(Bigstock)

Editor’s note: This is part 1 of a four-part series.

Correction: An earlier version of this post stated that Sean Groubert and Michael Slager both initially faced murder charges. However, the man Slager shot, Levar Jones, survived the shooting. Groubert was initially charged with assault and battery of a high and aggravated nature, the same charge to which he pleaded guilty. 

“To this day, if you live in Kershaw County and saw those media reports, you’d still think Lori Jean Ellis was just some crazy old black lady who opened fire on a bunch of cops,” says Robert Phillips. “You’d think she got what she deserved.”

Phillips is an attorney in Rock Hill, S.C., just across the border from Charlotte. He represented Ellis’s estate in a lawsuit after her death. Ellis, 52, was killed on April 21, 2008, when an officer from the state’s Department of Natural Resources (DNR) and two Kershaw County sheriff’s deputies served a series of warrants on her home at about 11 pm.

In the ongoing national discussion about police brutality and lethal force, a common reform recommended by advocacy and activist groups is that police shootings be investigated by an outside organization. Police agencies shouldn’t be trusted to investigate internally, the thinking goes, and prosecutors are too close to and dependent on local police to be impartial. Wisconsin recently passed a bill requiring all police shootings investigations be led by a law enforcement agency other than those involved in the shooting. The new policy has been well-received by both police groups and advocates for reform.

But the practice of outside agencies investigating officer-involved shootings has already long been common in South Carolina. With the exception of Richland County, all police shootings in the state are investigated by the South Carolina Law Enforcement Division (SLED), the free-floating police department that’s commonly seen as the state’s version of the FBI. In addition to officer-involved shootings, SLED has a panoply of other responsibilities, including terrorism investigations, technical training for other agencies, drug investigations, support for local agencies, arson investigations and alcohol regulation. SLED also conducts statewide investigations at the request of the governor or attorney general. Currently, local police agencies invite SLED in to investigate officer-involved shootings as a sort of custom, but the South Carolina legislature is now considering a bill that would require this in all officer-involved shootings in the state.

But within South Carolina, attorneys who handle police abuse cases say that not only isn’t the model working, it may be allowing wayward law enforcement officers and agencies to operate with a sense of invincibility. Policing experts who reviewed the Ellis shooting and several other cases for this series seemed to agree: The model isn’t working. There is no doubt that a proper oversight agency can do a lot of good. But an inept one can do a lot of harm — especially an inept agency that’s perceived as competent.

On the night Lori Jean Ellis died, DNR Officer Gregg Lowery and Deputy William Sowell arrived at her home first, then radioed Deputy Tyrell Coleman for backup. According to the officers, they then rode Lowery’s mostly unmarked DNR truck over Ellis’ fence and around to the back of her home. They then maced her dog. They claimed to have called out to Ellis several times. When she didn’t respond, they ripped open her back door with an animal control pole. That’s when they say Ellis fired a high-powered rifle at them, and they responded with gunfire. Lowery fired two shots: one missed, and one grazed the top of Ellis’s shoulder. Coleman fired the fatal shot, which struck Ellis in the back of the head.

The next day, Kershaw County Sheriff’s Department spokesman Cpl. David Thomley told WISTV in Columbia that Ellis had an “extensive criminal history” dating back to 1972. He told WLTX TV that the officers “knew of her violent past so it was absolutely a necessity to have more than one officer go out there.” He added that the county’s residents should be both thankful for the officers’ sacrifice and grateful that they escaped the raid unharmed. “It’s also unfortunate for these officers that had to go through that and it’s gonna be a tough time for them as well, but at the same time, you have to stop and give thanks that it didn’t turn out any differently,” he said.

Within hours SLED investigators were on the scene in Kershaw County. The three deputies signed their statements at 2:20 a.m. the morning after the raid. Lowery described seeing a puff of smoke from the barrel of Ellis’s gun. Coleman stated that he saw a muzzle flash. Sowell wrote that he heard a shot, and would later say his face was less than 10 feet away from the gun barrel when she fired what sounded to him like an SKS semi-automatic rifle, a weapon similar to an AK-47. All three officers claimed to have seen Ellis pointing the gun at them.

Three-and-a-half months later, SLED sent its report to W. Barney Giese, then the solicitor for South Carolina’s Fifth Judicial District. (In South Carolina, district attorneys are called solicitors.) Based on SLED’s report, Giese determined that the two officers who shot Lori Jean Ellis “were in reasonable fear for their lives,” and therefore “were justified in discharging their firearms.” All three officers were cleared of any wrongdoing. (Giese, incidentally, is now in private practice. He’s currently the defense attorney for Sean Groubert, the former state trooper now awaiting sentencing for shooting unarmed motorist Levar Jones during a traffic stop. He did not return a request for comment.)

A Google Earth view of Ellis' home in 2008. A Google Earth view of Ellis’s home in 2008. (Courtesy of Robert Phillips)

Despite Giese’s decision, one needn’t delve too far into the case to see some red flags. The warrant for which Lowery went to Ellis’s home that night was a bench warrant for failure to appear in court for an open-container violation. The officers found three other outstanding warrants for Ellis, but all for what might be called “poverty crimes” — nonviolent crimes more the result of having little money than criminal intent. One was for driving on a suspended license, another for failure to turn over a suspended license and the last was for bouncing a $219 check.

Oddly, Lowery was off duty on the night of the raid. According to deposition transcripts, earlier in the day he had attended a city employee appreciation event on the other side of the county. Phillips estimates that Lowery drove about two hours out of his way to pull and serve the open-container warrant on Ellis.

The biggest red flag of all was that high-powered rifle that all three officers claimed they saw, the one that allegedly produced a muzzle flash, smoke and a 140-decibel boom — the equivalent of standing about 10 feet behind a jet engine. But Ellis’s gun wasn’t an an SKS rifle, or an AK-47 or a firearm at all. It was a BB gun. Because a pellet gun uses air pressure instead of gunpowder to propel its projectile, it isn’t capable of producing a muzzle flash, a loud boom or a plume of smoke. The state crime lab also found no fingerprints on the weapon, according to the SLED report, despite the fact that the shellacked wood from which it was made should have been conducive to retaining prints. In fact, the lab couldn’t even get the gun to function.

The more Phillips dug, the more problems he found. There was plenty of evidence inconsistent with the officers’ narrative. The bullet trajectories (as calculated by SLED investigators, then confirmed by investigators for Phillips) rendered the officers’ version of events next to impossible. And then there were the tactics themselves: Why did the officers drive a black, mostly unmarked DNR truck at 11 p.m. at night through Ellis’s fence and into her back yard, mace her dog, and rip open her door with an animal-control pole, all to serve misdemeanor warrants for nonviolent crimes?

Such stories about botched raids, questionable police shootings and implausible statements from law enforcement officers aren’t uncommon. But there was an additional and especially troubling component to Ellis’s death. SLED investigators are supposed to be objective and impartial. Yet when looking into Ellis’s death, they didn’t pursue any of these inconsistencies, contradictions or questionable tactics.

LJETrailer Lori Jean Ellis’s trailer at night, shortly after her death. (Courtesy of Robert Phillips)

Ellis certainly did have a criminal record, mostly for bouncing checks and failure to pay fines for traffic violations. She had a couple arrests for domestic violence, which Phillips says was the result of a relationship that while sometimes volatile, wasn’t actually violent. “They’d have arguments, get angry, and one would call the cops on the other,” Phillips says. None of those arrests resulted in a conviction. Ellis’s only felony conviction was possessing a car that had been reported stolen. Phillips looked into the allegation and thinks that too was the result of an argument. “In poorer parts of South Carolina, people don’t sell cars with all the paperwork. So someone doesn’t pay someone back, or bounces a check for the agreed amount of the sale, and the seller might report the car stolen. It isn’t like she hot-wired some car off the street.” Ellis was convicted and given a sentence that was suspended if she paid the balance of what she owed on the car — about $400.

Ellis was also a veteran who served as a nurse in the 1991 Gulf War and suffered from PTSD as a result, according to Phillips. She appears to have medicated with alcohol. One neighbor who otherwise remembered Ellis fondly added that she was often intoxicated, although not in a threatening way. She mostly lived off her pension and disability benefits. Neighbors described her as kind and generous with what she had. Ellis loved animals, and kept a lot of them on her property — chickens, pigs, goats, the occasional llama. In one deposition given several years later, a neighbor likened Ellis’s home to Noah’s ark. Ellis had no children of her own, but she welcomed neighbor children to her house. Photos of the inside of her apartment the night she died showed a collection of toys. She would sometimes surprise her neighbors by mowing their lawns while they were away.

In 2013, Kershaw County settled with Ellis’s estate for $2 million — a hefty sum for a county of 60,000 people. It’s an especially large settlement given that Ellis had no close relatives and died almost instantly. Damages are typically based on the survivors’ relationship to the victim, and the amount of pain and suffering the victim endured. In Ellis’s case, the money went to several half-siblings she hadn’t seen in years. Nor did Ellis suffer much. She died within seconds.

“The county knew this was bad,” Phillips says. “They wanted this case to go away.”

Outside oversight: no panacea

“There are two things you’re looking for by having an outside agency investigate these shootings,” says Seth Stoughton, a former police officer who now studies law enforcement and police law at the University of South Carolina School of Law. “You want the appearance of independence — you want the investigation to be legitimate in the eyes of the public. But you also want actual accuracy and objectivity. You need to have both.”

An outside agency that doesn’t achieve both may be worse than no outside agency at all. If the investigating agency has only the reputation of integrity and independence, but doesn’t demonstrate either in practice, misconduct doesn’t just go unpunished, it gets papered over. The public gets a false sense of security. Watchdog groups, journalists and social justice groups may be less likely to question shootings, or they’ll be taken less seriously when they do. That can create a culture in which abuse and corruption flourish.

Phillips, other attorneys across the state who have represented victims of police brutality and policing experts consulted for this series say there are signs that this is what’s happening in South Carolina.

“I like most cops,” says Brian Gambrell, an attorney in Columbia who has litigated police abuse cases. “We need cops. I think most cops are hard-working people who have tough jobs and try to do the right thing. But we let the bad cops get away with things here. The policies I’m talking about don’t protect the well-intentioned, hardworking cops. They protect the bad ones. Some of them start to think they’re bulletproof.”

Judging by recent headlines, South Carolina would seem to be a place where police shootings are taken seriously and where officers who shoot in error are held accountable. In just the past two years, two high-profile shootings resulted in criminal charges for a police officer: the shooting of Walter Scott by North Charleston police officer Michael Slager last April, and the shooting of Levar Jones by former state trooper Sean Groubert in September. Both shootings were caught on video. In March, Groubert pleaded guilty to assault and battery of a high and aggravated nature. He faces a possible sentence of up to 20 years in prison. Slager was indicted for murder in June. His trial is scheduled for October.

There have been other incidents, too. In late 2014, Eutawville’s then-police chief Richard Combs was charged with murder in the death of an unarmed Walter Bailey near a courthouse in 2011. After one mistrial, Combs in September pleaded guilty to misconduct. He was sentenced to a year of home detention. And North Augusta Officer Justin Craven is due to be tried on felony charges in the death of 68-year-old Ernest Satterwhite. Craven had tried to pull Satterwhite over on suspicion of driving under the influence. Satterwhite instead drove home. As he approached Satterwhite’s car, Craven claims to have seen Satterwhite reaching for a gun. He fired through the passenger-side window, killing Satterwhite. But Satterwhite was unarmed. In each of these cases, the police officer was white, the victim was black, and the victim was unarmed. (UPDATE: In April, Craven pled guilty to “misconduct in office” — a misdemeanor.)

These particular high-profile cases from just the past few years seem to suggest that at least parts of South Carolina are more willing than most jurisdictions around the country to charge police officers with crimes after unjustified shootings and other incidents of excessive force. But critics such as Phillips say they’re the exceptions that prove the rule. They argue that these are just especially egregious examples that attracted national attention and scrutiny, and that — with the exception of the Bailey shooting — have been captured on video.

Richard Harpootlian, a former Fifth Judicial Circuit solicitor who now works in private practice, says SLED investigations can be hit or miss, depending on the investigating agents. “You have regional officers who handle these cases, sometimes involving a department they used to work for,” he says. “It used to be that cops thought internal affairs departments were out to get them, that the investigators were looking to put a notch in their belt. SLED often errs too far the other way. We need something in between.”

SLED investigations also only look at whether police officers committed a crime. They don’t look for civil rights or constitutional violations, or for patterns of excessive force with a particular police officer or agency, and they don’t make judgments about questionable tactics or use-of-force policies. But they’re commonly seen as the last word in these cases. A SLED investigation may only clear a police officer of criminal wrongdoing, but the public perception is that if SLED investigated and the cop was cleared, there’s nothing more to discuss.

To complicate matters, the final word on whether or not to charge a police officer rests not with SLED, but with the local solicitor. So when it comes to making the decision on whether to press charges, the alleged independence that comes with a SLED investigation is no longer relevant. “The local solicitors work with law enforcement every day,” says Harpootlian. “Those relationships can distort the sort of justice that you get in these cases.”

A solicitor can decline to press charges even when SLED finds significant wrongdoing on the part of a police officer. The Zachary Hammond case is a good example. Hammond was shot and killed by Seneca police officer Lt. Mark Tiller in a fast-food restaurant parking lot last July. According to Eric Bland, who represented Hammond’s family in their lawsuit, the SLED investigation found that Tiller put himself in harm’s way, then engaged in unnecessarily reckless behavior when he shot and killed Hammond as the teenager drove away. But two months later, 10th Judicial Circuit Solicitor Chrissy Adams declined to press charges against Tiller.

“We were mostly satisfied with the SLED report,” says Bland. “There were some problems with transparency, but the report itself was comprehensive. It was the solicitor’s refusal to press charges that was baffling.”

This can create a double layer of protection from criminal liability for the state’s law enforcement officers. If SLED finds no evidence of a crime, it’s unlikely that a local prosecutor will conduct a separate investigation. But even when SLED finds wrongdoing on the part of a police officer, which is rare, the local prosecutor — the same prosecutor that works with that department on a daily basis — can still decline to indict.

“In the end, I think police officers in South Carolina get the sort of presumption of innocence that everyone ought to get,” says Harpootlian. “Unfortunately, it’s only police officers who seem to get it.”

An empty investigation

The death of Lori Jean Ellis is a vivid illustration of the problem. Though SLED investigators showed up at the scene of the shooting within hours, the three officers were then permitted to sit in a room together unmonitored before a SLED investigator interviewed them and took their statements. That already is problematic. “You just don’t do that,” says law professor Stoughton. “It’s investigation 101, whether you’re investigating police officers, suspects, or witnesses. You separate them to be sure you get independent statements.”

But a recent investigation by a team of reporters at the (Charleston) Post and Courier found that SLED routinely lets officers collaborate, talk to their attorneys and take a night or two to “sleep on it” before they’re asked to give a statement or sit for an interview. One officer was allowed to wait 28 days before giving a statement. Others gave statements via email.

The truck used to raid Lori Jean Ellis' home. (South Carolina Law Enforcement Division) The truck used to raid Lori Jean Ellis’s home. (From the SLED report)

Despite the fact that there were five houses around Ellis’s, SLED investigators only interviewed one of her neighbors. One neighbor they didn’t interview, a juvenile justice officer and former corrections officer, would later testify in a deposition that she heard the raid go down. By her account, there were just three shots fired, not four, which would have been necessary to verify the officers’ account. (She did say she heard an initial shot well before she heard the subsequent three shots in a row. Phillips says he thinks one of the deputies initially shot out the security light in Ellis’s yard. The officers say it was shot out by the SWAT team that responded after the raid. The neighbor’s account corroborates Phillips’s theory.)

The SLED investigators took measurements of the bullet trajectories, but they made no apparent effort to determine whether the trajectories were consistent with the officers’ accounts. A more thorough investigation would have found that trajectories strongly suggest that the raid couldn’t have happened the way the officers claim it did. Deputies Coleman and Sowell said Ellis was standing at the door with her gun when she fired at them; Lowery claimed she pointed it at them as she charged down a short hallway. But based on the trajectory and height of the bullet that grazed her shoulder, for Ellis to have been at the door when the officers fired their weapons, her shoulder would had to have been less than three feet off the ground. When standing, Ellis’s shoulder was 5 feet from the ground. By the officers’ version of events, she would have had to have been in a semi-squatting position. That isn’t a natural position to stand under any scenario, but certainly not while holding a rifle. On the night of the shooting Ellis’s blood-alcohol percentage was 0.28. So it’s also unlikely stance for a woman who was severely intoxicated.

According to an animation of the raid based on the bullet trajectories commissioned by the officers’ own attorneys, it also seems unlikely that the officers could have seen Ellis through her doorway at all. From the position where Lowery and Coleman fired their guns, there’s no line of sight to the back doorway. In fact, the only officer who could have seen through the doorway was Sowell, who was in the back of the truck. He was the one officer who didn’t fire.

This reconstruction, commissioned by the Kershaw County officers' own attorney, is based on measurements taken by South Carolina Law Enforcement Division investigators. (Robert Phillips)

The trailer itself was also pitch black. There were no lights on inside, and Ellis’s home is in a rural area with no ambient light. One plaintiff’s expert who visited the trailer at night described the conditions as “total darkness.”

Phillips and the experts he hired believe Ellis was likely in a drunken sleep in her bedroom when the officers arrived, shot out her security light and maced her dog. She likely awoke when they ripped open her door and proceeded to her bedroom door to see what was happening. They say the forensic evidence suggests she was standing behind her bedroom door when the first bullet went through the outer wall to the trailer, through the door and grazed the top of her shoulder. That’s consistent with the bullet trajectories, as well as a fresh scratch on her face that was consistent with a piece of wood found in the bedroom that was ejected from the door as the bullet went through it. Phillips and his experts believe Ellis opened the bedroom door and walked toward the back door to investigate the commotion, then likely turned and retreated back to her bedroom when she realized she was being fired upon. That’s likely when Coleman’s bullet struck her in the back of the head.

As for the BB gun, it was found in the doorway. That’s consistent with testimony from Ishmail Stokes, Ellis’s boyfriend at the time, that she kept the gun leaning against the door frame so it would be handy when she needed to scare off wild animals. “My guess is that when they pried open the door, the gun fell, and one of the officers yelled ‘Gun!,’ ” Phillips says. “And then Lowery and Coleman panicked and fired their weapons.”

Ellis’s gun is also the subject of some of the more disturbing shortcomings in the SLED investigation. Once the agents learned that the rifle was a pellet gun, not a high-powered firearm, impartial investigators should have tried to re-interview the deputies to ask them to explain the discrepancy. How could they have seen a muzzle flash, seen smoke and heard a deafening sound from a pellet gun that was incapable of producing any of those things?

The SLED investigators never asked any of these questions. In incredible depositions taken of the SLED investigators in 2011 and 2012, Phillips tried to figure out why they didn’t. In one exchange, Phillips asked SLED captain Paul Grant, who supervised the SLED investigators in the case, if the officers would have been justified in shooting Ellis if she had never fired her “weapon.” Grant refused to speculate. Phillips pressed him. What Grant said next is remarkable.

Phillips: Let’s just assume these cops aren’t telling the truth . . .

Grant: But they are telling the truth.

Phillips: How do you know that?

Grant: Because they’re police officers and I believe what they’re telling me . . .

Excerpt from video deposition of South Carolina Law Enforcement Division captain Paul Grant, taken in the case Blakely vs. Kershaw County Sherriff's Office, et. al. (Robert Phillips)

Grant believes them, simply because they’re police officers.

“I understand why an investigator might feel some some camaraderie with the police officers he’s investigating,” says Stoughton. “But his job isn’t to clear these officers. It’s to develop an objective record of facts to determine if they committed a crime.”

Stephen Downing, a former deputy chief of police for the Los Angeles Police Department who now advocates for police and drug reform, is far more blunt. “He is incompetent,” Downing said after reviewing the case. “He doesn’t know the difference between an opinion and a fact, yet he treats a police officer’s statement as fact. His bias is transparent and rather ugly.”

Later in the deposition, Phillips asked Grant about the muzzle flash, the sound and the smoke the officers claimed to have seen from the pellet gun. “So when this one says ‘I saw something like smoke come out of the end of the gun,’ that didn’t cause you any concern at all?” Grant replied, “No.”

Phillips then turned to Coleman’s claim to have seen a muzzle flush.

Phillips: Have you ever seen a BB gun or a pellet gun create a flash?

Grant: I haven’t.

Phillips: Have you ever seen a high-powered rifle create a flash?

Grant: I haven’t seen one, but I know they probably do.

Phillips: So the deputy involved in the shooting says, “I saw a flash,” and then the next day we find out that it wasn’t a high-powered rifle, it was actually a BB gun. That didn’t cause you any concern?

Grant: No.

Phillips: And why not?

Grant: The officer thinks he’s being shot at by a high-powered rifle. We’re all unable to see the same things. I’ve seen that over my years of experience. If something happened in this room right there, all of us are not going to recount it the same way.

Phillips tried the question from a different angle. He got the same result.

Phillips: Okay, let me ask if you this. If something happens in this room, and one of us testified to something that was physically impossible, do you think all of us are being truthful?

Grant: I can’t answer that either. I’m saying that you see things a certain way than somebody else sees it.

Phillips pressed on. He asked if Grant believed the statements from the officers about seeing a flash, seeing smoke and hearing a shot from the BB gun were “highly unlikely.” Grant replied, “No. I would not agree.” Phillips asked why not. Grant replied, “Because people see things differently. We’ve got three differently people who see things differently. The bottom line is, they all saw a high-powered rifle being pointed out the door at them.”

Excerpt from video deposition of South Carolina Law Enforcement Division captain Paul Grant, taken in the case Blakely vs. Kershaw County Sherriff's Office, et. al. (Robert Phillips)

Grant didn’t say the officers claimed to see a high-powered rifle. He said they did see one. He took everything the officers said at face value. Their word was truth, and the investigation proceeded from there.

“All he is doing here is shrugging and saying, ‘That’s what he said,’ ” says Downing. “If that is all that is needed in a competent investigation, then you don’t need an investigator with 20 years’ experience. A note taker would suffice.”

Stoughton sees the exchange similarly: “This isn’t a matter of trusting or believing the officers. If they’re claiming to have seen things that couldn’t have happened, you have to at least try to determine if those claims are reasonable. If they aren’t, then you have a problem.”

Paul Grant was later promoted, and is now listed as the assistant chief of SLED.

Phillips’s deposition of SLED chief case officer James Flowers produced the same sorts of answers. Flowers wouldn’t admit that a pellet gun is incapable of producing a flash. He conceded only that he personally has never seen it happen. Phillips continued:

Phillips: So did anything prevent you, from the moment that you found out it was a mere BB gun, to say, “I want to go back and talk to this deputy . . .”

Flowers: Nothing prevented me from doing that.

Phillips: Okay. Why didn’t you go back?

Flowers: Because I didn’t feel it necessary.

Phillips: So someone telling you something that you’ve never seen before, that doesn’t compel you to maybe follow up?

Flowers: No. Not in all cases . . .

Phillips:  . . . so if I tell you something that can’t physically happen, you’re just going to take my word for it?

Flowers: See, here’s the thing. As the lead investigator for the state’s premiere law enforcement agency, it is my responsibility to put this case together. After looking at this information, I deemed that it was not necessary to interview that officer again. And that was the decision that I made.

Excerpt from video deposition of South Carolina Law Enforcement Division Agent James Flowers, taken in the case Blakely vs. Kershaw County Sherriff's Office, et. al. (Robert Phillips)

“The arrogance here is stunning,” Downing says. “This response either reveals Flowers’s incompetence or his bias. Either way, he should not be conducting investigations of officer-involved shootings.”

Later, Phillips asked Flowers what decision-making process he went through to conclude that it wasn’t necessary to follow up on the portions of the officers’ statements that directly contradicted the forensic evidence. Flowers replied, “I looked at the information, and I didn’t deem it necessary. It’s pretty much — it’s pretty simple, honestly.” Flowers gave a similar answer when Phillips asked why he only interviewed one neighbor when there were five houses that had lines of sight to the shooting. He simply didn’t think he needed to.

Flowers also admitted he wasn’t aware of the bullet that grazed Ellis’s shoulder. He thought the bullet that killed her was the only bullet that struck her. He also didn’t know which officer fired the fatal bullet, and he mistakenly believed that one of the bullets had gone through Ellis’s front door. (They all went through the wall to the side of the door, which is why it would have been difficult for the officers to have seen Ellis when they fired their weapons.) Flowers also admitted he never saw or inspected Ellis’s gun. He admitted that he hadn’t read the entire forensics reports — nor does he in most of the cases he investigates.

Phillips: Well, when you signed off on [the SLED report] —

Flowers: Uh-huh.

Phillips:  — had you read the firearms, the latent prints, and the fingerprints —

Flowers: No.

Phillips: — report?

Flowers: I don’t read those whole reports, no.

Phillips: You don’t read these reports?

Flowers: No.

Phillips: Why is that?

Flowers: I get a synopsis of those reports from the people who provide — who do those reports.

Phillips: That’s not procedure for you to read what’s in these reports?

Flowers: Not those full reports, no.

Phillips: Why is that?

Flowers: That’s just how it is.

This testimony would be unnerving enough if it had come from a sheriff or police chief. But Flowers was working for the agency in charge of investigating police shootings. And he was talking about a case for which he was the lead investigator.

“You absolutely have to read the forensic evidence,” says Stoughton, the law professor. “You can’t conduct a proper investigation if you don’t.”

Two months prior to the raid on Ellis’s home, Lowery was involved in another incident in which he discharged his weapon but failed to report it. He was suspended for five days by the Department of Natural Resources. The SLED investigators didn’t look into that incident as part of their investigation of the Ellis shooting, even though it may have raised questions about Lowery’s credibility.

Ken Katsaris, a former law enforcement officer and trainer with over 30 years of experience, was another of Phillips’s experts for Ellis’s family. Katsaris is no hired gun for plaintiffs. He testifies in defense of police officers about three times more often than he testifies against them. He also wrote one of the reports that cleared the officer who shot Tamir Rice. In this case, Katsaris seem baffled by what he learned. In his deposition, Katsaris said the idea that an officer investigating a police shooting wouldn’t read all of the forensics reports is “almost beyond incredible.”

Then Katsaris laid into Flowers:

[Flowers] says it is not important to interview the officers about whether Ms. Ellis actually fired the gun or not. Not important. He never looked at the fingerprint reports. He has no opinions about whether the gun was or was not fired. He has never even seen the gun that Ms. Ellis had. He has no idea whether the gun that Ms. Ellis has can smoke or flash. He’s the case agent. He’s turning this over to the solicitor.

You’ve got a gun that conceivably doesn’t smoke and certainly doesn’t flash, and he doesn’t even know. He didn’t ask questions about whether the gun flashed or smoked or that the officer said that the gun flashed and smoked . . .

He got information after the fact, and even though it was contraindicated to what the officer said, he said he didn’t go back and ask any follow-up questions. He said his decision about following up is just something that is not necessary no matter what you find out.

He found it irrelevant to interview witnesses at the scene who may be there or who may have heard something. He doesn’t even have an answer as to why the neighbors weren’t interviewed . . .

He doesn’t actually know which officer actually shot Ms. Ellis. He’s giving the solicitor information about whether a crime or not has occurred. He doesn’t even know at the time of his deposition which officer actually shot. He says it’s not important to know the trajectory or shot placement. Not important. He doesn’t know about the bullet holes or the trajectory. He said angles of the trajectory are not troubling to him. He said he never looked for an explanation on how her gun could have flashed.

Lowery also told SLED investigators that he had made several prior attempts to serve the warrants (the first of which was issued on April 2) on Ellis during the daytime, but she had avoided him. But while Lowery’s worklogs for April 2008 included such items as inspecting boats, litter duty, checking the bait in turkey traps and attending city employee appreciation parties, they make no mention of attempts to serve the bench warrants for Ellis. That’s surprising, given that the police would later describe Ellis essentially as a violent fugitive. Notably, it was Phillips who obtained Lowery’s worklogs during discovery for the lawsuit. It’s another inconsistency in the officers’ story that SLED investigators never checked out.

More protectors than investigators

Last year, an investigation by Clif LeBlanc of the State newspaper in Columbia found that of the 209 incidents in which South Carolina police officers fired their weapons over the previous five years, only a few resulted in criminal charges. Those few cases had only occurred over the previous year or so, and none to that point had resulted in a conviction. The paper consulted attorneys who in prosecutor’s offices and others who defend police officers accused of misconduct. All told the paper that until 2014, they couldn’t remember a single instance in which a police officer had been charged for a shooting.

Of course, that could be because all or almost all of those police shootings were justified. The State noted that “the vast majority of the suspects shot at in South Carolina during the past five years have been armed.” That’s consistent with national FBI data showing that the vast majority of police shootings involved armed suspects. But that would also mean that the state’s law enforcement officers only recently began shooting people without justification. That seems unlikely. The other, more likely, possibility: It’s been happening all along, but the emergence of video has forced state officials to take a closer look.

The State also noted that in most shooting cases, “records … are sometimes incomplete, which leaves key details to be ferreted from individual police agencies.”

The Ellis shooting demonstrates why those details are so critical to assessing the legitimacy of a police shooting — and why “armed suspect” statistics can be so misleading. For the purposes of the FBI data, Ellis would have been considered “armed.” That says very little about her death, the unanswered questions about the events leading up to it, or whether the officers were justified in shooting her.

“Too often, SLED investigators don’t question inconsistencies in reports or statements from police officers,” says defense attorney Christopher Mills. “They’ll let the officer corroborate their stories. They’re great at finding evidence that clears police, but you don’t see much effort to find evidence that incriminates them. There was a time when internal affairs officers were disliked by other police officers. In South Carolina, other cops tend to see SLED investigators as their protectors.”

The current head of SLED, Mark Keel, has said made some public statements that seem to support Mills’s characterization of the agency as more of a protector of law enforcement officers. As part of its investigation last year, the State newspaper interviewed Keel. Paraphrasing him, the paper wrote that he “worries  . . . whether increased public scrutiny might result in more officer deaths as people second-guess the officers’ split-second decisions.”

That sentiment — that merely scrutinizing police shootings will result in more officer deaths — is one commonly expressed by police unions and advocacy groups. Agree with it or not, it’s a strange view coming from the head of an agency entrusted to investigate police shootings fairly and objectively. Keel then told the paper, “as long as individuals continue to not follow the instructions of law enforcement … there’s going to continue to be these confrontations. The public has to understand to be more compliant.”

Melvin Tucker is former police chief in Tallahassee, Fla., and now works as a litigation consultant. He has consulted in over 500 cases, including 21 in South Carolina. He was a consultant for Phillips in the Ellis case. “To hear that from the head of SLED bothers me,” Tucker says. “It makes an assumption right off the bat that any shooting or incident of excessive force was the citizen’s fault, not the police officer’s. You have to question whether there’s any objectivity.”

Tucker and other policing experts interviewed for this series reviewed several cases, many previously unreported, and found a pattern of inaccurate statements, misleading reports and at times what appeared to them to be rather brazen attempts to cover up misconduct. The reviewers generally concluded that the SLED investigations into these cases were too deferential to the accused officers, incomplete and at times showed little interest in looking beyond the statements of police officers. As readers will see over the course of this series, the false statements and reports were often not just from individual officers, but occurred at an institutional level.

“Practice here long enough, and you start to see patterns of deceit,” says Mills. “In some of these departments, there’s little hesitation about hiding or even manipulating evidence.”

SLED’s investigations generally don’t look into that, either. In short, while SLED meets the first of Stoughton’s criteria for credible outside investigations — the appearance of independence — there’s good evidence that it falls short on the second, actual independence, accuracy and objectivity.

The Post and Courier investigation last year, for example, concluded that SLED’s handling of police shootings amounted to “an investigative system stacked in favor of officers who use their guns.” In one example, the paper looked at the shooting of Bryant Heyward, a 26-year-old black man in Hollywood, S.C. Heyward dialed 911 last May as his home was being invaded by two gunmen. Heyward quickly grabbed his handgun and exchanged fire with the men. When the police arrived, a deputy mistook Heyward for one of the assailants and shot him, paralyzing him. The discrepancy in how Heyward and the deputy were treated is telling. Heyward was interrogated by an officer who climbed into the ambulance with him on the way to the hospital. When the paper’s article ran nearly three weeks later, the deputy still hadn’t been questioned.

Of the 235 SLED investigations of officer-involved shootings, the Post and Courier found 168 in which investigators documented the amount of time that had lapsed before the officers were asked to give a statement. Of those, 104 officers were given two or more days before talking to investigators. Only 35 were required to give a statement the same day.

As Richard Harpootlian pointed out above, SLED typically assigns investigators by region — a SLED agent from the southern part of the state will usually get cases from that area. The theory is that those agents will have local knowledge of the area and local sources that could aid in their investigation. But Tucker explains that presents a lot of problems too.

“How is an agency like SLED staffed?” Tucker asks. “With police officers. Where do those officers come from? Other police departments. The same police agencies that SLED investigates. I had a case where the SLED investigator had just worked as a deputy for the same department he was investigating. You can say an agency like that is independent, I guess. But I don’t know that it means much.” The attorneys and experts interviewed for this series say while such a direct conflict of interest isn’t always the case, it also isn’t uncommon.

The Post and Courier review also found that as in the Ellis case, SLED officers rarely looked into an officer’s disciplinary history or at other evidence to determine if the officer is reliable. Nor did the agency look into the history and patterns of particular police departments. Each case was typically looked at in isolation, with a great deal of weight and trust put on the officers’ statements.

Phillips was so troubled by Ellis’s death that he brought it to the attention of South Carolina U.S. Attorney William Nettles. “He seemed disturbed by it, and said he’d look into it,” Phillips says. A couple months later, Phillips received a curt voice mail from the FBI agent assigned to investigate the case, telling him he had found no evidence that the deputies had violated Ellis’s civil rights. Phillips later discovered that prior to joining the FBI, the same agent had been a deputy with the Kershaw County Sheriff’s Department.

In the end, despite the $2 million settlement, neither Gregg Lowery, William Sowell or Tyrell Coleman was ever disciplined or reprimanded for the death of Lori Jean Ellis.

“It’s bad here,” Phillips says. “There’s a culture of not just brutality, but invincibility. There’s no oversight. You see blatant lying, not just from individual officers, but from entire agencies. They think they can get away with anything. And they usually do.”

(Note: The Kershaw County Sheriff’s Department and the South Carolina Department of Natural Resources did not respond to a request for an interview. A public information officer with the South Carolina Law Enforcement Division initially agreed to set up an interview with chief Mark Keel, but then didn’t call back.)

Coming in Part 2: The power of video.