Determining the Scope of Legal Holds: Waypoints for Navigating the Road Ahead

Editor’s Note: This is the second of two articles based on the author’s October 2007 research project for the ARMA International Educational Foundation (AIEF). The complete study, Legal Holds for Anticipated Litigation: New Case Developments to Determine Triggering Events and Scope of Production may be downloaded at www.armaedfoundation.org. The first article, “Braking for Legal Holds:How to Read the Signals,” appeared in the January/February 2008 issue of IMJ.

John J. Isaza, Esq.

Since the December 2006 revised U.S. Federal Rules of Civil Procedure went into effect, organizations have been struggling more than ever to understand litigation, or legal, holds, and, in particular, the scope of what to put on hold. Under the revised rules, they must hold from destruction all relevant electronically stored information (ESI).

Case law has shown that the consequences for untimely destruction of ESI can be dire for any organization, irrespective of size. Unfortunately, neither case law nor legal scholars have provided much guidance on what could be considered relevant, so organizations are left to make that critical determination themselves. Regrettably, the question of scope boils down to a case-by-case determination. An overview of general principles to determine scope, followed by examples of cases dealing with the degree of scope, may provide organizations some guidance on how to proceed.

General Principles Determining Scope of Legal Holds

Once the duty to preserve documents for a litigation hold is triggered, an organization’s primary obligation is to determine the scope of what to hold and so decide what documents and information it must preserve. Generally speaking, courts – such as in Mosaid v. Samsung, citing Scott v. IBM Corp – have held that the duty to preserve extends to what the organization “knows, or reasonably should know, will likely be requested in reasonably foreseeable litigation.” Samsung v. Rambus said that simply instructing employees to “look for things to keep” or not to destroy relevant documents is insufficient. In other words, once a party reasonably anticipates litigation, it must suspend its routine document retention/destruction policy to ensure that relevant documents are preserved, as in In Re NTL, Inc. Securities Litigation, et al., citing Zubulake vs. UBS Warburg. In practice, determining the appropriate scope of the duty to preserve can be difficult, particularly with respect to electronic data.

Two main sources that outline principles for determining the scope of legal holds are Sedona Principles 2007: Best Practices Recommendations & Principles for Addressing Electronic Document Production and the U.S. Federal Rules of Civil Procedure.

The Sedona Principles

The Sedona Principles and related guidelines are the result of a think tank of attorneys, legal academicians, and judges looking for solutions to electronic discovery issues. Until the adoption of the 2006 revised rules, the Sedona Principles, along with the Zubulake case, were the most cited authorities on the issue.

Sedona Principle No. 8 attempts to address the types of electronic documents that must be preserved. It states:

The primary source of electronic data and documents for production should be active data and information purposely stored in a manner that anticipates future business use and permits efficient searching and retrieval. Resort to disaster recovery backup tapes and other sources of data and documents require the requesting party to demonstrate need and relevance that outweigh the cost, burden, and disruption of retrieving and processing the data from such sources.

Furthermore, under Sedona Principle No. 9, the panel states that “absent a showing of special need and relevance, a responding party should not be required to preserve, review, or produce deleted, shadowed, fragmented, or residual data or documents.”

Finally, Sedona Principle No. 12 gives guidance regarding metadata, stating that “[u]nless it is material to resolving the dispute, there is no obligation to preserve and produce metadata absent agreement of the parties or order of the court.” This principle may fly in the face of some case law, but at least it sets an approach for metadata. Even so, there is still ambiguity regarding the so-called materiality discussed in Principle No. 12.

At present, The Sedona Conference is circulating for public comment a document entitled The Sedona Conference Commentary on Legal Holds, August 2007, available at www.thesedonaconference.org. Even this document provides only general guidelines, and no specifics are addressed:

In determining the scope of information that should be preserved, the nature of the issues raised in the matter, experience in similar circumstances and the accessibility of the information are factors that may be considered.

The Revised Federal Rules of Civil Procedure

The Sedona Principles essentially are codified in the revised rules, with the exception of Principle No. 12 regarding metadata.

Specifically regarding scope, Federal Rule 26(a) requires parties to disclose “a description by category and location of … electronically stored information.” This includes, of course, identification of sources under Rule 26(b)(2) that are “not reasonably accessible.” The rule exempts the party from having to produce discovery from such inaccessible sources unless the adversary moves to compel.

Under Rule 26(b):

A party need not provide discovery of electronically stored information from sources that the party identifies as not reasonably accessible because of undue burden or cost. On motion to compel discovery or for a protective order, the party from whom discovery is sought must show that the information is not reasonably accessible because of undue burden or cost. If that showing is made, the court may nonetheless order discovery from such sources if the requesting party shows good cause, considering the limitations of Rule 26(b)(2)(C). The court may specify terms and conditions for the discovery.

Despite the attempts to codify and explain the scope of legal holds for litigation, the practical application of such guiding principles leaves lingering doubts about the precise scope. Legal holds and the duty to preserve information can potentially cripple an organization, and the courts and other experts are keenly aware of the dilemma. For instance, Sedona Principle No. 5 recognizes that “it is unreasonable to expect parties to take every conceivable step to preserve all potentially relevant data.”The revised rules echo this sentiment with the parameters they have set for non-discovery of “inaccessible” information.

Case Examples Dealing With the Scope of Litigation Holds

Because guiding principles on scope are still vague and subject to much dispute, organizations must revert to case-by-case analysis to determine the scope of the legal holds. Four scope scenarios are discussed covering such commonly sought records as

  1. Key player documents
  2. Employment decision documents
  3. Computer networks and hard drives
  4. Server log data or random access memory

1. Key Players and Their Documents

In Consolidated v. Alcoa, Consolidated contended that Alcoa deleted relevant e-mails and other electronic data for years after it reasonably anticipated litigation – and after the litigation actually commenced – in violation of a duty to preserve all potentially relevant evidence. The primary issue was whether Alcoa took all reasonable steps to preserve the necessary evidence. The court from the outset acknowledged that Alcoa, upon recognizing the threat of litigation, did not have to preserve “every shred of paper, every e-mail or electronic document, and every backup tape … [T]he duty to preserve extends to those employees likely to have relevant information, i.e. the ‘key players’ in the litigation.”

Consolidated issued a demand letter to Alcoa in November 2002, almost a year before the lawsuit was filed. Alcoa’s counsel met with and participated in telephone conferences with four individuals to establish a document hold in conjunction with the demand letter. One of several spoliation claims that later arose pertained to retention of records for other individuals beyond the original four. In attempting to explain why it did not preserve the e-mails of other individuals, Alcoa contended they were plant-level employees with minimal interaction with the corporate-level key players. Alcoa argued that for every e-mail sent and received by plant-level individuals, there was a corresponding key player recipient. Thus, Alcoa claimed, there was no reason to preserve non-key player records such as those involving the plant-level individuals. In addition, Alcoa asserted that the only pertinent information a certain plant employee had with respect to this case related to his historical knowledge of the facility, which could be obtained only through deposition testimony regarding his memory of events, not through contemporary documents or e-mails.

The court disagreed with Alcoa’s view about the plant employee. Statements by Alcoa indicated that he was one of the key players likely to have relevant information in his e-mails. Alcoa had a duty to preserve his e-mails from the time the litigation became reasonably anticipated, even if he was a plant-level employee who was not directly involved in the ongoing environmental investigations underlying this case.The court also disagreed with Alcoa’s argument that any information contained in his e-mails had actually been preserved because his e-mails to and from their designated key players had been preserved. Since the Court construed the plant employee to be a key player as well, all e-mails sent and received by him, not just those involving the designated key players,would be relevant to this matter and should have been preserved. To make matters worse, Alcoa’s initial disclosure in February 2005 identified approximately 100 individuals as being “likely to have discoverable information that Alcoa may use to support its claims and defenses.” The Court found that instructing only 11 Alcoa employees to abide by the litigation hold in May 2005 was inadequate to fulfill its preservation obligations.

Thus, under Consolidated v. Alcoa, organizations must pay particular attention to the employees deemed key players, especially if during the onset of litigation counsel errs on the side of caution by naming as many individuals as possible to be potential witnesses at trial. See also In Re NTL, Inc. Securities Litigation; Gordon Partners, et al. v. Blumenthal, et al. (in class action for securities violations court granted discovery sanctions for destruction of evidence including e-mails of approximately 44 of the defendants’“key players”).

2. Employment Decision Documents

In Scott v. IBM, the plaintiff brought suit against his former employer IBM, alleging that it discriminated against him on the basis of his race, age, and disability when he was laid off. At issue were flip charts and other documents used as part of the decisionmaking process for a large layoff at IBM. The flip charts were the sole documentation of how managers intended to implement the lay-off directives. However, the flip charts used to arrive at the layoff determination were destroyed the day after the meeting. This denied the plaintiff the ability to discover exactly what the managers wrote down during the decision-making process. Significantly, there was no original documentation as to nine other potential layoff candidates; nor was there documentation as to where Scott ranked on this list in relation to the others.

At trial, Scott argued that IBM committed evidence spoliation by destroying several key pieces of evidence in this case. The court found that IBM’s decision to destroy these documents could give rise to a “spoliation inference.” The plaintiff contended the destruction of the flip charts and name ranking deprived him of any objective evidence concerning whether IBM truly followed its internal parameters when deciding to lay him off. The court found that because the flip charts were destroyed, there was little documentation as to what really went on during the decision-making process.

There was no direct evidence that the documents were knowingly destroyed after the plaintiff filed his U. S. Equal Employment Opportunity Commission complaint, nor did they “disappear” after a court order for their production. However, IBM was aware that any documents relating to the lay-off process would be subject to discovery. While litigation was not guaranteed, it could be viewed as reasonably foreseeable. IBM managers knew that Scott, if not others, was protected by federal employment discrimination laws. Scott had made previous claims of race discrimination within IBM, and thus IBM had ample notice that it was discharging a potentially litigious employee when it fired him.

The court deemed that common sense would dictate preserving all helpful documentation when dealing with the discharge of an employee with a litigious history. Because IBM did not do so, the inference was that the destroyed documents were unhelpful to IBM’s case. The court found that IBM should have foreseen that litigation might follow the lay-off and should have foreseen that all documentation concerning the decision-making process could become relevant.

3. Computer Networks and Hard Drives

In St. Andrews Park v. Corps of Engineers, the plaintiffs were the owners of a parcel of property known as the St. Andrews Park Site. The U.S. Army Corps of Engineers (Corps) asserted jurisdiction over the isolated wetlands on the site, and this suit challenged the Corps’ assertion of jurisdiction.

In December 2001, prior to filing the suit, the plaintiffs e-mailed multiple Freedom of Information Act (FOIA) requests under 5 U.S.C. § 552, seeking records related to the site. In a March 2002 letter, plaintiffs’ counsel reactivated the FOIA requests and asked for all file documents, papers, and correspondence, including e-mail communications, relating to the St. Andrews Park Site. The FOIA request extended to any e-mails or electronic communications still stored on intermediate servers used by the Corps to transmit e-mail and to any computer hard drive upon which such e-mails or messages still existed. The Corps failed to search for documents located on individual computer hard-drives of its employees. The Corps stated that the entire network server was searched, specifically including requested e-mail accounts of identified Corps personnel.

But because the defendants did not search individual computer hard drives, the court found that they did not meet their burden of “thoroughly search[ing] for the requested documents where they might reasonably be found.” For more information, see St.Andrews Park v.Corps of Engineers, citing Miller v.United States Dep’t of State; see also Oglesby v. United States Dep’t of Army, which shows an agency cannot limit its search to only one record system if there are others that are likely to turn up the information requested.

Thus, under St. Andrews Park v. Corps of Engineers, individual computer hard drives are subject to search, in addition to network systems. In this case the FOIA requests set in motion the duty to preserve for the defendants. Destruction of the records, by extension, could be construed as spoliation when litigation is anticipated.

4. Server Log Data

In the copyright infringement case Columbia Pictures Industries v. Bunnell, the court addressed the plaintiffs’ discovery motion to compel the defendants to preserve and produce electronic data and their request for evidentiary sanctions. Although the court did not issue sanctions in this case, the preservation duty may create an obligation to modify or suspend features of an electronic information system’s routine operation to prevent the loss of potentially discoverable evidence.

The plaintiffs filed a complaint for copyright infringement in February 2006, asserting that the defendants’ website operation, which downloaded dot-torrent files, involved piracy of the plaintiffs’ copyrighted works. (Dot-torrent files enable a user’s computer application to locate and access content on the Internet.)

Columbia sent notice to the defendants in May 2006 to preserve potentially discoverable evidence but failed to specifically mention or identify server log data (SLD). A website’s server log, if enabled, copies download requests into a log file and also logs that requested files were delivered. The website’s logging function was not enabled – a decision made to make the website more attractive to users who wanted to remain unidentifiable – and the SLD was not affirmatively retained. But the log data passed through the website server’s random access memory (RAM) and was temporarily stored for approximately six hours.

The court evaluated the nature of SLD and reached several conclusions. Absent specific precedent on the issue, the court concluded that data in RAM qualifies as electronically stored information within the scope of F.R.C.P. section 34(a). Specifically, the court found that SLD was electronically stored information because the data is copied temporarily to the RAM while user requests are processed. Second, the court determined that the SLD was within the “possession, custody, or control” of the defendants whether it was routed to the defendants’ server directly or to the defendants’ partner’s server. The defendants at all times had the ability to manipulate at will how the SLD was routed.

The court expressly noted that its ruling should not be read to require litigants in all cases to preserve and produce electronically stored information that is temporarily stored only in RAM – they made mention of the specific facts in the case and the unique nature of the legal claims (i.e., copyright infringement). However, it is certainly possible that data in other cases would fit this court’s criteria for whether to determine that the information was within the scope of a litigation hold. In other words, electronic data temporarily stored in RAM may be within the scope of a litigation hold when it is key and potentially “dispositive” to a case, is otherwise unavailable, and there is no evidence that preserving and producing the data would be unduly burdensome and costly.

Conclusion

The law is far from clear as to what documents to hold in cases of anticipated or contemplated litigation, particularly in an age of increased complexity and reliance on electronic records and storage. Courts and experts are aware of the potentially crippling effect of forcing an organization to put a hold on all potentially relevant information. The question is how far to go in satisfying the revisionist and hindsight view of the courts. The cases surveyed illustrate how courts have ruled in the past and may help organizations set parameters around these thorny issues in the future.

John Isaza, Esq., can be contacted at jisaza@hilawgroup.com.

 From March - April 2008