87 Harv. L. Rev. 957, 957-961 (1974)

While another elaborate argument for the reform of hearsay law might be in order, the purpose of this Comment is a more modest one. It will present a heuristic device which I believe can aid in the analysis of the hearsay rule and its exceptions by exposing the rule's structure and the underlying values at stake. The goal, in the first instance, is to assist students of the law in understanding an otherwise complex area. No less important, by directing the attention of students, practitioners, and judges toward structure and policy and away from cookbook formulas of rule and exception, the device may dramatize the inconsistencies and oddities that pervade hearsay law. And that may be a first step to future reform.


The basic hearsay problem is that of forging a reliable chain of inferences, from an act or utterance of a person not subject to contemporaneous in-court cross-examination about that act or utterance, to an event that the act or utterance is supposed to reflect. Typically, the first link in the required chain of inferences is the link from the act or utterance to the belief it is thought to express or indicate. It is helpful to think of this link as involving a "trip" into the head of the person responsible for the act or utterance (the declarant) to see what he or she was really thinking when the act occurred. The second link is the one from the declarant's assumed belief to a conclusion about some external event that is supposed to have triggered the belief, or that is linked to the belief in some other way. This link involves a trip out of the head of the declarant, in order to match the declarant's assumed belief with the external reality sought to be demonstrated.

The trier must obviously employ such a chain of inferences whenever a witness testifies in court. But the process has long been regarded as particularly suspect when the act or utterance is not one made in court, under oath, by a person whose demeanor at the time is witnessed by the trier, and under circumstances permitting immediate cross-examination by counsel in order to probe possible inaccuracies in the inferential chain. These inaccuracies are usually attributed to the four testimonial infirmities of ambiguity, insincerity, faulty perception, and erroneous memory. In the absence of special reasons, the perceived untrustworthiness of such an out-of-court act or utterance has led the Anglo-Saxon legal system to exclude it as hearsay despite its potentially probative value.

There exists a rather simple way of schematizing all of this in terms of an elementary geometric construct that serves to structure its several related elements. The construct might be called "the testimonial triangle." By making graphic the path of inferences, and by functionally grouping problems encountered along the path, the triangle makes it easier both to identify when a hearsay problem exists and to structure consideration of the appropriateness of exceptions to the rule that bars hearsay inferences.(1) Double-click to see diagram

If we use the diagram to trace the inferential path the trier must follow, we begin at the lower left vertex of the triangle (A), which represents the declarant's (X's) act or assertion. The path first takes us to the upper vertex (B), representing X's belief in what his or her act or assertion suggests, and then takes us to the lower right vertex (C), representing the external reality suggested by X's belief. When "A" is used to prove "C" along the path through "B," a traditional hearsay problem exists and the use of the act or assertion as evidence is disallowed upon proper objection in the absence of some special reason to permit it.

It is of course a simple matter to locate the four testimonial infirmities on the triangle to show where and how they might impede the process of utterance. To go from "A" to "B," the declarant's belief, one must remove the obstacles of (1) ambiguity and (2) insincerity. To go from "B" to "C," the external fact, one must further remove the obstacles of (3) erroneous memory and (4) faulty perception.

When it is possible to go directly from "A" to "C" with no detour through "B," there is no hearsay problem unless the validity of the trier's conclusion depends upon an implicit path through "B."(2)9 Suppose, for example, that the issue in a lawsuit is whether the Government took adequate safety precautions in connection with the nuclear test at Amchitka in 1971. James Schlesinger, then Chairman of the Atomic Energy Commission, "told reporters at Elmendorf Air Force Base outside Anchorage that he was taking his wife ... and daughters ... with him [to the site of the Amchitka blast] in response to Alaska Gov. William E. Egan's invitation. Egan strongly disapprove[d] of the test." In these circumstances, the trip from "A," the Chairman's proposed travel with his family to the site of the blast, to "C," the conclusion that the blast was reasonably safe, may appear at first to be purely "circumstantial," but in fact that trip requires a journey into the Chairman's head and out again--a journey through the belief "B" suggested by his willingness to be near the blast with his family. The journey from "A" to "B" involves problems of possible ambiguity and of insincerity in that the Chairman was apparently seeking to dispel fears of danger, so that his act may not bespeak an actual belief in the test's safety. And the journey from "B" to "C" involves problems of memory and perception in that he may not have recalled all the relevant data and may have misperceived such data in the first instance, so that his belief in the test's safety, even if we assume the journey from "A" to "B" safely completed, may not correspond to the facts sought to be demonstrated. On both legs of the triangle, therefore, there are testimonial infirmities that cross-examination contemporaneous with the act "A" could help to expose.

By contrast, when the trier's inference can proceed from "A" directly to "C," the infirmities of hearsay do not arise....


1. wpe2E.jpg (40252 bytes)

2. 9. An uncompromising behaviorist might insist that no detour through mental states is ever necessary because every trip from an act or utterance "A" to a conclusion "C" is reducible to a circumstantial inference about the statistical frequency with which "C" is present when "A" is present. There are difficulties with accepting the behaviorist perspective as a coherent one. See Chomsky, A Review of B.F. Skinner's Verbal Behavior, 35 Language 26 (1959); Chomsky, The Case Against B.F. Skinner, New York Review of Books, Dec. 30, 1971, at 18. But even if one does adopt such a perspective, it does not follow that the trier's way of using the evidence "A" will in fact mirror that perspective, for the trier is likely to reason about states of mind even if it is in some sense incorrect or unnecessary to do so. Moreover, the connection between "A" and "C" may well be such that the frequency with which the latter accompanies the former depends upon the actor's testimonial capacities so that, even from a behaviorist perspective, information about a declarant's use of language, tendency to lie, eyesight, and so forth, may increase or decrease the statistical correlation between the utterance and the fact reported.