January 18, 2018
E-discovery in 2017 featured increasing stability and maturity, due in large part to the continuing impact of the 2015 federal rule amendments addressing sanctions and proportionality. Yet, many challenges remain. Here are some of the highlights from the past year:
As always, the year was an interesting one for e-discovery. We invite you to read our more detailed analysis and observations below.
Amended Federal Rule of Civil Procedure 37(e) continues to have a substantial impact on sanctions for failure to preserve ESI.
Most courts are faithfully applying the requirements of amended Rule 37(e) to sanctions motions, only awarding the most serious sanctions where the responding party destroyed evidence with intent to deprive, tailoring sanctions to be proportionate to actual prejudice, and denying sanctions where there was no prejudice. Nevertheless, a surprising number of courts still relied on common law pre-dating the 2015 rule amendments, apparently unaware of amended Rule 37(e) and its requirements.
Under amended Rule 37(e), courts can only issue the most serious sanctions—e.g., case terminating sanctions or an adverse inference jury instruction—where a party acted with the intent to deprive another party from using the ESI in the litigation.
In Organik Kimya, San ve Tic. A.S. v. Int’l Trade Comm’n, 848 F.3d 994, 103 (Fed. Cir. 2017), the defendant presented evidence that, days before an investigation was to take place, the plaintiffs intentionally began overwriting their laptops to delete what the court estimated to be hundreds of thousands of relevant files. Applying Rule 37(e), the court found that the plaintiffs acted with intent to deprive and held that a default judgment was appropriate “not merely to penalize those whose conduct may be deemed to warrant such a sanction, but [also] to deter those who might be tempted to such conduct in the absence of such a deterrent.”
In Basra v. Ecklund Logistics, Inc., No. 8:16-cv-832017, WL 1207482, at *1, *4 (D. Neb. Mar. 31, 2017), which arose out of an accident involving two trailer-tractors, the plaintiffs alleged the defendant had intentionally destroyed relevant ESI, including accident logs and reports. The plaintiffs requested an adverse jury instruction and attorneys’ fees. The court found that, “although [the] defendant’s record-keeping [was] less than meticulous,” the plaintiffs did not establish that the defendant had destroyed evidence with an intent to suppress the truth. The court therefore held that the defendant did “not engage in conduct that would warrant the sanction of an adverse jury instruction for spoliation of evidence,” and did not issue any sanctions. The court did not explicitly reference Rule 37(e), but appeared to apply its requirements.
In Jackson v. Haynes & Haynes, No. 2:16-cv-01297-AKK, 2017 WL 3173302, at *3–4 (N.D. Ala. Jul. 26, 2017), the court found that the plaintiff failed to take reasonable steps to preserve relevant ESI on her smartphone when she relinquished it to her provider after having retained counsel to pursue the litigation. The court denied the defendants’ request for default judgment or an adverse inference jury instruction, however, because the plaintiff had not acted with intent to deprive the defendants of the evidence. The court reasoned that being “negligent and irresponsible in maintaining the information” and “knowing of her obligation to preserve the integrity of the information” are “not sufficient to show an intent to deprive[.]”
Some courts have found an intent to deprive based on inferences drawn from conduct that might reasonably have been interpreted as negligent, at worst. For example, in Moody v. CSX Transp., — F. Supp. 3d —, No. 07-CV-6398 P, 2017 WL 4173358, at *15 (W.D.N.Y. Sept. 21, 2017), a case arising out of railway accident, the court granted the plaintiff’s motion for an adverse inference instruction where the defendant transferred information from an event data recorder saved on a laptop computer to a central repository, permitted the data on the recorder to be overwritten and recycled the laptop, only to later discover that the data in the repository was unreadable. The court found that the defendant’s conduct supported an inference that it acted with the intent to deprive plaintiff of the event recorder data.
Absent evidence of actual prejudice, courts continued to deny sanctions under amended Rule 37(e)—even in the face of an intentional failure to preserve evidence. For example, in HCC Ins. Holdings, Inc. v. Flowers, No. 1:15-cv-3262-WSD, 2017 WL 393732, at *2-*4 (N.D. Ga. Jan. 30, 2017), the defendant and her husband ran several computer cleaning programs on her personal laptop after a court ordered her to produce her computer. The court concluded that, although the couple’s actions were “troubling, and in breach of [their] duty to preserve,” spoliation sanctions were “not warranted” because the presence of any trade secrets or other information that was relevant to the case was merely “speculati[ve].”
Similarly, in Simon v. City of New York, No. 14-CV-8391-JMF, 2017 WL 57860, at *7 (S.D.N.Y. Jan. 5, 2017), the court refused to impose sanctions against the plaintiff for failing to retain a cell phone video of the events giving rise to an alleged false arrest. The court held there was no prejudice under amended Rule 37(e) because there was no evidence that the video would help the defendants and arguments regarding the contents of the video amounted to “pure speculation.”
In Eshelman v. Puma Biotechnology, Inc., No. 7:16-cv-18-D, 2017 WL 2483800, at *5 (E.D. N.C. June 7, 2017), the plaintiffs sought an adverse inference jury instruction due to the defendant’s failure to preserve internet web browser and search histories relating to an alleged defamatory investor presentation. In refusing to sanction the defendant, the court first noted that, despite the loss of the internet browser history, “other avenues of discovery [were] likely to reveal information about the searches performed.” For example, the defendant could seek such information from people who previously had worked with the plaintiff and assisted her in preparing the investor presentation. The court also found that the defendant had failed to present any evidence “regarding the particular nature of the missing ESI in order to evaluate the prejudice it [was] being requested to mitigate.”
In Crow v. Cosmo Specialty Fibers, Inc., No. 3:15-cv-05665-RJB, 2017 WL 1128505, at *1, *5 (W.D. Wa. Mar. 24, 2017), a court refused to sanction a party under amended Rule 37(e) for its failure to produce an email, where the email was later produced after a more careful search, finding only “meager prejudice.” The moving party was able to conduct several depositions in which it explored topics in the email, and there was no showing that delayed receipt of the email had affected any aspects of the case.
In Edelson v. Cheung, No. 2:13-cv-5870 (JLL (JAD), 2017 WL 150241, at *2-*4 (D. N.J. Jan. 12, 2017), the court awarded an adverse inference jury instruction sanction against the defendant for deleting emails from his personal computer. The plaintiff presented evidence that the defendant had opened a second email account, which he did not disclose even to his own counsel, for the purpose of evading discovery, and then deleted key emails when it was discovered. The plaintiff pointed to an email from the undisclosed account obtained from a third party that stated, “don’t forget to use only gmail account . . . Do not use frontier email. They read everything.” The defendant, for his part, testified that it “didn’t occur” to him to disclose the email account and that he deleted the e-mails because his computer “was running very sluggish” and someone recommended that he delete “certain items” from his computer in order to increase its speed. The court did not find the defendant’s explanation credible.
Pursuant to amended Rule 37(e), courts have continued to order remedies no greater than necessary to cure the prejudice that the moving party suffered. For example, in Edelson, supra, 2017 WL 150241 at *1, *4, the plaintiff sought a default judgment, or, in the alternative, an adverse inference jury instruction, where the defendant deleted key emails from his personal computer. The court found that the defendant had intentionally deleted the emails in an attempt to deprive the plaintiff of relevant information. Nevertheless, the court held that the plaintiff had “failed to demonstrate that he ha[d] suffered a degree of prejudice that merit[ed] the imposition of a default judgment against [the] defendant.” Other evidence besides the emails at issue was available for use at trial to support the plaintiff’s allegations. Thus, the court adopted the “more appropriate sanction [and] instruct[ed] the jury that it [could] presume the information was unfavorable to [the] defendant.”
Despite fairly broad application of amended Rule 37(e) in 2017, a surprising number of courts failed to apply it in spoliation sanctions motions. In many, but not all, of the cases, it nevertheless appears that the sanctions decision would have been the same under Rule 37(e).
For example, in Dallas Buyers Club, LLC v. Huszar, No. 3:15–cv–907–AC, 2017 WL 481469 (D. Or. Feb. 6, 2017), the plaintiff claimed that the defendant illegally downloaded its eponymous movie. The defendant denied doing so, and subsequently destroyed his computer’s hard drive. He claimed the computer began exhibiting signs of failure, at which point he took it to a technician and the content was lost. Id. The court found the defendant credible but still issued an adverse inference jury instruction, finding that “although an adverse inference instruction is not as drastic a remedy as a default order, it is still a harsh remedy and will sufficiently compensate for the potential prejudice suffered by [the plaintiff].” Id. The Court did not consider amended Rule 37(e). Had it done so, the court’s finding that the defendant’s explanation was credible may have precluded a finding of intent to deprive, which would have been necessary to award an adverse inference instruction, and its finding of “potential prejudice” rather than actual prejudice would have been insufficient for any sanction under Rule 37(e).
In Redzepagic v. Hammer, No. 14-civ-9808-ER, 2017 WL 780809, at *4, n. 9 (S.D.N.Y. Feb. 27, 2017), the court refused to issue spoliation sanctions for the plaintiff’s deletion of text messages following commencement of the lawsuit, despite the defendant’s argument that a “very strong inference” could be drawn “that the information [the] plaintiff had would support [the] defendant’s position.” Without reference to amended Rule 37(e), the court found that an employee of the defendant had separately preserved the relevant text messages, and the employee voluntarily turned over those texts to the court. The court reasoned that “because these documents were preserved by an employee . . . and were available to both parties in the action, there [was] no reason to infer that the text messages [the plaintiff] deleted would support [the defendant’s] position.” Thus, the court “decline[d] to impose sanctions or grant an adverse inference,” a result that would likely have been the same under Rule 37(e).
Brown v. Certain Underwriters at Lloyds, London, No. 16-cv-02737, 2017 WL 2536419, at *2–6 (E.D. Pa. Jun. 12, 2017), arose out of a fire that occurred at plaintiffs’ property. The defendants suspected that the plaintiff was involved in setting the fire. They were interested in examining his cell phone to determine whether it contained any evidence that would tend to corroborate their suspicion. A day before the plaintiff was scheduled to produce the contents of his cell phone, he claimed for the first time that he had lost it “months ago.” He provided no details, however, regarding how he lost the phone or his attempts to preserve or recover its contents. The court failed to reference Rule 37(e) and instead relied on common law superseded by the rule. Finding that the defendant’s explanation lacked credibility, the court awarded an adverse inference jury instruction and attorneys’ fees.
Finally, in Charles v. City of New York, No. 12-cv-6180 (SLT) (SMG), 2017 WL 530460, at *25-26 (E.D.N.Y. Feb. 8, 2017), a wrongful arrest case, the court declined to apply Rule 37(e) to a video recording on a smart phone. The defendant sought case terminating sanctions because the plaintiff had lost the smart phone on which she recorded video of her interaction with the police. Noting that the smart phone was not the only evidence in the case, and that there was no evidence of intentional destruction, the court refused to issue sanctions, finding that the plaintiff’s actions at most amounted to “mere negligence, not gross negligence.” The court did not apply amended Rule 37(e), reasoning that amended Rule 37(e) only applies to ESI and that neither the phone nor the video constituted ESI.
Many had expected that the December 2015 amendment to Rule 37(e) would eliminate courts’ inherent authority to impose sanctions for preservation failures, particularly in light of the statement in the Committee Notes that the amended rule “forecloses reliance on inherent authority or state law to determine when certain measures should be used.”
Yet, the language of the amended rule itself did not address the issue. And, barely a month after the amendment’s effective date, Magistrate Judge James C. Francis IV held in Cat 3 LLC v. Black Lineage Inc., 164 F. Supp. 3d 488 (S.D.N.Y. 2016), that if a party’s apparent alteration of e-mails was not sanctionable under amended Rule 37(e), then the court could still impose sanctions pursuant to its inherent authority. Judge Francis subsequently co-authored an article laying out his case for the survival of inherent authority. See Hon. James C. Francis IV & Eric P. Mandel, Limits on Limiting Inherent Authority: Rule 37(e) and the Power to Sanction, The Sedona Conference Journal (Vol. 17, No. 2, p. 613) (2016).
Following Judge Francis’ opinion in Cat 3, Judge Paul Grimm, who was a member of the Civil Rules Advisory Committee, stated that “[w]hen the drafters were crafting Rule 37(e), we did so with a desire to occupy the field.” To obtain spoliation sanctions under inherent authority, according to Judge Grimm, you would “have to argue that in some way, the existing Rule is insufficient and you also have to be faithful to the law of inherent authority,” meaning “you would need to show bad faith.” Tera Brostoff, Reports of Death of Inherent Judicial Authority Exaggerated?, Bloomberg BNA Electronic Discovery and E-Evidence (Nov. 15, 2016).
Judge Grimm’s statement is reminiscent of the Supreme Court’s statement in Chambers v. NASCO, a key case regarding inherent authority, that courts ordinarily should rely on the Rules in imposing sanctions, but “if in the informed discretion of the court, neither the statute nor the Rules are up to the task,” the court may rely on inherent authority. Similarly, Judge Francis has stated that “[t]he point is, if there is a gap in the rule, then the exercise of inherent power is appropriate[.]” Views from the Bench: Leading Federal Judges in Conversation on EDiscovery and More, 34 (R. Hilson & C. Sullivan eds., 2017).
Nevertheless, it appears to be Judge Francis’ view that inherent authority exists even if a matter is covered by Rule 37(e). See id. at 34-35. That view is not shared by all others. See, e.g., id. at 35 (Hon. Frank Maas, ret., quoted as stating “I’m far less sure than Judge Francis is that inherent authority lives on in cases that fall within the four corners of Rule 37(e).”) See also Gareth Evans and Phillip Favro, Unfinished Business: A Holiday Wish List For New E-Discovery Centered FRCP Amendments, LegalTech News (Dec. 15, 2017) (calling for moving to the text of the rule the language in the Rule 37(e) Committee Note foreclosing reliance on inherent authority).
In 2017, the Supreme Court addressed courts’ inherent authority to impose discovery-related sanctions in Goodyear Tire & Rubber Co. v. Haeger, __ U.S. __, 137 S.Ct. 1178 (2017). The Court held that sanctions imposed under inherent authority must be compensatory rather than punitive and must have been “causally related to the sanctioned party’s misconduct.” The case did not involve spoliation, however, and the court did not address whether amended Rule 37(e) forecloses reliance on inherent authority. Thus, it appears unlikely that Goodyear has resolved the issue whether courts may rely on inherent powers in awarding sanctions for a failure to preserve ESI.
Meanwhile, some courts continued to rely upon inherent powers in issuing sanctions for preservation failures. In Hsueh v. New York State Dept. of Financial Servs., 15-civ.-3401-PAC, 2017 WL 1194706, at *4, *6 (S.D.N.Y. Mar. 31, 2017), for example, the court found that amended Rule 37(e) did not apply to the destruction of ESI where the party had “intentionally deleted” the information (despite the fact that Rule 37(e) expressly applies where a party acted with intent to deprive). The court stated that “[b]ecause Rule 37(e) does not apply, the Court may rely on its inherent power to control litigation in imposing spoliation sanctions” in granting an adverse inference sanction for spoliation.
The court in Hsueh observed that amended Rule 37(e) is aimed at “serious problems resulting from the continued exponential growth in the volume of ESI as well as excessive effort and money that litigants have had to expend to avoid potential sanctions for failure to preserve ESI.” In this case, the court reasoned, the ESI was not lost on account of “improper systems in place to prevent the loss of the recording” but rather “because she took specific action to delete it.”
The court concluded, however, that under either amended Rule 37(e) or the court’s inherent authority an adverse inference and attorneys’ fees were appropriate because (i) the plaintiff was under an obligation to preserve the recording, (ii) there was no doubt the destroyed evidence was relevant to the claims in the case, and (iii) the plaintiff acted in bad faith and with an intent to destroy the ESI.
Accordingly, the debate continues over whether inherent authority survives as a basis for spoliation sanctions. At least some of the discussion, however, has shifted to limits on the circumstances under which inherent authority may be invoked (assuming that it can be invoked at all)—for example, that Rule 37(e) must not provide an adequate remedy and bad faith conduct must have been involved. In any event, we doubt that we have heard the last of this issue from courts, commentators and possibly even drafters of future rule amendments.
Proportionality as a limit on the scope of discovery continues to gain traction following its incorporation into Rule 26(b)(1)’s definition of the scope of discovery in the 2015 rule amendments. Of particular note in 2017, the Sedona Conference released its Commentary on Proportionality in Electronic Discovery, 18 Sedona Conf. J. 141 (2017), which sets forth six “Principles of Proportionality” pertaining to the amended rule’s proportionality factors and courts’ application of them since the 2015 rule amendments.
These principles consist of the following: (1) “[t]he burdens and costs of preserving relevant electronically stored information should be weighed against the potential value and uniqueness of the information when determining the appropriate scope of preservation;” (2) “[d]iscovery should focus on the needs of the case and generally be obtained from the most convenient, least burdensome, and least expensive sources;” (3) “[u]ndue burden, expense, or delay resulting from a party’s action or inaction should be weighed against that party;” (4) “[t]he application of proportionality should be based on information rather than speculation;” (5) “[n]onmonetary factors should be considered in the proportionality analysis;” and (6) “[t]echnologies to reduce cost and burden should be considered in the proportionality analysis.”
The discussion in the Commentary on Proportionality reflects that the evaluation of whether discovery is “proportional to the needs of the case” is highly dependent on the specific facts of any given case, and it is the parties’ burden to provide evidence and educate the court on their specific situation. Additionally, proportionality does not merely involve an analysis of the cost of collection and production compared to the need for the documents—it extends beyond this, taking into account the good faith of the parties, the parties’ comparative access to information, and the importance of the issues. Further, the Commentary advocates that parties work together and utilize appropriate technologies in the discovery process.
Judicial decisions in 2017 continued to reflect that proportionality in discovery has gained traction since the 2015 federal rule amendments. In Solo v. United Parcel Service Co., No. 14-12719, 2017 WL 85832 (E.D. Mich., Jan. 10, 2017), for example, the court considered whether UPS should be compelled to produce information stored on backup tapes because their billing system only maintained live data for a short period of time. Id. at *2. UPS submitted a declaration attesting that it would take six months and $120,000 to recover the data from the back-up tapes. The court held that restoring back-up tapes was not proportional to the needs of the case not only because of the expense, but also because the data would only be relevant if the plaintiffs prevailed on certain issues on the merits.
In Scott v. Eglin Fed. Credit Union, No. 3:16-CV-719-RV-GRJ2017, 2017 WL 1364600, at *3 (N.D. Fla. Apr. 13, 2017), an employment discrimination case, the defendant (the plaintiff’s former employer) moved to compel the plaintiff’s current employer (a third party) to produce emails and text messages with the plaintiff. Noting that “emails and text messages may be fair game for discovery in most cases,” the court nonetheless denied the motion to compel, explaining “[b]alancing the marginal relevance of information in emails and text messages against the time and expense that would be involved for a small business … in searching cellular telephones, servers and other electronic storage facilities makes little sense and would cause Plaintiff’s current employer to incur an expense that ultimately will have little or no impact on the outcome of this case.” Id. at *3.
In Simon v. Northwestern Univ., No. 1:150-CV-01433, 2017 WL 467677 (N.D. Ill. Feb. 3, 2017), the court engaged in a substantial proportionality analysis, including analyzing the importance of the issues (“The court finds the importance of the issues at stake in this action extremely high”); the amount in controversy (“the Court finds this amount to be high as well”); the relative burden on the defendants (the court determined it was high as to the individuals but “relatively low” as to the university); and the parties’ access to relevant information (determining that the university had the greatest access).
In Crabtree v. Angie’s List, Inc., No. 1:16-CV-0087-SEP-MJD, 2017 WL 413242, at *3 (S.D. Ind. Jan. 31, 2017), a wages and hours action, the defendant requested a forensic examination of the plaintiffs’ electronic devices to determine how many hours the plaintiffs were working offsite. The court denied the request as not proportional to the needs of the case. Notably, as part of its proportionality analysis, the court considered the plaintiffs’ privacy and security interests.
In Gordon v. T.G.R. Logistics, Inc., 321 F.R.D. 401 (D. Wyo. 2017), the defendant moved to compel production of an electronic copy of the “entire Facebook account history” from the plaintiff’s two Facebook accounts on the ground that the information would be relevant to her claims of physical and emotional injury resulting from a motor vehicle accident. The court engaged in a proportionality analysis, stating that “[s]ocial media presents some unique challenges to courts in their efforts to determine the proper scope of discovery of relevant information and maintaining proportionality.” While it is conceivable that almost any post to social media will provide some relevant information concerning a person’s physical and/or emotional health, it also has the potential to disclose more information than has historically occurred in civil litigation.
Whether a party has “possession, custody or control” over relevant and responsive documents—and therefore an obligation to preserve and produce them—continued to be an important issue in 2017. A split in authority has persisted between courts applying the “legal right” test (i.e., finding that a party has control over documents in the possession of others only when it has the legal right to the documents) and those applying the “practical ability” (i.e., finding that a party has control when it has the practical ability to obtain the documents, even if it does not have a legal right to them).
In Parris v. Pappas, No. 3:10-cv-1128 WWE, 2017 WL 3314001, at *2 (D. Conn. Aug. 3, 2017), the court applied the practical ability test in denying a motion to compel the defendant to produce documents in the possession of his girlfriend. The court held that the plaintiff had failed to sustain her burden of establishing that the documents were in the defendant’s possession, custody or control because the defendant attested that he had asked his girlfriend for the documents, but she had refused to provide them. The court noted, however, that the plaintiff could subpoena the documents from the girlfriend pursuant to Rule 45.
By contrast, the court in Ronnie Van Zant, Inc. v. Pyle, No. 17 Civ. 3360-RWS, 2017 WL 3721777, at *8-*9 ( S.D.N.Y. Aug. 28, 2017), also applying the practical ability test, imposed sanctions on a defendant for its failure to prevent a third-party independent contractor from destroying relevant text messages on his smart phone. The lawsuit arose out of a “blood oath” among the surviving members of the band Lynyrd Skynyrd and the family members of band members who had been killed in a 1977 plane crash that none would seek to profit from the band’s name or story. Despite the oath, which was later reflected in a consent order, the band’s drummer—Artemis Pyle—worked with the defendant film company to produce a film about the band.
In the ensuing lawsuit for breach of the consent order, the court awarded an adverse inference jury instruction holding the defendant film company responsible for the failure of the film’s director—an independent contractor—to preserve relevant text messages that were lost when he turned in and upgraded his personal smart phone. The court reasoned not only that the film company had the ability to ensure that the director preserved relevant data on his smart phone, but also that its failure to do so coupled with the director’s actions “evince the kind of deliberate behavior that sanctions are intended to prevent and weigh in favor of an adverse inference.”
In Williams v. Angie’s List, No. 1:16-00878-WTL-MJD, 2017 WL 1318419, at *2-*3 (S.D. Ind. April 10, 2017), a wage and hours action, the court applied the legal right test. The plaintiffs—who often worked from home and, accordingly, their hours were not reflected in badge-swipe data—sought from the defendant background data automatically recorded while they were working on Salesforce, a sales platform utilized by the defendant. The court rejected the defendant’s argument that it did not have possession, custody or control of the Salesforce data, citing the defendant’s contractual relationship with Salesforce giving the defendant the right to the data.
It is not an overstatement to say that social media has become an integral part of modern life. Social media has played an important role for a number of years in keeping us in touch with friends and family. In recent years, social media applications have also played an prominent role in professional networking and, increasingly, in workplace communications and collaboration. Not surprisingly, therefore, the discovery of social media is also becoming increasingly commonplace.
As social media has expanded into many different areas, conceptions of what it exactly is are becoming somewhat blurred. No longer just Facebook, but numerous other social and professional networking and communication applications may be considered social media. The Oxford English Dictionary defines “social media” as “websites and applications used for social networking” and “social network,” in turn, as “the use of dedicated websites and applications to communicate with each other by posting information, comments, messages, images, etc.” See Concise Oxford English Dictionary (12th ed. 2011). Many social media applications have their own direct and group messaging functions, and many instant messaging applications have features that are common to social media.
As social media is becoming ubiquitous, early notions that social media might have a special status because of privacy concerns (leading to, for example, a requirement of a threshold showing before discovery could be propounded) are giving way to social media being treated no differently from other forms of evidence. See, e.g., United States ex rel Reaster v. Dopps Chiropractic Clinic, LLC, No.13-1453-EFM-KGG, 2017 WL 957436, at *1-*2 (D. Kan. Mar. 13, 2017) (“while information on social networking sites is not entitled to special protection, discovery requests seeking this information should be tailored so as not to constitute the proverbial fishing expedition in the hope that there might be something of relevance in the respondent’s social media presence”) (internal quotations and citation omitted).
Proportionality and relevance requirements can play a particularly important role in discovery of social media. Because social media accounts usually contain a substantial amount of irrelevant and personal information, courts must balance legitimate rights to discovery against overly broad and intrusive inquiries. See, e.g., Brown v. Ferguson, No. 4:15-cv-0083-ERW, 2017 WL 386544, at *1-*2 (E.D. Mo. Jan. 27, 2017) (rejecting disclosure of social media passwords as constituting unfettered access, but also rejecting a distinction between private messages and public content on Facebook).
Gordon v. T.G.R. Logistics, Inc., 321 F.R.D. 401 (D. Wyo. 2017), illustrates the challenge facing courts in determining the appropriate scope of social media discovery. In Gordon, the defendant brought a motion to compel the production of the “entire Facebook account history” of the plaintiff’s two Facebook accounts on the ground that the information would be relevant to her claims of physical and emotional injury resulting from a motor vehicle accident.
The court engaged in a proportionality analysis, observing that “[s]ocial media presents some unique challenges to courts in their efforts to determine the proper scope of discovery of relevant information and maintaining proportionality.” The court continued that “[w]hile it is conceivable that almost any post to social media will provide some relevant information concerning a person’s physical and/or emotional health, it also has the potential to disclose more information than has historically occurred in civil litigation. While we can debate the wisdom of individuals posting information which has historically been considered private, we must recognize people are providing a great deal of personal information publicly to a very loosely defined group of ‘friends,’ or even the entire public internet.”
The court explained that the relative ease and low cost of downloading a user’s Facebook history would not itself resolve the issue. The court observed that, in the past, “[n]o court would have allowed unlimited depositions of every friend, social acquaintance, co-employee or relative of a plaintiff to inquire as to all disclosures, conversations or observations. Now, far more reliable disclosures can be obtained with a simple download of a social media history.”
The court reasoned, on the one hand, that even though producing the plaintiff’s Facebook history would involve very little time or expense, it could nevertheless have a very significant impact in generating additional discovery and in lengthening testimony. “It’s not difficult to imagine a plaintiff being required to explain every statement contained within a lengthy Facebook history in which he or she expressed some degree of angst or emotional distress or discussing life events which could be conceived to cause emotion upset, but which is extremely personal and embarrassing.”
On the other hand, the court recognized that “Defendant has a legitimate interest in discovery which is important to the claims and damages it is being asked to pay. Information in social media which reveals that the plaintiff is lying or exaggerating his or her injuries should not be protected from disclosure. Courts must balance these realities regarding discovery of social media and that is what most of the courts which have addressed this issue have done.”
In the end, the court denied the defendant’s request for the entirety of the plaintiff’s Facebook history and instead limited the scope of the discovery to Facebook posts after the accident that relate to the accident and her resulting physical and emotional injuries and any posts relating to other events that could reasonably be expected to result in emotional distress.
A noticeable practice trend in 2017 has been that the use of technology assisted review (“TAR”)—also known as predictive coding—to search and review large document populations appears to be more widespread than in past years.
We are seeing requesting parties more frequently using TAR in their review of substantial incoming productions, where the TAR protocol and training of the TAR tool will not be subject to challenge from the opposing party. We are also seeing TAR used more often in symmetrical litigation, where both sides have large production obligations and both use TAR—or want to have the option to use TAR—in their document search and review process.
That is not to say that the use of TAR is commonplace, as many had anticipated would be the case by now. Rather, within a relatively small slice of litigation matters—those that involve particularly massive amounts of ESI to search and review—it appears that TAR is being used more than in the past.
A substantial body of case law has developed regarding issues relating to the use of TAR. See The Sedona Conference TAR Case Law Primer, 18 Sedona Conf. J. 1 (2017). Yet, many issues remain unresolved—except that TAR is generally accepted by the courts as a legitimate search and review methodology. There was a dearth of case law in 2017 involving disputes over TAR, perhaps reflecting that TAR is most being used on incoming productions and pursuant to stipulated protocols in symmetrical litigation. The two decisions in 2017 regarding TAR disputes dealt with the extent of transparency required regarding the TAR process and the use of search terms to cull a document population before the use of TAR.
In Winfield v. City of New York, No. 15-cv-05236 (S.D.N.Y. Nov. 27, 2017), the plaintiffs argued that the defendant’s TAR model was improperly trained because its reviewers had over designated documents in the seed and training sets as non-responsive. The plaintiffs argued—and the court agreed—that several inadvertently produced documents designated as non-responsive used to train the TAR model were actually responsive. The plaintiffs sought both to bar the defendant from continuing to use TAR and to require disclosure of information about the TAR process—including the defendant’s coding of seed and training documents, how the defendant trained its document reviewers, and detailed information about the ranking system used in the TAR process (i.e., what relevance score cut-off was used, and how many documents were deemed responsive and unresponsive at each ranking level).
The court referenced Sedona Principle 6, which provides that the producing party is in the best position to “evaluate the procedures, methodologies, and technologies appropriate for preserving and producing their own electronically stored information.” Id., slip op. at 20; see also The Sedona Conference Principles, Third Edition, 19 Sedona Conf. J. 1, 118 et. seq. (forthcoming 2018) (available at www.thesedonaconference.org).
The court stated that, “[t]raditionally, courts have not micro-managed parties’ internal review processes for a number of reasons.” Those reasons include that “attorneys, as officers of the court, are expected to comply with Rules 26 and 34 in connection with their search, collection, review and production of documents, including ESI.” Additionally, the court stated that “internal attorney ESI work processes may reveal work product” and noted that “perfection in ESI discovery is not required[.]” Nevertheless, the court asserted, “parties cannot be permitted to jeopardize the integrity of the discovery process by engaging in halfhearted and ineffective efforts to identify and produce relevant documents.” Id., slip op. at 20-21.
The court reviewed information about the defendant’s TAR process in camera—including information about the seed and training sets, its training of reviewers, and the validation process the defendant used. The court concluded that “the City’s training and review processes and protocols present no basis for finding that the City engaged in gross negligence in connection with its ESI discovery—far from it.” Id., slip op. at 23.
Additionally, with respect to detailed information about the defendant’s TAR process—such as the cut-off used and the number of responsive and unresponsive documents at each ranking level—the court stated that it “views this information as protected by the work product privilege and, accordingly, [it] is not subject to disclosure.” Id., slip op. at 27; see also John M. Facciola and Philip J. Favro, Safeguarding the Seed Set: Why Seed Set Documents May Be Entitled to Work Product Protection, 8 Fed. Cts. L. Rev. 1 ( Feb. 2015). Nevertheless, because there was some evidence of “human error” in the training process, the court ordered the defendant to provide the plaintiffs, on an attorneys’ eyes only basis, with a random sample of 300 non-privileged documents from the population of documents the TAR process determined to be non-responsive. Id., slip op. at 25-26.
The only other reported or widely publicized TAR decision in 2017, FCA US LLC, v. Cummins, Inc., No. 16-12883, 2017 WL 2806896, at *1 (E.D. Mich. Mar. 28, 2017), involved a dispute over “whether the universe of electronic material subject to TAR review should first be culled by the use of search terms.” Without any substantive discussion, other than to cite materials that it reviewed, the court stated that “[a]pplying TAR to the universe of electronic material before any keyword search reduces the universe of electronic material is the preferred method.”
The consolidation among medium-sized and large e-discovery service providers, usually financed by private equity funding, that has been going on for several years now only seemed to accelerate more in 2017. It is not apparent whether this consolidation is fundamentally altering the market for e-discovery services, other than to possibly result in greater stability in the space once all of the M&A dust settles.
Generally, the market appears to be settling into several different segments: (1) large vendors with a national and often international footprint providing basic, commodity services using mostly standard technologies; (2) medium-sized vendors—also with a national and global footprint—focused on providing both expert e-discovery consulting and professional services as well as standard and more advanced technologies; (3) vendors of “do it yourself” online e-discovery software services (i.e., “SAAS,” aka software as a service), usually targeted at small and medium-sized law firms that now, increasingly, must deal with e-discovery; and (4) traditional local and regional vendors providing basic services, much as they have in the past.
The local and regional vendors seem to be increasingly squeezed in this market, either being acquired by or not able to compete with the large vendors providing commodity services. Notably, it appears that there are far fewer new entrants in e-discovery services market—which used to have relatively low barriers to entry—than in the past. Also, there appears to have been significant maturation of some of the SAAS providers, which appear to be finding a solid niche in a potentially large market segment—small and medium-sized law practices—often not previously serviced by e-discovery providers.
Other noteworthy developments in the vendor space have been the challenges posed by mobile devices, social media and ESI stored in the cloud—often requiring advanced tools and significant expertise to collect, process and search—and the more widespread availability of analytics applications that vendors can license and provide to their clients rather than having to develop in-house.
The past year showed once again that e-discovery continues to progress, but also continues to face new and pre-existing challenges. We hope that you found our 2017 Year-End E-Discovery Update informative. We invite you review further the many articles, client alerts and updates that our attorneys have published by going to the Gibson Dunn Electronic Discovery Practice Group’s page on the Firm’s website.
The following Gibson Dunn lawyers assisted in the preparation of this client update: Gareth Evans, Jennifer Rearden, Heather Richardson, Chelsea Mae Thomas and Natalie Dygert.
Gibson Dunn’s lawyers are available to assist in addressing any questions you may have regarding the issues discussed in this update. The Electronic Discovery and Information Law practice group brings together lawyers with extensive knowledge of electronic discovery and information law. The group is comprised of seasoned litigators with a breadth of experience who have assisted clients in various industries and in jurisdictions around the world. The group’s lawyers work closely with the firm’s technical specialists to provide cutting-edge legal advice and guidance in this complex and evolving area of law. For further information, please contact the Gibson Dunn lawyer with whom you usually work or the following leaders of the Electronic Discovery and Information Law practice group:
© 2018 Gibson, Dunn & Crutcher LLP
Attorney Advertising: The enclosed materials have been prepared for general informational purposes only and are not intended as legal advice.