'Stop And Frisk' Partially Stopped

I remember being taught in school about the "Stop and Frisk" concept, also known as the "Terry Stop," although in my entire life I've never actually encountered or witnessed it. It is chiefly practiced in big cities of the type I don't especially enjoy beyond the briefest of visits, where it is alleged to be necessary.

It would need to be necessary to be acceptable, because it sounds pretty dodgy on the surface, at least as I was taught to think about it. Apparently your 4th Amendment rights weren't being violated by being physically stopped and physically searched, including having the police order you into a humiliating stance and pat you down. This was supposed to be true even though the standard for such a search -- supposedly not a "search" for 4th Amendment purposes -- was not probable cause or a warrant, as the 4th requires. It was a lesser standard called a "reasonable suspicion."

New York took it a step further, to hold that police could stop and frisk you without any cause at all. Sounds like a Federal judge has decided she doesn't buy the expansion.
During the trial, Judge Scheindlin indicated her thinking when she noted that the majority of stops result in officers finding no wrong doing.

“A lot of people are being frisked or searched on suspicion of having a gun and nobody has a gun,” she said. Only 0.14 percent of stops have led to police finding guns. “So the point is suspicion turns out to be wrong in most cases.”
It's good to see the expansion rolled back, at least.

44 comments:

Texan99 said...

I never could see why it was acceptable when it was pitched to me in law school, either. If there's a minimal interference and convenience, I guess I can see why it would be constitutional (though I don't care for it) if it's based on minimal, but soundly based, suspicion. Stop-and-frisk for no better reason than that you hope you'll get lucky really stinks, and if it turns out to be exercised 80% of the time on dark people, that stinks, too. I'm glad it was tossed.

Bloomberg seems to think all he needs to do to justify it is to point out that sometimes they catch some contraband.

Grim said...

It's a standard I can only hope will become less popular over time.

Eric Blair said...

The "stop and frisk" tactic in New York has been seen to have reduced crime.

Sure, they're not finding much. But everybody knows the cops are looking, so they don't pack the way they used to.

I don't really support it, because it is wasting time playing with a symptom instead of addressing the cause.

But I'm seeing the Chiefs point to it as a metric of success.

Texan99 said...

I hope the question whether it reduces crime won't be the only thing we consider. There are ways of reducing crime that are worse than living with crime.

Cassandra said...

I'm with Eric. To people who don't live in a high crime area, it seems unnecessary but we're not the ones who need protecting.

I've never been a big fan of simply assuming that there's some better way (or that it's better for innocent people to live with violent crime on a daily basis than for a few young men to be stopped and frisked on reasonable suspicion and have less crime as a result).

As Eric says, things have gotten a LOT better in NYC. It's for the people of NYC to decide, not for me living in the suburbs. That's kind of what federalism is all about.

And "the people" didn't decide, here. This is essentially a disparate impact claim. I rather doubt anyone has presented actual evidence that they were stopped bc they were black.

But then this was never about evidence.

Texan99 said...

No, it's not about evidence, or about any other method of evaluating effectiveness. We can probably all agree that it was effective. From a purely statistical point of view, we could improve the crime rate by authorizing the government to exterminate all black males under the age of 40. We're not entitled to focus only on effectiveness, not even out of compassion for people who are more victimized by crime than ourselves.

Grim said...

I think Tex is quite right here.

It's for the people of NYC to decide, not for me living in the suburbs. That's kind of what federalism is all about.

Well, it is, but not because you live in the suburbs -- because you live in Maryland.

However, you raise a point that I've been thinking about a lot lately. Increasingly federalism doesn't really capture the chief divide in American life, which does have to do with the question of whether you live in dense cities (which trend blue, high-crime, and low social trust) or in less-dense arrangements (which trend red, low-to-vanishing-crime, and are full of churches and clubs).

You really need very different social structures, including especially laws, in those cases. We don't have a mechanism for making that division formal, the way that the federalist principle divides the Federal government from the states. Typically states can impose mandates freely on counties/cities, indeed to the point that they are free to create or destroy them.

I was reading that somewhere in rural Ireland, they were doing away with DUI laws to enable people to drive into the local town and drink at the pub. Density was low enough that the one car on the road wasn't very much more dangerous even with a drunken driver, and they thought there was greater harm in the damage to the social fabric of losing pub life to having people drinking at home alone.

That actually sounds pretty sensible to me. You couldn't make it work in a big city, though: it doesn't scale. Dense traffic vastly multiplies the effects of the slow or bad decisions of one drunken driver.

Grim said...

The danger of not making that division formally, I should add, is that you get a race to the bottom. There are two ways this can happen.

One of them -- which I will call "Cassandra's Race," not to imply that she favors it but because it's the race that threatens on her side of the argument -- is that the law sets the standards for everyone according to the needs of the place(s) where things are worst. That means a race to the bottom in the sense of collapsing political liberty and personal freedom, even though in places outside the worst places such a loss of liberty is completely unwarranted.

The other race -- "Tex's Race," for the same reasons -- is that you race toward the protection of political liberties because you value them over the lives of distant people in cities you don't inhabit anyway. Their inability to treat people in cities more like presumed criminals means that many of the actual criminals in those cities will prey on the weak. That will result in a race to the bottom of practical liberty for many citizens: their political liberties will be safe enough, but they won't be practically free to walk to the store without getting robbed.

The division preserves the opportunity to maximize both political and practical liberties as much as is possible in the two cases.

Texan99 said...

I don't object to treating people like presumed criminals if there is a principled way to distinguish them from law-abiding citizens before the search begins. The problem with stop-and-frisk, if carried too far, is that it gives up on the notion of being able to make that distinction, and settles instead for randomly intruding on everyone just in case. That makes total sense from a pragmatic point of view; it's the same approach followed in customs like random drug tests on limited populations that can be deemed to have consented to the intrusion. But it's a bad approach to let the police follow for the general population, even in crime-ridden areas, because there's no end to it. The same principle would permit the police to search your house randomly, because statistics show that such random searches turn up really dangerous stuff in x percent of cases.

We have to balance the benefit to society of catching the occasional contraband against the danger to society of letting the police intrude randomly. There was a good reason that the Founding Fathers set such store on the concept of unreasonable search and seizure, and that the courts have constructed a careful system of "probable cause" before an intrusion can be warranted.

It's true that crime-ridden neighborhoods are unusually at risk of the nasty contraband that we'd like to find, but at the same time they're unusually at risk of intrusion from police who have carte blanche to search and seize, unfettered by inconvenient standards of probable cause.

There are tons of gradations of probable cause, with exceptions for exigency. I'm not arguing that the police can never do anything without a cumbersome hearing before a judge. I just think they ought to have to articulate something more convincing than "it's a bad neighborhood" or "my experience tells me that black and Hispanic young men often are up to no good" or "I know for a fact that x percent of people in any random neighborhood will be found to be carrying illegal weapons, though I can't say which ones."

On the other hand, I totally agree with Cassandra that disparate-impact standards are hazardous. It's a tough line to draw, when the police can't really say what it was that caused them to frisk these guys and not these other guys, and you observe that it always just seems to happen the darker ones looked more friskable. It's actually one of the few areas where I think the disparate-impact analysis makes sense. The better the cops are at articulating some other reason for the decision, the less I'd want to look at the disparate impact.

Yu-Ain Gonnano said...

The problem, Tex, is that if 80% of the perpetrators are "dark skinned" then 80% of the stop-and-frisk being "dark-skinned" would actually be race-neutral. If 90% of the perpetrators are "dark skinned" that would actually show an abundance of caution with respect to race.

The 80% is meaningless by itself.

Somewhat similarly, judging whether a search could be reasonably founded you would need to know how often a truly random sample would have yeilded a weapon. Your "reasonable suspicion" method should drastically outperform the random search. I'd love to have data, but a 1/1000 sucess rate sounds pretty poor to me for a high crime neighborhoods. Hell, in some areas, a random search might have been better.

Texan99 said...

It wouldn't be a disparate impact if it were in line with the background population. If the S-and-F is limited to dark-skinned areas, the police are responsible for articulating a reason for that scope without simply citing race as an aggravating circumstance.

I don't think it's right approach to consider how often they find a weapon. Again, it's not the efficacy of the search that's in question, it's whether the police had probable cause beyond assumptions about race: that is, something specific about behavior.

We have examples of societies in which the police can accost you at any time and demand to see your papers just on general principles. Preventing crime is a great thing, but it's not the only thing.

Cass said...

Again, it's not the efficacy of the search that's in question, it's whether the police had probable cause beyond assumptions about race: that is, something specific about behavior.

This is nearly impossible to prove. The only "proof" we have that something "bad" is going on is that more black people are being stopped and frisked than their proportion of the general population.

That's not proof that there was no behavioral reason for the stop/frisk. It's an indirect accusation.

Extrapolating from a relatively harmless tactic to exterminating a whole race makes no sense.
Everyone loves the slippery slope when they're arguing against something and abhors it when they're arguing for something. Where we draw the line in balancing the rights of law abiding citizens against the rights of criminals (or the rights of people who are stopped improperly - and we don't actually know what that number is - against the rights of people who would be robbed/raped/murdered/drive-by shot if LE was less effective) isn't a simple task, and shouldn't be reduced to "if you're willing to frisk someone and let them go, why not kill them all (even the folks who weren't even stopped).

Texan99 said...

It's true that I don't know in detail what happened in court, but the way it's supposed to work is that the cops articulate their grounds for choosing whom to frisk as a way of defending against the charge of unconstitutional action. It's very possible that they did so here and the court just refused to listen. If so, they should win on appeal. If, on the other hand, all they can say for themselves is that they were randomly stopping people because that's proved effective for them, then if I were in charge of the decision, they'd lose.

I got the impression from news reports that they were doing random screening, but I'll be the first to admit I haven't read up on it closely and that news reports are unreliable, anyway. If they had reasonable criteria for stopping specific people and are now being harassed because the results weren't race-neutral, then I'm of your way of thinking entirely. I'm not against police searches, only against searches without specific probable cause.

Cassandra said...

I don't object to treating people like presumed criminals if there is a principled way to distinguish them from law-abiding citizens before the search begins.

This is a great point, because I don't think there is one. And the logical implication of such a requirement would be to hamstring cops by imposing a set of arbitrary rules made by people with no experience in law enforcement.

Which is pretty much everything we hate about big government.

Here is all the 4th Amendment actually says on the subject:

The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.

What is "reasonable" search and seizure, though? Imagine a situation where aliens land on planet Earth and can take on the appearance of humans. The only way to tell a real human from a soul-sucking alien who will eat your face off is to pat him down or something similar.

Would we really say that it was "unreasonable" to pat people down? It's a balancing test - change the circumstances and you change the calculation.

This is what bugs me so much about Constitutional absolutism - in order to say what violates the Constitution, you absolutely have to define "reasonable".

Cassandra said...

Not saying anyone here is being an absolutist - I'm sandwiching comments between soul sucking spreadsheet nonsense and testing :p

So please read my quick comments charitably as only half my brain is working...

Yu-Ain Gonnano said...

It wouldn't be a disparate impact if it were in line with the background population.

I don't agree. We already know that crime is not in line with the background population, so I see no need for the searches to align with the background population. If, they could be right a large percentage of the time, it should align with the offender population, not the population at large.

Cassandra said...

FWIW, I do see the danger in stop and frisk. I read an article early this morning that made some of the stops sound frivolous or even malicious, but the same article at times also made the guy complaining about it sound unreasonable. Which only underscored for me the difficulty of designing top down policies that don't do more harm than good.

The same is probably true of bottoms up policies, but I tend to trust the folks actually doing the job more than I do some federal judge. That's my bias :)

Texan99 said...

C, you don't think there's a principled way to distinguish between criminals and law-abiding citizens before the search occurs? I have to disagree with you there. There's no way to be sure, obviously, but the whole jurisprudence of probable cause (or "unreasonable search") is about the many principled ways police can use objective observations to distinguish between someone who should be subjected to a search and a random citizen plucked out of a phone book.

It's never been the constitutional tradition of this country that police can search just anyone at any time. Nor, on the other hand, do the police have to prove that someone is a criminal first. They just have to be able to explain what about this person, in these circumstances, distinguishes him from people at large--whether it's his behavior, or a tip from a credible source, or whatever. Courts use a sliding scale, so that a minimally invasive stop-and-frisk requires only a slight cause, while a body-cavity custodial search demands much more.

YAG, I don't think the right test is whether the searches align with the racial profile of the criminal population. The right test is whether the searches are triggered by objective criteria in the absence of race. If they are not, and the searches fall more heavily on a suspect class (such as a racial minority) than would be suggested by the racial makeup of the whole population, the court should cast a cold eye on them.

It may be that certain criminal behaviors are overwhelmingly more present in one race than another; but even so, then it should be possible to trigger searches on the basis of some kind of marker of behavior other than skin color. It's never going to be a good idea to say "most druglords are black or Hispanic, so we'll frisk all the blacks and Hispanics and leave the Caucasians alone"--even though, statistically, that might yield very good results from a law enforcement perspective.

Cassandra said...

C, you don't think there's a principled way to distinguish between criminals and law-abiding citizens before the search occurs?

I'm not sure there is, in the sense that a police officer could say definitively "this person looked furtive to me" or could explain a gut feeling that something just wasn't right about a particular person. It could be any number of small things that combine to make a person stand out and an officer could very well not be able to articulate it.

I have had that experience with guys who approached me: some combination of factors either says, "This person is not a threat" or "This person is someone I don't trust". It's not always rational, nor do I believe such instincts (often based on long experience) can be encapsulated into law.

Cassandra said...

There's no way to be sure, obviously, but the whole jurisprudence of probable cause (or "unreasonable search") is about the many principled ways police can use objective observations to distinguish between someone who should be subjected to a search and a random citizen plucked out of a phone book.

But probable cause is a higher threshold than reasonable suspicion. It requires more evidence to support. That's a big part of it.

Do we really want to write into the law that only probable cause is sufficient?

Cassandra said...

It's never going to be a good idea to say "most druglords are black or Hispanic, so we'll frisk all the blacks and Hispanics and leave the Caucasians alone"--even though, statistically, that might yield very good results from a law enforcement perspective.

But is this really what's happening? How many Black toddlers or grandmothers or women are stopped and frisked? Clearly we're not talking about race as the sole criteria. If it were, a cop would be equally likely to stop any person who was Black, regardless of other characteristics. It's not as though kids never beat anyone up or rob them. And it's not as though women do, either.

But I'm guessing neither kids nor women are stopped as often as young men. Is that profiling?

Texan99 said...

I was being careless in my terminology. "Reasonable suspicion" is indeed what we call it when it's at the very weak end of what I think of as the probable-cause spectrum. But constitutionally, RS is one thing and PC another, and you're correct that RS is the right standard for a frisk. Nevertheless, even RS is more than an unarticulated hunch. Private people are completely entitled to rely on hunches, but that doesn't mean it's a good idea to let police do it -- or, at least, they're totally free to do it as part of their private sizing up of a situation, but if they want to advance to searching or even frisking, I think we should require more of them. It's not that hard to articulate what it was about a suspect that triggered the desire to search; police are used to it.

It's true that the criterion is never purely race, but if the other non-race attention-triggers raise concern only when coupled with a particular race, then it bothers me. The police should not randomly frisk every young black man they see and no white young men. It's OK with me if they're going to frisk every young man they see who appears to be engaging in clandestine hand-offs on the street corner while keeping a lookout, even if that standard happens to nab 90% blacks in a neighborhood that's got a 50/50 black/white mix.

Grim said...

Do we really want to write into the law that only probable cause is sufficient?

After Illinois v. Gates, 1983, that standard is down to 'a substantial chance,' but not a 'better-than-even' standard. Meaning you don't even have to get to a preponderance to get to probable cause, the way the courts are cutting the government ever-greater authority. If we don't even have to demonstrate 'a substantial chance,' say 30%, well.

Tom said...

It seems wrong to me that one restriction of citizens' rights (stop & frisk) was the solution for problems possibly caused in part by an earlier restriction of citizens' rights (denial of the right to bear arms).

Also, do we know that this particular policy actually reduced crime? If we can't show that, then is it still reasonable?

Grim said...

To some degree, that depends on what you mean by "crime." If you regard restrictions on citizens' rights to bear arms as illegitimate, as you say, then a lot of the NYC reduction falls to the side -- after all, they aren't really reducing legitimate crime, they're reducing citizen use of their natural rights.

douglas said...

Given the preponderance of concern over police powers around here, is it reasonable to suggest increasing police patrols in lieu of Stop and Frisk tactics to maintain the peace? Does that carry risks greater than enabling police to S.a.F.? I'm assuming here that the S.a.F. are for reasons beyond race and have achieved the level of 'reasonable'.

Cassandra said...

The police should not randomly frisk every young black man they see and no white young men.

I'm not sure this was even alleged, but I agree with you.

It's OK with me if they're going to frisk every young man they see who appears to be engaging in clandestine hand-offs on the street corner while keeping a lookout, even if that standard happens to nab 90% blacks in a neighborhood that's got a 50/50 black/white mix.

Again, I agree.

This reminds me a lot of the Zimmerman trial. Most of the conversation is about rather extreme scenarios that as far as I can see, were not actually alleged in this suit.

If we want to evaluate the reasonableness of SaF, then we need to discuss it as it was actually implemented across the board, not as the worst of worst case scenarios. Every single law out there can be abused, but the presence of abuses doesn't necessarily mean the law should be abolished.

Any law implemented by flawed humans will sometimes be flawed. The question in my mind is, "Do the harms outweigh the benefits?". That's not actually a totally objective question, as some people would rather see 10 innocent people killed than ten innocent people frisked and allowed to go on their merry way. I don't think any of you fall into that category (this is a worst case scenario itself, which in my mind shows that they can be useful to illustrate a larger point, but aren't the larger point).

Texan99 said...

I wasn't talking about the case, I was talking about policy.

I agree there's a cost-benefit analysis. I probably have much stronger feelings than you about the dangers of a state that's given too broad powers to search and seize. I think it's worked out badly for every country that tried it, and it was a big part of the American experiment in the 18th century: making sure the government couldn't intrude into private lives whenever it thought it would be best. Few governments can behave well with that temptation; they always seem to escalate as they see one more wrong they could cure if they could just get more control.

Grim said...

Once again, I agree with Tex.

Yu-Ain Gonnano said...

YAG, I don't think the right test is whether the searches align with the racial profile of the criminal population. The right test is whether the searches are triggered by objective criteria in the absence of race.

The problem is that the better the test does conform with objective standards the closer the distribution will align with the criminal population.

Both racism and a proper objective test can give the exact same distribution.

Texan99 said...

That's OK with me. I just want the rationale to be capable of being stated without reference to race. It's only when the rationale is entirely subjective and inarticulable that I worry about disparate impact.

Yu-Ain Gonnano said...

Got no problem with that. I'm just explaining why "OMG, 80% of those stopped have dark skin" falls on deaf ears.

To you, it sounds like disparate impact. To me, it sounds like their standard very well might be working properly.

The number, by itself, is completely meaningless.

Texan99 said...

It's not the disparate impact that gets my attention, it's the randomness of the decision whom to stop and frisk and the inability to articulate a standard that's not based on race. As soon as I see an articulable non-racial standard, I lose all interest in any disparate impact.

Yu-Ain Gonnano said...

it's the randomness of the decision

But that's just it. It doesn't look random to me.

Texan99 said...

Can you put into words the pattern it followed? That's all I would ask of the police.

Cassandra said...

... it's the randomness of the decision whom to stop and frisk and the inability to articulate a standard that's not based on race.

You mean they can't tell you when they stopped beating their wives? :p

The police aren't articulating a standard based on race. The problem is that opponents of SaF are claiming they have no other standard (a claim for which they have exactly NO evidence other than a disparate impact argument).

Texan99 said...

What is the standard that the police are articulating?

Cassandra said...

Let's ask Ray Kelly:

"The fact that they often do not lead to arrests or summonses misses the point. When a police officer stops and makes inquiry of an individual about to burglarize a location the officer has stopped a burglary. When officers stop and make inquiry of young men about to strong-arm a bodega owner as he leaves his store late at night they've stopped a robbery or perhaps worse.

"But obviously stops for suspicious behavior do lead to good arrests as well. We've taken about 8,000 guns from individuals over the last decade this way. Something our critics sneer because the number is 'too low' compared to the number of stops, despite the fact that stops average, as the Mayor said, less than one per week per officer on patrol.

"On Saturday a man threw a duffle bag into the trunk of a double-parked car in Washington Heights. When officers approached the man, he ran. That's suspicious behavior. They found $750,000 worth of heroin in the trunk. They arrested the suspects and spared the untold misery that three-quarters of a million dollars worth of drug addiction can cause to families who can least afford it.


Read more: http://www.nydailynews.com/news/politics/new-york-police-commissioner-ray-kelly-comments-stop-and-frisk-decision-article-1.1424689#ixzz2c2movS1S

FWIW, if the police are taking legally purchased guns from people, I'd have a problem with that. But as has been noted many times by gun rights activists, most urban crime is committed with illegally purchased or stolen weapons.

Texan99 said...

I have no objection to stopping and frisking a man who runs when the police approaching him while he's stuffing a duffel bag into a car. That's an articulable suspicion.

Cass said...

I didn't think you did, but I'm trying to make a point here. Grim, for instance, wants to refer to these as "baseless searches", but what Kelly is describing really isn't a baseless search.

Most people would oppose "baseless searches". The real question is whether this is what's actually happening.

At this point, the usual rebuttal is to point to a few instances that clearly violate even the NYPD's guidelines. But that isn't really a valid measure of whether the policy as a whole is objectionable, because it violates the policy.

Sexual assaults are occurring the the military. Does that prove the military's policies in this area are wrong? No, it proves that people break rules. Even when they're repeated reminded not to.

If the argument here is that, even though the police have repeatedly stated that racial profiling violates their stop and frisk policy, and even though their police force is very diverse (like Montgomery cty MD, where a study of "driving while black" stops showed that black police were stopping black drivers at the same rate white police were), disparate impact somehow proves racism?

Texan99 said...

I'd be hard pressed to say that anything but openly stated racism proves racism; normally we have to rely on a kind of indirect evidence, in the form of a pattern of racial impact combined with a complete absence of any credible non-racial explanation.

I think it's dangerous to authorize police to stop and frisk or search people unless they can articulate a reasonable suspicion--and it should go without saying that the suspicion must be something entirely independent of race, just as it should be independent of any purely personal animus a policeman might hold.

An astute and experienced policeman may in fact have a pretty good track record of simply glancing at a guy on the street and intuiting whether he's likely to be a criminal without being able to explain his reaction before or after. It is nevertheless a dangerous practice to allow frisks/searches on no other basis than that. It is bound to lead to abuse, because no objective standard can be applied to it, and therefore any personal animosity of an individual policemen will be refractory to discipline or control. Racist abuse is one of the likely forms for this kind of abuse to take, and happens to be one concerning which there is a rich constitutional jurisprudence, but racism is not really the point: unpredictable personal arbitrariness of state power is.

Grim said...

Grim, for instance, wants to refer to these as "baseless searches", but what Kelly is describing really isn't a baseless search.

That's bait-and-switch. I was talking about the TSA, who themselves defined their searches as 'suspicionless searches.' Kelly gives an example of a case with actual probable cause.

Eric Blair said...

Well, it's baseless, because the damn drug shouldn't be illegal in the first place.

George Schulz (yes that one) told Moynihan as much in 1969, and it's still true today.

Your cure is worse than the disease.

Ymar Sakar said...

Not even the Japanese would randomly or on suspicion go around touching males and females out on public avenues.

Interviewing them and questioning them, or holding them in detention/interrogation cells, of course. But only time is wasted then.

I suspect the NY police unions are behind on this.