THE RECENT murder of Gloria Whipple, a 51-year-old Washington artist, has prompted fresh outcries about the failures of our criminal justice system.

It was not only the nature of the murder that angered so many people, though that was disturbing enough: The victim was brutally kicked and beaten after leaving a Smithsonian art class one night, and she died later in the hospital. But even more troubling to the public was the fact that two of the men charged with the murder had been free on bail at the time, awaiting trial in a separate rape case. In fact, one of the men was out on bail even though, when charged in the rape case, he was already on probation following a conviction for assault.

Why was that man still on the streets? Why wasn't he returned to jail when arrested on the rape charge while he was still on probation? Why wasn't he held under the District of Columbia's preventive detention law? Those are some of the questions that citizens have been asking.

Their anger about the crime is understandable. The question is whether it is sustainable. For it is a replay of what happens almost every time a prominent person becomes a victim of a vicious crime. The public fury erupts, the questions are asked -- and then the issue is largely forgotten until we are faced with the next brutal crime.

The point is that the community has to face and make choices in the various steps that could be taken if it is serious about such matters, not merely wait for the next crime to be perpetrated. Those steps are varied, and some are also costly -- and not only in money.

Some might involve amendements to the law or different enforcement. Some consist of nothing more than a tightened administration of existing programs. Others would require substantial outlays of funds, especially if the public is serious about detaining many more potentially dangerous persons. Still others would require fundamentals changes in philosophy and public attitudes.

Consider, for example, the District of Columbia's preventive detention law. It was enacted in the early days of the Nixon administration after the federal Bail Reform Act of 1966 barred the use of money bond as a bail device except to assure presence at trial. Such figures as Deputy Attorney General Richard Kleindienst and D.C. Police Chief Jerry Wilson argues that some alternative method was still needed to hold those who might stay and be a danger to the community, not only those likely to flee.

So it was that the D.C. preventive detention law was passed, the only one in the nation directly allowing detention without bail in non-capital cases, and all sides predicted that it would lead to incarceration of large numbers of defendants. As it turned out, they were wrong. For years the statute was hardly ever used, and while the prosecutor's office recently has stepped up its efforts in this area, detention still is sought in only a fraction of the eligible cases.

"Why? One reason is that high money bond continues to be used by judges, albeit improperly, as a less complicated alternative to detention. A judge can still say that a defendant "is likely to flee" and set bond rather than consider detention proceedings.

Prosecutors themselves have been reluctant to seek preventive detention for several reasons. One is the restrictive time limit involved: Cases must be tried within 60 days once preventive detention is granted. This concern about the speed with which those cases can be brought seems legitimate, and it is unfortunate that bills addressing those issues have not been enacted.

The prosecutors also worry about the requirement that there be an early hearing to demonstrate that a person with a serious criminal record probably committed the offense with which he has been charged. But proposals to eliminate a judicial hearing as a condition to pretrial jailing raise far more troublesome questions.

The presumption of innocence, a cornerstone of American law, extends its basic protection to everyone, including those with criminal records. Not only are a substantial number of those arrested ultimately exonerated, but it seems simply unfair to keep an individual in jail on nothing more than a policeman's or prosecutor's accusation.

The U.S. attorney usually has argued that court hearings, because they must be held early in the process, will force the government to tip its hand, and that this could lead to witness intimidation or worse. But prosecutors in most other metropolitan centers have long adhered to a policy of opening their evidence files to the defense, on the general premise that criminal trials have ceased to be games of surprise, and the more practical notion that when the accused knows of the strength of the evidence against him he will often conclude that his defense is hopeless and plead guilty.

There is somewhat more substance to the witness intimidation argument, but this practice is clearly not so widespread that it should outweigh the very significant policy reasons against jailing a citizen without any impartial decision based on evidence that there is at least a probability that he committed an offense.

What else, then, can be done to improve the present system? In fact, a number of measures might be taken.

Several years ago, the D.C. Superior Court adopted a policy of moving to revoke the release status of any probationer rearrested for a serious offense, and the Board of Parole later followed suit with respect to individuals on parole. Such actions with respect to probationers and parolees raise no substantial constitutional or philosophical problems, and there is no reason why the policy should not be rigorously enforced. Yet this has apparently not been done, at least not as consistently as it might be.

Likewise, energetic measures should be taken against those on bail who violate the terms of thier release. All too often, requirements of abstinence from narcotics, the securing of steady employment, and regular reporting -- conditions designed to keep defendants out of further trouble -- are ignored by both the defendants and the system. A substantial proportion of repeat offenders are found in these categories, and systematic efforts to deal with violations they commit are likely to have a significant impact on the crime problem.

It is also clear that crime on bail is not made sufficiently costly for the perpetrators. While mandatory sentences are not desirable as a general proposition -- largely because they eliminate the flexibility essential to sentencing fairness -- they are an appropriate and necessary deterrent to crime by individuals on bail. To that end, the law might well be amended to provide that an individual who commits a crime while on bail shall on that account alone be given a short but mandatory sentence, that the bail violation offense may not be plea-bargained away, and that the sentence may neither be suspended nor be made concurrent with any other sentences for any other crime.

Several other things could be done, but all require substantial funds. It can confidently be predicted, for example, that a reduction in the period between arrest and trial will proportionately diminish the incidence of new criminal acts by those on pretrial release. As a practical matter, however, that period can be shortened only by providing the additional judges, prosecutors and others needed to deal with the flood of criminal cases more swiftly than is now possible.

In a similar vein, crime by individuals on pretrial release could be significantly reduced by effective supervision. At present, those released on bail theoretically are supervised by the Pretrial Services Agency and by various private organizations. In reality, insufficient funds and personnel render this supervision largely meaningless. Yet experience with juveniles has shown that when intensive supervision and counseling are provided -- at a ratio, say, of one probation officer for every five defendants -- new crimes decline dramatically.

At the other end of the ideological spectrum, if it be the judgment of the public and its elected representatives -- although it is not mine -- that many more suspects than now should be incarcerated while awaiting trial, whether by broadening preventive detention categories or otherwise, that, too, could be accomplished, but again only at substantial expense. The jails, here and elsewhere, normally are filled to capacity, and unless the additional prison population were to be warehoused in totally inadequate space -- a notion that offends civilized values and that is not, in any event, permitted by the Constitution -- more pretrial detention could not be achieved without the construction and maintenance of vast additional prison facilities.

All of that seems obvious. Yet many of those who raise an outcry about crimes committed by defendants on bail are equally vocal in their resistance to providing the funds necessary for these kinds of solutions. When the inconsistency of these positions is pointed out, it is usually suggested that if the criminal justice system managed its affairs more efficiently, additional resources would not be necessary.

Any organization, including courts, prosecutor's offices and correctional departments, can benefit from improved performance, and the public has a right to hold all of them to the highest standards in that regard. But in the end, reliance on more efficiency alone is less a solution that an excuse for refusing to make the choice between crime and the taxes necessary to pay for measures to curb it.

Hardly any candidate for public office fails to promise that through better management he will be able to avoid the hard choices before the American public -- inflation vs. unemployment, energy vs. environmental values, guns vs. butter. Yet experience has shown again and again that, while more efficient administion can provide some improvements in service, in the final analysis they cannot save us from these hard choices. The same principles apply to the criminal justice process. There, too, a choice between difficult alternatives has to be made.

Similar questions govern the resolution of the philosophical questions surrounding the bail-preventive detention problem. Again the public signals are contradictory. When an individual on pretrial release commits a serious offense, there is an immediate demand for a tougher bail policy. On the other hand, reports of overcrowding at the jail or publication of statistics indicating that many of those jailed while awaiting trial are eventually exonerated, are followed by agitation for alternatives to detention.

The fundamental problem, at which there can be no blinking, is that no one, whether judge, prosecutor, social worker or psychologist, can predict with accuracy whether a defendant produced in court on a criminal charge will become a repeater should he be released pending trial. Every effort naturally should be made to attain as much certainty as is humanly possible, and the risk of error can no doubt be minimized by such devices as limiting the categories of offenses and offenders covered by a detention law and requiring that no one may be detained without a prior judicial hearing. But public policy cannot realistically be based on the assumption that even with such safeguards, predictions of future behavior can be made with anything approaching scientific accuracy.

Given that uncertainty, a basic choice must be made between two conflicting policies. One alternative is embodied in the traditional maxim that it is more important to let a hundred criminals go free than to take a chance on unjustly punishing one innocent man. The other is to heed the cry of those who insist that defendants who might be dangers to the community must be kept off the streets at all costs, including the cost that some, or many, who are innocent will be caught in the net. while there may be some constitutional constraints enforceable through the courts, the decision rests primarily with the public and those who formulate and legislate its policies. If their choice among the alternatives is made after a full debate and exploration of the issues, another incident, whether involving crime or unwarranted detention, will still cause frustration and anger. But the community will at least know that it had weighed the risks and had struck a balance expressive of its priorities and its deepest values.