The Email String and Assertion of Attorney-Client Privilege
Introduction
In recent years Federal Rule of Civil Procedure 26(b)(5) has proven to be complex, particularly when considering one of the most fundamental privileges in the legal field – attorney-client privilege – and one of the most basic components of the ever-expanding world of technology – the email. The legal field has been affected by the expansion of technology; while new developments have brought tremendous advantages, they have also brought about change and given rise to the problem of interpreting and implementing old world rules within new boundaries. The court system’s struggle with the imposition of attorney-client privilege within email strings is perhaps the best illustration of how ‘simple’ concepts are no longer so simple after all.
Rule 26(b)(5)
Under Federal Rule of Civil Procedure 26(b)(5), a party may withhold information that would otherwise be discoverable by asserting a privilege over such information (1). In withholding this information, the party must expressly make the claim and describe the undisclosed information in a manner that will enable opposing parties to assess the proper assertion of the privilege at hand, while not revealing the privileged information. This information is contained in a privilege log that is then turned over to the opposing party. (2)
Opposing views
Like many other professionals, lawyers send and receive hundreds of emails to and from colleagues, clients, and potential clients every day. These emails may consist of a single message or multiple messages from the same person, forwarded messages, or a message to which the lawyer is replying. These messages may deal with legal advice, mundane business matters, or a mixture of the two. As a result, the information contained therein may be privileged, non-privileged, or both. Thus, when entering litigation, these ‘simple’ messages that can be sent at the click of a button transform into a mountain of questions that the courts have been trying to resolve, but have not yet been able to settle. While it has been debated whether widespread dissemination of an email or the lack of a further dissemination limitation clause results in a waiver of the right to claim that information is privileged, perhaps the most basic question in this area is whether an email string should be considered to be a single communication, similar to a conversation and thus logged as such, or whether each individual email should be considered to be a separate communication in regard to which privilege must be evaluated.
Finding it misleading to break down a series of emails into individual messages against which privilege must be assessed, the US District Court for the Northern District of California took a strong stance in favour of the single communication theory in United States v Chevron Texaco Corporation.(3) The court reasoned that each email/communication comprised the sender’s final message and all preceding emails that were attached to this message. Thus, if the final email in an email string was privileged, then all prior emails attached to the final email were also privileged. If privilege was waived with regard to the final email, the ability to assert a privilege over all prior attachments in the string was waived. (4)
Analogising email strings to a conversation, the defendant in In re Universal Service Fund Telephone Billing Practices Litigation shed more light on the reasoning behind the single communication theory. (5) Pointing out that, just as each individual’s lines in a conversation would not be separated and assessed independently with regard to privileged information, the defendant argued that neither should an email string, no matter how many individual emails are involved.(6) Despite the defendant’s arguments, the judge for the Northern District of Kansas could not be swayed in this matter, ultimately stating that the inclusion of individual emails in a privilege log was “essential”.(7) Finding the prior rulings on which the defendant built its argument to be unpersuasive, the court preferred to err on the side of caution rather than to support the defendant’s “risky yet reasonable” arguments. The court rejected the defendant’s analogy for three reasons:
• The court found that the prior opinions were construed too broadly and that they lacked the in-depth analysis that would be required to be persuasive.
• It is believed that the well-established case law concerning privilege logs and privileges provided better guidance on this topic.
• If the defendant’s argument was to be upheld, such a rule would result in stealth claims of privilege which, intrinsically, could never be the subject of meaningful scrutiny by a judge or a meaningful challenge by opposing counsel. (8)
While it may be easy to understand how two different courts from different states can have varying perceptions when it comes to an area of unchartered territory, perhaps the clearest illustration of opposing views and varying concerns comes from the conflicting opinions of the magistrate judge (9) and the district judge (10) for the Northern District of Illinois in Muro v Target Corporation. There, on reviewing the facts, the magistrate judge held that where a defendant’s privilege log merely identified the most recent email in a particular email chain and failed to mention information regarding prior emails which were forwarded on as a part of the chain, the requirements of Rule 26(b)(5) had not been met.(11) Stating that the “mere assertion of a privilege is not enough to establish the privilege”, the judge focused on the withholding party’s burden to describe the communications and nature of the documents being withheld in a way that did not reveal the privileged information but enabled the opposing parties to evaluate the applicability of the privilege asserted.(12) In failing to include information regarding the prior communications comprising the email chain, individuals who received earlier messages but who might not have been sent the final message were not included on the defendant’s privilege log, thus making it impossible for the opposing parties to assess the privilege asserted when considering each individual email in the chain. The court, therefore, concluded that each communication not produced must be described, whether it consisted of an entire email string or simply one part of the string.(13)
Reading Rule 26(b)(5) with more emphasis on the requirement of sufficient information for an opposing party to evaluate the applicability of privilege without revealing privilege information, the district judge came to the opposite conclusion.(14) Although the concerns raised by the magistrate judge were warranted, the district court judge highlighted the intrinsic risk of requiring the defendant to log each individual email as a separate communication on the privilege log. There, the district court judge recognized that in making a party include already produced documents in its privilege log, the party may be forced to waive attorney-client privilege where the opposing party would then be able to gather enough information from the log and the previously produced materials to discover the information forwarded to the opposing party’s attorney.(15) Relying on Upjohn Company v United States, (16) which provides that non-privileged information, when communicated to an attorney, may be privileged even if the primary information remains unprotected, the district court judge likened the prior emails in an email string to documents or prior conversations that are later quoted by an individual to his or her attorney or letters with attachments.(17) In addition, the district court judge went as far as to say that compelling a party to include previously produced documents in its privilege log would be “confusing”. (18)
Seemingly in agreement with the district judge’s application of Rule 26(b)(5) in Muro, in Rhoads Industries, Inc v Building Materials Corp of America the court elaborated on the underlying components of an email string in a more thorough explanation of Muro, before evading the establishment of a “broad, black-letter rule”.(19) Highlighting the complexity of the email string, the court noted that an email chain that consisted of four email messages would be comprised of four distinct versions of the email:
• the final complete string, which would consist of the most recent version of the email, as well as the previous three emails;
• the string consisting of the third, second, and first email;
• the string consisting of just the second and first email; and
• the original email from which the string eventually grew.(20)
The court recognized that in Muro the district judge had found that each prior version of an email string was considered to be a unique document, and thus must be produced or, if privileged, included on a privilege log. However, despite his understanding that the district judge in Muro disagreed with the view that Rule 26(b)(5) required separate itemization of each individual email comprising an email string, the judge in Rhoads seemingly took a step back from this approach in stating that if a document was not produced, each email must be logged individually in order to claim the privilege.(21) With that, as soon as Muro’s seemingly balanced approach came into existence, its illusion of a resolution was again thrown off kilter by the Eastern District of Pennsylvania.
Following suit of those that had come before, the courts continue to flounder, flipflopping back and forth when confronted with the complex dilemma that the email string presents when included in a privilege log. In the most recent ruling on this issue, in Benefitvision Inc v Gentiva Health Services, Inc a magistrate judge for the US District Court for the Eastern District of New York proved that the issue may be no closer to being resolved once and for all. (22) Finding in favour of a plaintiff regarding its motion to compel a defendant to produce email communications that were being withheld based on assertions of privilege, the magistrate stated that the defendant’s proposition of producing a supplemental privilege log regarding intermediary emails that were withheld, but not privileged, was unacceptable.(23) Rather, the judge required the defendant to produce the intermediary emails comprising email chains if such emails were not the subject of attorney-client privilege or work product, yet made no reference to prior court decisions on the issue.(24) Although no precedent had been set regarding this matter, in choosing to disregard the opinions of those that came before, the judge for the Eastern District of New York provided the perfect example of how the difficult issue of email chains in privilege logs is being perpetuated.
Practical implications
As previously stated, while email has provided legal professionals with numerous advantages on a daily basis, it has also generated a neverending list of questions that have gone unanswered. While courts struggle to define the term ‘communication’ in an effort to enforce properly assertions of privileges under Federal Rule of Civil Procedure 26(b)(5), their battle against the email string has often resulted in wasted time, money, and effort as attorneys are forced to draft privilege logs, spending tremendous amounts of time going through each document and providing its description only to find out that, in a best-case scenario, the resulting product must be changed in order to meet a specific court’s obscure requirements. In an even worse instance, a less forgiving court could find that the party had failed to assert privilege properly where the description of email strings in a privilege log are found to be unsatisfactory, and thus the party is found to have waived these privileges. An accidental waiver of a litigant’s fundamental privileges could then result in damaging the participating attorney’s reputation as well as his or her credibility.
Moreover, while many negative implications of the email string’s presence within privilege logs are apparent, others require more thought. For example, when briefly examining attorney-client privilege, it is possible to gain a simplified understanding of the insurmountable task at hand which courts currently face. The ease with which an alleged privileged message can be disbursed to an infinite number of people at the click of a button causes courts to question whether each individual present on an email string was a person directly concerned with the communication, and thus was a privileged party to the same. The courts must scrutinize the description of these individuals, including their job titles and their possible roles as agents to privileged parties, when making these decisions. Furthermore, the courts must determine whether such widely disseminated communications were made with an actual expectation of continued confidentiality. In essence, courts are forced to dissect every element of the attorney-client privilege in a new and painstaking manner, despite the fact that this process used to be something more like second nature. Having knowledge of the courts’ newly increased standards of review, attorneys must act with greater care to inform their clients of the risks of waiver and how waiver may occur. As a result, many attorneys are left wondering whether email is a blessing or a curse.
Looking forward
Although courts still struggle with the proper inclusion of an email string in a privilege log, until this conflict is resolved there are steps that can be taken to safeguard clients’ privileges while also saving time, money, and energy.
A first, although unlikely, proposition would be to avoid using email. While this option is obviously not the most convenient solution, claims were filed and cases were resolved before the evolution of email. The email could also be used simply to set up meetings or conference calls during which confidential information could be discussed. In addition, rather than avoiding the use of email strings as a whole, lawyers could simply keep in mind that it is their duty to keep their clients informed of their privileges and the proper steps that need to be taken in order to protect those privileges. An attorney should begin by explaining to a client the dangers of email strings and should email only small groups of privileged parties. These emails should include a definite statement of confidentiality which should be explained to clients ahead of time so as to ensure that emails are not being forwarded to non-privileged parties after being launched into cyberspace. As a result of these steps, privileged email strings and their dissemination could be reduced.
A second, and possibly more advantageous, way of tackling a privilege log in light of the existence of email strings is to strike an agreement among all parties to a suit regarding the content and format of privilege logs to be produced in a particular matter. While litigants could agree to treat email strings as a single communication in one case, they could agree to treat them as separate communications in another, but also to identify each communication by already existing metadata that could simply be exported to a privilege log rather than manually entering a description of the same.(25) As a result, each litigant would know what to expect and what he or she must create, reducing the wasted time, cost, and energy of drafting a privilege log that must be revised. According to the Notes of the Advisory Committee on the 1993 amendments to Rule 26(b)(5), Federal Rule 26(b)(5) makes no attempt to define the specific information that must be provided by a party in a particular case when a party asserts a privilege, leaving lines of opposing party communication open for negotiation.(26) However, in creating these guidelines, the parties to a matter must continue to abide by Rule 26(b)(5).
Endnotes
*Co-authored by Courtney Pasquariello.
(1) Fed R Civ P 26(b)(5).
(2) Id.
(3) 241 F Supp 2d 1065, 1074 n6 (ND Cal 2002).
(4) Id.
(5) 232 FRD 669, 672 (D Kan 2005).
(6) Id.
(7) Id at 674.
(8) Id at 672¬-674.
(9) Muro v Target Corp., 243 FRD 301, (ND Ill 2007).
(10) Muro v Target Corp, 250 FRD 350, (ND Ill 2007).
(11) Muro v Target Corp, 243 FRD 301, 307 (ND Ill 2007).
(12) Id at 305.
(13) Id at 307.
(14) Muro v Target Corp, 250 FRD 350, 363 (ND Ill 2007).
(15) Id.
(16) 449 US 383 (1981).
(17) Muro, 250 FRD at 363.
(18) Id.
(19) 254 FRD 238, 241 n4, 241.
(20) Id at 241 n4.
(21) Id at 240-241.
(22) 2011 US Dist LEXIS 71510.
(23) Id at 6-7.
(24) Id.
(25) Elizabeth T Timkovich and Meghan A ODonnell, “Alternative Privilege Log Techniques in an E-discovery World”, 2 Commercial and Business Litigation 9, (Winter 2009).
(26) Fed R Civ P 26(b)(5).