It’s one of the first steps. Upon learning of an event—an accident, a breach—that makes one contemplate litigation, a party’s lawyer or the lawyer’s investigator interviews nonparties with relevant knowledge. These interviews produce facts, of course, but also could contain, directly or indirectly, the lawyer’s mental impressions or legal strategies about important things such as the strength of a potential claim, liability exposure, or damages.
The interview results come in different forms. Some lawyers audio- or video-tape a question-and-answer session with the potential witness; others translate the witness’s comments into a sworn declaration; while still others memorialize the interview in a lawyer-prepared memorandum. And when litigation erupts and the adversary moves to compel the recorded interview, the declaration, or the memorandum, the party’s lawyer chuckles while asserting the work-product doctrine.
The chuckling stops, however, when the lawyer sees how courts have ruled in a seemingly complex maze of fact work-product and opinion work-product. Isn’t a recorded interview simply a recitation of facts devoid of an attorney’s mental impressions? Same goes for a signed witness statement, right? “Facts aren’t privileged” is a familiar if inapposite refrain. “Just redact the lawyer’s comments and produce the facts” is another.
Maybe one court has answered these questions and solved the nationwide inconsistency. Let’s discuss.
Of common-law origins, the work-product doctrine protects from discovery a lawyer’s mental impressions about the legal dispute at issue. The idea is that “[d]iscovery was hardly intended to enable a learned profession to perform its functions … on wits borrowed from the adversary.” Hickman v. Taylor, 329 U.S. 495 (1947) (Jackson, J., concurring).
In 1970, the Supreme Court approved a work-product amendment to the civil-procedure rules. Residing in FRCP 26(b)(3), the rule provides a step-by-step analysis for determining whether to produce or protect work-product materials. The Advisory Committee followed Hickman’s mantra “that one side should not automatically have the benefit of the detailed preparatory work of the other side.” It also noted that witness statements are not automatically discoverable, but could become so if the witness, for example, is reluctant or hostile, or has a lapse of memory.
Yet, as one treatise explains, federal courts provide lawyers inconsistent guidance on this first-step, witness-statement issue. Some federal courts hold that the work-product doctrine protects witness statements, while others do not. 10 Fed. Proc., L. Ed. § 26:188 (2019).
The Eastern District of Tennessee, in a series of cases, noted a split “across the country” whether the work-product doctrine protects signed witness statements—before deciding that it does not. Nam v. U.S. Xpress, Inc., 2012 WL 12840094 (EDTN May 15, 2012); Trustees of Plumbers v. Crawford, 573 F.Supp.2d 1023, 1028 (EDTN 2008) (stating that the work-product doctrine does not protect a sworn witness statement).
Now we have an differing opinion from EDTN’s Judge Steger that provides a deceptively simple solution to the pervasive inconsistency. In Mitchell v. Archer Daniels Midland Co., 2019 WL 117386 (EDTN Jan. 4, 2019), available here, the plaintiff’s lawyer retained an investigator to interview five individuals with factual information.
The investigator audio-recorded the interviews, and the defendant moved to compel them with a credible argument—previous courts in this district ruled that witness statements that merely recite the facts are unworthy of work-product protection.
Rather than perpetuate the inconsistency, the Mitchell Court simply applied Rule 26(b)(3). First, were the audio-recordings “documents” or “things” “prepared in anticipation of litigation”? The Court answered that, under FRCP 26(b)(3)(A), “recordings of witness interviews by or at the direction of an attorney in anticipation of litigation are generally not discoverable…. [a]nd this is true whether the recordings contain attorney mental impressions or whether they do not.”
The discoverability of “trial preparation materials”—as the Rule labels it—“does not turn upon the degree to which those materials reflect” opinion work-product. The initial question is whether the recordings constitute “trial preparation materials, not whether they contain opinion work product.” The Court refused to depart from the 26(b)(3) procedure because the recordings contained “only facts from non-party witnesses” sans “attorney mental impressions or trial strategy.”
Second, the Court analyzed whether it should override the recordings’ non-discoverability under 26(b)(3)(A)(i)—are the recordings relevant and proportional; and 26(b)(3)(A)(ii)—did the defendant show a substantial need. While the recordings were certainly relevant, the defendant failed to show it could not obtain the same information by other means, such as conducting its own interviews or taking their depositions.
This ended the analysis with the work-product doctrine protecting the witness recordings from discovery. But the Court helpfully provided us what would have been the third part of the analysis. If the defendant had shown a substantial need, the Court would have compelled production but protected the plaintiff’s opinion work-product as required under 26(b)(3)(B).
It would have done so through a redaction of the attorney’s mental impressions, but if the recordings “were so replete with opinion work product that redaction would be ineffective,” the Court would have denied the motion to compel.
So, there you have it—a quite simple, rule-based application of the work-product doctrine to nonparty witness statements. With the myriad, incongruent federal-court decisions on this issue, is the Mitchell decision the catalyst for clarity?
Time will tell.