Prosecution: Prosecutorial Discretion

The legal philosopher Ronald Dworkin has distinguished several senses in which the word "discretion" is used in legal discourse. Sometimes the word is used in a relatively weak sense, signifying that "the standards an official must apply cannot be applied mechanically but demand the use of judgment." (Dworkin, 1977, p. 31). For example, we might say that the lieutenant left the sergeant a great deal of discretion if she ordered him to select the five most experienced soldiers for a particular mission, since the criterion of experience could be applied in different ways. In a different weak sense, we sometimes say an official has discretion when we mean that his decision cannot be reviewed and reversed by a higher authority: although the rules of baseball clearly define the strike zone, the umpire could be said to have discretion over the call of balls and strikes, because no higher power can overrule his call, whether or not a videotape replay shows that the decision was inconsistent with the rule. We also sometimes use the word in a very strong sense, to mean that the official is simply not bound by any standard at all. If the university registrar is told simply to divide the students taking chemistry into two sections, she might have complete discretion to divide them alphabetically, or by pulling names from a hat, or by assigning the first students to register to the more popular instructor; there are no governing criteria by which her decision can be said to be wrong.

Even in this strongest sense, discretion is always conferred for a particular purpose, and operates within some limits. The registrar in our example has been given no authority to assign students who have not registered for chemistry to one of the two sections. Moreover, there may be some assumed or external constraints even on extremely broad discretion of this type: for example, the registrar might be in violation of law if she divided the sections by assigning students to sections according to their race.

Although prosecutorial charging discretion is extremely broad, there are some limits of this kind. First, prosecutors' discretion operates within the universe of cases in which the minimum requirements for legal punishment exist; while the prosecutor has a wide discretion not to bring charges that could be supported, she has no right to bring charges that are not supported by sufficient evidence, and legal checks are in place to absolve defendants of such charges. When a prosecutor decides to bring a charge, that decision is ultimately subject to a judge's authority to determine whether the legal elements of a crime have been alleged, and to a jury's ultimate power to decide whether the facts have been established beyond a reasonable doubt. Indeed, in the vast majority of U.S. jurisdictions, the prosecutor's assessment that the evidence warrants a felony charge must be submitted for review by a judge or grand jury before a trial or full judicial proceedings are instituted. (As a practical matter, however, such reviews are rarely a very significant check on the prosecutor. The evidentiary standard at this threshold stage is low, grand juries in particular operate under rules that permit prosecutors to dominate their proceedings, and in virtually every system, the pressure of case load requires that the review be cursory in most routine cases.)

Second, it is well established that a prosecutor's decision to bring charges against an individual may not be based on discriminatory grounds such as race, religion, or the expression of political opinion ("selective prosecution"), or as retaliation for the successful exercise of legal rights, such as the right of appeal ("vindictive prosecution"). The Supreme Court has ruled that when defendants claim to have been singled out for prosecution for discriminatory reasons, the prosecutorial decision is subject to the ordinary constitutional standard of equal protection. But this standard is extremely difficult to meet. It requires not merely a statistical showing that, for example, the vast majority of those prosecuted for a particular crime are members of minority groups, but also proof that prosecutorial decision-making was actually motivated by a discriminatory purpose. Even a statistical showing of disparate effect may well be difficult to establish, because the defendant must show not only that those prosecuted are members of a disfavored group, but also that there were cases in which members of favored groups were known to the authorities to have committed the offense, but were nevertheless not prosecuted. The latter information is not usually publicly available, the courts have resisted defendants' efforts to obtain discovery of statistical information from prosecutors about cases in which charges have not been filed, and it is difficult to control for the bewildering factual variations among different cases to eliminate possible nondiscriminatory explanations for suspicious patterns of results.

The Supreme Court has also held that the constitutional guarantee of due process protects a defendant against prosecutorial vindictiveness, that is, against a prosecutor increasing the charges in retaliation for a defendant's exercise of a statutory or constitutional right. For example, a prosecutor would be forbidden to bring a more serious charge in retaliation for the defendant's having appealed a conviction on a lesser charge. Indeed, in that situation, courts even apply a presumption of vindictiveness, requiring prosecutors to bear the burden of proving that the decision to increase the charge was not retaliatory. But the circumstances in which such a presumption is applied are extremely limited, and in the absence of a presumption, establishing that prosecutors acted from a deliberately vindictive motive is difficult. Moreover, the Supreme Court has held that the ordinary trade-offs of plea bargaining, in which a defendant forgoes a right in exchange for a reduction in charges, are legitimate, and do not constitute the equivalent of vindictively punishing those defendants who do choose to exercise their rights. Thus, the scope of this constraint also is more theoretical than practical.

Finally, it must be emphasized that even these limited legal constraints, even in theory, only permit a challenge to a charge that is brought for improper reasons; they do not provide a basis for requiring action by a prosecutor who has decided not to bring charges. Victims, police agencies, and members of the public have not been permitted, for example, to use selective prosecution arguments in an effort to force prosecutors to be more aggressive against favored groups.

Thus, when a prosecutor chooses not to bring a case, it can probably be said that he exercises discretion that has elements of all three of Professor Dworkin's types. Even where there is general agreement on the standard prosecutors should apply—for example, that a prosecutor should not bring charges where the evidence would be insufficient to support a conviction—the prosecutor exercises substantial discretion in Dworkin's first weak sense, since determining the quantity and quality of evidence necessary to convict requires the exercise of substantial experience and judgment, and similarly qualified lawyers might well disagree about the decision in a particular case. Similarly, like the baseball umpire, the prosecutor's decision not to proceed in a particular case is within her discretion in the second weak sense, because no court has the authority to reverse that judgment, however mistaken it might be. Finally, the prosecutor has, for the most part, discretion in the strong sense as well, because, outside the limited zones in which the prosecutor's judgment might in principle be regarded as unlawful, it is up to the prosecutor herself to decide what principles should influence the decision whether to proceed and how much weight should be given to each.