Tag Archives: Nick Selby

Cross: Nick Selby, A Cop At The Crossroad of Tech and Reform

September 7, 2016 (Fault Lines) — Ed. Note: Scott Greenfield and David Meyer-Lindenberg cross Nick Selby, a Dallas-area police detective, co-author of In Context: Understanding Police Killings of Unarmed Civilians and CEO of StreetCred.

Q. Before you became a cop, you spent twenty years at a bewildering array of jobs. You were a sound engineer for CBS, wrote guidebooks for Lonely Planet, consulted with businesses and governments on information security, became a pilot and co-founded businesses, including a publishing firm in Russia and the information security practice at The 451 Group. Where were you heading? Did you have a plan or did you go where the wind blew you? You certainly had an interest in IT security from the early days. Did you ever consider becoming a lawyer, a prosecutor?

A. Well, to hear me tell it, the arc of my career is really smooth and predictable: I specialize in helping people understand highly complex systems. Doesn’t matter if your network’s been hacked, you’re lost in Syktyvkar, or you’re trying to figure out officer performance. And yet, I admit that on paper, I have the world’s weirdest resume. You left out that in 1987 I designed the sound for Penn & Teller on Broadway.

So it wasn’t about going where the wind blew, it was about doing what my dad always told me, which was, “Do what you love and the money will follow.” That’s pretty great advice, and I’ve tried to do it. I can honestly say I have the best job in the world, because it’s the one I created to take immensely deep dives into things that fascinate me: can data, through statistics, data-mining and basic math mixed with sociology, philosophy and a basically scientific approach, actually explain how the criminal justice system works in the same ways that it has allowed, for example, retail or car sharing to work? I think that it can.

The challenge I address is the asymmetric understanding among people in and outside of the criminal justice system of law enforcement process, procedure, workflow and dataflow. We are at a very strange time in our country: people are trying to apply statistical analysis to law enforcement data without the fundamental understanding that law enforcement data is a siren song, urging people to aggregate that which cannot be correctly aggregated due to intrinsic flaws in the data. I talk about this a little later in some detail.

One thing my wife and I are thrilled that we did was our concept of “pretirement.” When we were in our 20s, supporting our travels around the world through writing books for Australia-based publisher Lonely Planet, we decided that most people work a lifetime and hope they’re healthy and wealthy and of sound enough mind to book tickets to someplace fantastic. We decided that before we get fat and sick, we should take that time. For the next couple of years we bounced around South East Asia, Australia, Eastern and Western Europe and South America. A quarter-century later I can’t say I think we made a mistake in that decision, and it looks as if we may even stay thin, healthy and (one hopes) wealthy enough to do it again when we’re old.

My mom is a writer. My dad was a corporate and securities attorney in New York, and I worked in law firms on information technology for a couple of years. The only thing that would make me consider law school would be the opportunity to be a prosecutor. I have a whole lot of respect for prosecutors, and just as much for public defenders (my friend John Cannel was a public defender in Hudson County for years before becoming the Commissioner of the New Jersey Law Revision Commission – John has done so much for the people of New Jersey).

Public defenders and prosecutors are so necessary to our system of justice; they are treated so badly by our system, and required to give so much more than the job description calls for. I always have time to hear of ways that technology might be of use to prosecutors or defenders and I stand ready to help. I am working right now with a partner on getting training in cyber crime (economics, logistics, laundering) to state attorneys. This is a hugely important thing that has been ignored for years, and I look forward to helping there in a significant way. Stay tuned.

Q. You founded StreetCred in 2010, to help police make sense of the mass of data now being collected. One product gives cops easy access to information about people with outstanding warrants, and another is supposed to help chiefs and administrators evaluate their cops’ performance. It was the same year you became a cop. Did you become a police officer because of StreetCred, or was StreetCred the result of your interest in law enforcement? Do you consider StreetCred part of the legal tech movement, with its attendant problems in getting its target demo to use it? Are street cops open to new tech? Are they willing to use it? Do they use it the way it was intended? Is there an issue with databases being misused, abused or sold? Is privacy adequately protected?

A. I first became interested in solving law enforcement data problems in 2008. I was consulting Fortune 500 corporations and governments (both US and foreign) on technology issues (like identity, persona, semantic search, data exfiltration, and data theft) and realized that the only time we ever hear of data and technology in law enforcement in our country is when it relates to terror. That’s really silly, because America’s cops spend about a tenth of a percent of their time on terror.

Having spent years in the tech industry, I recognized that a big problem in police technology was that people create technology for other markets and then try to repurpose it to sell to cops – that’s one reason why police tech is so bad. Another is that police take a long time to adopt technology; the excuse is they need to understand the way it works under all conditions, and more importantly, make policies for its use. That takes years. Police are about 15 years behind the mainstream. After almost two years of work, I started to find problems that were acute in law enforcement but which had been solved – really, solved – in the private sector.

That’s when I came up against the one thing that is very true about the popular perception of cops: cops generally trust cops, and distrust non-cops. Especially if it is a non-cop telling a cop what the cop should be doing to solve a problem that the non-cop has never personally faced, in the sense that it was his ass on the line.

That is one of the biggest challenges: it is all fine and good for civilians to discuss, in the comfort of their homes, cars, offices or even courthouses or statehouses, what they think the cop should have done in a given situation. It is quite another to be standing alone on the street at night facing someone who is angry, bigger than you, pumped-up and ready to fight or flee based on the fact that he’s done something illegal. Or when you happen into a situation in which there is a body on the floor, and the family is screaming, and the children are terrified, and the neighbor is shouting. At these moments, the theory flies out the window.

Now, tell me how your iPhone app is going to change my life.

It was that understanding that got me to the point that I needed to go to the academy and get a job, but it was also something else: I am one of a small percentage of cops with training and experience in information technology, and we have a growing cyber crime problem in this country. I am writing now about how this all comes down to training of prosecutors (in a nutshell, they won’t take a case they don’t think they can win, and how can they win if they don’t understand even the basic economics of how cyber crime works), and I felt I had a civic responsibility to lend a hand.

In 2010, I found an agency that would tolerate my unique situation and sponsor me to the academy, and I’ve worked part-time as an officer since. For the past three years, I’ve been a part-time (paid) detective, and my work is almost entirely organized retail crime (centered on credit card theft and fraud against point of sale systems) and child pornography.

Your question raises some of the set of problems I have been working on since joining up, that has gotten worse since Ferguson: there is an exquisite temptation on the part of non-statisticians and even some statisticians, but usually reporters with the thinnest of statistical training (and no sociological training), to mix and match police, crime, court, jail, prison and corrections statistics and then reach a general conclusion. This is a huge mistake, and it leads to really, really bad conclusions and worse social and policing policy decisions.

Anyone can easily make broad statements about broadly aggregated statistics. Here’s an example: there was a substantial reduction in 2014 in seizures by the Drug Enforcement Administration (DEA) of hallucinogens (to 48,970 doses, from 119,507 doses in 2013 and 872,366 doses in 2012), and methamphetamine (to 2,946 kg in 2014 from 4,227 kg in 2013 and 4,813 kg in 2012). Seizures in the same period of heroin and cocaine were relatively flat: heroin seizures were at 1,020 kg in 2014, 1,044 kg in 2013, and 1,010 kg in 2012), and cocaine seizures were at 33,770 kg in 2014, from 24,103 in 2013 and 36,736 kg in 2012[1].

The thing about those kinds of statistics is that they feel really detailed. And in a way, they are: they are indeed the fodder for a lot of completely valid analysis about a lot of things, from the efficacy of the drug war to the consideration of racism in its lopsided prosecution. Consider that, despite widely stated “epidemics” of heroin and opioid overdoses and methamphetamine overdoses across the country, at the level of the President of the United States, actions against traffickers of those narcotics are flat over time. It is also interesting to note that, despite DEA marketing and promotion of its continuance of classification of marijuana as a dangerous drug, marijuana enforcement in the United States is down significantly: seizures of marijuana fell in 2014 to 74,225 kg from 270,823 kg in 2013, 388,064 kg in 2012, 575,972 kg in 2011 and 725,862 kg in 2010[2]. This is a pattern that belies the agency’s rhetoric.

Okay, so all that is excellent, and it seems incredibly detailed. Yet it has nothing to do with criminal justice statistics at the ground level. At the ground level, of the 40 million people who had a face-to-face interaction with a police officer in 2008[3], every single encounter was unique. Sure, we can extrapolate by use of some very clumsy pivots: incidents with use of force by the officer versus those with no use of force; incidents which ended in an arrest or citation, versus those that ended with a verbal or written warning. But in each and every face-to-face interaction, human beings interact, and things are never the same.

Was the cop in a bad mood, and being a dick? Was the person being stopped in a bad mood and getting lippy? Did the cop ask the person, “Where are you coming from?” and expect an answer? Did the person express their stern conviction that the officer has no right to stop them, because they are a free citizen? Were the police called to the scene by a concerned neighbor, who gave a report of a crime and a description of the suspect, or did the officer self-dispatch? Was the suspect high? Mentally ill? Drunk? Sad? Maudlin? I don’t care how good you are with statistics, you can’t answer these questions just as I can’t answer them, because the method of data capture is hopelessly shabby and inconsistent across our nation’s nearly 18,000 agencies. And remember – about one in four people who come in contact with a cop do so more than once in a year. That skews statistics, too.

So, while the temptation is tantalizingly strong to draw conclusions when a white officer shoots dead an unarmed black man, there is always more to the case than “race,” “armed status” and “dead.” We count more than 70 such data-points[4], and each point is a pain in the ass to get.

When Muhammad Youssef Abdulazeez[5] opened fire on military personnel at a Chattanooga recruiting station, I don’t think anyone thought the cops shooting him to stop that deadly threat were not justified. That, statistically, is the majority of cases in this country, no matter what source you look at (more on this later). But I am absolutely interested when a cop shoots someone like Walter Scott[6] in the back.

I challenge you to find a reasonable person who thinks Officer Slager was justified.

My concern, my goal, is to accurately count the Walter Scotts of the world.

This stuff is controversial. I am absolutely, positively certain that the numbers of Walter Scotts are dramatically lower than people think. As in, significantly less than 10 percent of all people killed by police are killings that are not legally and even ethically justified.

Getting people to look at data when the police data to date has been so dirty, so badly gathered and analyzed, so seemingly dishonest, is really difficult. It serves us damned right as police that now we are the ones having to defend against shitty work by charlatans or well-intentioned but incorrect reporters. But that is where we are.

But – and this is a huge ‘but’ – unless you are operationally familiar with the nuance and order of every single step in the workflow between first contact and ultimate disposition, you simply don’t know what happened by looking at the stratospherically high-level statistical aggregations available about law-enforcement interactions in America.

I’ve assigned myself the job of helping. Wish me luck.

Q. You’re one of a very few cops willing to publicly support police reform. But when you do speak out, you’re careful to express admiration for the police, praise their heroism, and you more often try to gently nudge cops in a certain direction than outright criticize conduct. Your average Black Lives Matter protester would probably call you a cop shill. Why are you being so careful? Your co-bloggers, like Peter Moskos – himself a former cop – reject the notion that there’s a problem with police culture. Do you? Are you a believer in the “one bad apple” view of police misconduct?

A. I’m going to push back on several aspects of this question. I don’t think that there are so few cops willing to publicly support police reform. As I have written[7], it is my genuine belief that when experts on law enforcement critique the work that police officers do, cops listen. Many police reform issues raised during the past five years have led to profound[8] changes[9] in departments[10] across[11] the country[12]. But cops, like anyone else, don’t like hearing that they’ve done a bad job when they have done a good job[13], and that (like cops testifying against other cops[14]) happens more often than you might think.

I do think that there is a tide of very poorly informed media reporting on the actual division between police and the community. This generation of reporters doesn’t work the police beat, doesn’t learn the streets, doesn’t understand the law, doesn’t have contacts like the last one or two ago – that is not a bad thing at all. Those generations never held the cops to the kind of scrutiny that this one rightfully does. But for every great scoop, there’s a load of dung to sift through, and people are getting really poor civic understandings from this generation of clickbait-headlined, bullshit coverage[15].

I think there is a lot, too, that I don’t understand about Black Lives Matter protesters – and I’ve reached out. In my personal experience, those to whom I have reached out have remained aloof and unengaged. It takes all my control to not get cynical about that. All I can do is, once again, say I remain willing to talk and learn more.

One of the most interesting things we found in our research on police killings was that it was members of the community who called the police in almost all the non-traffic incidents (and such calls comprised the vast majority). They described the person the police actually confronted. This is actually evidence that the police did not select those people on the basis of race – they responded to calls. This is a very important finding. We cannot tell you (no one can) whether the police treated black people differently in these cases. The data does not exist. But we can definitely say that the police were not targeting based on the race of the decedent. That’s huge.

In the just over 50% of the cases, there were non-police, civilian witnesses present. Most witnesses gave a story that was consistent with the official police account – and in the very small minority that did not, they often did not dispute the actions of the decedent, but rather the level of force used by the officer. People do not understand use of force, or deadly force policy or case law, and they make really uninformed assumptions – like, “Why didn’t they shoot him in the leg?”

There’s a huge disconnect between perceptions and reality, and I think that is largely responsible for tensions between police and communities. Another part is that police are uncommunicative and tend to clam up when people accuse them of things. Rather than explain, the police retreat to one certain redoubt: no one gets in trouble for saying nothing. That’s a problem, too. We have one side with misinformation and another side with negative incentives to speak. I’ve said repeatedly that a “conversation” on policing means that both sides listen, and both sides talk[16].

Second, I don’t think my words have been ‘gentle nudges,’ and anyone who thinks I’m a shill can meet me behind the bicycle racks after school. In one recent interview[17], I said of things cops have to do to reform:

It begins with honesty, transparency and a timely release of information, which includes a timeline, recordings of 911 calls, a narrative, any video. Without that, the public has nothing but the word of police to go on. And as we have seen, police sometimes lie. Not always, not even a lot, according to our research. But consider again the Walter Scott case and the lies that Officer Slager told. That makes it easy for citizens to lose faith in the system. Sunlight is the best cure.

I also said:

Police agencies have been uniformly terrible at releasing information and data on officer-involved deaths. That must stop. Don’t tell me about integrity of the investigation. When a citizen dies and we don’t have a chief explaining within the first 48 hours everything he knows, then the public and media are rightfully skeptical. They fill in the blanks.

I’m gentle? Not so much. And I don’t think your characterization of Professor Moskos (with whom I have written an article for the Washington Post[18]) is accurate.

As for comments about ‘police culture,’ again, this is something we think we know more about than we actually know. Consider this 2004 article in George Washington Law Review by Barbara E. Armacost. It contains two important things: first, Armacost refers to “police culture,” and second, she states something I find remarkable:

In the face of outside criticism, cops tend to circle the wagons, adopting a ‘code of silence,’ protecting each other, and defending each other’s actions. If the misconduct is found to be true, moreover, their departments deem the miscreants ‘rogue cops’ whose conduct does not reflect negatively on the organization from which they came[19].

Let me review: In a scholarly paper that contains 445 footnotes, that was printed in Georgetown Law Review, a paragraph fundamental to Armacost’s thesis is presented without a single supporting citation.

It is a concept, but it is presented entirely without substantiation. There is not only very little empirical evidence of the code of silence about which she speaks, but there is also no evidence whatever of a pervasive and homogeneous “police culture[20].”

That said, I had a conversation on this the other day with Seth Stoughton, a former cop and an assistant professor at USC Law. Stoughton said that he, sadly, didn’t have the same experience with cops showing solidarity with the prosecutor. He was more used to seeing them show up to support the defendant cop (including recently with Slager, the guy who shot Walter Scott). All this is par for the course – police cultures change depending on where you are, the presence or absence of a union, rural or urban… You get it.

In fact, I submit there are perhaps as many police cultures as there are police departments; there are more than 12,000 of those in the United States, not including county sheriffs, state police and federal agencies. To me, it really seems that Armacost can, through her work on that article, be the tail that wags the dog: academic statements that, if you ask me, were founded in Armacost’s regular viewing of Starsky & Hutch.

And because of where it exists, and academia’s dirty little secret about law reviews, it is itself citable. Meaning that one can now say, in an academic work,

“…cops tend to circle the wagons (Armacost, 2004)”

and be accurate! That’s bullshit.

There is a huge difference between what we know about policing and what we think we know about policing. The fact is that, despite our certainty that we understand policing and how we are policed, most of our truth is lost in a blur of confirmation bias. Our work shows very clearly that the number of truly unjustified killings of unarmed people is dramatically lower than citizens think, and is significantly higher than cops would think. This means that both cops and citizens are being led by narrative, not data. It’s not being helped by bad research conducted by journalists and shouted by CNN and Fox and repeated by politicians.

Consider the widely reported statement by The Washington Post that unarmed black men are “seven times more likely to die” at the hands of police than are white men[21]. In that statement, the Post reporters have leaped to some rather sophomoric statistical conclusions. This statement has been repeated by many well-intentioned journalists and by Secretary Hillary Clinton, and it is demonstrably misleading and unhelpful.

The figure was arrived at through a ham-fisted attempt to control for the population disparity in the United States between black and white people: While the raw number of white people shot dead by the police is higher than the raw number of black people[22], remember that black people only comprise about 13 percent, and black males only 6 percent, of Americans. The reporters (after the release of a report by ProPublica that claimed black males were 21 times more likely to be killed by police than were white males[23], written by reporters I happen to know were warned before publication by scholars that their numbers were inaccurate and misleading) sought to adjust the numbers to represent a count.

The mistake here, of course, is that by leaving out the context of what the decedent was doing at the time of his death, the calculation becomes truly meaningless. In the words of Joseph Cesario, director of the Social Cognition Laboratory at the Michigan State University, “To adjust the raw shooting numbers on population proportions assumes that … an officer buying a cup of coffee is as likely to shoot the cashier selling him the coffee as he is to shoot a citizen with an outstanding warrant who has just been pulled over for speeding. Not only does common sense suggest this is wrong, the data do not support this assumption[24].”

Now, I didn’t choose Stoughton at random to speak with about this. His recent papers – in the Harvard Law Review and in a forthcoming paper in Wake Forest Law Review (properly cited, I might add) – reveal tremendous thoughtfulness on these very issues that make my anecdotal stories somewhat less compelling. Stoughton takes great care in mentioning that there is no “universal” police culture, but also avers that is not the issue:

…[A]lthough I attempt to distill a coherent set of principles from modern policing, policing is not one dimensional. I make no claim that the whole of modern policing reflects a universal approach to law enforcement or that the principles I identify, both those that I contend are problematic and those I believe are curative, are either overwhelmingly present or entirely lacking at any given agency or within any given officer. Despite what I acknowledge are very real variations between agencies and officers, there is a coherent and identifiable set of principles that pervade modern policing; it is those principles that concern me here[25]

Yet Stoughton also believes that the warrior mindset of police, as opposed to the guardian mindset, has taken the kind of toll that your question raises. He does a great job of describing why the cops feel as though they must be warriors:

Officers take an oath to protect society and view themselves as part of the thin blue line, but they are exposed to situations that leave very little room for faith in human decency or the value of society itself. They see families turn on each other. They see the most vulnerable members of society brutally victimized. They also see their efforts to fight crime and disorder undermined by what they perceive as legal technicalities—the rules that a corrupt society has adopted to protect the very criminals who prey on it[26].

Remember, officers see the thin blue line as a very good thing – we are the line that stands between good people and chaos; between good citizens and the sociopathic, and yes, predatory, criminals; and all of this butch imagery “…[depicts] law enforcement as standing alone, the only barrier that protects an otherwise helpless society[27].”

And in that culture, which I admit is damned near universal in law enforcement, we see the unintended consequences. If we are warriors, then we are at war; if we are at war, there is an enemy. And if we are not careful, the enemy becomes everyone who is not us.

This is all pretty metaphysical, and the fact is that I don’t have the data to argue it other than to say I recognize it can exist, I think that current thinking in the popular press is as bad as current thinking in the police press (that is to say, data-free), and I welcome the work of professors like Stoughton and Moskos and Kennedy and others who are helping us think about these things in the context of police service.

Q. You’re steering a difficult course. Your support of police reform isn’t strident enough for critics of the police, but the fact that you’re doing it at all can’t be winning you many friends among your fellow cops. How have they reacted? Do they see you as a traitor, or has there been more tacit support than we’d expect? In Chicago, cops have turned on their own when they blow the whistle on misconduct. Do you count as a whistleblower? Are you afraid of being “Serpicoed”? What are the risks a cop takes when he suggests there might be ways to change the way law enforcement functions, and it doesn’t sit well with the blue team?

A. Do I get relentless crap about being on CNN, where they called me a “Manhunt expert?” Oh, absolutely. But the conversations I have had with my fellow officers as well as cops from around the country have been very, very positive.

But the question presumes an answer that’s actually unsupported – that police think that admitting problems and pointing out bad behavior is something bad. I have literally never met that guy. I think that Americans get so much of their information about police culture and police behavior from police shows that they forget that they got their information from police shows.

Not for nothing, but I happen to know Frank Serpico a little – he is a neighbor of mine in a small town in upstate New York where my wife and I own a house. When I told Paco that my police academy in Texas taught about him with nothing short of hero worship, he literally didn’t believe it. I told him several times. Times have changed since he was a cop (I also find it very funny that the thing that seems to bother him most after all these years was the incident in which he was accused of committing a homosexual act in a precinct toilet). You will be pleased to know that he’s still a massive flirt, with a twinkle in his eye, and he still scores with young and highly attractive women.

Not only am I winning friends among my fellow cops, I have the full support of my chief, assistant chief, commanders, lieutenants, sergeants and colleagues. I circulate my articles and media appearances among a group of almost 1000 officers in North Texas, as well as the International Association of Crime Analysts, and I have literally never heard a negative word – in fact, I most often get praise and encouragement.

Wait! I just remembered a negative word. A Deputy Chief in a mid-sized city excoriated me for saying that his agency had a culture that led directly to the death of an unarmed person in 2015. He was furious at me, and treated me in the worst possible way: he treated me like a journalist.

Q. When Moskos reviewed Radley Balko’s “Rise of the Warrior Cop,” he took particular offense at the claim that cops have been known to lie. From “testilying” to making false reports, from misleading judges about Stingrays to outright planting drugs and weapons on innocent people, there are many ways for police officers to fudge the facts. Are these the outliers, as apologists would have it, or is bending the truth an endemic problem? If the former, what about the cops who bear witness to the lies of the “bad apples” and say nothing? If the latter, what can be done to stop it? Or is this just the “price” of getting the bad dude off the street?

A. So maybe we read a different review, but Peter Moskos and I both thought that Radley’s book is excellent and important. That was probably why Peter troubled to say at the beginning of his review that the book is, “fascinating and sometimes terrifying.” That is a high compliment. I’ve said publicly things very similar.

I’ve worked with Radley Balko on one large article[28] that I guest-wrote for his Washington Post column, and we’ve communicated and collaborated on others, despite fundamental differences in opinion. What Moskos said about the lying was actually quite similar to my critique about Armacost, above:

Balko asserts that most police officers regularly commit felonious perjury. Lying, he writes, is ‘routine,’ ‘expected,’ and ‘part of the job.’ He supplies little evidence for this claim – an absence that is particularly notable because the rest of his book is so meticulously researched and thoroughly footnoted.

That is a pretty darn reasonable statement – what he said was, “There’s no data.”  You know why? Because there’s no data. The idea that it is routine and expected and part of the job is, to me, an insult. My agency considers lying a fireable offense, and I support that. I know of agencies in which that is not true.

Do cops lie? Of course they do, because they are human beings. There are many occasions in which it is acceptable for a cop to lie.

Widespread systematic perjury? Please.

Q. Why are so few cops willing to even mildly criticize the state of American policing? Is there a fear that any acknowledgement that cops are less than perfect will open the floodgates? What, then, makes you willing to stick your toe into waters that are off limits to other cops? Is it your diverse experiences? Your education? Or are you just a radical by nature? Are you hoping your fellow cops will follow suit, join you? Is there any movement growing within law enforcement to sincerely change course?

A. I think this is really the same question again, sorry. In other words, I think it’s repetitive, redundant, asking the same thing.


Q. In the last twenty-five years, the Supreme Court has given cops a great deal of leeway to mess with people without violating the Fourth Amendment. Thanks to Whren v. United States, Heien v. North Carolina and Utah v. Strieff, the police get to seize people on a pretext, a reasonable mistake of law, or even conduct that’s unlawful but not “flagrantly” so. And thanks to Graham v. Connor, they enjoy exceptional deference when it comes to determining whether they used excessive force. Have the Supremes gone too far and enabled bad cops? Is it good for cops, for society, when the Fourth Amendment is this squishy? Is it a good thing that the ostensibly objective reasonableness test at the heart of Graham v. Connor hinges on the testimony of a “cop expert”? Is it just too easy to be a bad cop?

A. I recently said that policing involves small and heroic acts every day that never make the news. Have you ever considered the danger inherent in moving a vomiting, spitting intravenous-drug user from a dangerous place, giving them Narcan and bringing them to the hospital[29]? Fifty-two percent of unarmed people who died after a police encounter last year were on drugs, suffering from mental illness, or physical disability, or two of those, or all three[30]. Chief Brown’s comments after Dallas[31] were spot on: the collateral duties thrown upon cops mean we are animal control, hospitals, firefighters, mental health counselors, and cops all at once. And it’s never enough. That makes this job very tough.

To answer the question about the reasonableness test, let me talk about what we found in our study of unarmed civilians who died after police encounters.

We selected unarmed cases because they were the most likely place we would discover malfeasance.  We determined that citizens who were unarmed were the most likely group to contain unjustified use of deadly force by the cops—or at least that is the impression of the layman. As cops, we know that unarmed doesn’t mean not dangerous, but this was the best group to start with to maximize our chances of finding unjustified killings.

Based on our independent review of information available in the public domain at the time of writing our book, using the standards you just described, of the 153 cases in the database, we independently concluded that the same ten (6.5%) cases appeared to involve the unjustified use of deadly force by a police officer. We found that an additional number of cases – 9 (Singleton), 12 (Selby) and 14 (Flosi) – involved police actions that either were “partially justified,” or were considered justified but we expressed some reservations about aspects of the case or the information available for review. These are preliminary opinions, based on the limited amount of information that has been released publicly, and the authors reserve the right to return to each case should more information become available[32].

The cases with unjustified or partially justified use of deadly force highlighted for all of us the inherent value of this kind of data-driven examination of police behavior. Unless officers, administrators, civilian oversight commission members, city leaders, policymakers, journalists and activists can examine data in an open and transparent way, we can’t learn the lessons of incidents, whether they were mistakes or “by-the-book” incidents that confirm the value of, or suggest a needed update to, a policy or procedure.

That’s not a lot of cops going bad. As I said above, it’s between 6 and 7 percent. And before you say I’m biased, consider that our numbers jibe really closely with those of the Washington Post. They’re not exactly the Patrolman’s Benevolent Association. The Post said of its own research that 74 percent of those fatally shot by the police in 2015 had already fired shots, brandished a gun or attacked a person[33]. Another 16 percent of the shootings came after other potentially dangerous threats. These shootings were most commonly of individuals who brandished knives and refused to drop them[34]. Note that these analytical conclusions were not a featured story on their own in the Post’s coverage of police shootings[35], but rather buried in another.

But to get back to my point: According to the Post, that’s nine out of ten. These numbers are about right. So no, I do not think for one second that the Supreme Court has made it easier to be a bad cop, nor do I think that expecting that there actually can be a “reasonable officer” is a bridge too far. In fact, what I believe is that you can take it to the bank that more than nine times out of ten, in the worst case scenario – deadly force situations, which are the most rare in policing – we have cops doing the right thing. These numbers add up quickly on both sides of the equation: let’s ask ourselves whether 100 questionable incidents (out of the about 1000 people counted by the Washington Post) in a nation of 321 million people[36], in which there are 40 million annual encounters[37] between citizens and the 800,000[38] police from 18,000[39] agencies, rises to the level of epidemic that some have claimed.

But let’s also not forget that there are incidents in which cops do the wrong thing, through training or through malice. And let’s work on it.

Q. One of your specialties is busting received wisdom in the police reform movement, like the idea that cops racially profile when they choose whom to stop or maliciously fine black and Hispanic drivers more. In a WaPo op-ed, you contended that the first claim relies on a faulty metric (most drivers stopped come from outside the city, so comparing them to the ethnic breakdown of the city’s residents makes little sense). You also argued that higher fines for minorities is due to factors outside the cops’ control: as a result of their poverty, blacks and Latinos are more likely to be stopped for more serious offenses like driving on a suspended license than their white counterparts.

So what about stop-and-frisk, the tactic, not the legal doctrine? In cities like New York, blacks are disproportionately stopped and frisked. Because they’re pedestrians, the resident metric makes sense. And even allowing for the fact that stops and frisks are more subjective than speeding, the failure rate is immense and cops’ justifications are consistently flimsy. Is there a legit explanation for this?

A. I think that Stop and Frisk as a tactic was fatally flawed but well-intentioned. It is true that in New York, black people have been stopped and frisked disproportionately to their representation among the racial composition of the city’s population. Here’s something else: black people in New York are described as suspects by 911 callers at a disproportionate rate to their representation within the city’s racial composition.

So my question to you is, under what science have you determined that racial representation should remain proportionate to criminal actions? That’s based on too few variables. How many lawyers mug someone? How many dentists or college professors jack someone on the street?

To go back to your question: Is there a legit explanation? The court found that the practice was unconstitutional, so the program was not legit in any way. Now that they’ve stopped it, I think it’s important to recognize that at the outset, they had the best of intentions. I think that the NYPD, led by CompStat – which did help save lives and reduce crime in New York City throughout the 1990s and early 2000s – thought that it would be effective. They were wrong – despite getting guns off the street, it was unconstitutional, so the juice was not worth the squeeze.

I want to widen the aperture through which we view problems like stop and frisk. Throughout most of the neighborhoods in which there was stop and frisk, there was also crappy mental health care, very limited pre-K and early childhood care and education, highly limited resources for public education, horrible nutrition, obscenely bad health care and dental care, a prevalence (still) of lead paint, dangerous parks and declining music and art training in schools – in short, there was stop and frisk and all the conditions that sociologists have been telling us since the 1960s lead to crime.

These are not police problems, they are problems in our society. To blame the police may make us feel better, but is certainly won’t solve inequity and injustice at levels of government including education, welfare, health, parks, infrastructure and other areas.

To make a difference we really need to better understand and make more nuanced analysis of data around poverty and crime and race and policing. Writing about the connection between race and poverty and poverty and crime, Ben Singleton found[40] that, “while not all Americans who live in poverty commit crime, it is interesting that if only the demography of impoverished Americans (70 percent white, 30 percent black) is compared to the demography of criminals (69 percent white, 28 percent black), there is little disparity. Such data lends credence to a hypothesis that poverty leads to crime.”

Singleton compares the median income of the nation’s most dangerous cities ($31,775) to that of the safest cities ($84,879). And then he looks at race, and finds that in the five most dangerous cities, African Americans make up 50 percent of the population, whereas in the safest, they make up 3 percent.

Is Singleton conflating correlation with causation? Quite possibly – but he does say it is only a hypothesis. But everywhere we look, we find this same correlation – and as Eric Olson[41] says, correlation does mean correlation – each year, corporations in America make billions of dollars correlating Zip code, web pages you view and what kind of laundry detergent you buy to determine what products to send you coupons for.

It needs more study. It is nothing near as simple as, “Police are stopping too many black people,” because they compare the percentage of the population to the percentage stopped, as if the only things that matter are, “race” and “being stopped.” Those are not the only things that matter, and until we as a nation stop looking at everything as a matter of race and black and white, we simply will keep repeating our mistakes.

You want justice? Ask uncomfortable questions, use science and get the nuanced answers. They make shitty sound bites, but they work.

Q. You’re an aberration: a cop who can write about the doings of the police without resorting to the passive voice or word jumbles. Are you as frustrated as everyone else by the way police spokespeople obfuscate? Would plain speaking to the media and public mean a little more comity? Or would it lead to mindless anti-cop anger? Whether at StreetCred or in the articles you write, you try to make the processes that make cops act the way they do transparent. Should police departments themselves embrace transparency? Do they fear the public knowing what they’re up to? Is it that nobody understands a cop except another cop?

A. Look (again) at how Chief Brown handled the shootings in Dallas. I wrote in the New York Times the next day that, “Even as his officers fought terror in the streets — the worst loss of life for law enforcement since Sept. 11, 2001 — Chief Brown maintained his commitment to transparency, briefing reporters while the bullets were still flying[42].”

But another thing that I included in that article was this hugely important paragraph:

Friday morning, after our brothers were assassinated for being white and for being officers, the word was sent out: more protests are expected, and we must not interfere with them. And that is the way it should be.

I think police departments absolutely are embracing transparency, and I think that body-worn video and dashcams are helping. I’ve been calling for more transparency as loudly as I can, and the important thing is that we do it quickly, because sunlight is the best disinfectant. But I think that police communications has gotten tremendously better. I disagree that they are obfuscating – I go to public information officer conferences and I see genuine efforts to be more open, through online tools, social media and other outreach programs.

Again, I think this comes to a perspective uninformed by appreciation for the role of police in our society. It is highly important to remember the pressure that police are under when asked a question about an ongoing investigation.

Mistakes in law enforcement are not well tolerated. The wrong words can destroy a life. The wrong words can destroy a case.

The cost of wrong is awesome, in the true sense of the word. That’s why cops are not good at speaking about things until they are 100% certain of their facts – and that is exactly as it should be.

Think about any press conference you are complaining about and ask yourself whether the cop was being difficult, or whether he was trying to navigate through a set of highly complex facts and make statements that will not blow a case or destroy a life, and then we can talk more.

Q. You’ve got an unusual approach, to say the least, to the police reform movement – taking on the cops’ outdated technology as well as the outdated parts of their mentality. Do you ever feel you’ve got too much on your plate? Could departing from police orthodoxy have negative consequences for your business? Are there circumstances under which you’d shut up, fall back in line? What’s in your future? Are you a cop for good, or have you got an exit strategy? Is Dallas-area detective where you end up, or just one more step in your journey?

A. If we want to improve things, we need to invest more in law enforcement technology, especially in data capture, normalization, access, storage and collaboration in the CJIS[43] environment. We need open standards across the law enforcement, court and corrections information technology fabric, and should require vendors to provide open application programming interfaces to allow data to get out of silos and into consideration. We need to ensure that officers have the data capture tools to provide contextually relevant and accurate data about the interactions between police and the public. And we need new laws about data in law enforcement to encourage police departments to discover new patterns of data without being penalized through immediate open record requests and demands for a fix to new problems that are uncovered through more sophisticated use of data.

As for me, I will never shut up and never fall in line as long as I see that either side is lying about how we are policed. This is the best opportunity in my lifetime to change fundamentally the relationship between police and the communities they’re sworn to protect, and wasting it with a bunch of meaningless shouting about narrative is such a hindrance.

And how long I stay in policing is unknown. Six years of public service is a lot more than most people accomplish in a lifetime. I have no intention to quit at all – not while I can be valuable in helping to curb human trafficking and child abuse; not while I can be helpful in creating or evangelizing for great new technologies to change the way we explain how police do what they do. I am blessed with the opportunity to be paid to police and paid to do other things at the same time.

This is a hugely rare gift. I don’t intend to squander it.

[1] Drug Enforcement Agency, United States of America (2016) “DEA Domestic Drug Seizures.” Resource Center, Statistics & Facts, Drug Enforcement Agency, United States of America. Accessed 4 September 2016. Available: http://goo.gl/rBz3Rm

[2] Ibid.

[3] US Department of Justice (2011). “Contacts between Police and the Public, 2008.” Office of Justice Programs, Bureau of Justice Statistics, Rep. No. NCJ 234599 at 1. Accessed 5 Sept 2016. Available http://www.bjs.gov/content/pub/pdf/cpp08.pdf

[4] http://streetcredsoftware.com/PKIC

[5] https://en.wikipedia.org/wiki/2015_Chattanooga_shootings

[6] https://en.wikipedia.org/wiki/Shooting_of_Walter_Scott

[7] http://thecrimereport.org/2016/07/08/national-conversation-on-policing-means-everyone-listens-too/

[8] http://www.nydailynews.com/new-york/stop-and-frisk-continues-cops-don-reforms-article-1.2533429

[9] http://wspa.com/2016/03/14/greenville-police-chief-changes-use-of-force-policy-in-wake-of-national-outcry/

[10] http://www.latimes.com/local/lanow/la-me-ln-lapd-rules-change-20160315-story.html

[11] http://newsok.com/article/feed/1002499

[12] https://www.minnpost.com/politics-policy/2016/03/avoiding-lawful-awful-how-minneapolis-and-st-paul-police-officials-are-looki

[13] http://tucson.com/online/video/graphic-content-another-angle-of-tucson-police-shooting/youtube_068a704b-fc69-51ff-b0f0-8724bfd84b40.html

[14] https://www.washingtonpost.com/news/true-crime/wp/2016/06/24/ex-fairfax-police-officer-who-killed-john-geer-returns-to-court-friday-for-sentencing/

[15] See, e.g., my critique of The Guardian’s The Counted and its attack on Kern County police, calling them the deadliest department in the country, at https://medium.com/@nselby/kern-county-s-murderous-raping-stabbing-criminals-and-the-lethal-cops-who-stand-between-you-fc15853abdb5#.ftc4e4663

[16] http://thecrimereport.org/2016/07/08/national-conversation-on-policing-means-everyone-listens-too/

[17] http://thecrimereport.org/2016/05/31/doing-the-right-thing-2/

[18] https://www.washingtonpost.com/posteverything/wp/2016/01/15/just-counting-people-killed-by-police-wont-fix-problems-we-need-better-data/

[19] Armacost, Barbara E., Organizational Culture and Police Misconduct (2004). George Washington Law Review, Vol. 72, No. 3, 2004; UVA School of Law, Public Law Working Paper No. 03-6. Available at SSRN: http://ssrn.com/abstract=412620 or http://dx.doi.org/10.2139/ssrn.412620

[20] I don’t know how it worked out for you when you arrested a cop, but when I helped do it, the entire department stood at the trial behind the prosecutor to send to the jury the unmistakable signal that we stood against the cop we had arrested and testified against. I wrote a lot about this at https://nselby.github.io/Are-You-Certain/

[21] Somashekhar, S., Lowery, W., Alexander, K., Kindy, K., & Tate, J (2015). “Black and Unarmed: A year after Michael Brown’s fatal shooting, unarmed black men are seven times more likely than whites to die by police gunfire.” Washington Post, August 8, 2015 (Online). Accessed 4 Sept 21015. Available: http://www.washingtonpost.com/sf/national/2015/08/08/black-and-unarmed/

[22] That number refers to killings by police overall; in fact, in terms of unarmed people, the numbers are different: more blacks were killed by police than whites. One hundred and fifty three cases fit the criteria for inclusion in the StreetCred Police Killings in Context Database in 2015. Of these, two (1.3%) were Asian or Pacific Islander; 59 (38.56%) were black or African American; 22 (14.37%) were Hispanic or Latino; 5 1 (0.65%) was Native American; 2 (1.3%) were Semitic, 12 (7.84%) were of unknown race; and 55 (35.94%) were White.

[23] Gabrielson, R., Grochowski Jones, R., Sagara, E. (2014). “Deadly Force, in Black and White.” ProPublica, October 10, 2014 (Online). Accessed 4 Sept 2016. Available: https://www.propublica.org/article/deadly-force-in-black-and-white

[24] Cesario, J. (2016). “Are Black Americans killed by police more than we would expect?” Social Cognition Laboratory, Michigan State University. Accessed 5 Sept 2016. Available http://goo.gl/cAzXR8

[25] Stoughton, Seth W., Principled Policing: Warrior Cops and Guardian Officers (August 26, 2016). Wake Forest Law Review, Forthcoming. Available at SSRN: http://ssrn.com/abstract=2830642

[26] Ibid.

[27] Ibid.

[28] https://www.washingtonpost.com/news/the-watch/wp/2016/06/20/guest-post-the-low-hanging-fruit-of-police-reform/

[29] http://thecrimereport.org/2016/05/31/doing-the-right-thing-2/

[30] https://www.washingtonpost.com/news/the-watch/wp/2016/06/20/guest-post-the-low-hanging-fruit-of-police-reform/?utm_term=.4e76c777f314

[31] http://dfw.cbslocal.com/2016/07/11/dallas-police-chief-were-asking-cops-to-do-too-much-in-this-country/

[32] Selby, N., Singleton, B., and Flosi, E. (2016) “In Context: Understanding Police Killings of Unarmed Civilians.” p. 66. St. Augustine, FL: Contextual Press/Calibre Press.

[33] http://www.washingtonpost.com/sf/investigative/2015/10/24/on-duty-under-fire/

[34] Ibid.

[35] If you were among the millions of Americans who missed this important analysis, I don’t blame you; The Washington Post didn’t say this loudly. The story that announced this analysis was not by-lined by Wesley Lowery, the reporter behind the Post’s coverage of police shootings. It was almost as if Lowery didn’t want us to notice. In fact, this analysis was fairly well hidden in the tenth paragraph of an article about something else. It was, in fact, so well hidden that my own Post editor couldn’t find the reference – he challenged my assertion – until I gave him the specific URL—but there it was:

74 percent of those fatally shot by the police in 2015] had already fired shots, brandished a gun or attacked a person with a weapon or their bare hands… These 595 cases include fatal shootings that followed a wide range of violent crimes, including shootouts, stabbings, hostage situations, carjackings and assaults… Another 16 percent of the shootings came after incidents that did not involve firearms or active attacks but featured other potentially dangerous threats. These shootings were most commonly of individuals who brandished knives and refused to drop them.

[36] https://www.census.gov/quickfacts/table/PST045215/00

[37] http://www.bjs.gov/content/pub/pdf/cpp08.pdf

[38] http://www.bjs.gov/content/pub/pdf/nsleed.pdf

[39] http://www.bjs.gov/content/pub/pdf/csllea08.pdf

[40] Singleton, B. (2016). “North Texas police officer explains why poverty is the missing link in our discussion of race and police.” Dallas Morning News, August 3, 2016. (Online) Available: http://www.dallasnews.com/opinion/latest-columns/20160803-ben-singleton-missing-link-in-discussion-of-race-and-police-is-poverty.ece

[41] https://digitalwaterblog.com/about/

[42] Selby, N (2016) “Police and Protesters can Co-Exist” The New York Times, July 9, 2016, page A21. Print. Available: http://www.nytimes.com/2016/07/09/opinion/bad-guys-win-if-the-police-reject-protests.html

[43] Criminal Justice Information Services, the security framework under which law enforcement data is managed