I. Introduction

The need for an effective federal obstruction of justice[1] statute has never been greater. In recent years corporate con artists have often been able to hide the evidence necessary to convict them. By shredding numerous incriminating documents many wrongdoers have been able to evade conviction and, thanks to generous loopholes in federal obstruction statutes, they have avoided any serious consequences for blatantly destroying evidence.[2] After federal authorities narrowly succeeded in applying the federal obstruction statutes to convict the accounting firm of Arthur Andersen for its destruction of Enron-related documents, Congress interceded to modernize and strengthen federal power to prosecute the destruction of evidence.[3] The Sarbanes-Oxley Act[4] produced, among other things, 18 U.S.C. § 1519, which criminalizes any "knowing" destruction of documents with the intent to impede a "contemplated" investigation.[5]

Although § 1519 closes many loopholes existing under prior obstruction statutes, its broad language may in fact go too far in punishing the destruction of corporate documents. Even before its passage, many critics worried that businesses would be severely punished for destroying documents that they did not realize were necessary for a "contemplated" investigation.[6] To make matters worse, many businesses routinely purge superfluous electronic data, including emails and redundant documents from their computer systems.[7] Such minor deletions, which are a necessity in modern data storage, could result in severe criminal sanctions against a business or its operators.

Although no cases have yet been decided under this new statute, the first indictments have been made, and federal courts will likely begin hearing these cases and interpreting the terms of § 1519 within the next year. This article argues that courts should take a balanced approach in defining the scope of § 1519 by adopting the type of foreseeability test that federal courts already use in analyzing civil spoliation cases. Such an approach will punish those who destroy documents when a federal investigation is foreseeable, but will not punish the routine destruction of documents under a reasonable retention program when no investigation is foreseeable.

Part II of this article will discuss the history and development of federal laws prohibiting document destruction, and Part III will describe the recent revision of these statutes in the Sarbanes-Oxley Act of 2002. Part IV then argues that the revised provisions should not be interpreted too broadly to avoid exposing companies to unnecessary criminal sanctions. Part V provides a brief conclusion.

II. Federal Obstruction of Justice: Origins and Overview

A. Development of Federal Obstruction Power

Federal obstruction of justice laws generally prohibit any attempt to interfere with the administration of justice.[8] Obstruction can come in many forms: attempts to influence witnesses, jurors, and court officers, destruction of evidence, or interfering with law enforcement investigations.

Federal power to prosecute obstruction of justice has its origin with the federal courts' common law contempt power, which gives judges general authority to assure that no person wrongfully interferes with the process of adjudication.[9] These powers were gradually codified,[10] with the modern obstruction statutes first appearing in 1948 after a massive congressional revision of the criminal portions of the U.S. Code.[11] This revised obstruction scheme prohibits a variety of conduct including efforts to influence, threaten, or injure jurors or officers of the court,[12] as well as interference with legislative or administrative proceedings.[13] The 1948 law also retained the prior obstruction language as an "omnibus" clause that would provide a catch-all for wrongful activities that were not expressly prohibited by the statutes.[14]

Federal obstruction laws were further developed in 1967 and 1982. In 1967, Congress added a further section prohibiting the use of bribery to prevent third parties from providing information to authorities.[15] In 1982, Congress passed the "Victim and Witness Protection Act," which prohibits anyone from (1) improperly influencing a witness's testimony, (2) inducing another to withhold testimony or to destroy or alter records, (3) otherwise hindering, delaying, or preventing a person from providing information to authorities,[16] and (4) retaliating against witnesses, victims, and informants.[17]

Until the passage of the Sarbanes-Oxley Act, none of these statutes directly addressed document destruction as an act of obstruction. Most of the case law regarding document destruction developed under the "omnibus" clause, and, after 1982, the Victim and Witness Protection Act's provision against inducing others to destroy documents. The caselaw derived from these two provisions, therefore, constitutes the corpus of federal law regarding destruction of documents in criminal cases.

B. The Omnibus Provision (18 U.S.C. § 1503(a))

1. General Provisions

Section 1503 deals primarily with acts of improperly influencing or injuring officers of the court or jury members.[18] However, woven into the complex language[19] of the statute is the "omnibus" clause, which provides that "[w]hoever . . . corruptly . . .obstructs, or impedes, or endeavors to . . . obstruct, or impede, the due administration of justice, shall be punished . . . ."[20] The Supreme Court noted that this provision is intended as a "catchall, prohibiting persons from endeavoring to influence, obstruct, or impede the due administration of justice."[21] Although the federal circuits have read the clause differently, most have held that this section generally prohibits destroying or concealing relevant documents[22] and falsifying records.[23] Some courts have also held that it applies to the destruction of documents in civil proceedings.[24]

2. Intent Element

On its face, § 1503 seems broadly to prohibit improperly destroying documents: the government need only show that the defendant acted "corruptly" to "obstruct" the "due administration of justice." Nonetheless, the courts have severely limited its scope in significant ways. First, the Supreme Court has used a "nexus" approach to find that § 1503 only applies to actions occurring after a judicial proceeding has actually commenced. Second, the circuit courts have further narrowed the scope of § 1503 by applying stringent definitions of "proceedings" and the type of intent required.

a. The "nexus" requirement. The Supreme Court limited the scope of § 1503 in United States v. Aguilar by confirming that § 1503 only applies to conduct occurring after a judicial proceeding had already commenced.[25] In that case, federal judge Robert Aguilar was charged with obstructing justice when he lied to FBI agents about his involvement with Michael Rudy Tham, who had been convicted of embezzling from a teamsters account and was suspected of racketeering.[26] After a grand jury began to investigate, the FBI questioned Aguilar about his involvement with Tham. According to the government's brief, one agent specifically mentioned that a grand jury investigation had commenced: "[T]here is a grand jury meeting. Convening I guess that's the correct word. Um, some evidence will be heard I'm . . . I'm sure on this issue."[27] Aguilar denied any substantial relationship with Tham.[28] However, it was later discovered that Aguilar had in fact warned Tham's associates of an FBI wiretap and that he had sought to pressure another judge to set aside Tham's conviction.[29] Aguilar was subsequently convicted of obstructing justice.[30]

Writing for the Court, Justice William Rehnquist reversed the § 1503 conviction. The majority reasoned that that under § 1503 "a person is not sufficiently charged with obstructing or impeding the due administration of justice in a court unless it appears that he knew or had notice that justice was being administered in such court."[31] The Court went on to adopt a "nexus" approach to defining intent under § 1503, which requires that "the act must have a relationship in time, causation, or logic with the judicial proceedings."[32] In other words, the defendant's conduct must have the "natural and probable effect" of interfering with the due administration of justice.[33]

In applying this analysis to the facts of the case, the majority found that Aguilar could not have had the necessary intent under § 1503 because, even if he were aware that a grand jury had convened, he did not know that his statements to the FBI agents would be reported to the grand jury.[34] Giving false testimony to a federal agent was not obstruction because "[w]e do not believe that uttering false statements to an investigating agent . . . who might or might not testify before a grand jury is sufficient to make out a violation of the catchall provision of § 1503."[35] "Such conduct," the Court continued, "falls on the other side of the statutory line from that of who delivers false documents to or testimony to the grand jury itself."[36] Finally, the Court concluded that a narrow rule was necessary because a broader reading would make it a crime for a husband to give false information to his wife on the grounds that the wife might in turn provide the false information to a grand jury.[37] Three justices dissented.[38]

b. Narrow interpretation of judicial proceeding. Since Aguilar, circuit courts have applied § 1503 to only a limited range of proceedings.[39] These have included grand jury proceedings[40] but not routine investigations,[41] unless it is part of a grand jury investigation.[42]

c. Narrow interpretation of "corrupt" intent. Following Aguilar, courts have also persistently interpreted "corrupt" intent to require the defendant to have specific intent to affect a particular judicial proceeding. Therefore, the requisite intent cannot be established where the defendant destroyed evidence or made false statements if no judicial proceedings had commenced.[43] Similarly, intent cannot be established when the court determines that the defendant was unaware that such a proceeding had commenced or when the investigating agents do not indicate that the evidence will be used in a judicial proceeding.[44] Finally, even when a defendant has received a grand jury subpoena for certain documents, he does not obstruct justice if he makes false statements regarding items that were not subpoenaed.[45] However, some courts have focused on Aguilar's language that the "natural and probable" effect of a defendant's actions is to obstruct a judicial proceeding.[46] Therefore, an "endeavor" or willingness to obstruct a proceeding is sufficient, even if it ultimately fails.[47]

C. The "Corrupt Inducement" Provision (§ 1512(b)(2))

Section § 1512(b) punishes those who command or induce others to destroy documents relevant to a proceeding:

Whoever knowingly uses intimidation, threatens, or corruptly persuades another person, or attempts to do so, or engages in misleading conduct toward another person, with intent to . . . cause or induce any person to . . . alter, destroy, mutilate, or conceal an object with intent to impair the object's integrity or availability for use in an official proceeding . . . .[48]

Therefore, in order to prove a § 1512(b) offence in the case of document destruction, the government must prove that the defendant (1) knew or had notice of the likelihood of an "official proceeding"; (2) knowingly engaged in intimidation, threats, corrupt persuasion, or misleading conduct; (3) with the intent to cause or induce any person to alter or destroy documents relevant to the proceeding.

The burden of establishing the existence of an "official proceeding" and the defendant's knowledge of it is greatly reduced under § 1512. The statute specifically defines "official proceeding" as a proceeding before a judge, a grand jury, Congress, or a federal agency.[49] The statute also provides that an official proceeding need not have commenced or even been scheduled at the time of the offense.[50] Therefore, § 1512(b) punishes obstructive efforts occurring during the investigative phase.[51]

However, courts differ as to how much notice the defendant must have about a possible proceeding. Some courts have held that § 1512 does not apply in cases where no proceeding is "ongoing or scheduled for the future."[52] However, other courts-especially those considering § 1512 in the context of evidence destruction, have found that § 1512 applies when there is "direct evidence that [the defendant] in fact expected [a proceeding] in the foreseeable future, and that his intent was to make the items unavailable for us in such proceeding or proceedings."[53]

III. The Sarbanes-Oxley Reforms

A. Background

The Sarbanes-Oxley Act was enacted largely in response to the Enron crisis, the largest bankruptcy in U.S. history until the WorldCom collapse of 2002.[54] Enron's underhanded accounting practices, which had made the company appear successful by siphoning losses to so-called "special purpose entities,"[55] raised awareness that current accounting and enforcement practices were incapable of detecting large-scale misstatements.[56]

The goals of the Sarbanes-Oxley Act are to remedy these problems and adopt better provisions to punish corporate wrongdoing, while protecting the innocent.[57] These goals are accomplished by a number of means, including improved financial reporting requirements,[58] whistleblower provisions,[59] and revised provisions governing the destruction of documents.[60]

B. Concern for Document Destruction

1. The Arthur Andersen Case

The evidentiary problems of document destruction became a major concern for Congress after Arthur Andersen LLP, the company which had audited Enron's finances, was indicted under § 1512(b)(2) for destroying its Enron auditing papers.[61] The indictment claimed that Arthur Andersen had advised Enron to make use of "special purpose entities" and that, with the ensuing collapse of Enron, Arthur Andersen began the "wholesale destruction of documents." The government claimed that this included several urgent meetings during which employees' shredded documents in anticipation of a subpoena, which was served in November, 2001.[62] After the subpoena was served, Arthur Andersen alerted its audit teams that they must stop shredding because they had been officially served with documents.[63]

Additionally, the government produced evidence that Nancy Temple, in-house counsel for Arthur Andersen, became aware of Enron's demise on October 9, 2001.[64] She was aware that the firm had been previously fined by the SEC for prior audits and, as part of an agreement with the agency, it would be held in contempt if it violated further SEC laws.[65] Accordingly, on October 12, Temple emailed Michael Odom, who was also in-house counsel. "It might be useful to consider reminding the engagement team of our document retention policy. It would be helpful to make sure that we have complied with the policy. Let me know if you have any questions."[66] Temple attached Arthur Andersen's document retention policy to the email,[67] which provided for the destruction of documents as well as for the suspension of routine destruction because of "regulatory agency investigations . . ., or other legal action in connection with which the files would be necessary or useful. In such cases the materials in our files cannot be altered or deleted. Other reasons may include litigation involving [the company] or the client . . . ."[68]

Odom forwarded the email to David Duncan, the Houston partner in charge of the Enron audits.[69] The email contained only one line: "more help."[70]On October 16, Duncan learned that the SEC was conducting an informal investigation into Enron's transactions.[71] On October 22 he met with other members of his team, including Rick Causey, Enron's chief financial officer, to discuss a course of action.[72] The next day, Duncan organized an extensive effort to shred documents and delete emails.[73]

The jury found Andersen guilty of the charges, concluding that under § 1512(b)(2), Temple was the "corrupt persuader."[74] However, the jurors did not find the email of October 12 to be the key inducement. Instead, they focused on another email mentioned by the government in which Temple asked Duncan to alter a memo he had written for his files about the way Enron characterized its losses in public statements.[75] The Fifth Circuit has affirmed Andersen's conviction.[76]

2. Congressional Response to Andersen

Both Enron and Arthur Andersen illustrated an important point: not only were large companies able to work around accounting rules, but they were also able to circumvent the obstruction of justice laws and to clean up the trail of evidence. In a committee report, Senator Patrick Leahy noted that the obstruction statutes created a sizeable loophole for corporations.[77] He explained that the ambiguous provisions of § 1503 have been interpreted narrowly so that the government must prove that the destruction of documents was "closely tied" to a judicial proceeding.[78] Moreover, he explained that the Andersen case underlined the great anomaly of § 1512(b): in relying on the "corrupt inducement" statute, the government had "to proceed under the legal fiction that the defendants are being prosecuted for telling other people to shred documents, not simply for destroying evidence themselves."[79] He went on to explain that although the Andersen prosecution was proceeding, the law technically allowed a person to destroy as many documents as he wanted, as long as no judicial proceeding had commenced and he was acting alone.[80] The committee sought to close this loophole by enacting a statute that would "apply broadly to any acts to destroy or fabricate physical evidence so long as they are done with the intent to obstruct, impede or influence the investigation or proper administration of any matter . . . or the contemplation of such a matter or investigation."[81]

C. The Statutes

1. Section 1519

The Sarbanes-Oxley Act broadens the power of the federal obstruction statutes and modifies the prior criminal obstruction statutes in several important ways. It creates § 1519, which imposes criminal penalties for the destruction of documents related to actual or "contemplated" investigations conducted by a federal department or agency:

Whoever knowingly . . . destroys . . . any record, document, or tangible object with the intent to impede, obstruct, or influence the investigation or proper administration of any matter within the jurisdiction of any department or agency of the United States . . . or in relation to or contemplation of any such matter or case, shall be fined under this title, imprisoned not more than 20 years, or both.[82]

It is notable that § 1519 makes several significant departures from § 1503. First, it adopts a "knowingly" standard in place of "corruptly," which broadens the scope of the act to include all purposeful destruction of documents, regardless of whether the destruction is legitimate or not. Second, the section expands the intent requirement. Unlike § 1503, which after Aguilar, requires that the defendant intentionally obstruct an on-going judicial proceeding, § 1519 punishes the person who intentionally destroys documents related to a "contemplated" investigation or administration.

2. Section 1512(c)

The Act also amends § 1512. Where § 1512 formerly only applied to "corrupt inducement" in which the defendant coerced another party to destroy documents, the revised section punishes any person who "corruptly" destroys documents:

Whoever corruptly . . . destroys . . . conceals, covers up, falsifies, or makes a false entry in any record, document, or other object, or attempts to do so, with the intent to impair the object's integrity or availability for use in an official proceeding . . . or attempts to do so, shall be fined under this title or imprisoned not more than ten years, or both.[83]

However, note that § 1512(c) still retains the limitations of the existing sections. For one, it still requires that the documents be destroyed with "corrupt" intent-not merely "knowing" as § 1519 provides. Moreover, it still only applies in the context of "official proceedings," which means that the ongoing circuit splits regarding the precise scope of the law will likely continue to affect this statute.

IV. Evaluating the Sarbanes-Oxley Statutes

A. Advantages of § 1519

Section 1519 is likely to be prosecutors' preferred obstruction statute in document destruction cases because it has several advantages over other statutes. Perhaps most important is the fact that it comes into effect as soon as law enforcement "contemplates" an investigation into a matter." We have already seen that the "judicial proceeding" or "official proceeding" elements of § 1503 and § 1512 severely limit their scope to, at best, destruction occurring during the investigative phase of an inquiry. The language of § 1519 goes well beyond this limit to punish destruction of documents occurring before authorities have officially commenced an investigation.

Also, unlike its § 1503 and § 1512 counterparts, § 1519 only requires that the documents be destroyed "knowingly." The "corrupt" destruction of records required by § 1503 and § 1512 is not easy to prove if the documents are destroyed as part of a routine practice, for example, pursuant to a reasonable document retention/destruction program. In addition, as we have seen, the "corrupt" standard further burdens prosecution because its meaning changes from jurisdiction to jurisdiction; there is no uniform definition of what "corrupt" destruction requires. Adopting a "knowingly" standard, § 1519 resolves these problems for prosecutors. "Knowingly" is a widely recognized state of mind that courts will likely apply with a higher degree of uniformity. "Knowingly" is generally understood to mean that the actor is "aware that it is practically certain that his conduct will cause . . . a result."[84] In terms of § 1519, the actor need only be aware that his conduct will result in the destruction of documents-regardless of whether the destruction is ultimately corrupt or not. Such a standard greatly increases the scope of the statute to include routine acts of document destruction carried out pursuant to a routine document retention/destruction program.

Finally, the "knowing" standard may also assure that § 1519 supercedes the § 1512(b)(2) "corrupt inducement" statute. Because "knowingly" requires only that the actor be "practically certain" that his actions will result in the destruction of documents, there may no longer be a need to expressly prohibit corrupt inducement. Therefore, in scenarios where, for example, an officer of a company commands a subordinate to destroy documents, or even gives a strong hint that it should be done, the officer still "knowingly" destroys the records even if he does not place them in the shredder himself.

B. Problems with § 1519

1. Congressional Concerns

Although Senator Patrick Leahy's report urges a broad interpretation of § 1519 to end "technical distinctions" that have created loopholes,[85] others, including Senator Orrin Hatch, stressed their concern that § 1519, though closing an important gap in federal obstruction law could nonetheless be "interpreted more broadly than we intend."[86] Hatch suggested that the section should not cover the destruction of evidence in the "ordinary course of business, even where the individual may have reason to believe that the documents may tangentially relate to some future matter within the conceivable jurisdiction of an arm of the federal bureaucracy."[87] Although no case has yet been decided under § 1519, recent fact patterns involving document destruction suggest that a broad construction of this section could harm innocent companies.

2. Recent Cases

a. The Quattrone Case. The case of Frank P. Quattrone, former banker at Credit Suisse First Boston (CSFB) illustrates the possible complexities of document destruction cases. Quattrone was charged in April 2003 with obstruction of justice under sections 1503, 1505, and 1512.[88] Beginning in 2000, the U.S. Attorney in Manhattan issued subpoenas to CSFB requiring documents relating to allegations that the bank had been allocating shares in hot initial public offerings (IPOs) to investors who would pay inflated commissions.[89] CSFB's document policy requires the preservation of key documents relating to IPOs, but all ancillary documents are to be destroyed, unless a civil or criminal subpoena is issued, in which case the destruction process is halted.[90]

The subpoena related to matters handled by CSFB's equity division, which was unrelated to Quattrone's Global Technology Group.[91] However, according to prosecutors, on December 4, 2000, Richard Char, a member of Quattrone's staff, suggested that a memo should be sent to employees of the Global Technology Group reminding them of the firm's document policy and the possibility of damaging civil litigation based on the accusations.[92] The email concluded, "Today it's administrative housekeeping-in January it could be improper destruction of evidence."[93] Quattrone responded with a one-line email: "You shouldn't make jokes like that on email."[94] On December 5, Quattrone sent Char's email to the rest of his staff and added the line, "Having been a key witness in a securities litigation case in South Texas, I strongly recommend you follow this advice."[95] On December 7, the general counsel of CSFB also sent an email to staff notifying them, according to the document policy, of a freeze on all routine document destruction.[96]

Two years later, federal prosecutors charged Quattrone with obstruction, claiming that the email exchanges were evidence of an effort to encourage document destruction.[97] The jury failed to agree on a verdict, and the case resulted in mistrial.[98] On retrial, a jury convicted Quattrone,[99] and he was sentenced to eighteen months in prison.[100]

b. The Trauger Case. The first case to be tried under the Sarbanes-Oxley obstruction statute, § 1519, was filed against Ernst & Young (E&Y) partner, Thomas Trauger.[101] The complaint, filed on September 4, 2003, claimed that Trauger altered key documents relating E&Y's audit of NexCard, a credit card company.[102] The government contends that beginning in 2000, NextCard began classifying some of its credit losses as fraud losses, which triggered an investigation by the Office of the Comptroller of the Currency (OCC) in August 2001.[103] The government claims that Trauger was aware of this investigation and even helped draft the company's written response.[104] Allegedly, in response to OCC's request for certain working papers of NextCard's financial statements, Trauger and a colleague, Oliver Flanagan, altered several of the documents and saved them in the E&Y database as if they were the originals.[105] As part of the effort to alter the documents, Trauger and Flanagan allegedly asked a former auditor to re-write her handwritten notes on some of the documents.[106] Trauger then asked Flanagan to destroy all documents and emails relating to their alterations and to destroy any copies of the original papers.[107]

3. Problems for Companies

These cases illustrate several of the issues involved in modern destruction practices. In cases like Trauger, if the allegations are proven to be true, they are likely to result in conviction under § 1519. The facts indicate that Trauger "knowingly" altered and destroyed several of the working papers relating to the audit. The fact that he provided the OCC with the altered copies indicates that his intent was in fact to impede the agency's investigation. However, the Quattrone case and the Andersen case discussed earlier both raise some of the complex issues involved in document destruction cases. These concerns reveal that the broad scope of § 1519 may in fact harm innocent businesses.

a. Companies produce a staggering number of "documents." In the era of electronic information, companies are able to produce many more "documents" than they were able to in the print era. This is largely due to the prevalence of email communication, which documents a variety of personal, business, formal, and informal exchanges that were not traceable twenty years ago. Email communication has largely replaced telephone conversations and letters as the day-to-day means of business communication. It has been estimated that at least two-thirds of American business employees use email, which means that approximately 130 million employees send over 2.8 billion emails a day.[108] In 2000, this added up to about 1.4 trillion email messages sent from U.S. businesses.[109] Email creates unique problems in document destruction cases because, unlike informal telephone conversations or office meetings, each email sent creates a "document" stored on the company databases.

In most companies, each time an employee sends or receives an email on the corporate computer system, the system will generate a file, which is saved on the company's computer system and possibly on the employee's personal computer.[110] Therefore, even when an employee deletes an email from her personal computer, the file is stored in the company's central system.[111] This stored file is considered a discoverable "document" for use in civil trials[112] and, under the Federal Rules of Evidence, is afforded the same evidentiary status as paper documents in both civil and criminal cases.[113]

b. Many companies destroy documents as part of a formal company document policy. Because companies accumulate vast amounts of documentation, including email files, many companies have policies for routinely destroying non-essential or outdated files. These policies are tailored to the specific needs of the company, providing guidelines for what types of documents should be held and procedures for storing and destroying documents.[114] Such policies are drafted by the company's attorney and usually require companies to retain certain administrative, accounting, financial, tax, pension, payroll, legal, insurance, contracts, and correspondences.[115] Some such documents may be critical to the company; others are required by law.[116] Generally the policy will require that the company hold these documents and destroy all outdated or non-essential documents. Document policies generally require the suspension of routine destruction when the company becomes aware of or involved in an investigation or litigation.[117]

Most document retention policies delete emails on a thirty-day cycle, unless litigation is pending. Considering the large volume of emails generated by companies each day, most delete those emails that are not saved by their employees.[118] This is not only because the cost of email storage is considerable,[119] but the cost of producing discoverable documents can be debilitating. In a recent sex discrimination case, UBS Warburg estimated it would cost approximately $175,000 exclusive of attorneys fees to retrieve archived emails.[120] In a more extreme example, a magistrate judge in the Eastern District of Louisiana confirmed that one responding corporation would have to pay $6.2 million in order to retrieve and review the emails of 650 employees.[121] Companies avoid these costs by implementing a routine process of destroying email files on a thirty-day cycle, unless impeding litigations requires suspending the process.[122]

c. Intent in obstruction cases is generally inferred from limited direct evidence. Both the Andersen and Quattrone cases have an interesting similarity: in both cases, the government's key evidence of intent centered on a single email sent by the defendant urging members of the company to comply with the document retention policy. In the Andersen case, the only direct evidence of intent was provided by a single email sent by Nancy Temple urging members of the firm to implement the firm's document policy.[123] Similarly, the Quattrone case depends on the intent inferred from an email Frank P.Quattrone sent to his staff urging them to implement the firm's document policy.[124] In both cases, the brief emails, combined with circumstantial evidence can be understood to mean one of two things; either (a) the emails were a sincere reminder of the firm's document retention policy, or (b) the emails were a coded sign to begin active document destruction. On one hand, it seems logical that a partner in an accounting firm would want to remind his staff of the document policy on the eve of an investigation in an effort to preserve documents. However, it seems equally logical that a partner would also want to remind his staff of the technicalities of the document policy and urge his staff to destroy documents before it technically became illegal.

IV. Defining a "Contemplated" Investigation

Section § 1519's protection of documents relevant to a "contemplated" investigation can be broadly interpreted to punish almost any form of document destruction. However, in light of business concerns and the general inability to preserve every document, the section should be understood in a more limited sense.

A. Aguilar Revisited: The Need for "Nexus"

The Supreme Court's holding in United States v. Aguilar suggests that the intent element in obstruction statutes should be understood in terms of foreseeability. Although Senator Leahy specifically expressed his hope that § 1519 would circumvent the holding in Aguilar, that case should not be quickly abandoned. [125] Leahy in fact, did not take particular issue with the "nexus" requirement discussed in Aguilar, but only its narrow application.[126] In fact, the "nexus" rule of Aguilar provides some general insight as to how a specific intent statute like § 1519 should be applied, indicating that the foreseeability of a probable investigation should be an element of the offense.

1. Analyzing § 1519 as a "Specific Intent" Statute

Section 1519 is essentially a "specific intent" statute because it punishes certain actions taken to inflict wrongful consequences. [127] In other words, the defendant must first commit an initial act that may or may not itself be illegal, in the hopes that such an act will result in certain illegal consequences.[128] In order to establish a § 1519 offense, the government must prove that the defendant (1) knowingly destroyed documents (2) with the intent to obstruct a contemplated investigation. The first of these requirements is straightforward: the defendant must know or be aware that his conduct will result in the destruction of documents. However, the second element is more difficult to establish because the statute does not define how to determine "intent."

Generally, intent is understood in two ways. Under a traditional specific intent approach, the second element of the statute is proven if the defendant consciously desires to obstruct an investigation or if he knows that by destroying documents he is practically certain to obstruct an investigation.[129] Similarly, under the more modern view adopted by the Model Penal Code, the second element is established if the defendant destroys documents when "he is aware of the existence of such circumstances or he believes or hopes that they exist."[130] Such intent is often proved through indirect circumstantial evidence.

2. Aguilar's Nexus Requirement and Foreseeability

The "nexus" approach used by the Supreme Court in Aguilar seems to summarize both of these definitions and may be useful in understanding § 1519. Aguilar's "nexus" rule provides that the "intent" element is met if the document destruction has a "relationship in time, causation, or logic" with an obstructed proceeding or investigation. [131] In other words, obstruction is the "natural and probable effect" of the document destruction.[132] Aguilar's "natural and probable effect" test is derived from the old criminal law maxim which states that a person is "presumed to intend the natural and probable consequences of his acts."[133] Such a standard is equivalent to the foreseeability tests used to gauge proximate causation in negligence cases because it defines intent in terms of what a reasonable person would have foreseen as probable.[134] Although criminal "intent" is thought to articulate a higher standard than criminal negligence, specific intent crimes present a unique dilemma because intent to obstruct justice, for example, often cannot be proven with direct evidence. In most cases such intent is established using circumstantial evidence. Therefore, in cases under § 1519, the probability that the destruction of documents will foreseeably obstruct an investigation is a proper standard.

B. Developing a Standard of Foreseeability:

1. The Federal Common Law Doctrine of Spoliation

In trying to define a "nexus" between the destruction of documents and intent to obstruct an investigation, the natural place to look is the common law doctrine of spoliation, which imposes sanctions on parties who destroy discoverable evidence prior to a proceeding. To be sure, the common law doctrine of spoliation is significantly different from criminal obstruction of justice, but they share several important points.

Federal common law provides sanctions against parties in civil litigation who despoil evidence. Spoliation is defined as a failure to preserve all documents relevant to pending or potential litigation:

While a litigant is under no duty to keep or retain every document in its possession once a complaint is filed, it is under a duty to preserve what it knows, or reasonably should know, is relevant in the action, is reasonably calculated to lead to discovery of admissible evidence, is reasonably likely to be requested during discovery, and/or is the subject of a pending discovery request.[135]

The destruction of evidence must be in "bad faith," meaning that the party destroyed the evidence even after it was aware that the evidence was potentially relevant.[136] Therefore a party has acted in bad faith when it destroys relevant evidence after being served with a civil complaint.[137] A company also acts in bad faith by destroying documents that would be important to potential litigation.[138] Similarly, destroying documents for the sole purpose of limiting prospective liability is also considered bad faith.[139]

Failure to preserve documents that a party knows or should have known to be relevant to pending or potential litigation is considered spoliation of evidence and the court may impose sanctions at its discretion.[140] Because sanctions for spoliation are imposed through the court's inherent power to manage its own affairs, the court may sanction a party even when no discovery order or rule of procedure has been violated.[141]

2. Special Rules for Document Retention Programs

Destruction of otherwise discoverable documents pursuant to a routine document destruction program may be a defense against a claim of spoliation. The Eight Circuit in Lewy v. Remington Arms Co.[142] developed three factors to be considered in assessing whether the execution of a routine document destruction program was in good faith and therefore not sanctionable:

1) Is the record retention policy reasonable? The reasonableness of a document destruction program is generally determined by the type of document or record involved.[143] Therefore, the blind destruction of all documents without screening them for their potential relevance to litigation is not a defense to spoliation.[144] Moreover, mere adherence to the minimal retention requirements of an industry is not enough to prove reasonableness.[145]

2) Have lawsuits been filed or are lawsuits commonly filed? If so, what is their frequency and magnitude?[146]

3) Was the document destruction policy instituted in bad faith? Even a document destruction program, no matter how reasonable, is not a defense to spoliation if it is created in bad faith. A document retention policy with the primary purpose of limiting liability is in bad faith.[147] Similarly, a destruction policy that is reasonable in most circumstances is used in bad faith if a party fails to suspend it when documents known to be material are scheduled to be destroyed.[148]

3. Using Spoliation Elements to Determine Improper Intent under § 1519

Interpreting "contemplated" investigations to include those reasonably foreseeable by the defendant is a feasible standard because it does not penalize companies who did not know or could not have known that the documents destroyed were relevant to a potential investigation. Drawing from Eight Circuit's holding in Lewy, the court should consider the following:

(a) Is the document retention policy reasonable?

(b) Did the company have notice of the investigation and properly suspend the destruction of potentially relevant documents?

(c) Could the company foresee an investigation because of past investigations of the company or current investigations of related companies?

Such a foreseeability test has several advantages. First, it depends on a rigid definition of "contemplated," that is, it does not define a point up to which a company may destroy documents without penalty. Second, it does define a relatively clear point at which companies should be wary of destroying certain documents. Under this standard, a business that has no indication of a potential investigation may still execute its routine document destruction policies without the fear of violating § 1519 merely because a law enforcement agency somewhere may be "contemplating" an investigation. Third, it imposes a reasonableness standard that makes the most of circumstantial evidence. If a company can prove it had no inkling that it would be investigated, then it will not be held liable for documents it destroyed as part of a routine program. However, if the company suspected that it might come under investigation, it will be responsible for any actions that a reasonable person would understand to be an effort to avoid getting caught.

4. Application of a Foreseeability Standard

Applying a foreseeability test similar to that used in federal spoliation cases would probably not change the outcomes in the Andersen, Quattrone, and Trauger cases. In both Quattrone and Trauger, the individuals destroyed or altered documents after receiving a subpoena. The destruction was therefore clearly in light of a "contemplated" investigation, as defined by the foreseeability test.

The foreseeability test is more pertinent in the Andersen case, where the company began destroying documents before being served with a subpoena. Had § 1519 applied to that case, there would be some doubt as to whether an investigation was "contemplated." However, under the foreseeability test, an investigation would be contemplated, even if Andersen had not been expressly notified, because it had been investigated by the SEC in the past under similar circumstances, and it was aware that the government was looking into the sudden demise of one of its clients, Enron. Furthermore, under the foreseeability test Andersen would not be justified in relying on its document retention policy because any policy that permits the destruction of documents when the company is aware of a possible investigation is not "reasonable."

The foreseeability test is also pertinent to a common hypothetical: a company with a sound document retention policy deletes incidental email communications from the company databases without any indication that they would ever be subpoenaed in an investigation. Under a broad reading of § 1519's "contemplated" investigation requirement, a company would be required to retain indefinitely all email communications because it could not be sure when or if an investigation involving the emails may be "contemplated." Under the foreseeability test, such routine destruction of electronic documents would not amount to obstruction when the company (1) uses a reasonable document retention program, (2) is not aware that the subject matter of the emails or the sender or recipient have been the subject of a related investigation in the past, and (3) is not aware of any other companies currently under investigation for related matters. Under such circumstances, a company can be assured that the documents are not the subject of a "contemplated" investigation and can spare itself the cost of retaining all such documents.

V. Conclusion

As we have seen, both the Andersen and Quattrone cases are good illustrations of the complexities of determining the intent of parties destroying documents. In both cases, the defendant sent an email to key members of the company urging them to adhere to the document retention policy. This evidence was used to establish an intent to obstruct a grand jury investigation on the presumption that the emails were in fact coded commands alerting the staff to begin destroying documents. This conclusion was supported by the fact that the staff did in fact respond by destroying documents.

Although these cases may present strong circumstantial evidence of an intent to obstruct a proceeding, subsequent cases may not be decided so easily. Section 1519's standard, if interpreted broadly, may subject companies to harsh criminal sanctions for innocent adherence to a reasonable document retention policy. Therefore, in interpreting § 1519, courts should consider the long-standing and carefully developed doctrines of spoliation to determine when document destruction obstruct justice and when it does not.