably to other risk assessment instruments used in criminal justice settings, and the predictive accuracy matches or exceeds that of many tools deployed across the country. Still, that doesn’t mean this tool, or any of the existing tools in use, are necessarily good or make peoples’ lives better — just that it meets existing validation standards.

We chose to model the risk of lengthy sentences among people prosecuted federally for several reasons. The most practical is simply that the data existed. In many criminal justice settings, the information advocates and researchers need most is not collected, or collected poorly, such as the details of plea bargains that make up roughly 95 percent of convictions. Lengthy sentences are also a particularly pernicious problem in the U.S. — far worse than in most other democratic nations. Norway caps sentences for most crimes at 21 years, Portugal at 25. Excessively long sentences are noted as a primary cause of mass incarceration, and substantial evidence suggests longer sentences can actually have a negligible or negative impact on the alleged goal of rehabilitating people sent to prison. Evidence also suggests that lengthy sentences do not prevent future crime. Finally, non-legally relevant factors impacting sentencing decisions are well documented: The U.S. Sentencing Commission concluded that similarly situated Black men received, on average, a sentence that is 19.1 percent longer than their white counterparts.

The process of creating this tool illuminated the choices embedded in the creation of other tools that are frequently used in parole and pretrial settings. For instance, we set multiple thresholds, such as the definition of an excessively long sentence and the probability boundary of when we considered someone especially likely to get one of those sentences. These are policy choices, much like the choices the Bureau of Prisons made to shift the thresholds on its own tool to reduce the number of medically vulnerable incarcerated people eligible for release during the pandemic, or ICE, which used a risk assessment tool to prevent the pretrial release of immigrants. In short, any time a tool is made, it is likely to center the viewpoint of its creators and create a new policy, which should make us wary of these tools and their applications.

The models we built — unlike existing risk assessment instruments — are built to aid public defenders and accused persons, instead of prosecutors and judges. If we were to provide public defenders with the risk that a defendant will get a severe sentence or just how far they are from other similar cases, perhaps it could help them to make informed decisions when navigating sentencing proceedings and plea bargaining.

There are other possible applications as well. The recently enacted First Step Act enabled incarcerated people, for the first time in history, to directly file motions with the court to seek a sentence reduction where “extraordinary and compelling” circumstances warrant a reduction. Since then, federal district courts across the country have granted thousands of such motions where the personal history of the defendant, the underlying offense, the original sentence, the disparity created by any changes in the law, and other factors warrant such a reduction. With our models, defendants could point to how far their sentence deviates from what the model would predict based on the characteristics of their case, including the ability to look at how their case may be resolved today versus when they were originally sentenced. Petitioners could also point to the legally irrelevant factors that may have influenced their sentencing.

The ACLU and our coalition partners have been pushing the Biden administration to use the presidential power of clemency. Though he recently commuted the sentences of 75 people, he is failing to use it systematically. Clemency is mostly used only for specific high-profile cases, and tends to provide relief to people sentenced under now-defunct criminal laws or charges now deemed overly punitive (e.g., non-violent drug offenses). These categories exclude many federally incarcerated people from even a slim possibility of mercy. We built the model in such a way that it can indicate unreasonably long sentences, even for people left out of many criminal justice reforms and clemency actions, such as those who have been sentenced for violent crimes.

We expect that there will be objections to the use of this model on the basis of technical limitations, the acceptability of using an algorithm for such a high stakes decision, or the subjectivity of the choices that were made. But those who might raise such concerns would do well to apply them equally to the tools currently used throughout the system.