When testimony is read back to the jury without the judge's approval, can it be characterized as a mere trial error?

When trial testimony is read back to a jury without the judge's approval, outside the judge's presence and when he is unavailable, the error cannot fairly be characterized as mere trial error. Trial error is error "which may ... be quantitatively assessed in the context of other evidence presented in order to determine whether its admission was harmless." Arizona v. Fulminante, 499 U.S. at 308, 111 S.Ct. at 1264. Structural error, on the other hand, is a "defect[ ] in the constitution of the trial mechanism, which def[ies] analysis by 'harmless-error' standards." Id. at 309, 111 S.Ct. at 1265.

A conviction obtained after a proceeding in which no judge presided and no judicial discretion was exercised is "abhorrent to democratic conceptions of justice." Hays v. Arave, 977 F.2d 475, 481 (9th Cir.1992). In these circumstances, there is a breakdown in the construct of the trial, a structural collapse so severe that its effect on the trial cannot be "quantitatively assessed in the context of the other evidence presented." In short, the error is structural and is not susceptible to harmless error analysis.