United States Ex Rel. Green v. Washington

FINDINGS OF FACT AND CONCLUSIONS OF LAW SHADUR, Senior District Judge. After the litigants had conducted the protracted discovery that was required for the full presentation of the issues in this class action brought under 42 U.S.C. § 2254 (“Section 2254”), this Court conducted an extended evidentiary hearing on September 26-29, October 2-8 and October 12, 1995 (the “Hearing”). Well after the completion of the Hearing, the parties simultaneously filed proposed findings of fact and conclusions of law, after which each filed a simultaneous response to the other side’s initial submission. Finally each side filed a reply, so that the matter has been fully briefed and is ready for disposition. In accordance with Fed.R.Civ.P. (“Rule”) 52(a), what follows are this Court’s Findings of Fact (“Findings”) and Conclusions of Law (“Conclusions”). 1 To the extent (if any) that the Findings as stated may be deemed conclusions of law, they shall also be considered Conclusions. In the same way, to the extent (if any) that matters later expressed as Conclusions may be deemed findings of fact, they shall also be considered Findings. In both of those respects, see Miller v. Fenton, 474 U.S. 104, 113-14, 106 S.Ct. 445, 451-452, 88 L.Ed.2d 405 (1985). Findings of Fact I. Description of the Petitioner Class 1. On January 24, 1994 this Court certified this action as a Section 2254 habeas corpus representative action. Petitioners are a class of prisoners in the custody of the Illinois Department of Corrections (“DOC”) (Petition ¶ 4): (a) each of whom has filed a currently pending appeal in the First District of the Illinois Appellate Court from a non-capital felony conviction in the Circuit Court of Cook County; *1241 (b) each of whom is represented in his or her appeal by the First District Office of the Office of the State Appellate Defender (“OSAD”); 2 (c) each of whom has been sentenced to serve no more than 20 years in prison; and (d) in each instance, as to whom either (1) his or her appeal has been pending for one year or more with no opening brief filed on his or her behalf, or (2) he or she has been advised that it will be more than one year after his or her notice of appeal before an opening brief will be filed in his or her ease. As of June 1, 1995 there were approximately 366 prisoners within the class as so defined (P.Ex. 104; Tr. 1031-32). II. OSAD Background Information A. OSAD’s Origin and Scope of Its Responsibilities 2. OSAD is a statewide indigent appellant defender agency that was established in 1972 with the goal of providing uniformly high quality appellate defense to indigent persons appealing their criminal convictions (Tr. 33-34). It was created pursuant to the Illinois State Appellate Defender Act (now 725 ILCS 105/1 to 105/11), which provides in part that it is OSAD’s duty to “represent indigent persons on appeal in criminal and delinquent minor proceedings, when appointed to do so by a court under a Supreme Court Rule or law of this State” (id. 105/10(a); see also 725 ILCS 5/121-13). 3. OSAD has been headed by Theodore Gottfried, Esq. (“Gottfried”) since its inception. Before becoming the State Appellate Defender, Gottfried was the Director of the Illinois Defender Project, and before that he was an assistant public defender in Cook County. Gottfried is an active member of the National Legal Aid and Defender Association (“NLADA”), having served on its Board of Directors, on its Executive Committee and as Chairman of its Defender Committee. During his involvement with NLADA Gottfried has worked on studies of various public defender offices nationwide and has assisted in developing policies for public defender offices. Gottfried has also written the criminal appeals chapter of the Illinois Institute for Continuing Legal Education series. Tr. 30-33. 4. OSAD has been recognized by NLA-DA and the American Bar Association (“ABA”) for its provision of public defender services. In addition Gottfried has received the Reginald Heber Smith Award from NLADA for his work in providing indigent defense. Tr. 32-33. Despite respondents’ cavil in this respect, it is fair to infer that OSAD has a national reputation as a well-qualified indigent appellate defender agency. 5. Statewide OSAD is organized into five district offices, corresponding to the five districts in the Illinois Appellate Court, with district offices in Chicago (First District), Elgin (Second District), Ottawa (Third District), Springfield (Fourth District) and Mount Vernon (Fifth District). OSAD also has a specialized Supreme Court Unit devoted solely to death penalty appellate litigation. Finally, its Administrative Office consists of three attorneys (the State Appellate Defender himself, the Legal Director and the First Assistant Appellate Defender) whose duties are primarily administrative and who are located in Springfield, Illinois. P.Ex. 109 at 12-15. 6. In Illinois counties other than Cook County, OSAD handles virtually every indigent appeal from criminal convictions (Tr. 36). 7. Both statewide and in First District, OSAD is also appointed from time to time to represent criminal defendants in appeals from proceedings filed in the Circuit Court pursuant to the Illinois Post-Conviction Hearing Act, 725 ILCS 5/122-1 to 122-7 (Tr. 34). Those appointments are made pursuant to Ill.S.Ct. Rule 651, which provides that if *1242 the Circuit Court determines that a Post-Conviction Hearing Act petitioner is indigent the Court “shall appoint counsel on appeal.” It is undisputed that in eases to which OSAD is appointed under Rule 651, it has no alternative but to handle those appointments (like all others) in a manner consistent with the Rules of Professional Responsibility (Tr. 108-13). In First District such post-conviction appeals make up slightly less than 20% of the OSAD caseload (Tr. 132). B. First District 8. First District handles appeals arising from criminal cases initiated in Cook County. It is headed by Michael Pelletier, Esq. (“Pel-letier”), who has held the position of Deputy Defender for First District since 1987. Pel-letier’s duties and responsibilities include hiring staff, assigning cases, overseeing the functioning of the office by monitoring work performance and handling the administration of the office (Tr. 119-22). 9. In Cook County, First District is appointed to represent indigent appellants in approximately 30 to 35% of the appeals from criminal convictions (Tr. 36). Of the indigent appeal cases to which First District is not appointed, the majority are handled by the Appeals Division of the Cook County Public Defender’s Office, the only other agency that handles indigent criminal appeals in Cook County (Tr. 130-31). 10. In addition to felony appeals, First District handles a small number of Class A misdemeanor convictions and appeals from juvenile delinquency findings. Misdemeanor appeals make up less than .2% of First District’s caseload. First District does not handle juvenile abuse and neglect cases, mental health commitment cases, Sexually Dangerous Persons Act cases, paternity cases or appeals in state habeas proceedings, and for some years it has not handled probation revocation appeals. Tr. 131-32. Although he has not studied the matter systematically, it is the considered view of State Appellate Defender Gottfried (and respondents have not disputed) that First District handles a tougher caseload than any other district office (Tr. 58-59). C. OSAD’s Funding, Appointments and Staffing I. Summary of Funding, Appointments, Staffing and Pending Unbriefed Cases in FYs 1990-95 II. OSAD is primarily funded on a statewide basis through an annual appropriation by the Illinois General Assembly (Tr. 45) for a State of Illinois fiscal year (“FY”) that commences on July 1 of the prior calendar year and ends on June 30 (so that, for example, “FY 1995” refers to the 12-month period ended June 30, 1995) (Tr. 46). This next table (based on P. Exs. 2 and 109) sets forth OSAD’s statewide budget, statewide number of appointments and statewide volume of pending unbriefed cases for each year from FY 1990 to FY 1995: OSAD Statewide FY Budget No. of Appts. Pending Unbriefed Cases — End of FY 1990 $5.722 million 1,593 908 1991 $6.307 million 1,866 1,055 1992 $6.079 million 2,046 1,284 1993 $4.996 million 1,929 1,512 1994 $5.180 million 2,073 1,987 1995 $6.285 million 2,211 2,188 And this next table (based on P. Exs. 2 and 3 and Tr. 126) sets forth First District’s staffing, appointments and pending unbriefed cases for the same FYs: *1243 First District [[Image here]] 12. In the FYs from 1984 through 1995 OSAD’s statewide responsibilities have increased steadily: In FY 1984 OSAD statewide was appointed to represent indigent appellants in 1,353 cases, while in FY 1995 OSAD was appointed statewide to represent indigent appellants in 2,211 cases (P. Exs. 2 and 109). 13. In dollar terms (not adjusted for inflation), OSAD’s statewide annual appropriation has increased from $3,611,182 in FY 1984 to $6,285,184 in FY 1995 (P.Ex. 2). But that increase has not produced significant increases in OSAD’s staff. On the contrary, in every one of the five OSAD district offices, the support staff has been reduced in FY 1995 from the support staff levels that existed in FY 1984 (id.), while there has been only a small increase in the total number of attorneys in the district offices: In FY 1984 there were 63 attorney slots in the five OSAD district offices, and that total had increased to only 67 in FY 1995. Indeed, even that increase is deceptive in terms of understanding the causes of the buildup of a backlog, for in the preceding year (FY 1994) there had been only 51 attorney slots in the district offices, 12 fewer than in FY 1984. P. Exs. 2 and 109. 14. In consequence of the constant increase in the total statewide volume of OSAD’s appellate appointments, as contrasted with the relatively static staffing levels, there have been alarming increases 3 in the statewide total number of pending unbriefed cases to which OSAD has been appointed. In FY 1984 OSAD had a total of 607 pending unbriefed cases in its statewide district offices, while that number had mushroomed to 2,188 as of the beginning of FY 1995 (P.Ex. 2). 2. History from, FY 1989 Through FY 1992 15.Each fiscal year the State Appellate Defender participates directly in the budgeting process for OSAD (a) by preparing a proposed budget request that is submitted to OSAD’s Board of Commissioners for approval, (b) by writing letters explaining OSAD’s budgetary needs to members of the General Assembly, (c) by seeking support from judges and bar associations and (d) by personally testifying before the appropriate committees of the General Assembly in support of OSAD’s budgetary needs (Tr. 46-48). For example, on March 18, 1988 Gottfried wrote to Illinois State Representative Peg Breslin, the legislative sponsor for OSAD’s budget request, explaining OSAD’s request for an appropriation for FY 1989 in the amount of $5,877,522 and pointing out that the number of OSAD’s staff attorneys had *1244 increased by only 1.4% from FY 1985 to FY 1987 despite the fact that OSAD statewide had been appointed to 19.4% more appeals in FY 1987 than in FY 1985, resulting in an increase of 22.7% in OSAD’s statewide volume of pending unbriefed cases during the course of FY 1987 (P.Ex. 16 and Tr. 47-49). 16. Despite that request and the reasons supporting it, OSAD’s appropriation from the General Assembly for FY 1989 was only $4,952,700 — nearly $1 million less than the amount requested from the General Assembly (P.Ex. 1 and Tr. 50). Understandably, during FY 1989 the volume of OSAD’s pending unbriefed cases statewide increased from 808 at the close of FY 1988 to 896 at the close of FY 1989 (P.Ex. 2 and Tr. 51). 17. By the end of FY 1991 OSAD’s volume of pending unbriefed cases had continued to increase — to 1,055, nearly double the volume of pending unbriefed cases as of the close of FY 1985 — as a result of further increases in the number of cases assigned to OSAD attorneys statewide without any corresponding increase in the size of OSAD’s staff (P. Exs. 2 and 22 and Tr. 52). 18. Despite its having been apprised of the growing problem (P.Ex. 22 and Tr. 52), the General Assembly actually reduced OSAD’s FY 1992 appropriation by approximately 3.8% from the FY 1991 appropriation (P.Ex. 2 and Tr. 53). As a result of that reduced appropriation, OSAD was forced to reduce its attorney staff statewide by six attorneys (P.Ex. 2 and Tr. 53). At the same time, during FY 1992 OSAD’s case appointments statewide rose to 2,046, about 180 more appointments than during the prior fiscal year, producing an increase in the statewide volume of pending unbriefed cases from 1,055 at the end of FY 1991 to 1,284 at the end of FY 1992 (P.Ex. 2 and Tr. 54). 19. As distressing as was that statewide pattern of an inadequate supply to meet an increasing demand, the trend of increasing case appointments from FY 1989 through FY 1992 was even more dramatic in First District than it was statewide. There OSAD’s appointments increased from 164 in FY 1989 to 582 in FY 1992. P.Ex. 4 at 411, P.Ex. 7 at 583 and Tr. 144-45. 3. Funding Reductions for FY 199S; Impact and Efforts To Deal with Reductions 20. Before the beginning of FY 1993 State Appellate Defender Gottfried took steps, including a personal appeal to House Speaker Michael J. Madigan (P.Ex. 26), to persuade the General Assembly that further cuts to OSAD’s funding would worsen the already unacceptable backlog of pending un-briefed cases (Tr. 55). Despite that and other efforts, the appropriation for FY 1993 actually reduced OSAD’s statewide budget by approximately 18% (OSAD’s total statewide funding for FY 1993 was $4,996,300), forcing OSAD to lay off 15 attorneys statewide (P. Exs. 1 and 8 and Tr. 56-57). 21. In First District the FY 1993 budget reductions forced OSAD to eliminate approximately four attorney positions and to cut the support staff approximately in half (P.Ex. 2 and Tr. 148-49). In addition to that reduced staffing, First District was forced to eliminate certain service contracts, to reduce its budget for travel to prisons and to limit the office’s ability to accept collect phone calls from clients (Tr. 149-50). 22. Because of the problems caused by underfunding (and consequent understaffing) at the same time that the number of assigned cases was increasing, in the FYs preceding FY 1993 First District had suffered an increase in the number of assigned eases in which no opening brief had been filed. That in turn meant that an increasing number of those cases had completed records but were awaiting the assignment of a First District lawyer to begin working on the case. Thus at the beginning of FY 1991 there were 194 pending unbriefed cases in First District, 71 of which had completed records and were awaiting briefing. At the beginning of FY 1992 there were 436 pending unbriefed cases, 93 of which had completed records. By the beginning of FY 1993 there were 560 pending unbriefed cases, 192 of which had completed records. P.Ex. 3 (June 1990, June 1991 and June 1992) and Tr. 141-42. 23. That upward trend had developed during a period when First District had approximately 20 lawyers on staff. It was thus *1245 obvious that if no action were taken to address the problem, the budget cuts imposed on OSAD in FY 1993 would have a dramatic adverse effect on the First District backlog of pending unbriefed cases (Tr. 148-52). In an effort to address the impending crisis in First District stemming from the FY 1993 budget reductions, Pelletier took the four separate actions described in Findings 24 to 27. 24. In May 1992 Pelletier wrote to the Chief Judge of the Criminal Division of the Circuit Court of Cook County to ask that, in light of the impending budget reduction and staff layoffs, the Circuit Court Judges not appoint First District to handle any more appeals (P.Ex. 113 and Tr. 150-51). As a result of that letter, appointments to First District were reduced but not stopped altogether (Tr. 151-52): It was appointed to only 358 new eases during FY 1993, over 200 fewer appointments than in the prior FY (P. Exs. 7 and 8 and Tr. 145-46). 25. In June 1992 and October 1992 First District filed motions to withdraw as counsel on appeal in approximately 80 cases to which it had been appointed. Those motions advised the court of OSAD’s funding reductions for FY 1993 and reported the then-existing volume of First District’s pending unbriefed cases. Each motion stated that OSAD could not remain as counsel without impairing the client’s rights to a timely appeal and to the effective assistance of counsel on direct appeal. Tr. 153-55 and P.Ex. 42. Without exception the motions to withdraw were denied (P.Ex. 43 and Tr. 155). Then when no additional funding for OSAD was forthcoming out of the General Assembly’s fall 1992 session, First District filed a second round of motions for leave to withdraw on the same basis as the earlier motions. Approximately 45 such motions were filed after obtaining permission of the clients. All of those motions were also denied. Tr. 153-55. 26. In January 1993 (midway into FY 1993) the State Appellate Defender transferred cases among the five OSAD district offices in an effort to equalize the backlog of pending unbriefed cases. Approximately 115 to 125 cases were transferred out of First District, which had the most severe backlog among OSAD’s district offices, to other districts. P.Ex. 8 at 587 and Tr. 155, 60. Despite those transfers First District’s backlog remained at too high a level for the prompt effective representation of the Office’s clients (P.Ex. 2 and Tr. 62). In FY 1993 OSAD was appointed to a statewide total of 1,929 cases (nearly 500 more appointments than the Office had received in FY 1989, the last prior fiscal year in which its budget had been below $5 million) (P.Ex. 2 and Tr. 57). During FY 1993 OSAD’s statewide volume of pending unbriefed eases increased from 1,284 at the close of FY 1992 to 1,512 at the close of FY 1993 (P.Ex. 2 and Tr. 57-58). As a consequence, no additional transfers of cases out of First District to other district offices would have been possible without increasing the backlog of pending unbriefed cases in those district offices (P.Ex. 2, P.Ex. 8 at 587 and Tr. 62). 27. In addition First District also worked with the Chairman of the Judicial Advisory Council of Cook County to push for pro bono representation of some of the clients in First District’s backlog. Although the Judicial Advisory Council established a pro bono committee and a small program to accept some indigent criminal appeals both from First District and from the Appeals Division of the Cook County Public Defender’s Office, that program had only a negligible impact on First District’s caseload problem. In the three years that the program has been in existence, it has been able to accept only a total of 30 cases from First District. Tr. 157-58. 28. Primarily as a result of the reduced level of appointments in FY 1993 and the transfer of cases to other districts in the middle of FY 1993, First District’s volume of pending unbriefed eases was reduced to 417 at the beginning of FY 1994. Neither the reduced appointments nor the transfers, however, produced a reduction in the number of First District cases awaiting briefing with completed records. During FY 1993 that number increased from 192 at the beginning to 240 at the end of the FY. P.Ex. 3 (June 1992 and June 1993) and Tr. 142. *1246 4. OSAD Funding, Appointments and Pending Unbriefed Cases Since the FY 199S Funding Reductions 29. In FY 1994 OSAD received a 3.69% increase in its appropriation above the FY 1993 level, to the sum of $5,180,657. That amount was still well below pre-1992 levels and did not enable OSAD to hire any additional staff attorneys to make up for the FY 1993 staff reductions. P.Ex. 1 and Tr. 63. 30. Duririg FY 1994 OSAD was appointed statewide in a total of 2,073 cases, about 150 more cases than in FY 1993. At the end of FY 1994 OSAD’s statewide volume of pending unbriefed cases had reached 1,987, up from 1,512 at the end of FY 1993. P.Ex. 2 and Tr. 63-64. As Finding 31 reflects, in First District there were dramatic increases in both appointments and backlog. 31. In December 1993 Cook County Public Defender Rita Fry wrote to Chief Judge Fitzgerald to complain that the Appeals Division of the Cook County Public Defender’s Office was being overwhelmed with an unprecedented number of appointments as a result of the reduction of appointments to First District that had occurred in response to Pelletier’s May 1992 request. Ms. Fry’s letter requested that the Chief Judge appoint the Cook County Public Defender’s Appeals Division in fewer cases. Tr. 758, 780. Thereafter the Circuit Court judges began to appoint First District in more cases and the Appeals Division of the Cook County Public Defender in correspondingly fewer cases. In FY 1994 First District was appointed in 542 cases, nearly 200 more appointments than in FY 1993, despite the fact that its attorney staffing remained unchanged. P.Ex. 9 and Tr. 146, 758, 780. As a result, at the end of FY 1994 there were 604 pending unbriefed eases in First District, 349 of which had completed records and were awaiting briefing. Tr. 143 and P.Ex. 3 (June 30, 1994). 32. In FY 1995 OSAD’s statewide appropriation was increased to $6,285,184, approximately the same dollar amount (not adjusted for inflation) as the FY 1991 appropriation. That increase enabled OSAD to hire additional attorneys in all of its district offices so as to achieve staffing levels close to the FY 1991 attorney staffing levels — but with substantially reduced support staffs. Because OSAD’s 1991 personnel had been unable to keep pace with the statewide volume of appointments in FY 1991 and because OSAD’s 1995 appointments (2,211) significantly exceeded the level of appointments in 1991 (1,866), the restoration of funding to OSAD in FY 1995 was clearly insufficient to enable OSAD to keep pace with its responsibilities. P.Ex. 2, P.Ex. 109 at 26 and Tr. 65-68. 33. As a result of the restoration of OSAD’s funding to pre-1992 levels, First District was able to hire four additional lawyers at the beginning of FY 1995 (Tr. 146-47). But during FY 1995 First District was appointed to an all-time high of 598 eases (P.Ex. 109 and Tr. 147). Thus despite the increase in staff, OSAD’s volume of pending unbriefed eases continued to grow during FY 1995. As of April 30, 1995 there were 684 pending unbriefed cases in First District, 439 of which had completed records and were awaiting the availability of a lawyer for briefing. Tr. 143 and P.Ex. 3 (April 1995). 5. Status of First District as of the Hearing 34. If trends as of the time of the Hearing were to continue, First District would likely receive more than 700 new appointments in FY 1996 — the largest volume of new appointments ever (see P.Ex. 96). But OSAD had not then received any increase in its appropriation for FY 1996 above the amount received for FY 1995 (Tr. 68). As of August 31, 1995 there were 758 pending un-briefed cases in First District. As of the time of the Hearing, First District’s volume of pending unbriefed cases was at 765 (the highest level ever) (Tr. 123), 522 of which had complete records and were awaiting briefing (P. Exs. 50 and 50(a)). 35. Findings 56 and 57 describe OSAD’s policy of briefing oldest cases first. Under that policy First District attorneys were, as of the late September 1995 Hearing date, working only on cases to which First District had been appointed during or before March 1994. Thus all of the cases that the First District attorneys were engaged in briefing as of the time of the Hearing had already *1247 been pending on appeal, with no opening brief filed, for at least 18 months. Tr. 165-66. 36. In accordance with OSAD’s statewide office policy referred to in Findings 35, 56 and 57, the 765 cases pending in First District as of the time of the Hearing are being-handled on a first-in-first-out basis, under which the oldest cases will be assigned for briefing first. Pelletier anticipates that each lawyer on his staff will file the opening substantive document — a brief, a dispositive motion or an “Anders motion” (a brief and motion under Anders v. California, 386 U.S. 738, 87 S.Ct. 1396, 18 L.Ed.2d 493 (1967)) — in approximately 20 cases each year, or the equivalent of 1.6 dispositions per attorney per month. Tr. 167-69. As of the time of the Hearing there were 19.50 full-time equivalent attorneys (taking account of part-time attorneys) on staff in First District (Tr. 126). 37. On the assumptions (a) that First District will continue with the equivalent of 19.50 full-time attorneys on staff, (b) that each of them can be expected to prepare 1.6 opening substantive documents per month and (e) that no pending cases are eliminated from the docket, Pelletier anticipates that it will take close to two years before an Assistant Appellate Defender will be available to begin working on the newest cases to which First District had been appointed as of the time of the Hearing (Tr. 169-71). Because of the length of time that it has taken and continues to take before a First' District attorney becomes available to begin work on a newly-assigned pending case, the length of time that it takes to prepare the record on appeal has not contributed to delays in the processing of First District cases. As of the time of the Hearing it took approximately six to eight months for First District to received completed records in its office. Tr. 219. 38. As part of its continuing effort to reduce the number of pending unbriefed cases, First District was at the time of the Hearing in the process of transferring 50 eases to OSAD’s Springfield Office (Tr. 68, 125). D. First District’s Success Rate 39. OSAD counts an appeal as “successful” if the appeal results in any modification of the trial court’s judgment or sentence or in any opportunity being afforded the appellant to obtain such a modification through further proceedings in the trial court (Tr. 184). In FY 1995 First District was successful, in that sense, in 30% of the cases that it handled on direct appeal (Tr. 99, 184). In FY 1994 First District had been successful in the same sense in 24% of its cases (Tr. 185). 40. Pelletier was personally aware of situations in which appellant clients of First District had completed their terms of incarceration and had been released before the First Appellate District had rendered decisions in their appeals, including instances in which those decisions had reversed the convictions' outright. 4 Although Respondent’s Response 2 has disputed the sufficiency of the Hearing record to demonstrate the frequency of such occurrences, it is surely an appropriate matter for judicial notice that given Illinois’ system of day-for-day credit for an inmate’s good conduct in prison, 5 an inmate who does not forfeit such credit and whose appeal will not have been decided until (say) 2% years have elapsed since his incarceration — an overly optimistic assumption — will have been released before that decision if his sentence had been five years or less. On a more realistic assumption of three years between notice of appeal and decision, any defendant with a prescribed sentence of six years of less and with no forfeiture of good time credits will have served his or her entire custodial sentence and been released before *1248 the decision is reached. And of course any decision that reverses a conviction means that the period of time that has been spent in serving the invalid sentence has been imper-missibly lengthened by any undue delay in the handling of the appeal. III. Response by the State of Illinois and the Illinois Courts During the Pen-dency of This Litigation 41. As heretofore and hereafter set forth, the Illinois legislature has been on notice of the severity of OSAD’s backlog of unbriefed cases for several years. In the spring of 1992, immediately prior to the time that the General Assembly reduced OSAD’s funding by 18%, the Speaker of the Illinois House was apprised that the backlog was most severe in OSAD’s First District Office and that further cuts would render competent and effective representation of First District clients impossible. P.Ex. 24 and Tr. 152-53. Since the filing of this action, the General Assembly has restored OSAD’s funding to close to the pre-FY 1991 level that existed prior to the FY 1993 budget cuts, but that funding level is plainly insufficient to enable OSAD to employ enough staff to address the growth in appointments and backlog that OSAD’s district offices have experienced in the last three years. 42. Less than a month ago (on January 30, 1996) Illinois Supreme Court Chief Justice Michael Bilandic transmitted the Court’s annual report to the General Assembly. It read in relevant part: The General Assembly in the past has recognized the importance of providing adequate resources to assure fair and timely handling of appeals from convictions of capital and other felony defendants. [This quotation omits the portion of the report dealing with death penalty appeals] The lack of resources for appeals by other criminal defendants has also become a critical problem. The General Assembly took a constructive step to address this problem in 1994. At that time, funds were appropriated to restore attorneys to the staff of the Office of the State Appellate Defender. However, this improvement was not fully maintained for the current fiscal year. A comparison of the staffing levels of the state appellate defender with the trend in cases to which the office is appointed shows a sharp contrast. For fiscal year 1990 the office had 72 attorneys assigned to Appellate Court cases; currently there are only 60. During the same six-year period the number of new cases to which the office was appointed on an annual basis increased from 1,593 to 2,211. This means that the number of attorneys decreased by 17 percent while the number of new appeals assigned annually increased by 39 percent. The consequence of these diverging trends is that the period of time needed for appellate defender attorneys to file a brief on a new case has lengthened from about 11 months to the current 19 months. The Federal District Court of Northern Illinois has determined that such lengthy delays deny due process to indigent defendants represented by the state appellate defender and the Cook County public defender in the 1st District of the Appellate Court. (U.S. ex rel. Green v. Peters, No. 93 C 7300, U.S.Dist.Ct., N.D.I11.) Further action by the District Court on this matter is anticipated in the near future. As noted in my message to the General Assembly in January 1994, under-staffing of criminal appeals often leads to more retrials and appeals claiming violation of constitutional rights because of “ineffective assistance of counsel.” We urge the General Assembly to renew its commendable previous efforts to restore the number of attorneys for the state appellate defender to levels sufficient to provide fair and prompt representation on criminal appeals. As indicated by earlier Findings, that report was more charitable in its reference to the General Assembly’s prior funding provisions than the facts would warrant, but that is entirely understandable in a message that seeks to obtain an appropriation in an era of governmental cutbacks. 6 It is of course too *1249 early to know whether the Supreme Court’s request will bear fruit. 43. As directed by this Court shortly after this action was commenced, petitioners’ counsel apprised the First District of the Illinois Appellate Court of the pendency of this litigation in a November 24, 1993 letter .directed to that Court. Counsel’s letter enclosed copies of individual habeas corpus petitions that had been filed by Hilberto Sanchez, Vernon Joy and Rodney Odell complaining of appellate delay in terms identical to the pending class action petition filed here by Mervin Green. Counsel’s letter communicated this Court’s request for information as to indigent criminal appellate case processing in the First Appellate District and as to whether private counsel could be appointed for petitioners Sanchez, Joy and Odell. 44. Pelletier has also advised the Illinois Appellate Court of the pendency of this litigation. On November 8, 1993 he wrote letters to the Presiding Justices in the divisions of the First Appellate District to which the Sanchez, Joy and Odell cases were assigned (P. Exs. 37, 38 and 39 and Tr. 162-64). Those letters reminded the Court that OSAD had filed motions to withdraw as counsel in each of the three cases because of the volume of pending unbriefed cases in First District, once again requesting that outside counsel be .appointed to represent the petitioners. Pel-letier’s letter also stated that First District had many other clients similarly situated to petitioners Sanchez, Joy and Odell and asked that action be taken for those other clients as well. Tr. 162-65. At some time thereafter private pro bono counsel were indeed obtained to represent appellants Sanchez, Joy and Odell. After such pro bono counsel had been obtained, the First Appellate District permitted First District to withdraw from those cases. Other than in the Sanchez, Joy and Odell cases, the First Appellate District has granted only one other motion to withdraw as counsel based upon First District’s backlog of unbriefed cases. Tr. 161-65. 45. On March 11,1994 the Justices of the First Appellate District responded to this Court with a letter (P.Ex. 40) that acknowledged that the Appellate Court “is aware of the problem” of excessive appellate delay for First District’s clients. In the letter the First Appellate District made several observations as to indigent criminal appellate case processing, stating in essence: (a) According to the Court’s informal research, OSAD then had larger backlogs in other districts than in First District, but motions to withdraw had not been filed in the other districts. (b) First District had filed proportionately fewer motions to withdraw under Anders v. California than the Appeals Division of the Cook County Public Defender, with the Court suggesting that First District may be “refusing to exercise discretion” to file Anders motions. (c) First District’s policy of briefing oldest cases first, counting “from the date of conviction or sentencing,” “gives preference to post-conviction cases.” (d) First District had filed fewer briefs per attorney than OSAD’s other district offices. One of the Justices of the Second District Appellate Court then expected OSAD attorneys to file 30 briefs per year per lawyer in Second District cases. (e) Based on a communication with Renee Goldfarb (“Goldfarb”), Chief of the Appeals Division in the Cook County State’s Attorney’s Office, it was suggested that OSAD’s success rate may be 15% rather than the higher success rate reflected in OSAD’s Annual Reports. (f) People v. Thurston, 255 Ill.App.3d 512, 515-16, 193 Ill.Dec. 393, 395, 626 N.E.2d 426, 428 (2d Dist.1994) had opined that “the appellate defender should reexamine his policy of processing such eases” (that is, cases that the court viewed as not raising substantial questions). In addition, the First Appellate District’s letter stated that it has “the inherent power to appoint private counsel to represent indigents in criminal eases.” It noted that the Chairman of the Judicial Advisory Council of Cook County has been recruiting pro bono lawyers to handle some indigent criminal appeals and expressed the hope “that we will *1250 be able to expand the volunteer program so that it will not be necessary to compel lawyers to represent indigent defendants on appeal without compensation.” There was no reference to any other possible solutions to the problem of appellate delay for First District clients. IV. Nature of OSAD’s Performance A. OSAD’s Workload Standards 1. Standards Themselves 46. OSAD has established a work standard that requires all of its attorneys to meet a minimum level of productivity. That standard, set forth in OSAD’s 1994 Employees’ Manual (P.Ex. 11), requires each assistant appellate defender with one year of service to complete, during each year, 24 “brief units” — a term defined as an appellate court brief in a direct appeal from a judgment entered following a criminal trial, in which the record on appeal is not less than 250 pages and not more than 500 pages. To refine that concept to reflect various relevant variables, a weighted brief-unit credit value is assigned to various tasks performed by OSAD’s lawyers (opening brief, reply brief, certain motions, etc.) and, as to the opening brief itself, the credit is weighted in terms of the length of the record (more than one brief unit is attributed to records greater than 500 pages, and less than one brief unit is attributed to records under 250 pages). P.Ex. 11 and Tr. 42,171-75. 2. Expert Testimony as to OSAD’s Workload Standards 48. Petitioners called two expert witnesses who opined as to OSAD’s work standards and as to the productivity of First District attorneys. Both experts concluded that OSAD’s attorneys must work hard to achieve the level of productivity called for by OSAD’s work standard. Respondents presented no evidence contesting those conclusions, and this Court credits the expert -witnesses’ testimony as summarized in Findings 49 through 53. 49. Indiana University School of Law Dean Norman Lefstein, who also now chairs the Indiana Public Defender Commission, testified as to the role of defense counsel in criminal proceedings, especially as it pertains to professional responsibility issues and criminal justice standards. Dean Lefstein has taught, written and lectured extensively on professional ethics and criminal justice issues. In addition to having served in the past as director of the Public Defender Service for the District of Columbia, he has been the reporter to the ABA Commission that promulgated the second edition of the ABA’s Standards for Criminal Justice. In that capacity, he authored the ABA’s standards on the prosecution and the defense functions, on providing defense services and on guilty pleas. Dean Lefstein has frequently been qualified as an expert witness on issues of professional responsibility and competence of representation in criminal matters. Dean Lefstein testified in these proceedings without compensation. Without objection from respondents, this Court found Dean Lefstein qualified as an expert in this area. Tr. 223-31. 50. Dean Lefstein opined that no more than approximately 25 indigent felony criminal appeals should be assigned to a single attorney in any one calendar year. That opinion, which was unrebutted by respondents, was based in part upon a 1992 NLADA publication titled Indigent Defense Caseloads and Common Sense, and was also informed by Dean Lefstein’s own experience in observing the work of the extremely well-qualified appellate lawyers under Dean Lefstein’s supervision in the Public Defender Service for the District of Columbia. P.Ex. 59 and Tr. 233-36. It is the thrust of Dean Lefstein’s opinion, and this Court finds, that the assignment of significantly more than 25 cases of average complexity to one attorney in a single calendar year would create an unacceptably high risk that the attorney would be unable to brief the cases competently within a reasonable period of time. 51. Robert Spangenberg (“Spangen-berg”) was petitioners’ other expert witness, this time as to the assessment of indigent defender systems. Spangenberg has studied indigent defense delivery both at the trial and the appellate level throughout the nation. *1251 For that purpose he has visited hundreds, of public defender offices throughout the country and has studied public defender offices in over 30 states in the last 20 years, having extensively analyzed the functioning of the appellate defender 'systems in California, Florida, Colorado, Wisconsin and Minnesota, among many other jurisdictions. In particular, under the auspices of the ABA’s Bar Information Program Spangenberg has consulted with numerous public defender offices regarding ways to improve the delivery of indigent defense services. Spangenberg was project director for the Department of Justice’s Bureau of Justice Statistics National Analysis of Criminal Defense Systems, the first exhaustive analysis of indigent defense systems performed in the United States. He has testified before courts as an expert witness and before numerous governmental bodies regarding indigent defense delivery. Tr. 291-301. As it did with Dean Lefstein, this Court found Spangenberg qualified as an expert in his area of testimony without objection by respondents. Tr. 302-04. 52. In preparation for his testimony here, Spangenberg performed an extensive analysis of First District’s functioning. For that purpose his assessment included review of OSAD’s Employee Manual (P.Ex. 11) and interviews with all of First District’s professional staff and with Gottfried. Tr. 306-08. 53. In Spangenberg’s unrebutted expert opinion, OSAD’s work standard constitutes a measure of attorney productivity that requires OSAD attorneys to work up to capacity in order to provide quality representation to the required number of clients, although in some respects it requires attorneys to forgo some aspects of optimum client representation. Tr. 371-73. OSAD’s standard was modeled upon the 1980 NLADA Standards for Appellate Defender Offices (P.Ex. 115), which have been a national baseline in defining the appropriate level of productivity for an appellate defender. NLADA’s Standards impose a significantly less demanding productivity requirement upon appellate defenders than does the OSAD work standard: NLADA’s Standards require appellate defenders to achieve 20 “work units” per year (in contrast to the 24 “brief units” required by OSAD’s work standard) and define work units in such a way as to allocate more work unit credit to certain tasks and certain types of records than does the OSAD brief-credit definition in the OSAD work standard. P. Exs. 11 and 115 and Tr. 574-79. B. OSAD Client Representation and Briefing Policy 1. OSAD’s Briefing Policy 54. OSAD’s Employee Manual (P.Ex. 11) also sets forth office-wide policies and procedures, including OSAD’s briefing policy, relating to effective client representation. After OSAD is appointed to a matter by the Circuit Court, it sends the represented person a form letter that attaches a description of the stages of the appellate process (Tr. 133-34). In part' the attachment describes the length of time that it generally takes to complete the record on appeal and to file the opening brief. In that respect clients of First District have been and are .presently being informed that they can expect to wait 16 to 22 months (many months after the filing of the record) before the opening briefs in their cases will be filed. P.Ex. 73 and Tr. 133-34, 158-59. 55. OSAD is professionally obligated — by its enabling statute (725 ILCS 105/1 to 105/11), by Illinois Supreme Court Rules and by the Rules of Professional Responsibility that bind its attorneys — to provide representation in every one of the cases to which OSAD is appointed. It is not possible for OSAD to decline to “process” a ease simply because someone within or outside of OSAD might believe that other matters are more deserving of the attention of OSAD’s lawyers (Tr. 73 and P.Ex. 32 at 5). This Court of course appreciates (and shares) the concerns sometimes expressed by Illinois appellate opinions as to the potential waste of scarce resources in addressing legal issues of minimal or even no significance — indeed, this Court itself regularly encounters the trial court equivalent of that problem in its having to address pro se filings (most frequently by persons in custody claiming what they may mistakenly perceive as violations of their constitutional rights). But just as this Court must expend its own resources (which could *1252 far more profitably be devoted to more constructive activity) in determining whether such pro se filings may be dismissed at the outset on grounds of “frivolousness” in the legal sense (see Neitzke v. Williams, 490 U.S. 319, 109 S.Ct. 1827, 104 L.Ed.2d 338 (1989) and Denton v. Hernandez, 504 U.S. 25, 112 S.Ct. 1728, 118 L.Ed.2d 340 (1992)), so the appointed OSAD lawyer is duty bound to review the record once he or she is appointed and then to file either an Anders brief or a brief on the merits — and those tasks alone require what may fairly be labeled as a waste of resources. It must be remembered that criminal appeals are not taken by the appellate defender but by the convicted defendant who has an unrestricted right to do so (just as, except in the most egregious eases of repetitive groundless litigation, this Court cannot forbid filings by the pro se litigant). In that sense OSAD attorneys should not suffer the fate of the Persian messenger for having to discharge their professional responsibilities by handling the cases that the courts have assigned to them. 7 56. As a matter of statewide agency policy, OSAD attorneys are assigned cases for briefing based upon the date of the Circuit Court’s final disposition of the case — which in an initial direct appeal is typically the date of sentencing — with the oldest eases being-placed first in line for briefing. As with the FIFO method of valuing inventory for accounting purposes, that is appropriately referred to as a “first-in-first-out” method of assigning cases for briefing. Tr. 37-38, 137. 57. In appeals of proceedings commenced under the Illinois Post-Conviction Hearing Act, the date of the Circuit Court’s final disposition of the post-conviction proceeding — and not the date of the original sentencing — is treated as the final disposition date for purposes of the assignment method described in Finding 55 (Tr. 137). Hence the concern expressed in the First Appellate District’s March 11,1994 letter (P.Ex. 40), which stated that OSAD’s briefing policies give preference to post-conviction appeals, was based on the Court’s mistaken belief that in those appeals the original sentencing date is used to place such appeals in line for briefing. In fact, post-conviction appeals are placed in line among direct appeals on a straightforward date-of-disposition basis, with no priority being given to either category of appeals. 58. If First District attorneys were to reserve the briefing of all post-conviction cases until the attorneys had discharged their responsibilities in direct appeal cases, the practical result would be that post-conviction appeals would never be briefed, because First District is appointed to more direct appeal cases than its attorneys can competently handle each year (P.Ex. 3 and Tr. 37-38). Even an extended delay, rather than such a total cessation, in the handling of such post-conviction cases in order to provide preferential treatment to other First District clients (those taking direct appeals) would also create ethical problems of the nature dealt with in Findings 60 through 63. 2. Expert Testimony as to Briefing Policy 59. Only Dean Lefstein among the Hearing witnesses addressed the ethical and professional responsibility implications of the first-in-first-out priority system followed by OSAD as against other possible alternatives. Findings 60 through 63 contain his unrebut-ted opinions in those respects, and Finding 69 contains his unrebutted opinion as to the ethical problems posed by the inappropriate overuse of Anders motions, 8 all of which opinions this Court credits not only by reason of Dean Lefstein’s expert testimony but also in consequence of this Court’s own ex *1253 tensive background (beginning well before it took the bench) in the area of professional responsibility. 60. If an appellate defender is appointed to a larger number of cases than he or she can appropriately handle, the lawyer’s responsibility is to inform the court of the problem, to seek relief from further appointments and, with respect to eases that cannot be reassigned, to brief them in the order in which they were decided by the trial court (Tr. 235-36). That approach is called for by ABA Standard 5-5.3 and its commentary relating to the provision of criminal defense services (P.Ex. 108 and Tr. 236-37). 61. First District has taken the best approach among those potentially available, as testified to during the Hearing, to discharging its professional responsibilities in light of the backlog and large volume of pending unbriefed cases in that office (Tr. 238-39). It was both appropriate and necessary for the First District to inform the First Appellate District of the First District’s problems by filing motions for leave to withdraw in cases that the office could not brief within a reasonable period of time (Tr. 240-41). 62. First District’s method of assigning the priority for briefing eases on a first-in-first-out basis is entirely appropriate (Tr. 239-41). Other suggested alternatives — such as establishing a higher priority for cases with short sentences — produce an impermissible conflict of interest on the lawyer’s part in applying preferential handling among the clients of the office, under which a subjective judgment is used to give one client priority over another client in the handling of their respective cases. Indeed, the Florida Supreme Court has specifically held that the fast-tracking of short-sentence cases poses an unacceptable conflict of interest (In re Order on Prosecution of Criminal Appeals by the Tenth Judicial Circuit Public Defender, 561 So.2d 1130, 1135, 1138 (Fla.1990) (per curiam)). Tr. 286-87. 63. Another suggested alternative of fast-tracking those cases that have “clearly winnable issues” poses the same conflict of interest problems as fast-tracking short-sentence cases. In addition, there is no plausible way, short of reviewing the entire record and beginning work on a case, to determine in advance whether it has a clearly winnable issue. Tr. 239-40, 276-77. Thus even if this alternative did not implicate ethical problems (as it does), it would hold out minimal prospects of assisting in the reduction of a major backlog such as that faced by First District here. 64. Respondents have tendered an offer of proof with respect to a portion of Span-genberg’s deposition testimony as to whether he would ever consider a fast-track alternative to a first-in-first-out system of priorities in the briefing of appeals (Dep. 182-84). In that regard Spangenberg testified that he was “impressed” by the Florida Supreme Court’s decision as to conflicts of interest referred to in Finding 62 (Dep. 182), and he then said (Dep. 184): A. I would be concerned about the conflict of interest problem. I would seriously consider fast-tracking if I thought it would not injure my clients or not be acceptable to the courts. This Court finds Respondents’ offer of proof irrelevant and, as Conclusions 25 and 47 reflect, it concurs in the Florida Supreme Court’s holding as to the ethical problems posed by a fast-tracking approach. 65. Even apart from the just-discussed ethical problems faced by First District in being appointed to represent a large group of clients such as the petitioner class, respondents’ plumping for the adoption of any type of fast-track approach misses the most fundamental fact that what is at issue here is a supply (comprising the total available time of First District’s lawyers and their staff support) that is hopelessly inadequate to meet the demand (comprising the total time needed to review the records and to brief the appeals of the petitioner class). Even on the most favorable assumption that it would be possible for First District’s lawyers quickly to identify which of their eases are capable of speediest resolution (without thereby occupying time that could be spent more efficiently in the actual substantive preparation of briefs in those or other cases on First District’s docket) and then to work on those cases first, the only thing that might be *1254 accomplished would be to create a deceptive set of statistics: a somewhat smaller number of cases that were still undisposed of, but with those remaining cases requiring a greater amount of lawyers’ effort and time per case to dispose of them. No increase in efficiency or in available resources would have been achieved — merely a reallocation of the same finite resources to benefit part of the class (a favored few) at the expense of the disfavored majority of the class members. And that type of reallocation would not alter the Findings or Conclusions reached here one whit. C. OSAD’s Anders Motion Practice 1. OSAD’s Anders Policy 66. Each year OSAD files Anders motions to withdraw in approximately 10% of its direct appeal cases where, after careful review of the record, it has been determined that there are no non-frivolous issues that could be raised in those cases (P. Exs. 2, 109 and 4 through 9 and Tr. 69). As a matter of statewide agency policy, OSAD does not permit its attorneys to file an Anders motion in any case unless the line assistant’s initial judgment that an Anders motion is warranted has been approved by a second attorney who has also independently reviewed the record and concurs that there are no non-frivolous issues in the case (Tr. 180-81). 67. Every proper Anders motion should be supported with an exhaustive analysis of every possible issue that could be raised in the case and a demonstration, as to each such issue, of why the issue is non-meritorious (Tr. 69-70, 242-44, 284-86; Anders v. California, 386 U.S. 738, 744, 87 S.Ct. 1396, 1400, 18 L.Ed.2d 493 (1967), and see Conclusion 30). It is thus often easier and less time-consuming simply to submit a one-issue brief in a case than to submit a properly supported Anders motion to withdraw (Tr. 69). 68. Nothing in the record suggests any legitimate basis for any criticism of OSAD’s practice in considering potential Anders motions, or for any criticism of the number or percentage of such motions that it files. Moreover, because of the considerations dealt with in Findings 66 and 67 there is no reason to believe that if OSAD were to file a larger percentage of Anders briefs in eases to which it had been appointed there would be any appreciable diminution (if indeed any diminution at all) in OSAD’s volume of pending unbriefed cases from the level that will exist under OSAD’s continuation of its present Anders policy (Tr. 69) — and in addition, the same point that has been made in Finding 65 would apply here as well. 2. Expert Testimony as to Anders Policy 69. It is not appropriate for an indigent appellate defender who is appointed to a larger number of cases than he or she can appropriately handle to seek to resolve the problem by filing Anders motions to withdraw from appointments. Anders is not a docket management tool. In addition, as Finding 67 reflects, an appropriately supported Anders motion is often more time-consuming to prepare than a brief on the merits raising one or two potential issues. Tr. 241-42. 70. There is a inherent peril (a) in not adhering to a strict requirement of obtaining a second opinion before filing an Anders motion or (b) in filing Anders motions in a large percentage of eases: the risk that non-frivolous issues may be missed (Tr. 243-44). Evidence at the Hearing demonstrated that such risks are not merely hypothetical. In People v. Thomas Jones, No. 92-3460 (Ill.App. 1st Dist.) the Cook County Public Defender’s Office Appeals Division, representing Jones as one of several co-defendants on appeal, filed an Anders motion to withdraw (P.Ex. 201). First District, which represented one of the other co-defendants, submitted a merits brief that raised issues that were applicable both to its client and to Jones. At First District’s urging the Cook County Public Defender’s Office then withdrew its An-ders motion and substituted a merits brief, raising an issue (which had been suggested to the Cook County Public Defender by First District) that ultimately prevailed in the Appellate Court. Tr. 284r-86. D. Summary 71. On the record before this Court, First District’s performance of its responsibilities *1255 as appellate defense counsel has been appropriate in all respects relevant to this litigation. It has not been shown to bear any responsibility for the systemic delays that form the gravamen of this action and that are dealt with in subsequent Findings. V. Expert Testimony as to Length of Appellate Process for First District Clients 72. Spangenberg on behalf of petitioners and Dr. Roger Hanson (“Hanson”) on behalf of respondents provided this Court with expert testimony regarding the length of the appellate process for direct appeal clients of OSAD. Findings 73 to 76 are this Court’s summary findings in that respect, while the more specific underlying testimony by the experts is set out in Findings 77 to 84. 73. As to cases disposed of by the First Appellate District during calendar year 1994, there was no dispute between the experts that delays of approximately 18 months from commencement of the appeal to the filing of appellant’s opening brief were typical. According to Spangenberg’s analysis the average delay between those events was approximately 18 months. According to Hanson’s analysis the median delay between those events was also approximately 18 months. 74. Again as to cases disposed of by the First Appellate District during calendar year 1994, there was no dispute that delays significantly in excess of two years from commencement of the appeal to the Appellate Court’s decision were typical. According to Spangenberg, for cases decided by the Appellate Court (excluding appeals that were dismissed or voluntarily withdrawn) the average delay between those events was approximately 30 months. According to Hanson the median delay from notice of appeal to decision was 28 months. 75. Petitioners’ evidence showed that the unprecedented volume of First District’s now pending unbriefed cases will result in significantly longer delays from the notice of appeal to the filing of the opening brief for current OSAD clients, including the members of the petitioner class, than were typical for cases disposed of by the First Appellate District during calendar year 1994. Nor would that situation be changed if OSAD were to set up a different system of priorities from its present policy of briefing the cases with the oldest trial court judgment dates first. Even apart from the already-found ethical problems posed by any alternative priority system, the effect of so doing would be to improve somewhat the situation of the preferred appellants at the cost of creating even greater delays for the rest of the plaintiff class. Although these last two sentences will not be repeated in the other instances where these Findings refer to delays in terms of OSAD’s present policy, the same principles are equally applicable in each such instance. 76. In consequence of what has gone before, it is equally undisputed that the delay from notice of appeal to decision for persons who are currently members of the petitioner class will substantially exceed two years. Based on Findings 73 and 75, the delay from notice of appeal to opening brief will itself be more than 18 months, and will in all likelihood be close to two years. Then on average an additional 10 to 12 months will elapse before a decision is rendered in those cases. See P.Ex. 1.14 and R. Ex. 23. 77. OSAD maintains a computerized docket tracking system that contains data as to each case to which OSAD is appointed, tracking the progress of each case from notice of appeal to final disposition. That system is a well-maintained and thorough source of information (Tr. 305 and P. Exs. 35 and 36) as to the processing through the appellate system of cases assigned to OSAD. OSAD’s computerized docketing records show as to every case to which OSAD was appointed from the mid-1980s through 1994 (P.Ex. 35 and Tr. 304-06, 993-99) the dates of (1) OSAD’s appointment, (2) the notice of appeal, (3) filing of the record, (4) where applicable, the filing and disposition of an Anders motion, (5) where applicable, the filing of a motion to withdraw or to dismiss the appeal, (6) filing of the opening brief, (7) filing of the State’s brief, (8) filing of the reply brief, if any, (9) oral argument, if any, and (10) the Appellate Court’s decision. In addition to those dates the computerized records show, using a numerical “appeal type” code, the type of appeal involved (e.g., direct *1256 appeal following a jury trial, direct appeal following a bench trial, direct appeal following a plea of guilty, appeal in a post-conviction proceeding, etc.). Tr. 305, 313, 1023. Using a separate numerical code, the computerized records also contain information as to the case (e.g., affirmed (Code 1) or dismissed (Code 8)). Tr. 321, 997. Finally, the database contains a notation as to the OSAD district office in which the appeal originated and a separate notation as to the OSAD district office that handled the appeal. Tr. 1024. 78. For every case in OSAD’s database, there is a record of whether OSAD’s attorneys filed a brief on the merits, a motion to withdraw under Anders, a motion to withdraw on some basis other than Anders or a motion to dismiss the appeal. OSAD’s database contains a separate field within which the date of filing of each of those different types of opening documents is recorded. P. Exs. 35 and 36 and Tr. 854-58, 995-97. 79. David Newhouse (“Newhouse”), Spangenberg’s computer analyst, performed an analysis of OSAD’s computerized docketing records for Spangenberg’s use. 9 After the elimination of eases transferred from First District for briefing by other OSAD district offices, the OSAD database contained reliable data on approximately 260 direct appeal cases to which First District was appointed and that were decided by the First Appellate District during calendar year 1994 (the “analyzed cases”). First District filed a brief in approximately 205 of the analyzed cases, while in the others the First Appellate District rendered a decision following' First District’s submission of a document other than a brief: a motion to withdraw under Anders, a motion to withdraw for reasons other than Anders or a motion to dismiss the appeal. Spangenberg’s analysis further showed that of the approximately 260 analyzed cases, 242 First Appellate District decisions were rendered after the filing of a record in those cases. P. Exs. 114 and 114(a) and Tr. 1021-30. 80. Spangenberg’s analysis of the OSAD database shows that for all 260 of the analyzed cases, the average (arithmetic mean) period of delay from the date of First District’s appointment to the date of filing the record was 8.09 months (P.Ex. 114 and Tr. 328). For the 205 analyzed eases in which OSAD filed an opening brief on the merits, the average delay from the date of First District’s appointment to the date of filing of the opening brief was 16.75 months (P. Exs. 114 and 114(a) and Tr. 324-25). There is no reason to believe that, under OSAD’s policy of handling oldest cases first, the period of delay from First District’s appointment until its filing of the opening substantive document in a case would vary substantially if the document were a motion to dismiss or to withdraw the appeal rather than a merits brief (Tr. 326-27). 81. For the same 205 analyzed cases (those in which an opening brief was filed rather than an Anders motion to withdraw, a motion to withdraw for reasons other than Anders or a motion to dismiss the appeal), the average delay from date of appointment to the date of decision by the First Appellate District was 29.10 months (P.Ex. 114 and Tr. 311-13, 319-21). For all of the approximately 260 analyzed cases (thus also including cases where the First Appellate District’s decision was a dismissal), the average delay from the date of First District’s appointment to the date of the decision was 25.73 months (P.Ex. 114 and Tr. 320). 82. Spangenberg’s analysis used the date of First District’s appointment as the starting point for measurement of appellate delay, solely because the data were more complete for the appointment date than for the date of the appellant’s notice of appeal. Where data were available for both the date of appointment and the date of the notice of appeal, the average date of appointment for the analyzed cases was approximately 23 days later in time than the date of notice of appeal as recorded on the database. Thus the average delays described in Findings 80 and 81 may *1257 be understated by approximately one month. P. Ex 114 and Tr. 317-19. 83. Spangenberg’s analysis of the delay on appeal for First District direct appeal cases decided by the First Appellate District in calendar year 1994 is generally consistent with the findings of respondents’ expert Hanson. Hanson found that the median — not the average or arithmetic mean — delay from the notice of appeal to the First Appellate District’s decision for direct appeal eases handled by First District and decided by the First Appellate District in 1994 was 28 months. For those cases the median delay from notice of appeal to filing of the record was 8 months, and the median delay from the filing of the record to the filing of the appellant’s brief was 10 months. R. Ex. 23. Thus the median delay from notice of appeal to opening brief was approximately 18 months (Tr. 865). 84. Hanson apparently analyzed not only cases that were assigned to First District and briefed by that office but also cases that after initial assignment to First District were transferred to other OSAD district offices. Hanson analyzed 319 eases decided in 1994, a figure that corresponds roughly with Span-genberg’s count of 330 total direct appeal cases (including those later transferred) that were initially assigned to First District and were then decided by the First Appellate District in calendar year 1994. 10 R. Exs. 23 and 10 and Tr. 1030-31 and 886-87. 85. In sum, there is no factual dispute in the record that the mean and the median delays for First District direct appeal cases decided by the First Appellate District in calendar year 1994 were significantly in excess of two years. Nor is there any factual dispute that for First District direct appeal cases decided by the First Appellate District in calendar year 1994, both the mean and the median delay from the notice of appeal to the filing of the appellant’s opening brief were nearly 1% years (compare P.Ex. 114 with R. Ex. 23 and Tr. 865). 2. Increasing Delays in the Appellate Process. 86. It is so obvious as to be tautological (see Tr. 383-84) that the delays experienced from First District direct appeal clients whose cases were decided in 1994, from the time of filing of their notices of appeal to the filing of the opening briefs in their cases, was the direct result of First District’s being appointed to so large a number of cases that many months elapsed before First District’s attorneys could begin working on the cases to which the office had been appointed (Tr. 333-36). Indeed, respondents’ expert Hanson did not dispute that it is the large volume of appointments to First District, with the resulting backlog, that produces the delays from notice of appeal to filing of the opening brief (Tr. 860-61). 87. For the most part, cases that the First Appellate District decided in 1994 were cases to which First District was appointed as counsel in 1991, 1992 and 1993. In those three years First District’s volume of pending unbriefed cases never exceeded 579. In the last of those three years (calendar year 1993), following First District’s transfer of some cases to equalize caseloads among the various OSAD districts, First District’s volume of pending unbriefed cases never exceeded 500. P. Exs. 3 and 97 and Tr. 333-34. 88. With more than 750 pending un-briefed eases on First District’s docket at the time of the Hearing, it is inevitable that for cases to which First District is appointed today the delays on appeal will substantially exceed the delays experienced by First District direct appeal clients whose cases were decided by the Appellate Court in calendar year 1994 — unless substantial measures are adopted to alter the already-described trends. That factual conclusion is wholly consistent with Pelletier’s informed prediction (see Finding 37) that unless staffing levels change or pending cases are eliminated from First District’s docket, it will take approximately two years before an Assistant Appellate Defender will be able to begin *1258 work on the newest cases to which First District is appointed today (Tr. 170-71). 89. Spangenberg and his associates prepared computer-generated projections to illustrate the trend of anticipated future delays for First District direct appeal cases (a) from the date of the office’s appointment to the filing of the opening brief and (b) from the date of the appointment to the First Appellate District’s decision. Those projects were prepared for all First District cases that were not dismissed (P.Ex. 93) and for all those not-dismissed cases in which the sentence length was 20 years or less (P.Ex. 94), and were based on actual historical data on (1) the number of new cases to which First District was appointed in each month from January 1990 through April 1995, (2) the number of First District’s attorneys for the same time frame, (3) the total number of cases pending in First District for the same time frame, (4) the total volume of pending unbriefed cases assigned to the office for the same time frame, (5) the average elapsed time from appointment to brief, month by month, for cases to which First District was appointed from January 1990 through June 1993 and (6) the average elapsed time, month by month, from such appointment to the First Appellate District’s decision from January 1990 through June 1992. R. Ex. 12 and Tr. 999-1021. 90. Spangenberg’s projections in P. Exs. 93 and 94 are based on assumptions (1) that from May 1995 forward First District will be appointed to 50.4 new cases per month (the average of the new monthly appointments for the 12 preceding months), (2) that from May 1995 forward First District will have 20.75 attorneys on staff (the number of positions that existed as of May 1995) and (3) that during each month from May 1995 forward each attorney will prepare a substantive initial document (an opening brief, an Anders motion to withdraw, a motion to withdraw for reasons other than Anders or a motion to dismiss the appeal) in an average of 2.01 cases (the average of such dispositions per month for the 12 months from April 1994 through March 1995). P.Ex. 93 and 94 and R. Ex. 12 and Tr. 999-1021. Those projections are driven by the assumption, explained by Spangenberg in his testimony (Tr. 333-36) and unrebutted by respondents’ expert Hanson (Tr. 860-61), that there is a direct relationship between the volume of pending un-briefed cases in First District and the length of delay from appointment to brief and from appointment to decision (R. Ex. 12). 91. Based on the just-described assumptions, the projections anticipate an increasing backlog of pending unbriefed cases into the future. Using the derived ratio between the length of time from appointment to brief and the volume of pending unbriefed cases, the projections anticipate future delays, month by month, for the cases to which First District is appointed in that month. Similarly, using the ratio between the length of time from appointment to decision and the volume of pending unbriefed cases, the projections anticipate additional delays from appointment to decision, month by month, for cases to which First District is appointed in that month. P. Exs. 93 and 94 and R. Ex. 12 and Tr. 999-1021. 92. This Court has not considered Span-genberg’s projections as a precise prediction of future backlog or future delay, nor does it appear that they were offered for that purpose (Tr. 383). But the projections do corroborate the unquestionable inference that, assuming that future appointments, future staffing levels and future attorney productivity are consistent with recent experience, the backlog of First District eases and the briefing and decision delays for First District’s clients (both generally and for the class members in particular) will be even greater in the future than they were in 1994. Respondents offered no evidence that calls into question the trends of increasing backlog and resulting increasing delays in the briefing and decision of First District cases overall and, in particular, those eases with sentences of 20 years or less. 3. First District Delays Are Excessive and Inordinate 93. As the ensuing Findings reflect, the delays shown by the parties’ analyses both of the First Appellate District’s calendar year 1994 decisions (P.Ex. 114 and R. Ex. 23) and of the significantly longer delays that will *1259 ensue for direct appeal clients of First District whose cases are just now entering the system are excessive and inordinate because they exceed every known normative reference point. 94. Those delays vastly exceed the appellate scheduling rules set forth by Illinois Supreme Court Rules 608 and 343 to govern the filing of the record and of briefs in appeals (including appeals in criminal cases) in the Illinois Appellate Courts (P.Ex. 60). 95. Those delays also greatly exceed the ABA’s Standard for Time on Appeal (ABA Appellate Standard 3.52), which provides that 75% of all appeals to intermediate appellate courts are to be decided within 290 days of the notice of appeal, 95% of all such appeals are to be decided within one year of the notice of appeal and the remaining 5% are to be decided as soon after one year as is possible. Although ABA Standard 3.52 is not empirically based and is subject to possible revision, the standard reflects the considered consensus judgment of a commission of distinguished and experienced persons as to how long appeals should take in the majority of United States appellate jurisdictions. P.Ex. 57 and Tr. 912-13, 356-57. 96. Those delays also substantially exceed the two year presumptive limit established by the Court of Appeals for the Tenth Circuit in Harris v. Champion, 15 F.3d 1538 (10th Cir.1994) for processing a direct felony criminal appeal from notice of appeal to final disposition — a limit conceded by respondents’ own expert Hanson to be an important reference point for determining when delay in appeal is excessive (Tr. 914). 97. Those delays also exceed the median delays in every jurisdiction as to which data concerning time on appeal in direct felony criminal appeals (including earlier data for Chicago) was presented by either side at the Hearing. Hanson has collected extensive data concerning the time on appeal in felony criminal appeals. In 1993 he produced a paper entitled Time On Appeal, which among other things reported the following median times (expressed in days) on the appeal of convictions during 1989 in 19 urban trial courts throughout the country (P.Ex. 64 at Table 6): Jurisdiction Median Days from Notice of Appeal to Decision Chicago 11 594 Cleveland 546 Colorado Springs 527 Dayton 376 Detroit 532 District of Columbia 642 Houston 400 Miami 376 Milwaukee 217 Phoenix 226 Pontiac 436 Portland 392 St. Paul 244 San Diego 418 Santa Ana 436 Seattle 552 Waukegan 602 Wheaton 625 Wichita 391 98.Hanson also testified about a preliminary report regarding median times on appeal that he had presented in August 1995 to the ABA’s Task Force on Reduction of Appellate Delay (P.Ex. 118 and Tr. 870-84). Hanson’s preliminary report described the results of analysis of computerized docketing records, reflecting the following median times (expressed in days) on direct felony criminal appeals in the following jurisdictions: Jurisdiction Median Days from Notice of Appeal to Decision Arizona Div. 1 (Phoenix) 319 California, 6th Dist. (in S. Cal.) 378 Georgia (statewide) 234 Idaho (statewide) 371 Iowa (statewide) 429 Maryland (statewide) 272 Minnesota (statewide) 246 Missouri (statewide) 471 New Mexico (statewide) 216 New York 1st Dept. (Manhattan) 757 New York 4th Dept. (Albany) 505 Pennsylvania (statewide) 266 Texas 5th Dist. (Dallas) 500 Texas 13th Dist. (rural Texas) 411 Mississippi Supreme Court (statewide) 619 Puerto Rico Supreme Court 1,398 Vermont Supreme Court (statewide) 273 Wyoming Supreme Court (statewide) 351 Except for the numbers for Puerto Rico (which Hanson characterized at Tr. 882 as an “oddball”) and Manhattan, the median delays in the other jurisdictions studied by Hanson from notice of appeal to the actual decision date are either substantially or significantly less than the two year delay that Pelletier *1260 anticipates, and that this Court finds, will occur for current First District clients from notice of appeal to the opening brief (Tr. 169-71). 12 Even in Manhattan the median delay to decision is only about one month longer than the projected delay to opening brief in First District cases, so that on Hanson’s own study the First District delay is significantly greater than in every other jurisdiction studied except for what he acknowledged as “oddball” Puerto Rico. 99. In addition, the anticipated delay of approximately two years from notice of appeal to opening brief for the newest cases appointed to First District vastly exceeds the delay incurred by other statewide indigent appellate defender agencies in preparing and filing their opening briefs, as reflected in the following information on other statewide indigent appellate defender agencies obtained by Spangenberg through a telephone survey (P.Ex. 61 and Tr. 357-60, 368): Jurisdiction State Agency’s Delay from Notice of Appeal to Filing of Opening Brief Colorado 320 days Connecticut 1 year Iowa 150 days Massachusetts 6 months Michigan 84 + days Minnesota 150 days New Hampshire 210 days New York 1 to Vk years North Carolina 45 to 50 days Ohio 60 + days Oklahoma 200 days Oregon 120 days South Carolina 135 days Wisconsin 330 days Nor could respondents’ expert Hanson point to a single other jurisdiction in which the median delay from notice of appeal to opening brief is as much as 18 months (Tr. 888-89, 891-92), as it was for First District cases decided in 1994. 100.As already indicated, Hanson testified as an expert witness on behalf of respondents in the area of appellate court systems (Tr. 814). His investigation in this case was limited to the OSAD computerized docket records and the Annual Reports of the Administrative Office of the Illinois Courts (Tr. 817-20). Hanson had no exposure to any information relating to OSAD’s backlog, volume of pending unbriefed cases, staffing levels or volume of new appointments (id and Tr. 866-67). This Court finds that without that information Hanson was in no position to opine as to the reasons why First District’s direct criminal appeals take as long as they do before they reach resolution, and he was also in no position to comment as to the acceptability or unacceptability of the level of appellate delay (Tr. 975-76). 101. In 1993 Hanson submitted to the Michigan State Court Administrative Office a pair of reports on the problem of delay in the Michigan Court of Appeals, produced as a result of lack of judicial resources to handle a growing appellate docket. Hanson’s reports documented median delays in criminal cases from notice of appeal to decision ranging from 746 days (in homicide cases) to 487 days (in burglary/theft cases). He also documented median delays from notice of appeal to decision in civil cases of 750 days (in general civil appeals), 779 days (in family law appeals) and 519 days (in administrative agency appeals). Hanson attributed those delays to a “bottleneck in the appellate process” produced by an inadequate level of resources in the Michigan Court of Appeals. R. Ex. 40 and Tr. 892-902. 102. Hanson’s analysis of the Michigan situation included the finding that a “bottleneck” was producing delays of up to approximately 500 days between submission of an appeal to the Court and the Court’s decision. His analysis was that the bottleneck was produced by a problem of understaffing in the Court. One of Hanson’s reports concluded that this situation in the Michigan Court of Appeals “threatens the basic integrity of appellate review” (R. Ex. 39). He further characterized the situation as a “crisis” that required “immediate attention” and “bold action” {id. and Tr. 892-902). 103. It is a reasonable inference from Hanson’s Michigan reports that if he had been apprised of the fact that the delays for First District clients — lengthier than those documented in his Michigan reports — were *1261 caused by chronic underfunding at First District, and that even lengthier- delays were imminent in First District (approximately two years from the notice of appeal before a lawyer can prepare his or her clients’ opening brief), respondents’ own expert Hanson would have reached the conclusion that the delays caused by the backlog and understaff-ing at First District also constitute a “crisis” that “threatens the basic integrity of appellate review.” Indeed, if Hanson were not to reach that conclusion in total consistency with his own characterization of the Michigan situation, his testimony would have to be discounted as lacking in credibility. 104. In consequence of the First District delays discussed at length in the foregoing Findings, approximately 45% — nearly one-half — of the members of the petitioner class will have completed serving their sentences before the First Appellate District issues its decisions in their appeals. 105. Respondent’s expert Hanson professed that he was unable to conclude one way or another whether the appellate delay to which the petitioner class is subjected is or is not inordinate (Tr. 972-74). But Hanson conceded that a delay on appeal is inordinate where the persons with cases on appeal have been released or have virtually served their prison time before the appellate court decides their cases (Tr. 916-17) — and as Finding 104 reflects, that is true as to an unacceptably large percentage of members of the petitioner class. 106. Because this action has been brought under Section 2254, the predicate of petitioners’ claim must be and is that their rights under the Constitution have been and are being violated. That claim is of course to be tested by a federal standard, in this instance involving the petitioners’ right to appeal a criminal conviction without being subjected to inordinate delay amounting to a deprivation of due process of law. Although related state law standards are therefore not determinative of the issue as such, they may clearly bear on the question of excessiveness. In that respect, the delays currently encountered by petitioners as reflected in these Findings and Conclusions are far in excess of the situation that existed a little' more than a decade ago, as reflected in P.Ex. 51. That exhibit is the June 19,1984 report of the “Ad Hoc Committee on Procedures To Eliminate Delays in Criminal Appeals” (the “Committee”), which had been created by the Illinois Supreme Court to “specifically address the matter of undue delay.” In P.Ex. 51 the Committee reported to the Supreme Court that the then-existing average time interval of 7.4 months between notice of appeal and opening brief (almost a full year less than, the corresponding delay period in 1994, as reflected in these Findings, and well over a year less than the delay period that is being encountered in the newest appeals assigned to First District) was “unacceptable.” If less than an eight-month period was “unacceptable” then, what can be said of delays that are three times that long (the 24r-month prospect faced by new members of the petitioner class)? 107. In summary, the delays in the processing and hence the disposition of the First Appellate District’s criminal appeals caused by underfunding and consequent understaff-ing in OSAD’s First District are excessive and inordinate, and those delays will increase substantially unless prompt action is taken to reverse the trend that has existed for the past several years. This is in no way negated — indeed it is really confirmed — by the evidence from respondents’ own expert Hanson. It should be made clear that on the record in this case no part of the impermissible delays is ascribable to the post-briefing handling of criminal appeals by the First Appellate District itself — no evidence adduced at the Hearing ascribed any lack of dispatch in the processing of appeals by that Court once the cases became fully briefed and thus ripe for decision. VI. Appeals in the Public Defender’s Office and the State’s Attorney’s Office A. Cook County Public Defender’s Office 108. James Reddy, Esq. (“Reddy”) is the Chief of the Appeals Division in the Cook County Public Defender’s Office, a position he has held since March 1990. In that position Reddy is responsible for the day-to-day operation of the Appeals Division, which *1262 briefs appeals on behalf of indigent persons to whose cases that office is appointed. Tr. 747-48. 109.In the years that Reddy has been in charge of the Appeals Division, it has suffered a decrease in its number of lawyers, an increase in the annual rate of appointments and a persistent and ineradicable backlog of cases with records ready for briefing in which an opening brief has not been filed. 13 In summary, the Cook County Public Defender’s staffing level, appointment level and calendar year-end backlog from 1990 through 1994 has been as follows (blanks indicate that there are no figures in the record, Tr. 772-83): Calendar Year Staff Appointments Year-End Backlog 1990 62 1,000 reduced to below 1000 1991 1,200 108 1992 48 1,600 around 300 1993 35 about 1,700 around 400 1994 about 1,400 just under 400 1995 330 Reddy characterized even the last-reflected caseload as “onerous” and as not permitting the assumption of cases by way of transfers from First District. 110. Reddy’s office disposes of an extremely high percentage of its eases by filing Anders motions to withdraw. For example, in 1993 the Public Defender’s Office filed motions in approximately 30% of the cases to which it was appointed. In 1994 that percentage increased to 41%. In December 1994 that percentage was 53% of all cases. Any such percentages impose a grave risk that viable issues that could be raised for clients of the office will be overlooked or disregarded. Tr. 787-96. 111. OSAD’s work standard is at least as demanding as the production policy to which lawyers in Reddy’s Appeals Division are required to adhere. Under the Appeals Division’s production policy, each appellate attorney is required to complete 16 “briefs” per year. For that purpose a “brief’ is defined as (1) an opening merits brief, (2) an Anders motion to withdraw or (3) a motion for voluntary dismissal of the appeal. No weight is allocated for case complexity. In addition, each assistant public defender in the Appeals Division must complete 30 “work units” per year (a system of weighted credits for various appellate tasks). R. Ex. 6 and Tr. 764-66. 112. In practice, over time each attorney in the Appeals Division -will complete on average 30 dispositions (merits briefs, Anders motions or voluntary dismissals) per year. According to Reddy’s testimony, that level of productivity has been achieved by his attorneys on a pretty constant basis. Tr. 769-71. In terms of the raw number of dispositions, that level of productivity is equal to what Pelletier’s staff at OSAD are achieving — but unlike OSAD, the Appeals Division’s caseload includes 10% misdemeanor cases, plus juvenile abuse and neglect cases, paternity cases and mental health commitment cases. Tr. 785-87. By producing a matching level of productivity despite having to handle a more complex caseload mix, OSAD’s attorneys are demonstrably more productive than the Cook County Public Defender’s Appeals Division. This should not be misperceived as an adverse finding as to the quality of the attorneys in the Appeals Division — it instead negates any adverse inference as to the quality of OSAD’s staff. 113. In light of the present staffing level in the Cook County Public Defender’s Appeals Division and the onerous caseload of that office, it would be impossible for the Appeals Division to accept additional appointments that would otherwise be directed to First District, as a potential way to relieve the backlog there (Tr. 783). To put the matter in terms of the issues in this action, any such reshuffling of the appeals of the petitioner class would not lessen the impermissible delays discussed in earlier Findings. 114. Although the Appeals Division does not refer to the practice as a “fast-track” approach, that office does have a practice of assigning higher priority to the disposition of short record cases that appear to be easily resolvable. That practice involves reviewing short records to determine whether they can be easily resolved, while cases with longer *1263 records are not reviewed. On occasion that practice can dispose of post-conviction appeals that appear to present easy issues for resolution ahead of direct appeal cases. Tr. 783-85. 115. There is no evidence that any net gain in the disposition of cases (and hence any net reduction in the scope of delay dealt with in these Findings) is accomplished by the approach of advancing certain records ahead of other cases because those records appear to be easily resolvable. All that such an approach does accomplish is to benefit certain appellants in terms of the priority of handling their cases, at the expense of all other appellants. As already found, such an approach poses conflict of interest problems and should not be employed by First District as a means of dealing with its backlog problem (even apart from the absence of any showing that the approach would produce more favorable results in that respect). B. Cook County State’s Attorney’s Office 116. Goldfarb is the supervisor at the Criminal Appeals Division of the Cook County State’s Attorney’s Office, a position she has held since 1989. That office is responsible for responding to all appellate briefs filed on behalf of criminal defendants in the First Appellate District. Tr. 705-06. 117. According to Goldfarb, the Cook County State’s Attorney’s Office is on top of its calendar and is able to respond to briefs from the defense bar with reasonable promptness. That office’s Appeals Bureau does not have a continuing problem of inadequate resources. Whenever Goldfarb has needed additional lawyers, they have ultimately been made available (though not necessarily as quickly as Goldfarb would have liked). Tr. 714-17, 731, 735. 118. Almost all of the State’s Attorney’s appeals work is in the role of an appellee responding to issues that have been framed by a defendant-appellant. Because of the appellate waiver doctrine, it is unnecessary for Assistant State’s Attorneys, who act in a ease on behalf of the State as appellee, to review the record for purposes of identifying and framing the issues in a case — instead any issue not briefed by the defendant-appellant is deemed waived. In addition, as appel-lee the State’s Attorney’s Office does not file reply briefs. Tr. 741-42, 743, 745. 119. Unlike OSAD with its hundreds of clients, the Cook County State’s Attorney has only one client, the People of the State of Illinois. Hence the Cook County State’s Attorney’s Appeals Division is free to establish priorities for the cases handled for that one client as it sees fit, without implicating the ethical conflict of interest concerns that OSAD faces. 120. Goldfarb has never been a defense lawyer and has never worked in a public defender agency. Although she has instituted several measures to make the Cook County State’s Attorney’s Appeals Division more efficient, she is in no position to say whether those measures would be useful, usable or practicable in an appellate defender agency. Tr. 732, 745-46. VII. Prejudice to the Class A. Expert Testimony 1. Dr. Robert T.M. Phillips 121. Dr. Robert T.M. Phillips is a forensic psychiatrist who has conducted numerous examinations for various persons under the control of the criminal justice system (Tr. 464-66, 469-71). Dr. Phillips has worked on behalf of both prosecution and defense (Tr. 468-69), and his work has given him extensive exposure to prison environments and to persons incarcerated within them (Tr. 469-71). For a number of years Dr. Phillips was the Chief Executive Officer of the Whiting Forensic Institute, which is the sole maximum security hospital for the State of Connecticut and is responsible for the 197 insanity acquitees in that State (Tr. 465, 468, 469). Dr. Phillips’ work has included giving testimony on behalf of the prosecution in the case of United States v. Francis Martin Duran, which involved a highly publicized weapons incident in the vicinity of the White House (Tr. 469; see also P. Exs. 69 and 69(a)). In addition, Dr. Phillips is a psychiatric consultant to the District of Columbia field office of the United States Secret Service (Tr. 466; see also P. Exs. 69 and 69(a)). Dr. Phillips *1264 was qualified as an expert in the area of forensic psychiatry without objection from respondents (Tr. 471). 122. Dr. Phillips opined that the members of the petitioner class, who are in a vulnerable state and are susceptible to stres-sors by virtue of their incarceration, are at an increased risk of suffering a range of adverse psychological reactions (including anxiety, mistrust, fear, hopelessness, and depression) because of the severe delays that are occurring in the appellate process (Tr. 472-73, 476, 480-81). Dr. Phillips assumed for purposes of his opinion that because of staffing problems the petitioner class members must wait approximately 18 to 20 months before a lawyer can begin working on their respective appeals (Tr. 472-73). 123. Dr. Phillips testified that the experience of incarceration, particularly at the initial stages, is emotionally devastating. For example, the risk of suicide in the group of persons who are in the first 48 hours of imprisonment is about nine times the risk of suicide among persons in the general population. Tr. 499. Because of the stresses imposed by the loss of liberty, the hope of regaining liberty becomes the single most organizing principle for persons in custody (Tr. 475-76). And because the appellate process — the opportunity to have one’s case reviewed — is perceived as a possible way to regain liberty, the appeal becomes a focus of that hope. When the appeal is delayed, the thwarting of that central hope creates a potential for great stress and emotional turmoil. Tr. 476-77, 480-81. 124. Dr. Phillips further opined that within a reasonable degree of medical certainty, the petitioner class members as a group are all at increased risk of developing adverse psychological reactions (Tr. 476-83, 497-500). Such an increased risk occurs because appellate delay thwarts the class members’ right to an appellate process, which is a right of central psychological importance to the class members (id.) Although it cannot be predicted who within the class will in fact develop adverse psychiatric reactions (Tr. 477-78, 485-86, 493), all of the class members are at increased risk as a result of the delays in the appellate process (Tr. 479-83, 497-500). Accordingly the increased risk of suffering adverse psychological reactions is not dependent upon whether the person is guilty or innocent of the crimes charged (Tr. 481-83, 497-98). 2. Dr. Jeffrey Metzner 125. Dr. Jeffrey Metzner testified as an expert witness in the area of forensic psychiatry for respondents (Tr. 633-34). Like Dr. Phillips, Dr. Metzner has testified on numerous occasions on various psychiatric issues relevant to determinations of the criminal justice system with regard to persons under its jurisdiction (Tr. 637-38). Dr. Metzner has written and lectured widely in this area (Tr. 635-36, 638). In the course of his work Dr. Metzner has also had substantial exposure to prisons and prisoners (Tr. 637-38). He was Chief of Psychiatry for the Colorado Department of Corrections for approximately one year (Tr. 637). His experience as an expert witness includes testimony regarding litigation (Tr. 637-38). Dr. Metzner was qualified as an expert in psychiatry without objection from petitioners (Tr. 639). 126. Dr. Metzner opined that some of the members of the petitioner class are at an increased risk of experiencing adverse psychiatric reactions as a result of delays in the appellate process (Tr. 640-41, 653). However, Dr. Metzner testified that in his opinion the members of the petitioner class at an increased risk would be “a small subset of the class” (Tr. 641). Dr. Metzner opined that the individuals within the subset are those individuals who have a clear chance of having their sentences significantly changed or reduced as a result of a successful appellate outcome, but did not include those individuals who simply believe they had a chance at a successful appellate outcome (Tr. 642). Dr. Metzner further testified that the persons within the subset would be on a continuum, with the persons at highest risk being those who were actually innocent of the crime for which they had been convicted (Tr. 641-42, 645, 649), with the middle of the continuum comprising persons whose conduct was criminal but whose convictions were improper because they were convicted of the wrong offense or were entitled to a reduced *1265 sentence, and with the other end of the continuum comprising persons who did commit the criminal conduct of which they were convicted but whose convictions were not proper for “technical reasons” such as the trial court’s receipt of improperly obtained evidence (Tr. 641-45, 653-54). 127. Dr. Metzner limited his definition of what constitutes “adverse psychological reactions” to include only those psychological reactions that would rise to the level of a diagnosable “mental disorder” (Tr. 648, 652). Although he conceded that other members of the class would feel distress and experience anxiety relating to the delays in the appellate process, he repeated that those reactions would not result in a “mental disorder” (Tr. 644). Thus Dr. Metzner would exclude from the members of the class who are at an increased risk of suffering adverse psychological reactions those individuals, like Vernon Joy, who believe there were errors committed with respect to their convictions or sentences, who believe they are entitled to relief and who express distress and anxiety about the delays in resolving their appeals, but whose convictions are ultimately affirmed by the Appellate Court (Tr. 643-45). 128. Dr. Metzner conceded that, while it is clear that some class members are at an increased risk, at this stage it is not possible to identify what individuals among the petitioner class are or are not within Dr. Metz-ner’s “small subset” of individuals who are at a higher risk of suffering mental disorders as a result of appellate delay (Tr. 654-56). Most critically as to Dr. Metzner’s testimony, the criteria that he employed were directly at odds with those that have uniformly been utilized by court decisions in defining the prejudice caused by excessive appellate delays (see Conclusions 37 and 39). 3. Testimony of Class Members a. Vernon Joy 129. Vernon Joy (“Joy”) is incarcerated in the DOC following his June 1992 conviction for burglary in the Circuit Court of Cook County (Tr. 583-84). Joy was sentenced to three concurrent 10 year terms of incarceration (Tr. 584). Following his conviction Joy filed a post-trial motion challenging the validity of his conviction and sentence (Tr. 586, 592-93 and P.Ex. 98). After the denial of his post-trial motion, Joy filed a timely notice of appeal (Tr. 587-88). 130. On August 4, 1992 Joy was advised that First District had been assigned to represent him in his appeal (Tr. 589-90 and P.Ex. 45). Attached to the letter was an information sheet explaining the procedures that First District would follow in handling Joy’s appeal and containing timetables setting forth the approximate length of time the appellate process could be expected to take (id), advising Joy in part that it would take approximately eight to ten months before First District would receive a complete record of proceedings in his ease (P.Ex. 45). 131. On August 30, 2992 Joy wrote a letter to First District paralegal Susan Carr (“Carr”), advising her that the transcript of the proceedings in his ease had already been delivered to First District and therefore requesting that an attorney be assigned immediately to represent him (Tr. 595-96 and P.Ex. 99). On September 14,1992 Joy wrote another letter to Carr, attaching a copy of his post-trial motion and offering to assist the office in drafting his brief in order to speed up the process (Tr. 592-93 and P.Ex. 98). 132. Joy was thereafter advised that due to First District’s backlog of unbriefed cases no lawyer could begin to work on his appeal for 20 months (Tr. 593-94). On October 5, 1992, at Joy’s direction Pelletier filed a motion to withdraw as counsel in Joy’s case, requesting the appointment of other counsel to represent Joy (Tr. 596-97 and P.Ex. 42). In part Pelletier’s motion advised the Appellate Court that due to the backlog of un-briefed cases in the office and the underfunding of attorney positions, First District would be unable to process Joy’s appeal in a reasonably timely manner (id). On October 15, 1992 the Appellate Court denied the motion without prejudice (Tr. 597-98 and P.Ex. 43). 133. First District filed a number of motions for extension of time to file Joy’s brief (Tr. 601-02). At no time did Joy authorize the filing of such motions on his behalf (id). 134. On February 1, 1993 Pelletier advised Joy by letter that “[w]ith our present *1266 staff it could be approximately two years before we could file your original brief’ (P.Ex. 46 and Tr. 158-59, 599). That letter further advised that First District would attempt to take steps to reduce the delay by transferring cases to other district offices, filing additional motions to withdraw as counsel and working to recruit lawyers to handle appeals on a pro bono basis (P.Ex. 46). Joy strenuously objected to the delay in completing his brief and expressed his displeasure to Pelletier (Tr. 600). 135. On March 11, 1993, with Joy’s consent Pelletier filed a second motion to withdraw as counsel and for the appointment of private counsel to handle Joy’s appeal (Tr. 600-01 and P.Ex. 47). As with the first motion to withdraw, Pelletier cited as reasons the reduced funding to his office and the tremendous backlog of unbriefed cases. As with the first motion to withdraw, on March 17, 1993 the First Appellate District denied the motion (Tr. 601 and P.Ex. 48). On March 28, 1993 Pelletier wrote Joy that it could still be “approximately 20 months before the original brief in your case is filed” (Tr. 602 and P.Ex. 49). 136. On July 8, 1993 Joy wrote Pelletier expressing his concern and frustration over the delay in filing the brief in his behalf (Tr. 602-07 and P.Ex. 102); Mr. Pelletier since I have been locked up this year alone, I’ve gone through some things that I wouldn’t wish on a damn dog, let alone a human being.... But wait, there’s more! My briefs were supposed to be filed over a year ago, but to this date still “NO” filed brief. I occasionally try to call my lawyer whom, I might add, told me and I quote “Know that you will not be abandoned. At all times you will be represented by counsel. Our hope your right to counsel on appeal is a meaningful one.” Joy stated that notwithstanding those representations: Ms. Salinger, who I believe is a good lawyer, tells me that she’s not even allowed to “look” at my case until she gets to it. Excuse me, but, when you said, if I have any questions about the appellate process, just call or write and ask, did anyone think that I just might ask a question about those- files that Ms. Salinger “can’t look at”. Joy further advised: As each day goes by in here it gets harder and harder to deal with and I don’t have any indication it’s coming to an end ever. ‡ Hí * H* # % This entire ordeal is beginning to take its toll on me sir and I’m asking you, no I’m begging you to tell me something or show me some type of action from your office. 137. Joy attempted on his own to find private counsel willing to handle his appeal (Tr. 607). He wrote the Illinois State Bar Association, Richard Phelan’s office and the University of Illinois, all to no avail (Tr. 607-09). 138. Joy then filed a petition for habeas corpus in this District Court challenging the constitutionality of the appellate delays in his ease (Tr. 610). Pelletier wrote a letter on Joy’s behalf to the First Appellate District advising it of Joy’s habeas corpus petition and asking that the court appoint private counsel, to represent Joy (Tr. 161-65 and P.Ex. 37). Thereafter First District filed a motion to withdraw in Joy’s case after the law firm of Keck, Mahin and Cate had agreed to represent Joy on a pro bono basis, and that motion was granted (Tr. 610-11). 139. On March 4, 1994 Joy’s new counsel filed his appellate brief (Tr. 612 and P.Ex. 204). On March 25 the new counsel filed a motion to expedite the appeal, which was granted in large part (Tr. 613-14 and P.Ex. 103). Then on August 16, 1994 Joy's conviction was affirmed by the First Appellate District, after which the Supreme Court denied a Petition for Leave To Appeal (Tr. 614, 626). 140. Joy testified as to the anger, frustration, distress and depression that he suffered because of the delays in processing his appeal (Tr. 616-17): To have to deal with incarceration in the first place is a tough thing.... I mean without the added, roller coaster ride of thinking that you’re going ... to get heard in the Appellate Courts or get a chance to tell your side. But to constantly be denied *1267 a right that you believed you had coming and know that you’ve done everything you can to get this, you know, and you’re still denied, I really don’t think we have enough time for me to go into details of how painful that is. And it’s personified when there’s personal things happening at home. Now I know that — that it’s not up to the courts to control what happens in my life— in my personal life, but the part of my life that they were in control of, they were screwing it up. The things they could change, they didn’t change. When I was found guilty of a crime, they sentenced me according to the law. But when it was their turn to follow the letter of the law, they were able to get continuances and leeway until it was fit — so they could fit it the way they wanted to. And [it] hurts.... 141. Joy was a sincere and highly credible witness and clearly articulated adverse psychological reactions, including anxiety, frustration and depression, as a direct consequence of the delays in adjudicating his appeal. And even if those adverse reactions were required to be discounted in terms of any claimed damages sustained by reason of those delays, on the basis that Joy’s conviction was affirmed when his appeal was ultimately considered on the merits, 14 it is more than reasonable to infer — and this Court finds — that all the same types of reaction are certain to be suffered by the substantial number of presently unidentifiable members of the petitioner class whose appeals will ultimately prove to have been successful (based on past statistics in that respect) and who will therefore have suffered needless harms because of the excessive and inordinate delay. b. Charles Smith 142. Charles Smith (“Smith”), who is incarcerated in East Moline Correctional Center, was convicted of burglary in the Circuit Court of Cook County on April 2, 1994 (Tr. 658) and was sentenced to a five year term of incarceration (Tr. 659). Taking into account statutory good time credit, his projected release date from DOC is April 2, 1996 (Tr. 659-60). 143. At Smith’s sentencing the trial court advised him that he had the right to appeal his conviction and sentence (Tr. 663). Smith filed a motion for a new trial, which was denied, and he then filed a timely notice of appeal because he believes he is not guilty of the crime of conviction (Tr. 660). First District was appointed to represent him (Tr. 663-65 and P.Ex. 70). 144. On April 4, 1995 Patricia Mysza (“Mysza”), the OSAD lawyer assigned to Smith’s case, filed a motion for extension of time to file appellant’s brief, advising the First Appellate District that “there are 459 unbriefed cases in this office with judgment dates prior to defendant’s judgment date. Given the size of the Office of the State Appellate Defender’s present staff, it will be ' approximately 14 months before an attorney will brief this case” (Tr. 665-66, 667 and P.Ex. 72). From that motion it became clear to Smith that no brief would be filed on his behalf until after he would be released from prison based upon his projected April 2,1996 release date (Tr. 666). When he read the motion he was “depressed and [I] felt like they were playing games with me” (id.). He attempted to telephone his lawyer but was unable to reach her (Tr. 675). 145. In a letter that Mysza received on May 2, 1995, Smith expressed his anger and disgust over the length of time that it would apparently take, based upon the time projected in the motion for extension of time, to file his brief (P.Ex. 75 and Tr. 667-69). Smith’s letter proclaimed that because he is innocent of the crime of conviction: I exercised my right under the guidance of an attorney to file an appeal to rectify the matter. Little did I realize how crowded and overworked the court system is in that said appeal would be heard long after my release on my 5 year sentence. Smith also asked: Why would I want to have an appeal heard in another possible 14 to 18 months[?] I’ll *1268 be on the beach in the carrib[b]ean by then, and I’ll be damned if I’m coming back to go to court. Those feelings were reinforced when, three weeks before the Hearing, another motion for extension of time to file the opening brief was filed in Smith’s case (Tr. 670). 146. Smith was also a sincere and highly credible witness and clearly articulated his upset, frustration and anger over the fact that he will be released from his confinement long before resolution of his appeal. At this time, of course, it cannot be foretold whether Smith’s appeal is or is not meritorious and therefore will or will not be successful, but what has been said at the conclusion of Finding 141 applies with equal force here. c. Thomas Powers 147. Thomas Powers (“Powers”), who resided in Lincoln Correctional Center at the time of the Hearing (Tr. 679), was convicted of attempted aggravated criminal sexual assault in the Circuit Court of Cook County on September 10, 1993. He was taken into custody immediately following his conviction (Tr. 680, 682) and was sentenced to a 12 year term of incarceration (Tr. 679). As of the date of the Hearing he had served approximately 25 months of that sentence, so that his projected release date (taking into account statutory good conduct credit) is June 13,1999 (Tr. 680). 148. Powers filed a pro se motion for a new trial and a motion for reduction of sentence (Tr. 682-83), then timely asserted his right to an appeal by filing a notice of appeal from his conviction and sentence in January 1994 (Tr. 680, 683). First District was appointed to represent him on appeal (Tr. 680). 149. On March 3, 1994 Carr of First District’s office wrote Powers that it would take approximately 8 to 10 months before the office would receive the transcripts of the proceedings in his case (Tr. 684 and P.Ex. 81). On March 6 Powers wrote Carr that the lawyer who had represented him at trial already had a copy of those transcripts (Tr. 685-86 and P.Ex. 82). Powers also attached a copy of the pro se post trial motions that he had filed and a draft of an appellate brief that he had prepared on his own (P.Ex. 82). On April 27 Powers again inquired of Carr about the status of his transcripts (Tr. 686-87 and P.Ex. 83). 150. Originally Linda Eigner was assigned to work on Powers’ appeal, but his case was then transferred to Yasemin Eken (“Eken”) because she could begin work on Powers’ case sooner (Tr. 687-89 and P.Ex. 84 and 85). Nonetheless, as of the date of the Hearing First District lawyers had filed five to six motions for extensions of time to file Powers’ brief on appeal (Tr. 691 and see, e.g., P.Ex. 76-80). 151. On December 8, 1994 Eken wrote Powers that she would probably not be able to file his brief by the January 31, 1995 deadline (Tr. 689-90 and P.Ex. 88). On February 4, 1995 Powers wrote Eken objecting to the recently filed motion for extension of time to file his brief, complaining about the length of time that it was taking to file his briefs on appeal and inquiring why it was taking so long (Tr. 692 and P.Ex. 90). In response Eken wrote Powers that there were 162 cases in the office with earlier judgment dates that would take priority over his case (Tr. 693 and P. ex. 91). Powers’ appellate brief was finally filed in early September 1995 (Tr. 694), nearly two years after his conviction. In the meantime Powers had written numerous letters to private lawyers attempting to obtain pro bono assistance with his appeal, but without success (Tr. 695). 152. Powers has had numerous conversations with prison personnel about the stress he was suffering because of the delays in processing his appeal (Tr. 696). In fact Powers has had some suicidal thinking while incarcerated (Tr. 700-701). 153. If Powers is granted a new trial following his appeal, a witness critical to his defense — his former wife — will be difficult to locate because of the length of time that has elapsed since his original trial. Powers does not know his ex-wife’s present whereabouts or how to reach her. Tr. 698. 154. Powers was also a highly credible witness and clearly articulated both adverse psychological reactions and potential impairment of his defense at retrial as a result of the delays in the appellate process. Again *1269 what has been said at the end of Finding 141 applies here with equal force. Conclusions of Law I. Exhaustion of State Remedies 1. Although the failure to exhaust available state remedies is not strictly jurisdictional in nature (see, e.g., Granberry v. Greer, 481 U.S. 129, 131, 107 S.Ct. 1671, 1673, 95 L.Ed.2d 119 (1987)), Section 2254(b) sets out this familiar precondition to a state prisoner’s invocation of federal habeas corpus: An application for a writ of habeas corpus in behalf of a person in custody pursuant to the judgment of a State court shall not be granted unless it appears that the applicant has exhausted the remedies available in the courts of the State, or that there is either an absence of available State corrective process or the existence of circumstances rendering such process ineffective to protect the rights of the prisoner. 2. This Court has earlier held, and it now reconfirms, that where as here the petitioner class’ Complaint addresses the constitutionality of the inordinate delay in the State’s appellate process itself, any requirement of exhaustion of state remedies in the ordinary sense — in this instance by requiring, before petitioners’ Section 2254 Complaint may be considered at all, the disposition of a pending appeal whose nondisposition is the very gravamen of the Complaint — should not be imposed. Although our Court of Appeals has not had occasion to decide the issue, Allen v. Duckworth, 6 F.3d 458, 459 (7th Cir.1993) has identified the anomaly that such a requirement would present: There is a nice question, unnecessary to decide, whether exhaustion of state remedies is required when a prisoner is complaining about inordinate delay in the state court system. Requiring exhaustion would add another layer of delay — would make the requirement of exhaustion literally exhausting, and might even threaten an infinite regress. It seems a case for invoking the exception to the exhaustion requirement when there are “circumstances rendering [state corrective] process ineffective to protect the rights of the prisoner.” 28 U.S.C. § 2254(b). Many cases so hold; illustrative are Harris v. Champion, 938 F.2d 1062, 1068-69 (10th Cir.1991), and Coe v. Thurman, 922 F.2d 528, 530-31 (9th Cir.1990). We have taken no position, Lane v. Richards, 957 F.2d 363, 365 (7th Cir.1992), and need not do so today, since if the case was moot the district court lost jurisdiction whether or not the petitioner was excused from seeking a state remedy for the state supreme court’s delay. 3. As Allen recognized, the decided cases almost without exception support this Court’s holding expressed in Conclusion 2. For example, in addition to “Harris I ” (the Tenth Circuit decision cited in Allen) and the other illustrative cases to which Allen refers, the second case in the Harris v. Champion trilogy (“Harris II,” 15 F.3d 1538, 1556 (10th Cir.1994) (which states, “Although § 2254 requires a habeas petitioner to exhaust his or her underlying claims before coming to federal court, it does not require a petitioner to exhaust the issue of exhaustion, itself’)) and a number of other cases cited in Harris II are all to the identical effect. 4. Harris II, 15 F.3d at 1556 holds that a delay of two years or more from the notice of appeal without the issuance of a state appellate decision in a habeas petitioner’s direct criminal appeal is such a long delay as to give rise to a presumption of ineffective state process, therefore excusing the exhaustion requirement of Section 2554(b). Here the current members of the petitioner class have shown that close to two years after their respective notices of appeal will elapse even before the opening briefs will be filed in their cases, with another 10 to 12 additional months required for the issuance of the Appellate Court’s decisions in their cases (see Finding 76). On the uncontradicted and overwhelming evidence adduced during the Hearing, the reason for the front-end delay from notice of appeal to briefing — the delay that causes the overall appellate process to exceed the Harris presumptive two-year threshold — by a substantial margin — is First District’s inadequate staffing that prevents the assistant appellate defenders from beginning their work on petitioners’ cases in time *1270 ly fashion. Because that is a wholly inadequate justification for the excessive delay, the exhaustion requirement is excused for petitioners. 5. Petitioners do not ask this Court to decide any underlying substantive claims that may be presented in their delayed state appeals. Instead they ask this Court to adjudicate their claim that the delay on appeal itself violates their constitutional rights and to issue an order directing an appropriate remedy for that violation. There is ample authority for the granting of such relief in federal habeas proceedings — see, e.g., Harris II, 15 F.3d at 1566-67; Coe, 922 F.2d at 532; Simmons v. Reynolds, 898 F.2d 865, 869-70 (2d Cir.1990). II. Deprivation of Due Process 6. Although Allen, 6 F.3d at 459-60 correctly notes that the right to appeal a state court conviction is not itself a federal constitutional right, it is equally correct in observing that such a right once conferred cannot be stripped away arbitrarily: We can assume, as a number of cases have held, though none in this circuit, e.g. Cody v. Henderson, 936 F.2d 715, 718-19 (2d Cir.1991); Coe v. Thurman, supra, 922 F.2d at 530, that excessive delay in the processing of a criminal defendant’s state appeal can be a denial of due process of law. The defendant has no federal constitutional right to an appeal, Ross v. Moffitt, 417 U.S. 600, 611, 94 S.Ct. 2437, 2444, 41 L.Ed.2d 341 (1974), but a state is not permitted with one hand to grant such a right and with the other to take it away in an arbitrary fashion, as by denying an indigent a free transcript of the trial, Griffin v. Illinois, 351 U.S. 12, 18, 76 S.Ct. 585, 590, 100 L.Ed. 891 (1956), or preventing a prisoner from filing a timely notice of appeal, Burns v. Ohio, 360 U.S. 252, 257, 79 S.Ct. 1164, 1168, 3 L.Ed.2d 1209 (1959)—or simply refusing to decide an appeal. So at least a number of courts have held, and we would have no occasion in this case to quarrel with them even if we were disposed to disagree, since the petitioner must lose in any event. Here Illinois’ granting of the right to appellate review and to the representation by counsel in the course of that review (Ill. Const, art. VI, §§ 4(b) and 6; 730 ILCS 5/5-5-4.1; Ill. S.Ct. Rule 607) makes such review “an integral part of the Illinois trial system for finally adjudicating the guilt or innocence of a defendant” {Griffin, 351 U.S. at 18, 76 S.Ct. at 590). 7. Given the right to appeal referred to Conclusion 6, an inordinate delay in the adjudication of that appeal is a paradigmatic example of a due process violation. As Evitts v. Lucey, 469 U.S. 387, 393, 105 S.Ct. 830, 834, 83 L.Ed.2d 821 (1985) teaches: Almost a century ago, the Court held that the Constitution does not require States to grant appeals as of right to criminal defendants seeking to review alleged trial court errors. McKane v. Durston, 153 U.S. 684 (1894). Nonetheless, if a State has created appellate courts as “an integral part of the ... system for finally adjudicating the guilt or innocence of a defendant,” Griffin v. Illinois, 351 U.S., at 18 [76 S.Ct. at 590], the procedures used in deciding appeals must comport with the demands of the Due Process and Equal Protection Clauses of the Constitution. Harris II, 15 F.3d at 1557 applies that principle to inordinate delays in the appellate adjudication so as to trigger the availability of federal habeas review: We start by noting that delay in adjudicating a state prisoner’s direct criminal appeal may do more than simply excuse exhaustion. It may also give rise to an independent due process claim. Harris I, 938 F.2d at 1068; accord United States v. Pratt, 645 F.2d 89, 91 (1st Cir.), cert. denied, 454 U.S. 881, 102 S.Ct. 369, 70 L.Ed.2d 195 (1981); Cody v. Henderson, 936 F.2d 715, 719 (2d Cir.1991); Burkett v. Fulcomer, 951 F.2d 1431, 1446 (3d Cir.1991), ce rt. denied, 505 U.S. 1229, 112 S.Ct. 3055, 120 L.Ed.2d 921 (1992) (Burkett II); Rheuark v. Shaw, 628 F.2d 297, 302 (5th Cir.1980), cert denied, 450 U.S. 931, 101 S.Ct. 1392, 67 L.Ed.2d 365 (1981); Coe v. Thurman, 922 F.2d 528, 530-31 (9th Cir.1990). Thus a habeas petition may be predicated on a due process viola *1271 tion arising from the State’s delay in adjudicating a petitioner’s direct criminal appeal even if the petitioner’s allegations of error at the trial level are based on state law and, therefore, not proper for federal habeas review. Accord on the due process issue, in addition to cases cited in Harris II, United States v. Tucker, 8 F.3d 673, 676 (9th Cir.1993) (en banc) and United States v. Johnson, 732 F.2d 379, 381-82 (4th Cir.1984). 8. Only the Harris v. Champion trilogy (the third opinion, “Harris III” is reported at 48 F.3d 1127 (10th Cir.1995)) addresses directly, in the due process and federal habe-as context, the problem of systemic indigent criminal appellate delay suffered by the clients of an understaffed appellate defender agency. All three of those thoughtful and thorough opinions, which deal with parallel difficulties faced by the indigent clients of the Oklahoma Appellate Public Defender System, are particularly pertinent to this case. In light of the quality of the analysis there, this Court has previously announced its adherence to the principles announced in, and it will continue to follow, the Harris opinions not only on the issue of exhaustion (see Conclusions 3 through 5) but on the due process and other issues posed by this litigation as well. 9. Harris II, 15 F.3d at 1559 holds that the balancing test for ascertaining violations of the constitutional right to a speedy trial, as established in Barker v. Wingo, 407 U.S. 514, 92 S.Ct. 2182, 33 L.Ed.2d 101 (1972), “provides an appropriate framework for evaluating whether a defendant’s due process right to a timely direct criminal appeal has been violated.” That framework calls for the consideration and balancing of (1) the length of delay, (2) the reason for the delay, (3) defendant’s assertion of his or her right to a timely appeal and (4) the prejudice to defendant (Harris II, 15 F.3d at 1558-59, citing Barker, 407 U.S. at 530-32, 92 S.Ct. at 2192-93). 10. Ample pre-Harris II authority from other Circuits had supported such reliance on the Barker factors in the appellate delay context (see Tucker, 8 F.3d at 676; Elcock v. Henderson, 947 F.2d 1004, 1007 (2d Cir. 1991); Johnson, 732, F.2d at 381-82; Rheuark, 628 F.2d at 303). Since Harris II the Second and Third Circuits have reconfirmed their adherence to the Barker factors in determining the constitutionality of appellate delay (Elcock v. Henderson, 28 F.3d 276, 279 (2d Cir.1994) and Simmons v. Beyer, 44 F.3d 1160, 1169 (3d Cir.1995)), as have a number of District Courts, including one from this Circuit (Jackson v. Duckworth, 844 F.Supp. 460, 465 (N.D.Ind.1994)). 11. Barker’s factors should not be applied mechanically, but rather “must be considered together with such circumstances as may be relevant” (407 U.S. at 533, 92 S.Ct. at 2193). Presence or absence of any one of the factors is not a “necessary or sufficient condition” of a due process violation (id). As Harris II, 15 F.3d at 1562 has said, the more substantial the delay the less significant the other factors become. In this instance, however, the Hearing established the existence of each of the four Barker factors. A. Length of Delay. 12. This Court holds, as did Harris II, 15 F.3d at 1560, that a delay in the adjudication of a state criminal appeal of more than two years from the notice of appeal to the appellate court’s decision is excessive and presumptively unconstitutional, and thus shifts the burden to the state defendants to show good cause for the delay. That holding in the context of systemic delays in the appellate process is entirely consistent with the decisions reached elsewhere in cases addressing individual claims of appellate delay (see, e.g., Burkett, 951 F.2d at 1445-46; United States v. Antoine, 906 F.2d 1379, 1382-83 (9th Cir.1990); Rheuark, 628 F.2d at 302; Einaugler v. Dowling, 862 F.Supp. 793, 799 (E.D.N.Y.1994)). 13. As the Findings have set out at length, the Hearing evidence established without contradiction that all persons in the petitioner class have in the recent past exceeded, and will in the future (unless present trends are altered) continue to exceed, the two-year Harris II threshold for inordinate and excessive delay. Both the mean and the median times from notice of appeal to deci *1272 sion for First District clients whose cases were decided by the Appellate Court in calendar year 1994 exceeded two years. For that same group of decisions, both the mean and the median times from commencement of the appeal to filing of the opening brief were approximately 18 months. For persons whose cases are now entering the appellate pipeline, including members or prospective members of the petitioner class, the time spans from commencement of the appeal to filing of the opening brief and then to actual decision will, unless a substantial alteration in present trends is occasioned by substantial changes that are nowhere in prospect, be materially greater than the times reflected in the 1994 decisions. Thus the delays to which petitioners are being subjected before the briefing and ultimate decision of their appeals are excessive and inordinate. 14. Several aspects of the evidence at the Hearing confirm the wisdom of the Harris II presumption of unconstitutional exeessiveness where there are delays of more than two years from the notice of appeal to the appellate court’s decision: (a) Delays of such magnitude produce an unacceptable threat to the integrity of the appellate process. In this ease approximately 45% of the members of petitioner class will likely have served all of their sentences even before a decision is rendered in their appeals. Many more will have served the principal part of their sentences prior to any appellate decision. This situation — aptly characterized as “outrageous” by one of petitioners’ expert witnesses — renders the appellate process “a meaningless ritual” in an unacceptably high proportion of cases. (b) Delays of the magnitude that Harris deems presumptively unconstitutional— and that the petitioner class is experiencing and will continue to experience — -vastly exceed the median delays that occur in other appellate jurisdictions in this country. Of the many jurisdictions discussed during the testimony of Spangenberg and Hanson, the median time on appeal in only one jurisdiction in the territorial United States, 15 New York’s First Department, comes even close to the time between the notice of appeal and decision for First District clients — even those whose eases were decided in 1994, before the situation here began to deteriorate even further. And the experience in New York’s First Department (with a median delay that is approximately three months shorter than the median time from notice of appeal to decision in the First Appellate District cases decided even in 1994) has been the target of justifiable criticism — see Mathis v. Hood, No. 87 Civ. 6234 (RPP), 1990 WL 100869, at *2 n. 4 (S.D.N.Y. July 11), aff'd., 937 F.2d 790 (2d Cir.1991): The delays in hearing indigent criminal appeals in the First and Second Departments have been recognized as deplorable and shocking by many courts, (c)Delays in excess of the Harris II threshold are fully twice as long as the time period that has been recognized as appropriate by the considered consensus judgment of distinguished and informed persons, as reflected in ABA Appellate Standard 3.52, which provides that in the majority of United States jurisdictions 95% of appeals should be resolved within one year of the notice of appeal. B. Reason for Delay 15. It was also established by the uncon-troverted Hearing evidence that the reason the petitioner class members’ appeals are being delayed for substantially in excess of two years is that First District is overwhelmed with more appointments than its staff can timely handle. Over the course of the past several years, the volume of total pending unbriefed cases assigned to OSAD— both statewide and in the First District — has continued to increase. Even more revealing of First District’s understaffing is the fact that the backlog of cases there — the group of cases with completed records that are aging in file drawers awaiting the assignment of a lawyer — has grown consistently (despite *1273 some reductions in the total volume of First District’s pending unbriefed cases resulting from (a) the transfer of some cases among OSAD’s district offices and (b) the obtaining of some intermittent relief from appointments from the Circuit Court). That number has increased from 71 at the beginning of FY 1991 to 522 at the time of the Hearing. Chronic underfunding of OSAD by the State of Illinois is the reason that the backlogs, with the resulting delay, were created and continue to grow. 16. Barker, 407 U.S. at 531, 92 S.Ct. at 2192 makes it plain that the real concern of the inquiry into the reason for appellate delay is whether the delay is the fault of the defendant or the fault of the state. As Barker, id. put it, the “ultimate responsibility” for systemic problems “must rest with the government rather than with the defendant.” In that respect Harris I, 938 F.2d at 1065 and such other cases as Muwwakkil v. Hoke, 968 F.2d 284, 285 (2d Cir.1992) and Coe, 922 F.2d at 531 hold that a public defender agency’s failure to brief its clients’ direct appeals in a timely manner cannot be attributed to the clients. Harris I, 938 F.2d at 1065, addressing the issue in the context of exhaustion, reasoned that such delays were “forced upon an unwilling petitioner by reason of his indigency.” Other courts have since agreed that delays in indigent defendants’ direct appeals should be attributed to the state, not to the defendants (see Simmons, 44 F.3d at 1170; Jackson, 844 F.Supp. at 465 n. 6)—and that is surely true where, as here, the delays are systemic. 17. Judicial attribution to the state of any systemic delays in direct appeals follows from the Supreme Court’s holdings that states that provide a right to a direct criminal appeal must also provide indigent appellants with counsel (see Douglas v. California, 372 U.S. 353, 355, 83 S.Ct. 814, 815, 9 L.Ed.2d 811 (1963)) and trial transcripts (see Griffin, 351 U.S. at 20, 76 S.Ct. at 591). Implicit in those decisions is the recognition that the denial of such services is nothing less than a denial of access to the courts. Where systemic problems inordinately delay an indigent appellant’s access to the appellate court, the right to an appeal truly becomes “the right to a meaningless ritual” CDouglas, 372 U.S. at 358, 83 S.Ct. at 817). To deny an indigent appellate access to his appeal and then to hold him or her responsible for that denial would be truly Kafkaesque. 18. Here it was uncontroverted that since at least 1988 the Illinois General Assembly has been on notice that OSAD statewide has had a growing problem of inadequate staff to handle a trend of increasing numbers of appointments — both in First District and elsewhere in the State. Despite a plea by the State Appellate Defender in the spring of 1992 that OSAD was unable to keep pace with its appointments at the then-current staffing levels, the General Assembly actually slashed OSAD’s statewide budget by 18%, forcing severe staffing reductions statewide. Although OSAD’s pre-FY 1993 funding was restored in FY 1995, that has plainly been insufficient to meet the demands of an increase in the level of appointments to new appeals. Most recently (less than a month ago) the Chief Justice of the Illinois Supreme Court has reconfirmed that the General Assembly has not appropriated adequate funds to enable OSAD to discharge its responsibilities statewide. Clearly the State of Illinois and not the petitioner class bears the responsibility for the delays in appeal that have been caused by the major underfunding. 19. There have been a number of ways in which the First Appellate District has also been apprised of the OSAD funding problems and the appellate delays caused by underfunding: (a) In response to the funding crisis of FY 1993, First District filed motions to withdraw from a large number of its pending cases, citing its inability to discharge its responsibilities to all of its clients in a timely fashion. (b) At this Court’s instruction, petitioner’s counsel wrote a letter to the First Appellate District advising it as to the pendency of this litigation. (c) Pelletier has corresponded with the First Appellate District seeking the appointment of private counsel in the cases of certain petitioners before this Court and in the eases of other OSAD clients. *1274 (d) Most importantly, the First Appellate District is regularly apprised of First District’s backlog and delay problems in motions for extension of time to file records and briefs that First District must frequently file with the First Appellate District. 20.Although the judicial branch of government lacks the power of the purse to provide OSAD with the added resources to cope -with the problem of appellate delay, it need not stand helpless in the face of that serious problem as reflected by the Hearing. Yet no substantial steps have been taken by the First Appellate District to address the problem. OSAD’s mass motions to withdraw in the face of the difficulties created by the FY 1993 funding crisis were denied without exception. Instead the First Appellate District has permitted the situation to become increasingly exacerbated by simply and routinely granting OSAD’s forced motions to extend time for filing its briefs, rather than taking substantial steps to increase the pool of lawyers involved in the handling of indigent criminal appeals. In response to the point made by this Court many months before the Hearing, the First Appellate District has acknowledged that it is “aware of the problem” of appellate delay and that it has “inherent authority” to appoint counsel to take on appellate cases that First District is not in a position to handle in timely fashion. But except for a very small number of cases that have been assigned to private counsel pursuant to a pro bono program implemented by the Judicial Advisory Council for Cook County, no action to appoint other counsel has been taken by the First Appellate District in exercise of the authority that appears to be conferred upon it by 725 ILCS 5/121-13 to appoint minimally compensated private counsel to handle the excess caseload. In the face of the protracted, serious and large-scale constitutional problem that has been identified in these Findings and Conclusions, it simply is not a sufficient response for the First Appellate District to be efficient in its own consideration and disposition of appeals (as it unquestionably is) once those appeals have finally become ready for such disposition by the belated completion of the briefing process. 21. In all events it cannot be said that the petitioner class bears any responsibility for any failure on the part of the First Appellate District to address the problem of appellate delay. As Harris I, 938 F.2d at 1065 has said, such delays that have been “endorsed and made possible by the willingness of [the appellate court] to grant ... lengthy continuances” and by that court’s failure to act are the responsibility of the State of Illinois, not of the petitioner class. 22. No evidence was presented at the Hearing that the delay in petitioners’ appeals is in any way attributable to petitioners themselves. Again any failure of First District — a state-funded indigent appellate defender agency — to brief petitioners’ direct appeals in a timely manner cannot be attributed to the clients (see Muwwakkil, 968 F.2d at 285; Harris I, 938 F.2d at 1065; Coe, 922 F.2d at 531). And that would be true whether the delays have been caused by the underfunding of First District or, as respondents have contended during the Hearing (unper-suasively, it should be said), by any failure of First District to utilize the funds that it has received in an appropriate and efficient manner (see Harris II, 15 F.3d at 1562 & n. 13) — either (a) by continued processing of cases that do not raise substantial issues or (b) by not placing certain cases on a fast track or (c) by an incorrect allocation of resources to post-conviction appeals or (d) by any failure to be sufficiently productive or (e) by failing to file more Anders briefs. 23. Although any asserted mismanagement on OSAD’s part would therefore not derogate from petitioners’ entitlement to relief in this action, it could of course bear on the nature of the relief to be accorded to petitioners. Hence these Conclusions will go on to address the issues advanced by respondents in criticism of OSAD generally and First District specifically for the handling of its caseload. As the Findings have already indicated and as the following Conclusions reflect, such criticisms — assertions that the agency has not managed its inadequate resources appropriately — are unwarranted. 24. OSAD’s enabling statute (725 ILCS 105/1 to 105/11) and the Illinois Rules of *1275 Professional Conduct require OSAD’s attorneys to provide loyal, competent and zealous representation to all the clients whose cases the agency is appointed to handle, so long as the cases are within the ambit of OSAD’s responsibilities as defined by the enabling statute. There is thus no legal basis upon which OSAD can decline to “process” a case to which it has been appointed, even if an OSAD lawyer may hold the view that other cases to which the office has also been appointed present more substantial issues. 25. When an agency such as OSAD is appointed to more cases than it can timely handle, as has been and continues to be the case here, conflicts of interest are necessarily created as a surfeit of clients compete for the scarce resources of available attorney time and attention. As In re Order on Prosecution of Criminal Appeals by Tenth Judicial Circuit Public Defender, 561 So.2d 1130, 1135 (Fla.1990) (per curiam) has accurately stated: When excessive caseload forces the public defender to choose between the rights of the various indigent criminal defendants he represents, a conflict of interest is inevitably created. As the court below stated, “The rights of defendants in criminal proceedings brought by the state cannot be subjected to the fate of choice no matter how rational that choice may be because of the circumstances of the situation.” Indeed, if anything that problem is less amenable to solution with indigent clients who have nowhere else to go than it might be with clients who have hired their lawyers and then, if dissatisfied with the attention that their cases are getting, may choose to go elsewhere for representation. Those conflicts of interest pose an inherently intractable dilemma that admits of only one possible solution — the agency must seek to withdraw from cases to which it has been appointed until there are sufficient available resources of attorney time and attention to eliminate the conflict (id. at 1138). 26. Respondents’ Hearing evidence suggested some possible alternative “solutions” to the ethical dilemma faced by a defender agency that has more clients than it can represent adequately. Those proffered “solutions” — working first on “clearly winnable” cases, working first on cases with short records, working first on cases with short sentences — are not capable of being implemented without heightening the conflict of interest among the agency’s clients. No better or fairer way to resolve the problem emerged from the Hearing than the first-in-first-out method of selecting the priority of cases for briefing that First District uses. For one thing, it is impossible to determine without substantial detailed review of a record whether a particular case presents “winnable” or “winning” issues. Indeed, even after such review the likelihood of success in a particular ease may be difficult to assess. Any system of assigning priorities to cases with short records or those involving short sentences is simply unfair and prejudicial to clients whose records or sentences are longer. 27. In summary, there is no basis on the record to conclude the OSAD’s policy of briefing eases on a first-in-first-out basis constitutes mismanagement of the agency’s resources. Just as importantly, the record is devoid of evidence that the adoption of any other method of case handling or priorities would reduce the backlog that has created and continues to cause the deprivation of petitioners’ constitutional rights (see Finding 65). 28. It could perhaps be argued that direct appeal cases are more deserving of immediate attention than are post-conviction appeals because in many, though not all, of the latter category of cases the post-conviction appellants have already had one trip to the Appellate Court on direct appeal. Even if that contention were to be accepted arguendo, First District cannot as a practical matter resolve its backlog situation by adopting a policy of briefing direct appeal cases before post-conviction appeals. It is undisputed that even if post-conviction appeals were eliminated entirely, First District’s attorneys could not keep pace with the current volume of new appointments. Moreover, the adoption of such a policy would result in post-conviction appeals never being briefed — an unacceptable derogation of OSAD’s ethical obligations to the clients and to the appoint *1276 ing Appellate and Circuit Courts in those matters. 29. It was established by the Hearing evidence that First District’s attorneys work hard to produce briefs for their clients, and nothing in the record suggests that the representation the attorneys provide is lacking in quality. OSAD has a work standard that requires more output of its attorneys than is required by NLADA’s nationally recognized baseline standard for appellate public defender productivity. With few exceptions, OSAD’s attorneys in fact meet the requirements of the OSAD work standard. Despite the fact that the Cook County Public Defender’s Office’s Appeals Division has a less difficult caseload, First District’s lawyers produce as many substantive opening documents (briefs, motions to dismiss and motions to withdraw) for their clients as do the lawyers in the Appeals Division. There is no basis in this record upon which to conclude that First District is mismanaging its resources by failing to be sufficiently productive. 30. It would be entirely improper for First District to seek to expedite its cases by filing more motions to withdraw from appeals under Anders v. California, 386 U.S. 738, 87 S.Ct. 1396, 18 L.Ed.2d 493 (1967). Anders is not a docket management tool. Motions to withdraw under Anders should be made sparingly and only after great care has been taken to assure that an appeal contains no non-frivolous issues. Moreover, a properly prepared Anders motion is often more time-consuming than a merits brief. As our Court of Appeals has said of the Anders practice in the criminal appeals with which it must deal (United States v. Kellum, 42 F.3d 1087, 1096 (7th Cir.1994)): Before granting counsel’s motion, we must be satisfied that counsel has conscientiously examined the ease. See Anders, 386 U.S. at 744, 87 S.Ct. at 1400. Only if we agree with counsel’s conclusion that the potential issues on appeal are “groundless in light of legal principles and decisions” will we grant the motion to withdraw and dismiss the appeal. United States v. Eggen, 984 F.2d 848, 850 (7th Cir.1993) (citing McCoy v. Court of Appeals, 486 U.S. 429, 436, 108 S.Ct. 1895, 1900, 100 L.Ed.2d 440 (1988)). Indeed, that points up the fact (also reflected in Seventh Circuit decisions) that Anders imposes special burdens not only on defense counsel, but also on the appellate tribunal itself, to scour the record to make certain that no non-frivolous issues lurk there to make a dismissal of the appeal (rather than affirmance) — the consequence of an Anders motion, see Kellum, id. at 1097 — appropriate. Although the Hearing did not of course examine the pending cases on First District’s docket to determine the appropriateness or inappropriateness of Anders filings either individually or collectively, nothing in the record calls for any conclusion that First District could in fact dispose of more cases by filing more Anders motions. Certainly nothing in the record calls for any conclusion that First District (or OSAD) is mismanaging its resources by failing to file a greater number of Anders motions in its cases. C. Petitioners’ Assertion of Their Rights to Timely Appeal 31. Petitioners’ class membership in this representative-action habeas proceeding is of itself enough to demonstrate that they have asserted their respective rights to a timely appeal (Harris II, 15 F.3d at 1563). Each petitioner has filed a notice of appeal from his or her conviction or sentence or both. There has been no evidence suggesting that any member of the petitioner class has intentionally failed to prosecute his or her appeal. Rather, each of them must wait in line for one of the lawyers at OSAD to become available to work on his or her case because of OSAD’s policy of briefing the cases in the sequence of their respective judgment dates. 32. Testimony at the Hearing by Joy, Powers and Smith, each of whom complained about the delays in his case, is illustrative of the point made in Harris II, 15 F.3d at 1563 (which also dealt with a first-in-first-out policy in the public defender’s office there) that “even if petitioners had complained vigorously about delays in prosecuting their appeals, those complaints probably would have been unavailing.” This Court concludes that the members of the petitioner class have suffi *1277 ciently asserted their constitutional right to a timely appeal. D. Prejudice Due to Delay 33. Harris II is the only reported decision that proposes an analysis for determining when a violation of the right to a reasonably timely appeal occurs in the context of a group petition. And Harris III, 48 F.3d at 1132 contains the most recent pronouncement on the appropriate standard for determining whether the prejudice component of the Barker analysis has been demonstrated: Our experience since Harris II now leads us to hold that a presumption of prejudice will arise when delay in adjudicating the appeal attributable to the state exceeds two years. As in the exhaustion context, this presumption is a rebuttable one. 34. That rebuttable presumption of prejudice where the delay in adjudicating appeals exceeds two years on an across-the-board basis is the only suitable standard for deciding whether or not there has been the necessary showing of prejudice, where (as has occurred in this case) such delays are chronic and systemic and have resulted in the wholesale denial of the right to a reasonably timely appeal to an entire class of petitioners. Any other approach to determining prejudice would unfairly saddle an appellant who- is already being denied access to the appellate court with the additional burden of proving that he or she will be harmed by continued denial of that access. 35. On the undisputed evidence introduced during the Hearing, it must be concluded that the class members have in fact been presumptively prejudiced by the excessive delays in adjudicating their criminal appeals. Conclusions 36 to 42 amplify on the reasons for that conclusion. 36.It is undisputed that all class members must wait at least 18 months (and the evidence as to the present delay on newly-assigned cases, and as to the projected delays on eases to be assigned in the future, is even more egregious) before a lawyer will begin to work on their cases. In all events the resolution of class members’ appeals will substantially exceed two years if current trends remain unaltered. As a consequence, some 45% of the class members (if they have earned their statutory good conduct credits) will have served their entire sentences before their appeals are resolved. For any individuals whose incarceration is in fact unlawful (it will be recalled that First District has a substantial success rate in its cases) and who have no hope of having the errors in their trials or sentencing determinations resolved in time to afford them any relief, the right to a reasonably timely appeal is truly a “meaningless ritual” (Harris II, 15 F.3d at 1558). Of the three class members who testified at the Hearing, Smith most graphically demonstrated that a delayed appeal can mean no appeal at all. Smith testified that he appealed his decision because he believes that he is innocent of the crime of conviction, yet he will in all likelihood complete his term of incarceration before his First District lawyer even files the original brief in his case. Pel-letier testified that Smith’s experience is one shared by other OSAD clients — and even worse, that it has occurred in situations where the Illinois Appellate Court has ultimately reversed the conviction without requiring a retrial. 37. Dr. Phillips’ expert testimony established that the members of the petitioner class as a group are at an increased risk of suffering a range of adverse psychological reactions (including increased anxiety, mistrust, hopelessness, fear, and depression), above and beyond the reactions suffered by incarcerated persons generally, because of the mere fact of their confinement, as a result of the delays in the appellate process. Numerous courts have held that adverse psychological reactions, such as anxiety, attributable to appellate delay can demonstrate prejudice (see Simmons, 44 F.3d at 1170; Harris II, 15 F.3d at 1563; Tucker, 8 F.3d at 676; Elcock, 947 F.2d at 1008; Johnson, 732 F.2d at 382; Rheuark, 628 F.2d at 303 n. 8; Jackson, 844 F.Supp. at 465). 38. By definition the most severe actual damage due to the delays at issue in this action is suffered by members of the petitioner class whose appeals would result in relief from their convictions or sentences or both. But the increased risk referred to in Conclu *1278 sion 37 is not dependent upon whether any individual petitioner’s appeal will in the final analysis ultimately prove successful. This Court credits Dr. Phillips’ testimony that increased risk results from the very thwarting of the hope that liberty will be restored through a right that the State has guaranteed — the appellate process. 39. Respondents’ expert Dr. Metzner does not dispute that some members of petitioner class are in fact at an increased risk of suffering adverse psychological reactions. But by limiting his definition of what constitutes an adverse psychological reaction to what clinically constitute “mental disorders,” Dr. Metzner substantially reduces the size of the group that sustains what he labels as harm, and he relatedly ties the level of the risk to both the likelihood of a successful outcome of the appeal and the magnitude of the relief obtained. That restrictive approach does not properly define the scope of the harm to the class members, for the “constitutionally cognizable anxiety” and “oppressive incarceration” recognized in Harris II, 15 F.3d at 1563 and the other cases cited in Conclusion 37 (criteria that are drawn from the Barker factors) are not so limited. 40. In the broader perspective of harm described in Conclusion 38, Joy’s testimony exemplified the emotionally devastating impact of the delays in the appellate process. Such a candid and sincere articulation of the anxiety, depression, hopelessness and anger produced by the denial of effective access to the appellate process — for an individual who believes, even mistakenly, that he will be released from his confinement once an appellate court reviews the legal errors in his case — is the “constitutionally cognizable anxiety” referred to in Harris II and other cases. 41. Apart from what has already been addressed, Powers’ testimony provided an example of yet another type of prejudice suffered by class members: impairment of the ability to present an appellant’s grounds for appeal or his or her defenses in the event of a retrial (Harris II, 15 F.3d at 1563). Powers testified that he does not know the whereabouts of, and therefore may be unable to locate, a crucial defense witness in the event of a retrial. No evidence was presented demonstrating that Powers’ concerns are unique to him and are not shared by other class members. 42. Although the presumption of prejudice resulting from delays in the appellate process is based on the failure to obtain a decision within two years after the filing of the notice of appeal (see Conclusions 33 and 34), that does not of course require that two years must have elapsed before a member of the petitioner class is presumed to have suffered prejudice. When as here the unrelieved burdens and uncured sources of delay assure that the two-year timetable will not be met as to any present member of the petitioner class, the constitutional right of every member of the class to a reasonably timely appeal has been violated and every member of the class has presumptively suffered prejudice. Nothing during the Hearing has suggested a rebuttal to that assumption, either as to the petitioner class as such or as to any member of the class. III. Considerations in Shaping a Remedy 43. Petitioners have established a clear violation of their constitutional rights and are entitled to an appropriate remedy for that violation. But shaping such an appropriate remedy has been rendered far more difficult by what appears to be a distressing lack of response to an obvious and growing problem on the part of the First Appellate District. To be sure, the core offender has been the Illinois General Assembly, whose continued underfunding of OSAD — the office that the General Assembly itself created to provide legal representation of indigent criminal appellants — has created and then exacerbated the intolerable delays reflected in these Findings and Conclusions. However, this litigation is now well over two years old. It has been more than two years since this Court issued its January 25, 1994 opinion granting the motion for class certification — and when it forwarded a copy of that opinion to the Chairman of the Executive Committee of the First Appellate District as a matter of courtesy, this Court referred to the point (emphasized in that opinion) that neither the petitioner class nor this Court was voicing any criticism as to that Court’s expeditious han *1279 dling of the appeals once they had become fully briefed. From the outset of this litigation this Court has viewed it as both seemly and appropriate to extend the highest respect to the members of the state judiciary who are confronted with such a difficult problem. 44. But the expeditious handling of appeals after they have become fully briefed does not address the seminal problem of “too little and too late” in the process of enabling the appeals to reach that fully-briefed stage. Earlier in January 1994 this Court had issued an opinion that, although it was highly critical of the Attorney General’s Office’s initial response to this litigation and to the problem that it posed, concluded with a most respectful request that the First Appellate District might furnish information that could help to lead to a potential solution to the serious problem that was identified in the Complaint, and that has now been confirmed by these Findings and Conclusions. For present purposes it is worth quoting at some length from that January 6, 1994 opinion— first its Background section (which was drawn, as it then had to be, from the Complaint’s allegations, but which has now been totally confirmed by the evidence at the Hearing) and then from the portion extending the request to the First Appellate District. 45. Here is the excerpt from the Background section of the January 6,1994 opinion (1994 WL 8258, at *2): There has been a systemic breakdown in the Illinois appellate system’s First District for the large percentage of convicted defendants who cannot afford to hire private lawyers. Although Illinois law guarantees every defendant the right to one appeal (Ill. Const, art. 6, §§ 4(b) and 6; 730 ILCS 5/5-5-4.1), in the First District that guaranty is currently honored more in the breach than in the observance because Illinois lacks enough lawyers in the Office of the State Appellate Defender and in the appeals division of the Office of the Cook County Public Defender (collectively “Defender Systems”) to handle the volume of appeals with which they are overwhelmed. And although Illinois law expressly allows the appointment of members of the private bar to handle the appeals of indigent defendants (725 ILCS 5/121-13), that option has not been exercised by the Illinois Appellate Court for the First District (at least not in the vast majority of cases) — on the contrary, in each of the cases before this Court and over a hundred other eases, the Appellate Court has expressly rejected the effort of the Office of the State Appellate Defender to withdraw (in order to permit substitution by an appointed private counsel) because of the enormous backlog and delays that office faces. 7 Mervin Green’s prospective representative action (Case No. 93 C 7300) and each of the three earlier-filed individual Complaints (in Case Nos. 93 C 5671, 5672 and 5673) tell essentially the same story. Although the Defender Systems are operating efficiently, they are hopelessly understaffed. Because of the tremendous volume of unbriefed cases, no staff attorney can file an opening brief for some two years from the date of conviction. That means that indigent criminal defendants may expect on average at least a three-year delay between their respective conviction dates and the date of resolution of their appeals. Under the Illinois system of day-for-day good time credit, many defendants will thus have served a substantial portion of their terms of incarceration before their appeals are decided. Indeed, many will have served all or virtually all of their terms — thus rendering their statutory right of appeal a nullity. *1280 46. Here is the concluding section of the January 6,1994 opinion (id. at *5-*6): This Court has consistently demonstrated during the pendency of these proceedings that it has no desire to create a set of constraints that would place unreasonable or unrealistic burdens on the state appellate system. Yet it cannot of course ignore the mandate of the Constitution that convicted felons too are entitled to the protection of due process of law — in this case, the elimination of excessive delays in the appellate process. Accordingly the Illinois Appellate Court for the First District is respectfully requested to respond to these questions to enable this Court to fashion an appropriate remedy if a representative action is indeed authorized on behalf of a certified class: 1. How soon can (a) appointments of private counsel be made to represent appellants in the oldest cases on appeal that are not yet being actively worked on by the Defender Systems and (b) a structure and procedure be established for the systemic appointment of private counsel in cases that are beyond the capacity of the Defender Systems to handle on a reasonable schedule? 2. What would be a reasonable target to establish as a timetable for the disposition of appeals if something similar to the type of remedy urged by petitioners were to be adopted? 12 This Court trusts that a 21-day period is not unduly short for the formulation of such responses. If such expectations are unrealistic because of the Appellate Court’s need to develop a collective response, this Court would appreciate an indication to that effect within the same time frame. 47. When this Court initiated that respectful inquiry, the First Appellate District had available to it the same type of vigorous response that the Florida Supreme Court had given in 1990 (see the case cited in Conclusion 25) to an identical problem — a “tremendous backlog of appeals [to an intermediate appellate court] by indigent defendants in which briefs are substantially overdue” (561 So.2d at 1131). Indeed, the intermediate appellate court in Florida had issued a sua sponte order identifying the inherent conflict of interest referred to in Finding 62 and in Conclusions 25 and 26 here (id. at 1132). After a thorough analysis, including the confirmation of the duty of an appellate public defender to ask leave to withdraw (just as First District has done here) “where the backlog of cases in the public defender’s office is so excessive that there is no way that he can timely handle those cases” (id. at 1138), the Florida Supreme Court had expressly identified the road that had to be taken (id.): This procedure, however, does not address the existing problem of the enormous backlog of appellate cases awaiting briefs in the office of the Public Defender for the Tenth Judicial Circuit. This situation demands immediate resolution; the constitutional rights of these indigent appellants are being violated. These delinquent appeals must be briefed promptly. We believe this situation can only be resolved by massive employment of the private sector bar on a “one-shot” basis. 9 Yet for these past two years the First Appellate District, despite having acknowledged its *1281 “inherent power” to appoint counsel other than First District (and see also 725 ILCS 5/121-13), has taken no steps even approaching that obvious response. 16 48. There is no satisfaction to be derived by this Court from having been prescient and from now having to deal, fully two years later, with a situation that has not only failed to improve but has actually deteriorated in consequence of a lack of constructive response from the judicial branch as well as the legislative branch of state government. As the preceding Conclusions reflect, this Court’s restrained invitation to the First Appellate District to look into utilizing any means available to it to remedy the problem — an invitation that was fully mindful not only of the respect due to a fellow court but also of the importance of inter-governmental respect in a federal system of government— has generated no effective response in the form of the First Appellate District’s appointment of counsel other than OSAD through the use of that court’s acknowledged “inherent power” or via the authorization granted under 725 ILCS 5/121-13 (or by any other means). 49. In the usual situation where a violation of constitutional rights has been caused or permitted to continue despite full knowledge and ample opportunity to cure on the part of governmental defendants, the appropriate remedy is one that would grant prompt (perhaps immediate) relief to the victim of that violation. To shift to the specific area of habeas corpus law, footnote 12 to this Court’s January 6, 1994 opinion has already said that the most common remedies afforded to persons in custody whose constitutional rights have been found to have been violated are either an order of immediate release or a conditional release order to take effect if the State does not provide a retrial within a specified time. Where as here the petitioners are convicted persons who are already in custody, and where their habeas claims do not address the merits of their convictions or sentences but rather their failure to get their guaranteed appellate hearings, those traditional remedies would pose serious risks— they would require the wholesale release into society of a large group of persons, as to whom the statistics tell us that the substantial majority have been lawfully convicted and sentenced. That would plainly appear to be too high a price to pay for the vindication of petitioners’ constitutional rights, notwithstanding the level of governmental misconduct that has been established on the part of the General Assembly and despite the demonstrated failure to act on the part of the First Appellate District, the two branches of state government that have respectively caused and failed to respond to the problem dealt with here. 50.At the same time, it must also be recognized that addressing the problem of petitioners’ constitutional right to prompt representation (as the First Appellate District needs to do) will also impact seriously on the resources of that court itself. As already found here, the First Appellate District has not added in any respect to the existing delays by its own processing of appeals once they have reached the fully-briefed stage — the time taken by that court in the decisional process has been entirely reasonable. And it may be presumed that such would have continued to be the case if the problem of impermissibly delayed representation had been addressed promptly after this Court issued its respectful invitation two years ago — or even earlier when First District was issuing its cries for help by filing its motions for leave to withdraw: If that had been done, the incremental increase in case dispositions needed to reduce the backlog might well have been absorbed by the First Appellate District on a gradual basis. Now, however, the situation is one of an exacerbated crisis. If massive appointments of counsel are in fact made by orders of the First Appellate District (following the lead of the Florida Supreme Court), the excessive back *1282 log will soon be shifted to that court’s docket — and it will have to devise means other than “business as usual” to grapple with that problem. This consideration adds another level of complexity to the devising of an appropriate remedy for the constitutional violations at issue here. IV. Remedy 51. As stated earlier, only Harris among the reported decisions has addressed a systemic problem presented in the same class action context as this case. And Harris II, 15 F.3d at 1566-67 has confirmed the existence of, and the propriety of exercising, judicial power to issue a conditional writ of habeas corpus directing respondents either (a) to resolve within a specified period of time the appeals of those class members whose direct state criminal appeals have been unconstitutionally delayed or (b) to release those petitioners from their unconstitutional confinement. That type of writ is comparable to what Herrera v. Collins, 506 U.S. 390, 403, 113 S.Ct. 853, 862, 122 L.Ed.2d 203 (1993) has termed the “typical relief granted in federal habeas corpus” (where a petitioner challenges his or her conviction as unconstitutional, that “typical relief’ is to issue “a conditional order of release unless the State elects to try the successful habeas petitioner” (id.)). And that type of writ would mirror the writs that other courts have entered as to individual petitioners who have been the victims of like delays—see, e.g., Coe, 922 F.2d at 532; Simmons, 898 F.2d at 869; Cameron v. LeFevre, 887 F.Supp. 425, 434-35 (E.D.N.Y.1995); and Jackson, 844 F.Supp. at 465. 52. Under Harris III, 48 F.3d at 1132 each of the members of the petitioner class whose appeal has already remained unresolved for two years or more after filing of the notice of appeal would be entitled to such a conditional writ, because respondents have not come forward with evidence rebutting the presumptive unconstitutionality of the continued confinement of those persons. Because this Court has not been apprised as to the numbers of petitioners who fit that description and as to the feasibility of resolving their appeals within a specified short time period, respondents are ordered to file in this Court’s chambers (and of course to serve on respondents’ counsel) a complete report providing that information on or before March 18, 1996 to permit this Court to shape an appropriate order in that respect. There will then be a status hearing at 8:45 a.m. March 22, 1996 to discuss the contents of respondents’ report and the nature of the order that appears to be called for at that time. 53. In addition, this Court also has the power and duty to craft a conditional writ that will address the State’s responsibility to afford meaningful appellate access to all of the other members of the petitioner class, including those whose efforts to obtain appellate review are still in the early stages and have not yet consumed two years. For the reasons discussed under Considerations in Shaping a Remedy, that problem poses added complexities that justify a bit more time in responding. Accordingly respondents are ordered on or before April 2, 1996 to file in this Court’s chambers (and to serve on petitioners’ counsel) a concrete and detailed proposal setting forth means, through the appointment of added counsel or otherwise, (a) to expedite the briefing of cases in First District’s backlog and (b) to provide for a continuing reduction in that backlog, while at the same time maintaining the effective assistance of appellate counsel for the petitioner class members. This Court will then determine what further procedures are required to address that problem. Footnotes [1] References to the parties' exhibits will take the form "P.Ex. — " ("P" standing for Petitioners) or "R.Ex. — ” ("R” standing for Respondents). All references to the Hearing transcript will take the form "Tr. — .” Where such a reference is enclosed in parentheses, the cited pages support the statement in the immediately preceding sentence. But where such a reference is made following several sentences or at the end of a Finding and is not enclosed in parentheses, the cited pages support all aspects of those sentences or of the Finding. [2] Because the First District Office is the focus of this lawsuit, as a matter of pure convenience these Findings and Conclusions will most frequently use the shorter term "First District” (omitting both "OSAD” and "Office”) rather than the full title. That should not occasion any confusion, because "OSAD” will consistently be used to denote the statewide office and all of its districts and because references to the Appellate Court for the First District (most often in the form "First Appellate District”) will always be made clear. [3] Again Respondents' Response 1 quarrels with that characterization (taken from petitioners' proposed Findings) as "neither a finding of fact supported by record citation nor a conclusion of law.” Quite apart from their misperception of the concept of Findings (in which they ignore the propriety of this Court’s drawing of inferences from the evidence of record), respondents' counsel betray an extraordinary lack of sensitivity in failing to acknowledge the distressing nature of the trend disclosed by the evidence. As reflected later in these Findings, the recent annual report from the Chief Justice of the Illinois Supreme Court to the Illinois General Assembly characterizes this as "a critical problem” — does the Attorney General’s office disagree? [4] Pelletier specifically recalled People v. DiGiaco-mo, a case in which an appellant client of First District had been convicted of attempted murder and sentenced to four years’ incarceration and had then been released several months before the Illinois Appellate Court issued its decision reversing the conviction outright and ruling that defendant DiGiacomo should never have been convicted (Tr. 185-86). [5] This does not take account of inmates’ other opportunities (1) for additional credit due to meritorious conduct or (2) for credit for time in custody before trial, factors that make the potential for their being forced to spend excessive time in custody, triggered by the delays at issue in this case, even more troublesome. [6] As also shown by earlier Findings, the delay referred to in the report also represents a substantial understatement. That factual error should not be faulted, for the Supreme Court did *1249 not have the benefit of the record developed at the Hearing here. [7] That situation renders wholly nonanalogous the point made in Thurston, 255 Ill.App.3d at 516, 193 Ill.Dec. at 395, 626 N.E.2d at 428: In the real world, a client would not pay a talented appellate lawyer to appeal such a minor point. Because Illinois provides criminal defendants not only with the guaranty of an appeal but also with an absolute right to representation by counsel in taking the appeal, such free market analysis simply does not apply in the context now under consideration. [8] None of those Findings includes the regular repetition of “In Dean Lefstein's unrebutted opinion,” as petitioners’ proposed Findings had done, but each of those Findings here should be understood in that manner. [9] Although the programming and technical work were performed by Newhouse, that analysis will be referred to as ''Spangenberg’s analysis” in these Findings — it was Spangenberg who provided the expert testimony as to the results and their significance. [10] That 11-case differential appears likely to be attributable to the experts' differing judgments in the exclusion of certain case records as flawed. In any event, the small difference could not affect the level of consistency that has been referred to in Finding 83. [11] This figure covers all appeals from 1989 convictions, without regard to whether the appeal was handled by OSAD, by the Cook County Public Defender or by a private lawyer (Tr. 862). [12] A fortiori, the delays from notice of appeal to actual decision in those other jurisdictions pale in comparison to the corresponding delays being suffered by the petitioner class here before their appeals are actually decided. [13] As the following text figures reflect, after a dramatic reduction in the backlog during 1991, the backlog rose to a substantial level by the end of 1992 and has continued to be substantial since then, [14] Any such discounting would essentially reflect a “no harm, no foul” approach. But it is unnecessary to decide the propriety of such a discounting, in light of the Finding next reflected in the text. [15] As already reflected in Finding 98, Puerto Rico is a total outlier, both literally and (more importantly) figuratively. [7] This Court has no way of knowing whether that pattern of rejections stems from the state system’s lack of a pool of experienced counsel available for appointment, or from a lack of funds for the payment of fees, or for some other reason. If the concern is for payment of fees, it would seem that the provision for allowing fees in a "reasonable amount" in 725 ILCS 5/121-13 could take the overall unavailability of funds into account in setting the award in each case. It is worth observing that in addition to this District Court's ability to appoint counsel on a compensated basis under the Criminal Justice Act, 18 U.S.C. § 3006A, it imposes on members of its trial bar the requirement that they be available for appointment to render services to indigent litigants on a pro bono publico basis — without any payment of fees (General Rules 3.82-3.90), *1280 and our Court of Appeals similarly provides counsel for indigent parties in both civil and criminal appeals. [12] This should not be mistaken as a ruling on the appropriate remedy. This Court is of course sensitive to the fact that decisions as to a defendant's release or imprisonment pending appeal are — or at least should be — made on a reasoned and individualized basis, taking all relevant factors into account. For that reason it is not an attractive prospect for a federal court to review and possibly to override such a decision just because an appeal has been delayed for an unconstitutionally excessive period. On the other hand, it must be remembered that the traditional habeas corpus remedy where there has been a violation of constitutional rights is actually to release the defendant (or to release the defendant unless a retrial is begun within a limited time)— the fundamental theory of habeas corpus being that a defendant is being deprived of his or her liberty unlawfully because of the constitutional violation. [9] It stands to reason that the employment of additional assistant public defenders to handle properly the appeals that are currently being filed will not take care of the backlog of delinquent appeals. [16] It is true that the Florida Supreme Court went on to say (561 So.2d at 1138-39) that the legislature there should appropriate funds for the adequate compensation of appointed private counsel — and the same concept should apply to the General Assembly as well. But nothing in this case suggests (particularly given the First Appellate District’s reference to "inherent power”) that such funding is a precondition to such appointments — and of course providing pro bono representation is in the best traditions of the Illinois bar.