Honestly, it seems to me that—given a hypothetical ideal jury with encyclopedic knowledge of the criminal code—it doesn't even make sense to have a trial that is "about" a particular crime. Rather, wouldn't it make more sense (again, given a jury that knows what the requirements are for every crime in the book) for the trial to just present the evidence in the case to the jury, and then the jury to come back with a list of zero or more crimes that they've decided that the defendant has committed?

It seems to me that the only reason we do things any other way, is that juries don't have encyclopedic knowledge of the criminal code, and so you have to teach them about what the requirements are for guilt in the particular case.

Given that the court system doesn't like jury nullification, and wants juries to be deciding solely on whether the facts presented to them match the legal requirements for the definition of a particular crime—why would it matter whether they know which crime they'll be asked to evaluate the facts against, before they know the facts? They can learn the facts first, or the law first. Either way, the last step is just matching one to the other. (In the Supreme Court's opinion, at least.)

> It seems to me that the only reason we do things any other way, is that juries don't have encyclopedic knowledge of the criminal code, and so you have to teach them about what the requirements are for guilt in the particular case.

The reason is more involved, because:

(1) law, like facts, is subject to question and answering those questions is out of scope of the jury, which is the trier of fact alone. Your model would necessarily transfer the judge's responsibility to the jury.

(2) The law creates permission for prosecution (bit civil and criminal , though only the latter is relevant here), but that permission is not a mandate. Judgement of the cake of prosecution is given to the offended party (in criminal law, the executive on behalf of the government). Your model would necessarily transfer the executives authority largely to the jury (though prosecutors could withhold inculpatory evidence to avoid undesired charges being triggered.)

(3) But, most critically, your model would provide the defendant with a vast surface to defend against, and require the defense to provide evidence against any offense the jury might infer from the evidence. The use of specific charges and specifications putd the defense on notice of the specific charges they must defend against.

(4) Related to (3), without specific charges, any trial is essentially a trial for all potential criminal conduct occurring before that point, which creates double jeopardy problems if concealed evidence of an early crime comes to light, as distinguishing between a trial addressing other charges (which would not foreclose a new trial for double jeopardy reasons) and a trial addressing the offense indicated by the new evidence but which merely failed to convict (which would foreclose a new trial) is impossible. Meaningful protection against double jeopardy while permitting trials for newly discovered offenses requires something like the current specific-charge model.

> Given that the court system doesn't like jury nullification,

The court system may not like it, but it has repeatedly been found to be part of the Constitutional order of government. By the court system, I might add, so obviously they like it enough to keep protecting it.

> law, like facts, is subject to question and answering those questions is out of scope of the jury, which is the trier of fact alone.

Like I said, I was presuming an ideal jury. (Picture a land where everyone has the legal experience and character of a Supreme Court justice—a land that makes Plato's Republic look populist.)

A real jury is, of course, Not Good at the law.

> Your model would necessarily transfer the executives authority largely to the jury (though prosecutors could withhold inculpatory evidence to avoid undesired charges being triggered.)

Not necessarily. The jury could return a list of crimes that have been proven (in Scottish justice-system terminology), and then the aggrieved could decide which crimes they want the defendant to be sentenced for.

Basically, extend the judge's ability to issue a "judgement notwithstanding" verdict, to be something done in cooperation with the aggrieved party.

Or, to put that another way... take a jury trial, and split it into two pieces:

• a jury-empanelled inquest—a truth-finding procedure—where the jury is there to be a truth-finding oracle, consuming 1. a criminal code and 2. a ream of testimony, and then outputting 3. a set of "proven" / "not proven" decisions for every crime in the criminal code, as they apply to the testimony.

• then, a bench trial, where the judge also has access to the testimony from the inquest, but where the law directs the judge to make their decision the same way they do in a jury trial: by—unless their reading of the transcript reveals that something has gone terribly wrong in the execution of justice—treating "the decision reached by the jury" as the only valid testimony. The judge then assigns a "guilty" / "not guilty" verdict for particular crimes that were being pursued, based on that testimony, and does sentencing.

I think this approach addresses most of your other points. Please poke holes in it :)

> By the court system, I might add, so obviously they like it enough to keep protecting it.

Jury nullification isn't really upheld by the court system (it's been repeatedly found that jurors have no right to it); its existence just comes down to the fact that criminalizing a juror's voting in one direction or the other would kind of break the justice system.

In a justice system that separated the "we declare that you screwed up" and the "so we will punish you" parts, I believe that jurors found to have "voted their conscience" would likely be found to have done something wrong in the eyes of the justice system—just something that cannot, necessarily, be criminalized.

About half of states require a grand jury to screen criminal indictments.

Your suggestion seems to be to coalesce the grand jury with the petit jury, and have the jury decide first that a criminal court should convene to conduct a trial, and then decide only later if the facts proved in that trial correspond to any particular crime or crimes.

The prosecutors could still press for a specific crime, but the defense would also be able to argue directly to the jury that "even if X and Y were true (which they aren't), that would only be a class N crime, because the prosecution hasn't proved Z, which is necessary for it to be a class M crime."