In the paper “Torture and Plea Bargaining”, John H. Langbein lays out the similar roots and usage of the two practices, in medieval times and today, respectively. On the face of it, the comparison may seem surprising. In the modern world, it is unlawful for punishment to be cruel or unusual, and it would be an impeachable offense for a judge to sentence someone to flogging for refusal to cooperate with the court; but while the method has grown more refined, the essence remains the same. He argues that they serve similar purposes, arose for the same reason, and can be solved by the same types of reforms.
The requirements of proof which the Roman Church required (beginning in 1215), were exceedingly strict; for a serious crime, you must have either two unimpeachable witnesses, or a voluntary confession. The eyewitnesses had to see the act itself; being seen by a crowd running with a bloody dagger away from a house where a murder had taken place was insufficient. Such a standard is obviously unworkable, and so courts would have to rely upon a voluntary confession. But how to get it? In a case such as the one described, the judge had sufficient proof to order the accused tortured. No confession obtained while under torture would suffice – they would have to be willing to swear in court they had done it – but such a protection was rendered moot by the fact that, were they to recant, they would simply be remanded to torture until they were willing to “voluntarily” confess. While there were some formal safeguards, (it was only supposed to be used to elicit information that no innocent could know, and suggestive questioning was not allowed), these were totally insufficient toward protecting the innocent.
In a like manner, Langbein argues, plea bargaining arose because the standards of proof and procedure had grown so much as to be unworkable. In the era of the Constitution, jury trials were a perfunctory matter; a court would get through a whole bundle in a single day. Trials lasting longer than were so non-existent that in 1794, during the Treason Trials in England, the Court had to seriously discuss whether it had any authority to adjourn for the evening at all! This is a lightning speed compared to the time when Langbein was writing (and things have not much changed since); if a felony case went to jury, as of 1968 in Los Angeles, it lasted 7.2 days. Notwithstanding the procedural guarantee in the 6th amendment, not everyone can get a jury trial; the system would quite simply collapse. Instead, about 95% of felony cases end in a plea bargain. We threaten overwhelming punishment in order to elicit a voluntary confession, and so our protection of rights – the careful accumulation of jurisprudence to safeguard the rights of the accused – are nothing but meaningless letters.
What is to be done? Langbein is not confident that reform will come soon; he notes the persistence of the medieval system long after its flaws were apparent, and given that 40 years on the problems persist, his cynicism was warranted. What he suggests is that the American system be run more along German (at the time West German) lines; trials are conducted by judges who are involved in finding the facts themselves, and are not merely referees for the benefit of the jury. Under this system, guilty pleas are not allowed; you may confess all you want (and indeed, in a large percentage of cases there are confessions tendered; a consequence, I think, of improved surveillance and policing increasing the proportion of people caught red-handed), but it will not result in your sentence being reduced one iota, and neither is a confession sufficient; the court must still find enough evidence for conviction. By getting through trials faster, it allows far more people to actually get a hearing at all.
I am no expert on alternative court systems. I will not, and cannot, vouch for any alternative at this time. I would only suggest that, if you drew a graph of effective procedural protections – what people actually get, not what they are merely promised – the line would be curved, not straight, and that rolling back our protections in theory would increase our protections in fact.
Robin Hanson noted that the reason trial by combat was popular in the medieval era was because their normal trials had become so cumbersome:
https://twitter.com/robinhanson/status/1116753635378126849