American Prosecutors Research Institute on Crawford

The American Prosecutors Research Institute has devoted an issue of its journal The Voice to Crawford v. Washington. As one might expect, it is an attempt to limit the impact of Crawford as tightly as possible, and in particular to preserve what it euphemistically refers to as "evidence based prosecutions" -- that is, prosecutions without evidence being given by the complainant subject to confrontation -- as they were allowed before Crawford.

Adam M. Krischer makes the rather extreme assertion that "domestic violence almost always involves forfeiture." I agree that forfeiture does occur frequently in domestic violence cases, and even that "the domestic violence itself may have procured the victim's unavailability." But the burden of proving forfeiture must remain squarely on the prosecution's shoulders. There should not be an attempt to create anything like a presumption that if the case involves domestic violence the accused has forfeited his confrontation right, or even more modestly that if the complainant is no longer willing to testify in accordance with her initial statement then intimidation is the cause. Mr. Krischer goes into considerable, and useful, detail as to how prosecutors may prove forfeiture. But he offers little guidance on what prosecutors can and should do, notwithstanding intimdation, to increase the chance that the complainant will testify live subject to cross-examination -- though he does say, "A properly supported victim, connected to services and counseling, is more likely to be cooperative than not, and a cooeprative victim, willing and able to testify, makes Crawford a moot point." When prosecutors have believed they needed to secure live testimony by the complainant or face dismissal of the charge, some have not hesitated to exert considerable pressure, including jailing the complainant. Whatever the merits of that approach, courts should not be so ready to reach a conclusion of forfeiture that prosecutors will have little or no incentive to preserve the accused's right to confront.

The issue also includes a list of predicate questions, prepared by Cindy Dyer, Chief of the Family Violence Unit of the Dallas County District Attorney's Office, to ask when examining at a preliminary hearing or at trial a police officer who has taken a statement from a complainant. If the complainant made statements in response to questions by an officer, then the prosecutor is advised to include the following in the examination:

1. Were the statements taken "during the course of an interrogation"?

2. What was the purpose of your questions?

3. Were your questions to her an interrogation or merely part of your initial investigation?

4. Were these questions asked to determine if a crime had even occurred?

Apparently the thought is that if the desired answers are received -- and even apart from the leading nature of three of these questions we can be sure that the police will understand what the desired answers are -- then the complainant's statements should not be deemed to be testimonial. For the reasons already discussed in my post on The Interrogation Bugaboo, this reflects a deficient understanding of the confrontation right.

Cross-examination by Another Party

The Wisconsin Supreme Court issued an opinion today, in State v. Hale, holding that the trial court had violated Crawford v. Washington in introducing against Hale testimony from the trial of another defendant, Jones, given by a witness, Sullivan, who was unavailable by the time of Hale's trial. Though this conclusion should be obvious, the court took a long time to reach it; when the court did get to the issue, though, it did not mince words, holding broadly that "prior testimony may be admitted against a criminal defendant only when that defendant has had a prior opportunity to cross-examine the witness giving that testimony."

Interestingly, Wis. Stat. § 908.45(1), the Wisconsin hearsay exception for former testimony, unlike Fed. R. Evid. 804(b)(1), does not prescribe this rule – i.e., that cross-examination by another person cannot act as a proxy for cross-examination by the defendant himself. Instead, it removes the hearsay bar from prior testimony so long as there was an opportunity for cross-examination by a party "with motive and interest similar to those of the party against whom [the statement is] now offered." One might have thought it obvious in this case that this rule was not satisfied – Sullivan's testimony put the murder weapon in Hall's hands rather than in Jones' – and one justice wrote a concurring opinion that emphasized this point. But the court's opinion instead addressed the constitutional question, and the rule comes through clearly: The right of confrontation is a personal one, and no matter how similar another party's motive may have been, the opportunity of that party to cross-examine does not satisfy the right of the accused.

Ultimately, the court determined that the error was harmless, and much of the justices' discussion concerned this question. One justice also wrote a long elaboration on forfeiture, in which two other justices joined, but no issue of forfeiture was presented and this opinion seems mainly to have been for the edification of the bar.

Mendez on Crawford

Miguel A. Méndez of Stanford Law School has published Crawford v. Washington: A Critique, 57 Stan. L. Rev. 569 (2004). The article covers a great deal of ground. Here are just a few points that jumped out at me, with my comments.

Prof. Méndez emphasizes the difficulties of determining, under either an objective or subjective test, the declarant's anticipation of whether a statement will be used as trial evidence. I think that probably prospective use at trial should not be the key; the anticipation of prospective use in the process of criminal justice should be enough, and with respect to child witnesses the anticipation of use in some punitive process should be enough. More on this in later posts.

He also complains about the Crawford Court's leaving the situation unclearwith rspect to dying declarations. But I think virtually all courts to deal with these since Crawford have let them in, though there is still uncertainty whether the traditional hearsay exception should be followed or rather, as I believe, this is a matter that should be resolved by forfeiture doctrine; see my prior post on forfeiture anddying declarations.

Prof. Méndez contends, as I understand it, that if the defendant's negligence causes the witness to be unavailable that is enough for forfeiture. That is a plausible view, but it is bound to be controversial.

He also says, "Whether hearsay is admissible against the accused under Crawford depends initially on the forum's rules of evidence." If I understand this statement correctly, I believe it is incorrect. It seems to be based on his perception that Crawford excluded business records from the category of testimonial statements. It is true that Crawford, in its historical analysis, spoke of business records as by their nature being non-testimonial. But I believe that reference was meant to be descriptive: Business records in general are not testimonial. I do not believe that Crawford meant to state a rule that, no matter what classes of dcouments a state defined to be business records, they would be deemed non-testimonial as a matter of confrontation law. A police record of an interview with a witness is inadmissible under Crawford no matter how the state characterizes it. Prof. Méndez and I do not disagree on the proper outcome here; our difference is just that he treats the Court's reference to business records as being more significant than I do.

With respect to conspirator statements, Prof. Méndez argues that often the defense has a strong interest in cross-examining the declarant, but these statements are not covered by Crawford. True, but in my view this is not a confrontation problem -- see the post on Non-testimonial Statements -- and Crawford essentially leaves the treatment of these statements unchanged.

Non-testimonial Statements

Crawford v. Washington left open the status under the Confrontation Clause of non-testimonial statements. While saying that the primary focus of the Clause is testimonial statements, it added:

Where nontestimonial hearsay is at issue, it is wholly consistent with the Framers' design to afford the States flexibility in their development of hearsay law--as does [Ohio v.] Roberts, and as would an approach that exempted such statements from Confrontation Clause scrutiny altogether.

I believe the law of the Confrontation Clause would be improved if the Court were to make clear that the Clause has no application to non-testimonial statements.

As a matter of principle, this is the right outcome. If a person is not acting as a witness within the meaning of the Clause -- and I take the definition of witness as being one who makes a testimonial statement -- then she simply is not covered by the Clause. Perhaps her statement ought to be excluded, and perhaps even (occasionally) it should be excluded as a constitutional matter, but not on the ground that to admit it would be to violate the accused's right "to be confronted with the witnesses against him."

Furthermore, I believe that the confrontation right, where it does and should exist -- that is, with respect to testimonial statements -- is more likely to be robustly protected if the right is not stretched beyond its proper domain. A large part of the problem before Crawford was that the confrontation right depended on hearsay law, and hearsay covers such a vast body of statements that a categorical rule of exclusion was not feasible. This made for a very complicated body of law, and obscured the nature of the confrontation right and its value. The basic rule of Crawford -- if the statement is testimonial, the accused has a categorical right to confront the maker -- is simple and relatively easy to articulate and understand, expressing a principle that has great intuitive appeal. If the rule is complicated by adding in a lesser adjunct -- that if the statement is not testimonial, the confrontation right may still require exclusion if the statement fails a multi-step test of reliability -- it will inevitably lose some clarity. And eventually, I believe, this would weaken the rule at its core.

Finally, I don't believe the Roberts test provides any substantial protection to defendants with respect to non-testimonial statements. Even before Crawford, the Roberts standard did not keep out many statements that were not testimonial. And since Crawford I don't know of a single case where the application of Roberts to non-testimonial statements has made a difference. (If there are any such cases, I would love to know; please post a comment!) This is not particularly surprising. It is highly unlikely that a court would hold that (1) a given statement is non-testimonial for Crawford purposes, (2) it satisfies the rule against hearsay, either because it (a) fits within an exception or (b) is supported by sufficient guarantees of trustworthiness to warrant admissibility, and yet (3) it is barred by the Roberts test because it is unreliable, neither (a) fitting within a "firmly rooted" hearsay exception nor (b) supported by sufficient guarantees of trustworthiness that are of a form satisfactory for constitutional purposes (i.e., circumstances surrounding the making of the statement, as opposed to corroboration). I am not saying that this is conceptually impossible, just that it hardly ever happens. I suppose it could more often happen that a state court would reach conclusion (2) and that a federal habeas court, without a basis for looking behind that decision, would yet reach conclusions (1) and (3). But I don't know of any cases fitting this mold, either, and I do not believe they are plentiful.

What does tend to happen, over and over again, is that courts determine that a given statement is not testimonial and then go through the whole Roberts analysis only to conclude -- surprise, surprise -- that the statement is sufficiently reliable to warrant admissibility. It is at least prudent for lower courts to go through this exercise. Arguably, it is required. The Supreme Court has said:

If a precedent of this Court has direct application in a case, yet appears to rest on reasons rejected in some other line of decisions, the [lower courts] should follow the case which directly controls, leaving to this Court the prerogative of overruling its own decisions.

Rodriguez de Quijas v. Shearson/American Express, Inc., 490 U.S. 477, 484 (1989); accord, Agostini v. Felton, 521 U.S. 203, 237 (1997). But I believe that the Court's treatment of Roberts in Crawford creates a situation different from those present in Rodriguez deQuijas or Agostini. (In Rodriguez deQuijas, it was a shifting attitude towards arbitration that signaled the probable overruling of a prior case; in Agostini, five members of the Court had, in separate opinions, indicated that the prior decision should be overruled.) Crawford explicitly overruled Roberts so far as it applied to testimonial statements. With respect to non-testimonial statements, the Court recognized that its holding "casts doubt" on Roberts, and, while recognizing that it "need not definitively resolve whether [that part of Roberts survives" the Crawsford decision, it explicitly said, as noted above, that exempting such statements from the Confrontation Clause altogether would be "wholly consistent with the Framers' design," as would this aspect of Roberts. A majority opinion of the Court has therefore explicitly given Roberts , so far as it applies to non-testimonial statements, no greater standing than a rule rendering the Confrontaiton Clause inapplicable altogether to such statements. At this point, it seems to me it could be appropriate for a lower court to choose the latter rule, which is certainly more in accordance with the logic of the Crawford opinion than continued application of Roberts.

But assuming the lower courts do not wish to be so adventurous, they will continue to go through the charade of analyzing non-testmionial evidence under Roberts and then deciding it is admissible. I suppose this is a harmless exercise, apart from the waste of time and energy it entails. It may go on indefinitely unless a prosecutor declines to argue that a given non-testimonial statement should be deemed reliable under Roberts and instead contends simply that Roberts should not apply to non-testimonial statements.

Assuming that the Confrontaiton Clause does not apply to non-testimonial statements, they are , of course, still subject to the rule against hearsay, and occasionally admission may even violate the accused's general right to due process. I will not attempt here to articulate a general theory of when the due process right might require exclusion of a non-testimonial statement, but here are some considerations: How important is the evidence to the prosecution? How likely is it that cross-examination of the declarant would be significantly informative? How pejudicial is the evidcne likley to be absent cross-examination? Did the accused have an opportunity to cross-examine the declarant? Is, or was, the prosecution better able than the defense to produce the declarant as a witness, either at trial or at a deposition? If the declarant is unavailable, did the conduct of either party contribute to that unavailability?

The Interrogation Bugaboo

Since Crawford v. Washington, some courts have said that a statement is not testimonial unless it is made in response to governmental interrogation. E.g., People v. Bryant, 2004 WL 1882661 (Mich. App. 2004); United States v. Webb, 2004 WL 2726100 (D.C. Super. Dec. 22, 2004). And indeed, some have gone further, refusing to characterize a statement as testimonial unless it meets a restrictive definition of interrogation as "structured police questioning." State v. Barnes, 2004 WL 1773301 (Me. 2004); Fowler v. State, 809 N.E.2d 960 (Ind. App. 2004), transfer granted (Ind. Dec. 10, 2004); People v. Newland, 775 N.Y.S.2d 308 (N.Y. App.), leave to appeal denied, 3 N.Y.3d 679 (2004). This idea has begun to distort police practices, as police try to act in such a way that prosecutors can later argue that statements made to the police were not in response to interrogation.

I believe that te whole supposed interrogation requirement is entirely mistaken. Interrogation is a factor that in some contexts supports an inference that the statement is testimonial, but the statement may be testimonial even though it is not in response to interrogation. In this post, I will not contend against the less extreme proposition that only if a statement is made to a government agent can it be testimonial. I believe that proposition is also erroneous, but I will address it in a later post.

Those who contend that interrogation is necessary for a statement to be deemed testimonial have language they can point to in Crawford, though it is quickly apparent that the language does not really support them. Sylvia Crawford's statements were made in response to police interrogation, and the Court held that, whatever else the category of testimonial statements might include, statements made in response to police interrogation certainly are. Here are the passages in question, with emphasis added in each case:

The Clause's primary object is testimonial hearsay, and interrogations by law enforcement officers fall squarely within that class.

Statements taken by police officers in the course of interrogations are also testimonial under even a narrow standard. Police interrogations bear a striking resemblance to examinations by justices of the peace in England.

In sum, even if the Sixth Amendment is not solely concerned with testimonial hearsay, that is its primary object, and interrogations by law enforcement officers fall squarely within that class.

Whatever else the term covers, it applies at a minimum to prior testimony at a preliminary hearing, before a grand jury, or at a former trial; and to police interrogations. These are the modern practices with to the abuses at which the Confrontation Clause was directed.

There is no indication, then, that statements not made during formal testimnial events -- a preliminary hearing, grand jury or a former trial -- must be in response to police interrogation to be considered testimonial. The Court is very clear that it is merely listing a core class of testimonial statements, a class that plainly includes the statements at issue in the case, and is deciding no more than that these statements are testimonial. Left for another day is the question of what additional statements, if any, shall be considered testimonial. It is true that the Court left open the possibility that it will not consider any statements beyond this core class to be testimonial. Indeed, the fact that the Court took the care, in footnote 4, to offer some elaboration on the meaning of "interrogation" -- saying that it was using the term in a colloquial sense, that it did not have to choose among definitions, and that "Sylvia's recorded statement, knowingly given in response to structured police questioning, qualifies under any conceivable definition" -- confirms that the Court preserved the possibility that the term would in some circumstances be decisive. But that is as far as the Court went in this direction. It offered no intimation that a statement not made in response to interrogaiton would not be considered testimonial. And it certainly did not suggest that if a statement was not "knowingly given in response to structured police questioning" it would not be testimonial; it merely said that a statement meeitng that standard "qualifies under any conceivable definition."

So Crawford does not tell us that a statement must be in response to interrogation to be characterized as testimonial. And common sense tells us that there is no such requirement. Suppose that at trial a prosecutor gives an observer an opportunity to come to the front of the courtoom and then says, "Ms. Observer, I invite you to tell us what you know about this incident." After the witness does so, the prosecutor says, "Thank you. You may go." Of course, defense counsel objects because of a lack of confrontation. "But," says the prosecutor, "this was no witness. I did not subject her to any interrogaiton." The prosecutor is right that there was no interrogation, but of course we would expect the legal argument to be rejected sneeringly. What Observer was doing was testifying. It does not matter that her statement was not given in response to questions; nor would it matter whether she or the prosecutor took the intiative in arranging for her to give the testimony.

So now suppose the invitation comes not at trial but at the police station.: "Ms. Observer, if you care to make a statement, please feel free to do so. I will videotape it, and when we this perpatrator stands trial I will give the prosecutor the tape so that she can play it in front of the jury." I think it is equally obvious that a statement made in response to this invitation is testimonial. And now suppose an observer walks into the police station and says, "You don't know about a crime that has been committed, but I am now going to tell you, and I expect that you will then want to prosecute. Please record what I am about to say, because I expect you will want to use it at trial -- I do not like the idea of being under oath and having to answer questions by some aggressive defense lawyer." I cannot see a plausible basis on which this statement should not be deemed testimonial. Or suppose the observer walks in to the police station with an affidavit completed, describing the crime. Does anyone seriously contend that this is not testimonial?

Now, of course, the statements in these hypotheticals are less formal than in the usual case, in which a witness makes a statement to a police officer in the field, perhaps before the officer is confident that acrime has been committed. But, for reasons that I have already analyzed in a post called The Formality Bugaboo, formality is not required to render a statement decisive. If the declarant in that field situation understands full well that, once the officer receives the statement, it is likely to be used for prosecutorial purposes, it is testimonial. The declarant is creating evidence, and this critical reality is unaffected by the facts that the police officer was not confident until the moment that the statement was made that a crime had been committed, and that structured questioning by the officer was not necessary to secure the statement.

The bottom line is that if the declarant is making the statement with the reasonable anticipation of prosecutorial use, it is testimonial, even if it is made without quesitoning by government authorities or entirely on the witness's own initiative. Interrogaiton may, however, be a significant factor in indicating that the declarant did have this anticipation, because if the authorities are interrogating that is a factor that would often convey to the declarant the likelihood of prosecutorial use. But when the declarant is reporting a crime this factor is not necessary to characterize the statement as testimonial; she knows that she is conveying ot the authorities information about a crime, and presumably she understands that they will use that information to invoke the machinery of criminal justice. To hold that such a statement is not testimonial is merely to try to avoid Crawford because it makes prosecutions more difficult.

Statements by Government Agents

I hope soon to post a discussion of the question of whether a statement must be made to a government agent to be considered testimonial; I don't think it should be. Here, though, I will discuss an issue that arises less frequently but that is still very important, and also very intriguing – the extent to which statements by government agents should be considered testimonial for Crawford purposes.

Such statements may come into play in a variety of circumstances. First, I wil lconsider statements in government reports and certificates. Then I will address statements made by confidential informants on tape-recorded statements.

Reports and Certificates: In Napier v. State, 2005 WL 22858 (Ind. App. Jan. 6, 2005), the principal issue was the admissibility of a certificate asserting that a machine used to test the accused's blood alcohol content was in good working order. In United States v. Rueda-Rivera, 2005 WL 39763 (5th Cir. Jan. 10, 2005), the defendant was convicted of re-entering the United States without consent after having been deported and removed. The prosecution proved the lack of consent by introducing a Certificate of Nonexistence of Record (CNR), signed by an official in the records office of the Immigration and Naturalization Service that a search of INS files revealed no evidence that Rueda-Rivera had been granted permission to re-enter. The CNR was admitted on the testimony of a Border Patrol Officer who admitted that he did not know what type of search of records had been made before the CNR was completed. In each case, the court held that the document in question was not testimonial and that its admission did not violate Crawford. Neither opinion offered an extended argument. The Napier court was at least candid in raising what it regarded as “the unreasonable alternative” to its holding, “to have a toxicologist in every court on a daily basis offering testimony about his inspection of a breathalyzer machine and the certification of the officer as a proper administrator of the breath test.” The Rueda-Rivera court did little more than cite a previous, unpublished opinion, which was equally conclusory, and liken the document to a business record.

Both these decisions are wrong, in my view. In each case, the document should be considered testimonial. In each case, the practical consequences could be mitigated substantially.

The situation in Rueda-Rivera is especially clear. Plainly, the CNR was completed solely for purposes of this prosecution, to be used as evidence at the defendant's trial. To say that this is a routinely kept record is only to say that prosecutors and courts have developed a routine of violating the confrontation right in cases of this sort. (Of course, the status of a dcoument under a hearsay exception does not affect its standing under Crawford. Most routinely kept records are not testimonial, but that does not mean that a statement is not testimonial because it is a routinely kept record.) I suppose that if in assault cases police officers took to filing Certificates of Guilt the prosecution would argue that the elements of the crime could be proven by introducing such a certificate, without need for proof.

That being said, we nevertheless must appreciate the difficulty from the prosecution side. All the prosecution wants to prove on this aspect of the case is the absence of a particular type of record; the defendant presumably makes no attempt to prove that the record exists, or that in fact consent to re-enter was granted, but presumably he is not willing to stipulate to the absence of consent. In such a case, does this mean that the case can only be proven by sending a clerk from Washington to the place of trial, by giving the defendant the opportunity to go to Washington for a deposition, or by arranging the immigration files so that they can be searched from near the place of trial? Actually, it seems to me the more sensible solution would be to place on the defendant the burden of producing evidence that he did have consent to re-enter. Most deported aliens are not given consent to re-enter, and it is far easier to prove an existential statement (in this case, that there was consent) if true than to prove the corresponding universal statement (in this case, that there never was such consent) if that statement is true. Of course, to impose the burden on the defendant might require a change of law -- whether it would require a statute or could be accomplished by decisional law I do not know -- but there is no constitutional obstacle to doing that. As long as the law places the burden of proving lack of consent on the prosecution, the prosecution ought to prove the point in a constitutional manner.

Note, by the way, that Rueda-Rivera seems not to have raised a contention that the documents proving the original deportation were testimonial, and any such contention would have been far weaker. Even though such documents are prepared in the knowledge that some deportees will attempt to re-enter and will be prosecuted, with documents proving deportation then introduced against them, such documents are not prepared in the context of the crime later charged, or of a course of conduct culminating with that crime. I think these documents probably should not be considered testimonial. (If they were, the consequences would not be devastating; the defendant could be asked to sign an acknowledgment of deportation and if he refused then offered cross-examination at the time the documents were completed.)

The certificates in Napier present what is in some sense an intermediate case: They were completed before the particular crime in question was committed, but they were completed in the clear anticipation that they would be used in the prosecution of some commission of the crime of driving with an elevated blood alcohol level. Indeed, one of the certificates specified that "[t]he original Letter of Certification, issued by the State Department of Toxicology, must be kept on file in the office of the Clerk of the Circuit Court and may be duplicated as needed for use in Court." The Napier court argued weakly that "the information contained in the certificates does not pertain to the issue of guilt." But of course it did. The only proof that Napier had an elevated blood alcohol level was the results of the test performed with the machine in question; unless the jury believed that machine to be accurate, it could not have voted to convict him. Beyond that, the only argument offered by the court was the one noted above, that it would be "obviously impractical" to require a toxicologist in court every case of this type testifying about the inspection of the machine and his own certificaiton.

That is an inadequate argument. It is always cheaper and easier to do without confrontation, and if the courts have gotten used in a given context to doing without it then the requirement that the right be respected will always seem expensive. If need be, there could be courtroom textimony about the machine, just as there is courtoom testimony about fingerprint or DNA results. In most cases, we could expect the testimony to be rather perfunctory.

Testimony at trial would not ordinarily be necessary, though. The prosecution could offer the defendant an early opportunity to take the deposition of the testing officer. In many cases, I suspect, the defendant would not bother. If the trial were held at a substantially later time, I think it would be appropriate to be rather generous to the prosecution in determining that the officer, having already been cross-examined, was not available to be a trial witness; in many cases, after all, the office would no longer have any recollection of the particular testing. Moreover, I am not sure that in this context the prosecution really should have to present affirmative evidence that the machine is working; in the absence of evidence to the contrary, I think the jury could conclude that the machine was in working order, and I believe procedural rules could properly provide that if the defendant wants to argue or present evidence to the contrary he must raise the contention early, well before trial.

At least two courts in New York have done better with this type of issue than did Napier and Rueda-Rivera. People v. Hernandez, a decision at the trial level, involves a latent fingerprint report prepared by a police officer who was retired and living in Ireland by the time of trial. The court held that the report was testimonial and so could not be introduced absent cross-examination. It relied on a decision of the Third Department, People v. Rogers, 8 AD3d 888 (3rd Dept. 2004), made to similar effect with respect to a report of blood alcohol content (of the alleged vicitm of rape, the question being whether she was capapble of consent). These decisions seem clearly correct to me. Rogers says: "Because the test was initiated by the prosecution and generated by the desire to discover evidence against defendant, the results were testimonial." I do not believe that initiation by the prosecution is essetnial, though of course it is indicative of the evidentiary anticipation. In Hernandez, the state tried to distinguish Rogers on the ground that the report was not initiated by the prosecution, but hte court properly rejected the attempt, as well as the assertion that because the police did not know who they were going after when they began the fingerprint test the statement could not be testimonial: "The fingerprints in question were not taken simply for administrative use. They were taken with the ultimate goal of apprehending and successfully prosecuting a defendant." At the same time, Hernandez presents the question of what to do when the government agent who makes a report is not likely to be available at trial. In most cases, the simple answer is to provide an opportunity for pre-trial confrontation. That is relatively simple when the defendant is in custody. If the defendant is identified but not in custody, I think it would suffice to give the defendant proper notice of the intention to use the report and of his opportunity to cross-examine; of course, there may be litigation over what constitutes proper notice. If the defendant is not yet identified, then there is a larger problem; I doubt that it would suffice to give notice by publicaiton to the person whose fingerprint matches a given description.

Tape-Recorded Statements by Confidential Informants: At least two cases of which I am aware (United States v. Hendricks, and United States v. Joval, recently argued in the Nionth Circuit) pose this situation: A confidential informant (CI) engages in a tape-recorded conversation with participants in an alleged conspiracy. By the time of trial, the CI is dead. The accused has not had an opportunity to cross-examine the CI, and there is no showing that the CI is dead as a result of wrongful conduct by the accused. The prosecution offers the tape.

Hendricks, I think, got the matter basically right. First, it acknowledged "appeal" to the ruling of the district court that the statements by the CI were testimonial: "Insofar as they contain the statements of CI Rivera, the conversations reasonably could be categorized as involving statements that Rivera expected to be used prosecutorially; obtaining evidence for the prosecution is, after all, the raison d'être of being a confidential informant." But then it contrasted the case with United States v. Bourjaily, 483 U.S. 171 (1987), which allowed a conversation between a co-defendant and a CI; Crawfordpreserves Bourjaily, regarding the statements there as non-testimonial. The Hendricks court acknowledged a key difference: "[A]lthough the Bourjaily Court addressed the Confrontation Clause implications of the admission of conversations with a government informant, it focused on the non-informant half of the conversation." Hendricks correctly concluded that Crawfordposes no barrier to the admission of statements made to the CI by the accused or by his conspirators. These statements, after all, are not made in contemplation of creating evidence. As for the statements by the CI, the Government disclaimed an intent ot introduce them for their truth. Rather, they were offered to "put the statements of the other parties to the conversations into perspective and make them intelligible to the jury and recognizable as admissions." (Internal quotaiton marks omitted.) For that purpose, the court held the statements admissible. The court did not reveal enough of the facts to allow a judgment of whether the standard was properly applied, but its articulation of a governing standard seems appropriate:

We thus hold that if a Defendant or his or her coconspirator makes statements as part of a reciprocal and integrated conversation with a government informant who later becomes unavailable for trial, the Confrontation Clause does not bar the introduction of the informant's portions of the conversation as are reasonably required to place the defendant or coconspirator's nontestimonial statements into context.

__________________

In short, many statements by government agents seem clearly testimonial, and some of them pose interesting procedural challenges. It may be that in some cases careful adherence to the confrontation right in this context will create substantial expense and inconvenience. But this is not inevitably so. And courts should not take the disingenuous path of least resistance by declaring that statements by governement officials are not testimonial even though such statements were clearly issued for prosecutorial use.

Victimless Prosecutions

Andrew J. King-Ries, who has extensive experience prosecuting domestic violence cases with the King County (Washington) Prosecutor’s Office, and who is now a visiting assistant professor at University of Montana School of Law, has published Crawford v. Washington: The End of Victimless Prosecution?, 28 Seattle U. L. Rev. 301 (2005). He contends broadly that Crawford should be applied in such a way that it does not interfere with victimless prosecutions of domestic violence. Put another way, he thinks that Crawford should not interfere with admission of 911 calls and statements to responding officers. He disagrees quite sharply with some of the conclusions that Bridget McCormack and I reached in Dial-In Testimony; I won’t attempt a point-by-point rebuttal here.

One basis for his conclusion is the view that in the domestic violence context these statements are not testimonial, “particularly under the narrowest definition of ‘testimonial’ proposed in Crawford.” I suppose it is true that under the narrowest definition most of these statements are not testimonial – but for reasons I have begun to argue on this blog, the narrowest definition is inadequate. Many of these statements – I am avoiding making a universal statement – are made with the anticipation that they will assist in prosecution for crime, and they should be considered testimonial.

Second, he contends that “the historical record supports the proposition that testimonial statements should not include excited utterances, present sense impressions, and statements to medical personnel.” It is difficult to square this discussion with footnote 8 in Crawford, which emphasizes how narrow the exception for spontaneous declarations was at the time of the framing of the 6th amendment, if the exception existed at all. I think it is clear, and Crawford regards it as clear, that at the time of the framing courts did not use an excitement of the speaker theory to allow use at trial of a statement made to an official describing a crime that had been committed on the speaker some time before.

Third, Prof. King-Ries argues that policy considerations support allowing victimless prosecutions. Part of his argument here is that the assailant’s wrongdoing may have intimidated the victim from testifying. That is often so, and it should be dealt with as a matter of forfeiture doctrine, which he acknowledges but barely addresses. To the extent that the argument is simply that domestic violence is a terrible crime and it should be prosecuted effectively, I offer two quick responses: First, it still must be prosecuted constitutionally – we cannot, for example, do away with the jury if we come to the conclusion that jurors are letting DV assailants off too readily – and practices that some courts have gotten used to in recent years have ignored the confrontation right. Second, prosecutors should consider more often offering an early, pretrial opportunity for confrontation.

The Formality Bugaboo

Some cases have indicated that a statement cannot be considered testimonial for purposes of the Crawford inquiry unless it was made formally. One such case is People v. Jimenez, 2004 WL 1832719 (Cal. App. 2d Dist. Aug. 17,2004), which has been mentioned in comments on this blog; another, in dictum, is the recent decision of the Second Circuit in Mungo v. Duncan, 2004 WL 2988301 (Dec. 28, 2004). I believe this view represents a misunderstanding of Crawford, and of the basic approach to the confrontation right that Crawford reflects.

This is not to say that courts adopting a formality rule can find no language in Crawford to cite in their support. First, drawing on a definition given by Noah Webster, Justice Scalia wrote that testimony “is typically ‘[a] solemn declaration or affirmation made for the purpose of establishing or proving some fact.’" Second, Justice Scalia then offered this contrast: “An accuser who makes a formal statement to government officers bears testimony in a sense that a person who makes a casual remark to an acquaintance does not.” Third, one of the three formulations of the class of testimonial statements presented by Justice Scalia is the one adopted by Justice Thomas (with Justice Scalia himself joining) in his separate opinion in White v. Illinois, 502 U.S. 346, 365 (1992): "extrajudicial statements ... contained in formalized testimonial materials, such as affidavits, depositions, prior testimony, or confessions."

Even on their face, none of these three passages adopts a formality rule. The Court did not say that testimony must be a solemn declaration; it said that it typically is such a declaration. The two polar categories, “a formal statement to government officers” and “a casual remark to an acquaintance,” plainly do not exhaust all possibilities, and so presenting these two does not indicate where the boundary between testimonial and non-testimonial lies. As for the Thomas formulation, it is only one of three alternatives presented by the Court, and the only one that includes a formality rule. Moreover, it is not clear whether Justice Thomas regards confessions as being a subset of “formalized testimonial materials”; if so, it is not clear why, because confessions can be very informal, and if not, it is not clear why the two sets, “formalized testimonial materials” and “confessions,” should be deemed to constitute the overall class of testimonial statements.

In short, nothing in Crawford compels the conclusion that only formal statements can be deemed to be testimonial. And courts should not adopt such a rule, most importantly because it makes no sense. Consider this exchange (I could make the same point by using the actual statements made by the informant in the decision of the Sixth Circuit in United States v. Cromer, 2004 WL 2711130 (6th Cir. Nov. 29, 2004), which rejected a formality requirement; see my posting on Cromer on this blog on December 9):

Police Officer: Please have a cup of coffee and make yourself comfortable.
If that chair is too hard, please let me know and I’ll get you a cushion.
Witness: Thanks so much. The chair is fine, but I’d love some milk if you
have it.
Officer: Sure. Here you go. You know, I’m collecting evidence for
the trial of Suspect on robbery charges. I know you’ll find it inconvenient and
unpleasant to testify in court, so why don’t you tell me everything you
remember, and then I’ll tell the jury everything you’ve told me. We can do this
very informally. In fact, I’m not even going to take notes. So just start
talking whenever you’re ready.
Witness: OK. Well, I was just walking down Main Street, minding my own business . . . .

It seems to me clear that this statement is testimonial. Clearly, Witness is making a statement for use at trial and (if it matters, which I don’t think it should) Officer understands that as well. But just as clearly, the statement seems informal – or, put another way, it cannot be considered formal without robbing that term of all meaning. Finally, it seems obvious that this type of statement should not be admitted against accused if he never has an opportunity to cross-examine Witness. And -- here is the crucial part -- it is inadmissible not despite the lack of formality but, one may say, in large part because of it.

What formalities is this statement missing? The presence of the accused and the opportunity for him to cross-examine, most notably. Those, of course, are the essence of the confrontation right. Clearly, the logic could not be that because of their absence the statement is informal and therefore the confrontation right does not apply, because that is a Catch-22 that would prevent the right from ever applying. Apart from those two, the most obvious formality is the oath. But we already know from Crawford itself that the absence of the oath will not make the statement be considered non-testimonial; the majority opinion was quite explicit on this point, see Crawford, 124 S.Ct. at 1365 n.3 ("We find it implausible that a provision which concededly condemned trial by sworn ex parte affidavit thought trial by unsworn ex parte affidavit perfectly OK."), and the statement at issue in the case was not given under oath. There are other formalities as well that usually accompany testimony -- the question-and-answer format and the general ceremonial nature of the courtroom -- but these are of lesser importance; I will explain in a separate post reasons why I do not believe interrogation is necessary to make a statement testimonial.

The bottom line is this: The absence of formalities does not render a statement non-testimonial. Rather, the absence of the most important formalities may make unacceptable as evidence a statement that is testimonial in nature. This casts a helpful light on dictionary definitions , like the one quoted by Crawford, that include formality as a component of testimony: Formality is an ideal, an aspect of testimony given in the optimal way, at trial in open court. The purpose of the Confrontation Clause, indeed, is to ensure that testimony be given in an acceptably formal way, in th presence of the accused and subject to cross-examination. To say that the absence of formality takes a statement that would otherwise be deemed testimonial outside the purview of the Clause would be to treat a defect of the statement into a virtue.

It would also give investigating officers precisely the wrong incentive. Whtever procedure is deemed to be a critical aspect of formality they would tend to avoid, so that statements given to them in full anticipation of evidentiary use would then be deemed non-testimonial and outside the rule of Crawford.

For reasons I hope to explore more fully in future posts, I believe the proper approach to the question of what is testimonial depends on the anticipation of the declarant, or on the reasonable anticipation of someone in the declarant's position, that the statement would be used evidentiarily. The presence of formalities may support a determination that the statement was testimonial; for example, in Crawford the statement was videotaped, with an introduction by the investigating officer that could leave no doubt about why the statement was being taped. But when a witness to a completed crime knowingly makes a statement to the police or other authorities describing the crime, the statement should be deemed testimonial, no matter how informally it was taken, because the likely evidentiary use is so clear. The presence of formalities can reinforce that determination, but they are not necessary to it.