Why does Kopf believe cops most of the time?

I told Scott H. Greenfield at Simple Justice that I would answer the question posed in the title. And so I shall.

The following is not intended as an excuse. Indeed, it may be viewed as an indictment. With the foregoing keenly in mind, and in no particular order, here are some of my thoughts on why I have tended to believe cops most of the time.

  • I am a shitty judge of credibility. Truly, I am. See here for what happened when I believed a defendant and it blew up in my face with an editorial cartoon and the whole nine yards. Thus, when forced to judge between a cop and a defendant it is safer to believe the cop than the defendant particularly if a judge cares about his or her reputation. While pleading the subconscious in mitigation, there was a period of time when I really thought I might make it to the Circuit if I were a good little boy. See what happened to Judge Baer when he “screwed” up.
  • As an empirical matter, the vast majority of defendants are guilty. Therefore most defendants arguably have a more intense incentive to lie compared to the incentive of a cop.  This is particularly true when it is doubtful (save for the instant case where the cops were caught on video) that the real “truth” will ever come out.  Both the defendant and the cop know that there is little likelihood that their testimony will ultimately be proven true or false. Prison for the defendant is far worse than dishonor for the cop. Ergo, most defendants lie more frequently than most cops.
  • I have very seldom been faced with a situation where the defendant testifies and so does the cop. In suppression hearings, for example, defense lawyers keep their clients off the stand in most cases. Consequently, credibility testing on a one-sided record lacks the data necessary to make an informed credibility determination.
  • Cops are frequently armed with credibility enhancers like written consent forms, in-car or on-uniform cameras or audio receivers, drug dogs, damning Miranda interview statements buttressed by written advisement forms and more than one cop saying the same thing.
  • The cops I tend to see in federal court are very experienced and most often extremely well-trained. For example, a state trooper making an interdiction stop of a car with 20 kilos of cocaine in a secret compartment will almost always be an expert at making the stop and subsequent consent search bullet proof.
  • I was a magistrate judge for about five and half years. Making good credibility determinations when cops are involved requires practice. In the District of Nebraska, we use magistrate judges to handle all pretrial criminal motions. The district judges review those rulings on the record made before the magistrate judge. Whatever talents I had as a magistrate judge in 1992 to judge credibility of cops has deteriorated over the ensuing two decades. Frankly, this had made me cautious.
  • While I have never personally taken the MMPI (Minnesota Multiphasic Personality Inventory), I would guess that I would score high for respect for rules (authoritarianism). Put more simply, my personality, formed as white middle class child, strongly incorporates a positive but stereotypical view that we need rules and cops follow them. Even more simply, I believed the Dick and Jane books.
  • I used to represent county sheriffs in federal civil rights actions when they were sued in their individual capacities. While I also had my fair share of criminal cases, my county sheriff clients were far more honest than my clients who were defendants in criminal cases. Experiences like that are hard to erase when one looks at the world.
  • While I do not think of myself as “pro prosecution,” I deeply fear for our society because of the many predators I see on a daily basis. I suppose that if I am going to err, I err on the side of what I see as order.

Can I still function as a decent federal judge in a criminal case where a cop’s credibility is at issue? My answer is yes.

I am aware of my inclinations (implicit bias if you like) and I have done, and will continue to do, my level best to hold these impulses in check.  That is not a perfect answer, but it is the best I can do with what little I have.*


*For what it is worth, in 1984 and as an active Republican (serving as county chairman), I prosecuted Nebraska’s Republican Attorney General in an impeachment proceeding before the Nebraska Supreme Court.  Here’s a New York Times article on that affair. I add this to illustrate that I have not been hesitant to take a swing at high-ranking law enforcement officers even though the personal consequences of doing so were not pleasant.

48 responses

  1. These posts make me think of Scott’s post from a few weeks ago about what to presume when a video goes missing.


    The ability to realtime capture audio and video of what happens around an officer is a huge change in how we see witness credibility. Suddenly, and for the first time, there is a nearly unimpeachable account of the truth of what happened. Scott’s argument that when that account could have, but does not, make it to the court’s hands, that the presumption should be against the officers seems persuasive. Considering his legal argument, especially viewed in light of your candid comments here about the non-evidentiary reasons you (and other judges) would tend to believe officers, I think it would be a productive change.

    One point where this seems particularly needed is the FBI’s totally absurd and indefensible practice of not recording interviews. A recording is miles better than even the best notes taken by an agent. This piece by Harvey Silvergate in the Boston Globe mirrors my thoughts pretty closely on this.


  2. I have often wondered if judges are any better at determining credibility of witnesses than the average person off the street. I know we think we are but I doubt we are, at least if better not by much. We tend to suffer from many blind spot biases. There is an interesting study were 90% of judges at a meeting believed they were in the top 25% of the judges in the room in terms of ability. Really? 😄 Further empirical research on determining credibility and instruction on how to improve it, if evidence based, would be excellent for judicial training at all levels. I also think a healthy dose of skepticism for all professional witnesses is not too bad a thing. As the jury instruction goes it’s risky to give any class of witnesses a presumption of truth telling.

  3. Terrific. I could go on for hours with you, in a bar or the Inns of Court. For now, just one comment: almost all of them did it, but they are not all guilty of what they were charged with and some, even if they did it, should not have been charged at all.

    That being said, just remember this is a country where a person goes to jail for possessing small amounts of drugs and not for ruining the economy, putting faulty cars on the roads or polluting the atmosphere. Not putting people in jail promotes the ideals of liberty. I err on the side of justice. Order be damned.

  4. Judge Bennett,

    I read (and wrote about) your Alabama Law Review article with Prof. Robbins addressing your empirical study of defendants’ statements at sentencing. Of particular note was your findings as to the important judges place on sincerity, in juxtaposition to their self-ascribed ability to detect sincerity:

    Given the importance that judges assign to sincerity and genuine remorse, judges presumably believe that they are adept at recognizing deceit and feigned remorse in their courtrooms. But are they really? Numerous studies have indicated that humans, in general, are not as adept as they think they are at detecting sincerity and, more specifically, that judges are able to identify deceit at a rate only slightly better than chance.

    Perhaps the ability to detect sincerity is akin to the ability to detect credibility.


  5. Agreed, I think RGK’s excuses are rationalizations, at best, and downright laughable, at worst. For example, “I am a shitty judge of credibility,” yet he was good at it when a Magistrate Judge? How does he know? He may have been lousy then too, or is better than he thinks now. There is no way to tell, other than a subjective impression, which RGK admits is not too good.

    But in the end, all of his bullets are beside the point and I believe he is handling it the right way. Jurors are told to do their best to be impartial, no different for judges who also view the world through their respective experiences and prejudices. No one is totally objective or a true seer, and everyone can be fooled. Recognizing your own biases can play a role in decision making is probably the best anyone can do.

  6. Mark,

    I have quibble. Fairly read, I did not say I wasn’t shitty at credibility when I was an MJ. What I did say was that making credibility determinations gets better with practice. Put another way, I was less shitty as an MJ.

    All the best.


  7. It is a presumption, but even more to the point the logic does not work. You can’t assume that the propensity to lie is greater for defendants simply because the consequences to them are greater, you must also take into account the disincentives to lie (that is, the consequences of being caught)

    If the disincentive to lie is low enough, (and my understanding is that perjury convictions are very very rare) then the marginal utility of lies for defendants over cops is immaterial, the floor is so low that the incentive provided to cops to lie is sufficient to bring rates for cops into parity with the defendants.

    The last bullet point is a popular meme but it has no basis in reality. All empirical evidence shows that we live in one of the safest societies ever to exist on the face of the Earth. Judge Kopf needs to get out more.

  8. Neither of your points strike me as very persuasive. The discentive to lie is obviously greater for a cop — who has a profession, reputation, etc. — than for a defendant, who has almost literally nothing to lose. Plus, a cop has to worry about his credibility over time, while most defendants are only going to be before a particular judge once, or maybe a couple of times. That’s true even if the odds of either getting caught in any one case are relatively low. I’m sure your response will be the usual demonization of everyone in law enforcement as pathological liars, etc, but I don’t think that persuades many outside the Radley Balko corners of internet.

    As to the second statement, the “reality” for anyone who works in or around the criminal justice system — particularly the federal system — is a realization that, while some defendants are very unlucky and get unfairly screwed, there are others who are seriously scary people who need to be taken out of circulation. The relative safety of our society may have more than you would like to admit to do with the fact that these people do tend to get locked up for a pretty long time. Of course, that doesn’t justify dishonesty or corner-cutting in individual cases, but it makes Judge Kopf’s observation perfectly reasonable, and makes me think that you’re the one who could stand to “get out more.”

  9. There was once a judge who was retiring after many years on the bench. A dinner was being given in his honor. In his remarks, he said the following: “When I was first appointed to the bench, I was probably more inclined to be pro-prosecution and I probably convicted people who were innocent. Later, I became more cautious, and probably acquitted some people who should have been convicted. So I’m pleased to say that between the wrongful acquittals and the wrongful convictions, the mistakes probably evened themselves out and it all came out in the wash.”

  10. Pingback: Why does Judge Kopf believe cops most of the time? |

  11. All the disincentives for cops you cite only come into play if they get caught, and absent any video or audio recording, in most cases the risk of getting caught is close to zero. Irving Younger wrote about how police testimony converged on a standard account, the “dropsy’ testimony.” This should strike any reasonable observer as being highly unlikely, and this serves as evidence that yes, law enforcement lying has at times been rampant.

    “The relative safety of our society may have more than you would like to admit to do with the fact that these people do tend to get locked up for a pretty long time.” I don’t dispute that there are nasty people in the world, but science does not support this position see http://www.oxfordscholarship.com/view/10.1093/acprof:oso/9780195181159.001.0001/acprof-9780195181159

  12. To bullet #3, I often advise that they can testify and give their account of what really happened, but the chances of a judge believing the word of a defendant over the word of a cop are minimal to nonexistent. Further, after the judge rules that the cop is more credible than my client, the judge will probably enhance the sentence because he finds the defendant committed perjury.

    So yes, you may believe the word of a cop because the defendant doesn’t testify, but the defendant doesn’t testify because the defendant and the defense lawyer know you will believe the word of the cop. If a defendant could testify without fear of enhancing their sentence then maybe more defendants would testify, but I believe the overwhelming consensus of the defense bar is that there is no potential benefit to a defendant testifying, as the defendant’s word never helps, but the potential cost is considerable.

    I don’t know how to fix this. Maybe if more defendants testified then judges might start believing some, but the defendants who tried to change the culture would suffer greatly. So if the judiciary wants more defendants to testify, then the judiciary has to change the cost-benefit analysis.

  13. The FBI doesn’t record statements because they know that if a defendant testifies to one thing and two FBI agents testify differently, the agents will be believed every time.
    We have a “problem” in Texas with malfunctions of police car video recorders, even if there are no records of maintenance of those failures. I have a case right now where I subpoenaed police video that starts two seconds, yes, two seconds, after my guy passed the cop car. It doesn’t show the presence or lack of probable cause.
    I think the courts should use spoilation charges when the cops can’t come up with their videos. Telling the jury they can presume that the missing video which was in the sole custody of the authorities showed something contrary to their case would go a long way to ending missing videos.

  14. I think you’re mistaken in assuming that perjury is the only, or even the primary, disincentive to lie in court. I suspect most people feel at least some moral obligation to tell the truth when testifying under oath in a trial. And, assuming equal disincentives to lying (equally little chances of perjury, equal risk to professional career, equal feelings of moral obligation to tell the truth, etc.) then it only makes sense that the greater the incentive to lie, the more likely one is to lie.

  15. It is so refreshing to see a federal judge speak candidly in a public forum that I hesitate to be critical, but it’s hard to read this list and not conclude that, really, everything on it is just a rationalization for the only thing that really matters:

    “I deeply fear for our society because of the many predators I see on a daily basis. I suppose that if I am going to err, I err on the side of what I see as order.”

    And if, ultimately, that’s the real rationale, what does one do when he’s got a pretty good idea the officer is lying (or would, if he gave the officer’s testimony half as critical a vetting as he would his teenage daughter’s story about where she was before she came home after curfew), but there’s no smoking gun video to force that finding? He swallows the lie and lets the machine roll on. I understand the temptation. I just have a hard time reconciling giving in to the temptation with taking an oath to administer justice without respect to persons.

  16. Most people, maybe. The question is do most cops feel that they have moral obligation not to lie in court. If so, does that obligation extend to not lying in evidence suppression hearings even though the Bad Guy will walk. I don’t know the answer to that question. I do know that a lot of attorneys and judges have stated that they suspect many cops routinely lie.

    However, the actual rate of cops lying is not material to my original point, which is that simply because defendants have more to lose than a cop, it does not follow that defendants lie more often. You have to account for disincentives.

  17. It is an acceptable explanation. It is also a good reason why we need more diversity, racial, gender, socioeconomic background, and life as a lawyer experience, on the bench. And as a Republican appointee you are highly enlightened and we are lucky to have you. But we still need Democratic Presidents to do more than reach into silk stocking firms and US Attorney offices for Article III nominations.

  18. Judge:
    As with others, I marvel at your candor. IMHO, credibility assessments, by adjudicators at any level, is probably the toughest part of the job. I can’t always be one hundred percent sure that a witness is credible. Yet, with experience, I’ve come close enough to sleep well at night. Hope you do, too.

  19. What do you think of the cops and judges in the innumerable ‘dropsy cases’ of the past in NY? Do you think the cops appearing in front of you are so dissimilar from them in ethics? It seems to me that those judges would have written a very similar article too.

  20. Robert,

    In my own way, I think I have come close enough to sleep well at night too. You are very kind.

    All the best.


  21. Roger,

    Thanks for writing. You and others have used the word “rationalization” to describe my post. If my post is a “rationalization,” what is my unstated or ulterior motive? My question is not rhetorical.

    All the best.


  22. Tom,

    I tend to agree with you that the FBI should record their interviews. They should do that for the sake of accuracy. Yes, it is true that the present practice makes it difficult for defense lawyers. So what? This is not about helping defense lawyers. It is, however, about accuracy.

    I also like your spoliation of evidence argument. I confront those issues all the time in the civil context. Maybe criminal defense lawyers should read the civil cases on spoliation to get the lay of that land. Very interesting.

    All the best.


  23. Paul,

    Thanks for your very thoughtful comment.

    I do not want more defendants to testify. I fully understand why they don’t. But, when it comes to judging the credibility of a cop who takes an oath, the absence of the defendant’s counter testimony is stark reminder that it is easy to make up stuff from the safety the defendant’s chair at counsel table.

    I don’t think things are going to change regarding whether defendant’s testify.The risk/reward ratio is, and I think will always remain, uninviting.

    All the best.


  24. Mark,

    Given your work on implicit bias, and your interest in using empirical methods to get at implicit bias, I think there is another axis for you to pursue. That is, the education of federal judges on how to gauge credibility. Such a targeted approach should start with “baby judges school” and then be supplemented from time to time through the Federal Judicial Center’s judicial education efforts.

    All the best.


  25. Pingback: Incentives Revisited: Which Lying Liars Do You Favor? | Simple Justice

  26. Thanks for replying. You say you’re shitty at judging credibility. I’m shitty at psychoanalysis, even of people I know, and certainly of people I don’t know, but since you said it’s not a rhetorical question I’ll try to respond.

    I don’t think we have to look very deep for an ulterior motive here. In my state, at least, a potential juror who admits that a police officer would “kind of start with something extra” in the competition for credibility should be stricken for cause. State v. Carnes, 945 S.W. 2d 594 (Mo. App. 1997). You say you choose to err on the side of what you see as order. I don’t know you, but from reading your blog you seem to be someone who cares about doing the right thing. I suspect that you would strike for cause a prospective juror who said that when in doubt she would believe an officer because she chooses to err on the side of order. I think it causes a good man a certain amount of cognitive dissonance to admit that he does what he forbids others to do. It would be wrong for you to believe the police because they are the police, so that must not be what you do.

    In short, I suspect your ulterior motive is the same as my ulterior motive when I rationalize: to allow me to keep feeling good about myself. I believe that Robert Heinlein got it it right when he said that man is not a rational animal, he is a rationalizing animal. I believe that applies to all of us, even federal judges. That is not meant to be snark, though as I reread it it may look that way. I appreciate your willingness to engage these issues.

  27. Please excuse the several politically incorrect references in the clip. Context is a bitch when dealing fictional Hollywood history onto modern day conundrums. I hope the never ending challenge and dreams come through it.

    Keep going there judge.

    There is plenty to learn along the trail.

    P.S. Function is not a function of decent just as position has never been a function of credibility.

  28. I believe the reason why FBI doesn’t record interrogations is so that their own creative writing skills will make 18 USC 1001 a slam dunk when the defendant tells a different story.

  29. Frank,

    Perhaps. I know when the two FBI agents interviewed me for the background check prior to my nomination they made sure to remind me of the false statement statute. Chilling.

    All the best.


  30. Your Honor,

    I respectfully disagree.

    Late 1970s, Denver, Colorado, a crime-wave of after-hour store robberies were plaguing Denver Metro businesses. The thieves would steal a car; use it to ram through a store front; and quickly grab the merchandise. Smash-and Grab. Police would be alerted by the silent alarm and be on the scene in minutes. They would pull up to the crime-in-progress; pop the trunk; and take their share of the loot.

    Law enforcement became such a joke that the local newspaper of the time, The Rocky Mountain News, had an editorial cartoon showing a woman in distress calling the 911 operator, “Help! I’m being robbed!” The 911 operator replied, “It’s all right Ma’am. Just take his badge number.”

    Eventually, the Feds arrested half of the Denver Police Department including the Chief of Police. The problem is that the entire police force was in on it. And those officers who were complicit and were not charged have raised the next generation of Denver Police.

    There was a motorcycle theft and high-speed chase that started in Denver and ended well outside of Denver Police jurisdiction. Several area Police departments participated in the chase including the Colorado State Patrol. The thief crashed and died at the scene. Several Denver officers were there first but the State Patrol took control soon after. When the Colorado State Patrol rolled the body over, there were 3 pistols and two knives under the body. All planted.

    The moral of the above examples is that where there is gain, there is temptation. With asset forfeiture, there is a great deal of temptation for Law Enforcement at all levels to lie-for-gain. With monthly arrest quotas, there is a lesser, but still present, temptation for the local Police to lie and plant evidence.

    No, your Honor, I would not believe any officer ‘most of the time’.

    JJ Swiontek
    Denver, CO

  31. Lets turn things around, if in a criminal case it is the defense who has more of an incentive to lie because they are facing prison, then wouldn’t it be true as well that in a civil case against an officer that you should believe the plaintiff instead of the officer?

    In the cases I have followed regarding violation of civil rights, the judge is actually more likely to believe the officer than in a criminal case.

    I have seen numerous criminal cases where evidence is thrown out or simply dismissed because of a violation of rights but when the officer is sued in federal court, the same circumstances result in the officer getting the case dismissed. This is true even in cases where the only relief asked for was attorneys fees and an assurance that officers would be better trained.

    So, no incentive for the plaintiff to lie and every reason for the cop to lie, but the judge sides with the cop anyway.

    Our judicial system is pretty much based upon the fact that unless there is obvious proof otherwise, judges will side with cops. This extends to precedent and judicially created immunity as well. So basically you could have just stopped with the first bullet point. It is the system that forces you to trust a cop over others or else you would no longer be a judge or a judge of someplace or court with very little to do with serious criminal matters.

    To me, the first reason listed is all that really needed saying.

  32. Roger,

    Now, I get what you mean. And, I agree that most of the time all of us reason (at the subconscious level) in ways that burnish our perceptions of ourselves. But, I am not sure that truism amounts to a “rationalization” as that word is frequently understood.

    As for striking a juror who said she would generally believe a cop over a regular citizen, I would not do so. I would inquire further, however. I would ask whether she would do her best to set aside that point of view and then follow up. Assuming the juror gave the “correct” responses, I would overrule a motion to strike for cause and let counsel use their peremptory challenge.

    All the best.


  33. Tom,

    What a great line. “Bad cops are good liars.” I will remember it both because it is pithy and, more importantly, because it is likely true.

    All the best.


  34. Stilgar,

    While it is true that civil cases don’t often succeed against the cops, you might be surprised with the number of times juries hit cops with verdicts for the plaintiff. That is particularly true when you get the cop and the plaintiff on the witness stand and the jury has the opportunity to view both. As for bench tried cases, I agree that judges are more likely to believe cops.

    All the best.


  35. I admire your candor here. Does it bother you, now that you’ve written this post, to reflect on the fact that for all intents and purposes you’re doing exactly what you regularly instruct juries not to do — i.e. drawing an adverse inference from the exercise of Fifth Amendment rights?

  36. Sir, just a few observations on your points.

    *Credibility. Most people are terrible at assessing this. Kinesthetic interviewing training helps, so long as you just focus on the “tells” and not working the techniques to obtain a confession (where it can be manipulated and misused).

    *Enhancers. I only comment that handlers know how to get their dogs to alert by the handler actions. Of course, they don’t have to show reliability or false alert data, so there is no check on this.

    *Officer testimony. Almost all officers are trained how to testify in the academy and practice makes perfect. I used to get copies of predicate questions so I could anticipate what the defense counsel would ask.

    Officers in general are more honest than criminals, but the more specialized the unit (SWAT, Narcotics, K9, etc.) the more likely that the officer will play word games or outright lie. Also, as a federal judge, you’re more apt to see it than a state judge, as the prosecutions will be for (typically) more serious offenses.

    It’s important to note that the officer almost never faces any repercussions. Officers are almost never prosecuted for these issues.

  37. Griff,

    Strictly speaking, I am not drawing an inference from the exercise of the Fifth Amendment. Strictly speaking, I don’t infer anything from the lack of testimony of the defendant when I judge the credibility of a cop in the absence of contrary testimony from the defendant. However, I do weigh the evidence. Technically, those are two different things.

    All the best.


  38. Ex-cop, Law Student,

    Coming from a former cop, your observations are very important to me. Thank you for providing them.

    All the best.


  39. I can vouch for ExCop-LawStudent. He is who he says he is, and like you, his candor has proven beyond reproach. He’s never shied from giving me a good spanking when it was warranted.

  40. Pingback: 5 Cops Caught Lying On The Stand When Defense Produces Recording Contradicting Their Testimony

  41. Judge Kopf, you impeach yourself with your own words. Not only do you admit that you are predisposed to authoritarianism but you also confess that you are a “shitty judge of credibility” and routinely exercise a prejudicial bias in favor of law enforcement because “Prison for the defendant is far worse than dishonor for the cop.”

    Had you actually been doing your job, the police officers who lied in court and sent people to prison would face prison too, not mere disapprobation. You should never be allowed near a bench again, every last decision you made over the course of your career should be reversed and you should be sitting in a jail cell along with all the police you have allowed to ruin defendants’ lives. Your Nuremberg-style mea culpa brings no clarity or closure and serves only to confirm what most who have experienced the effects of this nation’s court system have already known: you will find no justice in a courtroom, just us.

  42. Pingback: Links 24/4/2014: OpenPower Foundation, Core Infrastructure Initiative | Techrights

%d bloggers like this: