Prof. Friedman, could you provide an analysis of the following scenario and commentary on my suggested formulation:
An NYPD DNA expert (X) writes an unsworn report. In the report, X states that on January 1, 2011 she responded to a crime scene. There she observed a naked female, face down, dead with a cigarette butt on her back. She collected the cigarette butt and brought it back the the NYPD lab. She processed the cigarette butt, following the lab's standardized procedures for processing evidence for DNA, and extracted a tiny amount of saliva. She used the latest available scientific techniques and equipment. Her testing procedures resulted in raw data – graphs, charts, etc. (i.e., a DNA profile). She used her expertise to interpret the DNA profile. And reached the opinion that the DNA profile was from a male Caucasian. The tiny test sample was consumed during the testing process. X then entered the DNA profile into a nationwide DNA database.
30 years later, X died in a car accident. As of that date, NYPD had not yet identified a particular suspect, i.e., they didn't have "reasonable suspicion" to detain anyone, or "probable cause" to arrest anyone, there were no "hits" from the database.
10 years later, an NYPD DNA expert (Y) was notified that there was a "hit." And the defendant (Mr. Z) was identified as the suspect. Y obtained a blood sample from Z. He then determined Z's DNA profile. In Y's opinion, Z's DNA profile "matched" the DNA profile of the saliva, i.e., in Y's opinion there was a statistical match probability of 99.9%.
At trial, would the CC prevent the prosecution from introducing the following evidence (assume that there are no other constitutional or evidentiary hurdles to admissibility):
(1) The raw data establishing the DNA profile of the saliva?;
(2) Y's expert opinion testimony (based upon a review of X's report – the substance of which is not disclosed to the jury – and his own expertise) that (A) the DNA profile of the saliva "matches" Z's DNA profile; (B) the DNA profile of the saliva was accurate/reliable because X was qualified and followed all proper procedures?; and
(3) X's entire report setting forth her expert opinion that the DNA profile of the saliva was that of a male Caucasian?
For what is worth, I believe that X was not a CC "witness," and, therefore, all of the above evidence would overcome a CC objection. I believe that at least four justices in Bullcoming (and, perhaps, in Bryant) will redefine CC "witnesses" in a manner consistent with the following formulation:
Persons (who have personal knowledge of the crime allegedly committed by the defendant) who make out-of-court statements (i.e., an assertion of fact by a human being) -- whether testimonial or nontestimonial (under any of the three formulations in Crawford) -- and whether sworn or unsworn, are not CC "witnesses" unless each of the following requirements are met: (1) a government official (or their agent) responsible for the investigation and prosecution of crime, (2) by some affirmative action (i.e., questioning, "interrogation," etc.), was involved in the production of the particular statement, (3) at a time that an "adversarial relationship" existed between the government official (or their agent) and a particular person (i.e., at the time the statement was produced, the government had "reasonable suspicion" to detain (or "probable cause" to arrest) a particular person for criminal activity, and (4) viewed objectively, from the perspective of a competent adult in the position of the declarant, the declarant knew that they were providing a statement against someone.
The third element of this test replaces the "primary purpose"/"ongoing emergency" formulation set forth in Davis. Under the "adversarial relationship" element, once "reasonable suspicion" to detain (or "probable cause" to arrest) exist, it will be conclusively presumed that the government's "primary purpose" was to create evidence for possible use in a future criminal trial. Before the government has "reasonable suspicion" to detain a particular person, they have no motivation whatsoever to manipulate or manufacture evidence against that person by means of coercion, trickery or otherwise. However, once the "reasonable suspicion" threshold has been crossed (a determination that courts have routinely been making for Fourth Amendment purposes since Terry was decided) the police have a motive to start to build their case against the suspect. With that motive comes the distinct possibility that the reliability of the resulting evidence, and the truth-seeking adversarial process the CC is designed to protect, will be compromised.
The formulation that I set forth is entirely consistent with all of the Court's CC cases, before and after Crawford, including Melendez-Diaz as the "analyst" in that case didn't have any personal knowledge regarding the defendant's crime.
While the Framers didn't have to consider the constitutional ramifications of "pure cold case" prosecutions involving unavailable expert DNA analysts, the formulation that I have set forth is a "reasonable estimate" – as "reasonable" as any other estimate given the absence of any direct historical evidence – of how they would have treated these cases, at least under the CC.
If Giles's forfeiture doctrine would have been broader then, at least in "pure cold case" murder prosecutions, the Court would not be faced with the constitutional question that I raise. In Bullcoming (indeed, in Bryant) the Court has the opportunity to revisit the scope of the term "witnesses" in the CC. It is evident the framing-era historical record does not provide a clear cut answer. Any solid historical evidence that does exist lends itself to more than one "reasonable" interpretation of the original meaning of this text. Thus, policy considerations which are consistent with the truth-seeking purpose of the CC come into play. Four justices in Melendez-Diaz have voiced their dissatisfaction with the "testimonial" approach set forth in Crawford. It appears as though Justices Sotomayor (in Bryant and Bullcoming) and Kagan (in Bullcoming) might be the swing votes and might be instrumental in moving CC jurisprudence in a new direction that, in the absence of clear historical guidance, focuses more heavily on the practical, real life, impact of the constitutional rule on the lives of all Americans, not only the accused.
Criticisms, critiques and alternative formulations to the one I set forth above are welcome.
I will continue to avoid predictions, but I do think that Paul’s formulation would severely impair the confrontation right. It seems to be one more attempt by prosecutors (here a former one) to devise rules that would under cut down on the right and, to a substantial extent, nullify Crawford. (Having said that, I do continue to believe that the decision in Giles was most unfortunate.) I really wish all that mental energy went into seeing how those states that have routinely violated the confrontation right could efficiently bring themselves into line with states that have a good compliance record.
I’ll also say as a prefatory matter that yes, sometimes in long-delayed cases prosecutions are lost because by the time the prosecution is brought a crucial witness who has not been subjected to cross-examination is no longer available. This is a cost of our system that we have been tolerating for 500 years or so.
Paul has put forward before the idea that a statement is not testimonial unless a government official is involved in producing it. I think this is a non-starter, for reasons I’ve indicated before. It conflicts with historical practice; it’s important to understand that the confrontation right long predates the institutions of public prosecutors and police forces. The Court has already made it clear that volunteered statements, not made in response to interrogation, can be testimonial. If Paul’s suggestion were taken, there would be an army of private statement-takers who would enable witnesses to avoid confronting the accused.
Paul now suggests that for a statement to be testimonial there has to be an identifiable suspect before the statement is made. (I’m not sure whether in his view the suspect at the time of the statement has to be the ultimate accused.) I’m not sure where he gets that – so far as I know there is no historical basis for it. If adopted, this idea would distort police behavior; it would encourage police to avoid focusing on a suspect – and being in a position in which they could at least somewhat credibly contend later that they had not focused on a suspect – for as long as possible so that they can gather incriminating evidence. And the consequences of this idea would be stunning. Paul seems to acknowledge that, for example, a witness describing a crime scene would not be covered by the confrontation right. But more than that – nobody who made any statement in the 40 years between the crime and the time that the cold hit was made would be considered a witness against the accused for Confrontation Clause purposes – even if the statement was made to the authorities in the full hope and expectation that eventually it would be used as evidence at trial against the person the authorities identified as the killer. (This would include the victim if she made a statement describing the crime and giving a general description of the accused.) As a matter of principle, there is no justification for decimating the Clause in this way. The optimal rule is actually pretty simple. A person is a witness for purposes of the Clause if a reasonable person in her situation would realize that her statement would likely be used in investigating and prosecuting a crime. And if her testimonial statement is introduced against an accused, that makes her witness against the accused for purposes of the Clause.
What can the state do to minimize the danger of losing lab evidence if the there is a substantial chance that the case will not be tried for many years? Preserve some of the sample for retesting; Paul’s hypothetical is becoming increasingly unrealistic with improved DNA methods. If the case is important, have a second observer, as some jurisdictions routinely do for autopsies. Sometimes videotaping the testing might help. An extreme response – one of dubious constitutionality, but one with precedent and preferable to cutting the Clause down so close to its core – would be to appoint counsel shortly after the test for the as-yet unidentified suspect to depose the analyst; sometimes, notwithstanding the lack of identifying information at the outset, the interest of the ultimate accused is clear from the start.