In the Interest of: J.M.G., a Minor ( 2020 )

[J-76-2019] [MO: Mundy, J.]
IN THE SUPREME COURT OF PENNSYLVANIA
MIDDLE DISTRICT
IN THE INTEREST OF: J.M.G., A MINOR : No. 18 MAP 2019
:
: Appeal from the Order of the
APPEAL OF: J.M.G. : Superior Court dated May 18, 2018,
: at No. 476 MDA 2017, affirming the
: Order of the Cumberland County
: Court of Common Pleas, Juvenile
: Division, dated March 15, 2017 at
: Nos. CP-21-CV-0003322-2017 &
: CP-21-JV-0000206-2014.
:
: ARGUED: September 12, 2019
CONCURRING OPINION
JUSTICE TODD DECIDED: April 22, 2020
I concur in the result. Like Justice Baer, I cannot join the majority’s determination
that violations of the Section 5944 privilege in the context of Act 21 proceedings are not
amenable to a harmless error analysis. Indeed, for the reasons below, the privilege
breach in this case can and should be analyzed for harmless error, as we treat all other
privilege breaches, most notably including breaches of the attorney-client privilege.
Nevertheless, I concur in the majority’s mandate as I conclude, for the reasons Justice
Baer sets forth in his Concurring Opinion, that the breach in this case was not harmless.
As the majority discusses, some errors, principally constitutional errors in criminal
cases, are deemed beyond analysis for harmlessness. As the United States Supreme
Court noted, “[t]hese errors came to be known as structural errors.” Weaver v.
Massachusetts, 137 S. Ct. 1899, 1907 (2017).1 Such errors are distinguishable from
“trial error,” because trial errors may “be quantitatively assessed in the context of other
evidence presented in order to determine whether [they were] harmless beyond a
reasonable doubt.” United States v. Gonzalez-Lopez, 548 U.S. 140, 148 (2006) (quoting
Arizona v. Fulminante, 499 U.S. 279, 307-308 (1991)) (alterations original). That is, for
trial errors, a reviewing court “can make an intelligent judgment” about whether the error
might have affected the fact-finder. Satterwhite v. Texas, 486 U.S. 249, 258 (1988). Most
errors, and indeed most constitutional errors, fall within this category. See Gonzalez-
Lopez, 548 U.S. at 148. The defining feature of structural errors, on the other hand, is
that they “defy analysis by ‘harmless-error’ standards because they affect the framework
within which the trial proceeds, and are not simply an error in the trial process itself.” Id.
(internal quotation marks and alteration omitted). Such errors “infect the entire trial
process.” Brecht v. Abrahamson, 507 U.S. 619, 630 (1993).2
In Weaver, the high Court set forth three “broad rationales” for finding that an error
is not amenable to a harmless error analysis, and thus is structural error. The first is
where the right at issue is designed to protect an interest other than the correctness of
the proceedings, and thus harm is irrelevant, such as the right of a criminal defendant to
conduct his own defense. Weaver, 137 S. Ct. at 1908. The second is where the effects
1 Despite the name, “the term ‘structural error’ carries with it no talismanic significance as
a doctrinal matter. It means only that the government is not entitled to deprive the
defendant of a new trial by showing that the error was ‘harmless beyond a reasonable
doubt.’” Weaver, 137 S. Ct. at 1910 (quoting Chapman v. California, 386 U.S. 18, 24
(1967)).
2 Although courts find such errors primarily in the criminal realm, we have deemed errors
in civil cases to be akin to structural errors. See, e.g., Bruckshaw v. Frankford Hospital
of City of Philadelphia, 58 A.3d 102, 113-14 & n.6 (Pa. 2012) (removal of presumptively
competent juror, by a court officer, without notice to the court, without notice to parties,
and substituted by last alternate juror was error for which prejudice was presumed,
“suggestive” of structural error).
[J-76-2019] [MO: Mundy, J.] - 2
of the error are too hard to measure – for example, denying the right of a criminal
defendant to choose his own counsel. Id. Finally, the third category is where the error
“always results in fundamental unfairness,” such as a defendant’s right to a beyond-a-
reasonable-doubt instruction. Id. However, the Court has stressed that the category of
structural errors is small, and that “if the defendant had counsel and was tried by an
impartial adjudicator, there is a strong presumption that any other errors that may have
occurred are subject to harmless-error analysis.” Rose v. Clark, 478 U.S. 570, 579
(1986).
The majority essentially concludes that breach of the Section 5944 privilege herein
falls into the last Weaver category, reasoning that “[s]imilar to the types of constitutional
errors the United States Supreme Court has deemed so basic to a fair trial that application
of the harmless error doctrine is inappropriate, we deem scrupulous adherence to the
psychotherapist-patient privilege to be basic to fair Act 21 proceedings.” Majority Opinion
at 18. While I agree that protecting the privilege is vital to ensuring effective mental health
treatment, I cannot agree that violation of the privilege necessarily results in an Act 21
proceeding that is fundamentally unfair.
Act 21 proceedings are fact-finding proceedings focused upon determining
whether a juvenile “has a mental abnormality or personality disorder which results in
serious difficulty in controlling sexually violent behavior that makes the person likely to
engage in an act of sexual violence.” 42 Pa.C.S. § 6403(d). But, as Justice Baer notes
in his concurrence, only a subset of the information potentially disclosed in a therapeutic
setting would have relevance to this fact-finding. See Concurring Opinion (Baer, J.) at 3-
4.
Moreover, in my view, breach of the Section 5944 privilege is not the type of error
that “affects the framework” within which Act 21 proceedings occur. Rather, it seems to
[J-76-2019] [MO: Mundy, J.] - 3
me that breach of the privilege as occurred here, like evidentiary errors committed in a
criminal trial, may be “quantitatively assessed,” Gonzalez-Lopez, 548 U.S. at 148, against
any other evidence presented in an Act 21 proceeding to determine the affect the breach
had on the relevant fact-finding. The majority does not contend otherwise.
Furthermore, I cannot agree with the majority that maintaining the Section 5944
privilege in the Act 21 context is on par with ensuring those rights – the bar on admitting
coerced confessions, the right to counsel, and the right to an impartial presiding judge –
the high Court in Chapman parenthetically identified as essential to a fair trial. See
Majority Opinion at 16 (citing Chapman, 386 U.S. at 23, as listing these three examples).
Initially, I note that the high Court later disavowed the first of these examples and held
that the introduction of a coerced confession is subject to a harmless error analysis,
ultimately viewing it as a routine, albeit serious, evidentiary error. See Arizona v.
Fulminante, 499 U.S. 279, 310 (1991) (“When reviewing the erroneous admission of an
involuntary confession, the appellate court, as it does with the admission of other forms
of improperly admitted evidence, simply reviews the remainder of the evidence against
the defendant to determine whether the admission of the confession was harmless
beyond a reasonable doubt.”).
More centrally, unlike violation of the right to counsel or the right to an impartial
trial judge, which have systemic impact, see Fulminante, 499 U.S. at 311 (deeming the
admission of an involuntary confession not to be the type of error which “transcends the
criminal process”), merely because Section 5944 privileged information has been
introduced does not imply that an Act 21 proceeding can no longer “reliably serve its
function” to determine whether the juvenile has a mental abnormality or personality
disorder necessitating treatment. See id. at 310 (observing that structural defects affect
framework within which a trial proceeds, such that the “trial cannot reliably serve its
[J-76-2019] [MO: Mundy, J.] - 4
function as a vehicle for determination of guilt or innocence”). Although the admission of
such privileged evidence manifestly undermines the privilege, with respect to the Act 21
proceedings themselves, it is much like any other evidence; its admission does not alter
the framework for the proceedings, alter how the judge performs his fact-finding function,
or undermine the ability of a reviewing court to assess its impact relative to the other,
properly admitted, evidence. Cf. Sullivan v. Louisiana, 508 U.S. 275, 281 (1993) (an
erroneous beyond-a-reasonable-doubt instruction is structural error because it “vitiates
all the jury's findings. A reviewing court can only engage in pure speculation” (emphasis
original)). The majority does not entertain this analysis.
Moreover, the majority’s holding elevates the treatment of the Section 5944
privilege above all other privileges. As the majority concedes, the Superior Court has
analyzed a breach of the Section 5944 privilege for harmlessness in the criminal context.
See Commonwealth v. Fewell, 654 A.2d 1109, 1115 (Pa. Super. 1995); Commonwealth
v. Flynn, 460 A.2d 816, 823 (Pa. Super. 1983). More critically, Section 5944 itself
instructs that this privilege “shall be on the same basis” as the attorney-client privilege.
42 Pa.C.S. § 5944. However, despite its status as “the most revered of the common law
privileges” Commonwealth v. Flor, 136 A.3d 150, 158 (2016), we nevertheless apply a
harmless error analysis to violations of this venerated privilege. See Commonwealth v.
Chmiel, 889 A.2d 501, 528-29 (Pa. 2005) (finding that trial court error in allowing
prosecutor to question defendant regarding prior statements whose only source were
confidential communication defendant had with prior attorney was harmless beyond a
reasonable doubt); see also Commonwealth v. Stantz, 509 A.2d 351, 356–57 (Pa. Super.
1986) (even if attorney-client privilege was infringed, no apparent harm resulted); see
generally United States v. Bauer, 132 F.3d 504, 510 (9th Cir. 1997) (holding that, because
[J-76-2019] [MO: Mundy, J.] - 5
violation of the attorney-client privilege is not an error of constitutional magnitude,
harmless error analysis applies).
Indeed, this approach is entirely consistent with how courts of this Commonwealth
treat all privilege breaches. See, e.g., Commonwealth v. Small, 980 A.2d 549, 562 (Pa.
2009) (violation of spousal privilege subject to harmless error analysis); Commonwealth
v. May, 656 A.2d 1335, 1343 (Pa. 1995) (same); Commonwealth v. Reese, 31 A.3d 708,
719 (Pa. Super. 2011) (same); Commonwealth v. Moyer, 595 A.2d 1177, 1180 (Pa.
Super. 1991) (admission of privileged Mental Health Procedures Act records subject to
harmless error analysis); Commonwealth v. Musolino, 467 A.2d 605, 610-11 (Pa. Super.
1983) (violation of priest-penitent privilege subject to harmless error analysis). Indeed,
my research did not disclose a single instance where a court found the violation of a
privilege not to be subject to a harmless error analysis.
The majority’s answer to this is that Act 21 proceedings are different from criminal
trials, “that the harm to the therapeutic relationship and the efficacy of mental health
treatment that Section 5944 is designed to protect[] is not entirely tangential to the factual
burden in Act 21.” Majority Opinion at 18. I am hard-pressed to agree with this implicit
contrast. In my view, the Section 5944 privilege’s centrality to an Act 21 proceeding, with
its focus on assessing the existence of a juvenile’s “mental abnormality or personality
disorder,” is not qualitatively different than the centrality of the attorney-client privilege to
a criminal trial, with its focus on guilt or innocence. In each case, the assisting
professional, by virtue of his role with respect to his patient/client, may hold information
relevant to the ultimate factual questions at issue, but is under a foundational obligation
to hold that information in confidence. And that confidence is critical to ensure the vitality
of the professional relationship. Yet, and again, despite the attorney-client privilege’s
revered status, harmless error doctrine is routinely applied to its breach.
[J-76-2019] [MO: Mundy, J.] - 6
Furthermore, the majority appears to suggest that its holding is based, in part, on
a deterrence rationale. See Majority Opinion at 18 (“Erosion of the privilege can only
complicate and adversely affect the fundamental rehabilitative goals of the juvenile
system and any treatment ordered under Act 21.”). I can find no support for the contention
that harmless error doctrine is founded on such a consideration. Cf. Weaver (setting forth
three bases for deeming an error not to be amenable to harmless error analysis, none of
which is deterrence). Certainly, the Section 5944 privilege should be safeguarded as it
is critical to effective mental health treatment. However, as the majority concedes, the
privilege has already been breached in the instant case, and new Act 21 proceedings, in
this or any similar future case, will not cure the breach. A new proceeding can only require
new fact-finding shielded from the privileged material. But a harmless error analysis may
achieve the equivalent, without the cost of a new proceeding, where a court is convinced
that the erroneously-admitted privileged evidence could not have contributed to the
court’s factual determinations.
Accordingly, for these reasons, I cannot agree with the majority’s conclusion that
the breach of the Section 5944 privilege in the context of Act 21 proceedings is per se
harmless; rather, like Justice Baer, I would apply a harmless error analysis. Nonetheless,
I thoroughly agree with Justice Baer’s assessment that the breach was not harmless in
this case. See Concurring Opinion (Baer, J.) at 5-6. Thus, I join his Concurring Opinion,
and concur with the majority’s mandate to remand this matter for a new Act 21 hearing.
Justice Baer joins this concurring opinion.
[J-76-2019] [MO: Mundy, J.] - 7