Monday, March 28, 2011

Statements made in the absence of interrogation

Bryant concerns statements made in response to police interrogation. It remains clear, however, that interrogation is not a requirement for a statement to be considered testimonial in nature.

So how is the question of whether a statement is testimonial to be made if the statement is not testimonial? The principal theoretical issue that divided the majority and the dissent in Bryant – from whose perspective should the determination of whether a statement is testimonial? – does not arise if there is no interrogation: The choice must be made from the perspective of the speaker.

Bryant does appear to expand on Davis v. Washington by creating a general principle that “primary purpose” is determinative in all circumstances of whether a statement is testimonial. As I have written before, I think it would be better to speak of understanding or anticipation than of purpose, especially given that the test is supposed to be an objective one. Moreover, if determining the primary purpose requires a weighing of objectives – “Of these two (or more) purposes served by making the statement, which one would have been more important to a reasonable declarant in the position of the actual declarant?” – it is, as Justice Thomas wrote in his partial dissent in Davis v. Washington, “not reliably discernible,” and the test becomes inevitably “an exercise in discretion.”

But note how Justice Scalia handles “primary purpose” in Bryant. It is important to bear in mind that he was the author of both Crawford and Davis. In Bryant, of course, he was in dissent, but he was writing on a question that the majority does not address – how to determine the primary purpose of the declarant given that his purpose alone is determinative in the particular case. Scalia, of course, addressed that question because in his view only the declarant’s purpose is ever determinative. The majority did not address that question because in its view the purpose of the interrogator, if there is one, is also important, and indeed in Bryant the majority paid much more attention to the interrogator's perspective. But in a case without an interrogator, the majority will have to consider the speaker's perspective more carefully.

Justice Scalia wrote:

For an out-of-court statement to qualify as testimonial, the declarant must intend the statement to be a solemn declaration rather than an unconsidered or offhand remark; and he must make the statement with the understanding that it may be used to invoke the coercive machinery of the State against the accused.

That’s a useful, serviceable standard. It avoids the problems highlighted by Justice Thomas. And I believe it is consistent with Bryant. Courts should apply it.

Sunday, March 13, 2011

Due process confrontation rights

A California lawyer has written me raising an important issue on which I don't believe I've written. I'll only offer a few thoughts, because I'm sure others have thought more about this. The lawyer writes:

I am interested in any resources you know of which deal with the question of cross examination rights being denied in domestic violence, restraining order courts.

Here in California, we have courts hearing protective order applications in a "Jerry Springer" style, where in the majority of cases, one litigant, probably both, are not represented by counsel, and the judge takes a statement of accusation from the applicant, then asks the respondent about it, and goes back and forth until announcing a decision.

It seems this procedure, if it can even be called that, is commonplace in most California counties and probably around the country.

A common problem in all this is that, even if the accused person knows of questions which would greatly impeach the accuser, he is not asked if he has any questions, and often, as in the case of my appeal, if he asks the Court to ask questions, is simply blown off at the Court's whim, without any reason being given, unless of course he has an attorney.

In other words, the right to confrontation is often denied to the accused person. This is so even despite a California Supreme Court case, Elkins, which restates that litigants in family court, (where these order are issued under CA Family Code 6200) retain all procedural rights that other civil litigants have.

I am hoping you have knowledge of academics, judicial councils, state Bar associations, etc, who have expressed concern that basic due process rights, as I say confrontation being primary, are denied in these hearings.

Please let me know of anything that will help.

So, first off, the Confrontation Clause as such does not apply -- it only applies to criminal prosecutions. But in other contexts, a right of confrontation is incorporated in the right of due process. The key case is Morrissey v. Brewer, 408 U. S. 471 (1972), which involved revocation of parole. It seems to me that the situation posed by the reader – a hearing with significant consequences for the litigant, witnesses testifying from the stand, and the litigant not given an opportunity to pose any cross-examination – presents a strong case for holding that the due-process right to confrontation has been violated. The basic concept that if a witness testifies for one side the other side gets to cross-examine that witness is deeply established in Anglo-American jurisprudence.

The interesting problem, it seems to me, will arise if the state, in a non-criminal context, decides not to present a live witness because it is aware of the due-process confrontation right, and instead presents, say, a video-tape of a statement taken beforehand for use in the hearing. Effectively, the witness has been able to testify without coming to court; the idea that statements made out of court with litigation in mind may be deemed to be testimonial, and the focus of the confrontation right, lies at the heart of Crawford, and has force in this context as well. But presumably a narrower view of what should be considered testimonial, or a broader set of exceptions, or a looser sense of what the right means, or some combination of the three, must apply in this context than in the criminal context. If the approach the Supreme Court used in the Confrontation Clause context in Michigan v. Bryant had instead been applied in other contexts such as hearings for restraining orders or parole revocation, it wouldn't have been nearly so bad.

Friday, March 11, 2011

Doesn't the Confrontation Clause apply in Kansas?

Gee, I realize that some prosecutors are holding out hope that Melendez-Diaz will be confined or even overruled in Bullcoming, but it is the law, and I would have thought that by now word would have reached Kansas. But John Collins, director of the Michigan State Police Laboratory, has passed on this article from the Wichita Eagle, Bill lets forensic experts report in writing, skip testifying. Yoo hoo! You can't do that.

Linda Greenhouse on Scalia and Bryant

Linda Greenhouse of Yale Law School, and formerly the long-time Supreme Court correspondent of the New York Times, has written an interesting piece on Justice Scalia and Bryant. You can read it by clicking here.

Thursday, March 10, 2011

Confrontation, the Journalist's Privilege, and Harmless Error

Focus on Crawford-related issues should not obscure the fact that many significant Confrontation Clause questions concern the scope of the accused’s right to cross-examine a witness who actually testifies at trial. Adam Liptak of the New York Times has alerted me to an interesting decision issued yesterday by the Second Circuit in United States v. Treacy.

Treacy is a former president of the company that operates the job-hunting website Monster.com. He was convicted of securities fraud, the theory being that he participated in the back-dating of stock options. One of the witnesses against him was a Wall Street Journal reporter, Charles Forelle, who was subpoenaed by the Government to testify to the proposition that Treacy had made statements attributed to him in an article co-written by Forelle. Probably because the Second Circuit accords a journalist a qualified privilege protecting him from the compelled disclosure of even non-confidential materials, the district judge, while denying Forelle’s motion to quash the subpoena, tightly circumscribed both direct and cross-examination. For example, the court prevented defense counsel from asking Forelle about a follow up e-mail that he had sent to the company’s public relations representative and that arguably supported the defense’s version of the conversation between Forelle and Treacy. The Second Circuit held that the limitations on cross- examination were improper. It declared that

once a trial court has determined that the Government has made the required showing to overcome the journalists privilege and compel a reporter’s direct testimony, the trial court may not, consistent with the Sixth Amendment’s Confrontation Clause, thereafter employ the privilege to restrict the defendant’s cross-examination of the reporter to a greater degree than it would restrict such cross-examination in a case where no privilege was at issue.


That seems correct, or at least close to correct, to me. The case to bear in mind is Davis v. Alaska, 415 U.S. 308 (1974). There, the state had a general rule preventing the disclosure of a juvenile adjudication, but the Supreme Court held that this rule must give way to the confrontation right where the young witness’s history of trouble with the law gave him a possible motivation to point the finger for a theft to the accused.

In Treacy, the jurisdiction has a rule limiting the ability of a party to call on a journalist for testimony. But once a court decides that the constraints of this rule may be overcome in the particular case on behalf of the prosecution, the accused must be allowed robust cross-examination. Perhaps at the margin, the policy behind the qualified privilege may guide the trial judge’s discretion in limiting cross-examination – this is why I am not certain that the Second Circuit is correct in saying that the privilege washes out altogether – but in this case the cross-examination seemed a natural follow-up to the direct and should have been allowed.

The Second Circuit held that the error was harmless. In assessing harmlessness of a Confrontation Clause violation, it is important that the court not speculate as to how effective cross-examination would have been if it were allowed. Delaware v. Van Arsdall, 475 U.S. 673 (1986). The simplest way to do this is to examine the case on the hypothetical assumption that the direct testimony had never been introduced. Ultimately, this is what the Second Circuit did, concluding (whether accurately or not I don’t know) that “it is highly unlikely that [the Government] would have been unable to secure a conviction in the absence of Forelle’s testimony.”

Monday, March 07, 2011

Child Observers and Bryant — a GVR

This morning, the Supreme Court GVRed — granted, vacated, and remanded — the case of Allshouse v. Pennsylvania, No. 09-1396, for reconsideration in light of last week's decision in Michigan v. Bryant. Allshouse involves statements by a four-year-old to a child protection agency worker investigating allegations of abuse. The petition, seeking review of Commonwealth v. Allshouse, 984 A.2d 847 (Pa. 2009)(unanimous court, one justice not participating; majority opinion, two justices concurring, and two concurring partially and in the result), was brought by Jeff Fisher; an amicus brief in support of it was filed by the National Association of Criminal Defense Lawyers. Interestingly, Pennsylvania acknowledged in its Brief in Opposition that the case was worthy of certiorari — so the very short reply brief in support of the petition argued that cert should be granted immediately rather than after the decision in Bryant. But not surprisingly, the Court held the case pending Bryant, and now it has signaled, wisely I think, that before stepping into the very difficult area of children's statements it wants to see how Bryant plays out in the lower courts.

A few reactions:

First, the GVR is one more indication that the impact of Bryant may be less dramatic than some observers have thought. If the Court thought it was obvious given Bryant that, as the Pennsylvania Supreme Court held, the statements were not testimonial, then it could simply have denied cert. Also, if it thought that Bullcoming might have any bearing on the outcome of this case — say, by narrowing the meaning of "testimonial" — it could have just held this case pending its decision in Bullcoming.

Second, statements by children are one of the few contexts in which taking a questioner's perspective into account makes it more likely that a court will deem the statement to be testimonial; the child presumably does not know the legal consequences of the statement, but the questioner certainly does.

Third, having said that, I suppose that the greater number of lower courts, and perhaps eventually the Supreme Court, will hold statements of this sort to be non-testimonial, on the grounds that the child did not anticipate prosecutorial use (it will be interesting to see whether the courts emphasize the speaker's perspective in this context more than the Supreme Corut did in Bryant!) and that the social worker was focused on therapeutic and protective goals rather than on gathering evidence — no matter how frequently the given social worker and her colleagues have just happened to find themselves testifying for prosecutors about statements made by children in similar situations. If I am right in this supposition, it demonstrates the manipulability and inadequacy of the Bryant approach.

Fourth, I am pleased to say that I have begun work on an article in this area with Steve Ceci, one of the world's leading developmental psychologists. I anticipate we will call it The Child Quasi-Witness. Here are the main points I hope we make:

1. Some very young children, even though capable of purposive communication, should not be deemed capable of being witnesses for purposes of the Confrontation Clause. Their statements, even though made in contexts that would lead to the statements being characterized as testimonial if they were made by an adult, should therefore be deemed beyond the scope of the Confrontation Clause. I hope to offer several perspectives on the standards courts might use in determining whether a child should be deemed capable of being a witness for Confrontation Clause purposes.

2. Even if a child is not capable of being a witness for Confrontation Clause purposes, her statements may have substantial probative value.

3. If a child is not capable of being a witness for Confrontation Clause purposes but a prosecutor offers her statement against an accused, then the accused should have a right — both as a matter of ordinary procedural law and as a matter of due process — to examine the child, as he would an inanimate object that is the source of crucial evidence against him. But this examination would not be by personal confrontation, under oath and subject to cross-examination in open court. Rather, it would be by a qualified expert — presumably a child psychologist — in a controlled environment.

4. This procedure is far better for truth determination than is cross-examination in open court of a very young child, and offers the accused a better opportunity to explore weaknesses in the child's account.

Defense counsel might consider making an argument along these lines as an alternative to an objection to a child's statements based on the Confrontation Clause; I think courts will find the relief called for by this argument much more appealing than a holing that the child must be subject to confrontation in open court.

Meanwhile, here is a set of old posts bearing on statements by children; some of these anticipate these arguments.


Children and forensic interviews, revisited
, Jan. 7, 2008

Further developments and thoughts on child witnesses, Oct. 26, 2007


Child Witnesses on the Academic and Judicial Front
, Sept. 7, 2007

An interesting sidelight on Crawford and Craig, June 26, 2006


R. v. Brasier — a classic case from 1779
, Dec. 24, 2005

Maryland's Highest Court on "Tender Years" Statements, Feb. 7, 2005.

Sunday, March 06, 2011

Paul Vinegrad on "Government Coercion & The Confrontation Clause: Getting To The Heart of The Matter"

Blog reader (and retired prosecutor) Paul Vinegrad has asked me to put the following post on the blog. I’m happy to do it, though as usual I disagree with just about everything Paul has to say, creative though it is. Paul’s post begins after the first row of x's and after the second row I present my own comments in response.

Any reader who has a confrontation-related message too long for the comments section of the blog is welcome to submit it to me; so long as it is appropriate for the blog, I I will post it.

x x x x x x

I take the position that only hearsay that is created by sufficiently coercive government conduct is barred by the Confrontation Clause if the accused has not had an opportunity to cross-examine the declarant.

The following hypothetical illuminates my argument:

Gang warfare is rampant in South Central LA. The Bloods are shooting at the Crips. And the Crips are returning fire. Innocent bystanders have been maimed and killed in the ensuing battle, struck by a hail of bullets fired from automatic weapons stuck out of car windows by unidentified perpetrators.

Mr. V and his family, unfortunately, have to live and work in this gang infested area.

One evening, Mr. V is exiting his business – a local grocery market – with his wife. Crip gang members are in the parking lot. A car pulls up. Everyone in the car is dressed in blue attire. One of the occupants sticks a Mac-10 automatic weapon out of the window and opens fire on the Crips. The car speeds off. Mr. V is struck by several rounds from the Mac-10. He sustains very painful, but not life-threatening, wounds to his legs and arms. Mr. V recognizes the shooter from the neighborhood as Psycho Dan.

Mr. V makes the following statements after being shot:

(1) Upon being struck he immediately blurts out "Psycho Dan why did you do this!" This statement is heard by a bystander;

(2) His wife runs to his side. She asks him if he is ok. He says "yes." She then asks him who shot him? He says "Psycho Dan. But don't tell anyone! He will kill our family if he finds out that I was a snitch!"; and

(3) Several beat cops who were in the area, heard the shots, and are aware of the ongoing gang war and its toll on innocent people caught in the cross-fire, arrive. They reasonably believe the shooting was part of the ongoing gang war. They approach Mr. V. They ask him "Who shot you?" Mr. V says "I can't tell you. They will kill my family. Please get the paramedics!" The cops tell Mr. V that he will live, his wounds are not life-threatening, whoever did this is very dangerous and may kill innocent women and children as he moves through the area trying to shoot rival gang members. They plead with Mr. V to help them protect the public from this menace. They plead with him to tell them "Who did this?" Mr. V says "I can't tell you. They will kill my family. Please get the paramedics!" One of the cops then tells Mr. V "If you don't tell us who did this, we won't call the paramedics" Mr. V, in excruciating pain, says "Psycho Dan. Please don't tell anyone that I told you! He will kill my family!"

The prosecutor calls Mr. V to testify at Psycho Dan's trial. Mr. V, because of fear that his family will be hurt, refuses to testify. He is held in contempt by the judge. The prosecution then seeks to introduce each of Mr. V's statements identifying Psycho Dan. The defense objects on Confrontation Clause grounds.

It is my opinion that the judge should rule as follows:

(1) The blurted out identification is not barred by the CC. It was, in the words of Bryant, "reflexive." It was not made with any particular "purpose," or even anticipation of subsequent use by the government to investigate or prosecute the perpetrator. It was, to use Prof. Friedman's example, analogous to a drug sniffing dog barking when it smelled cocaine. Most importantly, there was no conduct by any state actor that "created" this statement. Despite Justice Scalia's suggestion in Bryant, and during argument in Bullcoming, that the CC would be applicable to "blurted out" statements (and neither "interrogation," nor any questioning, is required), I believe the overwhelming majority of the Court will reject his position, at least with respect to unsworn statements. And, ultimately, will hold that, absent any conduct by state actors, unsworn hearsay is outside the scope of the CC.

(2) The identification made to his wife (a private citizen) in response to her pointed question – a question (when viewed objectively) that was unmistakably designed to identify the shooter – is also outside the scope of the CC. Unlike the blurted out statement, and viewed objectively, it is reasonable to conclude that this interaction between Mr. V and his wife, and the resulting statement, had some purpose. Clearly, the "primary purpose" of the question and the answer was to identify the shooter, and not to address any "emergency," i.e., threat to the victim or any other member of the public. However, once again, no state actor participated in the "creation" of this statement. Absent that element, the CC is inapplicable to this unsworn statement. The Court, as it has done in the context of the Fourth and Fifth Amendments, will ultimately hold that state action is required before the CC can apply. Just as state action is required for the Fifth Amendment's prohibition on "compelling" a person to be a "witness" against himself in a criminal trail to apply, I believe, ultimately, the Court will hold that the Sixth Amendment's CC incorporates the same requirement.

(3) The identification made to the cops is barred by the CC, despite the fact that, objectively viewed, the "primary purpose" of the "interrogation" was to address an "ongoing emergency" – a more real and pressing "ongoing emergency" than existed in Bryant. The CC applies to (and bars admission of) this statement because of the coercive conduct by the cops in extracting the words from the declarant's mouth. I believe that it is this element of coercion by state actors that is at the heart of the CC.

Coercion by the government was utilized against Cobham in the form of "interrogation," while in custody, and accused of being a co-conspirator. It was used, albeit to a lesser extent, against Sylvia Crawford as she was "interrogated" in custody at the police station after being advised of her Miranda rights. It was used, to an even lesser extent (but still present), by the government against Amy Hammon as the police "interrogated" her, one (or two)-on-one, face-to-face, in a confined location not open to the public, and after she initially stated that "nothing happened."

The government created coercive conditions in these situations placed varying degrees of pressure (in the words of Fifth Amendment jurisprudence, "compulsion") on the declarants to tell the government actors what they believed they wanted to hear. Whether it was that Raleigh was conspiring against the King, Mr. Crawford was lying when he asserted self-defense, or Mr. Hammon did in fact beat Amy, despite her initial denial. In each instance the government coercion created a motive for the declarant to fabricate a story in order to gain some perceived benefit. Cobham's shifting of blame to Raleigh and providing details of the alleged conspiracy would, from the reasonable perspective of someone in his position, inure to his benefit. The same is true in the case of Sylvia Crawford and Amy Hammon – viewed objectively, each could have felt it would be in their interest to tell the police what they believed the police wanted to hear (regardless of the veracity of their statement) in order to end any "pressure" that the police created and were exerting.

This element of state created coercion was not present in Davis where there was physical separation between the government and the declarant. It is hard to pressure someone on the telephone to say what you want when they can simply hang up. The lack of resemblance between Davis and Raleigh's Trial really has nothing to do with the presence of an "ongoing emergency" in the former and no such emergency in later case. (In this regard I agree with Justice Scalia that one can make a reasonable argument that the "emergency" in the case of Raleigh – the potential overthrow of the King – was just as real and significant, if not more so, than that in Bryant, yet the circumstances under which Cobham's accusatory statements were created by the Crown were the paradigmatic CC violation.) Rather, despite the presence or absence of an "emergency" (ongoing or otherwise), the critical factor for purpose of determining the applicability and scope of the CC is whether the statements were uttered in response to a government created coercive situation.

Sufficient coercion did not exist in Bryant, given the very limited nature of the questioning by the cops, the public setting, and the absence of any external "pressure" by the police to get Covington to speak, i.e., to tell them what a reasonable person in his position would believe they wanted to hear. Nor does any coercion remotely exist in Bullcoming, where the declarant is simply asked, via a written request, to test a sample of blood and report back what, if any, alcohol is present. The analyst Caylor in Bullcoming was under no "pressure" by anyone to have the gas chromatograph generate a particular result. As Justice Alito alluded to in the Bullcoming argument, analyst Caylor's livelihood and/or promotions did not depend upon the results of testing that he performed.

Government created coercion/pressure does, however, permeate the statement by Mr. V identifying Psycho Dan to the cops. The fact that the cops were confronted with a very real "ongoing emergency," of which Mr. V was acutely aware, is irrelevant to the determination of the CC's applicability.

Coercion – not allowing any hearsay that is tainted by the possibility of government pressure – is at the heart and soul of the CC. It is the presence or absence of that coercion (when viewed objectively) that determines whether the "resemblance test" (approvingly cited by 6 justices in Bryant) has or has not been met.

x x x x x x x x

There are several problems with the coercion test proposed by Paul.

Perhaps the most obvious one is that it would be extremely difficult to apply, and so highly subjective I think it would provide very little protection.

Second, it has absolutely no grounding historically. We have to bear in mind that the confrontation right long pre-exists the institutions of a police force or a public prosecutor; most prosecution until the late eighteenth century was by private persons. The confrontation right would have been virtually empty if it only applied to instances of coercion by public officials.

Third, there isn’t even a whisper of such a limitation in the text of the Confrontation Clause.

Fourth, such a rule would yield very odd results. A person who observes a crime could write an affidavit, perhaps with the assistance of a private NGO, and send it to a prosecutor, and the confrontation right would never come into play.

Fifth, the Supreme Court has already rejected such a rule. As the last point suggests, a coercion rule is a more restrictive version of a test limiting the confrontation right to interrogations, and the Court has explicitly said that the Confrontation Clause is not so limited.

So far as the state action requirement is concerned, it is satisfied by the fact that the state is trying and presumably convicting an accused in part on the basis of testimony that it introduces without offering the accused an adequate opportunity for confrontation.

Thursday, March 03, 2011

Impressions of the Bullcoming argument

I attended Wednesday’s argument in Bullcoming v. New Mexico. Recall the facts in brief: The analyst who performed a test on Bullcoming’s blood was on unpaid leave at the time of trial. Instead of presenting his live testimony at trial, the state presented his report through a supervisor from the lab who had not observed performance of the test. Here are some impressions of the argument.

First, the reports of the death of the Confrontation Clause in light of Michigan v. Bryant appear to be greatly exaggerated. The New Mexico Attorney General, Gary King, cited Bryant in the very first sentence of his argument, and referred to it a couple of times later, though he actually tried to distance himself from the case by saying that the purpose test of Bryant did not apply absent an interrogation. No member of the Court seemed to be interested in any possible ramifications of Bryant for this case. Nor did any member of the Court suggest that the report ought to be admissible because it supported an expert opinion offered by the testifying supervisor, as countenanced by Fed. R. Evid. 703. Indeed, King emphasized the lack of human analysis in the report. Nor was there any suggestion that I picked up that any justices were considering overruling Melendez-Diaz.

Second, some of the justices continue to be troubled by the multiple-witness problem. This came up several times in various contexts. It is difficult to know whether they were persuaded by the fact that many states have long followed the rule that if a lab analyst’s report is admitted that analyst must testify in court, absent stipulation, and these states have not faced intolerable difficulties. But I think that enough empirical evidence has now been gathered demonstrating this point that it will be difficult for them to write an opinion predicting disaster if Bullcoming wins.

Third, much of the questioning focused on factual settings not quite like the one presented by the case. The nature of the questioning appeared to be an attempt to determine how states might cope efficiently with a ruling for Bullcoming. In the actual case, the prosecution introduced a report by the absent analyst, which is why this should be, as Jeff Fisher, arguing for Bullcoming, asserted at the outset, an easy case; an exchange between Jeff and Justice Alito clarified that the report asserted the identity of the sample tested, the lack of tampering, the procedures used, and the result. But what if the state had not introduced the report? Some of the justices were interested in the possibility that a witness could testify to a lab’s general procedures, from which the jury could infer that the procedures were followed in the case at hand. Jeff agreed that the Confrontation Clause would not pose an obstacle to doing that. But the problem would still remain of proving the results in the particular case. If, for example, one analyst enters the defendant’s name in a machine and the machine performs the test and spits out a report with the defendant’s name on it, there would still be a testimonial assertion by that analyst that this particular report is the result of a test performed on that defendant’s blood; I don’t believe a second analyst should be allowed to testify, in the absence of the first, that the practice of the lab is to enter accurately the name of the source of the blood in the machine and then perform the test.

But in any event, as I’ve said, that case is not the one before the Court, in which the prosecution clearly introduced a full report by the absent analyst. One possibility is that the Court will simply resolve the case before it, saying that this was a violation, and leave to future cases the development of the bounds of acceptable procedures.

Finally, Justice Kagan was essentially silent. She apparently began to ask one question but was interrupted and did not persist. So that makes any prediction even more speculative, and I won’t speculate.

Wednesday, March 02, 2011

Preliminary thoughts on the Bryant decision

Here are some preliminary observations on the opinions in Michigan v. Bryant. First, a brief review of the facts.

Responding to a 911 call, police found Anthony Covington on the ground near a service station in Detroit, profusely bleeding from a gunshot wound. As each officer arrived, he asked Covington what happened. Covington said that a man named Rick – the accused, Bryant – had shot him through a door at Bryant’s home, about six blocks away and 25 minutes earlier; Covington had managed to drive himself to the spot where he was found. Covington died several hours later of his wounds. Bryant was eventually extradited from California and tried for murder. The first jury hung, and a second one convicted him. Ultimately, the Michigan Supreme Court held that admission of Covington’s statements violated the Confrontation Clause. The United States Supreme Court has now reversed that decision, in an opinion by Justice Sotomayor. Justice Thomas concurred in the result (on the ground, expected given his prior expressions, that the statements were not formal). Justices Scalia and Ginsburg wrote separate dissents.

Now, two over-arching thoughts:

(1) I believe the decision is a very unfortunate development for the Confrontation Clause. The approach that emerges is remarkably mushy, unjustified by any sound reasoning and virtually incoherent. It leaves courts ample room in many types of cases to characterize almost any type of statement as non-testimonial. It will be easily manipulable by governmental authorities and at times may distort their behavior.

(2) I believe this decision is in large part a result of the Supreme Court’s error in unduly restricting the scope of forfeiture doctrine in Giles v. California. In this case, there was substantial evidence, easily enough to justify a finding, that Bryant had killed Covington and therefore that he himself was at least the initial cause of his inability to cross-examine Covington. Given Covington’s condition, his death within hours, and Bryant’s flight, it does not appear that there was a reasonable opportunity to mitigate the problem by arranging a deposition. Accordingly, a court easily could have held that Bryant forfeited the confrontation right – had Giles not foreclosed the possibility by holding that even a defendant who murders a witness forfeits the right only if he commits the murder for the purpose of rendering the witness unavailable. The bottom-line result of the Michigan Supreme Court’s decision – that Covington’s statements were inadmissible – is singularly unappealing at a gut level, and I think it was inevitable that courts would compensate for the unavailability of forfeiture in cases like this by narrowing the confrontation right. See my post of June 29, 2008, Reflections on Giles, Part 2: Is Giles bad for defendants?

Some more particularized comments:

For the first time, the Court has purported to give a broad, general approach to determining what is testimonial. It takes the “primary purpose” language of Davis and expands on it. Now, “primary purpose” is not simply a test to choose between whether a statement is testimonial or instead made in response to an “ongoing emergency”. Rather, it appears, the accused now has the burden of establishing that the primary purpose of the conversation in which the witness’s statements were made was to “creat[e] an out-of-court substitute for trial testimony.” Determining primary purpose is, of course, a very tricky matter, as Justice Thomas emphasized in his dissent in Davis and repeats here. Indeed, the concept is very difficult to nail down; if a person makes a statement for the purpose of creating trial testimony and also for another purpose, how do we determine which one is the primary purpose? By which one she would be more willing to give up if she had to choose? But beyond that, let’s say that we are able to analyze the situation precisely and conclude that 48% of the person’s purpose was testimonial and 52% not (sounds silly, I know, but it’s not my idea to determine which is primary); why is that not testimonial? I’ve argued elsewhere, and won’t repeat the argument here, that the real question is not purpose but anticipation. Justice Scalia’s dissent adopts this view. He tips his hat to his language in Davis by saying that for a statement to be testimonial the speaker has to intend to make a solemn declaration, but the test that carries the bite for him is whether the speaker has “the understanding that it may be used to invoke the coercive machinery of the State against the accused.”

The more serious problem is that the Court adopts what it calls “a combined approach” in determining the perspective from which the question of whether the statement is testimonial should be determined. That is, it looks to the purpose of both the speaker and the interrogator (if there is one). Now, I don’t think there is any real objection to taking all circumstances into account in determining whether a statement is testimonial, and if there is a questioner the purpose for which the questioning is conducted is certainly a factor that might affect the understanding of the speaker. But that is a different matter from what the majority does, which is to make the test an indeterminate combination of what did the speaker intend and what did the questioner intend: A court is supposed to determine “the purpose that reasonable participants would have had.”

That the proper perspective is that of the speaker – the lack of a testimonial purpose on the part of a questioner should not make a statement non-testimonial if the speaker knows she is creating evidence, and a questioner’s hope to gather evidence does not make another person’s statement testimonial (though in some cases of trickery an estoppel should operate) – is another point that I have argued repeatedly, and I won’t go through the matter again here. (For those who are interested: The majority cites my amicus brief as making an argument that it rejects, and Justice Scalia cites one of my articles, Grappling with the Meaning of "Testimonial", 71 Brook. L. Rev. 251 (2005), in agreeing with the point.)

But note two problems with the majority’s test. First, because the Court properly adopts an objective test, it cannot ask what the actual participants intended. It has to put it in terms of reasonable participants. But because it has phrased the test in terms of purpose rather than understanding, it has to ask a baffling question: What purpose would reasonable participants have had? The problem is that purpose is a matter of desire, not simply understanding, and equally reasonable people might have different desires in a given situation. So I guess what a court has to do is posit a reasonable person in the position of a participant and, drawing on all the facts of the incident, take a stab at guessing what the primary purpose of most reasonable people in that situation would be. That’s pretty awkward.

More seriously, what on earth could the majority mean by the combined purpose of “the interrogation”? Justice Sotomayor says that simpler is not always better, and that the Court is “unwilling to sacrifice accuracy for simplicity.” But accuracy in determining what? It appears to me that the majority has adopted a standard that is not only the wrong one but is really empty. As Justice Scalia points out, there is a glaringly obvious problem that the majority never touches on – what if the speaker has one purpose and the questioner another? That of course is entirely plausible, especially in cases like this one in which the statements are made right after police officers come up to a scene: It may be that the officers have little idea what is going on, whether a crime has been committed or whether there is a grave threat to the public safety, and yet the speaker knows perfectly well that a crime has been committed, that there is no imminent threat to the public safety, and that what she is about to say will provide information useful in bringing a criminal to justice but will have little value other than that.

So in a circumstance like that, what is a court supposed to do, given the Bryant decision. The majority denies that it “intend[s] to give controlling weight to the ‘intentions of the police,'” and that is good. But look what they did in this case. They emphasized at length the uncertainties and potential dangers facing the officers as they arrived at the scene. I don’t think that even so they are able to make a plausible argument that their “primary purpose” was to resolve an ongoing emergency – the officers were doing a lot of evidence gathering and not a whole lot of public protection at that point – but this is their best shot. As for the speaker, Covington, all the Court does is emphasize his dire situation at the time and pronounce conclusorily that it “cannot say that a person in Covington’s situation would have had a primary purpose to establish or prove past events potentially relevant to later criminal prosecution.” Well, why not? Badly as Covington was hurt, he was coherent in speech and seems to have been thinking straight. He was giving information describing the background and commission of a serious firearms assault. Presumably he was strongly motivated to see that the person who had shot him was brought to justice. And what else could he have been attempting to accomplish? The majority does not even offer an alternative suggestion.

So one of my concerns is that police officers will quickly learn that they can get statements characterized as non-testimonial if they testify, in effect, “I came up to the scene and didn’t know what was happening. My principal concern was securing the public safety. What this person told me was very important for that purpose.” They will also have an incentive to gather as much information as possible before the situation is fully under control; thus, the decision in this case distorts their incentives in performing their policing function. And once they do control the situation, if they can no longer make a credible contention that they had some primary purpose other than evidence gathering, then they can pass the witness – so I may continue to call the person who makes a statement while understanding its likely future prosecutorial use – on to a social worker, whose "primary purpose," of course, will be therapeutic, notwithstanding the fact that in performing that function she repeatedly learns, and relays to juries, information that turns out to be useful in prosecuting crime.

Perhaps in emphasizing some of the problems in this decision I am being unduly pessimistic. Perhaps the Court’s emphasis on the context-dependence of the determination of whether a statement is testimonial will ultimately limit the impact of the decision. Perhaps, indeed, this decision will be understood over time to be a product of the unfortunate inability of the Court to apply forfeiture doctrine to this case. Perhaps the Court’s insistence that the speaker’s intent is one consideration will become more important over time, and perhaps the Court will recognize what a giant opening for manipulation and distortion it has left by making the purpose of the questioner, when there is one, a significant factor.

What is more, there are a few good points in the decision. Establishment of an objective test was not a surprise, but the clarification is useful. The Court explicitly rejects the notion of an absolute formality requirement, even though it gives formality considerable force. It continues to preserve the possibility that statements made to persons other than government officials might be considered testimonial in some circumstances.

So we’ll have to see over time. But this decision strikes me as a giant step backwards towards a morass like that of Ohio v. Roberts, which gave the courts considerable leeway to let almost any statement in. Indeed, I have not even begun to address the jarring suggestions, which read at their most dangerous seem to augur a return of Roberts, that a factor in determining the Confrontation Clause issue may be the reliability of the statement and that hearsay law may be a guidepost in that determination. More on that later.