Further developments and thoughts on child witnesses

Yesterday, the Ohio Supreme Court decided State v. Siler. In my view, this is a case of considerable interest, significance, and difficulty.

The facts are gruesome and hard to stomach. Brian Siler is charged with murdering his estranged wife Barbara, who was found hanging by a cord in the garage. Medical evidence indicated that she had been choked to death from behind and then hanged. Among the evidence against Siler was the testimony of Detective Larry Martin, a plain-clothes police officer and trained child interviewer who was called to the scene, as to statements the couple’s 3-year-old son Nathan made soon after the body was discovered. According to Martin, Nathan said that his father had scared him the night before by banging loudly on the front door, that his parents had argued loudly in the garage, that his father had hurt his mother, by grabbing her from behind above the shoulders, and that"the yellow thing" that was holding his mother up was put on his mother by his father.

In a pre-Crawford trial, these statements were admitted as – surprise – excited utterances, though apparently Nathan did not exhibit distress until well into the interview. The appellate court, pre-Crawford, affirmed, and the state supreme court, soon after Crawford, denied review; even on a motion for reconsideration, highlighting Crawford, the state supreme court declined review. But then the U.S. Supreme Court – in a decision noted in the very first posting on this blog – vacated and remanded for reconsideration in light of Crawford. Siler v. Ohio, 543 1019 (2004). On remand, the appellate court held unanimously that Nathan’s statements were testimonial and that admitting them in the absence of an opportunity for cross-examination violated the confrontation right. State v. Siler, 843 N.E.2d 863 (Ohio App. 5th Dist. 2005). And now the Ohio Supreme Court has reached the same conclusion, also unanimously on this point. So that is a measure of how far matters have progressed from the pre-Crawford era, and perhaps the immediate post-Crawford era as well.

In State v. Stahl, 855 N.E.2d 834 (Ohio 2006), a case involving an adult declarant, the Ohio Supreme Court had previously applied (or more accurately mis-applied) the objective-witness test. In Siler, that court has now held that this test applies even if the declarant is a child – but only so long as the statement is not made in response to interrogation by "police or those determined to be police agents." Relying on Davis, the court held that when a child does make a statement in response to such interrogation, the primary purpose of the interrogator is the decisive question. Accordingly, the age of the child does not matter in that context. (Two justices issued a separate opinion. They concurred in the judgment, but would have declared that the primary-purpose test applies to statements made in response to police interrogation, whether the declarant is a child or an adult, and they would not have declared that the objective-witness test applies to statements by children to persons other than the police and their agents.) Applying the primary-purpose test in this case, the court easily held that the statement was testimonial, because Detective Martin’s purpose was so clearly investigation of a crime.

Several points:

1. Even assuming this decision stands, the state is not precluded from introducing Nathan’s statements on retrial. It could make Nathan a witness. How much good that would do – he is now 9 and the murder happened six years ago – is another matter. But even if he said he remembered nothing, just putting him on the stand would presumably satisfy the United States Supreme Court, given Crawford’s explicit statement that " when the declarant appears for cross-examination at trial, the Confrontation Clause places no constraints at all on the use of his prior testimonial statements," 541 U.S. at 59 n.9, and the fact that Justice Scalia, author of Crawford, also wrote United States v. Owens, 484 U.S. 554 (1988), in which the fact that the witness had virtually no memory because his head was bashed in was not deemed to undercut the value of cross. But perhaps the Ohio Supreme Court – which adopted the appellate court’s formulation of the declarant being "present at trial to defend or explain" the statement – will be more demanding.

2. There is an odd inversion with respect to child statements. In the context of statements by adults, the defendant usually favors the objective-witness standard. True, it is plenty manipulable (and Stahl is a good example; there, the court characterized as non-testimonial a detailed statement alleging sexual molestation and kidnapping, made to a nurse practitioner the day after the alleged incident and after signing a form consenting to "this forensic examination and collection of evidence"). Nevertheless, it is more likely than the primary-purpose-of-the interrogator standard to yield a characterization of a statement as testimonial, in part because often the declarant knows she is reporting a crime before the receiving police agent does and in part because the latter standard is so easily manipulable by a police agent testifying as to a supposedp urpose other than law enforcement. By contrast, in the context of statements by children, the defendant often prefers a test based on the purpose of the interrogator, because often it is clear the interrogator is gathering evidence for prosecution but the child has at best a dim understanding of what is going on. That is the case in Siler. If the primary-purpose-of-the interrogator test applies, the statement is clearly testimonial; if a declarant-based test applies, the matter is more complicated, as discussed below.

3. To underline the point about manipulability made in the last paragraph: Although this statement was undoubtedly testimonial under the primary-purpose-of-the interrogator test, because it was clear that Detective Martin was gathering evidence, I wonder how clear it will be in such cases in the future, now that police agents and prosecutors are forewarned. "What was your primary purpose in asking Nathan these questions, Officer?" "My primary purpose was to protect this child and other people as well. I knew that Nathan's mother had been killed, that the killer was at large, and that Nathan himself would be at risk, especially if the killer was his father or if the killer realized that Nathan had observed some of the incident. I knew also that so long as a killer who had acted with such rage was at large other persons might also be at risk, both people known to the killer and strangers. I needed as much information as possible to help us protect Nathan and the rest of the community."

4. It is certainly understandable that in light of Davis the court – like many other courts – would adopt an approach based on the primary purpose of the interrogator when there is a police interrogator. But as I have argued at length, the proper perspective should be that of the declarant. Grappling withthe Meaning of "Testimonial", 71 Brook. L. Rev. 251, 255-59 (2005). And I have also argued that Davis is perfectly compatible with this approach. Crawford, Davis, and Way Beyond, 15 J. L. & Pol. 553, 557-63 (2007).

The fact that a statement is made in response to a police agent who is preparing prosecution of a crime is a critical factor in determining whether the statement is testimonial, but not because the agent’s purpose is itself the criterion; rather, if the agent is gathering evidence, that fact will usually (at least in the case of an adult) be apparent to a reasonable person in the declarant’s position, and so such a person would necessarily anticipate the likely use of the statement.

5. A significant problem with the Ohio Supreme Court’s approach – which may reflect a developing trend – is that it enunciates two different standards in different settings, which suggests that it is not supported by any underlying theory. The more difficult it is to state a comprehensive standard for what is testimonial, the more the Confrontation Clause will appear to be a bundle of rules without a clear rationale – the more it will look like Roberts.

I wonder, for example, how courts that apply a primary-purpose-of-the interrogator standard in a case like Siler will handle conspirators’ statements made to undercover cops. I do not believe that the Supreme Court would have adopted the testimonial approach in Crawford if it thought that a result would be to render such statements inadmissible – and yet such statements are often gathered for the primary purpose of preparing a criminal prosecution. One could try to cover for this situation by arguing that such statements are not sufficiently formal – but as I have also argued, I believe an independent formality test is misbegotten, e.g., Way Beyond, supra, at 567-71, and it would be so mushy that it would severely weaken the confrontation right.

6. If a court does take the declarant’s perspective with respect to children’s statements, then one cannot simply ask why the police agent interviewed the child. I have already written about some of the issues in a recent, long posting on child witnesses, so I will not go into much depth here.

a. One possible issue is whether the child should be considered so immature that he is deemed incapable of engaging in the act of testifying; see par. 4b of the prior post, and note that in my view, if the child is not capable of being a witness that does not eliminate all constitutional issues, because the defendant may have a due process right to examine the child, perhaps through a child psychologist.

b. Assuming the child is capable of being a witness, there is a question of whether the child’s age should be taken into account in applying the reasonable-declarant standard top the particular statement. This issue is addressed briefly in par. 5 of the prior post.

c. To determine that the child is capable of being a witness, as suggested in paragraph 6.a above, or to determine that the particular statement is testimonial, if the age of the child is taken into account, what level of understanding must the child have? I don’t believe that the child must understand the nature of criminal prosecution and the adjudicative system; it should be enough that he understands that is statement might cause adverse consequences to someone. Perhaps also, as Sherman Clark has argued, the child’s level of moral development ought to be taken into account; see par. 4b of the prior post.

7. How to take all this into account in Siler? Perhaps Nathan was old enough to understand that he was "telling" on daddy, and maybe that is enough for his statement to be considered testimonial. But I am not sure how much good is going to be accomplished by having him take the witness stand now, or even how much would have been accomplished by having him testify at the original trial, at least months after the nurder. Perhaps the way out of this is to say that the state, knowing the limitations on children’s memories and the seriousness of the situation, should have offered Siler a chance to cross-examine Nathan while the memory was fresh. An alternative result would be to hold that Nathan was too immature to be deemed a witness, but because he was a source of evidence the defense should have had a chance to examine him, on videotape, with a child psychologist asking questions in an informal setting.

* * *

for the most part I have avoided commenting on decisions of intermediate state appellate courts, but some time ago reader Justin Eisele posted (in a comment attached to my posting of Sept. 20) another child-witness case, Seely v. State, 2007 WL 2781965 (Ark. App. Divs. I & IV Sept. 26, 2007), and he and another reader posed questions about it. Seely is an interesting case, not because the facts are unusual but rather because (unlike Siler) they reflect a very common pattern and the trial court and two groups of appellate judges each responded in different ways.

A young child -- also three at the time -- made statements to her mother and to a social worker accusing her father of sexual abuse. The trial court held the child, four by then, incompetent to testify at trial, but the mother and social worker were allowed to testify to the statements made by the child. The defendant was convicted. The appellate court held that the statements made to the social worker were testimonial. Accordingly, because the defendant had never had an opportunity to cross-examine the child, admission of the statements was a violation of the Confrontation Clause (and it was not harmless, so the conviction was reversed). But the majority also held that the statements to the mother were not testimonial. Two judges would have held that the statements to the mother, as well as those to the social worker, were testimonial.

First, in response to the anonymous commentator (please do identify yourselves, if possible): I don't think there would be any inconsistency between a holding that a given statement is testimonial and a holding that the declarant is incompetent to testify at trial. (In any event, in this case, it was the trial court that held that the child could not testify at trial and the appellate court that held that some of her statements were testimonial.) The first holding focuses more on the nature of the statement and the second on the capacity of the declarant. Together, these holdings can reflect a perception that the declarant was acting as a witness when she made her statement, but that she lacks the capacity to testify in a way that is acceptable to the judicial system. There is no logical inconsistency there.

Having said that, I’m inclined to think that courts should avoid that pair of characterizations. As noted above, I have held out the possibility that a very young child can have such a dim understanding of the likely consequences of her statement that she should not be considered capable of being a witness at all within the meaning of the Confrontation Clause. (Again, she doesn’t have to understand the legal system for her statements to be testimonial, but if she doesn’t understand that she is making a statement accusing someone of wrongdoing, with the likelihood of that person suffering some kind of adverse consequences, arguably that isn’t testifying.) In other words, there are three possibilities – (a) child incapable of being a witness, or put another way, of making statements deemed testimonial; (b) child capable of making statements testimonial in nature but not of giving testimony in an acceptable way, perhaps because of lack of understanding of the obligation to tell the truth; (c) child capable of testifying acceptably. Personally, I am inclined to think that if a child is capable of making testimonial statements then her testimony ought to be heard in court, even if she doesn’t understand the obligation to tell the truth; plenty of adult witnesses flout the obligation, even assuming they understand it. In other words, I would be inclined to eliminate category #2.

On the merits of the case, I certainly agree with the minority in rejecting a formality test and a standard that statements by a child "to loved ones or acquaintances" are not testimonial.

* * *

I bleieve the confrontation issues posed by statements made by children are enormously important, complex, and troubling. Sooner or later, the Supreme Court will have to begin resolving many of these issues. I believe that, given the limitations on the Court's resources, the chance that the Court will achieve a sound resolution of these issues will be greater if the Court first focuses on basic issues in the context of adult witnesses, and once the framework is built addresses applications to child declarants.

Second Brooklyn Symposium now available on-line

Papers from the second post-Crawford symposium organized by Bob Pitler at Brooklyn Law School are available on-line, and you can reach them by clicking here. Though the symposium was held just over a year ago, I believe most of the papers are still timely. They represent a wide divergence of points of view. I have already commented on Jennifer Mnookin's paper, in a post on expert evidence. I may comment on others, as time allows.

A view of Crawford from a Massachusetts judge

A Massachusetts judge, Davd Lowy, and Katherine Bowles Dudich, who has served as a law clerk for his court, recently published an article titled After Crawford: Using the Confrontation Clause in Massachusetts Courts, 12 Suffolk J. Trial & App. Advoc. 1 (2007). The article was brought to my attention by a recent law school graduate who worked on it and who suggested that I post a link to it, on the thought that it might be helpful to lawyers and judges, especially but not exclusively in Massachusetts. I am happy to do so, because I agree that the article may be useful -- but posting of course does not indicate endorsement, and there are in fact aspects of the piece with which I disagree.

Conflict deepens on expert evidence

I have already commented on issues raised by expert evidence in several posts. See The Not-for-the-Truth End Run; The Expertise End Run and People v. Goldstein; and The Expert Opinion Problem. One of the recurrent issues is whether a statement that is testimonial in nature may be introduced against an accused, notwithstanding the absence of an opportunity for cross-examination, on the ground that it is admitted only in support of an expert’s opinion. As the Goldstein case, from the New York Court of Appeals, made clear, when the statement supports the expert opinion only if it is true, this theory is a sham, and to countenance it allows the expert opinion to be used as a conduit for the admission of testimony that has not been subjected to confrontation.

Courts nevertheless continue to adopt this theory. A recent example is Szymanski v. State, 2007 WL 2428508 (Wyo. 2007). In proving that a fire was the result of arson, the state offered the testimony of a fire inspector to a statement made to him the night of the fire by the tenant of the apartment in which the fire was set. In the statement, the tenant, who died before trial of unrelated causes, described the condition of the apartment before she left it that evening. The state supreme court forthrightly held that this statement, which was made after the fire had been extinguished and as part of an investigation into a possible crime, was testimonial. (One would not guess this from West's headnotes to the case.) Nevertheless, the court affirmed the trial court's decision to admit the evidence not for the truth of the matter asserted "but only to show what the inspector did and what information he relied upon in forming the opinion that the fire was intentionally set by human hand."

The veneer here is awfully thin. What difference does it make "what the inspector did"? And it is obvious that the statement can help support the expert's opinion only if it is deemed to be true. It seems to me undeniable that the state was asking the jury to accept the tenant's unconfronted statement as the truth, and that if the jury did not do that the statement had no substantial value for the prosecution. This is what I have called the "expertise end run", meaning that the supposed need to allow the expert to provide the basis for his opinions is being used as a means of avoiding the Confrontation Clause.

If the logic of this case holds, one can also imagine rape counselors, domestic violence counselors, and child abuse counselors testifying that in their opinion a crime had been committed, and that one of the bases on which they draw that conclusion is the statement made to them by the victim, who has chosen not to come to court.


It is clear that, to prevent serious undermining of the confrontation right, the Supreme Court should rule that when a statement supports an expert opinion only if it is true, then admitting the statement for purposes of supporting the expert's opinion is equivalent to admitting it for the truth of what it asserts. The conflict among the states on this issue is clear and well established. I hope the Supreme Court resolves it soon.

More generally, I believe the Court needs to impose constraints, as a matter of constitutional law, on a court's ability to admit evidence of a testimonial statement for some reason other than the truth of what it asserts. The problem the Court must come to grips with is that if there are no such constraints then lower courts will be utterly free to exercise creativity in devising flimsy not-for-the-truth theories on which testimonial statements may be admitted. Cf. David Crump, On the Uses of Irrelevant Evidence, 34 Houston L. Rev. 1 (1997) (arguing that evidence is always relevant to a given proposition). Limiting instructions will provide no genuine protection for the confrontation right. I will not explore in detail here what constraints the Court should impose; that is a subject worthy of careful thought. But here is a first crack. It seems that a court considering the confrontation issue must first ask whether the testimonial statement has substantial value in proving the proposition for which it is supposedly offered without respect to whether the statement is true or not. If the answer is negative -- as it is in the Szymanski-Goldstein type of case -- then that proposition does not offer a basis for admitting the statement. If the answer is affirmative, then the court must assess whether that value warrants the risk that the jury will, notwithstanding an instruction to the contrary, use the statement to prove the truth of what it asserts.

In a new article, Expert Evidence and the Confrontation Clause after Crawford v. Washington, 15 J. L. & Policy (791), Prof. Jennifer Mnookin lucidly critiques the expertise end run and several other strategems under which expert witnesses have often been used as conduits to admit statements that were made by others and that should be regarded as testimonial. These include treating the business records exception to the hearsay rule as an exception to the rule of Crawford; treating cross-examination of the expert as an adequate substitute for cross-examination of the declarant whose statement comes in through the expert; emphasizing the supposed impracticality of holding the evidence inadmissible in the absence of the declarant; and distinguishing between factual findings and opinions, and holding only the latter to be testimonial. Her perspective suggests the need for the Supreme Court to step into this realm soon:

There are already sufficiently large numbers of cases relying upon precisely
these arguments that some courts are, unfortunately, beginning not even to argue
these points with care, instead merely relying on the allegedly persuasive
authority of other courts’ reasoning. If this trend continues, it will have a
real cost: a too-thoughtless pragmatism will have trumped principled application
of the underlying principle at stake in Crawford.

In the last part of her article, though, in focusing on two particular problems, Prof. Mnookin draws back somewhat. One of these problems is that of certificates routinely performed on certain frequently used appliances, such as speed radar guns and breathalyzer kits, to ensure that they are in good working order. If, during the period for which a given test is supposed to validate that the appliance is working properly, it generates evidence supporting prosecution of a crime, must the technician who tested the appliance and prepared the report testify at trial? Courts resist the conclusion; the costs of requiring testimony are substantial, and there may be little likely benefit. Those considerations are present with respect to many routine test reports, but they do not in themselves provide a principled argument for removing the reports from the Confrontation Clause. In the case of the certificates being considered here, though, there is a principled argument: As Prof. Mnookin points out, these certificates are not prepared with a "specific criminal act" in mind.

A holding that for this reason the Confrontation Clause does not apply to these certificates would not, in my view, do terrible violence to the Clause. On the other hand, the certificates almost certainly were prepared with an evidentiary use in mind, with the idea that they would likely be used in some prosecution, and as a matter of principle that should be enough to characterize them as testimonial. Also, if we are going to draw a boundary of the sort favored by Prof. Mnookin, we have to be careful just where to draw it. Sometimes, especially in the domestic violence context, a victim makes a testimonial statement after one alleged criminal act, and that statement is offered with respect to a subsequent act. I think the confrontation right should still apply (though it will not always matter; if the second act is homicide; it may be that the accused forfeited the right by rendering the witness unavailable). So instead of speaking only of a specific criminal act, it would probably be better to speak also of a series of acts.


But in any event, I believe that in this context we can avoid impractical results without drawing a line of the sort advocated by Prof. Mnookin, because the whole issue is constitutionally unnecessary. Nothing in the Constitution requires the state to prove that the appliance was recently tested. If a witness testifies to the time something occurred, having noted the time on a clock shortly before or after the event, there is no precondition that evidence of the clock’s accuracy be presented, much less that it be presented in a given manner. As with the clock, so with other instruments: If the state wanted, it could just let a witness testify to the evidence generated by the appliance in question, and leave it to the defense, if it wished, to raise questions about accuracy. And if the state wished, it could require the testing agency to furnish the certificate of accuracy to the defense. If the agency complied with this requirement, and the certificate appeared in order, the defense ordinarily would have no incentive to raise the issue before the jury. But if the agency failed to comply, or if on the face of the certificate the test was untimely or the appliance was defective, that would be a tip-off to the defense and to the court that a challenge to accuracy might be worthwhile; there would of course be no Confrontation Clause objection to the defense offering either a certificate or evidence of failure to furnish one.

The second problem addressed by Prof. Mnookin at the end of her article concerns autopsy reports in homicide cases in which the defendant is not apprehended for many years after the crime. She suggests that in such a case it would be tolerable for a "substitute expert" to testify to the report. I disagree. First, let’s note that the problem is a subset of a much more general problem, of witnesses who are no longer around to testify in delayed prosecutions. And in this case, the prosecution, knowing of the potential problem, can usually protect its interests quite well. If, as will usually be the case, there is a relatively small group of prime suspects, the prosecution can give notice to each of them and hold a deposition, offering to pay attorneys’ fees. (Prof. Mnookin briefly considers a "public deposition"; whether, and under what conditions, public notice could suffice in the relatively rare case in which the eventual defendant is not identified as a suspect soon after the autopsy is an interesting question. A deposition of this sort, taken of an elderly victim rather than of an expert and with counsel appointed to represent the eventual defendant, was approved in an unpublished case decided a couple of weeks after Crawford, People v. Wilkey, 2004 WL 576659 (Mich. Ct. Apps. 2004).) Also, much of the evidentiary value of an autopsy report can be gained by high-resolution video recording of the autopsy. So long as the video can be authenticated – which should be simpler than, say, authenticating DNA samples years after the fact – a pathologist would be able to testify in court on the basis of what is apparent from the video (what she sees rather than what she has been told), and there would be no confrontation problem, or even hearsay problem, at all.

There is no doubt that in the area of expert testimony, as in other areas, careful protection of the confrontation right increases the cost of prosecution. But of course that is true generally of the confrontation right, and all the rights on which our system of criminal justice is built; one could devise a much cheaper system, but it would not be a satisfactory one. Because the confrontation right was so limp in the years before Crawford, the costs of protecting it properly are often very salient, and that is probably particularly true with respect to expert evidence, which involves recurrent practices and professionally qualified (and often expensive) witnesses. With proper preparation and the creation of appropriate procedures, the judicial system can maintain rigorous adherence to the confrontation right in this area without incurring undue expense – and thus can avoid allowing expert evidence to be a force that erodes a sturdy framework of confrontation law.

Child Witnesses on the Academic and Judicial Front

Some of the most perplexing issues related to the Confrontation Clause involve statements by children. The Indiana Law Journal has recently published a symposium on the topic, under the sponsorship of Prof. Aviva Orenstein. Her introductory essay, Children as Witnesses: A Symposium on Child Competence and the Accused's Right to Confront Child Witnesses, 82 Ind. L.J. 909 (2007), summarizes the other pieces. At a glance, I can see that there is a good deal with which I agree and a good deal with which I disagree.

A recent opinion in the area is State v. Krasky, 2007 WL 2264711 (Minn. Aug. 9, 2007). This is another in a series of cases in which some courts have treated as non-testimonial an accusation of abuse made by a child to a care-giver, such as a nurse or social worker, a considerable time after the alleged incident. A few points about the case warrant mention and raise other issues.

1. The “purpose” evasion

The majority of the Minnesota Supreme Court reached this conclusion notwithstanding the fact that a police report indicated that a police officer and a child protection worker jointly concluded that “the best way to proceed with the investigation was to have [the Child Resource Center where the child made the accusation to a nurse] do an interview with [the child] along with a medical exam.” Picking up on the “primary purpose” language of Davis v. Washington, 126 S.Ct. 2266 (2006), and extending it beyond the emergency situation, the court deemed the primary purpose of the interview to be protecting the health and welfare of the purported victim. In my view, this result highlights one of the serious problems of a test depending on the primary purpose of the questioner. True, some courts have shown deftness in manipulating what I believe is the proper test, depending on the reasonable anticipation of the speaker. But a questioner-purpose test is much more easily manipulable. First, it allows the questioner herself to portray her purpose in a way supporting a conclusion that the accusation is not testimonial. Second, as in this case, such a test allows a routine in which accusations made to agents without law enforcement responsibility to be deemed non-testimonial, no matter how predictable it may be that the agent will relay the accusation to court if the accuser does not appear. Indeed, given the approach in Krasky, I am not sure what would prevent a state from referring all alleged crime victims to a victims’ counselor who would then take the accusation and transmit it to court if the victim does not testify there. So imagine this conversation between a counselor and an adult who has allegedly been the victim of a crime:

Counselor: Officer Tuesday has referred you to me so that I can help you deal with the considerable trauma that crime victims often suffer. I know you already told her about this incident, but for me to assist you, it is important that I hear from you exactly what happened to you, and who did it to you. If you don’t mind, I am going to videotape our session so that I can consult later with my colleagues about it.

Alleged Victim: No problem. And thanks so much. But tell me. If I understand correctly from TV shows I’ve seen, if I choose not to go to court, you could testify yourself and show the videotape. Right?

C: That’s true. Of course, that’s not our primary purpose in holding this session.

AV: Whatever. I just don’t want to be in the same room as that guy, or look at his face, or have to answer a lot of questions by his lawyer.

C: No, you wouldn’t have to.

That sounds to me as if we are entering death-of-the-confrontation-right territory.

2. Davis leeway as limited to emergencies.

Justice Page, with one of his colleagues, dissented in Krasky, which is interesting because he had written one of the prior decisions holding a child’s accusation in largely similar circumstances to be non-testimonial – but he regarded Davis as calling for a different result, because the statement was not made during the course of an ongoing emergency.

3. Non-testimonial statement, incompetent witness.

The child was apparently incompetent to testify at trial. This also was a basis for the Page dissent. But the matter is quite complex, in my view. Obviously, the combination of holdings that the prosecution may introduce the child’s out-of-court statement and that the defense cannot call the child to the stand places the accused in a theoretically disadvantageous position: The statement will be admitted against him and he has no way of examining the declarant.

How big a problem is this? As a practical matter, it is unlikely (under present procedures) that if the statement is admitted against the accused he would choose to call the declarant to the stand if he could; that doesn’t happen often, for reasons I have explored before. (See, e.g., these posts: Pending Cert Petitions, Shifting the Burden, Take 2, and Shifting the Burden.) Moreover, the accused is in a similarly disadvantageous position whenever a statement by an unavailable declarant is admitted against the accused, and if the statement is in fact non-testimonial (as the majority held it was here) there is no confrontation problem. So if we assume that the statement was not testimonial in nature, I don’t believe the extra consideration that the defense could not call the child to the stand is particularly worrisome.

4. Apparently testimonial statement, incompetent witness

But now suppose, along with Justice Page, that the out-of-court statement would be characterized as testimonial in nature if the child were a competent witness – but in fact she is incompetent to testify at trial. What are the consequences?

In the 18th century, before R. v. Brasier, 1 Leach 199, 168 E.R. 202 (K.B. 1779), changed the rules, young children were categorically incompetent to testify at trial, and their out-of-court accusations were sometimes admitted. (I discussed this fact briefly in my reply brief in Hammon, and Tom Lyon and Raymond LaMagna explore it at length in their contribution in the Indiana symposium, The History of Children’s Hearsay, from Old Bailey to Post-Davis, 82 Ind. L.J. 1029 (2007)). I don’t think this history is a good guidepost for decision today. Those old cases admitted the out-of-court statements as the best evidence available, but I believe they were accommodating to, and effectively avoiding and undercutting, an overly active rule of incompetence. That is, most of these children probably should have been deemed to be competent witnesses (or at least to be heard; see below). Admitting the out-of-court statement was a way around the rule of incompetence, though with the serious problem that the statement was not made under the conditions required for testimony.

In resolving the problem, I think it is useful to distinguish between two different levels of incompetence.

a. The child who is capable of testifying, but not in a satisfactory manner

On one level, the child is sufficiently capable of understanding the nature of her statement for it to be considered testimonial, but she lacks a sufficient sense of obligation to tell the truth for her testimony to be accepted in court. Thus, the out-of-court statement is clearly testimonial in nature, but the child is not capable of testifying in court in a satisfactory manner.

In this setting, it is clear that the out-of-court testimonial statement should not be admitted. Out-of-court testimony, without oath or cross-examination, is clearly inferior to in-court, testimony, and if the child’s testimony in court should not be admitted because she is incompetent to give it then a fortiori neither should her out-of-court testimony be admitted.

Now, in passing I’ll mention that I believe there is a plausible argument that the child should not be disqualified from testifying in court simply because she does not demonstrate a sense of obligation to tell the truth. We know that many witnesses, though taking an oath, are strongly motivated to lie and do lie quite deliberately, and we nevertheless admit their testimony – as we admit all sorts of non-testimonial evidence – for what it is worth; even if the evidence is not particularly reliable, it may be a net benefit to the truth-determining process. Indeed, as I understand it, the tradition in some Continental courts is that the accused does not testify under oath but nevertheless he tells his story. And this was the practice in common law courts for centuries – this is what John Langbein called the “Accused Speaks” model, The Historical Origins of the Privilege Against Self-Incrimination at Common Law, 92 Mich. L. Rev. 1047 (1994), which prevailed before defense lawyers tended to keep their clients quiet and before criminal defendants were allowed to testify under oath, beginning in the latter part of the 19th century. In the case of a child witness, administering the oath or an oath-like procedure might be considered an attempt to remind the child of the obligation to tell the truth and to reveal the child’s inclination to do so – and arguably that is all, so that any defects the child may have on this score should not prevent her from telling her story in court.

Putting this thought aside, though, the narrower point made here is clear: If a child makes an out-of-out testimonial statement and does not testify in court, the presumptive violation of the Confrontation Clause is not excused on the basis that she would not have been allowed to testify in court anyway because she lacks a sufficient obligation to tell the truth.

b. The child who is incapable of testifying

Now consider a deeper level of incompetence – the child is (or a child of ordinary understanding of her age would be) so insufficiently developed that the statement should not be deemed testimonial at all. Note that characterizing the child in this way would not be saying that she is acting as a witness but without satisfying the conditions for a proper witness; rather, the argument would be that she is just not capable of engaging in the kind of activity – witnessing – covered by the confrontation right.

I will not attempt to resolve here what kind and degree of development would be necessary for a person to be considered capable of witnessing. It may be that it is sufficient if the child understands that her statements are capable of causing adverse consequences to another person – or put another way, if the child has such limited understanding that she does not realize this, she should not be considered to be witnessing. I think it’s also worthwhile thinking about the suggestion made by my colleague Sherman Clark, An Accuser-Obligation Approach to the Confrontation Clause, 81 Neb. L. Rev. 1258, 1280-85 (2003), that in determining who is a witness within the meaning of the Confrontation Clause moral as well as cognitive development must be taken into account.

In his contribution to the Indiana symposium, Testing the Testimonial Concept and Exceptions to Confrontation: “A Little Child Shall Lead Them”, 82 Ind. L.J. 917 (2007), Bob Mosteller vigorously disagrees with the idea that some children should be considered so undeveloped that they are deemed out of the Confrontation Clause altogether. He seems to acknowledge that in one case that I have used to make this argument, State v. Webb, 779 P.2d 1108, 1109 (Utah 1989), in which an 18-month-old child said, “Ow bum daddy,” the argument “may be well taken.” But in the more usual case, he says, the child uses “purposeful communicative abilities.” I acknowledge that Webb is an unusual case, but that doesn’t eliminate the issue. I suppose that the child in Webb had and was using “purposeful communicative abilities.” More fundamentally, it strikes me as question-begging to assume that the use of such abilities renders the statement testimonial. Not all purposeful communication is testimonial, of course; the question becomes what additional conditions are required to make it so, and I believe it is at least plausible to hold that membership in the human species is not enough to qualify the maker of a statement as a witness. (If a trained bloodhound’s bark were determined to be purposeful communication to the effect, “This guy here is the source of the smell on that shirt you put under my nose a while back,” that would not render the dog a witness within the meaning of the Confrontation Clause.)

So at least arguably, I believe, in some cases a child who makes an out-of-court statement should not be considered a witness at all for purposes of the confrontation right. Now, this possibility might seem to lead to absurd results, because it would mean that the child’s incapacity is an argument favoring admissibility of her statement. I have a couple of responses.

First, I’m not sure that the result is indeed absurd on its face – let’s bear in mind that the confrontation right is not about sifting out good from bad evidence but about protecting the conditions for giving testimony.

Second, even if the confrontation right does not apply, on the ground that the child was not acting as a witness, this does not necessarily mean that the accused would not have a right to examine the child if the child’s statement were admitted. Let’s say the prosecution introduces blood that it claims came from the accused, or a pistol that is arguably the murder weapon, or a document the age of which is in question. The defendant has a right to examine this evidence, and it should be allowed to do reasonable experiments on it. I think the same principle ought to apply if the prosecution wishes to introduce an out-of-court accusation made by a child who is not deemed competent to be a witness. That is, the accused ought to have a right, recognized as a matter of due process, to examine the child – but I’m using “examine” in a non-technical sense. The child is not acting as a witness but is nevertheless a source of evidence, and so the defense has a right to examine her and try to generate information suggesting that the evidence she yields does not have the value the prosecution ascribes to it. In the case of a pistol, the principal examiner would be a ballistics expert; in the case of a child, it might be a child psychologist. Much of the examination would consist of questions, but it would not be cross-examination as we know it. Furthermore, in this context the accused would not generally have a right to be in the presence of the child. But presumably, the accused would have the right to have the encounter video-taped.

So I am suggesting that even though the Confrontation Clause does not apply in this setting, the accused may still have a constitutional right to have some sort of examination of the child. Does this undercut the clean principle (for which I’ve advocated) that if a statement is not testimonial the confrontation right does not apply? I don’t believe so. I’ve always thought that in particularly pressing cases the accused might have a due process right to demand production for cross-examination of the declarant of a non-testimonial statement, if the declarant is available. The procedure being discussed now would apply only if the child was available; the due process demand is particularly compelling in the context being examined here, in which the declarant is incompetent to be a witness; and the prescribed procedure is not even cross-examination.

Conceivably, the child could give her account, and the examination on behalf of the defense could occur, at trial. This still would not make the event testimonial if the child is deemed incapable of witnessing; this would be the counterpart of performing an experiment in the courtroom.

5. Age adjustment?

Even if a child is considered competent to be a witness within the meaning of the Confrontation Clause, there is a separate issue as to whether the child’s level of understanding should be taken into account in determining whether the particular statement is testimonial. This is a tricky question. I’m not entirely happy with a doctrine that asks about the expectations of a reasonable person when the speaker is in fact a young child. But neither does it seem satisfactory to break down the objectivity of the inquiry in this particular context – all the way by asking about the subjective anticipations of this child, or with a halfway measure such as asking about the anticipation of an eight-year-old child of ordinary understanding. Perhaps the best resolution is to consider the particular child subjectively in deciding whether she is capable of engaging in witnessing but to apply a straight objective test – referring to the standard “reasonable person” in determining whether the particular statement is testimonial. Readers’ thoughts on this question, as well as on any of the other difficult matters raised here, are of course welcome.

Opportunity for Cross-Examination at Preliminary Proceedings

I am writing this post to invite readers' comments on this set of questions: California v. Green, 399 U.S. 149 (1970), holds that an opportunity to cross-examine at a preliminary hearing satisfies the confrontation right. How, if at all, does this principle affect the conduct of lawyers and courts at preliminary hearings and other pre-trial evidentiary proceedings (such as depositions taken for discovery purposes) not held for the primary purpose of preserving testimony? For example, how often do defense lawyers conduct a full cross-examination – knowing that if they do not and the witness is unavailable to testify at trial, it may be that the court will admit the earlier testimony and reject a claim of the confrontation right by ruling that the defense already had an opportunity for cross-examination? And how willing are courts to allow a full cross-examination given this possibility?

My own slightly educated guess is that usually the prospect of trial admissibility does not cause defense lawyers to conduct extensive cross-examinations at preliminary proceedings, and that if they tried to do so the courts would constrain them. Preservation of testimony is not the purpose of these proceedings, and if the defense protected itself (though at the potential cost of losing benefits of surprise) by making extensive cross-examinations these proceedings would become much more time-consuming. But I would like to know more than I do now from those who deal with these situations regularly. Thanks!

Ninth Circuit decision in Yida -- on "reasonable means" and unavailability

Yesterday, a panel of the United States Court of Appeals for the Ninth Circuit issued its decision in United States v. Yida, 2007 WL 2325143. This case generated some anticipation because after oral argument the panel issued an order inviting amicus briefs from any interested person. The court's opinion is a good one, and has several interesting aspects to it.

Yida was tried on drug charges, but the jury hung. A key witness against Yida was Reziniano, an Israeli. The Government allowed Reziniano to be deported to Israel, accepting Reziniano's solemn promise that he would return to testify at a retrial. But soon after reaching Israel, on the eve of the retrial, surprise, surprise, Reziniano said he would not return, supposedly for medical reasons. The district court held that Reziniano's testimony from the first trial could not be admitted at the retrial. The Government appealed that ruling, but the Ninth Circuit has now affirmed. Judge Gould wrote the opinion for a unanimous panel, resting the decision squarely on a holding that Reziniano was not unavailable within the meaning of Fed. R. Evid. 804(a). He also added a brief concurrence (rather unusual, isn't it, the same judge issuing majority and concurring opinions at the same time?), emphasizing the constitutional significance of unavailability.

The main opinion has a discussion that I think is quite useful on the advantages of live testimony at a second trial as compared to the transcript of testimony from the first trial. (Some self-interest there, I suppose, because this discussion quotes approvingly from my amicus brief.) The most obvious advantage, of course, is that live testimony gives the jury the chance to observe the demeanor of the witness. Beyond that, requiring the presentation of live testimony when possible gives a defendant a second crack at the witness, and thus a chance to develop inconsistencies; it gives the defense an opportunity to cross-examine on the basis of all information available at the time of the second trial; and it deprives the prosecution of the opportunity to "stand pat" on the transcript when it has reason to believe that the witness would come off worse in live examination. An accused is not guaranteed a second chance to examine a witness, of course, but these are beneficial by-products of holding a second trial, and they should not be forsaken if the witness is available to testify live.

The principal issue at stale in Yida is whether the Government used "reasonable means," within the meaning of Fed. R. Evid. 804(a)(5), to procure Reziniano's attendance at trial. The court's discussion strikes me as very sound. (Same self-interest, same reason.) The court examined alternative choices that the Government might have made, even apart from keeping Reziniano in custody pending a second trial. It might have taken Reziniano's passport away or held him in electronic detention, or done both, to ensure his appearance. Or before deporting Reziniano it might at least have taken a video deposition, which would have offered most of the advantages of live testimony. To the Government's argument that a deposition would have been of no avail, because it could be admitted only if Reziniano was deemed unavailable, the court properly responded that taking a deposition would have altered the calculus of whether the Government's conduct in deporting Reziniano was reasonable.

More generally, the court squarely rejected the Government's contention – which did not have a lot to be said for it – that reasonableness ought to be determined within a time frame beginning shortly before trial. The measure of reasonableness should not be confined by artificial boundaries, and prosecutors should be charged with the ability to think and plan ahead. I hope this same principle will be applied in the somewhat related context of determining whether, for purposes of applying forfeiture doctrine, the accused should be deemed to have rendered a witness unavailable. It sometimes happens that a witness who otherwise would have been able to testify at trial is prevented from doing so by the accused's misconduct – most frequently intimidation or homicide – but if at an earlier time the prosecution had taken reasonable steps (such as holding a deposition) the witness could then have testified subject to confrontation. In this context, as in Yida, it should not be enough for the prosecution to focus on the time immediately before trial and proclaim that there was nothing it could then do to bring the witness to trial.

Standard of review for limitations on cross-examination

Greg May has brought to my attention a significant en banc decision, United States v. Larson, 2007 WL 2192256, issued Wednesday (Aug. 1, 2007) by the Ninth Circuit, addressing the question of the standard of review that applies when an accused contends on appeal that a trial judge’s limitation of questions on cross-examination violated his rights under the Confrontation Clause. (Greg’s comments are available on California Blog of Appeal. My views are in large part, but not completely, in accord with his.) This is an issue on which the Circuits, and indeed panels within the Ninth Circuit, have split in three basic groups.

Some courts have held that restrictions on cross should be reviewed de novo. Some have held that such restrictions should be reviewed only for abuse of discretion. And some courts, often suing rather mystifying language, have held that some combination of these standards applies. The Ninth Circuit now adopts the last of these approaches, with a prevailing opinion that offers considerable elaboration:

If the defendant raises a Confrontation Clause challenge based on the exclusion of an area of inquiry, we review de novo. In reviewing a limitation on the scope of questioning within a given area, we recognize that “trial judges retain wide latitude insofar as the Confrontation Clause is concerned to impose reasonable limits on such cross-examination based on concerns about, among other things, harassment, prejudice, confusion of the issues, the witness' safety, or interrogation that is repetitive or only marginally relevant.” Delaware v. Van Arsdall, 475 U.S. 673, 679 (1986).

I wonder how tenable that distinction will be; “area of inquiry” is not a self-defining term. In Larson itself, for example, the court concluded that the restrictions on cross-examination – preventing the defendants from asking about the mandatory minimum sentences that two witnesses would have received had they not cooperated with the Government – lay “within an area of inquiry,” specifically “the biases and motivations to lie” of those witnesses. “[B]iases and motivations to lie” constitute a mighty broad “area of inquiry”; I wonder how often a judge ever forecloses cross-examination of it. In this case, one might have said, it seems to me, that the relevant area of inquiry was “the sentence that the witness would have received absent cooperation.”

Moreover, immediately after articulating the distinction, the court immediately muddied the waters by adding, “A challenge to a trial court's restrictions on the manner or scope of cross-examination on nonconstitutional grounds is thus reviewed for abuse of discretion.” I don’t see any reason why it should matter whether the governing rule of law is constitutionally based or not.

And the court added further confusion by stating, “This standard . . recognizes that whether there has been a Confrontation Clause violation is ultimately a question of law that must be reviewed de novo.” I am scratching my head trying to reconcile that statement with the passage quoting Van Arsdall. Perhaps the Court means that in some circumstances de novo review incorporates a discretionary substantive standard of what constitutes a violation, but if that is so effectively there is review for abuse of discretion. In any event, I do not believe that the premise – that whether there has been a violation is a question of law subject to de novo review -- is completely accurate. Compare the case in which a trier of fact determines that there has been negligence. The reviewing court must determine by de novo review whether the standard applied for what constitutes negligence is the correct one. And the reviewing court might decide as a matter of law that the facts could not support a conclusion that there was negligence in the particular case; if there is a jury, the question of whether the case should have been left to it is subject to de novo review. But in many cases, the evidence could support a conclusion either way, and that includes some cases in which what happened is clear but how it should be characterized is not. De novo review does not apply there.

Having said all this, it seems to me that the en banc court had the right instinct, though its expression was somewhat clunky. As indicated above, I don’t think there is any sharp distinction between preclusion of an area of inquiry and restriction within an area. But there may be a significant continuum here, and perhaps the court is pointing in the general direction of it. Appellate review can be more or less deferential. It should be less deferential, all other things being equal, if a situation is likely to recur with some regularity in materially similar circumstances. Recurrence suggests value in uniformity, which can be gained only with de novo review. And recurrence in materially similar circumstances suggests the possibility of articulating workable rules of law. So the more the trial judge’s ruling sounds like precluding a well-defined area of inquiry, and the less it sounds like merely putting some limitations on the extent to which defense counsel can pursue that inquiry, the stronger the argument for giving little deference to the trial judge’s ruling.

On that basis, I think there may be good reason to conclude as a matter of law that the defense must be allowed a reasonable opportunity to explore what the cooperating witness’s expectations were with respect to the sentence he would receive absent cooperation. But, as noted above, the en banc court treated this as a matter subject to review for abuse of discretion.

In applying that standard, the court – unanimous up to that point, as I understand it – split three ways. The prevailing group concluded in the case of one witness that the defense had an adequate opportunity to demonstrate the bias of the witness. (This was in part, interestingly enough, because the witness actually answered the question as to what the minimum would have been absent cooperation, and though the trial judge intervened at that point and said this was not a proper subject of questioning, he never actually struck the answer; so what should defense counsel have done, say, “Judge, if you’re going to rule against me, please do it clean, and tell the jury to disregard that helpful answer I just got”?) The judge’s ruling with respect to the first witness gave a clear signal, and defense counsel never attempted to reveal to the jury that absent cooperation the second witness faced a mandatory life sentence without the possibility of release. The prevailing group of Ninth Circuit judges concluded that the defense was thus improperly precluded from demonstrating to the jury the magnitude of the incentive that the second witness had to cooperate with the Government. But guess what – the error was harmless. One other group of judges, contending that there was no violation with respect to the second witness, concurred in the judgment; another group contended that there was a violation with respect to the first as well as the second witness, and that the error was not harmless.

Forfeiture: The Standard of Proof and the Reflexive Case

Here's a post I meant to put up several weeks ago:

The Washington Supreme Court has joined the minority of courts that have adopted an elevated standard of proof that a defendant engaged in the wrongful conduct that rendered a potential witness unavailable. In State v. Mason, 2007 WL 2051541 (Wash. Jul. 19, 2007), the court held that “in deciding whether to apply the doctrine of forfeiture by wrongdoing, the trial court must decide whether the witness has been made unavailable by the wrongdoing of the accused based upon evidence that is clear, cogent, and convincing.” The court held, with apparent justification, that the standard was met in the case before it. This was a murder case, and the statements at issue, which the court held were testimonial, were accusations by the victim of a prior assault on the victim. The court appears not to have been troubled by the fact that the accusations in question were made before the crime in question – the murder – was committed.

I think the Washington court is right that an elevated standard should apply in determining forfeiture, but the betting has to be that if the Supreme Court ever resolves the question as a matter of federal constitutional law it will adopt a preponderance standard; it certainly gave a big hint in that direction in Davis v. Washington, 126 S.Ct. 2266 (2006). If it does so, though, state courts will remain free to apply elevated standards as a matter of their own constitutional law.

In any event, I believe this issue is less important than might first appear. A court inclined to let evidence in on the basis of forfeiture will most likely be willing to recite the "clear, cogent and convincing" standard if it has to. I doubt this language will help very many defendants.

More important, it seems to me, is the question of what duty to mitigate the prosecution has when it seeks to invoke the forfeiture doctrine. Sometimes, when the wrongdoing that rendered the witness unavailable is murder, the prosecution has an opportunity between the time of the fatal blow and the death of the victim to take the victim's deposition; most often, as in Mason's case, that is not so. And, when the contention is that the accused forfeited the right by intimidating the witness, a slew of issues arises concerning what steps, if any, the prosecutor or court should take to ascertain how much of the confrontation right can be preserved. I have commented on some of these issues in prior posts -- for example, Forfeiture, the Prosecutorial Duty to Mitigate, and Rae Carruth.

In Mason, by the way, four justices refused to sign the majority opinion. The principal reason was that they believed the court should not adopt what I have called a doctrine of reflexive forfeiture – that is, applying forfeiture doctrine when the act that purportedly rendered the witness unavailable is the same as the act with which the defendant is tried. (Two judges dissented on this basis; the other two concurred because they believed the error was harmless.) That is, the accused was charged with murder, and the act that allegedly rendered the witness (the alleged murder victim) unavailable was the act of murdering him. These four justices regarded application of the doctrine in this circumstance as a violation of the presumption of innocence and the principle requiring proof of guilt beyond a reasonable doubt. In my view, this is simply a logical error. Because I've expressed this view often before, see, e.g., Forfeiture and dying declarations, I'll address it very briefly here.

In a case of this sort, two basic factual determinations must be made, and usually by two separate decision-makers. The jury must decide whether the defendant is guilty beyond a reasonable doubt of the crime charged. In the course of the trial the court must decide whether the criteria for the admission of evidence are met. If, say, the accused claims that admission of an out-of-court testimonial statement by a witness to a robbery would violate his confrontation right, but in fact the reason the accused has not had an opportunity to cross-examine the witness is that he murdered her, it seems obvious to me that the accused has forfeited the confrontation right. It seems equally obvious that it is the court that must determine whether the facts supporting a conclusion of forfeiture are true to the requisite degree of probability. The situation does not change when the act that purportedly rendered the witness unavailable is the crime being tried. If in fact the accused’s wrongful conduct is what kept the witness from testifying in court, then it would be abhorrent to keep that testimony out on the basis that the accused never had an opportunity to cross-examine. There are still two fact-finding functions that are separate. The jury must determine guilt; to do so it must begin with the presumption of innocence, it can only use the evidence allowed by the court, and it can fund guilt only if it is persuaded beyond a reasonable doubt. The court has to determine whether the conditions for forfeiture are made out, and if it does it admits the evidence – but of course it does not say, “Jurors, the reason you’re hearing this evidence is that I’ve determined that the accused murdered this potential witness” – just as in a conspiracy case it does not say, “Jurors, the reason you’re hearing this statement is that I’ve concluded that it was made during the course of and in furtherance of a conspiracy of which both the declarant and the accused were members. Now it’s up to you to determine whether they were conspirators for the purpose of determining guilt.” In short, I believe this resistance to reflexive forfeiture is a red herring.

Cert petition by Missouri on the lab report issue

One of the most significant outstanding Crawford-related issues is whether lab reports and other reports routinely prepared as part of the prosecutorial process should be deemed testimonial. If such a report is considered testimonial -- as it should be -- then it may not be admitted against the accused unless the author of the report is subjected to cross-examination. The lower courts have sharply divided on the issue, but so far the Supreme Court has not taken an opportunity to resolve it. But that may soon change, because now a state, rather than a defendant, has asked the Court to take the matter up. You can read the petition for certiorari (passed on to me by Jeff Fisher) in Missouri v. March by clicking here.

Can testimony be preserved before arrest?

It often happens that for some time after commission of a crime, especially a homicide, a suspect is not apprehended, yet the state anticipates that eventually there may be an arrest and a trial. Now suppose that a prosecutor realizes that a key witness may not be available by the time of a trial. May the prosecution preserve the witness's testimony?

Let's divide the problem into three. In the first variation, the ultimate defendant has been identified as the principal suspect, but he has not yet been apprehended. I've speculated about the possibility that counsel could be appointed for the suspect, some form of notice be given to the suspect, and a deposition held. That, it seems to me, gives an adequate opportunity for confrontation, at least if the notice given was reasonable under the circumstances; perhaps the accused would have had to emerge from hiding to confront the witness and to consult with counsel, but he did not have a right to remain in hiding. If there isn't notice of the deposition, but the eventual accused knows that he is being sought, is that enough to constitute a forfeiture of whatever aspects of the confrontation right were not protected by the deposition procedure? I don't know.

In variation two, the ultimate defendant is one of several identified suspects shortly after the crime, and the authorities don't yet have enough evidence to arrest any of them. In this variation, it seems the solution is to give notice to all of them, to offer counsel to each of them, and to appoint counsel for any that do not appear.

The toughest variation arises when the authorities have not yet identified a suspect. May they still preserve the testimony by giving some kind of publication notice, appointing counsel for the ultimate defendant, and taking a deposition? That is considerably more difficult, but the matter is still worth considering. In some cases, and on some issues, counsel may be able to conduct suitable cross-examination even without knowing who his client is. But is this enough?

I've just stumbled across People v. Wilkey, 2004 WL 576659 (Mich. Apps. 2004), which highlights some of the issues with respect to pre-arrest preservation of testimony. This was a homicide committed in the course of a break-in. The victim's widow, who was present at the scene, was 82 years old. Wilkey was one of several suspects from the start, but no arrest was made for some time. So the state held a deposition of the widow. A lawyer, who later became a judge, was appointed to represent the interests of any future defendant. The widow's testimony was quite detailed about the event. She also gave some physical description of the perpetrator, who was wearing a mask. According to the appellate court, the defense lawyer, "in a very respectful manner, asked questions of [the widow] to show that she was of sound mind, e.g., what is the date, and she answered appropriately." At the close of her tesitmony, counsel offered the opinion that she was "not easily confused."

The widow died before Wilkey's trial, and the prosecution introduced the deposition transcript. Wilkey was convicted and the appellate court affirmed. Its decision came fifteen days after Crawford, but the court showed no recognition of Crawford, and held that the deposition was admissible under Michigan's residual exception to the hearsay rule. The court did take into account Wilkey's argument that the appointed lawyer had no motive to develop evidence to support Wilkey's theory that the perpetrator was one of two other men identified by Wilkey. The court acknowledged that "more precise questions could have been asked . . . if counsel was aware of the physical characteristics of defendant," but concluded that the lawyer had the tools to develop evidence about the perpetrator's physical characteristics., and did so.

I think the appointment of counsel in this case before the arrest or even definite identification of the accused is noteworthy. Courts ought to encourage prosecutors to preserve testimony where possible. But in this case, the court could have done better. This was a case fitting within the second variation described above -- Wilkey was one of several suspects identified at the time of the deposition. He, along with the others, should have been given notice before the deposition, and he should have had an opportunity fo separate representation, so his lawyer could have asked the widow questions that might have deflected suspicion away from himself. In a system that guarantees the rights of effective assistance of counsel and confrontation of adverse witnesses, it is not enough to be assured that a lawyer was appointed who had a disinterested opportunity to explore the facts.

Illinois Supreme Court deems statements to clinician and social worker to be testimonial

The Illinois Supreme Court issued a significant decision today in People v. Stechly (and thanks to my former student Brian Koch for pointing it out to me). It holds that a child's statement of abuse made to her mother was not testimonial, but that statements made to a clinical specialist in charge of a hospital child-abuse team and to a social worker at the child's school were testimonial -- and that the error in mischaracterizing these statements was reversible. The Court adopts the view of testimonial articulated in United States v. Cromer, 389 F.3d 662 (6th Cir. 2004). That's good.

The court also adopts a narrow view of forfeiture outside the context of murder, holding that intent (in the sense of purpose, not of anticipation of the natural consequences of one's actions) to procure the witness's absence is necessary for forfeiture. The court distinguishes murder cases -- without holding definitively that intent is not required in those cases -- on the ground that in a murder case the perpetrator has "absolute certainty" that the murdered witness will be unavailable to testify. Well, I don't see how that distinguishes murder -- it's at most just a very strong anticipation of unavailability. And in fact, the perpetrator doesn't have that certainty in most of the cases where the issue arises. The typical case is one in which the defendant (by hypothesis) cast the fatal blow, and the victim makes a statement afterwards; the accused can't necessarily know that the victim won't survive long enough for confrontation at least at a deposition.

Another state decides (at least partially) correctly on lab reports

The Oregon Supreme Court today decided State v. Birchfield (and thanks to Ryan Scott for calling it to my attention even before it was decided!), holding that the trial court violated the accused's right under Article I, sec. 11 of the state Constitution "to meet the witnesses face to face" when it introduced a lab report without presenting the testimony of the criminalist who prepared it or demonstrating his unavailability. The court also holds that a process giving he accused the right to subpoena the witness is no substitute for the constitutional right. The court does not reach contentions under the federal Constitution. Very nice result, crisply reached, immune from federal review, doesn't help set up a conflict for certiorari purposes. My enthusiasm is qualified because the decision leaves open the possibility that the court would say that if the criminalist is available the report can come in even absent an opportunity for confrontation. But clearly that issue is not reached in this decision.

California Supreme Court decision in People v. Cage -- commentary by Jeff fisher

The California Supreme Court issued its decision yesterday in the long-awaited case of People v. Cage. (Thanks to Paul Vinegrad for calling my attention to it.) It treats a statement by a young victim to a police officer, in the station-house and the hospital, as testimonial, and his statement to a doctor, which the court regarded as having been asked and given solely for medical purposes, as non-testimonial. The court acknowledges a conflict in reasoning with State v. Mechling, 633 S.E. 311 (W. Va. 2006), but does not purport to overrule People v. Sisavath, 118 Cal.App.4th 1396 (Cal. App. 2004). I'm sorry that I'm too rushed now with other matters to make a longer comment on the case at this time. But I've asked Jeff Fisher to expand on his initial comments, and here is what he has to say:

The California Supreme Court's opinion in Cage provides a platform to consider a very important category of post-Crawford evidence: statements to doctors and other medical personnel. It also highlights some of the confusion developing in the wake of the Davis v. Washington.

John Cage was involved in an argument and physical altercation with family members. A neighbor called the police to report the disturbance, and over one hour later a police officer found John alone in the hospital waiting room awaiting treatment for a serious cut on his face. The officer asked John to describe the incident to him. John told the officer that his mother, Lisa Cage, had cut him with a piece of glass from a table that had broken during the altercation. A doctor was present for "at least part" of that conversation and then took John back for an examination. The doctor's first question was, "what happened?" John repeated to the doctor the same thing that he had told the officer.

The California Supreme Court held that John's statement to the officer was testimonial because it was not materially distinguishable from the statement Amy Hammon gave the police in Hammon v. Indiana (decided in the same opinion as Davis v. Washington). In the more noteworthy portion of the opinion, the Court also held that the statement John made to the doctor was nontestimonial. In dealing with statements to doctors that describe criminal conduct, there are three general approaches a court might take:

(1) Statements describing criminal conduct as causing a physical condition and identifying an alleged perpetrator are testimonial. Courts, including the California Court of Appeal in an opinion that the California Supreme Court here did not call into question, have taken this approach when the doctors have examined victims as a coordinated part of a criminal investigation. See, e.g., People v. Harless, 125 Cal. App. 4th 70 (2004), rev. granted, 109 P.3d 69 (Cal.), rev. dismissed, 119 P.3d 962 (Cal. 2005) (statement to doctor "in the course of the district attorney's investigation of child abuse" testimonial); Medina v. State, 143 P.3d 471 (Nev. Oct. 5, 2006); Hernandez v. State, 946 So.2d 1270 (Fla. App. Jan. 26, 2007).

(2) Statements describing criminal conduct as causing a physical condition are not testimonial but statements identifying an alleged perpetrator are testimonial. Some courts have adopted this position in cases in which doctors were not coordinating their efforts with the police. See, e.g., State v. Slater, 908 A.2d 1097 (Conn. 2006); In re T.T., 815 N.E.2d 789 (Ill. App. 2004); State v. Bartholomew, 127 Wash. App. 1006 (2005). Some states, in fact, provide that statements identifying an alleged perpetrator to doctors do not even satisfy the medical diagnosis hearsay exception. See, e.g., Commonwealth v. DeOliveira, 849 N.E.2d 218, 224 (Mass. 2006).

(3) No statements made to doctors (at least when police are not also part of the interview) are testimonial. Some courts have adopted this position in cases, as in (2), in which doctors were not coordinating their efforts with police. See, e.g., People v. Vigil, 127 P.3d 916 (Colo. 2006); State v. Vaught, 682 N.W. 2d 284 (Neb. 2004); United States v. Peneaux, 432 F.3d 882 (8th Cir. 2005).

The California Supreme Court avoided approach (1) by stating that "[t]here is no evidence that Dr. Russell was acting in conjunction with law enforcement." This seems questionable. When the doctor asked John "what happened," he was already aware that the police were investigating whether John's injury was a result of criminal conduct. And, as the court acknowledged, California law required the doctor to report any evidence of abuse that came to light during his examination of John. But the court was unmoved, contending "[t]he mere fact that doctors must report abuse they see, suspect, or know of does not transform them into investigative agents of law enforcement," and that an investigative purpose is necessary to trigger the Confrontation Clause. It will be interesting to see how future courts deal with the "mandatory reporter" issue. If states can use such laws to gather accusatory statements and can steer witnesses to mandatory reporters without triggering the Confrontation Clause, they can go a long way toward creating systems that evade the confrontation right altogether.

Perhaps even more troubling, the California Supreme Court chose approach (3) over approach (2) for two reasons that appear illegitimate and one that Davis simply leaves unresolved. First, the court focused on the fact that the doctor's question to John was "neutral in form" and not "structured" to generate evidence. But Davis instructs that answers to "open-ended questions" are just as likely to be testimonial as answers to structured interrogation, for "it is in the final analysis the declarant's statements, not the interrogator's questions, that the Confrontation Clause requires us to evaluate." 126 S. Ct. 2274 n.1. And whatever relevance may pertain to a question that is asked, it is hard to imagine a question more designed to elicit a testimonial response than "what happened?" That is the most common question asked of witnesses giving direct testimony in court.

So what about John's statement itself? The California Supreme Court's second reason for saying that the statement is nontestimonial is because it was made for the purpose of medical treatment. The court never explains, however, how the part identifying the person who allegedly cut him was necessary for medical treatment. Nor can one perceive any reason why it might have been relevant to such treatment.

Third, the California Supreme Court asserted that John's statement was nontestimonial because it was given in an "informal" setting. Davis did not dispute that formality is essential to a testimonial utterance. But all we really know on that score is that a statement to a responding police officer or 911 operator is sufficiently formal to be testimonial, while a "casual remark to an acquaintance" is not. It seems to me that answering a doctor's question "what happened" is much closer to the former than the latter. Put another way, it does not seem at all strange to say that John was bearing witness when he told the doctor what happened. But we'll have to await further word from the Supreme Court on this.

Finally, it is worth noting that the California Supreme Court explicitly rejected the view that a statement is testimonial if a reasonable declarant would have expected it to be used prosecutorially. This holding diverges from several post-Davis cases that continue to use this formulation outside of the context of "interrogations." Again, the Supreme Court will have to answer this question sooner or later.

Jessica Smith's "Emerging Issues": Links Available

Jessica Smith, a member of the faculty at the School of Government of the University of North Carolina at Chapel Hill, has given me permission to post on the blog links to her monograph, Crawford v. Wsahington: Confrontation One Year Later (2005), and to a supplement about to be published, Crawford v. Washington: Emerging Issues in Confrontation Litigation. These are intended primarily to be practical guides for North Carolina judges, but I think they will be very valuable resources for anybody dealing with Crawford issues. Posting this link does not mean that I vouch for any statement made in these documents, of course.

"A Sixth Sense About Criminal Trials"

The ABA Journal E-Report has a nice write-up on Jeff Fisher, especially emphasizing Crawford and related issues, and discussing how the textually-based arguments that Jeff has made in Crawford and other cases have appealed both to Scalia and to the Court's liberals. You can get to the article by clicking here. There is one, uh, slight factual error in reporting, which I will leave to readers to find. It won't actually be all that hard.

Deepening Conflict on Lab Reports

Yesterday, in State v. March, 2007 WL 828156, the Missouri Supreme Court joined other jurisdictions in holding that a lab report prepared for use in prosecution is testimonial. I believe this result is clearly correct, and appropriately the court did not appear to endure much angst in reaching it. The decision sharpens the conflict among the states. The Supreme Court will have to resolve this matter, and in my view the sooner the better.

A Challenge to Maryland v. Craig

Lawyers from Baker Botts in Washington, D.C., have filed a well-crafted petition for certiorari asking the Supreme Court to take another look at Maryland v. Craig, 497 U.S. 836 (1990), which allows child witnesses in certain circumstances to testify against an accused outside his presence by electronic means. The petition presents a conflict among states in interpreting Craig – some states treat potential trauma to the child as sufficient in itself to justify such a procedure; others require that the trauma threaten to impair the child’s ability to testify. More broadly, the petition asks the Court to overrule Craig.

In arguing for adoption of the testimonial approach in Crawford, Jeff Fisher as counsel for Crawford and I as a supporting amicus emphasized that to do so would not require the Court to rethink Craig; the issues were orthogonal. Similarly, in an amicus brief written a few years earlier on behalf of the ACLU in Lilly v. Virginia, 527 U.S. 116 (1999), Margaret Berger and I wrote, referring in part to Craig, “Our focus here is on when confrontation is necessary, not on what is required for confrontation.” That having been said, I think that the majority opinion in Crawford bears much more affinity to the dissent written on behalf of four justices in Craig than it does to the majority opinion in Craig. This cannot be altogether surprising: Justice Scalia wrote the majority opinion in Crawford and the dissent in Craig, and Justice O’Connor, the author of the majority opinion in Craig, was one of the two justices who did not join the majority opinion in Crawford. Not only does Craig rely heavily on Ohio v. Roberts, 448U.S. 56 (1980), which is now discarded, but the Craig majority opinion reflects a balancing approach to the confrontation right, reflective both of the Roberts era and of Justice O’Connor, while Crawford reflects a more categorical approach favored by its author.

There can be little doubt that Justice Scalia, whose Craig dissent was bitter and pungent, would like to see that decision overruled. Would he now have the votes to do so? Will the Court decide that this is the time to consider the issue? We’ll know before long.

Confrontation Milestone?

I'm still working on my posting on Whorton v. Bockting. But meanwhile, Roger Kirst of the University of Nebraska College of Law, who follows confrontation developments very closely, has sent this message, under the title Confrontation Milestone?, to the evidence professors' e-mail list, and has given me permission to post it here.

For those who follow the details of confrontation doctrine in the Supreme Court there was an interesting denial of certiorari yesterday.

The high-water mark for the Supreme Court's reliability test as a substitute for actual confrontation was New Mexico v. Earnest, 477 U.S. 648 (1986). That was a GVR in which then-Justice Rehnquist added a concurring opinion joined by three others that asserted that Lee v. Illinois had adopted the reliability theory. This was also the last confrontation case the Court decided before Justice Scalia joined the Court. On remand the New Mexico Supreme Court retreated from its earlier conclusion that it was a violation of the right of confrontation to convict on the basis of the confession of an alleged accomplice the defendant could not cross-examine; they declared the confession "reliable" and affirmed the conviction.

After an unsuccessful further petition for certiorari and an unsuccessful postconviction challenge in federal court, Earnest did nothing until Crawford was decided. At that point Earnest filed a postconviction challenge in state court (represented by Tom Sullivan, now at Arkansas-Little Rock). Eventually the state trial court ordered a new trial and the New Mexico Supreme Court affirmed in an opinion that said basically "we were right the first time and Crawford now shows the Supreme Court was in error from 1986-2004 with its reliability test." [119 P. 3d 144 (N.M. 2005)] This time the State filed the petition for certiorari, arguing that Crawford was not retroactive. [New Mexico v. Forbes, No. 05-644]

The Supreme Court was holding No. 05-644 while it heard Whorton v. Bockting. During the wait the State tried to get an untainted conviction, but the alleged accomplice would not testify.

The Court's holding last week that Crawford is not retroactive might seem to suggest at least a GVR in No. 05-644, but that did not happen. Nothing more than a denial of certiorari, meaning the case is now over.

The reliability test was only part of Roberts, so we do not need to revive the debate about when Roberts died, but the decision yesterday is a reminder of how much confrontation doctrine keeps changing.

And Phil Pucillo, of Ave Maria School of Law, now but regrettably not for long of Ann Arbor, had this to say in response (and has also given me permission to post it here):

I just glanced at the New Mexico Supreme Court's opinion. What's fascinating is that the court did not claim to be applying Crawford retroactively. Instead, it emphasized that Crawford did not even announce a new rule, a proposition that the U.S. Supreme Court squarely rejected in Bockting.

It's almost as if the U.S. Supreme Court let this one go because explaining to Mr. Earnest why he won't get a new trial after all would be too embarrassing: "Well, sir, what happened was that the New Mexico Supreme Court was absolutely correct back in 1985 when it reversed your conviction and ordered a new trial for you. But you never got that new trial because, at that time, we thought that the New Mexico Supreme Court was wrong. Well, it turns out that we were wrong, but it took us another 20 years or so to figure it out. But see, the problem now is that you can't receive the benefit of our recent change of heart, because your direct appeal concluded a long time ago. So even though the New Mexico Supreme Court's 1985 decision to grant you a new trial was right on, you're pretty much stuck with our incorrect decision saying that you can't have a new trial. Sorry aboutthat!"

A Flurry of Forfeiture Decisions

There have been three significant decisions on forfeiture within the last few weeks: U.S. v. Martinez, 2007 WL 489217 (D.C.Cir. Feb 16, 2007); State v. Jensen, 2007 WL 543053, 2007 WI 26, Wis., February 23, 2007 (No. 2004AP2481-CR.), and, just today, People v. Giles (Ca. Mar. 5, 2007). I hope to post comments on all three (and also on Whorton) within a few days.

Crawford not retroactive

I'm on vacation but Ive just found out that the Supreme Court held today in the Whorton case that Crawford isn't retroactive. No surprise -- the consequences would have been enormous, more than the Court probably wanted to deal with. I hope to post something mroe substantive about the deciswion within the next few days.

Certiorari denied in Craig

Today the Supreme Court denied the petition for certiorari I filed in Craig v. Ohio. Obviously, I am disappointed. I think the issues that the petition raised are important ones that the Court must ultimately resolve, and I hope the time will come sooner rather than later. Most narrowly, the case posed the question whether an autopsy report in a murder case is testimonial; more broadly, the question is whether a state can immunize from Confrontation Clause review a statement routinely made by government officials in anticipation of litigation by characterizing it as a business or public record. Last week, the Court turned away another autopsy case, seeking review of the Second Circuit's decision -- one I thought was atrocious -- in United States v. Feliz. And several weeks ago, the Court rejected the petition in Campbell and Pinks v. North Dakota, which could have let it resolve the broader question and which more directly posed the question of whether the Confrontation Clause can be satisfied by giving the accused the oportunity to subpoena the witness. It may be that the Court saw problems in each of these cases making it a poor vehicle for review. It may be that the Court wants the lower courts to decide more "business records" cases before addressing the area. It may be that the Court wants to let more time go by before returning to the whole Crawford area altogether. And perhaps there are other explanations as well; as a rule, the Court gives no explanation for cert denials. These denials should not be read as an endorsement of the decisions of the lower courts, which I continue to believe are plainly wrong. There are lots of Crawford-related issues that only the Supreme Court can resolve, and eventually it will do so.

Harmless Error and Confrontation

It is clear that Confrontation Clause errors are subject to harmless error analysis. Delaware v. Van Arsdall, 475 U.S. 673 (1986). But it seems to me that there is a significant limitation on how the harmless-error inquiry may be conducted: If a testimonial statement is admitted against an accused without his having had an opportunity for confrontation, the court should not speculate that cross-examination would have been ineffective. On the contrary, the court should address the harmless-error issue by assuming that cross, if it had not been denied, would have been maximally effective.

Where prosecution testimony has been admitted in violation of the confrontation right because the accused never had an opportunity to be confronted with the witness, the reviewing court should consider the case as if that testimony had never been admitted, and declare the error harmless if it is confident beyond a reasonable doubt that even without that evidence the accused would have been found guilty. Van Arsdall, supra, 475 U.S. at 682 n.5 (describing Harrington v. California, 395 U.S. 250 (1969): “Noting that the wrongfully admitted evidence was cumulative and that the untainted proof of the defendant's guilt was overwhelming, the Court concluded that the error was harmless beyond a reasonable doubt.”). One might complain about the exercise that it involves some guesswork about what the jury would have done, but basically it is a practical and necessary judgment: “The tainted evidence is of trivial importance, and a jury that found the accused guilty with that evidence would not fail to do so without that evidence.”

But suppose that after considering the untainted evidence in this way the court cannot say with the requisite confidence that the outcome would have been the same without the tainted testimony. Then it is not appropriate for the court to say, in effect, “The error is still harmless, because if there had been an opportunity for cross-examination of that witness it would have done little or any good.” To do that would be not merely to assess what the jury likely would have done with a slightly altered body of evidence but to speculate what an absent witness would have said if he had faced the adverse questioning that the Constitution guarantees the accused. The Confrontation Clause is meant to prevent the need for speculation, by inducing the prosecution to produce the witness for confrontation.

Van Arsdall supports this interpretation. There, the accused had not been denied the confrontation right altogether; he had cross-examined the witness in question, but the trial court had erroneously prevented him from pursuing a line of questioning that might have revealed his bias; thus, the immediate error was not admitting prosecution evidence but excluding defense evidence. Extending Harrington to this situation, the Court held that the error was subject to harmless-error analysis, and so remanded to the state supreme court, perhaps with a hint that because the testimony of the witness in question was cumulative and peripheral it would be entirely proper to deem the error harmless. The Court said: “The correct inquiry is whether, assuming that the damaging potential of the cross-examination were fully realized, a reviewing court might nonetheless say that the error was harmless beyond a reasonable doubt.” (emphasis added)

Thus, the reviewing court must act under the assumption that the accused gets the full “damaging potential of the cross-examination.” It is appropriate for the reviewing court to say, “Even under that assumption, the error doesn’t matter, because the most that cross could have done would have been to nullify the testimony of this witness, and the case was still overwhelming.” Possibly, but much more dubiously, it is appropriate for the court to say, “Even if the cross-examination that the accused says he would have conducted had gone entirely his way, achieving its full damaging potential by showing the witness was willing to lie in favor of the prosecution, the testimony would have had some value because even a dishonest witness could not likely have told this story if it were not true, and given that the error is harmless beyond a reasonable doubt.” But what the court should not do is say, “The error is harmless because if the accused had had an opportunity for cross-examination the answers of the witness would in all probability have been unhelpful to the accused.”

Brief in Opposition and Reply Brief in Craig v. Ohio

You can read the state's brief in opposition to the petition for certiorari in Craig v. Ohio by clicking here, and my reply brief in support of the petition by clicking here. To see my prior post describing the case, and the petition, click here. I expect we'll know by the end of the month whether the Court will take the case.