PROSECUTION: COMPARATIVE ASPECTS

In the United States, the prosecutor is probably the most important decision-maker in the criminal process. The impetus to begin a criminal investigation usually emanates from a private complainant, and it is ordinarily the police who conduct the bulk of investigations; but the determinations whether to charge a suspect, what to charge him with, and what sanctions eventually to impose are made or substantially influenced by the prosecutor. The wide scope of American prosecutors' discretionary power has long remained undetected. It became the subject of scholarly discussion only in the 1970s, in the wake of seminal studies (Miller; Davis), and the debate about the need to limit and control prosecutorial discretion intensified in the 1980s when the introduction of limitations on judges' discretion in sentencing shifted even more power to the prosecutor (see Stith and Cabranes, pp. 130–142). Yet, even though express statutory authority is lacking, the existence of broad prosecutorial discretion is still widely accepted as part of the U.S. legal tradition. Several key features of the U.S. system of criminal justice are indeed dependent on prosecutors' freedom of action. Without it, selective enforcement of criminal laws and diversion of marginal offenders from the criminal process would be impossible, and plea bargaining would be likely to disappear. To demand the abolition or even a radical curtailment of prosecutorial discretion would thus not be realistic, and it would moreover not be desirable to have an overly rigid system of criminal justice lacking the prosecutor's office as a filter for those cases which are nominally criminal but, for various reasons, do not merit conviction and punishment.

This does not mean, however, that the American prosecutor's discretionary power should have to remain as broad and uncontrolled as it has traditionally been. In the United States, the fact that the great majority of state prosecutors are elected and thus responsible to the people is widely regarded as sufficient to control their powers. Although this form of control may be useful with respect to overall law enforcement policy, it has little impact on prosecutors' day-today decision-making and on their treatment of individual cases. The American prosecutor's political responsibility, which becomes effective only when and if the prosecutor seeks reelection, thus tends to mask the need for more direct limits on his discretionary powers as well as for controls upon decisions the prosecutor and his or her assistants make in individual cases. Foreign legal systems have developed various mechanisms in that regard. In some countries, statutes explicitly limit prosecutors' discretionary authority, and in many legal systems, the prosecutor is subject to various forms of control by courts, victims, or private citizens.

Restrictions on prosecutorial discretion of course do not exist in a vacuum. Their significance depends not only on the allocation of functions within the prosecutorial system but also on, for example, the general orientation of the criminal process ("truth-finding" versus conflict resolution), the relationship between prosecutors and courts, career structures within the prosecution service, case loads, crime rates, historical background, and popular expectations and conceptions of the criminal justice system. Before one makes changes based on a comparative perspective, one should therefore carefully assess the ramifications such changes may have for the system as a whole: even a slight reduction in the present scope of prosecutorial authority may lead to an unforeseen increase in some other agent's formal or informal powers. Only with that caveat in mind should recommendations be based on foreign ways of organizing prosecution, even when foreign models appear to be promising examples of more predictable and equitable ways of administering criminal justice.

This article focuses on the prosecutorial systems of four European countries: Austria, England, France, and Germany. Each of these countries has developed a different mode of organizing and controlling the prosecution function.

Who prosecutes?

In the United States, prosecution—except for the federal system—is organized locally; prosecutors are elected or appointed, frequently on the basis of political preference. By contrast, in Europe the prosecutorial corps is typically structured hierarchically on a nationwide or statewide basis. Prosecutors have the status of civil servants but lack judicial independence. Ultimate responsibility for the selection and appointment of prosecutors resides in a government-level official, usually the minister of justice. Appointment and advancement of local prosecutors are largely independent of political considerations and are based on performance in examinations and on merit in service. Young lawyers usually begin their prosecutorial career soon after completing their legal training, which includes some practical experience as an intern in a criminal court and/or prosecutor's office. Many prosecutors follow the prosecutorial career pattern until retirement. On the Continent, it is quite unusual for a prosecutor to become a private attorney, a law teacher, or a full-time politician.

According to legal theory, the prosecutorial function belongs to the executive branch of government. It is thus the minister of justice who is ultimately responsible for the organization of criminal prosecution. The minister also has the authority (rarely used) to issue general as well as specific instructions to prosecution personnel.

In marked contrast to American prosecutors, their continental counterparts display a strong affinity with the judiciary. They share attitudes and perceptions, and prosecutors see themselves more as detached officers of the law than as partisan advocates. In some countries, it is even customary for a career prosecutor to serve as a judge for a few years, and for a judge who aspires to higher office to do a stint as a prosecutor.

The proximity between the offices of judge and prosecutor in Europe has deep historical roots. In the traditional inquisitorial process, which prevailed on the Continent until the nineteenth century, the judge was responsible for investigating a criminal case as well as for deciding on the guilt or innocence of the defendant. He was thus called upon to determine the sufficiency of his own inquiry. The obvious conflict inherent in that dual role called for a separation of functions. In post-Revolutionary France, the procurator, who had since the Middle Ages represented the fiscal interests of the Crown, was entrusted with the prosecution of crime, thus leaving the trial judge to be a neutral arbiter. In the 1840s, this model spread from France to other European countries. The public prosecutor, who remained the sovereign's servant (and therefore continued to be regarded with due suspicion by many liberals), was created from the rib of the judge.

Several characteristics of the Continental prosecutor can be explained as part of that judicial heritage. In some countries, criminal procedure codes expressly require prosecutors to collect and consider exonerating as well as incriminating evidence (Strafprozessordnung 1960 in der Kundmachung der Wiederverlautbarung vom 9. Dezember 1975, Bundesgesetzblatt für die Republik Österreich 1975, stück 211 [Austrian StPO], para. 3; Strafprozessordnung vom 7. Januar 1975, Bundesgesetzblatt 1975 I, p. 129 [German StPO], para. 160, 2; Codice di procedura penale de 22 settembre 1988, Decreto del Presidente della Repubblica n. 447 [Italian CPP], art. 358), and under the French code the public prosecutor is described as a public officer who sees to it that justice is done (Code de procedure penale, Loi n. 57–1426 du 31 décembre 1957 [French CPP], art. 31). If it turns out in the course of the trial that the evidence does not support a conviction, the prosecutor is expected to withdraw charges or to ask the court to acquit the defendant. The prosecutor can (but rarely does) even file an appeal on behalf of the defendant if he regards a conviction or a sentence as unjustified (Austrian StPO, para. 282, 1; German StPO, para. 296, 2). The spirit of objectivity makes it easy for the continental prosecutor to assume a quasi-judicial role in pretrial proceedings (Brants and Field, pp. 144–145 [on Dutch prosecutors]; Frase, pp. 613–615: Goldstein and Marcus, pp. 249–250). Austrian and German prosecutors even assume the function of a judge in imposing noncriminal sanctions in return for refraining from pressing charges (see Austrian StPO, paras. 90a–90m; German StPO, para. 153a).

The English system differs strongly from the Continental approach. In theory, at least, any citizen can bring criminal charges (s. 6 (1) Prosecution of Offences Act 1985). In practice, however, it is the police who investigate the bulk of crime and initiate the prosecution by charging the suspect. At that stage, the case is turned over to the Crown Prosecution Service who determine whether to go forward with the prosecution or to refrain from pursuing it further. The Crown Prosecution Service, which was introduced in 1986, bases its decisions on an independent evaluation of the evidence as well as on policy considerations. With respect to the former, prosecutors are limited to reviewing the written materials presented by the police; they do not themselves talk with victims or interrogate witnesses or the defendant (see Ashworth, pp. 160–172). The Crown Prosecutors' discretion in determining the public interest in prosecution is guided by paragraphs 6.4 and 6.5 of the Code for Crown Prosecutors, which offers two sets of legitimate considerations, one pointing toward prosecution, the other indicating a lack of public interest. Examples of factors suggesting nonprosecution are the lack of serious harm caused by the offense, a long delay since its commission, the old age or bad health of the defendant, and the fact that the offender has put right the loss or harm that was caused. As a rule, however, the Crown Prosecution Service go forward with the prosecution whenever there appears to be sufficient evidence to convict (Fionda, p. 24). Because the police will already have weeded out many cases that do not merit criminal sanction or can be resolved by merely cautioning the suspect, and because police files will reflect the police's assessment of the remaining cases as serious (see Brants and Field, pp. 140–143), public interest considerations do not frequently move the Crown Prosecution Service to drop convictable cases. In 1998–1999, the Crown Prosecution Service discontinued 12 percent of prosecutions in Magistrates' Court; in a further substantial percentage of cases, the prosecution did not offer any evidence, thus bringing about a directed acquittal of the accused (Ashworth, pp. 175–176; Sprack, p. 61).

Although the police, in conjunction with the Crown Prosecution Service, are responsible for prosecuting the majority of English cases, about one-quarter of prosecutions are brought by other agencies or by private individuals. The Inland Revenue Department, for example, investigates and prosecutes cases of tax fraud, and the Department of Health and Social Security prosecutes fraudulent benefit claims (see Lidstone, Hogg, and Sutcliffe, pp. 34–94). Prosecutions by individuals are rare except in cases of common assault. When a private individual brings a case before the criminal court, the Crown Prosecution Service can at any time take over the prosecution, either in order to go forward with it or to discontinue it (ss. 6 (2), 23 Prosecution of Offences Act 1985). The private complainant has no legal recourse against such action (Sprack, p. 62).

Criminal investigation and the prosecutor

When one looks at the relevant law, the Continental prosecutor tends to play a leading role in the investigation process. His formal position depends on whether he is to share this role with the investigating magistrate (juge d'instruction in France, Rechter-Commisaris in the Netherlands), a judicial officer with a specific mandate to conduct pretrial investigations. But even when that is the case, the prosecutor is entrusted with the investigation of all but the most serious offenses. Although the law describes the prosecutor's role as "conducting" the investigation, the prosecutor generally delegates routine operations to the police. As a result, it is almost invariably the police who are in fact conducting the bulk of criminal investigation.

More specifically, legal arrangements in the three Continental countries treated here are as follows:

In Austria, police and all other public authorities are obliged to report to the prosecutor any suspicion of a criminal offense (Austrian StPO, para. 84). The prosecutor must then follow up on such reports by directing police or an investigating magistrate to obtain further information by interrogating witnesses and securing relevant physical evidence (Austrian StPO, paras. 87, 88). In the most serious cases, which are to be tried before a jury court, the prosecutor will then turn over the investigation to the investigating magistrate, who is responsible for determining whether there is sufficient evidence to make the defendant stand trial. Although the prosecutor is precluded from performing acts of investigation himself, he (as well as the suspect) can request the investigating magistrate to conduct particular acts of investigation (Austrian StPO, para. 97). The prosecutor can at any time terminate the magistrate's investigation by declaring that he no longer wishes to prosecute the case, and he can do so even when the magistrate has returned the file after closing the investigation with the indication that there is sufficient cause to proceed to trial (Austrian StPO, para. 112). The investigating magistrate, on the other hand, can dismiss the case even against the prosecutor's wishes (Austrian StPO, para. 109). The great majority of cases do not require involvement of an investigating magistrate. In these cases, it is the prosecutor who supervises and guides the investigation.

French procedure is similar in many respects. However, the French prosecutor can personally perform acts of investigation; he can, for example, interrogate witnesses and, if a suspect has been caught while committing the offense or soon thereafter, conduct searches and seizures on the spot (French CPP, arts. 54, 56). As is the case in Austria, investigations conducted by magistrates are in practice the exception rather than the rule (see Frase, p. 667), and it is thus the prosecutor who is in charge of investigating all but the most serious criminal cases.

Germany abolished the Office of Investigating Magistrate in 1975 and since that time places full responsibility for pretrial investigations in the hands of the prosecutor. Police are to undertake those investigative measures which are immediately necessary to avoid loss of critical evidence, but they must then report to the prosecutor without delay (German StPO, para. 163). The prosecutor conducts the investigation, calling upon the police for assistance to the extent necessary (German StPO, paras. 160, 161). As in other Continental legal systems, acts involving more serious intrusions into citizens' liberty or privacy (e.g., pretrial detention, wiretaps, and searches and seizures) require judicial permission prior to their execution or, if exigent circumstances made immediate action necessary, subsequent authorization by a magistrate.

The legal situation is somewhat different in England. According to English law, investigation is generally the task of the police, and the members of the Crown Prosecution Service are limited to requesting the police to undertake further investigatory acts if that is deemed necessary. The police are not compelled by law to honor such requests. The advantage of the English model is its clear separation of functions: the police investigate and initiate a prosecution, and the prosecutor makes all further determinations. The drawback of that model is that the prosecutor must accept the results of the investigation as presented to him and is dependent on the goodwill of the police for gathering further information he may regard as necessary for intelligent decision-making.

The Continental model avoids this problem by casting the prosecutor in the role of "conducting" the investigation as well as directing and supervising police activities. Continental laws describe the police in criminal proceedings as subservient to the prosecutor (see, e.g., French CPP, art. 41-2; German StPO, para. 161), and they are thus duty-bound to carry out all investigatory acts the prosecutor demands. The prosecutor's control over police is, at the same time, regarded as a guarantee of the police's adherence to relevant legal rules in pretrial proceedings.

In all Continental systems, however, reality differs from the statutory arrangement described above. In practice, the balance of powers established by the legislature is invariably tilted in favor of the police. At least in routine matters, the police clear up cases completely before even informing the prosecutor of their existence, and prosecutors typically defer to the police with respect to the investigation. This does not necessarily mean, however, that prosecutors always accept the findings of the police without further inquiry. In Germany, it has been found to be quite common for the prosecutor to request police to undertake further investigations if he or she is willing to press charges but perceives a need to obtain additional information to strengthen the position of the prosecution (Steffen, p. 183). Yet it is only in very serious or spectacular cases that the prosecutor actually directs the police investigatory activities or takes the investigation into his or her own hands. German prosecutors may sometimes personally interrogate the victim or an important witness—witnesses are under a legal duty to appear before the prosecutor but need not talk to the police (German StPO, para. 161a)—but even such limited involvement in the investigation usually occurs at the request of the police. According to one German study, prosecutors personally conducted parts of the investigation in only 1–5 percent of the cases (Blankenburg, Sessar, and Steffen, p. 99). The only area in which Continental prosecutors tend to investigate on their own is white-collar crime. Several prosecutor's offices have established departments staffed with lawyers, accountants, and other specialists, which have monopolized the investigation of serious economic crime.

The fact that Continental prosecutors largely abstain from investigation has a significant impact on the allocation of powers in the pretrial process. Although the police lack official authority to dismiss cases, there is strong evidence that they can and do predetermine prosecutorial decision-making by the amount of investigative effort they invest in particular types of offenses. The police, moreover, collect and electronically store large amounts of information on crime and offenders, and they often shield this information even from prosecutors. As most Continental prosecutors are unable or unwilling to counteract such police strategies, prosecutors' impact on the case flow from detection of an offense to its adjudication is merely negative: they can screen out cases previously "cleared" by the police, but they cannot restore cases lost through strategic decisions of the police or through insufficient police work.

To a large extent, this state of affairs is inevitable because the police monopolize the manpower, expertise, information, and equipment necessary for successful investigation. Yet the possibilities of controlling police decisionmaking in the area of crime detection and investigation ought to be further explored. Because prosecution policy is heavily dependent on the allocation and deployment of police resources, these matters should not be left to ad hoc determination by police chiefs or individual officers. A thorough restructuring of decision-making and control processes in criminal investigations may well be necessary. Such reforms would have to start with accepting the fact that prosecutors do not conduct or even direct investigations. Shedding the fiction of prosecutorial domination of the investigation process might be the key to a realistic and feasible definition of the prosecutor's role. This role should primarily be supervisory. It could have the following features: police inform the prosecutor as early as possible of each prima facie plausible report or complaint of an offense; the prosecutor may give general or specific advice but leaves the conduct of the investigation to the police, who in turn make continuously available all relevant information to the prosecutor and confer with him when major strategic or tactical decisions have to be made. When the police deem the investigation complete, they submit their findings, and the prosecutor can demand further information if necessary. The decision whether to prosecute should be the prosecutor's alone.

The decision to prosecute

The crucial function of the prosecutor in any legal system is to determine which cases should be brought before the court for adjudication and which should be disposed of in other ways. In all legal systems, the prosecutor has some leeway in making this decision, yet the amount of freedom the prosecutor enjoys differs. American law grants the prosecutor almost unlimited discretion, whereas some European legal systems are considerably less liberal.

Discretion to dismiss. Because prosecutors do not in fact control criminal investigations, they receive from police many cases as "cleared" although the suspect's legal guilt is doubtful or sufficient evidence for establishing his responsibility in court is lacking. It would not only be a waste of judicial resources but would also be unfair to the suspect to make him stand trial if a conviction is unlikely. The prosecutor therefore has the important task of independently assessing the legal and factual circumstances of each case as well as the available evidence. He should dismiss as early as possible cases that cannot be successfully prosecuted. Standards of evidence required for preferring charges are formulated in different ways. A representative formulation might be that of the English Code for Crown Prosecutors, which proposes that charges should be brought only if there is a reasonable prospect of conviction, that is, if conviction is more likely than acquittal (Code for Crown Prosecutors, para. 5; see Ashworth, pp. 161–168).

Determining the chances of conviction implies predicting the behavior of decision-makers, that is, of the professional and/or lay judges comprising the court of first instance. Although the "prospect" test may be phrased objectively, that is, whether a "reasonable" court following the law would tend to convict, the prosecutor needs to make a prediction based on his or her personal assessment of the possible outcome of a trial—an assessment that is inevitably colored by the individual prosecutor's degree of confidence and willingness to take risks. This individual judgment on the strength of the prosecution case should, however, be distinguished from the concept of prosecutorial discretion. The prosecutor exercises discretion, in its proper sense, only when he or she regards the suspect's conviction as likely and yet refrains from filing charges. Discretion thus does not refer to the sufficiency of the evidence but is based on considerations of policy. The American system broadly permits discretionary dismissals of convictable cases and thereby casts the prosecutor in the role of an independent policymaker. In other systems, decisions on criminal policy are regarded as the exclusive domain of the legislature, and the prosecutor's task is consequently limited to determining the sufficiency of the evidence. One possible way of restricting or even eliminating prosecutorial discretion is adoption of the principle of "mandatory" prosecution. This principle, which is still recognized in a number of European systems (in Italy, the prosecutor's duty to file charges whenever there is sufficient evidence is even enshrined in art. 112 of the constitution), seeks to ensure the equal application of the criminal law by mandating its full enforcement. The law denies the prosecutor authority to make exceptions in individual cases, thus precluding favoritism or considerations of political partisanship. The source of this principle can be found in the nineteenth-century reformers' distrust of public prosecutors' impartiality—at that time, prosecutors were still regarded as obedient servants of the government, prone to abuse their power for political purposes.

Austrian law is an example of a system based on the principle of mandatory prosecution. The code of criminal procedure requires the prosecutor to do everything necessary to bring about the conviction of an offender whenever the prosecutor has received information of a criminal offense. Exceptions apply only to crimes committed abroad and to multiple offenses committed by one individual; in the latter case, the less serious charges need not be brought (Austrian StPO, para. 34). The law precludes the prosecutor from dismissing charges simply because he or she regards the case as unimportant or because the offender does not seem to merit punishment. Since 1999, the prosecutor can, however, withhold prosecution if the offender's guilt is not significant and the suspect made restitution to the victim or a payment to the state or did community service (Austrian StPO, para. 90a). This provision significantly limits the practical impact of the mandatory prosecution principle. Even before the introduction of the option of conditional dismissal, however, Austrian prosecutors did not in fact prosecute each and every criminal offense. One way to avoid wasting resources on trifles has been opened by a provision of the criminal code that permits the court to acquit the defendant when the offense does not merit punishment, especially in view of the fact that the offender had made restitution to the victim (Strafgesetzbuch der Republik Österreich of 1975, para. 42). If the conditions of this provision are met, the prosecutor can anticipate the court's decision and refrain from prosecution on the theory that the suspect would not eventually be convicted. In other cases, the prosecutor may conceal what is in effect a discretionary dismissal behind the label of "insufficient evidence"—he or she may cite doubts regarding the suspect's criminal intent or may decline to make the police follow up on an incomplete investigation (Driendl, pp. 254, 280–283; but see Nowakowski, pp. 272–275). Recognition of the principle of mandatory prosecution does thus, in effect, not preclude Austrian prosecutors from concentrating their resources on serious cases and from avoiding prosecution of even slightly doubtful minor offenses.

In Germany, the first nationwide Code of Criminal Procedure of 1877 (Strafprozessordnung vom 1. Februar 1877. Reichsgesetzblatt 1877) introduced the principle of mandatory prosecution, which can still be found in the actual version of the Code (German StPO, para. 152, 2). In the meantime, however, practical considerations led to the recognition of a growing number of exceptions (Frase and Weigend, pp. 337–339). The most important of these exceptions concerns "minor" offenses, that is, those with a statutory minimum sentence of less than one year imprisonment. With respect to these offenses—which comprise the great majority of cases reported to the police—the prosecutor can refrain from filing charges whenever the offender's guilt is "minor" and prosecution is not in the public interest (German StPO, para. 153). The prosecutor also has the option of imposing certain obligations on the suspect as conditions of nonprosecution; such obligations can be, for example, payments to be made to the victim or to a charitable organization (German StPO, para. 153a, see below). Further exceptions from the principle of mandatory prosecution exist for offenses committed abroad (German StPO, para. 153c), and for (even serious) offenses directed against interests of the state when prosecution and public trial might cause additional harm to these interests (German StPO, para. 153d). If the suspect is alleged to have committed more than one offense, the prosecutor can limit prosecution to the most serious charge and drop all others (German StPO, para. 154). Taken together, these exceptions almost seem to swallow the rule. German prosecutors are under an unqualified duty to prosecute convictable offenses only if the suspect has committed a serious felony with a statutory minimum penalty of one year imprisonment, such as murder, robbery, rape, or arson.

In fact, prosecutors make extensive use of their authority to dismiss cases on policy grounds. In 1996, prosecutors brought charges in only twenty-eight of one hundred cases in which the police were able to name a suspect. Twenty-seven out of one hundred cases were dismissed for lack of sufficient evidence, twenty-one cases were not prosecuted because there was no "public interest," and in six cases there was a conditional dismissal under para. 153a Code of Criminal Procedure (Statistisches Bundesamt, p. 118). (Eighteen out of one hundred cases were disposed of in other ways, for example, by referral to juvenile court or administrative proceedings.) Most German prosecutors nevertheless assume that prosecution should be the rule, at least with respect to non-trifle offenses. They actually welcome the principle of mandatory prosecution as a shield against pressure from outside. Yet they use their discretionary authority to get rid of the "small fry" that would only clog the court system without leading to significant sanctions.

The French Code of Criminal Procedure does not explicitly determine whether the prosecutor has discretion or is duty-bound to submit cases with sufficient evidence to the court. Yet Art. 40 CCP, which simply states that the prosecutor "receives complaints and denunciations and decides how to proceed," has generally been interpreted to confer broad discretion upon the prosecutor (Pradel, pp. 345–346). If the prosecutor does not wish to prefer charges, he or she simply lays the file aside, and prosecution can be resumed at any time. The code does not contain any guidelines on the exercise of prosecutorial discretion. Policy reasons that may prompt French prosecutors to dismiss convictable cases include restitution made to the victim, triviality of the offense, and availability of noncriminal sanctions (Frase, pp. 614–615). Prosecutors have formal authority to conditionally dismiss certain charges.

In England, both police and prosecutors have broad discretion whether to bring and sustain charges. Whereas police discretion is not formalized save by standing orders on a local level, the Code for Crown Prosecutors contains a list of considerations prosecutors are to take into account when determining whether to go forward with a case presented by the police. As a general guideline, the code states that prosecution in cases of any seriousness shall usually take place unless there are public interest factors against prosecution that clearly outweigh those tending in favor (Code for Crown Prosecutors, sec. 6.2). The list of factors suggesting a decision in favor of prosecution mainly contains criteria relating to the seriousness of the offense (for example, premeditation, a particularly vulnerable victim), whereas factors disfavoring prosecution include insignificant harm, expectation of a small penalty, negative effects of the prosecution on the victim's mental or physical health, and compensation paid by the defendant to the victim (Code for Crown Prosecutors, secs. 6.4, 6.5).

This brief overview shows that statutory rules and principles make little difference for actual prosecution practice. Prosecutors tend to follow simple rules of reason; they enforce the law vigorously with respect to serious crime, but they are selective in prosecuting minor offenses, giving due regard to characteristics of the offender and the victim as well as to noncriminal alternatives, most prominently restitution.

Informal sanctions. In many systems, prosecutorial discretion extends beyond the simple alternative between prosecution and complete inaction. Informal noncriminal sanctions are often imposed on suspects as a condition of non-prosecution. In the United States, such practices are known as pretrial diversion.

In the three Continental systems considered in this entry, the law authorizes the prosecutor to propose to the suspect that he fulfil certain obligations in exchange for having his case dismissed (Austrian StPO, paras. 90a–90m; French CPP, art. 41-1–41-3; German StPO, para. 153a). If the suspect consents and complies with these obligations, he will not be prosecuted. The theory of these provisions is that the suspect, by accepting and fulfilling the obligations, eliminates the necessity of punishment because the purposes of punishment (deterrence and/or rehabilitation) have already been met (see Burgstaller, p. 12). In the Austrian, French, and German systems, possible obligations include payments to the state or to a charitable organization (these are in practice imposed most frequently), restitution payments to the victim, and community work. In France, the suspect can also be required to give up his driver's license for a period of four months or less, or to turn over to the state objects that had been used in committing the offense (French CPP, art. 41-2). When the suspect has fulfilled the obligations he can no longer be prosecuted for the offense of which he had been suspected. While these features are identical in all three systems, rules differ as to the need for judicial consent to the prosecutor's disposition. In France, the president of the court has to ratify any agreement between the suspect and the prosecutor's office (French CPP, art. 41-2, 6); in Germany, judicial consent must be obtained (but is almost never withheld) only when the offense is fairly serious; in Austria, the prosecutor can impose the statutorily authorized conditions for dismissal without involving the court.

When the option of "conditional dismissal" was first introduced in Germany in 1975, the new law was heavily criticized on the grounds that it shifted sentencing power from the judiciary to the prosecutor and that undue pressure was put on the suspect to comply with the prosecutor's offer even when he was in fact innocent or when his guilt could not have been proven at trial. This criticism has since subsided in view of the overwhelming practicality of this instrument. As noted above, prosecutors employ this diversionary tool on a regular basis (see Weigend) and have even extended its use to large-scale white-collar crime where required payments sometimes reach extremely high amounts. The fact that Austria and France have adopted similar models in 1999 seems to indicate that the idea of prosecutorial sanctioning, in spite of its interference with the traditional judicial monopoly on sentencing, has internationally been regarded as a sensible way of disposing of minor cases.

In England, conditional dismissal of charges has not yet been adopted for adult suspects. The widespread practice of police cautioning, which consists of giving the suspect a formalized warning that his or her criminal conduct will not be tolerated further (Sprack, pp. 65–66), does not imply any immediate consequences for the individual concerned. Introduction of cautioning plus informal sanctions has been discussed, but it seems that such practice would be acceptable only if it was placed in the hands of a public prosecutor, not the police (Brants and Field, pp. 139–143; Fionda, pp. 39–40).

Discretion to select charges. One important aspect of prosecutorial discretion in the common law systems is the prosecutor's power to select the legal charges he brings against the defendant. To most instances of criminal conduct, more than one section of the criminal code can be applied, and each section may provide for a different measure of punishment. By choosing the amount and nature of charges, the American prosecutor often predetermines to some extent the sentence the defendant receives if convicted. For that reason, the selection of charges frequently plays an important role in the prosecutor's negotiations with defense counsel about the conditions of a guilty plea.

The inquisitorial structure of Continental criminal procedure makes it impossible for the prosecutor to bind the court by his legal assessment of the case. The formal accusation filed by the prosecutor determines the scope of the court's inquiry only with respect to the factual situation (person of the suspect, time, and place of the conduct in question); that is, the court cannot punish the defendant for conduct on an occasion not mentioned in the formal accusation. But the court remains free to apply provisions of the law other than those cited by the prosecutor (Austrian StPO, para. 262; French CPP, art. 351; German StPO, para. 155, 2). Any "bargaining" that may occur in Continental systems thus cannot relate to the prosecutor's charge but must go directly to the sentence to be imposed by the court.

In England, courts are bound by the legal charges brought, yet they can convict for a lesser offense included (expressly or by implication) in the offense charged (Criminal Law Act 1967, sec. 6).

Control of discretion. The American system of prosecution relies almost exclusively on the political responsibility of prosecutors and provides few other checks on their discretionary powers. In Europe, two means of confining and controlling prosecutorial discretion exist: private prosecution and court supervision. Both are of interest especially insofar as they limit the prosecutor's inconspicuous power not to prosecute.

Private prosecution. Private prosecution has historically been the primary form of prosecution. In most legal systems, prosecution by an agent of the state has gradually superseded private prosecution. Yet the victim's (or any citizen's) right to bring criminal charges can serve as a check on the public prosecutor's inaction, and many modern legal systems provide for this possibility.

In England, private prosecution still forms the theoretical basis of the prosecution system. But in practice only a minuscule portion of prosecutions is actually brought by private citizens (see above). This is not surprising because private citizens typically lack the financial means and legal expertise necessary to launch successful prosecutions. Private prosecutions are further discouraged by the fact that the Crown Prosecution Service can at any time take the prosecution away from the private citizen and terminate it. The citizen has no legal remedy against this measure.

In Germany and Austria, some criminal offenses are labeled "private prosecution offenses" and can be brought to trial by the victim without any involvement of the public prosecutor. These offenses include slander, breach of confidentiality, and, in Germany, simple assault and trespass (Austrian StPO, paras. 2 and 46; German StPO, para. 374). In Germany, the public prosecutor can at any time take over the proceedings, and the court can dismiss the case if it deems the defendant's guilt to be minor (German StPO, paras. 377, 2 and 383, 2). For the same reasons as in England, private prosecutions in Austria and Germany are infrequently brought (in 1996, only 0.2 percent of German prosecutions were initiated by private individuals; Statistisches Bundesamt, p. 84), and even less often lead to conviction. The private prosecutor must gather and present evidence without assistance from the police, he must in some cases attempt reconciliation with the offender before he can bring charges (German StPO, para. 380), and he runs the incalculable risk of dismissal, with costs, in court. The public prosecutor, on the other hand, can remain inactive with respect to offenses subject to private prosecution, which is thus largely inefficient as a control mechanism.

A different approach is to accord the victim the right to request a judicial investigation. This model exists both in Austrian and in French law. In Austria, the victim can appear as a "subsidiary prosecutor" whenever the public prosecutor has declined to open a formal investigation or has closed it without filing charges, claiming lack of evidence. The victim can, however, not bring the case directly to trial but is limited to requesting an investigating magistrate to conduct or continue an investigation. The victim cannot insist on prosecution when the prosecutor has decreed a conditional dismissal (see above) (Austrian StPO, para. 48). French law goes one step further by permitting any victim personally harmed by the offense to declare himself or herself a "civil party" (partie civile ) and to set in motion a public prosecution (French CPP, arts. 1,2 and 2). The victim can do so by requesting the investigating magistrate to conduct an investigation (French CPP, art. 85) or, with respect to less serious offenses within the jurisdiction of the correctional or police court, by directly submitting his or her claim to that court (French CPP, art. 388). In either case, the public prosecutor is compelled to join in the victim's suit. The victim can thus initiate the criminal process without bearing sole responsibility for collecting the evidence necessary for conviction. This far-reaching right was originally limited to individual victims; since the 1980s, the right to act as partie civile in criminal proceedings has been extended to a large number of organizations dedicated to supporting victims of crime.

Court supervision. The prosecutor's affirmative decision to file criminal charges is always reviewable by a court. In most legal systems, summary judicial review of the strength of prosecution evidence is available even before the defendant is required to stand trial; in the absence of such review, the truth of the accusation is in any event tested in the trial itself. The defendant cannot claim, however, that the prosecutor should have refrained from filing charges on the basis of his or her discretionary authority.

Judicial review is limited with respect to the prosecutor's decision not to act. German law gives a victim who had reported the offense to the police the right to appeal to the appellate court when the prosecutor has declined prosecution on evidentiary grounds. If the court finds, on the basis of the victim's brief and the prosecutor's file, that sufficient grounds for prosecution exist, it orders the prosecutor to bring charges. The complaining victim can then join the prosecution as a "supplementary prosecutor" (German StPO, paras. 171–175 and 395, 1). The court cannot, however, review a prosecutor's refusal to prosecute if it was based on policy grounds.

Such review may be available in England. In a number of decisions since the 1960s, the Court of Appeal indicated that it was willing to review decisions not to bring prosecutions if the victim could establish that the decision in question was unreasonable (Ashworth, pp. 188–189, citing cases).

Granting the victim access to the courts for the purpose of checking the appropriateness of a prosecutor's decision not to go forward appears to be a more effective control on prosecutorial decision-making than requiring victims to bring a private prosecution. Meaningful review would, however, require that the prosecutor state in writing the general rules that guide the exercise of his discretion.

Thomas Weigend

See also Adversary System; Comparative Criminal Law and Enforcement: China; Comparative Criminal Law and Enforcement: England and Wales; Comparative Criminal Law and Enforcement: Islam; Comparative Criminal Law and Enforcement: Russia; Criminal Law Reform: Europe; Criminal Law Reform: England; Criminal Procedure: Comparative Aspects.

BIBLIOGRAPHY

Ashworth, Andrew. The Criminal Process. An Evaluative Study. Oxford, U.K.: Clarendon Press, 1994.

Blankenburg, Erhard; Sessar, Klaus; and Steffen, Wiebke. Die Staatsanwaltschaft im Prozess strafrechtlicher Sozialkontrolle. Berlin: Duncker und Humblot, 1978.

Brants, Christje, and Field, Stewart. "Discretion and Accountability in Prosecution." In Criminal Justice in Europe. Edited by Phil Fennell, Christopher Harding, Nico Joerg, and Bert Swart. Oxford, U.K.: Clarendon Press, 1994. Pages 127–148.

Burgstaller, Manfred. "Über die Bedeutung der neuen Diversionsregelungen für das österreichische Strafrecht." In Diversion. Ein anderer Umgang mit Straftaten. Edited by Roland Miklau and Hans Valentin Schroll. Vienna: Verlag Österreich, 1999. Pages 11–18.

Davis, Kenneth C. Discretionary Justice: A Preliminary Inquiry. Baton Rouge: Louisiana State University Press, 1969.

Driendl, Johannes. "Staatsanwaltschaft und Strafverfolgung in Österreich." Funktion und Tätigkeit der Anklagebehörde im ausländischen Recht. Edited by Hans-Heinrich Jescheck and Rudolf Leibinger. Baden-Baden: Nomos, 1979. Pages 191–327.

Fionda, Julia. Public Prosecutors and Discretion: A Comparative Study. Oxford, U.K.: Clarendon Press, 1995.

Frase, Richard S. "Comparative Criminal Justice as a Guide to American Law Reform: How Do the French Do It, How Can We Find Out, and Why Should We Care?" California Law Review 78 (1990): 539–683.

Frase, Richard S., and Weigend, Thomas "German Criminal Justice as a Guide to American Law Reform: Similar Problems, Better Solutions?" Boston College International and Comparative Law Review 18 (1995): 317–360.

Goldstein, Abraham S., and Marcus, Martin. "The Myth of Judicial Supervision in Three 'Inquisitorial' Systems: France, Italy, and Germany." Yale Law Journal 87 (1977): 240–283.

Kock, Gerald L., and Frase, Richard S. The French Code of Criminal Procedure. Translated from French by Gerald L. Kock and Richard S. Frase. Littleton: Rothman, 1988.

Lidstone, K. W.; Hogg, Russell; and Sutcliffe, Frank. Prosecutions by Private Individuals and Non-Police Agencies. Royal Commission on Criminal Procedure, Research Study No. 10. London: Her Majesty's Stationery Office, 1980.

Miller, Frank W. Prosecution: The Decision to Charge a Suspect with a Crime. Edited by Frank J. Remington. Boston: Little, Brown, 1970.

Nowakowski, Friedrich. "Die Behandlung der Bagatellkriminalität in Österreich." Zeitschrift für die gesamte Strafrechtswissenschaft 92 (1980): 255–294.

Pradel, Jean. Droit penal. Procedure penale, 9th ed. Paris: Cujas, 1997.

Sprack, John. Emmins on Criminal Procedure, 8th ed. London: Blackstone Press, 2000.

Statistisches Bundesamt. Rechtspflege. Fachserie 10, Reihe 2: Zivilgerichte und Strafgerichte 1996. Stuttgart: Metzler-Poeschel, 1997.

Steffen, Wiebke. Analyse polizeilicher Ermittlungstätigkeit aus der Sicht des späteren Strafverfahrens. Wiesbaden: Bundeskriminalamt, 1976.

Stith, Kate, and Cabranes, Jose A. Fear of Judging: Sentencing Guidelines in the Federal Courts. Chicago: University of Chicago Press, 1998.

Weigend, Thomas. "In Germany, Fines often Imposed in lieu of Prosecution." Intermediate Sanctions in Overcrowded Times. Edited by Michael H. Tonry and Kate Hamilton. New York: Oxford University Press, 1995. Pages 50–55.