May 1 and 2, 1997

The Civil Rules Advisory Committee met on May 1 and 2, 1997, at the LaPlaya in Naples, Florida. The meeting was attended by all members of the Committee: Judge Paul V. Niemeyer, Chair, Judge John L. Carroll, Judge David S. Doty, Justice Christine M. Durham, Francis H. Fox, Esq., Assistant Attorney General Frank W. Hunger, Mark O. Kasanin, Esq., Judge David F. Levi, Carol J. Hansen Posegate, Esq., Judge Lee H. Rosenthal, Professor Thomas D. Rowe, Jr., Judge Anthony J. Scirica, Judge C. Roger Vinson, and Phillip A. Wittmann, Esq. Edward H. Cooper was present as reporter. Sol Schreiber, Esq., attended as liaison member from the Committee on Rules of Practice and Procedure, and Professor Daniel R. Coquillette was present as Reporter of that Committee. Judge Adrian G. Duplantier attended as liaison member from the Bankruptcy Rules Committee. Peter G. McCabe, John K. Rabiej, and Mark D. Shapiro represented the Administrative Office of the United States Courts. Thomas E. Willging represented the Federal Judicial Center. Deborah Hensler attended and reported on the progress of the RAND Institute for Civil Justice class-action project. Observers included John Beisner, Sheila Birnbaum, Kathleen Blaner, Elizabeth Cabraser, Jonathan Cuneo, John P. Frank, Danita James, Beverly Moore, Ira Schochet, Fred S. Souk, and H. Thomas Wells, Jr. (liaison, ABA Litigation Section).

Chairman's Introduction

Judge Niemeyer opened the meeting by noting the progress of the Judicial Conference report to Congress on the results of the Civil Justice Reform Act. This Committee met in conjunction with the ABA conference on the CJRA in March. Members of this Committee had worked with members of the Court Administration and Case Management Committee to help shape the CACM draft report that was submitted to the Judicial Conference. The Judicial Conference Report has not yet been delivered to Congress, and remains "embargoed," but it is expected to report that it is good to establish early discovery cut-offs and to set a firm trial date early in the pretrial process. The Rules Committees may be asked to consider these techniques. The data gathered by the RAND study will in any event prove useful to the Discovery Subcommittee in its work.

The Policy and Agenda Committee met during the March meeting. Its work will be what its name implies. In addition to helping frame the Committee agenda, it will make recommendations on policy matters to be decided by the Committee. Several matters were discussed for consideration at the present Committee meeting.

The Policy and Agenda Committee noted that this Committee's work cannot focus only on consideration of a rule and recommendations for improvement. There are many constituencies to be addressed. These constituencies include the familiar constituencies that make up the Enabling Act process -- the Standing Committee, the Judicial Conference, the Supreme Court, and Congress. They also include an increased level of interest within the bar, in the business community, and in the media. Congress is taking ever greater interest in procedural matters. If this Committee believes in a proposal, it probably will have to work harder to encourage adoption, keeping many different groups informed of the proposal and the justifications for it. The ideal rule change is one that is purely procedural, that "creates peace," and is satisfactory to all sides of a dispute. Achieving such changes is difficult. There is a great risk that changes will be seen to favor one "side" or the other, whether or not that is so. Discovery is a good illustration. Enough discovery to uncover the proverbial "smoking gun" always seems to be a good thing. But that may not be the fair measure of a good judicial dispute-resolution procedure. Any proposals in this area are likely to be scrutinized closely from many different perspectives and positions of interest.

It is important that we keep Congress informed of what may be coming through the Enabling Act process. It also is important to keep in touch with Congress on legislative initiatives to revise procedure. The values of the Enabling Act process must be continually emphasized. At the same time, this Committee should be alert to the possibilities of enlisting the help of Congress with matters that seem to call for resolutions that lie, in part or entirely, beyond the reach of the Enabling Act process. The interplay of substance and procedure seems to be growing ever more persistent, and may increasingly call for joint solutions.

Press releases provide one means of reaching the more numerous constituencies. The chair has begun to issue press releases, believes them to be a good and healthy means of illuminating the Committee's work, and plans to continue to use them.

There are persisting questions as to the reach of public access to Committee work. Meetings of course are public "sunshine" meetings, unless special reasons require an executive session. But what about telephone calls on Committee business? Probably they are public. Every writing exchanged on Committee work is a public paper. When in doubt, questions should be resolved on the side of keeping correspondence and other writings in Committee records. At the same time, the Committee process is a deliberative one. There are some aspects that should not require immediate publication. The subcommittees should be able to have working sessions that are not open to the public. Further questions are raised by conversations over meals, even over who is allowed to attend meals that occur during Committee meetings; by work with consultants; by conversations with those who represent one or another "interest." Committee members and staff must be sensitive to any appearance of forming relationships with those who represent defined interest points of view, and should not accept entertainment or food gifts.

Discussion of these matters led the Committee to conclude that it is proper for a subcommittee to meet or confer without public notice when, as a working committee, the subcommittee believes that course desirable.

Legislation Report

As noted in the introduction, Congress is increasingly interested in procedural matters, and in the work of this Committee. The legislative process can bypass the Enabling Act process. Often Congress does not duplicate the advantages of the extremely deliberate but thorough procedures dictated by the Enabling Act. But Congress may feel it important to get results faster than the Enabling Act permits. Accommodation of these competing interests can be difficult. The Enabling Act process will be better served if most lawyers can be persuaded of the importance and benefits of the process.

The pending legislation that is closest to the present work of the Committee would create a permissive interlocutory appeal procedure for orders granting or denying class certification. The sponsors know that proposed Rule 23(f) is well advanced, and that the Standing Committee could recommend its adoption to the Judicial Conference this summer. Other bills would create new offer-of-judgment procedures; one of them would directly amend Civil Rule 68. The Sunshine in Litigation Act, which would impose new requirements and limits on discovery protective orders, has again been introduced.

Other pending legislation that directly affects the judiciary does not seem to fall directly within the scope of this Committee's work. Among other matters, pending bills would seek to control "judicial activism," create one judge-disqualification right for all parties on each side of an action, and place limits on orders that direct increases in state or local taxes. It was pointed out that several states have statutes that allow a party to disqualify one judge from hearing an action.

The Standing Committee and other Judicial Conference committees regularly write to Congress about bills that affect federal courts and federal procedure. Judge Niemeyer also reported on his meeting with Congressman Charles Canady on Rule 23(f). It was agreed that the Committee should make Congress better aware of the Committee's work and processes. Congressional aides should be invited to Committee meetings. It will help to show Congress that the Committee's process generally enlists the help of the most able legal minds in the country, and that advice from all sides of an issue is seriously considered. But the Committee must recognize the capacity of Congress for speedy action.

Standing Committee Report

Professor Coquillette, Reporter for the Standing Committee, summarized portions of the Standing Committee's work that affect this Committee.

The Standing Committee has some responsibilities that affect all of the advisory committees alike. The style project is one. The Standing Committee wants to have all the sets of rules conform to its style conventions. It hopes that as much of the work as possible can be done by exchanges between each advisory committee reporter and the Standing Committee style consultant, Bryan Garner.

Relations with Congress are the direct responsibility of the Standing Committee. Congressional aides are regularly attending the meetings of the Evidence and Criminal Rules Committees. The Standing Committee believes it important to resist any pressures to accelerate the Enabling Act process.

The Standing Committee has devoted much effort to the questions that arise from local rules regulating the conduct of attorneys. It may decide at its next meeting on the best form of response to these problems. Among the possible responses would be a uniform or model local rule; this and some other possible forms of response are likely to be handled directly by the Standing Committee and its Local Rules project. One possibility, however, is adoption of a uniform national rule. If that form is chosen, the Civil Rules Advisory Committee is likely to become involved.

The Civil Justice Reform Act local rules are approaching the statutory sunset date. Many of them have not been adopted under the regular local rules process. To endure, they will have to be readopted as local rules, subject to the constraints on local rules.

Discussion of local rules focused on the importance of national uniformity. It has been difficult even to achieve a uniform numbering system for local rules. There is likely to be even greater resistance to efforts to achieve national uniformity by superseding local rules. Wide differences in disclosure practice have emerged under Civil Rule 26(a)(1). The local bar culture in districts that have departed from the national rule is likely to generate pressure against an amendment of Rule 26(a)(1) that would remove the permission to depart. Local lawyers feel little sympathy for the needs and interests of the mobile, nationwide federal bar. This topic will be one of the items studied by the Discovery Subcommittee, beginning with the discovery conference in September.

It was further noted that the Judicial Conference is considering the possibility of establishing a "mass torts" committee. If there are to be recommendations for substantive law, they are not likely to come to this Committee. But this Committee has gathered much information about mass tort litigation in the course of its class action work. Some means should be found to bring this Committee's experience into the study. And if any proposals are made with respect to the class action procedure established by Civil Rule 23, this Committee certainly should be heard.

Report of Discovery Subcommittee

Discovery will be the topic of the September meeting, to be held as a conference at Boston College. The meeting will be primarily a "listening" meeting for this Committee. The work of sorting through the information and setting an agenda of specific topics for consideration by the Discovery Subcommittee will come at the October meeting.

The September meeting will begin each day with a brief review of discovery history. Most of each day's program, however, will focus on panel discussions. It is hoped to get as much input as possible from practicing attorneys. Invitations to participate have been extended to many bar groups; we hope for written presentations from each.

It was suggested that there have been marked changes in discovery practice in the last four or five years. The growing use of magistrate judges and the Civil Justice Reform Act have made a big difference.


The Minutes of the October, 1996, and March, 1997 meetings were approved.

Rule 23

Proposed Rule 23 amendments were published in August, 1996, for comment. The volume of written comments, statements, and testimony was impressive. All have been collected in a four-volume set of materials. The Committee is deeply grateful to the many lawyers, judges, and bar groups that expended great time and effort to share their experience.

Having gathered this information, the Committee now must report to the Standing Committee. Not only must the published proposals be reviewed; new proposals advanced in the process -- often reflecting proposals that this Committee had considered to some extent -- should be reviewed as well. At the March meeting, the Committee concluded that all Rule 23 issues should be considered open for further consideration or final action. The central issues are summarized in Judge Niemeyer's March 15 memorandum to the Committee.

Two proposals provoked the greatest volume of comments. Perhaps the greatest attention was drawn by proposed Rule 23(b)(3)(F), which would allow a court to deny class certification because the probable relief to individual class members does not justify the costs and burdens of class litigation. The reactions to this proposal demonstrated that it goes to the very heart of the purpose of Rule 23. The central question is whether -- and when -- Rule 23 should be used not for the purpose of providing meaningful individual relief but for the purposes of enforcing public values, of forcing wrongdoers to internalize the costs of their wrongs. Much comment also was directed to proposed Rule 23(b)(4), which would allow certification of a (b)(3) class for settlement even though the same class might not be certified for trial. Settlement classes raised complicated issues. Some of these issues are likely to be resolved, and others illuminated, by the Supreme Court decision in the pending Amchem litigation. It was agreed in March that it would be premature to act further on proposed (b)(4) before the Court has rendered its decision.

The agenda materials arranged the proposals in three groups for purposes of discussion. The first group included the least controversial matters, the proposed change in Rules 23(c)(1) and new Rule 23(f) for interlocutory appeals. The second group included rather more controversial matters, proposed factors (A), (B), and (C) for Rule 23(b)(3). The third group included the proposal to amend Rule 23(e) as well as the quite controversial Rules 23(b)(3)(F) and (b)(4).

A letter from Judge Patrick E. Higginbotham, former chair of the Committee, was summarized. Judge Higginbotham suggested that the (b)(4) settlement-class proposal must await the Supreme Court decision in the Amchem case. He further suggested that years of work would be required to establish the proper shape for the concept expressed in the small-claims proposal advanced in (b)(3)(F). The remaining proposals, however, can properly go forward now.

It was asked why the settlement hearing proposal in subdivision (e) had been grouped for discussion with the controversial matters. It was responded that the hearing requirement indeed could be handled independently. Rule 23(e), however, has been closely linked throughout this process with settlement-class proposals. The proposed hearing requirement, indeed, was added as part of the decision to recommend subdivision (b)(4). Much of the testimony bearing on settlement classes suggested greater changes in subdivision (e). Further consideration of settlement classes is likely to shape not only any settlement-class proposal that may yet be made but also subdivision (e).

Timing of Proposals

Discussion turned to the question whether any part of the Rule 23 proposals should be recommended for present adoption while the settlement class questions remain pending. This question might be affected by the choice whether to open up new proposals in addition to those published for comment in 1996.

The first new proposal discussed was the possibility of adopting an opt-in class for some situations. An opt-in class could be used to replace the opt-out nature of (b)(3) classes. Opt-in classes instead could be added as an alternative to (b)(3) opt-out classes, either generally or for specific situations. One recurring suggestion was that an opt-in alternative might be a good means of addressing classes involving claims so small that individual class members may not care about vindication of their claims.

Opt-in classes could bear on each of the new factors proposed as matters pertinent to the decision whether to certify a (b)(3) class. If a general opt-in class should be proposed, or if (b)(3) should be amended to provide for opt-in rather than opt-out classes, the factors could be affected substantially. At least as important, an opt-in proposal clearly would require publication for a new round of comment. Thought might be given to seeking advice outside the Committee even before publication.

A note of caution was sounded. The Committee has done a lot, and has worked a lot. There is no compulsion to do anything, much less to propound a comprehensive package of Rule 23 amendments. It would be proper to propose nothing for actual adoption. It would be appropriate to propose modest changes now, reserving the question whether more substantial changes should be proposed later. Substantial questions have been raised even as to proposals that were regarded as modest, and that probably should be implemented -- if adopted -- in a modest way. The opt-in question presents the possibility of drastic changes; if it is to be considered, it will be a long-range project. The Committee must be careful not to make a complex rule even more complex.

At the other end of the complexity spectrum, it was noted that the proposed amendment of subdivision (c)(1) and the new interlocutory appeal provision could reap significant benefits, and are easy to implement. Appellate courts have strained to take a more active role in class-action law in recent years, with good results. Affording a more regular means of involvement, increasing the opportunities for appellate review, may do much to simplify current law and make practice more nearly uniform. In addition, Congress is interested in the interlocutory appeal question; it would be good to demonstrate the responsiveness of the rulemaking process.

And so the question is what sort of package, if any, should be presented now. This question was seen to be one that could not be fully resolved before exploration of the several proposals. If some are ready to be advanced now, while others require further work, the wisdom of acting on some now will depend in part on the length of the anticipated delay. If there is likely to be a relatively extended delay, the risk of rapid successive amendments is much reduced.

The Committee agreed to consider first the (c)(1) and (f) proposals.

Rule 23(c)(1)

Rule 23(c)(1) now requires the court to determine whether to certify a class action "as soon as practicable after the commencement of an action." The published proposal would change this to "when practicable."

The first note was that current style conventions would support a substantial rewriting of this sentence to achieve the same meaning with fewer words. At the same time, the simple change from "as soon as" to "when" has at least two virtues. It emphasizes the nature of the one intended change of meaning. And it reduces the risk that arguments will be made to find unintended changes of meaning in other language changes. It was agreed that the special complexities and sensitivities of Rule 23 counsel restraint in style.

The proposed changes in (c)(1) were approved unanimously, to be recommended to the Standing Committee for submission to the Judicial Conference.

Several changes were made in the (c)(1) Note. In part, the changes reflected the anticipated decision not to send forward now the published changes in subdivision (b)(3) or the proposed settlement-class provisions of (b)(4). Other changes deleted proposed references to public comments, to deferring the certification settlement pending settlement attempts, and to decisions that have hesitated to decide Rule 12(b)(6) or Rule 56 motions before deciding whether to certify a class. With these changes, the proposed Note was unanimously approved.

Rule 23(f)

The permissive interlocutory appeal provision of proposed Rule 23(f) was approved unanimously as published. It will be recommended to the Standing Committee for submission to the Judicial Conference.

A possible revision of Rule 23(f) was noted. Several witnesses urged that a class certification decision may rest on misunderstandings that may yield to quick correction on motion for reconsideration. Rather than force an attempt to appeal for fear of losing the 10-day time limit, the rule should provide that the 10-day appeal period is suspended by a motion to reconsider made within the 10-day period. The appeal time would begin anew upon disposition of the motion to reconsider. This revision was found to raise complicated issues that could not be easily resolved in the text of the rule. The topic was moved to discussion of the published Note to Rule 23(f).

Several changes were made in the Note, responding to the public comments and testimony. The references to 28 U.S.C. § 1292(b) were revised to make it clear that none of the restrictions that surround § 1292(b) appeals apply to subdivision (f) appeals. Passages predicting that permission to appeal would be granted with restraint, and that the proposed change is modest, were removed. The suggestions that district courts may wish to comment on the desirability of appeal, however, were retained.

The value of district court suggestions on the desirability of appeal led discussion back to the effect of motions to reconsider. It was noted that there are no clear analogies in present practice. Section 1292(b) depends on certification by the trial judge, which can easily be withheld pending a timely motion to reconsider. The possible complications of motions to reconsider are further reduced for § 1292(b) appeals by the power of the district court to enter a new § 1292(b) certification after expiration of the time to appeal triggered by an initial certification. Since a Rule 23(f) appeal is not a matter of right, the rules that surround the effect that a notice of appeal has on continuing district court proceedings are not directly apposite. The question may arise, moreover, on motion made before a petition to appeal is filed, after a petition to appeal is filed, or after a petition to appeal has been granted. The effects on district court power to grant reconsideration and amend the certification decision may vary among these several situations. The occasions for district court reconsideration, moreover, may turn on incomplete presentation of the initial certification arguments, better knowledge of the issues that emerges as the action progresses, or partial pretrial dispositions or proposed settlements. Many of the possible problems are likely to be worked out in a pragmatic accommodation between the court of appeals and the district court. It was concluded that there are too many issues to be addressed cogently in the Note to Rule 23(f). They must be left for development as the courts of appeals find best.

Preliminary RAND Report

Deborah Hensler gave a preliminary report on the progress being made in a study of class actions undertaken by the RAND Institute for Civil Justice.

The study was initiated a year ago at the urging of Judge Patrick E. Higginbotham, then chair of this Committee.

The study is designed to build on, and supplement, the Federal Judicial Center study that was undertaken at the request of this Committee. Its goals are to describe the current class-action landscape; to describe the practices of attorneys and parties with as much objectivity as can be brought to a study based on their own descriptions of their own practices; and to assess the consequences of class actions for claimants, consumers, businesses, and society. The approach begins with creating a data base. Then practitioners are interviewed. And finally, selected cases will be studied intensively in an attempt to measure the costs and outcomes.

It was not feasible to determine the total number of class actions, either for federal courts or for state courts. Instead, an effort has been made to identify the range of class actions that are being filed. Reliance has been placed on a variety of data bases, all of them subject to electronic search. These sources will show what is going on now, but do not show changes over time.

As of April 30, interviews have been conducted with 34 different attorneys, some of whom asked associates to participate so that more than 50 attorneys have been involved. Many of the interviews have lasted several hours.

An early step was to group class actions into categories. The FJC data were consulted, with the help of the FJC, so that the FJC cases could be reshuffled into the same categories. The most numerous categories of class actions involved securities and civil rights. Employment cases, including employment discrimination, came next. The RAND data bases show a larger proportion of "consumer" class actions than the FJC data. The difference is thought to arise from the fact that state cases are included in the RAND data, and also from the fact that the RAND data cover a more recent period.

Reports of class actions in the general press for one year, from July 1, 1995 through June 30, 1996, show 3,080 class actions. This number includes cases that simply have class allegations; some of the cases may have disappeared without any action whatever, and many may never be certified.

Mass torts are a smaller proportion of class actions than the categories already noted, even including property damage cases in the mass tort category.

"Consumer" cases tend to include antitrust, fraud, and "fee" cases that involve charges by service providers -- insurers, credit-card issuers, and the like. Antitrust cases often follow on the heels of public actions, or stimulate public enforcement. "Fraud" cases involve a variety of deceptive practices.

The data bases do not provide complete information whether the class actions were filed in state or federal court. Over the same one-year period, 413 appellate decisions in class actions were found. A majority of these decisions were in state courts.

The interviews show a dramatic increase in class-action activities, suggesting a doubling or tripling in the last two or three years. All agree that the increase is mostly in state courts. Multiple competing suits are increasing, as are races to certification.

There has been a dramatic change in the landscape in another way. The "absent plaintiff" class -- often for an injunction, or for damages readily calculated by formula -- remains, but there are now more aggregations of individual claims that require individualized determinations of damages. Common-law claims are more prevalent. There are more cases that seek a mix of remedies.

It will be very difficult to identify categories of cases that are particularly likely to present problems. The new kinds of class actions "are all over the map." Class members' damages may span a spectrum from low to high in many subject areas.

The interviews with lawyers show disagreement as to what the problems are, and as to which problems can be fixed.

There is a lot of agreement that class actions sometimes are useful. Even defendants recognize this. And many, including even plaintiffs, agree that there are problems.

There is general agreement on many propositions. Fees play a primary role in plaintiff filing decisions. Settlements are driven by defendants' risk aversion. Unless judges exercise careful supervision, the system will encourage filing on weak claims, but also will encourage settlements that are too low in strong-claim cases.

There are attorneys who electronically scan news reports, government announcements, and like sources for events that can support class actions. Many of the resulting class actions are meritorious. Many are not. All agree that these activities help drive the surge in filings. The surge is in part fueled by the common judicial tendency to follow a "first filing" rule that gives precedence to the class -- and its attorney -- that first filed. This practice is a problem, however difficult it may be to correct. There also are problems arising from "copycat" filings that seek at least to become involved in the action. Some lawyers believe that it is possible, and clearly desirable, to adopt better rules as to priorities in the race to represent a class.

In private, defense lawyers will recognize that settlements are offered on the basis of little discovery, small class recovery, and nice fees for plaintiff counsel. Plaintiffs' lawyers will say that they never offer such settlements, but that defendants often approach them with such offers.

There is substantial criticism of fee awards based on noncash compensation ("coupon settlements"), or on hypothetical changes in the defendant's behavior, or on a hypothetical class size. Defendants recognize that these deals are often in their own interest. There is strong evidence that, although criticized, these practices are quite common.

If we think reform is desirable, what is the means? More demanding certification practices may help. Federal certification standards already seem to be tightening -- and many lawyers think that is the reason for the movement to state courts.

The resolution process also may be a focus for change. Whether plaintiffs or defendants are the initiators of bad settlements, it is agreed that the problem is that judges often are not discharging their responsibilities to review settlements, nor their responsibilities in scrutinizing fees in relation to work done. The need for judicial supervision exists in all class actions, not only settlement classes. The interests and incentives of the parties demand that judges take their Rule 23(e) responsibilities more seriously. The strong general favor of settlement, and the sense that judges should not intrude, do not work as well here.

The final questions address the consequences of class actions, good and bad. These will require measurements of many things. Transaction costs, compensation to class members, bad products improved or good products withheld, the public repute of courts, and so on are involved. These questions will be very difficult. Broad-scale judgments probably will prove impossible. The RAND study will attempt only to focus on what a select number of individual actions have achieved.

ADR methods have been used in the post-resolution phase of class actions, primarily to determine individual damages. RAND is struggling with the question of how many individual case studies it will be able to do within budget limits. They hope to complete at least half a dozen by this fall. These studies may shed some light on the use of nationwide classes in state courts.

The Committee expressed great thanks and appreciation to Dr. Hensler for her presentation.

Rule 23(b)(4)

It was agreed without further discussion that the settlement-class proposal published as Rule 23(b)(4) should be deferred to the next meeting that discusses Rule 23. The impending Supreme Court decision in the Amchem case will provide a much more secure foundation for further consideration of settlement classes.

Rule 23 General Discussion

The remaining published proposals include factors A, B, C, and F in Rule 23(b)(3). In addition, the hearings have renewed more philosophical questions as to the proper role of Rule 23. These questions are summarized in the March 15 Memorandum from Judge Niemeyer to the Committee. The possibility of adopting opt-in classes has been advanced by several witnesses, and provides a useful focus for the deeper questions.

The deliberations that led to adoption of Rule 23(b)(3) in 1966 included consideration of a (b)(3) class without an opportunity to request exclusion, but it was decided not to force people to be in such a class. Opt-out was adopted. No one knew what the effects would be. Professor Kaplan, the Committee Reporter, suggested that it would take a generation to learn the consequences. We have a generation of experience. It may be time to reconsider the assumption that class members are litigants until they opt out.

In a world of perfect communication, there would be no difference between opt-out and opt-in. Each class member would actually receive notice, would fully understand the nature of the litigation and the consequences of being in the class, would form a sophisticated judgment as to the desirability of being in the class, and would opt in or out as the rule might demand. All the testimony, however, confirms the Committee's sense that there is an enormous difference between opt-in and opt-out. The "default" mechanism is vitally important. Both the capacity of Rule 23 to accomplish important social ends and the untoward pressure that may be placed on defendants depend on how many people the class representative controls.

The present-day effects of opt-out classes have never been debated in the Enabling Act process. They were not foreseen. The question, however, is not what was intended, but what the rule has become. Plaintiffs welcome the aggregation. Defendants fear the pressure exerted by large classes, but also welcome the opportunity to achieve peace. The rule lies on the edge between substance and procedure. The proponents of the practice that has evolved around Rule 23(b)(3) argue that it is an essential public enforcement tool, a tool that Congress has relied upon. The rule in this form is not neutral in its substantive impact. Changes in the rule will have substantive impact. Should the rule have the substantive impact that comes from aggregating the "clout" of class members, or should it be only a device for aggregating the claims of those who deliberately choose to become involved? The time is right to consider this question, whether or not it proves desirable to do anything, or possible to do whatever may seem desirable.

It was suggested that opt-in classes would provide no real protection for defendants. There would be ongoing individual actions, or multiple opt-in classes. Defendants would be most unhappy with a general opt-in class provision. If opt-in classes were limited to "consumer" cases, it might not be undesirable. If more general, it would destroy the effective use of (b)(3).

John P. Frank reminded the Committee that the opt-out provision was proposed as a compromise to preserve adoption of the (b)(3) class rule. The Committee then was thinking of 100-person classes. But what they were thinking then is irrelevant now. He urged that if the action is not important enough to a person to warrant the investment of energy and postage to opt in, that person should not be a class member. Opt-in classes should be adopted, at least for the "consumer" class action. The private attorney-general notion is not a social policy that this Committee should make. Congress can do that. The virtually inadvertent creation of a rule in the 1960's should not make this Committee a tool of social regulation in place of Congress and the administrative agencies.

Support for the opt-in approach was expressed with the recognition that it will require more work. The work is worthwhile, because (b)(3) needs to be corrected.

The Committee was reminded that earlier drafts included a general opt-in provision that was an alternative to the opt-out class, not a substitute. Even if opt-in classes were adopted in place of opt-out classes, many opt-out class opportunities would remain available in the state courts. During the early stages of any study of opt-in classes, attention must be directed to the experience under the Fair Labor Standards Act procedure, adopted also for Age Discrimination in Employment Act cases. As to the "social policy" aspect of the change, it must be remembered that any substantive effects of the present rule emerged from the rule. If the rule could be propounded in the Enabling Act process, surely the Enabling Act process can amend the rule. Amendment, however, will affect the substantive effects wrought by the initial rule.

Thomas E. Willging reminded the Committee of an empirical study conducted in the 1970's of three cases in which, without apparent authorization in Rule 23, district courts certified opt-in classes. The opt-in procedure seemed to have a dramatic effect in reducing class size.

It was urged that proposed factor 23(b)(3)(F) represented this Committee's conclusion that meaningful individual relief is the proper goal of Rule 23, not private attorney-general enforcement.

The effect of an opt-in procedure on class size was attributed to several factors. Some class members may understand and make a deliberate choice to stay out. Many, however, would not understand the class notice. And even if the notice is understood, the average person fears litigation. Class members may fear exposure to costs, discovery, or even counterclaims if they come in.

Other questions about opt-in classes involve notice, and perhaps amount-in-controversy requirements.

The importance of an opt-in class was recognized, whether as an alternative to (b)(3) opt-out classes or as a limited or general substitute. It was suggested that it would be a mistake to go forward with the small-claims proposal embodied in published Rule 23(b)(3)(F), and then to follow it promptly with an opt-in proposal, whether or not the opt-in proposal was limited to small-claims classes. It well might be that the Committee will not want to adopt a general opt-in provision. There are real risks that opt-in classes will increase, not reduce, the burdens faced by the judicial system. The motive underlying proposed factor (F) was that Rule 23 sweeps in people who really do not want to be in a class, particularly in "consumer" class actions with very small individual stakes. Opt-in classes may make particular sense as an alternative for the "(F)" cases.

This discussion of opt-in classes led back to the package that might be presented to the Standing Committee with a recommendation for action now. Settlement classes have been deferred. It is likely that any eventual proposal will require publication for another round of public comment. An opt-in proposal surely will require publication and comment. Perhaps it makes sense to hold back as well on factors (A), (B), and (C), so that a second package can be presented as a whole. On the other hand, the (A), (B), and (C) factors were shaped in large part by the desire to respond to mass tort class actions. They might go forward now, if the Committee concludes that they should be adopted, either as published or with minor changes that do not require a second publication.

Discussion of the general approach to timing the (A), (B), and (C) proposals noted that the public comment provided much information. There was some controversy over these factors. If they were to go ahead with the change in subdivision (c)(1) and the interlocutory appeal provision of subdivision (f), the whole package might be affected by concerns directed primarily to (A), (B), and (C). A motion was made to separate (A), (B), and (C) for further study, but to defer sending them forward until the Committee has completed work on other Rule 23 issues. The motion carried by 8 votes for, 4 votes against.

Rule 23(b)(3)(A)

Proposed Rule 23(b)(3)(A) would direct the court to consider the practical ability of individual class members to pursue their claims without class certification. It was proposed primarily because of concerns that some courts may have been too eager to certify mass torts classes despite the prospect that individual class members would in fact do better by pursuing individual actions. Public comments and testimony, however, suggested several reasons for caution.

One concern, advanced in common for several of the proposals, is that changes that are intended to be modest may have unintended consequences. The most surely predictable consequence is that lawyers will seek advantage in any change, contending for unintended meanings. Years of litigation will be required to force the change into the desired shape, and even then unintended consequences may emerge.

It also was urged that proper administration of present factor (A) should accomplish everything intended by the proposed new (A). Present (A) directs attention to the interest of class members in individually controlling separate actions. A bare "interest" in separate actions means nothing if separate actions are not practically available. Practical ability, in short, already is in the rule.

Conceptual difficulties arise, moreover, in addressing the many classes whose members' claims span a wide spectrum of amounts. A single antitrust or securities class, for example, may involve members whose claims range from tens or hundreds of dollars to tens of millions. Refusing to certify a class because some members can practicably pursue separate actions does a grave disservice to the many members who cannot.

The range of claims encountered in many classes has other consequences. The 1995 securities litigation reform legislation stresses the importance of involving in class litigation the class members who have the largest claims. Unthinking application of proposed (A) could be inconsistent both with the letter and spirit of this legislation. The insights of the securities legislation, moreover, are important in all types of classes. It is important to involve the large-stakes claimants. They may become representatives, and in any event will monitor the representatives. Inclusion of the large-stakes claimants makes it easier to achieve settlement, in part because the defendant achieves a greater measure of repose.

Large-stakes claimants also are most likely to understand the nature of a class action, and to make intelligent determinations whether to opt out of the class. There are strong reasons to remain in a class even when individual litigation is practicable. There is greater efficiency in class litigation, and the class stakes may support more thorough litigation than could be supported by even the largest individual claims. Participation in class litigation also may reduce the risks of reprisal.

Individual litigation, moreover, may lack the capacity to achieve remedies that are available in a class action. Settlements of class actions can provide remedies beyond the reach of traditional judgments, and beyond the scope available in piecemeal settlement of individual actions.

Beyond these problems lie the administrative uncertainties opened up by the proposal. There is no definition of "practical ability." Does it refer to the size of individual claims? The cost of litigation, as it may be affected by the merits? The resources available to an individual to meet the costs of litigation? The sophistication and "savvy" of each individual class member? The actual level of desire to pursue separate actions? How far will discovery be available to test these factors as to each person included in a proposed class definition?

All of these uncertainties leave it probable that different judges will respond differently. Different perceptions about the desirability of class actions in general, or about the wisdom of a particular body of substantive law, or about the apparent strength of a specific claim, will mean that class certification will depend on the identity of the judge assigned to the case.

These difficulties led to the suggestion that the possible benefits of proposed factor (A) are not worth the costs. Present (A) is very close to it. At most, the proposal would accomplish very little. The burden of justifying change lies on the proponent. The language of the proposal is unclear, and the purpose has changed over time. Initially, this proposal was linked to the later-abandoned theory that (b)(3) should be altered to require that a class action be "necessary" for the fair and efficient adjudication of the controversy. Now it has come to be linked to mass torts and an exhortation designed to encourage small claims classes. The administrative problems are manifest. Implementation could mean not only that some class members are excluded from the class definition, but also that some classes are not certified at all. A single-event mass tort such as an airplane crash, for example, might involve claims that all support individual actions.

A motion to abandon proposed factor (A) carried unanimously.

Rule 23(b)(3)(B)

Proposed factor (b)(3)(B) revised present factor (A). The focus continues to be on class members' interests in separate actions. The present reference to "individually controlling the prosecution or defense" of separate actions would be deleted, however, in favor of a more open-ended reference to separate actions. The change is intended to remind courts to consider alternatives beyond simple two-party litigation. A proposed class should be considered in relation to alternative class definitions, aggregation of individual actions by voluntary joinder, intervention, consolidation for joint pretrial proceedings by the Judicial Panel on Multidistrict Litigation, and the like.

The proposed change of emphasis was thought to be too minor to justify an amendment. There was no suggestion that the purpose was unwise. Consideration of all alternative modes of adjudication seems an inevitable part of any superiority determination. But a subtle amendment of the present rule is likely to carry greater potential for mischief than for benefit. A motion to abandon the proposed revision of present factor (A) carried unanimously.

Rule 23(b)(3)(C)

Proposed factor (b)(3)(C) would amend present factor (B) by adding a reference to the "maturity" of related litigation. The impetus came primarily from mass tort cases that have presented great uncertainty as to the causal connection between exposure to a claimed harmful thing and later injuries. The fear has been that premature adjudication can enforce or defeat claims on grounds that are disproved by later scientific knowledge. Premature settlements present parallel dangers. The hope is that deferring class litigation while individual actions develop the information needed for well-informed adjudication or settlement will improve the process. The focus has been on situations in which the court can be confident that there will be substantial numbers of individual actions, has strong reason to fear the inadequacy of the evidence that can be adduced, and has good reason to hope that significantly better evidence will be developed in the reasonably near future.

Much of the opposition to the proposal arose from the possible consequences for regulatory enforcement actions. Securities law actions were often held up as examples. Even when the facts are intricate and the law unclear, a single class action often is the best means of adjudication. The resources needed to explore the issues in depth can be mustered for a class action, but not for any individual action. There is little reason to expect that delay will improve the basis for decision. There are few individual issues to becloud the picture; damages commonly can be calculated by a common formula.

Concerns also were expressed that there is no definition of maturity, and that the lack of definition could augment the problem of delay while individual actions are processed. Neither is there any definition of "related" litigation, of the closeness of the nexus that justifies deferring class certification.

As with the earlier proposed factors, it was observed that courts already are focusing on problems of maturity. And they may focus on maturity for different purposes than the simple advance of knowledge. In In re Norplant Products Liability Litigation, E.D.Tex.1996, for example, the court deferred certification so that trial of groups of individual claims would show whether individual issues predominate to an extent that defeats class certification.

Elizabeth Cabraser addressed the Committee, suggesting that more time is required to develop the proper role for a maturity factor. Recent appellate decisions have provided substantial guidance, whether or not the guidance is attractive to all observers. Lawyers and courts can extrapolate from the facts of recent cases. It may prove in the end that opt-in classes are desirable for mass torts, and that in this setting there is no need for a maturity factor comparable to the need that arises if mass torts are opt-out or mandatory classes. The plaintiffs will have individual lawyers, they can make well-informed decisions whether to opt in, and there will be substantial numbers of individual actions. So long as individuals are making a choice, there is no need to make them wait. And if we are asking courts to experiment, as they now are, it is better to let the jurisprudence develop. Further consideration of the maturity factor can be deferred, both in its own terms and until there is more consideration of opt-in classes.

Sheila Birnbaum also addressed the Committee. She agreed that there is a growing maturity jurisprudence, but suggested that to reject the published proposal will be seen as disapproval of the development. The Committee should not send this signal. In mass torts, opt-out works. But postponing the maturity factor for further consideration would not present this problem.

Possible ambiguities of the reference to the maturity of related litigation were noted. It was suggested that perhaps the reference should be to the maturity of the factual or legal theories advanced in the class action. This approach would avoid any implications that there must be related litigation, or that any related litigation have reached some point of maturity.

For working purposes, it was moved that this proposal be amended and carry forward on the Rule 23 docket as an amendment of present (b)(3) factor (B), approximately in these words: "the extent and nature of any related litigation, and the maturity of the issues in the controversy." The motion carried without dissent.

Rule 23(b)(3)(F)

Proposed factor 23(b)(3)(F) would make pertinent to the determination of predominance and superiority "whether the probable relief to individual class members justifies the costs and burdens of class litigation." This proposal was the subject of many comments and much testimony. Vigorous support and vigorous opposition were offered.

Discussion focused both on the proposal as published and on the possibility of tying the concern with small claims classes to an opt-in alternative. Rather than speculate about arguments whether relief on small-stakes claims is meaningful to individual class members, an opt-in class would provide direct evidence that individual class members believe the hope for relief is sufficiently important to justify involvement in the litigation.

The Committee was reminded that pre-publication discussion in the Standing Committee focused on the ambiguity of the focus on individual relief. If it is intended to focus only on individual relief, so that a $10 recovery must be balanced against the costs and burdens of class litigation, the factor would kill consumer class actions. If it is intended to allow consideration also of the aggregate class relief, so that 1,000,000 individual $10 awards -- an aggregate of $10,000,000 -- would justify costs and burdens estimated at $1,000,000, the rule should say so more clearly.

Factor (F) renews the question whether this Committee should make the value choices that surround consumer classes. Small-claims classes do play a regulatory function now. Congress is aware of this function, and has regulated it, with examples including recent Truth-in-Lending Act amendments dealing with mortgage lending. There are no clear empirical data to show that small-claims classes are somehow "out of control." All we have are anecdotes. There is no clear showing that would help separate actions that serve only enforcement purposes from those that also provide meaningful individual relief.

The Committee's focus has not been on ending the regulatory use of small-claims classes. Rather, it has been attempting to find a way to regulate abusive uses of small-claims classes. The public comments and testimony show that (b)(3) classes have come to serve many purposes; many choices have been made by the courts in the process of developing (b)(3). It may be possible to curb abuses without making the cosmic choices about public law regulation through Rule 23. And it may not be possible to curb abuses even if the big choices are made.

The impact of small-claims class actions on public perceptions of the judicial process has been a recurring theme. Many observers have stated that the public is repelled by a process that turns courts into agencies for administering small recoveries. The sense of revulsion is aggravated by the frequent award of large attorney fees in these cases. These actions breed more cynicism about courts than anything short of the most publicized and exploited criminal prosecutions. John P. Frank suggested to the Committee that the big abuse is the "lawyers relief act" aspect of small-claims classes. Proposed factor (F) is the most important of all the published proposals. Administration of present (b)(3) has unleashed a scandal.

These observations were met with the response that courts should police fees more directly and effectively. Large aggregate recoveries justify substantial fees. "Coupon" recoveries or other meaningless class relief do not. Judicial regulation of fees, however, is difficult. Even if a court pleads for participation and opposition in the fee-review process, there is little interest. Judges cannot regulate fees without help from objectors.

Support for further consideration of proposed factor (F) was expressed. The Committee gave birth to a rule that has profound impact. The Committee, however, is not the body responsible for determining how far the rule's unforeseen consequences are in the public interest. There should be some constraint on certifications. It is difficult to express them with precision. The published proposal is as good as can be managed.

The focus of (F) on "probable" relief was questioned. This word seems to invite consideration of the likely outcome on the merits, a consideration that the Committee once determined to add to (b)(3) and then deliberately abandoned. At the same time, the Committee refused to determine whether (F) permits consideration of the probable outcome on the merits. The time has come to decide, and the decision should be against confusing the certification process with an anticipatory trial on the merits. The rule should refer to "requested" relief rather than "probable" relief.

The Committee was reminded that "requested" relief had earlier been considered and rejected. The difficulty is that few if any complaints demand "coupon" or similarly small relief.

One compromise would be to delete "probable," so that the rule would refer only to "the relief to individual class members." This would reduce any implied reference to success on the merits, without binding the court to any extravagant demands set out in the pleadings.

It was observed that the language of proposed (F) is vague. Actual administration will give it many meanings. Each judge will be right, and each will be wrong. It may be better to address these problems by adopting an opt-in class as a means of making sure that the class-member plaintiffs really "are there."

Further doubts were expressed about the prospect that a rule can be drafted that will accomplish whatever purpose the Committee may finally reach. The hope is to separate out "bad" class actions that impose great burdens on courts and defendants to no real purpose. The amount of individual recoveries is at best a crude proxy for this purpose. The proxy, moreover, can work only in cases that involve uniformly small stakes; when some class members have larger claims, administration could become very difficult. Opt-in classes may help in this setting, if those with significant claims actually do opt in. On the other hand, if those with significant claims fail to opt in, the result may be defeat of important individual interests as well as significant public enforcement interests.

Turning to the means of separating opt-in classes from opt-out classes, it was suggested that the test should be whether individual claims are so significant that it is fair to infer consent to class litigation by class members who fail to opt out. Failure to opt out by those who hold significant claims often justifies the inference of knowing consent to be represented by the class action. Failure to opt out by those who hold small claims does not justify the same inference.

It was urged on the other hand that the small-claims class members are those who most need the protection of a class action. They have no realistic alternative means to assert their claims, but many are likely to fail to opt in for reasons that do not reflect disinterest.

Discussion of the disposition to be made of proposed factor (F) led back to the drafting questions. The focus on "individual" relief was challenged again, with the suggestion that this word be deleted, or that "aggregate" be added -- "whether the aggregate relief to [individual] class members" justifies class litigation. As an alternative, it was suggested that "individual" simply be removed: "whether the [probable] relief to class members" justifies class litigation. Deletion of "individual" relief would not make (F) meaningless; "coupon" cases and small total recovery cases would be denied certification. This alteration would completely change the intended character of (F). The focus was on individual benefits on the ground that class actions are justified only by individual benefits, not by public enforcement purposes. The Committee did decide that the private attorney-general function is not the stuff of Rule 23 procedure.

The reference to "probable" relief was defended on the ground that the class certification decision is made early in the litigation. An absolute prediction of relief cannot be made; only probable relief can be considered. The focus should be on the relief that will be given if the class wins on the merits. One suggestion was "the likely relief should the class succeed."

These problems led to the suggestion that (F) should be dropped entirely. Administration of any version that might be drafted will not be worth the effort. And any proposal will be challenged on the ground that it is advanced for substantive purposes. We know what the present factors identified in 23(b)(3) mean, even if the meaning is indefinite. The rule should not be complicated further.

The question was raised whether certification can be denied under present Rule 23(b)(3) on the ground that a class action is not superior to other means of adjudication when only de minimis individual benefits will follow. The Committee has regularly been advised that certifications are now denied on this ground. At the same time, if this ground of denial is proper, it should be expressed in the rule. Denial can then be stated openly, and reviewed in its own terms.

Discovery as to probable relief seems inevitable if factor (F) should be adopted. This will add administrative burdens and delay.

Factor (F) continued to stir discussion of the character of Rule 23. Is it simply a device for aggregating claims? Or does it also serve public enforcement purposes? How far can we adopt a "common sense" provision that authorizes rejection of proffered classes that "just ain't worth it"? If the problem is that some present class certifications should not be made, can a good cure be found that justifies the controversy that inevitably will surround any attempted cure? Adoption of a proposal aimed at restoring "common sense" to some portion of present practice inevitably will lead to a vast body of law seeking to define common sense. It is difficult to add precision to concepts that have been developed into well-defined contours of vagueness. Attempting to attain precision will be costly.

The question whether some version of factor (F) should continue on the agenda for further discussion at the October, 1997 meeting, despite misgivings about the published proposal and the variations that have been suggested, was resolved by consensus. The proposal should carry forward as part of the Rule 23 agenda, with five alternative forms: (1) add a reference to aggregate class relief, while dropping the reference to recovery by individual class members; (2) create an opt-in class that is available only as an alternative to opt-out classes in small-claims cases; (3) refer not to "probable" relief but to "likely relief if the class wins"; (4) the proposal as published; and (5) do nothing. It was agreed that information should be provided about experience with opt-in classes under the Fair Labor Standards Act procedures, including experience with Age Discrimination in Employment Act cases.

Other Rule 23 Proposals

Repetitive class attempts. One of the problems that has been noted is the succession of efforts to win class certification, or to foster parallel and overlapping classes. In addressing certification of a (b)(3) class, the court should be directed to consider "whether a class has previously been certified or denied for the same claims by any court." Even if this practice is adopted by most courts now, it should be in the body of the rule. It does not preclude a court from certifying a class that has been denied by another court, nor from certifying a class that has been certified by another court. It simply urges caution in considering the predominance and superiority questions. Although there may be similar problems with (b)(1) and (b)(2) classes, they have not been urged on the Committee and it is better not to tinker with those classes. The likely place for the amendment would be as an addition to present factor (B). A motion to consider this proposal as part of the Rule 23 agenda was adopted, 8 votes for and 1 vote against.

Common evidence. Several of those who commented and testified on the published proposals urged consideration of a "common evidence" provision to bolster the "predominance" requirement in Rule 23(b)(3). Classes have been certified in which individual elements predominate. One version would require that the trial evidence be substantially the same as to all elements of the claims asserted by class members. A less ambitious version would add a new factor asking "whether plaintiffs have demonstrated their ability to prove the fact of injury as to each class member, without making individualized inquiries as to class member injury." This version would focus on common evidence of the fact of injury, not the amount of injury. The point of separation is between liability -- which includes the fact of injury -- and proof of the individual quantum of damages.

This proposal would defeat the opportunity to include in a class members who have not yet experienced injury. Mere exposure would not establish the "fact of injury," unless state law allows compensation for the fear or risk of future injury.

This proposal is meant to restrict the use of issues classes. Whether common issues predominate depends on the purposes for which the class is defined. The focus on common proof as to the fact of injury would defeat certification of an issues class that is confined to a subset of the liability issues. All liability issues should be tried in the class, or there should be no class. "Issues" classes are a snare and a delusion. Causation questions, for example, are not as readily divided into "general" and "specific" issues as some would wish.

The proposal is not meant to defeat the use of subclasses when the predominance of common issues can be achieved in that way. Variations in state law, for example, may be met by grouping different states into a limited number of subclasses that account for the variations without forcing state-by-state subclasses.

Concern was expressed that this approach may require undue inquiry into the merits of the claims at the certification stage.

A motion to consider alternative forms of a "common evidence" requirement carried by 9 votes for, 2 votes against.

Notice: It was urged that the Committee take up for consideration the draft that would allow sample notice in actions that join in a (b)(3) class very large numbers of members who typically have small claims. A thoughtful letter from Professor David Shapiro has urged this proposal.

Due process concerns were expressed that sample notice may defeat any justification for an opt-out class.

It was noted that the earlier draft on notice addressed several questions. It sought to encourage notice language that is intelligible to class members. It provided for notice in (b)(1) and (b)(2) classes. It also provided for notice in opt-in classes.

A motion was made to add a full notice draft to the Rule 23 agenda. The motion was resisted on the ground that the Committee has been trying to narrow the scope of its further Rule 23 work. The motion failed. It was agreed, however, that should the Supreme Court find occasion to address notice questions, the Committee should study whatever the Court might say.

Burden of persuasion and legal indeterminacy: An observation was made that none of the many class-action proposals have suggested a satisfactory response to the central dilemma created by the public-enforcement function of (b)(3) classes. Private enforcement of the public interest is usually desirable when clearly established facts establish a clear violation of clear law. But the decision to file is not made by public officials charged with discretionary responsibility to determine whether unremitting enforcement of the law in a particular situation serves the public interest. Private decisions to file may be motivated by misguided views of the public interest, or by less worthy motives. Defendants may be at an unfair disadvantage because a class action multiplies the risks of mistaken factfinding or law application. Defendants are at risk in another way as well. Even with the best of intentions and the most sophisticated legal advice, it is easy to violate the rules in many areas of modern regulatory law that abound in shades of gray uncertainty. The great stakes at risk, the high costs of litigating, and the inescapable residual fallibilities of adjudication create irresistible pressures to settle even ill-founded claims. These dangers could be addressed directly. Plaintiffs in a (b)(3) class could be required to prove the facts by clear and convincing evidence, and to show that the clearly proved facts establish violation of clear legal rules. These elevated showings would be required only in (b)(3) class actions, as the price for use of a procedural device that combines the potential for great benefit with the risk of great injustice. Ordinary standards of persuasion and enforcement against good-faith violations would continue to apply in other settings. This observation was not made as a suggestion for present action or even further consideration. It was offered instead as an illustration of the forms of direct response that might be made to the persisting dilemmas posed by Rule 23(b)(3).

Rule 81(a)(2)

Rule 81(a)(2) states the time to return a writ of habeas corpus. The statement is directly inconsistent with the later and superseding provisions of Rule 4 of the rules that govern habeas corpus proceedings governed by 28 U.S.C. § 2254. It also is probably inconsistent with the rules that govern other habeas corpus proceedings, at least to the extent that § 2254 Rule 1(b) authorizes resort to Rule 4 in proceedings not governed by § 2254. Something must be done to bring Rule 81(a)(2) into conformity with the later § 2254 rules.

It is not clear what is the best course. It may be that habeas corpus proceedings governed by the general provisions of 28 U.S.C. § 2241 should be controlled by the § 2254 rules for all purposes. It may be that distinctions should be drawn, as they now are. And it may be that a distinction in the time to answer is appropriate. The possible need for distinctions is most likely in petitions brought by persons in federal custody but outside the provisions of 28 U.S.C. § 2255.

All of these matters must be worked out with the Criminal Rules Advisory Committee, which has had primary responsibility for the § 2254 Rules. The Committee directed its Reporter to work with the Reporters for the Standing Committee and the Criminal Rules Committee in developing a solution to these questions.

Admiralty Rules B, C, and E

Mark Kasanin reported as chair of the Admiralty Subcommittee. Substantial progress has been made in working on proposed amendments to Admiralty Rules B, C, and E. Delay has been required to allow consideration of a revised forfeiture Rule C(6)(a) proposed by the Department of Justice. It is hoped that a final draft of the proposed revisions will be ready by early summer, to be reviewed by the full subcommittee and refined in time to be presented to this Committee in the fall.

Future Meetings

The September meeting will be devoted to the discovery conference. The October meeting will review the September discovery conference and select proposals to be developed by the Discovery Subcommittee for further consideration at the following spring meeting. In addition, the October meeting will resume consideration of Rule 23 proposals and such other matters as prove ready to be considered.

Respectfully submitted,

Edward H. Cooper