The special challenges of electronic records discovery, including the exponential growth of electronic data, present novel and complex questions about what records litigants should be required to retain and retrieve and how courts should manage the often staggering cost of responding to requests for electronic discovery. Litigants and courts continue to struggle with these important questions in an ad hoc and sometimes inconsistent manner. As a result, at the federal level the Federal Rules Advisory Committee is considering rule changes designed to provide more guidance on how to handle electronic discovery.
This article is part one of a two article series dealing with the emerging area of electronic discovery. Part two of this series will address how to organize an effective response to electronic discovery — from assembling the right team of professionals to the basic steps that should be considered at the preservation, collection and production stages of electronic records discovery. This article discusses why, in most cases, the discovery of electronic records is more complex and expensive than the discovery of paper records. For example, this article cites some shocking statistics about the proliferation of electronic data, which is exceeding in a very short period of time the volume of paper records acquired by institutions over decades or even centuries. It also discusses why electronic records are generally more widely disbursed and individually controlled than paper records and why electronic records typically require the assistance of skilled computer technologists to locate and assemble for production. In addition, this article explains how backup tapes, increasingly sought after in discovery, were designed as a safety net strategy for disaster recovery, not for retrieving information in the ordinary course of business, and why such disaster recovery media often can be restored and searched only at very significant time and cost to an organization. Finally, this article also addresses briefly why the risk of inadvertent waiver of privileges is greater when producing electronic records and how the increased risk is contributing to the high costs of electronic discovery.
Discovery of electronic records is generally more complex and expensive than discovery of paper records.
The volume of electronic information available through discovery has grown dramatically and has exceeded in a very short period of time paper records acquired over decades. Today’s technologies are resulting in the generation of electronic data compilations at breathtaking speed. The fastest-growing component of electronic communication today is e-mail. (Dreyer, When the Postman Beeps Twice: The Admissibility of Electronic Mail Under the Business Records Exception of the Federal Rules of Evidence, 64 Fordham L. Rev. 2285 (1996)). As of 1996, there were approximately 20 million users of e-mail in the United States. Ibid. The International Data Corporation estimates that 31 billion person-to-person e-mails were sent each day in 2002. This number is expected to grow to 60 billion each day by 2006. Johnston, You’ve Got Mail!, IDG News Service, (Sept. 26, 2002).
The average office worker in the United States sends an estimated 60 to 200 e-mail messages each day. Craine, Here Come the Lawyers. Is Your IT Department Ready?, at http://solution/educomts.com/pr ess/artarchive.asp (June 20, 2002) An industry technologist for a large financial services company acknowledged that the company’s amassed e-mail equaled in volume all of its paper records combined, for all of its customers, for the company’s over 100-year existence. The bulk of the company’s e-mail was generated after 1996. In 1996, the company had around 6,000 users. Four years later, it had in excess of 78,000 users. The staggering increase in e-mail seems more comprehensible when considering that e-mail is, in many instances, the memorialization of what used to be a telephone conversation. Imagine the quantity of data stored in a 10-year period if every telephone conversation was memorialized in a document.
While the other forms of electronic records may not have received the attention and experienced quite the growth of e-mail, clearly electronic records of all kinds are growing with similar speed. For example, from 1998 to 1999, the worldwide software management market revenue increased 47 percent. That same market is expected to grow from $4.2 billion in 1999 to $14.7 billion by the end of 2004. Press Release, Gart-nerGroup Dataquest, Gartner’s Dataquest Says Worldwide Storage Management Software Industry to Reach $14.7 Billion in 2004. It has been estimated that within 10 years, the total number of electronic records on the planet could double every 60 minutes.”
Lysakowski, Titanic 2020 — A Call To Action, at www. censa.org/ html/f_news.html (December 28, 1999).
The sheer volume of electronic records, and the extent to which they are dwarfing the volume of paper records in our new economy, often makes electronic discovery more challenging than paper discovery. Unfortunately, the volume of electronic records is not the only new challenge facing litigants.
Electronic records are often more widely disbursed and individually controlled than paper records and typically require the assistance of skilled computer technologists to locate and assemble for production.
In the paper world, records are typically managed by requiring individuals to move their records to a centralized storage facility. This centralized management of paper records is almost entirely non-existent in the electronic world. As a result, electronic records are often more widely dispersed and individually controlled than paper records, which means that such records can be more difficult and expensive to locate and assemble.
For many institutions, there are dozens — sometimes hundreds — of different computer systems, exclusive of PCs, laptops and hand-held computers. For example, companies often have not only mainframe computers, but local area networks, wide area networks, storage area networks, various servers and other systems. In addition to the number of different systems that store electronic information in different places, these various systems often store information in different ways, which complicates the task of searching for records. Another factor complicating the search for electronic records is that computer systems sometimes move data around throughout an institution to the most under-utilized location, which may change several times during the course of a week. Typically, all of these computers and their output are out of the control of the centralized records management program that manages paper documents.
The diversity of document or file names also contributes to the challenge of finding widely dispersed electronic records. “Naming conventions” are hardly universal due to variant business needs. This search difficulty is compounded because many electronic records are only identifiable by their file name, a name that is constrained by the number of characters in the name, the creativity of the author and the fact that, consequently, the creator of the record may be the only one able to recognize the record from looking at the file name. Without an index or some other means for getting to the record, finding it among thousands of other records stored on the same device presents additional challenges. In some cases, this task could be analogous to trying to locate a book in a library without a card catalogue and without titles on the books.
Another aspect of decentralization of electronic records is the continuing proliferation of mobile technologies being used by employees. For example, one laptop can store hundreds of thousands of pages, as can that same employee’s desktop. In addition, there are increasingly large volumes of data and electronic records being created and retained on a variety of hand-held computers, personal computers and diskettes, which may be in the control of individuals only.
In summary, an institution facing a broad request or order to produce electronic records not only has a virtual ocean of electronic data it could search through, the institution also must contend with such complicating factors as: the wide dispersion of data among numerous diverse computer systems; the movement of data to take advantage of storage space in under-utilized parts of a network or system; variations in document or file names which sometimes make searches for particular documents or types of documents akin to looking for a needle in a haystack; and the rapid growth of mobile technologies. This means that discovery in the electronic world requires a very different skill set than discovery in the paper world. In the paper world, all that may be needed to collect responsive records is a lawyer working with a user and records-management professional. In the digital world, searches for electronic records sometimes require a team of computer technologists, who may be forced to shut down systems, shift staff and abandon maintenance of mission critical systems in order to search for electronic records at significant expense to the management of an on-going business.
Computer back-up systems are designed for disaster recovery, not for retrieval in the ordinary course of business, and restoring and searching them can be very time consuming and expensive.
While many stored paper records are not accessed in the normal course of business, they are relatively accessible if it becomes necessary to search them. Electronic data on back-up tapes, however, can present an entirely different situation. Computer crashes, power outages and operator error can sometimes destroy documents that are being maintained on computer systems. Just as certain storage precautions are taken to minimize the risk of loss of paper records, computer systems use back-up tapes to minimize the risk of losing years of work and business advantage, and untold millions of dollars, if significant information is lost in a computer crash or other disaster. Therefore, back-up tapes typically are not intended for anything other than the mere possibility of a need to recreate information. Given the unlikely occurrence of either a natural or unnatural disaster, little regard is given to search capabilities since, post-disaster, the goal would be to restore all of the data rather than to search for and recover data on a piecemeal basis. These back-up systems sometimes “archive” the data in such a way that allows it to be compressed into as small a storage space as possible, which requires the data to be unarchived or restored before it can even be searched. The process of restoring archived electronic media into readable form and searching it can take skilled technologists lengthy periods of time, even if they know precisely what they are trying to find, and costs astonishing sums of money. For example, in Linnen v. A.H. Robins, 1999 WL 462015 (Mass. Super. 1999), the expense of restoring backup tapes cost an estimated $1.75 million.
The risk of inadvertent waiver of the attorney client privilege is greater when producing electronic records and is contributing to the high costs of electronic discovery.
The attorney-client privilege is the oldest of the privileges for confidential communications known to the common law. Its long-standing purpose is to encourage full and frank communications between attorneys and their clients and thereby promote broader public interests in the observance of law and administration of justice. The privilege, however, can sometimes be waived even through inadvertent disclosure. Although courts apply the doctrine of waiver differently, the inadvertent disclosure of a privileged document can sometimes lead to the other side being able to use the document in court, or even result in a finding (under the doctrine of subject matter waiver) that any privilege for documents on the same subject matter has been lost.
In the paper world, protecting against disclosure of attorney-client privileged documents is relatively straightforward: communications from lawyers typically arrive on law-firm letterhead, announcing to the reviewer that the document is likely to be privileged. Similarly, communications from a client to an attorney will usually contain title or address information that reveals that the recipient is an attorney. Copies of the documents involved will typically be kept in a relatively discrete and central location within an organization. Collection of the documents at issue for review, therefore, is relatively straightforward. Finally, paper records typically involve a relatively small number of copies, perhaps just one file copy with enclosure.
These clues to the privileged nature of paper communications, however, rarely exist in the electronic-records world. Communications via e-mail, a more casual mode of communication in the first instance and one that increasingly serves as a substitute for telephone conversations, often identify recipients and senders by name only — without the benefit of an attorney’s title or law-office designation. Perhaps more confounding is the lack of a discrete and central location for such records. The records may be residing on hard drives, back-up systems or on servers, among other places.
E-mail will often have attachments, named or unnamed, and will sometimes require the use of software other than what is used by the receiving party to open. For example, use of different word-processing software may change the format or content of information shared by e-mail or, may render it altogether inaccessible due to the incompatibility of the two e-mail programs, thereby rendering one party incapable of ever opening the attachment to determine its contents, whether for privilege or otherwise. Such information, even if deleted, may nonetheless continue to exist somewhere on the computer.
The proliferation of electronic records is often a function of the ease with which they can be sent. A communication in the paper world requires at least a modicum of effort: someone must prepare a letter; copy the enclosure to be sent therewith; prepare an envelope, label or transmittal sheet; affix postage; and physically transport that piece of correspondence to a mail repository or to a fax machine. A communication in the electronic world, however, is as simple as typing the name of the recipient in the “to box,” identifying an attachment of any size with the click of a mouse and clicking send. Moreover, mistakes in communication of a privileged document can be corrected in the paper world; documents can be retrieved or destroyed. In the electronic-records world, however, a record takes on a nearly indestructible life of its own; it is stored by the file server through which it is sent and will reside within the electronic files of the recipient and his or her file servers once a communication is opened by a recipient and may physically reside in a place thousands of miles away.
Because a claim of privilege traditionally must be made on a document-by-document basis, a company wishing to protect against the inadvertent waiver of the attorney-client privilege in the digital age may have to review tens of thousands or even millions of pages of electronic records which, as noted above, exist in scattered physical locations and may take numerous forms. Such a task can often be a practical impossibility or, at best, extraordinarily time consuming and expensive. Litigants facing this situation have to weigh the risk of inadvertent disclosure against the expense of reviewing large volumes of individual electronic records. Courts have ruled that where the producing party undertook reasonable steps to detect and exclude privileged materials from voluminous productions, the inadvertent production of a proportionally very small number of privileged documents did not waive the attorney-client privilege. See, Fleet Nat. Bank v. Tonneson & Co., 150 F.R.D. 10 (D.Mass 1993) (two privileged documents out of 50,000 documents reviewed for production); Int’l Digital Sys. Corp. v. Digital Equip., 120 F.R.D. 445 (D.Mass 1988) (20 privileged documents out of 500,000 ); Commerce & Indus. Ins. Co. v. E.I. Du Pont De Nemours & Co., 2000 WL 33223235 (Mass. Super. 2000) (1,500 pages privileged documents out of 300,000 documents).
Unfortunately, the balancing test used by some courts to evaluate whether inadvertent disclosure results in waiver does not provide much predictability or protection. For example, in CIBA-Geigy Corp. v. Sandoz Ltd., 916 F.Supp. 404 (D.N.J. 1995), the U.S. District Court of the District of New Jersey held that defendants waived attorney client privilege regarding documents by inadvertently producing the information. Plaintiff’s counsel had obtained a corporate database during discovery and had discovered “privileged” documents by conducting a search within the database. Defendants claimed that its failure to review documents for privilege prior to designating them as privileged exhibits was a result of their mistaken belief that all documents on the database had previously been produced to plaintiff and that none of the documents contained on the database were privileged. The court stated that absent reasonable precaution to preserve confidentiality, there is a presumption that inadvertent disclosure of a document falling within the attorney client privilege is the result of gross negligence or intentional conduct, waiving the privilege. Ibid; Accord U.S. v. Keystone Sanitation Co., Inc., 903 F.Supp. 803 (M.D.Pa. 1995).
With results such as the one in CIBA-Geigy Corp. v. Sandoz, the risk of inadvertent waiver of privileged communications is real and, consequently, it is driving up the cost of responding to requests for electronic discovery. Therefore, the Federal Rules Advisory Committee is evaluating the merits of Texas Rule of Civil Procedure 193.3(d), which explicitly abolishes the “inadvertent waiver” rule.
The proliferation of electronic information is having a profound impact on the management of information in the business world and, understandably, it is impacting the process of discovering information in litigation. For both the business world and our civil justice system, it will be a challenge to keep up with these changes. As a practical matter, with the amount of electronic information being generated today, much less with how much will be generated in the years to come, litigants simply cannot save everything, nor can they approach the discovery of electronic information in the same way litigants have approached the discovery of paper records in the past. The realities of electronic discovery require that litigants and courts develop practical solutions to these new challenges. In that regard, part two of this series will focus on practical steps to organizing an effective response to a request for electronic discovery.
This article originally appeared in the May 2004 issue of South Carolina Lawyer. It was reprinted with permission from the South Carolina Bar. See page 18 for part two.
J. Mark Jones is a partner, and John D. Martin is an associate with Nelson Mullins Riley & Scarborough, LLP in Columbia. They practice in the area of complex litigation — with an emphasis on technology, trade secrets and pharmaceutical litigation.