The Death of Fact-finding and the Birth of Truth

magnififying.jpgToday’s Supreme Court decision in Scott v. Harris is likely to have profound long-term jurisprudential consequences. At stake: whether trial courts, or appellate courts, are to have the last say on what the record means. Or, more grandly, does litigation make findings of fact, or truth?

The story itself is pretty simple. Victor Harris was speeding on a Georgia highway. Timothy Scott, a state deputy, attempted to pull him over, along with other officers. Six minutes later, after a high-speed chase captured on a camcorder on Scott’s car, Scott spun Harris’ car off the road, leading to an accident. Harris is now a quadriplegic. He sued Scott for using excessive force in his arrest. On summary judgment, the District Court denied Scott’s qualified immunity defense; the Eleventh Circuit affirmed.

Justice Scalia, writing for the majority, noted that the “first step is . . . to determine the relevant facts.” Normally, of course, courts take the non-moving party’s version of the facts as given. [Or, to be more precise, the district court resolves factual disputes in favor of the non-moving party.] But here, the videotape “quite clearly contradicts the version of the story told by respondent and adopted by the Court of Appeals.” Notwithstanding a disagreement with Justice Stevens on what whether that statement was accurate (“We are happy to allow the videotape to speak for itself.” Slip Op. at 5), the Court proceeded to reject the nonmoving party’s version of the facts. To do so, it relied on the ordinary rule that the dispute of facts must be “genuine”: the Respondent’s version of the facts is “so utterly discredited by the record that no reasonable jury could have believed him.” (Slip Op. at 8).

Let’s get a bias out of the way. At the Court’s suggestion, I watched the video. I lean toward Justice Stevens’ view: “This is hardly the stuff of Hollywood. To the contrary, the video does not reveal any incidents that could even be remotely characterized as ‘close calls.'” Such a dispute over a common story immediately highlights the most serious problem with the Court’s opinion: we all see what we want to see; behavioral biases like attribution and availability lead to individualized view of events. Where the majority sees explosions, Justice Stevens sees “headlights of vehicles zooming by in the opposite lane.” (Dissent at 2, n.1 – and check out the rest of the sentence for a casual swipe against the younger members of the court.) It brings to mind the Kahan/Slovic/Braman/Gastil/Cohen work on the perceptions of risk: each Justice saw the risk of speeding through his or her own cultural prism.

But even if I agreed with the majority on what the videotape shows, the Court’s opinion is disruptive to fundamental principles of American Law. Justice Stevens suggests that the majority is acting like a jury, reaching a “verdict that differs from the views of the judges on both the District court and the Court of Appeals who are surely more familiar with the hazards of driving on Georgia roads than we are.” (Dissent at 1). There are several problems with such appellate fact finding based on videotape that the Court ignores.

First, if we take seriously the idea that experience in evaluating evidence brings skill, then appellate judges are likely to be worse at evaluating video evidence than trial judges. Certainly, there is no reason to think they are better. So, imagine a world in which the police routinely videotaped their searches and seizures of homes, and all resulting conversations with suspects. (It isn’t too hard to imagine.) Trial courts would make rulings on the admissibility of evidence based on that videotape, supplemented, perhaps, by testimony from the police and accused. Assuming that such videos would be put into the record, what kind of deference should an appellate court give a trial court on such judgments after this opinion? Current practice accords the trial judge deference because she is “in the room,” and “smells the same air” as the testifying parties. But that approach would seem to be significantly undermined by the Court’s formalation of the purpose of litigation: to determine what “actually happened.” The Scott rule starts to make trial courts into something like magistrates: useful for moving paper and effecting settlement, but ultimately not decision makers.

Second, Scalia’s opinion demonstrates its weakness by telling the reader to make their own independent evaluation of the video evidence. Opinions should stand on their own feet. In world where all opinions were freely available online to all citizens, complete with video embedding, we might not care. That world, of course, is coming, but only for those who can afford computers and broadband connections. This characterizes almost every reader of this blog, but only a minority of the rest of the population. To take only a small example, consider prisoners in the state jail. Prisons strictly control internet access, and time, on the theory that the law in the books is an adequate substitute for constitutional purposes. If most courts begin throwing questions from the page to the tape, will prisoners gain a constitutional right to computer access?

But this project is misguided anyway. The majority as much as turns its back on the courts’ ordinary role to determine legal facts, instead of the truth of the “event.” We don’t read opinions to tell us what happened. We read them to tell us what facts the courts have found. That is why innocence is not, ultimately, a legal defense: the facts found by a jury control. This separation is a necessary part of the dispute resolution system, enabling finality. The court’s opinion has the potential to destabilize this delicate regime in areas that would likely matter to some folks more than the civil rights suit of one speeder.

(H/T: Orin Kerr; More commentary at SCOTUSBlog, which notes that “this decision is a constitutional holding more than it is an essay about facts, and very likely will be applied by lower courts beyond its specific factual setting.”; Marty Lederman has charactistically great comments here.)

(Photo: Garrison Photography, courtesy of SXC.)

You may also like...

22 Responses

  1. Alice Ristroph says:

    Great post, Dave. I watched the video and share the view that the chase is not obviously so dangerous that it warrants deadly force in response. The chase is a far cry from Justice Scalia’s characterization: “a Hollywood-style car chase of the most frightening sort.”

    But I disagree that respect for the fact-finding role of juries must mean that “innocence is not, ultimately, a legal defense.” It seems to me that we can embrace juries as fact-finders and nonetheless recognize that they will sometimes get it wrong. And it seems to me that some factual questions (such as, did the defendant actually do it?) are sufficiently serious that we may sometimes need to revisit these questions even after a jury determination. These issues are distinct from those raised in Scott v. Harris, where no jury will ever get to make a factual determination about the dangerousness of the car chase and the reasonableness of the police response.

  2. Orin Kerr says:


    I strongly disagree. At summary judgment, the district court doesn’t have the power to find facts. Why should the majority defer to the district court if the district court is not actually making any findings at all? And if you’re going to just make an argument on institutional competence — assuming that you have some authority for the view that this is legally relevant, which I don’t think you do — shouldn’t you consider the relative amount of time that the different courts gave to the issue? I would guess the Justices and their clerks spent about 100x more person-hours on this issue than did the District Court. Third, I think you’re misrepresenting Scalia by claiming that he just tells the “to make their own independent evaluation of the video evidence.” He says only in a footnote, after a thorough analysis, in response to a claim that he is misrepresenting what the video shows.

  3. rick swedloff says:

    I initially agreed with the underlying premise of your post: that this opinion neuters trial courts, because the courts of appeal no longer owe their factual findings any deference. Upon further reflection, Scott v. Harris may, in a funny way, give more discretion to trial courts. As you initially wrote and as the S.Ct. explicitly stated, prior to this opinion, courts “usually” felt constrained in the qualified immunity context on summary judgment to “adopt[] the plaintiff’s versions of the facts.” Slip op. at 5. After this opinion, however, courts are given explicit permission to disregard the plaintiff’s version when it is “blatantly contradicted by the record, so that no reasonable jury could believe it.” Slip op. at 7. Maybe what the court is saying is that it doesn’t believe that either the trial or appeallate court reviewed the video tape at all, and thus made a de novo evaluation of the contents. If that is the case, I don’t know that I am as concerned about the deference questions.

  4. dave says:

    Orin: On SJ, the court actually has to make a finding – what is in dispute and what is not, looking at the parties’ briefs (the court can’t create dispute where none is raised, and visa versa). Where the parties do dispute, the court must defer to the non-moving party’s version of the facts, assuming that a reasonable jury could so find. (As to whether this genuineness “out” is wise, or constitutional, I’ll leave to another post or another author).

    The issue is whether a reasonable jury would be compelled to find that the tape precludes plaintiff’s version of events. This genuineness finding is obviously a type of fact-finding, and deserves deference. The factual nature of this inquiry can be most easily seen if you imagine that the question is one of the genuineness of a confession. The trial court looks at the tape, and based on his or her experience, says: “I’ve seen what people look like after being coerced into waiving their rights, and this isn’t it.” The appellate court (or scotus) looks at the same tape and sees coercion. To which, you respond “but they’ve spent more time on the case!” And that is right – in this case – but the problem seeps outwards, and calls into question every district court judgment. Obviously, appellate courts can’t give 100x man hours that district courts can for every appealled case, so they will give it in, what, the cases that catch their fancy? Those with good video resolution? For what it is worth, I disagree with you that Scalia’s refutaion of Stevens’ (and respondents) description of the events is thorough. The proof is in the tape, which, I think, lends itself best to the minority’s viewing.

    Rick: But the 11th circuit did rely on the tape. See note 12 in the appellate opinion.

    Alice: Great point. I think I should have been more clear: the distinction between legal facts and truth is one way that folks defend putting innocents to death. It is a bad distinction in that context.

  5. Matt says:

    The idea that a tape (especailly one like this) can “speak for itself” is a pretty silly view. But even beyond that, as far as I could tell the tape didn’t actually show the crash (which happend with a car other than the one taping, it seemed) and it was quite hard to tell how fast things were really going, how crowded the road was, etc. I’ve been in the front of police cars in chases that certainly seemed to be going faster than that tape looked and where no one even considered crashing a car to stop it. Maybe that was required in this case, but you certainly could not tell that by wathing the tape. It’s really, really odd to imply that you could.

  6. Orin Kerr says:


    I’m afraid you misunderstand my position. Obviously I am not arguing that the Supreme Court should not defer just because the Justices spent more time on the case. (???) My point is that no one is entitled to deference here, and it has nothing to do with any issues “seeping outwards” about broader questions of the role of appellate courts and district courts.

    This was really just a simple issue: the district court is not entitled to deference on what a reasonable juror would conclude except on the narrow question as to whether and what specific facts are in dispute. The district court did not make findings as to what specific facts are in dispute that are implicated in this appeal, so appellate courts could look at the record just like the district court could.

    It may be that this case touches on some broader concerns you have about the role of appellate courts. But this is a narrow case, not a broader symbol, and as far as I can tell there’s nothing particularly new or remarkable about the specifics of what the Court did.

  7. Orin Kerr says:

    Oh, and Matt: Watch the whole tape. The crash is at the end. And I believe the average speed was about 100mph during the highway portion.

  8. Dave Hoffman says:


    I’m sorry if I misunderstood your post. I thought you were, in part, suggesting that institutional competence arguments would push us to think that appellate courts ought to have the right to overturn district court fact finding based on the time they spend on cases. On re-reading, I see that you weren’t making that claim.

    I think we’re still in disagreement, however, on the effects of the case. Few Supreme Court decisions, in my view, have narrow doctrinal effects, or can be understood without thinking about the messages they send to lower courts about authority, law, and the judicial system. Here, you’ve got the court saying, essentially, that we took a fresh look at the record, and based on the videotape, which speaks for itself, we disagree that the lower courts’ view of the “disputedness” was reasonable. Justice Breyer’s concurrence makes this particularly clear.

    I am still confused as to the root of your position. Imagine a different cause of action, where the issue was, say, whether a particular remark was made in the workplace. Plaintiff claims that he was called “over the hill,” and that remark forms the basis for an age discrimination suit. The plaintiff submits an affidavit to that effect, which is disputed by the employer’s affidavit. (Assume that such self-serving affidavits get you past SJ, because there is some other documentary supporting evidence on the side). As it turns out, the employer (contra solove) is enthralled by big brother, and has the place wired for sound and video. The employer submits a video, and argues that the conversation depicted in the video contains the statement “you are over him Bill,” referring to a recent breakup. The DCT denies SJ, stating there is a GDMF on what was said. The Court of Appeals, relying on Scott, reverses, based on its independant reading of the video

    There are two ways to think about such a decision being in error:

    1. It doesn’t put enough weight on the jury trial right. One might argue that only if the video shows that the sun was rising when the plaintiff says it was setting should the case go away. And even in such a circumstance, the best argument for permitting SJ is judicial efficiency, at the trial court docket level. If the trial court thinks it has time for a jury trial, does the appellate court know better?

    2. Worse, it suggests that there is some kind of “objective truth” about the events that we can know by looking at the video, and the fact that eight justices agree on viewing the video that no jury could find Harris’ version of events credible is dispositive. This is the nub of the mistake for me. Videotape evidence, just like a deposition, or trial testimony, needs to be evaluated contextually, a job that trial judges are good at and appellate judges are bad at. Five of the judges looking at this case thought no reasonable jury could agree with Harris. Five (Stevens, Three in the Eleventh, One Distict Court) disagreed. I’m with the losing five.

  9. Matt says:

    I did watch the whole tape- but the crash didn’t seem to be caused by the car that was filming- another car comes in front of it and then the car filming comes up to the crash. So, it’s hard to see how that can really “speak for itself”. (I’d say again that it takes a pretty naive person to think any such tape could do so, but this one perhaps especially.) And around 100mph isn’t really that fast for a high-speed chase. I come from a family of police officers and have been in the car when a chase has gone faster than that. Quite a lot of people drive that fast just on their own on rural roads and in the west. The decription of the chase by the majority seems pretty silly to me. Certainly it is not obvious from the video (or the speed) that ramming the car was reasonable. It’s stupid or dishonest to think so, I think. (That it was obviously not unreasonable- perhaps it was necessary, but you can’t tell that from the tape, for sure.)

  10. Dave Hoffman says:

    For what it is worth, there were four tapes made of the chase. See the Respondent’s brief.

  11. Orin Kerr says:


    My view is that this case makes only a narrow point. A plaintiff in a civil action can’t make a claim about the facts and get past a videotape that obviously shows the falseness of his allegations so long as the plaintiff doesn’t claim the videotape is inaccurate in any way. Your hypo doesn’t work, I think, because the plaintiff is contesting the facts in the video.

    I think the true difficulty with this portion of Scalia’s opinion is that there wasn’t actually a conflict between the facts in the videotape and the facts portrayed by the lower courts. My post on that is here.

  12. Anonymous 1L says:

    Will the next supreme court reporter have an embedded video player in it?

  13. Richard says:

    Maybe I’m missing something here, but isn’t part of the standard for granting summary judgment that there are no material facts at issue? Seems to me a videotape of the sort in this case gives rise to all sorts of material facts.

  14. Dave Hoffman says:

    Orin, that is a really helpful post, thanks. Maybe I got a little misled by Scalia’s rhetoric.

    And, obviously, congratulations on the win!

  15. MJG says:

    Note – This was cross posted to the Volokh site, but the issues are all the ones Dave H. has gone into great detail about.

    This opinion troubles me. Although I agree that fleeing car suspects do pose risks to bystanders (unlike Tenn. v. Garner), I think Scalia gives insufficient (or even no) weight to the fleeing suspect’s interests in his own health. This may simply be evidence of my own shortcomings, but I have a hard time explaining why this is not a per se rule that the police may hit anyone who begins to flee, at almost any speed.

    And, whether or not the videotape really does show that the lower court’s factual rulings were flatly wrong (I do not think it does), this is still a SJ question. Unless I’m mistaken, the question is whether “no reasonable jury” could find for Harris. The majority says no. I find that a stretch. No reasonable jury could look at that same tape and say, “Hmm, Harris was going fast, but, you know, ramming him into that embankment at 90 mph and paralyzing him from the neck down was not reasonable”?

    Maybe none of this will matter because there won’t be police videotapes to “conclusively disprove” lower court findings taken most favorable to the plaintiff. But I just find it strange.

    I don’t think this was the case to grant cert on for this issue.

  16. Reader says:

    Here’s a question I have for Scalia & those defending his outrageous opinion: where does Scalia cite or deal with the restvery intent on error correction.

    What a goon.

  17. Read says:

    Oops, my comment came out wrong. Let me try again, sorry:

    Here’s a question I have for Scalia & those defending his outrageous opinion: where does Scalia cite or deal with the rest of record? The deposition testimony, the expert submissions? For instance, the sheriff of the next county over — who was in the parking lot where the Coweta County officer collided into the suspect — described that incident as the officer “ramming” the suspect. All of that material formed part of the record that the District Court & CA11 had before it as well, and informed its decision to find that a reasonable jury could consider all the evidence & find for the P. Where was Scalia considering all this?

    The video is not the sum & complete total of the record.

    I heard Scalia talk a couple years ago. Some person in the audience began describing some criminal appeal w/ claims of innocence on which the Court had denied cert. Scalia brusquely cut him off, chiding that “we don’t do error correction.”

    Interesting. When a death row appellant claims an error in the trial that convicted him, Scalia says the Court’s job is not error correction. When a paralyzed kid seeks to recover from the vicious actions of an out of control police officer, Scalia is suddenly very intent on error correction.

    What a goon.

  18. Orin Kerr says:


    I gather that you don’t like Justice Scalia, but I don’t think your criticism of him here makes any sense.

    First, whether a person who saw the contact in the parking lot described it as a “ramming” or something else isn’t relevant. Harris did not argue that anything in the videotape was inaccurate, or that any other evidence in the case conflicted with the videotape. His claim was that watching the videotape, he was not dangerous. The court disagreed.

    Second, the Court did not take Scott v. Harris to perform “error correction.” The Eleventh Circuit’s opinion was an enormously important legal decision that was forcing police agencies around the country to reformulate their policies on police chases. The Court took the case because of the importance of the legal issues, not because it wished to perform some kind of error correction. (For those who don’t follow Supreme Court practice, “error correction” refers to correcting a court’s misapplication of settled law when the correction has no broader significance beyond teh facts of that particular case.)

  19. Reader says:

    With all respect Prof Kerr, I don’t think that Harris’ claim was “that watching the videotape, he was not dangerous.” I think the claim was that a reasonable jury, considering *all* the evidence, could conclude that he was not dangerous. The evidence consisted of more than the video. The video *cannot* possibly show everything perfectly. The recollections of other eye-witnesses are important pieces of evidence. Ditto the opinions of experts, say, regarding the feasibility (or not) of the officer’s alleged attempt to perform the swiping manuever.

    Scalia did *not* consider any of that, and it’s outrageous for him to have failed to do so. How can a court *possibly* feel entitled to reverse a determination that a party is not entitled to summary judgment when considering only a *fraction* of the record before the lower courts. It’s an assault on every tenet & principle of reasoned judicial decisionmaking of which I’m aware.

    And, sorry, but I don’t understand how this wasn’t error correction with no real application to any other facts. The opinion was *entirely* about how Scalia viewed *this* video. That’s an outrageous waste of the Court’s sparse docket space.

  20. View from the trenches says:

    I submitted a comment to this post characterized by staggering genius and insight. It was to be run by the “post reviewer” or whatever it is called, but never made it. What gives?

  21. lgg says:

    Sorry…there was not enough information I could find on this case to make a ‘Constitutional’ judgment…..speeding in most jurisdictions, unless over ‘certain’ speeds is a ‘civil’ not criminal violation….and discretion given to the citizen in ‘that which is prudent..’ I don’t see anything in that tape that would justify for a ‘speeding’ violation justifying an officer then endangering that individuals life…he had the license number, no doubt, and the tape in order to pursue a charge of ‘speeding’ and/or reckless endangerment. I also would bet there were other options of road block and backup. To ram someone’s car for a civil violation which the lower courts upheld due to their familiarity with the terrain speaks of ‘injustice’ rather than ‘justice’. ‘Warning’ the citizenry by turning on the lights first and foremost was an act of enforcement. If there had been a death or injury, there would have been criminal charges. We are over the top in this country in pursuing ‘civil’ traffic violations, at the cost of using our law enforcement for the blatant property thefts, and other more detrimental crimes in this country.

  22. lgg530 says:

    I finally had a chance to watch the tape…to me this was definitely ‘excessive force…’ Most local speeding laws are worded that the speed was ‘in excess of that which was prudent and reasonable.’ This incident happened, it appears, at almost 11:00 at night…and it appears the officers were endangering oncoming traffic much more than the citizen was…and there most definitely would have been other alternatives than pushing a man into an enbancment which resulted in his loss off his limbs for the rest of his life. Speed and regulatory signs are to regulate traffic and protect the citzenry and it doesn’t appear he was drunk or driving ‘recklessly’ other than exceeding the posted speed limit (a civil violation) in any manner.