E-Discovery Update: Finding the Line Between E-Discovery Expert and Fact Witness TestimonyBy Conrad J. Jacoby, Published on June 25, 2007
Digital information is an increasingly common part of civil and criminal litigation. Electronic mail messages and documents-or evidence that such materials are suspiciously missing-are powerfully persuasive pieces of evidence that can make or break a case. Courts recognize the importance of electronic files and their analysis, and they routinely admit such materials into evidence. However, where is the point where relaying objective facts about electronic documents crosses the line into expert opinion testimony?
The Sixth Circuit case, United States v. Ganier, 468 F.3d 920 (6th Cir. 2006), recently addressed exactly that issue. In a criminal matter, the prosecution announced its intention to present the testimony of an IRS agent regarding evidence that potentially relevant files had been located by running queries using Microsoft Windows utilities-and then deleted. This activity was uncovered through the use of forensic software that permitted the examiner to review Windows registry information and confirm the existence of the deleted files to the extent that they hadn't already been overwritten at the time of the forensic examination.
Defendant Ganier challenged the admissibility of this evidence on grounds that the government had failed to disclose expert testimony as defined by Federal Rule of Evidence 702 as was required by the Criminal Procedural Rules. The government, in turn, opposed defendant's motion by characterizing the proposed testimony as purely factual in nature and essentially similar to any lay witness testimony about what they saw in the course of their everyday work. In this regard, the government argued, the forensic software was no different from more common software, such as Microsoft Word or Outlook. The trial court sided with the defendant and granted his motion to exclude evidence of the forensic analysis, but also certified this question for interlocutory appeal.
On appeal, the Sixth Circuit reversed the trial court with respect to the specific relief that had been granted, but the Court affirmed the trial court's classification of the proposed IRS agent testimony as expert analysis that is clearly governed by F.R.E. 702. Key to this conclusion was the complexity of the tools and analysis performed by the agent. The Court found that unlike Microsoft Word or Outlook, programs that are used by millions of computer users each day, the agent's analytical software was a specialized program (or suite of programs) that required separate and uncommon expertise to operate. In the court's eyes, this additional technical proficiency and understanding of normally hidden Windows functionality elevated the agent's otherwise factual statements to the level of "specialized technical knowledge" governed by FRE 702.
From a practitioner's perspective, the Ganier opinion should raise few alarms that it is articulating a new and elevated standard for admitting electronic documents and analysis into evidence. Indeed, the Court's opinion may make it easier for lay witnesses (in addition to experts) to offer authenticating testimony about electronic files created by common software programs. The Ganier opinion even makes it more likely that lay witnesses can testify about easily visible document metadata such as document properties or tracked changes, since these are program functions that many people use on a daily basis. The Court noted the long-standing requirement that the lay witnesses must have made their technical observations in the course of their regular activities, not as "after-the-fact" investigations specific to the litigation action, but that does not prevent a lay witness from testifying, for example, about seeing a computer backup device write to a specific backup tape or what scripts were run to achieve certain results. In some circumstances, this can venture into fairly complex testimony.
The Ganier opinion also suggests that information found in corporate databases and content management systems could be validated by users who can describe what steps they take (or that the system takes) to ensure that this digital information is reliable for their work-so long as the underlying systems are generic commercial systems that are relatively popular in the market. This contrasts strongly with the position some courts have taken that detailed expert testimony is required to describe and validate not only a specific piece of digital evidence at issue, but also the hardware and software used to create and store it. See, e.g., In re Vinhnee, 2005 WL 3609376, 06 Cal. Daily Op. Serv. 146, 2006 Daily Journal D.A.R. 169 (B.A.P. 9th Cir. Dec 16, 2005) (detailed 11-part test required to authenticate electronic evidence).
On the other hand, the Sixth Circuit's analysis suggests that discussing the output and actions of less commonly used software programs should be classified as expert testimony that requires appropriate foundation and disclosure. This raises the question of whether the Sixth Circuit's test relies upon the popularity of the software and hardware at issue or its complexity. Though the Court cited the IRS agent's technical expertise as an important factor pushing his testimony towards the side of expert testimony, Microsoft Word and Outlook, programs cited by both the government and the Court as examples of mainstream software, are very sophisticated software programs that contain extensive features and functions that few users ever learn, much less master. However, the Ganier opinion suggests that lay witnesses could likely describe how these programs operate and sponsor files created by these programs for admission into evidence. One can certainly argue that the forensic software analysis required expert testimony because it is less common, not because it is more difficult to use.
Applying this analysis to less sophisticated but unpopular computer programs or obsolete mainstream programs like MultiMate or VisiCalc would seem to reinforce this interpretation. VisiCalc, released in 1979, was the first personal computer spreadsheet program, and some claim it launched the entire business PC industry. Twenty years ago, MultiMate was one of the most popular word processing programs in the world and held much the same prominence that Microsoft Word does today (at least until it was eclipsed by WordPerfect). These programs had very large user communities at their peak popularity, but these vintage programs are now abandoned and used by virtually no one. Many courts, not just the Sixth Circuit, would likely require expert testimony as foundation for any evidence about these systems today.
A common-sense approach consistent with the Sixth Circuit's analysis is to acknowledge that it is the uncommon nature of niche software and computer systems, not their sophistication, that requires expert, not lay, testimony about how they work. In the case of vintage software, a retired secretary who used such systems for many years might have sufficient experience and background to serve as an expert. Similarly, a "power user" of forensic software, such as the IRS agent in Ganier, might qualify as an expert for purposes of explaining the analysis generated by the software. This approach adds procedural steps to the party sponsoring the evidence, but, for the most part, it does not erect insurmountable legal barriers to introducing this information.
Counsel should carefully consider the type of electronic evidence they believe will be important to develop their case and whether expert testimony will be required to admit these materials into evidence. Typically, pursuant to case management orders, expert witnesses are disclosed on a fixed schedule so that their expert reports or testimony can be analyzed in advance of a hearing or trial, and Courts are generally ill-disposed towards permitting additional experts to be named at a late point in the litigation, seeing this as a potential ambush that is inconsistent with the philosophy of fact discovery. As a consequence, though, early case assessment and understanding are even more crucial than ever before. Without lists of potential exhibits and their associated foundational weaknesses, litigants may not discover the need for "expert" testimony until it's too late.