A jury has a hard enough (if not impossible) time following the presumption of innocence instruction when the defendant decides to testify; the odds are stacked even further against the defendant if the judge specifically points out that the defendant is an “interested witness.” The playing field was leveled somewhat yesterday by the Second Circuit’s decision in United States v. Brutus, Docket No. 06-2710-cr (2d Cir. October 2, 2007). In it, the Court was called upon to reconsider the validity of a jury instruction it had previously approved on how to evaluate a defendant’s testimony at trial.

Poor Ms. Brutus had an uphill battle in attempting to explain to the jury her compromising circumstances (arrested at the airport in possession of 12 men’s sandals filled with cocaine) and her even more compromising post-arrest statements (that she knew the sandals contained drugs, had been recruited to make the trip for $3,000, and was to hand the drugs over to a stranger in the airport lobby). Her plight was further magnified by the judge’s instruction to the jury – over defense counsel's objection – that she had “a deep personal interest in the outcome of her prosecution,” so unique, it “create[d] a motive to testify falsely.” The very same instruction had been approved by the Second Circuit in United States v. Talkow, 532 F.2d 853 (2d Cir. 1976). Overruling Talkow and in light of its decision on a similar instruction last year in United States v. Gaines, 457 F.3d 238 (2d Cir. 2006), the Second Circuit found the instruction in Brutus to be constitutional error: “Simply stated, an instruction that the defendant’s interest in the outcome of the case creates a motive to testify falsely impermissibly undermines the presumption of innocence because it presupposes the defendant’s guilt,” the Court held. The Court further noted that the error is not cured by additional language in the instruction that the defendant can still be truthful. Going forward, and quoting with approval the Seventh Circuit’s pattern instruction on the subject, the Court advises district courts that “if the defendant has testified, the charge should tell the jury to evaluate the defendant’s testimony in the same way it judges the testimony of other witnesses.” Practitioners should therefore closely scrutinize jury instructions addressing a defendant’s testimony for deviations from the approved instructions set forth in Brutus.

Unfortunately, the holding is a pyrrhic victory for Ms. Brutus, mother of two, currently serving a ten-year mandatory minimum sentence. The Court found the jury instruction error to be harmless beyond a reasonable doubt in light of Ms. Brutus’s “manifestly incredible” testimony, which the trial judge described as “the most incredible perjury” he had ever heard.

Courtesy of Hofstra law professor Eric Freedman on the Federal Trials Listserve, I learned of this fascinating article by two Cornell law professors, Theodore Eisenberg and Valerie Hans: Taking a Stand on Taking the Stand: The Effect of Prior Criminal Record on the Decision to Testify and on Trial Outcomes.Analyzing a unique set of data collected from 300 criminal trials in 4 large counties, including Bronx Co., New York, which included a quantitative measure of the jurors' and judges' overall estimate of the strength of the evidence, the authors conclude that statistically significant associations exist (1) between the existence of a criminal record and the decision to testify at trial, (2) between the defendant testifying at trial and the jury learning about the defendant's prior record, and (3) in cases with weak evidence, between the jury learning of a criminal record and conviction. Essentially, in cases where the evidence against the defendants is strong, learning of criminal records is not strongly associated with conviction rates. In cases where the evidence would not normally support conviction, however, juries appear to rely on criminal records to convict. In fact, "[t]he effect in otherwise weak cases is substantial and can increase the probability of conviction to over 50 percent, when the probability of conviction in similar cases without criminal records is less than 20 percent." The authors conclude that "[t]he enhanced conviction probability that prior record evidence supplies in close cases may well contribute to erroneous convictions."

While most defense lawyers know this intuitively, it is always helpful to have one's instincts confirmed by empirical evidence. More importantly, this research arms a defense lawyer with persuasive data to advise a client with a criminal record against testifying in a close case, or to persuade a judge in a pretrial motion to preclude the government from cross-examining the defendant about his/her record.