Monday, December 05, 2011

Thoughts on the brief of the New York DA and OCME in Williams

I am going to try over the next several days to post (rather belatedly) a series of entries analyzing some of the arguments made by the state-side briefs in Williams. I’ll begin with the brief filed on behalf of the New York County DA’s Office and the New York City Office of the Chief Medical Examiner (OCME); I’ll refer to this as the New York brief. I’m beginning with this one because it probably makes the most extravagant claims of all the state-side briefs. It is in large part an attempt to scare the Court into thinking that if Williams wins this case prosecution use of DNA and some other types of forensic evidence will become unfeasible. That’s just not true.

A. The brief goes into depth to show the number of technicians that, in some labs, work on a DNA sample. A few responses:

(1) The brief simply mischaracterizes the position taken by, and in support of, petitioner as an “all-technicians-must-testify” rule. Williams contends for nothing of the sort. The Confrontation Clause only applies to testimonial statements that are in some way presented to the tier of fact. (I say “in some way” because there can be a confrontation problem without formal introduction of the statement; I’ve discussed this issue in other posts and will again.)

So consider the stages of DNA analysis discussed in the brief, p. 7:

(a) Examination: A technician “examines the sample and takes cuttings for DNA extraction.” There’s no testimonial statement there – examining and cutting do not constitute a statement.

(b) Extraction: A technician adds reagents to the sample. Again, no statement.

(c) Quantitation: A technician measures the amount of DNA. Presumably this technician reports on that amount. But even assuming that this report is a testimonial statement, there’s no need for it to be presented to the trier of fact. The witness who reports on the profile found in the later part of the process does not have to convey to the trier of fact or even rely in her own testimony on the results of this stage; we know from the fact that a DNA profile was ultimately found that there was enough DNA to perform the analysis.

(d) Amplification: A technician copies specific portions of the DNA to raise them to sufficient levels for testing.

(e) Electrophoresis: Here at last we have the performance of the test that matters. A technician who performs this test must report on the results. That report, in a case in which the test is clearly performed for forensic purposes, is a testimonial statement, and it provides the essential information that the prosecution needs.

So even assuming Williams wins and some labs continue to adhere to the procedure described by the New York brief, the Confrontation Clause would say nothing about most of the technicians involved in that procedure. As a check on this, try this thought experiment: Assume for the moment that Williams wins this case. Does anyone think that the signatories to the New York brief would be proclaiming that all the technicians in this procedure would have to testify?

I have not said anything here about chain of custody. So long as a witness speaks only about what she knows from personal knowledge, chain of custody is not a confrontation problem per se. Melendez-Diaz makes clear that as an initial matter it is up to the prosecution to decide what witness's statements it wishes to present to establish the chain of custody. If the gaps in the chain are too great, there may be insufficient proof, and at some point that could be a due process violation. But reasonable inferences can bridge some substantial gaps. And I don’t believe the sample needs to have been sitting still during those gaps; technicians may have performed procedures on it other than letting it change naturally over time.

(2) Given modern DNA techniques, retesting is virtually always a possibility. Neither the NY brief nor any other brief on the state side gives any reason to suggest that it would not be routinely possible in a case like this one. Only a small minority of cases go to trial; in a given case, if the original technician could not conveniently testify at trial, a technician better placed to do so could retest the sample without adding great expense.

(3) The Sixth Amendment does not incorporate the Cellmark protocol. Much of the New York brief reads as if Confrontation Clause jurisprudence must take as given the procedures such as those used by Cellmark in this case. But other labs use different procedures. Note, for example, that only one technician from the Illinois State Police lab did the test on the blood sample taken from Williams. The Michigan State Police lab rarely involves more than three people in a given lab test.

Of course, such vertically integrated procedures might not be as efficient as those used by Cellmark. But bear in mind that the Cellmark procedures were designed with little or no much regard to the confrontation rights of the accused. It is hardly surprising that procedures designed without that constraint would be more efficient than procedures subject to it.

Neither the New York brief nor any of the other state-side briefs suggest any reason why states that do not already operate under constitutionally satisfactory procedures cannot emulate states that do.

B. Neither does the Sixth Amendment incorporate New York evidentiary law. The New York brief says, p. 9, that under New York law the prosecution only has to present the analyst who compared DNA profiles. But note the consequence: No witness with personal knowledge testifies to the facts underlying the comparison. In my amicus brief, I presented an analogy to which no one on the state side has responded: An expert witness testifies that someone else (who does not testify in court) has given him a description of the assailant, and in the expert’s opinion that description matches the accused. Plainly that would not be allowed, assuming the description was given in contemplation of use in prosecution. That hypothetical actually is more than analogy – it is exactly what happened here, but the description, instead of being the more familiar characterization of features, was a statement of the genetic identity of the assailant.

C. Nor does the Sixth Amendment incorporate Fed. R. Evid. 703, a creation of the late 20th century. Invoking the language of that Rule, the New York brief refers, p. 15, to the dangers of “a constitutional rule that would preclude a testifying expert from reasonably relying on information provided by others.” Of course, no one is contending for such a general rule. But I doubt that even the authors of the New York brief would have the effrontery to contend for the opposite rule – that there is no constitutional constraint on an expert’s relying on information provided by others so long as a court can characterize the reliance as reasonable.

Again, there is no confrontation issue unless the underlying statement is both (a) testimonial and (b) conveyed in some manner to the trier of fact. I think that in light of Melendez-Diaz and Bullcoming, the Cellmark report here must clearly be considered testimonial. And in this case, I think it is clear that substance of the statement was conveyed to the trier of fact: The in-court witness testified that Cellmark deduced a male DNA profile from the vaginal swab, and that this profile was such that both she and a computer program matched it to Williams’s acknowledged profile. Whether there is sufficient communication of the statement in other situations – say, where the in-court expert relies on information she learned in a testimonial statement but does not disclose it – is an issue the Court need not reach here.

D. The New York brief suggests, p. 12, that “if there were an all-technicians-must-testify rule, no competent defense attorney would stipulate until she had assured herself that all of the technicians who worked on the defendant’s case were available to testify.” Well, first, of course, the premise is wrong – again, no one is contending for such a rule. And, as indicated above, whatever burden there is on coming to court is mitigated by vertical integration. But beyond that, all the attempts at creating a scare continue to ignore the fact that states that have always played by the rules for which Williams contends have not found this to be an insuperable burden. Why not? I do not deny that sometimes defense counsel plays this game, but often they realize it is more likely to do harm than good. For example, they may recognize from experience that the prosecution will do whatever it takes to ensure that any necessary lab witnesses appear. And often they understand that their chance of reaching an acceptable plea bargain will be substantially impaired if they’re perceived as game-playing in hopes of imposing costs on the prosecution. (In his Melendez-Diaz dissent, Justice Kennedy argued that it would be unprofessional for counsel to waive a client's rights for fear of incurring judicial displeasure; I am putting aside the possibility that counsel would act in that way.)

E. The New York brief also suggests, p. 10, that adoption of an “all-technicians-must-testify” rule – same flawed premise – might “force the OCME to reduce the amount of DNA testing it conducts, and force prosecutors to forego forensic DNA analysis in cases where it might be highly probative.” Oh, really. Requiring New York to adhere to the Constitution, as other states do, might cost it more money, but it’s not going to cause the state to forgo use of this powerful and efficient tool. The New York brief offers no basis for concluding that states that use constitutionally proper procedures rely on DNA evidence less than states that do not. And, as Jeff Fisher emphasized at a conference at Brooklyn Law School last month, the Confrontation Clause says nothing about police and investigative procedure, but only about the procedure for presenting testimonial statements at trial; a Supreme Court decision will not diminish the authorities’ desire to identify the actual assailant, nor will it diminish the ability of DNA testing to do that.

E. In my amicus brief, I pointed out that an in-court expert could rely on non-testimonial statements by lab technicians. I suggested, for example, that the report of a routine blood test would not be testimonial unless the testifying analyst announced gratuitously that she was seeking the information for prosecutorial purposes. The New York brief scoffs at this position. P. 16 n.10. It points out that at autopsies the OCME relies on tests from various sources, such as metabolic testing from NeoGen Screening. So, it asks, does the medical examiner have to hide her purpose from the NeoGen technician, and what if that technician figures it out? And is the defendant entitled to a hearing on the matter? The brief announces that merely to ask these questions shows that my position should not be the rule. Well, I disagree.

I’ll address the last question first – the accused is always entitled to whatever procedure is necessary to resolve a Confrontation Clause objection. But the trial court does not have to rely on admissible evidence in resolving that matter.

Assuming that the metabolic test is the type of thing that the NeoGen technician does as a regular matter principally for non-forensic purposes, and assuming that the technician doesn’t know that in the particular case the test is for forensic purposes, then, yes, I’d say it’s not testimonial – and I would bet a lot of money that, if the question were actually decisive in a case rather than something to be waved around for scare purposes, the OCME would vigorously advocate this position. And it does seem to me that if in fact the technician learns that the particular test is to be used for prosecutorial purposes, then it does become testimonial, so if they like the OCME and NeoGen can devise blind procedures. (Such procedures are a standard part of some scientific techniques.) But hey, if you don’t like my standard, then go for another. Here's one possible option: if a given type of report is usually or non-prosecutorial purposes then it is not testimonial even if in the particular case it is sought for such purposes and the author of it knows this. That rule would, in my view, be well sub-optimal – but it would do a lot less violence to the confrontation right than the positions advocated by the New York brief.

Ultimately, though, if in a murder case – the type of case in which an autopsy is usually relevant – a lab technician performs a test and writes a report on it knowing that it is likely to be used in prosecution, and before the accused is convicted of murder that technician is required to testify live rather than simply mail the report in, this does not strike me as a terrible result.

Finally, a general word. When Crawford was decided, I had hoped, naively, that most prosecutors would say, “OK, it makes sense that, when I want to use against an accused a statement that the person made understanding that it would be used for that purpose, I have to bring that person to trial. Let’s see how I can do this most efficiently.” And some good prosecutors and forensic lab directors have indeed taken that attitude. The New York County District Attorney’s Office is a good prosecutor’s office; if you didn’t know that, they announce it on p. 1 of their brief. Unfortunately, their attitude mirrors the one that most prosecutors have taken every step of the way since Crawford – “Let’s see how we can minimize this thing so that we have to present as little live evidence as necessary and change our way of doing things as little as necessary.” I think it’s very unfortunate that this brief uses persistent scare tactics in an attempt to achieve that result.


pv said...

Prof. Friedman presents his take on the DA's brief. Here is my take on Williams' reply brief:

The sole hearsay "statement" that Williams can identify as having been introduced by the state in its case-in-chief is Lambatos's answering "Yes" to a question by the prosecutor that incorporated the fact that a male DNA profile was found in the semen from the victim's vaginal swabs.

First, it must be noted that the prosecution independently proved this fact circumstantially through admissible evidence that the swabs were sent to Cellmark, and returned by Cellmark with the electropherogram depicting the assailant's DNA profile. So any hearsay on this point that should not have been admitted against a CC objection was harmless. Indeed, it does not appear that Williams ever objected to the prosecutor's question to Lambatos and, thus, any CC objection was waived.

In any event, to the extent that this "statement" is deemed to be a "statement" by some unknown Cellmark employee that "I obtained semen with a male DNA profile from the victim's vaginal swab," it is not testimonial.

It is not testimonial under the "primary purpose" test set forth in Bryant.

What is the primary purpose of the state in asking Cellmark the following questions: What is in the swabs? Is there any DNA in the swabs? What is the profile of any DNA found in the swabs?

Certainly, the state's purpose (primary or secondary) was not to build a case against Williams, as Williams was not even identified by the victim as her attacker until a year after the crime. Likewise, the purpose of the Cellmark employee(s) who answered the state's questions was not to create evidence to be used against Williams (or anyone else) in lieu of their live testimony at trial. There simply is no resemblance whatsoever between the exchange between the state and Cellmark and the type of ex parte examinations that are at the core of the CC.

Moreover, the questions and answers in Williams do not come close to satisfying the "formality" that Justice Thomas requires. Nor is there any indication that the state strategically refused to call the Cellmark employee(s) to evade confrontation.

Simply put, the sole hearsay statement that Williams has been able to identify is not testimonial under the Bryant primary purpose test or under Thomas's more narrow formulation.

There will at least 5 justices who will hold that any hearsay from Cellmark (tangentially introduced via Lambatos) was nontestimonial. Indeed, the 4 M-D & Bullcoming dissenters wouldn't apply any CC bar even if any statement from Cellmark was testimonial.

The court will also make clear that an expert, like Lambatos, can "rely" upon testimonial hearsay in forming their opinion without offending the CC, so long as the substance of that hearsay is not placed before the trier of fact by the prosecution. And this is true regardless of the reliability or accuracy of the underlying hearsay -- reliability and accuracy not being concerns of the CC, but only the forum's evidence law and, in certain limited circumstances, constitutional due process.

This case can be decided by the court in favor of the state by applying the legal principles set forth in Crawford, Davis, Bryant, M-D & Bullcoming, without attempting to assess the practical impact that a decision in favor of Williams would have on the ability of prosecutors to ensure that justice is done, the guilty are punished and the innocent are exonerated.

Richard D. Friedman said...

A brief response to some of Paul's points: It's not just that a male DNA profile was found. What was crucial also was the substance of the profile -- that is, it was such that, as determined by both Lambatos and a computer system, it matched Williams'. However far Paul's circumstantial evidence argument could go, a question I won't discuss here, it can't go as far as proving the substance of the profile.

The fact that Williams was not yet a suspect is neither here nor there. On that basis, if an eyewitness to a street crime was asked by the police in the station-house to identify the perpetrator, a description given by the eyewitness (or any other statement describing the crime or the scene) would not be testimonial. And bear in mind that the Cellmark statement is in fact a statement describing the perpetrator.

The Cellmark report was written in full recognition that it might well be used in prosecuting a crime. I don't think any member of the Court is going to believe that it is insufficiently formal to be testimonial.