Opinion recap: Court strengthens municipal immunity for prosecutorial violations

On March 29, 2011 in Connick v. Thompson, the Court – by a vote of five to four, in an opinion by Justice Thomas – reversed a fourteen-million-dollar award in favor of John Thompson, who served eighteen years in prison for murder and armed robbery before his convictions were vacated.  The Orleans Parish District Attorney’s Office conceded that prosecutors failed to turn over, in violation of Brady v. Maryland, evidence to Thompson’s defense team that would have been exculpatory, but it maintained that it could not be held liable under Section 1983 for failing to properly train its attorneys.  The Court agreed, holding that  the district attorney’s office cannot be held liable under Section 1983 for failure to train its prosecutors based on a single violation of Brady.

To prevail on his failure-to-train claim, Thompson bore the burden of proving that (1) District Attorney Harry Connick, Sr., as the policymaker for the DA’s office, was deliberately indifferent to the need to train his prosecutors on Brady; and (2) the lack of training actually caused the Brady violation in Thompson’s case.  The Court began by emphasizing the rigorousness of the “deliberate indifference” standard, which requires proof that the municipal actor disregarded a “known or obvious consequence” of his action; the contours of this deliberate indifference standard were the central issue in Thompson.  In Canton v. Ohio, the Court held that deliberate indifference is ordinarily proven by showing a pattern of similar constitutional violations by untrained employees that is subsequently ignored by policymakers.  However, the Court also hypothesized in Canton that there might be a situation in which the need for training was so obvious that failure-to-train liability could be premised on a single violation – for example, if a city were to arm its police and send them out into the community to capture fleeing felons without training them on citizens’ constitutional rights.  Although the Court has never recognized such a situation in practice, Thompson argued that his case fit squarely within the Canton hypothetical: the violation was such an obvious consequence of failure to train on Brady that it substituted for the ordinarily requisite pattern of violations.

The Court roundly rejected the comparison of the Canton hypothetical to Thompson’s situation, emphasizing that the Canton example is fundamentally different from an alleged deficiency in Brady training of prosecutors.  Rather, the Court explained, the Canton example was meant to illustrate a situation in which the need for training in constitutional requirements is so obvious ex ante that failing to train employees amounts to deliberate indifference to constitutional violations.  First, the Court reasoned, while there is no reason to assume that police academy recruits are familiar with the constitutional restraints of the use of deadly force, a district attorney is entitled to rely on prosecutors’ professional training and ethical obligations unless there is specific evidence (e.g., a pattern of violations) which alerts him to the need for training.  Second, the Canton hypothetical assumes no knowledge whatsoever of the constitutional limits of deadly force, while in the instant case, the prosecutors were demonstrably generally familiar with the Brady rule, even if they were ultimately incorrect in their application of it.  Third and finally, unlike police officers, attorneys are trained during and after law school to “find, interpret and apply” legal principles.  Just because some Brady decisions are difficult or even unsettled does not make it “so obvious” that violations will occur without certain training.  Thus, absent a pattern of violations, Connick was not on notice that such violations were “highly predictable.”

Justice Ginsburg authored (and read from the bench) a dissent that was joined by Justices Breyer, Sotomayor, and Kagan.  The dissent countered that Thompson’s case actually does fall under the Canton example for single-incident failure-to-train liability.  Much of the dissent is dedicated to discussing the record and the multiple ways in which the DA’s office failed in its constitutional obligations.  Contrary to the majority’s view, the dissent contended, there are many ways to demonstrate deliberate indifference short of an established pattern of violations, and the evidence presented at trial was more than sufficient for the jury to find deliberate indifference.  Connick did not ensure that new prosecutors knew about their obligations under Brady, he did know of the need to train and monitor on Brady, and his “cavalier” approach to his staff’s Brady knowledge led to a “culture of inattention” to Brady in his office.  These factors, according to the dissent, easily rise to the level of deliberate indifference.

Justice Scalia authored a brief concurring opinion that was devoted primarily to addressing and refuting several points made in the dissent.  In particular, he complained about the dissent’s “lengthy excavation of the trial record,” emphasizing that the question presented was a legal one.  He also contended, however, that Thompson had failed to meet the “rigorous” standard of causation required by Section 1983, reasoning that no amount of training would have prevented the prosecutor’s willful suppression of the Brady evidence.  Finally, Justice Scalia contended, it is possible that there was no Brady violation at all; the blood evidence in question was untested, and this type of evidence, which is “on the frontier” of the court’s Brady jurisprudence, is not subject to obvious training standards.  Thus, if there were a Brady violation, Justice Scalia would not have found the causation necessary to assign liability.