The Kavanaugh Document Dump: An Ediscovery Perspective

You thought ediscovery disputes got heated.

The appointment hearings for Brett Kavanaugh’s nomination to the Supreme Court of the United States have taken political contentiousness to a whole new level. With both sides of the aisle flinging accusations of partisan gamesmanship, Kavanaugh’s nomination seems to be constantly perched on the brink of catastrophic derailment. Let’s step away from the politicking for a moment and look at one recent event — a last-minute document dump — from an ediscovery perspective.

What Happened: The Late Document Production

Kavanaugh’s first nomination hearing was scheduled to begin at 9:30 a.m. Eastern time on Tuesday, September 4, 2018. Mere hours before then — around 5 p.m. on Labor Day, September 3 — a lawyer representing George W. Bush released 5,148 documents totaling 42,390 pages to the Senate Judiciary Committee. These documents related to Kavanaugh’s service in President Bush’s administration. The public still has little information about the late production: all of the documents have been designated “committee confidential,” limiting access to the members and staff of the Judiciary Committee. Chuck Schumer, the Senate Minority Leader, promptly tweeted about the “absurd” production, claiming that “Not a single senator will be able to review these records before tomorrow.” The Senate Judiciary Committee accepted his challenge, tweeting its response just before 11 p.m.: “The Majority staff has now completed its review of each and every one of these pages.” To be clear, these pages were a drop in the bucket compared to earlier productions. Bush’s legal staff had already provided approximately 415,000 additional pages to the committee. (Nor is that all; the Trump administration has another 101,921 pages that it has withheld under executive privilege.) In a conflict that will be familiar to ediscovery counsel everywhere, Democrats have sought a full review of Kavanaugh’s documents before any committee hearings, while Republicans have argued that those documents are irrelevant. The National Archives, playing the role of discovery vendor here, has released many of its records already but states that it will be unable to release all of Kavanaugh’s documents before October. Predictably, the responses to this late document dump have fallen along political lines. Democrats stalled Kavanaugh’s hearing on Tuesday morning, demanding more time to review the new documents. Meanwhile, one conservative columnist objected that “the last-minute dump is a time-honored tradition in the nation’s capital and a common tactic in litigation.” But is it? Would this fly in ediscovery practice?

An Ediscovery Perspective on the Kavanaugh Document Production

In the world of ediscovery, a 42,000-page production isn’t so unusual. Earlier this year, in Motorola Solutions, Inc. v. Hytera Communications Corp., No. 17 C 1973 (N.D. Ill. May 17, 2018), the magistrate observed that the parties’ single-issue discovery had already amassed 700,000 documents totaling nearly 3 million pages. Of course, ediscovery attorneys bring an array of sophisticated tools to bear on these million-page productions. Moreover, even if the Senate Judiciary Committee has ready access to ediscovery processing software and technology-assisted review (TAR) platforms, it must still receive documents in a searchable electronic format for these tools to do much good. If it doesn’t, scanning thousands of pages of documents or running optical-character recognition software could, itself, take hours. Assuming that the production was a true document dump with mixed types of electronic files, the first stages of processing it would require basic organization efforts: deduplication of related emails, email threading, and concept clustering. While processing and initial keyword searches for relevant terms might be fairly speedy, subsequent review would be considerably slower: humans, however quickly they read, need time to do so, and predictive coding software takes time to train. Here, an enormous problem arises: in litigation, a limited number of issues are in dispute. There may be several complex issues in a case, but discoverable information must ultimately be relevant to some contested issue. Each of those issues can be distilled into a list of keywords and concepts that a human review team, with or without TAR, can watch for. In a Supreme Court nomination hearing, practically anything might be relevant. The primary legal issues that Democrats raised with Kavanaugh’s nomination ranged from abortion and impeachment to torture and executive power, with a side helping of racial bias and the Second Amendment. But as we’ve seen since, extrajudicial conduct and attitudes are equally likely to derail a court appointment. This wide-ranging relevance would slow down any review, technologically enhanced or otherwise. How would the courts respond to a last-minute production like this one? It’s hard to say for certain. Earlier this year, McCallan v. Wilkins, No. 2:18-CV-117-WKW (M.D. Ala. Mar. 19, 2018), reviewed a bankruptcy court’s decision to hold a party in contempt for discovery misconduct including repeated document dumps. Prior cases have resulted in monetary sanctions for document dumps where responsive documents have been “imbedded in large amounts of otherwise irrelevant documents.” But more recently, in Murphy v. Piper, No. 16-2623 (DWF/BRT) (D. Minn. Aug. 6, 2018), the court didn’t sympathize with a party that received over 2.5 million pages in a late-stage document dump. Instead, it advised the parties to “be careful what you ask for, you just might get it.” At this point, no one in the Kavanaugh hearings seems happy with what they’re getting.