When Federal Rule of Evidence 502 was amended in 2008 to make clear that the inadvertent production of privileged material will not constitute a waiver if the producing party took reasonable steps to prevent disclosure, there was a collective sigh of relief by inside and outside counsel. After all, with the explosion of e-mail and other forms of electronic material, in many cases the risk that privileged documents would slip through the cracks was real and substantial, and the costs to protect against disclosure (and a full subject matter waiver) were becoming prohibitive. Now, it seemed, so long as counsel took "reasonable steps" to prevent disclosure, litigants would have a free pass if privileged documents were inadvertently produced.
But what, exactly, are "reasonable steps" in the world of e-discovery, where the use of search terms and sampling has, by necessity, become routine among inside and outside counsel? The Advisory Committee has provided some guidance, recognizing that "a party that uses advanced analytical software applications and linguistic tools in screening for privilege and work product may be found to have taken 'reasonable steps' to prevent inadvertent disclosure."
But be wary: As a number of post-amendment cases illustrate, although the use of these tools will strengthen the "reasonableness" argument, their use is by no means sufficient. Consider a recent case from the Southern District of West Virginia, Mt. Hawley Ins. Co. v. Felman Production, Inc., 2010 WL 1990555, at *13 (S.D. W. Va. 2010). The plaintiff sought recovery on an insurance claim filed with the defendant. The plaintiff made a document production that included privileged evidence that the plaintiff sought to falsify its insurance claim. The defendant, without notice, promptly attached the email to a motion to amend its answer and assert counterclaims. Understandably, a big waiver fight ensued, which the plaintiff eventually lost.
What is most remarkable about the case is that the steps the plaintiff took to protect against disclosure - and that the court found to be insufficient - would sound familiar to most litigants: Among other things, hiring an outside e-discovery vendor; identifying, applying and testing search terms to locate privilege documents; identifying, applying and testing search terms for relevant, non-privileged documents; and conducting an attorney review of the relevant documents prior to production. Where counsel went wrong, though, was in assuming that the documents ultimately "hit" by the privilege search terms encompassed the universe of privileged documents. The testing of the privilege search terms turned out to be inadequate, and the review of the so-called "non-privileged" documents was not sufficiently robust to identify whatever privileged documents remained. In fact, more than 400 privileged documents made their way into the defendant's hands, and that - coupled with the high percentage of non-responsive documents in the production - was enough for the court to determine that the screening process was sloppy and reasonable steps were not followed. The court in Victor Stanley, Inc. v. Creative Pipe, Inc., 250 F.R.D. 251 (D. Md. 2008) put it best: "While it is universally acknowledged that keyword searches are useful tools for search and retrieval of ESI, all keyword searches are not equal; and there is a growing body of literature that highlights the risks associated with . . . relying exclusively on such searches for privilege review."
So how should litigants be guided by these cases? First, although there are a number of ways to reduce the costs of e-discovery, try to avoid cutting corners on the privilege review. Producing an irrelevant document is unfortunate, but manageable; a privilege waiver, on the other hand, can be devastating. Second, inside counsel should keep a running list of the names of current and former inside and outside attorneys, paralegals and legal staff. This is not too difficult a task, but will be enormously helpful when the time comes to conduct a privilege review. Third, use that list as the starting point for search terms, but add other relevant terms and do not assume that the segregation of documents hit by the search terms constitutes a defensible privilege review. Further testing, sampling and review of documents slated for production is essential, and attorney reviewers need explicit guidance on what constitutes a privileged document. Fourth, enter into an agreement setting forth the notice procedures upon the discovery of a privileged document. If the producing party fails to discover it first, above all that party needs to prevent the receiving party from using the document affirmatively (as the defendant did in Mt. Hawley). Finally, keep a checklist of all steps taken to protect against disclosure. The hope is that you never have to use it, but privileged documents get produced all the time and litigants are increasingly finding themselves fighting against a waiver.
These pointers are by no means exhaustive, but should provide some guidance as the courts' understanding of what constitutes "reasonable steps" continues to evolve.
Read Matthew Ingber's next column.