"The prosecutor has more control over life, liberty, and reputation than any other person in America."

-- Former U. S. Attorney General Robert H. Jackson

"What are all the limestone facades of Fifth Avenue and all the marble halls and stuffed-leather libraries and all the riches of Wall Street in the face of my control over your destiny and your helplessness in the face of the Power?"

-- Prosecutor XXXXX XXXXXX in Tom Wolfe's "The Bonfire of the Vanities"

THINK WHAT you wish about the outcome of United States vs. Marion Barry, it does highlight an issue that is much bigger than the culpability of one person: The arbitrary and excessive power of prosecutors to select certain individuals and reject others for prosecution. The process leaves prosecutors far too much room to act out of political motives rather than motives that are more purely legal, sensible and community-spirited.

This is not a complaint from a particular quarter of the political spectrum. It has been made by Kenneth Mundy in the defense of Marion Barry and taken up by much of the nation's African-American community, and it has been made just as fervently by The Wall Street Journal and other conservative beacons in criticizing the zeal of U.S. attorneys in prosecuting certain well-heeled individuals under the amorphous Racketeer Influenced and Corrupt Organization (RICO) statute. An issue that such diverse factions agree upon and regard as critical deserves to be taken seriously.

Prosecutors are sworn to bring to justice those who violate the law. Where the law is less than explicit, prosecutors are left to exercise discretion to ensure that cases are processed justly and effectively. The less explicit the law, the greater the discretion; the greater the discretion, the greater the opportunity for unevenhanded treatment of cases -- that is, for disposition in a manner driven by the individual prosecutor. Inevitably, then, the attorneys in some 8,000 state and local prosecution agencies and 93 federal offices weigh not only legal considerations, but practical and political ones as well, in deciding the fate of more than 12 million persons arrested each year.

The opportunities for unevenhanded action run through the entire process and provide a potential playground for politically motivated prosecutors. It is prosecutors who decide which arrests and police charges to file in court and which to reject, how much attention to give each case, whether to induce and accept a plea and reduce charges, how much to reduce charges, whether to induce one defendant to plead guilty to reduced charges in return for evidence against another defendant, and what sentence to recommend to the judge.

In addition, prosecutors' goals and internal guidelines differ widely. One office may emphasize broad crime control, for example, and go after long-term professional criminals; another may try to clean up a neighborhood and target street hoodlums; still another may focus on white-collar crime or political corruption. Prosecutors seldom spell out their philosophy publicly -- possibly so as not to tie their hands -- but their varying approaches clearly can result in uneven application of the law. Available evidence confirms such disparity. Prosecutors tend, for example, to engage in plea negotiations at significantly different rates -- even in neighboring federal regions with similar case mixes and with resources balanced by the Justice Department according to caseloads. In 1984, the 6th U.S. Circuit (Kentucky, Missouri, Ohio, and Tennessee) had 5.8 pleas per trial, while the neighboring 5th U.S. Circuit (Louisiana, Mississippi, and Texas) had 50 percent more -- 8.7 pleas per trial. The Southern District of West Virginia had 9.9 pleas per trial, while the neighboring Western District of Virginia had less than half that -- 4.2.

Variations in plea-bargaining practices are, if anything, more pronounced among local jurisdictions. A study conducted by the Institute for Law and Social Research in 1985 found that in Manhattan, for every felony arrest that went to trial 24 pled guilty; in Washington, only five pled guilty for every one that went to trial. (The difference is primarily a matter of policy rather than heavy caseloads per prosecutor in New York. In District Attorney Robert Morgenthau's Manhattan office, only 3 percent of all felony arrests are rejected; in Washington, U.S. Attorney Jay Stephens's local-crime unit rejects 15 percent.) Prosecutors generally aim to pleas, in a manner of speaking; but some, like Stephens devote considerably more effort to trials than others.

Disparities in prosecution need not be a fact of life, however; they can be brought under control. In fact a strong precedent already exists: sentencing guidelines for judges. Judicial excesses in the exercise of sentencing discretion were documented statistically in a series of studies funded by the Department of Justice in the 1970s and 1980s. These disclosures led to legislation that limited judges' discretion and established sentencing guidelines, first in several states and then, with the passage of the Sentencing Reform Act of 1984, at the federal level.

As one example of how the guidelines work, consider the case of Randall Cecola, a minor figure in the Dennis Levine insider-trading scandal. Cecola pleaded guilty and received a pre-federal-guidelines sentence of five years probation. Had the crime been committed after Nov. 1, 1987, the guidelines would have mandated a sentence of six to 12 months.

While the judiciary has not welcomed the guidelines, it has adapted to them and many judges express relief at not having to wonder about how their sentences stack up against those of their brethren. Moreover, the guidelines have been found to work: A National Institute of Justice compairson of 10,000 felony sentences in North Carolina -- 5,000 before that state's guidleines were enacted and 5,000 after -- revealed that the guidelines shortened the interquartile range of sentences (i.e., the distance from the 25th to the 75th percentile) from three to 10 years to two to six years. At the same time, the guidelines seem to have resulted in no apparent loss in either deterrent effectiveness or "pure justice." The laws that govern prosecution are every bit as broad in some areas as were pre-guideline sentencing statutes. RICO and drug-conspiracy laws, in particular, give prosecutors extraordinary latitude to pick and choose certain defendants and ignore others. Such laws are too looosely worded to prevent prosecutorial excess; they need to be more sharply defined, as do the prosecutors' targeting policies so that citizens can know more clearly the legal limits of behavior. Stephens and his assistants justified singling out Marion Barry from the immense crowd of prosecutable drug users by labeling him a turncoat general in the District's war against drugs, but nothing in the law explicitly supported such an argument. If there had been a set of published federal prosecution guidelines for targeting high public officials who break laws, the appearance that the prosecution was motivated primarily by racial or political considerations might have been substantially lessened.

Many prosecutors may regard the notion of limiting their discretion as an attempt to take the teeth out of the law, but the two are not at all the same. Reduced prosecutorial discretion would not give criminals more freedom to menace society any more than reduced sentencing disparity has necessarily resulted in lighter punishment. (Sentences are tougher under the federal guidelines; North Carolina's and Minnesota's guidelines have had the opposite effect.) The sharpening of laws and publication of prosecutorial guidelines would simply mean a narrowing of discretion in areas now too broadly defined -- areas in which complaints of arbitrary or "selective" prosecution have been strongest and most numerous.

Prosecutors may differ on a particular goal of prosecution, or on a particular method of achieving a goal, but it would seem difficult to object to a more coherent -- and consistent -- basis for making case-processing decisions. Yet prosecutors have tenaciously protected their powers and resisted attempts to commit their rules to widespread publication, arguing that separation of powers prohibits the courts or legislative bodies from altering the discretionary aspects of prosecution.

Continued abuses of discretionary authority will inevitably lead to calls for control by either or both of the other branches of government. Prosecutors themselves can take the initiative in identifying those areas in which guidelines are most needed and putting them in place, so that their practices can be made both more explicit and more evenhanded. If they do not, others are likely to do it for them.

If we are truly a "government of laws, and not of men," as John Adams wrote in 1774, we will begin to take measures to ensure that our citizens are less susceptible to arbitrary prosecution. Kafkaesque justice has no place in the United States.

Brian Forst is visiting lecturer at the George Washington University's School of Business and Public Management. He was, from 1977 to 1985, director of research at the Institute for Law and Social Research, and from 1985 to 1989, director of research at the Police Foundation, both in the District of Columbia.