ParfumGigi@aol.com

By Leonard Deutchman
Pennsylvania Law Weekly
April 12, 2007

The recent changes to the Federal Rules of Civil Procedure require litigants to have early discussions to address issues relating to the disclosure and production of electronically stored information (ESI). In the last several articles I have used preparing for the "meet and confer" as a model for understanding and addressing your case's e-discovery issues.

Parties at the meet and confer must address four issues: 1) mandatory disclosure of all ESI under Rule 26(a)(1)(B); 2) what will not be produced, or produced only if the requesting party bears some or all of the cost of production, because the ESI is not "reasonably accessible because of undue burden or cost" under Rule 26(b)(2)(B); 3) the form of production under Rule 34(b); and, 4) "claw back" agreements under Rule 26(b)(5) to provide for the return of privileged documents inadvertently disclosed.

In the last several articles we have discussed in detail the steps of e-discovery production, from inventorying, preserving and gathering ESI, to culling, searching, processing and producing it as e-discovery, as part of our discussion of how to address the first three meet and confer issues. In this month's column, we will discuss the fourth issue -- guarding against inadvertent disclosure of privileged materials -- as well as some interesting questions raised by the new "undue burden or cost" analysis called for by Rule 26. Next month, we will wrap up our discussion of the new rules with a hypothetical case that will call upon us to draw on all of the legal and technical discussions in this series of articles.

The proliferation of computers as a means of business and personal communication and storage of information led to a problem in civil litigation: discovery production took forever. Several factors contributed to the delays, the most basic of which was that the proliferation of computers led to a vertiginous increase in the volume of data to be reviewed by counsel prior to disclosure.

One of the goals of the Rules changes was to speed up review, but the Rules Committee recognized that instances of mistaken disclosure of privileged data have increased commensurately with the increase in data to be reviewed, speeding up the review process would further increase the likelihood of mistaken disclosure, and without any means to retrieve mistaken disclosures, counsel would have to review each file in excruciating detail, thereby insuring that e-discovery discovery production would remain painfully slow.

The Committee's solution was "claw -ack" agreements, which allow for the responding party who inadvertently discloses privileged documents to retrieve them from the requesting party. The procedure promises to speed up discovery production by guarding against dire consequences for inadvertent disclosure. Rule 16(b)(6) foresees that agreements for the return of privileged documents inadvertently disclosed will be honored.

Rule 26(f)(4) contemplates that the Pretrial Conference will "address any issues relating to claims of privilege or protection as trial preparation material including -- if the parties agree on a procedure to assert such claims after production -- whether to ask the court to include their agreement in an order." Rule 26(b)(5)(B) sets forth the procedure by which a party who believes he or she has inadvertently disclosed privileged information may retrieve it.

The suggestion that claw-back agreements can help speed up discovery production has been met with wide criticism and skepticism in the close to two years between the publication of the first draft of the Rules changes for comment and the adoption of the changes in final form in December 2006. The well-regarded Hopson v. Mayor & City Council of Baltimore, 232 F.R.D. 228 (D.Md. 2005), a case decided while the Rules changes were still pending, explored the shortcomings of claw backs in detail.

After reviewing the factors that have led to the need for such agreements, the court noted that critics questioned whether such agreements are effective, or even enforceable against third parties. The court noted that courts have developed three different approaches to waiver: 1) the "strict accountability" approach of the First and Federal Circuits, under which once confidentiality is breached it can never be restored; 2) the "lenient/'to err is human'" approach of the Eighth Circuit and some District courts, which excuses inadvertent waiver and finds waivers only when the waiving party does so intentionally or is grossly negligent; and 3) a "balancing test" in which each matter is judged on a case-by-case basis. Since under F.R.Civ.P. 501, waiver issues in diversity cases are to be analyzed under the principles of state law, the meaning of any claw-back agreement would depend upon the jurisdiction in which it is enforced.

The court proposed a solution to this problem by adopting the standard of waiver in the agreement and having the agreement made under the scheduling order under Rule 16, a protective order under Rule 26(c), and/or a court-ordered discovery management order under Rule 26(b)(2). The Rules Committee, however, responding in part to Hopson, recognized that claw-back provisions under Rule 26(f) (4), even embodied in a court order, cannot work without a change to the Federal Rule of Evidence.

It thus proposed F.R.E. 502, under which no waiver of privilege would occur if a "voluntary disclosure" was made, the discloser was "itself privileged or protected," the disclosure was "inadvertent" and made "during the discovery in federal or state litigation or administrative proceedings," and if the holder of the privilege or protection "took reasonable precautions to prevent disclosure and took reasonably prompt measure, once the holder knew or should have known of the disclosure, to rectify the error." Proposed Rule 502 is necessary for two reasons. As has already been discussed, the different approaches to waiver taken by the various Circuits must be reconciled.

Moreover, for the Rule to work it must apply to state court proceedings; otherwise, materials "clawed back" under federal agreements could be used in state court, and so such agreements would provide little real protection. For the federal rule to apply to state courts, Congress must enact it into law, unlike with the Rules of Civil Procedure, where changes will go into effect unless Congress vetoes them. As the Committee Note to Rule 26 discusses, Congressional authorization is required because: 1) it is specifically mandated by 28 U.S.C. §2074(b); and, b) it applies to state court proceedings and so must be enacted as law under the Commerce Clause, as was the case with Class Action Fairness Act of 2005, 119 Stat. 4, PL 109-2.

The foregoing discussion should make it clear that reliance upon a claw-back provision would be misplaced. Having said that, it is still probably best to agree to one at the meet and confer and have it made part of the discovery order. Having the agreement may make the difference between little and no relief, but it is always best to obtain the little relief the order may supply, and there is no downside to getting it.

In previous columns, we have discussed how, under Rule 26(b)(2)(B), a responding party may object to producing ESI or ask the requesting party to bear some or all of the cost of production on the ground that the ESI is not "reasonably accessible because of undue burden or cost." We discussed application of Rule 26 when we explored the step and costs of data gathering, processing and production.

There are some interesting issues arising from Rule 26(b)(2)(B)'s particular language that I have not seen addressed elsewhere. Whether seeking or producing e-discovery, prepared litigators should be aware of these issues.

Rule 26(b)(2)(B) tried to settle a question raised in case law and in comments to the initial drafts of the Rule: what does "inaccessible" mean? If it meant "technically" inaccessible, then data on backup tapes, for example, would be "accessible," as would be data on dozens of servers and personal computers across continents, even if it cost hundreds of thousands of dollars to have the tapes restored.

What constitutes an "undue" burden is subjective; the comedian Steven Wright has observed that "everything is walking distance if you have the time." To help define what is "not reasonably accessible," the drafters added the modifying phrase, "because of undue burden or cost," thereby making cost the key element of accessibility. By restricting the analysis of accessibility to one factor -- cost, in financial and other terms - presumably, over time, case law will establish guidelines for determining what burdens and costs are "undue."

One interesting consequence of defining accessibility in terms of cost was probably not considered by the drafters: some data which would not be accessible to the user would be accessible in e-discovery production, while other data that would be accessible to the user would not be accessible in e-discovery production. Deleted files are a perfect example of ESI inaccessible to the user but accessible when producing e-discovery. If data is gathered forensically, forensic tools will recover deleted files as easily as stored files, assuming the deleted files retain their "file attributes" (i.e., the metadata which tells the operating system that a file is a file).

Making the highly reasonable assumptions that it costs $250-$500 to create a forensic image of a hard drive and an additional $1,000 simply to cull the drive of all human readable files of interest (e-mails, e-documents, images), both stored and deleted, the cost of recovering deleted files as well as active ones is $1,500 a hard drive. Is that cost an "undue" one? It may depend upon how many drives are at stake: if fifty need to be imaged and culled, the cost is now $75,000. By the same token, it may also depend upon what is at stake in the litigation: if fifty drives need to be imaged, potential damages would most likely run well into the millions, and so $75,000 to gather all files may not seem "undue."

Regardless of how a court would decide, it is important to understand that the average user's understanding of what data is "inaccessible" has in Rule 26(b)(2)(B) been supplanted by a cost calculation. Since most attorneys are "average users" of computers, another way of stating the previous sentence is that the Rules changes will force you to think of the accessibility of data in terms of the cost of using data gathering tools, not in the terms that you as a computer user would think when contemplating whether you could get access to data on a hard drive.

Now consider when deleted files have lost their file attributes. Fragments of those files can be found in unallocated space on the hard drive using a keyword search, but each one of those fragments would have to be examined individually to determine its value. Such analysis is very time-consuming and thus costly. Deleted files recovered in this matter would in almost all cases deemed "not reasonably accessible," even though technically they can be recovered.

If some data thought by users to be "inaccessible" can be rendered accessible under Rule 26, so can the opposite happen. Consider some kinds of "metadata." Metadata ranging from the "MAC" times -- the date of file Creation and when it was last Accessed and Modified -- to comments and track changes are accessible to the user, but costly to gather and produce. As we discussed in previous columns, data must be gathered forensically to preserve metadata such as MAC times, and the cost of extracting, producing and reviewing comments and track changes for privilege can be very high. Thus, data that within an individual file can be readily accessible to the user can be very costly to preserve, gather, produce and review, and so "not reasonably accessible because of undue burden or cost," when it is not the user who needs access to the data but a third party, and when thousands of files, as opposed to one or two, are involved.

Through the last several columns we have used the model of preparing for the meet and confer to discuss the legal and technical issues attendant to requesting and producing e-discovery.

We have seen that addressing these legal issues requires that a litigator know a client's IT infrastructure, where all repositories of ESI are, who can produce them, how they can be obtained, what ESI needs to be preserved, what searched and what produced, how to ask for e-discovery and how production form can dictate what content is produced, how the cost of every step of e-discovery informs and influences production, and how the new Rules afford a limited protection against inadvertent waiver of privilege.

The litigator who understands how ESI is preserved, gathered, searched and produced understands how the Rules changes must be interpreted in light of the technical realities of ESI, and is in the best position to use those rules to his or her advantage.