(dissenting) — I respectfully dissent from the majority opinion on the issue of Finch’s restraint by the trial court. The majority essentially negates our decision in State v. Hutchinson, 135 Wn.2d 863, 887-88, 959 P.2d 1061 (1998), cert. denied, 525 U.S. 1157 (1999), and ignores the trial court’s findings of fact on the restraint questions we specially remanded to the trial court to decide.
Our decisions on in-court restraint of a defendant during *873trial are clear. The defendant’s right to a fair trial under the Sixth and Fourteenth Amendments to the United States Constitution is violated if the defendant appears before a jury in shackles or other restraints. Because restraints have implications as to a defendant’s dangerousness, an appearance in restraints as we noted in Hutchinson may deprive a defendant of the presumption of innocence before a jury.
We have also noted, however, this right is not absolute. A trial court has broad discretion to determine if a defendant’s conduct is so dangerous or disruptive as to require restraints in the courtroom. State v. Breedlove, 79 Wn. App. 101, 113-14, 900 P.2d 586 (1995). In the seminal case State v. Hartzog, 96 Wn.2d 383, 400, 635 P.2d 694 (1981), we enunciated the factors a trial court must consider before a defendant may be restrained in court:
[T]he seriousness of the present charge against the defendant; defendant’s temperament and character;[24] his age and physical attributes; his past record; past escapes or attempted escapes, and evidence of a present plan to escape; threats to harm others or cause a disturbance; self-destructive tendencies; the risk of mob violence or of attempted revenge by others; the possibility of rescue by other offenders still at large; the size and mood of the audience; the nature and physical security of the courtroom; and the adequacy and availability of alternative remedies.
By this time, trial courts should be well aware the Hartzog factors require an individualized assessment on the record of the necessity for restraint of a defendant in a courtroom proceeding. This is not a decision that can be delegated by the trial court to correctional personnel.
*874In the present case, there were ample grounds supporting an individualized decision that Charles Finch should be restrained at trial. Neither we nor the trial court should be oblivious to the fact Finch was charged with a double murder. He had a history of convictions for five prior felonies including manslaughter and assault with a dangerous weapon. Moreover, in the process of arresting Finch for these crimes, law enforcement personnel attempted to subdue him with tear gas. Despite the tear gas, three deputy sheriffs were required to subdue him. The trial court noted, “And the Court was aware of two of those officers as being very strong, very large deputy sheriffs.” Report of Proceedings (Remand) at 1. Finch is a large man who made threats to his wife in the course of the alleged crime. He took his ex-wife and her elderly mother as hostages. Armed, Finch kept law enforcement personnel at hay by shooting indiscriminately in their direction. Finch, while incarcerated, attempted suicide. The proposed jury room for trial was relatively small, giving Finch ready access to some of the witnesses, including his ex-wife against whom he had expressed violent feelings. All of these individualized factors supported the use of restraints under the Hartzog factors.
The trial court here erred, however, because it did not undertake the individualized assessment of Finch’s situation Hartzog requires. Contemporaneous consideration on the record of the Hartzog factors for the particular defendant is required to avoid error and a deprivation of the defendant’s constitutional rights to appear before the jury unrestrained.
A. HARMLESS ERROR
The key consideration in this case, and why I depart from the majority’s analysis, is the majority’s failure to apply harmless error analysis. In Hutchinson, we made it clear a harmless error analysis applies to the question of restraint:
A claim of unconstitutional shackling is subject to harmless *875error analysis. In order to succeed on his claim, the Defendant must show the shackling had a substantial or injurious effect or influence on the jury’s verdict. Because the jury never saw the Defendant in shackles, he cannot show prejudice. Rhoden v. Rowland, 10 F.3d 1457, 1459-60 (9th Cir. 1993). The Defendant does not argue persuasively that he was prejudiced in any way by the unseen restraints. Accordingly, we hold any error was harmless. See United States v. Collins, 109 F.3d 1413, 1418 (9th Cir.), cert. denied, 522 U.S. 870 (1997).
Hutchinson, 135 Wn.2d at 888. The majority simply ignores Hutchinson by failing to do a harmless error analysis of the restraints on Finch during the sentencing phase. All the majority does is conclusorily decide any restraints constitute prejudice and are never harmless error: “Undeniably, the presence of Mr. Finch in physical restraints indicated to the jury that his character is dangerous and prejudiced him in such a manner that warrants a reversal of his death sentence.” Majority op. at 865.1 disagree.
After finding Finch guilty of aggravated first degree murder, the jury had to decide whether to sentence Finch to death or life in prison without the possibility of release or parole. These were the only two sentences allowed under RCW 10.95.030.25 The majority decides the presence of restraints impermissibly induced the jury to choose the death penalty because the restraints indicated to the jury Finch was too dangerous to be turned loose on the prison population.
This conclusion requires a substantial suspension of *876disbelief. It requires one to assume that uppermost in the jurors’ minds when they were considering whether to sentence Finch either to death or life imprisonment without the possibility of parole was his future dangerousness to the prison population to which he would be exposed for the rest of his life.261 do not believe it plausible that solicitude for the safety of imprisoned criminals persuaded the jury to impose the death sentence on Finch.27
Equally implausible is the majority’s notion that but for the appearance of restraints on Finch, the jury would not have voted for the death penalty. Having found him guilty of aggravated first degree murder, the jury knew the following information about Finch: He admitted he had been planning to kill Modlin, a blind man who was essentially defenseless, for three months. 1 Trial Report of Proceedings at 64. He ambushed and killed a police officer for no reason other than “I don’t like the lousy cops.” 1 Trial Report of Proceedings at 92. He was without remorse, as exemplified by his statement in a letter: “I’m just sorry that I didn’t finish the job I started.” 11 Trial Report of Proceedings at 1984. He had committed violent crimes in the past, including manslaughter, first degree rape, and assault with a dangerous weapon, as well as two burglaries. 14 Trial Report of Proceedings at 2466. In spite of all that, the majority holds it was only the appearance of Finch in restraints that impermissibly planted the seed in the jurors’ heads that Finch might be a dangerous man, causing the jury to choose the sentence of death, rather than life in prison.
The majority’s refusal to consider harmless error is *877inexplicable. After engaging in a harmless error analysis to uphold the guilty verdict in the guilt phase, despite the jurors’ knowledge of the restraints on Finch, the majority rejects a similar harmless error analysis for the penalty phase verdict, giving as a reason for doing so only that “[t]he evidence considered during the special sentencing proceeding is of a more subjective nature dealing with not only the nature of the crimes involved but also with personal history and the character of the Defendant.” Majority op. at 863. The majority’s approach is illogical. If Finch’s restraint during the guilt phase, when the jury had yet to decide Finch was actually guilty, was harmless error, how can the presence of restraints during the sentencing phase not also be harmless?
Moreover, while it is true that consideration of mitigating factors during the capital sentencing phase calls for a “reasoned moral response,” In re Personal Restraint of Rupe, 115 Wn.2d 379, 387, 798 P.2d 780 (1990) (quoting California v. Brown, 479 U.S. 538, 545, 107 S. Ct. 837, 93 L. Ed. 2d 934 (1987) (O’Connor, J., concurring)), the majority offers no clue as to why this makes a difference in concluding we should not even do a harmless error analysis. The mere presence of some subjectivity in the sentencing decision does not logically compel the majority’s tendentious conclusion.28 Indeed, if we do the analysis, we may *878conclude the error was not harmless, but we should at least do the analysis.
The majority’s refusal to apply a harmless error analysis here is contrary to a huge body of law allowing it. The United States Supreme Court “has recognized that most constitutional errors can be harmless.” Arizona v. Fulmi-nante, 499 U.S. 279, 306, 111 S. Ct. 1246, 113 L. Ed. 2d 302 (1991). The rule is as follows:
[I]f the defendant had counsel and was tried by an impartial adjudicator, there is a strong presumption that any other errors that may have occurred are subject to harmless-error analysis. The thrust of the many constitutional rules governing the conduct of criminal trials is to ensure that those trials lead to fair and correct judgments. Where a reviewing court can find that the record developed at trial establishes guilt beyond a reasonable doubt, the interest in fairness has been satisfied and the judgment should be affirmed.
Rose v. Clark, 478 U.S. 570, 579, 106 S. Ct. 3101, 92 L. Ed. 2d 460 (1986).29
In Washington constitutional errors are generally deemed prejudicial, State v. Caldwell, 94 Wn.2d 614, 618, 618 P.2d 508 (1980), and the State bears the burden of proving harmless error beyond a reasonable doubt. State v. Guloy, 104 Wn.2d 412, 425, 705 P.2d 1182 (1985), cert. denied, 475 *879U.S. 1020, 106 S. Ct. 1208, 89 L. Ed. 2d 321 (1986). Most directly on point, however, is Hutchinson, where we said, “A claim of unconstitutional shackling is subject to harmless error analysis. In order to succeed on his claim, the Defendant must show the shackling had a substantial or injurious effect or influence on the jury’s verdict.” Hutchinson, 135 Wn.2d at 888. Despite Hutchinson, the majority simply concludes restraints always have a “substantial or injurious effect,” irredeemably prejudicing the jurors, hardening their hearts and implacably impelling them to the death penalty verdict.
Also missing from the majority opinion is a discussion of constitutional law. Where was the constitutional error here? Finch had no presumption of innocence at the sentencing hearing. The majority identifies no constitutional infirmity and does not say what constitutional right Finch was denied.
Because the majority does not permit a harmless error analysis, it establishes the principle that no matter how heinous a crime, no matter how certain the guilt, in all circumstances, without exception, if during the sentencing phase of a trial the jury learns the defendant is being physically restrained, the defendant is deprived of some right, the verdict is irretrievably tainted, and a retrial of the sentencing phase must occur. While such a conclusion may obtain, absent harmless error, for the guilt phase of a trial,30 during which the accused is presumed innocent until the verdict of guilty is actually pronounced, it is totally incomprehensible for the penalty phase, where the presumption of innocence no longer exists for the now-convicted murderer awaiting sentencing. Commonwealth v. Edmiston, 535 Pa. 210, 634 A.2d 1078, 1089 (1993) (“[Pjresumption of innocence which accompanies the accused throughout the proceedings to determine his guilt *880has no direct application to the sentencing determination.”).31 As the Utah Supreme Court said:
This presumption [of innocence] no longer applies during the penalty phase of a capital trial. Indeed, a jury sitting at a penalty phase recently found the defendant guilty of first degree murder during the guilt phase of the trial. As a result, the constitutional foundation for the right to be free from shackles or other restraint no longer exists, and we accord greater significance to public safety concerns. Where the defendant has already lost the presumption of innocence, it is within the sound discretion of the trial court to determine the safety measures necessary to insure the security of the courtroom and its occupants. These safety measures may include shackling a defendant in appropriate circumstances.
State v. Young, 853 P.2d 327, 350 (Utah 1993) (footnotes omitted).
The majority cites only one case Elledge v. Dugger, 823 F.2d 1439 (11th Cir. 1987), cert. denied, 485 U.S. 1014, 108 S. Ct. 1487, 99 L. Ed. 2d 715 (1988), to support its proposition that the presence of physical restraints during the sentencing phase of a capital murder trial always requires a retrial, and that harmless error analysis does not apply. Elledge is in fact the only case that so holds. All other courts to have considered the presence of physical restraints on convicted murderers during sentencing proceedings disagree with Elledge.32
First among those decisions was the Florida Supreme Court’s decision in Elledge v. State, 408 So. 2d 1021 (1981), cert. denied, 459 U.S. 981, 103 S. Ct. 316, 74 L. Ed. 2d 293 *881(1982).33 There, the trial judge ordered the convicted defendant to wear leg irons during the death penalty hearing after learning the defendant had stated his intention to attack the bailiff, and was proficient in karate. The Florida Supreme Court, after noting the presumption of innocence attaches to an accused person, said:
But appellant did not stand before the sentencing jury as an innocent man; rather he stood as a confessed murderer of three persons. The critical issue in a restraint case is the degree of prejudice caused by the restraints. Here, we can find very little prejudice since the appellant was an avowed dangerous individual.
Id. at 1022-23. Likewise, Finch was a convicted double murderer, a “dangerous individual” by any measure, at his sentencing hearing. It is difficult to understand how his jurors could have been first inclined to sentence him to life in prison but then changed their minds and unanimously sentenced him to death upon discovering he was wearing Velcro ankle restraints. But the majority, following its one, lonely case, Elledge, does not permit considerations of prejudice, despite the Hutchinson holding to the contrary.
In Bowers v. State, 306 Md. 120, 507 A.2d 1072, cert. denied, 479 U.S. 890, 107 S. Ct. 292, 93 L. Ed. 2d 265 (1986), Maryland’s highest court held the trial court had not abused its discretion by ordering a convicted murderer to wear leg irons during his death penalty sentencing hearing. Similarly to the Florida Supreme Court in Elledge, the Bowers court said:
In this case there is no presumption of innocence because Bowers was convicted at the earlier trial and we affirmed on appeal. He stands in the position of a convicted felon brought before a trial court for sentencing. He thus is unlike the ordinary defendant who at trial stands clothed with a presumption of innocence.
*882Id. at 1078. After a lengthy, scholarly review of the history of restraining accuseds at trial, the court concluded Bowers was not prejudiced by the restraints, and the trial court had not abused its discretion in ordering it. Id. at 1081.
A subsequent Maryland case reaffirmed the reasoning of the Bowers decision. In Hunt v. State, 321 Md. 387, 583 A.2d 218 (1990), cert. denied, 502 U.S. 835, 112 S. Ct. 117, 116 L. Ed. 2d 86 (1991), upholding the trial court’s restraint of a convicted murderer during the death penalty sentencing hearing, the court observed:
Shackling a defendant during the guilt/innocence phase of trial is inherently prejudicial because it highlights the “need to separate a defendant from the community at large. ...” Holbrook v. Flynn, 475 U.S. at 569, 106 S. Ct. at 1346, 89 L. Ed. 2d at 534. This concern is not as great during the sentencing hearing. It is clear that the defendant will be separated from the community. The only issue is whether the defendant will receive the death penalty or life imprisonment. The defendant’s guilt of first degree murder is established and the jury is less likely to be prejudiced by the defendant’s appearance in leg irons.
Id. at 228. Once again, a court, unlike Elledge and the majority, has taken into account prejudice to the convicted murderer in its determination of the propriety of physical restraints during a death penalty sentencing hearing.
Two state courts have specifically addressed the holding of Elledge and rejected it. The Nevada Supreme Court in Duckett v. State, 104 Nev. 6, 752 P.2d 752 (1988), held a trial court did not commit reversible error by refusing Duckett’s request to he free of manacles and prison garb during the sentencing phase of his trial. As had other courts, Duckett noted a convicted murderer’s constitutional right to appear in court in street clothes does not exist at a postconviction sentencing hearing. Id. at 755. In State v. Young, 853 P.2d 327, 350 n.95 (Utah 1993), the Utah Supreme Court noted, “A jury may expect that a person it *883has just found guilty of murder will be restrained in some fashion.”34 The court went on to note:
Further, defendant has shown no prejudice resulting from the shackling. Aside from showing that some jurors briefly viewed the shackles, defendant has offered no evidence of the effect of the shackles upon the jurors or of the likelihood that the shackles had a bearing on the verdict. A defendant is not necessarily prejudiced when jurors briefly view him wearing shackles.[35] The trial court took steps to minimize the effects of the shackles by placing them underneath defendant’s coat and by having those present remain seated while the court and the jurors entered the room. Therefore, the effect of the shackles did not likely prejudice defendant.
Id. at 351 (footnote omitted). Again, a court took possible prejudice into account.
In Duckett v. Godinez, 67 F.3d 734 (9th Cir. 1995), cert. denied, 517 U.S. 1158, 116 S. Ct. 1549, 134 L. Ed. 2d 651 (1996), the Court of Appeals for the Ninth Circuit considered Duckett’s habeas corpus petition and addressed the restraint question. The court largely adopted Elledge, with one important exception: it asked if Duckett had been prejudiced by shackling, and adopted a harmless error analysis for the determination. Id. at 749 (remanding for evidentiary hearing on the issue of prejudice).36
In summary, the majority ignores our recently reaffirmed *884rule in Hutchinson and adopts a singular minority case as authority for its conclusion we should not review the restraint of Finch during the sentencing phase of his trial for harmless error. In so doing, it elevates form over substance. In the words of the United States Supreme Court, the harmless error rule “ ‘promotes public respect for the criminal process by focusing on the underlying fairness of the trial rather than on the virtually inevitable presence of immaterial error.’ ” Rose, 478 U.S. at 577 (quoting Delaware v. Van Arsdall, 475 U.S. 673, 681, 106 S. Ct. 1431, 89 L. Ed. 2d 674 (1986)). There was no unfairness to Finch. The majority’s approach here will not promote public respect for the criminal process.
B. REFERENCE HEARING
Turning to the question of whether there was any error at all in this case, we ordered the unusual procedure of a reference hearing in a direct appeal. Specifically, we directed two questions to the trial court:
(1) Was it possible for the jury to see that the Defendant was restrained when entering, leaving or sitting in the hearing room?
(2) If the restraints were not visible, was it possible for the jury to know that the Defendant was restrained in some way based on the restriction of the Defendant’s movements?
Washington State Supreme Court Order, Feb. 20, 1998. The trial court conducted an April 9, 1998 hearing on this question and answered both questions directed by us in the negative:
1. At the hearing on April 27, 1995, it was not possible for the jurors to see that the Defendant [sic] was restrained when he was entering, leaving, or sitting in the hearing room.
2. At that hearing, the defendant’s movements were not restricted in any way that would make it possible for the jurors to know that the defendant was restrained. It is possible that some of the jurors may have speculated that he was restrained, but it is not possible that any of them knew.
Rather than relying on the trial court’s enunciated find*885ings of fact, the majority chooses instead to parse out sentences in the trial court’s oral ruling in order to discredit the trial court’s findings. Even though the trial court found “it is not possible that any of [the jury] knew” Finch was in Velcro leg restraints, the majority “finds” just the opposite: “it was clearly possible that the jury could have known that the Defendant was restrained.” Majority op. at 857. And based on the mere possibility the jury knew of the restraints, as opposed to the jury’s actual knowledge of the restraints, the majority decides to remand this case for a retrial of the sentencing phase.
Frankly, I find this appellate court fact-finding offensive to the trial judge. We asked the trial court to hold a hearing and make findings of fact in answer to two specific questions. This was done. Now the majority decides the trial court did not know what it was talking about because of what the majority decides are inconsistent statements in the oral ruling. Nobody on this court knows what the Ginni Stevens Hearing Room in the Snohomish County Courthouse looked like or even how it was arranged for the Finch proceedings. The trial court judge in this case does. Nevertheless, the majority decides not to believe him. A decent respect for the work of our trial court judges should preclude such tendentious second-guessing of their findings of fact where those findings are supported by substantial evidence.
Plainly, the trial court, with the benefit of the witnesses, demonstrations, and its own consideration of the factual circumstances of this case including actual, personal knowledge of the layout of the courtroom, determined the jury would not have seen Finch’s restraints before defense counsel specifically requested the trial court remove any skirting around the counsel table so the jury could see the restraints on the defendant. Until that time, the restraints were not visible to the jury. The majority presumes to substitute its understanding of the facts for the trial court’s, even though the trial court was obviously in a better position to see and understand the circumstances of the *886physical layout of the courtroom. What substantial evidence supporting its own findings of fact does the majority rely upon?
Given the majority’s analysis of the harmless error question, some might believe the majority had already determined any form of restraint of a criminal defendant during trial is per se prejudicial error in Washington. If so, then the special reference hearing we ordered the trial court to conduct was meaningless window dressing and a waste of everybody’s time.
The general rule trial courts should follow is that physical restraints of defendants are the exception and not the rule. If there are clear security concerns regarding a defendant’s presence in the courtroom, the trial court must undertake the individualized assessment of the circumstances required by Hartzog and undertake such an analysis on the record before ordering any kind of physical restraint of the defendant in the courtroom.
At the same time, courts cannot be oblivious to the realities of courtroom security. I am not so confident as the majority Finch was no threat simply because he had not recently expressed any violent tendencies or been a security problem. We should not announce a rule that requires trial court judges to wait until a defendant on trial commits a violent act in the courtroom before physical restraints may be considered. Rather, we should leave such evaluations to the discretion of the trial court judge, who, after all, is ultimately responsible for the safety of jurors, court personnel, and spectators, and review such evaluations only for abuse of discretion. It is easy to decide in our Olympian detachment, long after the fact and far from the courtroom, that an accused murderer posed no threat to anyone; it is not so easy to make such decisions when one is personally responsible—right now, on the spot—for preventing harm to others.
When the majority can believe a large man who has previously committed violent acts, including manslaughter, first degree rape, and assault with a deadly weapon, who *887was on trial for aggravated first degree murder after killing two people including a blind man and a police officer, who took hostages in the course of the crimes, who had threatened a key witness in the case, who had attempted suicide, and who, despite a discharged tear gas canister in a trailer, required three armed deputies to subdue him is not a person about whom the trial court judge could have security concerns, I express little wonder the public finds the judiciary so often out of touch with reality. I would affirm the judgment on the verdict of the jury, both as to Finch’s culpability and as to his sentence.
Durham, J., and Dolliver, J. Pro Tern., concur with Tal-MADGE, J.
24The majority opinion disregards this factor. Finch premeditated and in cold blood shot a blind man in the head, and then murdered a deputy sheriff from ambush. He was charged with aggravated first degree murder and the State sought the death penalty. Nevertheless, the majority says at 851, astoundingly, “[T]he crimes charged in this case do not warrant physical restraints.” One may well wonder what crimes, in the majority’s view, would be serious enough to warrant physical restraints. The majority’s statement here effectively abrogates the first of the Hartzog factors. If indeed the majority wishes to overrule this part of Hartzog, it should say so explicitly.
The Legislature enacted very strong language to ensure one convicted of aggravated first degree murder shall never leave prison. RCW 10.95.030(1) provides:
Except as provided in subsection (2) of this section, any person convicted of the crime of aggravated first degree murder shall be sentenced to life imprisonment without possibility of release or parole. A person sentenced to life imprisonment under this section shall not have that sentence suspended, deferred, or commuted by any judicial officer and the indeterminate sentence review board or its successor may not parole such prisoner nor reduce the period of confinement in any manner whatsoever including but not limited to any sort of good-time calculation. The department of social and health services or its successor or any executive official may not permit such prisoner to participate in any sort of release or furlough program.
Notably, Finch himself did not argue to the jury he posed no danger to the prison population. Finch may well have concluded such an argument coming from a double murderer is so absurd it would have angered the jury.
Although we held in State v. Gentry, 125 Wn.2d 570, 641, 888 P.2d 1105, cert. denied, 516 U.S. 843, 116 S. Ct. 131, 133 L. Ed. 2d 79 (1995), as an evidentiary matter, “it was not unreasonable for the prosecuting attorney to argue that the Defendant is a dangerous person who would continue to present a danger even while incarcerated[,]” I deem it highly unlikely that such considerations would sway a sentencing jury one way or the other after it had already found a defendant guilty of aggravated first degree murder.
That some element of subjectivity enters into a jury’s deliberations during the penalty phase is inevitable. Nevertheless, the United States Supreme Court has admonished the states to observe strict procedures to avoid caprice in sentencing. As we said in State v. Dodd, 120 Wn.2d 1, 13-14 n.2, 838 P.2d 86 (1992):
The Eighth Amendment, which prohibits “cruel and unusual punishments”, applies to the states through the Fourteenth Amendment. Robinson v. California, 370 U.S. 660, 8 L. Ed. 2d 758, 82 S. Ct. 1417 (1962). Infliction of the death penalty especially invokes the concerns of the Eighth Amendment because the penally is irrevocable. Furman v. Georgia, 408 U.S. 238, 306, 33 L. Ed. 2d 346, 388, 92 S. Ct. 2726, 2760 (1972). The Supreme Court, therefore, imposes procedural and substantive restrictions on legislatures and sentencing authorities, to ensure that “capital punishment is not imposed without . . . serious and calm reflection . . . .” Thompson v. Oklahoma, 487 U.S. 815, 856, 101 L. Ed. 2d 702, 733, 108 S. Ct. 2687, 2710 (1988). The sentencing scheme must not allow the death penalty to be wantonly or freakishly imposed, it must direct and limit jury discretion, to minimize the risk of arbitrary or *878capricious action, and it must allow particularized consideration of relevant aspects of the character and record of each defendant, and the circumstances of the offense, before imposition of the sentence. Gregg v. Georgia, 428 U.S. 153, 188-89, 49 L. Ed. 2d 859, 883, 96 S. Ct. 2909, 2932, reh’g denied, 429 U.S. 875, 50 L. Ed. 2d 158, 97 S. Ct. 197, 198 (1976); Woodson v. North Carolina, 428 U.S. 280, 304-05, 49 L. Ed. 2d 944, 960-61, 96 S. Ct. 2978, 2991 (1976).
Accord United States v. Smith, 987 P.2d 888, 892 (2d Cir.), cert. denied, 510 U.S. 875, 114 S. Ct. 209, 126 L. Ed. 2d 165 (1993); United States v. Casoni, 950 P.2d 893, 917 (3d Cir. 1991); United States v. Mackey, 114 F.3d 470, 473 (4th Cir. 1997); White v. Johnson, 153 F.3d 197, 202 (5th Cir. 1998); United States ex rel. Thomas v. O’Leary, 856 F.2d 1011, 1017 (7th Cir. 1988); United States v. Raether, 82 F.3d 192, 194 (8th Cir. 1996); Rice v. Wood, 77 F.3d 1138, 1141 (9th Cir.), cert. denied, 519 U.S. 873, 117 S. Ct. 191, 136 L. Ed. 2d 129 (1996); United States v. DeSoto, 950 F.2d 626, 630 (10th Cir. 1991); Luu v. People, 841 P.2d 271, 273 (Colo. 1992); State v. Gerardi, 237 Conn. 348, 677 A.2d 937, 944 (1996); Ex parte Fierro, 934 S.W.2d 370, 372 (Tex. Crim. App. 1996), cert. denied, 521 U.S. 1122, 117 S. Ct. 2517, 138 L. Ed. 2d 1019 (1997).
The majority cites pages of authority for this uncontroversial point, but that authority is applicable only to the guilt phase of the trial, and does not bear on the sentencing procedures at issue here.
Accord McNew v. State, 271 Ind. 214, 391 N.E.2d 607, 612 (1979); Bowers v. State, 306 Md. 120, 507 A.2d 1072, 1078, cert. denied, 479 U.S. 890, 107 S. Ct. 292, 93 L. Ed. 2d 265 (1986); Matter of Wayne M., 121 Misc. 2d 346, 467 N.Y.S.2d 798, 801 (N.Y. Fam. Ct. 1983); Duckett v. State, 104 Nev. 6, 752 P.2d 752, 755 (1988); State v. Wiles, 59 Ohio St. 3d 71, 571 N.E.2d 97, 119 (1991), cert. denied, 506 U.S. 832, 113 S. Ct. 99, 121 L. Ed. 2d 59 (1992).
The majority does not state Elledge is a minority rule and fails to discuss or at least cite to other cases with contrary views.
The Court of Appeals for the Eleventh Circuit in Elledge, the case the majority relies on, ordered a retrial upon Elledge’s writ of habeas corpus after the state court verdict.
Indeed, Finch’s jury may have devoutly wished for Finch to be restrained during the sentencing hearing, given Finch’s history of murderous violence. As Duckett observed, “Conceivably, after being convicted of two heinous crimes for which the death penalty could have been imposed, Duckett might have concluded that he had nothing to lose from further acts of violence.” Duckett v. State, 752 P.2d at 755.,
35The Court of Appeals for the Eleventh Circuit agrees with this statement. See Gates v. Zant, 863 F.2d 1492, 1501 (11th Cir.) (“defendant is not necessarily prejudiced by a brief or incidental viewing by the jury of the defendant in handcuffs.”), cert. denied, 493 U.S. 945, 110 S. Ct. 353, 107 L. Ed. 2d 340 (1989).
There are a prodigious number of cases considering the possibility of prejudice stemming from the restraint of defendants. See, e.g., United States v. Mayes, 158 F.3d 1215, 1226 (11th Cir. 1998) (defendant failed to show prejudice because jury did not see restraints); Johnson v. State, 704 N.E.2d 159, 162 (Ind. App. 1999) (“The fact that a defendant has been seen by jurors while being transported in handcuffs is not a basis for reversal, absent a showing of actual harm.”).