The Frame of Reference and Other Problems

Richard D. Friedman*

Jeffrey L. Fisher**

Given the pair of Fishers, we’ll use first names. And because we have more to say about George’s essay, we’ll concentrate our attention there.[1]

I. History

George argues that, centuries ago, jurists did not distinguish between testimonial and nontestimonial hearsay, and so the distinction cannot be a historically well-grounded basis for modern confrontation doctrine. The argument proceeds from an inaccurate frame of reference.

When the confrontation right developed, principally in the sixteenth and seventeenth centuries, and English defendants—Raleigh among them—demanded that adverse witnesses be brought face to face with them, they were making a procedural assertion as to how witnesses must give their testimony. (Giving testimony is what witnesses in litigation do.) Rarely did they phrase this claim in terms of hearsay, for the simple reason that there was no rule against hearsay in the modern sense. Similarly, numerous statutes protected the right of treason defendants to have witnesses brought face to face, and these statutes never mentioned hearsay.

True, Geoffrey Gilbert’s treatise from the early eighteenth century said that “a mere hearsay is no Evidence,” but that was a limp statement: neither he nor any other writer at the time elaborated on it.[2] (Contrast the dense discussion of the law governing witnesses.) They did not offer a definition of hearsay, without which an exclusionary rule is indeterminate, nor did they catalogue exceptions, without which such a rule would be impractically broad. The fact is that lots of hearsay was admitted in the early- and mid-eighteenth century, in criminal as well as civil cases.[3]

George certainly agrees that not all hearsay should be excluded. But the point appears to have escaped John Marshall; Marshall’s broad-brush 1807 condemnation of all hearsay (and exclusion of a conspirator’s statement) proves far too much and offers no support for George. Marshall’s statement does suggest that by then—thirty years after the early state constitutions articulated the confrontation right, some using the “face-to-face” formula—lawyerly recognition of, and broad opposition to, hearsay had taken hold.[4] Over succeeding decades, the wide scope of the hearsay rule tended to make it, rather than the confrontation right, the doctrinal focus, even as the rule was whittled and shaped by exceptions that made its scope more practical but obscured its rationale.

As for George’s argument based on the statement by the Portuguese gentleman in Raleigh’s case, Rich responded to this precise argument earlier in the year, so given word limitations we will offer only a citation here.[5]

And as for dying declarations: When the doctrine emerged, it was not conceived as an exception to the rule against hearsay, which was still in embryonic form. Rather, the stated rationale was that imminent death was as powerful a sanction as the oath—in other words, that a surrogate for the ordinary required conditions of testimony was present. (Rich has argued repeatedly that forfeiture provides a better rationale.[6]) Even as the hearsay rule began to gel, a leading case described dying declarations as an alternative (along with formal statements to a justice of the peace taken under statutory directive) to the ordinary way of giving evidence—live testimony subject to oath and confrontation in open court.[7]

II. Standards

Like George, we have been disappointed by the Supreme Court’s failure thus far to articulate a fully developed conception of what is “testimonial.” But the problem is fixable—it is not the fault of the underlying approach—because such a conception is available.

We agree that it is confusing to speak of an actor’s primary purpose “objectively considered.” Purpose is a subjective matter. But this aspect of the problem would disappear if the Court spoke, as we believe it should, in terms of reasonable anticipation—rather than purpose—of prosecutorial use.

Whose anticipation? Again, we agree that the Court has muddied the waters. Justice Scalia is right: it is the anticipation of the declarant, the purported witness, that should be decisive.[8] We believe that this is not inconsistent with the Court’s pronouncements that the perspective of the questioner, if there is one, should be taken into account: what the declarant understands the purpose of the questioner to be may serve as a key factor in determining likely use of the statement. (And there is nothing perverse, by the way, about the fact that a prosecutor might create evidence without being subject to the Confrontation Clause; evidence creation alone is not sufficient to invoke the clause, which is indisputably about witnessing.)

III. Reliability

George contends that the best rationale for Crawford is that “declarants who speak in anticipation of trial have reason to lie, rendering their statements unreliable,” but his argument misses the mark. Crawford, while recognizing that the Confrontation Clause creates “a procedural rather than a substantive guarantee,” spoke of its “ultimate goal” as being “to ensure reliability of evidence.” That’s not quite right. Eyewitness testimony is notoriously unreliable, and confrontation cannot ensure reliability; when two witnesses give conflicting evidence, both cannot be reliable. The most we can hope for is that confrontation will help the trier of fact make accurate findings out of an assemblage of evidence, much of which may be very unreliable. But whether or not that is true, the real purpose of the clause is to ensure that witnesses perform their function under historically prescribed conditions (that is, face to face, under oath, subject to cross-examination, and, if reasonably possible, in open court). When a witness testifies at trial, a judge does not say, “That was so reliable that no cross is necessary.” There’s no reason for a different result when a witness has testified out of court, with the expectation that the testimony will be used for criminal investigation and prosecution.

IV. Roberts Redux?

In the end, George’s proposed solution is essentially an attempt to create an ideal hearsay code and give it constitutional force. While creating such a hearsay code is a worthwhile project, we respectfully suggest that it has nothing to do with the confrontation right. At any rate, there is no reason to believe a second attempt at turning the confrontation right into a catalogue of supposedly reliable classes of evidence, backed up by case-specific assessment, would be any more successful than the first. Can we really say that custodial confessions are inherently unreliable? Even a confession acknowledging responsibility for a murder, as in Lee v. Illinois?[9] Someone nostalgic for a return to the era in which the Court transformed the confrontation right into an attempted sifter of good and bad evidence might take pause from the fact that, decades into the Roberts regime, lower courts mangled cases like Lee, Lilly v. Virginia,[10] and Crawford. In each of these instances, the lower courts held admissible against the accused a statement made out of court to the authorities by another person—even though for more than three centuries it had been obvious that one could not act as a valid witness against another by making a custodial confession.[11]

It is true, though unfortunate, that four justices think that the scientific underpinnings of forensic lab reports, at least in part, make such reports nontestimonial. But no member of the present Court has stated that he or she wishes to return to the Roberts regime. Small wonder; it yielded a doctrine without historical or textual basis, one that gave inadequate guidance to the lower courts and failed to provide clear protection in core cases.

The testimonial approach expresses a sound conception of the confrontation right. It has not yet developed as fully as it should. But the missteps that George identifies and some resulting opacity in the doctrine need not constrain the future.


*         Alene and Allan F. Smith Professor of Law, University of Michigan Law School.

**       Professor of Law, Stanford Law School.

[1].     We certainly agree with Deborah that it is important to recognize the public implications of domestic violence. We do not regard the holding in Hammon v. Indiana as violating this principle. There, the suspect was known and in the presence of the police; it was clear that there was no imminent danger to anybody.

[2].     Geoffrey Gilbert, The Law of Evidence 99 (Garland Publ’g 1979) (1754).

[3].     E.g., John H. Langbein, The Origins of Adversary Criminal Trial 234–35 (2003); T.P. Gallanis, The Rise of Modern Evidence Law, 84 Iowa L. Rev. 499 (1999); John H. Langbein, Historical Foundations of the Law of Evidence: A View from the Ryder Sources, 96 Colum. L. Rev. 1168, 1189–90 (1996).

[4].     See Gallanis, supra note 3, at 503 (“[T]he 1780s were a period of considerable activity and . . . by 1800 much of the modern approach to hearsay was already in place.”).

[5].     Richard D. Friedman, The Mold That Shapes Hearsay Law, 66 Fla. L. Rev. 433, 459–62 (2014).

[6].     E.g., Richard D. Friedman, Giles v. California: A Personal Reflection, 13 Lewis & Clark L. Rev. 733 (2009).

[7].     King v. Woodcock, 1 Leach 500, 502, 168 Eng. Rep. 352, 353 (1789).

[8].     We will not discuss here whether the ordinarily applicable approach should be adjusted, or replaced altogether, when the speaker is a child or a person with developmental deficiencies.

[9].     476 U.S. 530 (1986).

[10].     527 U.S. 116 (1999).

[11].     Tong’s Case, Kel. J. 17, 18, 84 Eng. Rep. 1061, 1062 (K.B. 1662).


Download as PDF.