Products Liability Admissibility of Test Evidence: A Fresh Approach By Michael Hoenig - New York Law Journal - May 12, 2008	Print article

Battles over the admissibility of test evidence often are pivotal confrontations in products liability trials. Test evidence can enhance an expert's qualifications, standing and credibility. It can tend to prove or disprove a fact crucial to the determination of the claim. Tests can illustrate important scientific or technical principles, make complex expert testimony understandable, and provide a basis for drawing or rejecting critical inferences urged by a litigant.

Moreover, since the modern jury is television-, video- and computer-oriented, demonstrative test evidence that has a visual, "touchy-feely" quality to it possesses an effectiveness level over verbal descriptions. The maxim, "a picture is worth a thousand words," is perhaps more emphatically true today than in prior generations.

Because of its potential to sway or mislead juries or to give one expert an advantage over his or her adversary, test evidence properly receives close attention by trial and appellate courts.

But what standards should trial judges use to monitor the admissibility of test evidence? Are there predictable, uniformly applicable guidelines so that the litigants' counsel will have assurance that tests conducted by experts at considerable expense will actually be seen by juries? Indeed, do the admissibility rules cited by courts when they accept or reject test evidence make sense today? Or, because they were fashioned many years ago, do they represent mere relics from a bygone era? Has robust judicial "gatekeeping" of expert testimony for "reliability" under Frye or Daubert admissibility criteria effectively made evidentiary standards such as "sufficient similarity of conditions" or "substantially similar tests" outdated?

Given the complexities of these questions and our space limitations in this article, we may not be able to answer such questions definitively here. But we can get readers thinking more analytically, creatively and energetically about rationales for offering or opposing test evidence. We can begin to stimulate newer, fresher, more incisive advocacy which, in turn, may attract judicial attention regarding present weaknesses in how courts seem to decide the admissibility of test evidence. One catalyst for doing so here is a new, thoughtful, provocative, well-researched and clearly written article published far away. It is not likely to appear on most lawyers' or judges' radar screens.

Yet, this article, hot off the press in the Nebraska Law Review, by noted trial specialist and author Jonathan M. Hoffman, should be required reading for those grappling with test evidence in important cases. Entitled, "If the Glove Don't Fit, Update the Glove: The Unplanned Obsolescence of the Substantial Similarity Standard for Experimental Evidence,"¹ we will survey a few of the article's highlights here.

Illustrative Cases

But first, briefly, a couple of case examples to set the stage. Vinci v. Ford Motor Co.² was an automotive products liability case tried in the Bronx to a defense verdict. On appeal in the Appellate Division, First Department, the issues involved the trial court's admissibility of defendant's crash test video and its refusal to permit plaintiff's expert to testify as a rebuttal witness regarding the video. The panel majority opinion, referring to the admissibility issue as "unquestionably a close call," concluded that admission of the test evidence was a proper exercise of discretion "despite certain dissimilarities between the accident conditions and the conditions under which the test was conducted." However, the majority also ruled that it was reversible error not to let the plaintiff's expert testify on rebuttal in order to exploit dissimilarities between the video and the accident.

The dissenting justices agreed that the crash test re-enacting the accident was properly admitted in exercise of the trial court's discretion. They noted the traditional approach, that evidence of experiments is properly admissible "so long as the proponent establishes a substantial similarity between the conditions under which the experiments were conducted and the conditions at the time of the accident, particularly where the opponent has an unrestricted opportunity to cross-examine." Thus, while the "test conditions were not identical, there was sufficient similarity to permit the inference that the results of the reenactment - which used the same vehicles with the same seat belt configurations, based on plaintiff's expert data and a reasonable approximation of what was known about the underlying accident - shed light on what occurred." In addition, plaintiff had been given the opportunity for cross-examination to exploit any alleged dissimilarities. Conceding that the "better course" would have been to allow recall of the expert as a rebuttal witness on the crash test, the dissenters nevertheless concluded that, given the extensive cross-examination conducted when the video was admitted, plaintiff did not demonstrate that the outcome of the trial would have been different had the rebuttal expert been allowed.

In Styles v. General Motors Corp.,³ a products liability case decided by the Appellate Division, First Department, in July 2005, a Chevrolet Suburban SUV overturned after being hit head-on by another vehicle. The roof collapsed on the front passenger side. This was a hard fought case with sharp disagreements about the SUV's crash-worthiness, how many rolls occurred, how much force was applied to the roof and whether any manufacturer could protect against such severe forces. Plaintiffs' experts did a test in two phases on the same test vehicle. They removed the windshield, lowered the vehicle upside down at an angle onto the junction of the A-pillar and the roof. After some time the structure crushed. Then they lifted the damaged SUV and dropped it from a height of six inches at an angle. Plaintiffs' experts claimed this testing was similar to the federal standard's roof crush test involving a quasi-static pushing on the roof with a metal plate. Defendants objected. The trial judge did not hold a Frye hearing. The jury awarded a substantial verdict.

Admissibility Questions

The appellate panel said that the variations in the two tests from the federal standard's test did not render plaintiffs' experiment a "novel" scientific test within the meaning of the Frye "general acceptance" test. However, although each phase of plaintiffs' test was, separately viewed, a widely accepted technique, plaintiffs nevertheless failed to demonstrate that the use of both tests, in combination, on the same vehicle, had gained "general acceptance" within the pertinent scientific community. It was self-evident that an automobile subjected to two roof-stress tests was more likely to suffer a collapsed roof than a vehicle that undergoes only one such test.

Where, as here, the trial court admits expert testimony without conducting a preliminary inquiry into the reliability of the procedures utilized by the experts, the proper course was to hold the appeal in abeyance while the matter was remanded for a post-trial Frye hearing. At such a hearing, plaintiffs' experts would need to establish, inter alia, the general acceptance of their combination of the tests, to substantiate how the precise measurements of angle, weight, height, time, and other components were taken. Plaintiffs' experts would be limited to discussing the experiment they presented at trial, and would be precluded from offering any new or supplemental tests.

The concurring justices in Styles wrote separately in order to point out that the record, in their view, supported ordering a new trial as to liability. The testimony and proof of the plaintiffs' experts was deemed deficient because the test performed by plaintiffs' experts was not conducted under conditions "sufficiently similar" to that of the accident in question. Further, they said, the test evidence should not have been admitted because the procedure and results were not generally accepted as reliable in the scientific community.

In Styles, we actually see judicial scrutiny focused on two separate aspects of admissibility of the plaintiff's test evidence: (1) was the testing sufficiently "reliable" under the Frye "general acceptance" test governing an expert's novel scientific evidence; and (2) were the test conditions "sufficiently similar" to the accident conditions as to render the evidence admissible? There may be overlap between threshold expert reliability issues regarding test evidence and foundational issues described in older case law in terms of the "sufficiently similar conditions" criterion.

This relationship becomes even more involved because case law after Styles, notably Parker v. Mobil Oil Corp.,⁴ established that scientific evidence questions can boil down to two fundamental threshold issues regarding experts' reliability: (1) the Frye "general acceptance" test governing admissibility of "novel" scientific evidence and (2) the "foundational reliability" standard for admitting non-novel expert evidence. Our February 2008 New York Law Journal column, "Judicial 'Gatekeeping': 'Frye,' 'Foundational Reliability,'"⁵ elaborated on this development as articulated by the Court of Appeals in Parker.

Hoffman's Observations

Now we can turn to Jonathan Hoffman's astute observations in his Nebraska Law Review article.⁶ Mr. Hoffman says that, in the wake of the U.S. Supreme Court's decisions in Daubert⁷ and Kumho Tire,⁸ more forensic experts than ever in scientific and technical fields are testing the theories about which they testify in major litigation. This means that test evidence is not merely a species of demonstrative evidence. "Testability" of a hypothesis or opinion goes to the core of the reliability question. The rigorous quest for reliability of expert testimony stimulated by these decisions, later codified in amendments to the Federal Rules of Evidence, has made obsolete nebulous tests of "substantial similarity" traditionally used for screening experimental evidence.

When the Federal Rules of Evidence are rigorously applied to ensure that test evidence is reliable and relevant (as required by Evidence Rules 702 and 401) and that risks of unfair prejudice do not outweigh that evidence's relevance (as required by Rule 403), then the imprecise substantial similarity standard "becomes unnecessary." In a review of case law development before the Federal Rules of Evidence were enacted, the author demonstrates that courts actually began to treat experimental evidence as a discrete category using "substantial similarity" to determine its admissibility. This approach carried over into the modern era even though the codified rules of evidence provided adequate tools to gauge relevance, reliability and prejudice issues.

Standard Obsolete?

Since the Federal Rules of Evidence, particularly in the rigorous post-Daubert/Kumho Tire gatekeeping era, has ample muscle to screen test evidence fairly, the "substantial similarity" standard is no longer necessary. Indeed, the old criterion has serious flaws that often conflict with admissibility of relevant, reliable and nonprejudicial test evidence. Thus, the "substantial similarity" standard:

(1) can distract courts from analyses it ought to conduct under applicable rules of evidence;

(2) is so vague that it can permit disregard of significant dissimilarities or rationalize exclusion of tests based on trivial differences;

(3) creates a false dichotomy between "re-creations" and "abstract principles," as when courts disallow case-specific test conditions but allow dissimilar enactments to "merely illustrate" technical principles;

(4) allows exclusion of demonstrations that are "too similar" even though they also may illustrate relevant principles urged by one party's perception of the evidence;

(5) can shift the foundational burden from the party who should have it as, for example, when the proponent of expert testimony has the burden of showing the evidence is relevant and reliable under Rule 702 - but, then, the opponent has the burden under Rule 403 to demonstrate that the probative value of the relevant evidence is "substantially outweighed" by various prejudicial factors.

Rigid application of a "substantial similarity" standard for test evidence can skew the appropriate foundational burden.

Conclusion

Accordingly, Jonathan Hoffman concludes that the rules of evidence already provide the necessary tools for assessing the admissibility of experimental evidence. "They offer a more detailed and systematic methodology for identification and analysis of the pertinent issues relating to such evidence." His "modest proposal," to follow the Rules of Evidence, is summarized on page 669 of his law review article.

It is a lucid, eye-opening approach which calls for abandoning an evidentiary standard that is an "artifact" of pre-Federal Rules evidence law. Given the similar movement of New York state courts towards more robust gate-keeping of expert testimony under Frye (for "novel" scientific evidence) and under Parker for foundational reliability of all scientific evidence, perhaps the continued use of the "substantial similarity" standard for screening test evidence ought to be re-examined.

 

Michael Hoenig is a member of Herzfeld & Rubin.

Endnotes:
_________________________________________________________________________________________

1. Jonathan M. Hoffman, 86 Neb. L. Rev. No. 3, pp. 633-671 (2008).
2. 45 AD3d 335, 846 NYS2d 9 (1st Dept. 2007).
3. 20 AD3d 338, 799 NYS2d 38 (1st Dept. 2005).
4. 7 NY3d 434 (2006).
5. New York Law Journal, Feb. 11, 2008, p. 3.
6. "If the Glove Don't Fit," etc., supra n. 1, 86 Neb. L. Rev. pp. 633 - 671 (2008).
7. Daubert v. Merrell Dow Pharm. Inc., 509 U.S. 579 (1993).
8. Kumho Tire Co. Ltd. v. Carmichael, 526 U.S. 137 (1999).