Evaluating the Proposed Changes to Federal Rule of Civil Procedure 37: Spoliation, Routine Operation and the Rules Enabling Act

¶ 1 For years scholars,¹ practitioners,² and the courts³ have been struggling with how best to manage electronic discovery under the current Federal Rules of Civil Procedure ("Rules").⁴ Many commentators have advocated that electronic discovery ("e-discovery") deserves individualized treatment in the Rules because managing e-discovery under the current Rules has proved problematic.⁵ This is because electronically discoverable material is different in several important respects from traditionally discoverable documents.⁶ Because of these differences, it is difficult to manage electronic discovery under rules developed to accommodate traditional discovery.

¶ 2 After five years of studying the issue of electronic discovery,⁷ the Civil Rules Advisory Committee ("Committee") proposed in 2004 a set of amendments to the Federal Rules of Civil Procedure aimed at the discovery of electronically stored information.⁸ The proposal was recommended for public comment in August of 2004⁹ and subsequently prompted considerable public comment.¹⁰ A revised version of the proposed changes¹¹ was passed to the Judicial Conference of the United States, who voted in September 2005 to approve the changes.¹² Since it is unlikely that the Supreme Court or Congress will oppose the proposed changes, they should go into effect on December 1, 2006.¹³

II. An Overview of the Proposed Changes

¶ 3 Although the proposed changes are multifarious,¹⁴ the various proposals can be categorized and described by the basic electronic discovery issue they seek to address.

¶ 4 Several changes reflect the principle that many electronic discovery disputes can be prevented, managed, or resolved through early and direct communication between the parties and their technology experts.¹⁵

¶ 5 Under the current Rules, the use of the term "document" to describe the wide range of currently discoverable material is, as a formal matter, too narrow and fixed to encompass the rapidly expanding and changing character of discoverable electronic information.¹⁶ Therefore, another set of changes essentially seeks to update the concept of traditional discovery in the Rules.¹⁷

¶ 6 Concerns regarding the form of production in the electronic discovery context animated changes to Rules 34 and 45.¹⁸ These changes reflect the Committee's acknowledgement that electronic discovery may be more burdensome and difficult to analyze if produced in a static form because massive amounts of electronic data are often easier to navigate if available in a live searchable form.¹⁹ Therefore, the proposed rules encourage parties to agree on an acceptable form while providing default production forms in situations where the form of production is contested.²⁰

¶ 7 Fulfilling an electronic discovery request often will mean that a party must hand over a massive amount of data.²¹ Such information transfers necessarily imply concerns about the unwanted dissemination of privileged information. The first concern is that a party might inadvertently transfer privileged information.²² The second concern is that exceedingly thorough efforts to preview the information add a considerable amount of incremental cost to the litigation process.²³ Therefore, another set of proposed changes allows for "claw back" and "quick peek"²⁴ agreements.²⁵

¶ 8 The life cycle of data in an electronic environment is markedly different than that of data in a typical paper document.²⁶ For example, one would not expect a discovery order requiring the producing party to search the local garbage facility for a potentially relevant document that had been disposed of years ago. Electronic data, however, is often requested from the virtual dump.²⁷ This is because in the electronic environment, information often lingers in proprietary systems long after it has been "thrown-away." Examples include data stored on back-up tapes, retained in legacy systems, and reserved for deletion on hard drives.²⁸ The proposed rules acknowledge that producing such material can be highly burdensome and, in cases where such data is not "reasonably accessible," should not be discoverable.²⁹

¶ 9 The final issue the proposed rules address relates to sanctions for the spoliation of electronic data. This highly controversial issue³⁰ is the primary focus of the discussion set forth in this Note. The Committee has proposed that sanctions cannot be imposed for material that is deleted as part of a "routine, good-faith [computer] operation."³¹

¶ 10 This "safe-harbor" has been criticized from all sides of the Bar.³² Unfortunately, these criticisms have not acknowledged a fundamental problem. A federal rule is limited in the problems it can permissibly solve. It is necessary to make normative assessments about the proposed changes in light of the limitations placed on the Rules Advisory Committee by the Rules Enabling Act ("REA") of 1938 (amended 1988).³³

¶ 11 The first sentence of the Act provides that the "Supreme Court shall have the power to prescribe general rules of practice and procedure... for cases in the United States district courts... and courts of appeals."³⁴ The second provision limits this power by requiring that "(s)uch rules shall not abridge, enlarge or modify any substantive right."³⁵ Most commentators agree that, generally, the Act provides that a Federal Rule of Civil Procedure must relate to procedural matters without trammeling too far on matters of substantive law.³⁶ However, where to draw the line between substance and procedure has been the subject of fervent debate.³⁷

¶ 12 This Note will examine one of the most controversial proposed changes—the spoliation safe harbor. Although commentators have criticized "routine operation" as too vague a standard,³⁸ this Note will focus the discussion by explaining that a more "certain" definition of routine operation would violate the Rules Enabling Act. Therefore, any attempt to define the affirmative duties of companies to retain electronic information should be left to legislatures or to common law rulemaking. Essentially, commentators should not ask for what they cannot get in a Federal Rule: substantive law making.

¶ 13 This Note shall briefly explain why Electronic Discovery poses unique problems in the spoliation context. A discussion of pre-proposal attempts to deal with e-discovery spoliation will follow. The Note will then detail the specific proposed changes to Rule 37. The case for why Federal Rules are, on some level, a good place to deal with e-discovery (uniformity), but are not a Mecca (framing why the substantive limitations imposed by the Rules Enabling Act are important not only from an academic perspective) will be made. The Note will then discuss in detail the substance versus procedure distinction in the Rules Enabling Act and apply it to Rule 37(f) as proposed by the Committee and by commentators. The Note will conclude by making a normative assessment of the proposed changes to Rule 37 in light of the REA.

III. Why electronic discovery should be treated differently than traditional discovery in the spoliation context

¶ 14 The use of computer systems by companies has significantly increased the volume of potentially relevant data:

The sheer volume of such data, when compared with conventional paper documentation, can be staggering. A floppy disk, with 1.44 megabytes, is the equivalent of 720 typewritten pages of plain text. A CD-ROM, with 650 megabytes, can hold up to 325,000 typewritten pages. One gigabyte is the equivalent of 500,000 typewritten pages. Large corporate computer networks create backup data measured in terabytes, or 1,000,000 megabytes: each terabyte represents the equivalent of 500 billion typewritten pages of plain text.³⁹

¶ 15 To further complicate matters, companies retain and use data in a variety of different forms:

1. Active or Online Data: This is the kind of electronic information—like e-mail, presentation software, spreadsheets, databases, and GIF files—that an everyday user consciously accesses and manipulates.⁴⁰

2. Near-Line Data: This is the kind of electronic information—like optical disk storage—that is located in a physical "library" and accessed by automated machines.⁴¹

3. Internet Data: This includes data like cookies, cache memory, favorites, history, autocomplete, etc.⁴²

4. Metadata: This is electronic information that is embedded into a particular document.⁴³ It is essentially "data about the data."⁴⁴

5. Replicant Data / File Clones: This is electronic data stored when the computer makes a flash copy of a document in use. The paradigmatic example is the automatic saving of a copy of a word processing document for document recovery purposes.

6. Residual Data: This is data the ordinary user has deleted.⁴⁵ Although the file no longer appears to the user to exist on the hard drive (e.g., does not appear in Windows Explorer), the file has merely been marked for deletion by the computer.⁴⁶ This marking process simply allows the computer to make available the space on the hard drive for overwriting should the memory be needed.⁴⁷

7. Backup Data: This data usually exists outside of the network in the form of large network back-up magnetic tapes.⁴⁸ The purpose of the storage is usually to restore the full system in the event of a massive system failure. These tapes are overwritten periodically to facilitate more recent back-ups.⁴⁹

8. Legacy Data: This is data that was generated by outdated software no longer in use.⁵⁰ Although the data still exists, the systems needed to read and manipulate the data are no longer used by the company.⁵¹

¶ 16 The intersection of volume and form is the heart of why spoliation is such a vexing problem in the electronic context. As the above descriptions illustrate, most electronic data serves a dynamic function and thus has only temporary existence on a given system. If the search for truth were the only goal of the litigation process, we might expect a company to suffer sanctions unless the company halted the operation of all computer operations once the company reasonably knew that litigation is likely. The litigation system, however, is not a "Terminator"⁵² programmed to find truth. The truth-finding goal of the system is balanced by the "utilitarian concern for efficiency, the need to preserve procedural-substantive balance, and the need to provide predictable standards of primary behavior."⁵³ Imposing sanctions on companies for spoliation due to routine system operations would have massive effects on primary behavior and thus violate the substance-procedure balance. Taken to the extreme, to avoid sanctions, large companies that are constantly under threat of litigation would have to store all data indefinitely or have departments of individuals dedicated to preserving all potentially relevant data.⁵⁴ Given the relatively low truth-finding payoff, when compared to the expense of preserving all electronic data, sanctioning all electronic discovery spoliation would be inefficient and significantly add to the costs of the litigation process for individual litigants. To prevent undue burden, expense, and inefficiency, sanctions for electronic discovery should not be imposed in all situations where data is lost due to system operations.⁵⁵

IV. The Current State of Electronic Discovery Spoliation: Are the Federal Rules an Appropriate Forum by which to Address the Problem?

¶ 17 Although the case has been made that spoliation poses unique problems in the electronic context, the question remains as to whether a Federal Rule addressing the problem is the most desirable solution. The basic case for the adoption of a Federal Rule of Civil Procedure is a desire for uniformity. One of the primary purposes of the Federal Rules of Civil Procedure is to cultivate uniformity in the federal courts by getting away from local rules.⁵⁶ Courts have been adopting local rules relating to electronic discovery.⁵⁷ Among those courts with local rules related to electronic discovery, several districts address spoliation⁵⁸ and several are silent on the issue.⁵⁹ Of those which address spoliation, the approaches vary. Some local rules rely on the agreement of the parties during meet-and-confer sessions to sort out what types of often overwritten information must be preserved.⁶⁰ Other districts require that the parties file a report with the court stating whether "reasonable measures have been taken to preserve potentially discoverable data from alteration or destruction in the ordinary course of business or otherwise."⁶¹

¶ 18 Differing local rules are not the only source of incongruity. In the absence of local rules, some courts have fashioned rules of decision related to spoliation.⁶² Further, several organizations have adopted best practices in an attempt to guide courts in their approach to dealing with the spoliation of electronic evidence.⁶³ Although these best practices are helpful, their opt-in advisory nature necessarily implies that while they may foster increased uniformity, they will not achieve uniformity at the same level as a mandatory Federal Rule. The adoption of local rules and court rules of decision, as well as the publication of independent advisory recommendations, prompted the Advisory Committee to address electronic discovery spoliation in the Federal Rules.⁶⁴

¶ 19 Although a Federal Rule would increase uniformity, the Committee is limited in its rule-making power by the Rules Enabling Act. The more definite and precise a rule, the more likely it is to foster uniformity. As this Note will show, however, a precise rule defining what electronic information is sanctionable if spoiled could violate the substantive limitation of the second sentence of the Rules Enabling Act.

V. Proposed Changes to Rule 37

¶ 20 Having made the case for individualized treatment and argued that a change to the Federal Rules would help ensure uniformity, this Note will now look deeper at the proposed changes to Rule 37. The Committee has proposed an entirely new subsection intended to specifically address spoliation sanctions:

(f) Electronically stored information. Absent exceptional circumstances, a court may not impose sanctions under these rules on a party for failing to provide electronically stored information lost as a result of the routine, good-faith operation of an electronic information system.⁶⁵

¶ 21 The rule acknowledges that a company does not have to stop breathing (e.g., halt all normal computer operations) at the commencement of suit just because some data might be lost through routine operation.⁶⁶ When viewed against the proposed Rule 26(b)(2) requirement that only "accessible material" is discoverable, the Rule essentially establishes that companies cannot be sanctioned for routine spoliation of material not reasonably accessible.⁶⁷ This establishes a two-tiered approach⁶⁸ to electronic discovery that acknowledges the problems of volume and form inherent with electronic discovery and ensures that the multiple goals of the litigation system are not subverted into an unrestrained attempt to seek truth at all costs.

¶ 22 The proposed change is also notable for what it does not do. The rule does not modify the sanctions currently available under Rule 37 if the spoliation of electronic discovery is found to be improper.⁶⁹ Under Rule 37, appropriate sanctions could include monetary penalties, the exclusion of evidence, adverse inference jury instructions, dismissal or default judgment, and potential criminal penalties.⁷⁰ The rule does not articulate a positive duty of preservation nor does it address the preservation or destruction of information prior to the filing of the lawsuit.⁷¹

¶ 23 There has been considerable debate regarding what level of culpability should strip the responding party of the safe harbor.⁷² The Committee changed the articulation of the standard from the 2004 version, to a good-faith standard.⁷³

VI. Spoliation and the Rules Enabling Act

¶ 24 Before analyzing the proposed changes to Rule 37 under the REA, it is necessary to gain an understanding of the actual limitations imposed by the Act. Unfortunately, this is no easy task. The Supreme Court has never declared a Federal Rule of Civil Procedure invalid under the Act.⁷⁴ Because the Court's jurisprudence in this area is thin at best, it is worthwhile to examine the contours of the procedural versus substantive limitation as interpreted by leading scholars.

¶ 25 In Sibbach v. Wilson & Co., the Court announced that the Rules Enabling Act is not violated as long as "a [Federal] Rule [of Civil Procedure] really regulates procedure, - the judicial process for enforcing rights and duties recognized by substantive law and for justly administering remedy and redress for disregard or infraction of them."⁷⁵ According to this test, a rule is valid as long as it is truly a procedural rule.⁷⁶ As a formal matter, this test is not a principled application of the Rules Enabling Act because it effectively erases the substantive limitation from the Act.

¶ 26 The second sentence of the Act indicates that the drafters of the Rules Enabling Act recognized and were concerned with the fact that a "truly procedural" rule may still have an impermissible effect on substantive rights.

¶ 27 If the Sibbach rule were to be taken seriously, any Federal Rule that had at least some procedural purpose related to the "judicial process"⁷⁷ of the litigation system would be valid.

¶ 28 Later cases have recognized that the substantive limitation is more than a mere surplus. The Court has also recognized that the substance versus procedure distinction is not mutually exclusive. In Burlington N. R.R. Co. v. Woods, ⁷⁸ Justice Marshall held that a rule does not violate the Rules Enabling Act if its only impact on substantive rights is incidental: "[r]ules that incidentally affect litigants' substantive rights do not violate this provision if reasonably necessary to maintain the integrity of that system of rules."⁷⁹

¶ 29 The Court's jurisprudence offers but a glimpse into the full implications of the substantive limitation of the REA. REA challenges have most often come before the Court because a particular Federal Rule of Civil Procedure arguably modifies, abridges, or enlarges an existing state legal right. One question left unanswered by the Court pertains to exactly what constitutes a substantive right: are the substantive rights referred to in the Act limited to existing state substantive rights? Professor Burbank has persuasively argued that, while the REA does have federalism implications, the REA was enacted to allay separation of powers concerns.⁸⁰ The history of the REA suggests that not only does the limitation imposed by the REA prevent court rules from trammeling on existing substantive legal rights, but the purpose of the limitation also serves "to foreclose the creation in court rules of rights that would approximate the substantive law in their effect on person or property."⁸¹ A finding that the limitation imposed by the Act only forecloses the infringement of existing legal rights would effectively turn the REA into a rule of priority, rather than a rule which actually limits the Court's substantive rule-making power in a way that respects separation of powers concerns.

¶ 30 Another question left unanswered by the Court's jurisprudence is how much of the effect on substantive rights is merely incidental. Although the Court has upheld rules as only affecting substantive rights incidentally,⁸² the Court has never invalidated a Federal Rule as having too much of a substantive effect. Therefore, a rule's validity can only be evaluated with confidence if its substantive effect is less than the rules discussed in the cases. If the substantive effect is arguably greater or difficult to analogize, the Court's jurisprudence is of little help.

¶ 31 One could argue that there is no real upper bound to the permissible substantive effects because the Court has explicitly stated that a promulgated rule is presumptively valid.⁸³ According to Marshall, "[T]he study and approval given each proposed Rule by the Advisory Committee, the Judicial Conference, and this Court, and the statutory requirement that the Rule be reported to Congress for a period of review before taking effect give the rules presumptive validity under... the... statutory constraints [of the Rules Enabling Act]."⁸⁴ Since the rules automatically promulgate in the face of congressional silence, often the last affirmative act before promulgation is the vote of the Supreme Court.⁸⁵ Therefore, the Court is essentially abdicating to itself. With the limits of the Rules Enabling Act acting as a veritable black box of statutory interpretation, it is possible to suppose that the substantive limitation of the Rules Enabling Act has no real force in practice. However, even a realist rendering of the Act must acknowledge that the Advisory Committee is concerned with the limits imposed by the Act and drafts rules in consideration of the substantive limit.⁸⁶ Since the contours of the substantive limitation are unclear in light of Supreme Court precedent, it will be helpful to consider how leading scholars have sculpted the outer contours of the Act.

¶ 32 Professor Ely accepts the Sibbach analysis as it pertains to the checklist-like power to proscribe procedural rules.⁸⁷ However, he asserts that a rule should be viewed as substantive if it has a purpose or purposes "not having to do with the fairness or efficiency of the litigation process."⁸⁸ As a paradigmatic example, Ely analyzes a statute of limitations.⁸⁹ Although such statutes have procedural purposes⁹⁰ and thus satisfy the first sentence of the Act, they also have the substantive purpose of guiding individuals' primary behavior by allowing them peace of mind to plan without fear of suit—and thus move on with their lives.⁹¹ Therefore, according to Ely, the statute violates the substantive limitation of the Act.

¶ 33 Professor Carrington⁹² also agrees that the substance and procedure distinction is not mutually exclusive.⁹³ Unlike Professor Ely's test, Carrington's second sentence test does not inhere in the purpose of the rule. Instead, his test says that a rule does not affect a substantive right if its "application is sufficiently broad to evoke no organized political attention of a group of litigants or prospective litigant who (reasonably) claim to be specially and adversely affected by the rule."⁹⁴ Carrington's test is essentially an interpretation which looks to political reaction as a litmus test for whether or not a rule extends beyond the substantive limitation.

¶ 34 As a threshold matter, the proposed rule satisfies the first sentence of the Act because it relates to the fair and efficient administration of the litigation process.⁹⁵ By guiding the court as to when sanctions are appropriate, the rule instructs as to situations where, for example, an adverse inference will apply to spoiled material. Such a rule is critical to ensuring that the litigation process can continue if the producing party spoils evidence. The rule is concerned with fairness because it acknowledges the unique problems associated with preserving electronic data and strikes a balance between the search for truth and the potential burdens of overly harsh electronic spoliation sanctions.⁹⁶

¶ 35 Commentators have criticized the proposed "routine operation" standard as too vague.⁹⁷ However, a specific definition of routine operation would violate the substantive limitation of the second sentence of the Act. By adopting a standard that simply considers the status quo, the Committee avoided violating the substantive limitation.

¶ 36 A specific definition of routine operation would effectively establish a positive duty of preservation.⁹⁸ Such a duty would have a significant impact on primary behavior unrelated to any specific litigation. Even though sanctions would be limited to spoliation that occurred after litigation was reasonably likely, companies would have to consider the specific duty of preservation when designing their computer systems. If a company's system routinely deleted information that did not fit within the specific safe harbor, a company would have to either redesign the system or halt normal operations to avoid sanctions. For companies under constant threat of litigation, such a rule would establish a quasi-regulatory requirement.⁹⁹ Despite the Supreme Court's lack of clarity, a positive duty such as this seems to clearly have more than an incidental effect on substantive matters outside the four corners of the courtroom. The drafters avoided substantial effects by pegging the availability of the safe harbor to what is already considered routine.

¶ 37 Even though a positive duty to preserve information would help companies decide how to configure their computer systems, such a purpose would also render the rule invalid under Professor Ely's interpretation of the second sentence of the Rules Enabling Act. This is because providing system building guidance to companies is not related to the fair and efficient administration of the courtroom.¹⁰⁰ By constructing the safe harbor in relation to the existing status quo, the drafters steered clear of acting as legislators because the rule does not have a purpose designed to address an issue outside of the four corners of the courtroom.

¶ 38 A puzzle exists when trying to interpret either the proposed rule as written or a rule with a specific preservation duty under Professor Carrington's interpretation of the Act. The proposed rule is the result of considerable advocacy by the defense bar.¹⁰¹ Clearly, the rule does not only attract political attention, but owes its very existence to focused efforts of the corporate defense bar. One problem with Carrington's formation is that it does not reflect a modern reality: the sophistication of both the defense bar and trial lawyers. Although Carrington is convinced that certain rules of general application would not attract intense political attention, his examples are procedural rules that have existed for so long that their lack of controversy is merely a function of their age. It is almost impossible to imagine a proposed new rule or change that would not receive considerable attention from one side of the bar or another. Therefore, it is unlikely that the Committee, the Supreme Court, or Congress will find the proposed changes to Rule 37 invalid under Carrington's test.

VII. Normative Analysis

¶ 39 Viewed broadly, the rule as proposed is a favorable solution to the vexing problem of spoliation in the electronic context. It addresses the problem of electronic discovery spoliation while effectively balancing the multiple aims of the litigation process. Since the safe harbor will permit unsanctioned spoliation when it occurs as the result of routine operation, the rule will reduce the truth efficacy of the system. However, it will do so in a way that balances the desire for truth against the desire for an efficient litigation system. This is appropriate because the marginal benefit of truth at the point of preserving, for example, all residual data, is outweighed by the financial burdens imposed on companies required to preserve all such information.

¶ 40 One argument against the spoliation provision reflects the concern that in some situations material with known relevance and importance to the litigation might be excusably destroyed. The rule, however, contemplates that when the truth finding stake is high, routinely deleted information may be subject to spoliation sanctions if protected under a specific court order.

¶ 41 Even if one does not accept that the Rules Enabling Act prevents the Rules Advisory Committee from promulgating a spoliation rule that defines a uniform preservation duty, there are persuasive functional concerns that favor the proposed rule as written. As a matter of managerial justice,¹⁰² the electronic spoliation problem is appropriately addressed through a directed-managerial model. The Advisory Committee seeks to send a policy-based message through the promulgation of the proposed rule—that electronic spoliation needs to be uniformly addressed and excused in certain limited circumstances.¹⁰³ Those in favor of promulgating a positive preservation duty might argue that the precision of such a rule would more effectively and uniformly further the Advisory Committee's policy.¹⁰⁴ However, as Professor Redish has noted regarding conditional cost-shifting in the electronic context, "a distinction must be recognized between policy discretion and implementational discretion."¹⁰⁵ Accepting that total policy discretion is not the best solution,¹⁰⁶ the rule does not have to be so precise as to stave away all judicial implementation discretion. Such a rejection of managerial judging¹⁰⁷ in the electronic discovery context does not necessitate a fetish for precision. By pegging the scope of excusable spoliation to what is routine, the Rules Advisory Committee can effectuate its broad policy goal without gutting the implementational discretion of judges.¹⁰⁸

¶ 42 Implementation discretion is appropriate in the electronic spoliation context for several reasons. Technology is rapidly changing.¹⁰⁹ A rule that would fix what is considered routine would be too static to remain relevant for very long. Indeed, by the time a rule is promulgated, it is conceivable that technology might outpace the positive preservation duty. Another practical consideration that favors implementation discretion is the routine operation standard's ability to account for differences in systems among industries. Not only is such a rule flexible and easier to apply to individual litigants, it invites less criticism from particular industries or interests. If the Advisory Committee were to promulgate a specific preservation duty, the across-the-board preferred system architecture would likely be more similar to some sector practices than others. Therefore, a more definite standard would favor some technologies over others and invite criticism from industries bearing a disproportionate amount of the conversion burden.

¶ 43 Even if one accepts that the proposed rule effectively balances the multiple aims of the litigation matrix and that a positive preservation duty violates the Rules Enabling Act or is functionally maladapted the reality of managing electronic discovery in the judicial context, one might argue that anchoring the scope of allowable spoliation to what is routine may still present some functional concerns. Allowing the contours of proscribable behavior to shift in accordance with what is customary or routine is not a novel concept to the law. Custom is often used by courts to help sculpt the standard of care in negligence actions at common law.¹¹⁰ The concept of custom is attractive because it seems grounded in fairness, respects the non-legal market rationales for engaging in certain primary behavior, and seems workable since an industry standard can arguably be revealed by expert testimony.¹¹¹

¶ 44 An examination of the tort case law, however, reveals a serious problem which may arise when proscribable behavior is defined by custom. Sometimes the industry practice as a whole may contravene the policy or purpose of pegging a standard to what is customary. The case of The T.J. Hooper¹¹² is illustrative. In The T.J Hooper, the court found that the defendants breached their duty of care even though they were following industry practice.¹¹³ The court reasoned that the industry as a whole had unduly lagged in the adoption of new and available safety devices.¹¹⁴ The industry's practice as a whole was no longer a proxy for the duty of care since "there are precautions so imperative that even their universal disregard will not excuse their omission."¹¹⁵ Since custom no longer defined what was reasonable—liability in the face of unreasonable behavior being the underlying policy of the rule—the anchoring of the duty of care to what was customary was no longer warranted.

¶ 45 Similarly, anchoring sanctions to routine operation could lead to problematic results if what is customary actually contravenes the purpose or rationale behind the spoliation safe harbor. Two scenarios are possible. First, a particular industry may have systems that are more destructive to data than other industries, not as a function of the industry, but because the industry has failed to adopt the available new technology.¹¹⁶ Following this logic, if a new technology less destructive of data¹¹⁷ is available, in industries affected by a high degree of litigation risk, the proposed safe harbor may actually create a disincentive to invest in such technology.¹¹⁸ This is problematic because one of the purposes of anchoring to routine operation is the desire to peg the standard to a criterion unrelated to the litigation process. The proposed rule would collapse into itself if industry practice was actually a function of industry reaction to Rule 37. Further, a positive articulated duty of preservation was disfavored partially because of concerns that it could fix industry practice. However, if the proposed rule's disincentive effects are strong, it could also lead to a freezing of industry practice or a slowing of new technology adoption.

¶ 46 Anchoring to industry practice raises a second concern if industries collude, either explicitly or tacitly, to reduce litigation risk by configuring their systems in a manner more destructive to data. Such collusion would lead to a broader safe harbor for the industry. This would be problematic because the purpose of the spoliation safe harbor is to respect the non-litigation business reasons for creating systems that routinely destroy data—not to create an incentive for industry to create systems that are more destructive of such data.

¶ 47 These concerns are not addressed in the proposed Rule 37 as written. The Advisory Committee should have addressed these concerns prior to submitting the final version of the Rule 37 amendment.

VIII. CONCLUSION

¶ 48 The Judicial Conference of the United States has approved and sent to the Supreme Court several amendments and proposed changes to the Federal Rules of Civil Procedure relating to the discovery of electronic information. The Rules Advisory Committee has recognized that, in relation to traditional documents, electronic data poses unique problems in the discovery context. These problems are largely attributable to the sheer volume of electronic data available compared to traditionally discoverable documents. To further complicate matters, most traditional documentation is stored for archival purposes; however, electronic data exists in many task specific forms. Therefore, a significant amount of electronic data exists, not as an archive, but for some other system-related purposes such as reserving hard-drive space for overwriting. Since certain data in a given system often serves a temporary purpose unrelated to the archiving of user data, the inadvertent spoliation of electronic information is likely and therefore poses a serious sanction risk for companies facing litigation.

¶ 49 Truth-finding is not the only goal of the litigation system. If it were, an electronic spoliation safe harbor might be unwarranted. However, the litigation system should respect the undue burden a harsh draconian spoliation rule would have on individual companies both in terms of business expense and litigation costs. Recognizing these concerns, the Rules Advisory Committee has proposed an amendment to Rule 37 that would provide a sanction safe harbor for electronic discovery spoiled as a result of routine operation. Some commentators are concerned that the rule is too vague and that the rule should specifically articulate the preservation duty potential litigants face. However, the routine operation safe harbor, unlike a rule specifying a positive preservation duty, does not violate the Rules Enabling Act, and is functionally sound. A positive duty of preservation would force high litigation risk companies that were non-compliant to either modify their systems at great expense or suffer sanctions. Although such a rule would help signal companies as to what is required of their systems, such a purpose and effect would be improper under the Rules Enabling Act because such a rule would essentially legislate system requirements for companies. Further, even if readers do not find the Rules Enabling Act limitation persuasive, from a functional prospective, a specific articulation of the duty of preservation would be inflexible to changes in technology, may actually provide a disincentive towards the adoption of new technology, and would favor some industries over others. The anchoring of the safe harbor to what is, and will be, routine avoids these negative functional consequences. However, the Advisory Committee should have addressed a possible issue that could arise—that industries could contravene the purpose of the rule by tacitly or explicitly colluding to adopt or retain industry-wide systems that are more destructive of data.

ENDNOTES

^*Student, Northwestern University School of Law, Candidate for J.D. 2006; DePaul University, B.S. 1999. Thank you to Professor Martin Redish for his valuable feedback and my wife Katharine for her support during the development of this Note.

¹See, e.g., Stephen D. Williger & Robin M. Wilson, Negotiating the Minefields of Electronic Discovery, 10 Rich. J.L. & Tech. 52 (2004); Robert Douglas Brownstone, Collaborative Navigation of the Stormy E-Discovery Seas, 10 Rich. J.L. & Tech. 53 (2004).

²See, e.g., Benjamin Sotelo & James A. Flanagan, How Firms Can Help Clients Prepare for Electronic Discovery, Los Angeles Lawyer, Dec. 2003, at 45 ("Although data discovery is like paper discovery, new methods must be developed."); Kris K. Haworth, Effectively Dealing with Large Scale Electronic Discovery, Glasser LegalWorks, 2002, at 273.

³See, e.g., the entire Zubulake line of cases: Zubulake v. UBS Warburg LLC, 231 F.R.D. 159 (S.D.N.Y. Feb. 3, 2005); Zubulake v. UBS Warburg LLC, 220 F.R.D. 212 (S.D.N.Y. Oct 22, 2003); Zubulake v. UBS Warburg LLC, 216 F.R.D. 280 (S.D.N.Y. Jul. 24, 2003); Zubulake v. UBS Warburg LLC, No. 02 Civ. 1243(SAS), 2003 WL 21087136, at *1 (S.D.N.Y. May 13, 2003); Zubulake v. UBS Warburg LLC, 217 F.R.D. 309 (S.D.N.Y. May 13, 2003).

⁴For the current Federal Rules of Civil Procedure related to discovery, see Fed. R. Civ. P. 26-37.

⁵See, e.g., Martin H. Redish, Electronic Discovery and the Litigation Matrix, 51 Duke L.J. 561 (2001) (discussing how courts or rule drafters should adopt a cost-shifting model and recognize that electronically stored information deserves special treatment in the spoliation context); Thomas Y. Allman, The Case for a Preservation Safe Harbor in Requests for E-Discovery, 70 Def. Counsel J. 417 (2003) ("Despite the Courts' increased attention to dragnet requests for production of electronic materials, the scope of preservation should be addressed."); Thomas Y. Allman, The Need for Federal Standards Regarding Electronic Discovery, 68 Def. Counsel J. 206 (2001).

⁶For a description of key differences, see infra notes 39-55 and accompanying text. Discoverable material is characterized under the current Rules as books, documents, communications, data compilations, and tangible things. See Fed. R. Civ. P. 26.

⁷For a lengthy discussion of the process employed by the committee over the five year period, see Kenneth J. Withers, Two Tiers and a Safe Harbor: Federal Rulemakers Grapple with E-Discovery, 51 Fed. Lawyer 29 (2004).

⁸See Letter from Lee H. Rosenthal, Chair, Advisory Committee on Federal Rules of Civil Procedure, to Honorable David F. Levi, Chair, Standing Committee on Rules of Practice and Procedure (May 17, 2004, revised Aug. 3, 2004), available at http://www.uscourts.gov/rules/comment2005/CVAug04.pdf (Report of the Civil Rules Advisory Committee Regarding Proposed Electronic Discovery Rules).

¹⁰Kenneth J. Withers, We've Moved The Two Tiers And Filled In The Safe Harbor, 52 Fed. Lawyer 50 (2005).

¹¹September 2005 Summary of the Report of the Judicial Conference Committee on Rules of Practice and Procedure, available at http://www.uscourts.gov/rules/Reports/ST09-2005.pdf [hereinafter 2005 Proposed Changes].

¹⁴The package of proposals aimed at electronic discovery consists of changes to Federal Rules 16, 26, 33, 34, 37, and 45, along with changes to Form 35. See Letter from Lee H. Rosenthal to David F. Levi, supra note 8, at 1; 2005 Proposed Changes, supra note 11.

¹⁵Withers, supra note 7, at 30 (The changes to Rule 26, 16, and 35 ensure that counsel address specific issues related to electronic discovery during the discovery planning meeting.).

¹⁶Id. at 33. The Rules were amended in 1970 to include "data compilations." See Fed. R. Civ. P. 34(a) advisory committee's note (1970). Compilations do not, however, necessarily encompass the mass of intangible data that can be presented in a dynamic or live format. See Withers, supra note 7, at 33.

¹⁷See the proposed changes to Federal Rules of Civil Procedure 33, 34, and 45. Letter from Lee H. Rosenthal to David F. Levi, supra note 8; 2005 Proposed Changes, supra note 11.

¹⁸See the proposed changes to Federal Rules of Civil Procedure 34 and 35. Letter from Lee H. Rosenthal to David F. Levi, supra note 8; 2005 Proposed Changes, supra note 11.

¹⁹See, Withers, supra note 7, at 34-35.

²⁰See the proposed changes to Federal Rules of Civil Procedure 33, 34, and 45. Letter from Lee H. Rosenthal to David F. Levi, supra note 8; 2005 Proposed Changes, supra note 11.

²¹See Richard Marcus, Only Yesterday: Reflections on Rulemaking Responses to E-Discovery, 73 Fordham L. Rev. 1, 12 (2004).

²²See Withers, supra note 7, at 36.

²⁴Under such agreements, "documents are produced to the opposing party before or without a review for privilege, confidentiality, or privacy." The (2004) Sedona Principles: Best Practices, Recommendations & Principles for Addressing Electronic Document Production, 5 Sedona Conf. J. 151, 188 (2004) [hereinafter Sedona Principles]. The parties agree to "stringent guidelines and restrictions to prevent the waiver of confidentiality and privilege[,] . . . [I]f the requesting party finds a document that appears to be privileged, the producing party can 'claw back' the document without having waived any privilege." Id. at 188-89.

²⁵See the proposed changes to Federal Rules of Civil Procedure 26 and 45; Letter from Lee H. Rosenthal to David F. Levi, supra note 8; 2005 Proposed Changes, supra note 11.

²⁶See Martin H. Redish, supra note 5, at 587 ("Electronic documents tend to remain on a computer system well after similar paper records would be discarded, because there is no obvious pile of paper to remove.").

²⁷See, e.g., Beyers v. Ill. State Police, 53 Fed. R. Serv. 3d 740 (N.D. Ill. 2002).

²⁸See infra notes 39-55 and accompanying text.

²⁹See Letter from Lee H. Rosenthal to David F. Levi, supra note 8, at Federal Rules of Civil Procedure Proposed Amendments 11. As the Committee notes, discovery of such material may be compelled for good cause. Id. at Federal Rules of Civil Procedure Proposed Amendments 12. For the good cause provision, see Proposed Change to Federal Rule of Civil Procedure 26(b)(2). Id. at Federal Rules of Civil Procedure Proposed Amendments 4-6. See also 2005 Proposed Changes, supra note 11.

³⁰See Withers, supra note 7, at 30; Withers, supra note 10.

³¹See discussion infra Part V.

³²See Withers, supra note 10, at 54 (discussing the criticisms offered by lawyers and judges). See also Anita Ramasastry, The Proposed Federal E-Discovery Rules: While Trying to Add Clarity, the Rules Still Leave Uncertainty, Find Law, Sept. 15, 2004, http://writ.news.findlaw.com/ramasastry/20040915.html (arguing that the committee should have proposed a rule that (1) clearly stipulates the number of years a company should retain data and (2) sets forth type of preservation safeguards a company must employ to take advantage of the safe harbor).

³³28 U.S.C. § 2072 (1982).

³⁴Id. at para. (a).

³⁵Id. at para. (b).

³⁶See discussion infra Part VI.

³⁷See, e.g., Stephen B. Burbank, Hold the Corks: A Comment on Paul Carrington's "Substance" and "Procedure" in the Rules Enabling Act, 1989 Duke L.J. 1012 (1989).

³⁸See Ramasastry, supra note 30.

³⁹Manual for Complex Litigation (Fourth) § 11.446 (2003).

⁴⁰Dale Cendali et al., Electronic Discovery, 797 Practicing L.J. 323, 335 (2004).

⁴³An example is the information which automatically populates the "properties" window associated with a Microsoft Word document (e.g., author, last updated, created, etc.).

⁴⁴Cendali, supra note 40, at 336.

⁴⁸Id. at 336-37.

⁴⁹Cendali, supra note 40, at 336-37.

⁵²The Terminator (Metro-Goldwyn-Mayer 1984).

⁵³Redish, supra note 24, at 622.

⁵⁵Some exceptions should apply. See discussion infra Part VI.

⁵⁶Hanna v. Plumer, 380 U.S. 460, 472 (1965) (Warren, C.J., quoting Lumbermen's Mut. Cas. Co. v. Wright, 322 F.2d 759, 764 (5th Cir. 1963)).

⁵⁷See, e.g., Miss. Court Order 13 (May 29, 2003) amending Miss. Rule of Civil Procedure 26; Tex. Rule of Civil Procedure 193.3(d) (privilege not waived by production) and Rule 196.4 (electronic or magnetic data); District of Ark., Eastern and Western, Local Rule 26.1; District of Del., Default Standards for Discovery of Electronic Documents; District of Kan., Electronic Discovery Guidelines; District of N.J., Local Rule 26.1; District of Wyo., Local Rule 26.1.

⁵⁸See, e.g., District of N.J., Local Rule 26.1; District of Wyo., Local Rule 26.1; District of Kan., Electronic Discovery Guidelines; District of Del., Default Standards for Discovery of Electronic Documents.

⁵⁹See, e.g., Miss. Court Order 13 (May 29, 2003) amending Miss. Rule of Civil Procedure 26; Tex. Rule of Civil Procedure 193.3(d) (privilege not waived by production) and Rule 196.4 (electronic or magnetic data).

⁶⁰See, e.g., District of N.J., Local Rule 26.1 (parties to decide whether Legacy data must be preserved); District of Wyo., Local Rule 26.1 (parties to decide whether back-up data or "deleted" residual data must be preserved); District of Kan., Electronic Discovery Guidelines (parties to agree as to what information is to be preserved to prevent spoliation sanctions); District of Del., Default Standards for Discovery of Electronic Documents (parties to discuss preservation issues and depose retention coordinators if necessary).

⁶¹District of Ark., Eastern and Western, Local Rule 26.1.

⁶²See, e.g., Zubulake v. UBS Warburg LLC, 220 F.R.D. 212, 217 (S.D.N.Y. 2003) (recognizing that "a party need not preserve all backup tapes even when it reasonably anticipates litigation").

⁶³See, e.g., Sedona Principal 5: "The obligation to preserve electronic data and documents requires reasonable and good faith efforts to retain information that may be relevant to pending of threatened litigation. However, it is unreasonable to expect parties to take every conceivable step to preserve all potentially relevant data." Sedona Principles, supra note 22; Jason Krause, Unlocking Electronic Evidence, 3 No. 5 A.B.A. J. E-Report 5 (2004) (discussion of ABA electronic discovery draft standards).

⁶⁴See Withers, supra note 7, at 30.

⁶⁵See 2005 Proposed Changes, supra note 11, at Appendix C p. 84.

⁶⁷See supra notes 26-38 and accompanying text.

⁶⁸See Withers, supra note 7.

⁷⁰See Lisa Arent et. al, Ediscovery: Preserving, Requesting & Producing Electronic Information, 19 Santa Clara Computer & High Tech. L.J. 131 (2002).

⁷¹See Withers, supra note 7, at 39.

⁷²See, e.g., Thomas Y. Allman, The Case for a Preservation Safe Harbor in Requests for E-Discovery, 70 Def. Couns. J. 417, 422 (2003) (willfulness required).

⁷³See Withers, supra note 7, at 39; Withers, supra note 10, at 54.

⁷⁴Leslie M. Kelleher, Taking "Substantive Rights" (in the Rules Enabling Act) More Seriously, 74 Notre Dame L. Rev. 47, 105 (1998).

⁷⁵312 U.S. 1, 14 (1941).

⁷⁸480 U.S. 1 (1987).

⁸⁰Stephen B. Burbank, The Rules Enabling Act Of 1934, 130 U. Pa. L. Rev. 1015 (1982).

⁸²See, e.g., Business Guides, Inc. v. Chromatic Communications Enterprises Inc., 498 U.S. 533 (1991) (upholding the validity of Rule 11 against a Rules Enabling Act challenge).

⁸³Burlington N., 480 U.S. at 6.

⁸⁵28 U.S.C. § 2074.

⁸⁶See Withers, supra note 7, at 39.

⁸⁷John Hart Ely, The Irrepressible Myth Of Erie, 87 Harv. L. Rev. 693, 718-19 (1974).

⁹⁰For example, keeping dockets down and preventing stale evidence. Id.

⁹²Paul Carrington, "Substance" and "Procedure" in the Rules Enabling Act, 1989 Duke L.J. 281 (1989).

⁹⁵See Ely, supra note 87.

⁹⁶See Redish, supra note 26.

⁹⁷See Ramasastry, supra note 32.

⁹⁹Even though such a quasi-regulatory environment might not abridge, modify, or enlarge existing substantive legislation, a positive duty of preservation would impermissibly "approximate the substantive law" in its foreseeable effect on corporate systems architecture. See Burbank, supra note 80, at 1114.

¹⁰⁰See Ely, supra note 87.

¹⁰¹See Withers, supra note 7, at 38 ("As soon as it became known that the advisory committee was studying the issue of electronic discovery, members of the corporate defense bar offered a variety of proposals to limit the power of courts to impose sanctions for the inadvertent or nonwillful destruction of electronically stored data.").

¹⁰²See generally Judith Resnik, Managerial Judges, 96 Harv. L. Rev. 374 (1982).

¹⁰³See supra notes 39-64 and accompanying text.

¹⁰⁴Ramasastry, supra note 32 (arguing that "[c]lear limits and rules would be helpful").

¹⁰⁵Redish, supra note 26, at 617.

¹⁰⁶See supra notes 39-64 and accompanying text.

¹⁰⁷See generally Resnik, supra note 102.

¹⁰⁸The 2005 changes make clear that judges are to exercise implementation discretion with respect to Rule 37(f). The rule makers achieved this by adding the modifier "[a]bsent exceptional circumstances" to Rule 37(f). It is up to judges to define and give content to exactly what exceptional circumstances will prevent a litigant from taking advantage of the safe harbor. See 2005 Proposed Changes, supra note 11; Withers, supra note 10, at 54.

¹⁰⁹A Business Week poll claimed that forty million people browsed the Web in 1997—almost double that of 1996. Joseph I. Rosenbaum, Privacy, Jurisdiction, and the Regulation of Free Expression PRIVACY ON THE INTERNET: WHOSE INFORMATION IS IT ANYWAY?, 38 Jurimetrics J. 565, 568 (1998).

¹¹⁰See, e.g., The T.J. Hooper, 53 F.2d 107 (S.D.N.Y. 1931); Brune v. Belinkoff, 235 N.E.2d 793 (Mass. 1968) (locality rule in medicine).

¹¹¹See generally Jay M. Zitter, Annotation, Standard Of Care Owed To Patient By Medical Specialist As Determined By Local "Like Community," State, National, Or Other Standards, 18 A.L.R. 4th 603 (2005).

¹¹²60 F.2d 737 (2d Cir. 1932).

¹¹⁶A similar lag compelled the Second Circuit to expand the duty of care beyond what was customary in The T.J. Hooper. 60 F.2d 737 (2d Cir. 1932).

¹¹⁷For example, a new database system might be available that compresses archived data and thus takes up the same or less storage space than the current system. Such a system would likely recycle data less often and lead to less spoliation risk.

¹¹⁸The power of such a disincentive is an empirical question. The increased litigation risk of storing more data with a hypothetical new technology would have to outweigh the non-litigation benefits of adopting such a technology for the disincentive to actually serve to prevent the new technology from being acquired.

I. INTRODUCTION