Social-media discovery during the social-media boom

The challenge of finding relevant information

2014 July

Are you on Facebook?  Are your kids, relatives, neighbors, schools and employers on Facebook?  If you are like 1.2 billion people, clearly the answer is “yes!”  It is equally likely that you use Instagram or SnapChat to communicate daily with friends and family.  Nearly 300 million people use LinkedIn to share their resume and related professional information.  More than 241 million individuals and companies are “tweeting” away on Twitter.

The use of social media has grown and is growing exponentially, even while you were reading the paragraph above.  What does the rapidly expanding use of social media, in a multitude of formats, mean for civil discovery? This article will describe how courts are attempting to draw some lines for litigants about the discoverability of social media.

Because the use of social media as evidence in litigation is relatively new, courts are taking a variety of approaches to the challenges presented by social- media discovery.  Federal Rule of Civil Procedure 26(b)(1) generally provides that parties “may obtain discovery regarding any nonprivileged matter that is relevant to any party’s claim or defense.”  This rule broadly means that relevant information posted on social networking sites is discoverable, like any other relevant information. However, due to the volume of information stored, the way sites are operated and the possibility of shared ownership of information, problems in social-media discovery are abundant.

What is relevant information?

Like all digital evidence, one of the most difficult challenges presented by social media is the volume of data. If an individual has maintained a Facebook account for a few years, thousands, if not millions, of “pages” of information are a realistic possibility.  Presented with the unique challenges of social media, how does a court balance the policy favoring broad discovery with the requirement that a party need only produce responsive, relevant information? The decisions below have attempted to answer that question.

A case that contains a frequently cited quotation regarding social-media evidence is Thompkins v. Detroit Metro Airport (E.D. Mich. 2012) 278 F.R.D. 287.  The Thompkins court summarized the problem well:

[M]aterial posted on a ‘private’ Facebook page, that is accessible to a selected group of recipients but not available for viewing by the general public, is generally not privileged, nor is it protected by common law or civil law notions of privacy.  Nevertheless, the [requesting party] does not have a generalized right to rummage at will through information that Plaintiff has limited from public view.  Rather, consistent with Rule 26(b) . . . [and decisional law]. . . there must be a threshold showing that the requested information is reasonably calculated to lead to the discovery of admissible evidence.  Otherwise, the [requesting party] would be allowed to engage in the proverbial fishing expedition, in the hope that there might be something of relevance in Plaintiff’s Facebook account.

(Id. at 288.)

Of course, the notion of what is “relevant” in the social-media context can be difficult to answer. For example, in an employment case where the plaintiff alleges that she was depressed because of the way her employer treated her, is a picture of the plaintiff smiling or frowning posted to a social- media account “relevant”?  Is the fact that the plaintiff is using social media at all relevant?  One of the earliest social-media decisions was E.E.O.C. v. Simply Storage Management, LLC (S.D. Ind. 2010) 270 F.R.D. 430, which permitted a very broad discovery request for information from social-media sites. The court required the plaintiff to produce any communications from social-networking sites that “reveal, refer, or relate to any emotion, feeling, or mental state, as well as communications that reveal, refer or relate to events that could reasonably be expected to produce a significant emotion, feeling, or mental state.” (Id. at 436.)

But when this identical discovery request was made in Mailhoit v. Home Depot (C.D. Cal. 2012) 285 F.R.D. 566, the court found that, under the circumstances of that particular case, the request was not sufficiently tailored to place the responding party on notice of what was requested.  The test for whether or not a discovery request is sufficiently tailored is whether the request places a responding party upon “reasonable notice” of what is called for and what is not. (Id. at 570.)  Accordingly, the Mailhoit court permitted discovery only of communications between plaintiff and defendant’s employees that specifically related to plaintiff’s employment. (Id. at 573.)

Many courts have attempted to narrow the discovery requests to focus on information that is relevant to the claims and defenses raised in the particular action.  These courts are wary of requiring parties to open their entire social-media history just because they are litigants in a lawsuit, even when the producing party is the plaintiff. (See, e.g., Giacchetto v. Patchogue-Medford Union Free School District (E.D. N.Y. 2013) 293 F.R.D. 112, 115 [“a plaintiff’s entire social networking account is not necessarily relevant simply because he or she is seeking emotional distress damages”]; Mackelprang v. Fidelity Nat’l Title Agency of Nevada, Inc. (D. Nev. Jan. 9, 2007) 2007 WL 119149 at *7  [“Ordering . . . release of all of the private email messages on Plaintiff’s Myspace.com internet account would allow Defendants to cast too wide a net for any information that might be relevant and discoverable.”])

Accordingly, some courts attempt to formulate a limiting construction on which social-media communications are properly discoverable.  (Holter v. Wells Fargo and Co. (D. Minn. 2011) 281 F.R.D. 340, 344 [where plaintiff places her mental disability and emotional state at issue, defendant is entitled to receive, among other things, any communications from the social-media Web sites that reveal or refer “to any events that could reasonably be expected to produce a significant emotion, feeling, or mental state”]; Robinson v. Jones Lang LaSalle Ams., Inc. (D. Or. Aug. 29, 2012) 2012 WL 3763545 at *2 [plaintiff must produce “all online social-media communications” that “reveal, refer or relate to . . . any significant emotion, feeling or mental state allegedly caused by defendant’s conduct.”]; Sourdiff v. Texas Roadhouse Holdings, LLC (N.D. N.Y Oct. 24, 2011) 2011 WL 7560647 at *1, [plaintiff must provide “any photographs, profile information, postings, messages, comments and status updates and/or other posts, including deleted content are in any way related to plaintiff’s emotional or mental state, her physical condition, activity level, employment, this litigation and the injuries and damages claimed by plaintiffs in their complaint in this action”].)

In at least two cases, the court conducted an in camera review to decide if the information on social-media sites was relevant. In Bass v. Miss Porter’s School (D. Conn. Oct. 27. 2009) 2009 WL 3724968 at *1, the court conducted an in camera review to compare pages of plaintiff’s Facebook account that plaintiff withheld against the pages that plaintiff produced. The Court found that some of the withheld communications were “clearly relevant to this action” and ordered plaintiff to produce all withheld Facebook documents, admonishing that “relevance of the content of Plaintiff’s Facebook usage as to both liability and damages in this case is more in the eye of the beholder than subject to strict legal demarcations, and production should not be limited to Plaintiff’s own determination of what may be ‘reasonably calculated to lead to the discovery of admissible evidence.’”

In contrast, although the court in Offenback v. LM Bowman, Inc. (M.D. Pa. June 22, 2011) 2011 WL 2491371, conducted an in camera review of plaintiff’s Facebook and MySpace accounts, it found that most of the communications were “routine communications with family and friends” and ordered production of only a handful of selected communications. Unlike the Bass court, the Offenback court also expressed “some confusion about why the parties required the Court’s assistance in deciding what information within Plaintiff’s Facebook account is responsive to Defendants’ discovery requests” because “Plaintiff is the party with the greatest familiarity with his own Facebook account . . . .”  (Id. at n.3.)

Other courts have permitted broader discovery, without narrowing the requests and without conducting in camera review. In Thompson v. Autoliv ASP, Inc. (D. Nev. June 20, 2012) 2012 WL 2342928 at *5, the court ordered plaintiff to provide all the information on her social-media sites to defendant’s counsel on an electronic storage device. However, even with this broad discovery scope, the court ordered defense counsel not to share the information and to return the storage device to plaintiff while identifying relevant information. (Id.; see also Ingrid & Isabel, LLC v. Baby Be Mine, Inc. (N.D. Cal. April 1, 2014) 2014 WL 1338480 at *10  [ordering that plaintiff’s expert receive access to all of defendant’s accounts, including Facebook and Twitter, because of court’s serious concern that defendants did not produce responsive documents].)

Finally, a few courts have denied access to social-media information altogether.  (Salvato v. Miley (M.D. Fla. June 11, 2013) 2013 WL 2712206 at *2  [denying plaintiff’s request for production of all statements that defendant made on social-networking Web sites about the incident at issue because the “mere hope” that such discovery might lead to relevant evidence is not a sufficient basis to require defendant to provide “open access” to his private communications]; Potts v. Dollar Tree Stores, Inc. (M.D. Tenn. March 20, 2013) 2013 WL 1176504 at *3, [denying social-media discovery because defendant failed to present any evidence showing that the plaintiff’s social media would contain relevant information]; Kennedy v. Contract Pharmacal Corp. (E.D. N.Y. May 13, 2013) 2013 WL 1966219 at *2 [same].)

Beyond deciding what social-media information is discoverable, courts must address the burden of discovery.  Voluminous information usually inspires objections based on the burden caused by collection, review and production.  The burden of production is typically related to cost issues.  However, because social-media sites typically provide account holders with the means to download their information easily, the cost of retrieving and downloading the information may not be a problem, particularly for a narrowly tailored request. For example, with just a few clicks you can download your information from Facebook as well as Twitter. (See Twitter Help Center, https://support.twitter.com/ articles/20170160-downloading-your-twitter-archive#); Facebook Help Center, https://www.facebook.com/ help/131112897028467). The effort and cost appear minimal, at least when a single account is involved.

But as is true with all electronic discovery, the broader the request, the greater the cost. Courts may reject requests for cost shifting when the requesting party has not adequately demonstrated the burden.  (See German v. Micro Electronics, Inc. (S.D. Ohio Jan. 11, 2013) 2013 WL 143377 at *7-8 [rejecting request for cost shifting for plaintiff’s production of online journals, blogs and social-media/networking Web sites].) Ultimately, the cost battles may arise from the cost of counsel’s review of the material, rather than the collection and production of social-media information.

It is fair to conclude that the majority of courts, applying traditional notions of relevance and reasonable notice, have required social-media requests to be specific and give appropriate notice to the responding party. The requests must target the type of information that would be probative of the claims and defenses raised in that particular case.

Who has “possession, custody or control” over the information?

Rule 34 provides that a responding party must produce information within the party’s “possession, custody or control.” (See Soto v. City of Concord (N.D. Cal. 1995) 162 F.R.D. 603, 619 [“The phrase ‘possession, custody or control’ is in the disjunctive and only one of the numerated requirements need be met.”]) However, social media raises confusing questions regarding who has “possession, custody or control” over the information. While courts may permit parties to obtain relevant social- media content from each other, the Stored Communications Act (“SCA”), 18 U.S.C. §§ 2701-2712 (part of the Electronic Communications Privacy Act of 1986), prohibits litigants from obtaining social-media content directly from a provider.  (See Crispin v. Christian Audigier, Inc. (C.D. Cal. 2010) 717 F. Supp. 2d 965, 971-972; Mintz v. Mark Bartelstein & Assocs., Inc. (C.D. Cal. 2012) 885 F. Supp. 2d 987, 993.) However, the SCA permits a provider to divulge subscriber records and related information. (Id. at 992.)  Content, however, may only be disclosed by a party to the litigation who controls the social-media account.

The sharing of a digital device or social-media account between two parties presents a difficult issue. For example, if an employee uses her employer’s computer to send e-mails or to post information on a social-networking site, then it is unclear whether the employee has complete control over the information, especially if the employee has signed an agreement giving the employer the right to review messages, e-mails, etc. sent on a company device.  (See Holmes v. Petrovich Development Company, LLC (2011) 191 Cal.App.4th 1047, 1068-9 [employer could access employee e-mails to attorney that were sent from employer’s computer].)

Is there a privacy right in social media?

Asserting privacy interests in social media has not been a successful tactic in most cases, at least in federal court. The court in Reid v. Ingerman Smith LLP (E.D. N.Y Dec. 27, 2012) 2012 WL 6720752 observed that “legitimate expectations of privacy may be lower in e-mails or other Internet communications,” and certainly this view would extend to social-media sites shared with multiple “friends.”

It is somewhat inconsistent to acknowledge that you have more than 1,000 “friends” and yet argue that you have an expectation of privacy in your Facebook posts.  In Simply Storage, the court concluded that privacy concerns are limited on social-networking sites because individuals have already shared information with others through private messages or with a larger number of people through postings.  (270 F.R.D. at 437.)  Courts also recognize that any lingering privacy concerns may be dealt with through appropriate protective orders. (Ibid.)

Even in cases where the California constitutional right to privacy is raised, it is unclear whether discovery of  information from social-networking sites will be precluded. Under California law, the right to privacy is not absolute.  (See TBG Ins. Services Corp. v. Superior Court (2002) 96 Cal.App.4th 443, 339.)  Although Federal Rule of Evidence 501 requires that state privilege law applies in diversity jurisdiction cases, federal courts have not found that the California right to privacy necessarily bars production of social-media posts. (Mintz, 885 F. Supp. 2d at 995.)  In federal-question cases, the court will “take into account state privileges as a matter of comity where it can do so ‘at no substantial cost to federal substantive and procedural policy.’”  (ASIS Internet Servs. v. Active Response Group (N.D. Cal. 2008) 2008 WL 2129417 at *2 [even though customers have some privacy interest in e-mail addresses, disclosure of e-mail address lists was ordered]; see also Oakes v. Halvorsen Marine Ltd. (C.D. Cal. 1998) 179 F.R.D. 281, 284 [where a party’s need for information outweighs privacy interests, production of private information may be ordered].) Thus, there are no guarantees that information stored in social-media sites will remain private after litigation begins, even if maintained under a confidential setting.

Discovery of social media? A work in progress

Problems raised by discovery of social media are an extension of the now historical problems faced by litigants in the broader world of e-discovery. As usual, courts are “reacting” to the problem and somewhat behind the development of the technology in their reactions. Courts will continue to struggle to adapt traditional discovery principles – relevance, burdensomeness, and cost – to the new and evolving world of social media.

Judge Segal wishes to recognize and thank her Spring 2014 extern, Angela Davtyan, a law student at Southwestern Law School, for her excellent research and other contributions to this article.