1 Discretion may be said to exist “whenever the effective limits on [a public officer's] power leave him free to make a choice among possible courses of action or inaction.” Kenneth Culp Davis, Discretionary Justice: A Preliminary Inquiry 4 (Baton Rouge: Louisiana State University Press, 1969).Google Scholar

2 Among the useful discussions and empirical examinations of the decision to charge a suspect are those found in Cole, George F., The Decision to Prosecute, 4 Law & Soc'y Rev. 331 (1970); Comment, Prosecutorial Discretion in the Initiation of Criminal Complaints, 42 S. Cal. L. Rev. 519 (1969); Frase, Richard S., The Decision to File Federal Criminal Charges: A Quantitative Study of Prosecutorial Discretion, 47 U. Chi. L. Rev. 246 (1980); Joan E. Jacoby, The American Prosecutor: A Search for Identity 195-215 (Lexington, Mass.: D. C. Heath & Co., Lexington Books, 1980); Kaplan, John, The Prosecutorial Discretion—A Comment, 60 Nw. U.L. Rev. 174 (1965); LaFave, Wayne R., The Prosecutor's Discretion in the United States, 18 Am. J. Comp. L. 532 (1970); Frank W. Miller, Prosecution: The Decision to Charge a Suspect with a Crime (Boston: Little, Brown & Co., 1970); David W. Neubauer, Criminal Justice in Middle America 113-31 (Morristown, N.J.: General Learning Press, 1974). For an annotated bibliography of additional literature examining the prosecutor's decision to charge a suspect with a crime see W. Randolph Teslik, Prosecutorial Discretion: The Decision to Charge: An Annotated Bibliography (Washington, D.C.: U.S. Department of Justice, Law Enforcement Assistance Administration, National Institute of Law Enforcement and Criminal Justice, 1975).Google Scholar

3 See, e.g., Alschuler, Albert W., The Prosecutor's Role in Plea Bargaining, 36 U. Chi. L. Rev. 50 (1968); LaFave, supra note 2, at 539–47; Neubauer, supra note 2, at 194-223; Donald J. Newman, Conviction: The Determination of Guilt or Innocence Without Trial (Boston: Little, Brown & Co., 1966); Note, Guilty Plea Bargaining: Compromises by Prosecutors to Secure Guilty Pleas, 112 U. Pa. L. Rev. 865 (1964); Rosett, Arthur, The Negotiated Guilty Plea, 374 Annals 70 (1967).Google Scholar

4 Ronald A. Anderson, Wharton's Criminal Law and Procedure 1724 (Rochester, N.Y.: Lawyers Co-operative Publishing Co., 1957).Google Scholar

5 Id. The information entails “a written accusation of crime preferred by a public prosecuting officer without the intervention of a grand jury.”Id. at § 1723.Google Scholar

6 The following list categorizes the states by their possible use of the grand jury to initiate prosecution. Requirements for Grand Jury Indictment to Initiate Prosecutions Grand Jury Indictment Required Optional Grand Jury Indictment Optional Grand Jury Lacks Authority to Indict All Crimes Arizona Pennsylvania New Jersey Arkansas South Carolina California Tennessee Colorado Virginia Idaho Illinois All Felonies Indiana Alabama Iowa Alaska Kansas Delaware Maryland District of Columbia Michigan Georgia Missouri Hawaii Montana Kentucky Nebraska Maine Nevada Mississippi New Mexico New Hampshire North Dakota New York Oklahoma North Carolina Oregon Ohio South Dakota Texas Utah West Virginia Vermont Washington Capital Crimes Only Wisconsin Connecticut Wyoming Florida Louisiana Massachusetts Minnesota Rhode Island SOURCE: Survey and analysis of state laws conducted by Abt Associates. NOTE: Reproduced with permission from Deborah Day Emerson, Grand Jury Reform: A Review of Key Issues 12 fig. 2.1 (Washington, D.C.: U.S. Department of Justice, National Institute of Justice, 1983) (footnotes omitted).Google Scholar

7 Id. Google Scholar

8 Not all states have an adversary preliminary hearing of the type described here. In Gerstein v. Pugh, 420 U.S. 103 (1975), the Supreme Court majority held that the Fourth Amendment does not entitle the defendant to the rights of compulsory process, confrontation, cross-examination, and counsel at a preliminary hearing. Even in states providing for an adversary type of proceeding, where hearsay is the heart of the prosecutor's case, there is no real confrontation of the accuser. In Illinois, e.g., in some jurisdictions the prosecutor's presentation may consist entirely of hearsay. See, e.g., Neubauer, supra note 2, at 131–33.Google Scholar

9 The grand jury's finding of no probable cause does not necessarily terminate efforts by the state to prosecute. A prosecutor may subsequently present the case to the grand jury again or (in dual prosecution jurisdictions) file criminal charges and present the case to a judge at a preliminary hearing. Very little has been written about this form of prosecutorial discretion. For states where the grand jury is often used to initiate prosecutions, there appears to be no data on how often or the reasons why the state reinitiates prosecution after a grand jury no bill. But the potential for abuse has been observed; one author, e.g., suggested that a prosecutor might find it “possible to continue to present a case to succeeding grand juries until he found one which through ignorance or indifference was willing to do his bidding.”Miller, R. Justin, Informations or Indictments in Felony Cases, 8 Minn. L. Rev. 379, 401 n.88 (1924). New Mexico's laws permit prosecution by information after a grand jury no bill. For a discussion of that state's laws and comments about the potential for those laws to result in prosecutorial abuse see Notes and Comments, The Use of an Information Following the Return of a Grand Jury No Bill: State v. Joe Nestor Chavez, 10 N.M.L. Rev. 217 (197980).Google Scholar

10 Very little has been written about the practice of proceeding to the grand jury after preliminary hearing discharge, but in one recent work Jacoby briefly discusses legitimate as well as questionable grounds on which prosecutors may so proceed and suggests further research before judgments about the practice are made. [T]he prosecutor has the power to obtain a grand jury indictment after a case has been dismissed at preliminary hearing for lack of probable cause…, [T]his is an allowable procedure and is used in many instances for legitimate purposes…. When the status of the case changes after arrest or after the preliminary hearing, the use of the grand jury to reinstate the case is justifiable. Controversy results when this authority is used because of differences in opinion, philosophy, or policy between the court and the prosecutor. The exercise of prosecutorial discretion to resolve these differences raises the question of whether the authority of the prosecutor should be used to overturn undesirable judicial findings by “end-running” the case through the grand jury. Since both accusatory routes are legitimately available, the circumstances under which they are used need to be identified before evaluations or judgments are made. Jacoby, supra note 2, at 157–58. The author identifies Boulder, Colorado, as one jurisdiction in which indictments are sought after preliminary hearing discharges. Id. at 266.Google Scholar

11 Most discussions about reinitiating prosecution after preliminary hearing discharge have focused on the practice of holding a second preliminary hearing. For a general review of the pertinent law and practice in Michigan, Wisconsin, and Kansas, see Miller, supra note 2, at 137–41. A lengthy discussion of the practice of refiling charges in Los Angeles and some proposals to limit the prosecutor's right to refile are found in Graham, Kenneth & Letwin, Leon, The Preliminary Hearing in Los Angeles: Some Field Findings and Legal-Policy Observations (pt. 1), 18 UCLA L. Rev. 635, 728–35 (1971). Also on the practice of refiling in Los Angeles, Lynn M. Mather, Plea Bargaining or Trial? The Process of Criminal-Case Disposition 51–52 (Lexington, Mass.: D. C. Heath & Co., Lexington Books, 1979). For other state laws on the prosecutor's right to refile charges after a finding of no probable cause at a preliminary hearing, see Notes and Comments, Criminal Procedure: Prosecutor's Right to Refile a Charge After Dismissal at Preliminary Examination—The Effect of Nicodemus, Jones, and Harper, 25 Okla. L. Rev. 259 (1972) (Oklahoma law); Note, Preliminary Examination Potential, 58 Marq. L. Rev. 159 (1975) (Wisconsin & California law compared).Google Scholar

12 E.g., a frequently quoted articulation of this view of the preliminary hearing is in Thies v. State, 178 Wis. 98, 103, 189 N.W. 539, 541 (1922): The object or purpose of the preliminary investigation is to prevent hasty, malicious, improvident, and oppressive prosecutions, to protect the person charged from open and public accusations of crime, to avoid both for the defendant and the public the expense of a public trial, and to save the defendant from the humiliation and anxiety involved in public prosecution, and to discover whether or not there are substantial grounds upon which a prosecution may be based. Historically, the grand jury also has been thought to act as a check against unfounded prosecutions by standing between the prosecutor and the accused and by determining “whether the charge was founded upon credible testimony or was dictated by malice or personal ill will.” Hale v. Henkel, 201 U.S. 43, 59 (1906).Google Scholar

13 Scigliano, Robert G., The Grand Jury, the Information, and the Judicial Inquiry, 38 Or. L. Rev. 303, 308 (1959). However, depending on the judge and the jurisdiction, the preliminary hearing can function as a rubber stamp of the prosecutor. E.g., in Illinois, testimony given by a detective from police reports that he may not have prepared makes it virtually impossible for the accused to effectively cross-examine the state's only witness. For a description of this practice and the practice in another Illinois town of not permitting the accused the right to cross-examine witnesses at the preliminary hearing, see Neubauer, supra note 2, at 131–33.Google Scholar

14 See, e.g., Antell, Melvin P., The Modern Grand Jury: Benighted Supergovernment, 51 A.B.A. J. 153, 154–55 (1965); Calkins, Richard M., Abolition of the Grand Jury Indictment in Illinois, 1966 U. III. L.F. 423, 431–32; Graham & Letwin, supra note II, at 681; Miller, supra note 9, at 397–99; Neubauer, supra note 2, at 134; Note, Should the Grand Jury Indictment Procedure Be Abolished in Illinois? 2 J. Mar. J. Prac. & Proc. 348, 354–60 (1969)Google Scholar

15 The low “no bill” vote is sometimes cited as proof of the grand jury's rubber stamp of the prosecutor's wishes. See, e.g., Neubauer, supra note 2, at 134. Neubauer does note (at 131) that prosecutors' review of cases before charges are filed may itself “eclipse” the screening function of the grand jury. See also, on how concurrence of grand jury and prosecutor may demonstrate only that a large majority of cases coming before the grand jury are “open and shut” cases on which disagreement would be unlikely, Dession, George H., From Indictment to Information—Implications of the Shift,42 Yale L. J. 163, 178–79 (1932).Google Scholar

16 This characterization of the grand jury as merely rubber-stamping the wishes of the prosecutor appeared as early as 1929 in an Illinois statewide survey: Every prosecutor knows, and every intelligent person who ever served on a grand jury knows, the prosecuting officer almost invariably completely dominates the grand jury…. The grand jury usually degenerates into a rubber stamp wielded by the prosecuting officer according to the dictates of his own sense of propriety and justice. (At 299) [T]he prosecution in Illinois is unduly handicapped by the constitutional requirement of an indictment by the grand jury. The innocent citizen need not fear unfounded prosecution by information. If the state's attorney wished to prosecute him, he could easily obtain an indictment from a grand jury which he dominates. (At 218) Illinois Association for Criminal Justice, The Illinois Crime Survey (Chicago: Illinois Association for Criminal Justice in cooperation with Chicago Crime Commission, 1929). See also Calkins, supra note 14, at 429, 431; Morse, Wayne L., A Survey of the Grand Jury System (pt. 2, 2d installment), 10 Or. L. Rev. 295, 329(1931).Google Scholar

17 Coates, Walton, The Grand Jury, The Prosecutor's Puppet: Wasteful Nonsense of Criminal Jurisprudence, 33 Pa. B.A.Q. 311 (1962).Google Scholar

18 In Los Angeles, where the preliminary hearing is used to reinitiate prosecutions, Graham & Letwin have similarly expressed a concern that prosecutors may reinitiate prosecutions by simply ignoring the ruling of one preliminary hearing judge and proceeding with the same evidence before another more sympathetic judge for a probable cause determination. Graham & Letwin, supra note 11, at 734. The prosecutor's absolute power to refile should be substantially limited. He should be barred from refiling (or going the alternate grand jury route) where he has no better justification than the hope that a second magistrate will look more kindly on the same evidence, or that, somehow, better evidence can be mustered the second time around. To permit relitigation for these reasons is to reward casual preparation, to create uncertainty, to permit harassment, and to waste judicial resources. Id. at 730–31 (footnote omitted). In Illinois, concerns about prosecutors' going on to the grand jury with the same evidence resulted in new legislation requiring the prosecutor to inform the grand jury of a previous finding of no probable cause in a case. In People v. Kent, 54 III. 2d 161, 295 N.E. 2d 710 (1972), the Illinois Supreme Court rejected the contention that a finding of no probable cause at a preliminary hearing precluded the prosecution's proceeding to the grand jury for an indictment. The Chicago Bar Association (CBA) saw this holding as perpetuating a potential source of prosecutorial abuse: “At present, upon finding no probable cause, the State oftentimes marches the same witnesses before the grand jury, repeats the testimony adduced at the preliminary hearing, absent what was elicited on cross-examination, and an indictment is returned.” Letter from Michael P. Toomin, CBA Criminal Law Committee, to Terry Murphy, Secretary, CBA Defense of Prisoners Committee, Re: Legislative Proposals Relative to Preliminary Hearings and Grand Jury (Feb. 1, 1979), at 5. The CBA proposed legislation to ensure that grand jurors were apprised of the prosecution's responsibility to produce additional or more persuasive evidence to show probable cause. After much negotiation between various sectors of the bar, and a veto by the governor, the legislature enacted the current law (effective Jan. 1, 1980): In cases where the initial charge has been commenced by information or complaint and a finding of no probable cause has resulted as to any offense charged therein, the Grand Jury shall be informed of the finding entered at the preliminary hearing and further advised that such finding shall not bar the State from initiating new charges by indictment, information or complaint if the State's Attorney has reasonable grounds to believe that the evidence available at that time is sufficient to establish probable cause. III. Rev. Stat. ch. 38, § 112–(b) (1981). Although that law was not in effect during the period of this study, this research could supply part of the data for studying the impact of the law.Google Scholar

19 Such a consideration, e.g., is reported to affect the initial charging decision. Cole, supra note 2, at 341. See also Comment, supra note 2, at 531. In this survey of Los Angeles County deputy district attorneys, two-thirds said that public opinion affected the decision whether to initially issue a criminal complaint. Important to note, 60% believed that public opinion should have an effect. Id. at 531.Google Scholar

20 This factor is also reported to affect the initial decision to charge. Cf. Kaplan, supra note 2, at 181–82 (reporting that in some cases prosecuted by the US. attorney's office the initial decision to charge was shaped by the prosecutor's belief that the accused was guilty of previous offenses and likely to engage in criminal activity in the future). According to a survey of deputy district attorneys in the Los Angeles County District Attorney's main office, a suspect's prior convictions did weigh into their decision to initially file a criminal complaint: 27% reported giving them “great weight,” and 47% reported giving them “some weight.” Comment, supra note 2, at 529.Google Scholar

21 See generally on the adverse byproducts of the pretrial process Malcolm M. Feeley, The Process Is the Punishment: Handling Cases in a Lower Criminal Court 199–243.278 (New York: Russell Sage Foundation, 1979). Although Feeley's study is of the misdemeanor trial courts, the punishments he outlines are not irrelevant in a felony court.Google Scholar

22 Arenella, Peter, Reforming the Federal Grand Jury and the State Preliminary Hearing to Prevent Conviction Without Adjudication, 78 Mich. L. Rev. 463, 505–6 (1980).Google Scholar

23 This is not implausible given various descriptions of prosecutors' plea bargaining practices. One national study has suggested that prosecutors at early stages of the process may fail to terminate prosecution where it is questionable whether the state could actually prove its case. Some prosecutors see nothing improper in convicting a person they believe is factually guilty even though the state may be unable with its evidence to secure a conviction; others view the legal guilt requirements as not applying to pleas of guilty—presumably a defendant willing to plead guilty agrees in effect to forgo the trial's formal requirements of proof; still others argue that even the weakest case may result in a jury conviction. Herbert s. Miller, William F. McDonald, & James A. Cramer, Plea Bargaining in the United States 101–2, 107–9 (Washington, D.C.: US. Department of Justice, Law Enforcement Assistance Administration, National Institute of Law Enforcement and Criminal Justice, Sept. 1978). The philosophy that half a loaf is better than none also may spur on the practice of bargaining in cases where the evidence is unlikely to sustain a conviction. Alschuler, supra note 3, at 59–60. Generally speaking, given prosecutors' keen interest in a conviction, they not only may offer concessions to procure a guilty plea from a defendant not otherwise likely to be convicted, but may make a weak case the object of intensified plea bargaining. As one author suggests, “prosecutors may find themselves offering the greatest incentives to those defendants with the greatest chance for acquittal at trial … The ironic result in these circumstances is that those most likely to benefit from trial are those the system works hardest to keep from trial.”Vorenberg, James, Decent Restraint of Prosecutorial Power, 94 Harv. L. Rev. 1521, 1535 (1981). To encourage pleas of guilty in cases where ability to convict is doubtful, the prosecutor may offer charge reductions and sentences so attractive that they far outweigh any advantages of going to trial. Such intensified plea negotiation can result in the conviction of factually and legally innocent defendants. Alschuler, supra note 3, at 60–63, 81; Mitchell, John B., The Ethics of the Criminal Defense Attorney—New Answers to Old Questions, 32 Stan L. Rev. 293, 308–9 (1980).Google Scholar

24 Various other reasons might also account for the indictment of cases with insufficient evidence to warrant prosecution. E.g., the grand jury can be a less thorough screen for cases than the preliminary hearing because of the state's more extensive use of hearsay evidence at the grand jury. The prosecutor, e.g., rather than present witnesses expected to testify at trial, may present the state's evidence to the grand jury entirely through the hearsay testimony of the investigating law enforcement officer. With the freedom to use hearsay extensively at the grand jury the prosecutor may have less incentive to fully investigate his case before the grand jury proceeding to know whether witnesses or material evidence he expects to rely on at trial will indeed be available. Particularly with a strong interest in securing an immediate indictment (such as wanting to prevent the defendant from leaving custody and disappearing after a preliminary hearing discharge), the prosecutor may permit the police officer in the case to testify as to witnesses or evidence the state expects will materialize before trial. Once having obtained an indictment, however, the state may be unable to locate the witnesses on whose statements it based its presentation to the grand jury. This is particularly a potential problem where the failure of the state's witnesses to appear at the preliminary hearing was the reason for the defendant's discharge, for that nonappearance may predict their unwillingness to cooperate with the state later.Google Scholar

25 Murder cases were studied because data for this offense were collected as part of another study in progress by the author. For a discussion of the data collection for both studies see pt. II and especially note 31 infra. Google Scholar

26 See note 68 infra. Pertinent questions beyond the limits of this study include these: How does the type of offense (apart from other homicides discussed in pt. VI) influence the prosecutor's decision to reinitiate prosecution after a preliminary hearing discharge? In other jurisdictions or in other preliminary hearing courts, are prosecutors more likely to bypass the preliminary hearing if the preliminary hearing judge allows wide latitude to the defense attorney, e.g., allows extensive cross-examination of the state's witnesses? Do practical constraints in a jurisdiction, such as the frequency with which the grand jury sits to hear cases, limit the exercise of discretion to bypass the preliminary hearing? And does the attitude of the prosecutor's office toward a particular preliminary hearing judge affect the decision to bypass the preliminary hearing or to reinitiate prosecution after a preliminary hearing discharge?Google Scholar

27 See pt. III.B.1 infra Google Scholar

28 See pt. VI. A infra. Google Scholar

29 See note 68 infra. Google Scholar

30 The data collected for the cases included but were not limited to information about the type of disposition at the preliminary hearing stage (finding of probable cause, discharge, dismissal), the type of disposition at the trial court stage (dismissal, acquittal, conviction), the method of disposition in the trial court (plea, trial), and the type of trial (bench, jury).Google Scholar

31 A “case” in this sample was defined as any complaint(s), grand jury indictment, or prosecutor's bill of information, involving one incident, filed against one person. I gathered the basic data for this research as part of a study in progress about the representation of indigent defendants. That study examines the implications of a shift in the Office of the Public Defender of Cook County, Illinois, from horizontal representation (the sequential assignment of different lawyers to represent a defendant at each stage of the case) to vertical representation (the assignment of one lawyer to represent a defendant from beginning to disposition), a shift that took place in 1974 for homicide cases. To study the implications of that change, I collected data on 3,997 homicide cases entering the court system in the 8 years 1971–78. The study reported in this article utilizes those data from the latter 3 &frac;14 years of that period. The original study collected data for all homicide cases that received final disposition at the preliminary hearing stage—that is, were dismissed or discharged and never indicted—whether representation was by public or by private attorneys. For cases that reached the trial court stage, the original study collected data for only certain cases: (1) those defendants represented vertically and horizontally by public defenders, (2) those defendants represented by different private attorneys at the preliminary hearing and trial court stages (horizontal private attorney representation), and (3) those defendants represented by the same private attorney throughout their prosecution (vertical private attorney representation). For the first two types of representation data were collected for all cases in the system. For vertical private attorney representation, because of the large number of defendants so represented, data were collected for a 50% random sample of cases for each of the eight years. For the present study I collected the additional case file data needed to adapt the original study to the new research focus. For cases dealing with crimes occurring on or after October 1, 1975, data were collected for all cases in which defendants were (1) represented by a public defender at preliminary hearing and by a private attorney at trial, (2) represented by a private attorney at preliminary hearing and by a public defender at trial, (3) discharged at preliminary hearing and subsequently indicted, and (4) initially prosecuted by grand jury instead of by preliminary hearing. This additional data collection resulted in a sample of 1,712 murder cases and 94 manslaughter and reckless homicide cases, for a total of 1,806 homicide cases in the study. This sample is about 88% of all homicide cases in which prosecution was initiated from Oct. 1, 1975 through Dec. 31, 1978. The homicide cases not included in the study were those not selected as part of the 50% sample of vertically represented private attorney cases and 60 cases in which the state obtained an indictment prior to an accused's arrest (see figs. 1 & 2 infra). Some juvenile homicide cases may have been inadvertently excluded from the study and not included in figs. 1 and 2—those cases not transferred from juvenile court to Branch Court 66, which heard preliminary hearings in all adult homicide cases. Figs. 1 and 2 and the statistical tables in the study display estimated frequencies based in part on multiplying by 2 the 50% random sample of vertically represented private attorney cases. It should be noted, however, that the calculation of the chi square statistic in tables 2–8 uses the unweighted 50% random sample. So that the reader may have the basic numbers used in calculating the chi square statistic, I have provided for each of tables 2-8 a general note on the unweighted numbers in each cell.Google Scholar

32 The files were stored in the First Municipal District's file vault in the Daley Center in downtown Chicago and in the County Criminal Division's file vault in the courthouse at 26th Street and California Avenue. The Municipal District's file vault contained preliminary hearing judges' court sheets. The court sheet is a daily record of what happened to cases in court on a given day, e.g., whether a case was continued, if there was a preliminary hearing or bail hearing, if defense counsel was appointed. I used these sheets for Branch Court 66 to identify cases dismissed without a preliminary hearing, cases discharged after a finding of no probable cause at a preliminary hearing, and cases for which grand jury was used instead of preliminary hearing. Also kept here were the criminal complaints filed against a defendant and branch court information for all cases receiving a final disposition at the branch court (i.e., cases dismissed or discharged at preliminary hearing and never indicted). From these materials I recorded for each case in the study such basic information as type of final branch court disposition, first and last court dates, and type of defense counsel. The Criminal Division file vault contained court case files for all cases reaching the trial court through a grand jury indictment or through a prosecutor's bill of information from a preliminary hearing. To identify the homicide cases in the vault I used microfiche sheets in the clerk's office, which listed each indictment or information number in a given year, the defendant's name, and the offense(s) charged. Court cases were arranged in the vault by indictment/information numbers. To make certain that the cases discharged at preliminary hearing but subsequently indicted were not counted also as having received a final disposition at preliminary hearing, the two sets of data (preliminary hearing and trial) were compared for duplicate names.Google Scholar

33 See text accompanying notes 4–50 infra. Google Scholar

34 See note 31 supra. Google Scholar

35 October 1, 1975, was the effective date for new legislation in Illinois that eliminated the prosecutor's being required after a finding of probable cause at a preliminary hearing to present the case to a grand jury for indictment. Under the new law, after a finding of probable cause at preliminary hearing a prosecutor could initiate formal prosecution proceedings by simply filing an information. III. Rev. Stat. ch. 38, § 111–2(1981). Thinking that this new legislation could increase the time the grand jury would have to consider cases (because fewer cases would come before it), which could affect the rate of grand jury no bills and true bills in cases for which prosecution was reinitiated after a preliminary hearing discharge, I decided to include in this study only cases charging crimes occurring on or after the effective date of the legislation. See generally on this legislation Kavanaugh, James R. & Jesser, Steven H., The Grand Jury Bypass, 57 Chi. B. Rec. 282 (1976).Google Scholar

36 See in table 9 infra the state's reasons for not further pursuing prosecution for these 57 cases. For a discussion of how this sample was obtained see note 162 infra. Google Scholar

37 See in table 10 infra the state's reasons for pursuing prosecution for these 134 cases. For a discussion of how this sample was obtained see note 170 infra. Google Scholar

38 See pt. VI.D infra for a discussion of the limitations and advantages of using these internal memoranda for such information.Google Scholar

39 See notes 163 & 170 infra. Google Scholar

40 See table 13 infra and the discussion in pts. VI.D.4 and VI.E about this higher rate of dismissal in reinitiated cases. As indicated in table 3 infra, cases discharged at preliminary hearing but subsequently indicted were also more likely to result in a higher rate of acquittal than cases prosecuted after a finding of probable cause at preliminary hearing. Unfortunately, the files did not contain enough information to study what accounted for the difference. The files of only 5 of 43 reinitiated cases contained a form filled out by the trial court assistant, “Evaluation of the Not Guilty Verdict,” which contained a checklist of reasons for the not guilty verdict, e.g., “testimonial problems,”“witness changes anticipated testimony,”“witness impeached,”“all available evidence not presented,” and a category “any other reason,” in which the assistant recorded any special or additional reasons.Google Scholar

41 See table 13 and note 200 infra. Google Scholar

42 The “Nolle Pross Form” requested the following information from the trial court assistant: Name of defendant(s) Reasons for nolle pross? Defense lawer(s) Victim's or victim's family response to nolle pross? Indictment number Assistant who put case through the grand jury or preliminary hearing? Charge(s) What could have been done to save this case? Term date Attach all investigator's request slips and investigation reports. Nolle date Make out form in duplicate and attach copy to blue back. Statement of facts Attach I.R. sheetGoogle Scholar

43 E.g., the following is a portion of the daily court summary in one dismissed case: 3–31–78 Investigators out to locate subpoena [witness] and we need paper detailing investigation. 5–23–78 Per [investigator] cannot locate needed witness. See Inv's report. Ok to nolle [assistant's superior]. 6–6–78 m/s [motion state] nolleGoogle Scholar

44 Although this study collected data about all homicide cases prosecuted in this court system during the period oct. 1, 1975-Dec. 31, 1978, 1 have confined my description of the case flow process in pt. III to murder uses only because they make up 95% of all homicides prosecuted and because some of the case flow features of murder cases are different from those of the other homicides. In pt.VI.C, I present (in fig. 2) case flow data for voluntary and involuntary manslaughter and reckless homicide cases and discuss differences in the rate of reinitiation between murder and other homicide cases. The 1,895 cases in fig. 1 in which criminal complaints were filed may include a few cases for which the prosecutor approved a preliminary hearing complaint and an arrest warrant application commanding the arrest of someone not in custody. This procedure was sometimes used in cases of “known-but-flown” offenders where one suspect in a case was in custody and another at large. At the time of examining court files I was not aware that complaints were occasionally issued prior to the arrest of a suspect. I therefore did not distinguish between cases having a complaint issued after a suspect's arrest and cases having a complaint issued before a suspect's arrest. In both situations the effect is similar: a case is docketed in branch court for a preliminary hearing.Google Scholar

45 In cases initially prosecuted by grand jury, there is the question of how frequently the grand jury returns a no bill (a finding of no probable cause) and how frequently the state reinitiates prosecution in these no bill cases by either presenting the case again to the grand jury or by proceeding to a preliminary hearing. That question did not arise in this study, for all murder cases presented directly to the grand jury for indictment resulted in a true bill.Google Scholar

46 For a discussion of the study's sample of 1,712 murder cases and 56 prearrest indictments on which this figure is based, see note 31 supra. Google Scholar

47 The 56 cases in fig. 1 may include not only those defendants arrested after a grand jury indictment, but also possibly some cases where a suspect was arrested and then indicted the next morning without a preliminary hearing complaint ever being filed. At the time I examined court files for this study, I was not aware of the latter practice by the prosecutor's office and therefore did not distinguish between these two ways in which an indictment may occur without a criminal complaint ever having been filed.Google Scholar

48 In Chicago, Felony Review began in 1972, but the prosecutor's role in deciding whether to file criminal charges in homicide cases predates Felony Review. It is believed that at least as early as 1965 homicide charges could not be filed without prosecutorial approval. For other offenses, however, before 1972 the police alone decided after an arrest whether to file criminal charges. For an early description and evaluation of the Felony Review Project, see Donald M. McIntyre & Raymond T. Nimmer, Survey and Evaluation of Illinois State's Attorneys Association Comprehensive Project: Second Year, Illinois Law Enforcement Commission Grant, at 6–21 (photocopied typescript; Hinsdale: Illinois State's Attorneys Association, [1973?]). For descriptions of the old system in Chicago of allowing the police to decide whether to prosecute, see McIntyre, Donald M., A Study of Judicial Dominance of the Charging Process, 59 J. Crim. L.C. & P.S. 463 (1968); Dallin H. Oaks & Warren Lehman, A Criminal Justice System and the Indigent: A Study of Chicago and Cook County 30 (Chicago: University of Chicago Press, 1968).Google Scholar

49 The Cook County state's attorney's office entered into an agreement with the police department that Felony Review (the prosecutor's office precharge screening unit) has the final word on whether to file charges in homicide cases. For other types of cases, if the police disagree with the prosecutor's decision not to file charges, they can go to the deputy superintendent of the police department and he can authorize filing a complaint.Google Scholar

50 In Illinois, a preliminary hearing is constitutionally required whenever prosecution of an individual will be by an information rather than by grand jury indictment. See People v. Audi, 29 111. Dec. 691, 73 Ill. App.3d 568, 392 N.E.2d 248 (1979).Google Scholar

51 According to the judge assigned to Branch Court 66 during the period of this study, dismissals at this stage normally were not initiated by him but were granted after the state's motion for dismissal.Google Scholar

52 On the high rate of dismissal of felony cases in preliminary hearing court in the time before initiation of prosecutors' precharge screening in Chicago, see McIntyre, supra note 48, at 476–79.Google Scholar

53 Although in rare cases prosecutors may decide to proceed to the grand jury for indictment after a finding of probable cause at preliminary hearing, in Illinois this is no longer required. See note 35 supra. In some cases the state is concerned that a witness may change his testimony at trial. To discourage this and presumably to have a basis for a perjury prosecution if it does occur, the state will use both proceedings to memorialize his statement.Google Scholar

54 This is in marked contrast to a custom of the prosecutor's office in the late 1960s of proceeding to the grand jury in lieu of the preliminary hearing in virtually all homicide cases. Theis, William H., Preliminary Hearings in Homicide Cases: A Hearing Delayed Is a Hearing Denied, 62 J. Crim. L.C. & P.S. 17, 19(1972). Theis's research predated the 1970 Illinois Constitution's provision for a prompt preliminary hearing to establish probable cause unless the initial charge was brought by grand jury indictment (art. 1, § 7). For several years after the constitution's effective date, July 1, 1971, prosecutors in Cook County rarely sought a direct indictment after a preliminary hearing complaint had been filed. In 1973 the Illinois Supreme Court construed this constitutional provision in People v. Hendrix, 54 Ill. 2d 165, 295 N.E.2d 724, ruling that when a suspect was charged by complaint, a prompt indictment was as good as a prompt preliminary hearing to initiate prosecution. Subsequently there emerged the current practice of using the grand jury in some cases in lieu of the preliminary hearing. These observations on the changing practice were provided by Northwestern University Professor James B. Haddad, former first assistant state's attorney during State's Attorney Bernard Carey's administration, 1972–1980.Google Scholar

55 For those who might undertake research in this jurisdiction in the future, it is worth mentioning that when a case is taken directly to the grand jury, a “direct report” is prepared by a prosecutor, which gives the reasons for bypassing the preliminary hearing. This is no more than a statement of one or two sentences written by the assistant state's attorney on an office form. How regularly these are filled out is not known, but one supervisor interviewed believed they were not prepared regularly by all assistants.Google Scholar

56 No information was available on how frequently the supervisor disagreed with his assistants' recommendations.Google Scholar

57 Ill. Rev. Stat. ch. 38, §1124(d) (1981).Google Scholar

58 In Illinois, a defendant is not entitled to a preliminary hearing, only to a determination of probable cause, which may be made either at a preliminary hearing or by a grand jury. People v. Arbogast, 41 Ill. App. 3d 187, 353 N.E.2d 434 (1976). In recent years, prosecutors' discretionary choice to proceed by either information or indictment has received increasing attention because of the perceived denial to indicted defendants of advantages inhering in prosecution by information after a preliminary hearing. In California, e.g., the perceived disparity between the procedural rights afforded those prosecuted by information (the right to be present, the right to confront and cross-examine witnesses, present evidence, etc.) and the rights afforded those prosecuted by indictment resulted in the state supreme court's decision that denial of the preliminary hearing procedure violated the state constitution's equal protection clause. The court concluded that the indictment procedure could continue but that indicted defendants must have the right to demand a postindictment preliminary hearing. Hawkins v. Superior Ct., 22Cal. 3d 584, 150 Cal. Rptr.435, 586 P.2d 916 (1978). For a discussion of the effects of Hawkins on the prosecutor's use of the grand jury, see Seth Rosenfeld, Justice and the Grand Jury, Cal. Law., Feb. 1984, at 32. Since Hawkins, in several Illinois cases the courts have considered the issue whether a defendant initially charged by criminal complaint but indicted by a grand jury is still entitled to a preliminary hearing. In each case the court has declined to follow Hawkins and has rejected the defendant's argument that the prosecutor's unlimited discretion to choose the forum for initiating prosecution results in a denial of equal protection to those denied a preliminary hearing. People v. Franklin, 35 Ill. Dec. 121, 80 Ill. App. 3d 128, 398 N.E.2d 1071 (1979); People v. Garza, 48 Ill. Dec. 44, 92 Ill. App. 3d 723, 415 N.E.2d 1328 (1981).Google Scholar

59 Ill. Rev. Stat. ch. 38, § 111-2(a) (1981) (hearing is required unless defendant waives it).Google Scholar

60 Ill. Rev. Stat. ch. 38, § 109-1(b)(2) (1981).Google Scholar

61 “All preliminary hearings on felony cases are to be held within 30 days of the arrest of an accused. Preliminary hearings may be continued beyond the 30 day period only upon a showing of exceptional circumstances which warrant delay.” Rules of the Circuit Court of Cook County, Adopted May 17, 1976, in Illinois Courts Rule Book, looseleaf (3d ed. Chicago: Law Bulletin Publishing Co., July 1, 1983).Google Scholar

62 The data in this study indicate that only 13% of the murder cases had a preliminary hearing on the first day the case was in Branch Court 66. The mean number of days murder cases were in Branch Court 66 before a preliminary hearing was held was 20, and the median was 16.Google Scholar

63 Occasionally in murder cases the accused will waive preliminary hearing because he has decided to plead guilty at that stage. E.g., in this study, in 3 of the 33 murder cases in which the preliminary hearing was waived, the defendant pled guilty at the preliminary hearing stage. Because there were only 3 guilty pleas at this stage, it appears that in every case pled guilty, the hearing was waived. The circumstances surrounding the waiver in the remaining 30 cases is unknown. In homicides other than murder the preliminary hearing was waived in about 10% of the cases (see fig. 2 infra.) In each of those 10cases the defendant subsequently pled guilty at the branch court level. There was a total of 12 pleas of guilty at this stage, but information was missing for 2 cases on whether the hearing was conducted or waived. For a discussion of a similar low rate of preliminary hearing waiver in Los Angeles and a thoughtful exploration of why variations may exist, see Graham & Letwin, supra note 11, at 646–48, 646 n.35. In a recent study of the grand jury in two Arizona jurisdictions, the authors found that in one, Maricopa County (Phoenix), the preliminary hearing is frequently waived. Maricopa is like Cook County (Chicago) in that the preliminary hearing is the preferred forum of prosecution. However, of the 9,400 cases in Maricopa in which a preliminary hearing was set, 33% of the accuseds waived a hearing; in only 13% was a hearing held; in 9% the prosecutor proceeded to the grand jury in lieu of the preliminary hearing. In the remaining 45% the prosecutor dismissed the charges, deferred prosecution, or negotiated a plea to a misdemeanor. Deborah D. Emerson & Nancy Ames, The Role of the Grand Jury 84 (draft; Cambridge, Mass.: Abt Associates, 1982). The authors attribute the high rate of waiver in Maricopa County to several factors. First, the defense bar does not generally consider the hearing a major discovery vehicle, or an important forum for moving to suppress illegally obtained evidence or for raising affirmative defenses. Moreover, the defense may find several advantages in waiving the hearing: additional or early discovery, release of the accused on his recognizance, and a more favorable deal in plea negotiations. Id. at 95. In Chicago, however, the preliminary hearing is viewed quite differently by defense lawyers. For a discussion of their use of the preliminary hearing to “lock” prosecution witnesses into a statement, and to discover the prosecution's case, see Janet Ann Gilboy, Perspectives and Practices of Defense Lawyers in Criminal Cases 79, 84-86 (Ph.D. diss., Northwestern University, 1976).Google Scholar

64 The preliminary hearing in Illinois is considered a critical stage in the prosecution process for which an accused individual has the right to counsel. Phegley v. Greer, 497 F. Supp. 519, 521 (C.D. Ill. 1980).Google Scholar

65 This task force and its members have been the subject of numerous popular articles: see, e.g., Paul Galloway, Winner for the Losers, Chi. Sun-Times Midwest Mag., June 5, 1977, at 8; Robert McClory, Murder Is Their Business, Student Law., Sept. 1979, at 13; Lauri Meyers, Dial Them for Murder, Reader, July 1, 1977, at 1.Google Scholar

66 This vertical representation was introduced in 1974 and supplants the horizontal mode for homicide cases. The impact of this shift in type of representation is the focus of another in progress study by the author (see note 31 supra.) For an earlier study describing horizontal representation in the Cook County public defender's office, see Gilboy, Janet A. & Schmidt, John R., Replacing Lawyers: A Case Study of the Sequential Representation of Criminal Defendants, 70 J. Crim. L. & Criminology 1, 5–7 (1979). For a more general discussion of the role of court organization in shaping defense services as a succession of different lawyers at each court stage, see Gilboy, Janet A., The Social Organization of Legal Services to Indigent Defendants, 1981 A.B.F. Res. J. 1023.Google Scholar

67 There have been very few published materials discussing vertical and horizontal prosecution. For the few I am aware of, see Graham & Letwin, supra note 11, at 645; id., (pt. 2), 916, 924–25; Brian A. Grosman, The Prosecutor: An Inquiry into the Exercise of Discretion 26–27 (Toronto: University of Toronto Press, 1969); Jacoby, supra note 2, at 248–49; Taylor, Robert E., Street-Crime Drive: D.A.'s Special Teams Win stiffer Sentences for “Career Criminals,” Wall St. J., May 11, 1982, at 1, col. 1.; U.S. Department of Justice, National Institute of Justice, Exemplary Projects 14–15 (Washington, D.C.: U.S. Government Printing Office, 1981); Weimer, David L., Vertical Prosecution and Career Criminal Bureaus: How Many and Who? 8 J. Crim. Just. 369 (1980). In passing, it is useful to note that the implications of horizontal prosecution for the decision to reinitiate prosecution after a preliminary hearing discharge are unclear. In horizontal prosecution, because the assistant state's attorney who recommends an indictment will not prosecute the case at the trial court level, he does not learn from handling cases about the likely results of decisions to reinitiate, nor is there written feedback about the case from the trial court assistant. However, preliminary hearing attorneys do learn about the outcomes of some of the cases they have recommended for grand jury indictment, for apparently everyone in the office knows when a case is on trial. Also they may learn through their supervisor of trial lawyers' complaints about cases further prosecuted after discharge. (See text accompanying note 151 Infra.) That “review” of their work may encourage them to consider the likely consequences of their decisions affecting stages of work they will not be handling themselves. It should be mentioned that typically in horizontal prosecution the preliminary hearing assistant has never prosecuted a case in trial court, so he lacks that experience to inform decisions. This is because the prosecutor's job is often his first employment, and the normal progression in the prosecutor's office is from preliminary hearing court to trial court. On the other hand, the office assigns a lawyer with considerable trial experience to be a supervisor in Branch Court 66. As supervisor, that experienced lawyer conveys to assistant prosecutors information about what is likely to happen to a case at the trial court level, as well as formally reviews and approves their recommendation (see pt. VI.A infra.)Google Scholar

68 During Oct. 1975-Dec. 1978 Maurice D. Pompey was regularly assigned to Branch Court 66. Other judges occasionally handled that court's call, but 87% of the homicide cases in this study were disposed when Judge Pompey sat.Google Scholar

69 For a description of the operation of other preliminary hearing courts in Chicago, see James Eisenstein & Herbert Jacob, Felony Justice: An Organizational Analysis of Criminal Courts 208 (Boston: Little, Brown & Co., 1977).Google Scholar

70 See Neubauer, supra note 2, at 131–33 (medium-sized Illinois industrial town: detective testifies about police reports he may or may not have prepared); Jacoby, supra note 2, at 225 (New Orleans: credible hearsay is admissible and the prosecutor usually calls only one witness, the arresting police officer). An Illinois circuit court judge has observed that preliminary hearing procedures vary widely not only between counties but also between judges within a county, differences he explained as the outgrowth of long-established practices of judges who have simply always done it one way or another. Steigmann, Judge Robert J., The Preliminary Hearing in Illinois, 66 Ill. B.J. 700, 700 (1978). Probably because of concerns about such variations, the Illinois Supreme Court appointed a committee of judges to study rules of evidence at preliminary hearings, considered the committee's findings (Report of the Study Committee on Rules of Evidence at Preliminary Hearings (Sept. 1982)), and decided to make no change in practice and procedure at that time.Google Scholar

71 In Illinois due process does not require that hearsay evidence be excluded at the preliminary hearing to establish probable cause. People v. Blackman, 46 Ill. Dec. 524, 91 Ill. App. 3d 130, 414 N.E.2d 246 (1980); People v. Velez, 72 111. App. 2d 324, 219 N.E.2d 675 (1966).Google Scholar

72 But the judge placed limitations on hearsay testimony even on the cause of death, e.g., would not permit a police officer to testify on the cause of death from information received from a clerk in the medical examiner's office (see text at note 181 infra.)Google Scholar

73 The judge described, e.g., why his views differed from that of some other judges: [Some judges believe] that if a police officer comes in and testifies that Jane Jones said that John Smith committed rape, that if the judge believes the police officer, that this is sufficient for probable cause. I cannot buy that, I want to hear from the witness's lips what the police officer is telling. I've heard 3,000 murder case5 and 2,000 rapes. I want to look at the witnesses and observe them testifying like at trial and decide whether to give weight or credibility to that witness's testimony. [Some judges believe] the state's attorney would never prosecute someone unless there were sufficient evidence. I don't start with that premise. Just because the police have arrested someone and the prosecutor has decided to file charges doesn't mean that there is sufficient evidence. That's a presumption of guilt. A person coming into preliminary hearing court, they're still innocent until proven guilty.Google Scholar

74 In some cases, however, the state attempts to introduce the evidence of its witnesses through the hearsay testimony of the investigating police officer in the case. E.g., see in pt. VI.D.2 infra the discussion of practices of the state in cases where its witnesses are not at the preliminary hearing for tactical or other reasons.Google Scholar

75 The same strategy has been observed in other jurisdictions as well. Miller, supra note 2, at 67 & n. 11, 140. In Los Angeles, however, it has been the custom of prosecutors to present much more of their evidence. This was thought to be due to the practice in that jurisdiction of submitting cases to a trial judge for adjudication based on the transcript of the preliminary hearing. Graham & Letwin, supra note 11, at 658. It is unknown whether the practice of presenting more evidence than is necessary to obtain a probable cause finding still persists in light of the virtual disappearance since 1974 of the “submission on the transcript” (S.O.T.) procedure. See Mather, supra note 11, at 20–21, for a discussion of the reasons for the declining use of S.O.T.Google Scholar

76 Convenience and limiting discovery are also prime reasons in other jurisdictions for holding back on evidence at preliminary hearing. Miller, supra note 2, at 67. See also the account by one supervisor of how this strategy may result in preliminary hearing discharges even though the state may actually have at the time of the preliminary hearing evidence sufficient to warrant further prosecution. See text following note 156 infra. Google Scholar

77 In the jurisdiction studied, a court stenographer records the preliminary hearing, and a transcript is available to the defense and prosecution for use at trial court. The defendant's attorney can use the transcript to “freeze” the testimony of a witness in order later to draw out inconsistencies between statements made at preliminary hearing and those made at trial. Illinois law, however, does not require that the court provide either a court reporter or a recording device for the preliminary hearing. People v. Hanson, 3 Ill. Dec. 778, 44 Ill. App. 3d 977, 359 N.E.2d 188 (1977); People v. Lewis, 37 Ill. App. 3d 870,346 N.E. 2d 377 (1976). Whether prosecutors present more of their evidence in jurisdictions where no recording of the preliminary hearing is prepared is unknown. In those jurisdictions, bystanders or other witnesses at the preliminary hearing possibly could be called to testify at trial about statements a witness made at the preliminary hearing. This also might encourage the state to limit its evidence at the preliminary hearing.Google Scholar

78 In Los Angeles the average preliminary hearing takes somewhat more than 30 minutes. Variations exist, however, between branches within Los Angeles County and between other counties in California. Graham & Let win, supra note 11, at 659.Google Scholar

79 Accused persons are not required to present their defense at the preliminary hearing. People v. Patterson, 3 Ill. Dec. 479, 44 Ill. App. 3d 894,358 N.E.2d 1164 (1976). It is not clear whether accused persons have the right to present evidence on their own behalf at the hearing. For one Illinois Circuit Court judge's opinion that the defendant has no constitutional or statutory right to present evidence, see Steigmann, supra note 70, at 703–5. In other jurisdictions in the country as well, accused persons rarely offer evidence of their own. Miller, supra note 2, at 65, 76; Graham & Letwin, supra note 11, at 658; id., (pt. 2), 916,923.Google Scholar

80 The general rule in Illinois regarding cross-examination at the preliminary hearing is “that it may not extend beyond the scope of direct examination and such further interrogation as is directed to show interest, bias, prejudice or motive of the witness to the extent that these factors are relevant to the question of probable cause.” People v. Horton, 3 Ill. Dec. 436, 438, 65 111. 2d 413, 358 N.E.2d 1121, 1123 (1976).Google Scholar

81 The legal literature contains numerous references to these trial preparation functions of the preliminary hearing. See, e.g., Jonathan Casper, American Criminal Justice: The Defendant's Perspective 52 (Englewood Cliffs, N.J.: Prentice-Hall, 1972); Graham & Letwin, supra note 11, pt. 2), 916, 920–31.Google Scholar

82 Graham & Letwin, supra note 11, (pt. 2), 916, 925.Google Scholar

83 See generally Gilboy, J. A., Guilty Plea Negotiations and the Exclusionary Rule of Evidence: A Case Study of Chicago Narcotics Courts, 67 J. Crim. L. & Criminology 89 (1976).Google Scholar

84 Thus in murder cases, the suppression of the prosecutor's evidence at the preliminary hearing is not a reason for his recommendation not to reinitiate prosecution after a preliminary hearing discharge (see table 9 infra.)Google Scholar

85 People v. Holland, 56 Ill. 2d 318, 307 N.E.2d 380 (1974).Google Scholar

86 Gilboy, supra note 83, at 96–98.Google Scholar

87 E.g., the following is one such defense argument, followed by the prosecutor's rebuttal. DEFENSE: We have two defendants who made statements. The statement of one [his client] never in any way implicated himself even on the theory of accountability. He is driving down the expressway. He never said that he knew anybody was going to do any shooting. He was driving an automobile. Even if the officer's testimony is taken in the light most favorable to the state, his car died and he was in traffic pulling over [when the victim was shot by the codefendant]. Negative acquiescence and mere presence is not enough to hold a person accountable for the crime of another. STATE: …[The defendant] went with his brother. [The defendant] knew his brother had a gun, [the defendant] picked up the victim, [the defendant] was driving the vehicle when the shooting occurred, and [the defendant] had knowledge of the source of the weapon and the shooter; [the codefendant], at that moment was looking toward [the victim), and his brother was looking …. DEFENSE: Objection. JUDGE: Sustained. STATE: [The defendant] stated to the officer and the officer testified that the shooting occurred on the shoulder of the expressway. Thereafter the body was dropped off or ejected …. I think these actions taken together uncontradicted would show that [the defendant] was doing more than being present. He assisted in picking up the victim. He was present for the shooting. He was, in fact, providing the scene of the shooting. He helped dispose of the body… All the facts taken together show there was intent on the part of [the defendant] to perfect this act. Extracted from preliminary hearing transcript of Circuit Court of Cook County Case No. 76–3946.Google Scholar

88 This statement of the judge's reasons for his finding appears in the preliminary hearing transcript of Circuit Court of Cook County Case No. 78–3679.Google Scholar

89 One reason for this reaction by the Branch Court 66 judge to reinitiation of prosecution after his discharge of the accused might be his knowledge that the prosecutor might not have presented all his evidence during the preliminary hearing. He mentioned this as one of the reasons for the difference between himself and the prosecutor on whether a case should be bound over for trial. On a legal basis, there is no difference at all [between the prosecutor's and judge's decision]. I am aware of political considerations in the state's attorney's office, and these are not part of the judicial office. Concerning the political considerations in the state's attorney's office, they are more subject to the pressures of the community than the judges are. Another basic is that I understand there are witnesses that the slate did not wish to put on at the preliminary hearing but wanted to hold for trial. This would be like in gang retribution cases. They do not wish to disclose their witness. The state on purpose would withhold the evidence.Google Scholar

90 See text accompanying note 139 infra. Google Scholar

91 In Illinois, the purpose of preliminary hearings is to determine whether the crime charged has been committed and, if so, whether there is probable cause to believe that it was committed by the accused person. People v. Puleo, 51 Ill. Dec. 859, 96 Ill. App. 3d 457, 421 N.E.2d 367 (1981); People v. Horton, 65 111. 2d 413, 3 Ill. Dec. 436, 358 N.E.2d 1121 (1976). By statute: “The judge shall hold the defendant to answer to the court having jurisdiction of the offense if from the evidence it appears there is probable cause to believe an offense has been committed by the defendant.” Ill. Rev. Stat. ch. 38, § 109-3(a) (1981).Google Scholar

92 It was the practice of the judge that if the evidence at the preliminary hearing was sufficient to warrant prosecution on a felony charge, his finding was probable cause. I f the evidence only warranted prosecution for a misdemeanor offense he would find no probable cause and discharge the defendant.Google Scholar

93 After a finding of probable cause for a felony offense less than murder, the prosecutor may proceed to the grand jury to seek an indictment for a more seriou5 offense. This study did not collect data on the number of cases in which the prosecutor attempted to do this. The data in fig. 1 supra, however, indicate that in 66 of the 86 cases in which probable cause was found for an offense other than murder, the state subsequently obtained a murder indictment. It may also be noted that in 4 of the 20 cases prosecuted as lesser offenses, the state sought and obtained an indictment for a more serious offense than that for which probable cause was found at the preliminary hearing—voluntary rather than involuntary manslaughter.Google Scholar

94 It has sometimes been suggested that because of screening by the police and prosecutor beforehand, the preliminary hearing no longer serves an important screening function. See Miller, supra note 2, at 78; and Neubauer, supra note 2, at 131.Google Scholar

95 Fig. 1 indicates that a true bill on a murder charge was obtained in 175 cases and on a lesser offense in 7 cases. A no bill was the outcome for the remaining 6 cases.Google Scholar

96 In Illinois, a finding of no probable cause at a preliminary hearing is not a bar to a second preliminary hearing as long as there is no evidence of “harassment, bad faith, or fundamental unfairness” by the state. People v. Overstreet, 21 Ill. Dec. 227, 229, 64 Ill. App. 3d 287, 289, 381 N.E.2d 305, 307 (1978).Google Scholar

97 Cole reports, e.g., that in King County (Seattle), Washington, prosecutors have some control over what judge conducts the preliminary hearing. Cole, supra note 2, at 338.Google Scholar

98 E.g., in about 7% of the reinitiated cases the state reinitiated because it disagreed with the judge's ruling in a case (see tables 10 & 11 infra.) With that reason for reinitiating, the state might be expected to present essentially the same evidence again.Google Scholar

99 For the analyses in the rest of the article I have eliminated all cases that were prosecuted for lesser offenses than murder-20 of the probable cause cases and 7 of the reinitiated cases (see fig. 1 supra), leaving 1,071 probable cause murder cases and 175 reinitiated murder cases. The purpose was to eliminate from the comparisons of the three categories in tables 2–8 the effects of distinct or special characteristics of manslaughter and reckless homicide cases; e.g., generally they may be more likely than murder cases to result in a plea of guilty because of a lower mandatory minimum sentence on conviction.Google Scholar

100 See text accompanying notes 14–17 supra. Google Scholar

101 See text accompanying note 14 supra. Google Scholar

102 See, e.g., Calkins, supra note 14, at 431; Illinois Association for Criminal Justice, supra note 16, at 218.Google Scholar

103 People v. Rodgers, 54 Ill. Dec. 409, 98 Ill. App. 3d 999, 424 N.E.2d 1312 (1981). Generally in Illinois a trial court may dismiss an indictment for any of the reasons given in Ill. Rev. Stat. ch. 38, § 114-1. The court also has “inherent authority to dismiss an indictment in a criminal case where there has been a clear denial of due process even though that is not a stated ground in section 114–1.” People v. Lawson, 367 N.E.2d 1244, 1246 (1977). These grounds for dismissal are narrower than, e.g., in California, where a prosecutor could less easily use the grand jury to “rubber-stamp shaky prosecutions,” because the trial court has authority by statute to dismiss an indictment if it is not based on reasonable or probable cause. Even there, however, the absence of cross-examination in the grand jury proceeding may allow the state to proceed to the trial court with cases having substantial constitutional problems that otherwise might be identified by judicial review at the preliminary hearing. See Graham & Letwin, supra note 11, at 680.Google Scholar

104 See note 116 infra. Google Scholar

105 In Illinois generally “the preliminary hearing is the rule and the ‘direct indictment’ is the exception.” Report of Study Committee, supra note 70, at 5.Google Scholar

106 Judge Antell, a former prosecutor in New Jersey, has made a similar observation: But recognizing that a self-seeking prosecutor may, on occasion, wish to indulge himself, it must be noted that this will happen only where there are unique motivations. In the ordinary case he simply has no incentive to burden himself with pointless work, especially where this involves cases which are doomed from the outset. This is practicality at work…. The return of an indictment no longer is merely prologue to a conviction. The burdens of law enforcement are now well known, and the logic of self-interest suggests that few prosecutors will happily involve themselves in so formidable an undertaking except where the evidence is at least sufficient to augur a respectable showing. He does not relish the kind of disrepute which comes with the vexatious use of public resources, and this is a safeguard against arbitrary action. Antell, supra note 14, at 156.Google Scholar

107 See pt. III.B.2 supra. Google Scholar

108 Several researchers have noted the prosecutorial practice of bypassing the preliminary hearing to avoid problems associated with continuances. See, e.g., Jacoby, supra note 2, at 245 (grand jury used if “continuances are wearing out witnesses”); Donald M. McIntyre, Variation in the Operation of Grand Juries, A.B.F. Research Rptr., winter 1977, at 1, 1–2, (preliminary hearing bypassed to avoid problems due to defense continuances—delay of trial or inconvenience to witnesses).Google Scholar

109 A public defender I spoke with during the research questioned whether in serious cases proceeding by grand jury indictment was really to hasten prosecution, since it sped up by only a couple of days the time it took to get the defendant into trial court. Instead, he suggested, the state was primarily responding to the media and attempting to create a public image that they were “on top of things” and were “being as expeditious as possible in trying to prosecute the case.”Google Scholar

110 This reason for preferring the grand jury over the preliminary hearing was observed by Jacoby also: Since the prosecutor exercises control over the grand jury through his presentation of evidence, he can control the amount of time spent in the accusatory stage. This power is particularly important when there are cases not suited to preliminary hearings. For example, it is far easier to let the grand jury examine complicated cases (such as those involving worthless checks or fraud) which require the testimony of a large number of witnesses and consume a lot of time. Using a preliminary hearing for these exceptional cases not only disrupts normal processing times but also limits the ability of the prosecutor to clarify complicated cases, evidence, or issues. This latter point is most crucial. Jacoby, supra note 2, at 157. In Maricopa County (Phoenix), the grand jury was the preferred method of prosecution in complicated cases or ones that could take a long time to present at a preliminary hearing. For a list of the Maricopa County prosecutor's written guidelines for deciding which cases should be taken to the grand jury and which to preliminary hearing, see Emerson & Ames, supra note 63, at 80–81.Google Scholar

111 This is a generally held opinion. One study reports that preliminary hearings typically take two to three times longer than grand jury proceedings. McIntyre, supra note 108, at 1.Google Scholar

112 Although not mentioned by prosecutors in this study, prosecution by indictment can also give the prosecution a tactical advantage over the defense: the state does not have to reveal its evidence or case strategy at an early stage of a case, which may ultimately increase its chances of obtaining a conviction. Jacoby, supra note 2, at 161. Jacoby also mentioned that grand jury indictment may allow the prosecutor to avoid a preliminary hearing judge “who is considered incompetent, unqualified, or philosophically at odds with prosecutorial policy.”Id. at 157.Google Scholar

113 In tables 2–8, the chi square (χ²) (a general test of significance) is used to examine whether the difference between two or more statistics is a real difference or is due to chance variation. In table 2, e.g., the chi square is used to test the null hypothesis that there is no significant difference between the outcomes of cases in the two prosecuted groups under study. For a more complete description of the chi square, see N. M. Downie & R. W. Heath, Basic Statistical Methods 160–73 (2d. ed. New York: Harper & Row, 1965).Google Scholar

114 See note 99 supra on the composition of cases included in the probable cause category in table 2.Google Scholar

115 The following table shows case disposition: comparing prosecution by grand jury and by preliminary hearing (probable cause and reinitiated cases combined). Dismissed Acquitted Convicted Total No. % No. % No. % No % No. % Grand jury cases… 37 10.4 70 19.6 250 70.0 357^c 100.0 Probable cause and reinitiated cases… 111 9.2 244 20.2 856 70.7 1,211^d 100.0 χ²= 0.37; p < 0.90 NOTE: The chi square (χ²) is based in part on the unweighted 50% random sample of vertically represented private attorney cases. For the chi square calculation, the numbers of probable cause cases were 97 dismissed, 204 acquitted, and 743 convicted. ^a372 cases prosecuted by grand jury indictment, excluding 15 pending cases. ^b1,005 cases with preliminary hearing finding of probable cause on a murder charge, 66 cases with finding of probable cause on a lesser offense but subsequently indicted for murder, and 175 cases discharged at preliminary hearing and subsequently indicted for murder, excluding 35 pending cases of the total 1,246.Google Scholar

116 One assumption underlying my interpretation of the statistical data in table 2 should be discussed. This comparison of grand jury and preliminary hearing cases is based on aggregate conviction data. The assumption is that similar conviction rates reflect similar groups of cases prosecuted through both forums. It is possible, however, that prosecutors may choose the grand jury in unusually strong as well as in unusually weak cases and that the conviction rate might even out and be the same as for preliminary hearing cases, even though the strength of the evidence in cases taken to the two forums might be quite different. Because I did not collect for individual cases prosecutors' specific reasons for choosing the grand jury, it cannot be said conclusively that the grand jury conviction rate does not reflect that the grand jury was chosen for unusually strong and weak cases. It may be noted, though, that even if that reason for favoring the grand jury was operating, the conviction rate data in table 2 suggest that the number of unusually weak cases going to the grand jury was not large enough to substantially pull down the overall conviction rate. Moreover, such an interpretation of the data is not consistent with other data in the study. Generally the statistical data present a mixed picture, but one which on the whole strongly suggests that the grand jury is not used to prosecute both unusually weak and unusually strong cases. If it were so used we might expect to see a number of distinct features in grand jury cases not found in probable cause cases. First, we might expect more trials than pleas of guilty in grand jury cases. That would be due to two factors. Weak cases may be taken to trial because the accused feels a strong incentive to have the state's weak evidence reviewed at trial. Strong cases may be taken to trial because the state would be unwilling to negotiate a guilty plea in exchange for a reduced charge or a lenient sentence. Because the mandatory minimum sentence is 20 years for murder, the defendant may conclude that he has little to risk by proceeding to trial. Although the data do indicate that grand jury cases are significantly more likely to proceed to trial (63%) than probable cause cases are (57%), other data do not suggest that the grand jury is used to prosecute unusually weak as well as unusually strong cases. The existence of unusually weak or unusually strong cases was not reflected in the acquittal rates at bench trials or at jury trials. If the state were taking unusually weak cases to the grand jury, we might expect to see among grand jury cases a significantly lower rate of conviction at bench trial, the type of trial to which the defense is likely to take weaker cases. (For a full discussion of the implications of selecting a bench trial rather than a jury trial see the text accompanying notes 130-33 infra.) The lower conviction rate was not in evidence: in both categories, 57% of the cases at bench trial were convicted. Likewise, if the state were taking unusually strong cases to the grand jury, we might expect to see among grand jury cases a significantly higher rate of conviction at jury trials, the type of trial to which the defense would be most likely to take cases in which the state has a very strong case against a defendant. Instead, 79% of the grand jury cases and 75% of the probable cause cases at a jury trial resulted in a conviction. This is not a statistically significant difference. These data on conviction rates at bench and jury trials are particularly important because they do not indicate that the overall conviction rate in grand jury cases compared with probable cause cases is due to the aggregation of cases with a significantly higher rate of conviction at jury trial and a significantly lower rate of conviction at bench trial. That is, the aggregation of conviction data does not appear to hide a more complex phenomenon of the state's bypassing preliminary hearing in unusually weak and in unusually strong cases.Google Scholar

117 See note 115 supra. Google Scholar

118 See text accompanying note 104 supra. Google Scholar

119 See text accompanying note 106 supra. Google Scholar

120 See note 99 supra. Google Scholar

121 There appears to be no published data for other jurisdictions comparing the conviction rate in reinitiated cases to that in probable cause cases.Google Scholar

122 See table 13 and the discussion on dismissals in pt. VI.D.4 infra. Google Scholar

123 See tables 10 and 12 infra; see discussions about reasons for reinitiating prosecution in pt. VI.D.2 infra, about the rate of conviction in prosecutions reinitiated for various reasons in pt. VI.D.3 infra. Google Scholar

124 Almost all the reinitiated case dismissals in table 3 were requested by the state: of the 32 dismissed cases for which prosecutor's files were found (see table 13 infra), only 3 were judicial dismissals.Google Scholar

125 The percentage of cases proceeding to trial is extraordinarily high in both categories. In part this is because these are murder cases. Generally defendants are unlikely to plead guilty to the murder charge because of the high mandatory minimum sentence in Illinois—20 years. It may be noted that the ra§e of proceeding to trial in cases where prosecution was initiated by grand jury indictment falls somewhere between these two categories of cases: 63.4% (203 of 320) of the grand jury cases went to trial.Google Scholar

126 E.g., as one Chicago criminal defense lawyer reported to Alschuler during his study of plea bargaining, “When a prosecutor has a dead-bang case, he is likely to come up with an impossible offer like thirty to fifty years.” See Alschuler, supra note 3, at 60.Google Scholar

127 See Ill. Rev. Stat. ch. 38 5 1005-8-1 (1981). This sentencing law became effective Feb. 1, 1978, when Illinois shifted from an indeterminate to a determinate sentencing scheme in felony cases. (Note that only the last 11 months of this study was under the new law.) In the former law, the mandatory minimum sentence for murder was 14 years. For a discussion of the current law see Aspen, Marvin E., New Class X Sentencing Law: An Analysis, 66 Ill. B. J. 344, 347 (1978).Google Scholar

128 Mather suggests, in her study of what determines whether a case will be settled by plea bargaining or by trial, that generally public defenders will recommend a trial to a client if the risks of trial are low and the possible gains are high. She goes on to describe the particular considerations in cases involving a serious offense. “In a serious case where there is a good chance of conviction (either on the original charge or on a lesser charge) and where a non-state prison sentence cannot be obtained through bargaining … [h]ere, the sentencing risks are low since the defendant goes to prison whether by trial conviction or by plea; the possible gain is acquittal which means complete avoidance of punishment.” See Mather, Lynn M., Some Determinants of the Method of Case Disposition: Decision-making by Public Defenders in Los Angeles, 8 Law & Soc'y Rev. 187, 209 (1973). A trial manual for defense lawyers also suggests that while a plea of guilty may normally be considered “when the prosecution's evidence of guilt is overwhelming,” the consequences of the plea “may be so awful, in the client's circumstances, that even the faintest ray of hope offered by a trial is magnified in significance. If, upon conviction, the defendant will have many years of back parole time facing him or become subject to an automatic life sentence or its equivalent, he has nothing to lose by denying his guilt and going to trial.” Anthony G. Amsterdam, Bernard L. Segal, & Martin K. Miller, Trial Manual for the Defense of Criminal Cases 1–187, 1–190 (3d ed. Philadelphia: American Law Institute, 1974).Google Scholar

129 The choice of bench trial in grand jury cases was even less frequent—60.1% (122 of the 203 that went to trial).Google Scholar

130 But see note 133 Infra. Google Scholar

131 Letter to the author from James P. Carey, Associate Professor of Law, Loyola University of Chicago School of Law (Aug. 25, 1982) (emphasis added).Google Scholar

132 In Heumann's study of plea bargaining, interviews with prosecutors indicated that even in a “sure” case there is always the possibility that the state will lose at trial because of the unpredictability of juries. See Milton Heumann, Plea Bargaining: The Experiences of Prosecutors, Judges, and Defense Attorneys 111 (Chicago: University of Chicago Press, 1978).Google Scholar

133 There are exceptions, of course, to these rules of thumb for choosing bench or jury trial. First, there may be circumstances in which a defense lawyer may choose a bench trial where the state's case is strong: e.g., if a client rejects his advice to plead guilty in a hopeless case, one solution might be to abandon any glimmer of hope for acquittal by a jury and select a bench trial, which would limit use of the court's time and public resources, an economy that could perhaps incline the judge to some vestige of mercy in sentencing. Second, there are circumstances for preferring jury trial even if the case against a defendant is weak. As a former defense lawyer suggests, a lawyer's thinking about the type of trial to select differs according to the kind of weaknesses in a case (see quotation at note 131 supra.) When he feels success at trial turns on the strength of the case he intends to present and on the credibility of his witnesses, he may consider a jury more likely than a judge to believe the witness and to acquit the defendant.Google Scholar

134 Grand jury cases differed similarly from reinitiated cases: of the 203 cases that went to trial only 70 (34.5%) ended in acquittal.Google Scholar

135 Grand jury cases had a jury trial acquittal rate fairly similar to that of reinitiated and probable cause cases, 21% (17 of 81 cases).Google Scholar

136 It should be noted that the relationship in table 7 is not statistically significant at the same levels as in the previous discussions. The rate of acquittal at bench trial for grand jury cases was also quite different from reinitiated cases: 43.4% (53 of 122) of cases that went to a bench trial resulted in acquittal.Google Scholar

137 At the time of preparing this article, the structure of the review process for homicide cases discharged at the preliminary hearing is that after the preliminary hearing assistant makes his recommendation on whether to reinitiate prosecution, the case is reviewed by the supervisor of the Homicide Unit, then by the supervisor of Preliminary Hearings, and finally by the chief of the Felony Trial Division. This system is used only for homicide cases; for all other felony offenses, typically the courtroom supervisor is the only person who reviews the preliminary hearing assistant's recommendation and makes a final decision on what to do in the case.Google Scholar

138 Referral of difficult problems to a superior is one form of supervision that can serve as a check against any arbitrary action by prosecutors (Davis, supra note 1, at 143). To the extent that supervision depends on the initiative of the prosecutor to refer problems it is not clear whether by itself it is a sufficient source of review of the prosecutor's discretion. During the period of this study, probably the most important control exercised by the Cook County state's attorney's office over the decision to reinitiate prosecution was selecting an experienced prosecutor to act as supervisor in Branch Court 66.Google Scholar

139 See text accompanying notes 88–90 supra. Google Scholar

140 One former supervisor reported that since these documents could be discoverable by the defense he was very cautious about what his assistants put in them.Google Scholar

141 If a key prosecution witness fails to appear for the hearing and the judge will not permit the state to continue the case to find its witness, the prosecutor may still go forward with the hearing. One supervisor reported that the reason he was inclined to proceed to a preliminary hearing rather than dismiss the case was that he believed there might be future problems in trying to reinstate the charges after a dismissal.Google Scholar

142 Although these memoranda are not used for other purposes in the state's attorney's office, they do become a permanent part of the case file.Google Scholar

143 A fourth option is to take the case to a second preliminary hearing, but in this jurisdiction it is the custom to seek an indictment rather than to refile the case in preliminary hearing court. See text accompanying notes 96–97 supra. Google Scholar

144 In other jurisdictions, a strategy of shopping for a state's attorney willing to reinitiate prosecution after a preliminary hearing discharge would not be surprising given reports of attempts by the police at the time of the initial charging decision to shop around for a prosecutor who will be “sympathetic to their view on a case” and willing to charge the suspect. See, e.g., Cole, supra note 2, at 335–36. On the other hand, the police might distinguish between decisions to charge initially and decisions to reinitiate prosecution. They may believe they are entitled to a preliminary hearing review of the case they have prepared, since a judge might assess the evidence differently from a prosecutor. But once there has been a judicial review and discharge of the accused person, the police may regard this as sufficient examination of the weight of their evidence and not shop around for a supervisor willing to reinitiate prosecution.Google Scholar

145 Generally, appeal “by the affected party from the subordinate to the superior” is a form of supervision that can act as a check on the prosecutor's discretion. Davis, supra note 1, at 143.Google Scholar

146 The Cook County state's attorney's office assigned some of its own investigators to Branch Court 66. They delivered subpoenas and were sometimes sent by supervisors to locate witnesses that needed to be interviewed before a decision could be made whether to reinitiate prosecution.Google Scholar

147 An original aim of the study was to obtain information about current prosecutorial decision making by observing and interviewing prosecutors making decisions in a sample of cases. The rate of discharge of murder cases, however, was too small to make that method feasible. Given the rate of discharge (377 cases in the 39-month period of the study, or about 10 cases per month), an observer would have had to spend approximately three months of full-time research in Branch Court 66 in order to observe decision making in 30 discharged cases. This would mean a sample of only about 15 reinitiated cases and 15 cases in which no prosecution was pursued (see fig. 1 supra: 49.9% rate of reinitiation). Thus, identifying the range and frequency of reasons for declining or reinitiating prosecution simply could not be readily developed for murder cases using those methods of data collection.Google Scholar

148 This appears to he true elsewhere as well. Mather, e.g., made a similar finding in talking with assistant district attorneys in Los Angeles. One district attorney reported to her: I look at the probability of conviction—how weak or strong the case is—and weigh that against the public interest or cost to the community of prosecution. A serious ruse s more likety to be refiled than a little one. This case [a small theft case that had just been dismissed] I don't think will be refiled. It isn't very important. You see, in a small case like this one, the cost to the community is outweighed by the weak probability of conviction. But if it were a big case, then we'd refile…. Prior record is also taken into consideration. For example, on a G.T. A. [auto theft] where the guy's got a long history of G.T.A.s, then even if it's a weak case, we refile. There the weakness of the case is outweighed by the public interest in prosecuting the guy—you know, a habitual criminal. Mather, supra note 11, at 52 (emphasis by Mather).Google Scholar

149 According to prosecutors, the cases received from Felony Review (the office's precharge screening unit) were supposed to he provable at the time charges were filed.Google Scholar

150 The period within which in Illinois an incarcerated defendant is to be brought to trial or else discharged from custody is 120 days. When the defendant is discharged from custody he is immune from further prosecution for the same offense. See Ill. Rev. Stat. ch. 38, § 103–5(a), (d) (1981).Google Scholar

151 In this jurisdiction, the desire of prosecutors not to burden themselves with cases that could not be convicted, rather than the need to conserve limited office resources (i.e., organizational efficiency), seemed to propel decisions not to reinitiate prosecution if there was insufficient evidence. One prosecutor, e.g., indicated that while he was a supervisor in Branch Court 66 there was a tremendous expansion in the number of trial courts and of assistant state's attorneys—the case load for a trial court assistant dropped from about 350 to about 125 cases. He stated that although he knew that with the office's resources the state could proceed with very weak cases, he believed it was a waste of lawyers' time and effort to do so.Google Scholar

152 This is not surprising. The seriousness of the offense has also been noted to be a factor affecting the prosecutor's initial decision to charge in cases where the likelihood of conviction is uncertain. See Kaplan, supra note 2, at 181 (“In the case of the more serious crimes, it was often felt that the accused should be put on trial even though prosecution might routinely be declined for a lesser crime where conviction was equally uncertain”).Google Scholar

153 Graham & Letwin's interviews in Los Angeles suggest similar situations in which a prosecutor might reinitiate prosecution: One prosecutor asserted that the policy of his office was to refile only in a limited category of cases. Prime candidates were said to be those dismissed because of a magistrate's “erroneous view of the law”, because of an easily remedied “technica1” flaw in the prosecution's case, or because a crucial witness was unable to testify. Where a defendant is discharged simply because of the lack of persuasive force of the prosecution's testimony upon the magistrate, the decision whether or not to refile was said normally to depend on whether the deputy prosecuting the case shared the magistrate's judgment. If the judgment was shared, the case would not likely be refiled, even though the prosecutor might believe that he had presented a technically adequate case to warrant holding the defendant to answer. Graham & Letwin, supra note 11, at 73 (emphasis added).Google Scholar

154 See note 61 supra. Google Scholar

155 See note 141 supra. Google Scholar

156 See pt. III. D supra. Google Scholar

157 See also pt. V1.D. 4 infra for an indirect examination through various prosecutor's case file data of the reasons for trial court dismissals in reinitiated and probable cause cases.Google Scholar

158 See text at notes 140–42 supra. Google Scholar

159 The cases are analyzed in pt. V supra. Google Scholar

160 But see note 140 supra, indicating that at least some memoranda were prepared with an eye toward potential discovery by defense lawyers. How that consideration affected the contents of the memoranda is unknown.Google Scholar

161 In 6 discharged cases in the sample the assistant proposed “no further action,” but the supervisor believed the case should be sent to the grand jury. The disagreements were known, e.g., from a supervisor's note to an assistant. No case studied showed the supervisor disagreeing with the assistant's recommendation to proceed to the grand jury. It is certainly possible that the study underestimates the amount of disagreement between assistants and supervisors since, e.g., internal memoranda may have been rewritten after the case was discussed.Google Scholar

162 From Oct. 1975 through Dec. 1978, 189 murder cases discharged at the preliminary hearing were never further prosecuted (see fig. 1 supra.) The Cook County state's attorney's office searched for the files of these cases but found only 60 of them, in the file drawers of Branch Court 66. The office thought those not found probably had been shipped at some time to a warehouse but did not know where. As far as the prosecutor's office knew there was no systematic bias to the files located, i.e., nothing to distinguish the cases found from those not found. Of the 60 cases, 3 contained no information about the reason for no further prosecution.Google Scholar

163 To determine the reasons for no further prosecution, for 55 cases internal memoranda in the prosecutor's office case files were used alone or in conjunction with other documents in the file (e.g., investigative reports discussing efforts to find a witness). For 2 cases (one murder by 2 codefendants) the preliminary hearing transcript was solely relied on; it showed that after the defense presented evidence that the homicide was committed in self-defense, the court discharged both defendants, saying the state offered no contradictory evidence. I classified both in category 2(d) in table 9, since presumably the state did not later have new or additional evidence to refute that testimony.Google Scholar

164 The small number of discharged cases that the supervisor sent on to grand jury for indictment against the assistant's recommendation not to prosecute further are included in table 10, not table 9. See note 161 supra and accompanying text.Google Scholar

165 In total there were 49 cases in the study's sample in which witnesses unexpectedly failed to appear at the preliminary hearing (table 9 items l(a)-(c), table 10 items l(a)-(b)). This suggests a significant problem that occurs within the first month of a case. Future research might examine what state's attorneys and police do (or don't do) to keep track of witnesses and to encourage their cooperation.Google Scholar

166 Unless otherwise indicated, when the internal memoranda are quoted, they are quoted in full with only identifying information such as names and dates deleted.Google Scholar

167 In some cases, however, the state does proceed to the grand jury without first locating its witnesses, with the intention of introducing a missing witness's statement through the hearsay testimony of the police officer. See table 10 item l(b) infra and accompanying discussion in the text.Google Scholar

168 This is not always true. In some cases when a witness is not present at the hearing it is for tactical or other reasons known to the prosecutor (see table 10 item 1(c) infro.)Google Scholar

169 The 88% figure was developed from the data in tables 9 and 10 and fig. 1. First, the 28 cases in table 9 for which the state attempted to determine whether its witnesses could be found and would be willing to cooperate (table 9 items l (a)-(c)) were 49% of the 57 cases in the sample of discharged cases never indicted. Applying that percentage to the total of discharged cases never indicted would mean that in about 93 of the total 189 (see fig. 1 supra) the state attempted to find its witnesses before deciding not to pursue prosecution (49%× 189 = 93). Second, table 10 item I(a) shows that in 11 other cases the state sought to locate its witnesses before seeking indictment. These 11 cases were 8% of the 134 reinitiated cases in the table 10 sample. Extrapolating from these data, this equals 14 cases of the total 175 reinitiated cases (see fig. 1 supra) (8%× 175 = 14). Third, there were 10 cases for which the state did not attempt to find its witnesses before seeking an indictment (table 10 item 1(b)). This also equals about 14 cases of the total 175 reinitiated cases. The extrapolations taken together come to 121 cases (93 + 14 + 14) in the population of discharged cases in which the state's witnesses unexpectedly failed to appear at the preliminary hearing. In 107 of these (88.4%) the state did attempt to locate the witnesses before deciding whether to seek indictment.Google Scholar

170 In determining the reasons for reinitiating prosecution given in table 10, I used several different sources of information, all from the files of reinitiated cases. For 80 of the 134 cases, internal memoranda prepared by the preliminary hearing assistant in no probable cause cases were the primary source. For 8 cases, the primary source was the “Direct Report,” a statement in one or two sentences, prepared by the preliminary hearing assistant, to give the reasons why the state was proceeding to the grand jury for indictment. For 16 cases, the primary source was a “Synopsis of Facts” sheet, a sheet containing 11 questions to be answered by the preliminary hearing assistant for the trial court assistant's use. Answers supplied pertinent information on (1) testimony of witnesses, (2) testimony of police, (3) evaluation of witnesses, (4) preliminary hearing testimony of defendant and defense witnesses, (5) ruling of court on motion to suppress, (6) prior felony convictions, (7) defendant on probation or parole, (8) other indictments pending, (9) additional investigation needed, (10) prior plea offer, (11) evaluation of case. Answers to questions 1–4 and 11 were the ones largely relied on. In 22 cases the preliminary hearing transcript was used, usually in conjunction with other sources in the file. For instance, the transcript for several cases indicated the prosecutor's key witness was not in court for the preliminary hearing. Those cases were coded in table 10 as (a), (b), or (c) of category 1 according to the circumstances of the witness's failure to appear: e.g., the transcript stated witness was unable to get a baby sitter but the state expected her to testify at trial; or witness was unexpectedly missing and other documents in the file indicated whether the state found the witness before proceeding to the grand jury. In other cases the transcript indicated this was a codefendant case in which one defendant was discharged. Because the state's routine practice was to further prosecute the discharged codefendant, I coded such cases in table 10 as category 8. In the 8 remaining cases an assortment of other sources were used to determine why prosecution was reinitiated, including in one case an interview with a prosecutor who had been the trial attorney in the case. Although I have noted above the primary document used in classifying each of the 134 reinitiated cases, typically the process of classification was more complex than simply using one source. In every case the entire file was examined in order to fully understand why the discharge occurred and why the case was reinitiated. Often several documents were used to classify the case—I compared memoranda and other documents in a file, and where memoranda were incomplete (e.g., only gave the reason for the discharge and what was to be done before going to the grand jury) or where memoranda were missing, several sources typically were used to piece together the reasons for the discharge and for reinitiating prosecution. In 24 of the above reinitiated cases (12 with memos), I asked a supervisory level prosecutor to examine the file and to discuss it with me in order to make certain that I fully understood the contents of the file and that I was not overlooking something critical to my classification of the case. Of the 24 cases, in 3 the prosecutor was familiar with the case; in the remaining 21 he based his comments solely on a reading of the file. From my examination of the file and discussion with the prosecutor, 1 classified the 24 cases as follows: 2 in category 1, 4 in category 2, 1 in category 4, 5 in category 5, 3 in category 6, 2 in category 7, and 7 in category 8.Google Scholar

171 The information presented in table 10 was gathered for a random sample of 139 of the 175 reinitiated cases. The table excludes 4 cases whose files had no information on the reason for further prosecution and one case where the state erred in proceeding by information after a preliminary hearing discharge.Google Scholar

172 In some cases the state presents various witnesses to the grand jury for information purposes only and not for an indictment. If the state subsequently seeks an indictment, it usually introduces a witness's previous testimony to the grand jury rather than recall the witness.Google Scholar

173 These cases would have been placed in 1 (b) of table 10 if there had been information in the file that the witnesses unexpectedly failed to appear but the state assumed they would be available for trial and so proceeded to the grand jury.Google Scholar

174 See pt. III.C supra. Google Scholar

175 See pt. IV. A supra. Google Scholar

176 See pt. III.D supra. Google Scholar

177 In 5 cases the additional evidence was physical evidence apparently available but not introduced at the preliminary hearing, e.g., results of blood tests, analysis of gun powder in the victim's hair, or blood and victim's watch found. In 2 cases a witness had other testimony—in one a police officer had not included in his testimony parts of a defendant's statement that contradicted the defendant's claim that he fired a gun in self-defense; in the other, a witness who had unexpectedly changed his testimony at the preliminary hearing was taken to the grand jury on the same day, repudiated his statement and testified for the state as had been expected at the preliminary hearing. In 4 others the state expected to introduce additional findings of the medical examiner, e.g., the angle of the knife wound that contradicted the defendant's story at the preliminary hearing. In at least 4 of these 11 cases, the state presumably would have introduced the evidence at the preliminary hearing. Specifically, in one case the medical examiner's protocol was missing when the preliminary hearing was held; in another, the judge was unwilling to allow the state's critical evidence to be introduced because the state was unable to establish the chain of custody of the evidence after it was removed from the scene of the crime; in one case the state would have preferred the police officer to testify fully about the defendant's statement to him; and in the last case, the state called a witness and anticipated certain testimony that it was apparently only later able to elicit at the grand jury.Google Scholar

178 Table 12 infra indicates, however, that the conviction rate was about 56% for cases reinitiated because of new or additional evidence. This rate of conviction is substantially higher than that for cases reinitiated for nonevidentiary reasons. See especially the discussion in part VI.D.3 infra. Google Scholar

179 See, e.g., the quotation from Graham & Letwin in note 153 supra. Google Scholar

180 Because it is unusual for the state to seek indictment without first locating witnesses who unexpectedly failed to appear at preliminary hearing, reinitiation without witnesses may be the result in some of these cases of special motivations to prosecute the defendant because of characteristics of the defendant or of the crime, or overzealousness of prosecutors who simply assumed their witnesses eventually would be found and would cooperate.Google Scholar

181 See text accompanying note 72 supra. Google Scholar

182 In Costello v. United States, 350 U.S. 359 (1956), the Supreme Court held that an indictment may be based solely on the presentation of hearsay evidence. Subsequently, in United States v. Calandra, 414 U.S. 338 (1974), the Supreme Court in the context of deciding whether to extend the exclusionary rule to grand jury proceedings indicated that due process standards applicable at trial need not apply to grand jury proceedings. The Illinois Supreme Court has considered on several occasions questions regarding indictments based solely on hearsay and has upheld its previous decisions that the presentation of hearsay evidence alone is sufficient to support the return of an indictment. See People v. Creque, 22 Ill. Dec. 403. 72 Ill. 2d 515, 382 N.E.2d 793 (1978). In one earlier lllinois case, People v. Jones, 19 Ill. 2d 37, 166N.E.2d 1(1960), the court specifically considered whether an indictment in a narcotics case (where evidence was suppressed at the preliminary hearing) was properly quashed by the trial court judge who had interpreted the requirement of “competent evidence” for an indictment to mean evidence admissible at the trial. The court concluded that the indictment was improperly quashed because the defendants had no constitutional right to challenge an indictment as being founded on wholly inadequate or inadmissible evidence.Google Scholar

183 People v. Jones, 19 Ill. 2d 37, 166 N.E.2d 1 (1960).Google Scholar

184 According to ABA Standards for Criminal Justice: “A prosecutor should present to the grand jury only evidence which the prosecutor believes would be admissible at trial. However, in appropriate cases, the prosecutor may present witnesses to summarize admissible evidence available to the prosecutor which the prosecutor believes he or she will be able to present at trial.” American Bar Association, Standards for Criminal Justice, Standard 3–3.6(a) (2d ed. Boston: Little, Brown & Co., 1980).Google Scholar

185 In some codefendant cases, however, immunity is offered to a codefendant at the preliminary hearing stage (see table 9 item 3).Google Scholar

186 The 80% was calculated in the following way. First, all the reasons for not reinitiating prosecution (including probably the grants of immunity) suggest some evaluation by the state of the strength of its case. Extrapolating from the table 9 sample would mean that all 189 discharged cases that were never further prosecuted (see fig. 1 supra) represented decisions not to reinitiate for reasons relating to the state's evaluation of the strength of its case. Second, cases in table 10 were reinitiated after the state knew: for 36 that its witnesses would be available (1(a), (c)), and for 43 others that it had new or additional evidence (2(a)-(f)). These 79 cases are 59% of the 134 cases in the table 10 sample. Extrapolating this to the total 175 reinitiated true bill murder cases (see fig. 1 supra) would mean that 103 were reinitiated for reasons arising from prosecutors' evaluating the strength of the case (59%× 175 = 103). Third, 189 + 103 = 292, or 80% of the total 364 discharged cases.Google Scholar

187 This 10% figure was calculated from table 10 items 1(b),3–5. The 27 cases in these 4 categories are 20% of the 134 cases in the table 10 sample. Extrapolating from the sample to the total: 20%× 175 = 35. These 35 cases are 10% of the total 364 discharged cases, i.e., another 10% involved decisions about the strength of the case.Google Scholar

188 This 2% figure was calculated from table 10 items 6, 7. The 7 cases in these 2 categories are 5.2% of the 134 cases in the sample. Extrapolating from these data: 5.2%× 175 = 9, or 2% of the total 364 discharged cases, that were reinitiated for decisions based on nonevidentiary factors. We could also perhaps include here cases in table 10 item 1(b) where the state did not locate its witnesses before proceeding to the grand jury. Because this was unusual for the state, it could suggest special motivations or overzealousness as factors in reinitiating prosecution. See note 180 supra. If this category were included, nonevidentiary factors may affect 6% of the decisions to reinitiate prosecution.Google Scholar

189 This 8% figure was calculated from table 10 item 8. The 21 cases in this category are 15.7% of the 134 cases in the sample. Extrapolating to the total (15.7%× 175 = 28) would mean about 28 of the total 175 reinitiated cases would involve a decision by the slate to reinitiate prosecution based on codefendant accountability. These 28 cases are 8% of the total 364 discharged cases.Google Scholar

190 In the conclusion of this article I discuss several means to restrain or eliminate the prosecutor's unlimited power to reinitiate prosecution as well as the potential consequences of these proposals.Google Scholar

191 The differences in the rate of conviction between cases further prosecuted for evidentiary reasons and those for nonevidentiary reasons would be even more exaggerated were we to add the table 10 item 1(b) cases, which may also represent nonevidentiary motivations. See note 193 infra. Google Scholar

192 Perhaps the routine taking of discharged codefendant accountability cases to the grand jury explains why 6 of the 20 cases in this category were missing internal memoranda as well as other documents (such as a “Direct Report” or “Synopsis of Facts” sheet) that would give the reason for reinitiating. See note 170 supra on the use of these documents.Google Scholar

193 As noted elsewhere, because it is unusual for the state to seek indictment without first locating its witnesses, the reinitiation of prosecution in cases in this category may also suggest special motivations or overzealousness of prosecutors.Google Scholar

194 Reasons for acquittal are not compared because there was insufficient information about them. See note 40 supra. Google Scholar

195 Unfortunately, I cannot include grand jury cases (discussed in pt. IV) in this comparison because I did not collect reasons for case dismissals for that group.Google Scholar

196 See text following note 122 supra. Google Scholar

197 See note 42 supra. Google Scholar

198 See note 43 supra and accompanying text.Google Scholar

199 See table 3 supra. Google Scholar

200 Reasons for dismissal were found for 32 of the 34 reinitiated cases, in various records in the files: for 17 cases, Nolle Pross Form; for 10, daily court summary; for 3, the word of an administrator (after he read the case files) that the state was unable to prove accountability of indicted codefendants; for 1, a defense document demanding dismissal because the defendant was never indicted after a finding of no probable cause; for 1, a sheet requesting follow-up investigation on state's 2 missing key witnesses. Reasons for dismissal were found for 30 of the 34 probable cause cases sampled, in the following sources: for 20, Nolle Pross Form; for 7, daily court summary; for 1, a memo saying state's key witness recanted; the judgment of an administrator (after he read the case files) that for 1, the state could not find key witnesses and that for 1, state's evidence would make it difficult to try both codefendants separately. It should be noted that the number of dismissed reinitiated cases is different in tables 12 and 13 because in table 13, I included all reinitiated cases for which information was available but in table 12 (based on table lo), only a sample of reinitiated cases.Google Scholar

201 In table 12, categories 1–5 had 50 dismissals and acquittals, 76% of the total 66 acquittals and dismissals in the table 12 sample. If we exclude the 1 (b) cases, in which the prosecutor did not locate his missing witnesses before proceeding to the grand jury (which, as stated earlier, may suggest special motivations or overzealousness), then the 76% drops to 61%.Google Scholar

202 See pt. VI. C supra. Google Scholar

203 In table 12, categories 6, 7, and 8 had 16 dismissals and acquittals, 24% of the total 66 acquittals and dismissals in the sample. See also note 201 supra. Google Scholar

204 Other researchers also have noted that career and professional aspirations of prosecutors generally militate against their being overzealous in their prosecution efforts. Malcolm M. Feeley & Mark H. Lazer-son, Police-Prosecutor Relationships: An Interorganizational Perspective, in Keith O. Boyum & Lynn Mather, eds., Empirical Theories About Courts 216, 231 (N.Y.: Longman Inc., 1983).Google Scholar

205 Cole, supra note 2, at 334–35, 337.Google Scholar

206 Kaplan, supra note 2, at 178–81.Google Scholar

207 See, e.g., Cole, supra note 2, at 334–35: A request for prosecution may be rejected for a number of reasons relating to questions of evidence. Not only must the prosecutor believe that the evidence will secure a conviction, but he must also be aware of community norms relating to the types of acts that should be prosecuted…. [T]he heinous nature of the crime, together with the expected public reaction, may force both the police and prosecutor to press for conviction when evidence is less than satisfactory.Google Scholar

208 Kaplan, supra note 2, at 180 (“[M]ost assistants felt that it was not right to use the prosecutorial system just to harass an individual, however guilty he might be and hence, unless the case could be won, it was morally wrong to prosecute it”).Google Scholar

209 Id. (“assistants regarded the time and money spent on an unsuccessful prosecution as completely wasted”).Google Scholar

210 Cole, supra note 2, at 335; Kaplan, supra note 2, at 181.Google Scholar

211 Cole, supra note 2, at 335; Kaplan, supra note 2, at 178, 180.Google Scholar

212 Kaplan, supra note 2, at 180.Google Scholar

213 See pt. V supra. Google Scholar

214 See note 188 supra and accompanying text.Google Scholar

215 See note 189 supra and accompanying text.Google Scholar

216 More than half the murder defendants prosecuted after their preliminary hearing discharge were convicted (see table 3 supra.)Google Scholar

217 See fig. 1 supra. In only 3% of cases discharged at preliminary hearing and later presented to grand jury for indictment was a no bill returned. In addition, an indictment was returned in every case in which initiation of prosecution was by grand jury.Google Scholar

218 An example of recent legislation designed to restrain improper exercise of the power to reinitiate is that enacted in Illinois since my study was completed (see note 18 supra.)Google Scholar

219 Albert W. Alschuler commented on this section of the manuscript as follows: You discuss whether new restrictions should be placed on prosecutors in a “dual system of prosecution,” but the options that you discuss are not the only ones. Can anything be said for the “dual system” itself? The paper demonstrates that things are not as had as they might be. Nevertheless, a system in which the prosecutor can choose between the grand jury and the preliminary hearing without restriction still seems strange and inappropriate. Indeed, one possible implication of your conclusion is that both the grand jury and the preliminary hearing should be discarded and the charging power vested unambiguously in the prosecutor's office. Certainly the paper reinforces the general view that the grand jury checks the prosecutor's power hardly at all. In terms of its supposed “shield” function, the grand jury seems almost a functionless appendage. And what function does the preliminary hearing serve? It provides some discovery to defense attorneys but only as much as prosecutors choose to provide. When a prosecutor wishes to avoid discovery altogether, he goes the grand jury route; and when he wishes to provide only a little, he puts on whatever evidence he likes at the preliminary hearing, confident that if he does not secure a bindover, he'll have a chance to remedy the defect later. We don't need preliminary hearings to enable prosecutors to make voluntary disclosures to defense attorneys; they can do that anyway. In terms of the screening function, the preliminary hearing does only a little more than the grand jury. In effect, the judge offers an advisory opinion to the prosecutor. The prosecutor may yield to this advisory opinion much of the time, but he can overrule the judge by going to the grand jury. Is it worth retaining the preliminary hearing simply as make-work for judges or simply to afford them the power to offer advice to prosecutors? Shouldn't prosecutors be given legal responsibility for the power that they exercise in fact? Memorandum to the author, Jan. 17, 1984.Google Scholar

220 See table 10 items 6–8.Google Scholar

221 See table 10 item 3.Google Scholar

222 See table 10 item 1(c).Google Scholar

223 Most of the cases in table 10 items 2(a)-(f) were reinitiated with additional evidence that was in the prosecutor's possession at the time of the preliminary hearing.Google Scholar

224 See table 10 item 2(a).Google Scholar

225 See table 10 item 1(a).Google Scholar

226 This accounted for 4 cases in table 10 items 2(a)-(f).Google Scholar

227 See table 10 item 5.Google Scholar

228 See Graham & Letwin, supra note 12, at 730–31.Google Scholar

229 Id. at 731 (footnote omitted).Google Scholar

230 id. at 732.Google Scholar

231 There were 7 cases of surprise (table 10 item 2(a)). There was an additional case of inconsistent testimony, which might constitute surprise (table 10 item 2(b)). In 4 cases there was new evidence that led the prosecutor to reinitiate prosecution. See text preceding note 178 supra. In 9 cases the prosecutor disagreed with the judge's ruling (table 10 item 5). Even if we count all 9 cases in category 5 as remotely involving issues of law, this would be a total of only 21 cases, or 15.7% of all reinitiated prosecutions.Google Scholar

232 See the discussion in pt. III.D. We know that in about 26% of the 134 reinitiated cases the prosecutor proceeded to the grand jury to present evidence in his possession at the preliminary hearing but not presented for probably tactical or efficiency reasons. Specifically, there were 28 cases where the state had additional witnesses it chose not to present with the rest of its evidence at the preliminary hearing (see text accompanying note 177). There were also 7 other cases where it appears the state had physical or other evidence that it held off introducing at the hearing. See note 177. This is an estimate of the number of reinitiated cases where the state had evidence in its possession but chose not to introduce it at the preliminary hearing. In numerous other cases where there was a finding of probable cause the state had evidence it chose not to introduce. Presumably in some of those cases if the state knows it cannot remedy a discharge later by going to the grand jury, it may feel compelled to present much more of its case at the preliminary hearing.Google Scholar

233 New Mexico's statutes require, e.g., that after a grand jury no bill “the same matter shall not be presented again to that jury or another grand jury on the same evidence.” N.M. Stat. Ann. § 31-6-11.1 (Cum. Supp. 1983). The statute, however, does not prohibit the prosecutor from proceeding by information with the same evidence. For a critical discussion of New Mexico's laws on reinitiation, see Notes and Comments, supra note 9.Google Scholar

234 See note 188 supra and accompanying text.Google Scholar

235 See text accompanying notes 174 & 175.Google Scholar

236 E.g., there is no research on the effects in New Mexico of not allowing the prosecutor to resubmit a case to a grand jury on the same evidence. See note 233 supra. Has this increased the prosecutor's use of the information to reinitiate prosecutions? Nor do we know the effects in Oklahoma of possibly requiring the prosecutor to produce new evidence (not easily acquirable before the first preliminary hearing) before he can resubmit the case to a second preliminary hearing. See Notes and Comments, supra note 11. Has this resulted in making the preliminary hearing into a minitrial where the state presents most of its evidence? Or does the prosecutor now reinitiate prosecutions by indictment? Nor do we know the effects of recent legislation in Illinois that requires the prosecutor to inform the grand jury of a previous finding of no probable cause. See note 18 supra. Does the prosecutor attempt to reinitiate prosecution in fewer cases? Or does the grand jury return more no bills? Finally, in Pennsylvania where prosecution cannot be initiated or reinitiated by indictment (see note 6 supra), how frequently are prosecutions reinitiated by an information? Does judge shopping occur? Are conviction rates in reinitiated cases similar to those in probable cause cases?Google Scholar