How to police the police is a question as old as civilization, now given special urgency by a St. Louis County grand jury’s return of a “no bill” of indictment for Ferguson, Missouri, police officer Darren Wilson in his fatal shooting of an unarmed teenager, Michael Brown. The result is shocking to many, depressingly predictable to more than a few.

Can the cops be controlled? It’s never been easy: according to one old sociological chestnut, the monopoly on the legitimate use of violence is what defines modern government, and this monopoly is jealously protected against the second-guessing of puny civilians. All over the country, the issue of restraining police power is framed around the retribution against individual cops, from Staten Island to Milwaukee to Los Angeles. But is this the best way to impose discipline on law enforcement and roll back what even Republican appellate court appointees are calling rampant criminalization?

Police shootings in America

First, the big picture. Last year, the FBI tallied 461 “justifiable homicides” committed by law enforcement—justifiable because the Bureau assumes so, and the nation’s courts have not found otherwise. This is the highest number in two decades, even as the nation’s overall homicide rate continues to drop. Homicides committed by on-duty law enforcement make up 3 percent of the 14,196 homicides committed in the United States in 2013. A USA Today analysis of the FBI database found an average of about ninety-six police homicides a year in which a white officer kills a black person.

The FBI’s police homicide stats are fuzzy, and they are surely an undercount, given that they come from voluntary reports to the FBI from police departments all over the country. That the federal government does not keep a strict national tally shows just how seriously it takes this problem. A crowdsourced database has sprung up to fill the gap, as has a wiki-tabulation.

Perhaps the most disturbing thing about these police killings, many of them of unarmed victims, is that our courts find them perfectly legal.

SCOTUS and the license to kill

Chapter 563 of the Missouri Revised Statutes grants a lot of discretion to officers of the law to wield deadly force, to the horror of many observers swooping in to the Ferguson story. The statute authorizes deadly force “in effecting an arrest or in preventing an escape from custody” if the officer “reasonably believes” it is necessary in order to “to effect the arrest and also reasonably believes that the person to be arrested has committed or attempted to commit a felony…or may otherwise endanger life or inflict serious physical injury unless arrested without delay.”

But this law is not an outlier, and is fully in sync with Supreme Court jurisprudence. The legal standard authorizing deadly force is something called “objective reasonableness.”

This standard originates in the 1985 case of Tennessee v. Garner, which appeared at first to tighten restrictions on the police use of deadly force. The case involved a Memphis cop, Elton Hymon, who shot dead one Edward Garner: 15 years old, black and unarmed. Garner had just burgled a house, grabbing a ring and ten bucks. The US Supreme Court ruled that a police officer, henceforth, could use deadly force only if he “has probable cause to believe that the suspect poses a significant threat of death or serious physical injury to the officer or others.” The ruling required that the use of force be “objectively reasonable.” How this reasonableness should be determined was established in a 1989 case, Graham v. Connor: severity of the crime, whether the suspect is resisting or trying to escape and above all, whether the suspect posed an immediate threat to the safety of officers or others. All this appeared to restrict police violence—even if, in the end, Officer Hymon was never criminally charged for fatally shooting Edward Garner.

“Objectively reasonable”—what could be wrong with that? But in actual courtroom practice, “objective reasonableness” has become nearly impossible to tell apart from the subjective snap judgments of panic-fueled police officers. American courts universally defer to the law enforcement officer’s own personal assessment of the threat at the time.

The Graham analysis essentially prohibits any second-guessing of the officer’s decision to use deadly force: no hindsight is permitted, and wide latitude is granted to the officer’s account of the situation, even if scientific evidence proves it to be mistaken. Such was the case of Berkeley, Missouri, police officers Robert Piekutowski and Keith Kierzkowski, who in 2000 fatally shot Earl Murray and Ronald Beasley out of fear that the victims’ car was rolling towards them. Forensic investigations established that the car had not in fact lurched towards the officers at the time of the shooting—but this was still not enough for the St. Louis County grand jury to indict the two cops of anything.

Not surprisingly then, legal experts find that “there is built-in leeway for police, and the very breadth of this leeway is why criminal charges against police are so rare,” says Walter Katz, a police oversight lawyer who served on the Los Angeles County Office of Independent Review until it disbanded in July of this year. According to Erwin Chemerinsky, dean of the UC Irvine Law School, recent Supreme Court decisions are not a path towards justice but rather a series of obstacles to holding police accountable for civil rights violations.

An officer’s personal threat assessment is often bolstered by the fact that there are between 270 million and 310 million guns in the United States. Take a grand jury’s failure to indict the police officers who fatally shot John Crawford III, the black man holding a BB gun in a Walmart in Beavercreek, Ohio. In a country where shooting sprees are a regular occurrence, where guns are widely available at Walmart and where fake guns that look very similar to real guns are sold in the same store, the police officers’ fears were deemed reasonable enough for the grand jury to find no probable cause of criminal wrongdoing. That is how the Supreme Court police violence jurisprudence works, and it was firmly on the side of officer Sean Williams, just as it has now been found to be on Darren Wilson’s. Given the deference and latitude hardwired into the law, “there is just an underlying assumption that the officer did not engage in criminal activity,” says Katz.

The first step to controlling the police is to get rid of the fantasy, once and for all, that the law is on our side. The law is firmly on the side of police who open fire on unarmed civilians.

The sick joke of self-regulation

The lethal use of police force typically sets off an internal police investigation to determine if departmental regulations were violated. The regs and the law are not the same thing. Case in point: the chokehold that NYPD officer Daniel Pantaleo used to strangle Eric Garner, suspected of selling loose cigarettes, on Staten Island last July. (The grand jury bill on that case has still not been decided.) The chokehold is not prohibited by law, but it is by departmental rules. The violation might earn a departmental censure of some kind, from loss of vacation days to getting fired, but they tend to be radically mild, when not nonexistent.

What about internal affairs investigations? On television they are aggressive, dogged, uncompromising. In real life they tend to insulate the police from serious external sanction. “I stopped cooperating with the IAB ten years ago,” says Jason Leventhal, a former assistant district attorney in Richmond County, Staten Island who now works as a civil rights litigator, often suing the police. “IA will never, ever credit the claim of police abuse. They hide witnesses, they push witnesses around. The only time I cooperate with them is when I know I have their hands tied behind their back.”

Are there any effective civilian oversight systems at any major police department in the US? Nobody I interviewed for this article could name one. New York’s Civilian Complaint Review Board occasionally docks vacation days from police officers but the board has no real teeth. Even staffers at the New York Civil Liberties Union have candidly told me that it’s more or less worthless. “I don’t have any faith in the CCRB or the Internal Affairs Bureau or any other internal mechanism,” says Ron Kuby, a civil rights and criminal defense lawyer in New York. Civilian complaints rarely even get in the way of an individual officer’s career. In New York, CCRB complaints don’t even go in a police officer’s file, says Kuby. “The PBA just says that the more aggressive officers will get excessive force complaints.”

Firing a police officer with a record of abusive behavior (or worse) is often extremely difficult and can carry a heavy political cost. Patrolmen Benevolent Associations, which have escaped the kind of resentment directed at other public-sector unions, tend to be powerful players in local politics able to inflict pain on any politico who would cross them. (Remember when Sarah Palin struggled to fire a state trooper and ex-brother-in-law who had allegedly acted like a thug towards her sister?)

The reality is, it is extremely difficult to get law enforcement to police itself, and self-regulation is here, just as it is in poultry processing or coal mining, a sick joke.

Civil suits

Civil suits for monetary damages require a lower standard of proof than criminal cases, but these suits are not a slam-dunk for victims of cop violence, either. The same jurisprudence that grants wide leeway to law enforcement still holds. Last March, one victim’s family lost a federal civil suit for wrongful death and civil rights violations brought against police officer Nicholas Bennallack for fatally shooting a fleeing and unarmed drug suspect. The jury believed the cop’s claim that he opened fire out of fear for his life.

What about all the times when excessive force suits get settled out of court? It turns out that massive payouts don’t deter police misconduct for one straightforward reason: neither individual officers nor police departments are responsible for coughing up the cash. The union covers the officer’s lawyer, and research from Joanna Schwartz of UCLA Law School found that governments, not individual officers, paid out 99.98 percent of the damages. Settlements and damages aren’t paid by the police department, whose budget will waltz by untouched, but typically out of the general municipal budget.

Kuby maintains that civil remedies will always fall short and thinks only criminal prosecution has a prayer of changing police behavior. “Prosecution works well with people who are not fundamentally criminal and have enough stake in the system to respond! Any response below that is an insult.” But he admits that the political will to make the criminal justice system restrain and regulate its own members and enforcers is consistently lacking—and has been as long as he’s been practicing law.

Don’t make a federal case out of it.

Occasionally the federal Department of Justice intervenes to prosecute individual cops for depriving a victim of his or her civil rights. If a state-level prosecution fails to secure a conviction, this can give the government a second chance to prosecute on different charges. Rodney King’s LAPD assailants were convicted in 1993 in just this way.

But the feds, like other prosecutors, only like to take on cases they can win, says Alan Vinegrad, who was part of the federal attorneys that prosecuted the NYPD officer who in 1997 sexually tortured Abner Louima with a broken-off broom handle. “In the Louima case it was painfully obvious that the use of force was extreme and willful,” says Vinegrad. “But in other cases, if the officer is acting more in the line of duty, a federal civil rights violation is a tougher thing to show.”

Will the Obama DOJ take on the Michael Brown case? It doesn’t look like it. Eyewitness testimony is conflicting, and so far the multiple forensic reports that have come out are not inconsistent with Darren Wilson’s version of the shooting—which would make the federal standard of willful misconduct difficult to prove. The Obama DOJ has signaled with leaks that it is backing away from a federal case against the officer.

Far more useful are the DOJ Civil Rights Division’s root-and-branch interventions into violently dysfunctional police forces, triggered by “patterns and practices” of systematic rights violations rather than any one particular incident. For instance, the DOJ just launched a major effort to reorganize and reshape the police department of Albuquerque, reforming its trainings, protocols and appallingly trigger-happy habits. (The Albuquerque police have shot thirty-seven people in the past four years, twenty-three of them fatally.) This is a well-established tool of the federal government: other police departments under federal supervision include those of Seattle, New Orleans, Puerto Rico and, until recently, Los Angeles and Detroit; the feds also nearly took over wholesale the Oakland police department in 2012. But note that these federal interventions do not entail punishments, civil or criminal, of individual police officers.

In the unusual instances when a law enforcement officer is convicted, the penalties tend to be remarkably light. Johannes Mehserle, the Bay Area Rapid Transit officer who shot the unarmed Oscar Grant III dead on New Year’s Day, 2009, served eleven months of a two-year sentence for involuntary manslaughter. Then there is Jon Burge, the Chicago police detective who led the torture of over 100 (mostly black) suspects, released last month after serving four and a half years in custody. (The statute of limitations had passed for the more serious offenses.) Burge still collects a $54,000 departmental pension.

Such Scandinavian-style lenity is quite different from the mind-numbing severity of sentences inflicted on non-cops. Consider the sixty years that potentially face Marissa Alexander for firing a warning shot at an abusive ex. Or the life sentences without possibility of parole that more than 3,000 Americans are serving for nonviolent crimes.

Video

Some observers see hope for police reform in the ubiquity of smartphone video recorders. It’s true that the ever-growing supply of police misconduct videos fuels the anger needed to sustain reform efforts, and even occasionally leads to police officers being disciplined or prosecuted. Examples include NYPD officers David Afanador and Tyrane Isaac, caught on camera pistol-whipping a teenage suspect last August, or South Carolina state trooper Sean Groubert who in September opened fire on a motorist heading into his truck to retrieve his driver’s license. Jason Leventhal told me he expects an indictment of the Staten Island police officer who strangled Eric Garner because the video evidence, caught on a cellphone, was so brutal. It would certainly be a good thing if police departments invested less in military gear like M-RAP armored vehicles and more in dashcams and GoPro cameras; with the right department protocols and practices, these cameras strengthen police accountability.

But just as often these videos end up illustrating just how much leeway police have in opening fire on a suspect. Take the police shooting of St. Louis resident Kajieme Powell, a mentally ill man allegedly holding a knife, a fatal shooting arguably less defensible than the Michael Brown shooting ten days before, and caught on a cell-phone video. Or, again, John Crawford III, whose slaying by police officers was caught on the Walmart security cameras. Or the July, 2012 video of eight members of the Saginaw, Michigan, police department, six of them firing forty-six shots at Milton Hall, a mentally ill homeless man, hitting him eleven times, after he took out a pocket knife when a police dog started to lunge at him. Although the latter horrific video picked up a fresh wave of publicity when screened at a hearing of the Inter-American Commission on Human Rights in Washington, DC, last month, such videos cannot undo the legally enshrined deference to the subjective feelings of police officers when they reach for their weapons. No criminal charges were even attempted by state prosecutors in any of these cases; the DOJ has announced it is looking into the Crawford shooting, but declined to prosecute in the other two.

(A note on the IACHR and other international forums: bringing these cases of police shootings to them is a canny way to generate publicity and raise consciousness, but no one should ever imagine for even a second that such bodies will ever wield any actual power in American courts.)

There really is no courtroom miracle or lawsuit solution, no matter how clever the litigator, no matter how deep-dish the foundation grant, that is going to discipline the police and break them of their trigger-happy habits.

Thinking big

Police shootings are only one function of living in one of the most heavily policed societies in the world. Any movement to roll back this creeping overcriminalization is going to have to look beyond criminal prosecutions of individual police and take in the big picture.

The militarized police response to the mostly nonviolent demonstrations in Ferguson and elsewhere has appalled not only progressives but many conservatives as well. The army’s provision of of weapons surplus, from MRAPs to tanks, to local police departments under the federal 1033 Program, ought to be easy to end. But departments are reluctant to give back new toys, and majorities of both the congressional Progressive Caucus and the Black Caucus voted against a bill that would have stopped the disbursement of some military goodies to local police departments. Gun control advocates would do well to lobby hard and publicly for arms control measures applied to the police as well—undisciplined but up-armored cops are part of what that drives many otherwise sane private citizens to build private armories.

Wanton overpolicing had poisoned relations between the people and their government well before Darren Wilson shot dead Michael Brown. Less mediagenic than police militarization and far more insidious is law enforcement’s daily harassment of citizens for petty offenses. The local government in Ferguson has been treating its residents and neighbors less like free people with rights than like revenue milk-cows to be exploited to the max. Citations and fines for petty offenses are profligately inflicted on residents, particularly black residents. According to a blockbuster report issued by St. Louis’s ArchCity Defenders advocacy group, over 20 percent of city revenue comes from municipal courts (making them the city’s second-largest source of revenue), which issued enough warrants last year to slap three warrants, $312 worth, on every household in the town.

Not surprisingly, high on the list of demands issues at one Ferguson community forum was an end to the “overpolicing and criminalization of poverty,” an amnesty for old unpaid warrants, new fines proportioned to income and a state law capping municipal revenue from court fees at 10 percent. (Terrified by the Ferguson unrest, the city of St. Louis decided to eliminate 220,000 open arrest warrants for traffic violations last month.) The feeling of being under occupation by an armed force that cares more about meeting revenue quotas than public security corrodes all trust in law enforcement, and is the sort of environment in which police are more likely to open fire.

The state of emergency that Missouri governor Jay Nixon declared on November 17 seems all too likely to encourage the police overkill, both petty and heavily militarized, that shocked the world over the summer, when much of the state’s use of force against demonstrators was of dubious legality.

New policing models like the problem-solving approach developed by David M. Kennedy and others have focused law enforcement and social work resources on a city’s small number of known and likely violent gang members, leaving the rest of the community in relative peace. This approach has won demonstrable results in Boston and Cincinnati without intrusive “broken windows” policing or the wholesale stop-and-frisk harassment of black and Latino youth. Kennedy, director of the Center for Crime Prevention and Control at John Jay College, writes in his memoir, Don’t Shoot, that this type of community policing doesn’t fix the entire economy, but it does reduce homicide—including police homicide—and at least create social peace.

Police demilitarization, the decriminalization of working-class people, new policing models: these are all projects that could work in Ferguson and thousands of other American cities. Although none of these large-scale ideas is explicitly race-conscious, they would most likely tighten the severe racial disparities in policing violence that exist all over the country, more so than pouring more money into racial sensitivity training for cops. (Changing residency requirements of municipal police officers to get a more ethnically representative force might help a little, though research shows that such requirements correlate with less confidence in the police, not more.)

These big-picture reforms are fundamentally political solutions that will require long-term effort, coalition politics that spans race, ethnicity and political affiliation—a challenge, but also a necessity. As police and prosecutors assume more and more power in the United States—regulating immigration (formerly a matter of administrative law), meting out school discipline, and other spheres of everyday life where criminal law was almost unknown even a generation ago—getting law enforcement on a tight leash is a national imperative. In the meantime, the constant stream of news reports of unarmed, mostly black and Latino civilians killed by police demands bigger, bolder approaches. They are the only available paths to getting the police under control.