October 6 and 7, 1997

The Civil Rules Advisory Committee met on October 6 and 7, 1997, at the Stein Eriksen Lodge, Park City, Utah. The meeting was attended by all members of the Committee: Judge Paul V. Niemeyer, Chair; Sheila Birnbaum, Esq.; 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; Chief Judge C. Roger Vinson; and Phillip A. Wittmann, Esq. Edward H. Cooper was present as Reporter, and Richard L. Marcus was present as Special Reporter for the Discovery Subcommittee. Sol Schreiber, Esq., attended as liaison member from the Committee on Rules of Practice and Procedure, and Professor Daniel R. Coquillette attended as Reporter of that Committee. Judge Eduardo C. Robreno attended as liaison member from the Bankruptcy Rules Committee. Leonidas Ralph Mecham, Director of the Administrative Office of the United States Courts attended, as did Administrative Office representatives Peter G. McCabe, John K. Rabiej, Mark J. Shapiro, and Mark Miskovsky. Thomas E. Willging represented the Federal Judicial Center. Observers included Alan Mansfield, Mark Gross, Fred S. Souk, Robert Campbell (American College of Trial Lawyers), Reece Bader (ABA Litigation Section), Beverly Moore, Alfred Cortese, Rod Eschelman, and Nick Pace.

Chairman's Introduction

Judge Niemeyer opened the meeting by welcoming Leonidas Ralph Mecham. He observed that the policy of rotating committee membership serves the good purpose of bringing new perspectives the committee work, but also carries a significant price. The committee has worked on Rule 23 for six years, accumulating much knowledge, and now the time has begun when experienced committee members will leave while Rule 23 remains on the agenda of active items. Carol Posegate is finishing her second three-year term. The committee expressed thanks to Ms. Posegate, who responded that work with the committee has been one of the highlights of her professional career. Sheila Birnbaum was welcomed as a new committee member, with the observation that her regular attendance at committee meetings over a period of several years will serve her and the committee well as she becomes an official member.

Mark Kasanin was appointed to the discovery subcommittee to fill Carol Posegate's place, since the work of the subcommittee is not finished.

The Standing Committee is paying close attention to this committee's work, as to the work of each advisory committee; its confidence in the committee must continually be earned to be deserved. Congress also is paying close attention to this committee's work; its respect and deference also must be continually earned by careful and responsible behavior.

A proposed amendment to Civil Rule 23(c)(1) and a proposed new Rule 23(f) were taken to the Standing Committee in June with a recommendation that they be advanced to the Judicial Conference to be adopted. Members of the Standing Committee raised concerns about the proposal that Rule 23(c)(1) be amended to require certification "when practicable," replacing the present "as soon as practicable." After some discussion, it was decided that this proposal should remain part of the full package of Rule 23 proposals still being considered by this committee. The proposed permissive interlocutory appeal procedure was approved and transmitted to the Judicial Conference. The proposal has been approved by the Judicial Conference as a consent calendar item, and will be sent on to the Supreme Court.

Judge Niemeyer met with the Judicial Conference Executive Committee before the Judicial Conference session, along with other committee chairs. This committee's agenda was described, with the observation that the committee understands the risks of undertaking controversial topics.

After the Judicial Conference meeting, Judge Niemeyer met with other committee chairs. He urged on them the importance of the national rules, not simply as a convenience for practitioners but as an intrinsically national body of federal law that should remain uniform throughout the country. The Boston discovery conference provided support for national uniformity. The disclosure rule amendments of 1993 effected a breach in the wall of uniformity. Although the permission for local rules departing from the national standard was prudent at the time, the result has been great diversity of practice. It is incumbent on the rulemakers to provide a national rule. Some reservation might be expressed on the ground that not enough time has yet been allowed for experimentation that may show the way to better disclosure practices. But disclosure has been studied by the RAND report on the CJRA, and by the Federal Judicial Center. Local CJRA plan studies also are being made, including detailed studies in the Eastern District of Pennsylvania. District judges should be enlisted in the quest for uniformity.

The report to the Standing Committee described the discovery project. The difficulty of persuading district courts to surrender adherence to local rules was observed. One of the committee chores -- as exemplified by the discovery project -- will be to get district courts to understand the need to adhere to uniform national procedure.

Judge Niemeyer met with the Long Range Planning Liaison Group. They were interested in creating an ad hoc committee on mass torts. This topic has been much in the public eye. Judge Hodges, chair of the Executive Committee, suggested an ad hoc committee. The advantages of consideration by this committee were considered, recognizing that it will be important to coordinate efforts with other committees. Other committees that may be interested include the Federal-State Jurisdiction Committee, the Judicial Panel on Multidistrict Litigation, the Bankruptcy Administration Committee, and perhaps the Court Administration and Case Management Committee. This committee has devoted many years to studying class actions, and in the process has heard much about mass tort actions. The difficulties of responsible change have become apparent, as has the futility of trivial change.

Judge Niemeyer further observed that this committee can no longer think of itself as having a constituency of lawyers, judges, and academics. There is more public scrutiny of court procedure and of the committee's work. The committee and its members must become leaders of a dialogue beyond the confines of the Enabling Act process. Congress is increasingly interested and active, at least as measured by the introduction of bills that would affect procedure. Many members of Congress remain sympathetic to the role of the Enabling Act process, but there also are signs of impatience, arising in part from the deliberately deliberate pace of the process. An illustration is provided by the proposal to amend Rule 23 to provide for permissive interlocutory appeals -- although the proposal is now on the way to the Supreme Court, a bill to establish the same appeal procedure remains pending in Congress.

Legislative Report

John Rabiej provided a report on pending legislation. There are 15 or 16 pending bills that directly affect the civil rules. It does not seem likely that action will be taken on any of them this year.

Hearings will be held on HR 903, which includes offer-of-judgment provisions, but the hearings will focus on the arbitration issues in the bill. Last spring a letter was sent to Congress indicating that the rules committees take no position on the merits of the offer-of-judgment provisions, but also noting that after substantial study of Rule 68 this committee concluded that this is a very complicated subject. Some technical problems with the bill also were pointed out. Judge Hornby will testify on the arbitration parts of HR 903 for the Court Administration and Case Management Committee.

Bills dealing with Rule 11 seem to lack momentum.

A question was asked about progress on HR 1512, the current embodiment of longstanding attempts to adopt a minimum-diversity jurisdiction basis for consolidating single-event mass tort litigation in federal courts. It was noted that this topic requires coordination with the Federal-State Jurisdiction Committee, but that it fits squarely within the mass torts topic that will continue to attract this committee's attention.

The committee noted with appreciation the good help that John Rabiej and the Administrative Office continue to provide in tracking relevant legislation.

Minutes Approved

The Minutes for the May and September committee meetings were approved.

Agenda Items

The Copyright Rules remain an enigma on the agenda. Further consideration of the proposal to rescind these rules is set for the spring agenda. Congress has shown an interest in the topic, reflecting concern that nothing should be done that will make it more difficult to enforce copyrights against pirate and bootleg infringers. Parallel concerns have been identified by those working with the TRIPS portion of the Uruguay round of the GATT agreement. GATT countries are required to provide effective copyright remedies. There is a fear that simple rescission of the Copyright Rules might seem to other countries to belie the United States commitment to vigorous enforcement. These fears will need to be addressed when the topic comes up for consideration. It must be made clear that any action taken will be designed to remove the doubts that now surround the continuing force of Copyright Rules that were adopted under, and refer only to, the 1909 Copyright Act, and that are subject to serious constitutional challenge.

It was observed that the docket of agenda items should not state that the committee "rejected" the proposed amendment of Rule 47(a) that would create a party right to participate in voir dire examination of prospective jurors. Although the committee elected not to pursue the proposal in light of substantial controversy, it did urge the Federal Judicial Center to frame its sessions for new judges to stress the importance of party participation. This has been done. Judge Patrick Higginbotham, the former chair of this committee, has spoken on the topic at several meetings.

Discovery Subcommittee

Introduction. Judge Niemeyer introduced the report of the Discovery Subcommittee by observing that the discovery project aims at three central questions. We hope to find out how expensive discovery is, both in general and in the most expensive cases; to decide whether the cost exceeds the benefits often enough to warrant attempts at remedial action; and if remedies should be sought, whether changes can be made that do not interfere with the full development of information for trial. The undertaking is more likely to focus on the framework of discovery than on attempts to control "abuses."

The Boston conference in September was as good as a conference can be. It was part of a process of generating a "smorgasbord" of ideas. The subcommittee has generated a comprehensive memorandum gathering the wide array of ideas that have been suggested. For this meeting, the objective is to explore the ideas to determine which of them deserve development through specific proposals to be considered at the spring meeting.

Judge Levi and Richard Marcus presented the work of the subcommittee. Judge Levi noted that the smaller January conference in San Francisco and the larger September conference in Boston had been the main work of the subcommittee to date. The purpose of these conferences has been in part to afford the bar an opportunity to take the lead on discovery reform, to advise the committee on what needs to be done and perhaps to suggest more detailed means of doing it.

The first big question is whether to do anything at all about discovery. Discovery seems to be working rather well in general, but there are problem spots. Lawyers are open to change, but doubt whether much can be accomplished. There may be a division between trial lawyers, who believe that real savings can be had in discovery, and litigators, who spend most of their time in preparing for trial and are inclined to doubt whether significant savings are possible. Many lawyers believe that the committee should not "tinker"; changes should be significant. At the same time, it is recognized that desirable technical changes should not be thwarted by fixing them with the "tinkering" label.

The Special Reporter was asked to list all of the many separate suggestions that have been made for discovery changes. The purpose of this list is to preserve the suggestions, not to imply that all of them should be adopted. As a guide to discussion, five central areas have been chosen as most deserving of attention.

The first central problem is uniformity. There is some chagrin among alumni of the 1991-1992 committee deliberations that the 1993 amendments deliberately invited disuniformity. Uniformity was thought desirable by many participants in the Boston conference. But it is not clear how broad or deep is the desire for uniformity. Many at the ABA Litigation Section meeting in Aspen this summer suggested that good local rules can be better than a blandly uniform national rule. The sense of that meeting was that it would be important to know what the national rule would be before deciding whether uniformity is a good thing.

If uniformity is to be pursued, the committee must address disclosure. The original wave of fear seems to be subsiding. It is agreed that all of the information that Rule 26(a) requires to be disclosed could properly be sought by interrogatory. But some lawyers like to have an interrogatory to show to the client to justify the need to reveal the information, and to demonstrate that the lawyer is not penalizing the client for the lawyer's better understanding of the case. Yet if Rule 26(a) has not been the disaster that some anticipated, no one thinks it has been a major improvement. The studies may show some cost saving -- it is too tentative to be sure -- but it is clear that nothing terribly significant has happened. And Rule 26(a) will not be much help in the problem discovery cases that are the focus of concern. The complex and contentious cases are likely to be exempted from disclosure in any event.

There may be support to limit disclosure to "your case" information. But it is difficult to know how meaningful it is to ask that each party reveal at the beginning of the litigation, before discovery, what information it plans to introduce at trial.

Another approach to disclosure is to view it as the first step in a staged sequence of managed discovery.

Managed discovery is a third area for study. The central idea is that discovery might proceed in three stages. First would be disclosure, however disclosure may be reshaped. Second would be some level of core discovery, defined to be available to the lawyers without court management. This stage might well include stricter limits on the numbers of interrogatories and depositions than those set by current rules. It also might include time limits on depositions, and even might include some attempt to limit the quantity of document exchange. The third stage would require court management when any party wishes to engage in discovery beyond the core limits. In many ways this would involve a party-selected means of tracking; court management would be provided at the request of any party coming up against the limits of core discovery. This managed discovery system could be viewed together with Judge Keeton's proposal, including changes in Rule 16, using the whole pleading-discovery-pretrial conference process to get a better definition of the issues.

The managed discovery approach is consistent with the frequent observations that discovery works well in most cases. It would mean that for most cases, the parties would be left alone to manage the litigation without need for judicial involvement.

Core discovery rules could be drafted to include a clear and firm cutoff on the time for discovery.

Pattern discovery also should be considered. It seems to have support from both plaintiffs and defendants. The project would be to develop pattern discovery requests for each of several distinctive subject-matter areas. The pattern requests would be agreed upon by working committees that include experienced lawyers from all sides of litigation in the particular subject area.

A fourth area of inquiry is the basic scope of discovery. The American College of Trial Lawyers has long supported the 1977 proposal to narrow the scope of discovery defined by Rule 26(b)(1). There is a related view that the major problem of discovery arises with document production, and that the scope of discovery should be narrowed only for document discovery.

The fifth major area of inquiry is document production. This seems to be the area of greatest concern. No specific proposal is ripe for discussion.

Document production involves particular questions about privilege. There seems to be a consensus that there is a problem with the effort required to protect against inadvertent waiver. There also may be difficulties arising in courts that disregard the terms of Rule 26(b)(5) and insist on privilege logs that both impose excessive burdens and threaten to reveal the very privileged information to be protected. It has been suggested that it works to provide for informal review of potentially privileged documents by the demanding party under a protective rule that this mode of disclosure does not waive privilege. The demanding party then specifies any of the examined documents that it wants to have produced, opening the way to formal assertion and litigation of the privilege claim. Apart from this privilege problem, there are continuing problems with the sheer volume of documents that may be relevant to a discovery demand. The problem of volume is exacerbated when the production demand is addressed to a multinational enterprise that has documents, often in many different languages, scattered around the globe. And the problem of volume may be further exacerbated by electronic storage and erasing techniques that may complicate determination of what "documents" a party actually "has." Information that has been erased often remains available upon sophisticated inquiry.

Beyond these five major areas, many other worthy suggestions were grouped into a "B" list of second-level priority. The most important idea on the list is the firm trial date, an item relegated to this list only because it is not a discovery matter, even though it is closely related to discovery cutoff issues.

There also is a "C" list of technical changes that need not be reviewed at this meeting.

Professor Marcus extended the introduction. The inquiry has followed an interactive process up to now. The subcommittee has been in a receptor mode. The time has come to switch to an action mode. Yet the subcommittee will remain open to receive further information. The Federal Judicial Center continues to analyze the data from the discovery survey it did at the subcommittee's request, and the several bar groups that participated in the Boston Conference have been invited to continue to provide further ideas.

The five items on the A list include three "bullet" items: uniformity; initial disclosure; and the scope of discovery.

"Tinkering" is in order if the committee decides to make one or more significant changes. Once the amendment process is launched, it is appropriate to act as well on any technical changes that have accumulated and that deserve attention.

There are two main themes that underlie these separate questions: Should the committee seek only to tinker, or should it seek global changes in discovery? And should the change process be launched now, or is it better to wait, recognizing that there have been many discovery rules changes over the last quarter-century?

There are other thematic questions as well. Uniformity creates tensions, not only with the desire for local autonomy but also with the more general managerial view that it is better to leave individual judges free to manage litigation as best they can. The experience with "high discovery" cases may suggest that the committee should turn back the clock on activities that the 1983 and 1993 changes require in all cases. And the consideration of "core" discovery proposals might move beyond limits on the number and extent of discovery requests that can be initiated without judicial involvement to describe what the requests can demand.

Judge Niemeyer stated that the subcommittee had done a splendid job. The committee should start with its recommendations. Although attention can properly focus initially on the major areas of inquiry identified by the subcommittee, the items on the B list should not be removed from the agenda. As the process continues, it may prove desirable to move some B-list items up for active discussion and adoption.

General discussion began with the observation that this list of topics for consideration is not a definitive proposal. There has not been time, nor committee discussion, to support a narrow focus. The purpose of the current report is to open the question whether the time has come to do anything with the discovery rules, and to begin to identify the areas that seem best to deserve more concrete proposals.

Uniformity: Disclosure. The need for uniformity was identified as a central issue. The view was expressed that there is no pressing need for uniformity. Lawyers have learned to live with their present situations. Frequent change of the rules is not desirable, not even when the object is to establish national uniformity.

It was asked whether uniformity is important even apart from whatever difficulties or frustrations may -- or may not -- face lawyers who move among different disclosure regimes. How important is it that there be a nationally uniform practice in all areas governed by national rules adopted under the Enabling Act? And there also is a need to serve the courts' interest in good policy, in having an effective procedure even if it makes lawyers unhappy. And the committee must recognize that it will be difficult to achieve much consensus among the bar on this topic, perhaps even as support for doing nothing.

It was urged that "we need to bring these horses back into the barn." The flirtation with local practice can intoxicate, and it will be increasingly difficult to restore uniformity. If uniformity is to be restored, the committee should move quickly.

Of course a decision to pursue uniformity in disclosure practice will entail determination of what the uniform practice should be. We cannot pursue uniformity in the abstract. If the only uniform rule that can be pursued successfully through the full Enabling Act process is one that uniformly abandons disclosure, or uniformly narrows disclosure, is uniformity worth the price? Before deciding whether uniformity is the most important goal, the committee must decide what disclosure rule would be best.

One sense of the importance of uniformity is that Congress was anxious in 1988 to move away from divergent local rules and practices. The Standing Committee local rules project has sought for many years to cabin diversity in practice arising from local rules. If the committee cannot successfully pursue uniformity, there is a prospect that Congress will. For that matter, Rule 26(a)(1) was proposed as a uniform rule. The local option was added from concern for the variety of practice that had emerged from Civil Justice Expense and Delay Reduction plans, some of it stimulated by the disclosure rule the committee had published for comment in 1991. In addition, there was substantial opposition to any disclosure rule; the opposition was so substantial that for a while the committee thought it should abandon disclosure.

An alternative to amending the national discovery and disclosure rules is to explore the opportunities for offering advice through the Manual for Complex Litigation. The Third Edition of the Manual contains many suggestions for regulating discovery practice similar to those offered to the committee. The subcommittee plans to study the Manual both as a source of ideas and as an alternative to further revision of the discovery rules.

A related opportunity is to expand the use of magistrate judges. The RAND study found that hands-on discovery management is important, and that litigant satisfaction increases when a magistrate judge is available to resolve discovery disputes. There are many very good magistrate judges, and there are many competing demands for their time. In some districts, magistrate judges are "on the wheel" for trial assignments. They do not view themselves, and their courts do not use them, primarily as discovery managers. Discovery management in a complex case, moreover, often goes to the heart of the dispute. The most important contribution a district judge can make may be to assume responsibility for managing discovery in litigation that will come to her for trial.

It was concluded that the subcommittee should bring back to the committee proposals to abandon all disclosure, to require uniform national adherence to the present rule, and to adopt the best identifiable modification of the present disclosure rules that might be adopted as a uniform national practice. It is hoped that information about the effects of present practice will continue to accumulate while the subcommittee and committee continue to study the issue.

Core discovery. Turning to core discovery, the first question raised was whether there is any need to tighten further the limits on the number of discovery events. The reality of discovery practice is not what might seem from talking with lawyers who pursue high-stakes and complex litigation in the major metropolitan centers. The reality is the small and medium case. In these cases, every study and much experience suggests that discovery is working well. And it seems likely that there is nothing the formal rules can do about the cases that now present problems. The rules provide ample power to control discovery; what is needed is actual use of the power.

The response was that there is no intention to affect discovery as it is practiced in most cases. All of the proposed limits on lawyer-managed discovery would permit discovery without judicial involvement at levels that include the vast majority of cases under actual present practice. Of course that leads to the question of identifying the cases in which the limits will be helpful, since it is highly probable that judicial management will be required in bigger cases under any likely variation of present rules.

The hope is to create a mechanism that develops a plan -- a track -- for the now-routine cases. These cases might proceed even more freely, more frequently, than under present practices. At the same time, limits that cannot be exceeded without judicial involvement create a system that makes it impossible for reluctant judges to avoid the obligation of involvement. All the studies show little or no discovery in most cases; this is true even of the Federal Judicial Center survey, which was designed to exclude categories of cases in which there is likely to be no discovery. The object is to identify a threshold that will require the court to become involved. And even that threshold can be made subject to party stipulations that allow discovery beyond the core limits when the parties are able to manage discovery without any need for further judicial involvement.

As an alternative, it might be possible to put aside the "core" discovery theory in favor of a system that allows any party to demand formulation of a discovery plan. This system would have the same advantage in requiring judicial involvement when the parties are unable to agree, without the need for elaborate changes in present discovery rules.

The opportunity for judicial involvement is amply provided by present Rule 16. No more may be needed than a mechanism that prompts actual use of Rule 16 powers. And Rule 26(f) conferences provide the framework for stimulating judicial involvement. Perhaps nothing more is needed. These observations were challenged by the suggestion that both the Rule 26(d) moratorium and the Rule 26(f) conference might be abolished for core discovery cases, and also by the observation that many lawyers are reluctant to approach a judge with a demand for judicial supervision.

The Rule 16(b) scheduling order requirement was discussed as part of this package. One judge observed that despite the language of Rules 16(b) and 26(f), he enters a scheduling order at the beginning of each lawsuit. Many cases involve out-of-town attorneys, making it costly and difficult to arrange conferences. Once a conditional scheduling order is entered, any problems are brought to the judge. But many cases do not require any action by the judge. Rule 26(f) accounts for much of the ability of lawyers to manage discovery without judicial involvement; it is the best part of the 1993 amendments. Others observed that such practices probably are common, and certainly have been followed by several committee members. In some courts, indeed, personnel from the clerk's office manage status calls. One approach would be to make these practices more explicit in the rules, going beyond the direct tie between Rules 16(b) and 26(f).

This discussion concluded with the suggestion that there is substantial support for the Rule 26(f) conference as it now stands, but that it may not be necessary to have the parties report to the court when they do not want judicial help.

It was suggested that if disclosure is retained, it could serve the role of core discovery. All discovery beyond that would require a plan, approved by the court unless the parties could agree.

Another suggestion was that the plaintiff could be required to file specified interrogatories with the complaint, with a like obligation on the defendant to file interrogatories with the answer. The questions would be limited to core discovery. Interrogatory answers would be stayed if there were a motion to dismiss. Many federal cases involve small claims. These routine interrogatories could save six months of discovery. The Rule 33 limits on numbers of interrogatories are a good thing.

A variation is provided by form interrogatories. California state practice includes three different sets of form interrogatories that ordinarily can be used in matching cases without fear that they will be held objectionable.

Judge Keeton has advanced a proposal to address the loose fit between notice pleading and discovery that also deserves attention.

The question of limitations on depositions, and particularly of duration limitations, came next. It was reported that in the Agent Orange litigation, there were 200 depositions conducted under a ruling that permission must be sought to extend any deposition beyond one day. To make this feasible, the deposing party was required to send the deponent all documents relevant to a deposition before the deposition was taken, so that the deponent could study the documents before hand. Under this system, 168 depositions were conducted in one day each. Most of the remaining depositions were conducted in two days; only a few required three days.

It was urged that some limit on deposition length is better than any further limit on numbers of depositions because it is difficult to plan the number of depositions at the beginning of an action. Even though number limits would be only presumptive, and any limits adopted under a case-specific plan also could be modified, the number of depositions may not be the best means of triggering judicial involvement. But it was urged in response that a more persuasive showing of need for discovery beyond the limits can be made after the limits have been reached and the need can be specifically identified.

A related question was whether a core discovery system would reduce the opportunities for judicial involvement now available so long as discovery remained within the core perimeters. In the same vein, it was asked whether there is any point in changing the present number of permitted interrogatories and depositions, if the goal of changing the numbers is to trigger judicial involvement, and there is little difficulty now with discovery in cases that fall within present limits. Present limits work. 85% of the cases go through the system without difficulty. The Rule 26(f) conference is a good thing; if you cannot afford the time for a simple meeting, you should not take your case to federal court.

Further in the same vein, it was suggested that the discussion of judicial management was moving the committee's focus away from the main point. There is no need for judicial management in the core case. It is the big case that needs it. There is not much need to worry whether there should be 25, or 20, or 15 interrogatories in a normal case. The problem is focusing discovery on the issues that may be dispositive in the big case. But it was suggested in return that there should be some form of judicial involvement -- even if only through the clerk's office -- in every case. A great majority of cases can be handled by some other court officer without a judge, although it is better to have a judge when that is possible. We should do nothing that might discourage judicial involvement.

This discussion led on to the observation that judicial management can be simple. It can be done on paper, by telephone, or by a courtroom deputy. The need is to ensure uniformly high quality and timely judicial management in cases that involve a potential for over-discovery. The key issue is what should command court time.

Given present limits on the numbers of depositions and interrogatories, and given Rule 26(f) conferences and Rule 16(b) scheduling orders, it was suggested that the remaining targets of stated discovery limits may be the duration of depositions and the quantity of document discovery. Rather than focus on the length of each individual deposition, it may work better to allocate a total number of deposition hours to each side, to be allocated among as many depositions as will fit. To be sure, lawyers operating under such rules have reported difficulties in allocating the time consumed by each party. But information will be gathered on actual experience under such systems. The subcommittee will frame proposals addressing both deposition length and quantity limits on document production.

It also was suggested that the subcommittee could look at Lord Wolfe's report in England. It includes provisions requiring a party to pay some of the costs of discovery beyond stated limits, a limited form of costshifting.

Discovery cutoff. The RAND report reflected substantial confidence that a combination of early judicial management with earlier discovery cut-offs and firm trial dates can reduce expense and delay without adverse impact. This topic clearly demands attention.

As attractive as early-set and relatively short discovery cutoffs may seem, there are substantial difficulties in attempting to set a uniform period in a national rule.

One difficulty is that cutoffs work only if discovery works. If one party deliberately delays, the discovery period may expire without allowing opportunity for necessary discovery. Many lawyers will say off the record that the famed "rocket docket" in the Eastern District of Virginia is administered in ways that defeat proper discovery in a significant number of cases; obstreperous lawyers are allowed to take advantage of the system by deliberate delay.

Another difficulty is that early discovery cutoffs make sense only if they are combined with reasonably proximate and firm trial dates. Completion of discovery should leave the lawyers ready for summary judgment motions, and then for trial. If these events cannot both be scheduled promptly, there is much waste and little advantage in the early cutoff. To the contrary, the early cutoff may force the parties into discovery that otherwise would not be undertaken at all. Individual case scheduling orders now can effect workable discovery cutoffs in relation to realistic trial dates. But a fictitious trial date, set in a uniform national rule, cannot do this. The circumstances confronting different districts vary widely. Any trial date set to conform to a uniform national requirement would be unrealistic in many districts.

In defense of possible uniform national time limits for discovery and trial dates, it was urged that the limits would exert pressure on judges to become involved in individual cases to set alternative and realistic dates. As with the proposed core discovery limits, the purpose would be to force judicial action, not to set limits that really can be met in most courts for most cases.

Thomas Willging noted that the RAND findings should be kept in perspective. RAND found that 95% of the variation in cost and delay is driven by factors independent of judicial management. There is only a limited amount of room for addressing the remaining 5% by improved judicial management. The Federal Judicial Center has continued to analyze the data in its discovery study. It has undertaken multivariate regression analyses of many procedures, including discovery cutoffs, meet-and-confer requirements, and other devices. No relationship could be found between any of these devices and cost or delay.

A motion was made to stop further consideration of discovery cutoffs, on the ground that Rules 16(b) and 26(f) provide ample and better means of addressing cutoffs. Differences in the docket burdens of different districts are alone enough to make a national rule unworkable.

Discussion of the motion noted that discovery cutoffs involve more than discovery alone. Unless there is an integrated plan, there is no point in hurry-up-and-wait. Increasing specificity in a national rule is not the answer.

In response, it was repeated that a national rule stating the need to "march along" with a case will serve as a default mechanism that forces recalcitrant judges to pay attention to the needs of cases that do require individual attention. A reply to this argument was that it is rare to find that attorneys are ready for trial, but not the judge.

The committee decided to defer action on the motion to terminate consideration of discovery cutoffs. It was recognized that many observers are keenly interested in discovery cut-offs, and that the subcommittee should explore further the possibility of creating a workable national rule. A close look should be taken, even if it proves impossible to do anything constructive. The subcommittee and the committee should explore all possibilities before giving up on this possible opportunity. But Judge Levi stated that the discovery subcommittee will not look at specific cutoff times.

Pattern Discovery. Pattern discovery might be pursued by developing protocols for acceptable discovery in particular subject-matter areas. Or general sets of interrogatories might be developed, consulting California practice, that are useful for many different types of litigation. Several bar groups and commentators have expressed support for some effort along these lines.

The California practice was described as involving sets of general interrogatories. A party can simply choose from among interrogatories in a set. It is generally accepted that these interrogatories are proper, and they are routinely used and answered. Further inquiries will be made into the nature of the California practice, the frequency of use, and the level of satisfaction with the results.

Grave doubts were expressed about the need for the committee to become bogged down in the enterprise of drafting form interrogatories. The system works well on its own. There is no lack of forms to be consulted by those who wish.

It was agreed that the subcommittee would further study the prospects of developing some system of discovery forms.

Rules 16(b), 26(d), 26(f). Discussion turned briefly to the interplay among Rules 16(b), 26(d), and 26(f). It was agreed that the subcommittee should consider the desirability of revising Rule 16(b) to clearly authorize entry of a conditional scheduling order before the Rule 26(f) conference. The Rule 26(d) discovery moratorium will be considered in conjunction with the review of disclosure. To the extent that Rule 26(f) ties to Rule 26(d), it will be implicated as well. But there was no sense of dissatisfaction with the general working of Rule 26(f); earlier discussion suggested that it may be among the most successful features of the 1993 amendments.

Scope of discovery. The American College of Trial Lawyers has renewed the suggestion that the Rule 26(b)(1) scope of discovery be narrowed to focus on claims (or issues) framed by the pleadings. The weight of this suggestion figured centrally in the decision to undertake the present discovery project. The specific proposal was first advanced by the American Bar Association Litigation Section in 1977, and was promptly taken up and published for comment by this committee in the form now advanced by the American College. The proposal was abandoned after publication. It has been considered repeatedly by this committee over the years, but never again has advanced as far as publication. Current discussion of the proposal has gone further, suggesting revision of the final (b)(1) provision that the information sought need not be admissible at trial if it appears reasonably calculated to lead to the discovery of admissible evidence.

This proposal has been much argued over the years. The committee agreed that there is little need for additional work by the subcommittee in preparation for the spring meeting. The subject will be discussed at the spring meeting. But the subcommittee should draft alternative proposals to modify the (b)(1) provision allowing discovery of information reasonably calculated to lead to the discovery of admissible evidence.

Documents. Document discovery is more a category of problems than a single proposal. It includes privilege waiver problems. It also includes costshifting, although costshifting can be studied for all discovery devices. Former Rule 26(f), governing "conference[s] on the subject of discovery," provided that the court should enter an order "determining such other matters, including the allocation of expenses, as are necessary for the proper management of discovery in the action." This provision seems not to have had any general impact on the practice of leaving discovery costs where they lie.

It was suggested that document discovery works well in ordinary federal cases. If change is needed for anything, it is only for the "big" cases.

It was asked whether it is possible to limit the volume of document discovery in any way analogous to the present limits on numbers of interrogatories and depositions.

A recurring suggestion has been that the scope of discovery could be narrowed for documents production, but not for other modes of discovery. The American College proposal, for example, could be adopted only as part of Rule 34. Robert Campbell stated that document production problems may be a dominant part of the concern underlying the proposal. But it was suggested that it may be difficult to implement rules that apply different tests for the scope of discovery to different discovery devices.

Notice was taken of the pre-1970 practice that required a court order on showing good cause for document production. The thought was ventured that if disclosure remains in the rules, good cause might be required for production of documents outside those disclosed. But all agreed that it would be a step backward to require a court order for document production. The pre-1970 practice should not be revived.

Costshifting was recognized as a very complex problem. Any adoption of costshifting could easily have unintended consequences. But it is good to be able to condition discovery on payment of the costs by the inquiring party -- this practice is authorized now by Rules 26(b)(2) and (c). Costshifting in general should remain open for further discussion, but the subcommittee should be responsible now only for drafting changes in (b)(2) to refer explicitly to the possibility of conditioning discovery on payment of the costs.

Privilege problems arise predominantly from the fear of inadvertent waiver by document production. It seems to be common, among parties of good will, to stipulate that production be made under a protective order providing that production does not waive privileges. It is uncertain, however, whether such orders protect against waiver as to nonparties; general opinion suggests that there is no sure protection against nonparties. Absent a stipulated protective order, the burden of screening to protect privileges is greatly enhanced and, in a "big documents" case, can impose untoward costs. This problem could be much reduced by a rule providing a procedure for preliminary examination of documents by the requesting party without waiver. The requesting party then would demand formal production of the documents actually desired, focusing the producing party's privilege review and paving the way for direct contest on whatever documents are thought privileged.

Questions were raised as to Enabling Act authority to act with respect to privileges. The Evidence Rules Committee should be consulted on any proposal that might emerge. Any rule "creating, abolishing, or modifying an evidentiary privilege" can take effect only if approved by Congress, 28 U.S.C. § 2074(b). Even if this committee and the other bodies charged with Enabling Act responsibilities conclude that a no-waiver rule that simply governs the effects of federal discovery practice does not modify a privilege, it would be important to state that conclusion and offer it for examination both by the Supreme Court and by Congress. And there may be some question whether "Erie" and Enabling Act concerns should deter action with respect to state-created privileges -- and state law governs most privileges. If state law forces waiver by any disclosure, even under a case-specific protective order or under a general procedure rule, does a no-waiver rule enlarge a state-created substantive right?

It was noted that there is some federal law on waiver, including waiver arising from public filings.

Experience often shows that overbroad assertions of privilege can be greatly reduced by scheduling a privilege hearing. Most of the assertions are abandoned before the hearing. But this approach does not alleviate the fear of inadvertent waiver by producing, rather than over-aggressive privilege assertions.

It was generally agreed that case-specific protective orders are a good device, and that a general procedure rule would be a better thing. The subcommittee is to consider these questions further.

Privilege log practice also has been identified as a potential problem. The suggestion is that some courts go beyond the limits of Rule 26(b)(5), demanding specific information about withheld documents that not only imposes undue burdens but that threatens to compel disclosure of the very information protected by the privilege. Some courts have exacerbated the problem by insisting on tight time schedules that cannot be met, and then finding waiver as a sanction for failure to timely produce the privilege log.

The question is whether anything should be done to amend (b)(5) to force all courts to honor its present meaning. One suggestion was that The Manual For Complex Litigation prescribes a good procedure that is easy to follow, and that the real problem is that many judges are too lenient, failing to demand even the level of detail required by (b)(5).

Another suggestion was that an effective protection against inadvertent waiver would greatly reduce the problems of compiling privilege logs. Privilege disputes would be much narrower and better focused. When lawyers are unable to stipulate to protective orders now, on the other hand, the privilege log can be a serious burden in the big documents case.

Further discussion reflected substantial uncertainty as to the dimensions of any privilege log problems that may exist. It was suggested that the 1993 Committee Note to Rule 26(b)(5) might be amplified, but the committee concluded that it continues to be inappropriate to attempt to modify a former Note when no action is taken on the underlying rule. In addition, it was concluded that the 1993 Note is all that could be asked. If there is a problem, it is not because of inadequacies in the Rule or the Note.

The committee concluded to suspend further consideration of the privilege log issues. The topic will be revived if additional information suggests the need for further action.

Failure to produce. Several participants in the Boston conference suggested that serious problems remain in failures to produce information properly demanded by discovery requests. The problem is not with the present rules but with failure to honor them. The question is whether there is anything to be done to enhance compliance. One suggestion has been that represented clients, as well as their lawyers, should certify the completeness and honesty of discovery responses under Rule 26(g). Another possibility is to generate still more sanctions.

It was asked why there is an asymmetry in the operation of sanctions. Rule 37(c) imposes sanctions directly for failure to make disclosure. The balance of Rule 37 imposes sanctions for failure to respond to discovery requests only if there is a motion to compel compliance, an order to comply, and disobedience to the order. Complete failure by a party to respond also can be reached under Rule 37(d).

The practical problem was identified as arising from the fact that the failures of discovery become apparent close to trial, or at trial. The disputes that arise then tend to make discovery the issue, not the merits. And "huge" fines are imposed. On the other hand, some cases deny sanctions because the demanding party waited too long to move.

Brief note also was made of the complaint that some lawyers seek to set deliberate "sanctions traps" by demanding production of documents they already have obtained by other means, hoping that the responding party will fail to produce them. Failure to produce even marginally relevant documents is then made the basis for sanctions requests and attempts to show the responding party in an unfavorable light.

These questions were put on hold. The subcommittee need not prepare more specific proposals to deal with failures to produce, nor to require party certification of discovery responses.

Rule 26(c). The committee twice published proposals to amend Rule 26(c) to specify procedures for modifying or vacating protective orders. Further action was postponed for consideration as part of this more general discovery project. Congress has been interested in the possibility that protective orders may defeat public knowledge of products or circumstances that threaten the public health or safety, and some in Congress fear that the committee has been considering these problems for too long without acting. The second published proposal also stirred concerns by expressly recognizing the widespread practice of stipulating to protective orders.

It was noted that protective orders relate to the broader problems of sealing court records and closing court proceedings. The Committee once considered a partial draft "Rule 77.1" that sketched some of the issues that must be addressed if these problems are to be covered by a rule of procedure.

It also was noted that practicing lawyers do not find any problems in Rule 26(c) as it stands.

Rule 26(c) will remain on the committee docket, but the subcommittee will not be responsible for considering this topic.

Document preservation. The committee has, but has never considered, a draft Rule 5(d) prepared to require preservation of discovery responses that are not filed with the court. It would be possible to consider a rule that prohibits destruction of discovery materials after litigation is commenced but before discovery is demanded. A beginning has been made in the Private Securities Litigation Reform Act of 1995. Special difficulties would arise with respect to electronic files. Present action does not seem warranted. The subcommittee need not prepare proposals on this topic.

Electronic Information Discovery. The Boston Conference sketched the problems that are beginning to emerge with discovery of information preserved in electronic form. These problems will evolve rapidly. Capturing solutions in rules will be particularly difficult as the pace of technology outdistances the pace of the rulemaking process. The committee must keep in touch with these problems, but it is too early for the subcommittee to attempt to find solutions. The technology subcommittee will be considering these and related problems; many of the problems will need to be explored through the Standing Committee's technology committee in conjunction with all of the several advisory committees.

Masters. The use of discovery masters was encouraged by some participants at the Boston conference. "Everybody is doing it, but Rule 53 does not address it." It was agreed that the role of special masters involves too many issues in addition to discovery issues to be part of the present discovery project. The committee has held a detailed redraft of Rule 53 in abeyance since 1994. The subcommittee need not address the matter further.

Objecting statement of withheld information. It has been suggested that a party who objects to a discovery demand be required to state whether available information is being withheld because of the objection. The underlying problem is that a party may object, force the demanding party through the work of getting an order to compel, and then reveal that there is no information available. The lack of information is not revealed even during the premotion conference. The difficulty with requiring a statement whether available information is being withheld is that the purpose of the objection may be to forestall the burden of finding out whether responsive information is available. It would be necessary to allow a statement that the party does not know without further inquiry whether responsive information is available, that further inquiry is possible, and that it is unwilling to undertake the inquiry before the objection is resolved.

Members of the committee observed that their practice is consistent with this suggestion. If they know that they have no responsive information, they say so at the time of objecting. If they do not know, they state that no search will be made until the objection is resolved.

The most aggravated form of this possible problem may arise when a party makes pro forma objections to all discovery demands, but also responds in terms that leave the inquiring party uncertain whether the responses are complete.

The dimensions of this possible problem remain uncertain. The costs of dealing with it are equally uncertain. For the moment, at least, the subcommittee will not be responsible for formulating a specific proposal.

Firm trial date. The committee turned to the "B" list of discovery subcommittee proposals.

The first of these proposals is that the national rules require early designation of a firm trial date in all actions. It was agreed that a firm trial date is a very good thing. Some courts are able to set firm trial dates, and the results are good. But there are great difficulties in requiring this practice by uniform national rule, recognizing the wide variations in docket conditions in different districts. The committee needs to choose between a national rule and recommending that these matters be handled by the Court Administration and Case Management Committee and the Federal Judicial Center as a judicial management problem. This choice can be made at the spring meeting without requiring further work by the discovery subcommittee.

Notice pleading. It was suggested that the vague notice pleadings authorized by Rule 8 are hopelessly at odds with the need to define and refine the issues for trial. Although disclosure may be used to amplify the pleadings without undoing the "great 1938 design," the role it will play depends on how disclosure practice evolves in conjunction with Rule 26(f) conferences and on further consideration of the disclosure rules. One approach would be to expand and emphasize the court's authority to order more definite statements of the issues after the initial pleadings. Although courts may order clear formulation of the issues under present Rule 16, perhaps more should be done. The subcommittee was not given any directions on this topic.

Other. It was observed that sets of interrogatories often are prefaced by elaborate definitions and instructions on how to answer. The practicing members of the committee all responded that they ignore these prefaces, choosing to answer the interrogatories as they actually are written.

Questions have been raised about the need to have a treating physician prepare an expert testimony report for disclosure under Rule 26(a)(2). The Rule is clear that such reports are not required, and the Note reinforces this conclusion. There is no need to make these provisions even more clear; if some courts misapprehend the clear rule, there is little to be done apart from pointing the judge to the clear language.

Rule 26(a)(2) does present a possible problem, however, because of the double expense that arises from requiring disclosure of an expert report, followed by deposition of the expert. Experts are being deposed after the reports. It is not clear whether this expense is justified. This topic will remain open to further consideration, but without directions for further work by the subcommittee.

The "C List" of technical discovery rule changes was left in the hands of the subcommittee for further consideration.

The discovery subcommittee is to prepare proposed rule amendments for consideration by the committee in the spring, including alternative formulations where that seems appropriate.

Rule 6(b)

The Supreme Court has sent to Congress a proposed amendment of Civil Rule 73, and proposed abrogation of Rules 74, 75, and 76. These changes reflect repeal of the statute that for some years permitted parties who agree to trial before a magistrate judge to agree also that any appeal will go to the district court, to be followed by the opportunity for permissive appeal to the court of appeals. During this process, Rule 6(b) was overlooked. Rule 6(b) prohibits extension of specified time periods, including the Rule 74(a) appeal time periods. The committee agreed that Rule 6(b) should be amended to conform to the impending abrogation of Rule 74(a). The amendment will be recommended to the Standing Committee, to be sent forward in the process when there is a suitable package of items to accompany it.

Attorney Conduct Rules

Professor Coquillette, as Reporter of the Standing Committee, described for the committee the Standing Committee's work on attorney conduct rules. Much of the work is gathered in a September, 1997 volume of Working Papers, "Special Studies of Federal Rules Governing Attorney Conduct." The Standing Committee has taken the lead on this project because it cuts across several sets of rules, and because it involves the work of the Standing Committee's Local Rules project.

The many inconsistent approaches taken by local rules to regulating attorney conduct have become a special focus of the broader local rules project. At the Standing Committee's request, Professor Coquillette has drafted a set of uniform rules to be adopted by every district court, focusing on the particular problems of attorney conduct that commonly arise and directly affect the district courts. Apart from these specific problems, the rules will adopt the rules of the state in which the district court sits (a choice-of-law provision is included for the courts of appeals). The Standing Committee will consider the draft at its January meeting. After Standing Committee approval, the matter will go to the relevant advisory committees.

The most likely form for implementing this project will be amendment of Civil Rule 83, Appellate Rule 46, and the Bankruptcy Rules. The courts of appeals do not encounter these problems frequently, making incorporation into the Appellate Rules an uncontroversial matter. The Bankruptcy courts, on the other hand, encounter many problems, particularly those involving conflicts of interest, and care a lot about the answers. They operate under the Bankruptcy Code, and are likely to want a special set of rules for bankruptcy.

It was suggested that it might be desirable to use the district court rules as the foundation for the bankruptcy court rules, with such supplemental rules as may be desirable.

Professor Coquillette said that the draft rules would not require a separate federal enforcement system in each district. The matters covered by the specifically federal rules will involve matters that can be directly enforced by the court. He also said that work is still being done on the problem of lawyers not admitted to practice in the district court's state.

Admiralty Rules B, C, E

Mark Kasanin introduced discussion of the proposed amendments to Admiralty Rules B, C, and E. He noted that these proposals began several years ago with the Maritime Law Association and the Department of Justice. Much of the work has been done by Robert J. Zapf, who attended this meeting as representative of the Maritime Law Association, and Philip Berns of the Department of Justice, who also attended this meeting. The Admiralty Rules subcommittee has worked with them, refining the drafts to remove most points of possible dispute.

Many of the proposed changes reflect changes in statutes or in Civil Rules that are explicitly incorporated in the Admiralty Rules. Styling changes also have been made, and are so extensive that it is not helpful to set out the changes in the traditional overstrike and underscore manner.

Perhaps the most important changes have been separation of forfeiture and admiralty in rem procedures in Rule C(6), and deletion of the confusing "claim" terminology from Rule C(6).

Philip Berns introduced the history of the changes, noting that the roots of this project began back in 1985 or 1986 with the need to relieve marshals of the requirement of serving process in all maritime attachments. Attachment of a vessel or property on board a vessel still demands a marshal, a person with a gun, because these situations can be sensitive and potentially fractious. The service requirements in fact were changed in Rule C(3), but for some unknown reason parallel changes were not made in Rule B(1).

Another need to amend the rules arises from the great growth of forfeiture proceedings. Forfeiture procedure has adopted the maritime in rem procedure of Rule C. But the admiralty procedure for asserting claims against property is not well suited to forfeiture proceedings. In addition, there is a greater need to move rapidly in admiralty in rem proceedings, so as to free maritime property for continued use.

Robert Zapf underscored these reasons for amending the rules.

The adoption of the alternative Rule C(3)(b) service provisions into proposed Rule B(1)(d) was discussed and approved.

Proposed Rule B(1)(e) responds to the problem arising from incorporation of state law quasi-in-rem jurisdiction in the final provisions of present Rule B(1). Rule B(1) now incorporates former Rule 4(e), failing to reflect the amendment of Rule 4(e) and its relocation as Rule 4(n)(2) in 1993. Rule 4(e) allowed use of state quasi-in-rem jurisdiction as to "a party not an inhabitant of or found within the state." It provided a useful supplement to maritime attachment under Rule B(1). New Rule 4(n)(2), however, allows resort to state quasi-in-rem jurisdiction only if personal jurisdiction cannot be obtained over the defendant in the district in which the action is brought. Because maritime attachment is available in many circumstances in which personal jurisdiction can be obtained in the district -- it is required only that the defendant not be "found within the district" -- substitution of Rule 4(n)(2) for Rule 4(e) would serve little purpose. Discussion focused on the argument that Rule B(1)(e) should incorporate state quasi-in-rem jurisdiction without any limitations, discarding reliance on Rule 4. Objections were voiced in part on the same grounds that led to the restrictions incorporated in Rule 4(n)(2), and also from doubt that the quasi-in-rem jurisdiction aspect of Rule B(1) needs to be expanded. Further discussion showed that the main use of state law is as a means of effecting security, not jurisdiction. Although present practice seems to recognize that state law security remedies are available in admiralty through Civil Rule 64, it was decided that the draft Rule B(1)(e) should be revised to incorporate Rule 64, deleting any reference to state-law quasi-in-rem jurisdiction. The Note will reflect that this incorporation is effected to ensure that repeal of the former Rule 4 incorporation is not thought to make use of Rule 64 inconsistent with the supplemental rules. It was further agreed that deletion of state law quasi-in-rem jurisdiction seems to justify abandonment of the present reference to the restricted appearance provisions of Rule E(8). This issue was delegated to the admiralty subcommittee for final action.

Draft Rule C(2)(d)(ii) adds a new requirement that the complaint in a forfeiture proceeding state whether the property is within the district, and state the basis of jurisdiction as to property that is not within the district. This requirement responds to several statutory provisions allowing forfeiture of property not in the district. The draft was approved.

The notice provisions of draft Rule C(4) include a new provision allowing termination of publication if property is released after 10 days but before publication is completed. This change simply fills in an apparent gap in the present rule, both for the purpose of avoiding unnecessary expense and for the purpose of reducing possible confusion as to the status of the seized property.

The draft divides Rule C(6) into separate paragraph (a) procedures for forfeiture and paragraph (b) procedures for maritime arrests. Two major distinctions are made. A longer time is allowed in forfeiture to file a statement of interest or right, and the categories of persons who may file such statements include everyone who can identify an interest in the property. In admiralty arrests, on the other hand, a shorter time is allowed for the initial response because of the need to effect release of the seized property for continuing business. The categories of persons who may participate directly is narrower than in forfeiture, being restricted to those who assert a right of possession or an ownership interest. Lesser forms of property interests can be asserted in admiralty arrests only by intervention, in keeping with traditional practice. The Maritime Law Association has urged that the reference to ownership interests in C(6)(b) include "legal or equitable ownership." The Reporter objected that it is better to refer only to "ownership," as a term that includes legal ownership, equitable ownership, and any other form of ownership recognized by foreign law systems that do not respond to the Anglo-American distinction between law and equity. The Note makes clear the all-embracing meaning of "ownership." After discussion it was agreed that the multiple meanings of ownership could be made secure by amending the draft to refer to "any ownership" in C(6)(b)(i) and (iv). It was emphasized that the Note discussion of the changes in C(6) is an important part of the process, making it clear that elimination of the confusing reference to "claimant" and "claim" in the present rule is not intended to change the substance of admiralty rights or the essence of the allied procedure.

It was noted that draft Rule C(6)(c), continuing the admiralty practice of allowing interrogatories to be served with the complaint, was expressly considered in relation to the discovery moratorium adopted by Rule 26(d) in 1993. It was concluded that the special needs of admiralty practice justify adhering to this longstanding practice.

Draft Rule E(3) was presented in alternatives, a Reporter's draft and an MLA draft. The MLA draft deliberately uses more words to say the same things, in order to emphasize that process in rem or quasi-in-rem may be served outside the district only when authorized by statute in a forfeiture proceeding. The MLA version was supported by the admiralty subcommittee, and adopted by the committee.

Draft Rule E(8) must be adjusted to conform to draft Rule B(1)(e). Incorporation of Rule 64 in Rule B(1)(e) requires deletion of the incorporation of former Civil Rule 4(e) in Rule E(8). If the reference to Rule E(8) is deleted from revised B(1)(e), there is no apparent need to refer to Rule 64 in Rule E(8). The admiralty subcommittee will make the final decision on this point.

Draft Rules E(9) and (10) were approved for the reasons advanced in the draft Note.

Changes to Civil Rule 14 to reflect the changes in Supplemental Rule C(6) also were approved.

The package of Admiralty Rules amendments was approved unanimously. It was agreed that it would be desirable -- if possible under Enabling Act processes -- to reduce the period required to make these changes effective. This question will be addressed in the submission to the Standing Committee with the request that the proposed rules be published for comment.

Assistant Attorney General Hunger reported on the status of pending statutes that would bear on the proposed forfeiture rule amendments. The Department of Justice will continue to work with Congress on these matters.

Mass Torts

This committee began to review Civil Rule 23 at the suggestion of the Standing Committee in response to the urging of the Ad Hoc Committee on Asbestos Litigation. Mass torts present problems that are inherently interstate in nature. There often are tensions among state courts, and between state and federal courts, arising from overlapping actions. Special problems arise from the strong need of defendants to achieve global peace; these defense interests affect plaintiffs who want to settle. There are many problems that have not been resolved. Bankruptcy is often held out as a model, with such intriguing variations as "product-line bankruptcy." Interpleader, "bill-of-peace," and other traditional models have been offered for reexamination and possible expansion.

Increasing opportunities to inflict widely dispersed injuries have increased the burden of dispersed litigation and the desire to find solutions. Many of the proposed solutions require legislation. Civil Rules amendments cannot alone provide solutions.

The Judicial Conference has considered appointment of an ad hoc mass torts committee. The work of any such committee would bear on the work of many other Judicial Conference committees, including the rules committees. It would be necessary to coordinate its work with these committees, and particularly to ensure that specific rules proposals be subjected to the full Enabling Act process for adoption. The committees most obviously affected include the Federal-State Jurisdiction Committee, the Bankruptcy Administration Committee, and the Judicial Panel on Multidistrict Litigation. The Court Administration and Case Management Committee also might become interested, and of course the Manual for Complex Litigation is involved. These problems have made the Executive Committee wary of appointing a new committee. At the same time, it is anxious that the Judicial Conference process be actively involved with these problems.

This committee has learned much about mass tort litigation in its Rule 23 inquiries, and is a logical focal point for further efforts. Judge Niemeyer has proposed that a Mass Torts Subcommittee of this committee be created, to include liaison members from the most directly involved Judicial Conference Committees. The subcommittee would be charged with sorting through recommendations for addressing mass torts by coordinated legislation, rules changes, and other means. The task is formidable, and success is by no means guaranteed. A special reporter would be needed. Judge Niemeyer has asked Judge Scirica to chair the subcommittee, if it is authorized, recognizing that this will be a long-range project. The work must be tentative at first, and slow. Although there is a natural reluctance to continue to develop subcommittees, there are too many large-scale projects for this committee to work on each one as a committee of the whole. Here, as with the admiralty and discovery subcommittees, the subcommittee can be put to work on a "task-specific" basis.

It was noted that the subcommittee must remain sensitive to the risk that enthusiasm for particular proposals may entice it toward rules that trespass over the line into substantive matters.

A prediction was made that unless Congress will enact substantive laws, the only workable answers will be found through amendment of Civil Rule 23 or development of a specific class-action procedure for mass torts.

Rule 23

The proposed new Rule 23(f) is on its way to the Supreme Court. Rule 23(c)(1) has been commended by the Standing Committee for further study in conjunction with remaining Rule 23 questions. At the May meeting, the committee voted to abandon the proposed new factors (A) and (B) for Rule 23(b)(3); the "maturity" element proposed for new factor (C) was redrafted and carried forward. Proposed factor (F), colloquially referred to as the "just ain't worth it" factor, remains on the agenda for further consideration. The proposed settlement-class provision, which would be new Rule 23(b)(4), also remains on the agenda, along with the proposed amendment of Rule 23(e).

"Factor (F)." At the May meeting, the committee determined to consider five alternative approaches to factor 23(b)(3)(F) as published in 1996. The published version added as a factor relevant to the determination of predominance and superiority "whether the probable relief to individual class members justifies the costs and burdens of class litigation." The first approach would be to adopt the factor as published. This approach would require several changes to the Committee Note to reflect concerns raised by the testimony and comments. There was a widespread misperception that this factor would require a comparison between the probable relief to be received by one individual class member with the total costs and burdens of class litigation. If a class of 1,000,000 members stood to win $10 each, the comparison would weigh the $10, not the $10,000,000 in a process that inevitably must find the individual benefit outweighed by the costs and benefits of class litigation. The Note would have to be changed to dispel any remaining confusion, making it clear that the aggregation of individual benefits is to be compared to the aggregate costs. In addition, the Note should be changed to take a position on an issue that the Committee had earlier voted to leave aside -- whether measurement of the probable relief to individual class members entails a prediction of the outcome on the merits. Many of those who testified or commented believed that the proposed rule would require such a prediction on the merits. Other issues as well might need to be addressed in the Note, responding to additional concerns presented by the testimony.

A second approach would be to abandon the published proposal.

Another approach would delete the reference to "probable relief," substituting some formula that does not seem to invoke a prediction of the outcome on the merits. One possible formulation would be: "whether the relief likely to be awarded if the class prevails justifies the costs and burdens of class litigation."

A fourth approach would eliminate the reference to individual relief, focusing only on aggregate class relief. This approach could be combined with the third: "whether the relief likely to be awarded the class if it prevails justifies the costs and burdens of class litigation."

The fifth approach would be to create an opt-in class alternative for situations in which the recovery by individual class members seems so slight as to raise doubts whether class members would care to have their rights pursued. Certification of an opt-in class would provide evidence of class members' desires; if they opt in, that is proof that they wish to vindicate their rights.

All of these approaches were discussed against the underlying purposes that led to proposed factor (F). We do not wish to foster lawyer-driven class actions, where the lawyer first finds a "claim" and then finds a passive client without any substantial purpose to advance the interests of class members or the public interest. But it is different if persons holding small claims desire vindication and seek out a lawyer. Rule 23 should be available for small claims that cannot be effectively asserted through individual litigation. Is it possible to distinguish these situations by rule? One possibility is to resort to the opt-in class alternative, providing direct evidence whether class members desire enforcement.

A new suggestion was made that all of these alternative approaches involve speculation about the outcome on the merits. Focus on cases of meaningless individual relief should instead be placed in Rule 23(e). The problems arise from settlements -- often the "coupon" settlements -- and they can be addressed by refusing to approve settlements that award meaningless relief to the class and fat fees to counsel.

It was suggested that the specter of fat fees and meaningless class recovery is only a myth. The Federal Judicial Center study showed what other studies show -- fee awards generally run in a range of 15% to 20% of the aggregate class recovery. Many cases now are denied certification because the judge thinks they are useless; the superiority requirement authorizes this. Adding any variation of factor (F) will destroy the consumer class; it is contrary to the philosophy of Rule 23. The opt-in alternative is a delusion. In California, once a statutory or constitutional violation by the state has been adjudicated, an opt-in class can be formed. Even in this situation, with liability established, lawyers do not resort to the opt-in class because it is too expensive in relation to the results. Potential class members simply do not undertake the burden of opting in.

It was responded that opt-in never has been given a chance. A class member who is not willing to opt in does not belong in court.

The rejoinder was that there is a vast difference between opt-in and opt-out. Most classes are lawyer driven. This is recognized by rules of professional responsibility that allow lawyers to advance the costs and expenses of the litigation.

It was suggested that the opt-in alternative should be separated. The first decision to be made is whether the merits should be considered as part of the (F) calculation.

Another observation was that there is a philosophical chasm on small-claims classes. Adoption of any of the (F) alternatives would be the death-knell of consumer classes. These alternatives should be considered before moving to consideration of the opt-in class alternative.

This discussion led to the plaint that the committee has pursued these issues around the same tracks for several meetings. After much hard work, there still is no clear definition of what the proposal is designed to accomplish. Comparison to the relief requested for the class will accomplish nothing, since no one begins by asking for coupons or other trivial relief. The opt-in alternative is odd, because with very small claims it is not worth it to opt in. The proposed draft that would incorporate the opt-in alternative in the Rule 23(c)(2) notice provisions turns on finding reason to question whether class members would wish to resolve their claims through class representation, but does not provide any guidance to the circumstances that might raise the question. There has been no definition of what is meant by the "costs and burdens" of class litigation. We do not know how to implement this concern. The effort should be abandoned.

A motion to abandon further consideration of proposed factor (F), keeping the opt-in alternative alive for further consideration, passed with one dissent.

Opt-in classes. Discussion of the opt-in alternative pointed to several issues that must be resolved. Some of the drafts were integrated with the now-abandoned factor (F) proposal, authorizing consideration of an opt-in class only after certification of an opt-out class had been rejected under factor (F). If (F) disappears, some other means must be found to distinguish the occasion for an opt-in class from the occasions for opt-out classes. Even the (c)(2) notice draft adopted for purposes of illustration one alternative formulation of the (F)-factor drafts: "When the relief likely to be awarded to individual class members does not appear to justify the costs and burdens of class litigation and the court has reason to question whether class members would wish to resolve their claims through class representation, the notice must advise each member that the member will be included only if the member so requests by a specified date." Any of the alternative (F) formulations would do, and some alternative switching point might do better. But some means must be found, unless opt-in is to replace opt-out for all (b)(3) classes, or unless the court is given a discretionary choice between opt-in and opt-out for all (b)(3) classes. And at some point, it may seem inappropriate to aggravate the already curious Rule 23 structure that incorporates the distinction between opt-out and mandatory classes only in the notice provisions of subdivision (c).

Opt-in classes also require attention to several subsidiary issues. It must be made clear that the "class" includes only those who in fact opt in, not those who were eligible to opt in but did not. The class notice must specify the terms on which members can request inclusion; it would be helpful to indicate, in Rule or Note, whether the terms can reach sharing of costs, expenses, and fees. It might be useful to address the effects of opt-in classes on statutes of limitations, and the availability of party-only discovery devices and counterclaims against those who opt in. Thought also must be given to the question whether the judgment in an opt-in class can support nonmutual issue preclusion in later litigation, whether brought by those who were eligible to opt in or by others.

The opt-in class alternative in (c)(2) raised the same question as the (F) factor: what level of individual recovery triggers the opt-in alternative? The "$300" that was the median recovery in one of the districts in the Federal Judicial Center study?

Even the opt-in alternative continues to present the question whether the merits should be considered, as a matter of likely relief or as a matter of justifying the costs and burdens of class litigation.

The opt-in approach was supported as a way of showing whether there is support for litigation among the supposed class members. This is better than present practice, which allows a lawyer to volunteer as a "private attorney general" on behalf of a class that does not care and in service of a public interest that public officials do not find worth pursuing.

It was urged that the opt-in approach should be applied to all (b)(3) classes, without the complications of attempting to separate opt-in from opt-out classes.

It was responded that opt-in classes are a revolutionary idea. The Supreme Court sang the virtues of small-claims classes in the Shutts decision. Even constitutional doubts might be raised about substituting opt-in for opt-out classes. Who pays for notice? What about repetitive classes, made up of those who choose not to opt in to the first class? In effect, settlement classes today ordinarily are opt-in classes because they reach only those who file proofs of claim.

The fear that due process might defeat opt-in classes was doubted by others.

Opt-in was further supported as simple and clear. The opt-out provision was a last-minute addition to (b)(3). We should find a device that avoids any preliminary consideration of the merits, and opt-in does it.

Another member suggested that the (c)(2) draft that would allow a judge to opt out of opt-out class certification in favor of an opt-in class is a worthy idea, but is overcome by problems. A rule of procedure can generate preclusion consequences -- Rule 13(a) and 41 are obvious examples. But we cannot allow nonmutual preclusion to rest on an opt-in class judgment. And we cannot bind those who choose not to opt in. The small-claim area, moreover, is the area where opt-in will work least well. And what is to be done under the draft when a small number of individual claimants in fact appear: does this upset the "reason to question whether class members would wish to resolve their claims through class representation"?

The fear that opt-in classes would spur successive class actions was met by the observation that multiple and overlapping classes occur now.

The private attorney-general function was brought back for discussion with the observation that the committee has never rejected this concept. Opt-in classes would greatly reduce this function.

It was predicted that adoption of an opt-in class alternative would drive small-claims classes to state courts. But federal courts should provide the forum for resolution of nationwide issues. Economically, moreover, a lawyer can afford to invest $200,000, $500,000, or $1,000,000 in notice to an opt-out class; the investment is not possible for an opt-in class, because there will not be enough opt-ins.

The fear of driving national classes to state courts was countered by the suggestion that amendment of the federal rules would lead to parallel amendments by many states, discouraging resort to state alternatives.

An alternative to opt-in classes to control lawyer-driven actions might be to base fees on the amount of relief actually distributed. It has been suggested that counsel fees are often based on the maximum possible distribution, and are a far larger percentage of relief actually distributed in small claims cases. The Committee has not been able to get any clear sense whether this suggestion is often borne out in practice; adoption of the fee rule might give better evidence.

The conclusion was that the opt-in issues should remain open for further exploration. Earlier committee proposals had envisioned opt-in classes as a promising approach to mass tort litigation. The Mass Torts Subcommittee may be the best place for the next phase of study.

Opt-in classes were further defended on the ground that collective action on behalf of many should turn on agreement to be included. The opt-out default presumes consent that is not real.

Settlement classes. In 1996, the committee published for comment a proposed Rule 23(b)(4) that would allow certification of a class when "the parties to a settlement request certification under subdivision (b)(3) for purposes of settlement, even though the requirements of subdivision (b)(3) might not be met for purposes of trial." This proposal followed a long period during which the committee repeatedly considered the problems of settlement classes but found no clearly sound approach to the many problems involved with drafting a rule to regulate the practice. The proposal was intended only to overrule the Third Circuit rule that a class can be certified for settlement purposes only if the same class would be certified for trial. See Georgine v. Amchem Products, Inc., 3d Cir.1996, 83 F.3d 610; In re General Motors Corp. Pick-Up Truck Fuel Tank Litigation, 3d Cir.1995, 55 F.3d 768. The Supreme Court affirmed the Georgine decision, but the opinion states that a (b)(3) class can be certified for settlement even though "intractable management problems" would defeat certification of the same class for trial. Amchem Prods., Inc. v. Windsor, 1997, 117 S.Ct. 2231, 2248. Although the Court took note of the published committee proposal, the opinion also notes that the proposal had been the target of many comments "many of them opposed to, or skeptical of, the amendment," 117 S.Ct. at 2247. The Court's opinion, moreover, discusses settlement classes in terms that are not clearly as limited as the published proposal. The opinion could be found to reach classes certified under subdivisions (b)(1) or (b)(2), and is not limited -- as the published proposal was -- to situations in which the parties agree on a proposed settlement before seeking class certification. The reach of the Court's opinion may be uncertain in other dimensions as well.

In these circumstances, it was urged that simple adherence to the committee's published proposal would be unwise. The central purpose has been accomplished by the Supreme Court. It is not clear whether adoption of the proposal would merely bring the Court's interpretation into the text of Rule 23. There is only minor benefit in adding this particular gloss to the text of the rule, when so many other important aspects of class-action practice have not been added to the rule. And there is great risk that inconsistencies may exist between what the Court intended and what the amended rule might come to mean. Because the Committee cannot be confident of what the Court intended, cannot be confident whether the published proposal means something else, and cannot be confident of the ways in which an adopted amendment might be interpreted against the background of the Court's opinion, further work is necessary if Rule 23 is to be amended to address settlement classes.

It was suggested that the Amchem decision means that a nationwide mass tort class action cannot be settled. Problems of conflicting interests within the class and related inadequacies of representation will be insurmountable.

This suggestion led to the more general suggestion that the time is not ripe for immediate action on settlement classes. District court decisions since the Amchem decision seem to be moving toward stricter certification standards. It will be desirable to give more thought to the problem, and to gain the benefit of greater experience. In the Amchem case itself, the result so far has been that individual claims are being settled according to the protocols of the settlement; the only difference is that far greater amounts are being devoted to attorney fees. Many of the settlement-class issues are properly considered with the problems of mass torts. There are genuine problems to be addressed. The "limited fund" problem is real in the most widespread mass torts. Transaction costs are a great problem, as reflected in the RAND study of asbestos litigation. The best solutions may lie beyond the limits of the Enabling Act.

It was observed that the Fibreboard settlement is back in the Fifth Circuit, and may return to the Supreme Court in a way that will shed light on use of limited-fund (b)(1) settlement classes. In the same vein, it was noted that the Court has twice granted certiorari in cases that were meant to present the question whether mandatory classes can be used for mass torts; this level of interest suggests that another vehicle soon may be found to address this issue.

These difficulties and opportunities led to a consensus that it is better to defer further consideration of settlement classes. The committee has never been able to find attractive proposals to do more than overrule the Third Circuit rule that limits settlement classes to those that could be tried with the same class definition. The Supreme Court has provided plenty of food for further lower court thought. Although further proposals are not precluded by the Supreme Court opinion, it is better to await developments. The Mass Torts Subcommittee is likely to be considering these issues. If problems emerge as lower courts develop the Amchem opinion, the committee can return to the issue.

Other Rule 23 issues. The committee considered briefly two drafts that it requested at the May meeting. One provided alternative approaches to enhancing the "common evidence" dimension of Rule 23(b)(3) classes. The more demanding approach would require that for certification of a (b)(3) class, "the trial evidence will be substantially the same as to all elements of the claims of each individual class member." The softer approach would add a new factor, focusing on "the ability to prove by common evidence the fact of injury to each class member [and the extent of separate proceedings required to prove the amount of individual injuries]."

The other draft dealt with repetitive requests to certify the same or overlapping classes. It would add a new factor to (b)(3), allowing consideration of "decisions granting or denying class certification in actions arising out of the same conduct, transactions, or occurrences."

It was asked whether data can be got on the frequency of multiple certification attempts. Thomas Willging observed that the Federal Judicial Center study had some data, that showed at least one overlapping action in 20% to 40% of the classes, varying from district to district.

State court class actions were again noted as an alternative to federal actions, with the suggestion that changes in Federal Rule 23 might be followed by many states.

It was suggested that both drafts were interesting and deserved study. It was noted that the committee still has on its agenda the proposal to amend Rule 23(c)(1) to allow certification "when practicable," and the revised "maturity" factor for (b)(3) classes. Settlement classes and opt-in questions remain on the table, but are not ready to go ahead with recommendations for publication of specific proposals.

Brief discussion of the (c)(1) proposal asked whether "practicable" is the best word to use. It was noted that during the Standing Committee review of (c)(1), it was suggested that the key is to identify the purposes underlying the desire for early determination of certification requests. It also was suggested that these purposes may implicate so many different factors that it will be difficult to find a better single word.

These Rule 23 issues were continued on the agenda.

Judicial Conference CJRA Report

The Judicial Conference CJRA Report was summarized in the agenda materials. Each of the recommendations that bear on the work of this committee were included. Most of the recommendations were discussed extensively during the report of the discovery subcommittee because they bear directly on its work. All of the recommendations will be subjected to prompt and thorough continuing study.

Certificate of Appreciation

A certificate signed by all committee members was presented to Carol J. Hansen Posegate, commemorating and thanking her for six years of great service on the committee.

Electronic Filing

Peter McCabe presented a report on the status of electronic filing experiments, observing that developing experience is revealing many areas in which the Civil Rules must be studied to ensure effective application to electronic filing and, eventually, electronic service. The report was illuminated by a presentation by Karen Molzen on the Advanced Court Engineering project. Among the practical problems discussed were the use of the log-in and "key" for the attorney's signature; means of covering filing fees -- credit cards and attorney deposit accounts are the most likely means; difficulties confronting pro se litigants; and systems for detecting attempts to alter filed documents. The work of the clerk's office has already been affected; the need for paper has been reduced significantly. An attorney who submits an affidavit electronically must retain the original. When a judge authorizes filing, a facsimile signature is affixed to the order. There is a "firewall" system to ensure security. Different persons are allowed different and controlled levels of access to the system. FAX and email noticing are being used; if the message does not go through in three tries, a notice is printed out with a mailing label. A list of potential problems with the rules of procedure is being developed; it will be sent on to Judge Carroll as chair of the Technology Subcommittee.

Next Meetings

The date for the next meeting was set at March 16 and 17, 1998. It was agreed that if a second spring meeting becomes necessary -- most likely because great progress has been made with Discovery Subcommittee proposals that might be made ready to recommend for publication with one more meeting -- it will be held on April 30 and May 1. Locations were not set for either meeting.

Respectfully submitted,

Edward H. Cooper, Reporter