Saturday, August 09, 2014

Two new lab report cases from New Jersey

The New Jersey Supreme Court issued a couple of decisions earlier this week on lab reports.  They're worth comment, in part because they show the state of affairs in this still contentious area.

In one, State v. Michaels, a lab supervisor wrote a report based on blood tests performed by other analysts in the lab; he had not observed the performance of the tests.  The supervisor testified at trial; the others did not.  Over one dissent, the court held that there was no confrontation violation, though it acknowledged that the lab report was testimonial.  The majority emphasized that the results of the test were machine generated, that the testifying witness was familiar with the lab's procedures and the particular test at issue in this case, that he had supervisory authority over the performance of the test, and that he had given his own interpretation of the results.
 
The fact that the supervisor gave his own interpretation to the data really shouldn't matter (and I think five justices in Williams v. Illinois indicated that it shouldn't).  There are data and then there is analysis, and those are entirely separate.  The fact that the witness who provides the analysis testifies subject to cross-examination provides no relief from a confrontation problem created by a testimonial statement as to the data. And even though the tests were performed by machine, it does appear that a couple of humans made statements on which the testifying analyst relied (though the opinions are not entirely clear on this point, I don't believe).

The witness's supervisory authority and familiarity with the lab procedures also shouldn't really matter.  It doesn't distinguish this case from Bullcoming, and it does nothing to relieve the problem that the data were provided by others.

There is also a conceptual problem or misunderstanding in the majority opinion, it appears to me, indicated by the court's use at one point of the phrase "testimonial fact."  Facts are not testimonial -- statements are.  There is no plausible contention that anybody who handled the materials or observed the tests must testify.  The only real contention is that anyone who made a testimonial statement that is in some way presented to the trier of fact must testify.  (And it appears that there were at most one or two of those.)  I say "in some way presented" because formal admission of the statement is not necessary to create a confrontation problem.  If, for example, the testifying witness makes clear that he is relying on information provided by others, that is sufficient presentation to put the confrontation right in play, assuming the transmittal of that information was in a testimonial statement.

Having said all that, though, I think the result might have been justifiable given Williams -- which I continue to regard as a very unfortunate decision.   Understandably, the majority threw up its hands in trying to draw anything from Williams, but it does stand for something.   There is only a confrontation problem if the statements on which the testifying analyst relied were testimonial.  I think they should be -- they're statements made in the context of a blood test to determine whether a person was under the influence of drugs, and I think they were clearly made in anticipation of prosecutorial use.  But Justice Thomas presumably would regard those statements -- in contrast to the final report -- as non-testimonial because the analysts did not make them in a formal, certified report.  The plurality plainly would prefer to call those statements non-testimonial -- these are the Melendez-Diaz dissenters, after all -- but I'm not sure that their Williams opinion stands for the proposition that they are not testimonial:  These were statements directed at a targeted individual who was suspected of having committed a crime.  But in any event, the plurality would have concluded that there was no
confrontation problem with respect to these statements, on the basis -- properly rejected by five justices in Williams -- that the statements were not used for the truth because they underlay the opinion of the testifying expert.  In other words, I suppose the rationales of five justices in Williams support the bottom-line conclusion that there was no confrontation violation here.  

The other New Jersey case from this week, State v. Roach, was a prosecution for rape and associated crimes.  One analyst did a DNA test on a swab taken from the victim and wrote a report that included  a male profile.  Some time later suspicion fell on Roach, and a second analyst who had taken over the case when the first moved out of state did a DNA test on a buccal swab taken from him.  At trial, the second analyst testified that the two profiles matched.  This -- apart from the fact that the accused lived near the victim -- was essentially the only evidence of identity.  The prosecution made no attempt to bring the first analyst in.  Roach received a 40-year sentence.  The supreme court majority held that there was no confrontation problem.

A few points.  First, here, it does not appear that Williams could have done the majority any good:  As I understand it, the DNA reports here were certified, so that means that Justice Thomas would have lined up with the other four justices who were in the majority in Melendez-Diaz and Bullcoming.  In fact, the rationales of five justices from Williams clearly support the conclusion that there was a violation here; that's not a holding, though, because four of the five dissented from the result in Williams.  The case is in many respects similar to Williams -- DNA analyst who analyzed the sample taken from the suspect testifies, analyst who analyzed the crime scene sample does not -- but if this had been the one the US Supreme Court had heard presumably the result would have been the other way.

Second, a point emphasized by Justice Alito in Williams holds true here as well:  It would have been an extraordinary coincidence for analysis of the crime scene sample to happen to turn up the profile of a man who lived near the victim unless his DNA was in the sample and analyzed accurately by the lab.  But I think that if this were to be he basis for introducing the evidence without someone from that lab testifying live in court, it would have to be without relying on the proficiency of the lab.  The argument would be that, however bad that lab might be, the only way they could have generated those numbers was if they accurately analyzed a DNA sample from Roach.

Third, the supposed reliance in Michaels on the testifying witness's supervisory authority appears from this case to be rather hollow.  Here we had a non-supervisor tesitfying about a testimonial statement made by another analyst from the same lab, and the majority does not appear to have been particularly disturbed.

Fourth, for all the cries we keep hearing about the necessity of having multiple analysts working on DNA it appears from the state supreme court decision and also from the appellate division opinion that the one analyst who was in charge of the file at the time did all the procedures necessary for the DNA test.  Bravo!  It appears that the New Jersey State Police Forensic Lab -- which was the one involved in this case but not in Michaels -- had integrated vertically, having one analyst work on a case A to Z, and I surmise that they did so for the very purpose of minimizing who would have to testify.  If any readers have any further information on this, I'd be glad to know -- but from what I can tell, New Jersey has shown that this can be done.  If this kind of vertical integration is somewhat less efficient than an assembly line procedure, I think the response still has to be that if a state wants to   comply with the Confrontation Clause without multiple lab witnesses having to testify, then it should adjust its procedures, not that the demands of the Clause should be adjusted to conform to what some crime labs do.

Finally, bear in mind that the defendant got a 40-year sentence, that the DNA evidence is what proved identity, and that one of the two critical witnesses who made a testimonial statement used by the prosecution to prove that identity did not testify subject to confrontation because she had moved from New Jersey to Wisconsin.  There's not much doubt that if the state knew it had to bring the witness in from Wisconsin in order to secure a conviction, it would have found it well worthwhile to do so.  I really don't think it's too much to insist that the state bring such a witness in when it wants to put someone away for decades.