Writer's Direct Line (973) 426-3200
Facsimile No. (973) 426-3213
Internet Mail Address: allmant@basf.com

December 12, 2000

The Honorable John L. Carroll
Chief United States Magistrate Judge
United States District Court
Middle District of Alabama
PO Box 430
Montgomery, AL 36101

Re: The Need for Federal Standards regarding Electronic Discovery

Dear Judge Carroll:

I appreciated the opportunity to participate in the October 27, 2000, meeting of the Discovery Subcommittee of the Federal Civil Rules Advisory Committee (hereinafter, the "Brooklyn Conference"). I would like to take this opportunity to submit, by way of supplement, some specific reasons for the positions I advocated as well as some possible language for consideration by the Subcommittee.

Electronic Records are Quite Different from Paper Copies

As you will recall, the Subcommittee asked for comments on whether the use of electronic records constituted a mere evolution in or a true departure from past practice in ways relevant to discovery. In my view, there are two major differences that, I will argue, have created a need for the Subcommittee to act. First, the sheer volume of information available in the electronic context is materially different. Take, for example, the matter of routine interoffice communications. The use of electronic mail has increased geometrically the number of places where "copies" of those types of documents may be located. Such documents may exist both in systems designed to retain and manage such records as well as in systems only incidentally intended for that purpose. Moreover, copies may be heroically retrieved from locations not ordinarily available to the persons engaged in such communications. Part or all of "deleted" documents may be reconstructed from the hard drive of an originating personal computer ("PC") or from network servers relating to that PC, as well as PCs or servers of recipients. Copies also might be retrieved from "back-up" tapes created daily, weekly or monthly for purposes of disaster recovery, albeit at substantial cost.

A second major difference relevant to this discussion is the entirely new class of "embedded" information that is uniquely available in the electronic discovery world. It is yet to be seen if that type of electronic information is within the scope of discovery and it may be necessary to clarify the definitions relevant to that topic.¹

1 A member of the Federal Civil Rules Advisory Committee has noted that electronic records often contain a "new breed" of "embedded" information that may not even be covered by Rule 34 as it currently exists. See Scheindlin and Rabkin, "Electronic Discovery in Federal Civil Litigation: Is Rule 34 Up to the Task?," 41 Boston College Law Review 1(2000)(advocating amendments to Rule 34 to rectify issues identified).

Failure to Differentiate in Electronic Discovery Risks Confusion

The Federal Rules do not explicitly outline the required retrieval effort for producing parties. The burden is on the producing party to object to demands deemed to be onerous. Rule 26(c) allows issuance of "protective" orders and courts have been directed under Rule 26 (b)(2) to balance the likely benefit against the burden or expense to the producing party on a case-by-case basis. In the early phases of electronic discovery, this approach was sufficient since the 1970 amendment to Rule 34 added only "ordinary business records" kept by electronic means. See Pearl Brewing v. Jos. Schlitz Brewing, 415 F. Supp. 1122, 1136 (S.D. Tex. 1976); Manual for Complex Litigation (Third), Section 21.446 ("information generated and maintained in the ordinary course of business"). Courts routinely required that requesting parties pay some or all of the extraordinary costs associated with production. See Zonaras v. General Motors Corporation, 1996 U.S. LEXIS 22535 (S.D. Ohio 1996). However, as technical consultants have acquired greater influence and publicity has attached to extreme cases, requesting parties have increasingly argued that all electronic records - no matter how remote or difficult to locate - must be retrieved at the cost of the producing party. One oft-cited opinion that does not acknowledge the distinction among types of records asserts that such a retrieval obligation is entirely appropriate. See In re Brand Name Prescription Drugs Antitrust Litigation, 1995 WL 360526 *2 (N.D. Ill.)(characterizing the need to implement retrieval program as "an ordinary and foreseeable risk" of maintaining electronic systems). This approach is apparently based on a misplaced conviction that regardless of the burdens or complexities, the litigation process accords priority to discovery to the full extent of the technology. See Bills v. Kennecott Corporation, 108 F.R.D. 459, 463-64 (D.C. Utah 1985) (refusing to allocate costs of producing computer data to requesting party and holding that "information stored in computers should be as freely discoverable as information not stored in computers").

Adding to the problem is the misconception that the production of electronic records is easy to accomplish (the "press the button" syndrome). This, of course, is not the case. Computer systems are far more complicated and are operated for business needs that have nothing to do with litigation. A one-sided concentration on production of all electronic copies without acknowledging the differences between electronic and paper copies risks undermining the practical interpretation of Rule 34 adopted in paper copy discovery. Parties in that context routinely seek to produce in the first instance only those documents that are readily accessible to the identifiable personnel who may retain or have access to them in the ordinary course of business. Similarly, parties seeking to comply with their duty to preserve relevant records after notice of a claim or lawsuit appropriately notify only the same discrete and predictable class. Without clear guidance from the Rules, however, a serious risk exists that parties planning their compliance will be confused as to the continued acceptability of these practices if no explicit recognition of the differences in electronic records is established.

Traps for the Unwary

Parties proceeding in good faith to produce electronic records are at risk that perfectly appropriate business actions taken in accordance with past practice will later be deemed inappropriate. At any given time, large commercial and governmental users may be defending hundreds of cases, all started at different times and all alleging different claims. When each successive litigation requires restriction on the reuse of backup tapes, the entity finds itself in the unenviable position of converting its backup systems into defacto litigation storage systems, a burden never intended by the Federal Rules. See, e.g., Applied Telematics, Inc. v. Sprint, 1996 U.S. Dist. LEXIS 14053, *11 (Sept. 18, 1996)(holding defendant "at fault for not taking steps to prevent the routine deletion of the backup files"); In re Tyco Securities Litigation, 2000 U.S. Dist. LEXIS 11659, *9 (July 27, 2000) (finding that "large corporations typically overwrite and thereby destroy electronic data in the course of performing routine backup procedures"). Indeed, the failure to produce backup tapes held for one pending case in a successive case has led to sanctions - including spoliation inferences - when judged in retrospect. See Linnen v. A.H. Robins, 1999 Mass. Super. LEXIS 240, *31 (June 16, 1999) ("failure to preserve documents requested by a party is inexcusable conduct.") This is an unworkable standard that forces large users to choose between maintaining their normal business operations and surrendering valid claims or defenses.

Proposed Changes to Rule 34 regarding Electronic Discovery

The basic duty of production for electronic records should, like hard copy discovery, extend in the first instance only to those records that are reasonably available in the ordinary course of business. As you will recall, there was no real disagreement with this position expressed at the Brooklyn Conference, yet most testifying witnesses disagreed that this principle should be placed into the rules. Some asserted that most disputes were resolved by reaching a practical compromise and thus no need existed. Others pointed to the tools available to fashion relief, especially the balancing provisions of Rule 26(b)(2), and stressed the merits of a case-by-case approach. However, rather than requiring parties to guess at their responsibilities, it would be much fairer for everyone to have the understanding where the line exists without having to litigate it in each case. I pointed out that the Texas Rules of Civil Procedure had made a useful step in this regard by adopting a new rule devoted to the topic.² A similar provision could be incorporated into Rule 34(a) by adding a requirement that:

"The responding party must produce the electronic or magnetic data that is responsive to the request and is reasonably available to the responding party in the ordinary course of business. If the responding party cannot - through reasonable efforts - retrieve the data or information requested or produce it in the form requested, the responding party must state an objection complying with these rules."

2 Texas Rule of Civil Procedure 196.4 requires production of electronic or magnetic data that "is reasonably available to the responding party in its ordinary course of business" and provides that the party may object to such production if "reasonable efforts" do not allow for retrieval of the data or information requested.

Such a rule would help both requesting and producing parties, and would also guide and inform the Magistrate and District Court judges who might be called upon to resolve disagreements. Requiring a producing party to search its electronic records beyond those maintained and readily available in the ordinary course of business flies in the face of the fundamental principle that each party must bear the "ordinary burden of financing his own suit." Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 179 (1974). Naturally, as we discussed at the Brooklyn Conference, there are instances where special requirements of individual cases require production of electronic records not maintained in the ordinary course of business. In those circumstances when the case is made for an extraordinary production effort beyond that required as a matter of course, Rule 34 could explicitly incorporate a cost allocation that reflects that fact. Thus, a new provision, analogous to Texas Rule 196.4, could require that:

"If the court orders production by the party upon whom the request is served, the court must also order that the requesting party pay the reasonable expenses of any extraordinary steps required to retrieve and produce the information."

Anecdotal information from the Texas experience indicates that predictable allocation of costs has dramatically reduced the overbroad nature of many requests.

Spoliation and Preservation Issues

Large users of electronic records have legitimate business reasons to eliminate redundant or stale electronic records not intended for retention as business records. Some have argued, however, that the compact size and relative ease of retention of electronic records mean that producing parties have "no excuse" not to retain all such material forever. This argument is totally inconsistent with implementation of retention polices, see Lewy v. Remington, 836 F.2d 1104 (8th Cir. 1988), and has had a paralyzing effect upon development of such policies. It would be useful to have an explicit amendment to Rule 34 stating that parties should not be required to suspend the normal operation of reasonable document destruction without prior orders and limiting the spoliation sanctions to willful violations of such orders.

For example, Rule 34(d) could provide that:

"No sanctions or other relief predicated upon a failure to maintain or preserve documents or data, including electronically stored information, shall be entered in the absence of a discovery request that describes with particularity the specific documents or data requested and evidence that (1) the documents or data requested were relevant to the claim or defense of a party and (2) the party upon whom the request was served willfully failed to preserve such documents or data. Evidence that reasonable steps were undertaken to notify relevant custodians of preservation obligations shall be prima facie evidence of compliance. Nothing in these rules shall require the responding party to suspend or alter the operation in good faith of disaster recovery or other electronic or computer systems absent court order issued upon good cause shown."

This approach is consistent with ABA Civil Discovery Standard 29(a)(iii), which provides for the restoration of material deleted in the regular course of business only upon a showing of substantial need.

Conclusion

There are important differences between hard copy and electronic discovery and these differences have consequences in both the litigation process and business world. I would hope that as the Subcommittee performs its review function it would take into account how those differences are impacting the latter as well as the former.

If I can be of any further help to the Subcommittee, it would be my pleasure to do so. Thank you again for the opportunity to share these views with you.

Sincerely,

Thomas Y. Allman

cc: Prof. Richard L. Marcus; Hon. David Levi, U.S.D.J.