A couple weeks ago, I posted about a case in Kansas in which a couple was wrongly raided by a police tactical team. Robert and Addie Harte and their two children were held in their home by armed officers for over two hours as the officers searched the house for marijuana. They found no drugs. After spending $25,000 to get a judge to order the Johnson County Sheriff’s Department to turn over documents related to the search and investigation leading up to it, the Hartes discovered that Robert Harte, along with hundreds of other people, became a suspect when Missouri State Highway Patrol Sgt. Jim Wingo saw Harte, his son and his daughter emerge in August 2011 from a hydroponic gardening store and wrote down the Hartes’ license plate number.

About eight months later, Johnson County Sheriff’s Department deputies Mark Burns and Edward Blake conducted trash pulls at the Harte’s home. According to police, drug field tests on some wet plant material found in two of the trash pulls tested positive for marijuana. Based on those tests, the department raided the Hartes on April 20, 2012. The Hartes later learned that more conclusive crime lab tests on the “plant material” revealed it to be the looseleaf tea enjoyed by Addie Harte.

My post generated quite a bit of attention and outrage. Subsequently, Orin Kerr of the Volokh Conspiracy blog, hosted here at the Washington Post, took issue with the headline I wrote for that post, “Federal judge: Drinking tea, shopping at a gardening store is probable cause for a SWAT raid on your home.” Kerr chastised my characterization of the ruling. According to Kerr, the ruling by U.S. District Court Judge John W. Lungstrum ruled only on “a pretty technical Fourth Amendment issue that seems pretty far removed from what you would imagine from reading Balko’s post — and especially his headline.”

I don’t agree. It’s true that much of Lungstrum’s ruling focused on the field tests. And it’s true that Lungstrum evaluated the probable cause issue for the purpose of determining if the Hartes’ rights were violated, and if the police officers and/or Johnson County were liable for violating them. It’s true that I didn’t mention the field tests in my headline (though I mention them extensively in my post). It’s also true that my headline was intended to be provocative, and was written from the perspective of the Hartes. But none of that means the headline or the post were inaccurate or misleading. I also linked to Lungstrum’s decision in the post.

From the Hartes’ perspective, the headline covers precisely what happened to them. They were raided by a tactical team. The only things they did to cause that raid were shop at a hydroponic gardening store and drink loose-leaf tea. Those two actions are why the police began to investigate them (in a half-hearted sort of way), conducted field tests on the tea for which the officers had little training, and procured a search warrant. Those two actions are why a local judge signed off on the search warrant. It’s why a raid team was sent to the Harte home. And after the fact, it’s why a federal judge determined that the police had probable cause for the search, that the use of a tactical team was reasonable under the Fourth Amendment, and that the police didn’t violate the rights of the Hartes or their children.

I didn’t include information about the field tests in the headline because it would have made for a cumbersome headline, because they’re discussed in the post itself, and because I don’t think they diminish or substantially qualify what I stated in the headline.  Yes, someone viewing this case through the perspective of the police could have written a headline like, “Federal judge: Police not liable for relying on faulty field tests.” That would have been quite a bit different from my headline. And like my headline, it would also have been accurate. I chose to emphasize what happened from the Hartes’ perspective, because I find what happened to them to be outrageous, and that’s what I think is important about this story. My point was to emphasize just how little oversight there is when it comes to these raids, how few protections are afforded to potentially innocent people on the receiving end of them, and how when things go wrong the victims of mistaken raids have little recourse, and the cops who wrongly raided them face little accountability.

The disagreement here also reflects how differently legal academics sometimes see these cases than the rest of us. In his initial post Kerr, a self-described “Fourth Amendment geek,” was so focused on his very technical legal issue that he neglected to even acknowledge what happened to the Hartes, or the utter incompetence of the Johnson County Sheriff’s Department. After some criticism from his commenters, he later conceded both points in an update. But until that update, you could be forgiven for thinking that Kerr believed the only real outrage in this story was my headline, and the only real victim was Judge Lungstrum. It was a classic example of missing the forest for the trees, only I’m not at all convinced he was right about the trees.

Exchanges like these are common in online commentary about these issues. Someone will write about what they believe to be an outrageous example of police or prosecutor misconduct, only to be corrected by a legal scholar, who will then calmly and somewhat condescendingly explain how the police or prosecutor in question was acting entirely within the law, complete with strings of citations to court opinions. But that’s almost always beside the point.

To be clear, I don’t think Judge Lunstrum’s ruling was wrong on the law, nor did I state or imply as much in the post. (Although as is often true in these cases, and for the reasons I’ll lay out later in this post, I think Lungstrum could also have written an entirely justifiable opinion that allowed that Hartes to proceed with their lawsuit.) It is the law itself that is the problem.

The Fourth Amendment was a response to the invasive, sometimes violent searches of colonists’ homes by British soldiers armed with general warrants to look for untaxed goods. It requires probable cause for searches, requires that searches be reasonable, and requires specificity of the people to be searched and for what the government agents will be searching. Given the history of and debate around the ratification of the Fourth Amendment, it’s difficult to see how the men who drafted and ratified it could ever have envisioned a day in which soldiers (and that’s how they’d see today’s tactical teams) could force their way into a private home, then hold its occupants under armed guard and rifle through their belongings, all because they had some reason to believe that they may have been growing a forbidden plant. As the defense attorney Scott Greenfield writes in response to Kerr:

The conflict here derives from an inherent difference in perspective, both in how much of a slave one is to the court’s ruling, and whether one views the SWAT raid, the search warrant, the Fourth Amendment, from the point of view of innocent homeowners or police protecting society from the scourge of demon marijuana.  Judge Lungstrum’s ruling addressed the liability of the police for a raid that turned up nothing, but was performed with the benefit of a search warrant, which was obtained with the benefit of two positive drug tests. What more can one expect of the police?
The answer may well be found in whether one views the Fourth Amendment as a roadmap for the police in how to search and seize, or as a protection of individual privacy for innocent citizens from wrongful searches and seizures.

As I mentioned in the original post, I’ve been following this case since shortly after the raid took place. I’ve been to Kansas City, where I interviewed the Hartes, their attorney, and several other people in the area who claimed to have been wrongly raided by tactical teams. There’s a lot more to this case than what was in my original post, Lungstrum’s ruling, or Kerr’s response. I think it’s worth delving into exactly what happened in the months leading up to this raid, because while Kerr is certainly right that Judge Lunstrum’s opinion was well within the parameters of current Fourth Amendment case law, it’s also an illustration of just how much Fourth Amendment case law has failed to protect our rights. All the assertions that follow are either from my own reporting, from Lungstrum’s opinion, or from the Harte court filing linked in Kerr’s post. (I’ve also previously reached out to the Johnson County Sheriff’s Department to request an interview. They never responded to those requests.)

The hydroponic gardening store

Let’s start with the hydroponic gardening store. Kerr writes:

As the affidavit in support of the warrant explains, officers placed a suspect under surveillance after the suspect made a purchase at a particular hydroponics store that officers believed was largely used by those growing marijuana.

Perhaps it’s unintentional, but Kerr’s wording here seems to suggest that Harte was already a suspect at the time he made his purchases at the hydroponics store. It’s important to emphasize that it was Harte’s completely legal, completely innocuous purchase of gardening supplies to grow tomatoes and vegetables with his son that made him a suspect in the first place. Prior to that, he was unknown to the police. He had done nothing else to arouse suspicion. This is a major source of a lot of the outrage about this case. If Harte had never gone to that particular store, or had bought the supplies at Walmart, there’s no trash pull, no alleged false positives on the field tests and no raid.

Moreover, Harte was with his children, then ages 6 and 12, when he went to the store. I suppose that there are some marijuana growers out there who might take their young children with them when they buy growing supplies. But maybe, just maybe, the presence of Harte’s kids suggests he wasn’t the sort of criminal worthy of a tactical team.

But that isn’t how this investigation was conducted. Everyone observed by the officer staking out this store had his or her license plate recorded and turned over to local police. There was no effort to distinguish innocent customers from criminal suspects. (If there was, it’s hard to see what Harte did to put himself among the latter.) The officer was in the parking lot, so there was no effort to even make distinctions based on what the customers had purchased. Even here, Harte was seen carrying a “small bag,” which presumably wouldn’t be large enough hold equipment for a major operation. His mere presence at the store made him a target of investigation.

One more point here. Dep. Burns notes in his affidavit that he knows “from personal past experience,” this hydroponic gardening store sells equipment that customers use to grow marijuana. I’m sure that’s true. But so do Walmart and Home Depot. As we’ll see later in this post, of the hundreds of names Sgt. Wingo collected, the entire “Operation Constant Gardener” 2012 operation turned up less than a half dozen marijuana “grows.” And all but a couple of those were either inactive, or had less than 10 plants. You’d probably find just as high a proportion of marijuana offenders by taking down license plates at Bonnaroo, or at a Bernie Sanders rally.

The field tests

The field tests comprise the bulk of Lunstrum’s ruling and of Kerr’s analysis. So let’s go back to Kerr:

A few months later, officers searched the trash from the suspect’s house and found some plant material that in the lead investigator’s view “appeared to be wet marijuana plant material (leaves and stems).”

This isn’t the whole story. First, the trash pulls came nearly eight months later. This is significant because the grow cycle for marijuana is 60-90 days (a fact about which the officers who conducted the investigation were also ignorant). The eight months is significant also because it would only have taken the crime lab a few days to return conclusive results. They could wait eight months after Robert Harte visited the gardening store to start investigating him, but they couldn’t wait a few more days for the crime lab results, or to do some corroborating investigation.

Second, the officers actually conducted an initial trash pull which produced the same sort of “plant material” (which we now know was tea leaves) that they found in two subsequent pulls. They deemed the material found in the first trash pull to be “innocent.” It was the plant material they found in the second and third pulls that they believed to be marijuana (due to their “training experience”), even though it was tea every time. On all three occasions, they found the plant material in a bag of trash that had clearly come from the kitchen. It was surrounded by other items you might find in a typical bag of trash from a kitchen. There were no rolling papers. No burnt pot, roaches, or pot ashes. There was no other indication of criminal activity — only wet globs of plant matter.

Back to Kerr:

Based on the officer’s “law enforcement training and experience,” it appeared as though it had been saturated through a liquid to extract THC, the active ingredient in marijuana.

This too requires more context. In his deposition, Burns admitted that he had received no formal training and had no personal experience in investigating THC extraction methods. When asked about the “law enforcement training and experience” he referred to in the search warrant affidavit, he admitted that this training consisted entirely of watching a few videos on YouTube. He said he watched these videos on his own time, and never reported or documented his viewing them. So the Hartes were raided by a SWAT team because some wet tea leaves the cops found in their trash literally looked like something a cop once saw on the Internet.

Marijuana also has a very unique, pungent, and noticeable odor. Addie Harte has told me in previous interviews that the variety she drinks smells nothing like pot. Few teas do.  The tea smelled like tea, along with some floral and fruity scents. It also looked nothing like pot, which was noted both by the Johnson County crime lab analyst who later tested it and by the analyst hired by the Hartes. (The latter pointed out that the tea also included visual bits of fruit and flowers.) Dep. Burns and Sgt. Tom Reddin (the supervisor for the entire investigation) were later asked in a deposition if they ever attempted to smell the wet tea leaves they thought were marijuana. Reddin said he hadn’t, adding, “I don’t smell items that come out of people’s trash.”

Burns claimed he didn’t bother to smell the marijuana because in his experience “wet marijuana” doesn’t retain the distinctive smell of “raw marijuana.” He said the wet stuff smells “earthier.” Presumably, by “wet marijuana,” he was referring to pot that has been processed to extract THC. But it’s hard to fathom how he’d know how pot leftover from that process would smell, given that his training here consisted only of watching some YouTube videos.

Johnson County District Court Judge Peter Ruddick — the judge who signed the search warrant — later said that had he been told that the substance smelled nothing like marijuana, he’d have been less likely to authorize the search.

Oddly, though he was personally above smelling the plant material, Reddin (along with Burns) did later claim to have unrolled tiny wet leaves from the glob of tea. In depositions, he claimed some of the tiny leaves were serrated, which he was one reason why he believed the glob to be pot. Apparently neither officer, despite all their “training and experience,” bothered to observe that these leaves they were sorting through smelled more like tea than marijuana. This business about the serrated leaves also came out only in depositions. There’s no record of it. They didn’t photograph the leaves to document the serration, or write down their observations about the leaves at the time.

Back to Kerr:

The officer field tested the plant material, and it tested positive for THC.
The next week, the officers searched the trash again. Again, the officers found plant material that tested positive for THC. The affidavit describes the plant material found the second time as “1/4 cup saturated marijuana plant material (leaves and stems)” that was “consistent” with that found the previous time . . .
In his opinion, Judge Lungstrum concludes that having two positive field tests for THC, at different times, is enough evidence to avoid personal liability for the officers. Even if there was no other evidence at all, those two positive test results were enough: “The fact that Deputy Burns’ affidavit states that the material found in the Hartes’ trash on two consecutive weeks field-tested positive for the presence of THC,” the judge writes, “is a hurdle that is all but impossible for plaintiffs to overcome.”
In his blog post, Balko argues that the problem with relying on field test results is that the field tests used in this case were very unreliable.
That raises an important question: Just how accurate are drug field tests, and how much should the police be allowed to rely on them to get warrants? Maybe courts should scrutinize field test results more, much like some lower courts did for dog sniffs before the Supreme Court largely put the kibosh on that scrutiny in Florida v. Harris.
Those are really interesting issues. But based on my reading of the plaintiffs’ memorandum opposing the defendant’s motion for summary judgment, the plaintiffs didn’t squarely ask the court to get into them. The most relevant part seems to be on page 62, where the plaintiffs argue that the problem is that the agents must have known that their affidavit was misleading.

The Hartes address the field tests much earlier and much more thoroughly than Kerr suggests. As I’ll lay out, they offer a litany of reasons why they believe that the officers’ reliance on the field tests was reckless. Kerr and Lungstrum boil this case down to whether the officers should have known that field tests sometimes produce false positives. Lungstrum concludes that the officers not knowing this fact, and thus excluding it from the warrant, doesn’t amount to to the sort of reckless falsehood or omission that would exempt them from qualified immunity. Kerr appears to find Lungstrum’s conclusion reasonable.

I disagree. If a police officer is going to ask a judge to defer to the officer’s “training and experience” — which is essentially asking the judge to trust the officer’s observations and conclusions in the absence of other evidence — it doesn’t seem unreasonable to ask that the officer actually have accurate education and training about the tools he uses over the course of an investigation, particularly when ignorance about those tools could result in  terrifying raids on innocent people — which can and have resulted in injury or death.

For example, why is it too much to ask that the field test which a drug cop is relying upon to obtain a warrant for an potentially violent tactical raid not have an error rate as high as 70 percent? The error rate for this test is especially high for “kitchen botanicals,” a bit of knowledge that would have been especially pertinent to know, given that the material was found in the Hartes’ kitchen trash. You could even argue that not knowing these things ought to be be considered reckless. Burns and Dep. Edward Blake (who conducted the other positive field test) testified that they had no idea it was even possible for a field test to return a false positive result.

This video, sponsored by the Marijuana Policy Project, demonstrates how tests for illegal drugs could potentially be manipulated to generate false results. (Marijuana Policy Project and Mintwood Media Collective)

But even this really only scratches the surface of what went wrong here. Dep. Burns later admitted in depositions that despite the reference to his “training and experience” on the warrant affidavit, he actually had no formal training in the particular brand of field test he used in the Harte case. He said he just assumed one would “follow the directions.” But he didn’t even do that. Worse, if he had the raid likely would never have happened. From the Hartes’ brief:

Although Burns and the other deputies stated that they believed a positive field test could provide probable cause, the instructions provided by the manufacturer of the test kit, Lynn Peavey, expressly state in a lengthy instructional guide sent to the JCSO in 2007 that a positive result on a marijuana field test only provided “probable cause” to take the “sample in to a qualified crime laboratory…”

So the manufacturer of the test explicitly cautions against relying solely on the field test to establish probable cause.  Yet that’s exactly what these deputies did.

More troubling, according to Lungstrum Burns said in depositions that “he had no knowledge in April 2012 that anyone in the Sheriff’s Office ever submitted material from a trash pull to the lab for [further] testing.” As of that time, Burns had been working drug investigations for eight years. So to Burns’ knowledge, the department had been directly contradicting the explicit instructions on the test kit all that time. Perversely, this all then works to his advantage in defending himself from the Hartes. Lungstrum concludes that because to Burns’ knowledge the department never verified field tests with lab tests, he had no reason to doubt the accuracy of the field tests. Put more cynically (and I’d say accurately), because the entire department is incompetent, Burns can’t be held accountable for his own, personal incompetence.

Quick aside: In theory, this suggests that while Burns and the other deputies may not personally be liable, the Sheriff’s Department or Johnson County could be liable for improper training that resulted in foreseeable violations of constitutional rights. Sheriff Denning himself said that not only had he never heard of a false positive among the thousands of field tests done under his watch, he wasn’t aware such a result was even possible. The director of the Johnson County crime lab, on the other hand, said in a deposition that the possibility of false positives is “well known.” But because Lungstrum ruled that there was never any violation of the Hartes’ constitutional rights in the first place, this is all a moot point.

Another quick aside: It’s worth remembering here that the law enforcement officers’ claims about no prior record of a false positive is impossible to verify. This is because at the time, Kansas law permitted police agencies to warrant affidavits and other records of investigations secret, even from victims of police misconduct or mistaken raids. This is why the Hartes had to pay $25,000 in legal fees just to get a court to order the sheriff to turn over the documents related to their case. As I noted in a previous post, because of the Hartes, the Kansas legislature has since revised the law, but the new version is only marginally better. The targets of these investigations can now obtain documents related to their case without a court order, but police agencies can still deny open records requests from media outlets, watchdog groups, or attorneys interested in researching patterns or practices of possible misconduct.

The Hartes’ memorandum opposing the defendant’s motion for summary judgment also suggests that both Burns and Reddin didn’t even understand how the test works.

Despite Burns’ professed confidence in following the directions, he had difficulty explaining what particular colors and layers the Peavey KN test was supposed to produce to indicate a presumptive positive result for marijuana. He described it as a “bilayer red,” even though the instructions clearly state that a clear layer must be on top of an orange-red layer . . .
The field tests are done in small test tubes or in small “pouches,” and the diagram on the instructions explicitly shows a clear layer on top.
Burns also testified that, with field test results, the “color does not have to match exactly,” indicating he believed a degree of subjective interpretation might be necessary.

More than a year-and-a-half after the Harte raid, Burns took his first proficiency test on this variety of drug field testing kit. He had to take it again.

Reddin too seemed unclear on how the test works.

Reddin exhibited some confusion about what constituted a positive result on a test with the KN reagent, stating a positive result was a red layer over a clear layer, when the instructions state the opposite that the clear layer is on top. Reddin also testified that he never received any training about the possibility of false positive results and had never discussed false positive results with anyone.

This passage is all the more remarkable given that, again, Reddin was the supervisor of this investigation. The supervisor for a major drug operation wasn’t aware that the field tests his officers were using have high error rates and, at least in this deposition, couldn’t even accurately explain how to tell if the test was positive. This too seems reckless.

Incidentally, the officers never photographed or provided any other proof of the positive test results, nor were they required to. Judges, prosecutors, and the rest of us simply must take their word for it. This would be of less concern if the test results were only used to send the material to a lab for further testing. But the results, again, were the only evidence for which these officers were able to obtain a search warrant for a SWAT raid, in this case on a family that turned out to be innocent.

Worse yet, in the bizarro world of qualified immunity, the officers actually benefit from their ignorance about false positives. Lungstrum points this out in his opinion. As the law sees it, an officer can’t be recklessly indifferent to truth if he was never trained about, told about, or took the time to learn the truth in the first place.

For example, the fact that the plant material was found in the Hartes’ kitchen trash was critical because, as noted earlier, the particular variety of field test the deputies used has a 70 percent error rate on legal kitchen botanicals. The Hartes argue that Burns never mentioning in his search warrant affidavit that the plant material was found in the kitchen trash was a critical and reckless omission. So the judge who signed the warrant never knew of it. (And here again, Judge Ruddick said in a deposition that if had been made aware of this fact, it would have affected his decision to approve the warrant.)

Judge Lungstrum even points out in his opinion that Burns considered and dismissed the possibility that the plant material could have been legal — solely because of the field test results:

As explained by Deputy Burns, his suspicions were essentially confirmed—and he ruled out kitchen herbs or vegetables—once he tested the substance and obtained a positive result for the presence of THC.

Yet Lungstrum not only gives Burns a pass for the omission, he doesn’t even bother to address the fact that the test Burns used is particularly vulnerable to herbs, spices, and other plants used in the kitchen. He simply dismisses the Hartes’ argument about the significance of the kitchen trash in a footnote:

In the court’s view, there is nothing inherently unusual about discarding processed marijuana in the kitchen trash (as opposed to some other trash receptacle in a residence) and, for that reason, the fact that the vegetation was discovered in the kitchen trash is not material.

“Not material.” It’s critically material.

There has to be a limit to this practice of giving police officers immunity for their ignorance. If not, it’s essentially an incentive for officers to remain as ill-informed as possible about these tests — and about any other issues for which they could later be susceptible to lawsuits. It doesn’t seem all that controversial to say if police officers are going to use field tests as their only probable cause for raiding private homes, they ought to understand how those tests work, where those tests are vulnerable, and important bits of information such as the fact that the manufacturer itself says the test should only be used to determine what material is then sent to a lab for more conclusive tests. Here too, you could also make a strong argument that not knowing these things is reckless.

Finally, while these officers will plead ignorance when it suits them, they’ll also sometimes be quite aware — also when it suits them. Burns actually notes in his affidavit that the field tests used in the investigation are “presumptive, not conclusive.” That would seem to indicate that he understood that the tests could sometimes produce false positives. (Otherwise, why note the distinction between presumptive and conclusive?) Lungstrum actually credits him for this, writing that the statement “undermines any suggestion that Deputy Burns was representing to the judge that the material was conclusively marijuana.” And yet Burns was still willing to stage a SWAT raid on the Hartes, knowing that inconclusive field tests were the only evidence against them.

Somehow, Burns gets to claim that he both knew and didn’t know that these tests aren’t always accurate —and he benefits from claiming both.

Excessive force

The “raid” part of this story is a big reason why there was so much outrage. Most people aren’t aware that the police can wage a raid on their homes based on such little evidence, especially for such a paltry offense as growing pot. If the police had merely sent a couple officers to interview the Hartes, or even sent a couple uniformed cops to serve the warrant in a less aggressive manner, the story may still have been troubling, but it wouldn’t have spurred the outrage it did. Kerr doesn’t address this aspect of the case in his post, but I think it’s worth examining.

For the most part, the federal courts have been reluctant to weigh in on the amount of force police use while conducting searches. This is one reason why SWAT-like raids for such low-level drug crimes (and increasingly, other low-level offenses) have become so common. Indeed, in his opinion, Lungstrum cites a 10th circuit ruling upholding the use of a SWAT team for a misdemeanor offense, and another finding no rights violation by a police agency that had a blanket rule to use a SWAT team for every drug warrant, even if only for possession.

Lungstrum is rather dismissive of the trauma the Hartes and their children endured when an armed team of police came into their home, forced Robert Harte to the floor, and held the entire family under armed guard for two-and-half hours. He notes that there’s no evidence anyone pointed a gun at any of the Hartes, and that none of them were assaulted (or even touched), or taken from their home. He notes that while the raid team knocked loudly at the door, and while they had all the appearances of a tactical team, they weren’t wearing “helmets, hoods, kneepads or camouflaged clothing.”

All of that is true. It doesn’t mean that harm wasn’t done. It’s traumatic to be on the receiving end of one of these raids. It’s damaging for children to come down the stairs in the morning to see an armed man a black shirt and black bulletproof vest standing over their father as he lies on the floor.

The raid team also brought a battering ram with them, which means they were prepared to tear down the door if no one had answered it in time. (Indeed, the police battered down the door to a different home on the same day, as part of the same operation. They found only “traces” of marijuana in that home.) That’s when things get much more volatile. It’s when babies get maimed and both police officers and citizens get shot.

Defenders of “dynamic entry” tactics to serve drug warrants say the aggression is necessary because drug offenders tend to be violent, and a highly-trained, cohesive unit like a tactical team clearing a house quickly makes everyone safer. There are good objections to those claims, but for the purposes of this post, let’s take them at face value. Here are some other facts about the raid that came out in the depositions:

  • The supervisor at the scene of the raid — Sgt. James Cossairt — was brought on at the last minute, and had little knowledge of the case.
  • Though he was the supervisor of the raid team, Cossairt “had only been on a front entry team himself four or five times.”
  • Cossairt had never done a drug investigation, and said his experience with drug cases was very limited. Yet he was supervising a tactical team on a drug raid. (Amusingly, Cossairt — the cop with little drug experience — also said he was aware of the possibility of false positives on field tests.)
  • The seven-member raid team consisted of cops from several divisions who had never conducted a raid together before — not exactly the well-trained, cohesive unit defenders of these raids depict.
  • One member of the team had never conducted a drug search of a home.
  • Another member of the team brought his own AR-15 rifle from home to use in the raid. Cossairt had no problem with this.
  • Cossairt said he had no idea how long a raid team should wait after knocking before forcing entry. He said he knew of no formal training on this issue. In fact, how long police must wait before forcing entry has been litigated before the Supreme Court.
  • Though Cossairt was aware of the fact that two children lived in the home, he not only didn’t make allowances for that fact, he conducted the raid at a time (7:30 am) when the children were highly likely to be inside.
  • After the raid turned up nothing, Cossairt allowed a drug dog to come in and search the entire house when two of the officers claimed to have smelled marijuana in an upstairs hallway. He then let it stay longer than necessary in order to give it some “training and experience” — all while the Harte family was confined to a couch.
  • After the raid, one of the officers told the Hartes that they should take their son to a counselor to discuss his possible use of drugs.

Keep in mind, the police had zero evidence that the Hartes had ever sold or distributed marijuana. They did no surveillance on the house. They didn’t attempt a controlled buy. They didn’t check the Hartes’ utility bills for unusually high consumption of water or electricity. They didn’t talk to neighbors about unusual traffic to or from the house. Reddin later said in a deposition that even if he had received the tip about the Hartes much earlier, he still wouldn’t have done any additional investigation. He said doing so was “unnecessary.” Deputy Blake acknowledged that the Harte home had no outward indications of a marijuana “grow,” such as blacked out or shaded windows, fans, and so on. He also said that investigatory practices like looking at utility bills was “not a common practice.”

They raided the  home solely because a flawed drug testing kit suggested that, at worst, someone in the family might be growing pot for personal use.They not only had little evidence of the Hartes’ guilt, they either ignored or went out of their way to avoid finding any evidence of their innocence. What if Robert Harte hadn’t heard the knock on the door? What if their 7-year-old daughter had answered it? What if she had answered the door just as the police were swinging a battering ram at it? What if the police had broken down the door and stormed the house and Robert Harte had mistaken them for criminals? What if he’d had a legally owned gun nearby for home protection?

But it wasn’t just the Hartes. As I mentioned in a previous post, I was in the area last year to give a talk and to do some reporting on this story. I talked to a half dozen other people who said they were wrongly raided by area police agencies. They were just still too terrified to come forward. In his deposition, Sheriff Denning said that raids producing no contraband at all were “not uncommon.”

Given the weak evidence against the Hartes, instead of raiding them, the police might also have sent a couple of officers to conduct a “knock-and-talk,” or, again, send a couple uniformed officers to serve the warrant in a less aggressive and traumatic manner. The Hartes’ memoradum mentions other examples of “knock-and-talks” conducted with other customers of the gardening store. Some led to charges, some led officers to believe the suspects were innocent.

So why the raid? The Hartes’ brief indicates that many officers looked down on knock-and-talks because they were “low-key.” One deputy said he thought knock-and-talks had “ruined” some investigations. This presumably means the suspects destroyed the plants or stopped plans to grow pot before the police could conduct a search. Of course, if your goal is to stop people from growing pot, those investigations weren’t ruined, they were successful. They were only “ruined” if your definition of success is a raid that nets plants and arrests.

Which brings me to the next part of this story . . .

The 4/20 publicity stunt

“Operation Constant Gardener” was essentially a publicity stunt by a slew of law enforcement agencies in Kansas and Missouri. Sheriff Denning, along with other law enforcement officials across the state, had pre-planned press conferences on the afternoon of April 20th, which of course is the unofficial holiday of pot smokers. The press conferences were to announce all the busts and arrests produced by the operation. The public information officer for the Johnson County Sheriff’s Department had already pre-drafted a press release under the heading ““Law Enforcement Celebrates 420 with Multitude of Arrests.”

This seem likes the most logical explanation for why the deputies didn’t conduct a more thorough investigation of the Hartes. When asked why the investigation proceeded so quickly, Dep. Burns said in a deposition, that the “goal was to have it done on that day.” Though Robert Harte and his children had been spotted at the gardening store more than seven months earlier, the tip from Sgt. Wingo wasn’t passed on to them until March 20th. They didn’t conduct their first trash pull until April 3rd. This was the pull in which Burns found the “innocent” plant material. The next pull was on April 10th, and the final one on April 17th. The raid was on the morning of April 20th. That afternoon, Sheriff Denning and other law enforcement leaders touted their success. A rash of favorable media coverage followed. As Denning told one reporter, “They consider this a holiday to use marijuana openly, so we’ve decided to celebrate that day with them.”

But it really wasn’t much of a success. The raids conducted on April 20th didn’t result in the confiscation of a single marijuana plant. They did produce two misdemeanor possession charges and one felony for failure to procure a drug stamp (a bizarre law that requires pot users to pay taxes on the same marijuana that state law forbids them from possessing — which the state’s Supreme Court recently watered down.) Several days of raids across Johnson County leading up to April 20 found just two “grows,” resulting in the confiscation of four pot plants at one house, and 19 at another. In Olathe, the Johnson County seat, the raids produced no confiscations or arrests. In nearby Shawnee, there were arrests for low-level offenses, but no grow operations.

It’s impossible to say for certain how many of the “tips” from the gardening store led to confiscations, charges or successful prosecutions, because the police officials kept no record of investigations that were ended for lack of evidence. But Sheriff Denning himself acknowledged in a deposition that “there wasn’t much contraband recovered.” He didn’t mention that in his April 20 press conference, of course, nor did he mention that innocent people were raided.

The 2012 operation was actually the second edition of “Operation Constant Gardener.” The operation the previous year produced somewhat better results. But the Hartes’ attorney found internal communications between law enforcement officials after that operation which suggested that there was a lot of pressure to equal or improve those results in 2012. After the 2011 results were in, Sgt. Wingo, the Missouri Highway Patrol officer in charge of the campaign, sent out an email to participating police agencies to celebrate the success:

Thanks again to all of you who worked on this. I think for a first time even this was absolutely an outstanding success. The media coverage was 99% positive on this which is pretty darn good as well. We’ve had a lot of suggestions for this operation next year, such as having a telethon type billboard with a large green marijuana plant filling up as the pledges come in, making T-Shirts and whatnot. One agency had the observation that if this continues for a couple years 4/20 will be something to fear rather than something to celebrate. On another interesting note, Shawnee wins the IRONIC award for having a search warrant signed at exactly 4:20 p.m….

Note that eighth line. They want to make 4/20 “something to fear.”

But even the more successful 2011 campaign showed how easily innocent people could get ensnared in the rush to produce seizures and arrests in time to make the April 20th evening news. Wingo initially sent a message to all the agencies participating in the 2011 campaign telling them that he’d accumulated hundreds of names of customers who had been seen shopping at hydroponic gardening stores. He promised a “significant media event,” but also warned that they would only have time for a “brief investigation” before the 4/20 media blitz.

The press release sent out the morning of April 20 announced that police would be targeting 375 houses across both Kansas and Missouri. But Wingo later reported that police found just 52 marijuana grows. That’s better than single digit number of grows found in 2012. But it also means that more than 300 households were wrongly targeted. In Kansas City, the police searched 25 homes, but found just 7 in which someone was growing marijuana. In a foreshadowing of the Harte raid, the police raided one man who was growing only tomatoes. He became something of a running joke in the email chatter between law enforcement officials.

In addition to Sgt. Wing’s aim to make 4/20 “something to fear,” Sheriff Denning later said in a deposition that he wanted raids carried out that day to “send a message” to pot smokers. It’s clear, then, that there was a directive to law enforcement officers to produce as big a haul of drugs and arrests as possible, and that all the investigations needed to be wrapped up so that the raids could be carried out and the press conferences and news releases could be sent on April 20.  The Hartes very plausibly argue that all this pressure to produce results and produce them quickly caused the officers to take shortcuts in their investigations of the Hartes, to seize only on the light evidence of criminality they found, and to overlook the evidence of their innocence.

Lungstrum doesn’t buy any of this. He notes that Dep. Burns said in a deposition that he didn’t send the wet tea leaves to the crime lab because of department policy, not because he felt rushed to meet the 4/20 deadline. Likewise, Dep. Blake said they didn’t do any corroborating investigation of the Hartes because doing so “wasn’t common practice” at the department. I find both statements difficult to believe. But it also shouldn’t matter. The reason why they did no corroborating investigation or defied the explicit manufacturer instructions about sending field test results to the crime lab aren’t as important as the fact that both of these things happened. If anything, the fact that it was a matter of ongoing policy instead of the result of pressure for a one-time (or annual) campaign makes it all the worse.

No accountability

The law provides a number of ways to address a potential Fourth Amendment violation. This case illustrates as well as any that none of them are adequate.

The first way the Fourth Amendment is supposed to protect us is by requiring a judge to sign off on a warrant. In theory, judges will scrutinize warrants to be sure the police officer requesting it has established probable cause, has conducted a thorough investigation, etc. In theory, judges will apply extra scrutiny to warrants that will be served with added force and violence, such as a “dynamic” home entry or a no-knock raid. In practice, this isn’t how it works.

Despite the fact that the Operation Constant Gardener warrants would be served with dynamic entry tactics, for example, Judge Ruddick conceded in a deposition that he didn’t even recall the Harte warrant, but that it’s likely that he probably didn’t examine it for very long, and that his attention to it was likely “minimal,” “routine,” and “very short-lived.”

That isn’t at all unusual. After the fatal and mistaken raid on Ismael Mena in 1999, a newspaper investigation into drug warrants issued in Denver found that judges almost never declined a warrant and rarely asked the police officer requesting it any questions. Worse, the report found that judges had in some cases authorized no-knock warrants even when the police hadn’t requested one. (There’s supposed to be higher burden to obtain a no-knock.)

After the botched Atlanta raid that led to the death of 88-year-old Kathryn Johnston in 2006, an A.P. investigation found that police officers generally gave scant details in search warrant affidavits, including for searches to be conducted with no-knock or dynamic entry raids. The report found that in many cases, “officers only detailed their years of experience with the police force as justification for no-knock powers.”

I’m not aware of any recent studies attempting to quantify how frequently judges authorize warrants, but for my book, I did find a 1984 study which found that not only do judges rarely turn down warrants, the police quickly learn which judges are rubber stamps and adjust their behavior accordingly. The more skeptical judges are avoided. In a much more recent study from a trio of Duke law professors, the authors interviewed 30 federal magistrates and district judges from four states about their approach to evaluating warrants. Their conclusion:

The judges whom we interviewed did not see the probable cause decision as a probabilistic decision about facts but rather as a formalistic decision applying a pre-determined legal criteria. If the petition addresses all of the features of the existing legal criteria of probable cause, then it will be deemed sufficient. For them it appears to be more of an accounting task than a factual determination.

There’s very little effort to determine if the police officer is telling the truth, if the police officer is reliable, or if the boilerplate language about “training and experience” common to nearly all these warrant affidavits is actually backed up by training and experience.

A second way we attempt to protect the Fourth Amendment is by providing remedies after someone’s rights have been violated. In a case in which police find evidence of criminal activity, this is carried out through the Exclusionary Rule, which (in theory) bars such evidence from being used at trial. But there are some gaping exceptions to the rule. Police can argue that the violation wasn’t related to the search. They can argue that they were acting in good faith. In some cases, such as in Hudson v. Michigan, the Supreme Court has ruled that even if the police violate your Fourth Amendment rights, the Exclusionary Rule doesn’t apply. The theory here is that the rule serves as a deterrent to Fourth Amendment violations. The evidence on this is mixed at best.

The other remedy for a Fourth Amendment violation comes in cases like the Harte case, where the suspects are actually innocent. Here, they can in theory sue the police officers who violated their rights. But as we’ve discussed, the police have qualified immunity. You must not only show that your rights were violated (which Lungstrum found the Hartes failed to do), but that a reasonable police officer would have known that what he did was a violation of your rights.

Qualified immunity is a difficult hurdle to overcome. Mere negligence by a police officer is usually insufficient, even if it resulted in significant harm, or even death. Civil suits aren’t intended to be a deterrent; they’re supposed to compensate the victims. In most cases, individual police officers are indemnified from any personal liability. On the rare occasion that a plaintiff wins in court (most of the strong cases are settled before then), the taxpayers usually foot the bill. So the deterrent effect here is weak (witness the multitude of stories about cities that pay out millions in settlements, but still do little to nothing to hold bad cops accountable), although I have found a couple of examples in which a city’s municipal insurer demanded policy changes after multiple lawsuits.

In the update to his post, Kerr hints at the final approach to protecting the Fourth Amendment. He writes:

Just to be clear, then, this search should be an enormous embarrassment for the officers. The lead officer should probably be reassigned to a different job if he hasn’t been already. The officers should have investigated the case much more carefully, and it’s particularly bad that the officer looked at tea and thought it was marijuana. What kind of experienced officer can’t tell the difference?

This is similar to the approach Justice Scalia took in his majority opinion in the Hudson case. Instead of omitting evidence, or allowing police officers to be sued, we should just trust police agencies to discipline officers, or rely on shame or embarrassment to do the trick. Scalia referred to the “new professionalism” in police departments across America that is ushering in some new wave of police accountability. (The criminologist whose research Scalia cited in that opinion vehemently reject Scalia’s argument in a subsequent op-ed. Kerr then wrote a post defending Scalia.)

That was in 2006. If there’s anything we’ve learned from the last few years of reporting, government investigations, and coverage of police brutality, shootings, and misconduct, it’s that police officers are very rarely disciplined for even egregious offenses. Thanks to union contracts and laws and policies with names like the “Police Officers Bill of Rights,” it’s incredibly difficult to fire bad cops, even for offenses like wrongful shootings, beatings, or sexual assault. In many jurisdictions, the overwhelming majority of complaints are dismissed with little to no investigation. So the idea that internal discipline or professional embarrassment might nudge officers like those involved in the Harte case into more careful investigations or a better appreciation of Fourth Amendment rights just isn’t very convincing, particularly given that professional incentives (real or implied arrest quotas, performance reviews based on arrests, etc.), institutional incentives (federal anti-drug grants that are based on raw arrest figures), the desire to please superiors, and policies like qualified immunity give them a much more forceful nudge the other way.

In the Harte case, there isn’t even much remorse, much less accountability or change. Dep. Burns said in his deposition that in retrospect, he wouldn’t have done anything differently. He also said he did not have a “great deal of concern” for what the Hartes went through. Similarly, Sgt. Reddin said he had “no regrets” about the way the raid and investigation were handled. Sheriff Denning told Reddin he was “comfortable” with how the case was handled, and that he approved of all the decisions made by his deputies.

Since the Harte raid, Sgt. Tom Reddin has been promoted to lieutenant. Sgt. James Cossairt has also been promoted to lieutenant. And Dep. Mark Burns has been promoted to sergeant.

ADDENDUM: I should note that there has been one change as a result of the Harte lawsuit — the Johnson County Sheriff’s Department no longer uses these particular field tests, and apparently does wait for confirmation from the crime lab on other field tests before conducting raids. That’s a positive step, although as indicated, it’s a change that should have been the policy all along, and it only came about because the Hartes spent a great deal of money to discover the reasons for the raid, then filed a lawsuit based on what they had found.