Concept
Exceptions to the Clause
Right to confront witnesses
Exceptions to the Clause
Crawford v. Washington does not bar every testimonial statement that the defendant never had a chance to test. The Court read the Sixth Amendment as a reference to the right of confrontation at common law, subject only to the exceptions established when the Amendment was adopted. That reading closes one door and leaves another open. A court may not invent an exception because a class of statements looks trustworthy, since the Court rejected a judge’s finding of reliability as a substitute for cross-examination.1 A rule of admission settled at common law survives, however, because the Clause carries the old right forward together with its old limits.
Two such rules matter in practice. The first is forfeiture by wrongdoing: a defendant who procures a witness’s absence loses the right to confront that witness, but under Giles v. California only when the conduct was designed to keep the witness from testifying. The second is the dying declaration, a statement by a speaker who believes death is near. Both fit comfortably with Crawford, because neither asks a judge whether a statement seems believable. Forfeiture asks what the defendant did to the witness; the dying declaration asks whether the speaker was dying and knew it.
The Court has treated the two with unequal confidence. It has defined and applied forfeiture in Giles, but it has accepted dying declarations only tentatively, leaving open whether the Sixth Amendment incorporates them. The subsections below follow that order of firmness.
On an exam, an exception enters at the end of the analysis. Classify the statement as testimonial, confirm that the declarant does not testify, and check that the prosecution cannot show both unavailability and a prior opportunity to cross-examine. Only then ask whether the defendant’s own conduct or a settled founding-era practice supplies another ground for admission. An ordinary hearsay exception cannot do that work, because it is a rule of evidence and the Clause is a constitutional limit.
Forfeiture by Wrongdoing
A defendant who deliberately silences a witness cannot invoke the Confrontation Clause to keep that witness’s accusations from the jury, but the doctrine reaches only that kind of silencing. Giles held that forfeiture by wrongdoing admits testimonial hearsay only when the defendant engaged in conduct designed to prevent the witness from testifying. Merely causing a witness’s unavailability, even by killing the witness, does not suffice without that purpose. The Court reached the rule by asking whether the state’s broader theory was an exception established at the time of the founding, because Crawford read the Clause as carrying forward the common-law right together with only its common-law exceptions.2 Forfeiture is one of the two exceptions the Court found in that history, so the live question was how far the old doctrine extended. The rule has two working parts, wrongdoing that removes the witness and a purpose of preventing testimony behind it, and each rests on its own reason.
Equity supplies the first part of the rule. In Reynolds v. United States, the Court’s earliest treatment of the doctrine, a defendant whose own wrongful procurement kept a witness away could not complain when other evidence took the witness’s place, because no one is permitted to take advantage of his own wrong.3 Crawford accepted forfeiture on that footing, describing it as extinguishing confrontation claims on essentially equitable grounds and not as an alternative means of determining reliability.4 That basis explains why forfeiture survives a decision that discarded every reliability-based exception. A judge’s finding that a statement is trustworthy replaces cross-examination with a method the Constitution did not choose, while a finding that the defendant silenced the witness replaces nothing; it only denies the defendant the benefit of an absence he engineered.5 A rule covering conduct designed to silence witnesses also answers a practical danger, since without one defendants would have an intolerable incentive to bribe, intimidate, or even kill the witnesses against them.6
Giles shows where the doctrine stops. Dwayne Giles shot his former girlfriend, Brenda Avie, six times outside his grandmother’s garage, and at his murder trial he claimed self-defense.7 The prosecution offered what Avie had told a responding officer about three weeks before the shooting: that Giles had choked and punched her and threatened her with a knife. The trial court admitted those statements under a California hearsay provision for unavailable victims of inflicted or threatened injury.8 The California courts held that Giles had forfeited confrontation because he committed the very murder on trial, an intentional criminal act that made Avie unavailable.9 The Supreme Court vacated the judgment, because that theory asks only whether the defendant caused the absence. The state courts had treated his intent as irrelevant, a view the Court called error, though they remained free to consider evidence of intent on remand.10
The Court found the purpose requirement in the common-law sources themselves. Courts spoke of a witness “kept away” or “detained” by the defendant’s “means or procurement,” wording that points to a scheme to bring about the absence.11 How courts applied the rule is more telling. In the typical murder case, where the victim had made accusatory statements and nothing showed that the killing aimed at preventing testimony, courts excluded the statements unless they had been confronted or qualified as dying declarations.12 In King v. Woodcock, a beaten wife gave a magistrate a sworn account of the attack and died about two days later. The judge ruled that her statement could come in only if she made it believing death was near, so the test turned on her state of mind and not on her husband’s responsibility for her absence.13 American practice matched the English, since no court before 1985 invoked forfeiture outside deliberate witness tampering.14
The purpose requirement also does structural work that causation alone cannot do. If forfeiture followed whenever the defendant’s act caused the absence, the judge deciding whether to admit a murder victim’s statements would first have to find that the defendant committed the murder, and the jury would then hear the accuser’s account because the judge had already concluded that the defendant was probably guilty. The majority said that letting a judge strip a defendant of a trial right on the strength of a prior judicial assessment of guilt does not sit well with the right to trial by jury.15 Asking what the defendant meant to accomplish is a different question from asking whether the defendant is guilty, though the Court noted that a judge may still have to look at the charged offense where, for example, the defendant is on trial for murdering a witness to prevent testimony.16 The result was not unanimous. Justice Breyer, joined by Justices Stevens and Kennedy, would have found forfeiture on the Giles facts because the defendant’s own wrongdoing made the witness unavailable.17
Federal Rule of Evidence 804(b)(6) codifies the doctrine on the hearsay side, and the Court quoted it as applying only when the defendant “engaged or acquiesced in wrongdoing that was intended to, and did, procure the unavailability of the declarant as a witness.”18 The rule sets three things to prove: wrongdoing by the party or acquiescence in it, an intent behind that wrongdoing to make the declarant unavailable, and actual unavailability as the result. Acquiescence is enough, so the party need not personally commit the act that removes the witness, and the intent may be inferred from the wrongful activity itself.19 The hearsay rule reaches any statement of the declarant, whether or not another exception covers it.20 Because the constitutional exception demands the same intent, the hearsay and confrontation inquiries run on the same elements, and the Court found no authority for treating wrongdoing as forfeiting one objection while leaving the other.21
Two limits keep the doctrine in its place. Forfeiture matters only for statements that were never confronted, since a previously confronted statement of an unavailable witness comes in whoever caused the unavailability.22 And because the Clause governs only testimonial statements, a state hearsay rule is free to adopt a broader forfeiture rule for nontestimonial ones.23
Domestic violence prosecutions test the purpose requirement most often, and two decisions frame the issue. Davis rejected the argument that the Clause should bend for domestic violence cases, where victims are easily intimidated, and pointed to forfeiture as the answer when defendants procure or coerce silence; it left the forfeiture question in Hammon for the Indiana courts to take up on remand.24 Giles then showed that the purpose requirement is less of a barrier here than it first appears. Acts of domestic violence are often intended to dissuade a victim from seeking outside help, including conduct designed to prevent testimony to police or cooperation with a prosecution, so a murder that ends an abusive relationship may be found to express an intent to isolate the victim and stop her from reporting.25 Earlier abuse or threats meant to dissuade the victim, and pending proceedings in which the victim would be expected to testify, bear directly on that finding.26 Justice Souter, joined by Justice Ginsburg, added that the intent would ordinarily be inferable in the classic abusive relationship, which aims to cut the victim off from outside help, including law enforcement.27 A history of abuse proves little, though, unless the evidence ties it to a design to prevent reporting or testimony, and that tie is what the state courts had to look for on remand in Giles.
The doctrine’s procedure follows from its structure. Reynolds treated wrongful procurement as a question of fact to be settled as a preliminary matter before the substitute evidence is admitted, much like proof that a written instrument has been lost.28 Davis took no position on the standard of proof, but it observed that federal courts applying Rule 804(b)(6) have generally required the government to prove forfeiture by a preponderance of the evidence, that state courts tend to follow the same practice, and that the unavailable witness’s own out-of-court statements may be considered at the hearing.29 The judge holds the hearing outside the jury’s presence and, because the issue goes to admissibility, is not bound by the rules of evidence except those on privilege.30
Forfeiture should not be confused with judge-made waiver. Some courts had held that a defendant who creates a misleading impression opens the door to contrary evidence that would otherwise violate the Clause, and in Hemphill v. New York the Court declined to endorse that waiver.31 The two doctrines differ in what triggers them. Forfeiture turns on what the defendant did to the witness, a fact a court can find without weighing the evidence’s reliability, while an opened-door rule would make the judge decide when unconfronted evidence is needed to correct a distortion.
An essay stem that says the defendant killed the declarant does not end the analysis, because the usual error is to treat the killing as the whole test. The question is what the facts show the defendant meant the wrongdoing to accomplish. A killing that arose from jealousy or a dispute over money, with no link to a pending case or an expected report to police, leaves the testimonial statement excluded. A killing that follows threats about calling the police, a pending prosecution, or a pattern of isolating the victim supports an inference of purpose.
Dying Declarations
Crawford addressed dying declarations in a single footnote. The Court noted authority for admitting even dying declarations that are clearly testimonial, but it declined to decide whether the Sixth Amendment incorporates an exception for them, adding that if the exception must be accepted on historical grounds, it is sui generis, standing alone in a class of its own.34 That cautious wording governs how to read this subsection. The Court’s opinions recognize the exception, but the Court has never squarely held that it covers testimonial statements.
Giles later described the exception more concretely. It identified two forms of testimonial statement that the common law admitted without confrontation: declarations by a speaker who was both on the brink of death and aware of the dying, and statements admitted under forfeiture by wrongdoing.35 The first form has two conditions, and each does work. The speaker must be near death, and the speaker must know it. Awareness supplied the common law’s reason for trusting the statement, since a person who expects soon to face the Creator was thought to have every reason to speak truthfully. In Woodcock, the judge admitted the wife’s statement only if she “apprehended that she was in such a state of mortality as would inevitably oblige her soon to answer before her Maker for the truth or falsehood of her assertions.”36 Other courts excluded victims’ statements when the evidence that the speaker knew death was near was insufficient.37
An exception can survive a decision that rejected every reliability-based test because the two rest on different questions. Crawford read the Clause as preserving the common-law right together with the exceptions established when the Amendment was adopted, so the question for any proposed exception is what the common law admitted at the founding, not whether a judge finds the statement trustworthy.38 A dying declaration passes that inquiry because its admission rested on a historical rule with fixed conditions. The judge asks whether the speaker was dying and knew it, which differs from asking whether this particular accusation seems believable. The same history confines the exception. Whatever a state’s hearsay rule allows, the constitutional exception presumably reaches no further than the historical definition Giles set out.39
Giles itself shows the exception failing on its facts. The State offered what Brenda Avie had told a police officer about Dwayne Giles’s abuse about three weeks before he shot her, and Avie did not speak when she was dying, so her statements fell outside the dying-declaration exception.40 The State therefore leaned on forfeiture, arguing that Giles had forfeited confrontation by committing the murder on trial, and the Court held that forfeiture reaches only conduct designed to prevent the witness from testifying.41 The two doctrines fit together historically. Prosecutors in dying-declaration cases did not argue that the defendant’s guilt of the murder itself forfeited confrontation; instead, they went to the often considerable trouble of proving that the victim knew death was near.42 That awareness condition was a requirement prosecutors actually had to meet, and it separated the murder victim whose accusation could be heard from the one whose accusation could not.
Most dying declarations never need the exception, because most are nontestimonial, and the exception matters mainly when an official conducts a formal or structured interrogation of the dying speaker.43 In Michigan v. Bryant, a fatally wounded man named his shooter to officers in a gas station lot, and the Court held his answers nontestimonial because the circumstances objectively showed that the conversation was primarily aimed at quelling an ongoing emergency.44 The classification question therefore comes first, and the historical exception is reached only when the statement is testimonial.
Ohio v. Clark generalized the point beyond dying declarations. Because the Clause does not prohibit out-of-court statements that would have been admissible in a criminal case at the founding, the primary purpose test is a necessary, but not always sufficient, condition for excluding a statement.45 Two filters follow. A statement whose primary purpose was not testimonial never triggers the Clause, and a statement whose primary purpose was testimonial is excluded unless a founding-era rule admitted it. Clark also noted strong evidence that statements made in circumstances like those of the three-year-old declarant were regularly admitted at common law, though the Court decided the case by finding the statements nontestimonial.46
Justice Scalia, who wrote Crawford, concurred only in the judgment in Clark, and his objections state a separate view, not the Court’s position. He rejected the suggestion that the primary purpose test is merely one of several conditions, insisting that the Clause categorically entitles a defendant to be confronted with the witnesses against him and that the test only sorts out who is acting as a witness. He also objected to the burden he read into the majority opinion, under which a defendant must show that the founding generation understood the Clause to exclude what it regularly admitted. In his view, once a defendant shows that the State offers a testimonial statement without unavailability and a prior chance to cross-examine, the prosecutor must prove a long-established practice such as dying declarations. He suspected that the contrary dictum was a first step toward returning to the reliability approach of Ohio v. Roberts.47 Whether the carve-out stays confined to dying declarations or grows is a question the Court has left open.48
The exception’s own status is also unsettled. In her Bryant dissent, Justice Ginsburg said she would take up whether the exception for dying declarations survives the Court’s recent Confrontation Clause decisions if the issue were properly presented.49 Lower courts have continued to admit dying declarations: a Florida court held them admissible in 2016, observing that the majority rule still admits them, and the Supreme Court denied review.50 The Court appears likely to accept the exception, partly because Giles leaned so heavily on its existence when it rejected the broad forfeiture theory.51
On an exam, the dying declaration enters late in the analysis. After the statement is classified as testimonial and the prosecution cannot show unavailability plus a prior opportunity to cross-examine, ask whether the speaker was on the brink of death and aware of it. If the facts say yes, argue that the statement is admissible under the historical exception, and flag that the Court has recognized the exception without deciding that it covers testimonial statements. If the facts show a speaker who was not dying, the exception fails however damning the accusation, and the prosecution must turn to forfeiture, which requires conduct designed to prevent the testimony.