GUEST BLOG by Anonymous.
Editors Note: this is the first contribution the e-Discovery Team blog has accepted from an anonymous attorney. We are only authorized to tell you that the author is a partner in a large national law firm. The quality of this writing speaks for itself and convinced us to accept this Open Letter to the Judiciary for publication. This is a scholarly, well written article. Since it is almost 11,000 words in length, we will publish it in two parts. The Open Letter contains great research, including 34 footnotes at the end. It is our hope that lawyers and judges everywhere will take the time to read this important plea for reform in confidentiality designations.
______________________
AN OPEN LETTER TO THE JUDICIARY – CAN WE TALK?
The Need to Up-Date the Judicial Approach to
Confidentiality Designations in the Era of ESI Discovery
Dear Your Honors – both Federal and State Judges:
I am writing to you today in hopes that we can have an open and frank discussion about the need to up-date the judicial approach to confidentiality designations of documents produced during the discovery phase of litigation, which presently requires, according to many judges, the producing party to engage in a document-by-document, page-by-page pre-production examination. In the era of ESI discovery, the burden this requirement imposes is staggering in terms of time and expense. I am writing you anonymously as I have been involved in a matter where this issue arose and I do not wish to be seen as rearguing an already decided matter or as challenging any particular judge. I accept what is past, but hope for a changed future for us all.
Your Honors, the need for change is great and immediate – and the goal of this letter is to raise for consideration and discussion how best to address the realities of contemporary document production and concerns for protecting confidential information within the framework of existing case law and the confines of the present rules of civil procedure. I – along with many practicing attorneys – anxiously await your response. A wide-reaching “discussion” took place over recent years concerning the need to alleviate the burden of pre-production privilege review, ultimately resulting in a revision to Federal Rule of Evidence 502. I submit to you that a similar resolution can be reached vis-à-vis confidentiality designations – a resolution that will not require an amendment to the present rules. Please – we need help!
The Problems Wrought by ESI
There is no need to delve deeply into an examination of all of the practical problems that have been wrought on discovery by the advent of the computer era. Over the last five or so years, libraries’ worth of materials have been written by judges, lawyers, and scholars on the topic, and conferences on the subject abound. We all know that ESI discovery is not going away and will remain a constant challenge to all of us, especially as technology continues to evolve at such extraordinary speed that what seemed reasonable and valid only a two or three years ago now seems out-dated and unacceptable. Indeed, the 2006 amendments to the Federal Rules of Civil Procedure were designed to allow for flexibility to address a constantly changing technology environment.[1]
What has become reality today is that in a typical civil lawsuit, once the case is ripe to proceed – and even before that time for those of us who have heeded the numerous admonitions urging attorneys to learn as soon as possible the intricacies of their clients’ ESI systems[2] –counsel will have to deal with multiple thousands, and often millions of pages of electronic documents that must be identified, collected and reviewed during discovery. And, this is true even after we have consulted with opposing counsel; identified the “key custodians” who most likely have relevant, non-duplicative or non-cumulative documents; agreed to and sampled “search terms”; tried to limit the time frame to a reasonable period in light of the allegations; and, concurred that unless shown to be needed, recovery of back-up tapes is not necessary.
In practice, what this very often means is that the “key custodians” will number between 5-12 (if not more); the list of search terms, with all the “stemming” variations that are needed to capture singulars and plurals, abbreviations, nicknames, expected typos and a host of other permutations, will number a hundred or more, the time period will extend back in time several years and may continue into the future, and “live” emails, email folders, email archives, databases, and hard-drives, just to name a few of the typical repositories, will have to be searched for potentially responsive documents. The same is true for “e-docs” – i.e., letters, memoranda, Excel spreadsheets and PowerPoint presentations that each of the custodians’ authored or modified during the given time period. And, let us not forget that for any of the custodians who rely on an administrative assistant who may have copies of earlier iterations of “e-docs” or have shadow copies of a custodian’s email files, the same search will need to be performed on his or her electronic records – meaning that the 5-12 custodian suddenly can become in reality 9 -15 or more people. Oh – and just to complicate matters further, we should not ignore the possibility of having to retrieve, one way or another, voicemails, text messages, webpage iterations, blog postings, and social media communications. We also need to make sure that potentially responsive documents are being properly preserved on a going-forward basis usually from a pool of custodians significantly larger than the “key” custodians identified for collection – such preservation representing a whole other effort that presents its own unique issues and problems to be negotiated, and often is the source of later spoliation claims.
Having taken all of these steps, which are now regarded as de rigueur[3] – and which are designed to narrow the universe of documents for potential production down from the impossibly enormous to the “manageable” tens or hundreds of thousands or more[4] – the attorney will now have to oversee a review process designed to sift the wheat from the chaff in terms of responsiveness,[5] locate privileged documents or privileged portions of documents from regular business communications – and address the business sensitivity of the information contained in the emails and email strings that typically make up the vast majority of what is produced in today’s discovery.[6]
It was this very process and the resulting large volume of documents that led to calls for relief in terms of privilege review. There, the issue arose because it was possible that the release – even if inadvertent – of a single email containing a privileged communication could result, depending on the jurisdiction, in complete subject matter waiver of all documents/information relating to that topic. Thus, by necessity, counsel would have to examine very carefully all potentially responsive documents to ensure that no stray privileged communication buried in an email was produced – and would have to do so at great cost in terms of both time and expense to the client. As Magistrate Judge Grimm rightly explained in Hopson v. Mayor/City Council of Baltimore, 232 F.R.D. 228, 232 (D. Md. 2005), before the 2006 amendments to the Rules of Civil Procedure or the 2008 amendment to Rule of Evidence 502, the most challenging aspect of ESI discovery is “how to properly conduct . . . discovery within a reasonable pretrial schedule, while concomitantly insuring that requesting parties receive appropriate discovery, and that producing parties are not subjected to production timetables that create unreasonable burden, expense, and risk of waiver of attorney-client privilege and work product protection.” He also explained that it is important to have a clear appreciation of what is at stake – i.e., “an awareness of the enormous costs that would accompany a requirement that in all civil cases the production of electronically stored information could not be accomplished until after a comprehensive privilege review and particularized assertions of privilege and work product claims.” Id. at 238 (emphasis added). As to the actual costs involved, Judge Grimm correctly explained that “The cost of responding to a discovery request can be in the millions of dollars if several years’ worth of archived e-mail and files must be located, restored, sorted through and cleansed to remove non-relevant confidential material.” Id. at 239 (citation omitted).[7]
Judge Grimm’s words – as correct as they were and are in terms of ESI discovery and privileged documents – are even more accurate in terms of the identification of documents that contain confidential business information. This is true for two reasons.
First, based on the recognition of the expense and risks involved due to a fear of subject-matter waiver arising from inadvertent disclosure, in 2008, Federal Rule of Evidence 502 was amended such that an inadvertent disclosure of a privileged communication will not result in waiver if reasonable steps had been undertaken to prevent disclosure and the producing party acted promptly to locate and seek the return of the documents. Fed. R. Evid. 502(b).[8]
Second, although little about e-discovery is easy, there are relatively simple ways to conduct a search of electronic documents that will identify many, although not all, potentially privileged documents. For example, once the universe of potentially responsive documents is narrowed from “all” documents throughout the company relating to the document request topics, down to the 40,000, 80,000, 200,000 or 1,000,000 or more documents collected from the “key custodians” and subjected to search term filtering, a separate search then can be run using the names of in-house counsel, the outside lawyers used by the company, the domain names of law firms, specific topics on which legal advise was sought and other such privilege-specific filters. Again, while such a search will not be perfect,[9] it will tend to capture the vast majority of potentially privileged communications and, thereby extend well down the road toward the “reasonableness” standard incorporated into Rule 502(b).
Such relatively simple and limited filtering, however, cannot be done for business sensitive information because the topics discussed in a business context are so numerous, the language so broad, the concepts so amorphous and the scope so vague. In other words, when faced with a document request that seeks, for example — “all documents, for the period January 1, 2007- April 17, 2011, referring or relating to prices, pricing plans, pricing policies, pricing trends, pricing conditions, competitive prices, and price studies, whether or not commissioned by the company, relating to the sale of widgets: a) in the United States; b) in North America; and c) all other non-North American foreign countries” – and the “key” custodians include the company’s CEO, CFO, vice president of sales, vice president of marketing, four regional marketing managers, five national sales managers, the director of international sales and her three immediate subordinates, the volume of documents that will deal in some way with “prices” and “pricing” will be so numerous that an effort to implement refining filter(s) would be pointless. Moreover, emails discussing pricing will very often also contain language on topics responsive to other document requests concerning, for example, marketing, advertising, competition, sales performance, etc., such that each email will be responsive to multiple requests and cannot be searched and reviewed in isolation.[10]
In short, having taken all the steps the courts and the commentators urge counsel to take to narrow the universe of documents for review and production, the end result is still thousands or millions of pages of documents to be reviewed and culled for production – with the costs, just as Judge Grimm predicted in 2005, reaching to or exceeding seven figures. Meanwhile, of course, opposing counsel is chomping at the bit to start receiving discovery and is sending ever more pointed and snide emails about delay and the need to proceed quickly.[11]
Confidentiality Designations – Ay, There’s the Rub
So, what is the big deal with confidentiality designations? The answer is that companies regard their internal thinking about a subject, their ideas, their proposals, their draft plans, etc. as business proprietary and not something to be shared outside of the company. Such documents may not rise to the technical level of a trade secret, but they certainly represent the inner workings of a company trying to do business in a competitive environment – a company that wishes to and instructs its employees to conduct those efforts behind the company’s doors – legitimately so.
There is nothing new about these sentiments. They have long existed. What is different is that in the pre-computer era, there would have been much less of a written record about all of these activities. Notes may have been taken at a meeting, but discarded three months later. Multiple drafts of proposals would have been tossed once a document was final. Today, however, saved emails (and e-docs) serve as an ongoing record of each idea, thought, suggestion, plan and concept that may have existed at one point in time. Some of these might have been accepted, developed and implemented, others rejected outright, while yet others may show a momentary lack of judgment or just be downright personally embarrassing to someone. And companies do not share these with the public or with others in the business. Certainly, no company would invite in a competitor or members of the press and give them complete access to the daily emails that are exchanged concerning the company’s on-going business activities.
Should all these documents be shared with the opposing party in response to discovery requests? Yes – if they are responsive. Would a company want them to be disseminated by an opposing party to the company’s competitors or the press? No. Is an opposing party allowed to do so?[12] Yes! [13] Yes, that is unless there is a protective order prohibiting such distribution.[14] So, isn’t a protective order the answer to this problem? No – because most courts have so narrowly interpreted the “good cause” provision in Rule 26(c) that a business party winds up facing the choice of either not pursuing the benefits of a protective order, or finds itself in the exact position about which Judge Grimm warned in his Hopson opinion – having to postpone producing documents until after it has done a comprehensive confidentiality review, with particularized confidentiality notations made on a document-by-document, paragraph-by-paragraph, and/or line-by-line basis. In the case of a production that involves the review of 500,000 documents, of which 75,000 are deemed responsive, with 55,000 containing some language of a business confidential nature, counsel will have to expend an inordinate amount of time – and, therefore, money charged to the client – to painstakingly mark as confidential those sensitive portions of each iteration of each document – and do so at great cost and within the relatively short time frames that are permitted for discovery, especially in a class action context where “class discovery” occurs early in the process before “merits” discovery proceeds forward, assuming a class is certified.
Are there solutions? Yes – both as established by judicial decisions and by the language of Rule 26(c) itself.
A. The Discovery Phase of Litigation Can be Conducted in Private without Implicating the “Public’s Right to Know”
A core issue that seems to be driving the courts to hesitate before granting broad protective orders during the discovery phase of litigation is the long-established recognition that litigation should be conducted in public. As explained in Rushford v. The New Yorker Magazine, Inc., 846 F.2d 249, 253 (4th Cir. 1988), “[u]nder common law, there is a presumption of access accorded to judicial records.” Moreover, the courts have recognized a First Amendment right of access to documents that are filed with a court or made exhibits to judicial actions. See, e.g., In re Parmalat Securities Litigation, 258 F.R.D. 236, 243 (S.D.N.Y 2009) (collecting cases and discussing presumption of access to judicial documents under both common law and First Amendment). See also In re Terrorist Attacks on September 11, 2001, 454 F.Supp.2d 220, 222 (S.D.N.Y.2006) (strong presumption of public access to “ ‘judicial documents,’ – i.e., such “items filed with the court that are relevant to the performance of the judicial function and useful in the judicial process.”) (quoting SEC v.TheStreet.com, 273 F.3d 222, 231 (2d Cir.2001)).
However – and this is a key point – the courts, including the U.S. Supreme Court, have consistently stated that there is no presumption of access, under the First Amendment or otherwise, to documents exchanged between parties during discovery – which is regarded as a private part of the litigation process.[15] In other words, there is a judicially recognized distinction between the hundreds, thousands, or millions of pages of documents that may be produced during discovery versus the significantly reduced subset of those documents that may become trial exhibits, or exhibits to motions filed with the court for consideration of a preliminary issue. Specifically, in Seattle Time Company v. Rhinehart, 467 U.S. 20 (1984), the Supreme Court stated “[P]retrial depositions and interrogatories are not public components of a civil trial. Such proceedings were not open to the public at common law . . . and in general they are conducted in private as a matter of modern practice.” Id. at 33 (internal citations omitted.) Because of the private nature of the discovery process – in contrast to the public treatment of documents filed with a court – the Court then held that the First Amendment does not represent an impediment to courts issuing protective orders to prohibit disclosure of discovery materials. Id. at 37.
Following from this, more recent courts have emphasized that “there is no constitutional or common-law right of public access to discovery materials exchanged by the parties but not filed with the court. . . . Unfiled discovery is private, not public.” Bond v. Utreras et al., 585 F.3d 1061, 1066 (7th Cir. 2009) (holding journalist had no right to intervene post-settlement to obtain confidential documents exchanged in discovery, but never filed with court).[16] Indeed, “[s]ecrecy is fine at the discovery stage, before the material enters the judicial record.” Id. at 1075 (quoting Baxter Int’l, Inc. v. Abbott Labs, 297 F.3d 544, 545 (7th Cir. 2002)). See also United States v. Dimasi, Cr. No. 09-10166, 2011 WL 915349, at *3 (D. Mass, Mar. 16, 2011) (right to access to “judicial documents” does not extend to documents provided in discovery); Flagg v. City of Detroit, 268 F.R.D. 279, 293 (E.D. Mich. 2010) (where case in discovery phase and documents not yet filed with the court, protective order valid and does not implicate public right of access; citing numerous circuit court decisions); In re Zyprexa Injunction, 474 F.Supp.2d 385, 413-416 (E.D.N.Y 2007) (no presumption of private access to documents exchanged during discovery and discovery protective orders serve essential function to protect privacy and property rights, including sensitive business information).
In light of these decisions, the courts should not view broadly permissive discovery protective orders as infringing on a presumption of public access or otherwise threatening the judicial process. However, that is precisely what the courts have done. The time to change that perception has come because of the exploding volume of documents collected, reviewed and produced since ESI discovery has become the norm.
B. Rule 26(c) Provides Authority to Allow For Expansive “Umbrella” Protective Orders to Limit Burden of Undue Expense
Rule 26(c) provides: “The court may, for good cause, issue an order to protect a party or persons from annoyance, embarrassment, oppression, or undue burden or expense . . . . (Emphasis added.) Moreover, Rule 1 states that “These rules . . . should be construed and administered to secure the just, speedy and inexpensive determination of every action and proceeding.” (Emphasis added.) The courts have said that these two rules should be read together to ensure the laudatory goal set forth in Rule 1 is achieved during the discovery phase of litigation. Zyprexa, 474 F.Supp.2d at 414.
In applying these rules, the courts have focused very closely on Rule 26’s “good cause” language, recognizing that large portions of documents produced in discovery could cause embarrassment, harassment or oppression of a party (or a non-party) if the information were made public and that for those reasons “confidential” treatment during discovery is justified. In matters where many such documents likely will be produced, the courts find “good cause” to issue the protective order sometimes in the form of an “umbrella” protective order or much more frequently a “blanket” protective order. “Umbrella” protective orders designate all discovery as protected without any prior review by either the courts or the parties, whereas “blanket” protective orders place the initial burden on the parties to review the documents and determine “on a good faith basis” what is entitled to protection and so designate only those documents. Marolf v. AyA Aguirre & Aranzabal S.A., No. 4:09CV3221, 2011 WL 855676, *4 (D. Neb. Mar. 8, 2011). In other words, although a blanket protective order permits all documents to be designated as confidential, a party must only make such designations in “good faith” – on a document-by-document basis. Id.
In practice, however, the wide-spread use of broad umbrella protective orders is not favored.[17] Rather, the courts prefer blanket protective orders to address confidentiality concerns. However, a blanket protective order actually only increases the burdens on a producing party because they require a document-by-document review and designation – and woe to the producing party who is seen as over-designating, as demonstrated in Part Two of this letter.
I submit to Your Honors, that this result is unjust and forces a producing party to expend tremendous resources in time and money conducting the very type of line-by-line review that was deemed threatening to the entire discovery process when the focus was solely on privileged documents – as well as placing on the producing party’s shoulders an exercise that the courts recognized would be excessively burdensome if the court were required to engage in the same process.[18] See, e.g., Van v. Wal-Mart Stores, Inc., C 08-5296, 2011 WL 62499, at *2 (N.D. Ca. Jan. 7, 2011) (blanket protective orders avoid the undue cost and delay if courts had to make a good cause determination on a document-by-document basis for all documents exchanged in discovery) (emphasis added).
In other words, the courts should recognize because: a) there is no right of public access either by common law or under the First Amendment to documents exchanged in discovery; b) discovery is considered “private” and not part of the public aspect of the judicial process; c) the vast majority of documents produced in discovery will never be presented in court by way of motion or trial exhibit; d) companies have a legitimate interest in protecting their business confidential documents from dissemination to the public, competitors or the press; and e) the time and cost necessary to painstakingly review the extraordinary volume of documents that accompany ESI discovery is extremely burdensome, a producing party should be permitted to treat all documents produced as “confidential” at the on-set, with specific ultimate decisions made about confidentiality later in the process when a much smaller group of documents are set to be submitted in court for consideration on some issue. Moreover, such a procedure is permissible under Rule 26(c) given that it allows for protective orders to be issued to protect a party from “undue burden or expense” separately and apart from concerns as to whether a document that will never see the light of day in court actually meets the criteria to sustain a confidentiality designation. Thus, the “good cause” analysis should move away from considerations about public access to the judicial process – as none exists vis-à-vis the discovery process – and instead should focus on the costs and burdens associated with ESI discovery. This is not, however, what the courts have done.
______________
End of Part One.
Editors Note: Part Two will be published next week. It will contain the promised demonstration with case law discussion, plus an Analysis of the problem and the author’s proposed Solution. Know a judge? Please tell them about this letter.
Footnotes for Part One:
[1] Fed. R. Civ. P. 34(a) advisory committee’s note to 2006 amendments (“Rule 34(a)(1) is intended to be broad enough to cover all current types of computer-based information, and flexible enough to encompass future changes and developments.”)
[2] See Fed. R. Civ. P. Rule 26(f) advisory committee’s note to 2006 amendment (“important for counsel to become familiar with [client’s information systems] before the [Rule 26(f)] conference”); see also Magistrate Judge Andrew J. Peck, The Federal Rules Governing Electronic Discovery (revised) October 8, 2008, at 4 (“lawyers can no longer afford to be computer-illiterate”); Qualcomm Inc. v. Broadcom Corp., 2008 WL 66932, at *9 (S.D. Cal. 2008) (”[f]or the current ‘good faith’ system to function in the electronic age, attorneys and clients must work together to ensure that both understand how and where electronic documents, records and emails are maintained . . . .”) (emphasis added), vacated in part on other grounds by Qualcomm Inc. v. Broadcom Corp., 2008 WL 638108 (S.D. Cal. 2008).
[3] See, e.g., Trusz v. UBS Realty Investors, LLC et al., No. 3:09 CV 268, 2010 WL 3583064, at *5 (D. Conn. Sept. 7, 2010) (court expects counsel to reach practical agreement, without court having to micro-manage ediscovery, about “search terms, date ranges, key players and the like”) (quoting Allman, Conducting E-discovery After the Amendments: The Second Wave, 10 SEDONA CONF. J. 215, 217 (2009)), reconsideration granted in part on other grounds, 2011 WL 124504 (D. Conn. Jan. 13, 2011).
[4] The volume of documents that are retrieved can vary tremendously depending on a company’s electronic document retention policies. While some companies strictly limit the volume of emails that an employee may maintain and regularly purge the company’s systems for emails older than, for example, six month or some other short time period, many businesses allow for the creation of email folders which are not purged or utilize a form of email archiving that allows the employee to retrieve emails from prior periods. Similarly, there are people who are fastidious about deleting emails on a regular basis, essentially saving nothing, and then there are “hoarders” who keep nearly everything. If the “key custodians” include several “hoarders” the universe of emails to review even over a single year for each such person can be in the tens of thousands, thereby increasing substantially the time and expense necessary to review and produce.
[5] It should be understood that even the best crafted of search terms will result in the identification of many, many documents that contain one or more search term “hits,” but which are completely meaningless in terms of responsiveness. Sometimes such “non-responsive” documents relate to other parts of the company’s business and, unfortunately from a document collection and review perspective, simply contain some of the same key terms that are relevant vis-à-vis the portion of the business involved in litigation. Other times, search terms will lead to the collection of personal emails on family matters, soccer games, doctor’s appointments, etc., and then in almost every case in which I have been involved, the key word search process captures caches of pornographic materials of some sort or another. The limitations of keyword searching, even one utilizing sophisticated Boolean and proximity search methods, have been well-recognized by the courts and discussed in various commentaries. See, e.g., Victor Stanley v. Creative Pipe, Inc. (Victor Stanley I), 250 F.R.D. 251, 259-260 (D. Md. 2008) (discussing drawbacks and complexities with key word searching and citing other decisions); United States v. O’Keefe, 537 F.Supp.2d 14, 24 (D.D.C. 2008) (“Whether search terms or ‘keywords’ will yield the information sought is a complicated question involving the interplay, at least, of the sciences of computer technology, statistics and linguistics. . . . Given this complexity, for lawyers and judges to dare opine that a certain search term or terms would be more likely to produce information than the terms that were used is truly to go where angels fear to tread.”)
[6] As many of us are aware, more and more tools are being developed and marketed to help refine and “simplify” the review process. Emails can be “de-duped” electronically – both within a custodian’s files and across custodians’ files – although often opposing parties (or government attorneys) will insist that cross-custodian de-duplication is not acceptable from a production perspective. Likewise, duplicates or “near duplicates” can be identified electronically and gathered together around a “pivot” document so that all dupes and “near” dupes can be reviewed and tagged at the same time and consistently for responsiveness, specific claims, issues and defenses, privilege or redaction. In addition, we are told that “concept searching” and “predictive coding” will soon make eyes-on review a thing of the past – or will result in a greatly reduced exercise. Perhaps that will be true, but as of yet, those cutting-edge tools remain on the horizon for most practitioners and certainly have not yet entered the main stream of legal practice. Thus, at least for the time being, the process described above will remain the most typical.
[7] While these staggering costs are daunting enough to the parties in active litigation, they represent an even higher burden on the recipient of a third-party subpoena under Rule 45 as that third-party has no stake in the litigation and yet finds itself having to incur significant costs to preserve and produce documents in response to a subpoena. And while Rule 45 includes language to protect the subpoena recipient from facing undue burden and expense, even a relatively limited request for ESI can still cost thousands of dollars. See generally Commentary on Non-Party Production & Rule 45 Subpoenas (The Sedona Conference Working Group Series, April 2008).
[8] In a recent article penned by Magistrate Judge Grimm, Lisa Yurwit Bergstrom and Matthew P. Kraeuter, the authors explain that while Rule 502(b) requires that to avoid waiver of an inadvertently produced privileged document, the producing party must have taken “reasonable steps” to prevent disclosure and acted promptly to rectify the error, Rule 502(e) (read along with Rule 16(b) and 26(f)) recognizes that parties may enter into their own agreement and that nothing in this subsection of the rule requires “reasonable precautions.” Indeed, “these rules would permit the parties to agree that discovery material could be produced without any pre-production screening at all, but be “clawed” back upon demand after production.” Paul W. Grimm, Lisa Yurwit Bergstrom and Matthew P. Kraeuter, Federal Rule of Evidence 502: Has It Lived Up to Its Potential, 17 Rich. J.L. & Tech. 8, at *76 (Spring 2011). The authors note, however, some judges decline to enforce such agreements because of the lack of “reasonable steps” to avoid disclosure – and posit that such a reading of Rule 502(e) is not correct. Id.
[9] Aside from the fact that no key word search will likely ever result in “perfect” results, one of the reasons that such a search will typically not capture all privileged communications is because an attorney may have a common name – Joe, Bob, Deb, Anne – and internal email between two business people can say nothing more than “I spoke with Bob today and he has concerns about the most recent draft of the new language. Let’s talk.” Assume outside counsel’s name is Bob Brown, is this the “Bob” referenced in the email? If so, is this a privileged communication? Maybe . . . Maybe not. But, someone needs to consider the question. And because a search for “Bob” throughout the entire universe of emails would almost certainly identify thousands of irrelevant emails from, to or referring to a number of different “Bobs” in the company or customer representatives named “Bob,” etc. – not to mention someone’s cousin Bob, someone getting a new bob hair cut, and plans to bob for apples at an upcoming Halloween party – a filtered search for just “Bob” would be futile and a waste of time. Thus, this type of email communication will most likely only be identified as potentially privileged through eyes-on review – after which dupes or near dupes can be located and a privilege determination made.
[10] Another thing that will increase the volume of documents for review and then production are the documents that are technically responsive, but utterly meaningless in terms of substance. For example, Sam sends an initial email to four people saying “Can we get together this afternoon to go over pricing for next summer’s sales promotions? What’s a good time?” – Sue writes back to all four people “Fine. How about 3:00?” . . . Dave responds “3:00 won’t work, I have a call – can we say 3:30? . . . Sue writes back “3:30 is good for me.” . . . Mike then chimes in with “Whoa – no can do 3:30, but I clear up at 4:30” . . . Sue shoots back “That is good for me too” . . . Sam responds with “OK to 4:30” . . . Sue emails “OK” . . . Mike emails “Great” . . . Dave responds “Good for me” . . . and finally Sally, the fifth person who previously had not joined the discussion gets back to her desk, reads all of these exchanges and emails everyone saying “I can be there too.” Sally then forwards the entire chain to her secretary and says “Put this on my calendar for 4:30. Thanks.” While only Sam’s initial email contains anything substantively responsive (albeit marginally so), because each new exchange in the email chain contains a copy of that initial email, all eleven of the subsequent emails, shared among all five people, plus Sally’s secretary, are technically responsive and, therefore, should be produced. Even if cross-custodian de-duping is permitted, each iteration with each subsequent response will have to be produced at least once. What a waste of time – but very typical of the kind of thing one sees in reviewing documents and in making production decisions.
[11] The impatience of opposing counsel is especially so where one party has very limited documents – such as is usually the case in the consumer or employment class action context – or when documents are being sought by a government agency. See e.g., Hopson, 232 F.RD. at 245 (electronic discovery not played on level field in class actions as plaintiff typically has relatively few documents while defendant has an immense volume).
[12] In a situation where two business entities are opposing parties, there will be less incentive to release the other party’s information as “sauce for the goose is sauce for the gander.” However, counsel for an individual or a consumer class would have no such countervailing constraints or incentives.
[13] See, e.g., Phillips v. General Motors, Corp., 307 F.3d 1206, 1210 (9th Cir. 2002) (citing San Jose Mercury News, Inc. v. United States Dist. Ct., 187 F.3d 1096, 1103 (9th Cir.1999)) ( “[i]t is well-established that the fruits of pre-trial discovery are, in the absence of a court order to the contrary, presumptively public); U.S. ex rel. Davis v. Prince, 753 F.Supp.2d 561, 567-68 and n.7 (E.D. Va. 2010) (citing numerous cases for proposition that absent protective order, litigants may use information learned in discovery in any way they wish, including public disclosure).
[14] Issues of dissemination typically do not arise in the case of a government investigation where a company receives an informal letter inquiry requesting documents, a civil investigative demand, an administrative subpoena or some other mechanism by which the government requests information and documents. In such situations, there are statutory or regulatory provisions that serve to allow the information and documents produced to be designated confidential such that they are not otherwise distributed. See, e.g., 17 C.F.R. §200.83; 15 U.S.C. § 18(a)(h). Thus, a respondent to a government inquiry typically designates all documents as “confidential” in one way or another. Of course, if private litigation follows or commences along side of the governmental inquiry, the private plaintiff very typically will begin discovery by asking for all documents submitted to the government – at which point the issue of confidentiality and protection of that information arises.
[15] Best Practices Addressing Protective Orders, Confidentiality & Public Access in Civil Cases, 8 SEDONA CONF. J. 141, 145 (2007) (“Principle 1: There is no presumed right of the public to participate in the discovery process or to have access to the fruits of discovery that are not submitted to the court.”)
[16] The Bond court further explained that while various decisions speak broadly of a presumption of public access to discovery materials, those decisions sweep too broadly because “while the public has a presumptive right to access to discovery materials that are filed with the court, used in a judicial proceeding, or otherwise constitute “judicial records,” the same is not true of materials produced during discovery but not filed with the court.” 585 F.3d at 1073 (emphasis added).
[17] See, e.g., Citizens First National Bank of Princeton v. Cincinnati Insurance Co., 178 F.3d 943, 945-946 (7th Cir. 1999) (recognizing a minority judicial view that umbrella protective orders as “unproblematic aids to the expeditious processing of complex litigation,” but finding the majority of courts endorse a presumption of public access to discovery materials). Aside from the fact that later courts have adopted a recognition that there is no public right of access to discovery materials, the “majority” view described in this case developed prior to the advent of massive and voluminous electronic discovery.
Let’s hope this is the start of a dialogue. There are many issues similar to the issue of confidential information in the discovery process which would benefit from streamlining. If this one can be addressed, perhaps others can as well.
Ralph, this is a great letter. Thank you for sharing it with your audience. Hopefully some of the issues raised by the author will start to gather momentum through subsequent discussions. I’m looking forward to part II.
This issue represents a tempest in a tea pot.
For 40 years my grandmother was secretary to the Dean of Georgetown Law School. She taught me to type, but more importantly, she taught me how to structure a business communication and how to file. Confidential information either had a ‘Confidential’ designation in the subject line of the letter’s salutation or had ‘Confidential’ typed as a footer, and was then filed appropriately.
When computer companys developed hardware and software for business, the promise was to replace secretarial and clerical workers plus increase everyone’s productivity. They developed tools that mimiced everyday life. Hence we have ‘edocs’ and the like. One of the things they provide us is metadata fields for office documents and fielded flags for messaging communications (emails, etc.) Any email can be flagged confidential and any document can be meta tagged in a similar fashion.
If a litigant fails to make use of the tools, or fails to train and manage their employees to consistently use the tools they 1) either bear the cost of their failure as a confidentiality review or 2) run the risk of waiver. This is no different now than it was in 1960 when my grandmother was typing her letters.
Many lawyers fail to understand these simple concepts as soon as you put ‘e’ in front of anything. As long as prominent lawyers continue to wear computer illiteracy as a badge of honor, we will continue to see these well meaning but ignorant discussions. The only thng worse is when a ltitgant does a good job here only to have the flags lost because lawyers refuse to properly handle the metadata.
Every person and business today bears some risk of involvement in litigation. Every person or business shares the responsibility to manage their information assets properly. The tools are routinely provided. If they fail to act responsibly, why should they be given a ‘pass?’
Dear WTKJD,
I suspect your grandmother was a wonderful person and the Dean was lucky to have her do all of his typing and filing for him. I recall that only 30 years ago I had a secretary who typed and filed paper letters for me too. On a busy day we would send out two, maybe three letters, and she would file for me the four or five others letters I would receive in a day. It was customary to respond to a letter in a week, and the mail took another two to four days. We had time to think, to do draft after draft.
But those days are long gone. Now we all get 100 emails a day and some people expect a reply within hours, not days. The overall volume of our written communications has increased thirty-fold, if not much, much more. To make matters worse, we have also become our own typists and our own record keepers. None of us have a wonderful person like your grandmother anymore to carefully type and legend each and every email and attachment. Not only that, the country is much, much more litigious than in your grandmother’s day.
In the real world of today’s fast paced technology this is a serious problem. We are now all part of global commerce where U.S. companies no longer dominate like they did in your grandmother’s day, or even your mother’s. We must be sensitive to runaway litigation costs. We cannot continue to handicap U.S. companies by subjecting them to a higher litigation tax than anywhere else in the world, by far. This is especially important during this drag-on recession we are still in.
The relief requested here is reasonable. It does not give companies a “pass,” as you suggest, it gives them a fair chance to have disputes decided on the merits. That is a concept both your grandmother and her boss would approve. Let’s be realistic, in today’s world it is easy to make an over-broad request. All too often these requests are approved by courts and a company is forced to spend millions just to reply without waiving their legitimate rights to confidentiality. It is naive not to realize that many of these requests have ulterior motives, either to inflate settlement value, go on a fishing expedition, or both.
Let’s get back to focused, reasonable discovery. Let’s not just talk about Rule 1, let’s see it enforced. Let’s be much more liberal in granting umbrella orders. Let’s stop kidding ourselves about the so called “protection” of blanket orders.
Yours Truly,
Ralph Losey
Excellent post. Await Part II.
I will likely be using this issue for developing case studies in judicial education for Florida lawyers.
Ralph,
Thank you again for posting such a great masterpiece.I got the chills when I read this posting. This is so true. Knowing the cost associated with preservation, collection, culling, reviewing and production, there has to be a supreme rule that dictates the discovery as a private channel. I really like the paragraph that begins with ” what is the big deal with confidentiality designations? The answer is that companies regard their internal thinking about a subject, their ideas, their proposals, their draft plans, etc. as business proprietary and not something to be shared outside of the company. Such documents may not rise to the technical level of a trade secret, but they certainly represent the inner workings of a company trying to do business in a competitive environment – a company that wishes to and instructs its employees to conduct those efforts behind the company’s doors – legitimately so.” Providing a blanket confidentiality provision to discovery material may even expediate the whole discovery process. Further, protective order should have clauses in them that dictates return of confidential documents and provision of sanctions if violation of such term occurs. How painful it is to apply redaction on each and every single piece of document if no such protection is granted- only a Litigation paralegal or litigation support team will know!!!
WTKJD’s analysis above posits that because every document COULD be tagged with a “confidential” metatag, that every document WILL be so tagged by users and that if they are not, it is by definition the user’s fault. This flies in the face of reality and human nature. If the writer’s grandmother was truly as diligent as described, then she was an exceptional and very rare person.
What the writer fails to note is that while his/her grandmother’s ‘Confidential’ designations would have been a useful starting point for a review, those designations would be neither necessary nor sufficient. Despite her diligence, both untagged confidential documents and overclassified documents undoubtedly existed. No responsible lawyer would have depended on such a tag as the sole authority even in an all-paper environment. The problem now is worse.
The normal employee is overwhelmed with responsibilities and pressured by time. Applying confidentiality flags to documents is and must be a manual process. (As the author notes above, there are far too many topics and ways of discussing them for automated confidentiality tagging to be feasible anytime in the near term.) Applying confidentiality flags to documents intended for purely internal consumption is arguably an irresponsible use of scarce resources. If the document is already protected from disclosure or leakage by the business’ IT defenses and practices, the tag is unnecessary. (The theoretical possibility of future litigation should NEVER be the driver of business process. It is, for example, specifically excluded as a required basis for a company’s records retention standards.)
The writer further assumes that the tag will be usable for confidentiality filtering in a litigation context. The truth is that in any organization with a structured classification system, significant errors occur. Again, users tag the documents by hand. Even among the most diligent, some will be missed. More commonly, users will over-classify their documents, alleging confidentiality over communications which are not really sensitive. This can occur through a misplaced sense of self-importance, a misunderstanding of the situation or a sincere desire to be “better safe than sorry”. The few companies which do use confidential flags, in my experience use them only because they are set as a default to a document’s template. This does not meet the current “good faith” standard which the anonymous author described so well.
The reality is that unless you are working with the Department of Defense (and often even then), confidentiality coding is and always will be a low priority for users. Trying to assume differently and blaming the clients when their (perfectly legal) business processes don’t match YOUR priorities is irresponsible.
[…] Open Letter to the Judiciary – Can We Talk? (Part One) – http://tinyurl.com/3p8hbzp (eDiscovery […]
[…] Open Letter to the Judiciary – Can We Talk? (Part One) – http://tinyurl.com/3p8hbzp (eDiscovery […]
[…] please share this Open Letter to the Judiciary with state and federal judges that you may know. Here is the link to Part I of the Open Letter, which should, of course, be read […]
Can we talk, catchy title and good read!
[…] PART I: “An Open Letter to the Judiciary – Can We Talk? (Part One)” […]
[…] an enormous problem in litigation. See Anonymous, An Open Letter to the Judiciary – Can We Talk? Parts One and […]
[…] An Open Letter to the Judiciary: Can We Talk? (part one) Read article » […]