1 Burger, Warren E., Annual Report to the American Bar Association by the Chief Justice of the United States, 67 A.B.A. J. 290, 292 (1981). For a similar but more general criticism of criminal due process developments, see Macklin Fleming, The Price of Perfect Justice (New York: Basic Books, 1974). See also note 288 infra.Google Scholar

2 E.g., Editorial, Changing Court Rules, Wall St. J., Sept. 14, 1982, at 30, col. 1 (suggesting that court rules have “tilted justice heavily against successful prosecution” and alluding to “the number of convictions … overturned on appeal”). See also the examples of political law-and-order rhetoric in Fred P. Graham, The Due Process Revolution: The Warren Court's Impact on Criminal Law 11–25 (New York: Hayden Book Co., 1970). Cf. J. Woodford Howard, Jr., Courts of Appeals in the Federal Judicial System: A Study of the Second, Fifth, and District of Columbia Circuits 51–52 (Princeton, N.J.: Princeton University Press, 1981), contrasting the actual low reversal rate in criminal appeals to “popular impressions.”.Google Scholar

3 The phrase has been popularized by Glazer, Nathan, Towards an Imperial Judiciary, 41 Pub. Interest 104 (1975); see also Donald L. Horowitz, The Courts and Social Policy (Washington, D.C.: Brookings Institution, 1977); Wasby, Stephen L., Review of Horowitz, The Courts and Social Policy, 31 Vand. L. Rev. 727 (1978).Google Scholar

4 E.g., Paul D. Carrington, Daniel J. Meador, & Maurice Rosenberg, Justice on Appeal 60 (St. Paul, Minn.: West Publishing Co., 1976); see also the works cited in note 12 infra.Google Scholar

5 Wold, John T., & Caldeira, Greg A., Perceptions of “Routine” Decision-Making in Five California Courts of Appeal, 13 Polity 334, 339, 344 (1980), quoting intermediate appellate justices who describe their work in terms of “get[ting] rid of the garbage” and dealing with “these rotten little cases.” See also Wold, John T., Going Through the Motions: The Monotony of Appellate Court Decisionmaking, 62 Judicature 58 (1978); Howard, supra note 2, at 268, 277–78.Google Scholar

6 Cf. James Eisenstein, Politics and the Legal Process 197–98 (New York: Harper & Row, 1973).Google Scholar

7 E.g., Hufstedler, Shirley M. & Hufstedler, Seth M., Improving the California Appellate Pyramid, 46 L.A. B. Bull. 275 (1971). More generally, see Wright, Charles Alan, The Doubtful Omniscience of Appellate Courts, 41 Minn. L. Rev. 751, 751 (1957) (arguing that, as Leon Green had said in 1930, “appellate courts have drawn unto themselves practically all the power of the judicial system”). The assumption of hierarchical authority is also found in many works describing court systems; e.g., James A. Cramer, Introduction, in James A. Cramer, ed., Courts and Judges 11, 12 (Beverly Hills, Cal.: Sage Publications, 1981) (“Whether the unit of analysis is a specific court, the body of courts in a particular jurisdiction, or the complete court system on the state or federal level, a logically consistent and integrated organizational structure and hierarchy exists”).Google Scholar

8 E.g., Hopkins, James D., Small Sparks from a Low Fire: Some Reflections on the Appellate Process, 38 Brooklyn L. Rev. 551, 568 (1972). Several commentators have also compared the “supervisory” role of appellate courts to that of supervision or management in large, bureaucratic organizations; e.g., Note, Courting Reversal: The Supervisory Role of State Supreme Courts, 87 Yale L. J. 1191, 1193 (1978) [hereinafter cited as Yale Note], stating that “judicial supervision resembles the supervision that takes place in other large organizations” but noting that judicial supervisory communications and the scope of judicial review are subject to “severe constraints”; Dean J. Champion, The Organization of Trial Judges, in Cramer, ed., supra note 7, at 59, 67, asserting: “In many respects, judicial organization from the Supreme Court to the lowest municipal tribunal resembles closely the bureaucracy of the business world.” I think that the comparison to bureaucratic supervision tends to considerably overstate the intensity of appellate supervision over the trial courts and that it is not a useful analogy. Cf. James Eisenstein & Herbert Jacob, Felony Justice: An Organizational Analysis of Criminal Courts 9 (Boston: Little, Brown & Co., 1977), arguing that it is not appropriate to view trial courts as bureaucratic organizations. Compare Baum, Lawrence, Implementation of Judicial Decisions: An Organizational Analysis, 4 Am. Pol. Q. 86, 89–91 (1976), arguing for a more sophisticated organizational model in which “lower-court judges” are treated “as independent actors, who will not follow the lead of higher courts unless conditions are favorable for their doing so.”.Google Scholar

9 See notes 27, 30 infra.Google Scholar

10 Cf. Hopkins, James D., The Role of an Intermediate Appellate Court, 41 Brooklyn L. Rev. 459, 464, 473 (1975).Google Scholar

11 The research described in this article is also discussed in Thomas Y. Davies, Organizational Behavior and the Distribution of Case Outcomes in a California Court of Appeal (Ph.D. diss., Northwestern University, 1980) (Ann Arbor, Mich.: University Microfilms International, 1980). A number of significant changes have occurred in the First District since the data for this study were collected. Some of the justices interviewed for the study have retired and the number of justices on the court has been increased (see note 44 infra). In addition, there have been some important changes in substantive law, especially regarding sentencing (see notes 83–84 infra), and the reversal rate may have increased slightly (see note 141 infra). See also notes 39, 311, 313 infra.Google Scholar

12 E.g., Daniel J. Meador, Appellate Courts: Staff and Process in the Crisis of Volume (St. Paul, Minn.: West Publishing Co., 1974 & Supp. 1975); Carrington et al., supra note 4; Robert A. Leflar, Internal Operating Procedures of Appellate Courts (Chicago: American Bar Foundation, 1976); B. E. Witkin, Manual on Appellate Court Opinions (St. Paul, Minn.: West Publishing Co., 1977); Stephen L. Wasby, Thomas B. Marvell, & Alexander B. Aikman, Volume and Delay in State Appellate Courts: Problems and Responses, Pub. No. R0048 (Williamsburg, Va.: National Center for State Courts, Appellate Justice Project, 1979).Google Scholar

13 For a similar assessment, see Shapiro, Martin, Appeal, 14 Law & Soc'y Rev. 629, 630–31 (1980). The judicial administration literature does discuss certain procedural “imperatives of appellate justice,” but these do not address the purposes of appellate review itself; e.g., Carrington et al., supra note 4, at 8–12.Google Scholar

14 E.g., Carrington et al., supra note 4, at 60, 91–96. For a discussion of the introduction of expedited processing techniques in the California Court of Appeal for the First Appellate District, see Davies, Thomas Y., Gresham's Law Revisited: Expedited Processing Techniques and the Allocation of Appellate Resources, 6 Just. Sys. J. 372 (1981).Google Scholar

15 For a brief summary of the legal realist critique, see Theodore L. Becker, Political Behavioralism and Modern Jurisprudence: A Working Theory and Study in Judicial Decision-Making 40–66 (Chicago: Rand McNally & Co., 1964). For a more extensive treatment, see William Twining, Karl Llewellyn and the Realist Movement (London: Weidenfeld & Nicolson, 1973).Google Scholar

16 Karl N. Llewellyn, The Common Law Tradition: Deciding Appeals 19–61 (Boston: Little, Brown & Co., 1960).CrossRefGoogle Scholar

17 Howard, supra note 2.Google Scholar

18 E.g., C. Herman Pritchett, The Roosevelt Court: A Study in Judicial Politics and Values, 1937–1947 (New York: Macmillan Co., 1948).Google Scholar

19 E.g., Glendon Schubert, The Judicial Mind Revisited: Psychometric Analysis of Supreme Court Ideology (New York: Oxford University Press, 1974).Google Scholar

20 E.g., Walter F. Murphy, Elements of Judicial Strategy (Chicago: University of Chicago Press, 1964).Google Scholar

21 States with some type of intermediate appellate court, in descending order by population according to their rank in the 1980 Census, are as follows: California (1), New York (2), Texas (3), Pennsylvania (4), Illinois (5), Ohio (6), Florida (7), Michigan (8), New Jersey (9), North Carolina (10), Massachusetts (11), Indiana (12), Georgia (13), Missouri (15), Wisconsin (16), Tennessee (17), Maryland (18), Louisiana (19), Washington (20), Alabama (22), Kentucky (23), Oklahoma (26), Iowa (27), Colorado (28), Arizona (29), Oregon (30), Kansas (32), Arkansas (33), New Mexico (37), Hawaii (39), Idaho (41), and Alaska (50). The largest states without intermediate appellate courts are Virginia (14) and Minnesota (21). Virginia has a unique system in which virtually all appeals are discretionary to the Virginia Supreme Court. See Thomas R. Morris, The Virginia Supreme Court: An Institutional and Political Analysis (Charlottesville: University Press of Virginia, 1975). Minnesota recently considered a proposal for an intermediate appellate court. See Lawscope: The Judiciary, 67 A.B.A. J. 1248 (1981).Google Scholar

22 This article does not discuss the characteristics or processes of state supreme courts in the single-tier appellate systems in the smaller states; henceforth, all references to “supreme courts” are to those in two-tier appellate systems. For a description of a state supreme court in a single-tier system, see Beiser, Edward N., The Rhode Island Supreme Court: A Well-Integrated Political System, 8 Law & Soc'y Rev. 167 (1974 Google Scholar

23 For descriptions of the caseload strains leading to the creation of state intermediate appellate courts around the turn of the century, see generally Roscoe Pound, Organization of Courts (Boston: Little, Brown & Co., 1940). For the relationship between state population and the creation of state intermediate appellate courts, see Davies, supra note 11, at 22–25; Fair, Daryl R., State Intermediate Appellate Courts: An Introduction, 24 W. Pol. Q. 415 (1971); Kagan et al. Business of State Supreme Courts, infra note 28, at 130–31.Google Scholar

24 For a description of the subject matter allocation of appellate jurisdiction that was common in early two-tier appellate systems, see Pound, supra note 23, at 225–41. Note, however, that the actual distribution of appellate work did not always accord with the formal categories of jurisdiction. In California, the Supreme Court used its authority to transfer cases between itself and the courts of appeal to establish a de facto discretionary review process long before the 1966 revision of the constitution which formally adopted a largely discretionary jurisdiction for the Supreme Court. See Cal. Jud. Council, 1st [Biennial] Rep. 21 (1927) (noting that the Supreme Court was “compelled to assign to the District Courts of Appeal practically all the cases within its own original appellate jurisdiction”); Gibson, Phil S., The California Constitution and Its Judicial Article, 29 S. Cal. L. Rev. 389, 394 (1956); Taylor, B. Grant, Supreme Court Clears Its Calendar, 17 J. St. B. Cal. 157 (1942); Davies, supra note 11, at 115–18.Google Scholar

25 For a description of the allocation of appellate jurisdiction in states having two-tiered appellate systems, see Marlin O. Osthus, State Intermediate Appellate Courts (Chicago: American Judicature Society, 1976, & rev. ed. by Mayo H. Stiegler, 1980). Note that the American Bar Association, Commission on Standards of Judicial Administration, Standards Relating to Appellate Courts §§ 3.00-.11 & commentary (Approved Draft 1977), calls for giving a discretionary jurisdiction to the supreme court in a two-tier appellate system.Google Scholar

26 E.g., in California, appeals from judgments imposing a death sentence are within the original appellate jurisdiction of the Supreme Court under art. 6, § 11, of the California Constitution, and are automatically appealed to that court under § 1239 of the Penal Code (West 1970 & Supp. Pamph. 1981). Similarly, review of the decisions of the Public Utility Commission is by the Supreme Court under § 1756 of the Pub. Util. Code (West 1975). The Supreme Court also exercises exclusive jurisdiction over cases involving judicial or attorney discipline.Google Scholar

27 Being the most populous state, California presents an extreme picture of the shift of the appellate caseload to the intermediate appellate courts. In fiscal 1976 the five courts of appeal decided 5,943 appeals and writ petitions by written opinions (1977 Cal. Jud. Council Ann. Rep. 188) and disposed of another 3,448 writ petitions without opinion (id. at 251 table 5). During the same year 2,894 petitions for hearing were filed in the Supreme Court (id. at 179). The Supreme Court granted only 229 petitions (id. at 181) and decided only 191 cases by written opinion (id. at 179). The Supreme Court thus agreed to rehear only about 2 to 4 percent of the decisions of the courts of appeal, depending on whether one includes the writ petitions summarily denied in the courts of appeal. In fiscal 1976 the five courts of appeal decided 2,896 criminal appeals with opinion (id. at 251 table 5); in the same year the Supreme Court agreed to review 35 of the 1,077 criminal appeals it was petitioned to hear—3.2 percent of its petitions, or 1.2 percent of criminal appeals decided in the courts of appeal (id. at 182 table V).Google Scholar

28 For the most extensive research on the evolution of state supreme courts, see three articles reporting the findings of a study of 16 state supreme courts conducted by Kagan, Robert A., Cartwright, Bliss, Friedman, Lawrence M., and Wheeler, Stanton: Kagan et al., The Business of State Supreme Courts, 1870–1970, 30 Stan. L. Rev. 121 (1977); id., The Evolution of State Supreme Courts, 76 Mich. L. Rev. 961 (1978); and Friedman, Lawrence M. et al., State Supreme Courts: A Century of Style and Citation, 33 Stan. L. Rev. 773 (1981). Several other studies have also suggested that the creation of an intermediate appellate court tends to increase policy articulation in the state supreme court: Canon, Bradley C. & Jaros, Dean, External Variables, Institutional Structure and Dissent on State Supreme Courts, 4 Polity 175 (1970); Atkins, Burton M. & Glick, Henry R., Environmental and Structural Variables as Determinants of Issues in State Courts of Last Resort, 20 Am. J. Pol. Sci. 97 (1976); Groot, Roger D., The Effects of an Intermediate Appellate Court on the Supreme Court Work Product: The North Carolina Experience, 7 Wake Forest L. Rev. 548 (1971); Project, The Effect of Court Structure on State Supreme Court Opinions: A Re-examination, 33 Stan. L. Rev. 951 (1981). See also Hopkins, supra note 10, at 462 (the creation of an intermediate appellate court “releases the highest court to address itself solely to the determination of questions of law, with a particular view toward the development of the law as a whole”).Google Scholar

29 One California Supreme Court justice interviewed during this study described that court's decisions on petitions for hearing as follows: The test is that we don't take the case over [i.e., grant a petition for hearing] unless either the result is wrong or the language is so bad that we don't think that it should remain on the books. See also Traynor, Roger J., Some Open Questions on the Work of State Appellate Courts, 24 U. Chi. L. Rev. 211, 214–15 (1957). The result of this monitoring approach is that a large percentage of the decisions of supreme courts in two-tier appellate systems reverse or otherwise disagree with a decision below. See Kagan et al., Evolution of State Supreme Courts, supra note 28, at 995. A similar pattern is found in the federal system; see Howard, supra note 2, at 59, noting that the United States Supreme Court disturbed more than two-thirds of the decisions it reviewed in contrast to the federal courts of appeals which affirmed two-thirds of the decisions reviewed. For studies of the factors affecting the selection of cases for hearing in the California Supreme Court, see three articles by Baum, Lawrence: Decisions to Grant and Deny Hearings in the California Supreme Court: Patterns in Court and Individual Behavior, 16 Santa Clara L. Rev. 713 (1976); Policy Goals in Judicial Gatekeeping: A Proximity Model of Discretionary Jurisdiction, 21 Am. J. Pol. Sci. 13 (1977); Judicial Demand—Screening and Decisions on the Merits: A Second Look, 7 Am. Pol. Q. 109 (1979).Google Scholar

30 Kagan et al., Business of State Supreme Courts, supra note 28, at 131–32, 156, note that the number of full opinions issued by state supreme courts may actually decline following the introduction of an intermediate appellate court. The average length of opinion may tend to increase slightly, however. Kagan et al., Evolution of State Supreme Courts, supra note 28, at 991, 999.Google Scholar

31 See Kagan et al., Evolution of State Supreme Courts, supra note 28, at 987–90; id., Business of State Supreme Courts, supra note 28, at 145–50; Davies, supra note 11, at 120–22.Google Scholar

32 Kagan et al., Business of State Supreme Courts, supra note 28, at 148 table 2, indicate that in the period 1940–70 criminal cases made up between 9.5 percent and 31.0 percent of the cases decided by state supreme courts. If these percentages are applied to a caseload of 200 cases they result in the estimated range of 20 to 60 criminal cases per year.Google Scholar

33 I prefer the term law articulation because it covers the full range of appellate activity from interstitial interpretation to creative law making.Google Scholar

34 E.g., Howard, supra note 2, at 7; Carrington et al., supra note 4, at 2–3; Hopkins, supra note 10, at 460; Hufstedler & Hufstedler, supra note 7, at 278–79, 296. Shirley M. Hufstedler, then a federal court of appeals judge, described the intermediate appellate function as follows: Intermediate appellate courts have functions very different from either trial courts or courts of last resort. Their duties are a mixture of error correction in the individual case and institutional functions, which include supervising lower courts, filling interstitial spaces in statutory and case law, and occasionally striking out a few paces on a new jurisprudential path. Broad-gauge policy making is seldom a part of these courts' institutional concerns. Hufstedler, , The Appellate Process Inside Out, 50 Cal. St. B. J. 20, 23 (1975). Compare this description to the text accompanying notes 159–71 infra.Google Scholar

35 Hurst, James Willard, The Functions of Courts in the United States, 1950–1980, 15 Law & Soc'y Rev. 401, 446–49 (198081).Google Scholar

36 For a history of the creation and organizational evolution of the California courts of appeal, see 1973 Cal. Jud. Council Ann. Rep. 23–25.Google Scholar

37 Meador, supra note 12, at 7–12, indicates that the First District was the first state intermediate appellate court to combine a screening process with the operation of a central staff; he traces the introduction of screening to the United States Court of Appeals for the Fifth Circuit in 1968 and the use of a central staff to the Michigan Court of Appeals in the same year.Google Scholar

38 The California courts of appeal have been the subject of several studies: Wold & Caldeira, supra note 5; Wold, supra note 5; National Center for State Courts, The California Courts of Appeal (Full Report), Pub. No. R0013 (Denver: National Center for State Courts, 1974); Meador, supra note 12, at app. C; Ronald H. Beattie, Criminal Appeals in California, 1964–1968 (Sacramento, Cal.: Department of Justice, Division of Law Enforcement, Bureau of Criminal Statistics, n.d.); id., Report on California Criminal Appeals, 1929–1935, in Cal. Jud. Council, 6th [Biennial] Rep., app. (1936); id., Criminal Appeals in California, 24 Cal. L. Rev. 623 (1936). See also the following articles by California appellate judges: Gustafson, Roy A., Some Observations About California Courts of Appeal, 19 UCLA L. Rev. 167 (1971); Christian, Winslow, Delay in Criminal Appeals: A Functional Analysis of One Court's Work, 23 Stan. L. Rev. 676 (1971); Thompson, Robert S., Mitigating the Damage—One Judge and No Judge Appellate Decisions, 50 Cal. St. B.J. 476 (1975); id., Selection of Judges of the California Court of Appeal, 48 Cal. St. B.J. 381 (1973); Molinari, John B., The Decisionmaking Conference of the California Court of Appeal, 57 Calif. L. Rev. 606 (1969); Peters, Raymond E., Five Years of the Conference System, 19 J. St. B. Cal. 399 (1944)Google Scholar

39 For a map of the appellate districts of California as they existed at the time of the study, see Wold, supra note 5, at 63. Recent legislation has created a sixth appellate district in California which includes several counties formerly in the First District. See 1982 Cal. Jud. Council Ann. Rep. 22–23.Google Scholar

40 Cal. Const, art. 6, § 14, requires: “Decisions of the Supreme Court and courts of appeal that determine causes shall be in writing with reasons stated.”.Google Scholar

41 Although there are legal rules that apply to the use of writ petitions, the justices interviewed described the potential scope of review by writ as being quite broad. One said: In thinking about the writ system, you have to realize that discretionary review is what it really amounts to. Realistically, anything can be handled by a writ if the court wants to handle it. See also Fowler, Carlo S., Mandamus as an Original Proceeding in the California Appellate Courts, 15 Hastings L.J. 177 (1963).Google Scholar

42 Since the summary denial of a writ petition without opinion is not viewed as the “determination” of a cause, it is viewed as falling outside the constitutional requirement for written opinions cited in note 40 supra. As one justice put it, “If we summarily deny [a writ petition], that is the end of it and there is no opinion.” However, the formal distinction between the decision whether to hear the merits of a petition (i.e., whether to issue an alternative writ or show cause order) and the decision on the merits of petitions accepted for review tends to blur in practice. Most petitions that the court agrees to hear are ultimately granted—so the decision to hear petitions is, in substantial part, a tentative decision on the merits. See Davies, supra note 11 at 294–96.Google Scholar

43 The justices of the Court of Appeal were appointed by the governor and subject to nonopposed retention election every 12 years. See generally Thompson, Selection of Judges, supra note 38. By and large, the justices exhibited strong ties to California; 8 were born in the state, and all but 1 went to law schools in the San Francisco area. While all of the justices appear to have had some experience in private law practice early in their professional careers, most also had been either a prosecutor (5) or had held some other public office (2 as state legislators, 1 as mayor, 1 as executive secretary to the governor, and 1 as inheritance tax commissioner) prior to judicial appointment. Ten of the 12 justices had been superior court judges, with a median of 7.5 years of service, and 1 other had been a municipal court judge. With regard to political affiliation, 8 labeled themselves as Republicans, 2 as Democrats, and 2 gave no party identification. See Davies, supra note 11, at 36–39. The high percentage of justices with prior trial court judging experience and with experience as prosecutors appear to be common features of the intermediate appellate bench. Berg, Larry L. et al., The Consequences of Judicial Reform: A Comparative Analysis of the California and Iowa Appellate Systems, 28 W. Pol. Sci. Q. 263, 264–65, 276 & tables 4 & 5 (1975), report that 82.2 percent of California court of appeal justices between 1934 and 1971 had prior judicial experience and 36.7 percent had been prosecutors.Google Scholar

44 Because justices are permanently assigned to one of the four divisions, it is not unreasonable to view the First District as four courts. During interviews, the justices remarked that there was little interchange between the four divisions with regard to their cases or procedures. Moreover, there is no provision for the First District to sit en banc. An additional justice was authorized for each of the four divisions in 1976, but those positions were not filled immediately. 1980 Cal. Jud. Council Ann. Rep. 62 table XI at note a. More recently, an additional division has been authorized. 1982 Cal. Jud. Council Ann. Rep. 22–23.Google Scholar

45 See text accompanying notes 274–84 infra. See also Davies, supra note 14.Google Scholar

46 E.g., Meador, supra note 12; Carrington et al., supra note 4, at 4–5.Google Scholar

47 See the comparison of criminal and civil appeal filings in California in Davies, supra note 14, at 384 fig. 1. See also 1982 Cal. Jud. Council Ann. Rep. 52 fig. 1.Google Scholar

48 Carrington et al., supra note 4, at 60, 91–93. Cf. Wold & Caldeira, supra note 5, at 335–36, 338–39.Google Scholar

49 372 U.S. 353.Google Scholar

50 386 U.S. 738.Google Scholar

51 E.g., 1970 Cal. Jud. Council Ann. Rep. 35–36. See also comments justices made in interviews quoted in Davies, supra note 11, at 333, 338, 339–40.Google Scholar

52 E.g., the annual reports of the California Judicial Council have reported an increase in the rate of appeal from “contested superior court disposition” of criminal appeals from 35.7 percent in fiscal 1968 (the first year for which the data is reported) to 90.2 percent in fiscal 1981: 1975 Cal. Jud. Council Ann. Rep. 74 table VII, 1982 Cal. Jud. Council Ann. Rep. 52 table VII. The latter source also shows that the rate of appeal from “convictions after contested trial” has increased from 77.3 percent in fiscal 1976 (the first year for which the data is reported) to 110.3 percent in fiscal 1981. As indicated below, these attempts to measure the rate of appeal produce an exaggerated image of the supposed increase in the tendency of convicted criminal defendants to appeal.Google Scholar

53 The principal source for comparison is Beattie, Criminal Appeals in California, 1964–1968, supra note 38, for information on several characteristics of criminal appeals in California during the years of that study. See notes 85, 95 infra.Google Scholar

54 Fiscal year 1974 was chosen for the study because it was the most recent fiscal year for which substantially all filed appeals had been disposed of when data collection began in early 1975.Google Scholar

55 One criminal appeal was excluded from my data collection because no decision had been rendered by the time data collection was finished, People v. Corona, 1 Crim. 12401. (All case references beginning with “1 Crim.” are to the case numbers in the First District's criminal registers.) This appeal involved a conviction for the murder of 25 field workers that had received a great deal of national and local press coverage. Subsequently, the Court of Appeal reversed the conviction on the ground that Corona had been denied adequate representation because his attorney had interests in publication rights relating to the case. People v. Corona, 80 Cal. App. 3d 684, 145 Cal. Rptr. 894 (1st Dist. 1978). Corona has been reconvicted after a 7-month retrial and 10 days of jury deliberations. See Katherine Bishop, Corona Convicted Again in 25 Farm Worker Slayings, N.Y. Times, Sept. 24, 1982, at 9, col. 1. I chose not to add the Corona appeal to the study data for two reasons. First, it was not practical to review the case file so I could not collect data in the same manner as for other cases. Second and more important, I had learned in interviews that the Corona appeal had received an extraordinary amount of attention and resources; for that reason, it was unique and quite outside the typical variations in appeal processing. Including such an extreme case would not advance the purpose of describing the usual meaning of error correction. See also note 254 infra.Google Scholar

56 My classification of writ petitions by the subject matter of the underlying case differs from the classification used by the Administrative Office of the Courts in California. The latter classifies only postconviction petitions (e.g., habeas corpus) as “criminal”; all pretrial writ petitions are classified as “civil original proceedings” and filed in the civil registers—regardless of whether they relate to civil litigation or to criminal prosecutions—because they take the form of civil suits against the superior court. Fortunately, however, the subject of the pretrial writ petitions can be determined from register entries showing the subject matter of the case (criminal matters were designated as “995,”“1538.5” or “X”; see note 71 infra). In addition, the registers record that the real party in interest in criminal defendant petitions is the “People of the State of California.”.Google Scholar

57 Prosecution writ petitions were identified as described in note 56 supra except, of course, that the people were identified as petitioner.Google Scholar

58 To conserve research resources, I randomly chose 25 of every 50 sequential civil appeals or civil writ petitions. While not a truly random sample, this procedure should have eliminated any potential sources of sampling bias. To preserve the sense of the relative volume of civil and criminal matters, the discussions of civil appeals in the text and tables are estimated for the entire population by doubling the observed distribution of cases. Of course, this estimation process introduces some unknown degree of sampling error.Google Scholar

59 Because review of criminal appeal case files proved quite time consuming, I limited my review of case files in affirmed criminal appeals to a randomly chosen set of 109 of the 468 affirmed appeals. Of course, this means that there is a substantial possibility of some degree of sampling error. In order to analyze the relationships between various factors and dispositions, I have weighted the affirmed cases so that they can be used to estimate the distribution of the full set of affirmed cases. In other words, the raw numbers of affirmed cases have been multiplied by 4.29 in order to create an estimate of the distribution of affirmed appeals that can be compared with the data regarding the 26 reversed and 50 modified appeals. All statistics derived by this procedure are designated as estimations in the tables. Note, however, that data was collected for all affirmed apeals with regard to characteristics that could be coded from the registers.Google Scholar

60 For more information about prosecution appeals and civil appeals, see Davies, supra note 11, at 254–70, 270–88.Google Scholar

61 Generally, the court clerks were directed to identify appeals in which the time for filing a brief had expired or other appeals that failed to comply with standards for appeal, and those were then routinely dismissed by order. But see note 80 infra for an example of a more fluid decision regarding dismissal that occurs in a small number of appeals from guilty plea convictions.Google Scholar

62 Technically, the failure of an appellant to complete the procedure for requesting counsel resulted in a dismissal based on appellant's failure to file a brief within the required time, pursuant to Rule 17(a) of the California Rules of Court (Cal. Civ. & Crim. Ct. R. 17(a)). See Davies, supra note 14, at 386–92. For a more comprehensive discussion of dismissals of criminal appeals in the First District, see Davies, supra note 11, at 344–48.Google Scholar

63 The California Judicial Council's annual reports also exclude dismissed criminal appeals from the discussion of outcomes in criminal appeal decisions.Google Scholar

64 A variety of decisions must be made about how to define an “appeal” for quantitative analysis. The large majority of appeals present no difficulty since they involve one appellant and one criminal incident. In a few instances, however, there were multiple appellants who had been tried together. Because I was primarily interested in the decisions of the court, I treated a multiple-appellant case as a single appeal. In a few other instances it was apparent that sequentially numbered entries in the court register had in fact been treated as a single case by the court; that is, they involved appeals from codefendants who had been tried together. Because it was apparent that such entries had been assigned, briefed, argued, and decided together, I also treated them as single appeals. As a practical matter, those choices had a de minimis effect on the statistical analysis. For a general discussion of the difficulties encountered in defining a “case” for empirical analysis, see Cartwright, Bliss, Conclusion: Disputes and Reported Cases, 9 Law & Soc'y Rev. 369 (1975).Google Scholar

65 Under a 1959 constitutional amendment, decisions of the appellate departments were subject to discretionary review by the courts of appeal if the appellate department certified that the case involved an important legal issue. See Cal. Jud. Council, 18th Biennial Rep. 105–14 (1961). As a practical matter, however, the largest number of such cases reviewed by the courts of appeal was 18 in 1968; thereafter, the number of appellate department decisions accepted for review declined. See Davies, supra note 11, at 364–65. I identified only one municipal court decision that Court of Appeal agreed to review in the fiscal 1974 cases, a drunk-driving conviction that resulted in one of the three reversals that reinstated probation.Google Scholar

66 In a submission on transcript (SOT) proceeding, the case is submitted to a judge for decision on the basis of the transcript from the preliminary hearing. This type of proceeding was used extensively in Los Angeles and to a much lesser extent in San Francisco in the early 1970s. See Herbert Jacob, Urban Justice: Law and Order in American Cities 109 (Englewood Cliffs, N.J.: Prentice-Hall, 1973).Google Scholar

67 See note 120 infra.Google Scholar

68 See text accompanying notes 93–95 infra.CrossRefGoogle Scholar

69 Lynn M. Mather, Plea Bargaining or Trial? The Process of Criminal-Case Disposition 27, 60–63, 98 (Lexington, Mass.: D.C. Heath & Co., Lexington Books, 1979).Google Scholar

70 See Cal. Penal Code § 1237 (West 1970 & Supp. Pamph. 1981).Google Scholar

71 In all, 355 pretrial writ petitions were filed by criminal defendants in fiscal 1974, but only 16 (4.5%) were granted. Of those, 3 involved review of motions to suppress evidence under Cal. Penal Code § 1538.5 (West Supp. Pamph. 1981), and 5 involved review of motions to quash prosecution under § 995 of the Penal Code (West 1970). If there were any petitions granting bail, they would have to be found in the remaining 8 granted petitions, which were classified only as “X” in the registers. A writ attorney for one division who worked on pretrial writ petitions stated that in two years he had seen only one petition granted involving pretrial bail.Google Scholar

72 The enactment of § 1237.5 appears to have been a legislative response to the Warren Court's 1963 decision in Douglas v. California, 372 U.S. 353. Under Douglas, indigent appellants were entitled to appointed counsel in all appeals of right. By making appeal from guilty pleas conditional, § 1237.5 eliminated the need to appoint counsel in attempted appeals from guilty plea convictions. There does appear to have been a decline in the number of appeals from guilty plea convictions following the effective date of § 1237.5 in the fall of 1965. Beattie, Criminal Appeals in California, 1964–1968, supra note 38, at 7, 16 table IV, reports that of the 2,214 criminal appeals filed in the state in fiscal years 1965 and 1966 combined, 297 (13.4%) were from guilty plea convictions but that of the 3,072 criminal appeals filed in fiscal years 1967 and 1968, only 240 (7.8%) were. A similar restriction on the scope of Douglas was implemented by the California Supreme Court in 1965 in People v. Shipman, 62 Cal. 2d 226, 397 P.2d 993, 42 Cal. Rptr. 1, which held that orders rejecting error coram nobis petitions were not appealable of right. See Davies, supra note 11, at 359 n.43, 361 n.45. The Burger Court has explicitly declined to extend the right to appointed counsel to discretionary appeals; see Ross v. Moffitt, 417 U.S. 600 (1974).Google Scholar

73 See also Cal. Civ. & Crim. Ct. R. 31(d).Google Scholar

74 There were 12 appeals from guilty plea convictions filed in fiscal 1974 that were dismissed for lack of certification, and there were other unsuccessful attempts to appeal; see Davies, supra note 11, at 216 n.18, 221–22 & n.25; Comment, Remedies for Reneged Plea Bargains in California, 16 Santa Clara L. Rev. 103 (1975). For a general discussion of judicial supervision of the plea process, see Abraham S. Goldstein, The Passive Judiciary: Prosecutorial Discretion and the Guilty Plea (Baton Rouge: Louisiana State University Press, 1981); Comment, Judicial Supervision over California Plea Bargaining: Regulating the Trade, 59 Calif. L. Rev. 962 (1971).Google Scholar

75 Under § 1538.5, criminal defendants can seek appellate review of denials of motions to suppress evidence either by pretrial writ petitions before conviction or by appeal after conviction. This section also allows the prosecution to file a writ petition to challenge an order suppressing evidence. This opportunity is especially important to the prosecution since the prohibition against double jeopardy might prevent the state from challenging an order suppressing evidence on appeal. In fact, in the fiscal 1974 cases, criminal defendants were successful in only 3 writ petitions involving search and seizure issues (3% of 99 petitions) but the prosecution was successful in 7 (53.8% of 13). See also note 286 infra.Google Scholar

76 Mather, supra note 69, at 55. See also Gilboy, J. A., Guilty Plea Negotiations and the Exclusionary Rule of Evidence: A Case Study of the Chicago Narcotics Courts, 67 J. Crim. L. & Criminology 89 (1976).Google Scholar

77 395 U.S. 238 (1969).Google Scholar

78 The California Supreme Court implemented Boykin in In re Tahl, 1 Cal. 3d 122, 460 P.2d 449, 81 Cal. Rptr. 577 (1969), and extended the application of the Boykin standards to submission on transcript proceedings, in In re Mosley, 1 Cal. 3d 913, 464 P.2d 473, 83 Cal. Rptr. 809 (1970).Google Scholar

79 For an example of the transcript of a typical guilty plea ceremony, see Eisenstein & Jacob, supra note 8, at 227–30. For an empirical study of the degree of compliance with the Boykin standards for plea waivers, see James A. Cramer, Judicial Supervision of the Guilty Plea Hearing, in Cramer, ed., supra note 7, at 173, reporting that most such hearings are perfunctory and that the degree of compliance varies considerably.Google Scholar

80 One justice gave this description of how the Court of Appeal decides whether to treat an appeal from a guilty plea as properly filed, regardless of the lack of certification by the trial judge: To tell if the notice of appeal is operative, we look at the record to see if issues are being raised about section 1538.5 or if orders entered after judgment are involved. If there is a [motion to suppress under section] 1538.5, then our practice is to construe the notice of appeal broadly at that point and treat it as being operative—even if the wording is incorrect—as to that search and seizure issue, and the court informs the party of that in its order. Lawyers will continue to raise Boykin issues anyway in such cases—I guess out of lack of knowledge. When a Boykin issue is raised in the brief, the court will consider it and usually will decide it on the merits, because by doing so we think that it becomes the “law of the case” so it can't be collaterally attacked on habeas corpus. So we avoid the later attack by meeting it now—if we dismissed [the appeal on that issue] then it could be subjected to habeas corpus review. But if the Boykin issue is the only issue raised, then we usually dismiss the apppeal. If the Boykin issue turns out to have merit, however—if it is a good ground for appeal—then the court can consider it and act on it despite the notice of appeal problem because this can be considered a constitutional issue, and as such it can be raised at any time. In addition, National Center for State Courts, supra note 38, at 43, notes that the plea conviction appeal certification standards were inconsistently applied by the five courts of appeal.Google Scholar

81 For a brief description of the California indeterminate sentencing process in effect at the time of this study, see Mather, supra note 69, at 28–30.Google Scholar

82 See the discussion of modifications of sentences infra.Google Scholar

83 A “determinate sentencing” law became effective in California July 1, 1977. See 1977 Cal. Jud. Council Ann. Rep. 154. It should be noted that this change in the law may have altered the dynamics of plea bargaining somewhat—and thus may have affected the flow of cases to the Court of Appeal. See generally the discussion of the effects of the determinate sentencing law in 1982 Cal. Jud. Council Ann. Rep. 3–9. This source indicates that a number of appeals now raise sentencing issues, id. at 5.Google Scholar

84 1981 Cal. Jud. Council Ann. Rep. 74 indicates that there has been an increase in appeals from guilty pleas in the aftermath of the enactment of the determinate sentencing laws. Such appeals may be more common in other states; see, e.g., Hopkins, supra note 8, at 560–62, noting in 1972 that appeals from guilty pleas raising sentencing issues were common in New York. But see Carrington et al., supra note 4, at 97, commenting that: In the American judicial process there is no greater anomaly than the widespread lack of availability of review of trial court decisions fixing sentences. The sentence is the most important aspect of most cases—to the defendant and to society. Yet in the majority of American states, as well as in the federal courts, there is little or no room for appellate scrutiny of the sentencing decision if the sentence does not exceed the statutory maximum.Google Scholar

85 Data reported in Beattie, Criminal Appeals in California, 1964–1968, supra note 38, at 16 table IV, allow computation of the percentage of appeals resulting from the various modes of conviction. Guilty pleas accounted for 13.4 percent of all appeals in fiscal years 1965–66 and 7.8 percent in fiscal years 1967–68 (see also id. at 8); submission on transcript proceedings accounted for 12.5 percent and 18.7 percent in the two periods respectively; bench trials accounted for 27.0 percent and 22.3 percent; and jury trials accounted for 47.1 percent and 51.2 percent. Thus, appeals have become even more concentrated in jury trial convictions over time.Google Scholar

86 Issue “topics” rather than specific issues were counted because appeals may raise several closely related or overlapping issues that relate to a single topic such as alleged illegal search. There are multiple topics in many appeals, hence the sum of topics exceeds the number of appeals.Google Scholar

87 Commentators have often been critical of appeals raising only sufficiency of the evidence issues. E.g., Hopkins, supra note 8, at 555. In the study, however, appeals raising only insufficiency of evidence issues were infrequent. Moreover, while insufficiency of the evidence seldom leads to a reversal, the strength of the evidence can be very important in the application of the harmless error rule, as discussed infra. See especially note 232 infra.Google Scholar

88 Mather, supra note 69, at 27, 33, 53–54.Google Scholar

89 Offenses were coded from the trial court's registers. In cases of conviction for more than one offense, interviewees indicated that the court's practice was to list on the register the most serious offense the defendant was convicted of.Google Scholar

90 Mather, supra note 69, at 101–2.Google Scholar

91 Id. at 33, 63–64.Google Scholar

92 See Davies, supra note 11, at 233–34.Google Scholar

93 In several attempts to appeal from guilty plea convictions that I reviewed, the dispute appeared to center on the defendant's claim that he had a plea bargain for county time (jail) but had been sentenced to prison. See also Milton Heumann, Plea Bargaining: The Experiences of Prosecutors, Judges, and Defense Attorneys 103–9 (Chicago: University of Chicago Press, 1977); Mather, supra note 69, at 94.Google Scholar

94 Adult Prosecution Program Report, 1973, at 3 (Sacramento, Cal.: Department of Justice, Division of Law Enforcement, Bureau of Criminal Statistics, n.d.). Of the remaining defendants convicted in superior courts, 70 percent received probation and 7 percent jail terms, id. at 33 table 14.Google Scholar

95 Compare Beattie, Criminal Appeals in California, 1964–1968, supra note 38, at 20 table VIII, indicating that in calendar years 1964 through 1967 there were 6,144 appeals. Of those, 4,501 (73.3%) involved prison sentences, while an additional 406 (6.6%) involved commitments to CRC, DMH, or the Youth Authority; only 1,237 (20.1%) involved only probation or jail sentences.Google Scholar

96 Although it may appear counterintuitive, research has disclosed that many, perhaps most, of the felony prosecutions that result in jury trials involve “dead bang,” serious cases in which there is little or no basis for doubting factual guilt. Mather, supra note 69, at 41–42, found that few prosecutions in the Los Angeles superior court were “reasonable doubt” cases in which there was a chance of acquittal. Most cases were either dead bang cases in which there was no basis for doubt about the defendant's guilt, or “overfiled” cases in which there was a dispute only about the level of seriousness of the offense. Mather argues that the structure of incentives in plea bargaining are such that the dead bang, serious cases are among the most likely to be tried simply because they do not provide any basis for bargaining, id. at 60–64.Google Scholar

97 See generally Jerome H. Skolnick, Justice Without Trial: Law Enforcement in Democratic Society (2d ed. New York: John Wiley & Sons, 1975); Herbert L. Packer, The Limits of the Criminal Sanction (Stanford, Cal.: Stanford University Press, 1968).Google Scholar

98 The indicators of the strength of the evidence of guilt employed in this analysis are essentially the same as those used by court personnel to estimate the likelihood of conviction; see Mather, supra note 69, at 42–43. See also Eisenstein & Jacob, supra note 8, at 182–83, for a similar approach to measuring the strength of the evidence in criminal prosecutions.Google Scholar

99 See also Seymour Wishman, Confessions of a Criminal Lawyer 10, 16, 41–42, 221 (New York: Times Book Co., 1981).Google Scholar

100 E.g., see note 52 supra.Google Scholar

101 From 1905 to 1930 criminal appeal filings were tabulated from the reports on criminal appeals in the biennial reports of the California Attorney General; from 1931 to 1978 the figures are compiled from the biennial and annual reports of the California Judicial Council.Google Scholar

102 From 1905 to 1952 the numbers of nonplea convictions were tabulated from the reports of the district attorneys in the biennial reports of the California Attorney General; from 1953 to 1978, the figures are compiled from reports of the California Bureau of Criminal Statistics.Google Scholar

103 The sources used are the same as those in note 102 supra.Google Scholar

104 Figures on the number of felony complaints are compiled from the reports of the California Bureau of Criminal Statistics.Google Scholar

105 Population figures are from the United States Bureau of the Census.Google Scholar

106 Compare the use of logarithmic scaling of appeal filings in the annual reports of the California Judicial Council starting in 1975. See also 1976 Cal. Jud. Council Ann. Rep. 87 n.8.Google Scholar

107 Cal. Penal Code § 17 (West 1970) was amended in 1969 to allow certain less serious felonies to be prosecuted as misdemeanors in the municipal courts. See 1977 Cal. Jud. Council Ann. Rep. 199–201 & fig. 11. However, the percentage of felony complaints resulting in superior court dispositions began to decline in the mid-1960s—from 63.3 percent in 1964 to 53.2 percent in 1968 to 38.7 percent in 1971—suggesting that the modification of § 17 was part of a larger adjustment of the operative jurisdiction of the superior courts. See Davies, supra note 11, at 334–35; see also fig. 3 in this article.Google Scholar

108 Ken Olsen, The Drug Diversion Program—An Initial Report (Sacramento, Cal.: Department of Justice, Division of Law Enforcement, Bureau of Criminal Statistics, Oct. 1973).Google Scholar

109 Total superior court criminal case dispositions remained fairly level throughout the late 1970s as did contested dispositions; the former increased substantially in fiscal 1981 but the latter did not. 1982 Cal. Jud. Council Ann. Rep. 4 (table showing criminal case dispositions in superior courts).Google Scholar

110 The percentage of total criminal case dispositions in the superior courts that were disposed of by guilty pleas increased from 68.7 percent in fiscal 1975 to 77.3 percent in fiscal 1981, while the percentage disposed of in contested trials declined from 14.8 percent in fiscal 1975 to 9.0 percent in fiscal 1981. Id.Google Scholar

111 Table 6 compares dispositions of appeals in terms of interventions rather than reversals and modifications because it is difficult to classify the outcomes of civil appeals in ways that are comparable to those in criminal appeals on the basis of the limited information on civil appeal dispositions in the court's registers.Google Scholar

112 Table 6 excludes dismissed appeals, since they typically do not involve a decision by the court, but includes summarily denied writ petitions because those cases usually do involve some consideration of the merits of the petition. See notes 42, 61 supra.Google Scholar

113 Compare National Center for State Courts, supra note 38, at 305, reporting that in all California courts of appeal combined, 29 percent of civil appeals were “reversed” compared to 12 percent of criminal appeals. Note, however, that the study does not indicate how modifications were counted and that prosecution appeals have apparently been combined with defendant appeals (i.e., no separate prosecution category is reported). See also Howard, supra note 2, at 51–52, reporting that the federal courts of appeals “overturned” the decision below in 29 percent of civil appeals compared to 17 percent of criminal appeals. Again, however, it appears that the prosecution appeals and defendant appeals were combined. The combined reversal rate will be higher than for criminal defendant appeals alone, given the higher reversal rate that is usual in prosecution appeals. See also Kagan et. al., Evolution of State Supreme Courts, supra note 28, at 981 n.54, indicating that in the state supreme courts during the period 1940–70 criminal appeals had the lowest reversal rate of all major categories of cases (33.8% compared to an average of 40% for all civil categories). Thus, while the rate of reversal of criminal convictions is much higher in the decisions of the state supreme courts than in the state intermediate appellate courts, even in the higher courts criminal appeals are less likely to be successful than civil appeals are.Google Scholar

114 See note 56 supra.Google Scholar

115 The control over the filing of prosecution appeals was particularly strong in the First District because the attorney general's office handles and screens all such appeals. But see also the text accompanying notes 285–87 infra.Google Scholar

116 One type of intervention that was difficult to classify was found in three appeals that overturned orders revoking probation. I included these three cases as reversals because they seemed to have a comparable effect. Because of the context in which they arise, however, they are quite different from the other reversals and are excluded from certain aspects of the analysis below, as noted.Google Scholar

117 Three appeals ending in reversal involved multiple appellants: in one, convictions of both appellants were reversed; in another, convictions of two appellants were reversed but the appeal of a third was dismissed; in another, conviction of one appellant was reversed but those of the other two were affirmed. The incidence of multiple appellants in reversed and modified appeals is roughly the same as that found in affirmed appeals, so the analysis of the distribution of decisions would be essentially the same if it were done on the basis of individual appellants. See note 64 supra.Google Scholar

118 My definition of reversal is somewhat narrower than that used by the Court of Appeal and the Administrative Office of the California Courts. For example, if the lesser of two counts in a conviction was overturned, that might be described officially as a reversal, but I would classify it as a modification. The definition I have employed is most troublesome with regard to two dispositions in which convictions were reduced from second degree murder to voluntary homicide. In both cases, the degree of the offense was really the only issue. In one appeal, 1 Crim. 12932, the court used the term reversed to describe its disposition, but did not use it in a very similar decision in 1 Crim. 12753. I have counted both as modifications.Google Scholar

119 I examined the superior court files of 22 of the 26 reversed appeals (excluding the three probation cases and one other reversal) to determine dispositions on remand. In 10 the prosecution was dismissed and in 3 there was no entry subsequent to the appeal; many of these 13 cases involved suppression of illegally obtained evidence. In 4 cases, a guilty plea was entered with a sentence of time served, and in 2 a guilty plea was entered with some additional sentence. In the remaining 3, retrials were undertaken, but no judgments had yet been entered.Google Scholar

120 See Christian, supra note 38 at 677, indicating that the average delay between judgment in the superior court and a decision in the First District in 1970 was longer than 16 months.Google Scholar

121 In fact, in two appeals a copy of the Court of Appeal's opinion reversing a conviction that had been sent to the prison address of the appellant was found in the case file stamped “return to sender” (1 Crim. 12063, and 1 Crim. 12570). In another instance the copy of the opinion reversing a conviction was returned from the previous home address of the appellant with the notation “doesn't live here anymore” (1 Crim. 12751).Google Scholar

122 Two of the 50 modifications were discovered during reanalysis of data from the case file sample of affirmed appeals. These 2 modifications, both of which involved the striking of a mandatory minimum prison sentence, were not identified in the disposition entries in the registers. Hence, it is possible that a few additional modifications of this type may be hidden in the affirmed appeals that were not included in the case file sample.Google Scholar

123 Mather, supra note 69, at 98, notes that many Los Angeles defense attorneys did not view reductions in the degree of a conviction or the striking of prior felony convictions as significant changes in the severity of a sanction because they believed that the California Adult Authority fixed sentences according to facts of the offense, rather than according to the formal conviction. See also note 81 supra.Google Scholar

124 See Johnson, Phillip E., Multiple Punishment and Consecutive Sentences: Reflections on the Neal Doctrine, 58 Calif. L. Rev. 357, 390 (1970).Google Scholar

125 Appeals are considered filed and are entered in the register when the Court of Appeal receives the trial record and transcript from the superior court. Examination of registers indicated that there was a delay of about three to six months between the date of judgment in the trial court and the date the appeal was filed. (See also Christian, supra note 38, at 677–78, estimating that in 1970 the average delay between judgment and the filing of the record in the First District was 101 days.) Since the fiscal year used in the California courts ends on June 30, the six-month difference between calendar and fiscal years is roughly comparable to the delay between judgment and the filing of the record. Hence, superior court cases decided in calendar year 1973 become the appeals filed in the courts of appeal in fiscal year 1974 (i.e., ending June 30, 1974).Google Scholar

126 In fact, 6 of the 26 reversals occurred in appeals from noncontested dispositions, 3 from guilty plea convictions, and 3 from submission on transcript convictions. All involved search and seizure issues.Google Scholar

127 Two sources provide data on comparative reversal rates in the five California courts of appeal. Beattie, Criminal Appeals in California, 1964–1968, supra note 38, at 14 table II, provides district-by-district breakdowns of reversal rates for fiscal years 1965 and 1966 combined and for 1967 and 1968 combined (note that I have recalculated the reversal percentages to exclude dismissed appeals); 1977 Cal. Jud. Council Ann. Rep. 191 table X-a also reports the reversal rates of criminal appeals in each of the districts for the period September 1975 through June 1976 (note that I have combined the “no retrials” reversal category with the “for retrial” category):Google Scholar

It may be significant that the First District has the lowest or near lowest reversal rate at each time period. Generally speaking, the First District bench was described by some interviewees as being somewhat “conservative” on criminal justice matters, though the interviewees generally hesitated to compare justices among different divisions of the First District, or to compare the First District to other districts. On the other hand, it is also possible that these differences may be at least partly the result of various measurement artifacts. In particular, the 1976 data reflects the start of reporting criminal appeal dispositions, and since this material is reported by clerks in the various districts, the possibility of definitional discrepancies is present. In addition, it is unclear how prosecution appeals were treated in the 1976 figures (they are excluded in the 1965–66 and 1967–68 figures). Unfortunately the Judicial Council has not reported reversal rates by district in later annual reports.Google Scholar

128 All of the reversal and intervention rates in fig. 4 are calculated against the number of decided appeals only—dismissed appeals are excluded.Google Scholar

129 To compute this reversal rate, appeals were tabulated according to the year in which they were filed. Because of the small number of criminal appeals filed each year during this period, I added together appeal entries for several consecutive years in order to have a more stable estimate of the reversal rate. Years were combined as follows: 1904–8, 1909–13, 1914–18, 1919–23, 1924–28, 1929–32, 1933–36.Google Scholar

130 Appeals were tabulated according to the year in which they were filed. In order to obtain sizable numbers of cases for the earlier years in this period, selected years sampled were combined or used singly as follows: 1940–41, 1946–47, 1950–51, 1954, 1956, 1958, 1960, 1962, 1964, 1966, 1968, 1969, 1970, 1971, 1972, 1973, 1974.Google Scholar

131 Beattie, Report on California Criminal Appeals 1929–1935, supra note 38, at 100.Google Scholar

132 Vernier, C. G. & Selig, Philip Jr., The Reversal of Criminal Cases in the Supreme Court of California, 20 J. Crim. L. 60 (1929). Data for that study are reported broken down by decade, so I have plotted the reversal rates reported at the midpoints of the decades: 1905, 1915, and 1925.Google Scholar

133 Vernier and Selig report that the criminal appeal reversal rate in the California Supreme Court was 52.4 percent for the decade 1870–79. Id. at 63.Google Scholar

134 The California harmless error rule was adopted in 1911. See note 205 infra. Vernier and Selig attribute the decline in the reversal rate to the adoption of the harmless error standard. Id. at 79. Note however, that their data show that the reversal rate was declining steadily in the California Supreme Court after the 1880s. See curve d on fig. 4 for the decline prior to the adoption of the harmless error rule. But note that curve a does show a sharp decline in reversals in the First District following the enactment of the rule.Google Scholar

135 In addition to Gideon v. Wainwright, 372 U.S. 335 (1963), it appears that Escobedo v. Illinois, 378 U.S. 478 (1964), and People v. Dorado, 62 Cal. 2d 338 (1965), both of which established new standards regarding suspects' rights to counsel during interrogation, may have resulted in a substantial number of reversals during the mid-1960s. For example, Beattie, Criminal Appeals in California, 1964–68, supra note 38, at 15 table III, indicates that 40 percent of all appealed homicide convictions were reversed in California in fiscal years 1965–66, combined, but that this dropped to 13.5 percent in fiscal 1967–68, combined. (Because confessions may play an unusually important role in homicide prosecutions, the reversal rate in homicides may be a good indicator of the impact of these two decisions.) See also Tobriner, Matthew O., Individual Rights in an Industrialized Society, 54 A.B.A. J. 21 (1968), noting that the impact of Dorado on criminal appeal reversals was short term in nature. Note that in fig. 4, above, because the reversal rates are plotted according to the year in which cases were filed rather than year decided, the impact of the 1965 Dorado decision should be visible in the appeals filed in 1964.Google Scholar

136 One justice who had served on the Court of Appeal during much of the due process revolution period described the effect of newly announced due process standards on the reversal rate as follows:. As new supreme court decisions changed the rules in criminal procedure, they had the effect of causing reversals of convictions in which an appeal was pending at the time. In terms of reversal rates, the big impact is the first one—on pending cases. And if a new rule is retroactively applied [as several early Warren Court decisions were] then the spurt is bigger yet. If there is no retroactive application then there is no big build-up of reversals.Google Scholar

137 Note that the Bureau of Criminal Statistics study of criminal appeals during 1964–68 (Beat-tie, supra note 38), shows that the decline in the reversal rate between fiscal years 1965–66, and 1967–68 occurred in all five courts of appeal in California, as discussed in note 127 supra.Google Scholar

138 The Supreme Court has produced some innovative due process rulings during the Burger Court period, of course. For example, the Court's 1972 decision in Morrissey v. Brewer, 408 U.S. 471, extended due process standards to certain postconviction proceedings, but generally commentators agree that the Burger Court has not advanced beyond the due process positions taken by the Warren Court, and in many cases has retreated substantially from earlier standards.Google Scholar

139 As noted in part II infra, reversals occur disproportionately among the minority of less serious convictions that do not exhibit the predominant characteristics found in criminal appeals; that is, reversals are more common among the small number of appeals that involve less serious convictions or that involve some basis for doubt about factual guilt. As a result, it would be a plausible hypothesis that the contraction of the superior courts' caseloads, as described above, might have deflected a few appeals of the type most likely to be reversed. For example, a drug diversion program might have diverted away some potential appeals from convictions involving search and seizure issues that would have been likely candidates for reversal. Thus, the apparent contraction of the caseload in superior courts may have resulted in an apparent increase in the rate of appeal but an apparent decrease in the rate of reversal. Obviously, further research would be needed to test these speculations.Google Scholar

140 While proponents of central staff processing have asserted that it has no effect on the outcomes of the appeals decided, no empirical test of that proposition has been conducted. From the standpoint of contemporary organization theory, it can be hypothesized that introducing the staff process—based as it is upon the expectation that it would be used only for routine or hopeless appeals—would be likely to produce an even higher rate of affirmance than the traditional process would. See the discussion in the text accompanying notes 283–84 infra. See also Davies, supra note 14, at 395–403; Haworth, Charles R., Screening and Summary Procedures in the United States Courts of Appeals, 1973 Wash. U.L.Q. 257, 309–19, noting that the reversal rate also declined in the Fifth Circuit Court of Appeals following the introduction of a screening process.Google Scholar

141 The statewide affirmance, modification, and reversal rates for courts of appeal reported in the annual reports of the California Judicial Council are as follows (note that I have combined the Judicial Council's “reversed for expected retrial” and “reversed, no retrial possible” categories into a single reversed category):Google Scholar

It is interesting to note that the reversal rate increased substantially in fiscal year 1978, the first year that the determinate sentencing law was in effect. See note 76 supra. Further research is needed to determine whether there is a relationship between that change and the reversal rate. Note, too, that the source does not state whether prosecution appeals are excluded from these figures.Google Scholar

142 At first appearance, the determinate sentencing law would not appear to have direct implications for the reversal rate since most sentencing issues would only result in modifications of convictions. It may be, however, that the changes in the sentencing law may have changed the plea bargaining process in the trial courts (since it would be possible to bargain over sentence severity and length to a greater degree than was possible under the indeterminate sentencing process), and that may have altered the mix of cases going to trial and most likely to be appealed. Thus, like the measurements of the “rate” of appeal discussed in the text supra, the “reversal rate” as I have measured it is also susceptible to changes in the composition of the base line used in the calculation (i.e., the number of decided appeals).Google Scholar

143 The purported complexity of the law regarding search and seizure has been one of the recurring themes in commentary criticizing the exclusionary rule.Google Scholar

144 The court's references to harmless errors are tabulated in table 9 infra.Google Scholar

145 See notes 18, 19 supra.Google Scholar

146 According to unpublished statistics provided to the author by the California Administrative Office of the Courts, there were only 28 dissenting opinions and 27 concurring opinions filed in the First District in fiscal 1976. There were a total of 1,815 written decisions in the First District that year (1977 Cal. Jud. Council Ann. Rep. 190 table IX). Hence, a dissent was entered in only 1.5 percent of the court's decisions. Even if one assumes that all of the concurring and dissenting opinions were filed in different cases (actually there was some overlap), some difference of opinion was expressed in only 3 percent of the court's decisions. To put it another way, the 12 justices on the First District bench averaged only 2.3 dissenting opinions and 2.3 concurring opinions per year. The First District is not unique in this very low level of dissent: all five districts in the same year, with a total of 5,943 written opinions (id.), had only 97 dissenting and 59 concurring opinions (according to the unpublished statistics mentioned above)—an expression of some degree of disagreement in a maximum of only 2.6 percent of all written opinions.Google Scholar

147 Although appellate decisions are usually said to be collegial in nature, studies have identified the strong influence of the judge assigned to write the court's opinion. E.g., Sickels, Robert J., The Illusion of Judicial Consensus: Zoning Decisions in the Maryland Court of Appeals, 59 Am. Pol. Sci. Rev. 100 (1965) (the court involved is a state supreme court). See also Thompson, Mitigating the Damage, supra note 38. There was no strong relationship between the judge who authored the opinion and the outcome of criminal appeals in the First District, however. Of the 26 reversals, 17 were signed by an individual judge (the rest were “by the court”). One justice signed 3 reversals, four signed 2 reversals, seven signed 1 reversal (including two judges sitting pro tempore), and two did not sign any. The two justices who signed none were among those often described as conservatives, but the two justices sometimes described as liberals on criminal matters only signed 1 and 2 reversals, respectively.Google Scholar

148 Wold, supra note 5, at 65, also concludes that the “ideological outlooks of the justices are not and cannot be critical factors in the vast majority of cases the [California] courts of appeal handle.” Note, however, that Wold attributes the absence of ideological decision making to the “routine” nature of the cases without discussing the institutional norms discussed below.Google Scholar

149 See note 16 supra.Google Scholar

150 See note 2 supra.Google Scholar

151 During interviews, several judges commented that, even as trial judges, they had little knowledge or understanding of the appellate process prior to their appointment to the Court of Appeal, and that their views of the appellate process had changed substantially since their appointment.Google Scholar

152 For more complete analysis of the environmental sources of constraints on the Court of Appeal that utilizes the organization theory concept of organizational domain, see Davies, supra note 11, at 73–78, 107–92; see also Davies, supra note 14, at 376–83.Google Scholar

153 Mather, supra note 69.Google Scholar

154 Heumann, supra note 93, discusses the socialization process by which lawyers and judges adjust to the differences between the informal norms that support and facilitate plea bargaining and the formal adversary model of criminal justice.Google Scholar

155 Eisenstein & Jacob, supra note 8, at 244–52, found in their comparative study of felony courts in three cities that plea bargaining rates were highest in Chicago and Detroit where the membership of courtroom workgroups (i.e., the judge, prosecutor, and public defender in a particular courtroom) was most stable over time. Since the stability of workgroup membership is a condition that would facilitate the development of shared understandings and expectations, this finding indirectly supports the significance of shared norms in the plea bargaining process. See also id. at 109, describing the accommodative norms that develop among members of the workgroup.Google Scholar

156 The suggestion that informal norms exist in intermediate appellate courts is hardly novel; intermediate appellate judges have recognized such norms. E.g., Hopkins, supra note 10, at 460: “Apart from the formal delimitations supplied by rules, there is additional modulating force in the subtle accommodations arising out of usage and experiment, conscious or unconscious.”.Google Scholar

157 All of the interviewees were assured that their remarks would be anonymous. To preserve the anonymity of the clerk of the court and principal attorney, I have adopted the convention of referring to each of them as a justice in the few instances in which either may be quoted.Google Scholar

158 The interviews were conducted as “elite interviews” following the techniques suggested in Lewis A. Dexter, Elite and Specialized Interviewing (Evanston, Ill.: Northwestern University Press, 1970). Notes were taken during the interviews and these were written out fully immediately at the end of each interview. While the interview materials may not be entirely verbatim, they do capture the substance and tone of the interviewees' remarks. In a few instances, material that would have identified the interviewee has been deleted or changed to protect anonymity. All quotations without citations in the text and notes are from these interviews.Google Scholar

159 I refer to law-articulating opportunities because publishing opinions that contribute to the development of the law is one of the most satisfying aspects of appellate work to many judges. See Howard, supra note 2, at 247; Davies, supra note 14, at 381.Google Scholar

160 At the time of this study, the granting of a petition for hearing by the Supreme Court vacated the Court of Appeal decision as legal precedent. If the opinion of the Court of Appeal had been certified for publication, it would sometimes be deleted from the advance sheets and would not appear in the official reporter of appellate opinions. Moreover, unlike the practice of the United States Supreme Court, the California Supreme Court's practice is to make no reference in its opinions to the previous decision in the court of appeal.Google Scholar

161 The California Supreme Court began to use this nonpublication technique in 1971. From 1972 through 1976, the court ordered nonpublication of an average of 20 court of appeal opinions per year. See Note, Decertification of Appellate Opinions: The Need for Articulated Judicial Reasoning and Certain Precedent in California Law, 50 S. Cal. L. Rev. 1181, 1200–6 (1977). This source indicates that 3 court of appeal opinions were ordered nonpublished in calendar 1971, 20 in 1972, 22 in 1973, 12 in 1974, 13 in 1975, 32 in 1976, and 24 through August 23, 1977.Google Scholar

162 The justices of the First District also said that they generally did not feel free to articulate rules different from precedents written by other courts of appeal in California. In this sense, the California intermediate appellate judges probably have less opportunity for law articulation than the judges of the federal courts of appeals where the doctrine of the “law of the circuit” allows each circuit to develop its own precedents until such time as the Supreme Court addresses the issue.Google Scholar

163 1977 Cal. Jud. Council Ann. Rep. 251 table 6.Google Scholar

164 Compare Wold & Caldeira, supra note 5, at 339–41.Google Scholar

165 There was a dissenting opinion on this point. When asked if the Court of Appeal had a greater law-articulating role in civil than in criminal cases, one justice replied:. No, there aren't more precedents for us to write in civil than in criminal cases. The fact is that we don't have that many great precedents to write in either area. And when they do come up, [the California Supreme Court] takes over those cases.Google Scholar

166 Cal. Civ. & Crim. Ct. R. 977.Google Scholar

167 1977 Cal. Jud. Council Ann. Rep. 195 table XV.Google Scholar

168 In fiscal 1976, 17.4 percent of the First District's opinions in civil appeals were published, compared to 5.1 percent of criminal appeals, and 41.2 percent of writ petition opinions (this last figure is quite high because most writ petitions are summarily denied without any opinion). Id. In the same period, the First District wrote 972 opinions in civil appeals, 714 in criminal appeals, and 129 in writ cases. Id. at 251 table 6. Hence, the First District published approximately 169 opinions in civil appeals (972 ×.174), 36 in criminal appeals (714 ×.051), and 53 in writ petitions (129 ×.412), for a total of 258 published opinions. Thus, published opinions in criminal appeals accounted for only 14 percent (36 + 258) of all the opinions published by the court. In the fiscal 1974 criminal appeals I counted 35 in which the opinion was certified to be published (see note 226 infra).Google Scholar

169 Compare the description of a state intermediate appellate court's law-articulating work offered by James D. Hopkins, a judge of the New York intermediate appellate court. Although Hopkins argues that the volume of appellate cases and the limited capacity of the state supreme court call for the intermediate appellate court to play a larger role in law articulation, he rejects the appropriateness of the intermediate appellate court's distinguishing away prior supreme court precedent. Rather, he describes the proper role of the intermediate appellate court primarily in terms of identifying issues that should be addressed by the higher court. Hopkins, supra note 10, at 464–68.Google Scholar

170 See note 15 supra.Google Scholar

171 Hopkins, supra note 10, at 466–67, notes that intermediate appellate courts are often able to avoid the application of “repugnant doctrine” and supreme court precedents through the technique of developing factual distinctions, although he disapproves that practice. See also McBarnet, infra note 265.Google Scholar

172 Compare the following description of appellate review given by Frank M. Coffin, a judge on the United States Court of Appeal for the First Circuit:. Deciding an appeal is not a matter of approaching a problem as if for the first time. It is determining whether another, earlier, carefully structured decision should be upheld. That earlier decision, if by a court, is already the full-bodied product of a formal adversary hearing, held with all the garnishments of due process…. Such a decision ought to be more likely to be just than the great mass of decisions that govern us in our everyday living……. [S]o much has been invested in the first decision that intuition tells us it should not be too easily discounted. There is no public policy for preferring a second decision that differs from the first only because the second decider weighs the evidence differently. Nor does what hindsight reveals to be a mistake by the trial judge, but one that has not significantly infected the final result, count enough in the scales of justice to warrant a second trial. There must, therefore, be conscious and controlled deference to the decision of the lower court…. [A]n appellate court …. is restricted in two ways: it must confine itself to the factual record established in the trial court or administrative agency, and it must generally recognize only those legal issues which were raised in the trial court. Coffin, The Ways of a Judge: Reflections from the Federal Appellate Bench 52–53 (Boston: Houghton Mifflin Co., 1980). This description incorporates the harmless error rule, the substantial evidence rule, and the principle of abstention discussed in the text infra. See also Yale Note, supra note 8, at 1193 n.10, 1194 n.ll, 1199 n.32.Google Scholar

173 For an indication of the expectation of affirmance in criminal appeals, see Coffin, supra note 172, at 90, in which the author contrasts the case under discussion with “a routine criminal case with an affirmance.”.Google Scholar

174 As Lawrence M. Friedman has observed, “The basic problem of review or appeal is how to avoid doing everything over again—which would be a tremendous waste—but at the same time make sure that lower-court mistakes are corrected.” Friedman, A History of American Law 131–32 (New York: Simon & Schuster, 1973).Google Scholar

175 See, e.g., Coffin, supra note 172, 52–54.Google Scholar

176 E.g., Carrington et al., supra note 4, at 57, 109–10. See also note 172 supra.Google Scholar

177 See generally James D. Thompson, Organizations in Action: Social Science Bases of Administrative Theory, esp. chs. 3, 7 (New York: McGraw-Hill Book Co., 1967).Google Scholar

178 Id. at 91–92.Google Scholar

179 Except for a few justices who had been superior court judges in San Francisco, the justices said that they had seldom had personal contact with superior court judges. One justice explained:. I really don't [have much contact with superior court judges]. Not much at all. The very nature of the relationship between the courts of appeal and the trial courts mitigates against it, because we really sit in judgment of their decisions. So there is no tendency to socialize. We tend to be critical of them—it's our job—and they criticize us—but not in writing like we do.Google Scholar

180 There is relatively little systematic information on the way trial judges view appellate reversal; however, the existing sources do tend to confirm that trial judges react negatively to reversal. E.g., Rubin, Alvin B., View from the Lower Court, 23 UCLA L. Rev. 448, 448–49 (1976), in which the author, then a federal trial judge, comments on the “natural antagonism between judges who sit on high (or midway) and their (sometimes) learned brethren below.” Generally, empirical studies of criminal trial courts have little to say about trial judges' reactions to reversal (an understandable omission in view of the infrequency with which reversal actually occurs), but an exception is Heumann, supra note 93, at 66, 144, noting that the desire to avoid exposure to possible reversal motivates some judges to press for guilty pleas. Heumann quotes one trial judge as follows:. Oh yes, we do worry about reversals on appeal, certainly. Some judges are apprehensive about their reputations, not only among fellow judges and the bar, but also on the Supreme Court. If you make a lot of wrong decisions, the Supreme Court keeps sending down “error, error, error,” and, you know, you get very concerned about that. Some judges keep a tally sheet of the cases they've won and the cases they've lost, and they're concerned about what the Supreme Court thinks about them. Judges up there know the good judges and the phony judges, and judges that'll settle a case so that the case won't go up and there won't be any chance of being criticized. Id. at 144. Two surveys of trial judges have produced strikingly different views of the significance to trial judges of appellate review. Jackson, Donald W., Salient Interactions: The State Trial Judge and the Legal Profession, 1 Just. Sys. J. 24, 27–32 (1975), reports that when trial judges were asked to rank 25 categories of persons according to the importance of their views regarding the behavior appropriate for trial judges, 48.4 percent of 182 respondents said state appellate judges were “extremely important” (more than for any other category) while another 32.4 percent said they were “very important.” But Lenore Alpert, Learning About Trial Judging: The Socialization of State Trial Judges, in Cramer, ed., supra note 7 at 105, 132 table 3, reports that Florida trial judges ranked “appellate judges” near the bottom of a list of 13 categories of “role sources” for trial judges. See also William I. Kitchin, Federal District Judges: An Analysis of Judicial Perceptions 128–32 (Baltimore: Collage Press, 1978), reporting that while several federal district judges interviewed complained about too frequent reversals, they generally reported they were quite independent of supervision in their work.Google Scholar

181 The notion that relations with lower level courts may constrain the actions of higher level courts is not as novel as it may first sound. Organization theorists have argued that most organizational relationships are essentially two-directional. That is, “orders” from the upper level executive are effective only to the extent that they are accepted by lower level organization members. Acceptance of directives depends, in turn, on the perceived legitimacy of the order and on the incentive system of the lower level of the organization. Since the upper levels of an organization depend on lower levels in a variety of ways, this means that some effective constraints flow from bottom to top as well as from top to bottom. See Chester I. Barnard, The Functions of the Executive (Cambridge, Mass.: Harvard University Press, 1938). See also James Q. Wilson, Political Organizations (New York: Basic Books, 1973); Michel Crozier, The Bureaucratic Phenomenon (Chicago: University of Chicago Press, 1964). Of course, the lower courts can escape the control of the higher courts to a substantial degree. E.g., Murphy, Walter F., Lower Court Checks on Supreme Court Power, 53 Am. Pol. Sci. Rev. 1017 (1959); Baum, supra note 8, at 89–91.Google Scholar

182 See also Coffin, supra note 172, at 87, noting that “appellate courts rarely fault trial courts for rulings made in the absence of any objection.” See also id. at 165.Google Scholar

183 E.g., in 1 Crim. 12230 a defendant appealed from a submission on transcript conviction on the ground that he was misinformed as to the possible sentence when he waived a jury trial. Apparently the defendant's lawyer misstated the possible sentence of 2 to 10 years as 1 to 10 years and the trial judge repeated the 1-to-10-year statement. The Court of Appeal affirmed on the ground that the mistake was invited by defense counsel.Google Scholar

184 The case is 1 Crim. 12182. In addition to declining to reach the lineup issue, the Court of Appeal found that there had been no error in excluding defense evidence to the effect that there had been a previous and mistaken identification of another suspect in the case.Google Scholar

185 As noted in table 6, relief was granted in only 20 of the 396 habeas corpus petitions filed in fiscal 1974. Moreover, 8 of the 20 related to the issue of granting credit for time served prior to conviction, under In re Kapperman, 11 Cal. 3d 542, 522 P.2d 657, 114 Cal. Rptr. 97 (1974). So far as I could determine, none of the granted habeas corpus petitions reversed a conviction. Counsel was appointed only in petitions in which a show cause order was issued. See also Shapiro, David L., Federal Habeas Corpus: A Study in Massachusetts, 87 Harv. L. Rev. 321 (1973). Compare the infrequent success in pro se prisoner suits discussed in Turner, William Bennett, When Prisoners Sue: A Study of Prisoner Section 1983 Suits in the Federal Courts, 92 Harv. L. Rev. 610 (1979).Google Scholar

186 Counting instances of abstention on issue topics is complicated because the Court of Appeal sometimes coupled abstention with a statement that the allegation of error was without merit or that the error alleged would be harmless in any event. The distribution of instances of abstention by issue topics was as follows:Google Scholar

Note that issue topics that did not involve any abstention are omitted; compare to table 2 in text supra. Instances of abstention are estimated from the weighted case file sample; see note 59 supra.Google Scholar

187 The court found inadequate representation in only two appeals. See text preceding note 230; table 9 and accompanying text infra.Google Scholar

188 Cf. Roger J. Traynor, The Riddle of Harmless Error 76 (Columbus: Ohio State University Press, 1970): “All too frequently, … appellate courts apply these doctrines [of waiver and invited error] mechanically to inappropriate situations.”.Google Scholar

189 Criminal appeals are limited to “questions of law alone” under § 1235 of the Penal Code, enacted in 1872. Of course, there is no clear separation of law and fact questions in many instances. For example, whether the amount of evidence of guilt is sufficient to allow a fact finder to convict a defendant is often said to be a legal question. Similarly, probable cause issues are often said to be “mixed” questions of fact and law.Google Scholar

190 Unpublished opinion in 1 Crim. 12029, citing People v. Newland, 15 Cal. 2d 678, 680, 104 P.2d 778, 780 (1940). See also note 232 infra. See generally Carrington, Paul D., The Power of District Judges and the Responsibility of Courts of Appeals, 3 Ga. L. Rev. 507, 517–27 (1969), discussing the substantial evidence doctrine in the federal courts. But compare the standard for reviewing the sufficiency of the evidence set forth in Jackson v. Virginia, 443 U.S. 307 (1979).Google Scholar

191 See Traynor, supra note 188, at 27, offering the following caution about the application of the substantial evidence rule:. This seemingly sensible rule has taken some strange twists. Occasionally an appellate court affirms the trier of facts on isolated evidence torn from the context of the whole record. Such a court leaps from an acceptable premise, that a trier of fact could reasonably believe the isolated evidence, to the dubious conclusion that the trier of fact reasonably rejected everything that controverted the isolated evidence. Had the appellate court examined the whole record, it might have found that a reasonable trier of fact could not have made the finding in issue. One of the very purposes of review is to uncover just such irrational findings and thus preclude the risk of affirming a finding that should be disaffirmed as a matter of law. [Footnotes omitted.].Google Scholar

192 Traynor, supra note 29, at 220, describes the effect of the substantial evidence rule:. [Appellate courts] give finality to the disposition of cases on the facts when they are in conflict. Under the substantive evidence rule it is ordinarily futile to base an appeal on insufficiency of the evidence whether the burden in the trial court was to prove the facts by a preponderance of the evidence, by clear and convincing evidence, or beyond a reasonable doubt. Thus, the importance of winning in the trial court looms large. Compare Coffin, supra note 172, at 146, “All that is needed to support a jury's verdict is just enough evidence to enable a rational jury to find as it did, even though the court would have taken the opposite position.” He also notes that neither he nor his colleagues are “enthusiastic about reversing a judgment because of insufficient evidence to support a jury verdict of guilty.”Id. at 175.Google Scholar

193 The case is 1 Crim. 12269. In discussing probable cause, the court did not address the apparently undisputed fact that the police radio report on the robbery had described the robber's car as blue or green, while the defendant's car was bronze.Google Scholar

194 Mather, supra note 69, at 72–73, describes two throwing cases she observed, both of which appear to involve police perjury. In one, police testified that they approached the defendant and found a snuffbox containing marijuana after the defendant took it from his pocket and threw it. The defendant testified that the police stopped him as he was walking and said they were looking for a man with weapons and that they then searched him and found the snuffbox in one of his inside pockets. The defendant's story about the police giving a weapons frisk justification for the search has a note of credibility. It seems unlikely that a defendant would be sophisticated enough to concoct such an explanation—though it is possible he had been stopped before for such a reason. In the second throwing case described by Mather, the police later informally admitted falsifying their testimony that the marijuana was thrown by the defendant, saying the marijuana was thrown, but by someone other than the defendant. Mather notes that in such cases the trial court judge routinely accepts the police testimony. Id. at 73, 146–47. Indeed, she reports that public defenders often choose jury trials over bench trials in cases involving conflicts in testimony between defendants and police because they believe trial judges are less willing than juries to conclude that the police lied. See also Graham, supra note 2, at 136–38; Barlow, Sarah, Patterns of Arrests for Misdemeanor Narcotics Possession: Manhattan Police Practices 1960–62, 4 Crim. L. Bull. 549 (1968); Chevigny, Paul G., Police Abuses in Connection with the Law of Search and Seizure, 5 Crim. L. Bull. 3 (1969). See also Jonathan Rubinstein, City Police 389–90 (New York: Farrar, Straus & Giroux, 1973), for a description of police perjury regarding warrant proceedings.Google Scholar

195 See Note, Effect of Mapp v. Ohio on Police Search-and-Seizure Practices in Narcotics Cases, 4 Colum. J.L. & Soc. Probs., Mar. 1968, at 87. According to that study, the New York police claimed that contraband was in view or dropped in only 27.5 percent of all narcotics searches prior to Mapp's application of the exclusionary rule to the states, but that percentage rose to 39.6 in the next six months after Mapp and to 72.7 by 1964. The researchers concluded: “[P]olice allegations as to how evidence was obtained changed after the Mapp decision. There is some indication, however, that police practices in the field have not changed substantially, and that police officers often merely fabricate testimony ….”Id. at 87.Google Scholar

196 1 Crim. 12576.Google Scholar

197 Id. A similar instance occurred in 1 Crim. 12614.Google Scholar

198 1 Crim. 12473.Google Scholar

199 See Graham, supra note 2, at 137–38, for an anecdotal description of how trial court judges routinely accepted the validity of Miranda warnings during a period immediately after the decision when it was unlikely that the police had enough training to give the warnings properly.Google Scholar

200 California adopted the “reasonable probability” standard for determining the harmless nature of legal errors in People v. Watson, 46 Cal. 2d 818, 299 P.2d 243 (1956). See also Traynor, supra note 188, at 34. Compare the development of a similar harmless error standard in New York discussed in Note, People v. Crimmins: The New Prejudice Rule, 40 Alb. L. Rev. 405 (1976).Google Scholar

201 The historical background and emergence of the harmless error rule are discussed by Traynor, supra note 188, at 3–51.Google Scholar

202 See note 133 supra.Google Scholar

203 See Friedman, supra note 174, at 131–33, 347–48 (1973). Of course, as Friedman notes, appeal was infrequent and available on a less than equal basis in that period.Google Scholar

204 Roscoe Pound, Criminal Justice in America 161 (New York: Henry Holt & Co., 1930). But see Lawrence M. Friedman & Robert V. Percival, The Roots of Justice: Crime and Punishment in Alameda County, California, 1870–1910 at 261–87 (Chapel Hill: University of North Carolina Press, 1981), concluding from a systematic review of turn-of-the-century criminal appeals filed in Alameda County (Oakland)—one of the counties now in the First District—that “There was, on our evidence, no more than a pinhead of ‘hypertrophy,’ and precious little sign of appellate coddling. There may have been a case of ‘hypertrophy’ here and there; these could well mislead an observer. Bad cases make bad law—and deceptive impressions.”Id. at 287.Google Scholar

205 California's harmless error rule was originally enacted in 1911 as Cal. Const, art. 6, § 4 1/2; since the 1966 revision of the judicial article it is in art. 6, § 13. Traynor, supra note 188, at 8–9, indicates that the miscarriage of justice standard was also used in the English Judicature Act of 1873. Note, however, that as early as 1872 California had enacted a milder statutory prohibition against reversals based on “technical errors or defects … which do not affect the substantial rights of the parties” in § 1258 of the Penal Code (West 1970).Google Scholar

206 Traynor, supra note 188, at 33–51. Traynor advocates the most stringent standard.Google Scholar

207 See text accompanying note 200 supra. Note, however, that there has been an extensive debate over the appropriate threshold for harmless errors when a federal constitutional right is involved. See especially Chapman v. California, 386 U.S. 18 (1967), Harrington v. California, 395 U.S. 250 (1969), and Schneble v. Florida, 405 U.S. 427 (1972). See also Traynor, supra note 188, at 37–48, 55–64. There was also a notable debate between Judges Hand and Clark—advocating a less stringent harmless error standard—and Judge Frank—advocating a more stringent standard—in the decisions of the United States Court of Appeals for the Second Circuit, discussed in Marvin Schick, Learned Hand's Court 267–76 (Baltimore: John Hopkins Press, 1970). For a list of the key Second Circuit cases on this issue, see Traynor, supra note 188, at 94 n.85.Google Scholar

208 See Goldberg, Steven H., Harmless Error: Constitutional Sneak Thief, 71 J. Crim. L. & Criminology 421 (1980). For a discussion of the application of the harmless error rule in federal criminal appeals, see Winslow, Donald A., Harmful Use of Harmless Error in Criminal Cases, 64 Cornell L. Rev. 538, 545 n.36 (1979). Winslow employed a LEXIS computer search of reported federal appeals and concluded that the rate of the involvement of harmless error issues in such cases has been increasing (from .79% in 1966 cases to 3.4% in 1977 cases). The validity of this finding is unclear, however, since a large and growing proportion of federal appeals, especially criminal appeals, are unpublished and may not necessarily be included in the LEXIS data base.Google Scholar

209 E.g., Traynor, supra note 29, at 222, comments about the relationship between the substantial evidence rule and the harmless error rule:. The substantial evidence rule, working in uneasy harness with the rule that only prejudicial error is reversible, calls a halt to much litigation. The wayward application of the two rules poses a riddle. Although courts invoke the substantial evidence rule to abdicate review of the record as a whole, that abdication is counteracted by the prejudicial error rule whereby the court seizes upon an error to reverse an unjust judgment against the weight of the evidence. The court is constantly weighing evidence to determine whether error is harmless or prejudicial. See also Traynor, supra note 188, at 27–30.Google Scholar

210 Cf. Yale Note, supra note 8, at 1194 n.11.Google Scholar

211 See table 9 infra, for a tabulation of harmless errors relating to various issue topics in the appeals studied.Google Scholar

212 The case discussed is 1 Crim. 12542. Harmless errors were also present in three other affirmed appeals in which the record disclosed a prior hung jury trial, 1 Crim. 12353, 1 Crim. 12376, 1 Crim. 12786, and present in one appeal that ended in a modification after a prior hung jury, 1 Crim. 12373.Google Scholar

213 See Traynor, supra note 188, at 57.Google Scholar

214 See note 76 supra.Google Scholar

215 See Hopkins, supra note 10, 470–72, in which Hopkins, a New York intermediate appellate judge, observes “a tendency at the appellate level to defer to the judgment of the sentencing judge” except in extreme cases. More generally, see Rosenberg, Maurice, Judicial Discretion of the Trial Court, Viewed from Above, 22 Syracuse L. Rev. 635 (1971).Google Scholar

216 The discretionary nature of many family law questions was also given as a reason for the fact that many appeals from divorce and other family law cases were assigned for staff processing, as discussed infra. See also Yale Note, supra note 8, at 1210, noting that the reversal rate in family law cases in the state supreme courts tends to be lower than average.Google Scholar

217 Traynor, supra note 188, at 99 n.146.Google Scholar

218 E.g., 1 Crim. 12459, citing People v. Ibarra, 60 Cal. 2d 460, 466, 386 P.2d 487, 490, 34 Cal. Rptr. 863, 866 (1963). My treatment of the legal doctrines regarding ineffective assistance of counsel is necessarily oversimplified here. For a discussion of the diversity of standards that may be applied, see Comment, Defects in Ineffective Assistance Standards Used by State Courts, 50 U. Colo. L. Rev. 389 (1979). The “farce or sham” standard was the standard I found most often in the appeals I reviewed.Google Scholar

219 E.g., in 1 Crim. 12512, a rustling conviction, the court rejected the argument that trial counsel had been ineffective in failing to seek suppression of statements made by defendant under circumstances that allegedly violated the Miranda standard. The court said that even if the statements were suppressible, the failure to seek suppression might have been a trial tactic. The remaining question, of course, is whether an attorney who would follow such a “tactic” could be deemed effective.Google Scholar

220 Compare Jerome Frank, Courts on Trial: Myth and Reality in American Justice 222–24 (Princeton, N.J.: Princeton University Press, 1949), discussing what Frank called “the upper-court myth,” the notion that appellate courts are at the center of courthouse government.Google Scholar

221 Similarly, some of the judicial administration works that refer frequently to the “routine” or “hopeless” nature of most criminal appeals do so without any discussion at all of the norms of affirmance that make the appeals routine. E.g., Carrington et al., supra note 4. See also Wold & Caldeira, supra note 5.Google Scholar

222 One example of the Court of Appeal's reluctance to reverse for supervision occurred in 1 Crim. 13216, a case in which the defendant's “rap sheet” was included in papers before a judge during a bench trial. Under existing precedent, the Court of Appeal said that this practice had been found to be “normally improper.” In fact, the court went so far as to suggest that the result of the procedure used by this trial court was “to subject every criminal judgment of the Alameda County Superior Court to unpredictable hazards of attack, in writ proceedings as well as appeals.” Since there was no showing the trial judge in the case actually read the rap sheet, however, the court found no prejudicial error and affirmed. Interestingly, this opinion was “by the court,” indicating that it was probably a staff-processed case. It was not published.Google Scholar

223 See also the interview quotation in the text following note 273 infra.Google Scholar

224 See note 43 supra.Google Scholar

225 The absence of an enforcement posture toward supervision seems to be a general phenomenon in judicial administration. For example, Peter Graham Fish, The Politics of Federal Judicial Administration 418 (Princeton, N.J.: Princeton University Press, 1973), reports that the federal circuit councils very seldom enter formal orders directed to subordinate judges. Compare Ward, Brent D., Can the Federal Courts Keep Order in Their Own House? Appellate Supervision Through Mandamus and Orders of Judicial Councils, 1980 B.Y.U. L. Rev. 233. See also the discussion of appellate supervision through the promulgation of rules in Arlene Sheskin & Charles W. Grau, Judicial Responses to Technocratic Reform, in Cramer, ed., supra note 7, at 225, a study of the implementation of the Ohio Rules of Superintendence adopted by the Ohio Supreme Court in 1971 to increase trial court productivity. The authors report that virtually no actual sanctions were employed to enforce the rules. Although one Ohio chief justice had been somewhat aggressive in demanding that late judges catch up, another justice was quoted as follows:. As far as the [trial judges'] reports go, we're not policemen and it doesn't mean a thing to me whether a judge falsifies his reports or not. That's between him and his God … the purpose of those reports is for the judge to know how he stands on his own docket, not for me to be a policeman and call him up and say, “Hey, you only got rid of 30 files and picked up 50 last month and you're falling behind.”. Id. at 237–38. Nevertheless, the authors indicate that some trial judges did view the rules as a source of stress and thought that the rules may have had some effect on trial judge productivity.Google Scholar

226 The distinction between assisting the trial judges and supervising them may also be reflected in the Court of Appeal's publication practices. Although one might expect reversals to account for most publication in the court if it were primarily concerned with supervision, most published opinions occur in affirmed cases. The distribution of published opinions in the fiscal 1974 criminal appeals by dispositions is as follows:Google Scholar

(The Supreme Court subsequently vacated publication in one affirmed opinion and one modification opinion.) Although the rate of publication is somewhat higher in reversals than in affirmances, two-thirds of the published opinions are in affirmances, fewer than one-tenth are in reversals.Google Scholar

227 Several justices characterized the trial courts as “chaotic.” One also made the following observation about the San Francisco County Superior Court located in the Hall of Justice. This setup they have down at the Hall of Justice is just a racket. The same people are always involved—same judge, prosecution, and public defender in the same courtroom all the time. Their loyalties and allegiance aren't to the client or anyone else but to the arrangement itself.Google Scholar

228 See also the interview quotation in the text following note 208 supra and the second interview quotation following note 257 infra.Google Scholar

229 By testing the consistency of the data with the description of the court's norms, I mean that I will use the standard scientific test of whether the data falsifies the hypothesis. In other words, if the distribution of the case data is consistent with the relationship predicted on the basis of the norms, then the null hypothesis that there is no relationship can be rejected. Note, however, that the tests used do not “prove” the existence of any particular causal relationship. It is always possible that the relationships observed in the case data could be caused by other factors than those discussed. Unfortunately, except in truly experimental research (i.e., involving random assignment of cases for control) it is generally not possible to do more than test the consistency of the data with the hypothesis. See generally Donald T. Campbell & Julian C. Stanley, Experimental and Quasi-Experimental Designs for Research (Chicago: Rand McNally & Co., 1963).Google Scholar

230 1 Crim. 12570. Probably the most aggressive enforcement of search and seizure law in the appeals studies occurred in 1 Crim. 12780, in which the defendant and his wife were convicted of fraud for selling diluted oxygen, one of the few white collar crimes found in the fiscal 1974 appeals. Although no motion to suppress was made in the trial court, the Court of Appeal held that the search warrant involved was overbroad. It also held that trial counsel (who was privately retained) was ineffective in failing to challenge the warrant. Probably the most serious case reversed on illegal search and seizure grounds was 1 Crim. 12447, in which the defendant was convicted of possession of a shotgun and assault with a deadly weapon. Purportedly responding to a police radio broadcast describing an armed robbery and indicating that the suspect was hiding in a basement storage area, police broke into a basement storage area and found defendant hiding under a mattress with a shotgun. There was a dispute as to whether the defendant had resisted arrest. The Court of Appeal held that the breaking into the storage area was without probable cause since there was no showing of any reasonable basis for the information allegedly broadcast on the police radio. In addition, the court found that the defendant had been deprived of due process because three different trial judges had refused to give him a hearing about his persistent complaints about the public defender appointed to represent him. Other search and seizure reversals related to excessive searches following traffic arrests (1 Crim. 12001, 1 Crim. 12063), violation of statutory requirements for searches of residences at night pursuant to § 844 of the Cal. Penal Code (West 1970) (1 Crim. 12231), and rejection of search warrants based on conclusory allegations that failed to adequately state the justification for the search (1 Crim. 12310, 1 Crim. 12526). All of these last five cases involved drug convictions.Google Scholar

231 The two confrontation cases were 1 Crim. 12006 and 1 Crim. 13166. In the former, the court ruled that the name of the informant who had provided information leading to a warrant search of defendant's flat had to be provided to the defendant in the circumstances of the case—because the informant was a material witness with regard to possession of the drugs and the defendant was entitled to know his identity so that he could be examined at trial. In the latter case, the court held that it was error to admit at trial a witness's out-of-court statement adverse to the defendant when the witness invoked the privilege against self-incrimination when called to testify, thus preventing the defendant from cross-examining him. The deficient Boykin warnings occurred in 1 Crim. 12621 (the appellant subsequently pleaded guilty and was sent to prison). And the speedy trial issue was in 1 Crim. 12751, in which an indictment accusing the defendant of furnishing drugs was issued in October 1969 but the defendant was not arrested until January 1972—despite the fact that the police knew his whereabouts during that entire period.Google Scholar

232 In 1 Crim. 12413 defendant was convicted of second-degree murder on an aiding and abetting theory and sentenced to prison. The case arose out of a fight between six or seven people in which one person had been shot. Two persons with the defendant pleaded guilty to misdemeanor voluntary manslaughter and were sentenced to jail. The conviction was reversed because the trial court had not instructed the jury on the showing of intent and knowledge required for aiding and abetting. This decision shows the indirect effect that doubt about factual guilt had on the Court of Appeal's decision. Although the court found that the defendant had “made a persuasive argument of the general weakness of the case against him,” the court declined to find that the evidence was legally insufficient for conviction, noting that “the Attorney General relies upon the well-established rule [i.e., the substantial evidence rule] that a judgment of conviction must be upheld on appeal unless it clearly appears that upon no hypothesis whatever is there sufficient substantial evidence to support the trier of fact's conclusion that the defendant is guilty.” Nevertheless, because of the “ambiguity of the evidence” and because “in no sense can the case be characterized as a particularly strong and convincing one,” the court found that the instruction error could not be harmless, and reversed. Note that the differing sentences imposed on the appellant and his two friends may also have inclined the court to reverse. In 1 Crim. 12601 defendant was convicted of selling LSD to undercover police. Before allegedly selling drugs to the undercover police, however, the defendant had them sign documents indicating that they were law enforcement officers and that the defendant was dealing in drugs for a law enforcement investigation. The Court of Appeal held that it was error not to instruct the jury that if the defendant was deputized he was immune from prosecution.Google Scholar

233 In 1 Crim. 12144 defendant was convicted of assault on a police officer. The Court of Appeal reversed for failure to instruct the jury on the lesser charge of assault or on the issue of whether the police action involved was lawful. The incident involved began when the defendant “popped a wheely” at an intersection and continued with a running brawl in which the defendant was pushed off his motorcycle by the police, hit with a flashlight and shotgun butt, and hit twice with a police car. The defendant claimed self-defense. The court may have concluded that the defendant had been punished enough. In 1 Crim. 13053 appellant had been convicted of possession of a blackjack because he had been showing a “nunchaku,” an oriental martial arts device, to some young people in the park. The Court of Appeal reversed because the jury had not been instructed on the definition of a blackjack and because the defendant had not been allowed to explain the circumstances involved.Google Scholar

234 In 1 Crim. 12619 appellant was convicted of lewd conduct with his nine-year-old stepdaughter, largely on the basis of the mother's testimony about what the daughter had told her. The daughter did not testify. The court held that this hearsay testimony was prejudicial in view of the fact that the mother appeared to bear ill will against the defendant, that the daughter had not told her mother about the alleged incident for six weeks, and that the daughter had a reputation for lying in school and for being “inclined to manufacture incidents to get attention.”. In 1 Crim. 12656 the use of hearsay statements by alleged conspirators was found improper in a case arising from a check forgery by several coroner's investigators who found a cashier's check on a corpse. Nevertheless, the court found that the error was harmful as to only one of three defendants, the one against whom the evidence of guilt was weaker.Google Scholar

235 1 Crim. 12005.Google Scholar

236 The case was 1 Crim. 12914. The Court of Appeal opinion in this case notes that the record indicated that jury members had been polled by the trial court following the newspaper incident and had been unable to say flatly that the article had not influenced them.Google Scholar

237 In 1 Crim. 12437 defendant, a school principal, was convicted of sodomy for having homosexual relations with a 19-year-old who worked for him. The Court of Appeal found that the “case is extremely close on the facts” because of disputed testimony and that a variety of trial errors were more critical because of the fact issues. The court found that the prosecutor had acted improperly by asking questions in bad faith for the purpose of asserting facts otherwise inadmissible and for cross-examining witnesses in excess of the scope of direct examination. The prosecutor also referred to a suppressed lie detector test. And the court found it was prejudicial to admit evidence that the defendant was gay for “identification” purposes. In 1 Crim. 12708, a conviction for selling heroin, the prosecutor elicited inadmissible evidence regarding defendant's prior record on four occasions during the trial. In addition, the trial court ruled that the defendant's prior conviction for selling narcotics was admissible for “impeachment,” thus effectively preventing the defendant from testifying; but the trial court refused to hold a hearing on the legality of the prior conviction. The Court of Appeal held that these errors “when considered cumulatively require a reversal,” noting that the entire case against the defendant consisted of the testimony of the one police officer who had acted under cover as a heroin buyer. In 1 Crim. 13080 appellant had been convicted of attempted rape and assault with intent to commit rape. The facts showed that defendant had pushed the victim into bushes at a bus stop and kissed and fondled her, but that he had run off when a motorist stopped. The prosecutor had told the jury that it would lack “intestinal fortitude” and would be “copping out” if it only found the defendant guilty of simple assault. Despite the absence of any objection by defendant's trial counsel, the court held that, in view of the close evidentiary question on the intent of the defendant, the prosecutor's comments were not harmless error.Google Scholar

238 Of course, even percentages based on 20 observations are highly unstable, especially in view of the fact that the distribution of issue topics is estimated as described in note 59 supra.Google Scholar

239 Measuring the incidence of legal errors poses much the same kind of problems for empirical research that measuring factual guilt does. In both cases, the empirical researcher has no way of knowing what the true incidence is; rather, the researcher can only quantify tangible though admittedly imperfect indicators of error or guilt. While it is tempting to second-guess the court about its determination of certain issues, it simply is not practical to justify the researcher's disagreement with the court in each instance without writing a series of notes of the law review type. As a result, I have chosen to restrict myself to an analysis of those instances in which the court itself recognized a legal error. Of course, there is a wide range of qualitative differences regarding the significance and obviousness of the errors recognized by the court. And of course it is possible that different judges would disagree to some extent over the errors recognized.Google Scholar

240 See Coffin, supra note 172, at 90, a trial judge's “failing to sanction a prosecutor for not backing up his opening [statement] with evidence would rarely be reversible error.” See also Johnson, Alan V. & Southard, Jeffrey S., Prosecutorial Misconduct in Closing Argument: Does Harmless Error Mean Never Having to Say “Reversed?” 49 J. Kan. B.A. 205 (1980).Google Scholar

241 Commentators on the law of evidence have observed the low likelihood that appellate courts will reverse trial verdicts on the basis of errors regarding the admission of evidence. See Richard O. Lempert & Stephen A. Saltzburg, A Modern Approach to Evidence: Text, Problems, Transcripts and Cases 2 (St. Paul: West Publishing Co., 1977):. [T]o the extent that rulings on evidence law determine the outcome of [litigation], [litigation] is almost always won or lost at the trial rather than the appellate level…. The reason for this is that appellate judges in reviewing cases seek generally to ensure that justice was done below; they are less concerned with upholding the letter of the law. Consequently, except to some extent when constitutional values are implicated, appellate courts will rarely reverse trial courts for mistaken rulings of evidence law unless they have some reason to suspect the substantive justice of the trial result.Google Scholar

242 There were no reversals based on admission of illegally obtained confessions in the 1974 appeals. As table 9 shows, only two confessions or defendants' statements were found to have been illegally obtained, and in both instances that error led only to a modification overturning one count of a multicount conviction. There are indications that such issues may have caused a number of reversals in the mid-1960s, however. See note 135 supra.Google Scholar

243 An instruction error on a basic element of the crime charged would arguably be reversible per se, but a more peripheral error could be harmless. See Traynor, supra note 188, at 73–74.Google Scholar

244 Similarly, it appears there were no reversals of convictions among the granted habeas corpus petitions, only modifications. See note 185 supra.Google Scholar

245 There are certain differences in the treatments of these issues that could affect the frequency of errors. For example, search and seizure issues, and to some extent confession issues, may be more likely to be decided in a hearing on the issue before trial, whereas issues of admissibility or of improper comments by the prosecutor tend to arise during trials, when the judge has less time to reflect upon his ruling. Note too that in some instances of harmless errors regarding prejudicial evidence or improper comments by prosecutors the trial judge did give a curative instruction to the jury to disregard the erroneous evidence or statement. Of course, it is difficult to evaluate how effective such curative instructions actually are; see Traynor, supra note 188, at 78–80.Google Scholar

246 Id. at 58–63 observes that errors in admitting confessions, and to a lesser extent admissions, are unlikely to be harmless.Google Scholar

247 See the interview quotation in the text following note 209 supra.Google Scholar

248 Hurst, supra note 35, at 446–49.Google Scholar

249 Id.Google Scholar

250 Compare this description to Howard's description of the “result orientation” he found among certain federal courts of appeals judges. Howard, supra note 2, at 128–29. Even one of the judges Howard labeled as having a “lawmaking” orientation said, “The first consideration is doing justice to the parties.”Id. at 133.Google Scholar

251 See also Coffin, supra note 172, at 147, for an example of how the court's decision of a specific legal issue can be affected by the court's view of the overall outcome appropriate to the case.Google Scholar

252 Compare the following comment by the Chief Justice of West Virginia:. Higher courts and lower courts are natural enemies; the lower courts always consider that they are “closer to the ground” and must respond to practical problems, such as lack of money, poor personnel, and inferior institutions, by bending the law in order to achieve a practical result in each case, while the higher courts believe that they are guardians of sacred principles and can never respond to practical problems without compromising their pristine principle-articulating function. I have seldom known a trial judge who thought well of the court above him; a majority of my own trial judges regard me as something between a fiend and a fool. Richard Neely, How Courts Govern America 162–63 (New Haven, Conn.: Yale University Press, 1981).Google Scholar

253 But, of course, supreme courts may also avoid reversals in particularly unsavory cases by declining to accept the case for hearing.Google Scholar

254 “Show case” trials that receive a good deal of media coverage may create an exceptional situation, however, in which an intermediate appellate court may feel more called upon to uphold principle. One attorney commented that the California courts of appeal seemed to be more inclined to reverse in cases that had received media coverage before reaching the court:. It's odd, but the court seems to be more inclined to reverse in notorious cases. Then they reverse on grounds that they wouldn't in ordinary cases. For example, the Inez Garcia case [a woman who killed a man who had raped her a short time before] or Ruchel McGee or Huey Newton or something. It's like the court wants to show that they are very proper and very tough minded then—aware of all the legal so-called technicalities. But it doesn't happen that way for the rest of the defendants. See also Friedman & Percival, supra note 204, at 311–16 for a discussion of the distorted image of criminal justice created by “show case” trials and cases.Google Scholar

255 See Davies, supra note 14, at 382.Google Scholar

256 Similarly, the particularistic orientation of the Court of Appeal and the absence of law-making concerns appear to contribute to the very low rate of dissent in the court, as described in note 146 supra. During interviews, the justices indicated that they did not believe that the work of the Court of Appeal brought out the ideological differences between the justices. One justice contrasted the low rate of dissent in the Court of Appeal to the much higher rate in the California Supreme Court as follows:. The low dissent rate here is partly a result of the hat we wear as justices of an intermediate court. I sit on the Supreme Court on occasion as a pro tern justice, and then I put on a different hat and questions of policy are more important. The routine nature of some cases, the limited time available for consideration of each case, and the close interaction among the judges in each division probably also reduce dissent. See Atkins, Burton M., Judicial Behavior and Tendencies Towards Conformity in a Three Member Small Group: A Case Study of Dissent Behavior on the U. S. Court of Appeals, 54 Soc. Sci. Q. 41 (1973).Google Scholar

257 Of course, the difference between the decisions of the Court of Appeal and the Supreme Court could also be caused by differing ideological orientations between the judges of the two levels of appellate courts. At the time of the study, however, the justices interviewed said either that there was no consistent ideological difference or that the difference was slight. One commented:. It is definitely true that a pattern exists where the Supreme Court reverses criminal convictions where the courts of appeal have affirmed. I think it is because they are evolving a more rigorous rule in their decisions. Our function is to find Supreme Court authority and apply it to a case where it exists. But their role is really much more of a legislative act. They have the role of making new law or changing older rules. So the pattern is because of the difference in functions between the courts. I don't know, but if you took all the court of appeal justices in the state—less than a hundred—and you were to somehow weigh everyone, it might be that the court of appeal judges would be a little more conservative than the Supreme Court justices on balance, but it is really very, very hard to say.Google Scholar

258 It is necessary to use the Supreme Court registers rather than published decisions because the Supreme Court in California formally reviews the decision of the trial court, not the court of appeal. As a result, the Supreme Court opinions rarely discuss the prior decision by the court of appeal. See note 160 supra.Google Scholar

259 I have confined my comparison of cases to those that the Supreme Court actually heard because the significance of the court's denial of a petition for hearing is ambiguous. It is undoubtedly true that the Supreme Court would have agreed with the decision of the Court of Appeal in the majority of cases in which hearing was denied, but it would be invalid to assume that the denial of a hearing necessarily meant that the Supreme Court was in agreement with the decision below. The judges interviewed all agreed that the Supreme Court did not (and could not) accept all of the cases in which a majority of its justices would be likely to disagree with the outcome of the case below. Rather, interviewees said that the Supreme Court grants a hearing only in those cases involving significant or unsettled policy issues. See also the comments of Justice Stanley Mosk of the California Supreme Court, Foreword: The Rule of Four in California, 63 Calif. L. Rev. 2 (1975).Google Scholar

260 There are only 129 cases because no prior court of appeal decision is recorded regarding many of the writ petition cases decided by the Supreme Court; in most of those instances, however, the writ was previously presented to a court of appeal but was summarily denied. Of course, there also was no prior court of appeal decision in cases within the exclusive jurisdiction of the Supreme Court, for example, death penalty cases or review of Public Utilities Commission decisions.Google Scholar

261 The data in the lower portion of table 10 are reconstructed from Yale Note, supra note 8, at 1202 & n.44, which presents sufficient percentages and margin data to reconstruct the full cross-tabulation. Of course, only some of the 16 states studied had two-tier appellate systems, and even those that did had them for only part of the 100-year period studied, hence the relatively small sample size (n= 240) compared to the much larger number of supreme court decisions that directly reviewed the decisions of trial courts in the study (n= 4,766).Google Scholar

262 Unlike the California data presented in the top part of table 10, the Yale Note treats reversed-in-part or modified cases as affirmances. Id. at 1197 n.26.Google Scholar

263 Note, however, that the Yale Note drew the opposite conclusion to mine from the data in part II of table 10. The note concludes:. Strikingly, the reversal rate differed dramatically depending on the IAC [intermediate appellate court] disposition of the lower court decision. In cases in which the IAC affirmed the lower court, the supreme court reversed the IAC 38.6% [i.e., 64 of 166] of the time; if the IAC reversed the lower court, the [supreme courts'] reversal rate [of the IAC] was 52.5% [i.e., 31 of 59]. Thus the supreme courts seem to side with the trial courts, a finding that was not expected and is hard to explain. Id. at 1202 (emphasis added). I think that this interpretation is not merely hard to explain but inappropriate because it is based on a very incomplete reading of the data. In choosing to focus only on the row percentages discussed in the quote above, the Yale Note looks only at a comparison of the rate at which the supreme courts disagreed with the intermediate appellate courts depending on whether the latter courts had affirmed or reversed. It is not overly surprising that these rates are different because it seems likely that the supreme courts might apply different criteria in deciding whether to accept a case for hearing depending on whether the intermediate appellate court had reversed or affirmed the trial court. I think that the distribution of the margin data is far more significant here because it shows that, after choosing the cases to be reviewed, the net effect of the supreme courts' review is to reverse more trial court decisions than the intermediate appellate courts had reversed.Google Scholar

264 It is sometimes suggested that supreme court decisions after intermediate appellate decisions are superfluous for the purpose of correcting the trial court. E.g., Shapiro, supra note 13, at 632: “The third opinion rendered by the Supreme Court is superfluous in checking the arbitrariness of trial judges: surely it is no more likely to be just than the judgment rendered by the [intermediate] appellate court.” The data in table 10 indicate otherwise. The “third opinion” of the supreme court is superfluous only if one assumes that the internal norms and perspectives of the intermediate appellate court are equivalent to those in the supreme court so that there is no systematic difference in the two courts' evaluation of cases. My argument and table 10 indicate that there are such systematic differences.Google Scholar

265 Doreen J. McBarnet, Conviction: Law, the State and the Construction of Justice 158 (London: Macmillan Press, 1981). For a comparison of McBarnet's views with the present research, see Davies, Thomas Y., Do Criminal Due Process Principles Make a Difference? A Review of McBarnet's Conviction: Law, the State, and the Construction of Justice, 1982 A.B.F. Res. J. 247.Google Scholar

266 44 Cal. 2d 434, 282 P.2d 905 (1955). Cahan created a search and seizure exclusionary rule in California.Google Scholar

267 See also the discussion of the effect of guilt on appellate decisions in Harris B. Steinberg, The Criminal Appeal, in Arthur A. Charpentier, ed., Counsel on Appeal 1 (New York: McGraw-Hill Book Co., 1968); Whitman Knapp, The Civil and the Criminal Appeal Compared, in id. at 63.Google Scholar

268 But see Justice Robert Jackson's comments on the relationship between due process standards and the gravity of the crime in his dissent in Brinegar v. United States, 338 U.S. 160, 183 (1949):. [I]f we are to make judicial exceptions to the Fourth Amendment …, it seems to me they should depend somewhat upon the gravity of the offense. If we assume, for example, that a child is kidnaped and the officers throw a roadblock about the neighborhood and search every outgong car, it would be a drastic and undiscriminating use of the search. The officers might be unable to show probable cause for searching any particular car. However, I should candidly strive hard to sustain such an action, executed fairly and in good faith, because it might be reasonable to subject travelers to that indignity if it was the only way to save a threatened life and detect a vicious crime. But I should not strain to sustain such a roadblock and universal search to salvage a few bottles of bourbon and catch a bootlegger.Google Scholar

269 Several studies have found higher than average reversal rates in crimes against public morality such as nonviolent sex offenses. For example, Yale Note, supra note 8, at 1210 & n.81, reports that the reversal rate in supreme court cases for sex crimes (excluding rape) was 51.7 percent compared to an overall average of 35.6 percent (id. at 1209). Similarly, Beattie, Criminal Appeals in California, 1964–1968, supra note 38, at 7, 15 table III, indicates that “sex offenses” showed somewhat higher reversal rates than most other offenses in fiscal years 1967–68 combined, but note that that study may include rape with other types of sex offenses.Google Scholar

270 The somewhat higher incidence of reversal in rape convictions may reflect the particularly problematic evidence in those cases. See Mather, supra note 69, at 43, 110. The two reversals of rape convictions in the fiscal 1974 appeals both involved conflicting testimony. Of course, the number of cases in this category is small and the percentage is quite unstable.Google Scholar

271 See note 69 and accompanying text supra.Google Scholar

272 Mather, supra note 69, at 66 table 5–1, indicates that “light” cases in which there is a dispute about the evidence of guilt (i.e., “reasonable doubt” cases) are one of the types of cases that may go to trial, whereas light “dead bang” cases will almost always end in a guilty plea.Google Scholar

273 See also Traynor, supra note 188, at 50, where the former Chief Justice of California observes:. The failure [of an appellate court) to [inquire whether an error affected the judgment in a case] is particularly likely in criminal cases. Appellate judges, persuaded by the record that the defendant committed some crime, are often reluctant to open the way to a new trial, given not only the risk of draining judicial resources but also the risk that a guilty defendant may go free.Google Scholar

274 See also the description of staff processing in the First District in Meador, supra note 12, at 209–16 (app. C).Google Scholar

275 See Molinari, supra note 38; Peters, supra note 38.Google Scholar

276 See Thompson, Mitigating the Damage, supra note 38.Google Scholar

277 Interviewees estimated that the justices disagreed with or changed the proposed staff opinion to some extent in approximately 5 percent of the staff-processed appeals. Most interviewees felt that this indicated that the screening process and staff processing were working satisfactorily, but others suggested that the low level of disagreement was a reflection of the superficial nature of the justices' review of the staff work product.Google Scholar

278 Interviewees indicated that a “by the court” opinion usually indicated that an appeal had been staff processed. However, there are several imperfections in this indicator. First, at the time of the study Division Three followed the practice of having a justice sign all opinions as the author, regardless of staff processing. As a result, it is not possible to measure the incidence of staff processing in Division Three. Since the same screening and case assignment process was used for all four divisions, however, it is reasonable to assume that Division Three probably received the same proportion of staff appeals as the other three divisions. Second, there were reportedly instances in which a staff-processed case was converted into a signed and sometimes even published opinion. There were also reportedly a few instances in which judge-processed opinions were issued by the court. In sum, it appears that a by the court opinion is generally a valid indicator of staff processing, but there will be instances in particular cases where it is misleading. The principal attorney for the court did maintain a list of all staff cases at the time of the study, but the court did not view that list as a public document and I was not permitted to review it.Google Scholar

279 See Davies, supra note 14.Google Scholar

280 Because Division Three at the time of study followed the practice of having an individual justice sign each opinion, regardless of whether it was staff processed or not, as discussed in note 278 supra, a simple count of by the court opinions underestimates the actual number of staff-processed cases by roughly one-quarter. To correct for this, the figures in the text represent the actual number of by the court opinions in the fiscal 1974 criminal appeals multiplied by 1.33. This same procedure was used to estimate the total number of staff-processed civil appeals, but note that the civil appeal figures are based on a 50 percent quasi-random sample as discussed in note 58 supra. Meador, supra note 12, at 214, used a similar approach to estimating the volume of staff-processed cases in the First District in 1972, but he apparently did not correct for Division Three's practice of signing all opinions. Thus, it appears that his estimate that 53 percent of all criminal appeals and 24 percent of all civil appeals were staff processed should be increased to approximately 70.6 percent and 32 percent, respectively.Google Scholar

281 See Davies, supra note 14, at 396–97. Compare to Howard, supra note 2, at 279–80, indicating that screening and staff processing in the federal courts of appeals he studied are most common in criminal appeals and in civil appeals involving social security cases, prisoner petitions, diversity jurisdiction cases (many of which are personal injury), and tax cases.Google Scholar

282 E.g., see generally Meador, supra note 12; Carrington et al., supra note 4.Google Scholar

283 Division Three appeals are excluded from this comparison since they cannot be classified by the type of opinion issued. See note 278 supra.Google Scholar

284 I.e., 544 § 9.2%= 50.Google Scholar

285 Division Three appeals are excluded from this comparison since they cannot be classified by the type of opinion issued. See note 278 supra.Google Scholar

286 In the fiscal 1974 cases, defendants initiated and won 8 appeals and 3 pretrial writ petitions on search and seizure issues for a total of 11 successes on that topic. The state initiated and won 5 appeals and 7 pretrial writ petitions on search issues, however, for a total of 12 successes.Google Scholar

287 Several of the successful prosecution appeals relate to reversals of orders quashing prosecutions. Significantly, the norms of review applicable to orders dismissing a prosecution before trial are considerably different from those that apply to judgments after trial. For example, neither the substantial evidence rule nor the harmless error rule has any application to an appeal from an order quashing a prosecution; instead the standard for review in that context is for the court to assume all factual premises in favor of the losing party, that is, the prosecution. There is no defendant counterpart of an appeal from a dismissal. By the same token, prosecution appeals after an acquittal are typically barred by the prohibition against double jeopardy. Thus, the difference in the types of issues raised in defendant and prosecution appeals probably contributes to the difference in outcomes; but it does not explain the prosecution's high rate of success in search and seizure issues as indicated in note 286 supra.Google Scholar

288 One reason for the persistence of complaints about criminals being freed because of technicalities is the tendency of many critics of due process to cast their discussion in terms of a parade of theoretical horribles and extreme cases rather than in terms of typical processes and outcomes. Fleming, supra note 1, is perhaps one of the more extreme examples of this type of discourse. Fleming discusses a host of theoretically possible abuses of due process, but the actual cases discussed are drawn primarily from either unrepresentative “showcase trials” or death penalty cases. (See note 254 supra.) What is interesting about Fleming's work in the context of the present study, however, is that he never discusses any of the norms of affirmance that apply to criminal appeals. The omission is striking because Fleming has been a justice of the Second District Court of Appeal.Google Scholar

289 See McBarnet, supra note 265.Google Scholar

290 For example, the social costs attributed to the exclusionary rule in political rhetoric seem to far exceed the rule's measurable effect on criminal prosecutions. See Bradley C. Canon, Ideology and Reality in the Debate over the Exclusionary Rule: A Conservative Argument for Its Retention, S. Tex. L.J. (forthcoming).Google Scholar

291 For example, Brian Forst, Judith Lucianovic, & Sarah J. Cox, What Happens After Arrest? A Court Perspective of Police Operations in the District of Columbia, PROMIS Research Project Pub. No. 4 (Washington, D.C.: Institute for Law and Social Research, 1977), a study of arrests in Washington, D.C., described the reasons prosecutors declined to prosecute arrests as follows:. [I]n the vast majority of all arrests rejected at the initial screening stage [by prosecutors], the prosecutor specified either a witness problem (such as failure to appear, refusal or reluctance to testify, and lack of credibility) or a problem connected with nontestimonial evidence (such as unavailable or insufficient scientific or physical evidence)…. Also evident … is a low rate of rejections at screening due to improper police conduct. Less than 1 percent of all arrests were refused by the prosecutor with an indication that the police failed to protect the arrestee's right to due process (e.g., no probable cause for making the arrest, unlawful search for or seizure of evidence, inadmissible confession or statement). And 77 percent of the 168 rejections that did occur at screening with the prosecutor indicating a violation of due process were in the victimless crimes group, primarily narcotics cases. Id. at 67–68. The same study also concluded that due process problems constitute a small share (1%) of the dismissals by prosecutors subsequent to the initiation of prosecutions (id. at 69 exhibit 5.2). Similarly, see Kathleen B. Brosi, A Cross-City Comparison of Felony Case Processing (Washington, D.C.: Institute for Law and Social Research, 1979), a study of the causes of the attrition of criminal prosecutions in five jurisdictions (District of Columbia, Los Angeles County, New Orleans, Salt Lake County, and Cobb County, Georgia), which concluded that all due process considerations explained only 1 to 9 percent of the potential cases that prosecutors declined to accept (id. at 19), and only 2 to 12 percent of the prosecutions that were dismissed in trial court (id. at 20 table 3). The author of the study concluded: “The exclusionary rule and other due process related issues have been the topic of much attention among criminal justice practitioners and scholars alike. While these issues may be substantial in terms of legal theory, they appear to have little impact on the overall flow of criminal cases after arrest” (id. at 18). She also concluded: “Most of the reasons [for the attrition of prosecutions] are witness or evidence related; few cases are dropped because of due process inadequacies” (id. at 25). And a Vera Institute of Justice report on the dispositions of felony arrests in New York City reached the following conclusion:. Are serious criminals “getting away with it” in the adult criminal court system and being turned loose or being given inappropriately light sentences? Our probe suggests that this is not the case, and that we may arrive at a different conclusion: Where crimes are serious, evidence is strong, and victims are willing to prosecute, felons with previous criminal histories ended up with relatively heavy sentences. One is thus tempted to conclude that if criminals are “getting away with it,” they may be getting away with it more on the streets than in the courtroom. Felony Arrests: Their Prosecution and Disposition in New York City's Courts 134 (New York: Vera Institute of Justice, 1977).Google Scholar

292 See text accompanying notes 240–41 supra.Google Scholar

293 See text accompanying notes 245–46 supra.Google Scholar

294 See text accompanying notes 193–95 supra.Google Scholar

295 Of course, a third goal that could be assigned to appellate review would be the correction of unduly harsh or inappropriate sentences. Unfortunately, that task cannot be evaluated in the context of the present study because of the extremely limited review of sentences under the indeterminate sentencing law, as discussed in the text accompanying notes 81–84 supra.Google Scholar

296 Cf. Frank, supra note 220, at 222–24, describing the “upper-court myth.”.Google Scholar

297 For a discussion of the relationship between appeal and the legitimacy of political systems, see Shapiro, supra note 13; see also Wold & Caldeira, supra note 5, at 346.Google Scholar

298 See, for example, Traynor, supra note 188, at 50, commenting on the dangers of an appellate court being too concerned with substantive justice in terms of the risk of freeing a guilty defendant:. The very reluctance of judges to confront such risks, however, serves to condone errors that may affect a judgment and thus engenders a still more serious risk, the risk of impairing the integrity of appellate review. Nothing is gained by running such a risk and much is lost. If appellate judges forthrightly opened the way to a new trial whenever a judgment was contaminated by error, there would be a cleansing effect on the trial process. A sharp appellate watch would in the long run deter error at the outset, thereby lessening the need of appeal and retrials. Cf. Neely, supra note 252, at 150–69, arguing that the due process revolution was essentially aimed at institutional reform of the criminal justice process. See also Bazelon, David L., Foreword—The Morality of the Criminal Law: Rights of the Accused, 72 J. Crim. L. & Criminology 1143 (1981).Google Scholar

299 For an expanded discussion of the environmental constraints on the Court of Appeal, see Davies, supra note 14, at 376–80.Google Scholar

300 Traynor, supra note 188, at 80–81, advocates the highly probable test for harmless error in order to increase appellate supervision over the trial courts:. The highly probable test proposed in this essay would set bounds to an appellate court's discretion to dismiss error as harmless whenever it was disposed to do so merely because it had no quarrel with the result in the trial court. Moreover, the test would deter an appellate court from sterilizing an error as harmless if there were consequent risk that its own tolerance would in the long run weaken vigilance against error in the trial courts. Concerned though the test is with the effect of an error on a judgment, it is also sensitive to what imprint an error leaves on the judicial process. Note, however, that Traynor's discussion of a highly probable standard actually incorporates more than simply an outcome-oriented threshold. Rather, in his emphasis on preserving the “integrity” of the judicial process and of assuring “justice” in addition to substantive correctness (id. at 17–19), Traynor argues that errors involving certain basic due process rights should be evaluated so stringently that, in that limited set of cases, the harmless error rule would seldom if ever be appropriately applied. See, e.g., his discussion of automatic reversal for errors involving ineffective assistance of counsel, denial of right to jury trial, jury bias, double jeopardy, confessions (id. at 57).Google Scholar

301 See note 191 supra.Google Scholar

302 See for example the reference in Coffin, supra note 172, at 117, describing the “plain error” exception to the abstention principle in the federal courts as “something so bad that the conviction must be reversed even though no one objected.” See generally, Wangerin, Paul T., “Plain Error” and “Fundamental Fairness”: Toward a Definition of Exceptions to the Rules of Procedural Default, 29 DePaul L. Rev. 753 (1980).Google Scholar

303 See Chapman v. California, 386 U.S. 18, 43 (1967), suggesting that a harmless error analysis is inappropriate to errors involving deprivation of the right to counsel, a position followed in Holloway v. Arkansas, 435 U.S. 475, 489 (1978). See also note 300 supra; Comment, Principles for the Application of the Harmless Error Standard, 41 U. Chi. L. Rev. 616 (1974).Google Scholar

304 See Dworkin, Roger B., Fact Style Adjudication and the Fourth Amendment: The Limits of Lawyering, 48 Ind. L.J. 329, 333–34, 364–68 (1973). See also text accompanying note 265 supra; Davies, supra note 265, at 261–63.Google Scholar

305 See text accompanying note 210 supra.Google Scholar

306 One of the more extreme examples of the use of a by the court opinion in a case addressing a serious and potentially frequent error occurred in the case described in note 222 supra. How could it be appropriate for the practice in that case to be the subject of an unsigned opinion apparently prepared by the staff?.Google Scholar

307 See note 226 supra.Google Scholar

308 See text accompanying notes 240–41 supra.Google Scholar

309 See text accompanying notes 179–81 supra.Google Scholar

310 See text following note 221 supra.Google Scholar

311 The Appellate Defender Office in California was not yet in operation at the time of the study. See Cal. Penal Code § 1240 (West 1982). See also Comment, Statewide Public Defender Organizations: An Appealing Alternative, 29 Stan. L. Rev. 157 (1976).Google Scholar

312 E.g., Osthus, supra note 25, at 14–16, would limit the issuance of writs by an intermediate appellate court to those ancillary to the exercise of its jurisdiction; John T. Wold, The California and the United States Courts of Appeals: Problems and Proposals (Washington, D.C.: Federal Justice Research Program, Office for Improvements in the Administration of Justice, United States Department of Justice, 1979), assessing the California writ system in terms of the burden it imposes on the courts of appeal.Google Scholar

313 Of course, the political feasibility of due process reform is always an uncertain matter, especially in California in the context of the passage of the Gann Initiative in the spring of 1982. That initiative, known as Proposition 8, is a simplistic but wholesale attack on a wide range of criminal justice procedures and due process standards. The meaning of virtually all of the provisions of the initiative is presently unclear. See Bob Taylor, Experts Differ in Assessing Impact of Gann Initiative, Cal. Law., July/Aug. 1982, at 46; Gerald F. Uelmen, Prop. 8 Casts Uncertainty over Vast Areas of Criminal Law, Cal. Law., July/Aug. 1982, at 43; Schrag, Peter, The Will of the People? Ballot Measures Like Propositions 13 and 8 Leave an Impossibly Difficult Mess, Cal. Law., July/Aug. 1982, at 51.Google Scholar

314 Cf. 1982 Cal. Jud. Council Ann. Rep. 59, recognizing that the low reversal rate in criminal appeals does not mean such appeals were frivolous:Google Scholar

The low percentage of reversals does not necessarily indicate that the appeals were generally unmeritorious. Affirmances include cases in which there was error deemed to be harmless under all the circumstances, and those in which significant legal questions were decided adversely to the appellant.Google Scholar