I recently finished the new book by the late Bill Stuntz, The Collapse of American Criminal Justice. My co-blogger Paul Cassell reviewed it here in the Wall Street Journal, and I wanted to blog my own thoughts about the book. My basic take is that it’s a great book, with many important insights. It’s the most interesting book on the criminal justice system that I’ve read in a long time, and perhaps ever. At the same time, I’m not sure I’m entirely convinced: Stuntz makes a number of claims that strike me as questionable. Plus, his recommendations for reform stuck me as a bit utopian. In this post, I’ll introduce Bill’s basic theory, and then offer some thoughts in response. In a future post, I’ll turn to Stuntz’s recommendations, and again offer my own reaction.
I. The Basic Argument of Stuntz’s Book
Stuntz’s argument starts with a widely-heard set of complaints about today’s criminal justice system: Too many people are in prison, too many laws are too punitive, and criminal cases are generally resolved by guilty pleas guided by prosecutors rather than trials resolved by juries or judges. The question is, why? In Stuntz’s view, the major reason is that criminal justice has lost its local character.
In the 19th Century, Stuntz explains, criminal law and justice was largely local. There was a lot of variation by region, to be sure. But on the whole, local police captured local criminals, who were charged by local prosecutors and tried before local juries using criminal law standards from the common law that were vague and left considerable discretion to the jury. In Stuntz’s view, that system worked pretty well – much better than we realize today. Criminal punishment was relatively rare, and punishment reflected community norms and senses of justice. According to Stuntz, a range of forces have slowly eliminated the local character of criminal law. And losing the local character of the law has made it more punitive and less fair.
What changed? First, the shape of criminal law doctrine came to be seen as a legislative question, not a judicial question. Criminal law used to be fixed: It was settled by the basic doctrines of the common law. As criminal law became seen as a field subject to control of the state and federal elected branches, criminal law began to expand. Because most see themselves as potential crime victims rather than potential criminals, especially in the aggregate at the state and federal level, politicians have a strong incentive to be seen as “tough on crime” by making criminal law broader and harsher. And because most voters don’t directly experience crime levels or the effect of punishment on their own communities — most voters don’t live in high-crime neighborhood — they tend to be receptive to the “tough on crime” message even if the law is already tough. The system thus tends towards harsher and harsher punishment, especially at the state and federal level. Even local prosecutors have new incentives to be extra harsh, as state governments rather than localities have come to pay for the prisons: Local prosecutors can get all the public relations benefits of “locking them away” without having to pay the bill for costly imprisonment.
This trend was facilitated by the Supreme Court’s failure to keep legislatures in check. In Stuntz’s view, the Equal Protection clause should have been a major check on legislative and executive action in the criminal justice field. The Courts should have interpreted it to ensure substantive fairness in the laws, and especially to eliminate racial discrimination. But the Court instead gutted the Equal Protection clause in the 19th Century in United States v. Cruikshank. At the same time, the Supreme Court adopted an expansive interpretation of the Commerce Clause as far back as the 19th Century, allowing Congress to federalize morals legislation based on a simple showing that something crossed state lines. By taking a deferential position that allowed racial discrimination to flourish at the state level and enabled Congress to federalize criminal law, the Supreme Court wrongly let the political branches do as they pleased. Legislatures could raise punishments as high as they liked; police could engage in pretext prosecutions, and target minority groups; and prosecutors could add lots overlapping charges to induce guilty pleas.
Stuntz argues that the harshness of the criminal justice system was inadvertently aided by well-meaning progressive reformers. Reformers tried to make the system more fair, but their efforts backfired. In the 19th Century, criminal law was vague and procedural rights were few: Trials were cheap, they occurred all the time, and juries had a lot of discretion under common law standards. In response, 20th Century reformers tried to rationalize and clarify criminal law standards by replacing vague common law definitions of crimes with new clear ones (such as the Model Penal Code). The idea was to make criminal law more predictable and rational, but the perverse effect was to greatly diminish the role of the jury: Whereas traditional criminal law doctrines had left lots of discretion for the jury, the new clear standards gave defense attorneys very little to argue about. The result was more guilty pleas and fewer trials.
At the same time, the Warren Court’s criminal procedure revolution gave defendants more rights with the aim of making the criminal justice system more fair. To some extent, this was a belated response to the 19th century Supreme Court’s failure to take the Equal Protection clause seriously. But in Stuntz’s view, using the Bill of Rights to focus on the procedure of criminal justice (the law of investigations) rather than using the Equal Protection clause to focus on the substance of criminal law (definitions of crimes) only made the problem worse. More procedural rights raised the costs of trials. Because legislatures were free to alter the substance of criminal law however they pleased, legislatures responded to the new procedural rights by broadening criminal laws, jacking up criminal penalties, and giving prosecutors wide discretion to pressure defendants to plead guilty in exchange for a “deal.” Perversely, the result of giving defendants more rights was to create an environment in which most defendants were pressured into waiving all their rights in practice by pleading guilty. The Supreme Court’s regulating procedure instead of substance just made the substance worse instead of making the procedure better.
II. A Few Thoughts In Response
I find Stuntz’s argument fascinating. He makes a lot of intriguing claims, and there’s a ton to chew on here. It’s the kind of book I want to mull over and read again, which is about the highest compliment I can give. For what it’s worth, my initial “mull” finds me partially persuaded and partially unpersuaded. Some of what Stuntz says rings true, but some of his claims raise a lot of question marks.
Stuntz’s basic narrative of the need for localism strikes me as persuasive and really important. His focus on the need for criminal justice to stay attentive to the needs of the community — and the dangers when decisionmakers don’t reflect the views of the community most directly impacted by crime — rings true to me. To some extent, Stuntz’s view is the classic argument for local decisionmaking: People can take care of themselves better than others they don’t know, as they are more closely tied to the facts of what is happening and they can exercise judgment based on local opinions. So on the basic gist of the argument, I’m pretty impressed.
Stuntz’s effort to link the Supreme Court’s caselaw on substantive criminal law and criminal procedure is also a terrific and insightful move — one that is based on and extends what what is probably Stuntz’s most important article, The Uneasy Relationship Between Criminal Procedure and Criminal Justice, published in the Yale Law Journal in 1997. Uneasy Relationship is a favorite of mine, and I was glad to see its insights take a prominent place in Stuntz’s book.
At the same time, I had some significant concerns with parts of the argument. Let me focus on two parts in particular. First, I thought Stuntz had a rather rosy view of the past of criminal justice. His conclusion that the system more or less worked in the past strikes me as optimistic. Much of his argument was based on data like crime rates and imprisonment rates that I found hard to assess. I found myself unsure of whether to accept the data as accurate, and I was dubious that we can say very much about how the criminal justice system worked assuming all the data is right Take the fact that crime rates have plummeted since the early 1990s: Does that really suggest that we’re doing something right today that we were doing wrong twenty years ago? Or is this just the fortunate byproduct of banning lead, or something else? Stuntz doesn’t have much of an answer for why crime rates have plummeted in the last 20 years — he discusses it at length, and concludes it is a real puzzle — which to my mind reflects a skepticism that could be equally well applied to the causes of crime rates in the past.
Similarly, I think the differences between the common law of crimes and the more modern approach to criminal law are much more modest than Stuntz suggests. There are some differences, yes. But on the whole, they’re minor: The basic elements of the basic crimes today are pretty similar to what they were in Blackstone’s time. And some defenses, like necessity, are generally considered broader today than they were in the past (to the extent they existed at all). As a result, it seems dubious to me that changes in approaches to criminal law played a significant role in changing the nature of the criminal justice system. It’s an intriguing idea, and fits the classic Stuntzian unintended consequences mold. But I’m just not sure it works. If there were such an important difference between the common law standard and the modern standard, wouldn’t we expect to see a difference between the outcomes in jurisdictions that today still retain the gist of the common law approach and those that have widely adopted the modern approach? Stuntz doesn’t suggest that there is such a difference, and I don’t know of one. So I’m skeptical that that there is an effect such as what Stuntz mentions, or at least that it is significant.
Of course, it may be that the law has operated differently, rather than the doctrine is different. Slight differences in trial procedure can have a huge difference in how the law is applied. Take the case of the trial judge’s gate-keeping function. These days, trial judges play a major gate-keeping role: If a defendant wants to put on a defense, for example, the trial judge has to find enough evidence to support it before it can be argued to the jury. If there were different gate-keeping functions in an earlier time, then the law could be applied very differently even if the formal doctrine were the same. If judges used to give all the issues to the jury, for example, the jury would have a lot more discretion than they do today because they would be invited to consider more defenses, not because the details of the jury instructions on the law would be any different. I don’t know enough about the history of criminal trials to know if that gate-keeping function has changed much, but I suspect that is more likely source of jury discretion than changes in doctrine.