Over the past four years, I have seen a surprising number of new software products enter the eDiscovery market for identifying, collecting, processing and reviewing electronically stored information (ESI). So far, I have counted more than 70 new eDiscovery-related software products. When a company invests in a new in-house eDiscovery technology, the legal department should always inquire about whether the software has ever been challenged in a court proceeding because there is a possibility that they will need to defend the technology in court.
I am starting to see a trend of attacking the technology used to perform the collection and processing as a way to distract lawyers from the merits of the case. Unchallenged products are highly susceptible to this because the collateral attack on the technology usually focuses on raising doubt about the software’s accuracy and reliability.
Attacking the technology is different from attacking the effectiveness of the search strategy used to target responsive documents on the enterprise. We have seen the consequences of attorneys employing poor search strategy in opinions such as Victor Stanley v. Creative Pipe, Inc., in which the court came down on the party for not using “careful thought, quality control, testing and cooperation with opposing counsel in designing search terms.”
Checking for Technology Citations
When checking for cases that have cited particular applications, there is an important distinction between a court’s “vetting” the use of a technology product and the court’s merely “referencing” or “mentioning” its use. Mentioning is where a court merely calls attention to a technology used for the investigation. While a “mention” presumes the court approved of its use, in fact that is not the same as “vetting” it for use in obtaining ESI.
The cases in which the court legally vets an application are strong precedents because the court is satisfied that the technology is sound, and not just for use in the case before it, but for use by other courts in the jurisdiction in future cases as well.
The best form of validation a court can give is when it takes judicial notice of the established reliability of the software. Black’s Law Dictionary defines judicial notice as the act by a court to “recognize the existence and truth of certain facts, having bearing on the controversy at bar, which, from their nature, are not properly the subject of testimony, or which are universally regarded as established by common notoriety.” Such notice excuses a party from having the burden of establishing fact from the necessity of producing formal proof.
When opposing counsel questions the reliability of the computer program that generated or processed the ESI in question, the proponent of the evidence must testify to the validity of the program or programs used in the process. Daubert v. Merrell Dow Pharmaceuticals, Inc. is a landmark Supreme Court decision that sets forth a legal test to determine the validity of scientific evidence and its relevance to the case at issue. Corporate legal departments should become familiar with the basic elements of the Daubert analysis:
1. Whether a “theory or technique … can be (and has been) tested;”
2. Whether it “has been subjected to peer review and publication;”
3. Whether, in respect to a particular technique, there is a high “known or potential rate of error;” and
4. Whether the theory or technique enjoys “general acceptance” within the “relevant scientific community.”
On the other hand, in state courts we see the Frye v. United States test used to determine whether a scientific technique for obtaining, enhancing or analyzing evidence is generally accepted within the relevant scientific community as a valid process. Many other countries with a common law system also utilize standards with many similarities to Daubert and Frye.
Additionally, Federal Rule of Evidence 901(b)(9) provides a presumption of authenticity for evidence generated by or resulting from a largely automated process or system that is shown to produce an accurate result. This rule is often cited in the context of computer-processed evidence. As noted by the court in Lorraine v. Markel, Rule 901(b)(9) was designed in part to account for computer evidence. The Lorraine court noted that this was an important method to authenticate electronically stored information.
In complex eDiscovery cases with a large data volume, evidence concerning the search and recovery function and resulting visual outputs and printed reports is often as important as the recovered data itself. Understanding the tests used to validate the technology and how to properly defend it under a Frye/Daubert analysis is just as important as understanding its features and functions.
The “wow” effect of a new product demonstration will be quickly dampened when a company learns the software has no judicial track record or may be difficult to defend. Unless a company that sells you the technology is willing to stand behind it in court and has the necessary documentation to support a Daubert challenge, buyer beware!