Gregory P. Joseph*

The Advisory Committee on the Federal Rules of Civil Procedure has published draft amendments addressing electronic discovery. The proposals extend beyond electronic discovery in some respects. None of the proposals will take effect before December 2006, and all are subject to review and reconsideration.

Parties’ Initial Discovery Conference (Rule 26(f)). Rule 26(f), which governs the parties’ initial discovery conference, would be amended to add three topics to the discussion:

(1) ‛any issues relating to preserving discoverable information;“

(2) ‛any issues relating to disclosure or discovery of electronically stored information, including the form in which it should be produced;“ and

(3) ‛whether, on agreement of the parties, the court should enter an order protecting the right to assert privilege after production of privileged information.“

The proposed amendments to Rule 26(f) are sound. Preservation (or spoliation), electronic discovery, and privilege issues should be addressed in the initial discovery conference, and the draft amendments remind counsel to do so. Two aspects of the draft amendment to Rule 26(f) merit particular attention.

First, items (1) and (3) are not limited in impact to electronic discovery. Item (1) addresses spoliation generally and applies equally, for example, to the tangible product at issue in a products liability action as to electronically stored design data for that product. Item (3) governs the commonly-entered agreement that inadvertent production of a privileged document will not effect a waiver as well as the more esoteric agreements that permit the intentional disclosure of potentially-privileged electronic data without waiver.

Second, note that item (2) assumes that a party may specify a particular format when requesting production of electronically stored information. Proposed Rules 34(b) and 45(a)(1) confer this right on the requesting party.

Initial Court Conference (Rule 16(b)). Parallel to Rule 26(f), proposed Rule 16(b) adds two optional topics for the scheduling order entered at the initial pretrial conference:

• ‛provisions for disclosure or discovery of electronically stored information,“ and
• adoption of the parties' agreement for protection against waiving privilege.“

This should be uncontroversial. Electronic discovery and non-waiver agreements are important subjects, and reminding the court (like counsel) to consider them at the inception of the litigation is prudent.

‛Reasonably Accessible“ Electronic Information (Rule 26(b)(2)). Rule 26(b)(2) would be amended to permit a party to object to a discovery request that calls for electronically stored information which is not ‛reasonably accessible,“ requiring a motion to compel to obtain the data:

A party need not provide discovery of electronically stored information that the party identifies as not reasonably accessible. On motion by the requesting party, the responding party must show that the information is not reasonably accessible. If that showing is made, the court may order discovery of the information for good cause and may specify terms and conditions for such discovery.

This proposal has at least four important features.

First, it focuses solely on access — whether electronically stored information is ‛reasonably accessible.“ There are many other costs and burdens associated with electronic discovery apart from inaccessibility — most prominently, those generated by the sheer volume of data, which often are in active use and fully ‛accessible.“ To the extent that current Rule 26(b)(2)(iii) is deemed adequate to address all issues concerning the burden and expense of electronic discovery other than accessibility, it raises the question why a new provision is necessary to deal solely with accessibility. One answer could be that a different standard is needed to decide burden issues raised by inaccessibility. As discussed below, however, it does not appear that any different standard is being introduced.

Second, the new standard— ‛reasonably accessible“ — is presumably different from the criteria currently contained in Rule 26(b)(2)(iii) or there would be no reason to introduce it. What this difference may be is not explained in the text of the Rule, and it is not entirely clear from the Committee Note. A persuasive argument can be made, based on the Note, that the two standards are substantially the same, and that there is no need for the new locution.

Currently, Rule 26(b)(2)(iii) excuses production when ‛the burden or expense of the proposed discovery outweighs its likely benefit, taking into account the needs of the case, the amount in controversy, the parties= resources, the importance of the issues at stake in the litigation, and the importance of the proposed discovery in resolving the issues.“ Litigants frequently collapse these factors into an objection that responding to a particular discovery request would be ‛unduly burdensome,“ echoing the ‛undue burden“ language of Rules 26(c) and 45(c)(1).

Compare current Rule 26(b)(2)(iii) with the discussion of the ‛reasonably accessible“ standard contained in the draft Committee Note. The Note provides a series of examples that focus on ‛burden“ and ‛expense,“ and largely makes ‛not reasonably accessible“ sound the same as ‛unduly burdensome:“

Many parties have significant quantities of electronically stored information that can be located, retrieved, or reviewed only with very substantial effort or expense. For example, some information may be stored solely for disaster-recovery purposes and be expensive and difficult to use for other purposes. Time-consuming and costly restoration of the data may be required and it may not be organized in a way that permits searching for information relevant to the action. Some information may be ‛legacy“ data retained in obsolete systems; such data is no longer used and may be costly and burdensome to restore and retrieve. Other information may have been deleted in a way that makes it inaccessible without resort to expensive and uncertain forensic techniques, even thought technology may provide the capability to retrieve and produce it through extraordinary efforts. Ordinarily such information would not be considered reasonably accessible.

The Note consistently seems to equate ‛not reasonably accessible“ with ‛substantial effort and expense,“ clouding any distinction between the new standard and current Rule 26(b)(2)(iii):

Whether given information is ‘reasonably accessible“ may depend on a variety of circumstances. One referent would be whether the party itself routinely accesses or uses the information. If the party routinely uses the information — sometimes called ‛active data“ — the information would ordinarily be considered reasonably accessible. The fact that the party does not routinely access the information does not necessarily mean that access requires substantial effort or cost.“

Based on the contents of the Committee Note, ‛not reasonably accessible“ does not sound different from ‛unduly burdensome.“ If there is no substantial difference, introduction of the phrase may breed confusion in typical electronic discovery cases.

Consider, for example, the objection of a responding party who complains of the burden and expense of: (i) accessing back-up tapes, legacy data, and deleted data, and (ii) retrieving, converting and producing millions of pages’ worth of electronic data in current use (and thus ‛accessible“) from around the world. As far as the responding party is concerned, time and money are the issue with respect to both prongs of this objection. Is the standard different for each? If so, how? If not, why introduce a new phrase? Undue burden seems to capture everything. If it does not, substantially more guidance is required in the Rule and the Note to make it clear why not.

The draft Advisory Committee Note does provide valuable insights in its explication of the accessibility issue. Ultimately, it recognizes that case law is necessary to give shape to the principles that animate the proposal. That case law is currently developing without the nominally different standards governing electronic discovery ‛accessibility“ issues and those governing all other discovery issues (including all other electronic discovery issues). There seems to be every reason to expect that the case law will continue to evolve if the ‛reasonably accessible“ proposal is not adopted.

Third, the proposal’s exclusive focus on access is a product of the present state electronic information storage. Just as the current proposals are deleting references to ‛phonorecords,“ it may be that this proposal will soon be dated.

Fourth, this proposal creates a two-tiered approach to deciding ‛accessibility“ issues, which has the virtue of procedural clarity. Electronically stored information that is ‛reasonably accessible“ is subject to discovery through a simple request. Electronically stored information that is not ‛reasonably accessible“ is discoverable only by court order on a motion to compel production. On the motion to compel, the responding party carries the burden of proving that the information is ‛not reasonably accessible.“ (Of course, absent discovery on that issue, only the responding party has knowledge of the relevant facts.) If the responding party convinces the court that the information is not ‛reasonably accessible,“ the requesting party must show ‛good cause“ to obtain an order requiring production.

As a matter of procedure (putting aside the new, ‛reasonably accessible“ standard), this change to a two-tiered approach is more apparent than real, but the amendment nicely clarifies respective burdens. Under current practice (and as regards all types of discovery), there is in effect a two-tiered approach — a motion to compel is always required whenever a responding party objects to production on grounds of undue burden or expense. On the motion to compel, once the movant has demonstrated sufficient relevance, the responding party carries the burden of proving that discovery should be limited under Rule 26(b)(2) or 26(c). The new rule crystallizes current practice and clarifies, with the new ‛good cause“ standard for overcoming presumptive non-production, an important part of the calculus currently employed by the courts in deciding whether, or on what terms, to grant discovery.

Given the desirable clarity of this aspect of the amendment, one is tempted to ask why the Committee limits its reach to disputes over the accessibility of electronic discovery. It would be a useful paradigm for deciding Rule 26(b)(2) (and Rule 26(c)) issues generally, regardless of whether the ‛reasonably accessible“ locution survives the amendment process.

Claw-Back of Privileged Information (Rule 26(b)(5)). The proposed amendment to Rule 26(b)(5) would renumber the existing provision Rule 26(b)(5)(A) (Privileged Information Withheld), and add a new Rule 26(b)(5)(B):

Privileged information produced. When a party produces information without intending to waive a claim of privilege it may, within a reasonable time, notify any party that received the information of its claim of privilege. After being notified, a party must promptly return, sequester or destroy the specified information and any copies. The producing party must comply with Rule 26(b)(5)(A) with regard to the information and preserve it pending a ruling by the court.

The most notable features of this proposal are: (1) it is not limited to instances of inadvertent production; (2) it extends beyond electronic discovery to all documents and things; and (3) it does not address whether a waiver of privilege has been effected (in fairness, it cannot do so because 28 U.S.C. § 2074(b) bars the Supreme Court from adopting a rule that modifies a privilege without affirmative Congressional approval). This is quite significant because, in some jurisdictions, the fact of production of privileged information ordinarily effects a waiver, regardless of intent. This draft Rule requires the return of information even in such jurisdictions, subject to subsequent decision of the court.

Consider four different scenarios to which the proposed amendment would apply:

(1) The producing party grants the requesting party access to a computer database (or other voluminous set of documents or data) and allows the requesting party to review it first to identify what it wants produced. Only then does the producing party conduct a privilege review of the data that the requesting party wants produced, identifying those documents it deems privileged from among those sought by the requesting party. This approach, at a minimum, requires a court order to avoid waiver. Even with a court order, a waiver may have been effected as to third parties in other actions.

(2) The producing party inadvertently produces a privileged document as part of a large production. Whether a waiver has been effected depends on the circuit.

(3) The producing party intentionally produces a document after review by counsel who did not appreciate that the document was privileged. Again, whether a waiver has been effected depends on the circuit and proof of inadvertence.

(4) The producing party demands back a damaging document claiming that it is privileged even though there is no basis for the claim apparent on the face of the document. The Rule applies to this scenario, subject to a subsequent judicial determination.

The proposal requires the recipient of the information to ‛return, sequester or destroy the specified information and any copies,“ and the producing party must ‛preserve it pending a ruling by the court.“ In other words, the recipient may not present the documents to the court for decision, and argue from their terms. That is not optimal. The terms of the document may matter. The rule should permit the requesting party to present the document to the court promptly after the request for return is made.

Interrogatories (Rule 33). Rule 33 would be amended to make it clear that a party may answer an interrogatory by referring to specific ‛electronically stored information,“ just as it may with any other business records. The draft does so by providing that ‛electronically stored information“ is included within the definition of ‛business records.“ It is interesting that Rule 33 expands the phrase ‛business records“ to encompass ‛electronically stored information“ while, at the same time, draft Rule 34 carves ‛electronically stored information“ out of the definition of ‛document.“

Document Requests (Rule 34). Rule 34 would be amended in several ways:

First, it expressly distinguishes between electronically stored information and ‛documents.“ Rule 34(a) would read:

Any party may serve ... a request (1) to produce and permit the party making the request ... to inspect, and copy, test, or sample any designated electronically stored information or any designated documents (including writings, drawings, graphs, charts, photographs, sound recordings, imagesphonorecords, and other data or data compilations in any mediumCfrom which information can be obtained, translated, if necessary, by the respondent through detection devices into reasonably usable form) , or to inspect, and copy, test, or sample any designated tangible things....

This proposal is problematic. As a matter of drafting, it is unclear whether the parenthetical refers only to the immediately-preceding term ‛designated documents“ or to the broader phrase ‛electronically stored information or any designated documents.“ The parenthetical includes ‛sound recordings, images, and other data or data compilations in any medium.“ These clearly subsume electronic data.

If these items are ‛documents,“ then ‛document“ includes at least some types of ‛electronically stored information,“ blurring any distinction. If these items are not ‛documents“ — because the parenthetical refers to the entire phrase ‛designated electronically stored information or any designated documents“ — then the Rule makes no distinction between the two. At this point, the benefit of creating what the rulemakers believe to be an ‛express distinction“ between ‛electronically stored information“ and ‛documents“ appears to be lost.

The proposed distinction between ‛electronically stored information“ and ‛documents“ appears to be a solution in search of a problem. Present practice is to include electronic data within the definition of ‛document“ in Rule 34 requests for production and Rule 45 subpoenas. Some local rules do so, negating any need for the parties’ document requests to include any such definition. See, e.g., S.D.N.Y. and E.D.N.Y. Local Rule 26.3(c)(2). The draft Advisory Committee Note observes that ‛[i]t is difficult to say that all forms of electronically stored information fit within the traditional concept of a .document.“ Whatever the ‛traditional concept“ might be, it is not the concept entertained by bench or bar for the past decade or so.

If the amendment to Rule 34(a) is adopted, document requests must be re-written so that ‛electronically stored information“ is distinctly requested. Otherwise, ‛documents“ may not be read as requesting electronic data. Since it takes years for practice to catch up to changes to the rules, this distinction could be expected to wreak some havoc as practitioners gradually become aware of the need to change their form definitions.

The Advisory Committee Note provides that: ‛[A] Rule 34 request for production of ‘documents’ should be understood to include electronically stored information unless discovery in the action has clearly distinguished between electronically stored information and ‘documents.’“ The initial, independent clause appears incompatible with the ‛express distinction“ that the rulemakers are attempting to articulate. The subsequent, dependent clause, optimistically, does not mean that, if one party is aware of the new rule and makes the distinction in its requests, the other party is burned for failing to do so. It would be preferable to define documents to include ‛electronically stored information“ rather than to carve them out separately.

Second, Rule 34(b) would be amended to provide that ‛[t]he request may specify the form in which electronically stored information is to be produced,“ and the response may include an ‛objection to the requested form for producing electronically stored information.“ This codifies present practice. Query whether, as a drafting matter, these sentences should be limited to electronically stored information. It is common for document requests to specify, e.g., an electronic production format for hard copies. This provision presumably should not be read as limiting that practice.

A new Rule 34(b)(ii) would be added to provide that:

if a request for electronically stored information does not specify the form of production, a responding party must produce the information in a form in which it is ordinarily maintained, or in an electronically searchable form. The party need only produce such information in one form.

Note that the first alternative in the first sentence is that the information must be produced in ‛a“ form — not ‛the“ form — in which the information is ordinarily maintained. This affords the responding party some latitude. If the data are maintained only in a malleable form (e.g., Word or Word Perfect text, Excel numbers), production in that form is often undesirable. Such data are more difficult to authenticate — or even maintain the integrity of —because they are subject to easy alteration (intentionally or unintentionally). An electronically-searchable image (e.g., searchable PDF) has many practical advantages.

The last sentence of proposed Rule 34(b)(ii) provides that, absent agreement or court order, information need be produced in only ‛one form.“ Electronic information and hard copies are not equivalent in the information they convey. Sometimes, electronic data provide more information than hard copies — e.g., reflecting edits or the date/time of creation, editing, receipt or opening. Sometimes, hard copies provide more information than electronic data — e.g., association by rubber bands, staples or clips, inclusion in a common folder, or other indicia of file structure. Requesting parties should bear in mind that producing parties are making judgments when deciding on the form of production.

Sanctions Safe Harbor (Rule 37(f)). A new Rule 37(f) includes a safe harbor from sanctions relating exclusively to electronically stored information:

(f) Electronically Stored Information. Unless a party violated an order in the action requiring it to preserve electronically stored information, a court may not impose sanctions under these rules on the party for failing to provide such information if:

(1) the party took reasonable steps to preserve the information after it knew or should have known the information was discoverable in the action; and

(2) the failure resulted from loss of the information because of the routine operation of the party=s electronic information system.

There are at least three important aspect of this proposal.

First, this safe harbor appears to be illusory. The proposal affords a safe harbor only from sanctions under the Federal Rules of Civil Procedure. But there is no provision in the Federal Rules of Civil Procedure authorizing the imposition of sanctions — absent violation of a court order — for failing to produce information despite all reasonable efforts to do so. Rule 11 does not apply to discovery (Rule 11(d)); Rule 26(g) applies only to written discovery requests, responses and objections (Rule 26(g)(2)); and Rule 37, in relevant part, authorizes sanctions only for: (1) noncompliance with a discovery order (Rule 37(b)), (2) failure to serve a written response to a Rule 34 request for production (Rule 37(d)), and (3) failure to make any Rule 26(a) disclosure or amend a prior response to discovery as required by Rule 26(e)(2).¹ Nor does the draft Advisory Committee Note suggest that any judge has ever imposed sanctions for the behavior covered by the proposed safe harbor.

Second, the exclusive focus of paragraph (2) on routine deletion of data is problematic. This proposal suggests that, if a party fails to provide electronic information for any reason other than ‛routine operation of the party=s electronic information system,“ sanctions may be appropriate despite the fact that the party has acted reasonably. There are many other reasons why data disappears over time. Suppose that, early in the litigation, all relevant data have been housed on special servers but the servers are destroyed through no fault of the producing party (Hurricane Frances). That certainly is not sanctionable.

Third, this proposal would change corporate behavior. Parties who are routinely sued will accelerate the routine deletion of data. Is this behavior that should be encouraged?²

In a footnote, the Advisory Committee is also soliciting comments for an alternative safe harbor that would excuse negligence and require the producing party only to have refrained from violating a court order ‛intentionally or recklessly fail[ing] to preserve the information.“ This would reverse a two decades-old trend of requiring reasonableness of parties in their litigation-related conduct; would not address either the second or third points above; and would have the additional demerit of potentially requiring hearings on the subjective bad faith of the producing party, as bad faith is ordinarily considered a matter of fact.

Form 35 (Report of Parties’ Planning Meeting). Form 35 would be updated to parallel changes to Rule 26(f) are to form part of the discovery plan presented to the court before the initial pretrial conference.

Subpoenas (Rule 45). Rule 45 would be amended to capture the principal changes to Rules 26 and 34.

The draft can be viewed at http://www.uscourts.gov/rules/comment2005/CVAug04.pdf. The Advisory Committee seeks input from bench and bar on all of these proposals.