The Catholic University of America

NACUA Notes (National Association of College and University Attorneys), April 23, 2013. Volume 11, Number 10.




David Kessler, Fulbright & Jaworski L.L.P.

Caroline Mew, Fulbright & Jaworski L.L.P.



Over five years have passed since a NACUANOTE last addressed the topic of e-discovery. [1] Since that time, courts and litigants have had many occasions to wrestle with electronic data preservation and production issues in the context of contentious litigation matters. This Note provides an overview of the general issues relevant to the preservation and production of electronically stored information (“ESI”) during civil litigation discovery and touches upon some of the recent case law that reflects the potential perils of failing to properly prepare for or participate in e-discovery.

Since the Federal Rules of Civil Procedure and the cases decided thereunder tend to be at the forefront of these issues and thus set the standards, this Note will generally concentrate on lessons learned in the context of federal litigation. Most state courts have followed the course set by the federal rules and courts, so the principles discussed apply broadly, but by no means universally. Equally important to a heavily regulated industry like higher education, the e-discovery principles established by federal courts are becoming generally applicable to other legal processes, such as administrative investigations and procedures and audits, so campus counsel would be well served by reviewing these principles and considering their application in the broad context of requests for electronic records.



Brief Review: Preparing for E-Discovery

The 2008 NACUANOTE provided a basic overview of e-discovery, including what constitutes ESI, where ESI can be found, and steps that institutions can take to prepare for e-discovery before litigation even begins. That information remains relevant today. In the 2008 Note, we encouraged institutions to:


Review their records management policies and procedures. [2]


Do everything possible, subject to any pre-existing litigation preservation demands, to limit their inventory of back-up tapes.


Create catalogues of their IT architecture.


Have in place an effective preservation notice policy.


Consider policies to manage, and be actively aware of, the use of developing technology by faculty and employees.


Since 2008, courts have made clear that many of these steps are more than just encouraged, they are required. As such, institutions should be knowledgeable and proactive with regard to these matters before litigation or other legal processes arise.

Best Practices and Emerging Court Guidance for Document Preservation and Production


A. Timing

It is a generally accepted principle that a duty to preserve relevant evidence arises once a party reasonably anticipates litigation. [3] For a potential plaintiff, the duty to preserve will normally arise before a complaint is filed. For example, retention of outside counsel to investigate a potential claim and/or prepare for litigation means that a party reasonably anticipates litigation, and relevant evidence should be preserved. [4] For a potential defendant, the duty to preserve also may arise before service of a complaint. For example, receipt of a letter from a lawyer representing an employee or student that raises specific complaints and affirmatively threatens legal action may trigger a duty to preserve.

The determination of when a party reasonably anticipates litigation is often fact specific, as reflected in the following example cases:


In Scalera v. Electrograph Systems, Inc., 262 F.R.D. 162 (E.D.N.Y. 2009), the plaintiff sued her former employer under the Americans with Disabilities Act (“ADA”) and the New York Human Rights Law for failure to accommodate. Plaintiff sought spoliation sanctions for defendant’s alleged failure to preserve relevant evidence. [5] Plaintiff offered three theories as to when defendant’s duty to preserve arose: (1) after plaintiff fell down steps for which she purportedly had been seeking a railing as an accommodation; (2) after plaintiff hired an attorney and filed for worker’s compensation; or (3) after plaintiff filed her EEOC charge. The court rejected plaintiff’s first two theories. It was a disputed issue whether plaintiff had actually sought a hand railing as an accommodation, and plaintiff’s attorney wrote a letter to defendant’s landlord, not defendant, asserting negligence after the fall and making a claim for her injuries. The court therefore concluded that no duty to preserve arose just because of plaintiff’s fall. The court also rejected plaintiff’s broad argument that an employer should reasonably anticipate a disability discrimination lawsuit each time an employee files a worker’s compensation claim under facts similar to this case. Instead, the court held that defendant’s duty to preserve arose upon receipt of plaintiff’s EEOC charge. Plaintiff also argued that defendant had a duty to retain all documents relating to her employment, disability, and requests for accommodation pursuant to ADA regulations. The court agreed that, if plaintiff had made a request for installation of a handrail as a disability accommodation, documents evidencing the request should have been preserved for one year in accordance with ADA regulations. [6]


In Siani v. State University of New York at Farmingdale, 2010 WL 3170664 (E.D.N.Y. Aug. 10, 2010), the pro se plaintiff had settled a lawsuit with his university employer in early November 2007. But in March 2008, Siani sent a letter to the President and the Provost and Vice-President of Academic Affairs alleging further discriminatory acts and stating that he would be pursuing various paths of investigation. Less than a week later, Siani made a New York Freedom of Information Law (FOIL) request seeking electronic e-mail activity logs for several individuals. A separate dispute arose with regard to the FOIL request. In early 2008, the defendants hired an outside law firm “to obtain legal advice in connection with issues arising under [FOIL] and employment issues, including those related to [Siani].” In late June 2008, Siani filed a complaint with the EEOC, and defendants were notified of the claim in July 2008.

The court held that the duty to preserve evidence arose in early 2008, when outside counsel was retained by defendants, particularly because defendants were asserting work product protection over documents created at that time. In light of defendants’ prior litigation experience with Siani, his March 2008 letter also put defendants on notice.


Scalera illustrates that not every adverse event or threat of litigation necessarily triggers the duty to preserve—in that case, the plaintiff’s pre-litigation “threats” were not directed at defendants and were not specifically related to her eventual disability discrimination claim. But in Siani, the plaintiff, who had already proven to be litigious, had raised specific pre-litigation claims against the university before filing suit, giving rise to a duty to preserve. If an institution intends to assert work product protection over materials prepared in responding to such claims, then it is anticipating litigation and should take reasonable steps to preserve relevant evidence. [7]

B. How to Preserve

Once the duty to preserve arises, an informed decision should be made as to a reasonable and cost-effective approach to preserving evidence for the particular dispute at hand. [8] “Whether preservation or discovery conduct is acceptable in a case depends on what is reasonable, and that in turn depends on whether what was done—or not done—was proportional to that case and consistent with clearly established applicable standards.” [9] Generally speaking, once a duty to preserve arises a litigation hold directive should be issued and routine document retention/destruction policies covering potentially relevant material must be suspended. [10]

1. Litigation Hold Notices

In most cases, it will be necessary to perform an initial investigation to identify potential custodians of relevant evidence. Those custodians should then generally receive written litigation hold notices to alert them of the possibility of litigation and the need to preserve potentially relevant evidence. The notice should direct the recipients to immediately suspend the destruction of paper and electronic materials described in the notice, and to identify, segregate, and preserve all potentially relevant materials, including existing documents and documents created in the future. Relevant information to include in a notice, if known, includes the parties to the litigation, the claims at issue, and the time period for potentially relevant records. Specific instructions should normally be given for preserving potentially relevant materials (e.g., moving e-mails to a folder specifically created for the litigation and which is not subject to automatic deletion policies) and for the collection of such materials. In many cases, it will be necessary to interview key employees regarding their particular methods for maintaining their computer files to adequately instruct them on how to preserve such files. The expertise of information technology experts will often be helpful to ensure that preservation is performed securely.

It is prudent to require document custodians receiving the litigation hold notice to acknowledge by e-mail or other writing their receipt of the notice and willingness to abide by its directions. Counsel should keep copies of all such responses. In most cases, counsel or her agent (e.g., a paralegal) should then have a follow-up conversation with the custodians who received the hold notice (or in cases with dozens of potential document custodians, the “key” custodians) shortly after dissemination to ensure that all potentially relevant materials have, in fact, been identified, segregated, and preserved. Employees should be over-inclusive in the documents they preserve, and only counsel should decide what is ultimately relevant and responsive in the litigation. [11] Depending on the nature and course of the litigation, it may be necessary to periodically remind custodians of their obligations under the hold notice, and the scope of the notice should be modified, as needed, as the litigation evolves (e.g., a broader time period, additional claims, new custodians).

The following cases addressed issues that can arise with respect to litigation holds:


In Pension Committee of the University of Montreal Pension Plan v. Banc of America Securities, LLC, 685 F. Supp. 2d 456 (S.D.N.Y. 2010), counsel for plaintiffs contacted their client regarding document collection and preservation as they began work on drafting the complaint. Plaintiffs were instructed to be over-inclusive in gathering materials and to include e-mails and electronic documents in the production. The court found that this instruction failed to meet the standard for a proper litigation hold: “It does not direct employees to preserve all relevant records – both paper and electronic – nor does it create a mechanism for collecting the preserved records so that they can be searched by someone other than the employee.” [12]


In Siani v. State University of New York at Farmingdale, supra, the defendants instituted various document preservation measures. The Assistant Vice President of Administrative Services, who oversaw the defendant college’s IT department, backed up the electronic mailboxes of all individuals named in the EEOC charge. He also directed a litigation hold memo to 11 people, setting forth specific instructions about the preservation of records, including e-mails. A second litigation hold was sent out less than two weeks later by the college’s Associate Counsel. A third litigation hold was issued by counsel and distributed to all defendants approximately seven months later. Defendants were reminded repeatedly of their discovery obligations, and a second back-up of all electronic mailboxes was performed approximately 14 months after the first back-up. Despite defendants’ efforts to institute and enforce a litigation hold, some documents were deleted by individual custodians in the context of routine clean-up procedures, because routine file destruction procedures were not suspended as part of the litigation hold. The court found that defendants were negligent, if not grossly negligent, in the implementation of their preservation efforts, but declined to issue an adverse inference ruling because there was no “willful” spoliation by the defendants and Siani failed to show through extrinsic evidence that any of the missing e-mails were likely to be relevant and favorable to him.


As both Pension Committee and Siani show, litigation hold notices should explicitly and clearly instruct all document custodians not to destroy potentially relevant records. It may be just as important to periodically reissue and enforce litigation holds by following up with key custodians to ensure that they understand this directive and are not inadvertently deleting potentially relevant e-mails in a good faith—but misguided and potentially disastrous—attempt to clean up their inbox.

2. System-Wide Document Preservation

Some cases will require additional steps to reasonably preserve potentially relevant ESI. When litigation involves one or more employees who may have an incentive to destroy evidence once given notice of the litigation (e.g., an employment discrimination case where the employees are the accused discriminators), timely steps should be taken to independently preserve potentially relevant evidence in their possession. This typically involves taking a “snapshot” or otherwise archiving the relevant records (particularly e-mail). In certain cases, it also may be necessary to take a forensic image of the employees’ hard drive for future authentication purposes (for example, if you have reason to believe they already may have attempted to destroy relevant evidence).


In Voom HD Holdings LLC v. EchoStar Satellite L.L.C., 93 A.D.3d 33 (N.Y. App. Div. 2012), the defendant failed to timely institute document preservation measures. The court also criticized, in dicta, the steps that the company did take. The court found that the company failed to suspend the automatic deletion of e-mail until four months after the litigation commenced, causing the loss of e-mail documents due to the company’s seven-day auto-delete policy. Although the company took a snapshot of relevant custodians’ e-mail accounts, the court faulted the company for not doing so until four days after the lawsuit was filed.


The Voom court took a rather extreme position in its document preservation expectations. [13] In most cases, if a timely preservation notice is issued and appropriate guidance and monitoring is given by counsel, costly steps such as taking snapshots of e-mail records or suspending auto-delete procedures should not be necessary to preserve potentially relevant evidence. As long as employees are not actively deleting e-mails, the e-mails will be preserved. If, however, the institution has a particularly aggressive automatic deletion policy for existing e-mails that extends to e-mails kept in a recipient’s “inbox,” it may be advisable to suspend any “auto-delete” function for key document custodians. Care also should be taken to cease the deletion of e-mail or otherwise preserve ESI for former employees whose ESI may be relevant to a dispute or potential dispute.

As for backup tapes, the general view is that backup tapes are “inaccessible,” and therefore do not need to be preserved, when they are maintained solely for disaster recovery purposes. If an institution actively uses backup tapes for information retrieval, however, the tapes generally would be considered “accessible” and would be subject to a litigation hold. [14] Backup tapes also should be preserved in the uncommon event “they are the sole source of relevant information or when they relate to key players, if the relevant information maintained by those players is not obtainable from readily accessible sources.” [15]


Preservation remains the aspect of discovery that poses the most significant risk to litigants, because failures in preservation can be difficult to mitigate (if the data truly is gone, it may be impossible to recover) and can lead to case-altering sanctions. However, the retrieval and production side of discovery can be just as tricky and unnecessarily expensive if done inefficiently or – worse yet – defectively, requiring steps to be redone at significant expense.

Institutions should focus on (1) the process they use to search and cull electronic documents such as e-mails, Word documents and .PDF documents; (2) potentially relevant data that may be stored on data sources that are not reasonably accessible; (3) the form and organization of the production of materials; and (4) protecting against the inadvertent production of privileged documents.

A. Searching Electronically Stored Information

For more than a decade, litigants have used keywords and other selection criteria (e.g., date ranges and document types), to search stores of data to find potentially relevant documents. However, when used improperly, search terms can identify an excessive number of irrelevant documents (leading to unnecessary review costs) and, worse, can fail to identify a reasonable number of responsive documents (leading to the expense of re-searching the data for these additional documents).

The key to developing good search terms is to move beyond the scenario where a group of lawyers and paralegals brainstorms words and takes for granted that terms selected are both reasonably precise and complete. Many litigants have discovered that neither is true. The best search terms are developed after counsel has talked to witnesses (and discovered how they talked about the issues) and reviewed documents (to see the actual vocabulary at issue). Perhaps most importantly, the terms should be tested on a sample to determine how they actually react to the documents they are being applied against.

From a defensibility perspective, counsel should determine the recall of the chosen terms (what percentage of the documents that he is looking for is actually identified by the terms). From a cost perspective, counsel should determine the precision of the chosen terms (of the documents identified by the terms, what percentage is actually what counsel wants). If the terms are only capturing 10% of the documents of interest, the terms would likely be open to challenge, and if only 10% of the documents identified are of interest, that means 90% of the review time will be wasted on reviewing irrelevant documents. By looking at the sample and how the terms interact with the documents, counsel can calibrate the terms and improve recall and precision—thus improving defensibility and lowering cost. There is no magic threshold for recall and precision and it is impossible, absent reviewing everything, to capture 100% of all responsive documents; but keep in mind, the standard is not perfection, but reasonableness.


In National Day Laborer Organizing Network, 877 F. Supp. 2d 87 (S.D.N.Y. 2012), Judge Scheindlin discussed a party’s obligation to search for ESI in the FOIA context. Importantly, under FOIA, the government has the burden to show that its searches for responsive information were reasonable, while in discovery, the requesting party on a motion to compel or for sanctions has the obligation to show that the responding party’s searches were unreasonable. [16] Judge Scheindlin opined that “most custodians cannot be ‘trusted’ to run effective searches because designing legally sufficient electronic searches in discovery or FOIA contexts is not part of their daily responsibilities.” She also chastised several government agencies for not properly testing their search terms to determine how well they would identify responsive documents.


Judge Scheindlin’s position, stated in the National Day Laborer case, that most lay employees are not trained to conduct complex Boolean (or other) searches potentially applies in both FOIA and e-discovery contexts. But this cannot be taken to infer that employees in general cannot be trusted with document preservation, especially where employees not only know their documents better than anyone else, but are generally capable of differentiating relevant documents from irrelevant documents given the proper instruction. The more important take-away from the National Day Laborer case is Judge Scheindlin’s focus on the importance of iteratively testing selection criteria and not relying solely on brainstormed search terms when searching ESI.

Moving beyond reliance on only search terms, predictive coding is a newer search technology that, when implemented properly, measures and controls both the recall and precision of a search into a single quantifiable score called “F1.” [17] This allows litigants to prioritize automatically the documents in their review set based on what the lawyer believes are the most important, and, potentially, to exclude from review documents that are unlikely to be responsive. Predictive coding analyzes the coding made by reviewers on the document set and then suggests un-reviewed documents that it believes are similar to the documents the reviewers thought were important or responsive. As the reviewers continue to code documents, the technology refines its suggestions and generally gets better at predicting the documents that the reviewers want to see. Eventually, the technology can get to the point where it predicts that there are no (or very few) relevant documents that are of interest. If used properly, predictive coding promises to reduce the cost of document review while maintaining quality and defensibility and helping “secure the ‘just, speedy, and inexpensive’ (Fed. R. Civ. P. 1) determination of cases in our e-discovery world.” [18] At the same time, predictive coding has not made search terms obsolete, and in many cases, predictive coding and well-crafted search terms can and should be used together for maximum effect.


In Da Silva Moore v. Publicis Groupe SA, 2012 WL 607412 (S.D.N.Y. Feb. 24, 2012), Magistrate Judge Peck (an e-discovery thought leader) recognized that “computer-assisted review is an acceptable way to search for relevant ESI in appropriate cases[,]” in connection with defendant’s proposed process for using predictive coding in a gender discrimination class action. Although the parties agreed to the use of predictive coding to assist in the review of documents, they disputed the proper methodology. Judge Peck found that predictive coding could be used in a reasonable discovery process, but that the defendant’s request to search only a limited number of documents was arbitrary. Instead, the decision as to how many documents in the collection would need to be reviewed should be determined based on how many responsive documents were uncovered and by sampling the presumably non-responsive documents. Importantly, Judge Peck made it clear that, even if a few responsive documents were found in the “non-responsive” sample, this would not necessarily mandate further review of the “non-responsive” documents and that it would depend on not only the number of responsive documents, but also if they were cumulative of information already reviewed and produced. Perfection is not the standard.


In Global Aerospace, Inc. v. Landow Aviation, L.P., 2012 WL 1431215 (Va. Cir. Ct. Apr. 23, 2012), the Court, in a very concise order, overruled the objections of the plaintiffs to the defendant’s use of predictive coding as part of its discovery process. The court held that the plaintiffs could object to the completeness of the production once the defendants had produced their documents. Note that, after the defendant’s production, the plaintiffs did not raise any objections.


B. Not Reasonably Accessible Information

The Federal Rules of Civil Procedure allow responding parties to refrain from searching for, or producing documents from, data sources that are not reasonably accessible because of “undue burden or cost.” [19] Thus, even though a party may reasonably believe that relevant and responsive documents could potentially exist on a particular data source, such as a disaster recovery tape or in the unallocated space on a hard drive, the party would not need to search for and produce that data if the cost and burden to do so would be excessive in relation to its probative value or the nature of the claim. It is important to note, however, that information identified as “not reasonably accessible” must be difficult to access by the producing party for all purposes, not just for litigation.

Invoking the protections of Rule 26(b)(2)(B) requires the responding party to identify the data sources that it believes are not reasonably accessible. Although the rule does not provide a deadline for doing so, a party who fails to raise an objection to producing documents from data sources that are not reasonably accessible when it responds to written document requests risks waiving the protection. Moreover, parties are encouraged to discuss “not reasonably accessible” data sources during the initial meet and confer discovery process.

Once a party has identified data sources that are not reasonably accessible, it does not need to search or collect information from those sources unless the opposing party establishes good cause for the discovery. [20] If the requesting party can establish good cause, then the court can order collection and production of the data with (or without) limitations, including cost sharing. [21] Remember, however, that The Federal Rules Advisory Committee Notes also caution that “[a] party’s identification of sources of electronically stored information as not reasonably accessible does not relieve the party of its common law or statutory duties to preserve evidence.” As discussed above, a party does not generally need to preserve data that is maintained on not reasonably accessible data sources unless it contains unique, material information. [22] For example, in one of the Zubulake decisions, Judge Scheindlin found that a party would not need to preserve disaster recovery tapes unless they contained a key person’s data that was not otherwise available. Likewise, the Sedona Conference has found there is no obligation to preserve all potentially relevant information or take heroic steps to preserve. [23] Another common example of inaccessible data is deleted, fragmentary, and residual data on employee hard drives. Although it is likely that marginally relevant information exists in the unallocated space of hard drives used by relevant employee custodians, a party is not obligated to create forensic images of these computers to capture this information absent knowledge that relevant files have been deleted or there is reason to believe that an employee/custodian may have incentive to delete such information from his/her files (e.g., accused of embezzlement or theft of intellectual property).

C. Form and Organization of Production

How an institution produces its electronically-stored data can lead to excessive costs if not done properly. Under the Federal Rules of Civil Procedure and most state rules, a party is required to produce its data in a proper format and with proper organization. If an institution fails to do so, then it could be required to reproduce (or re-review) the data at great expense.

In federal court litigation, absent an agreement of the parties or an order of the court, a party responding to document production requests must produce either (1) in a form or forms in which the information is ordinarily maintained (native format) or (2) in a form or forms that are “reasonably usable.” [24] Production in native format may not be a good choice for a number of reasons: the documents cannot be Bates-labeled; they cannot be redacted; they can be inadvertently modified; and many lawyers are not facile with those types of files in depositions or in court. But producing documents in a “reasonably usable” form does not allow a responding party “to convert electronically stored information from the form in which it is ordinarily maintained to a different form that makes it more difficult or burdensome for the requesting party to use the information efficiently in the litigation.” [25] Thus, voluminous paper print-outs of otherwise searchable native electronic data likely will not be considered “reasonably usable” (unless the parties agree in advance that producing documents in paper format is agreeable). Conversely, a format like .pdf with searchable text and appropriate metadata [26] likely would be acceptable.

Additionally, in federal court litigation, a responding party must produce documents as “they are kept in the usual course of business or must organize and label them to correspond to the categories in the request.” [27] Some courts have found that producing documents as “they are kept in the usual course of business” requires the responding party to at least identify the custodian of the file and provide copies of any red-wells, folders or binders. Some parties have been pushing for even more information, such as complete file paths (e.g., C:/MyDocuments/Project/Task/File.doc). If this information is not obtained with the initial document collection, it can be very difficult to gather it later, which could force a party to re-review documents and categorize them by request in order to satisfy its obligations under the Federal Rules.


In Valeo Electrical Systems, Inc. v. Cleveland Die & Manufacturing Co., 2009 WL 1803216 (E.D. Mich. June 17, 2009), the plaintiff produced more than 270,000 pages of e-mails and other ESI. Defendants complained that the plaintiff had not produced the documents as they were kept in the ordinary course of business and moved for an order requiring the plaintiff to produce them according to defendants’ categories of document request. The Court denied the motion, and explained that a party produces documents as they are kept in the usual course “by revealing information about where the documents were maintained, who maintained them, and whether the documents came from one single source or file or from multiple sources or files.” For e-mails, the party must arrange “the responsive emails by custodian, in chronological order and with attachments, if any.” For other ESI, the party “must produce the files by custodian and by the file's location on the hard drive—directory, subdirectory, and file name.”


D. Inadvertent Production of Privileged Documents

Given the explosion in the volume of documents and the fallibility of even the most diligent reviewers, sooner or later privileged documents may accidentally be produced in discovery. However, there are steps institutions can take to protect themselves from waiving their privilege if and when an otherwise privileged document is inadvertently produced.

First, institutions can enter into “quick peek” or “claw back” agreements with opposing parties. Under a quick peek agreement, the responding party can provide certain requested materials for initial examination without waiving any privilege or protection. The requesting party then designates the documents it wishes to have produced and the responding party then reviews those documents and asserts privilege, where appropriate. Although this can limit the volume of documents that must be reviewed for privilege, it has the decided drawback of allowing the responding party to rummage around in the documents—both responsive and non-responsive—and, even if the privilege is not waived, the privileged information has been seen and cannot be removed from the opposing party’s memory. Claw back agreements allow for the return of documents that are inadvertently produced without waiver of any privilege or work product protection. Although this does not reduce the document review effort, because most parties want to avoid the disclosure of privileged documents in the first place (and avoid the production of irrelevant or private data altogether), it limits the number of privileged documents reviewed by the opponent as compared to the quick peek method and provides insurance against the inadvertent production of privileged material. Such agreements are commonly entered into by parties with the approval of the Court. [28]

Second, in federal court litigation, institutions should, usually as a matter of course, take advantage of Federal Rule of Evidence 502. Rule 502 provides that disclosure of a communication or information covered by the attorney-client privilege or work-product protection shall not constitute waiver if (1) the disclosure is inadvertent; (2) the holder of the privilege or protection took reasonable steps to prevent disclosure; and (3) the holder promptly took reasonable steps to rectify the error. This creates a significant incentive to establish a reasonable document review process (before documents are produced) and a proper quality control process. Perhaps even more useful, Rule 502(d) allows federal courts to enter orders providing that the disclosure of privileged materials does not constitute waiver in any circumstance, regardless of the efforts made to withhold or rectify the error (which can eliminate discovery disputes about such efforts).


In Radian Asset Assurance v. College of the Christian Brothers of New Mexico, 2010 WL 4928866 (D.N.M. Oct. 22, 2010), the defendants produced un-reviewed copies of their disaster recovery tapes to the plaintiffs pursuant to a Rule 502(d) court order, which provided that the production of any privileged documents on the tapes would not constitute waiver of the privilege. The plaintiff argued that the court’s order constituted an improper cost shifting because by producing the disaster recovery tapes to the plaintiff in their entirety, the expense of processing and reviewing the responsive documents was being shifted to the plaintiff. The court disagreed, finding that the cost of restoring and searching the back-up tapes would put an undue burden on the defendant college and, therefore, the tapes were not reasonably accessible, and the court was not shifting costs by merely providing that the production of privileged documents therein would not constitute waiver. Interestingly, it is not clear what the college was doing to prevent the disclosure of data that would otherwise be subject to privacy regulations (for example, depending on the records involved, a “quick peek” could potentially violate FERPA), and this issue is not addressed by the Court.


In Brookfield Asset Management v. AIG Financial Products Corp., 2013 WL 142503 (S.D.N.Y. Jan, 7, 2013), Magistrate Judge Maas encouraged the parties to enter into a 502(d) stipulation that he ordered. He found that this prohibited any waiver of privilege from the production of documents in the case, regardless of “what the circumstances giving rise to their production were.”


Managing the Costs and Risks of E-Discovery

The excessive costs and hazards of e-discovery are real, but that does not mean that institutions are powerless to minimize their risks and reduce costs. When dealing with reasonable opposing counsel, many of the procedural aspects of discovery can be addressed through the ordinary meet and confer process and cooperation throughout discovery. Cooperation does not mean capitulation, but it is a tool that can be used to focus discovery disputes on real issues. The judiciary, the Federal Rules of Civil Procedure, and more and more state rules, are encouraging (if not requiring) that parties discuss discovery – including potential problems in discovery – as soon as possible in litigation. Notably, judges across multiple circuits, including Scheindlin and Peck, have invoked the The Sedona Conference Cooperation Proclamation [29] in orders involving e-discovery. [30] The Federal Rules, for example, mandate that a number of discovery issues be discussed at the initial Rule 26(f) discovery conference between the parties. Institutions should take the time and effort to prepare well for these conferences so that they can set the agenda and ensure they enter into agreements with the other side that minimize both their costs and their discovery risks. Institutions also should realize that e-discovery issues can be complex, so they may not know enough at the beginning of a case to agree on all issues. Multiple conferences that continue throughout the litigation may be the best way to move forward.

The keystone for all discovery is reasonableness and proportionality and the ability to convincingly demonstrate the reasonableness and proportionality of your actions. Discovery is a means to an end, not the reason we litigate. Whether as requester or responder, litigants should always ask themselves if the cost of doing the discovery is worth the expected benefit. When acting unilaterally, responding parties generally need to be conservative because their decisions will need to withstand the scrutiny of 20/20 hindsight. Thus, where appropriate, it may be more cost effective to engage opposing counsel and attempt to set reasonable limits to discovery. This may be particularly effective in symmetrical cases where both sides have significant amounts of relevant data (and thus, similar risks and costs). In non-symmetrical cases, it may be useful to raise cost shifting. An outgrowth of proportionality, cost shifting essentially argues that we do not believe that the benefit of the discovery sought outweighs the cost and, if you do, you should be willing to pay for it. [31]


E-discovery is a fact of modern litigation, and you must be proactive to avoid the potential costs and risks if you are not prepared to meet the challenges of e-discovery head-on. Technology is quickly advancing, data volume is expanding, and courts expect litigants to keep up. A few concluding practical pointers:


You should be on top of your system architecture and existing records retention policies before disputes arise, so that you can respond quickly and efficiently once litigation is threatened. You do not need to know every facet of every system, but you should be well versed on the high value, high risk systems that are frequently part of litigation.


An ounce of prevention is worth a pound of cure: a thorough, timely, and appropriately enforced litigation hold can potentially save thousands of dollars down the road when it comes time to collect and produce documents.


Use the meet and confer process at the beginning of discovery to your advantage. Come armed with robust knowledge of your institution’s IT architecture and the ESI potentially relevant to your dispute, and try to reach meaningful and realistic compromises with opposing counsel to manage the burden of ESI discovery.


Keep in mind—and remind opposing counsel, when necessary—that all discovery should be conducted under the principles of reasonableness and proportionality.


Make sure that your legal services provider is well-versed in e-discovery so that document collection and production can be done right the first time.



1. Wendy Butler Curtis and Caroline M. Mew, Preparing for E-Discovery, NACUANOTES Vol. 6, No. 2 (Feb 22, 2008).

2. Outside of the litigation context, institutions should make sure that they are following their own standard document retention policies with respect to ESI. In reviewing a motion for sanctions for alleged document spoliation in Jones v. Bremen High School District 228, 2010 WL 2106640 (S.D. Ill. May 25, 2010), the court expressed its concern over the defendant’s failure to follow its own document retention policy, which was published (and touted) on its web site.

3. See, e.g., In re Pfizer Inc. Sec. Litig., 2013 WL 76134, at *12-13 (S.D.N.Y. Jan. 8, 2013); Pension Comm. of the Univ. of Montreal Pension Plan v. Banc of America Sec., LLC, 685 F. Supp. 2d 456, 461 (S.D.N.Y. Jan. 2010) (citing Fujitsu Ltd. v. Federal Express Corp., 247 F.3d 423, 436 (2d Cir. 2001)), abrogated on other grounds, Chin v. Port Auth. of New York & New Jersey, 685 F.3d 135, 162 (2d Cir. 2012).

4. See, e.g., Pension Comm., 685 F. Supp. 2d at 466.

5. “A party seeking an adverse inference instruction as a sanction for the spoliation of evidence must establish that: (1) ‘the party having control over the evidence had an obligation to preserve it at the time it was destroyed,’ (2) ‘the records were destroyed with a ‘culpable state of mind,’ and (3) ‘the destroyed evidence was ‘relevant’ to the party’s claim or defense such that a reasonable trier of fact could find that it would support that claim or defense.’” Scalera, 262 F.R.D. at 166.

6. See 262 F.R.D. at 173-74 (discussing 29 C.F.R. § 1602.14).

7. See Siani, 2010 WL 3170664, at *5 (“If [litigation] was reasonably foreseeable for work product purposes, . . . it was reasonably foreseeable for duty to preserve purposes.”).

8. “While a litigant is under no duty to keep or retain every document in its possession . . . it is under a duty to preserve what it knows, or reasonably should know, is relevant in the action, is reasonably calculated to lead to the discovery of admissible evidence, is reasonably likely to be requested during discovery and/or is the subject of a pending discovery request.” Zubulake v. UBS Warburg LLC, 220 F.R.D. 212, 217 (S.D.N.Y. 2003) (quoting Turner v. Hudson Transit Lines, Inc., 142 F.R.D. 68, 72 (S.D.N.Y. 1991)).

9. Rimkus Consulting Grp., Inc. v. Cammarata, 688 F. Supp. 2d 598, 613 (S.D. Tex. 2010) (citing The Sedona Principles: Second Edition, Best Practices Recommendations & Principles for Addressing Electronic Production 17 cmt. 2.b (2007)).

10. See Scalera, 262 F.R.D. at 172 (citing Zubulake v. UBS Warburg LLC, 220 F.R.D. 212, 218 (S.D.N.Y. 2003) and Toussie v. County of Suffolk, 2007 WL 4565160, at *7 (E.D.N.Y. Dec. 21, 2007)).

11. See Pension Comm., 2010 U.S. Dist. LEXIS 4546, at *38 (criticizing litigation hold for failing to direct employees to preserve all relevant records or create a mechanism for collecting preserved records; “the directive places total reliance on the employee to search and select what that employee believed to be responsive records without any supervision from Counsel”). But see Chin v. Port Authority, 685 F.3d 135, 162 (2d Cir. 2012) (rejecting Pension Committee’s holding that failure to issue written hold notice was per se gross negligence and, instead, holding that it depends on the “totality of the circumstances.”).

12. 685 F. Supp. 2d at 473.

13. One of the authors of this article has criticized the Voom opinion for its dicta holding that Echostar was grossly negligent for how it preserved documents once it believed the duty was triggered. See David J. Kessler and Emily Johnston, Is the Gap Between Perfection and Negligence Closing? Stripping the Dicta Out of Voom v. Echostar, BNA Digital Discovery & e-Evidence 2012.

14. See, e.g., Forest Labs., Inc. v. Caraco Pharm. Labs., Ltd., 2009 WL 998402, at *3-4 (E.D. Mich. Apr. 14, 2009) (citing Zubulake, 220 F.R.D. at 218).

15. Pension Comm., 685 F. Supp. 2d at 471.

16. Compare Nat’l Day Laborer Organizing Network, 877 F. Supp. 2d at 95, with The Sedona Conference’s Principle 7 (“The requesting party has the burden on a motion to compel to show that the responding party’s steps to preserve and produce relevant electronic data and documents were inadequate”).

17. Document Categorization in Legal Electronic Discovery: Computer Classification vs. Manual Review (see Resources section at conclusion of article).

18. Andrew Peck, Search, Forward, L. Tech. News, Oct. 2011, at 25, 29.

19. See Fed. R. Civ. P. 26(b)(2)(B).

20. See Fed. R. Civ. P. 26(b)(2)(B).

21. See id.; 2006 Advisory Note to Rule 26(b)(2)(B).

22. Zubulake v. UBS Warburg LLC, 220 F.R.D. 212 (S.D.N.Y. 2003) (Zubulake IV).

23. The Sedona Conference, The Sedona Principles, Princ. 5 and Comments (2d Ed. 2007).

24. Fed. R. Civ. P. 34(b)(2)(E)(ii).

25. Committee Note to Fed. R. Civ. P. 34(b).

26. Metadata is often defined as “data about data,” but from a practical perspective, it can be described as the information about documents that helps us sort, categorize and search them efficiently. While computers and applications create a great deal of metadata for individual documents, the metadata that is most often of interest in litigation and, consequently, most often exchanged in discovery includes: date created, date modified, author, to, from, cc, bcc, subject line, file name, and pathname.

27. Fed. R. Civ. P. 34(b)(2)(E)(1).

28. A model Rule 502(d) and claw back agreement is attached as an appendix to this article.

29. The Sedona Conference Cooperation Proclamation, 10 Sedona Conf. J. 331, 331 (2009).

30. William A. Gross Constr. Assoc., Inc. v. American Mfgrs. Mut. Ins. Co., 256 F.R.D. 134, 136 (S.D.N.Y. 2009) (Judge Peck); SEC v. Collin & Aikman Corp., 256 F.R.D. 403, 415 (S.D.N.Y. 2009) (Judge Scheindlin).

31. See, e.g., Oracle v. SAP AG, 566 F. Supp. 2d 1010 (N.D. Cal. 2008).


        NACUA Resources

NACUA Electronic Discovery and Electronically Stored Information Resource Page.

Wendy Butler Curtis and Caroline M. Mew, “Preparing for E-Discovery,” NACUANOTES Vol. 6, No. 2 (Feb 22, 2008)

        Additional Resources

K&L Gates: Electronic Discovery Law.

ABA Legal Technology Resource Center, Electronic Discovery.

ARMA International.

Electronic Discovery Reference Model (EDRM).

Federal Judicial Center, Education Programs and Materials
(including “Managing Discovery of Electronic Information: A Pocket Guide for Judges,” Barbara J. Rothstein; Ronald J. Hedges; Elizabeth C. Wiggins, 2007.) (2d ed. 2012 )

The National Archives Records Management.

The Sedona Conference.

The Sedona Principles, “Principles Addressing Electronic Document Production” (2d ed. June 2007)

EDUCAUSE: ESI and E-Discovery Resources.

Herbert L. Roiblat, Anne Kershaw, and Patrick Oot, “Document Categorization in Legal Electronic Discovery: Computer Classification vs. Manual Review,” 61 J. of Am. Soc’y for Info. & Tech. 70, Jan. 2012, pp. 70-80.

Robert Owen and David Kessler, “Who Knows Best?Law Technology News, Feb. 2011.

David J. Kessler and Emily Johnston, “Is the Gap Between Perfection and Negligence Closing? Striping the Dicta Out of Voom v. EchoStar” BNA, Digital Discovery & e-Evidence, Vol. 12, No. 4, Feb. 16, 2012, pp. 1-6.

David Kessler and Florinda Baldridge, “Predictive Coding: Five Things You Should Know,” Fulbright & Jaworski L.L.P.

David Kessler, “Debunking The Five Biggest Myths of Predicative Coding,” LJN’s Legal Tech Newsletter, Vol. 29, No. 2, June 2012.

Gene Eames, David J. Kessler and Andrea D’Ambra, “Once is Not Enough: The Case for Using an Iterative Approach to Choosing and Applying Selection Criteria in Discovery,” EDRM, July 21, 2010.



Model Rule 502(d) and Claw Back Stipulation

Inadvertent Production of Privileged Material. If a Party inadvertently or mistakenly produces Privileged Material, such production shall in no way prejudice or otherwise constitute a waiver of, or estoppel as to, any claim of privilege or work-product immunity for the inadvertently produced document or any other document covering the same or a similar subject matter under applicable law, including Federal Rule of Evidence 502. The Parties agree that pursuant to Federal Rule of Evidence 502(d), the inadvertent production of a privileged or work product protected ESI is not a waiver in the pending case or in any other federal or state proceeding.

If a Party has inadvertently or mistakenly produced Privileged Material, and if (1) the requesting party identifies it based on the information on the face of the document; or (2) the Party makes a written request for the return of such Privileged Material, the inadvertently produced Privileged Material (including any analyses, memoranda or notes which were internally generated based upon such inadvertently-produced Privileged Material), as well as all copies, shall be either sequestered or returned within ten (10) business days regardless of whether the Receiving Party disputes the claim of privilege. The Party shall provide sufficient information to the Receiving Party regarding the asserted privilege(s), in the form of a privilege log. If the Receiving Party disputes the assertion of privilege, the Receiving Party may move the Court for an order compelling production of the material, but such motion shall not assert the fact or circumstance of the inadvertent production as a ground for entering such an order. Subject to the Court’s direction, resolution of the issue may include the Court’s review of the potentially Privileged Material in camera. Notwithstanding this Order, no Party will be prevented from moving the Court for an order compelling the production of documents for which the privilege has been waived for any other reason, for example pursuant to the crime-fraud exception.

Permitted Uses of NACUANOTES Copyright and Disclaimer Notice


View this document in PDF or Word

Contact Us | NACUA Home Page

"To advance the effective practice of higher education attorneys for the benefit of the colleges and universities they serve."