<< News Scan | Main | Kevin Cooper, Finally >>

A Ninth Circuit Stinker Goes to SCOTUS

Judicial opinions can be roughly classified into three categories: the good, the bad, and the putrid. One of the latter from the Ninth Circuit will go before the Supreme Court next term. The "top side" amicus briefs, including ours, were filed last week in McDaniel v. Brown, No. 08-559, reviewing the Ninth's decision in Brown v. Farwell, 525 F.3d 787 (2008).

There are typically three reactions one has upon reading an opinion. Most are correct, and the bulk are obviously so. That is why most opinions at the intermediate appellate level are not published. A few strike the reader as wrongly decided, but in most of those there is room for reasonable disagreement. Then there are a few that make you want to throw the opinion against the wall and ask, "What planet are these people living on?!" The 2-1 decision in Brown, written by Judge Kim Wardlaw, is one of those. Judge Diarmuid O'Scannlain dissented.

The Brown case involves a crime of nearly unspeakable depravity, and I say that from the somewhat jaded perspective of one who deals with cases of depravity all the time. In Carlin, Nevada, a remote little town on I-80 about halfway between Reno and Salt Lake City, an adult man raped a nine-year-old girl so brutally that she "lost fifteen percent of her blood and required vaginal reconstruction." The primary question at trial was whether Troy Brown was that man.

On DNA evidence alone, even without all the other evidence, the state had a compelling case. Brown matched the RFLP test on a sample from the victim's underwear. The chance of that match occurring at random is only 1 in 3,000,000. Defense counsel did not pursue further testing pretrial because his own expert advised him (correctly, in turns out) that each further test would further inculpate Brown.

Posttrial, defense counsel did have another DNA test done. It was done on a separate sample, a vaginal swab, and it was performed by an independent lab of the defense's choosing, using a different technique, PCR. The lab was Forensic Science Associates, and the report is co-signed by the well-known Dr. Edward Blake. This test also came back a match, with a random match probability of 1 in 10,000. That rules out, to near certainty, explaining away the first match as contamination or lab error. Each of the two tests individually rules out the possibility that some unrelated person committed the crime and just happens to match Brown. Even the lesser random-match figure is many times the number of men who were in Carlin that night.

What about a related person? At trial, defense counsel brought up the possibility that a brother of Brown might match. Brown has four brothers, but three of them are excluded by other evidence. The state's expert testified at trial that the chance of a brother matching at random was 1 in 6500.

On appeal, Brown claimed, among other issues, that the evidence was insufficient for conviction. He did not, however, cite Jackson v. Virginia, 443 U. S. 307 (1979), the Due Process Clause, or any federal law whatever. The Nevada Supreme Court rejected the argument, using a standard from its caselaw that predates Jackson, "whether the jury, acting reasonably, could have been convinced of the defendant's guilt beyond a reasonable doubt."

On state habeas, Brown claimed [big surprise here] ineffective assistance of counsel. One might think he would claim trial counsel was ineffective for not sufficiently pursuing the brother-perpetrator hypothesis, given that DNA conclusively excludes all unrelated suspects. In fact he claimed exactly the opposite: counsel was allegedly ineffective for bringing it up at all.

When the case got to federal habeas, Brown asked to expand the record (Habeas Rule 7) with an affidavit that simply contradicts the state's trial expert. The new expert criticized the state's expert's response to a question recharacterizing the "random match" probability with "chance that someone else did it" probability. He did not, however, refute the original random match testimony. He also criticized the brother-match testimony. The chance of a single brother matching at random is 0.0038, not 0.00015. Assuming for the sake of argument he is correct, that is the difference between highly unlikely and extremely unlikely, neither of which is reasonable doubt. He also criticized the test of the esteemed Dr. Blake's outfit, claiming they couldn't really separate the perpetrator from the victim with the method they used. Without taking any further evidence, the district judge granted habeas relief on both sufficiency of the evidence and ineffective assistance grounds.

A divided Ninth Circuit panel affirmed on the sufficiency ground. How is Judge Wardlaw's opinion wrong? Let me count the ways.

First, she says that Brown fairly presented his Jackson claim to the state supreme court simply by making a generic sufficiency claim, even though he made no citation to federal authority. This is flatly, clearly, unmistakably contrary to Supreme Court precedent in two cases reversing none other than the Ninth Circuit, Duncan v. Henry, 513 U. S. 364 (1995) (per curiam), and Baldwin v. Reese, 541 U. S. 27 (2004). It is one thing to discuss Supreme Court precedents that appear to be on point and explain why they really are not. The district judge at least tried to do that. It is quite another to just ignore them and rule contrary to them without explanation or even any apparent awareness of them. That is either incompetence or dishonesty.

Next, the opinion goes into the supplementation of the record. The AEDPA issues here are sufficiently unclear that I won't trash this portion of the opinion, although I disagree with it.

Next we get to the merits. The standard the Nevada Supreme Court applied, Judge Wardlaw tells us quoting the district court, is contrary to Supreme Court precedent (Jackson), because it "only requires a reasonable jury--not a rational one." No, I'm not making that up. The opinion actually says that. First, if the words "rational" and "reasonable" are used in senses that they differ, review for reasonableness would be more favorable to the defendant, not less, than review for mere rationality. But we know Jackson did not use these words in any sense they differ, because Jackson itself uses the terms interchangeably. "After Winship, the critical inquiry on review of the sufficiency of the evidence to support a criminal conviction must be . . . to determine whether the record evidence could reasonably support a finding of guilt beyond a reasonable doubt." 443 U.S., at 318 (emphasis added).

To assert that a decision is contrary to Jackson because it uses the word "reasonably" is to assert that Jackson is contrary to itself. That is irrational.

But wait, there's more. Judge Wardlaw finds reversible error under the Jackson standard by reviewing the evidence at trial in light of evidence never presented to any state court. Did she even read the Jackson opinion? Over and over, the Jackson Court refers to the record of the evidence produced at the trial. Nowhere in that decision or subsequent decisions is there any authorization for a federal court to strike the portions of the evidence it does not think should have been admitted and then reweigh a balance of evidence different from what was before the jury.

We're not done yet. At the end of the opinion, the disposition is, "Respondents shall retry Troy within 180 days or shall release him from custody." Does Judge Wardlaw not know about Burks v. United States, 437 U. S. 1, 18 (1978)? Reversal for insufficiency does not get the defendant a new trial. It means he walks, protected by the Double Jeopardy Clause from retrial.

So, if Judge Wardlaw's opinion stood, a man that we know to a near certainly committed an unspeakably brutal rape of a helpless little girl would walk free. Fortunately, it is not going to stand. Although I rarely predict the outcome of a Supreme Court case this far in advance, I will on this one. It will be reversed, possibly unanimously.

How did an opinion so wrong in so many ways issue out of a federal court of appeals? It is not that the court did not have notice of how badly it was erring. One need only read the dissent to see some of the ways. Do the judges in the majority just not care? That is hard to believe in a crime of this magnitude. Do they actually believe Brown is innocent? To form that conclusion and bend (probably break) the law to spring him based on one affidavit from one expert contradicting another expert of greater reputation is negligent, at the very least.

We will probably never know what happened internally at the Ninth Circuit. Judges don't normally talk about their cases publicly. We do know that something went very, very wrong here. This case illustrates once again that Congress was correct in 1996 when it rejected the premise that the federal court always knows best. Now we need that law to be genuinely respected and enforced, not evaded as it so often is.


"Despite all these citations of, and quotations from, Strickland, the Ninth Circuit concluded that the California Supreme Court had held respondent to a standard of proof higher than what that case prescribes for one reason: in three places (there was in fact a fourth) the opinion used the term “probable” without the modifier “reasonably.” 288 F.3d, at 1108—1109, and n. 11. This was error. The California Supreme Court’s opinion painstakingly describes the Strickland standard. Its occasional shorthand reference to that standard by use of the term “probable” without the modifier may perhaps be imprecise, but if so it can no more be considered a repudiation of the standard than can this Court’s own occasional indulgence in the same imprecision. See Mickens v. Taylor, 535 U.S. ___, ___ (2002) (slip op., at 3) (“probable effect upon the outcome”); Williams v. Taylor, 529 U.S. 362, 393 (2000) (“probably affected the outcome”)."

The quoted material, from Visciotti, seems to nuke the "rational"/"reasonable" issue. And Visciotti was a Ninth Circuit case.

I am guessing (and it's an ill-informed guess) that the Ninth wanted to send a message about DNA testing, and these particular judges couldn't wait for a more appropriate vehicle. Liberal judges itch to make law, and I suspect that the worse the crime the more a liberal judge sees favoring a criminal defendant a badge of honor.

Great brief--you must have found yourself like a predatory fish attacking a school--so many targets as to make the exercise a difficult one. Seems like the use of Jackson was patently unfair. State's case gets fixed as of trial, whereby, if Jackson is expanded, then state would never have the opportunity to expand its case, as it does in Schlup-type cases. That, coupled with the harsh remedy of no retrial, is completely unfair to the state.

Wow, I can't even get past the idea that someone would think restriction fragment length polymorphism somehow didn't settle the issue (unless there was evidence of contamination).

Leave a comment

Monthly Archives