Law, Psychology, and Impartiality

Donald Trump's suggestion that Judge Curiel should recuse himself seemed obviously wrong to me (and apparently to his own lawyers, who, as Neal Goldfarb pointed out in response to my prior post, did not even bother to raise the issue by motion). But today's Supreme Court decision in Williams v. Pennsylvania  raises all the difficult questions that the Trump University lawsuit does not about when disqualification is required because a judge's "impartiality might reasonably be questioned." Richard Re also has more here at SCOTUSBlog.

On the face of it, the main questions are fairly straightforward. The District Attorney who personally approved the decision to seek the death penalty in Williams' case was later elected to be Chief Justice of the Pennsylvania Supreme Court. Decades after the original conviction, he was part of a panel that ruled against Williams' subsequent habeas petition. The Pennsylvania Code of Judicial Conduct--based on the ABA's Model Code--forbids judges from acting in a case where they  "served as a lawyer in the matter in controversy, or w[ere] associated with a lawyer who participated substantially as a lawyer in the matter during such association." Chief Justice Castille participated in the Pennsylvania Supreme Court's ruling shortly before his retirement. After his retirement, the Court considered--and rejected--a motion for rehearing. 

The first question before the Court was whether the violation of this rule would amount to a violation of constitutional due process. The Court, in an opinion authored by Justice Kennedy, held that it did: "Where a judge has had an earlier significant, personal involvement as a prosecutor in a critical decision in the defendant’s case, the risk of actual bias in the judicial proceeding rises to an unconstitutional level."

The second question that the Court had to decide was whether the judge's participation in the Pennsylvania Supreme Court decision amounted to harmless error, given that the ruling was unanimously decided by a multi-member court. This, to me, is a much more difficult question. The Court held that harmless-error review was not appropriate, as the disqualified judge could have influenced the other members of the panel, and " it is neither possible nor productive to inquire whether the jurist in question might have influenced the views of his or her colleagues during the decisionmaking process."

Chief Justice Roberts, joined by Justice Alito, wrote a dissenting opinion arguing that due process does not require recusal when the petition does not allege that the judge had  "any previous knowledge of the contested facts at issue in the habeas petition, or that he had previously made any decision on the questions raised by that petition." (emphasis in original).

Justice Thomas also wrote a dissenting opinion. He pointed out that the due process requirements of criminal proceedings are and should be different that those required by later habeas proceedings (a form of civil action), and argued in favor of greater deference to the relevant rules and legislative enactments. He also pointed out that the Pennsylvania Supreme Court's earlier decision to deny rehearing in the case--a decision made after Chief Justice Castille had retired from the court--might have "cured" the alleged due process violation.

So what, in my mind, makes this case so hard? Some of it involves law, politics, and difficult recusal issues.

(1) The Commonwealth argued that it was unreasonable to think that Castille would be biased given the amount of time that had gone by and given his relatively limited participation in the case. The murder at issue occurred in 1986, and the Pennsylvania Supreme court decision was issued in 2014. And although Castille had to personally sign off on the decision to seek the death penalty, the Commonwealth's brief called that an "administrative act," as Castille would have deferred to the prosecutors who worked up the case. But, as Justice Kennedy points out in the Supreme Court's opinion, this characterization of Castille's role is at odds with how he portrayed his role as district attorney when he ran for judge, as "multiple news outlets reported his statement that he 'sent 45 people to death rows' as district attorney." Perhaps this was mere campaign puffing, but it is hard for the state to walk back Justice Castille's involvement at this point. I am also personally troubled that the death penalty has become so politicized that these decisions have become effective fodder for judicial campaigns.

(2) The Supreme Court's decision points to the possibility of unconscious bias, a topic very deftly argued in an amicus brief by Yale's Ethics Bureau (and spearheaded by Lawrence Fox). But if the Court takes seriously the risk that a judge, in the Court's words,  "would consciously or unconsciously avoid the appearance of having erred or changed position" taken as a prosecutor, than isn't that doubly true of the current justices on the Pennsylvania Supreme Court? If we believe that Chief Justice Castille could have tainted the panel's decisionmaking, then wouldn't we expect the remaining judges on the Pennsylvania Supreme Court to be subject to the same unconscious bias hindering a change of position? One of my favorite social science articles suggests that people are very bad at identifying their own biases--and, in fact, that further reflection upon possible biases only makes people believe even more strongly that their views are neutral and unbiased--even when the evidence would suggest otherwise.  See Cynthia McPherson Frantz, I AM Being Fair: The Bias Blind Spot as a Stumbling Block to Seeing Both Sides, 28 BASIC & APPLIED SOC. PSYCHOL. 157 (2006). 

(3) The underlying facts of the case are especially tragic, and show how various factors (including social stigma, overworked defense lawyers, and overly zealous prosecutors) can combine to create a miscarriage of justice. Williams, the defendant, was subjected to sexual abuse between the ages of 13 and 17. He later committed two murders--one at age 17 and one at age 18--and there was substantial evidence that both of the men he murdered were among those who had raped him. More importantly, the prosecutor's office knew that history, according to later-released notes stating that Williams' "relationship" to the two victims was "substantially similar." Williams' first attorney used that information as mitigation evidence in the trial for the first murder, and the jury returned a verdict of third-degree murder. In the trial for the second murder, however, Williams had new counsel and seemingly did not tell his attorney of this history. Perhaps a better attorney would have done a more careful job with the client interview and drawn out that information; certainly a better attorney would have examined the earlier proceedings. But for whatever reason, Williams' second attorney did not, and Williams himself testified falsely at trial that he had no earlier connection with the murder victim. All of this led to issues in the later habeas proceedings, including whether the defense attorney provided ineffective assistance of counsel and whether the prosecutor committed a Brady violation by failing to turn over evidence of the prior sexual abuse. Justice Thomas notes, in footnote 2 of his dissent, that it's an interesting question whether "a prosecutor could violate Brady by failing to disclose information to the defendant about the defendant’s motive to kill." Clearly, Williams knew of the prior abuse. But the stigma surrounding that abuse--particularly in the 1980s--may have hindered his willingness to raise the issue even when faced with a potential death penalty. But untangling the web of responsibility here is a difficult one: Williams admittedly committed perjury in the second proceeding; his attorney, at a minimum, did a bad job investigating the facts of the case; and the prosecutor knew of relevant mitigating evidence that defense counsel did not.

(4) Finally, I have to note one of the most offensive items from the briefing in the case. The Commonwealth's brief to the Supreme Court, at page 10, gratuitously  goes out of its way to describe Williams as "a double murderer who had sex with men for money." Given that he was between the ages of 13 and 17, Williams was below the age of consent for the majority of the time period at issue. Not only is this statement offensive (as has been repeatedly noted, "Sex without consent isn’t sex. It’s rape."), it is also terrible advocacy. Point out Williams' responsibility to testify honestly; point out that sexual abuse does not justify murder. But disparaging a child as young as thirteen for his own rape only perpetuates the stigma associated with sexual abuse and trafficking and stops other victims from coming forward before the tragedy cascades further.