Is Digital Different?
Electronic Disclosure and Discovery in Civil Litigation
By Kenneth J. Withers1
A new phenomenon has surfaced in civil litigation in the United States, or rather in the media coverage of significant cases. In the Microsoft antitrust litigation,2 the investigation of President Clinton by Judge Starr,3 Raytheon Corporation's suit against its own employees for libel,4 countless employment-related actions,5 and even routine divorce cases,6 the evidence takes a new form. E-mail, chat room transcripts, databases, spreadsheets, web browser history files, and information derived from system backup tapes are replacing conventional paper documents.
This phenomenon is not confined to the United States and its tradition of broad discovery. In 1997, the British press reported that insurance giant Norwich Union paid a 6 450,000 settlement to a rival company after disclosure of internal e-mail containing libellous allegations.7 In 1999, BG (formerly known as British Gas) paid more than 6 225,000, half of which represented legal fees, to settle a similar case.8 As in the United States, e-mail is emerging in UK employment-related litigation.9
The legal and business press in other common law countries, such as Canada,10 South Africa,11 and Australia,12 have noted the phenomenon. Even in Scotland, proud of its civil law heritage, a commentator in the popular press wrote that "these cases emphasise the growing importance of companies adopting policies to deal with requests [for electronic documentation] that might arise as a result of legal action."13
The emergence of evidence in electronic form, and the emergence of entirely new forms of evidence, present a number of cultural, practical, and legal challenges to both bench and bar. This paper concentrates on one area: discovery. Part Two of this paper summarises the rules of discovery, and some of the conflicting philosophies and cultural attitudes towards discovery, in the United Kingdom and the United States. Part Three asks whether "digital is different," and summarises the recurring themes found in approximately 200 academic, professional, business and popular press articles on electronic discovery. Finally, this paper proposes areas in which further research is needed, particularly if changes in civil procedure rules or judicial management are contemplated to address electronic discovery challenges.
II. Civil Discovery in the UK and US
The United Kingdom and United States share a common law tradition of private civil litigation in which the parties themselves investigate, develop, and present the facts necessary to resolve their dispute, under rules promulgated and enforced by the court. This stands in contrast to civil law jurisdictions, such as France and Germany, in which the judge plays the role of investigator.14 Central to the common law method of dispute resolution is the process of collecting information that may be presented as evidence in settlement negotiations or at trial. The term "discovery" is used in both the United Kingdom and the United States by practitioners to refer to this process, although the recently promulgated Civil Procedure Rules in the United Kingdom now call this process "disclosure."15 While there are significant differences in discovery rules and even the definitions of common terms, recently enacted or pending rules reforms on both sides of the Atlantic are bringing the concepts closer together.
A. Discovery under the Federal Rules of Civil Procedure in the United States
In the United States, the process of investigating and developing facts prior to trial is governed by Rules 26 through 37 of the Federal Rules of Civil Procedure (FRCP).16 While these rules are in force only in federal courts, most of the state court systems in the United States have adopted civil procedure rules that closely track the federal rules.
The process outlined in the FRCP is driven by Rule 8, which allows "notice pleading," that is, that a litigant may commence an action by filing no more than a "pleading which sets forth a claim for relief... [consisting of] (1) a short and plain statement of the grounds upon which the court's jurisdiction depends... (2) a short and plain statement of the claim showing that the pleader is entitled to relief, and (3) a demand for judgement for the relief the pleader seeks."17 Statements of specific facts are not required at the outset, necessitating a pre-trial fact-developing procedure. It commonly takes the form of depositions (formal interviews of potential witnesses, conducted under oath with counsel for all parties present),18 interrogatories (written questions served on the opposing party to ascertain the factual basis for a claim or defence),19 and requests for production (requests to produce documents or other tangible objects "relevant to the subject matter" of the dispute).20 Included within the definition of "document" found at Rule 34 is language adopted in 1970, "data compilations from which information can be obtained, translated, if necessary, by the respondent through detection devices into reasonably usable form."21
In the adversarial American system, a party has no duty to come forth with information before trial unless the opposing party specifically asks for it, so much of the discovery process is spent defining the universe of potentially relevant information. The scope and timing of discovery events is largely up to the parties. While rules, local practice, and judicial management orders may set outside limits on the number of interrogatories or depositions, or set the deadline to complete discovery, judges are generally not involved in the day-to-day conduct of discovery, intervening only when one or another party objects to specific questions or requests.
B. Discovery under the Civil Procedure Rules in the United Kingdom
Civil litigation in the courts of England and Wales is governed by the new Civil Procedure Rules (CPR), 22 which have only been in force since April of 1999. Similar to Rule 8 of the FRCP, the new CPR adopt a streamlined process to initiate of actions. CPR Rule 7.2 (1) states "[p]roceedings are started when the court issues a claim form at the request of the claimant."23 Rule 16.2(1) states that the claim form must contain "a concise statement of the nature of the claim" and "specify the remedy which the claimant seeks."24 Under Rule 16.4, the form should also state the "particulars of claim," including "a concise statement of the facts on which the claimant relies."25 The Practice Direction accompanying Part 16 sets out the information to be included in the particulars.26 Lord Woolf envisaged the parties developing these basic facts through pre-action exchanges.27 Thus, while the CPR do not go quite as far as the FRCP in adopting "notice" pleading, the resolution of most disputes still depends on a pre-trial fact-developing process.
This fact-developing process, called "disclosure," is primarily document-based and governed by CPR Part 31. Rule 31.4 defines a "document" as "anything in which information of any description is recorded."28 In contrast to the American approach of waiting for a document request from the opposing party, disclosure under Part 31 is pro-active. The parties themselves, after making a "reasonable" search, determine what documents are relevant to the action and generate a statement disclosing their existence.29 The opposing party then has the right to inspect and make copies of any disclosed document, unless the disclosing party states that the document is no longer is within his or her control, the document is privileged, or disclosure would be "disproportionate to the issues in the case."30
C. Differences and Convergence
As outlined above, there have been significant differences between British and American discovery practices since at least the late 1930’s.
Perhaps more significant than the actual differences are the perceptions of difference between the two cultures in regards to discovery.
The actual differences between the two cultures may not be that extreme, and recent civil procedure reforms (or pending proposals) may narrow them further. Discovery in the United States is not as large an industry as the legal and popular press portray. Empirical studies show that discovery occurs in only about half of civil cases filed in United States federal courts, and that document discovery is generally not the most costly and time-consuming activity. On average, deposition discovery costs twice as much. The median cost of discovery, as a percentage of the total amount at stake, is quite small, around 3% to 5%. Most significantly, a majority of attorneys believe that discovery produces an appropriate amount of information, with only 10% on either end of the spectrum believing that discovery in their experience results in the production of either too little or too much information. The widespread perception of burdensome discovery could be derived from the minority of high-stakes, big-money cases, approximately 5% of the docket, in which discovery can exceed 30% of the total claim.36
The Americans are moving closer to the British "disclosure" model of discovery. Pre-discovery disclosure was first incorporated into the FRCP in 1993,37 but was made optional, allowing each of the more than 90 federal district courts to either adopt the new rule, follow the old discovery rules, or fashion their own disclosure procedure by local rule. Only one-third of the district courts chose to adopt the new FRCP Rule 26 in its entirety, but in the years that followed the experiment seemed relatively successful. On September 15, 1999, after extensive discussion and debate within its Advisory Committee on Civil Rules, the Judicial Conference of the United States voted to recommend pre-discovery disclosure uniformly across the nation, with one important modification: attorneys will have no obligation to disclose documents or other evidence adverse to their client’s case.38 This recommendation must still be approved by the United States Supreme Court and ultimately by Congress.
On that same day, the Judicial Conference made another historic decision, to change the "scope" language of Rule 26 from "any matter, not privileged, which is relevant to the subject matter in the pending action…" to "any matter, not privileged, that is relevant to the claim or defence of any party."39 If approved by the Supreme Court and Congress, this would bring the scope of discovery in the United States closer to that of the United Kingdom.
Both the United States and the United Kingdom have recently gone through a series of civil procedure reforms with a common theme: an increased role for judicial management as a tool to control the delay and expense of civil litigation, particularly discovery. In the United States, this has been done by amending particular rules to give judges greater power to set discovery timetables and limits.40 In the United Kingdom, the reform was much more dramatic: an express declaration in the CPR that judges have a duty to engage in "active case management" to achieve the "overriding objective" of dealing with the matter expeditiously, fairly, and "proportionately" to the complexity of the issues and the amount at stake.41 The United Kingdom’s new CPR then goes one step further, leap-frogging over the recent civil procedure reforms in the United States, by declaring that "active case management" includes "making use of technology."42
These developments come at a critical juncture. The rapid spread of digitally-based information technology in international commerce, and computer-mediated communication in everyday life, mean that those legal systems that are best able to integrate computer-generated documentation into their dispute-resolution mechanisms, take full advantage of the potential time and cost savings that digital technology offers, and produce results that are viewed as just and predictable, will have an advantage. Perhaps closer to home, those court systems that are not overwhelmed by the demands of a relatively small number of high-stakes, high-profile "electronic discovery" cases will be better able to allocate precious judicial resources to meeting the everyday needs of justice.
III. Is Digital Different?
The 1970 amendment to FRCP Rule 3443 in the United States and practice under the old RSC Order 24 (which has continued under CRP Part 31)44 firmly established that for the purposes of discovery, information in electronic form is to be considered the same as information in conventional paper form.
A growing body of legal, business, and popular literature on electronic discovery, however, presents perceptions that "digital is different." This section is based on analysis of nearly 200 writings on electronic discovery and evidence, ranging from ponderous academic law review articles to mass-circulation press reports, radio, and television.45 Some of the perceptions that "digital is different" arise from the volume, portability, and volatility of digital media. Others arise from the new forms of documentation and communication that digital media have fostered. Perhaps the most significant perceptions of difference arise from the clash of paper and digital cultures. Technologists and other experts may legitimately assert that these perceptions are inaccurate or exaggerated, or that the problems presented can be solved with existing or future technology. However, perceptions of discovery, particularly those held by the more vocal sections of the legal profession and the public at large, have been historically more influential in policy formation than empirical data.46 In this section, we will state some of the fundamental perceptions of difference, explore their rationale, and ask if they present any serious civil justice policy ramifications .
A. Increased Volume
Electronic discovery is almost universally perceived in the legal and popular literature as something voluminous. The numbers of electronic records that may be subject to discovery are reported to be huge, even in cases involving parties of relatively modest size. This perception is supported by the nature of computerised business processes and the internal logic of computerised document creation and storage, especially in modern client/server network environments.
1. Copy Proliferation
The electronic discovery literature reports, and common experience confirms, that digital documents are easier and cheaper to copy and distribute than documents committed to paper. Exact duplicates of electronic documents can be created with the touch of a button and distributed world-wide, without regard to the length or complexity of the document. Since these digital documents take up virtually no space, little effort is made to destroy them when they are no longer needed. Most document-creation software, such as word-processing, incorporates redundancy as a matter of course.
2. Multiple Locations
The electronic discovery literature reports that the portability of digital documents has dramatically increased the number of locations where potentially discoverable documents may be found. A desktop computer may have several versions or copies of a document on its hard drive. Other versions or copies may be located on the network server. Others may be downloaded and copied onto other desktop computer hard drives. The document may have been copied onto disks, CD-ROMS, or tapes for a variety of purposes. It may be loaded onto a laptop computer for on-the-road use, or transmitted via the Internet to an employee’s home computer for weekend editing. Unlike a paper document, which will usually be found in an identifiable file cabinet, the digital document can, and will, be found anywhere.
3. Backup and Archiving
The data on computers, especially the data on network servers, will usually be duplicated onto digital storage media on a regular schedule for disaster-recovery purposes, thus creating another copy set of the document (including all pre-existing copies and versions). Ideally, the backup media will be retained for only a few days or a month, and then be destroyed or recycled as is made redundant by subsequent backups. Unfortunately, the legal, popular, and even technical press report that these backups are usually saved. Often the data on the backups (and sometimes the backups themselves) are treated as archival. The data are transferred to more permanent digital or optical media, usually without any selectivity or reference to conventional records-retention policies.
4. The Vampire Effect
One of the most popular themes in the electronic discovery literature is the recovery of so-called "deleted" data. According to every expert in the field, it is virtually impossible to destroy a virtual document. Like the vampire of legend, it will always return. This effect is attributed to the method used by most computer operating systems to "delete" a document, which is to rename the file and remove it from the internal directory, designating the physical space that it occupies available to be overwritten by new data. The literature points out that such "deleted" data are almost always recoverable, as complete overwriting of every copy and every variation of a document at every location it may be found seldom occurs.
5. The Policy Implications of Volume
There is general agreement that today in most businesses, government offices, and institutions, the volume of electronic information greatly exceeds the volume of information kept in conventional paper form. Often the issue of volume is confused with the issue of scope, as if the volume of information alone renders the information less relevant to the claims or defences of the parties. However, proposals to address the issue of volume by reducing scope would be misplaced. While a reduction in scope would probably have the effect of reducing volume, it would do so at the expense of the truth-seeking function of discovery.
An alternative managerial approach to the volume problem, avoiding the question of scope, is for the judge to engage in the cost-benefit analysis, called "proportionality," The concept is found in FRCP Rules 26(b)(2)47 and stated more explicitly in CPR Rule 31.7(2),48 which lists "the number of documents involved" and "the ease and expense of retrieval" as factors to determine the "reasonableness" of a document search. There is a tremendous danger, however, for volume to become an excuse for lack of diligence by the parties in discovery. If volume is viewed in isolation, without reference to the state of the art in electronic records management, retrieval technologies, and the needs of both business and justice, the courts could fall into the trap of rewarding organisations for having outdated, bloated, and poorly organised information systems by exempting them from discovery.
B. New Forms of Computer-mediated Communication
Another popular theme throughout the electronic discovery literature is the emergence of new forms of computer-mediated communication. In computerised workplaces tied into world-wide networks, people communicate in ways they have never done before. From a litigation prospective, the most important features of computer-mediated communications are that they are recorded, distributed, and preserved. While these attributes of computer-mediated communication play an important role in American discovery, the effect on British discovery could be revolutionary: Communications, information, and relationships that were heretofore unavailable in discovery because they were undocumented are now routinely documented and subject to disclosure.
In the more recent electronic discovery literature, and in nearly all of the literature found in the popular press, discovery of e-mail is almost synonymous with electronic discovery. The discovery of databases, spreadsheets, or econometric models understandably pales in comparison with the sensational nature of e-mail.
Estimates of the volume of e-mail traffic vary wildly, but even conservative estimates are staggering. In the United States, the press estimates the number of e-mail messages transmitted daily in the billions. Americans are portrayed as addicted to e-mail, and the British are portrayed as catching up quickly. Even for modest-sized organisations the theoretical numbers mount up quickly. One hundred employees in a workplace, each handling an averaging of ten to fifteen e-mail messages per workday, will generate or receive 240,000 to 360,000 messages in a year, before factoring in mass distributions, system redundancies, backups, and other mechanisms for duplication. Unlike the telephone conversations, short meetings, and hallway encounters that e-mail is replacing, these informal exchanges are preserved for discovery purposes.
Much is made in the electronic discovery literature of the informal, revealing, and often embarrassing nature of e-mail. E-mail is considered the window into the corporate soul, and is therefore a highly sensitive area for discovery. The literature also notes a deep disconnect between individuals’ perceptions of e-mail as private and transitory, and the reality of e-mail as a permanent and discoverable corporate record, although only rarely does the literature suggest that this phenomenon has been studied empirically in any other discipline, such as linguistics or psychology.49
c. Lack of Management
The literature agrees that e-mail is routinely mishandled or ignored by businesses and government agencies as a records management issue. Surveys presented in the legal and business press indicate a minority of large organisations have e-mail management policies and procedures, even though e-mail is increasingly being used as a medium for conducting substantive business. Therefore, while e-mail may be relevant and important to the issues of the case, e-mail databases usually have no useful organisation or structure to aid discovery efforts. Squarely relevant and wildly irrelevant material will be co-mingled. Privileged and non-privileged communications will be indistinguishable without individual inspection. Significant documents missing from the conventional records collection may be found attached to e-mail.
d. Policy Implications of E-mail
In the United States, with its tradition of broader discovery, e-mail is inescapable. One theme in the electronic discovery literature is that e-mail must be pursued, as often evidence (particularly of the "state-of-mind" nature) will exist nowhere else. A strong counter-theme is that discovery of email is tantamount to blackmail, as the cost of discovery exceeds the value of the case and forces settlement. This counter theme has found its way into the British legal and popular press. Both of these opposing views are presented with a small number of often-repeated e-mail incidents. No empirical evidence has been brought forward to support the view that the discovery of e-mail, in itself, has tipped the scales of justice in a general way to warrant changes in the scope of discovery or the definition of a "document."
In Britain, the "proportionality" equation considered above in relation to volume will likely come to play in relation to e-mail. But there is a chance again that the courts, if they consider each case in isolation, will reward those organisations with poor records management practices. Part of the proportionality equation must be an assessment of what records management practices a party should be expected to have in place, given the resources and sophistication of the organisation and the environment as a whole.
Some courts have expressed a willingness to reject e-mail entirely as evidence on the basis that it is too informal and unstructured to be considered a "business record" and overcome hearsay objections.50 This may also be the view of some attorneys, who are more than willing to ignore it if they could. However, it is too late in the game for this approach to be taken. The electronic records management literature shows that e-mail has moved from an informal message transmission system to an integral part of the business process for many corporations.
2. Other New Forms of Computer-mediated Communication
a. World Wide Web
E-mail is not the only new form of computer-mediated communication, although it is portrayed in the literature as the most significant. Second to e-mail in the electronic discovery literature is use of the World Wide Web. In both the United States and the United Kingdom, discovery has uncovered cache files (files recording the Internet addresses of Web pages visited and the graphic elements on those pages), cookies (information about the user placed in a file by a Web site operator for quick recall when the user visits again), bookmark files (one-click shortcuts to favourite Web sites, created by the user), and downloaded text and images stored on local hard drives, floppy disks, and even servers. These communications may serve as substantive evidence of wrongdoing (such as copyright infringement), present circumstantial evidence of wrongdoing (such as gambling or pornography distribution), or simply be embarrassing.
b. Chat Rooms and Bulletin Boards
Chat rooms and bulletin boards are relatively new forms of computer-mediated communication subject to discovery. Like Web use, exchanges in these forums may be substantive evidence of libel, fraud, or breach of confidence. Participation may be evidence of infidelity, insider trading, or criminal activity, or discovery of participation could simply be embarrassing. Chat room and bulletin board discovery presents logistical problems, however. Complete transcripts of conversations or postings are seldom kept for more than a few days, so discovery may be confined to records of participation. While it is not unusual for corporations, educational institutions, and even government agencies to sponsor and host chat rooms and bulletin boards, most are hosted by commercial third parties, particularly online service providers. In the United States, discovery of documentation held by third parties in civil cases is difficult,51 and in the United Kingdom discovery documents held by third parties is rare.
c. Voice Mail and Collaboration Tools
Other forms of computer-mediated communication for which there are no conventionally discoverable counterparts are digital voice mail and collaboration tools. Voice mail is only mentioned occasionally in the literature, and there are no instances cited in which it has actually been disclosed in discovery or used at trial. Collaboration tools, such as virtual sticky notes and virtual white boards, are mentioned only in passing as potentially discoverable documentation. However, there is nothing in theory and little in substance to distinguish these forms of communication from e-mail or chat room participation for the purposes of discovery.
d. The Policy Implications of Other Computer-mediated Communication
New forms of computer-mediated communication are giving rise to new avenues of discovery that are essential to the claims and defences of both parties in civil litigation. Indeed, many causes of action may be based entirely on evidence to be found in the documentation that these communications generate. As computer-mediated communications become the dominant forms of communication in business, education, government, and even family life, the need for discovery of the resulting documentation will be compelling. There is no doubt that the evolution from informal, unrecorded face-to-face or voice communication to computer-mediated communication increases the volume of material subject to discovery and increases the number of locations which an attorney must check in order to conduct the "reasonable" inquiry required by the discovery rules in both the United States and the United Kingdom. This naturally has an effect on costs and may invoke the "proportionality" provisions of CPR Part 31 or FRCP Rule 26. However, as with the general problem of volume and the problem of e-mail, proposals to address the problems associated with new forms of computer-mediated communication by limiting the scope of discovery would be misplaced.
C. New Areas for Discovery
The growth of computerisation in all aspects of life has resulted in the growth of new areas for discovery. Computers have not only created new forms of substantive communication and facilitated the documentation of conventional forms of communication, computers are busy generating documentation of their own processes. Much of the literature on electronic discovery written for the legal audience stresses the need for lawyers to conduct a preliminary layer of discovery prior to substantive discovery. This involves discovery of a client’s (or opponent’s) computer system itself, the business processes associated with it, and the resulting system documentation. Particularly in the United States, bar journals abound with articles on how to conduct this preliminary discovery. While new forms of substantive computer-mediated communication may increase the volume of discovery, inquiry into system documentation increases the scope of discovery. The question is whether this increased scope is an unavoidable consequence of electronic discovery.
Conventional paper-based document discovery is relatively straightforward. Lawyers and their clients understand and can readily see storage boxes, file cabinets, file folders, and the papers contained therein. The filing system and records management process for paper documentation is generally known or at least documented in most organisations, if for no other reason than the need to store and retrieve the physical files on a day-to-day basis. Unlike conventional discovery of paper documents, the sources of electronic documentation, the storage systems, and the mechanisms of retrieval are not readily visible, and most lawyers and their clients are unfamiliar with the workings of their own desktop computers, let alone a computer network.
One lesson repeated in the literature is the need to discover metadata, or information about computer-generated documents, in addition to the documents themselves. Metadata are information contained within the electronic version of a document that may not be apparent in a print-out of the same document. The date the document was created, the identity of the author and subsequent editors, the distribution route of the document, and even the history of editorial changes may all be part of the metadata. Metadata are important when viewing a word-processing document, and considered essential when viewing an e-mail as the only method of authenticating the sender, route, and content.
2. Network Records
Computer networks automatically generate copious records of their own activities, and the electronic discovery literature stresses the need for attorneys to obtain and understand this information. Network records may include information about who had access to what data or equipment at any given time; the structure of the files; what documents were distributed to whom, when, and in what form; when backups were performed; and even when the network or various parts of it were inoperable. It should be noted that none of this information touches on the substantive content of any documents on the network, archives, or associated storage media.
3. Ghost Data
As discussed above in relation to the increased volume of discovery data, one of the favourite themes of the electronic discovery literature is that routine deletion of a computer file does not destroy the file, but merely marks the disk space the file occupies for overwriting if needed. Given the ever-expanding memory capacity of today’s computers and the fact the file itself may be found on several places on a disk, the chance that a "deleted" file would actually be entirely overwritten is slight.
The recoverability of deleted data is a prominent theme in the electronic discovery literature. In both the United States and the United Kingdom, a small army of computer forensic consultants has sprung up to answer the perceived need among attorneys for data recovery services. Alternatively, a review of the literature could lead one to conclude that a small army of computer forensic experts has created a perception through the electronic discovery literature that their data recovery services are needed.52
4. The Policy Implications of New Areas of Discovery
Metadata, system records, and deleted files are new sources of discoverable information with few counterparts in the conventional world of paper discovery. While metadata may be viewed as simply extensions of relevant documents, the discovery of metadata and other system records expands the scope of discovery from documents containing substantive information about the dispute to documents containing information about substantive information about the dispute.
Some level of discovery beyond the substantive documents themselves may be necessary to establish the context or authenticity of the substantive documents. Since this can now be done in documentary form (as opposed to oral testimony, the usual method of establishing the authenticity and context of conventional paper documents) it may be less costly and more efficient to allow this discovery in some cases. It does not necessarily follow, however, that because we now have the capability of reconstructing the past in minute detail using electronic documentation, there is a need to do so in every case.
Unless the attempted deletion of electronic documents is itself an issue under dispute, the discovery of deleted documents will reveal only that the readily-available electronic documents were preceded by drafts or accompanied by copies, or that in the normal course of business prior to litigation, many routinely-generated documents were considered non-essential. These are not surprising revelations.
Therefore, the discovery of metadata, system records, and deleted files, beyond that necessary to authenticate or place the substantive documentation into proper context, may be viewed as an expansion of scope, and such discovery may be conditioned on a showing that it is likely to result in substantive, material evidence not otherwise available.53 The explicit authorisation in CPR Part 31 to determine "proportionality" in setting the limits of discovery and disclosure, coupled with an explicit instruction to use technology whenever it may reduce costs and speed justice, may place British judges in a better position to deal with inevitable disputes over the scope of electronic discovery. Although American judges are authorised to engage in a similar balancing of costs and benefits, they are traditionally loath to do so.
D. Legacy Data
1. The Legacy Data Problem
In conventional, paper-based discovery, if a document exists in a file, it usually can be read without any difficulty. On occasion, paper may deteriorate or ink may fade, but seldom are the documents written in a dead language. The electronic discovery literature, on the other hand, often focuses on the problem of discovering unreadable data.
Computer-based information systems change every few years. Unless careful attention is paid to converting old data into current formats as part of the system upgrade, vast amounts of "legacy" information will exist in electronic form but be costly or impossible to retrieve.54 The information may be stored on media that can no longer be accepted or organised in a format that can no longer be read using current software. Computer museums must be consulted to locate the appropriate hardware to accept the outmoded media, and special programs must be written to convert the data into useable form before any determination can be made that the information is relevant to the pending action. The computer forensic experts who specialise in data recovery report in the electronic discovery literature how costly and time-consuming this process can be.
2. The Policy Implications of Legacy Data
The discovery of legacy data is analogous to the discovery of deleted files. A proponent should be willing to provide convincing evidence that the restoration of legacy data will result in the discovery of substantive, material evidence not otherwise available. The holder of legacy data also may have an interest in restoration and review for defence purposes, opening up the possibility for a compromise in the way of cost-sharing or co-operation in locating the most relevant tapes or disks and performing "test drills."
But as with voluminous active data discussed above, the holder of legacy data should not be allowed to avoid the duties of discovery by having a poorly planned or poorly maintained information system. If the failure of the party to routinely update legacy data is unreasonable, viewed in the light of business or regulatory needs without reference to the pending litigation, the party should be expected to pay the costs of that failure.
E. Danger of Spoliation
The literature on electronic discovery often focuses on the issue of "mutability," that is, the apparent ease with which computer files may be altered or destroyed. This is in direct contrast to the literature’s alternate focus in electronic file indestructibility, discussed above. It is sometimes difficult to square these two apparently contradictory positions.
1. Routine Data Destruction
Active files on computers, especially shared data on network servers, are constantly changing. In the normal course of business, files are opened, edited, added to or subtracted from, or deleted. They are subject to manipulation in ways that static paper documents, stored in file cabinets, are not. Word-processed documents or e-mail messages may take on a more-or-less permanent form once completed, but financial records, customer databases, and other data compilations are dynamic. Even the act of opening files for review in the course of discovery may fundamentally change an important characteristic, such as automatically-generated dates or calculations of interest. In the discussion of electronic discovery volume above, it was noted that many computer network administrators will routinely create back-ups of network data. Many administrators keep these backups forever, but others will do exactly what they are supposed to do, and overwrite them in the normal course of business with more current data.
This routine, systematic churning of data means that when a legal action is filed, or when it is apparent that an action will be filed, the information system must be frozen in time to preserve the discoverable evidence, without crippling the organisation in the process. This may be accomplished by creating "images" of the disks and other data storage media, and basing discovery on the images, rather than the active files. There may be a tendency for attorneys to either delay issuing instructions to their clients to preserve data or disagree over the procedure. Meanwhile, the clock may be ticking, as especially in large, decentralised organisations, the evidence may be disappearing.
2. Apparent Ease of Wilful Destruction
Distinct from the routine churning of data resulting in the inadvertent destruction of potential evidence is the perception in the electronic discovery literature that widespread spoliation is occurring. It should be noted that there is no empirical evidence for this, but the literature portrays a world in which individuals faced with potential discovery immediately delete damaging e-mail and word-processing files. The perception is supported by ample anecdotal evidence and by assumptions that electronic evidence is easier to destroy than conventional paper evidence.
3. The Policy Implications of Potential Spoliation
There may be a tendency to dismiss the fears associated with the mutability of electronic evidence by noting that all electronic data is recoverable, and that nothing is really lost. This is not entirely true. Multiple overwriting can, indeed, make data unrecoverable, and routine mutation of the data can render it unreliable or take it out of context. Even if the destruction of electronic evidence, inadvertent or wilful, is ultimately unsuccessful, the cost of locating and recovering the evidence will increase substantially. This fact squares the apparent contradiction in the literature: while the complete physical destruction of electronic data is difficult or impossible, it is relatively easy to render the data inaccessible or too costly to be retrieved. This form of evidence destruction is just as likely to occur inadvertently in the routine course of business as it is to occur at the hands of a wilful spoliator.
From a policy perspective, it is important that the courts recognise this danger of increased costs and delay, and be open to early action to preserve data. Such early action may take the form of an immediate conference of the parties to identify areas for electronic discovery and preserve the necessary evidence, the development of a data preservation protocol, or the appointment of a neutral expert to oversee the parties’ data preservation activities. One of the unique attributes of the new Civil Procedure Rules in the United Kingdom is the concept of pre-action disclosure under pre-action protocols.55 While it is contemplated that protocols will be developed for routine cases, a protocol for preservation of data may be useful, especially if such a protocol can be scalable to reflect the needs of a variety of cases. The Federal Rules of Civil Procedure in the United States have no equivalent method of influencing the actions of the parties prior to filing, but individual judges may use their discretion under FRCP Rule 16 to issue instructions on data preservation to the parties early in the procedings.56
F. Electronic Records Management
There is a widespread perception that the management of electronic records in most organisations is woefully inadequate, and that gives rise many of the problems associated with electronic discovery . This perception is confirmed by empirical research presented in the professional records management literature. The research indicates that few organisations manage their electronic information with the same attention that they formerly paid to their paper documents, compounding the problems of volume and scope when the electronic records become subject to discovery. In addition, the management of conventional paper records is also suffering.
1. The Demise of Records Management
Conventional paper-based records management stressed storage for the purpose of retrieval (necessitating human application of logical indexing systems) and keeping the physical collection as lean as possible, while fulfilling the essential mission of documenting business decisions and long-term operations. Professional records managers were in an excellent position to answer discovery demands quickly, inexpensively, comprehensively, and reliably.
The demise of records management as an important part of corporate culture in the United States and the United Kingdom has been a long-term process attributable to several factors, most importantly the ascendance of computer-based information technology. Computer technology was introduced in most corporations and government offices in the 1960's by the finance departments or their equivalents. Computers were seen as tools to automate the creation of conventional paper documents, initially financial reports, statements, invoices, and the like. Once reduced to paper, the computer output was incorporated into the conventional records management process. When word processing was introduced in the 1970's and 1980's, the centralised computers were still under the control of the financial department and their subsidiary, Management Information Services (MIS). The computer was still seen as a tool for the creation of paper documents, which would then be incorporated into the conventional records management process. Drafts, back-ups, and other by-products of computer applications were considered the property of the MIS department, to dispose of as they wished, with no reference to corporate records management policies or procedures.
With the introduction of client-server networks and other distributed models of corporate computing, documents and data in electronic form took on a life of their own. Communication took place without reduction to paper, and new desktop applications, such as databases and spreadsheets, proliferated. By this time, computer technology and the data it generated was controlled by independent departments whose primary responsibility was the acquisition, application, and maintenance of hardware and software. The new Information Services (IS) or Information Technology (IT) departments had no concern for records management. As memory became cheaper, electronic data was saved on vast, unindexed backup and archival media, usually in direct contradiction to the conventional records management policy of the organisation.
The spread of distributed models of corporate computing lead to the downsizing of many corporations and government offices. Among the first staff to go were the now-marginalised and invisible records managers. With the reengineering of business processes, the secretaries and clerks (who previously served as shop-floor enforcers of the records management policies), were also made redundant. In many organisations, professional records management disappeared at the same time that virtual documentation came to dominate corporate business processes.
2. Lack of Corporate and Technological Response
Corporate culture has been slow to recognise the consequences of the demise of records management. The literature of the past three to four years, however, indicates that corporations and government agencies are beginning to recognise that electronic document proliferation, particularly the retention of unnecessary and potentially damaging e-mail, must be checked. Still, only a minority of major corporations and institutions in the United States and the United Kingdom have operating electronic records management programs.
The predominate response in the corporate world has been to look to the IS and IT department for technological solutions: Software that will somehow identify mission-critical information and store it properly while weeding out the irrelevant or outdated information. So far, no solution has been developed that fills the role of professional records management.
3. The Policy Implications of the Demise of Records Management
The demise of professional records management may have serious policy implications for electronic discovery. Unless sound management techniques are applied to the ever-expanding mass of electronic data and communications, or a technological breakthrough is made, litigants will face ever-spiralling discovery costs, which the courts will be powerless to control. Case-by-case application of the proportionality principles of both the American in British civil procedure rules will severely curtail discovery in many cases. As discussed above, the net effect may be to reward those organisations with inadequate corporate controls, while subjecting others to a broader scope of discovery, simply because they have the organisational ability to handle it.
Therefore, the courts, when faced with litigants objecting to the costs of electronic discovery, must develop tests to discern between those costs which are self-inflicted and those which are truly out of proportion with the stakes of the case at hand. An objective test would be to determine what records management capabilities are reasonable for a party to possess, given the day-to-day needs of the organisation, the developing state-of-the-art in records management and information technology, and the stakes of the case.
G. Form of Production or Disclosure
1. The Costly Relict of Past Practice
The legal literature on electronic discovery reflects the continuing preference of attorneys to print out discovery documents, even if they were generated and stored electronically, for inspection by the opposing party. Attorneys complain that electronic discovery forces them to create or copy documents, at their own parties’ expense, rather than simply make them available for inspection, as they did with conventional paper discovery. In larger disputes, when one or both parties is utilising computerised litigation support, this bias in favour of paper has resulted in a diseconomic pattern of "electronic to paper to electronic to paper," in which producing parties print out their clients' electronic documents to allow inspection of the resulting paper, and requesting parties digitally scan the paper, ultimately printing out the images for trial preparation.
But most attorneys are unwilling to break with past practice and make documents available for inspection in digital form. Their reasons are not irrational. Allowing an opposing party access to one’s computer system for the purposes of inspection almost guarantees that privileged and irrelevant information will be exposed, important evidence will be altered or compromised, and the disclosing party will suffer business disruption costs. Therefore many attorneys see no alternative but to engage in costly paper-based review and production.
2. Inadequacy of Print-outs
However, reduction of electronic information to paper may not be acceptable. It has been noted above that the routine visual and print manifestations of electronic documents such as word-processing and e-mail lack vital metadata, the information about the author, date of creation, and history of distribution and editing. To produce these electronic documents in their entirety, the disclosing party must make them available in electronic form, preferably on media as close to the original as possible. The requesting party must then have the ability to copy these to portable media in a manner that preserves the original electronic files to the greatest extent possible while protecting them from further mutation.
Databases and spreadsheets are dynamic methods of organising, processing, and presenting data and are not suited to being captured as static documents, in paper or electronic form. Spreadsheets, in particular, are slippery concepts, as each cell may contain static data, a formula with which to process data, or a dynamic result of data and formula. Unless a specific database report or spreadsheet view is at issue, production must be in electronic form, capable of being manipulated by the requesting party (with appropriate measures to assure the preservation of any static data).
3. Advantages to Digital Production
The problems associated with the production of electronic data may be viewed as opportunities to make dramatic changes in day-to-day practice to reduce costs and speed discovery. While the literature in the United States tends to emphasise the strategic aspects of electronic discovery and the potential cost disadvantages, the literature in the United Kingdom places relatively more emphasis on the management aspects and potential cost advantages. This may be a result of the climate produced by the Woolf proposals and resulting rules changes, which explicitly encourage the use of information technology to improve the civil justice system. But the British literature indicates that a major cultural change needs to occur before clients and solicitors see the potential advantages.
a. Reduced Processing Costs
The cost of manipulating paper documents, especially in large-scale litigation, can become prohibitive, particularly in multi-party cases. Once documents are identified, they must be sequentially numbered or otherwise uniquely identified, listed or indexed, photocopied, the photocopies inspected for accuracy, shipped, and stored. Costs of 6 1.25 or $2.00 per page of documentation through the life of the litigation are not unusual, and a litigation support service bureau industry has grown up in both the United States and the United Kingdom to meet the demand for paraprofessional assistance.
If the documents and data subject to discovery are generated originally in electronic form, and if clients and their attorneys normally store and use these in electronic form, there is no fundamental need to create paper versions simply for the purposes of discovery and disclosure. Copying costs may be reduced dramatically, particularly for multiple copy sets. The costs of quality inspection, creation of rudimentary unique identification or indexing, and shipping can be reduced to a pittance. The speed at which documents may be processed can be increased dramatically, and transport of entire document sets to multiple parties can take place in an instant.
b. Increased Utility
The attributes of electronic documents and data which give rise to so many worries on the part of attorneys may be used to the litigants' advantage. Electronic discovery and disclosure can be integrated with two other advances in law-related information technology to streamline the dispute-resolution process: computer-based litigation support databases and computer-based trial presentation technology.
Litigation support databases are essentially digital catalogues of discovery material, created to assist attorneys to access, analyse, and organise evidence in preparation for settlement negotiations and trial. In both the United States and United Kingdom, the value of this type of litigation support, especially for the "big document" cases, has long been recognised. But the utility has been perceived as limited to the "big document" cases because of the cost of physically entering data from paper documents or scanning paper documents to populate the database. If litigation support databases were integrated with document disclosure and production in electronic form, these costs could be reduced dramatically, and the advantages of litigation support made available in more routine cases. The parties may even contemplate creating a shared virtual document repository and bibliographic litigation support database, a secure Internet site from which documents may be accessed, viewed, and downloaded.
Electronic trial presentation technology is also gaining increased acceptance in both the United States and United Kingdom. Judges are acknowledging the cost, speed, and efficiency advantages to "paperless" trials, during which documentary evidence, precedents, expert reports, and other material is available on CD-ROM or other secure electronic media, and even live testimony is transcribed in real time and incorporated into the electronic "bundle."
3. The Policy Implications in the Form of Production
In recognition of the differences between electronic documents and data, and in order to realise the potential advantages of electronic discovery and disclosure, the courts should actively encourage the exchange of documents and data in electronic form whenever possible. This means creating a climate in which the electronic exchange of information is presumed, and overcoming the real objections that attorneys have to producing their clients' material in electronic form.
a. Reversing Presumptions
As stated previously, there is a long-standing bias in favour of paper-based document discovery, even when the documents have been created and are normally stored in electronic form. In the United States, there is explicit language in the rules of civil procedure and the rules of evidence stating this presumption.57 Judges have the power to change this aspect of legal culture by issuing orders, early in the case, stating a preference for electronic exchange and laying the groundwork for electronic trial bundles.
b. Developing Protocols
Electronic exchange is technically more complicated than paper exchange. There must be agreements in place as to the format for the exchange, the method of document or data selection, the time and place of the exchange, the minimisation of business disruption, the preservation of the data after exchange, and a host of other details. Working these details out on a case-by-case basis would be a costly waste of judicial energy, but protocols may be developed to cover the majority of situations. Alternatively, the court could call upon neutral electronic discovery technicians whose mission would to facilitate electronic document and data exchange between the parties, and who would reduce any resulting dispute to alternative procedures for judicial consideration.
c. The "Inadvertent Waiver of Privilege" Problem
Perhaps the most important obstacle to the co-operation and mutual access necessary for electronic document and data exchange is the traditional resistance attorneys have to letting opposing parties view their clients' document collection without thorough pre-screening for privilege. There are several strategies for minimising this problem, including careful segregation of responsive material prior to inspection and the use of a neutral technician to actually perform searches. But the most important way to diffuse the perception of danger in the minds of attorneys to remove the dire consequences of inadvertent waiver of privilege.
While doctrine of waiver is not applied consistently in the United States, the essential fear is that if a privileged document is inadvertently disclosed to the opposing party, that document may either be used in court by the party (or others), or worse, that the entire class of documents from which it is derived may lose privilege in this (or subsequent) litigation. Judges must work to clarify how the doctrine of inadvertent waiver will be applied in the future, recognising the differences between electronic and conventional paper discovery. Whenever possible, the parties should be encouraged to enter into judicially-sanctioned and binding agreements to limit the application of the doctrine in their case.58
IV. Areas for Further Research
Since litigation trends tend to follow trends in society, it is almost axiomatic that electronic discovery will eventually overshadow paper-based document discovery. Yet litigants and courts continue to express more anxiety than optimism over electronic discovery and its anticipated consequences. The ultimate question to be addressed is whether there is anything about the nature of electronic discovery that necessitates any changes in our rules of evidence, rules of civil procedure, or case management strategies. Before we can reach this ultimate question, we need to further explore some fundamental questions.
We may want to consider testing the assumption found in much of the electronic discovery literature that electronic discovery is costlier or causes more delay in litigation than paper-based discovery. This would be a difficult proposition to test, as we are not in a position to litigate the same cases twice and compare the costs. However, we may be able obtain data or design a survey that compares discovery costs in two statistically valid samples of cases, one of cases involving electronic discovery and another of cases involving exclusively paper-based discovery. The same method could be used to compare the length of time over which discovery took place, and the number of discovery disputes.
If such studies indicated that there were actual cost differences between the two discovery environments, the next questions would be why the differences arise. There may be several factors, not all of which are within the control of the court:
Developing and executing research strategies to test each of these hypotheses, and the many others that could be formulated, would be a formidable task. However, small-scale research projects could be designed that begin to look behind the perceptions stated in the electronic discovery literature and place our understanding of this issue into a more practical context.
Using existing court records, recently-closed cases most likely to have involved some discovery activity may be identified by subject matter, amount in controversy, docket entries, or a combination of indicators. From that sample, the attorneys and judge in each case could be contacted with a simple survey asking whether or not there was disclosure of electronic evidence, if so, whether they would be willing to participate in a more detailed survey.
A more detailed follow-up survey could ask the participants their impressions regarding electronic discovery:
Finally, interviews with the judges who respond could be conducted to obtain "neutral" views of the problem and gather ideas on judicial management.
The data gathered from such a series of surveys would probably not be considered "scientific" as much as exploratory. The response to the initial survey may indicate the extent to which electronic discovery is or is not a common element in certain types of litigation, a question which will be mooted over time. The goal of the second survey would not be to quantify various aspects of electronic discovery, but to perform a sounding to determine if there is a widespread belief that there are problems with, or advantages to, electronic discovery that are within the power of the judiciary to address. The participants in the second survey will be self-selected for those who have something to say, and perhaps an axe to grind, in relation to electronic discovery, and the resulting data would need to be analysed with that assumption clearly in mind. The goal of the third survey would be to assess the "state of the art' in terms of judicial management of electronic discovery, and begin to assemble a set of procedural and management tools for immediate use by both judges and attorneys.