Snitching

Criminal Informant Law, Policy, and Research

Incentives & Payments

Snitch deals in Kilmar Abrego Garcia case

by

Last week, a federal judge refused to detain Kilmar Abrego García, a Maryland resident who was wrongly deported to El Salvador and now faces criminal charges, based on what she deemed to be unreliable testimony from three different government informants. Judge Barbara Holmes described the lead cooperator as “a two-time, previously-deported felon, and acknowledged ringleader of a human smuggling operation, who has now obtained for himself an early release from federal prison and delay of a sixth deportation by providing information to the government.” Court order here. A Washington Post article — Star witness against Kilmar Abrego García was due to be deported. Now he’s being freed — describes the informant as having a long criminal history, including gun violence, in contrast with Abrego Garcia’s lack of any criminal record at all. The other two informants were family members of the main informant; in return for their testimony one is seeking release from custody (on unrelated federal criminal charges), and both are seeking deferred action on their pending deportations.

Many have commented on the counterintuitive fact that the government’s star witness has already been convicted of worse crimes than Abrego Garcia is accused of. But the informant market often flips the conventional rules of crime and punishment: letting the government use the proverbial “big fish” as an informant to catch smaller, less culpable fish. The practice is prevalent in, although not limited to, drug enforcement. This New York narcotics unit, for example, helped their informant drug dealers stay in business in exchange for turning in their low-level client drug users. An ACLU report described the same dynamic in New Jersey, under which “more culpable leaders of drug networks — the ‘kingpins’ — may get less severe punishment than underlings who play a lesser role in the drug trafficking operation.”

Perhaps the highest profile critique of this practice issued in 1966 from none other than U.S. Supreme Court Chief Justice Earl Warren, himself a former prosecutor. In its efforts to prosecute Teamsters Union President Jimmy Hoffa for a misdemeanor, the FBI dug up an incarcerated informant who was facing charges on manslaughter, kidnapping, embezzlement, assault and perjury. The informant was freed and all charges eventually dropped in exchange for his help prosecuting Hoffa. The Supreme Court validated the deal in the case Hoffa v. United States, but Chief Justice Warren dissented, complaining that “the Government reache[d] into the jailhouse to employ a man who was himself facing indictments far more serious … than the one confronting the man against whom he offered to inform.” He worried that “this type of informer … evidence[s] a serious potential for undermining the integrity of the truth-finding process in the federal courts.” Chief Justice Warren lost that argument, which is why the government is legally permitted to use, reward, and even forgive serious criminal offenders when they serve as informants against less blameworthy or dangerous people.

Refusing to snitch

by

The Marshall Project has published this remarkable story about George Hall who refused to serve as a jailhouse informant against David Wood. David Wood is on death row and is scheduled to be executed next month in Texas.

Hall, Wood, and two other men were all incarcerated together back in 1990. According to Hall, Hall and the two other men were offered deals by Texas law enforcement to testify falsely against Wood. Hall refused; the other men agreed and helped convict Wood. From the story:

“As Hall tells it, several El Paso detectives took the three men, without handcuffs, to a hamburger joint and a police station, seating them in a room with photographs of the victims, a large coffee pot, cigarettes and snacks. They handed the men case files with crime scene photos and interview notes with other witnesses, Hall said. “David Wood is our suspect,” he recalls the detectives saying. “It’d be best if you tell us something, because we can’t let this guy walk.” Plus, there was reward money for people who helped them.”

Stories like Hall’s are rare for a number of reasons. The government only rewards inculpatory evidence — evidence that builds the state’s cases and makes defendants look guilty — not evidence that might help exonerate a defendant. Conversely, defendants can’t offer leniency at all, and offering money or a reward looks like witness tampering. Which means that all the incentives run in one direction, towards snitching for the prosecution and away from contradicting the government’s story. When someone like Hall refuses to snitch, moreover, they will not be called as a witness by the government, which means the defense might never learn about them. And coming forward like Hall did can be risky for people facing their own criminal cases: they run the risk of law enforcement disfavor or even retaliation. (This reality sits in considerable tension with the Second Circuit’s holding that prisoners have a First Amendment right against being forced to act as an informant.)

For all these kinds of reasons, when people refuse to snitch, we are unlikely to learn about it, which makes this particular story even more revealing. Story here: He Refused to Become a Jailhouse Snitch. Can He Stop David Wood’s Execution?

U.S. Supreme Court decides case on expert admissibility

by

The Supreme Court decided a case on Thursday that is not about informants but that could have implications down the road for informant expert testimony. In Diaz v. United States (June 20, 2024), the Court held that a federal agent could testify as an expert that “most drug couriers” know they are transporting drugs. The defendant, Delilah Guadalupe Diaz, claimed that she did not know that drugs were in the car that she was driving. The Court reasoned that the agent was not testifying explicitly about what Diaz knew or did not know — that would have been prohibited by the Federal Rules of Evidence, Rule 704(b)— but only what “most people” in that group know. The majority rejected the dissent’s argument that the expert “functionally” stated an opinion about Diaz’s state of mind, because “[t]hat argument mistakenly con­flates an opinion about most couriers with one about all cou­riers.”  Rather, the Court held that “an expert’s conclusion that most people in a group have a particular mental state” is not an opinion about a particular individual in that group.

Diaz potentially enhances the admissibility of defense expertise regarding informants, specifically about whether informants know that they will receive benefits in exchange for the information they provide. As I and many other scholars have pointed out, most informants know that they will be rewarded for incriminating other people, even if the government has not expressly or formally promised them anything up front.  Nevertheless, the government often maintains that informants come forward for no reward, and informants will often tell the jury that they do not expect any reward.  In a similar vein, prosecutors often reassure juries that informants are reliable because lying informants will fear prosecution for perjury, but most informants (as well as lawyers and judges) know that such prosecutions are extremely rare. Expert testimony on what “most informants know” could thus be helpful to juries in deciding whether informants should be believed.

Courts are typically more willing to allow government experts than defense experts in criminal cases.  Indeed, the National Academies of Sciences issued a report in 2009 worrying that “trial judges rarely exclude or restrict expert testimony offered by prosecutors.” But Diaz cuts both ways. In her concurrence, Justice Ketanji Brown Jackson emphasizes that the decision is “party agnostic” and that “[b]oth the Government and the defense are permitted [] to elicit expert testimony ‘on the likelihood’ that a defendant had a particular mental state.”

Here is a link to an explainer about informant expert testimony that I wrote for The Appeal.  And here is a link to my colleague Professor Noah Feldman’s op-ed arguing more generally that Diaz was wrongly decided, and that it impermissibly waters down the prosecution’s obligation to prove its case beyond a reasonable doubt “because [the decision] invites the jury to conflate abstract statistical probabilities with the specific circumstances of the individual case.”

Deep dive into informant practices in Cincinnati

by

According to a recently released year-long investigation by the Cincinnati Enquirer, local police and prosecutors quietly used informants in dozens of homicide cases, many of which later fell apart. See Enquirer investigation: Cincinnati homicide cases unravel after deals with informants. Here’s just one example from the article:

Before his violent death in 2012 – prosecutors say he was shot for “being a snitch” – [Quincy] Jones became a prolific police informant, joining a network of informants and cooperating witnesses who for years helped Cincinnati law enforcement close homicide cases. The informants sometimes testified in multiple cases and, like Jones, worked with detectives and prosecutors who vouched for their reliability. But an Enquirer investigation found several of those cases later unraveled, raising the possibility that unreliable informants helped send innocent people to prison and allowed others to get away with murder.

Jones began cooperating in 2008 when he was charged with multiple murders. He fled to Seattle; when he was brought back to Cincinnati he met with Police Detective John Horn and offered to cooperate. As the article describes it, “On the day Jones signed the deal, prosecutors dropped one of the murder charges against him and reduced the other to involuntary manslaughter. [] Judge Beth Myers then sentenced him to four years in prison, far less than the 20 years to life he would have faced with a murder conviction.” Two years later Jones cut another deal which permitted him to walk free in 2010.

The article also reveals the kinds of sleight-of-hand used to conceal informant deals from defendants, courts, and the public:

Because [Jones’s] deal was confidential, defense attorneys, judges and juries in future cases wouldn’t know its terms. They also wouldn’t know that if Jones went back on the deal, all the original charges, including a possible life sentence, could be reinstated. Of the 12 homicide cases Jones cooperated on, The Enquirer found, he testified in court about at least five. Each time, Jones said he’d been promised nothing in exchange for his testimony. 

According to his written plea deal, that was true. Because the agreement didn’t identify specific cases, Jones could say his testimony in those cases wasn’t connected to his plea deal. 

[Detective] Horn, who is now retired, said that’s a common arrangement. “You haven’t done anything for me until, you know, you do something for me,” Horn said. “There’s never been any promises made.” 

Paying jailhouse snitches with conjugal visits

by

A Miami prosecutor has resigned from a death penalty case after Judge Andrea Ricker Wolfson found evidence of various types of “witness testimony manipulation” and “severe recklessness” by prosecutors stretching back decades. AP story here: Veteran Miami prosecutor quits after judge’s rebuke over conjugal visits for jailhouse informants, and ABA Journal story here.

As I have written many times, there are almost no limits on the kinds of rewards that the government can offer an informant in exchange for information. Typically it is the failure to disclose the benefits, not the propriety of offering the benefits themselves, that constitutes prosecutorial misconduct.