THE STATE OF SOUTH CAROLINA
In The Court of Appeals
The State, Respondent,
v.
Randy Wright, Appellant.
Appellate Case No. 2017-002130
Appeal From Berkeley County
Maitѐ Murphy, Circuit Court Judge
Opinion No. 5782
Heard August 19, 2020 – Filed November 18, 2020
REVERSED AND REMANDED
Appellate Defender Joanna Katherine Delany, of
Columbia, for Appellant.
Attorney General Alan McCrory Wilson and Senior
Assistant Deputy Attorney General Deborah R.J. Shupe,
both of Columbia; and Solicitor Scarlett Anne Wilson, of
Charleston, all for Respondent.
HILL, J.: After deliberating two hours in Randy Wright's trial for assault and
battery of a high and aggravated nature (ABHAN), the jury signaled it had reached
a verdict. When the jury returned to the courtroom, the trial court directed the
courtroom clerk to publish the verdict. The clerk read the verdict form, announcing
the jury had found Wright guilty of ABHAN and that the form had been signed by
the forelady. The clerk then stated: "Ladies and gentlemen of the jury if this is your
verdict, would you please signify by raising your right hand?" In response, each
juror raised his or her right hand. Wright then asked the trial court to poll each juror
individually. The court declined, explaining that, in response to the clerk's inquiry
"each of the jurors raised their hand individually." Wright appeals, asserting the
clerk's collective inquiry did not satisfy his polling right. We agree and reverse.
I.
The custom of polling a jury after a verdict developed in English practice, although
no precise method predominated. See Matthew Hale, Pleas of the Crown, 299–300
(Vol. II, 1800) ("[I]f the jury say they are agreed, the court may examine them by
poll . . . ."). Early South Carolina cases permitted polling in the trial court's
discretion. State v. Wyse, 32 S.C. 45, 10 S.E. 612, 615 (1890); State v. Allen, 12
S.C.L. (1 McCord) 525, 526–27 (1822). The trial court's discretion ended in State
v. Linder, which held a poll must be taken if requested and implied each juror must
be polled individually. 276 S.C. 304, 309, 278 S.E.2d 335, 338 (1981).
The right to poll the jury is not in itself a constitutional right but a procedural
protection of the defendant's constitutional right to a unanimous verdict. State v.
Pare, 755 A.2d 180, 188 (Conn. 2000). It also safeguards the right to a public trial.
If the poll reveals the jury's announced verdict is not in fact unanimous, the verdict
cannot stand, and the trial court may, as circumstances warrant, direct further
deliberation or declare a mistrial. State v. Kelly, 372 S.C. 167, 170–71, 641 S.E.2d
468, 470 (Ct. App. 2007). Just as trial counsel has no duty to request a poll, the trial
court has no duty to conduct one without a request. Green v. State, 351 S.C. 184,
196, 569 S.E.2d 318, 324 (2002); Linder, 276 S.C. at 308–09, 278 S.E.2d at 338.
Besides shoring up these rights, individual polling supports several other interests of
justice. The courtroom air thins when the jury returns to deliver its verdict. No other
trial moment demands the solemn clarity individualized inquiry provides. Individual
polling promotes finality and accountability of the verdict stage and enhances the
integrity of the deliberative process by ensuring no juror was coerced in the jury
room. See Humphries v. District of Columbia, 174 U.S. 190, 194 (1899) (observing
object of poll "is to ascertain for a certainty that each of the jurors approves of the
verdict as returned; that no one has been coerced or induced to sign a verdict to
which he does not fully assent"). We note Rule 31 of the Federal Rules of Criminal
Procedure was amended in 1998 to require individual rather than collective polling.
Fed. R. Crim. P. 31(d). The Advisory Committee note accompanying the change
points out collective polling "saves little time and does not always adequately ensure
that an individual juror who has been forced to join the majority during deliberations
will voice dissent from a collective response." Fed. R Crim. P. 31 Advisory
Committee Notes. We agree; as the proverb goes, valor delights in the test.
Even before the 1998 rule change, several federal circuits held a collective question
to the jury asking them to affirm their verdict (by show of hands or by verbal assent),
even if asked in open court, is not the best method for accomplishing the purpose of
a jury poll. United States v. Miller, 59 F.3d 417, 421 (3d Cir. 1995); United States
v. Carter, 772 F.2d 66, 68 (4th Cir. 1985). In states requiring individual polling
upon request, it has been held a collective polling question does not suffice. State v.
Coulthard, 492 N.W.2d 329, 333 (Wis. Ct. App. 1992); Miles v. Com., 256 S.W.3d
46, 46 (Ky. Ct. App. 2008). As the Connecticut Supreme Court has observed:
These cases reflect the understanding, based on common
human experience, that members of a group may react
differently when addressed as a group, and when
addressed individually. They also reflect the notion that
the concept of jury unanimity is sufficiently significant so
as to require that, upon request, each juror be required to
state his or her verdict in open court—individually—to
face the defendant and the state, and confirm, on his or her
own, that the collectively reported verdict is truly his or
hers.
Pare, 755 A.2d at 193 (collecting cases).
Linder did not endorse a particular method of individually polling the jurors, stating
only, "Polling is a practice whereby the court determines from the jurors individually
whether they assented and still assent to the verdict." 276 S.C. at 308, 278 S.E.2d at
338. Because Linder provided no guidance on the mechanics of proper individual
polling, we understand how the trial court could have concluded the clerk's inquiry
was enough. But it was not. We conclude individual polling means each juror must
be separately asked to confirm verbally on the record that the verdict announced is
still his or her verdict. We believe this person-by-person inquiry best advances the
prime reason for individual polling: "to dispel any doubt a party might entertain as
to the propriety of a jury verdict as rendered." 276 S.C. at 309, 278 S.E.2d at 338.
The trial court therefore erred in denying Wright's request for such a poll.
II.
Whether the denial of a jury poll request automatically requires a new trial is a novel
question in South Carolina. Linder does not say, and we have no controlling statute
or rule.
In the past half century, courts have attempted to divide constitutional errors into
two categories: trial errors and structural errors. In general, trial errors that are
harmless do not justify reversal. Structural errors, on the other hand, are reversible
per se, unredeemable by the harmless error doctrine. See Weaver v. Massachusetts,
137 S. Ct. 1899, 1908 (2017) (delineating three broad rationales for classifying an
error as structural); State v. Rivera, 402 S.C. 225, 246–47, 741 S.E.2d 694, 705–06
(2013) (differentiating between structural errors and trial errors subject to harmless
error review). The structural/trial error dichotomy does not cover all trial mistakes;
some, like the polling error here, elude neat classification. See, e.g., Weaver, 137 S.
Ct. at 1911 (noting the Court "has granted automatic relief to defendants who
prevailed on claims alleging race or gender discrimination in the selection of the
petit jury, though the Court has yet to label those errors structural in express terms"
(citations omitted)); State v. Short, 333 S.C. 473, 476–78, 511 S.E.2d 358, 360–61
(1999) (holding no showing of prejudice required when trial court erred in denying
defendant's right to exercise peremptory challenges—a statutory procedural right
designed to ensure constitutional right to an impartial jury; instead, error was
reversible per se).
The structural/trial error distinction is not pivotal to Wright's appeal, for a polling
error is not a pure constitutional error, and resembles both an error affecting the
"framework within which the trial proceeds" (structural error) and "an error in the
trial process itself" (trial error). Arizona v. Fulminante, 499 U.S. 279, 310 (1991).
The reasoning of Weaver, however, offers a rational way out of our classification
dilemma. In Weaver, the Court noted that in addition to relating to a trial's
framework, an error is structural if: (1) the right at issue is designed to protect an
interest other than the defendant's interest in being wrongly convicted; (2) the effects
of the error are "simply too hard to measure"; or (3) the error always results in
fundamental unfairness. 137 S. Ct. at 1908. We believe the denial of the right to
individual polling bears all three of these traits. The polling right protects not only
the defendant from being wrongfully convicted, but also the public's interest in
ensuring the outcome of the criminal trial process is reliable. Denial of the polling
right also defies harmless error analysis. To find the error harmless, we would have
to conclude the lack of a valid poll was an "error which occurred during the
presentation of the case to the jury, and which may therefore be quantitatively
assessed in the context of other evidence presented in order to determine whether
[the error] was harmless beyond a reasonable doubt." Fulminante, 499 U.S. at 307–
08 (1991). We cannot say the lack of a valid poll contributed to the verdict, as the
error occurred after a verdict was announced. It would be an odd end to the matter
to deem it harmless, for in effect we would be presuming the unanimity of the verdict
while simultaneously denying the defendant the only real right he has to check
behind the presumption. Finally, the denial of the polling right caused fundamental
unfairness by "undermining . . . the systemic requirements of a fair and open judicial
process." Weaver, 137 S. Ct. at 1911. If an announced verdict lacks unanimity in
fact, then the harm to the integrity and fundamental legitimacy of the entire trial is
total.
We are mindful of the general rule that a conviction may not be reversed due to
"insubstantial errors not affecting the result." State v. Chavis, 412 S.C. 101, 109,
771 S.E.2d 336, 340 (2015). It is our firm view that depriving a defendant of his or
her polling right is not a technicality, but a material and prejudicial error. See id. at
110 n.7, 771 S.E.2d at 340 n.7 ("[W]e readily acknowledge that there are some
errors, particularly errors of law, which cannot be rendered harmless by
overwhelming evidence."). The individual poll is the best chance the trial court and
the parties have to ensure the sanctity and unanimity of the verdict. It is not enough
to say, as the State does, that jurors seldom recant upon polling. Experience—and
case law—proves they do. Kelly, 372 S.C. at 171–72, 641 S.E.2d at 470–71
(involving a juror's recant and citing other cases where it has occurred). The rarity
of an episode so threatening to the vital center of our jury system is no reason to
ignore it.
Because of the importance of the polling right and the difficulty of deciphering the
harm its denial has caused, many federal circuits and state appellate courts have
deemed the denial reversible per se. United States v. F.J. Vollmer & Co., 1 F.3d
1511, 1522–23 (7th Cir. 1993); Virgin Islands v. Hercules, 875 F.2d 414, 419 (3d
Cir. 1989); Miranda v. United States, 255 F.2d 9, 18 (1st Cir. 1958); Pare, 755 A.2d
at 194; Commonwealth. v. Downey, 732 A.2d 593, 595–96 (Pa. 1999); Miles, 256
S.W.3d 46, 46–47. We are persuaded by the sound reasoning of these decisions and
therefore hold the denial of the defendant's substantial right to an individual poll of
each juror in open court—where each juror must express his or her continued assent
in the announced verdict—is reversible error per se, not subject to a harmless error
analysis. We are convinced such a rule incentivizes compliance with proper polling
procedure and best honors the value of the right itself.
We are aware retrials are costly and impede judicial efficiency. We are equally
aware that appellate reviews requiring lengthy searches through thick transcripts to
sense the net impact of an error on the whole trial can also be costly and inefficient
where, as here, the error is not in what was done, but what was not done.
REVERSED AND REMANDED.
WILLIAMS and KONDUROS, JJ., concur.