TAR Caselaw Update: Out West and Down Under
Here on the DiscoverReady blog, we report periodically on significant developments in the case law regarding the use of Technology Assisted Review (TAR) in legal discovery. This update, which recaps three cases decided in the last several months, takes us out to the Left Coast, and then to the land down under.
Out West: In re: Viagra Products Liability Litigation
In October of 2016, in the Viagra Products Liability Litigation, the Northern District of California issued an opinion on the use of TAR—the first one from a California court in more than two years. Citing to Judge Peck’s opinion in Hyles v. New York City, Magistrate Judge Sallie Kim denied the requesting party’s motion to compel the use of TAR and/or predictive coding. Pfizer, the responding party, objected to Plaintiffs’ insistence that it use TAR, preferring instead to use an iterative search term process—validated by statistical sampling and measurement—to identify responsive documents. As the court explained, “the responding party is the one best situated to decide how to search for and produce ESI responsive to discovery requests.” The court concluded that, “even if predictive coding were a more efficient and better method, which Pfizer disputes, it is not clear on what basis the Court could compel Pfizer to use a particular form of ESI, especially in the absence of any evidence that Pfizer’s preferred method would produce, or has produced, insufficient discovery responses.” In re Viagra Prods. Liab. Litig., Case No. 16-md-02691-RS (SK), slip. op. at 2-3 (N.D. Cal. Oct. 14, 2016).
Down Under: Money Max and McConnell Dowell
The last half of 2016 saw the international acceptance of TAR in legal discovery expand to the southern hemisphere, when two Australian courts addressed its use. In the first, Money Max v. QBE Insurance Group, the responding party had already used TAR to provide discovery documents. In its order, the Federal Court of Australia required the responding party to make certain disclosures about its TAR process, including:
- (a) the nature and technical parameters of the TAR algorithm used;
- (b) the process for selecting and coding the training set of documents;
- (c) the process for selecting and coding the validation set of documents;
- (d) the process for training the algorithm to identify relevant documents for production, including the level of relevance applied;
- (e) the process for validation and testing, including disclosure of analyses relating to the accuracy, validation or quality of documents produced;
- (f) the number of documents in the complete data set identified as relevant and irrelevant following the application of TAR and, with respect to the relevant documents, the number of documents withheld on the basis of privilege;
- (g) the search terms applied in conjunction with TAR; and
- (h) the process followed with respect to potentially privileged documents.
The court also ordered the parties to meet and confer about any disputes regarding the TAR process. Money Max Int Pty Ltd. v. QBE Ins. Group Ltd,  FCAFC 148 at 3-4 (Austl.).
In the second case, the Supreme Court of Victoria approved the use of TAR in McConnell Dowell Constructors v. Santam Ltd. The producing party, McConnell Dowell, identified a population of 1.4 million potentially relevant records. After estimating almost 600 hours of manual review to complete the project, the parties—with the help of a Special Referee appointed by the court—agreed to use predictive coding. Citing to cases from the U.S., the U.K., and Ireland, the court noted that a “TAR process is far more sophisticated than a word search facility.” Accordingly, the court accepted the Special Referee’s recommendation and permitted the parties to use TAR. McConnell Dowell Constructors (Aust) Pty Ltd v. Santam Ltd & Others (No 1)  VSC 734 (Austl.).