In both the constitutional law of American criminal justice and the scholarly literature that law has generated, substance and procedure receive radically different treatment. The Supreme Court, even in this conservative political period, continues to require costly procedural safeguards that go beyond what elected legislatures have provided by statute. The Court, however, has shown great deference to the choices these same legislatures have made about what conduct may be made criminal and how severely it may be punished.

The distinction between substance and procedure pervades academic thinking all the way down to its foundations. Substantive criminal law still holds its place in the sacred precincts of the first year curriculum. Criminal Law's cognate discipline is philosophy; the standard method of analysis is to measure general principles according to how well they track intuition's response to hypothetical cases. Criminal Procedure's cognate discipline is Constitutional Law; the standard method of analysis is to subject the operation of the criminal justice system to the same rhetoric of text, history, and precedent that frames the issues in separation of powers or freedom of speech cases. The philosophy mediated by doctrine is political, rather than moral theory.

In trial level courthouses, however, the distinction fades, as the defendant trades his procedural rights for reductions in his substantive liability. The substantive law endows the prosecution with the ability to charge the same conduct at many different levels of potential punishment. The procedural law also endows the defense with its stock in trade - the rights to suppression motions, discovery, elaborate jury selection procedures, confrontation of the victim, and so on.

These endowments are dynamic rather than static. A legislature that adopts a three-strikes law increases the prosecution's bargaining power. A court that reads the confrontation clause to bar excited utterances from the government's proof increases the defendant's bargaining power. In the trenches of criminal justice, these entitlements may well be traded off, erasing the distinction between substance and process.

Admirable scholarship has exposed this basic dynamic. Debate continues about two great issues. First, is this state of affairs normatively defensible or not? Second, if the present relationship between substance and procedure is undesirable, what, if anything, can be done about it?

In this paper I take up the second question, which seems to me to have drawn too little systematic attention (perhaps because it is so daunting). The literature has devoted considerable debate to alternatives to plea bargaining. But these discussions have been self-contained; they do not take account of the substance/procedure feedback loop already in place. The principal point against proposals to ban bargaining is not that we should not but that we cannot; self-interested, repeat-playing actors in the criminal justice process will find ways to bargain. The debate, naturally enough, has not gotten to the point of "what if we succeeded in banning plea bargaining?"

As things stand, the prohibition of bargaining would leave prosecutors with unregulated discretion to select charges from overbroad and draconian criminal codes. Prohibiting bargaining would mean that defendants could not trade their constitutional procedural entitlements off against the state's substantive criminal law entitlements. The new model would be one in which defendants, facing decades in prison for relatively modest crimes, would stand trials they have little chance of winning.

The discussions on plea bargaining have the same isolated quality as the discussions on individual bodies of criminal procedure doctrine. Of course they matter, in some cases; but the bigger picture is the relationship between substantive criminal law sentencing and the procedural rights of the defendant. So serious are the difficulties that I shall not - yet - defend any doctrinal reform on the ground that the relation between substance and procedure would be harmonized thereby. My task is one more modest, but I hope still useful. I aim to survey the possible strategies by which the system might escape the current impasse.

The possible strategies fall into five basic categories. First, we might continue what we seem to be doing now: increasing constitutional procedural entitlements in the hope of mitigating the excesses of the substantive criminal law. Second, we might give up on the constitutional distinction between substance and process by deconstitutionalizing procedure altogether, or at least to a dramatic degree. Responsible then for both substance and process, legislatures might strike a better balance than is produced by the current division of labor. Third, we might achieve the same sort of unification by constitutionalizing substance. Robust judicial review of substantive criminal legislation might curb overcriminalization, which might in turn lead the courts to develop a more rational body of procedural rights. Fourth, we might look for more rigorous restrictions on prosecutorial discretion, building on administrative law and experience with sentencing guidelines. Fifth, we might look for more rigorous restrictions on the defendants' right to waive procedural rights for substantive advantage.

What I hope to add to the scholarly conversation is a brief assessment of the promise and pitfalls that attend each of these strategies.


Criminal Law | Criminal Procedure | Public Law and Legal Theory

Date of this Version

June 2005