Judge Kozinski on juries

Judge Alex Kozinski of the U.S. Court of Appeals for the 9th Circuit (official photo).

On Tuesday, I began serializing Criminal Law 2.0, a new article by Judge Alex Kozinski—for whom I clerked 20 years ago, who is one of our nation's most prominent appellate judges and has long been seen as on balance a libertarian-ish conservative (appointed by President Reagan). The introduction gave 12 reasons to worry about our criminal justice system; yesterday's post discussed wrongful convictions; this post discusses the jury system; and I'll post other parts of the article in the days to come. I've added some paragraph breaks and removed the footnotes (which are available in the PDF version), but otherwise this is as Judge Kozinski wrote it:

Lawyers and judges are inculcated with the notion that the system works well and there is nothing to worry about. And perhaps it's true. But there are far too many uncertainties for us to be complacent. Criminal trials as we know them were developed centuries ago at a time when life and technology were very different. The process has remained essentially unchanged since time out of mind. While we have much experience with the process, we know very little about how well it works. We tell ourselves that the system works, and we really believe it, but this is largely based on faith. When all is said and done, we have only a guess.

Below I offer some suggestions on how the system might be improved and validated. I do not suggest how these changes are to be implemented: Some may require legislation; others a change in judicial practices; still others constitutional amendments. Nor do I insist that all my suggestions be implemented immediately. Some may deserve closer attention, and some should be delayed while others are accelerated. There may well be good reasons that my suggestions are unworkable, and perhaps others will come up with better ones. If my proposals raise controversy and opposition, leading to a spirited debate, I will have achieved my purpose.

A. Juries

Juries matter. They obviously matter in the relatively few cases that are actually tried to them, but they also matter in the multitude of cases that are pled or settled. To the extent the jury is viewed as an unpredictable, erratic force, it increases the uncertainty of the outcome and thus considerably raises the stakes for the parties, especially criminal defendants. If a prosecutor can make a credible case that a jury might return a verdict calling for life without parole, he is very likely to extract a plea deal involving a "mere" 15 or 20-year sentence.

Most judges, especially trial judges, express satisfaction with the operation of the jury system. I've heard judges say that they seldom or never think juries reach the wrong outcome. I am skeptical of such claims.

To begin with, judges don't know any better than anyone else what is the correct verdict in a case. The most they can say is that they would have reached the same verdict as the jury. But judges are not usually called upon to make findings when they are presiding over a jury trial; their function is to determine whether there is sufficient evidence to support a guilty verdict, a process which presupposes that the prosecution's witnesses are believed by the trier of fact. This is a very different and much less rigorous process than figuring out who's lying and who's telling the truth, and I doubt that judges routinely go through that process in parallel with the jury. I certainly don't.

Actual observation of behavior in the jury room is rare, but it does exist. As cameras have become smaller and less obtrusive over the last quarter century, we've had several instances where we have been able to observe jury-room behavior. [Footnote: TV cameras have entered the jury room on at least three occasions. First, in 1986, PBS aired a broadcast showing footage of deliberations in a Wisconsin criminal trial. Then, in 1997, CBS aired a 2-hour documentary consisting of footage of jury deliberations in four Arizona trials. Most recently, in 2004, ABC aired a 7-part TV series following six homicide cases from the pretrial stage to the jury deliberations and final verdict.] The results are not particularly reassuring.

There is at least one case of documented jury nullification, with every juror expressing the belief that the defendant was guilty yet acquitting him nevertheless. In another case, jurors misconstrued the judge's denial of their request for a certain transcript to mean that they were not entitled to any transcripts from the trial.

And in a high profile murder trial, one juror experienced aggressive pushback for expressing skepticism of the defendant's guilt because the medical examiner had switched her diagnosis from accidental drug overdose to homicide only after listening to a tape recording where the defendant said "I got away with it." The juror asked the judge (unsuccessfully) to excuse her from the jury because she was uncomfortable with being pressured, and eventually voted with the majority to convict the defendant while protesting that she was doing so in a "bullied manner."

Worse still, in deliberations after the penalty phase of the trial, a different juror expressed a complete change of heart from the jury's guilty verdict the day before, emphatically maintaining that she never believed the defendant committed one of the two murders of which he was convicted. In short, "[e]ven with the camera rolling, jurors compromised on verdicts, allowed personality conflicts to interfere with the deliberations, and oversimplified the judge's instructions."

Anecdotal accounts tend to support this view. I always debrief my jurors after they return a verdict (I've never had a hung jury) and try to get them to talk about what happened in the jury room. Some of the comments seem entirely rational, but much of what jurors describe looks like a fun-house mirror reflection of Twelve Angry Men.

There was one case I remember where the jury acquitted despite what I thought was an iron-clad prosecution case. I was a bit shocked and entirely puzzled about what had happened. When I debriefed the jury, I got somewhat muted responses from most of the jurors but one gentleman, who turned out to be the foreman, had very strong views. He thought the government was wasting taxpayer money in prosecuting this defendant who had been caught red-handed with a suitcase full of some 10,000 Ecstasy pills just imported from Europe. The foreman was a large man and quite vociferous—almost belligerent—about it.

Of course, during voir dire I had asked the usual questions about whether any of the panel members had philosophical objections to our drug laws, and he had answered in the negative. The reality was different; he had strong objections to the war on drugs and managed to pull the jury with him.

That one strong personality can dominate the jury room is consistent with my own experience. I've sat on two (state) juries. One of the cases was not close by any measure, but the other one hinged on the testimony of a single witness, as there was no physical evidence whatsoever. The police had not even managed to recover a large bag of coins that the accusing witness claimed he had handed over to the defendant during a store holdup, even though the defendant was apprehended within 20 minutes of the robbery, just a couple of blocks away.

Having been elected foreman, I spoke after every one of my fellow jurors had expressed the view that the defendant was guilty. I reminded my colleagues of the prosecution's heavy burden of proof and questioned whether the complaining witness's identification could be trusted given the missing coins. If the defendant was, in fact, the perpetrator, he couldn't have spent a bag of assorted coins in the time it took to apprehend him, and he couldn't easily have hidden it when he saw the police approaching. And wouldn't he have gotten much farther from the scene of the crime if he were carrying a bag of stolen coins?

One by one, all but one of the other jurors switched their votes to acquit. The one exception proved impossible to budge so we eventually asked the judge to declare a mistrial, which he did.

The simple truth is that our confidence in juries rests largely on faith, and our processes are not designed to help us improve the functioning of the jury because there is no systematic feedback mechanism to help us figure out what works and what doesn't. I've recently suggested that we introduce cameras into the jury room, and I will not rehash the arguments I've advanced in support. Suffice to say that viewing what juries do in actual cases will give us a much better understanding of jury behavior and provide valuable information for different techniques in presenting evidence, instructing juries and jury management.

Seeing what juries do in actual cases can also ameliorate or eliminate the endless speculation about which trial errors are harmless and which are prejudicial. Why shoot in the dark when a man's liberty or life is at stake? The same is true where there is a claim of improper juror conduct. In such circumstances, the trial judge will hear conflicting accounts about what happened in the jury room. Wouldn't it be better to see and hear what actually transpired?

Videos of jury deliberations could be sealed and preserved for viewing by researchers only after the case is final. Or they could be made available to the trial judge and a reviewing court as needed to resolve questions involving the jury's conduct. Or they could be made available to the lawyers immediately after the verdict. Disclosure could be implemented incrementally over time and rolled back if the process is found to interfere with the jury's function. But we need to get a close look as to what's going on in the jury room before we can even begin the process of meaningful reform.

In the meantime—or in addition—I offer the following suggestions for reform:

1. Give jurors a written copy of the jury instructions. Jury instructions are often lengthy and difficult to follow. Jurors are expected to absorb them by listening, which is probably the worst way to learn new and complex subject matter. Many judges try to ameliorate this problem by sending a copy of the instructions into the jury room when the panel retires to deliberate, but some judges refuse to do so.

It should be reversible error for a judge to fail to send a full set of jury instructions with the jury when it retires to deliberate. Pre-instructing the jury on key concepts and giving them those instructions in writing is also a good idea.

2. Allow jurors to take notes during trial and provide them with a full trial transcript. Most judges now allow note-taking and provide writing materials for the jury to use, but a minority refuse to do so. This should be reversible error. Consulting notes during deliberations is immensely useful when the jurors' memories differ as to what a witness has said. Forcing jurors to rely on their recollections alone exacerbates the distorting memory effects discussed above.

In fact, I would go a step farther and give jurors transcripts of the proceedings to consult during deliberations. This was not possible when transcripts had to be transcribed laboriously by hand. But real-time transcripts are now pretty much standard and available for the judge and lawyers to consult while the trial is going on. I can see no justification for keeping jurors in the dark.

3. Allow jurors to discuss the case while the trial is ongoing. Most jury trials now start with a stern admonition that jurors not discuss the case until they are sent out to deliberate. It's unclear why we do this except that we've always done it that way. When I served as a juror, this restriction seemed unnatural and counterproductive. My guess is that it exacerbates the distorting effects of memory. Allowing jurors to discuss what they've heard could give them a chance to express doubts and to remind each other of the need to keep an open mind.

4. Allow jurors to ask questions during the trial. I've been doing this for some years in civil cases and it seems to work well. I ask jurors to put any questions in writing and hand them up to me. I then share these questions with the lawyers and let one or both use them during their examinations. Other techniques are possible, including having jurors pose questions to the witnesses directly and letting the lawyers follow up in light of the answers.

5. Tell jurors up-front what's at stake in the case. In most jurisdictions, jurors in non-capital cases are not told what the likely punishment will be if the defendant is convicted. In fact, we tell jurors not to consider punishment in deciding guilt.

I don't understand why this is appropriate. In making most life decisions, we consider the consequences in determining how much effort to put into deciding and the degree of confidence we must feel before we go forward. Whether to get married or have a risky operation obviously requires a greater psychological commitment than choosing between Starbucks and Peets.

Jurors should be told the gravity of the decision they are making so they can take it into account in deciding whether to convict or acquit. As representatives of the community where the defendant committed his crime, the jury should be allowed to make the judgment of whether the punishment is too severe to permit a conviction. Having to confront the jury with the severity of the punishment they are seeking to extract may well deter prosecutors from using overcharging as a bargaining tool.

6. Give jurors a say in sentencing. Except for capital cases, we have turned our sentencing process over entirely to experts and professionals. We have mandatory minimums, sentencing guidelines, probation officers and judges at all levels involved in the decision, but we studiously ignore the views of the very people who heard the evidence and are given the responsibility to determine guilt or innocence while reflecting the values of the community in which the offense occurred.

This is a system only a lawyer could love. Jurors should be instructed on the range of punishments authorized by law and, if they find the defendant guilty, entrusted to weigh in on the appropriate sentence within that range. And I would make that the absolute upper limit of what punishment the judges actually impose, overriding any sentencing guidelines, mandatory minimums or their own considered judgment.