By Logan Dwyer, ACLR Featured Blogger

          This post begins by examining the two kinds of silence present in People v. Clary.[1] The first is a defendant’s choice not to testify at his or her first trial. Evidence of this choice is admissible in a second trial on the same charges under Raffel v. United States.[2] The second kind of silence is a suspect’s refusal to speak to the police once he or she is read the Miranda warning. Evidence of this kind of silence is not admissible should the suspect be tried, as the Supreme Court held in Doyle v. Ohio.[3] At the trial level in Clary, the Superior Court of Michigan incorrectly allowed evidence of both kinds of silence to be presented to the jury.[4] The Michigan State Supreme Court ultimately reversed Clary’s convictions, citing Doyle.[5] Michigan sought review by the Supreme Court, seeking to have Doyle “revisited and overruled” in light of the original meaning of the Fifth Amendment’s self-incrimination clause.[6] As the relevant precedent shows, the Supreme Court was correct in declining to grant certiorari.

           At the highest level of generality, Clary addresses a very old question: under what circumstances, if any, should evidence that a defendant kept silent be admissible against the defendant in court? This seemingly simple question has generated a plethora of different answers, depending on the context in which it is asked.[7]As a purely evidentiary matter, silence can be barred as hearsay, qualifying as a non-verbal assertion.[8]A defendant’s silence, however, can also be useful in impeaching the defendant—and testimony used to impeach is not hearsay.[9] When a defendant presents an alibi at trial that the defendant did not tell the police, for instance, it is easy to infer that the alibi was fabricated “to fit within the seams of the [prosecution’s] case.”[10] It should be noted that defense lawyers often advise their clients against talking to the police and against testifying at trial as a matter of course, so any inference from that silence likely uncertain.[11]

          The Supreme Court has addressed this problem with its rulings on the Fifth Amendment’s self-incrimination clause. In Griffin v. California,[12] the Court held that a defendant’s decision not to testify at trial generally cannot be used by the prosecution.[13] Nor may the trial court instruct the jury that the defendant’s silence can be taken as evidence of guilt.[14] Justice Stewart, joined by Justice White, dissented in Griffin on the grounds that a prosecutor mentioning a defendant’s silence does not compel the defendant to testify, rather the commentary merely “articulate[es] and bring[s] to the light of rational discussion a fact inescapable impressed on the jury’s consciousness.”[15]

           The one exception to Griffin’s rule comes from Raffel.[16] Raffel was tried twice for violating Prohibition in the 1920’s.[17] During the first trial, the arresting agent testified that Raffel admitted his crime; Raffel did not testify.[18] At the second trial, the agent gave the same testimony, but Raffel decided to testify.[19] The prosecutor then cross-examined Raffel, asking him why he kept silent at the first trial.[20] The Court, reasoning that “[t]he safeguards against self-incrimination are for the benefit of those who do not wish to become witnesses in their own behalf, and not for those who do,” held the cross-examination did not violate the Fifth Amendment.[21] Thus, a defendant’s silence in a prior trial is fair game, provided the defendant testifies at the subsequent trial.

            Though its reasoning was constitutional, Raffel’s holding also makes sense from an evidentiary perspective. Seeing the prosecution present its case likely demonstrates its holes more vividly than pretrial proceedings could. Defendants may consider those holes revealed by the first trial in developing strategy for a second trial. Therefore, it seems likely that when a defendant testifies at a second trial after keeping silent at the first, the decision is strategic instead of genuinely exculpatory.

          The admissibility of a defendant’s silence is further complicated by Miranda v. Arizona,[22] which famously held that the Fifth Amendment compels police officers to inform suspects of their right to remain silent and that “anything said can and will be used against [them] in court.”[23] These warnings are meant to reduce the “inherently compelling pressures” of custodial interrogation, reminding suspects that the Fifth Amendment protects them from self-incrimination.[24]Miranda itself held that “[t]he prosecution may not . . . use at trial the fact that [the defendant] stood mute . . . in the face of accusation”[25] Ten years after Miranda, the Court reaffirmed this part of its holding in Doyle v. Ohio.[26] There, the State argued that silence afterMiranda warnings should be admissible to impeach a defendant’s alibi.[27] While it recognized “the importance of cross-examination,” the Court rejected the state’s argument.[28] First, it noted that silence after the Mirandawarnings is “insolubly ambiguous.”[29] It further found an “implicit . . . assurance” in the warnings that “silence will carry no penalty.”[30] In light of this implied understanding, the Court reasoned it would be an unfair violation of due process to allow post-Miranda warning silence to be admitted against the defendant at trial.[31]

          Doyle’s prohibition is narrower than it may seem at first. In order for a defendant’s silence to be excluded under Doyle, two elements must be met. First, the silence must occur after arrest and warning; the Court has held that silence before arrest and silence after arrest but before Miranda warnings can both be used to impeach a defendant at trial.[32] Second, the Miranda warnings must be complete and in the proper form to exclude a defendant’s silence at trial; silence, or even statements, after defective warnings are still admissible.[33]Additionally, the prosecution may only use the defendant’s post-Miranda silence to impeach the defendant. Because there would be nothing to impeach should the defendant not testify, these decisions are only relevant in trials where a defendant takes the stand.

          The prosecution used both Raffel and Doyle silences in Clary.[34] During his state court appeals, Clary argued that the trial court erred in allowing the prosecution to use his silence at his first trial and his silence at arraignment (after he was given Miranda warnings) against him.[35] Michigan’s Court of Appeals agreed;[36] the Michigan Supreme Court found only use of the Doyle silence impermissible and remanded for a new trial.[37]Michigan then appealed the case on the theory that Doyle is inconsistent with the Framers’ intended function of the self-incrimination clause.[38]

          Doctrinally, the major problem with this argument is that Doyle is not a self-incrimination clause case. Michigan’s misreading is understandable, since Doyle is about the Miranda warnings, and Miranda was designed as prophylactic protection of the right against self-incrimination.[39] But the Doyle court grounded its holding in the Due Process Clause.[40] Essentially, Doyle stands for the proposition that if police tell a suspect “Anything you say will be used against you,” it is unfair to use the suspect’s decision not to say anything against them at trial.[41] Therefore, even if the Court had decided in Griffin that silence was not testimony, the rule in Doylewould still apply. While it is related to the self-incrimination clause, Doyle is primarily about fairness and ensuring that the government does not trick suspects when it tells them their rights.

           Even if Doyle had been about the self-incrimination clause, Michigan’s argument would still have failed. Focusing on the history of the Fifth Amendment can be illustrative when novel questions come before the court, but Clary was a different matter entirely. First, since Doyle was decided there have been no Circuit Court opinions to the contrary, and no confusion about how to apply that decision.[42] Second, cases like Griffin andMiranda illustrate how the Fifth Amendment has evolved since the founding. In the face of this well-established precedent, reinforced by stare decisis, historical arguments are simply not enough.[43] Moreover, Michigan provided no policy arguments against Doyle in general or as applied in Clay.[44] Because Michigan failed to offer any argument against Doyle short of noting its dissonance with the original purpose of the self-incrimination clause, the Supreme Court was correct to reject Michigan’s petition for certiorari.