The Supreme Court's consolidated ruling in United States v. Davis and United States v. Hammon is a classic of the genre of consensus opinions to which the Roberts Court aspired in its first, transitional term. The opinion, authored by Justice Scalia, contains practical accommodations unusual in a decision by the Court's fiercest proponent of first principles. The restraint that characterized the term is, of course, more about considerations of logistics (including the desire to avoid re-arguments after the mid-term replacement of Justice O'Connor) than about the alignment of logic. Because it reflects temporary institutional constraints rather than intellectual agreement, the much-talked-of consensus, already less evident in the final decisions of the term, will decline further in the term to come. But Davis was considered and decided at a cultural moment when non-unanimous decisions were closely watched, and dissent was taken as a signal that the majority opinions somehow carried less precedential weight. Because compromise was not only the aspiration but also the expectation, Justice Scalia found himself engaged in strategic and incremental decision-making, in an attempt to count to nine instead of five. That attempt narrowed the class of cases in which confrontation rights will apply. Whether, in those cases, the right will have meaningful content depends in part on the Court's next steps with regard to the requirement of cross-examination itself.
Davis concerns the scope of the renewed right to confrontation announced in United States v. Crawford and the extent to which statements to law enforcement fit within the definition of "testimonial." The Court was unanimous in ruling in Davis (with Justice Thomas's concurrence in the judgment only) that a 911 call seeking emergency assistance did not produce the kind of testimonial statements with which the Confrontation Clause is concerned. And in Hammon the Court held, in an 8-1 decision from which Justice Thomas dissented, that statements to police officers investigating a crime on the scene are testimonial in the absence of an ongoing emergency. The test the Court announced focuses on the purpose of the questioning by law enforcement, viewed through an "objective" lens. Statements are nontestimonial, even when made in the course of police interrogation, if made "under circumstances objectively indicating that the primary purpose of the interrogation is to enable police assistance to meet an ongoing emergency." Such statements are "testimonial when the circumstances objectively indicate that there is no such ongoing emergency, and that the primary purpose of the interrogation is to establish or prove past events potentially relevant to later criminal prosecution."
This formulation of testimonial closely resembles the "emergency response" test advanced by the federal government in the briefing and argument. The test did not find much favor with the members of the Court in the course of the argument, and the decision thus serves as a reminder that reading the tarot cards of oral argument is never a safe bet. It also raises the question of how the Court's thinking evolved between oral argument and the decision, or for that matter between the categorical approach to testimonial taken in Crawford and the pragmatic working definition that emerged from Davis. The answer lies at least in part in Justice Scalia's desire to assemble the votes to definitively overrule Ohio v. Roberts, which most lower courts had continued to apply to nontestimonial statements post-Crawford.
The Court buried the lede, but Davis does pronounce Roberts dead. According to the opinion, testimonial statements are not merely the "core" but also the "perimeter" of the category of prior statements that implicate the Confrontation Clause. And, in a footnote and an aside at the end of the opinion, the Court uses the magic words and states that Crawford "overruled" Roberts, something only the two Crawford dissenters (Chief Justice Rehnquist and Justice O'Connor) had been willing to say at the time. The "unpardonable vice" of Roberts's reliability analysis was so antithetical to Justice Scalia's view of the Confrontation Clause that the opportunity to dispense with it may have made him compliant at the margins about the test articulated to take its place.
The Court had to grapple, as well, with the application of Crawford's uncompromising reasoning to the realities of messy human situations like domestic violence. The span of just two years that separates the promulgation of the testimonial standard from its testing in the context of 911 calls compresses the usual course of constitutional analysis. It is rarely the same Court, and often not even the same generation, that draws the lines when exigent situations call for after-market modifications of constitutional rights. The Court's effort here has to be considered a first pass. Rather than adopt any of the formulations that had evolved in the lower courts, the Supreme Court acknowledged a spectrum of nontestimonial emergency responses but did not identify a clear endpoint, such as the identification of the attacker or the arrival of the police on the scene. Important considerations cited by the Court include the nature of the information (whether it concerns past facts or ongoing activity), the necessity of law enforcement's questions, and the formality of the inquiry, but none of those factors is dispositive. The assailant in Davis was in retreat, but the Court still deemed his identification to the 911 operator nontestimonial. An unspecified amount of aftermath questioning will thus escape the testimonial label, and statements made in the course of protracted emergencies like kidnappings might as well. The opinion also fails to account for the possibility that law enforcement will simply manage investigations in order to maximize admissible statements. And it is too facile to suggest a clear distinction between statements about what "is happening" and what "has happened," when the latter is almost always necessary to explain the former.
Davis also narrows Crawford considerably, and perhaps inadvertently. Although interrogation itself was not defined, the Crawford Court stated that "[w]hatever else [testimonial] covers," it applies to "police interrogations." And the Court sounded a particularly categorical note about this aspect of the definition, repeating that statements from interrogations would be testimonial "under even a narrow standard" and that "interrogations by law enforcement officers fall squarely within that class." Davis adds an "except when," not by defining interrogation itself, but by actually contracting the definition of testimonial. The Davis Court says that it is determining "which police interrogations produce testimony" and adds that "in some circumstances" interrogations "tend to generate testimonial responses." Now, instead of the endpoint of the inquiry, interrogation is the baseline from which courts begin to examine whether a statement is testimonial. Crawford can be read, and Davis could have been decided, to hold that all accusations to government agents that a person committed a crime are testimonial. But now some accusations-those that are the product of unstructured interactions in emergency situations-fall outside that category.
Furthermore, Davis may burden defendants with demonstrating the absence of an emergency. George Fisher has described the Davis definition as an "only-purpose" test: To the extent that there is an exigent situation that interrogation helps somehow to resolve, law enforcement's "primary" purpose is unlikely to be proving past facts, or at least that is a very difficult argument to make. Davis expressly sets aside the question whether statements made outside of the law enforcement context can be testimonial, but this primary-purpose test is a model that could be exported. Its logic would seem to apply as well to statements whose primary purpose is to seek medical treatment, even where medical personnel are asking questions that also gather investigative details. Likewise, Davis could shorten Crawford's reach in another category of cases where it has had high impact: statements made by victims of child abuse to their parents. As Fisher points out, parents' desire to put an end to abuse and seek help for their child will always be a sufficient purpose to trigger questioning, so a child's statements in this context may well be deemed nontestimonial, even if they are relayed to authorities as well.
Chief Justice Roberts has publicly avowed his desire to achieve unanimity through narrow rulings offering "clarity and guidance." In Davis, the decision is narrow and consensus was largely achieved, but the end result is not particularly clear. Stating seven times in the opinion that it is an "objective" test does not make it so. And declaring that "testimonial statements are what they are" helps the lower courts not one whit when they actually have to determine where the lines are drawn. Ironically, the Davis formulation is a totality of the circumstances test in many ways more similar to the Roberts reliability inquiry than the bright lines of Crawford.
But setting aside the unanswered questions and unintended consequences, the Davis decision does craft a reasonable compromise between two competing extremes. The broad reading of Crawford would draw the line at accusations of any stripe; the narrowest reading is limited to "formalized testimonial materials, such as affidavits, depositions, prior testimony, or confessions." Justice Scalia had joined Justice Thomas's concurrence in White v. Illinois, advancing the latter definition. In Davis, however, Justice Scalia declared that "[r]estricting the Confrontation Clause to the precise forms against which it was originally directed is a recipe for its extinction." Of course, Justice Scalia often takes the view that the Constitution should be interpreted to mean precisely what it originally meant. But this is a case that splits the hair separating the consequence of Justice Scalia's "faint-hearted" originalism from the effect of Justice Thomas's dogged intentionalism. It is also a case in which Justice Scalia-after sniping at oral argument that "[m]aybe we should just suspend the Confrontation Clause in spousal abuse cases"-acknowledges in the opinion that domestic violence is "notoriously susceptible to intimidation or coercion of the victim to ensure that she does not testify at trial." The net result of those shifts is a purposive test. It has the virtue of focusing on the intention of law enforcement and therefore on the potential abuse of state power.
With that aversion to inquisitorial methods and concern about government manipulation of out-of-court statements in mind, the Court should now give content to the cross-examination right itself. Having exerted itself deciding when the Confrontation Clause applies, the Court has failed to notice that applying it does not mean much. Crawford and Davis both left unchanged the rule of California v. Green: The Confrontation Clause permits the use of prior testimonial statements where the declarant either appears for cross-examination at trial, or is unavailable to testify but has been subject to a prior opportunity for cross-examination by the defendant. In United States v. Owens, the Court held that a witness has "appeared" for cross-examination and vindicated the Confrontation Clause right when he has been placed on the stand and under oath, and has responded willingly to questions (i.e., not invoked a privilege or been held in contempt), even if he has no memory of the underlying facts, and minimal information about the circumstances surrounding the prior statement. In the lower courts, Owens has led to the conclusion that total denial of making a prior statement or of its substance still affords a constitutionally adequate basis for cross-examination; that memory lapse (feigned or real) has no bearing on availability for cross-examination; and that, as in Vaska v. Alaska, a witness who "could not shed any light on whether the incident about which the statement was made occurred, whether she made the statement, or the circumstances under which she made the statement" appeared for cross-examination within the meaning of the Confrontation Clause.
These decisions are often justified with the truism that the Constitution guarantees an opportunity to cross-examine but not successful cross-examination itself. But many cases make reference to the "opportunity for effective cross-examination," including Green and Owens. To be sure, effectiveness itself is not required, but the chance to question (even if it is an opportunity missed or inartfully pursued) is the constitutional baseline. And to afford such an opportunity, a witness must demonstrate what Robert Mosteller has called "minimal testimonial adequacy."
The Court's focus, however, has been on the hollow formalism of physical presence. Justice Scalia recently dropped a hint that he may take aim at another Confrontation Clause opinion with which he has long and vigorously disagreed: Maryland v. Craig. In Craig, a child abuse victim was permitted to testify by one-way closed-circuit television. Because the child was under oath, cross-examination was contemporaneous, and the jury could observe the witness's demeanor, the Court found that the purposes of confrontation were served. In the United States v. Gonzalez-Lopez decision this term (an otherwise unrelated case affirming a defendant's right to choose paid counsel), Justice Scalia reiterated his disagreement with Craig and underscored his victory over Roberts, criticizing, as Richard Friedman noted on his Confrontation Blog, the "unrestrained functional approach" of both cases. But Justice Scalia's single-minded attention to face-to-face confrontation has neglected the substance of adequate cross-examination. The "main and essential purpose," as the Court stated in Davis v. Alaska, "means more than being allowed to confront the witness physically."
Looking a defendant in the eye-or perhaps more importantly allowing the jury an opportunity to look a witness in the eye-is an important element of the confrontation right, but it is not sufficient to satisfy the requirement of cross-examination. If a witness were to walk into court, sit in the witness chair, and look the defendant straight in the eye while invoking the Fifth Amendment privilege not to testify, everyone would agree that the defendant had been deprived of her constitutional right to cross-examine. In cases of hybrid witnesses, neither wholly unavailable nor wholly available, who are physically present in the courtroom but remote, recanting, or recalcitrant, looking the witness in the eye does not accomplish much either, because there is nothing behind the eyes to confront. It is quite difficult to conduct an effective examination of a hostile witness who is no longer hostile, but has previously made statements to your detriment.
As the Court continues to circle around the meaning of testimonial, it should also ensure that testimonial statements are in turn subject to constitutionally adequate cross-examination. Richard Friedman has reproached the Court for its "longstanding unwillingness to recognize that a party may be substantially hindered in attempting to examine a witness with respect to a prior statement the truth of which the witness no longer affirms." If the Court is poised now to strengthen the requirement of face-to-face confrontation, it is incoherent to do so without ensuring that the defendant will obtain some answers to his questions in the process. The salient fact of Owens, ignored by the lower court extensions of the holding, is that the witness in that case had a clear memory of at least one thing: the prior identification of the defendant. The witness did not, as would have been preferable, remember the circumstances surrounding the statement, but that gave the testimony a self-impeaching quality. Owens is a sub-floor on which to build a meaningful but still workable requirement. While it does not exactly comport with Crawford's ideal of the "crucible of cross-examination," a witness has been confronted within the meaning of the Clause if she at least acknowledges making the prior statement, which would permit some infirmities to be probed and the hearsay to be impeached. The goal of confrontation, according to Crawford's reasoning and Davis's purposive test, is to expose any governmental coercion or manipulation. But total evasion with regard to the production of a prior testimonial statement leaves the hearsay freestanding and unchallenged; it does not discharge the defendant's right to cross-examination. If physical presence alone truly meets the standard then the reinvigorated Confrontation Clause has reach but no force.