Predictive Coding Endorsed by the Southern District of New York.

Predictive Coding | People Plus Technology Equals Smart Search 

April 30, 2012| Law & Forensics
Part 1 of 1

Daniel B. Garrie, is the  Senior Managing Partner  at Law & Forensics LLC and works out of the Seattle, Los Angeles, and New York offices. He focuses on e-discovery, digital forensics, cyber security and warfare, data privacy, and predicitive coding working with law firms, governments, companies, and non-profits globally. 

Judge Carter published his opinion (1:11-cv-01279-ALC-AJP) in Monique da Silva Moore, et al. v. Publicis Group SA, et al. (hereinafter “Da Silva”).

In the opinion, Judge Carter affirmed Magistrate Judge Peck’s order approving (and arguably outright advocating for) the use of predictive coding, if both parties elect to use it.  The bottom line is that predictive coding is not going to be banished from the judicial cannon.

In Da Silva, both parties were amenable to establishing a protocol for the use of predictive coding, so long as it met their respective standards for reliability. Magistrate Judge Peck ordered both parties to come to an agreement on protocol, but the parties were unable to agree. In Da Silva, Magistrate Judge Peck ultimately set forth his own protocol, which plaintiffs argued was aligned distinctly with the defendants‘ protocol.  As a result, plaintiffs appealed Magistrate Judge Peck’s oral and written opinions to the District Court.

Practitioner Point:

Attorneys should not agree to the use of predictive coding or other search methodology or protocol without thinking it through.  Attorneys should develop and evaluate their electronic discovery protocol and evaluate the role of discovery technologies prior to meeting with the opposing party. The protocol should cover multiple areas including define the parameters they are willing to consider with regards to reliability, recall, and new technology. This protocol will assist the attorneys in ensuring that they know and can express what is acceptable to them prior to any agreement.

In Da Silva, plaintiffs’ brief objected under Rule 72(a) of FRCP and asked Judge Carter to set aside Peck’s ruling as “clearly erroneous or contrary to law”. The plaintiffs alleged that Magistrate Judge Peck violated FRCP 26, FRE 702 — because he improperly relied on outside evidence, biased defense experts, and otherwise adopted the defendants’ protocol based on insufficient evidence.  The District Court disagreed, and affirmed Magistrate Judge Peck’s protocol.

As Judge Carter notes several times, Rule 72(a) is a highly deferential standard and the Magistrate Judge is afforded a high degree of latitude with regards to non-dispositive decisions (i.e., discovery issues). Judge Carter affirmed Magistrate Judge Peck’s rulings finding “they are well reasoned and they consider the potential advantages and pitfalls of the predictive coding software.” Da Silva Moore v. Publicis Groupe, 11 Civ. 1279, at *3 (S.D.N.Y. Apr. 26, 2012)

Judge Carter also rejected plaintiffs’ contentions they would be denied reasonable discovery, because Magistrate Judge Peck’s protocol provided multiple points during the collection process where plaintiffs would have the ability to inspect what is produced and raise any such issue with Magistrate Judge Peck before final production. Magistrate Judge Peck’s protocol also allowed the parties, after the final production, to raise or revisit issues as to whether the software did the job. Judge Carter found that at this point in the litigation there was insufficient evidence to show that plaintiffs will be denied reasonable discovery.

Practitioner Point:

Counsel should engage experts early to effectuate a strategy. People that are intimately familiar with these complex new technologies will help avoid wasting valuable resources, energy, and efforts.

Judge Carter’s ruling recognizes the validity of the role of technology in mitigating the costs of e-discovery.  Discovery is no small task, and the costs associated with collecting and culling electronic documents can be monumental.   However the aim should still be making the litigation about the issues and not about discovery.  The goal of adding technology to the discovery process should be to provide a solution that will allow lawyers to focus on issues and not discovery.

Perfection is in the eye of the beholder, or as Carter so clearly states, “[t]here is no review tool that guarantees perfection”. Id. at 4. It has been our experience that expertise is critical when it comes to introducing new technology to the courts.  This decision affirms the belief that the tool is only as good as the people that use it.  The question for counsel to consider is as follows: Either method of computer assisted review is capable in the right hands, but which is more capable in lesser-trained hands?  The answer will increasingly be predictive coding.

Author: Daniel Garrie, Esq. is a renowned e-discovery special master and thought leader in the fields of information security, forensics, e-discovery, information governance, and digital privacy