For your delectation and delight, desultory dicta on the law of delicts.

Unlearning The Learned Treatise Exception

August 21st, 2010

The time has come to abandon or modify the learned treatise exception to the rule against hearsay. The federal version of this rule, Federal Rule of Evidence 803 (18) provides:

“Learned treatises.—To the extent called to the attention of an expert witness upon cross-examination or relied upon by the expert witness in direct examination, statements contained in published treatises, periodicals, or pamphlets on a subject of history, medicine, or other science or art, established as a reliable authority by the testimony or admission of the witness or by other expert testimony or by judicial notice. If admitted, the statements may be read into evidence but may not be received as exhibits.”

The rule is great fun for trial lawyers; it holds open the possibility to embarrass a witness on the stand with the statements of a “learned treatise,” written by a leading light in the witness’s field. Learned treatise cross-examination is not only fun, but it is practical and efficacious. The jury does not need to understand the statement from the witness or the contradictory statement from the text; it need understand only that the witness, who typically has never written on the subject, has been contradicted by a published statement. The learned treatise exception, however, is not well designed to advance scientific fact finding or the discovery of the truth. The exception uses publication, and endorsement by a witness, as a proxy for scientific reliability without any meaningful examination of the epistemic warrant for the treatise’s statement.

The learned treatise rule, incorporated into Federal Rule of Evidence 803(18), is inconsistent with the requirements of Rules 702 and 703. The point of these rules is that expert witness opinion must be helpful, and thus reliable, and that it must reasonably rely upon facts and data. Rule 705 permits the cross-examiner to insist that expert witnesses disclose the bases for their opinions. Statements from “learned treatises” have none of these evidentiary requirements or safeguards, and thus are suspect. Even if one were to adopt a Reaganesque “trust but verify” approach to expert witnesses, there is no way to verify the reliability of many statements from “learned treatises” because there is no practical method of inquiring into what the absent author has relied upon, or the reasonableness of that reliance. Furthermore, there is no way to inquire into the biases, prejudices, conflicts of interests, and enthusiasms of the absent author.

The problems engendered by Rule 803(18) have been put into bold relief by Supreme Court’s reading of Rule 702, and the subsequent Congressional revision of Rule 702. Rule 803(18) is a rule that excuses the absence of the hearsay declarant on grounds of the “trustworthiness” of the hearsay, but Rules 702 and 703 impose different, more stringent requirements for expert witness opinion testimony.

The comments to Rule 803(18) attribute the following basis for the rule to Wigmore:

“The foundation of the minority view is that the hearsay objection must be regarded as unimpressive when directed against treatises since a high standard of accuracy is engendered by various factors: the treatise is written primarily and impartially for professionals, subject to scrutiny and exposure for inaccuracy, with the reputation of the writer at stake. 6 Wigmore § 1692.”

Wigmore offered no empirical support for his high regard for treatises. Perhaps as a treatise writer he was biased in his assessment! Other evidence writers similarly favored the admissibility of learned treatises, and certainly their use for impeachment purposes. See, e.g., E. Cleary, et al., eds., McCormick on Evidence § 321, at 899 (3d ed. 1984); Morgan, Basic Problems of Evidence 366 (1962). The suggestion that a treatise writer has no bias with respect to a particular case misses the point that the writer may have a bias that applies to the class of cases, which he or she hopes to influence. The quaint notion that treatise writers seek to state their opinions accurately and truthfully ignores the obvious reality that writers’ opinions may be unreliable or unsupported. Even though publication may lead to reading and critical evaluation by some experts in the field, this eventuality is clearly insufficient to inhibit writers from stating opinions with incomplete and incorrect bases. Indeed, Wigmore’s assertions about the nature of treatise writing are wonderfully self-referentially illustrative of the flaws in opinions advanced in treatises.

The publication of professional books and journals publishing, at the turn of the last century, were certainly constrained by greater expense and smaller audiences than they are today. With the proliferation of publishers and journals, professional authors are rarely unable to get the most tendentious statements into print. Furthermore, there is now a great deal of empirical evidence that peer review is an extraordinarily weak screen for error and unreliability in published works.

Wigmore, in seeking to make scientific opinion more accessible, misrepresented scientific method and thinking, which is based upon facts, data, and reliable inference, not upon authority. The Royal Society, chartered in 1661 for “improving natural knowledge,” adopted its motto, “Nullius in Verba,” (On no one’s word) in 1663. The founders of the Royal Society meant to elevate the role of facts, determined experimentally, over authority in scientific method and discourse. [last visited August 20, 2010] A more contemporary statement from the last century comes from Albert Einstein. When Einstein’s detractors wrote a book, One hundred Authors Against Einstein, to deprecate and disprove his theory of relativity, Einstein reportedly stated, “If I were wrong, then one would have been enough.”

The learned treatise exception distracts the fact finder from the data and the reasoning behind the opinion at issue, and focuses on the opinions of out-of-court declarants. This is bad scientific practice and bad law. There are better ways for the law to structure cross-examination. Rule 26 of the Federal Rules of Civil Procedure requires disclosure of all the materials that expert witnesses have considered in arriving at their opinions. The cross-examiner thus has available the basis for showing that witnesses have failed to consider facts and data that cut against their opinions. To the extent that state law does not require similar disclosures, it should.

The learned treatise exception is a relic of an old way of thinking about scientific opinions. Its time has come and gone; now it should leave.