Correction: This post stated that the FBI raided Paul Manafort’s house without knocking. In court filings the FBI has stated that that is not the case.

The news Wednesday that the FBI waged a no-knock raid against former Trump campaign manager Paul Manafort has generated discussion about these raids. The consensus in this case seems to be that the early-morning no-knock raid is a good indication that special counsel Robert S. Mueller III has strong evidence of criminal wrongdoing, and that he feared Manafort might have destroyed some of that evidence if he had not sent FBI agents to retrieve it.

Unfortunately, the discussions have included broad generalizations that, although perhaps technically accurate, give a misleading impression of how no-knock raids are typically utilized. This CNN segment with legal analyst Jeffrey Toobin is an example:

Around the 4:30 mark, Toobin says, “Magistrate judges don’t give authorizations for searches of people’s homes lightly. I mean, this is a big deal.”

Toobin’s comment is technically true, if you analyze it in the context of an investigation into someone such as Manafort — a wealthy politician who once held a powerful position. Those sorts of people aren’t typically subjected to early-morning no-knock raids. But lots of other people are. And these sorts of comments not only erase their experience, but they also give a false impression of how and against whom these tactics are used.

They’re used, for example, against people such as the woman and her 3-year-old daughter who were held at gunpoint after FBI agents entered her home by taking a chainsaw to her door. They were looking for a drug dealer who lived in the building. They had the wrong apartment. They’re used against families such as the three children in New Mexico injured by flash grenades when FBI agents conducted a no-knock raid on their father, who was suspected of being a street-level drug dealer. Or the Quincy, Mass., couple raided last year by FBI agents and local police who took a battering ram to their door and ransacked their home. The agents found nothing incriminating and left a search warrant that included only the home’s address.

Sen. Richard Blumenthal (D-Conn.) also weighed in on the Manafort raid, with a series of tweets.

These tweets are technically true, provided you limit your pool of “targets” to people like Manafort. But again, that paints a misleading picture of how these tactics are used. They aren’t typically used against people like Manafort, and for most people against whom they are used, the target isn’t given a chance to cooperate. As for judges, I’ll let former federal prosecutor Ken White at Popehat elaborate:

In the federal system, federal agents present search warrant applications to United States Magistrate Judges for review. Magistrate Judges aren’t nominated by the President and confirmed by Congress like United States District Court judges — they are appointed by other federal judges for set terms, and have a reduced level of authority and responsibility. They do a lot of the unglamorous day-to-day work of the federal judiciary.
The magistrate judge reviews the search warrant application and, almost always, signs the warrant approving it . . .  I think that magistrates can be a little rubber-stampy at times. But probable cause is a pretty low bar.

Emphasis mine. “Almost always.” White adds that federal courts are generally better than state courts at scrutinizing warrants, and that federal prosecutors are typically better at reviewing warrants than district attorneys. But he adds that the most scrutinized warrants are in complex, white-collar cases. If Mueller does bring a case against Manafort, it will undoubtedly be complex. So again, in that context, comments such as those from Toobin and Blumenthal aren’t technically wrong, but they give the impression that all federal warrants for no-knock raids are carefully scrutinized and waged against only people for whom there is a significant amount of incriminating evidence. For most such warrants, that just isn’t true.

Move from the FBI to the Drug Enforcement Administration, and the problem only grows. From the mid-1990s through about 2010, the DEA conducted hundreds of full-on SWAT raids of medical marijuana clinics in states that have legalized the drug for medicinal purposes. These businesses were technically breaking federal law. But they weren’t heavily armed drug dealers. They were businesses licensed by the states in which they were operating. Through the 2000s, the DEA also sent SWAT teams to raid the offices of doctors whom drug cops suspected of over-prescribing opioid painkillers. Whatever you make of the current opioid crisis, these weren’t well-armed kingpins. They were doctors operating in the open.

But the more typical targets of a no-knock DEA raid aren’t doctors, medical marijuana clinics or kingpins. They’re people accused of low-level drug crimes. The evidence can be flimsy, or indiscriminate. The DEA will sometimes join state and local police to conduct multiple, even dozens of raids in a single city or neighborhood. Those raids might produce enough illicit drugs, cash or guns to make them seem worthwhile, but the sheer size of the operations can rope in innocent people, too. In one such case we covered here at The Watch, an elderly Detroit woman and her daughter were raided and roughed up by masked DEA agents during a massive sweep of the city’s Eight Mile neighborhood. The women never got their day in court because the DEA never revealed the identities of the masked agents.

A quick list of other incidents that come to mind:

  • DEA agents shot a New Hampshire woman through the arm as she was picking up her infant grandchild during a botched raid looking for diverted oxycodone. They had the wrong residence.
  • In 2011, DEA and San Francisco police raided a UC-Hastings law professor under the mistaken belief that he was growing marijuana.
  • In 2013, DEA agents and local officials raided an Illinois woman’s home based on evidence that she had shopped at a hydroponic store and that a substance in her trash came up positive for marijuana on a drug field test. Regular readers will know that such tests have notoriously high error rates.
  • In 2012, the DEA defended in federal court the practice of agents pointing their guns at children’s heads during drug raids. In that particular case, the agents had raided the wrong home, due to the sloppy recording of a license plate. Presumably, a federal magistrate signed off on the warrant.

Finally, most federal drug raids are also carried out not by a well-trained DEA or FBI SWAT team, but by multi-jurisdictional narcotics task forces, or local cops participating in the Justice Department’s adoption program. The training these task forces undergo is inconsistent at best, and they’ve been the subject of too many scandals and controversies to count.

Unfortunately, there isn’t a lot of reliable data on just how often judges sign off on search warrants. I don’t know of any such studies at the federal level (if readers are aware of one, please feel free to leave a link in the comments). There have been a few studies at the state and local levels, but most are pretty old. Still, they do suggest what current and former police officers, defense attorneys and more than a few prosecutors have told me over the years: In most jurisdictions, it’s incredibly easy to get a warrant. Most judges exercise little to no scrutiny at all, and the few who do can be circumvented. You just take your affidavit to another judge.

I looked at a couple of those studies in my 2013 book, Rise of the Warrior Cop. The first and most comprehensive was about 30 years ago.

A survey of the warrant process in seven US cities by the National Center for State Courts found that magistrates spend an average of two minutes and forty-eight seconds reviewing warrant affidavits before (almost always) approving the warrant. The study also found evidence that police “magistrate shop”—they seek out magistrates with a reputation for approving warrants quickly and with no hassles, and avoid those who ask questions. In one city, a single magistrate approved 54 percent of the search warrants over the period the study was conducted. The most popular magistrate in another city had rejected just one search warrant in fifteen years on the bench. Not surprisingly, “most police officers interviewed could not remember having a search warrant turned down.”

And from another in study in Denver in the late 1990s.

After the botched raid that ended the life of Ismael Mena in 1999, the Denver Post looked into how judges in the Mile High City handled requests for no-knock warrants. Again, the results were unsettling. Over a twelve-month period, police in Denver requested 163 no-knock warrants. The city’s judges granted 158 of them. Defense attorneys told the paper they were surprised to learn that the judges had rejected even five. Perhaps Denver police had come to the judges with more than adequate probable cause? Perhaps. But the paper also found that, astonishingly, many of the city’s judges would sign off on no-knock warrants even though the police hadn’t requested one. In fact, about 10 percent of the no-knock warrants were changed from knock-and-announce warrants merely by the judge’s signature—the police hadn’t presented any additional information establishing exi- gent circumstances. The paper also found that in eight of ten raids over that period, police assertions in affidavits that they would find weapons during the search turned out to be wrong. In only seven of the 163 no-knock affidavits did police present any evidence that the suspect had been seen with a gun. Of those seven raids, just two turned up an actual weapon. The Denver Police Department requires that all no-knock raids be preapproved by the DA’s office. In about one-third of the raids, that never happened. And nearly all the no- knock warrants were granted on little more than a police officer’s as- sertion that a confidential informant had told him the suspect was armed or likely to dispose of drug evidence, with no additional corroborating information.
When confronted with the results of the investigation, the presiding judge over Denver’s criminal court system wasn’t particularly reassuring. “We are not fact gatherers,” Judge Robert Patterson said. “It’s pretty formulaic how it’s done.” On how a judge could possibly inadvertently approve a no-knock warrant when the police hadn’t even asked for one, Patterson said, “If you sign your name 100 times, you can look away and sign in the wrong place. We read a lot of documents. We may, just like anyone else, sign something and realize later that it’s the wrong place or the wrong thing. Is it wrong not to be paying attention? No. It’s just that we’re doing things over and over again.”

After the death of 57-year-old Alberta Spruill during a botched NYPD drug raid in 2003, Newsday tried to do a similar survey, but found it impossible — New York City’s courts didn’t even keep no-knock warrants on file after the raids were carried out. (In my book, I also point out that when the FBI was on the trial of the notorious gangster Whitey Bulger — a man with bona fide record of violence —  they did not use a SWAT team. Instead, they lured him out with a ruse.)

Sometimes, officials act without even a warrant. In 2014, DEA agents raided a shop in Alpine, Tex., and an adjacent apartment, apparently in a search of synthetic marijuana. They bashed in doors, swore at the occupants inside and allegedly smashed a woman in the neck with the butt of a rifle. They obtained a search warrant for the apartment only after the raid. Incredibly, a federal magistrate — those who are allegedly so careful and cautious — then ordered one occupant to recant her story about the rifle butt as a condition of her bail.

Last summer, the FBI conducted no-knock raids on the homes of people who protested at the Republican National Convention based on almost no evidence at all. One man who was raided was suspected, at worst, of throwing urine-filled balloons during protests. I’d imagine it isn’t pleasant to get hit with a urine-filled balloon. But I’m not sure it merits a no-knock SWAT raid. An FBI agent then told a local TV station, “It was a protective sweep, that’s all it was. No search was conducted. There’s no warrant that’s needed for that.” There have been similar raids in recent years before or after major-party conventions. FBI and local police conducted peremptory raids of activists’ homes before the 2008 Republican National Convention in Minneapolis. In the end, authorities dropped the charges against 442 of the 672 people arrested.

I realize that Blumenthal and commentators such as Toobin are merely making the point that in cases such as Manafort’s, the use of a no-knock warrant is unusual and suggests that Mueller’s investigation is getting serious. But these are incredibly volatile and violent tactics. They have a very thin margin of error. You could make a good argument that the tactics themselves are a form of punishment, which is particularly troubling, given that a search warrant is served merely to collect evidence of a possible crime — no one has yet even been charged, much less convicted. In making the point about Manafort, they’ve implied that these tactics are used much more sparingly, under much more scrutiny and oversight, and only with much more incriminating evidence than is typical. And that just isn’t true.

The frustrating thing about high-profile investigations and court cases is that by their very nature, they’re atypical. It’s a bit like the laws of particle physics: The mere act of watching the system in action causes it to change. And yet these cases tend to be the primary means by which much of the public learns about courts, cops and prosecutors. And we’re all a little dumber for it.