Random selection is necessary to create stable meritocratic institutions

> Juries, widely trusted to impartially deliver justice, are the most familiar instance.

Trusted by those that have not looked into whether this is actually the case. The first prime minister of Singapore, Lee Kuan Yew, was famously against trial by jury, because of how easily lawyers can abuse biases in multiracial societies, based on his first-hand experience [1].

A UK study found his experience is the norm, not the exception - Black and minority ethnic (BME) jurors vote guilty 73% of the time against White defendants, but only 24% of the time against BME defendants [2]. (White jurors vote 39% and 32% for convicting White and BME defendants, respectively. You read that correctly - Whites are also biased against other Whites, but to a much lesser degree)

Edit: To answer what is the alternative to juries: Not all countries use juries, in some the decision is up to the judge, and in some, like France, they use a mixed system of judges and jurors on a panel [3]. The French system would be my personal preference, with the classic jury system coming in second, despite my jury-critical post. Like democracy, it's perhaps the least bad system that we have, but we shouldn't be under any illusions about how impartial and perceptive a group of 12 people selected at random is.

[1] https://postcolonialweb.org/singapore/government/leekuanyew/...

[2] https://www.ucl.ac.uk/judicial-institute/sites/judicial-inst... - page 165 (182 by pdf reader numbering), figure 6.4

[3] https://en.wikipedia.org/wiki/Jury

And what is the other option? Just led the judge alone decide?

Yes I trust judges more than I trust juries.

And it usually isn't a single judge. There is a panel of judges or en banc.

And juries aren't universal either. Lots of other countries don't have juries but they have a fair and equitable justice system. Look up civil law vs common law.

That statistic could also be the result of excessive prosecution against black/minorities and not necessarily just jury bias. (Which would also explain the white bias against whites)

In case I'm not the only ignorant one:

BME = black and minority ethnic

That's not evidence of jury bias on its own. You have to control for prosecution rates and rates of actual guilt.

I want the judge to keep the lawyers in check so they cannot. Judges are trustable because the jury limits their power. If I am a lawyer I know whothe judges are and it is to my advantage to figure out their bias (including judge shopping if there is more than one in the area), looking for embaressing things or blackmail material, what bribes they will accept (often in form of donation to a family charity) and so one.

which is to say the reason I trust judges is the jury keeps them in check by ensuring there isn't value in the above corruption.

What is the reasoning, judges are above racial bias?

Judge shopping is also a thing, if the judge was the only person in power to make the decision we'd be completely screwed.

I trust an individual judge's opinion on almost any topic to be more intelligent than that of an individual jury member.

But there's huge selection bias in who becomes a judge, and so we end up with a pool of people who are mostly former prosecutors, which is another pool with a huge selection bias.

All of the judges I know personally (though not all I've been around) are well-meaning, fair-minded people, but with maybe one exception they're all true believers in the fairness of the system, and all tend to give tremendous unearned deference to prosecutors. We should absolutely not make them the finders of fact in criminal cases.

If you're curious about this topic, I'd recommend you look up interviews with the jurors in the OJ Simpson trial. Many were black and by their own admission made their decision about OJ's guilt-based entirely on a feeling of racial justice. They considered it “payback.”

https://youtu.be/BUJCLdmNzAA?feature=shared

A universal counterargument that works on any data. But unlikely to be true, given that the UK sentenced a BME perpetrator to a short 2 years for one-punch-killing an 82-year-old veteran [1], while "threatening gestures" at police and chanting "who the f- is Allah" earn the White perpetrator 18 months in prison [2], and merely being present at a protest, while not engaging in any violence, earns 32 months in prison [3].

We also have to ask - if the biases in that study were flipped, if White jurors were far more likely to convict BME defendants, and pardon White defendants, and BME jurors were the more even-handed ones, would this not be trumpeted as conclusive evidence of racism?

[1] https://www.bbc.com/news/uk-england-derbyshire-66959198

[2] https://www.newhamrecorder.co.uk/news/24515551.london-disord...

[3] https://www.theguardian.com/uk-news/article/2024/aug/08/pens...

This is known as a "bench trial" and is a legal concept.

They are more likely than juries to be above racial bias. Not 100% but I trust them more due to the training and education needed to become a judge.

I’m curious, had any legislatures tried blinded trials, where judges and/or juries don’t see the litigants, don’t know their names or location details, and otherwise only have access to the information on a need-to-know basis?

Sort of like how removing names, ages and photos from resumes removes demographic biases and makes one focus on the actual skillset.

(I’m not sure if this is a good idea, merely wondering if it was tried.)

One possibility would be to use juries to just decide facts, then a panel of judges applies the law to those facts.

If there are many factual disputes in a case maybe use multiple juries with each jury only deciding on a subset of the facts, chosen so that no jury sees the entire case. They are less likely to be biased if they don't see the entire case.

What quantitatively separates "excessive prosecution" from "prosecution"?

The sentence isn’t relevant to the jury bias discussion (unless the jury is involved in sentencing in the UK?)

> We also have to ask - if the biases in that study were flipped, if White jurors were far more likely to convict BME defendants, and pardon White defendants, and BME jurors were the more even-handed ones, would this not be trumpeted as conclusive evidence of racism?

Yes, but why is this relevant? That’s not the case in the statistics you cited.

My comment was pointing out that there are multiple possible (probably simultaneous) causes for the jury statistics.

Anything that is statistically visible?

I’m not making a value judgement; I was just pointing out other explanations of the statistics.

> The sentence isn’t relevant to the jury bias discussion

It's relevant as an indicator of the bias, or lack thereof, of the system as a whole.

The very study you cited states in their initial summary of findings that "The study provides the first evidence to support a widely held belief: that racially mixed juries do not discriminate against defendants based on the defendant’s ethnic background. While the assumption has been that racially mixed juries will not discriminate against ethnic minority defendants, this study showed that racially mixed juries also did not discriminate against White defendants.[0]"

[0]https://www.ucl.ac.uk/judicial-institute/sites/judicial-inst... - page iv

So, like the excessive prosecution of men, who receive sentences about twice as long as women for committing the exact same crimes, on average? Men are prosecuted at far higher rates than women, too.

Bench trial and civil law are actually the most used legal systems worldwide, not jury trial and common law.

OJ Simpson is was such a famous case that I'd be inclined to treat it as an outlier in many ways, not the norm.

Wouldn't that be kind of tough since one of juries main function is evaluation of witnesses. Without seeing the witnesses face or hearing their voice, how could one do that?

I don't think one of the most high-profile and racially charged cases in history can serve as a reasonable benchmark for how the bulk of cases are handled.

Edit: Not sure why I am being downvoted, I tried to say the same thing dmonitor said.

> A universal counterargument that works on any data

How? My statement could not be correct if the data was instead: BME jurors vote guilty 73% of the time against White defendants, but only 24% of the time against BME defendants; and White jurors vote guilty 39% of the time against White defendants and 52% (instead of the cited 32%) against BME defendants.

And the LAPD seems to have decided OJ was guilty based largely on a feeling of racial prejudice. The fact that he actually did it was coincidental. I've often seen that trial described as the LAPD framing a guilty man. The prosecution did a terrible job and I'm not at all convinced the acquittal wasn't the correct verdict, even if it's pretty clearly contrary to the facts of what happened.

It seems safe to assume that the LAPD also did/does this to less famous people of color, in which case a higher rate of voting to acquit would not indicate bias by the jury.

Agreed. Outliers are a thing. As is cherry-picking data to try and prove a point.

Sucks, as this level of cherry-picking heavily biases me against the premise. If someone has a good data set, they don't need to drive anecdotes from the outliers. And if they are, is it an attempt to hide that the overall data paints a different picture?

Why not a jury of 12 judges for the best of both worlds?

The whole point is that they would be evaluating the witnesses' testimony, not their mannerisms and appearance (read: how culturally and racially similar they are to the jury, or to the "high class" ideal of their society).

Norway has a tradition of using panels of judges that uses a mix of professional judges and lay judges drawn from the jury pool. I don't recall the specifics on when this systems used vs. a regular jury. They deliberate with the professional judge, has the majority, but can be overridden if their reasoning is blatantly contrary to the law.

Not only that but today you have all kind of analytics for courts, and judges that you can use in your favour.

> A universal counterargument that works on any data.

Well, yeah, that’s why data itself doesn’t “show” anything on its own. You first need competing explanations of reality, and then data might help you choose one explanation over another.

> and merely being present at a protest, while not engaging in any violence, earns 32 months in prison [3].

> [3] https://www.theguardian.com/uk-news/article/2024/aug/08/pens...

From the article: William Nelson Morgan, 69, was sentenced to 32 months in prison, having previously admitted violent disorder and carrying a cosh during a riot on County Road in Liverpool on Saturday.

Sounds like he engaged in violence and was carrying a weapon.

> (White jurors vote 39% and 32% for convicting White and BME defendants, respectively. You read that correctly - Whites are also biased against other Whites, but to a much lesser degree)

This assumes that Whites and BMEs going to trial are equally likely to be guilty.

Shouldn't we assume there would be some hidden delta?

> Sounds like he engaged in violence

Seems that you didn’t read the article you have linked or your definition of engaging in violence is rather obtuse.

Surely you would not equate standing still when a policeman orders you to move and then resisting when they to forcibly move you to murdering a random person for no particular reason? Nor would you agree that the second violent offense deserves a significantly more lenient punishment than the first?

Lay judges are also an option and might offer a reasonable balance.

It would be very expensive but yes seems like a good option. Of course a problem in some places and systems is that judges are often political appointees which has its own implications

Jury trials are not unique to common law systems, though. Many civil law systems either have lay judges or allow jury trials under certain circumstances.

Of course you are not entitled to a jury trial in e.g. France unless you are accused of something very serious.

But I would argue that jury mentality is universal.

Any juror who knows about the concept of jury nullification is more likely to use it when the defendant reminds them of themselves or when the prosecution has so vastly disproportionate resources over the defendant that the trial can’t possibly be fair.

I would argue OJ would have been arrested far earlier for DV (long before the murder) if the LAPD didn’t have such a hard-on for the celebrity.

>I would argue that jury mentality is universal

Then argue it, because that's a pretty large thing to say unsupported

> My comment was pointing out that there are multiple possible (probably simultaneous) causes for the jury statistics.

Sure, but this is a non-statement without qualifying anything behind it. You can defeat any argument by claiming its "multi-faceted". Just like how I am doing to you right now, but instead forcing you into the position where you lack evidence to dismiss.

The idea is to filter out bias-introducing information with low relevance (like gender, appearance, or accent) and focus on the actual events that took place.

Otherwise the court starts to include elements of theatrics and objective truth starts to give way to how one presents their case, such as what sort of appearance litigants make. E.g., whenever they're speaking confidently or, say, stuttering nervously. While this can be relevant information (e.g. if someone refuses to look in the eyes it could be a sign one's lying), there are multitude of ways it can be deceiving (e.g. if someone refuses to look in the eyes it could be that they find eye contact generally uncomfortable, for example folks with anxiety disorders do that).

Presenting both litigants through a Vtuber-like interface that re-synthesizes voices, adjusts some patterns of speech (like replacing names with placeholders, or making language gender-neutral), reduces non-verbal signalling, and provides neutral appearances to both parties, feels like something that can make litigants, judge and juries all focus on the abstract ideas of what took place, potentially allowing for a more clear and neutral judgement.

But - of course - it's also perfectly possible that it would fail in some way I fail to foresee.

Because the bias was diluted by the other jurors - from the passage on page iii, I assume the juries in the study were only 10-33% BME. The study follows your quote with:

Even though the defendant’s ethnicity did not have an impact on jury verdicts, the research found that in certain cases ethnicity did have a significant impact on the individual votes of some jurors who sat on these juries. Statistical analysis of the individual votes of all 319 jurors who took part in the case simulation showed that in certain cases BME jurors were significantly less likely to vote to convict a BME defendant than a White defendant. [..]

The report concludes that this highlights the benefits of permitting majority verdicts and of having 12 member juries. The fact that 12 jurors must jointly try to reach a decision and that majority verdicts are possible meant that more verdicts were achieved and individual biases did not dictate the decision-making of these racially mixed juries. If juries were smaller or if unanimous verdicts were required, then individual juror bias might potentially have a greater impact on jury verdicts.

The examples you use are disingenuous and irrelebant. They were judge decisions, not jury ones, one was a minor with mental issues.

Over-prosecution of minorities is a documented bias.

No, if the biases were flipped, this kind of study would still not be evidence of racism. In the context of all the other studies and the history of slavery and racism, it is reasonable to _guess_ that white jurors are not discriminating against white defendants, and that sampling bias is to blame.

We'd need a full overhaul of the legal system from the ground up to get to a place where we could implement the better versions of lay judge participation, and if we got a full overhaul of the legal system that's not what we'd land on. And lay judges aren't really a balance between juries and judges in most implementations.

There is a belief that one can accurately determine guilt by looking at someone's face, seeing their body language, and hearing them speak. I myself sometimes think this is true.

But the truth of the matter is that in the United States at least, this is all irrelevant. No one gets a trial anymore, not in practice. Everything is plea bargain. Between 95 and 99 out of 100 cases is resolved in that manner, and the common opinion is that there is no capacity to give anyone trials at all.

Creating an obverse in your mind generally discards the context that brought the subject into being.

> not at all convinced the acquittal wasn't the correct verdict

I agree with you. I followed it closely at the time, and thought acquittal was the correct verdict based (only) on what the jury saw. We court-watchers, of course, saw everything that the defense managed to exclude, and came to a different (and I do believe more accurate) conclusion, but the jury got it - from their (deliberately constructed by the defense) point of view - right.

It was a formative episode in my civic understanding.