Article Bank # A-39 [Per Se Reversal Rule Applies To “Structural” Errors And Errors Which Preclude Meaningful Review].)
Harmless Error Analysis
93 Review
Dallas Sacher
II. SULLIVAN V. LOUISIANA: THE U.S. SUPREME COURT PROVIDES EXTREMELY
HELPFUL GUIDANCE ON THE MANNER IN WHICH HARMLESS ERROR
ANALYSIS IS TO BE CONDUCTED WHEN A FEDERAL CONSTITUTIONAL
ERROR IS FOUND.
In recent years, the California appellate courts have increasingly found more and more errors to be harmless. In so doing, the courts have frequently provided little analysis on exactly why a particular error was not prejudicial. Indeed, the likeliest line to be found in an opinion is the pithy conclusion that “the evidence was overwhelming.” While such a conclusion may be sufficient to dispatch an error of state law, the United States Supreme Court has recently advised us that a much more critical and sophisticated analysis is required when a violation of the federal Constitution has been found. In Sullivan v. Louisiana, supra, 124 L.E.2d 182, the issue before the court was whether a constitutionally deficient reasonable doubt instruction may ever be harmless error. While the court concluded that such an error is always reversible per se, the court nonetheless provided an extremely clear description of the type of harmless error analysis which is required for the review of those federal constitutional errors which do not compel per se reversal.
In this regard, the court first noted that harmless error analysis must be conducted with adherence to the Sixth Amendment’s guarantee that “the jury, rather than the judge, reach the requisite finding of ‘guilty.’ [Citation.]” (Sullivan, supra, 124 L.E.2d at p. 188.) Consistent with the jury-trial guarantee, the appropriate harmless error inquiry “is not what effect the constitutional error might generally be expected to have upon a reasonable jury, but rather what effect it had upon the guilty verdict in the case at hand. [Citation.] ” (Id., at p. 189.) Thus, the appellate court must look “to the basis on which ‘the jury actually rested its verdict.’ [Citation.]” (Ibid., emphasis in original.) Given this focus, the inquiry “is not whether, in a trial that occurred without the error, a guilty verdict would surely have been rendered, but whether the guilty verdict actually rendered in this trial was surely unattributable to the error.” (Ibid., emphasis in original.)
As is readily apparent from the language quoted above, the rule set forth in Sullivan is that an appellate court is powerless to excuse a federal constitutional error on the basis of its conclusion that the evidence was overwhelming. Indeed, although this point has been lost over the last quarter century of jurisprudence, the Supreme Court has always intended that the mere strength of the evidence, standing alone, cannot be a sufficient reason to find an error harmless. (Chapman v. California (1967) 386 U.S. 18, 23, fn. omitted; criticizing the California courts’ neutralization of harmless error analysis by “overemphasis, upon the court’s view of overwhelming evidence.”‘)In short, Sullivan requires an appellate court to closely and carefully assess the actual impact which an error has had on the jury’s deliberative process. In so doing, the court must be ever aware that the government bears a heavy burden of persuasion in showing that the error did not affect the jury. In this regard, the Supreme Court has made the difficulty of the government’s task quite clear: the guilty verdict must have been “surely unattributable to the error.” (Sullivan, supra, 124 L.E.2d at p. 189, emphasis added.)
Although it predates Sullivan, Arizona v. Fulminante (1991) 499 U.S. 279 provides an excellent example of the manner in which federal harmless error analysis is to be conducted. In Fulminante, the defendant was charged with the murder of his 11- year-old-stepdaughter. While in prison on an unrelated matter, Mr. Fulminante confessed to one Anthony Sarivola that he had sexually assaulted and killed his stepdaughter. Six months later, Mr. Fulminante gave a second detailed confession to Sarivola’s wife, Donna. On appeal, the Arizona Supreme Court held that the confession to Mr. Sarivola had been coerced and was inadmissible. However, based on the confession to Ms. Sarivola, the error was deemed to be harmless “‘due to the overwhelming evidence adduced from the second confession, . . .'” (Fulminante, supra, 499 U.S. at p. 297.)
Notwithstanding the Arizona Supreme Court’s conclusion that the judgment was supported by “overwhelming evidence,” the U. S. Supreme Court examined the record in much closer detail. In so doing, the court reversed based on three considerations: (1) the government had little evidence to corroborate the confession to Ms. Sarivola; (2) the confession to Ms. Sarivola might not have been believable in the absence of the earlier confession to her husband; and (3) the erroneous admission of the confession to Mr. Sarivola led to the introduction of information that he had connections with organized crime which led to the prejudicial inference that Mr. Fulminante sought out the company of criminals. (Fulminante, supra, 499 U.S. at pp.297-300.) In short, upon close examination of the record, the Supreme Court determined that the dynamics of the trial were such that the government had not met “its burden of demonstrating that the admission of the confession to Sarivola did not contribute to Fulminante’s conviction. [Citation.]” (Id., at p. 296.)
While the analysis in Fulminante speaks for itself, the essence of the case should not be lost. Notwithstanding the fact that the jury heard a complete and detailed confession to the sexual assault and murder of an 11-year-old, the U.S. Supreme Court refused to conclude that such evidence was so “overwhelming” as to excuse federal constitutional error. Given this result and the underlying reasoning employed by the Supreme Court, no objective observer can fail to deduce the lesson that the court fully expects and intends that a finding of reversible error is to be the rule, not the exception.
Indeed, at least one justice of the California Supreme Court has paid heed to Sullivan. In People v. Sims (1993) 5 Cal.4th 405, the majority held that the admission of the defendant’s confession was harmless error. (Id., at pp. 447-448.) However, in a ringing dissent, Justice Mosk concluded to the contrary.
In this regard, Justice Mosk first emphasized that Chapman “is intolerant and unforgiving of error.” (Sims, supra, 5 Cal.4th at p. 474 (dis. opn. of Mosk, J.).) Then, per Sullivan, Justice Mosk observed that the “focus under Chapman is what the jury actually decided and whether the error may have tainted its decision . . . [Par.] As a consequence, the focus under Chapman is not what a reviewing court might itself decide on a cold record . . . By its very terms, Chapman precludes a court from finding harmlessness based simply ‘upon [its own] view of “overwhelming evidence. “‘ [Citation.]” (Id., at pp. 475-476.) Given these precepts, Justice Mosk then proceeded to castigate the majority’s harmless error analysis:
“the majority rely on what they deem ‘overwhelming’ evidence of guilt apart from defendant’s confessions. [Citation.] But, as explained, Chapman effectively prohibits an appellate court from indulging in its own views as to the weight of the properly admitted evidence. Rather, it requires the court to concentrate on the improperly admitted evidence from the perspective of the jury. This the majority do not even attempt.” (Id., at pp. 476-477.)
In short, both Justice Mosk and the Sullivan opinion provide clear and specific guidance that intermediate appellate courts are to engage in a close study of the actual effect that a federal constitutional error has had “‘on the factfinding process at trial.’ [Citation.]” (Sullivan, supra, 124 L.E.2d 182, 189.) Given this guidance, it is now the duty of defense counsel to ensure that California appellate courts engage in the searching inquiry which is mandated by the United States Constitution.
