Commonwealth v. Hamlett, J., Aplt. ( 2020 )

[J-93-2019] [MO: Saylor, C.J.]
IN THE SUPREME COURT OF PENNSYLVANIA
WESTERN DISTRICT
COMMONWEALTH OF PENNSYLVANIA, : No. 8 WAP 2019
:
Appellee : Appeal from the Order of the
: Superior Court entered September
: 11, 2018 at No. 1172 WDA 2016,
v. : vacating the Judgment of Sentence
: of the Court of Common Pleas of
: Allegheny County entered June 30,
JAMES CALVIN HAMLETT, JR., : 2016 at No. CP-02-CR-0014824-
: 2015 and remanding.
Appellant :
: ARGUED: October 16, 2019
DISSENTING OPINION
JUSTICE WECHT DECIDED: JULY 21, 2020
Does the Commonwealth have a burden to establish harmless error? Unless the
Majority is inclined to overrule our seminal decision on the harmless error doctrine, the
answer incontrovertibly is “yes.” Commonwealth v. Story, 383 A.2d 155, 162 n.11 (Pa.
1978) (“[T]he burden of establishing that the error was harmless beyond a reasonable
doubt rests with the Commonwealth.”). Why, then, may a court raise the issue of
harmless error sua sponte, and then proceed to rule in the Commonwealth’s favor
thereon, despite the Commonwealth’s failure to carry its burden, and despite any attempt
by the Commonwealth to raise, assert, cite, discuss, or apply the harmless error doctrine
to the facts of the case? The Majority appears to suggest that it has resolved this second
question today, but its answer is irreconcilable with the settled answer to the first question.
A burden must be satisfied by the litigant who bears it, not by the adjudicator. Otherwise,
it is no burden at all.
The essential meaning of a “burden” in the law is readily understood by judges,
lawyers, law students, scholars, television commentators, and laypersons alike. It means
that there is some kind of obligation to carry one’s point, lest the point be lost. This might
make the question before us appear to have an obvious answer. Not so. The
jurisprudential waters have significantly muddied over the last two decades. The crux of
today’s appeal is the “tension” that has grown within our harmless error jurisprudence, a
tension that manifests itself in two divergent lines of precedent. On the one hand, there
is the fundamental principle, recognized by the Supreme Court of the United States in
Chapman v. California, 386 U.S. 18 (1967), and by nearly every court that has followed,
that it is the prosecution’s burden to establish the harmlessness of an error that potentially
tainted a defendant’s conviction. See id. at 24; Story, 383 A.2d at 162 n.11;
Commonwealth v. Davis, 305 A.2d 715, 719 (Pa. 1973).1 On the other hand, there is this
1 State courts overwhelmingly recognize this burden. The cases are legion. See,
e.g., State v. James, 443 P.3d 1063, 1079 (Kan. 2019) (“The burden of demonstrating
harmlessness is on the party benefiting from the error, which, in this case, is the State.”);
Coleman v. Binion, 829 S.E.2d 1, 26 (W.Va. 2019) (“In a criminal case, the burden is upon
the beneficiary of a constitutional error to prove beyond a reasonable doubt that the error
complained of did not contribute to the verdict obtained.”) (citation and internal quotation
marks omitted); State v. Larkin, 183 A.3d 589, 595 (Vt. 2018) (“The State bears the
burden of proving that an error . . . is harmless.”); State v. Rivas, 398 P.3d 299, 313-14
(N.M. 2017) (“For a non-structural, constitutional error as has been established here, the
State bears the burden of proving beyond a reasonable doubt that the error was harmless
to the outcome.”); State v. DeLeon, 374 P.3d 95, 100 (Wa. 2016) (“The State bears the
burden of showing that the constitutional error was harmless.”); State v. Perry, 245 P.3d
961, 973 (Idaho 2010) (constitutional error necessitates reversal “unless the State proves
beyond a reasonable doubt that the error complained of did not contribute to the verdict
obtained”) (citation and internal quotation marks omitted); State v. Sayles, 49 S.W.3d 275,
280 (Tenn. 2001) (“Once a constitutional error has been established . . . the burden is
upon the State to prove that the constitutional right violation is harmless beyond a
reasonable doubt.”) (citation and internal quotation marks omitted); Dawson v. State, 608
A.2d 1201, 1204 (Del. 1992) (“The defendant has the initial burden of demonstrating error.
If the defendant is successful in discharging that obligation, the burden of proof shifts to
the State.”).
[J-93-2019] [MO: Saylor, C.J.] - 2
Court’s more recent suggestion that an appellate court’s sua sponte invocation of the
harmless error doctrine is a permissible exercise of the court’s prerogative to apply the
right-for-any-reason doctrine—a notion floated in an undeveloped footnote in
Commonwealth v. Mitchell, 839 A.2d 202, 215 n.11 (Pa. 2003), and applied by this Court
on a handful of occasions since.
Where our precedents hold simultaneously that there is a burden placed upon a
party and that the court may elect at will to shoulder that burden itself, the precedents are
incompatible. I would preserve the foundations of the harmless error doctrine, adhere to
Story, and disavow our Mitchell line of cases. I would enforce the burden recognized in
Chapman.
Every United States Court of Appeals that adjudicates criminal cases also has
come to the same conclusion. See, e.g., United States v. Rivera-Carrasquillo, 933 F.3d
33, 47 (1st Cir. 2019) (“The government bears the burden of proving harmlessness.”);
United States v. Groysman, 766 F.3d 147, 155 (2d Cir. 2014) (“With respect to harmless-
error analysis, the government bears the burden of proof.”); United States v. Franz, 772
F.3d 134, 151 (3d Cir. 2014) (“The government bears the burden of establishing
harmlessness.”); United States v. Garcia-Lagunas, 835 F.3d 479, 488 (4th Cir. 2016)
(“The burden rests on the government, the beneficiary of the error, to show
harmlessness.”); United States v. Gutierrez-Mendez, 752 F.3d 418, 426 (5th Cir. 2014)
(“The government has the burden of establishing harmlessness beyond a reasonable
doubt.”); United States v. Susany, 893 F.3d 364, 368 (6th Cir. 2018) (“The government
bears the burden of proof on harmless error . . . .”); United States v. Robinson, 724 F.3d
878, 888 (7th Cir. 2013) (“The burden of demonstrating harmlessness rests with the
government.”); United States v. Davis, 859 F.3d 592, 597 (8th Cir. 2017) (“The
government bears the burden of proving an error is harmless.”); United States v. Esparza,
791 F.3d 1067, 1074 (9th Cir. 2015) (“The government bears the burden of proving that
the error was harmless beyond a reasonable doubt.”); United States v. Russian, 848 F.3d
1239, 1248 (10th Cir. 2017) (“The government bears the burden of making this
[harmlessness] showing.”); Bester v. Warden, 836 F.3d 1331, 1338 (11th Cir. 2016) (“The
government, not the defendant, bears the burden of establishing that a constitutional error
is harmless.”); United States v. Burnett, 827 F.3d 1108, 1119 (D.C. Cir. 2016) (“The
government bears the burden of proving harmless error.”) (internal citation and quotation
marks omitted).
[J-93-2019] [MO: Saylor, C.J.] - 3
The Majority chooses a different path, opting to provide no standard at all.
Ultimately, the Majority holds that a court may assume the Commonwealth’s burden “in
appropriate cases.” Maj. Op. at 9. The court may conduct a harmless error analysis on
its own initiative, the Majority reasons, pursuant to its “discretionary prerogative.” Id. at
10. Both the Majority and the Concurrence suggest some understanding that our
appellate courts are expected to glean from today’s decision, pursuant to which jurists will
recognize their sua sponte invocation of harmless error to be an “exception to the ordinary
rule,” id. at 9, one that must be used with caution in response to the “extremely occasional
need,” Conc. Op. at 2 (Donohue, J.), to do the Commonwealth’s job for it. I too am
concerned by the “superficiality” with which harmless error analyses are commonly
conducted. Maj. Op. at 12. But I find no practical limitations here, no standard, no rubric
by which to distinguish an “appropriate” case from an inappropriate one, and no reason
to expect that this purported “exception” to a litigant’s burden will prove to be anything
other than the norm, to the extent that the exception has not swallowed the rule already.
There is a straightforward path toward mitigating many of the problems that
Appellant Hamlett has brought to our attention—a path grounded in the foundational case
law from which we have departed. We should preserve the burden, in a form that is
recognizable as such.
The Majority does not discuss much of the jurisprudence underlying this inquiry.
The Majority does not discuss the facts or history of this case, nor the intermediate court’s
application of the law to the facts. See infra Part IV. Aside from a footnote providing
citations to the decisions of this Court that Hamlett specifically challenges, Maj. Op. at 4
n.3, the Majority does not meaningfully confront the precedents at issue. The harmless
error doctrine has a long history; the approach that this Court embraces today is of a more
[J-93-2019] [MO: Saylor, C.J.] - 4
recent vintage. In order to frame the question presented,2 to appreciate Hamlett’s
criticisms of the approach that the Court embraces today, and to articulate the grounds
for my profound disagreement with the Majority, it is necessary to set forth some
background. It is imperative to understand the competing interests at play, to review the
guiding principles of the doctrines at issue, and to consider where our jurisprudence has
brought us and where it will go from here.
I.
Harmless error often is treated as a routine matter in the law, but we must remain
cognizant of the stakes of the inquiry before us. To the individual appellant whose case
is under review, the importance of the inquiry is immense—a harmless error analysis
often represents the line between an accused’s entitlement to relief and the failure of his
appeal. A finding of harmlessness deals a knockout blow, simultaneously acknowledging
the merit in the appellant’s assertion of legal error while nonetheless informing the
2 The Majority does not quote the question presented. We granted allowance of
appeal on the following question, which we rephrased for clarity:
Can the tension between the well-settled rule that the Commonwealth bears
the burden of demonstrating harmless error beyond a reasonable doubt and
the contradictory principle that an appellate court has the ability to affirm a
valid judgment or verdict for any reason appearing as of record be
reconciled? If these conflicting principles must be reconciled in favor of the
Commonwealth proving harmlessness beyond a reasonable doubt, did the
Superior Court err in finding harmless error sua sponte?
Commonwealth v. Hamlett, 202 A.3d 45 (Pa. 2019) (per curiam).
The Majority states that, due to Hamlett’s framing of the appeal, “the issue does
not encompass the narrower question of whether the Superior Court may have erred in
the substantive aspects of its harmless-error review.” Maj. Op. at 3 n.2. However,
because this Court rephrased the question presented, this conclusion is somewhat
suspect. Because it is emblematic of certain problems in our jurisprudence, and because
the Majority holds that such analyses may be conducted sua sponte in any event, I believe
that the Superior Court’s present application of the harmless error doctrine warrants at
least brief comment. See infra Part IV.
[J-93-2019] [MO: Saylor, C.J.] - 5
appellant that no remedy for that error will be forthcoming. And beyond the individual
case lie important institutional concerns. In Story, we cautioned that “there is the danger
that a lenient harmless error rule may denigrate the interests and policies which both
constitutional and non-constitutional rules promote.” Story, 383 A.2d at 164. We further
admonished that “courts must be careful in applying the harmless error rule, for if the
violation of a rule is too readily held harmless, the importance and effectiveness of the
rule is denigrated.” Id. This unpleasant side effect not only can dilute and defeat
important rights, but also can undermine the deterrence value of appellate reversal,
demonstrating to future parties and courts that similar errors are inconsequential, and that
scrupulous care need not be taken to avoid them.3 A rule that frequently is broken without
consequence soon becomes no rule at all.
There is, of course, another side to this coin. It would be unreasonable to demand
perfection throughout the entire course of a criminal prosecution and trial. For that
reason, it is a longstanding tenet of our law that, “although an accused is entitled to a fair
trial, he is not entitled to a perfect one.” Story, 383 A.2d at 164. Modern harmless error
doctrine developed in the early twentieth century as a response to a “widespread and
deep conviction” that appellate review of criminal cases had grown too demanding, and
that appellate courts had become “‘impregnable citadels of technicality’” that often
reversed hard-won convictions on the basis of trivial irregularities. Kotteakos v. United
States, 328 U.S. 750, 759 (1946) (quoting Marcus A. Kavanagh, Improvement of
3 See Hon. Harry T. Edwards, To Err Is Human, But Not Always Harmless: When
Should Legal Error Be Tolerated?, 70 N.Y.U. L. REV. 1167, 1170 (1996) (hereinafter
“Edwards”) (“When we hold errors harmless, the rights of individuals, both constitutional
and otherwise, go unenforced. Moreover, the deterrent force of a reversal remains unfelt
by those who caused the error. In his seminal book on harmless error, entitled The Riddle
of Harmless Error, the late Justice Roger Traynor aptly observed that ‘[i]n the long run
there would be a closer guard against error at the trial, if appellate courts were alert to
reverse, in case of doubt, for error that could have contaminated the judgment.’”) (quoting
ROGER J. TRAYNOR, THE RIDDLE OF HARMLESS ERROR 23 (1970)) (footnotes omitted).
[J-93-2019] [MO: Saylor, C.J.] - 6
Administration of Criminal Justice by Exercise of Judicial Power, 11 A.B.A.J. 217, 222
(1925)). Undoubtedly, it is inefficient to order the repetition of an entire trial because of
an immaterial error that could not possibly have affected the outcome. Thus, the
“harmless error rule can save the time, effort and expense of unnecessary retrials where
the defendant has not been prejudiced by an error.” Story, 383 A.2d at 164. But because
the individual and institutional stakes are high, a balance must be struck: the rule must
aim to “save the good in harmless-error practices while avoiding the bad, so far as
possible.” Chapman, 386 U.S. at 23.
Both the Supreme Court of the United States and this Court historically struck that
balance by constraining the harmless error doctrine with two fundamental rules. First,
before an error may be deemed harmless, “the court must be able to declare a belief that
it was harmless beyond a reasonable doubt.” Davis, 305 A.2d at 719 (quoting Chapman,
386 U.S. at 24). “The second general precept” is that “the burden is on the
Commonwealth to establish that the error was harmless.” Id. at 719; see Chapman, 386
U.S. at 24; Story, 383 A.2d at 162 n.11. The beyond-a-reasonable-doubt standard is
commensurate with the prosecution’s burden at trial, and it would make little sense for an
appellate court to apply a lesser standard to the harmlessness inquiry. Story, 383 A.2d
at 162. Moreover, this Court has explained, the beyond-a-reasonable-doubt standard
reflects the prudent belief that “it is far worse to conclude incorrectly that the error was
harmless than it is to conclude incorrectly that the error was reversible.” Davis, 305 A.2d
at 719.
It is the “second general precept” of harmless error jurisprudence that is on the
chopping block today. Davis, 305 A.2d at 719. Fundamentally, the purpose of assigning
the burden to the prosecution is to minimize the potential for unfairness to the appellant,
for it is the prosecution that benefits from the finding of harmlessness. The Supreme
[J-93-2019] [MO: Saylor, C.J.] - 7
Court of the United States has explained that trial error “casts on someone other than the
person prejudiced by it a burden to show that it was harmless.” Chapman, 386 U.S. at
24. “It is for that reason that the original common-law harmless-error rule put the burden
on the beneficiary of the error either to prove that there was no injury or to suffer a reversal
of his erroneously obtained judgment.” Id. Although there are additional practical
challenges in the application of the doctrine, these two essential safeguards—the
governing standard and the assignment of the burden—maintain the relative balance in
the competing interests that underlie harmless error, and restrain the doctrine from
embracing too much of “the bad” about which Chapman warned us.
In the past two decades, however, this Court began to distort this balance, tipping
the scales in favor of the prosecution. In Mitchell, a capital direct appeal, this Court
agreed with the appellant that the trial court had erred in permitting the prosecutor to
impeach the appellant’s testimony at trial by reference to his exercise of his right to remain
silent. The Court proceeded to a harmlessness inquiry. In a footnote that ultimately would
engender the so-called “tension” in our jurisprudence that we confront today, this Court
stated:
Inexplicably, in this case the Commonwealth offers no alternative argument
that the error was harmless, as it simply argues that no error occurred. We
remind the Commonwealth that the burden of establishing harmless error
rests squarely upon its shoulders. Story, 383 A.2d at 162 n.11. Despite
this lapse by the Commonwealth, we are not without advocacy on this issue
as the question of harmless error was directly raised and addressed by
Appellant. Jurisprudentially, we can affirm the action of the court below on
other grounds. Bearoff v. Bearoff Bros., Inc., 327 A.2d 72, 76 (Pa. 1974).
Mitchell, 839 A.2d at 215 n.11(citation modified).
Although the Mitchell footnote is brief, there is much to unpack in it. It is evident
that sua sponte invocation of harmless error was not commonplace at the time, as
demonstrated by this Court’s scolding of the Commonwealth for its “lapse” in failing to
[J-93-2019] [MO: Saylor, C.J.] - 8
carry its burden—an oversight that this Court found “inexplicable.” Nonetheless, with a
touch of judicial sleight of hand, this Court escaped the suggestion that it had raised the
issue sua sponte at all, capitalizing upon the appellant’s foresight to have argued that the
error was not harmless, and proceeding then to deem that sufficient “advocacy” to decide
the issue. In one sentence, this Court flipped the longstanding burden on its head. The
Commonwealth wholly failed to address, let alone meet, its burden; the appellant not only
met his own burden to prove error, but went above and beyond to preempt the
harmlessness argument that the appellant anticipated but which the Commonwealth itself
never advanced. As a reward for his diligence, this Court held that the appellant
effectively had satisfied his opponent’s burden. It appears that the appellant would have
been better served by less punctilious advocacy. Finally, without any reasoned analysis
or discussion, this Court cited a single civil case referring to the right-for-any-reason
doctrine, ostensibly as a justification for its decision to engage in harmless error analysis
despite the Commonwealth’s failure to litigate the matter at all. This Court did not
acknowledge that its approach facially conflicts with the universal understanding of a
“burden” in the law as an obligation upon a party to prove the claim, or else to suffer
defeat.
Mitchell and its progeny embody the dangers of bad precedent. As sometimes
happens, what began as merely an underdeveloped excuse for an unsound decision soon
became the norm. In Commonwealth v. Moore, 937 A.2d 1062 (Pa. 2007), this Court
again confronted an error as to which the Commonwealth had failed to advance a
harmlessness argument. Citing Mitchell, this Court again paid lip service to the waning
premise that the burden to prove harmless error falls upon the Commonwealth. Yet, the
Court again invoked the right-for-any-reason doctrine as a basis to overlook the
Commonwealth’s omission. See id. at 1073. Mitchell had metastasized into a line of
[J-93-2019] [MO: Saylor, C.J.] - 9
precedent, binding upon this Court and all below, despite its patent but unacknowledged
conflict with Story, Davis, Chapman, and the foundational tenets of harmless error
doctrine. By the time this Court decided Commonwealth v. Allshouse, 36 A.3d 163 (Pa.
2012), Mitchell’s approach was so engrained that the Court started simply citing the right-
for-any-reason doctrine as a justification for sua sponte consideration of harmless error
without even referring to the Commonwealth’s burden as such. See id. at 182 (“It is well
settled that an appellate court has the ability to affirm a valid judgment or verdict for any
reason appearing as of record . . . . This Court may affirm a judgment based on harmless
error even if such an argument is not raised by the parties.”) (citation and internal
quotation marks omitted).
This Court’s treatment of harmless error grew increasingly casual. In
Commonwealth v. Hitcho, 123 A.3d 731 (Pa. 2015), another capital direct appeal, this
Court not only ignored the existence of a burden on the question of harmlessness, but,
unlike earlier cases in this line, glossed over the fact that the Commonwealth had not
advanced a harmlessness argument.4 After spending over a decade with the Mitchell
footnote, it was no longer “inexplicable” that the Commonwealth had failed to carry its
burden; the prosecution now could count on this Court to do the job for it. In Hitcho, this
Court further abbreviated the analysis. The Court not only performed a harmlessness
assessment on the Commonwealth’s behalf, it also declined even to consider a claim of
error. This Court held that “we need not resolve the issue of whether the trial court abused
its discretion in denying Appellant’s motion in limine . . . . Assuming, arguendo, the trial
4 The Commonwealth’s brief in Hitcho made a single, passing reference to harmless
error in the context of one discrete claim, but did not develop an argument on the matter,
and did not refer to the “overwhelming evidence” standard that we would deem dispositive
of a wholly different claim of error. Brief for Commonwealth, Commonwealth v. Hitcho,
691 CAP, at 29 (“Even if [Pa.R.E. 106] was applicable, which it is not, the error would be
harmless error.”). The analysis that I discuss was undertaken by this Court sua sponte.
[J-93-2019] [MO: Saylor, C.J.] - 10
court erred, given the overwhelming evidence establishing first-degree murder, we find
any error to be harmless, and thus not warranting relief.” Hitcho, 123 A.3d at 748. Putting
the proverbial cart before the horse, the Court found harmlessness without finding error.5
Absent any advocacy from the parties on the question, and instead forging ahead
sua sponte, this Court then disregarded that the “overwhelming evidence” variety of
harmless error necessitates that the evidence be uncontradicted—a requirement that
prevents the reviewing court from making questionable assessments of the credibility of
competing evidence from its inherently limited appellate perspective.6 As Story stated:
The requirement that the “overwhelming” evidence relied upon be
uncontradicted follows from the principle that an error cannot be harmless
if “‘honest, fair minded jurors might very well have brought in not guilty
verdicts.’” Davis, 305 A.2d at 721 (quoting Chapman, 386 U.S. at 18). A
jury has the duty to weigh the evidence and resolve conflicts therein. E.g.,
Commonwealth v. Murray, 334 A.2d 255 (Pa. 1975). Unless the evidence
5 We took a similar approach in Allshouse, 36 A.3d at 182. Although we are not the
first court to conduct such an analysis, and certainly will not be the last, this practice also
raises self-evident doctrinal concerns. See Edwards, supra n.3, at 1182 (“Another
troubling aspect of this trend is judicial use of the harmless-error rule to avoid reaching a
difficult issue in a case. Courts sometimes openly decline to decide whether a
defendant’s rights have been violated, instead evading the issue by stating that any error
that might have occurred was harmless. This practice leaves unresolved the question of
whether an error even occurred, thus offering no guidance to trial courts. What may be
an important question of trial error is therefore sidestepped by the application of a doctrine
that itself presupposes the existence of such an error. Nothing suggests that the
harmless-error rule was meant to serve such a purpose.”).
6 For ease of reference, this Court traditionally recognizes three varieties of
harmless error, which we have gleaned from our seminal discussion in Story:
Harmless error exists if the reviewing court is convinced from the record that
(1) the error did not prejudice the defendant or the prejudice was de minimis,
(2) the erroneously admitted evidence was merely cumulative of other
untainted evidence, or (3) the properly admitted and uncontradicted
evidence of guilt was so overwhelming and the prejudicial effect of the error
was so insignificant by comparison that the error could not have contributed
to the guilty verdict.
Commonwealth v. Petroll, 738 A.2d 993, 1005 (Pa. 1999) (emphasis added).
[J-93-2019] [MO: Saylor, C.J.] - 11
is uncontradicted a fair minded juror may well choose to credit the
defendant’s, rather than the Commonwealth’s evidence.
The principle is in accord with the proper function of an appellate court. An
appellate court is ill equipped to resolve conflicts in the evidence or make
findings of fact.
Story, 383 A.2d at 167-68 (citations modified). Although the Hitcho Court quoted
precedent that identified the correct standard, it nonetheless ignored this important
element, and instead expressly rested its “overwhelming evidence” assessment, in part,
upon an appellate credibility finding. The Court concluded that the defendant’s testimony,
which “offered an explanation for his actions to the jury” and, thus, ostensibly contradicted
at least portions of the Commonwealth’s evidence or the inferences derived therefrom,
was “fraught with inconsistencies and based on an unbelievable version of events.”
Hitcho, 123 A.3d at 748. In other words, this Court deemed a (potential) error harmless,
sua sponte, based at least in part upon its own post hoc assessment of the defendant’s
credibility, which the Court somehow gleaned on appeal from a cold record. Such
disregard for established standards sets a poor example for our courts, and does not
inspire confidence in the soundness of this line of precedent.
The Mitchell line of cases never overruled our landmark decision in Story, never
disapproved of its progeny, and never disputed the longstanding and well-settled principle
that the burden to prove harmlessness rests with the prosecution. See supra at 2 & n.1.
The cases also never recognized the contradiction—they merely cited the right-for-any-
reason doctrine, then proceeded sua sponte as though the tension were resolved. But
the tension remains. I addressed it directly in dissent in Commonwealth v. Hicks, 156
A.3d 1114, 1158 n.1 (Pa. 2017) (Wecht, J., dissenting), and the Superior Court in this
case quoted my concerns before proceeding sua sponte to address harmlessness. See
Maj. Op. at 2; Commonwealth v. Hamlett, 1172 WDA 2016, slip op. at 28 n.8, 2018 WL
[J-93-2019] [MO: Saylor, C.J.] - 12
4327391, at *13 n.8 (Pa. Super. Sept. 11, 2018).7 Hamlett identified the need for this
Court to confront the matter, brought us this important appeal, and has directly framed
this significant problem in our law with a challenge to the conflicting precedent. And so,
here we are.
II.
The Majority does not concern itself with the soundness of the Mitchell line of
cases, see Maj. Op. at 4 n.3, nor does it address the interests of the accused that are
implicated by harmless error review. Although it rejects various of Hamlett’s arguments,
in a manner that I address seriatim below, the Majority articulates one central conclusion
to explain why sua sponte review for harmless error is permissible, notwithstanding the
Commonwealth’s failure to make any attempt to carry its burden of persuasion: “[t]he
harmless-error doctrine functions as the underlying substantive principle of law, and the
right-for-any-reason precept merely provides the explanation for when and why an
appellate court may exercise its discretionary prerogative to proceed of its own accord to
preserve a valid verdict in appropriate circumstances.” Maj. Op. at 9.
This is not a satisfactory answer. Whether the verdict to be preserved is “valid” is
precisely the question that a harmless error analysis seeks to answer, and the answer
7 Although my dissent in Hicks expressed skepticism about this facet of our
harmless error jurisprudence, I felt bound there to “adhere to our precedential declaration”
that sua sponte invocation of harmless error is permissible under the Mitchell line of
cases. Hicks, 156 A.3d at 1158 n.1 (Wecht, J., dissenting) (citing Allshouse, 36 A.3d at
182). Upon review of the authorities and Hamlett’s focused, well-developed advocacy, I
am compelled now to take a more exacting view. We should disavow the flawed
precedent that generated the present controversy.
As the Majority notes, Justice Baer’s concurrence in Hicks also recognized our
precedential conflict, but resolved it in favor of sua sponte invocation of harmless error
via the right-for-any-reason doctrine. Maj. Op. at 3; Hicks, 156 A.3d at 1140 (Baer, J.,
concurring). Because this rationale is duplicative of that of the Mitchell line of cases, and
reflects the same reasoning that the Majority adopts today, I will not separately address
Justice Baer’s comments in Hicks.
[J-93-2019] [MO: Saylor, C.J.] - 13
cannot be known before the inquiry is conducted. Sua sponte application of the doctrine
disregards a crucial component of the “underlying substantive principle of law.” It is the
prosecution’s burden to establish harmless error. The court undertaking the analysis sua
sponte is not so much applying the substantive principle as it is applying part of that
principle, and dismissing the inconsistent part. The Majority characterizes the challenged
practice as a straightforward, sua sponte application of the right-for-any-reason doctrine,
but it is more accurately characterized elsewhere by the Majority as an “exception” to the
“ordinary rule,” id., that the Commonwealth bears a burden on the question of whether a
given error was harmless, or instead whether legal error, once identified, warrants relief.
With no discernible standard to constrain the exception, I fail to understand what is left of
the “ordinary rule.”
Additionally absent from the Majority’s analysis is any recognition of the oft-
repeated reasons that sua sponte decision-making is roundly disfavored in the law. Even
setting aside the express burden applicable in the harmless error context, it is axiomatic
that “[s]ua sponte consideration of issues deprives counsel of the opportunity to brief and
argue the issues and the court of the benefit of counsel’s advocacy.” Wiegand v.
Wiegand, 337 A.2d 256, 257 (Pa. 1975). Raising and addressing issues sua sponte
“disturbs the process of orderly judicial decision-making by depriving the court of the
benefit of counsel's advocacy, and depriving the litigants the opportunity to brief and
argue the issues.” Johnson v. Lansdale Borough, 146 A.3d 696, 709 (Pa. 2016) (citations
omitted). But beyond mere prudential challenges, and of particular relevance given
Hamlett’s specific argument to this Court, sua sponte decision-making may implicate due
process concerns, to the extent that it deprives litigants of notice and an opportunity to be
heard on potentially dispositive matters. See infra Part II(D). Before delving further into
broader concerns, however, I turn next to the Majority’s various rationales for rejecting
[J-93-2019] [MO: Saylor, C.J.] - 14
several of Hamlett’s specific arguments, and the reasons for my disagreements with these
rationales.
A. Judicial Economy
The Majority first invokes principles of judicial economy, noting the “systemic
interest in avoiding costly and unnecessary proceedings before the judiciary” and the
“social costs of retrial, including those of the judicial system at large, jurors, victims, other
witnesses, and the general public.” Maj. Op. at 10. Due to these interests, the Majority
asserts, “[j]udicious recourse to the discretionary prerogative to review for harmless error
may appropriately proceed” sua sponte, when the court deems it appropriate. Id.
First, the Majority’s justification begs the question inasmuch as it relies upon the
interest in avoiding “unnecessary” retrials, which again is precisely the question that a
harmless error analysis seeks to answer. Through the harmless error doctrine, the parties
litigate, and the court decides, whether it is a “valid verdict,” id. at 9, that the prosecution
seeks to preserve despite an error, such that a retrial becomes “unnecessary.” Id. at 10.
The answer to the question is meant to be provided through the parties’ advocacy—the
Commonwealth’s, specifically—not through judicial intuition.
Moreover, the Majority refers only to interests that underlie the harmless error
doctrine generally, but which have little relation to the question of the Commonwealth’s
burden. As it concerns the question before us, the typical framing of the interest in judicial
economy is largely beside the point. It is no secret that the harmless error doctrine, at
bottom, is informed significantly by principles of judicial economy. But that interest is in
no way incompatible with the Commonwealth’s advocacy. Indeed, requiring the
Commonwealth to carry its burden advances judicial economy in many ways, while sua
sponte consideration of harmless error directly and palpably disserves the interest in
judicial economy. Precisely because it is a highly record-intensive inquiry, harmless error
[J-93-2019] [MO: Saylor, C.J.] - 15
analysis is a labor. If it is to be conducted with due rigor, the burden to establish the claim
must fall somewhere. As I explain below, in allowing the Commonwealth’s burden to be
relieved in the appellate court’s discretion, the Majority spares none of the resources of
our trial courts, but allows a heavy workload to fall upon our already-busy appellate courts.
Setting aside precedent, fairness, important underlying interests, and institutional
concerns, perhaps it is helpful to discuss the reasons that the Commonwealth is better
suited than the appellate court to advocate for harmless error, as a purely practical matter.
Although appellate courts regularly review evidentiary records, we necessarily maintain a
relatively detached relationship with them. When we first open the papers filed in a given
appeal, we have no prior knowledge of the case. We were not in the courtroom at trial,
nor have we observed any witness’ demeanor, nor have we seen the parties’ pre-trial or
post-trial motions, nor have we heard their objections or offers of proof. We and the
record simply are not acquainted. We rely heavily upon the lower courts’ summaries of
the facts and proceedings, and to a lesser extent the parties’ representations, in order to
orient ourselves to the legal questions presented for review. Naturally, those questions
generally require us to consult the record, but we tailor our review to the issues before
us, and we require that the parties provide us with citations to the portions of the record
that contain the pertinent material. See, e.g., Pa.R.A.P. 2119(c) (“If reference is made to
the pleadings, evidence, charge, opinion or order, or any other matter appearing in the
record, the argument must set forth, in immediate connection therewith, or in a footnote
thereto, a reference to the place in the record where the matter referred to appears”). We
require such citation precisely because it is both onerous and inefficient for an appellate
court to embark upon a self-guided tour through an unfamiliar fact record.8
8 Cf. United States v. Hasting, 461 U.S. 499, 516-17 (1983) (Stevens, J., concurring
in the judgment) (“This Court is far too busy to be spending countless hours reviewing
trial transcripts in an effort to determine the likelihood that an error may have affected a
[J-93-2019] [MO: Saylor, C.J.] - 16
Unlike the appellate court, the parties are intimately familiar with the fact record.
They built it. They know how its pieces fit together. The parties were in the courtroom.
They filed the motions. They presented the evidence and questioned the witnesses.
They argued the case to the fact-finder. They studied the materials in preparation for
appeal. It is they who have command of the record, and it is they who can direct the
reviewing court to the portions thereof that support or undercut their respective positions.
Accordingly, it is the Commonwealth, not an appellate court reviewing a cold record
months, years, or even decades later, that is in the best position to establish the
harmlessness of an error. The Commonwealth is well-equipped to identify and draw the
court’s attention to, for instance, the remainder of the uncontradicted evidence asserted
to be overwhelming, or the properly introduced evidence of which the erroneous evidence
was merely cumulative. See supra n.6.
In the absence of any argument from the Commonwealth, the appellate court is
left in unfamiliar terrain without a guide. In order to render its judgment with any degree
of confidence, the court seeking to establish harmless error on its own initiative must
commence an exacting, systematic, and laborious review of the record with no direction
from the parties. This is plainly a burdensome exercise for any appellate court. See, e.g.,
United States v. Giovannetti, 928 F.2d 225, 226 (7th Cir. 1991) (per curiam) (noting that
sua sponte consideration of harmless error “would place a heavy burden on the reviewing
court, deprived as it would be of the guidance of the parties on the question whether
particular errors were harmless”). This is not a revolutionary observation.
jury’s deliberations. . . . I have spent several hours reviewing the one copy of the trial
transcript that has been filed with the Court. But I have not read all of its 1,013 pages,
and I have read only a few of the 450 pages of the transcript of the suppression hearing.
The task of organizing and digesting the testimony is a formidable one.”).
[J-93-2019] [MO: Saylor, C.J.] - 17
Certainly, in both the right-for-any-reason and harmless error contexts, the interest
in judicial economy often refers to the avoidance of unnecessary retrials—surely a
considerable burden upon our trial courts. See Maj. Op. at 10. But, from the trial court’s
perspective, it does not matter at all whether it is the Commonwealth or the appellate
court that makes the harmlessness showing. If an error is to be deemed harmless, the
end result—affirmance—is the same, and the resources of the trial court are spared to
precisely the same degree regardless of whether the Commonwealth satisfies its burden
or the appellate court shoulders that burden itself. The difference is that, where the
parties litigate the matter and guide the court through the fact record, the analysis can be
completed much more quickly, efficiently, and with greater confidence in the outcome. By
contrast, where the Commonwealth fails to carry its burden, and the appellate court
undertakes a self-directed harmless error analysis, the court must enmire itself in the
record, parse it for any and all potentially relevant details, and thereby expend a significant
amount of the court’s time and resources, at the expense of other cases on its docket.
There can be no serious argument that it is not at least preferable for the
Commonwealth to provide the reviewing court with meaningful advocacy on a potentially
challenging, highly record-intensive inquiry such as harmless error. It is for this reason
that the crux of Hamlett’s argument regarding judicial economy is that, although “the
practice of sua sponte review for harmless error might hypothetically advance judicial
economy in the trial courts, it does so at the expense of judicial economy in the appellate
courts.” Brief for Hamlett at 38. This is plainly true, but with one caveat. Condoning the
practice of sua sponte review advances judicial economy in the trial court only if we
assume that the Commonwealth, if required, would regularly fail to perform its duty as an
advocate, and that it somehow becomes the appellate court’s role to do the
Commonwealth’s job as an alternative to the consequence that should rightly follow from
[J-93-2019] [MO: Saylor, C.J.] - 18
the Commonwealth’s dereliction. The procedure by which a judgment is affirmed on
appeal is of no particular concern to the trial court.
The Majority’s focus upon judicial economy extends only to the harmless error
doctrine’s standard, general admonitions relating to the costs of retrial. None of that is
inconsistent with the placement of a burden upon the Commonwealth, nor with requiring
the Commonwealth to carry that burden. The Majority criticizes a number of Hamlett’s
arguments as “impugning harmless-error review as such, as distinguished from the sua
sponte aspect.” Maj. Op. at 11 n.9. Yet, a number of the Majority’s “primary lines of
attack,” id., also rely upon principles applicable to harmless error review as such, as
distinguished from the sua sponte aspect. Avoidance of the “societal costs” of
unnecessary retrials is wholly compatible with the Commonwealth advocating for its
position. The sua sponte component of this jurisprudence has never concerned itself with
the fact that the appellate court is doing the actual work of conducting self-guided
harmless error analyses, which speaks volumes about the law’s concern for judicial
economy. By contrast, if the Commonwealth advocates for its position, every step of the
appellate process becomes more efficient.
The most reasonable solution to these problems, from a judicial economy
perspective, is that the Commonwealth must be made to carry its burden. If we begin to
enforce the burden, the Commonwealth will learn to carry it again. It was once
“inexplicable” for the Commonwealth to so fail. Mitchell, 839 A.2d at 215 n.11. Opting
instead to find harmlessness sua sponte changes nothing in the trial courts, but shifts a
heavy workload onto our busy appellate courts. It saves no precious judicial resources.
It expends them. The entire exercise spares the effort of one party alone—the
Commonwealth. Sua sponte harmless error review is not judicial economy. It is
prosecutorial economy.
[J-93-2019] [MO: Saylor, C.J.] - 19
B. Right-For-Any-Reason
I would prefer not to dwell at great length upon the right-for-any-reason doctrine,
nor upon its conflation with the harmless error doctrine, because I think it plain that the
two doctrines are indeed distinct, that they have distinct purposes, and that there is a
reason that they were given distinct names in the first place. I also believe that the sua
sponte practice that the Majority approves is not actually an application of the right-for-
any-reason doctrine, but rather that it is simply an attempt to apply harmless error
principles on the court’s own initiative, with the Commonwealth’s burden conspicuously
absent from the analysis. But the ostensible connection between these doctrines is the
jurisprudential anchor of the Majority’s approach, so it is worth considering the ways in
which harmless error and the right-for-any-reason doctrine converge and diverge.
To separate the right-for-any-reason doctrine from harmless error, Hamlett
proposes a taxonomy to this Court, suggesting that trial error be classified as either error
of “admission,” “use,” or “rationale.” Brief for Hamlett at 17-23. He argues that only errors
in rationale are amenable to disposition under the right-for-any-reason doctrine, because,
in such a circumstance, the admission or use of the evidence was not erroneous; the only
deficiency was the trial court’s reason for a correct action. But if the admission or use of
the evidence had no valid justification at all, then it is truly erroneous, and the error may
find no safe harbor under the right-for-any-reason doctrine.
The Majority rejects Hamlett’s framework, opining that “the selection of an incorrect
basis for admitting evidence can be viewed as employment of an erroneous rationale,”
and, thus, that Hamlett’s categories are “not mutually exclusive.” Maj. Op. at 11.
Rejecting Hamlett’s lexicon, however, does not result inexorably in a functional
equivalence between the harmless error and right-for-any-reason doctrines. I find that
Hamlett’s classification scheme presents a useful exercise, but it merely places a label
[J-93-2019] [MO: Saylor, C.J.] - 20
upon elementary right-for-any-reason principles. It is not necessary to the analysis,
because differences between the doctrines abound.
Of course, an immediate and dispositive distinction is that the harmless error
doctrine always has imposed a burden upon the prosecution, while the right-for-any-
reason doctrine places no parallel burden upon the party seeking its benefit. But that is
not even the most obvious difference. As evidenced by their very names, the doctrines
pose different questions to the court. When an appellate court applies the right-for-any-
reason doctrine, it ultimately finds no reversible error, because the challenged decision
was “right,” for a different reason than was articulated below. The right-for-any-reason
doctrine asks the question: “Was the challenged action erroneous?” Harmless error, by
contrast, is a question of effect—once an error has been found, harmless error doctrine
asks the question: “Did the error change the outcome?”
The right-for-any-reason doctrine is best conceptualized as allowing for the
identification of an alternative, valid, legal justification for a challenged action. “We have
often stated that where a court makes a correct ruling, order, decision, judgment or
decree, but assigns an erroneous reason for its action, an appellate court will affirm the
action of the court below and assign the proper reason therefor.” Bearoff, 327 A.2d at
76 (emphasis added; capitalization modified). It is self-evident that an “error” is not a
“correct ruling” suitable for disposition under the right-for-any-reason doctrine. Rather,
an error is an error, suitable for analysis under the harmless error doctrine.
An archetypal example of the right-for-any-reason precept occurs when the trial
court offers a flawed rationale for its decision to admit a challenged piece of evidence, but
the facts nonetheless support the admission of that evidence for a different reason or
pursuant to a different rule. To perform the right-for-any-reason analysis in such a case,
the appellate court merely must assess whether the established evidence of record
[J-93-2019] [MO: Saylor, C.J.] - 21
satisfies an existing legal rule—a familiar task for appellate courts. If the facts fit the rule,
then we may conclude that the challenged action was proper, not erroneous. See, e.g.,
In re A.J.R.-H., 188 A.3d 1157, 1176 (Pa. 2018) (“The [right-for-any-reason] doctrine thus
may be applied by a reviewing court if the established facts support a legal conclusion
producing the same outcome.”). If such is an error, it is an error only in form, not in
substance. See Thomas v. Mann, 28 Pa. 520, 522 (Pa. 1857) (“The only error upon the
record is a wrong reason for a right judgment; but, as we review not reasons but
judgments, we find nothing here to correct.”). With harmless error, by contrast, there is
something to correct. There is a legal error that may have impacted the outcome.
The right-for-any-reason doctrine does not require the appellate court to consider
the effect of an error on the totality of the proceedings. The analysis does not require a
speculative inquiry into the likely impact upon a juror of hearing erroneously admitted
evidence, or an examination of whether the evidence may have strengthened certain
arguments in the eyes of a juror or called others into question, or—ultimately—a
determination that the jury still would have returned a guilty verdict absent the evidence.
These latter inquiries are hallmarks of the challenges underlying harmless error. Where
the right-for-any-reason doctrine is concerned with legal justifications, harmless error is
essentially more of a factual judgment than a legal one—whether, under the facts and
circumstances presented, beyond a reasonable doubt, a jury would have returned the
same verdict regardless of the error.
Answering the harmlessness question, moreover, poses intractable difficulties
stemming from the inherent limitations of appellate review. Quoting California Chief
Justice Roger Traynor’s famous treatise on harmless error, this Court in Story observed:
The appellate court is limited to the mute record made below. Many factors
may affect the probative value of testimony, such as age . . . intelligence,
experience, occupation, demeanor, or temperament of the witness. A trial
court or jury before whom witnesses appear is at least in a position to take
[J-93-2019] [MO: Saylor, C.J.] - 22
note of such factors. An appellate court has no way of doing so. It cannot
know whether a witness answered some questions forthrightly but evaded
others. It may find an answer convincing and truthful in written form that
may have sounded unreliable at the time it was given. A wellphrased
sentence in the record may have seemed rehearsed at trial. A clumsy
sentence in the record may not convey the ring of truth that attended it when
the witness groped his way to its articulation. What clues are there in cold
print to indicate where the truth lies? What clues are there to indicate where
the half-truth lies?
Story, 383 A.2d at 168 (quoting ROGER J. TRAYNOR, THE RIDDLE OF HARMLESS ERROR 20-
21 (1970)).
This Court has developed certain guidelines in order to make our journey into these
unfamiliar waters more comfortable—such as requiring that harmlessness be predicated
upon uncontradicted evidence so as to avoid the morass of appellate credibility
determinations. See Story, 383 A.2d at 166-68. This Court also has overlooked those
guidelines in our sua sponte pursuit of the doctrine. See Hitcho, 123 A.3d at 748. The
truth of the matter is that harmless error analysis remains a challenging task in a
significant number of cases. The endeavor is unavoidably speculative. The doctrine asks
appellate judges to evaluate and assess the persuasiveness of the entirety of an
evidentiary record, to run a hypothetical trial in our minds sanitized of the legal error before
us, while setting aside our own suspicions as to the guilt or innocence of the accused
(which may indeed be influenced by the error at issue), and to opine conclusively as to
how a hypothetical jury would have viewed a hypothetical body of evidence. 9 This is
9 Although such analyses pose unique difficulties on appeal, it would be an
overstatement, of course, to suggest that appellate courts are wholly incapable of
rendering judgments about the potential impact of certain evidence upon the fact-finder,
or the ultimate likelihood that a different result would have obtained. Indeed, in modern
post-Chapman jurisprudence, harmless-error-type inquiries have been engrafted upon a
number of substantive legal rules, such as the standard for demonstrating prejudice in
the ineffectiveness of counsel context, and the test for materiality under Brady v.
Maryland, 373 U.S. 83 (1963). See Edwards, supra n.3, at 1178 (conceptualizing the
development of these standards as an outgrowth of the harmless error rule). For better
[J-93-2019] [MO: Saylor, C.J.] - 23
palpably dissimilar from many of the purely legal considerations that might be resolved
easily and fairly under the right-for-any-reason doctrine. Fortunately, although the
harmlessness inquiry can be challenging, the court need not go it alone. The court can
be aided greatly by the parties’ advocacy, see supra Part II(A), and, in particular, by an
effort from the Commonwealth, as the beneficiary of the error, to demonstrate that the
error could not have contributed to the verdict.
The Majority may dismiss Hamlett, or me, as “impugning harmless-error review as
such, as distinguished from the sua sponte aspect.” Maj. Op. at 11 n.9. But by these
observations, I mean to reveal disparities between the doctrines before us, to the extent
that those disparities tend to undercut the Majority’s thinly supported assertion that one
doctrine may be viewed as a mere application of the other.10 Given the specific challenge
or worse, this has become a familiar exercise. A critical difference in the Brady and
ineffectiveness areas, however, is that the burden to make the showing is placed upon
the appellant instead of the prosecution, and this Court, at least, strictly requires the
appellant to carry that burden, and will not hesitate to deem the appellant’s claim waived
if it lacks sufficient development. I further address concerns regarding issue-preservation,
development, and waiver below. See infra at 25-27; 38-40.
10 Although it favorably cites a Georgia decision, the Majority acknowledges that its
approach—collapsing the harmless error rule into the right-for-any-reason doctrine and
approving limitless sua sponte invocation of either—is not common among American
jurisdictions. Maj. Op. at 9 n.6. The Majority deems all other approaches to sua sponte
consideration of harmless error to be simply a “close cousin” of its own approach, but it
does not elaborate upon this assertion. I question the degree of relation.
For its recourse to Georgia law, the Majority invokes Jones v. State, 802 S.E.2d
234, 237 (Ga. 2017) (admission of prior bad acts evidence “was harmless as to
appellant’s conviction and sentence . . . and so the Court of Appeals’ judgment is affirmed
as right for any reason”); see Maj. Op. at 9. However, the same tension that the Mitchell
footnote has engendered in Pennsylvania appears to afflict Georgia’s jurisprudence in
precisely the same way. The Jones Court did not refer to the harmless error rule as
imposing a “burden” at all, despite the Georgia Supreme Court’s recognition of that
burden with respect to evidentiary errors like that found in Jones, even in cases decided
in the same year as Jones. See Bozzie v. State, 808 S.E.2d 671, 677 (Ga. 2017) (“For
nonconstitutional harmless error, the State has the burden to show that it was highly
probable that the error did not contribute to the verdict.”). Although the Majority
[J-93-2019] [MO: Saylor, C.J.] - 24
before this Court—which reveals significant difficulties with the sua sponte aspect of the
challenged jurisprudence on multiple fronts, and which suggests a simple solution—the
remainder of one’s opinions about the balance of the harmless error rule would seem to
be of little consequence.
Perhaps some additional insight into the Majority’s reasoning may be gleaned from
its summary of the Commonwealth’s arguments, inasmuch as the Majority “agree[s] with
the Commonwealth in all material respects.” Maj. Op. at 8. Accordingly, one may
presume that the Majority endorses at least all of the Commonwealth’s arguments that it
summarizes in its Opinion. Concerning the absence of the Commonwealth’s burden
when the court pursues harmless error sua sponte under the right-for-any-reason
doctrine, the Majority favorably quotes the Commonwealth’s reliance upon a “core
precept underlying the right-for-any-reason doctrine, which is that only appellants are
charged with issue preservation obligations, whereas appellees bear none.” Id. (citing
Brief for Commonwealth at 42) (emphasis in original). Hamlett’s position, the Majority
and the Commonwealth fret, might “upend this conventional approach to issue
preservation by imposing waiver, or its functional equivalent, on the government as
appellee.” Id. (citing Brief for Commonwealth at 7, 42).
First of all, it is curious that the Majority would deem “issue preservation
obligations” to be a “core precept” of the right-for-any-reason doctrine, for it apparently
does not view the prosecution’s burden of persuasion to be a “core precept” of the
harmless error doctrine. The prosecution’s burden is indeed a “core precept” of harmless
error; in fact, it is the “second general precept” underlying the doctrine. Davis, 305 A.2d
at 719; see supra at 7; see also supra n.1. When harmless error “functions as the
establishes that we are not alone in rendering incompatible pronouncements on harmless
error, nothing in Jones helps to reconcile the incongruity between unfettered sua sponte
review and the recognition of a burden imposed upon a party.
[J-93-2019] [MO: Saylor, C.J.] - 25
underlying substantive principle of law,” Maj. Op. at 9, to which the court resorts through
its discretionary prerogative to employ the “right-for-any-reason precept,” id., what
happens to the general precepts of the harmless error doctrine? It seems that a core
precept of the harmless error doctrine is jettisoned, yet what the Commonwealth simply
insists is a “core precept” of the right-for-any-reason doctrine mysteriously remains.
In any event, the Commonwealth’s focus upon the absence of an issue-
preservation obligation upon appellees is a non sequitur. Maj. Op. at 8; Brief for
Commonwealth at 42, 46. We have never suggested that the Commonwealth’s burden
is one of issue-preservation; the Commonwealth need not raise the issue before the trial
court in the first instance on pain of waiver. Its burden is one of persuasion, and may be
satisfied by offering argument to the reviewing court for the first time on appeal. The
absence of an issue-preservation burden upon appellees, thus, is irrelevant. The point
would be roughly equivalent to suggesting that a court may research, litigate, and rule
upon a claim that the appellant fails to assert in its brief, simply because the appellant
preserved an objection and included the issue in its Pa.R.A.P. 1925(b) statement. Issue-
preservation is beside the point.
We speak of a burden of persuasion. Even if one assumes that it would be absurd
or unreasonable to impose an issue-preservation burden upon the Commonwealth as to
the harmlessness question, the suggestion of the Commonwealth’s “waiver” would not
follow from failure to preserve an issue. A finding of waiver is not a ludicrous response
to an observation that a litigant has failed to develop advocacy on a question as to which
it bears the burden. Even guiding federal jurisprudence, to which the Majority adverts but
does not discuss, recognizes the government’s failure to advocate as precisely this sort
of “waiver.” Giovannetti, 928 F.2d at 226; see infra at 38-40; but see Maj. Op. at 10 n.7.
The point that the Majority appears not to appreciate is that we will enforce an appellant’s
[J-93-2019] [MO: Saylor, C.J.] - 26
“waiver” for failure to develop a claim upon which the appellant bears the burden of
persuasion, but we are appreciably hesitant to impose any consequence whatsoever
upon the Commonwealth’s identical failure, or even to acknowledge it as such. We have
hesitated so long that a “tension” with fundamental principles has arisen in our
jurisprudence, which we now have been asked to confront and to resolve.
Commonwealth v. Hamlett, 202 A.3d 45 (Pa. 2019) (per curiam).
After its diversion into issue-preservation concerns, the Majority next favorably
quotes a set of excuses that the Commonwealth provides as to why it is onerous for it to
carry its burden of persuasion—a point that strikes me less as concerning the right-for-
any-reason doctrine and more as an argument in favor of overruling Story, but I digress.
The Majority summarizes the Commonwealth’s assertion that “there are many reasons
why advocates representing appellees may refrain from presenting certain alternative
arguments, such as out of a concern that such references might dilute the strength of
primary contentions, or on account of word-count limitations on briefing.” Id. (citing Brief
for Commonwealth at 43, 50). The Commonwealth, and by extension the Majority, seems
to question whether our courts are capable of effectively weighing a legal argument that
is followed by an alternative analysis. I am not so skeptical of our appellate jurists’
abilities. Able lawyers themselves, I am confident that they can identify and recognize
the purpose of an alternative argument. There is no reason to believe that courts will
suddenly begin to overlook meritorious legal arguments because the Commonwealth also
argues, in the alternative, that the asserted error was harmless. As for word-count
limitations, we do not allow such requirements to serve as an excuse for any other litigant
to fail to advocate for his or her position, and they manage to do so even in highly complex
cases. And, of course, if the Commonwealth fears that presenting arguments to the court
will cause it to expend the 14,000 words allotted to it, it remains free to seek the court’s
[J-93-2019] [MO: Saylor, C.J.] - 27
leave for an exception.11 I am further certain that our procedural rules could respond to
any systemic concerns that such advocacy may present. Pointing to word-count
limitations is little more persuasive than would be an assertion that we must have due
regard for circumstances in which the Commonwealth’s dog eats its homework.
The right-for-any-reason and harmless error doctrines may ultimately serve similar
ends—affirmance—but they are not the same doctrine. In the Majority’s conflation of
them, the obligation of the Commonwealth to advocate for its position has gone missing.
The Majority asserts that what it describes is an invocation of the “right-for-any-reason
precept” to apply the harmless error doctrine as the “underlying substantive principle of
law,” Maj. Op. at 9, but I fail to understand why the “second general precept” of the
harmless error doctrine, Davis, 305 A.2d at 719, simply disappears from the analysis.
Because core precepts of these doctrines stand in clear tension with one another,
perhaps the analysis is not as straightforward as the Majority suggests.
C. Judicial Impartiality
Hamlett further contends that sua sponte harmlessness findings are in tension with
the fundamental obligation of the judiciary to remain neutral and impartial. He reminds
us that the adversarial process, essential to the American model of criminal justice, has
two central components: “(1) neutral and passive decision makers and (2) party
presentation of evidence and arguments.” Brief for Hamlett at 38 (quoting Adam A. Milani
& Michael R. Smith, Playing God: A Critical Look at Sua Sponte Decisions by Appellate
Courts, 69 TENN. L. REV. 245, 272 (2002)). Hamlett argues that the practice of raising
and deciding issues sua sponte, harmless error in particular, stands in tension with the
well-understood role of appellate courts, articulately described by then-Judge Antonin
11 Pa.R.A.P. 2135(a)(1) (“Unless otherwise ordered by an appellate court . . . a
principal brief shall not exceed 14,000 words . . . .”).
[J-93-2019] [MO: Saylor, C.J.] - 28
Scalia as bodies that “do not sit as self-directed boards of legal inquiry and research, but
essentially as arbiters of legal questions presented and argued by the parties before
them.” Id. (quoting Carducci v. Regan, 714 F.2d 171, 177 (D.C. Cir. 1983)).12 “Put
simply,” Hamlett summarizes, “where the Commonwealth fails to take an adversarial
position on the matter of harmless error, it is not up to the appellate courts to pick up the
slack.” Id. at 40. The now-prevailing approach, Hamlett observes, “confers upon
reviewing courts essentially unfettered discretion” in deciding whether to reverse or to
pursue harmless error. Id. at 42. This, in turn, “imperils our jurisprudential ideals of
evenhandedness, predictability, consistency in the development of legal principles,
reliability on judicial decisions, and the actual and perceived integrity of the judicial
process.” Id. at 41.
The Majority rejects Hamlett’s argument without substantial discussion, adverting
again to principles of judicial economy and the “larger concerns” of the “social costs of
retrial.” Maj. Op. at 10. I believe that those latter interests are largely the responsibility
of the Commonwealth to advance, if it so desires. I also find the perceived neutrality of
the judiciary to be an exceptionally weighty concern. Hamlett’s observations ring true in
many respects. Although there are other issues that a court may raise sua sponte
pursuant to longstanding precedent, I can conjure no other circumstance where a party
has an express and established burden to prove a particular assertion, yet the court may
excuse that burden, take it upon itself, and declare victory for the party who has done
nothing to claim it. Earlier in this opinion, I characterized the finding of harmlessness as
a “knockout blow,” inasmuch as it adversely disposes of the appellant’s claim of error.
12 See also McNeil v. Wisconsin, 501 U.S. 171,181 n.2 (1991) (“What makes a
system adversarial rather than inquisitorial is . . . the presence of a judge who does not
(as an inquisitor does) conduct the factual and legal investigation himself, but instead
decides on the basis of facts and arguments pro and con adduced by the parties.”).
[J-93-2019] [MO: Saylor, C.J.] - 29
Supra at 5. But when the court finds harmlessness without being asked or without the
matter being litigated, that final punch comes not from one of the boxers; it is thrown by
the referee. The appellant has proven the existence of an error—the Commonwealth is
on the ropes. Yet, in comes the referee with the harmlessness haymaker, and the
appellant is down for the count. Certainly, no one watching that bout would characterize
the referee as a passive and neutral arbiter.
At the very least, the continued absence of guidance from this Court as to when
the harmlessness question may or should be raised sua sponte—a problem that persists
through today’s decision—allows for inconsistent application and potentially arbitrary
results. As it stands now, different panels of our Superior Court may take wholly different
approaches. One panel finding an error may observe that the words “harmless error”
appear nowhere in the Commonwealth’s brief, and simply reverse. Another panel may
take that same appeal and same error, and proceed to a harmlessness inquiry sua
sponte. Further, without guidance or argument from the parties, panels undertaking the
inquiry sua sponte may emphasize different aspects of the case, or find something in the
record that others might overlook on their self-guided tour, leading panels to reach very
different conclusions on the question. The fault for inconsistency in the case law,
however, lies not with the Superior Court. The Commonwealth is not to blame either,
even though the routine absence of its advocacy is the source of many of the identified
ills. The fault lies with this Court, which gradually has eroded the Commonwealth’s
burden, but has left our lower courts without guidelines for sua sponte application of the
doctrine.
Concern for evenhandedness would appear to be particularly acute in this context,
inasmuch as a finding of harmlessness always redounds to the benefit of the prosecution,
never to the defense. Perhaps this is why a portion of the Commonwealth’s argument
[J-93-2019] [MO: Saylor, C.J.] - 30
endorsed by the Majority can glibly suggest that, “[s]urely, the appellate courts of this
Commonwealth do not take lightly their duty to make fair and just decisions when the lives
and liberty of criminal defendants are at stake.” Maj. Op. at 8 (quoting Brief for
Commonwealth at 27). Naturally the Commonwealth is not particularly concerned; it
stands to lose very little from deficient findings of harmless error. I believe that Hamlett
has identified, and I have discussed, cases in which even this Court has given questions
of harmless error less than the full consideration that they may have deserved. The fact
that the Commonwealth can provide a list of cases that it deems to contain “good” uses
of sua sponte harmless error review, id. (citing Brief for Commonwealth at 27 n.2), does
not remove the potential for arbitrariness and abuse that inheres in the practice. It is
facially suspect that a court is empowered to intervene on behalf of, and to assume the
role of an advocate for, one party but not the other. This may be regarded as a question
of appearance, regardless of whether the resolution of any particular case is driven by
unseemly motivations. Cf. In Interest of McFall, 617 A.2d 707, 711 (Pa. 1992) (“We hold
herein that the impartiality of the court, which is a fundamental prerequisite of a fair trial,
must be deemed compromised by appearance alone, thus eliminating the need for
establishing actual prejudice.”). A truly impartial arbiter requires each party to carry its
own burden.
D. Due Process
The Majority rejects Hamlett’s due process argument in one paragraph, with two
indirect responses. Hamlett develops an intriguing claim. He begins with the most
fundamental command of due process: “that deprivation of life, liberty or property by
adjudication be preceded by notice and opportunity for hearing appropriate to the nature
of the case.” Brief for Hamlett at 30 (quoting Mullane v. Cent. Hanover Bank & Trust Co.,
339 U.S. 306, 313 (1950)). The critical right to be heard “has little reality or worth unless
[J-93-2019] [MO: Saylor, C.J.] - 31
one is informed that the matter is pending and can choose for himself whether to appear
or default, acquiesce or contest.” Id. at 31 (quoting Mullane, 339 U.S. at 314). When an
appellate court performs a harmless error analysis sua sponte, after the parties have filed
their briefs, without informing them that the issue will be dispositive of the appeal, and
without providing the appellant with an opportunity to argue against the finding, the
appellate court deprives the appellant of due process. Moreover, such a circumstance
deprives the appellant of the opportunity to be heard by counsel, in further tension with
due process norms. Id. at 33-35. Hamlett reminds us that the “right to be heard would
be, in many cases, of little avail if it did not comprehend the right to be heard by counsel.”
Id. at 34 (quoting Powell v. Alabama, 287 U.S. 45, 68-69 (1932)).
Notably, unlike the United States Constitution, the Pennsylvania Constitution
expressly guarantees the right to appeal.13 Accordingly, in this Commonwealth, a
convicted defendant has an express right to appeal his judgment of sentence, and has a
right to due process, which, at a minimum, contemplates notice and an opportunity to be
heard before the disposition of a matter in which his life or liberty is at stake. At least in
this context,14 merely combining a few fundamental propositions of law reveals that
Hamlett’s suggestion is compelling.
13 See PA. CONST. art. V, § 9 (“There shall be a right of appeal in all cases to a court
of record from a court not of record; and there shall also be a right of appeal from a court
of record or from an administrative agency to a court of record or to an appellate court,
the selection of such court to be as provided by law; and there shall be such other rights
of appeal as may be provided by law.”).
14 The suggestion that sua sponte decision-making may raise due process concerns
is not unheard of in Pennsylvania law, nor in broader legal scholarship. See Fallaro v.
Yeager, 528 A.2d 222, 224 (Pa. Super. 1987) (“Due process requires that the litigants
receive notice of the issues before the court and an opportunity to present their case in
relation to those issues. It is even more e[g]regious an error when the lack of notice,
through variance from the pleadings, is the court’s doing. For when the issue is first stated
only in the court’s resolution of it, the unsuspecting party has no opportunity during the
proceedings to voice his objections or match his case to the altered issue.”) (quoting In
[J-93-2019] [MO: Saylor, C.J.] - 32
The Majority gives two reasons for rejecting Hamlett’s position, neither of which
responds directly to Hamlett’s constitutional grievance. First, the Majority concludes that
sua sponte consideration of harmless error does not deprive an appellant of due process
because, if this were the case, then the federal statute and rule that provide for harmless
error review would be “patently unconstitutional as applied to their authorization of
harmless-error review by the federal appellate courts acting of their own accord.” Maj.
Op. at 10-11. Although this is a worthy matter to ponder, it is ultimately beside the point,
for the constitutionality of federal practice is not the question here, and, in any event, the
right to appeal is not enshrined in the federal Constitution, as it is in ours. Supra n.13.
Regardless, as the Majority seems to acknowledge, recognizing a due process
component of the procedure by which harmlessness is determined would not necessarily
render the federal enactments facially unconstitutional, inasmuch as the Majority refers
to as-applied challenges. Perhaps unbounded discretion to resort to harmless error sua
sponte is also problematic in federal court. See infra Part III. We are not called upon to
so decide; it is Pennsylvania law that is before us today. Because the Majority’s
observation does not substantively respond to Hamlett’s contention regarding the
contravention of due process norms, I am uncertain as to why it should dispose of
anything here.
The Majority provides one additional reason to dispense with Hamlett’s due
process argument. The Majority states: “In cases in which harmlessness may be a
Interest of M.B., 514 A.2d 599, 601 (Pa. Super. 1986), aff’d, 538 A.2d 495 (Pa. 1988));
see generally Barry A. Miller, Sua Sponte Appellate Rulings: When Courts Deprive
Litigants of an Opportunity to Be Heard, 39 S.D. L. REV. 1253 (2002).
Hamlett’s invocation of due process also is echoed by the only amicus curiae
participating in this appeal. See Brief for Amicus Curiae Pennsylvania Association of
Criminal Defense Lawyers at 4-12 (arguing that sua sponte findings of harmless error
deprive the appellant of the right to counsel and the due process right to notice and an
opportunity to be heard).
[J-93-2019] [MO: Saylor, C.J.] - 33
factor . . . able defense attorneys can well anticipate that appellate courts may be
legitimately concerned with the societal costs attending a new trial, and those attorneys
are free to address the prejudicial impact of trial errors in their own appellate briefs.” Maj.
Op. at 11. This point likewise fails to respond directly to Hamlett’s grievance, but it is also
highly problematic for at least three additional reasons.
First, this is nothing more than improper burden-shifting. The Majority seems to
be giving standing instructions to defense attorneys that, if they wish to represent their
clients “ably,” perhaps effectively, they had better argue against harmlessness as a matter
of course, lest the Commonwealth fail to meet its burden and the appellate court choose
to tag itself in. This strikes me as no different than inverting the Commonwealth’s burden
and placing it upon defense counsel. The Majority cautions against the right-for-any-
reason doctrine imposing a “high barrier to reversal,” id. at 11, but it imposes ever more
hurdles to relief. Where once the Commonwealth had a burden to establish harmless
error, now the Majority expects “able” defense counsel to anticipate that the
Commonwealth will fail to advocate for its position, and it instructs defense counsel, on
counsel’s own initiative, to expend additional resources to rebut an argument that has not
yet been made, as to which the Commonwealth nominally has the burden of persuasion
and the defense accordingly should stand in a responsive posture, so as to persuade the
court that the substantive legal error, which defense counsel will have just proven on the
merits, is worthy of relief—all of this depending, of course, upon whether the court
employs its discretion to consider harmless error at all. The Majority instructs defense
attorneys not only to satisfy their own burdens to prove legal error, but also to rebut, in
advance, the argument that the Commonwealth might or might not choose to make, and
that the court might or might not choose to undertake. The Majority not only shuffles the
responsibility for the doctrine over to defense counsel, it asks defense counsel to go first.
[J-93-2019] [MO: Saylor, C.J.] - 34
Without knowing what the Commonwealth would argue, or if it will argue, counsel filing
the principal brief for the appellant can offer only speculative defenses to hypothetical
arguments about harmless error.
Is the Commonwealth not an “able” advocate for its own position?
Second, the Majority’s comment on the duties of able defense counsel arises in
the context of a due process claim regarding the deprivation of notice and an opportunity
to be heard on a critical matter. To my knowledge, due process jurisprudence has never
placed the onus upon the individual subject to the deprivation to anticipate such
deprivation and launch a prophylactic challenge thereto. To the contrary, it is inherent in
the concept of “notice” that the individual is to be provided with notice of adverse action;
he is not expected to divine and preempt it. See, e.g., Mullane, 339 U.S. at 314 (“An
elementary and fundamental requirement of due process in any proceeding which is to
be accorded finality is notice reasonably calculated, under all the circumstances, to
apprise interested parties of the pendency of the action and afford them an opportunity to
present their objections.”) (emphasis added).
Third, the Majority’s suggestion to able defense counsel overlooks our own
precedent. In the Mitchell footnote, itself—the beginning of this line of cases that has
usurped Story—this Court used an able defense attorney’s proactive advocacy as a
substitute for the Commonwealth’s burden, and went so far as to suggest that our
consideration of harmless error therefore was not even an action taken sua sponte. See
Mitchell, 839 A.2d at 215 n.11; supra at 8. We then overlooked the Commonwealth’s
dereliction, and ruled against the appellant. Today’s Majority does not suggest its
disapproval of the Mitchell footnote, and indeed it directly embraces the legal approach
that Mitchell pioneered, or clumsily engrafted upon our case law. Against the backdrop
of the precedent challenged in the very case before us, the Majority’s suggestion that able
[J-93-2019] [MO: Saylor, C.J.] - 35
defense counsel should anticipate and argue against harmless error strikes me as a cruel
sort of irony.
The Majority speaks of what it deems to be the current state of Pennsylvania law,
and lauds its “due and appropriate regard for defendants’ substantive rights.” Maj. Op. at
6-7. I see due and appropriate regard in Story. I do not find it in the Mitchell footnote, or
in Hitcho. I see a credible assertion of a substantive constitutional right in this case, which
the Majority dodges with arguments that do not respond to the deprivation that both
Hamlett and amicus have identified and placed squarely within this Court’s view. I do not
doubt that the law on this question can reflect a better balance between “the promotion
of judicial economy” and the “due and appropriate regard for defendants’ substantive
rights.” Maj. Op. at 7. The law can require the Commonwealth to advocate for the
Commonwealth’s position, can demand that the defense be given notice and an
opportunity to respond, and can tell the court to simply adjudicate rather than attempt to
play judge, prosecutor, and defense counsel all at once.
Harmless error is often the line between a new trial and many years in prison.
Where the stakes are such, it is only fair to provide the interested party with notice that
the issue is under consideration, and with an opportunity to speak on the matter through
counsel. Regardless of whether the deprivation of that notice and opportunity constitutes
a due process violation, it certainly is less than a completely fair practice. The Majority
seems to recognize as much, at least by implication. See Maj. Op. at 12 (suggesting that
a court considering harmlessness sua sponte in “close cases” may order supplemental
briefing in order to “enhance fairness to the defendant”). Every defendant should be
entitled to fairness; fairness should not need to be “enhanced” in any particular case. The
way to encourage evenhanded treatment is to set a consistent standard and then apply
it rigorously. Due process or not, that is simply a matter of good judicial practice.
[J-93-2019] [MO: Saylor, C.J.] - 36
III.
All participants here—the parties, the Majority, the Concurrence, and myself—
have made some references to federal harmless error practice. Both the Majority and the
Concurrence refer to the standard for sua sponte consideration of harmless error outlined
in the Seventh Circuit’s oft-cited, per curiam decision in Giovannetti, 928 F.2d 225, though
they do not discuss its details. See Maj. Op. at 12 n.10; Conc. Op. at 2 (Donohue, J.).
Giovannetti also appears in the advocacy of the parties. Accordingly, it is worth
considering the legal analysis to which the parties, the Majority, and the Concurrence all
refer, alongside a recognition of the ways in which Pennsylvania jurisprudence differs
from its federal counterpart on matters of harmless error.
Federal criminal law provides for harmless error analysis by statute and rule,
neither of which have a corollary in Pennsylvania criminal law. Both the federal statute
and rule direct a reviewing court, in mandatory language, to disregard any error that does
not affect the “substantial rights” of a party. Brief for Hamlett at 26-27 & nn. 17-18 (quoting
Fed.R.Crim.P. 52(a) (“Any error, defect, irregularity, or variance that does not affect
substantial rights must be disregarded.”); 28 U.S.C. § 2111 (“On the hearing of any
appeal . . . the court shall give judgment after an examination of the record without regard
to errors or defects which do not affect the substantial rights of the parties.”)). Hamlett is
quick to point out that Pennsylvania has no such statute, and although similar “substantial
rights” provisions appear elsewhere in our procedural rules, there is no such language in
our Rules of Criminal or Appellate Procedure. Id. at 28. Harmless error is judge-made
law in Pennsylvania.
Also noteworthy, in my view, is that the Supreme Court of the United States has
described the harmless error inquiry under the Federal Rules of Criminal Procedure as
the inverse of the appellant’s burden to prove plain error. See United States v. Olano,
[J-93-2019] [MO: Saylor, C.J.] - 37
507 U.S. 725, 734 (1993) (“When the defendant has made a timely objection to an error
and Rule 52(a) [(harmless error)] applies, a court of appeals normally engages in a
specific analysis of the district court record—a so-called ‘harmless error’ inquiry—to
determine whether the error was prejudicial. Rule 52(b) [(plain error)] normally requires
the same kind of inquiry, with one important difference: It is the defendant rather than the
Government who bears the burden of persuasion with respect to prejudice.”). Unlike the
federal judiciary, this Court does not recognize the plain error doctrine; rather, we strictly
adhere to a paradigm of issue-preservation and waiver. See generally Pa.R.A.P. 302. In
this way, federal jurisprudence reflects a symmetry wholly lacking in Pennsylvania law.
Earlier I quoted the Seventh Circuit’s per curiam decision in Giovannetti for its
observation that conducting a harmless error analysis sua sponte “would place a heavy
burden on the reviewing court.” Giovannetti, 928 F.2d at 226; supra at 17. The court
discussed this labor while rejecting the government’s primary argument: that the
government’s harmless error arguments, themselves, are “nonwaivable.” Id. at 226. The
court rejected the government’s contention that the mandatory language of Fed.R.Crim.P.
52(a)—errors which do “not affect substantial rights shall be disregarded”—meant “that
even if the government does not argue harmless error, we must search the record—
without any help from the parties—to determine that the errors we find are prejudicial,
before we can reverse.” Id. The court was troubled by the burdens that would fall to the
court instead of the government, and cautioned against the “salami tactics” that the
government could use to hedge against having to argue the issue at all. Id. “Such tactics
would be particularly questionable in a case such as this,” the court emphasized, “where
the defendant goes out of his way to argue that the error of which he complains was
prejudicial, and the government by not responding signals its acquiescence that if there
was error, it indeed was prejudicial.” Id. But see Mitchell, 839 A.2d at 215 n.11.
[J-93-2019] [MO: Saylor, C.J.] - 38
This satisfied the Giovannetti court that “harmless-error arguments can be waived.”
Giovannetti, 928 F.2d at 226.15 It was a “separate question” whether the government’s
waiver “always binds the court.” Id. Thus, looking elsewhere in the federal rules, and
finding a general rule of construction in favor of simplicity, fairness, and the elimination of
unjustifiable expense, the Giovannetti court concluded that its powers were discretionary;
that courts “are not required to scour a lengthy record on our own, with no guidance from
the parties,” but that they may deem themselves “authorized, for the sake of protecting
third-party interests including such systemic interests as the avoidance of unnecessary
court delay, to disregard a harmless error even though through some regrettable
oversight harmlessness is not argued to us.” Id.
Despite its deep reservations about sua sponte use of the harmless error doctrine,
and despite its refusal to excuse the government’s waiver in the very case before it, the
Giovannetti court nonetheless developed a multifactorial test pursuant to which a court
may “overlook” the government’s waiver and proceed sua sponte. Id. at 227. The
“controlling considerations,” the court explained, are: (1) “the length and complexity of
the record”; (2) “whether the harmlessness of the error or errors found is certain or
debatable”; and (3) “whether a reversal will result in protracted, costly, and ultimately futile
proceedings in the district court.” Id. Applying that standard, the Giovannetti court
ultimately concluded that the case before it did not warrant sua sponte consideration of
harmless error, and it accordingly “decline[d] to relieve the government from the
consequences of its failure to raise the issue of harmless error in its brief on appeal.” Id.
15 But see Maj. Op. at 8 (agreeing with the “the Commonwealth’s view” that Hamlett
“seeks a novel decisional rule that would necessarily upend this conventional approach
to issue preservation by imposing waiver, or its functional equivalent, on the government
as appellee”).
[J-93-2019] [MO: Saylor, C.J.] - 39
The federal intermediate courts widely have adopted some version of the
Giovannetti test, but the Supreme Court has never endorsed its rubric. See, e.g., United
States v. Gonzales-Flores, 418 F.3d 1093, 1099-1101 (9th Cir. 2005); United States v.
Rose, 104 F.3d 1408, 1414-15 (1st Cir. 1997); United States v. Pryce, 938 F.2d 1343,
1347-48 (D.C. Cir. 1991) (plurality). I do not find the test persuasive. Even if all of its
factors counsel in favor of sua sponte review, it nonetheless suffers from the same
fundamental deficiency as the Majority’s analysis herein—it allows the court to excuse
and then assume a litigant’s burden of persuasion. Furthermore, the Giovannetti test is
not harmonious with our general approach to waiver in Pennsylvania. Although the
Majority summarily states that Giovannetti’s discretionary approach “is also the case in
the Pennsylvania state courts,” Maj. Op. at 10 n.7, we do not generally instruct our courts
to “overlook” a party’s waiver. Pennsylvania law treats waiver most strictly. Moreover,
there is a degree of internal tension within the Giovannetti test, inasmuch as avoiding a
“protracted” and “costly” retrial would appear to cut in favor of sua sponte review, but such
a retrial likely would be protracted and costly precisely because of the “length” and
“complexity” of the record—a factor that Giovannetti suggests would militate against sua
sponte review. Conversely, a short and simple record seemingly would suggest that sua
sponte review is more manageable, but for precisely that reason, a retrial would be
relatively less burdensome and costly, thus counseling against sua sponte review.
Regardless, the third Giovannetti factor—avoidance of “costly” but “ultimately
futile” retrials—is merely a reference to the concerns with judicial economy that undergird
the whole doctrine, not a helpful or case-specific consideration. The Majority refers in
passing to the second Giovannetti factor, which I take to be the essence of the test: that
the court should be “certain” that the error is harmless, and should not conduct the inquiry
if it is “debatable.” Giovannetti, 928 F.2d at 227. It remains mysterious to me how one
[J-93-2019] [MO: Saylor, C.J.] - 40
could purport to be certain that an error was harmless in the course of “deciding whether
to exercise [the court’s] discretion” to consider the doctrine sua sponte in the first place.
Id. One seemingly must be certain of an error’s harmlessness before even considering
the harmless error doctrine. The Majority suggests that the requirement of “certainty” is
a “more discerning review” than the court might apply if it was not proceeding sua sponte.
Maj. Op. at 12 n.10. I fail to see how elevating the standard to “certainty” is meaningfully
different than applying the beyond-a-reasonable-doubt standard that ordinarily governs
the inquiry. See supra Part I. We will never be “certain” of what twelve jurors would have
done if the trial had been different. The ordinary beyond-a-reasonable-doubt review is
essentially as “discerning” a standard as the law can provide. To demand the court’s
assertion of “certainty” means only that the court must utter a magic word before it
proceeds to do precisely what it would have done otherwise.
Contrary to what may be the tacit suggestion of either the Majority or the
Concurrence, or both, the Giovannetti test is not fit for this inquiry. A notable feature of
many decisions applying Giovannetti or inquiries like it in other jurisdictions is some
measure of caution in the approach, and some recognition of the potential downsides of
sua sponte review. See, e.g., Gonzales-Flores, 418 F.3d at 1101 (“[W]e are particularly
sensitive to the . . . concerns that sua sponte consideration of harmlessness will often
burden reviewing courts and give the government too many chances to argue harmless
error. Even more troubling, the practice may unfairly tilt the scales of justice by
authorizing courts to construct the government’s best arguments for it without providing
the defendant with a chance to respond. . . . We therefore conclude that sua sponte
recognition of an error’s harmlessness is appropriate only where the harmlessness of the
error is not reasonably debatable.”) (emphasis in original); People v. Sandoval, 363 P.3d
41, 78 (Cal. 2015) (“Courts in other jurisdictions have warned of ‘the dangers of allowing
[J-93-2019] [MO: Saylor, C.J.] - 41
sua sponte consideration of harmlessness,’ including ‘the potential burden on reviewing
courts of searching the record without guidance from the parties and encouragement of
sloppy practice by lawyers.’”) (quoting Gover v. Perry, 698 F.3d 295, 300 (6th Cir. 2012)).
Indeed, in Story, this Court similarly urged care and circumspection in the application of
harmless error analysis, albeit in response to a different set of concerns underlying the
doctrine. See Story, 383 A.2d at 164 (“[C]ourts must be careful in applying the harmless
error rule, for if the violation of a rule is too readily held harmless, the importance and
effectiveness of the rule is denigrated.”).
The Majority attempts to echo some of this caution, noting that many federal courts
generally find sua sponte harmless error review to be “extraordinary” and that it “should
be disfavored.” Maj. Op. at 12.16 The Majority urges that its approach “not be routinely
or liberally employed to impose a high barrier to reversal of criminal convictions.” Id. at
11. However, the Majority sets no meaningful standard, and it proposes no guardrails to
ameliorate the worst of the dangers of its holding. The Majority does not even attempt to
set forth any specific considerations along the lines of the Giovannetti test, flawed though
that test may be. The Majority seemingly has left us with among the most amorphous of
rules in the nation, leading the Concurrence to note the Majority’s reference to
Giovannetti, and to ponder a “question left unanswered” by the Majority. Conc. Op. at 2
(Donohue, J.). The question that the Concurrence believes the Majority has left
unanswered is not an insignificant one: “What is the relevant test for determining whether
an error affected the outcome when harmless error is invoked sua sponte?” Id. I wish to
16 Given the Majority’s recognition of the comments of federal courts as a general
admonition that the practice is extraordinary and should be disfavored, Maj. Op. at 12,
the Majority’s observation that federal courts are not “necessarily required to invoke
harmless-error precepts sua sponte,” id. at 10 n.7, is something of an understatement.
[J-93-2019] [MO: Saylor, C.J.] - 42
know the answer to this question as well. Unlike the Concurrence, however, the absence
of an answer to that question precludes me from supporting the Majority Opinion.
IV.
The Majority does not speak to the facts, and alludes only generally to the
substance of the Superior Court’s sua sponte application of the harmless error doctrine
in this case. See supra n.2; Maj. Op. at 1-2, 3 n.2. Although the Majority considers the
substance of the case before us to be outside the scope of the question presented, its
approval of sua sponte review for harmless error suggests that we may consider the
matter irrespective of the parties’ advocacy. I do not feel it improper, then, to exercise
my discretion to consider the very finding of harmless error that precipitated the instant
appeal. But we need not dwell on the matter, for only a small amount of additional detail
reveals significant problems.
The Majority relates that the Superior Court found error in the admission into
evidence of a video-recorded forensic interview with the victim of the alleged sexual
assault, but the Superior Court deemed the error harmless, sua sponte, because it found
the video to be merely cumulative of properly-admitted evidence in the form of the victim’s
testimony. Maj. Op. at 1-2. To reach that point, however, the Superior Court undertook
a thorough analysis of Pa.R.E. 61317 and related case law, and it found error in the
17 Rule 613 provides, in relevant part:
(c) Witness’s Prior Consistent Statement to Rehabilitate. Evidence of
a witness’s prior consistent statement is admissible to rehabilitate the
witness’s credibility if the opposing party is given an opportunity to cross-
examine the witness about the statement and the statement is offered to
rebut an express or implied charge of:
(1) fabrication, bias, improper influence or motive, or faulty memory and the
statement was made before that which has been charged existed or arose;
or
[J-93-2019] [MO: Saylor, C.J.] - 43
admission of the video because it contained a prior consistent statement offered merely
for the impermissible purpose of corroboration, and not to rehabilitate the credibility of a
witness who had been impeached. Hamlett, slip op. at 18-27, 2018 WL 4327391, at *8-
12. But when the court turned to harmless error, sua sponte, it held that the error was
harmless because the prior consistent statement was merely cumulative of the victim’s
testimony—the very testimony that it improperly sought to corroborate. Id. at 27-30, 2018
WL 4327391, at *12-14. How can an error be dismissed as merely cumulative, and thus
harmless, when its cumulative nature was the very reason that it was error? The Superior
Court stated that the same testimony that rendered the admission of the video erroneous
also simultaneously made the error of its admission harmless. Surely that cannot suffice
to allow an appellate court to conclude, beyond a reasonable doubt, that the absence of
the error—video corroboration of critical testimony—would not have influenced the jury’s
assessment of the credibility of competing evidence.
The court further noted that Hamlett testified in his own defense, and it did not
purport to base its conclusion upon overwhelming and uncontradicted evidence of
Hamlett’s guilt, but rather upon the “merely cumulative” variety of harmless error. Yet,
the court saw fit to predict the manner in which the jury would have viewed the credibility
of competing testimony, but for the error. Hamlett understandably asserted that the
impermissible corroboration of the victim’s in-court testimony may have prejudiced him in
the eyes of the jury, inasmuch as the video could tend to bolster the credibility of the
victim’s in-court testimony. The Superior Court disagreed:
(2) having made a prior inconsistent statement, which the witness has
denied or explained, and the consistent statement supports the witness’s
denial or explanation.
Pa.R.E. 613(c).
[J-93-2019] [MO: Saylor, C.J.] - 44
The jur[ors] heard from [the victim] under oath and their fundamental task
was to weigh the credibility of her story against that of [Hamlett], who
testified in his own defense. While we agree that our evidentiary rules prefer
to avoid prior consistent statements, as set forth at length supra, we do not
believe that the jury would be unduly swayed by the knowledge [the victim]
previously related the same story. Indeed, the jury presumptively assumed
that was the case. Thus, while there was no need to buttress her testimony,
the recorded statement was merely cumulative and harmless beyond a
reasonable doubt. [Hamlett] is therefore not entitled to a new trial despite
the error.
Hamlett, slip op. at 30, 2018 WL 4327391, at *14 (citation omitted).
This was not a sufficient application of any of the harmless error standards
discussed in Story. See supra n.6. The analysis of the cumulative nature of the evidence
was nonresponsive to the nature of the error, and the evidence that otherwise might have
been deemed “overwhelming” was expressly contradicted. It requires an appellate
credibility determination to say that an error such as this was harmless. To do so, the
appellate court must step into the juror box. The “superficiality” that the Majority rightly
criticizes is present in the very case before us. Maj. Op. at 12. Here, the “right-for-any-
reason doctrine” was indeed “routinely” and “liberally” employed to “impose a high barrier
to reversal” of a criminal conviction. Id. at 11. Hamlett merely points to ills in our
jurisprudence that appear plainly in the pages before us, which continue to echo the flaws
in cases like Mitchell and Hitcho.
Future courts and litigants will research this question, they will find our decision
today, and they will rely upon it. Particularly given this Court’s admonition that “decisions
are to be read against their facts,” Maloney v. Valley Med. Facilities, Inc., 984 A.2d 478,
489 (Pa. 2009),18 I can only presume that many of those courts and litigants will consult
18 See also Tincher v. Omega Flex, Inc., 104 A.3d 328, 378 (Pa. 2014) (“This case
speaks volumes to the necessity of reading legal rules—especially broad rules—against
their facts and the corollary that judicial pronouncements should employ due modesty.”)
(citing Maloney, 984 A.2d at 489-90).
[J-93-2019] [MO: Saylor, C.J.] - 45
the facts of this case and the lower court decision that we affirm, as they struggle to
determine how to apply the law that the Majority expounds today. When they do, they
will find a self-contradictory harmless error analysis, which weighs the credibility of
contradicted evidence, which eschews the safeguards designed precisely to avoid the
defects from which it suffers, and by which this Court apparently is not troubled. They will
see this Court’s approval and perpetuation of an oxymoronic statement of the law: that
the “harmless error doctrine places the burden on the Commonwealth to prove beyond a
reasonable doubt that the error could not have contributed to the verdict,” but that
simultaneously, “our jurisprudence does not require the Commonwealth to raise the
matter” at all. Hamlett, slip op. at 27, 2018 WL 4327391, at *12. Litigants and courts alike
now will offer pro forma quotations from today’s Majority Opinion in their briefs and
opinions, stating that the Commonwealth’s burden to establish harmless error is the
“ordinary rule,” Maj. Op. at 9, before moving immediately to the “exception,” i.e., the
court’s unbounded discretion to ignore the ordinary rule in “appropriate” cases. Id. My
impression would be that the Commonwealth’s burden is gone in all but name, and that
this Court does not plan to police the use of the harmless error doctrine any longer.
V.
Our harmless error jurisprudence has gone astray because we have eroded one
of its foundations. We have before us the opportunity to rein in the worst of its potential
for misuse. The problems are apparent in our own precedent and in the very case before
us. The Majority sees at least some of these problems, and it criticizes the “superficiality”
with which harmlessness is commonly discerned. Maj. Op. at 12. But today the Majority
declines the opportunity to do anything about it.
Instead, the Majority places all of these weighty considerations within the sole
discretion of the appellate court, but it offers no test, no set of factors, no interests to
[J-93-2019] [MO: Saylor, C.J.] - 46
balance. Indeed, the Majority’s analysis barely acknowledges that a burden exists, or
existed, at all. Its substantive discussion mentions the Commonwealth’s burden once,
which it characterizes as the “ordinary rule” to which discretionary sua sponte invocation
of harmless error serves as an “exception” to be used “in appropriate cases.” Id. at 9.
The Majority cautions the Commonwealth that it “proceeds at its own risk,” id. at 11 n.8,
when it declines to satisfy its burden, but the Majority does not meaningfully discuss the
consequence that might befall the Commonwealth should the court choose not to come
to its aid, nor does the Majority provide guidance as to when a court should decide that
the Commonwealth’s “risk” should become a loss. As far as I can tell from the Majority’s
holding, the appellate court may overlook the Commonwealth’s burden whenever it
wants. That is not a “burden” upon a litigant to establish harmless error. It is a suggestion,
at best; it is an invitation for arbitrariness in adjudication at worst. Our courts need and
deserve more guidance.
To that end, I reiterate that the most straightforward solution is simply to resuscitate
and then enforce the Commonwealth’s burden of persuasion on appeal. The
Commonwealth’s principal brief is the best place to carry its burden, thereby allowing the
appellant to address the Commonwealth’s argument in a reply brief in the ordinary course.
The Commonwealth is the best and most able advocate for its own position,
particularly as it concerns a highly record-intensive inquiry such as harmless error. Our
appellate courts are busy, and I see no compelling reason why the Commonwealth’s work
should ever fall to the court, for that does not advance judicial economy. Thus, should
the Commonwealth fail to advance a harmless error argument, or should it be expected
to fail so regularly as to require us to place a safety valve in our jurisprudence, I view a
focused briefing procedure to be the only possible solution that would achieve a more fair
balance of harms, relative to the alternative of standardless exercises in sua sponte
[J-93-2019] [MO: Saylor, C.J.] - 47
decision-making.19 Generally speaking, some type of briefing procedure tailored to the
harmlessness question could: (1) preserve the Commonwealth’s burden of persuasion;
(2) relieve the court of the weighty and ill-defined question of when it should take up the
matter sua sponte, mitigating the potential for arbitrariness and abuse along with it; and
(3) provide the appellant with notice and an opportunity to respond, thus going a long way
toward resolving the problems identified in this appeal. Whatever our approach, however,
it should be applied uniformly and consistently between cases, not in an ad hoc manner
that depends upon whether the court deems it “appropriate” to “enhance fairness to the
defendant” in a given case. Maj. Op. at 12.
A supplemental briefing procedure also would have clear downsides, in that it
would result in quite a bit of additional delay for appellants, who often are incarcerated
and eagerly awaiting resolution of their appeals. A significant increase in the volume of
briefs filed also surely would place an additional administrative burden upon our
prothonotaries. Accordingly, the Commonwealth’s principal brief remains the best place
for the argument, and the court should not have to invite the Commonwealth to carry its
burden. To the extent that our courts lack the will to enforce the Commonwealth’s burden
with the threat of reversal, however, I would not exclude the possibility that certain
deficiencies in our practice perhaps could be mitigated by a nuanced and informed
rulemaking process. The Court need only find the will to try.
I would insist upon a clear and predictable standard that calls for adversarial
presentation of argument to the court. Such a standard would require the party with the
burden to develop the issue, provide the party with the rights with an opportunity to
19 Notably, the Commonwealth presently advocates, in the alternative, for a
supplemental briefing procedure. Brief for Commonwealth at 54-56. The
Commonwealth’s alternative argument has not distracted me from analyzing the asserted
merits of its primary legal contentions. See supra at 27; Maj. Op. at 8 (citing Brief for
Commonwealth at 43, 50).
[J-93-2019] [MO: Saylor, C.J.] - 48
respond, and call upon the court to neutrally decide which side has the better of it. That
way, the court would not bear the weight of deciding whether to take all of these
responsibilities upon itself, and it could then decide the fate of cases based upon
advocacy, rather than judicial intuition or navel-gazing.
Because the Majority has a very different vision and arrives at a markedly different
conclusion, I respectfully dissent.
[J-93-2019] [MO: Saylor, C.J.] - 49